Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Functions of kindlin-1 in human keratinocytes and its immunolocalization in human oral mucosal tissues Petricca, Giorgio Maximillian 2005

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

Item Metadata

Download

Media
831-ubc_2005-0594.pdf [ 12.78MB ]
Metadata
JSON: 831-1.0092142.json
JSON-LD: 831-1.0092142-ld.json
RDF/XML (Pretty): 831-1.0092142-rdf.xml
RDF/JSON: 831-1.0092142-rdf.json
Turtle: 831-1.0092142-turtle.txt
N-Triples: 831-1.0092142-rdf-ntriples.txt
Original Record: 831-1.0092142-source.json
Full Text
831-1.0092142-fulltext.txt
Citation
831-1.0092142.ris

Full Text

Functions of Kindlin-1 in Human Keratinocytes and its Immunolocalization in Human Oral Mucosal Tissues  By  GIORGIO M A X I M I L L I A N PETRICCA BSc, The University of British Columbia M S c , The University of British Columbia D M D , The University of British Columbia  A THESIS IN P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T FOR THE D E G R E E OF M A S T E R OF SCIENCE In T H E F A C U L T Y OF G R A D U A T E STUDIES (Dental Science)  T H E UNIVERSITY OF BRITISH C O L U M B I A July 2005  © Giorgio Maximillian Petricca 2005  11  ABSTRACT  The primary protein structure of the kindlin-1 polypeptide reveals a number of features predicting that the function of this molecule relates to anchorage of the actin cytoskeleton to the plasma membrane. Kindlin-1 has C-terminal homology to talin and N-terminal homology to filopodin. Kindlin-1 also possesses a centrally-located pleckstrin homology (PH) domain and two regions of homology with the F E R M (filopodin and ezrin/radixin/moesin) domain. Interestingly, kindlin-1 is homologue of Mig-2, a focal adhesion protein, which recruits migfilin, an actin-binding protein, to ECM-cell adhesions. Cells probe, respond to, and remodel the E C M using integrin-actin cytoskeleton adhesion complexes. Identification of the molecular components of cell-cell and cell-ECM adhesions and the structural determinants that control their localizations to these two distinct structures are therefore of critical importance to our understanding of tissue morphogenesis, cell growth, and differentiation in various biological processes including cancer, wound healing, and development. In order to increase our understanding of how kindlin-1 localizes to these ECM-cell adhesions and how knockdown of kindlin-1 expression alters keratinocyte cell behaviour, human and Kindler syndrome oral mucosa was stained for immunolocalization of kindlin-1 and siRNAmediated gene silencing was used to knockdown it's expression in HaCaT cells. Normal human oral palatal and gingival mucosa demonstrated localization of kindlin-1 and two of its potential binding partners, migfilin and paxillin, to the basement membrane zone (BMZ). Only migfilin and paxillin immunolocalized to blood vessels. In junctional epithelium from a non-periodontally diseased tooth, migfilin, paxillin, and kindlin-1 immunolocalized to the B M Z of the external basal lamina (EBL) and the internal basal  Ill  lamina (IBL) zone against the tooth with ILK-1 and Mig-2 also localizing to the IBL zone. Both immunolocalizations suggest a possible interaction between paxillin, migfilin, and kindlin-1 and possibly a role in maintaining the integrity of the epithelial attachment against a tooth. In HaCaT cells, actin filaments terminated at sites of migfilin and kindlin1 focal adhesions and knockdown of kindlin-1, via siRNA-mediated gene silencing, resulted in decreased localization of kindlin-1 and some of it's potential binding partners, ILK-1, Mig-2, paxillin, and actin. Knockdown of kindlin-1 was confirmed by RT-PCR analysis (approximately 70%) and resulted in significant reduction in HaCaT cell spreading on different E C M matrices, migration during in vitro wounding, and proliferation when grown in serum. To conclude, the present study provides sufficient evidence to suggest that a loss of kindlin-1 likely results in altered keratinocyte cell adhesion at the level of the basement membrane, which clinically results in skin fragility and photosensitivity as well as aggressive periodontitis.  iv  TABLE OF CONTENTS Page ABSTRACT  11  T A B L E OF CONTENTS  iv  LIST OF T A B L E S  Vll  LIST OF FIGURES  Vlll  ABBREVIATIONS  xi  ACKNOWLEDGEMENTS  xm  INTRODUCTION CHAPTER  I  Review of the Literature  1.1  Kindler Syndrome - A Historical Perspective on its Diagnosis  3  1.2  Kindler Syndrome - Histological Features  5  1.3  Kindler Syndrome - The Genetic Mutation  11  1.4  Cell-Cell and Cell-Matrix interactions  16  1.5  Kindlin-1, a Novel Focal Adhesion Protein, its Homologue, MIG-2, and Migfilin  21  1.6  Focal Adhesion Proteins, ILK-1 and Paxillin - Potential Signaling Partners of Kindlin-1  27  CHAPTER  II  A i m of the Study and Hypotheses  35  CHAPTER  III  Materials and Methods  36  3.1  Reagents  36  3.2  Experimental In Vivo Oral Mucosal Wounding  37  3.3  Immunofluorescent and Immunohistological Staining of Frozen Sections  38  3.4  Cell Culture  41  CHAPTER  III  Materials and Methods Page  3.5  Kindlin-1 SiRNA Transfection of HaCaT cells  41  3.6  Semi-quantitation of Kindlin-1 mRNA by RT-PCR  42  3.7  Immunocytochemistry - Localization of ILK-1, Mig-2, Migfilin, Kindlin-1, Paxillin, and Actin  43  3.8  Cell Migration - In Vitro Wounding  45  3.9  Cell Spreading Assay  46  3.10  Cell Proliferation Assay  47  3.11  Statistical Analysis  48  IV  49  CHAPTER  Results  4.1.1  Immunolocalization of ILK-1, Mig-2, Migfilin, Paxillin, and Kindlin-1 in Human Oral Mucosa  49  4.1.2  Immunolocalization of ILK-1, Mig-2, Migfilin, Paxillin, and Kindlin-1 in Human Junctional Epithelium  51  4.1.3  Immunolocalization of Type IV Collagen, ILK-1, Mig-2, 53 Migfilin, Paxillin, and Kindlin-1 in Human Oral Palatal Epithelium at Day "0" of Wounding and Day "3" after Wounding  4.2  Immunolocalization of Actin, ILK-1, Mig-2, Migfilin, and Kindlin-1 in HaCaT Keratinocytes  56  4.3  RT-PCR Confirmation of Kindlin-1 Knockdown via siRNA-mediated Gene Silencing of KIND 1  59  4.4.1  Effect of Kindlin-1 siRNA Transfection on the Expression of Actin, ILK-1, Mig-2, Migfilin, and Kindlin-1 in HaCaT Keratinocytes  60  4.4.2  Effect of Kindlin-1 siRNA transfection on the Expression of Actin, 65 ILK-1, Mig-2, Migfilin, and Kindlin-1 in HaCaT Cells Spread on Type I Collagen, Laminin 10/11 and Fibronectin  VI  CHAPTER  IV  Results Page  4.4.3  Effect of Kindlin-1 siRNA Transfection on HaCaT Cells Spreading on Type I Collagen, Laminin 10/11 and Fibronectin  70  4.4  Effect of Kindlin-1 siRNA Transfection on HaCaT Cell Proliferation and Migration  76  V  79  CHAPTER  Discussion  5.1  Immunolocalization of ILK-1, Mig-2, Migflin, Paxillin, and Kindlin-1 in Normal and Wounded Oral Mucosa  79  5.2  Immunolocalization of ILK-1, Mig-2, Migflin, Paxillin, and Kindlin-1 in HaCaT cells and the effect of Kindlin-1 siRNA Transfection  87  5.3  Kindlin-1- and Integrin-mediated ECM-cell Signaling in HaCaT Cell Spreading on Type I Collagen, Laminin 10/11, and Fibronectin  90  5.4  Knockdown of Kindlin-1 decreases HaCaT Cell Spreading, Proliferation, and Migration  93  5.5  Limitations of the Study  98  CHAPTER  VI  Conclusions  99  CHAPTER  VII  Future Directions  100  BIBLIOGRAPHY  101  Vll  LIST OF TABLES  Page  Table 1:  Reported Clinical Features and Differential Diagnoses of Kindler Syndrome  5  Table 2:  Summary of Reported Histological Findings for Kindler Syndrome  11  Table 3a: Salient Characteristics of Focal Adhesion Proteins Investigated  32  Table 3b: Salient Characteristics of Focal Adhesion Proteins Investigated  33  Table 4:  List of Primary Antibodies Used for Immunofluorescent and Immunohistochemical Staining  40  Table 5:  List of Secondary Antibodies Used for Immunofluorescence  40  Table 6:  Localization and Relative Staining Intensity of ILK-1, Mig-2, Migfilin, Paxillin, and Kindlin-1 Focal Adhesion Proteins in Oral Gingival and Palatal Epithelium, 3 Day-old Palatal Wounds, and Junctional Epithelium of a Non-Periodontally Diseased Tooth.  55  LIST OF FIGURES  Page  Figure 1:  Schematic Diagram of the Kindlin-1 Protein with F E R M and P H 15 Domains as well as Regions of Homology with Filipodin and Talin. Locations of reported positions of all loss-of-function mutations within the KPND1 gene, including nonsense, frameshift or splice mutations (indicated by black arrows; Jobard et al., 2003; Siegel et al., 2003; Ashton, 2004).  Figure 2:  Drawing Depicting Proposed Interaction between Focal Adhesion Proteins  Figure 3:  Immunolocalization (VIP Staining) of ILK-1 (B,H), Mig-2 (C,I), 50 Migfilin (D,J), Paxillin (E,K), and Kindlin-1 (F,L) in Human Oral Palatal Epithelium (B-F) with a Magnified View of their Localization in Blood Vessels (H-L).  Figure 4:  Immunolocalization (VIP Staining) of ILK-1 (B,H), Mig-2 (C,I), Migfilin (D,J), Paxillin (E,K), and Kindlin-1 (F,L) in Human Oral Junctional Epithelium (JE).  Figure 5:  Immunolocalization (Alexa Fluor® 546 Immunofluorescent Staining) 54 of Type IV Collagen (A,G), ILK-1 (B,H), Mig-2 (C,I), Migfilin (D,J), Paxillin (E,K), and Kindlin-1 (F,L) in Human Oral Palatal Epithelium at Day "0" of Wounding (A-F) and Day "3" after Wounding (G-L).  34  52  Figure 6:  Immunolocalization of Filamentous Actin (C,D) and Focal Adhesion Proteins ILK-1 (E,F), Mig-2 (G,H), Migfilin (I,J), and Kindlin-1 (K,L) in HaCaT Keratinocytes Spread for 16 hours (overnight), in the presence of 10 ng/mol TGF-P (A,C,E,G,I,K) or Control (no TGF-P; B,D,F,H,J,L).  57  Figure 7:  Migfilin and Actin Immunofluorescent Double Staining. HaCaT keratinocyte spread for 16 hours (overnight) with double staining of migfilin (Alexa Fluor® 488; Green) and actin (Rhodamine; Orange/Red) in the presence of 10 ng/mol TGF-p.  58  Figure 8:  Kindlin-1 and Actin Immunofluorescent Double Staining. •HaCaT keratinocyte spread for 16 hours (overnight) with double staining of kindlin-1 (Alexa Fluor® 488; Green) and actin (Rhodamine; Orange) in the presence of 10 ng/mol TGF-p.  58  Page Figure 9:  Figure 10:  RT-PCR for the Evaluation of Kindlin-1 mRNA Reduction by siRNA Mediated Gene Silencing of KIND 1.  59  Immunolocalization of Filamentous Actin (A,B) and Focal 62 Adhesion Proteins ILK-1 (C,D), Mig-2 (E,F), Migfilin (G,H), Paxillin (I,J), and Kindlin-1 (K,L) in Kindlin-1 siRNA Transfected (A,C,E,G,I,K) HaCaT Keratinocytes in the Presence of 10 ng/mol TGF-p.  Figure 11a) and b): Spreading of HaCaT Keratinocytes over 24 Hours. Cells were 63 transfected with kindlin-1 siRNA or F-RNA (control) for 48 hours, trypsinized, and then allowed to spread for 24 hours on glass coverslips in the presence of 10 ng/mol TGF-p. The total number of cells attached (a) and the percentage of spread cells of the total number attached (b) was calculated. Figure 12a) and b): Spreading of HaCaT keratinocytes over 48 hours. Cells were 64 transfected with kindlin-1 siRNA or F-RNA (control) for 48 hours and allowed to spread during this period of transfection on glass coverslips in the presence of 10 ng/mol TGF-P. The total number of cells attached (a) and the percentage of spread cells of the total number attached (b) was calculated. Figure 13:  Immunolocalization of ILK-1 (A-F) and Mig-2 (G-L) in Kindlin-1 siRNA Transfected (A,C,E,G,I,K) HaCaT Keratinocytes. Controls (B,D,F,H,J,L) were transfected with Stealth™ R N A i negative control (N-RNA). Cells were transfected for 48 hours, trypsinized, seeded, and then allowed to spread for 4 hours on 20 ug/ml type I collagen (A,B,G,H), 5 pg/ml laminin 10/11 (C,D,I,J), and 20 ug/ml bovine fibronectin (E,F,K,L).  67  Figure 14:  Immunolocalization of Migfilin (A-F) and Paxillin (G-L) in 68 Kindlin-1 siRNA Transfected (A,C,E,G,I,K) HaCaT Keratinocytes. Controls (B,D,F,H,J,L) were transfected with Stealth™ R N A i negative control (N-RNA). Cells were transfected for 48 hours, trypsinized, seeded, and then allowed to spread for 4 hours on 20 fig/ml type I collagen (A,B,G,H), 5 ug/ml laminin 10/11 (C,D,I,J), and 20 ug/ml bovine fibronectin (E,F,K,L).  Page Figure 15:  Immunolocalization of Kindlin-1 (A-F) in Kindlin-1 siRNA 69 Transfected HaCaT keratinocytes. Controls (B,C,F) were transfected with Stealth™ R N A i negative control (N-RNA). Cells were transfected for 48 hours, trypsinized, seeded, and then allowed to spread for 4 hours on 20 [ig/ml type I collagen (A,B), 5 pg/ml laminin 10/11 (C,D), and 20 p.g/ml bovine fibronectin (E,F).  Figure 16:  Effect of Kindlin-1 siRNA Transfection on Short-term HaCaT Cell Spreading. Cells were transfected with kindlin-1 siRNA or negative control (N-RNA) for 48, trypsinized, seeded, and then allowed to spread for 45 minutes and 60 minutes on 20 pg/ml type I collagen.  72  Figure 17:  Effect of Kindlin-1 siRNA Transfection on Short-term HaCaT Cell Spreading. Cells were transfected with kindlin-1 siRNA or negative control (N-RNA) for 48, trypsinized, seeded, and then allowed to spread for 45 minutes and 60 minutes on 5 u.g/ml laminin 10/11.  73  Figure 18:  Effect of Kindlin-1 siRNA Transfection on Short-term HaCaT Cell Spreading. Cells were transfected with kindlin-1 siRNA or negative control (N-RNA) for 48, trypsinized, seeded, and then allowed to spread for 45 minutes and 60 minutes on 20 u.g/ml bovine fibronectin  74  Figure 19:  Effect of Kindlin-1 siRNA Transfection on Short-term HaCaT Cell Spreading. Cells were transfected with kindlin-1 siRNA or negative control (N-RNA) for 48, trypsinized, seeded, and then allowed to spread for 4 hours on 20 pg/ml type I collagen-, 5 u.g/ml laminin 10/11-, and 20 p.g/ml bovine fibronectin-coated glass coverslips.  75  Figure 20:  Proliferation of Kindlin-1 siRNA Transfected HaCaT Keratinocytes 77 using the Promega CellTiter 96® Non-radioactive Cell Proliferation Assay.  Figure 21a) andb): Effect of Kindlin-1 siRNA Transfection on HaCaT keratinocyte Migration.  78  Figure 22:  86  Visual Summary of the Ultrastructural and Molecular Features of Kindler Syndrome in Comparison to Normal Mucosa.  ABBREVIATIONS  a.a  amino acid  ANOVA  analysis of variance  BLAST  basic local alignment search tool  BMZ  basement membrane zone  Bov-FN  bovine fibronectin  BSA  bovine serum albumin  cDNA  cloned D N A  DMEM  dulbecco's modified Eagle's medium  DNA  deoxyribonucleic acid  dNTP  deoxyribonucleotide triphosphate  ECM  extracellular matrix  EDA  extra Domain A  EDB  extra Domain B  FBS  fetal Bovine Serum  FCS  fetal Calf Serum  FN  fibronectin  F-RNA  BLOCK-iT™ Fluorescent oligo siRNA  H &E  hematoxylin and eosin  HaCaT  human immortalized adult skin keratinocyte cell line  HB-EGF  heparin-binding-epidermal growth factor related peptide  HEPES  4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid  IgG  immunoglobulin G  ABBREVIATIONS ILK-1  integrin-linked kinase-1  JE  junctional epithelium  kb  kilobases  KIND  kindlin gene  KS  kindler syndrome  mAbs  monoclonal antibodies  MIG-2  mitogen-inducible gene-2  mRNA  messenger R N A  N-RNA  Stealth™ R N A i Negative Control  OCT  optimal cutting temperature  PBS  phosphate-buffered saline  PCR  polymerase chain reaction  RNA  ribonucleic acid  RT-PCR  reverse transcriptase- polymerase chain reaction  s.e.m.  standard error of the mean  SiRNA  small interfering R N A  TGF-P  transforming growth factor-beta  TRITC  tetramethylrhodamine isothiocyanate  VIP  Vector® VIP immunoperoxidase stain  X  times  Xlll  ACKNOWLEDGEMENTS  I would like to thank Drs. Hannu Larjava and Lari Hakkinen for their guidance, expertise, and continuing support throughout my studies. I would also like to thank Dr. Ed Putnins who encouraged me to pursue periodontics and for his useful input and words of wisdom.  I am very thankful for the works of Dr. Guoqiao Jiang who performed the siRNA transfection of kindlin-1 and RT-PCR analysis and the advice and input of Dr. Leeni Koivisto.  I would like to thank Dr. Chuanyue Wu of the Department of Pathology, The University of Pittsburgh (Pittsburgh, P A , U S A ) , for his kind donation of antibodies for Mig-2, migfilin, and kindlin-1.  I would also like to thank Mr. Christian Sperantia for his technical support.  Lastly, to my parents, I thank them for their love, support, tolerance, and patience in my pursuit of achievement and excellence throughout life.  1 INTRODUCTION Periodontal diseases affect over half the adults in the U.S. (Tanner et al., 2005). In a homogeneous group of subjects in Bocas del Toro, Panama, patients with Kindler syndrome (KS) have periodontitis with an early onset (teenage years) and with its rate of progression rapid, resembling aggressive periodontitis, compared to non-Kindler individuals of the same geographic and ethnic origin (Wiebe et al., 2003). Since Kindler's first report (Kindler, 1954), more than 100 cases have been described (Ashton, 2004a). Recently, two separate research groups identified the gene responsible for K S localized on the short arm of chromosome 20, designated as kindlerin by one group (Jobard et al., 2003) and KFND1 by another (Siegel et al., 2003). B L A S T analysis (Basic Local Alignment Search Tool, Altschul et al., 1990), a software tool to view sequences aligned with each other or to find homology, shows that kindlin-1, the protein expressed by the KIND1 gene, has C-terminal homology to talin and N-terminal homology to filopodin. Both talin and filopodin are involved in anchorage of the actin cytoskeleton in focal contacts (Critchley, 2000), which are membrane-extracellular matrix (-ECM) attachment structures observed in cultured cells (Zamir and Geiger, 2001a,b). Interestingly, kindlin-1 demonstrates closest homology (30% a.a identity) to the Caenorhabditis elegans (C. elegans) protein UNC-112 (Siegel et al., 2003), which in turn has closest homology with Mig-2 in humans. Kindlin-1 has 62% a.a identity with Mig-2 (Tu et al., 2003). Small interfering R N A (siRNA) is a novel technique used to silence specific genes of interest as a means of investigating their function (Elbashir et al., 2002; Harmon, 2002; Hudson et al., 2002). Using this siRNA, one group showed that Mig-2 localizes to cell-ECM adhesions and is essential for cell shape modulation (Tu et al., 2003), as are  2 other focal adhesion proteins. The authors further elucidated mechanisms by which Mig-2 regulates cell shape and found Mig-2 to interact with migfilin, also a novel focal adhesion protein. The N-terminus of migfilin binds to filamin, a component of actin filaments, which suggest that the interaction with filamin likely mediates the association of migfilin with actin filaments (Tu et al., 2003). As one begins to put the pieces of this complex puzzle of interacting focal adhesion proteins together, the question arises, how does kindlin-1 fit into the  puzzle? In the  present study, we examined the  immunolocalization of kindlin-1 along with it's homologue, Mig-2, and some of its potential binding partners, ILK-1, paxillin, migfilin, and actin in attempt to put the pieces of the extracellular matrix (ECM)-cell interaction puzzle together. At a cellular level, using the aforementioned siRNA-mediated gene silencing technique we sought to examine how knockdown of kindlin-1 expression would affect the behaviour of human keratinocytes, using the HaCaT cell as a model. The hypothesis is that kindlin-1 colocalizes with one or more of its potential binding partners and that knockdown of kindlin-1 will decrease cell spreading and migration. In further examining how kindlin-1 fits into the ECM-cell interaction puzzle, it is our goal to increase our understanding of the molecular mechanisms behind the clinical presentation of Kindler syndrome in both the skin and the mouth.  3  CHAPTER I Review of the Literature 1.1  Kindler Syndrome - A Historical Perspective on its Diagnosis Kindler syndrome (KS) was first described in 1954 by Theresa Kindler. Kindler  described K S as a progressive poikiloderma with marked diffuse cutaneous atrophy associated with congenital acral bullae and photosensivity in a 14-year old girl (Kindler, 1954). Poikiloderma is extra pigmentation of the skin demonstrating a variety of shades and associated with widened capillaries (telangiectasia) in the affected area. Congenital acral bullae are blisters found on a individual's palms and soles at the time of birth. Later in life, Kindler syndrome is characterized by generalized poikilodermatous changes and cutaneous atrophy. The first patient described by Theresa Kindler demonstrated gingivae which readily bled, skin of the dorsa of her hands and feet having a thin, wrinkled appearance, and webbing between the second and third toes on both feet. By the age of 10, the blistering and sun sensitivity had resolved, but the skin remained thin and fragile. In the original case and other cases of Kindler syndrome, gingival fragility as well as early and rapidly progressive periodontitis (aggressive periodontitis) are common features (Wiebe et al., 2003). Other features vary between cases and include photosensivity, acral hyperkeratosis, nail dystrophy, webbing and contractures of the fingers and toes, alopecia, actinic changes, and mucosal involvement including urethral, vaginal, anal, esophageal, and oral commissural stenosis (narrowing or constricting) as well as ectropion (eversion) of the eyelids, pigmentation of the lips, and onychodystrophy (malformation of the nails). Common dermatological findings include: congenital skin blistering that resolves slowly with age, mild photosensitivity that improves with age, and  4 early, generalized, progressive poikiloderma with extensive atrophy (Patrizi et al., 1996). Since Kindler's first report (Kindler, 1954), more than 100 cases have been described, but clinical overlap with hereditary acrokeratotic poikiloderma (an autosomal disorder described by Weary et a l , 1971) and the inherited skin blistering disease, dystrophic epidermolysis bullosa, has often caused confusion in making a clear diagnosis (Forman et al., 1989). Differential diagnoses thus include acrokeratotic poikiloderma (WearyKindler syndrome), hereditary sclerosing poikiloderma, Rothmund-Thomson syndrome, xeroderma pigmentosum, dyskeratosis congenita (Zinsser-Cole-Engman syndrome), Da Costa's  syndrome,  dermatopathia  pigmentosa reticularis,  Franceschetti-Jadassohn  syndrome, Degos-Touraine syndrome, and Braun-Falco-Marghescu syndrome, the clinical features of which all have been discussed in detail previously. Kindler syndrome and Weary-Kindler syndrome were reported to be variants of the very same disease, but have also been perceived as distinctive entities by numerous other authors (Lanschuetzer et al., 2003). A summary of clinical features and differential diagnoses for K S is provided in Table 1.  5 Table 1:  Reported Clinical Features and Differential Diagnoses of Kindler Syndrome  Most Commonly Reported Clinical Features  Reference  Thin, wrinkled appearance; webbing between second and third rows on both feet. By age 10: blistering and skin sensitivity resolved, but skin thin and fragile  Kindler, 1954  Photosensitivity, acral hyperkeratosis, nail dystrophy, webbing and contractures of fingers and toes, alopecia, actinic changes, mucosal involvement (urethral, vaginal, anal, esophageal, and oral commisure stenosis), ectropion of eyelids, pigmentation of lips, onychodystrophy  Patrizi etal., 1996  Congenital skin blistering resolving slowly with age; mild photosensitivity improving with age; early and generalized, progressive poikiloderma and extensive atrophy  Patrizi etal., 1996  Gingival fragility and early and rapidly progressive (aggressive periodontitis)  periodontitis  Differential Diagnoses Acrokeratotic poikiloderma (Wear-Kindler Syndrome) Heriditary Sclerosing Poikiloderma Rothmund-Thomson Syndrome Xeroderma Pigmentosum Dyskeratosis Congenita (Zinsser-Cole-Engman Syndrome) Da Costa's Syndrome Dermatopathia Pigmentosa Reticularis Franceschetti-Jadassohn Syndrome Degos-Touraine Syndrome Braun-Falco-Marghescu Syndrome  1.2  Wiebe et al., 2003  Reference  Lanschuetzer et al., 2003  Kindler Syndrome - Histological Features Identification of the gene mutated in dystrophic epidermolysis bullosa, COL7A1  encoding type VII collagen, has been excluded as the cause for Kindler syndrome by linkage and mutation studies (Shimizu et al., 1997; Yasukawa et al., 2002). Inheritance of Kindler syndrome seems to be autosomal-recessive and consanguinity has been found in over half of the families, although O M I M indicates autosomal-dominant transmission (MIM 173650). Genotype-phenotype correlations are difficult to assess, because of the variability of clinical severity both between patients at different ages and among patients  within the same family with the same mutation (Siegel et al., 2003). In view of the skin blistering observed in younger patients with Kindler syndrome, one author speculates that kindlin-1 (protein encoded by the Kindlin gene) is a structural molecule, although this mechanical defect could also be secondary to a signaling or organizational/recruitment role (Siegel et al., 2003). A n example of this organizational/recruitment role has been suggested for Mig-2 (mitogen-inducible gene 2), a focal adhesion protein believed to serve as a "docking protein" for the recruitment of migfilin, another focal adhesion protein, which in turn is linked to actin in the cytoskeleton (Tu et al., 2003; Wu et a l , 2004). Interestingly, Mig-2 has 62% a.a identity with kindlin-1 (Siegel et a l , 2003). Histologically, skin lesions of patients affected by Kindler syndrome are characterized by epidermal atrophy, focal vacuolization of the basal layer of the epithelium and pigmentary incontinence in the upper dermis in addition to a mild lymphocytic infiltrate, consistent with poikiloderma. Electron microscopy shows extensive reduplication and disruption of the lamina densa along the dermoepidermal junction beneath the basal cells, and cleft formation, which occurs in the lamina lucida (Wiebe et a l , 1999; Yasukawa et al., 2002; Jobard et al., 2003; Forman et al., 1989). Desmosomes, hemidesmosomes, tonofilaments, anchoring filaments and fibrils appear normal. In immunofluorescence studies by Wiebe and Larjava (1999), the basement membrane components including the integrins (a3pi and a6p4) are normal, with the exception of type VII collagen that is found in abnormal locations deep in the connective tissue stroma. Immunofluorescent labeling using a panel of basement membrane structural protein antibodies does not show any reduction or major alteration in staining intensity. However, according to Shimizu et al. (1997), there is broad, reticular labeling at  7 the dermal-epidermal junction with anti-type IV and VII collagen antibodies, in keeping with the ultrastructural findings of lamina densa reduplication. The authors suggested that such abnormalities of adhesion molecules linking the lamina densa to basal cells provide the pathological mechanism for Kindler syndrome (Shimizu et a l , 1997). It is possible that minor differences in the aforementioned findings can be explained by the fact that the group of Wiebe and Larjava study oral epithelium, which is unique to the preponderance of literature on Kindler syndrome, whereas other investigators have investigated dermal epithelium and/or their connective tissue stroma. In a recent review article on Kindler syndrome (Ashton, 2004a), the author describes clinical features including poikiloderma with hyperkeratosis, epidermal atrophy, hyper- and hypopigmentation and telangiectasiae. Histologically, the same author shows that areas of cleavage at or close to the dermal-epidermal junction may be present (Ashton et al., 2004b) in Kindler syndrome. These may be within the basal keratinocyte layer or beneath it. Additional features may include numerous melanophages (depending on skin type) and disruption of collagen and elastic fibres in the papillary dermis (Hovnanian et al., 1989; Patrizi et al., 1996). Despite the clinical similarities, there are marked ultrastructural differences between dystrophic epidermolysis bullosa and Kindler syndrome skin (Shimizu et al., 1997; White et al., 2005). Specifically, in dystrophic epidermolysis bullosa there is a plane of cleavage beneath the lamina densa with morphological abnormalities in anchoring fibrils. However, Kindler syndrome skin may show multiple planes of split within basal keratinocytes and/or in the lamina lucida, as well as extensive reduplication of the lamina densa (Patrizi et al., 1996; Hovnanian et al., 1989; Shimizu et al., 1997; White et al., 2005). This latter feature, although not  8 specific to Kindler syndrome, is useful in distinguishing the disorder from dystrophic epidermolysis bullosa. Before identification of mutations in the KIND1 gene, the ultrastructural reduplication of the lamina densa was one of the most useful clues in diagnosing Kindler syndrome. One of the purposes of the present investigation surrounding this thesis is to provide insight into the localization of kindler protein (kindlin-1) and some of its homologues in oral epithelium. Recently, there has been a report that the diagnosis of Kindler's syndrome can be unequivocally established on characteristic immunofluorescence (IF) antigen mapping (Hintner et al., 1981; Lanschuetzer et al., 2003) and typical transmission electron microscopy (TEM) findings, which agree with the observations published by Shimizu et al. (1997). However, the study was based on a case-report (Lanschuetzer et al., 2003). In the report, histologic evaluation revealed an atrophic epidermis with vacuolar degeneration along the basal cell layer. Apoptotic keratinocytes and keratin (cytoid and colloid) bodies were found in epidermal and dermal locations; the PAS-positive basement membrane (BM) showed irregularities in thickness. Melanophages were located along the dermoepidermal junction zone. Elastotic globules due to degeneration of elastic fibers were found throughout the papillary dermis in the elastica-stained section; normal elastic fibers were missing. Immunofluorescence studies revealed focally an extensively broadened, partly reticular staining pattern in the dermo-epidermal basement membrane zone (BMZ) with antibodies against laminin-5 and type IV as well as type VII collagen. Anti-a6 and 04 integrin staining revealed small gaps in the linear reactivity in the B M Z . Abundant keratin bodies, as detected by antiimmunoglobulin M (IgM) staining, were focally present in the dermis, indicating prominent epidermal apoptosis. The authors  9 (Lanschuetzer et al., 2003) verified these findings using a histochemical apoptosis stain [terminal deoxynucleotidyl transferase-mediated  dUTP nick-end labeling (TUNEL)  reaction]. The most striking ultrastructural characteristic in the skin specimen from the inner aspect of the upper arm was an excessive reduplication of the otherwise normalappearing lamina densa with branching, folding, and formation of circles and loops. There were also areas where the lamina densa was attached to basal keratinocytes by normal hemidesmosomes with mature dense plates, and desmosomes, tonofilaments, and anchoring fibrils were completely normal. The lamina lucida appeared to be focally widened; possibly indicating the beginning of subepidermal split formation. Otherwise, no ultrastructural clefts were found. Extensive lamina densa reduplications could be traced as deep as 6 um below the B M Z . T E M examination showed manifold reduplications of the lamina densa (with attached anchoring fibrils) as well as a keratin body surrounded by a fibroblast in the upper dermis. Because of the aforementioned heterogeneity of morphological findings in Kindler's syndrome, some have suggested that the disorder might be a heterogeneous condition caused by mutations in different genes, which encode either for a structural protein or one or more functional proteins, possibly including D N A repair enzymes, or for anti-inflammatory cytokines (Shimizu et al., 1997). In this regard, the exaggerated apoptosis observed in the skin of Lanschuetzer et al.'s (2003) Kindler patient may be of interest since the marked, focal apoptosis of basal keratinocytes demonstrated in this report has not been described in any form of hereditary E B , which are known to be due to genetically caused changes in structural B M Z proteins. The authors concluded that Kindler's syndrome might primarily be an apoptotic disorder of basal keratinocytes and that the hypothesis that Kindler's syndrome  10 is based on a functional rather than a structural defect becomes conceivable when considering the fact that areas of severely disturbed B M Z architecture alternate with stretches of morphologically normal B M Z . These findings are in agreement with the findings in oral epithelium of a Kindler patient with periodontal disease (Wiebe and Larjava, 1999). Discontinuities of the B M Z were seen in immunostaining of the adhesion molecules and electron microscopic studies for a604 integrin and laminin-5 molecules. The discontinuities, however, did not appear to result from altered organization at the level of the hemidesmosome  (Wiebe and Larjava, 1999). Instead, disorganization  occurred at the B M Z with blistering appearing evident at the level of the lamina lucida. The former investigative group (Lanschuetzer et al., 2003) concluded that, excluding mosaicism, these findings support the idea that there is capacity to accurately generate a normal B M Z in Kindler's syndrome; a genetic structural defect seems to be unlikely. Alternatively, it is possible that Kindler syndrome is due to a genetic structural defect that only manifests clinically in the presence of trauma or stress. Such stresses may include activities, which are normally sustainable  such as friction from tooth brushing,  mastication, or gingivitis in the oral epithelium and friction or sun exposure from daily contact in the skin. A n example can be elucidated from the finding of a group of more than 20 subjects with Kindler syndrome identified in the rural area of Bocas del Toro province of Panama (Wiebe et al., 2003). In this Kindler group, periodontitis had an early onset (teenage years) and its rate of progression was rapid, resembling aggressive periodontitis, compared to non-Kindler individuals of the same geographic and ethnic origin. With identification of kindlin-1 as the target protein in Kindler syndrome, more  11 specific and discriminatory patterns of immunostaining are now possible using an antikindlin-1 antibody. A summary of the histological features of K S is provided in Table 2.  Table 2 :  Summary of Reported Histological Findings for Kindler Syndrome  C o m m o n l y Reported Histological Features Epidermal atrophy, focal vacuolization of incontinence in upper dermis; mild lymphocytic Extensive reduplication and disruption of dermoepidermal junction beneath basal cells; lucida  basal layer; pigmentary infiltrate lamina densa along the cleft formation in lamina  Desmosomes, hemidesmosomes, tonofilaments, anchoring filaments and fibrils appear normal. Immunofluorescence reveals normal B M components including ct3Bl and oc6p4 integrins, except type VII collagen found abnormally, deep in the connective tissue stroma and discontinuities in laminin-1 and-5 Laminin 5 and collagen types IV and VII present in broad reticular pattern, corresponding to reduplication of lamina densa  Reference Formanetal., 1989; Yasukawa et al., 2002; Wiebe etal., 2003; Jobard et al., 2003; (Skin)  Wiebe etal., 1999 (Oral Epithelium)  Shimizu et al., 1997 (Skin)  Areas of cleavage at or close to dermal-epidermal junction (within basal keratinocyte layer or beneath it); abundant keratin bodies present focally in the dermis incidating apoptosis; extensive reduplication of normal appearing lamina densa with branching, folding, and formation of circles and loops; lamina lucida focally widened suggesting subepidermal split formation  Ashton, 2004a & b (Skin)  Exaggerated apoptosis in skin, reduplication of normal appearing lamina densa; focally widened of lamina lucida suggesting subepidermal split formation  Lanschuetzer et al., 2003 (Skin)  Numerous melanophages (depending on skin type); disruption of collagen and elastic fibres in papillary dermis  1.3  Patrizi etal., 1996; Hovnanian et al., 1989 (Skin)  K i n d l e r Syndrome - T h e Genetic Mutation  Recently, the gene responsible for K S was localized on the short arm of chromosome 20, and mutations in a new gene, designated as kindlerin, were identified in four families (Jobard et al., 2003). In the same year, another research group demonstrated homozygous nonsense and frameshift mutations in the same gene (which they called  12 KIND1) in 17 families with K S (Siegel et al., 2003). The former group (Jobard et al., 2003) performed a genome-wide scan of a consanguineous Algerian family with three affected children. This resulted in localization of a gene on chromosome 20pl2.3 in a 7cM interval, which later was refined to around 800 kb by analyzing two supplementary consanguineous families. Mutation analysis in five consanguineous families from North Africa and Senegal revealed four homozygous mutations in a new gene, which belongs to the  FERM  family  (band  four-point-one/ezrin/radixin/moesin;  aka  filipodin/ezrin/radixin/moesin). In one of the Algerian families no mutation was found, which the authors thought could suggest genetic heterogeneity in K S . The authors later identified mutations in the C20orf42 gene, which they proposed to name kindlerin. To elucidate the gene structure, Jobard and associates performed overlapping RT-PCRs covering the entire sequence of kindlerin and determined a gene structure with 15 exons as predicted by the Sanger Institute and confirmed by the submission of a complete cDNA (URP1). The total length of the cDNA is 4720bp. As of 2003, two transcripts for kindlerin have been described in public databases; one is predicted to code for a protein of 230 amino acids (a.a, BAA90957) and the other for a protein of 677 a.a (AAN75822). Similarly, the latter investigative group (Siegel et al., 2003) performed a genome wide scan on D N A from 24 individuals from a Panamanian cohort, 16 with K S and 8 unaffected family members. A cluster of markers (D20S846, D20S115, and D20S851) on 20pl2.3 cosegregated with the disease. The authors also found that patients with KS from consanguineous families from Italy, Oman, Jordan, Turkey, Afghanistan, and Saudi Arabia showed homozygosity across the same genomic region (0.6-cM area of homozygosity between D20S95 and D20S192) as the Panamanians with KS. Six genes  13 were identified within this interval followed by determination of their exons. Mutations in only one of the six were identified: FLJ20116, subsequently renamed KIND1 encoding the protein kindlin-1. This gene spans 48.5 kb of genomic D N A and the predicted ORF is 2,034 bp, encoding a protein of 677 a.a with a molecular weight of 77.3 kDa. Most recently, there has been a report of an 11-year-old boy with Kindler syndrome from a consanguineous Indian family, who has a new homozygous nonsense mutation in KIND1 (Sethuraman et al., 2005). The nonsense mutation was designated C468X and results in loss of a cut site for the restriction endonuclease, Nspl. The authors (Sethuraman et al., 2005) claim that the mutation (C468X) is the first pathogenic sequence variant to be described in exon 12 of the KIND1 gene. A n up to date summary of all the reported mutations for the KIND1 gene is depicted in Figure 1 below (modified from Ashton, 2004a and Siegel et al., 2003); they include, nine nonsense, seven frame-shift, and two splice-site mutations. Siegel and associates' (2003) analysis of the primary protein structure of the kindlin-1 polypeptide revealed a number of features predicting that the function of this molecule relates to anchorage of the actin cytoskeleton to the plasma membrane. B L A S T analysis shows that kindlin-1 has C-terminal homology to talin and N-terminal homology to filopodin. B L A S T (Basic Local Alignment Search Tool) is a software tool to view sequences aligned with each other or to find homology by locating regions of sequence similarity with a view to comparing structure and function (Altschul et al., 1990). Both talin and filopodin are involved in anchorage of the actin cytoskeleton in focal contacts (Critchley, 2000), which are membrane-extracellular  matrix (-ECM) attachment  structures observed in cultured cells (Zamir and Geiger, 2001a,b). As well, B L A S T  14 analysis revealed kindlin-1 possesses a centrally located pleckstrin homology (PH) domain.  PH  domains  mediate  associations  with  specific  phosphatidylinositol  phospholipids of the plasma membrane and are a common feature of cell-signaling molecules such as integrins (Maffuci and Falasca, 2001; Itoh and Takenawa, 2002). Thirdly, Siegel D H and associates (2003) found that kindlin-1 also possesses two regions of homology with the F E R M (filopodin and ezrin/radixin/moesin) domain, which is shared by erythrocyte protein 4.1, E R M (ezrin, radixin, and moesin) proteins, and a number of proteins that mediate anchorage of the cytoskeleton to the plasma membrane as well as several tyrosine kinases and phosphatases and the tumor suppressor protein merlin (Chisti et al., 1998; Takeuchi et a l , 1994). The F E R M domain has also been called the amino-terminal domain, the 30-kDa domain, 4.1N30, the  membrane-  cytoskeletal-linking domain, the ERM-like domain, the ezrin-like domain of the band 4.1 superfamily, the conserved N-terminal region, and the membrane attachment domain (Chisti et al., 1998). Ezrin, moesin, and radixin are highly related proteins (ERM protein family), but the other proteins in which this domain is found do not share any region of similarity outside of the domain. E R M proteins are made of three domains, the F E R M domain, a central helical domain and a C-terminal tail domain, which binds actin. The amino-acid sequence of the F E R M domain is highly conserved among E R M proteins and is responsible for membrane association by direct binding to the cytoplasmic domain or tail of integral membrane proteins. E R M proteins are regulated by an intramolecular association of the F E R M and C-terminal tail domains that masks their binding sites for other molecules. For cytoskeleton-membrane  crosslinking, the dormant  molecules  become activated and the F E R M domain attaches to the membrane by binding specific  15  membrane proteins, while the last 34 residues of the tail bind actin filaments. Aside from binding to membranes, the activated F E R M domain of E R M proteins can also bind the guanine nucleotide dissociation inhibitor of Rho GTPase (RhoDGI), which suggest that in addition to functioning as a crosslinker, E R M proteins may influence Rho signalling pathways (Pearson et al., 2000). Binding to integrin p subunits via a subdomain named F3 within the F E R M domain of talin, an actin-binding protein has recently been described and results in the activation of integrins (Calderwood et al., 1999 and 2002). Based on amino acid residue homology, the authors (Siegel et al., 2003) determined that kindlin-1 possesses an unusual, bipartite F E R M domain that is interrupted by a P H domain. Both of these structural features imply that kindlin-1 is involved in membrane and/or cytoskeleton association.  676insC 464deIA  L302X  R271X  1089deIG  Q65X  H,N Filopodin linf  ( FERMQ  __jjyy^ H Q263X  W205X 373delT  E304X  1909delA  1161delA  1714delA  PH  Q F E R M 3  _  Talin 1 COOH  T T _ ^ T ^ ^ ^ ^ C468X  W616X  R288X  Figure 1: Schematic diagram of the kindlin-1 protein with F E R M and P H domains as well as regions of homology with filopodin and talin. Locations of reported positions of all loss-of-function mutations within the KJND1 gene, including nonsense, frameshift or splice mutations (indicated by black arrows; Jobard et al., 2003; Siegel et al., 2003; Ashton, 2004).  16 1.4  Cell-Cell and Cell-Matrix interactions Cell-cell and cell-ECM adhesions are constituently and functionally distinct  subcellular structures that are essential for the assembly of cells into tissues, and for communication between neighboring cells and between cells and the E C M . Cell-cell and cell-ECM adhesions are mediated primarily by transmembrane adhesion receptors of different families (i.e. cadherins and integrins) and multiple cytoplasmic protein-protein interactions, which in-turn connect the plasma membrane to the actin cytoskeleton (Burridge et al., 1996; Geiger et al., 2001; Hynes, 2002). Cells reside in a protein network, the E C M , which they secrete and mold into the intercellular space. The E C M exerts profound control over cells. The effects of the matrix are primarily mediated by integrins, a family of cell surface receptors that attach cells to the matrix and mediate mechanical and chemical signals from it. These signals regulate the activities of cytoplasmic kinases, growth factor receptors, and ion channels and control the organization of the intracellular actin cytoskeleton. Many integrin signals converge on cell cycle regulation, directing cells to live or die, to proliferate, or to exit the cell cycle and differentiate (Giancotti and Ruoslahti, 1999). Adhesions with the E C M are formed by all types of adherent cells, but their morphology, size and subcellular distribution can be quite heterogeneous. The extracellular ligands that anchor these adhesions include fibronectin, vitronectin and various collagens. The best-characterized adhesions are the 'classical' focal adhesions, and variants include fibrillar adhesions, focal complexes and podosomes (Geiger et al., 2001). Cells probe, respond to, and remodel the E C M using integrin-actin  cytoskeleton  adhesion  complexes.  Identification  of the  molecular  components of cell-cell and cell-ECM adhesions and the structural determinants that  17 control their localizations to these two distinct structures are therefore of critical importance to our understanding of tissue morphogenesis, cell growth, and differentiation in various biological processes including cancer, wound healing, and development. Integrins and cell-ECM adhesions provide dynamic, bi-directional links between the E C M and the cytoskeleton. As the E C M provides the physical microenvironment in which cells live, it provides a substrate for cell anchorage and serves as a tissue scaffold, guides cell migration during embryonic development and wound repair as well as morphogenesis. The E C M is also responsible for transmitting environmental signals to cells, which in turn affect all aspects of a cell's function including proliferation, differentiation, and death (Giancotti and Ruoslahti, 1999). Adhesive interactions occur via a variety of molecular mechanisms. These include different integrin receptors that bind to E C M molecules via their extracellular domains and interact via their cytoplasmic moieties with the actin cytoskeleton (Hynes, 2002). Integrin a and (3 subunits are type I transmembrane proteins expressed in surface membranes as heterodimers. Each consists of a large extracellular domain, a single transmembrane domain segment, and a relatively short cytoplasmic tail (Calderwood et al., 2000). p-cytoplasmic tails are necessary and sufficient to link integrins to the actin cytoskeleton. Adhesive interactions are mediated by a network of 'anchor proteins', some of which directly mediate linkage between actin and the cell membrane, whereas others play a more regulatory role (Geiger et al., 2001). Adhesion is also mediated by a variety of membrane- or matrix-bound proteoglycan and glycosaminoglycan molecules (Woods et al., 1998). Integrin-mediated adhesions are molecularly heterogeneous, appearing in different forms such focal adhesions (FAs), fibrillar adhesions (FBs), focal complexes (FXs) and podosomes. Cell motility is  18 mediated in part by the spatial and temporal regulation of the formation and dissociation of matrix adhesions (Zaidel-Bar et al., 2004). FAs serve at least two significant cellular functions: 1) to transmit force or tension at adhesion sites to maintain strong attachments to the underlying E C M and 2) to act as signaling centers from which numerous intracellular pathways emanate to regulate cell growth, survival, and gene expression (Sastry and Burridge, 2000). Focal adhesions are also dynamic structures that assemble, disperse, and recycle (turnover) as cells migrate or enter into mitosis. Focal adhesions are flat, elongated structures and are often located near the periphery of cells (Zamir and Geiger, 2001a,b; Sastry and Burridge, 2000). They can anchor bundles of actin microfilaments through a plaque consisting of several different proteins. Formation of focal adhesions is stimulated by GTPase Rho-A and is driven by actomyosin contractility. Commonly found plaque proteins include paxillin, talin, and vinculin as well as tyrosine-phosphorylated proteins (Geiger et al., 2001). By Definition, focal adhesions are formed by cells in culture spreading on solid surfaces; however, such structures have also been observed in vivo (Turner et al., 1991). Fibrillar adhesions on the other hand are found more centrally in the cells; they are either elongated or dot-like structures associated with E C M fibrils (Zamir and Geiger, 2001a,b; Geiger et al., 2001). Components of fibrillar adhesions are typically, extracellular fibronectin fibrils, the fibronectin receptor a 5 p i integrin, and cytoplasmic tensin (Zamir and Geiger, 2001a,b; Geiger et al., 2001). Yet another type of matrix adhesion is the focal complex. Focal complexes are small, dot-like adhesions present primarily at the edges of lamellopodia (Geiger et al., 2001). These sites can play a role in cell migration or serve as  19 precursors of focal adhesions; their formation is stimulated by the Rho-family GTPase, Rac (Rottner et al., 1999). Lastly, podosomes represent another type of E C M adhesion. They are small (~ 0.5 um diameter) cylindrical structures containing focal contact proteins such as paxillin and vinculin. Podosomes are commonly found in various malignant cell-types and normal cells such as osteoclasts and macrophages alike (Gaidano et al., 1990). Proteins invariably found in podosomes are gelsolin and membrane invaginationassociated dynamin (Geiger et al., 2001). The composition of the matrix controls the recruitment of specific integrins into focal adhesions, suggesting that the conformational changes associated with ligand binding help to regulate integrin cytoplasmic linkages. Both ligand binding and clustering are necessary for full integrin function and the recruitment of several focal adhesionassociated proteins (Miyamoto et al., 1995). Integrin binding to the E C M alone is not sufficient to induce integrin clustering and focal complex formation. Rather, signaling events driven from the inside of the cell are required (Hotchin and Hall, 1995). The early finding that Rho-mediated contractility drives focal adhesion formation from the 'insideout' led to the theory that focal adhesion formation is a bi-directional event that requires intracellular signaling events in addition to extracellular integrin-ligand (ECM) binding (Chzanowska-Wodnicka and Burridge, 1996). For migration of a cell to occur, a cell must extend a protrusion to make an initial contact with the E C M . Some of these Racdependant contacts develop into Rho-dependant focal adhesions that stabilize the cell during migration (Wozniak et al., 2004). At the leading edge of a migrating cell, membrane protrusion is stabilized by small adhesive foci that initially contain paxillin  20 followed by a-actinin (Laukaitis et al., 2000). Initial foci of adhesion form just posterior to the actin network, and grow into focal complexes abundant in tyrosine phosphorylation containing integrin, paxillin, vinculin, and talin and focal adhesion kinase (FAK) (Laukaitis et al., 2000). Subsequently, zyxin and tensin are recruited to these complexes as they remodel into focal adhesions, which stabilize the protrusion (Zaidel-Bar et al., 2003). Such cell-ECM adhesion dynamics are likely controlled by the small GTPases, Rac (early events of membrane protrusion and focal complex formation) and Rho (later events necessary for maturation of focal complexes into larger focal adhesions to stabilize the cell during cell migration) (Rottner et al., 1999; Wozniak et al., 2004). Notably, a cell must be able to continuously remodel focal complexes into focal adhesions, and visa versa, to migrate. Others have reviewed the mechanisms responsible for such focal adhesion turnover/disassembly in more detail (Sastry and Burridge, 2000; Wozniak M A et al., 2004). Briefly, FAs form during spreading or migration on flat, rigid substrates to which E C M components become adsorbed (as in cell culture). The assembly of FAs in response to adhesion to the E C M is gradual, usually occurring within 1 to 2 hours after cell attachment (Sastry and Burridge, 2000). Initially, nascent cell-matrix adhesions, or focal complexes, form at the cell periphery as a cell spreads or at the leading edge as a cell migrates. Focal complexes mature into FAs as cells become stably attached to their substrates and tension is exerted on these adhesions. Actin filaments are indirectly linked to integrins at FAs (Critchley, 2000). In migrating cells, FAs can provide traction on the substrate over which cells crawl, although some cells can migrate without FAs and large FAs retard motility due to excessive adhesion (Huttenlocher et al., 1995). FAs disassemble or disperse under a number of physiological circumstances. For example,  21 adhesions to E C M are released at the rear of a migrating cell. As well, during mitosis, F A disassembly occurs to allow the cell to loose their attachment to the E C M seen as the cell adopts a round morphology. Finally, during oncogenic transformation, F A integrity is often compromised (Hynes, 2002).  1.5  Kindlin-1, a Novel Focal Adhesion Protein, its Homologue, Mig-2, and Migfilin Kindlin-1 demonstrates closest homology (30% a.a identity) to the Caenorhabditis  elegans (C. elegans) protein UNC-112 (Siegel et al., 2003). UNC-112 contains three band 4.1 domains as well as a P H domain; it has been shown to colocalize with (3-integrin in muscle cell membrane in C. elegans (Rogalski et al., 2000). When searching the human genome sequence, two further proteins with significant overall homology to UNC-112, Mig-2 (42% a.a identity) and Siegel et al's predicted protein MGC10966 (with 38% a.a identity), are found. Thus, the three human proteins, kindlin-1, Mig-2, and M G C 10966, and UNC-112 are very similar in size and share the same domain organization (Fig. 1) and talin/filopodin homology regions. Siegel and company found that evolutionary analysis using unweighted pair grouping (Nei, 1987) reveals that kindlin-1 and Mig-2 are more closely related and more recently diverged from M G C 10966. Overall, Mig-2 shows the closest a.a similarity to UNC-112. Mig-2 has 62% a.a identity with kindlin-1 and is located on 14q22.1. However, Mig-2 is human and not to be confused with mig-2 in C elegans (Zipkin et al., 1997). Mig-2 (mig-2) is neither structurally nor functionally related to human Mig-2 (Wick et al., 1994). Given their similarities, Siegel et al. (2003)  22 suggested that human Mig-2 and MGC10966 be renamed KIND2 and KIND3, respectively. Siegel and associates (2003) looked at tissue expression of KJND1 via hybridization of a KIND1 c D N A probe to a multiple-tissue northern blot. The KIND1 transcript was 4.9 kb with a transcript of this size also found in cultured epidermal keratinocytes, colon, kidney, and placenta with lower levels expressed in heart, skeletal muscle, liver, and small intestine.  In particular, when multiple-tissue cDNA panels  showed significant differences in expression of the three kindlin genes (KIND1, KIND2, KIND3), KIND1 was found to be expressed highly in keratinocytes, with lower expression in prostate, ovary, colon, kidney, and pancreas and weak expression in spleen, thymus, testis, heart, brain, placenta, lung, liver, and fibroblasts and no expression found in small intestine, peripheral blood leukocytes, or skeletal muscle. The contrasting level of expression of KIND 1 in keratinocytes versus fibroblasts was suggested by the authors using quantitative P C R (Siegel et al., 2003). Small interfering R N A (siRNA) is a novel technique used to silence specific genes of interest as a means of investigating their function (Elbashir et al., 2002; Harmon, 2002; Hudson et al., 2002). Using this siRNA one group showed that Mig-2 localizes to cell-ECM adhesions and is essential for cell shape modulation (Tu et al., 2003). Mig-2 interfering R N A and a 21-nucleotide irrelevant R N A as a control, respectively, were introduced into cells. At 30 minutes post-seeding transfected cells remained unspread whilst control transfectants and wild-type cells exhibited spread morphology. The authors further elucidated mechanisms by which Mig-2 regulates cell shape by yeast two-hybrid screens aimed at identifying proteins that interact with Mig-2. Mig-2 was found to  23 interact with migfilin, a novel protein that consists of three L I M domains at the Cterminus, an N-terminal region, and a proline-rich region located between these termini (Tu et al., 2003). The L I M domain is a zinc binding, cysteine rich motif consisting of two tandemly repeated zinc fingers. L I M domains do not seem to bind D N A but instead appear to mediate protein-protein interactions. Functionally, L I M domain containing proteins have been implicated in a variety of biological processes including cell lineage specification, cytoskeletal organization, and organ development (Bach, 2000). PCR revealed that there are two different lengths of migflin cDNAs and that cDNAs encoding proteins that are homologs to human migfilin are present in other species (mouse; GenBank AAHo477; 78% aa homology). Both forms interact with Mig-2. Monoclonal antibodies (mAbs) against Mig-2 were generated and immunofluoresence stained human cells used to determine the subcellular localization of Mig-2. Mig-2 co-aligned with actin stress fibres and clustered at cell-ECM adhesion sites where actin stress fibers were anchored. Double staining of cells with rhodamine-conjugated anti-migfilin and FITCconjugated anti-Mig-2 mAbs confirmed that migfilin was colocalized with Mig-2 in cellE C M adhesions. Phalloidin staining confirmed that migfilin aggregates also contained filamentous actin (F actin; Tu et al., 2003). The conformation of actin is different, depending on whether there is A T P or A D P in the nucleotide-binding site. G-actin (globular actin), with bound ATP, can polymerize to form F-actin (filamentous actin). Actin filaments may associate into bundles or networks, via cross-linking proteins. Filamin is a V-shaped cross-linking protein that causes actin filaments to associate in loose networks that give some areas of the cytoskeleton a gel-like consistency.  24 Tu and company (2003) demonstrated that while Mig-2 is not absolutely required for the formation of focal adhesions, it is required for the recruitment of migfilin into focal adhesions. Using siRNA, depletion of migfilin, like that of Mig-2, significantly impairs cell spreading; however, migfilin- or Mig-2-deficient cells are able to spread after prolonged incubation. The authors concluded that migfilin likely works in concert with Mig-2 in cell shape modulation. To determine how migfilin functions in cell shape modulation, the authors (Tu et al., 2003) looked at proteins, which interact with the N-terminus of migfilin, which had showed to mediate association with actin filaments. The N-terminal domain of migfilin interacts with the C-terminal region of filamin A / C . Mammals have three filamin genes: filamins A and B are expressed ubiquitously, whereas filamin C is muscle specific (Stossel et al., 2001). The finding that the filamin binding N-terminus of migfilin mediates the association with actin filaments and since filamin is a component of actin filaments suggest that the interaction with filamin likely mediates the association of migfilin with actin filaments (Tu et al., 2003). In light of filamin's central role in the coordinated assembly of the actin cytoskeleton (Stossel et al., 2001), depletion of migfilin significantly reduced the amount of F actin in cells, suggesting a significant role of migfilin in the regulation of actin assembly (Tu et al., 2003). In a recent review (Wu et al., 2004), the author states that migfilin [also known as FBLP-1A (Takafuta et al., 2003) or Cal (Akazawa et al., 2004)] is emerging as a key regulator of a variety of fundamental cellular processes, including shape modulation, motility and differentiation. To date, seven migfilin-binding proteins have been identified. Staining with anti-migfilin mAbs reveals that, in fibroblasts, migfilin is highly concentrated at cell-ECM adhesions,  25 although a fraction is also detected along the actin stress fibers linking the cell-ECM adhesions (Tu et al., 2003). In epithelial and endothelial cells, migfilin localizes not only to cell-ECM adhesions but also to cell-cell adhesions (Gkretsi et al., 2004). Recent studies have demonstrated that mutations in the genes encoding migfilinbinding partners cause several human disorders. In Kindler syndrome for example, Wu et al. (2004) proposes that the skin blistering is probably caused by, at least in part, the weakening of the actin-cytoskeleton-membrane link in the basal keratinocytes. Although we do not yet know how kindlin-1 is linked to actin filaments, given the significant homology between kindlin-1 and Mig-2, the author (Wu et al., 2004) proposes that migfilin, which is also clustered as part of focal adhesions in keratinocytes, might link kindlin-1 to filamin-containing actin filaments. Tu and Wu (personal communication) claim that  kindlin-1, like Mig-2, possesses migfilin-binding activity, which would  support their hypothesis. Kloeker et al. (2004) demonstrated in the immortalized epidermal keratinocyte cell line, HaCaT, that endogenous kindlin-1 (also known as kindlerin) localizes to focal adhesions and that after transforming growth factor-p (TGF-P) stimulation, kindlin-1 is also detected at membrane ruffles particularly at sites of cell-cell contact. This latter finding supports Wu's aforementioned hypothesis (Wu et al., 2004). It is possible that mutations in other genes encoding focal adhesion proteins that are physically or functionally associated with kindlin-1 could also lead to disease resembling Kindler syndrome. As suggested by Wu (2004), perhaps loss of kindlin-1 alters the distribution of migfilin between the basal cell-ECM adhesions and the cell-cell adherens junctions, which manifests as Kindler syndrome clinically. In contrast, upregulation of kindlin-1 expression in epithelia of the lung and colon is associated with  26 carcinoma in these organs (Weinstein et al., 2003). Perhaps this phenomenon is explained, at least in part, by the significant increase in kindlin-1 and a concomitant reduction in E-cadherin levels in human mammary epithelial cells treated with TGF-P demonstrated by Kloeker in 2004. Gkretsi and associates (2004) identified the determinants that control migfilin localization to cell-cell and cell-ECM adhesions, two functionally coordinated but structurally distinct adhesion structures. Both LIM2 and LIM3 are crucially involved in the localization of migfilin to cell-cell and cell-ECM adhesions, whereas LIM1 is not required for migfilin localization to either adhesion site. Mig-2, a docking adhesion protein for migfilin, is present exclusively in cell-ECM adhesions. It is possible that Mig-2 competes for migfilin at adherens junctions. The same authors (Gkretsi et a l , 2004) proposed that there are two 'pools' of migfilin, one that binds Mig-2 at cell-ECM adhesions, while the other associates with cell-cell adherens junctions. These two pools might serve to facilitate different cell functions during biological and pathological processes such as epithelial-mesenchymal cross-talk, wound healing, and carcinoma development and spreading. For example, upregulation of migfilin at cell-ECM adhesions with concomitant down-regulation at cell-cell junctions could allow cells to proliferate and migrate across an epithelial wound enabling wound healing.  27 1.6  Focal Adhesion Proteins, ILK-1 and Paxillin - Potential Signaling Partners of Kindlin-1 Paxillin was initially characterized as a 68-kDa focal adhesion protein (Glenney  and Zokas, 1989). Later, it was purified to homogeneity from chicken gizzard smooth muscle tissue, an abundant source of cytoskeletal proteins, and identified as a novel binding partner for the focal adhesion and actin binding protein vinculin (Turner et al., 1990). It was named paxillin, derived from the Latin word, paxillus, which means a stake or peg, consistent with its proposed function of linking actin filaments to integrin-rich cell adhesion sites (Brown and Turner, 2004). Several authors have reviewed paxillin's established identity as a focal adhesion protein (Zamir et al., 1999; Critchley, 2000; Sastry and Burridge, 2000; Brown and Turner, 2004). Analysis of m R N A expression profiles of adult human tissues shows paxillin is expressed abundantly in most tissues with the exception of the brain (Brown and Turner, 2004). The critical importance of paxillin was demonstrated via the generation of a mouse knockout of the paxillin gene (Hagel et a l , 2002). The result was embryonically lethal despite the expression of Hic-5 and leupaxin, which are paralogs of paxillin (Brown and Turner, 2004). Expression of paxillin family members in hematopoetic cells shows a tremendous amount of regulation. For example, leupaxin is upregulated in differentiating lymphocytes (Lipsky et al., 1998), while Hic-5 is upregulated during the transition of megakaryoctyes to platelets with a concomitant downregulation of paxillin (Hagmann et al., 1998). Regulation of paxillin is also associated with disease. Hic-5, for example might have tumor-suppressor functions; it is absent in prostate cancer lines (Fujimoto et al., 1999), epithelial carcinomas (Zhang et al., 2000), and many breast  28 cancer cell lines (Magklara et al., 2002). Additional roles for paxillin in cancer have been proposed based on the affinity of paxillin for phosphorylation by integrin and growth factor receptor ligation. These transmembrane receptors have established roles in tumorigenesis and metastasis (Brown and Turner, 2004). Integrins, which lack intrinsic enzymatic activity, transmit intracellular signals by interacting with various effector proteins (Liu et al., 2000). Integrin-linked kinase (ILK) is one such cytoplasmic effector that has emerged as an important component of integrin signaling complexes (Dedhar, 2000; Wu and Dedhar, 2001). ILK was described in 1995 as a serine/threonine (Ser/Thr) kinase that binds to the cytoplasmic tails of p i , P2, P3integrin subunits (Hannigan et al., 1996). The protein is composed of distinct domains, including four N-terminal ankyrin repeats followed by a central PH-like sequence and a C-terminal region that is homologous to the catalytic domain of protein kinases. Kinases activated by various cell signals regulate formation or disassembly of focal adhesions. ILK binds to plasma membrane integrins and to various actin-binding proteins, thereby mediating attachment of actin to the plasma membrane in focal adhesions. I L K is reportedly involved in several physiological and pathological processes including cell adhesion, E-cadherin expression, fibronectin matrix assembly, transformation and myogenic differentiation (Hannigan et al., 1996; Wu et al., 1998). Downstream targets of ILK signaling that are of importance include protein kinase B (PKB), glycogen synthase kinase 3 (GSK-3), p-catenin, mitogen-activated protein kinases (MAPKs), and the myosin light chain (Wu, 2001). Upstream events that regulate ILK activity include cellE C M interactions and insulin stimulation (Wu, 2001). In cultured mammalian cells, ILK has been localized to cell-matrix adhesion sites by immunofluorescence staining  29 (Mackinnon et al., 2002).  Several studies have identified I L K interaction with focal  adhesion components such as the adaptor protein paxillin, the L I M domain proteins PINCH 1 and 2, the actin-binding proteins CH-ILKBP/actopaxin/a-parvin and (3parvin/affixin (for review, see Wu and Dedhar, 2001), and more recently, the F E R M domain protein UNC-112/Mig-2 (Mackinnon et al., 2002). The authors showed that PAT4, a C. elegans homologue of ILK, interacts with UNC-112 (C. elegans homologue of Mig-2) in integrin-containing adhesion structures in C. elegans (Mackinnon et al., 2002). In turn, others have found that Mig-2 is able to interact with I L K (Tu Yizeng and Wu Chuanyue, unpublished data; Tu et al., 2003). Immunofluorescence microscopy of fibroblasts demonstrates that endogenous I L K as well as transfected green fluorescent protein-ILK co-localizes with paxillin in focal adhesions (Nikolopoulos and Turner, 2001). The authors analyzed the deduced amino acid sequence of ILK and identified a paxillin-binding domain in the carboxyl terminus of ILK. In contrast to wild-type ILK, paxillin-binding subdomain mutants of ILK were unable to bind to the paxillin LD1 motif in vitro and failed to localize to focal adhesions. The authors concluded that paxillin binding is necessary for ILK-mediated focal adhesion formation (Nikolopoulos and Turner, 2001). In addition, I L K associates with a Ser/Thr protein phosphatase of the PP2C family, referred to as ILK-associated phosphatase (ILKAP; Leung-Hagesteijin et al., 2001). Several additional regulatory proteins critical to integrin function, including the tyrosine kinases, focal adhesion kinase (FAK) and SRC family kinases, bind to regions within the amino terminus of paxillin adjacent to the ILK binding site (Turner, 2000). Accumulating evidence from a variety of cell types indicates that I L K can function as a kinase and as a scaffold protein mediating interactions between integrins  30 and the actin cytoskeleton through interaction with multiple binding partners. The regulation and function of ILK was recently investigated (Vespa et al., 2003). The epidermis is formed by keratinocytes at different stages of differentiation (Fuchs, 1990). Undifferentiated progenitor keratinocytes reside in the basal cell layer and are attached to the basement membrane separating them from the underlying dermis (Jones and Watt, 2003). After appropriate environmental (ECM) stimulation, basal keratinocytes initiate terminal differentiation, characterized by loss of proliferative capacity, decrease in integrin expression, detachment from the basement membrane, and migration upward to form post-mitotic suprabasal keratinocyte layers. In culture, keratinocytes can be induced to differentiate by elevating the extracellular C a  +2  concentration. Intercellular adhesion in differentiated keratinocytes occurs through tight junctions, cadherin-mediated formation of adherens junctions, and desmosomes (Jamora and Fuchs, 2002). In turn, epithelial integrity and barrier function rely on these cell-cell junctions. Vespa and colleagues (2003) looked at the function of I L K in mouse keratinocytes and showed that I L K protein expression is independent of integrin expression and signaling, since it's level remains constant during Ca -induced differentiation. Instead, I L K activity was markedly decreased during keratinocyte differentiation accompanied by a change in its distribution. I L K distributes in close apposition to actin fibers along intercellular junctions in differentiated, but not undifferentiated keratinocytes. Such localization to cell-cell borders occurs independently of integrin signaling and requires both Ca  and an intact actin cytoskeleton. As well,  contrary to other epithelia, I L K overexpression in differentiated keratinocytes does not  31  promote E-cadherin down-regulation and epithelial-mesenchymal transition involved in tumorigenesis (Vespa et al., 2003). In mouse, loss of ILK expression leads to peri-implantation lethality. Around implantation the primitive endoderm develops on the surface of the inner cell mass (ICM) of the blastocyst and lays down a basement membrane (BM) that is required for adjacent ICM cells to polarize and establish the epiblast (primitive ectoderm), and for remaining ICM cells to undergo apoptosis resulting in formation of the proamniotic cavity (Li et a l , 2002). The underlying cause of the polarization defect of ILK-deficient epiblast is abnormal F-actin reorganization (Sakai et al., 2003).  32 Table 3a:  Salient Characteristics of Focal Adhesion Proteins Investigated*  *Note: this list does not include all focal adhesion proteins involved in Cell-Cell or ECM-Cell Signaling  Focal Adhesion Protein  ILK-1  Paxillin  Mig-2  Salient Features  Reference  Ser/Thr kinase; binds to cytoplasmic tails pi, (32, p3-integrin; 4 distinct domains (4 N-terminal ankyrin repeats followed by a central PH-like region and a C-terminal region homologous to catalytic domain of protein kinases).  Hannigan et al., 1996 Wuetal., 1998  Localized to cell-ECM adhesion sites; interacts with FERM domain protein UNC-112/Mig-2.  Mackinnon et al., 2002  Interacts with focal adhesion components including paxillin, PINCH 1 & 2, actin-binding proteins CH-ILKBP/actopaxin/aparvin & P-parvin/affixin.  Wu and Dedhar, 2001  Colocalizes with Paxillin in focal adhesions.  Nikolopoulos et al., 2001  Associates with a Ser/Thr protein phosphatase referred to as ILK-associated phosphatase (ILKAP).  Leung-Hagesteijin et al., 2001  Can function as kinase and scaffold protein mediating interactions between integrins and actin cytoskeleton through interaction with multiple binding partners. Protein expression is independent of integrin expression and signaling; ILK distributes in close apposition to actin fibers along intercellular junctions in differentiated, but not undifferentiated keratinocytes.  Vespa er al, 2003  Necessary for epiblast polarization and remaining inner cell mass cells to undergo apoptosis resulting in formation of proamniotic cavity.  Li et al, 2002 Sakai et al, 2003  Expressed abundantly in most tissues except the brain; mouse knockout is embryonically lethal.  Brown and Turner, 2004  Hic-5, a paralog of paxillin, might have tumor-suppressor functions; Paxillin has affinity for phosphorylation by integrin and growth factor receptor ligation, which have roles in tumorigenesis and metastasis. Localizes to cell-ECM adhesions and is essential for cell shape modulation.; not absolutely necessary for formation of focal adhesions; but is required for recruitment of migfilin into focal adhesions (docking adhesion protein for migfilin). Interacts with migfilin and co-aligns with actin stress fibers and clusters at cell-ECM adhesion sites where actin stress fibers are anchored.  Hagel et al, 2002  Brown and Turner, 2004  Tu et al, 2003 Gkretsi et al, 2004  33 Table 3b:  Salient Characteristics of Focal Adhesion Proteins Investigated*  *Note: this list does not include all focal adhesion proteins involved in Cell-Cell or ECM-Cell Signaling  Focal Adhesion Protein  Salient Features  Reference  3 LIM domains at C-terminus; N-terminal region; proline-rich region located in between.  Tu et al., 2003  LIM domain: consists of two tandemly repeated zinc fingers; do not seem to bind DNA, but instead mediate protein-protein interactions including those involved in cell lineage specification, cytoskeletal organization, and organ development.  Migfilin  Colocalizes with Mig-2 in cell-ECM adhesions and likely works in concert with Mig-2 in cell shape modulation; migfilin aggregates contain F actin; N-terminus of migfilin mediates associates with actin via interaction with C-terminal region of filamin A/C. Depletion of migfilin reduces amount of F actin in cells. Migfilin localizes not only to cell-ECM adhesions, but also to cell-cell adhesions; suggested to link kindlin-1 to filamincontaining actin filaments. Closest homology, 30% a.a identity, with C. elegans UNC-112; similar in size and shares same domain organization and talin/filopodin homology regions as Mig-2, MGC10966, and UNC-112; has 62% a.a identity with Mig-2.  Bach, 2000  Tu et al., 2003  Gkretsi etal., 2001  Siegel et al, 2003  KIND1 (kindlin-1 gene) transcript found in cultured epidermal keratinocytes, colon, kidney, and placenta with lower levels in heart, skeletal muscle, liver, and small intestine.  Kindlin-1  Proposed to be linked tofilamin-containingactin filaments via migfilin. Endogenous kindlin-1 (kindlerin) localizes to focal adhesions; after TGF-P stimulation can also be detected at sites of cellcell contact. Upregulation in epithelia of lung and colon is associated with carcinoma in these organs. Concomitant increase in kindlin-1 and reduction in Ecadherin in mammary epithelial cells treated with TGF-P.  Gkretsi et al, 2001  Kloeker et al, 2004  Weinstein et al, 2003 Kloeker et al, 2004  34 Figure  2: Drawing Depicting Proposed Interaction between Focal Adhesion Proteins  and does not include all focal adhesion proteins or secondary messenger systems involved in Celk->ECM signaling  Drawn by: Giorgio M. Petricca  35  CHAPTER II Apriori Hypothesis and Aim of the Study Kindler syndrome represents the first demonstration of a defect in ECM-actin interactions compared to that between ECM-keratin causing a skin fragility disorder. More than one group has already demonstrated the termination of cytoskeletal actin filaments into focal adhesions of kindlin-1 (Siegel et al., 2003, Lanschuetzer et a l , 2003). The lack of lethality as a result of mutations in KIND1 suggests that other focal adhesion proteins known to link integrins to actin, ILK-1, paxillin, Mig-2, and migfilin, are also crucial to maintaining the integrity of the ECM-Cell interaction. However, the defect in kindlin-1 clearly has clinical relevance, dermatologically, namely, it leads to epithelial blistering in childhood, skin fragility and sun sensitivity (Ashton, 2004a,b). Of particular interest to our laboratory, a defect in kindlin-1 appears to render subjects more susceptible to periodontal disease, which appears earlier in life (teenage years) and progresses at an aggressive rate (Wiebe et al., 2003). It is possible that altered expression of kindlin-1 weakens the integrity of the ECM-Cell interaction (i.e. basement membranebasal keratinocyte adhesion in skin and oral epithelium) such that an individual with Kindler syndrome is rendered more susceptible to trauma from routine daily activities. One means of testing this hypothesis is to examine how keratinocytes, using the HaCaT cell as a model, behave when kindlin-1 expression is knocked down via siRNA-mediated gene silencing (Kloeker et. al., 2004) and to look at how kindlin-1 is distributed immunohistologically in normal human and wounded oral mucosa.  36  CHAPTER III Materials and Methods 3.1  Reagents Mouse monoclonal antibodies (mAbs; primary antibodies) recognizing Mig-2 and  migfilin were prepared and generously provided by the laboratory of Dr. Chuanyue Wu (Department of Pathology, University of Pittsburgh, P A , USA). Briefly, the antibodies were prepared using GST fusion proteins containing Mig-2 residues 218-486 and migfilin residues 1-189, respectively, as antigens based on previously described methods (Tu et al., 1999; 2001; 2003). Hybridoma supernatants, screened for anti-Mig-2 and antimigfilin activities by ELISA were further tested by Western blotting and then used for immunofluorescent and immunocytochemical staining. ILK-1 primary antibody was purchased from Upstate Cell Signaling Solutions, Lake Placid, N Y , U S A (mouse monoclonal IgG2b (clone 65.1.9; 200 ug/206 ul). Actin was stained in all cases using Phalloidin [tetramethylrhodamine isothiocyanate- (TRITC)-labeled; mixed isomers from amantia phalloides; Sigma Chemical Co., St. Louis, M O , USA]. Paxillin was purchased from C H E M I C O N International, Inc. (Temecula, C A , USA). Secondary antibodies used for immunofluorescent tissue and cytological staining included Alexa Fluor® 488 (goat antimouse IgG(H+L); 2 mg/ml; Molecular Probes, Invitrogen® Detection Technologies, Eugene, Oregon, U S A ) , Alexa Fluor® 546 (goat antimouse IgG(H+L); 2 mg/ml; Molecular Probes, Invitrogen® Detection Technologies, Eugene, Oregon, USA). Primary and secondary antibodies and Phalloidin were diluted appropriately for the following immunostainings as listed in Tables 4 and 5 below, respectively. Phalloidin and other  37 fluorophores (above) were always filtered immediately prior to use using a Nylon Acrodisc® syringe filters (0.45 um pore size x 4mm diameter; Gelman Sciences, MI, USA) fitted with a 1 ml Tuberculin syringe and 23 Gauge Precision Glide® needle, both purchased from Becton, Dickinson and Company, Franklin Lakes, NJ, USA. Bovine serum albumin (BSA, electrophoresis grade, initial fractionation by heat shock, fraction V ) , used in blocking solutions, was purchased from Sigma-Aldrich Co., St. Louis, M O , USA. Triton®-X-100 (electrophoresis grade), used in buffer solutions, was purchased from Fischer Scientific, NJ, USA. Glycine (electrophoresis purity), used in blocking solutions, was purchased Bio-Rad Laboratories, Hercules, C A , USA.  3.2  Experimental In Vivo Oral Mucosal Wounding The experimental protocol for human oral wound samples was approved by the  Clinical Research Ethics Board of the University of British Columbia and included twenty-three volunteer subjects. The subjects were healthy, non-smoking individuals ranging from 24 and 66 years of age. Standard full thickness wounds (10 mm long x 2mm wide) were made in the palatal mucosa, using parallel blades. Tissue harvested from the initial wound (control) was rinsed in saline, embedded in Tissue-Tek O.C.T. (Sakura Finetek, Inc, Torrance, CA), and immediately frozen in liquid nitrogen (Day 0 or control). After 3 and 7 days of healing, punch biopsies (4mm in diameter) were obtained and frozen in liquid nitrogen (as above). Frozen sections (8-10 um) cut from the punch biopsies (control-Day 0, Days 3, and 7) were used for routine histology and immunohistochemistry described below.  38 3.3  Immunofluorescent a n d Immunohistological Staining of F r o z e n Sections  The control and wound specimens were sectioned at 8-10 um thickness in a cryostat and stored at -86°C until used. Samples were allowed to thaw by air-drying at room temperature and then fixed with -20°C acetone for 5 minutes. Sections were encircled using a grease pen and then rehydrated in phosphate-buffered saline (PBS: 137 mM NaCl, 2.7 m M KC1, 4.3 m M N a H P 0 ' 7 H 0 , 1.4 m M K H P 0 , pH 7.3) containing 2  4  2  2  4  0.01% Triton-X-100 for 5 minutes. Non-specific protein-binding sites were blocked by a 30 minute incubation with 10 mg/ml bovine serum albumin (BSA; Sigma-Aldrich Co., St. Louis, M O , U S A ) and 0.01% Triton-X-100 in PBS. For immunohistochemical staining, sections were incubated in normal blocking serum (Vectastain® A B C Kit Mouse IgG; Vector Laboratories, Inc., Burlingame, C A , USA) for 20 minutes at room temperature. For both immunofluorescent and immunohistochemical stainings, sections were then incubated in a humidified chamber overnight with primary antibody at 4°C. For immunofluorescent staining, the primary antibodies (listed in Table 4) were diluted in cell culture medium. Cell culture medium consisted of Dulbecco's modified Eagle's medium ( D M E M ; Gibco B R L Life Technologies, Rockville, M D , USA) supplemented with 23 m M sodium bicarbonate, 20 m M HEPES, antibiotics (50 ug/ml streptomycin sulfate, 100 U/ml penicillin) and 10% heat-inactivated FCS (Gibco B R L , Rockville, M D , USA). For immunohistochemical stainings, primary antibodies (listed in Table 4) were diluted in 1 mg/ml B S A and 0.01% Triton-X-100 in PBS. A l l sections were then washed three times with 1 mg/ml B S A and 0.01% TritonX-100 in PBS. For immunofluorescent staining, sections were then incubated with secondary antibody for 1 hour at room temperature in a humidified chamber using the  39 appropriate Alexa Fluor® 546 antibody or Alexa Fluor® 488 secondary antibody conjugates as described above. Filamentous actin was stained using Phalloidin (1:100 dilution; TRITC-labeled; mixed isomers from amantia phalloides, Sigma Chemical Co., St. Louis, M O , USA). Secondary antibodies were diluted in cell culture media (as above). However, for immunohistochemical stainings, sections were incubated with biotinylated anti-mouse secondary antibody (Vectastain® A B C Kit Mouse IgG; Vector Laboratories, Inc., Burlingame, C A , USA) for 1 hour at room temperature in a humidified chamber. These sections were rinsed again with 1 mg/ml B S A and 0.01% Triton-X-100 in PBS and then reacted with A B C avidin/peroxidase reagent (Vectastain® A B C Kit Mouse IgG Vector®,  Vector  Laboratories,  Inc.,  Burlingame, C A ) for  30  minutes  for  immunohistochemical stainings. The sections were then reacted with Vector VIP substrate (Vector® VIP Peroxidase Substrate Kit, Vector Laboratories, Inc., Burlingame, CA). The colour development was followed by viewing the sections under a light microscope, and the reaction was stopped when the desired color was detected by immersing  the  sections  in  distilled  water  for  five  minutes.  Sections  for  immunohistochemical staining were allowed to air dry and then mounted using Vecta Mount™ Permanent Mounting Medium purchased from Vector Laboratories, Inc. (Burlingame, C A , USA). For immunofluorescent stainings, after secondary antibody incubations, sections were washed (as above) and mounted using the ProLong™ Antifade kit (Molecular Probes Inc., Eugene, OR, USA) and allowed to air-dry in the dark. Control immunostainings to indicate non-specific staining were performed without adding primary antibodies.  40 Immunofluorescent  staining was observed and photographed  using a Zeiss  Axiophot epifluorescence microscope fitted with a 63 (numerical aperture 1.4, oil) objective lens, whereas immunohistochemical (VIP) stainings were viewed with a light microscope adapted with a Nikon Coolpix® digital camera for recording images.  Table 4:  List of Primary Antibodies Immunohistochemical Staining*  Used  for  Immunofluorescent  and  Reference/Manufacturer  Antibody  Dilution  Anti-ILK-1  1:550  Upstate™ Cell Signaling Solutions, N Y , USA  Anti-Mig-2  1:500  (ascites preparation)'  Anti-migfilin  1:1000  (ascites preparation)'  Anti-paxillin  1:100  Anti-kindlin-1  undiluted  (Clone 65.1.9)  C H E M I C O N International, Inc., C A , U S A (serum supernatant)'  *Note: all antibodies were monoclonal except for kindlin-1, which was polyclonal; 'Antibodies provided by Dr. Chuanyue Wu, Dept. of Pathology, University of Pittsburgh, Pittsburgh, PA, USA  Table 5:  List of Secondary Antibodies Used for Immunofluorescence*  Antibody-Conjugated Fluorophore  Dilution  Alexa Fluor® 546' (orange/red)  1:100  Molecular Probes, Invitrogen® Detection Technologies, Eugene, Oregon, U S A  Alexa Fluor® 488' (green)  1:100  Molecular Probes, Invitrogen® Detection Technologies, Eugene, Oregon, U S A  'The number refers to the wavelength  Manufacturer  41 3.4  Cell Culture  The immortalized epidermal keratinocyte cell line, HaCaT was obtained as a generous gift from Dr. Norbert Fusenig (German Cancer Centre, Heidlberg, Germany). The cells were maintained in Dulbecco's modified Eagle's medium ( D M E M ; Gibco B R L Life Technologies, Rockville, M D , USA) supplemented with 23 m M sodium bicarbonate, 20 m M HEPES, antibiotics (50 ug/ml streptomycin sulfate, 100 U/ml penicillin) and 10% heat-inactivated fetal calf serum (FCS; Gibco BRL). The HaCaT cell line has been shown to mimic many of the properties seen in normal epidermal keratinocytes, it is not invasive and is able to differentiate under appropriate conditions (Boukamp et al., 1988). Cells were always subcultured at least 2 days before experiments to maintain consistency throughout the study. HaCaT cells were cultured in T25 (50 ml) or T75 (150 ml) Tissue Culture Flasks (Polystyrene; fitted with 0.2 um vented seal cap; Becton, Dickinson and Company, Franklin Lakes, NJ, USA).  3.5  K i n d l i n - 1 S i R N A Transfection of H a C a T cells  In order to investigate the effect of small interfering R N A (siRNA) mediated gene silencing of kindlin-1 on HaCaT cell spreading, a 21-nucleotide siRNA targeting the kindlin-1 ORF (Open Reading Frame): r ( G A A G U U A C U A C C A A A A G C U ) d ( T T ) (sense strand), originally described by Kloeker et. al. (2004), and a 21-nucleotide irrelevant siRNA, Stealth™ R N A i Negative Control (non-sense R N A , N - R N A ;  Stealth™ R N A i  Negative Control Kit, Invitrogen™) were individually synthesized by and purchased from Invitrogen™. BLOCK-iT™ Fluorescent oligo siRNA (F-RNA; Invitrogen™, Burlington, ON, Canada) and N - R N A , described by Wienke D et al. (2003), which are not  42 homologous to any human genes, were used as negative controls. For the siRNA complex, siRNA and lipofectamine 2000 (Invitrogen™, Burlington, O N , Canada) were mixed with an equal volume of 150 uL and incubated for 20 minutes at room temperature; they were later diluted with Opti-MEM® (Gibco B R L , Rockville, M D , USA) individually and preincubated at room temperature for five minutes. When HaCaT cells were 70% confluent, approximately 6 X 10 cells were 5  harvested by trypsinization and centrifugation, and suspended in 1.2 ml serum-free, antibiotic-free D M E M medium. Subsequently, cell suspensions were combined with 300 pi of 150 p M siRNA-lipofectamine 2000 complex and mixed gently with rotator at room temperature for 30 minutes. These cells were then seeded onto 6-well plates (Becton Dickinson, Franklin Lakes, N J , USA) and incubated at 37°C with 5% C 0 for 3 hours. 2  At this point the cells were supplemented with 120 ul of FBS and continued incubation for 24 hours and 48 hours at 37°C with 5% C O 2 for R N A extraction at these different time-points. Western blot analysis was used to confirm the knockdown of the kindlin-1 protein (Fig. 8 of Results).  3.6  Semi-quantitation of Kindlin-1 m R N A by R T - P C R In order to determine whether siRNA-mediated silencing of the KIND1 gene  (using kindlin-1 siRNA ) semi-quantitation of the product, kindlin-1 mRNA, relative to F-mRNA or N-mRNA negative controls, was performed as follows. Total R N A was extracted from the transfected cells with RNeasy Mini Kit (Qiagen). R N A integrity was assessed  by R N A formaldehyde  agarose gel electrophoresis,  and concentration  quantitated spectrophotometrically. One microgram of total R N A was reverse-transcribed  43 using  Oligo(dT)  primer  with  the  Superscript  First-Strand  Synthesis  System  (Invitrogen™). For each P C R reaction, 1 pi of reverse-transcribed R N A was combined with a total volume of 50 pi containing 20 m M Tris-HCl (pH 8.8), 10 m M KC1, lOmM ( N H ) S 0 , 2 m M M g S 0 , 0.1% Triton-XlOO, 10 m M dNTPs, 10 u M each of 4  2  4  4  oligonucleotide primers and 1 pi of 1 unit/ul D N A polymerase (total volume of 50 uL). An initial denaturation step at 94 °C for 90 seconds was followed by different cycles consisting of denaturation at 94 °C, annealing at 62°C, and extension at 72°C, each for 40 seconds. A final extension step was carried out at 72°C for five minutes. The designed primers for kindlin-1 were 5' - T G G T T C A G T G A C A G C C C T T T G A 3' and 5 ' - C A C A A C T T C G C A G C C T C T A A G - 3 ' corresponding to 613-635 and 12581279 of kindling-1 cDNA, respectively.  The primers for actin (control), 5'-  G A G A C C T T C A A C A C C C C A G C C - 3 and 5 ' - G G C C A T C T C T T G C T C G A A G T C - 3 , were expected to generate a 311 -bp band (Redlich M et al., 2001). RT-PCR products were separated by agarose gel electrophoresis, stained with ethidium bromide for visualization under  U V light, photographed,  and  quantified  using N I H I M A G E  software  (http://rsb. info.nih. gov/ni h-image) as shown in Figure 8 of the Results section.  3.7  Immunocytochemistry - L o c a l i z a t i o n of I L K - 1 , M i g - 2 , M i g f i l i n , K i n d l i n - 1 , Paxillin, and Actin  Immunolocalization was carried out as described previously (Larjava et al., 1990). HaCaT cells were cultured in T25 flasks (Invitrogen, Burlington, O N , Canada) in Dulbecco's modified Eagle's medium as described above. The cells were either transfected by kindlin-1, F-RNA, or N - R N A as described above or left non-transfected  44 (control). For control samples, HaCaT cells were trypsinized before exceeding 75% confluence. SiRNA-transfected cells were trypsinized 48 hours after transfection, seeded on glass cover slips and allowed to spread overnight at 37°C (time allowed for focal complexes to form) in the presence or absence of transforming growth factor P (TGF-P; 10 ng/ml; Chemicon, Temencula, C A , USA). The kinderlin gene (also known as KIND1) has been shown to be upregulated by TGF-beta (Jobard F et al., 2003). Kloeker et al. (2004) has demonstrated that kindlerin (also known as kindlin-1) expression is responsive to TGF-p levels. Cells were washed two times with PBS and then fixed with 4% formaldehyde/5% sucrose in PBS for 20 minutes at room temperature for non-transfection experiments. For siRNA transfections cells were fixed and permeabilized using acetone prechilled at -20°C for 4 minutes. Cells used in non-siRNA transfections were permeabilized with 0.5% Triton X-100 in PBS+ (PBS containing 1 m M C a  2+  and 0.5 m M M g ) for four minutes at 2+  room temperature. After rinsing in PBS the cover slips blocking solution (10 mg/ml B S A plus 1 mg/ml glycine in PBS+) was added for 30 minutes. Cover slips were then washed with PBS and incubated with primary antibodies against focal adhesion proteins ILK-1, Mig-2, migfilin, kindlin-1, and paxillin, as well as filamentous actin (F-actin) for 1 hour at room temperature. Stainings were conducted as described in the tissue section immunofluorescent staining above. For double staining, FITC and Alexa Fluor 594 were used in combination via separate 1-hour incubations. F-actin was stained with FITCphalloidin. The cells were then washed twice with PBS. Coverslips were mounted onto microscope slides using 50% glycerol in PBS, blotting the excess mounting media and  45 then sealing the edges with nail polish and stored at -20 °C. Immunofluorescent staining was observed and photographed as above.  3.8  Cell Migration - In Vitro Wounding  Scratch-wound migration assays were performed as described Koivisto et al. (personal communication). HaCaT cells were seeded on 24-well plates (400,000 cells per well; Becton Dickinson, Franklin Lakes, NJ, USA) in D M E M containing 10% FCS. The cells were grown to confluence for two days after which the cultures were scratched to create a wound using a 200 ul pipette tip. Loose, poorly attached cells as a result of the scratching were removed by washing with PBS. Then, 300 ul Ca -free minimum +2  essential medium ( E M E M ; Bio-Whittaker, Walkersville, M D , USA) supplemented with 1% Ca -fee FCS (Hakkinen L et al., 2002) and 1 ng/ml epidermal growth factor (EGF)+2  related peptide (heparin-binding EGF; HB-EGF; Invitrogen™) was added to the culture media and cells were allowed to migrate across the wound for 24 hours. After 24 hours of HaCaT cell migration, cells were fixed as described above, stained with crystal violet, and analyzed for migration. To this end, 8 fields per treatment group (10X magnification; identical dimensions chosen per field; light microscope) were randomly chosen and the number of cells per field counted. The experiment was repeated 3 times for reproducibility for a total of n = 24 fields per treatment group. Treatment group comparison was made between kindlin-1 SiRNA transfected versus F-RNA or N - R N A .  46 3.9  Cell Spreading Assay HaCaT cells were cultured as described above. In order to control for the effect of  transfection per se, comparison was made between kindlin-1 siRNA transfected versus FR N A or N - R N A . After 48 hours of siRNA transfection with kindlin-1 siRNA, F-RNA, or N - R N A , HaCaT cells were trypsinized and seeded on glass coverslips in 24-well plates (50,000 cells per well; Becton Dickinson, Franklin Lakes, N J , U S A ) and allowed to spread overnight at 37°C. In a set of experiments, after kindlin-1 siRNA transfection of HaCaT cells for 48 hours and trypsinization, cells were seeded on coverslips pre-coated with bovine fibronectin (Bov-FN; 20 pg /ml in PBS+, from bovine plasma; Chemicon, Temencula, C A , USA), type I collagen (from bovine skin type I collagen, Vitrogen; 20 Ug/ml in 0.01M Acetic Acid), or laminin 10/11 (5 ug/ml in PBS+, Gibco B R L Life Technologies, USA), and heat-denatured B S A (10 mg/ml) as control. Coatings were performed the day before seeding of HaCaT cells and allowed to dry in a laminar flow cell culture hood prior to use at which time the coverslips were rehydrated by washing with PBS three times. HaCaT cells were only allowed to spread for 4 hours in this experiment based on previous results with coated coverslips whereby maximal spreading was consistently found at 4 hours (data not shown). To quantify the effect of altered kindlin-1 R N A expression on HaCaT keratinocyte cell spreading, the total number of cells attached (total number of cells visible per unit area in a 40X field; n = 27 fields for each treatment group) as well as the percentage of cells spread (ratio of total number of cells considered spread to the total number of cells present in a 40X field; n = 27 fields for each treatment group) was calculated. Cells were considered spread if they demonstrated clearly visible lamellopodia.  47 Finally, in order to examine the short-term effects of altered kindlin-1 R N A expression on HaCaT keratinocyte cell spreading (i.e. their ability to attach), HaCaT cells were seeded in wells of 96-well plates (10,000 cells per well; Flacon, Becton Dickinson, Franklin Lakes, N J , USA). Plates were pre-coated with either type I Collagen, Bov-FN, or laminin 10/11 as described above. Cells were fixed at two time points, when at least 50% of cells in the well appeared spread (as above) and when 75% of cells appeared spread. These time points corresponded to 45 minutes and one hour for type I collagen and laminin 10/11 and 45 minutes and 90 minutes for Bov-FN, respectively. The number of cells attached was calculated as above. In order to calculate the percentage of cells spread, the ratio of cells spread to the total number of cells present (attached) was calculated. Cells were considered spread i f they clearly demonstrated the presence of lamellae.  3.10  Cell Proliferation Assay To estimate the effect of altered kindlin-1 R N A expression on HaCaT  keratinocyte cell proliferation, kindlin-1 SiRNA transfected HaCaT cells were seeded on 96-well plates (20,000 cells/well; 50 ul/well) in D M E M and allowed time to attach at 37°C with 5% C O 2 for 3 hours. After three hours, FCS was added to the media to a final concentration of 10% FCS and 7.5 ul of Promega® reagent (Promega CellTiter 96® Non-Radioactive Cell  Proliferation  Assay, Promega, Madison, WI, U S A ) . The  proliferation assay was terminated at 6-, 9-, 12-, 24-, 48-, and 72-hours (equivalent to time elapsed from the time of transfection; from seeding; time "0") and the number of viable cells measured by reading absorbance at 595 nm.  48  3.11  Statistical Analysis  A l l graphical and tabular data are expressed as the mean ± standard error of the mean (s.e.m.). Differences between two individual means were calculated the using unpaired t-tests with Welch's correction. Bonferroni's multiple comparison one-way analysis of variance ( A N O V A ) was used to compare means of multiple treatment groups included in the same experiment.  Differences between  means were considered  statistically significant i f the p-value was < 0.05. A l l statistical calculations were made using Prism® (version 2; GraphPad Software Inc, San Diego, C A , USA) software and all graphics were constructed using Cricket Graph III (version 1.5.3; Computer Associates International, Inc., Islandia, N Y , U.S.A.).  49  CHAPTER IV Results 4.1.1  Immunolocalization of ILK-1, Mig-2, Migfilin, Paxillin, and Kindlin-1 in Human Oral Mucosa VIP immunostaining of human oral palatal epithelium (Fig. 3) and oral gingival  epithelium (Fig. 4, A-F) did not demonstrate gross differences in the localization of focal adhesion proteins, ILK-1, Mig-2, migfilin, paxillin, and kindlin-1 with one exception. ILK-1, Mig-2, migfilin, paxillin, and kindlin-1 had the same distribution of localization in gingiva that was distant from the tooth per se. However, gingiva nearer the tooth (the free gingiva) demonstrated lower intensity of immunostaining up to the junctional epithelium (JE, Fig. 4) where the level of intensity was restored to levels similar to that of palatal epithelium. A novel finding was the immunolocalization of migfilin, paxillin, and kindlin-1 to the basement membrane zone (BMZ) of oral palatal epithelium (Fig. 3). A n additional novel finding was the immunolocalization of ILK-1, Mig-2, migfilin, and paxillin to blood vessels within the connective tissue, except for kindlin-1. ILK-1 and Mig-2 predominantly localized in suprabasal cell layers between cells, while migfilin and kindlin-1 localized to both the B M Z and between cells suprabasally. Paxillin was most specific in it's localization, found virtually exclusively at the B M Z of the epithelium and blood vessels. A summary of the distribution of these focal adhesion proteins is presented in Table 6.  50  Figure 3: Immunolocalization (VIP Staining) of ILK-1 (B,H), Mig-2 ( C J ) , Migfilin (D,J), Paxillin (E,K), and Kindlin-1 (F,L) in Human Oral Palatal Epithelium (B-F) with a Magnified View of their Localization in Blood Vessels (H-L). Sections (A) and (G) represent H & E stained controls. A magnified view of the basement membrane zone (BMZ) and basal cell layer is provided in the top-right corner (A-F). ILK-1 (B) and Mig2 (C) do not localize to the basement membrane zone or the basal cells, but are found in between cells in the suprabasal layers and blood vessels (H and I, respectively). Migfilin (C), paxillin (E), and kindlin-1 (F) localize strongly to the basement membrane zone and between cells throughout all layers of the epithelium with migfilin (C) and paxillin (D) also localizing to blood vessels. Kindlin-1 (L) does not localize to blood vessels.  51 4.1.2  Immunolocalization of ILK-1, Mig-2, Migfilin, Paxillin, and Kindlin-1 in Human Junctional Epithelium  The junctional epithelium (JE) faces both the gingival connective tissue and the tooth surface. A basement membrane, sometimes referred to as the "external basal lamina" (EBL, Schroeder, 1996), is interposed between the basal cells of the JE and the gingival connective tissue. Between the tooth and the basal cells of the JE lies a basal lamina (also known as the internal basal lamina, IBL), which forms together with hemidesmosomes (Listgarten, 1972a) the interface between the tooth surface and the JE and is named 'epithelial attachment' (Schroeder and Listgarten, 1997). VIP immunostaining of human junctional epithelium (Fig. 4) did not demonstrate gross differences in the localization of focal adhesion proteins, ILK-1, Mig-2, migfilin, paxillin, and kindlin-1 when compared to that observed in oral mucosa (Fig. 3). As in oral palatal epithelium migfilin, paxillin, and kindlin-1 localized to the B M Z , in this case, that of the EBL. ILK-1 and Mig-2, on the other hand, did not localize to the B M Z or to the basal cell layer of the E B L . Interestingly, all focal adhesion proteins investigated, ILK-1, Mig-2, migfilin, paxillin, and kindlin-1, localized to the IBL zone. As well, comparison between E B L and IBL staining intensity showed migfilin, paxillin, and kindlin-1 to be greater at the IBL zone (against the tooth) than at the E B L . Only ILK-1, Mig-2, migfilin, and paxillin localized to blood vessels. The difference in localization between these focal adhesion proteins in different areas of the JE may translate into differences in their function. A summary of the distribution of these focal adhesion proteins via immunolocalization is presented in Table 6.  52  Figure 4: Immunolocalization (VIP Staining) of ILK-1 (B,H), M i g - 2 (C,I), Migfilin (D,J), Paxillin (E,K), and Kindlin-1 (F,L) in Human Oral Junctional Epithelium (JE). Location of the JE is indicated by a rectangle in whole tissue sections (B-F). A magnified view of their localization in JE is provided in sections H - L with a closer look at the B M Z of the external basal lamina (EBL) at the top left corner (H-L). Sections (A) and (G) represent H & E stained controls. ILK-1 (H), M i g - 2 (I), migfilin (J), paxillin (K), and kindlin-1 (L) immunolocalize to the internal basal lamina (right side of JE, adjacent to tooth). ILK-1 (H) and M i g - 2 (I) do not localize to the B M Z of the E B L (left side of JE, connective tissue side) or the basal cells, but are found in between cells in the suprabasal layers and blood vessels. Migfilin (J), paxillin (K), and kindlin-1 (L) localize strongly to the B M Z of the E B L and between cells throughout all layers of the epithelium with migfilin (J) and paxillin (K) localizing to blood vessels.  53 4.1.3  Immunolocalization of Type IV Collagen, ILK-1, Mig-2, Migfilin, Paxillin, and Kindlin-1 in Human Oral Palatal Epithelium at Day "0" of Wounding and Day "3" after Wounding Immunofluorescent staining of human palatal epithelium (Fig. 5, A-F) revealed no  gross differences with that of VIP staining. Type IV collagen, which is typical found in the basement membrane was used as a control and intensely localized to the B M Z of the epithelium and endothelium of blood vessels within the supepithelial connective tissue. At Day "3" after wounding (Fig. 5, G-L) type IV collagen was missing from the B M Z at the leading edge of the migrating tip of wounded epithelium. On the other hand, patterns of immunolocalization of ILK-1, Mig-2, migfilin, paxillin, and kindlin-1 appear to remain consistent, with no localization of ILK-1 or Mig-2 to the B M Z , while that of migfilin, paxillin, and kindlin-1 remain continuous from the adjacent (Fig. 5, G-L, top and left) non-wounded epithelium and throughout the B M Z of the migrating tip of epithelium where it terminates into the granulation tissue (Fig. 5, G-L, bottom and right). A summary of the distribution of these focal adhesion proteins via immunolocalization is presented in Table 6.  V  54  *•«  »'  1  I  i*>\  i,  1  i  • •  B.?;,.-"c ,r  ;.? v<<-.  • ».  ;  ;  H  ;  • •,..  •  > X''  £  : ' -<lD i  f M' • ^ _ -v.  F  , ' 'K  - • iu*.  •s  ;  y  /  ,.  E  j #  I  :  '  Figure 5: Immunolocalization (Alexa Fluor® 546 Immunofluorescent Staining) of Type IV Collagen (A,G), ILK-1 (B,H), Mig-2 (C,I), Migfilin (D,J), Paxillin (E,K), and Kindlin-1 (F,L) in Human Oral Palatal Epithelium at Day "0" of Wounding (A-F) and Day "3" after Wounding (G-L). A magnified view of the B M Z and basal cell layer is provided in the top-right corner of each section except for type IV collagen (A,G; control) at day "3". ILK-1 (B, H) and Mig-2 (C, I) do not localize to the B M Z or the basal cells, but can be found between cells in the suprabasal layers at both at days "0" and "3" of wounding. Type IV collagen (A,G) and migfilin (C,I) localize to the B M Z at day "0". Paxillin (E,K) and kindlin-1 (F,L) localize strongly to the B M Z and lightly between cells suprabasally at both days "0" and "3" of wounding.  55 Table 6: Localization and Relative Staining Intensity of ILK-1, Mig-2, Migfilin, Paxillin, and Kindlin-1 Focal Adhesion Proteins in Oral Gingival and Palatal Epithelium, 3 Dayold Palatal Wounds, and Junctional Epithelium of a Non-Periodontally Diseased Tooth." ILK-1  Mig-2  Migfilin  Paxillin  Kindlin  +++  +++  +++  +  +  +  +  ++  Gingival Epithelium (VIP Stain)  Suprabasal epithelial cells Basal cells BMZ Subepithelial connective tissue Blood vessels  -  -  ++  +++  ++  +  ++  ++  -  -  +++  +++  +++  +++  -  +++  +++  ++  +  ++  -  -  ++  +  ++  -  -  ++  +++  ++  +  ++  +  -  -  +++  ++  ++  +++  -  ++  +++  ++  +  +  -  -  +  +  +  Palatal Epithelium (VIP Stain)  Suprabasal epithelial cells Basal cells BMZ Subepithelial connective tissue Blood vessels Day "O" Palatal Epithelium (Immunofluorescent Stain)  Suprabasal epithelial cells Basal cells BMZ Subepithelial connective tissue Blood vessels  -  -  ++  +++  +++  +  +  ++  -  -  +++  +++  +++  +++  -  ++  +++  +++  ++  +  -  -  -  +  -  -  -  +  +++  ++  +++  +++  +  +  Day " 3 " Palatal Migrating Wound Tip (Immunofluorescent Stain)  Suprabasal epithelial cells Basal cells BMZ Junctional Epithelium (VIP Stain)  +++ +++ +++ Internal basal lamina (IBL) Zone ++ ++ ++ Suprabasal epithelial cells + Basal cells ++ BMZ of external basal lamina (EBL) ++ ++ ++ Subepithelial connective tissue +++ +++ ++ Blood vessels "Relative intensity of immunostaining: +++, high; ++, moderate; +, low; -, no staining.  +  +  ++  ++  -  -  +++  56 4.2  Immunolocalization of Actin, ILK-1, Mig-2, Migfilin, and Kindlin-1 in HaCaT Keratinocytes Immunofluorescent staining HaCaT cells spread over night (Fig. 6)  demonstrates the localization of ILK-1, Mig-2, migfilin, and kindlin-1 to cell-ECM adhesions and the presence of filamentous actin seen terminating at cytocellular processes (i.e. lamellopodia). Addition of TGF-p (10 ng/mol) to the culture media resulted in a qualitative increase in the expression of actin as well focal adhesion proteins, ILK-1, Mig-2, migfilin, and kindlin-1 as seen by an increase in their number and staining intensity when compared to non-TGF-p treated controls. In most cases, TGF-P stimulation resulted in less diffuse cytoplasmic staining suggesting a reorganization of the focal adhesions as a result of increased cell spreading induced by TGF-P stimulation. As noted by Kloeker et al., (2004), TGF-P resulted in an enrichment of kindlin-1 along the confines of the cell membrane (Fig. 6; K,L). The author (Kloeker et al., 2004) also noted 'a notable increase in kindlerin expression in a more diffuse cytoplasmic staining pattern', which was also frequently noted in the present study. Some authors have demonstrated that filamentous actin terminates at sites of both migfilin (Tu et a l , 2003) and kindlin-1 (Siegel et al., 2003; Kloeker et al., 2004). These findings were confirmed in the present study as shown in Figures 7 and 8, respectively.  57  Figure 6: Immunolocalization of Filamentous Actin (C,D) and Focal Adhesion Proteins ILK-1 (E,F), Mig-2 (G,H), Migfilin (I,J), and Kindlin-1 (K,L) in HaCaT Keratinocytes Spread for 16 hours (overnight), in the presence of 10 ng/mol TGF-(3 (A,C,E,G,I,K) or Control (no TGF-P; B,D,F,H,J,L). Secondary antibody alone (A,B) was used as a negative control. A magnified view of the focal adhesions is provided in the bottom-right corner of each section. TGF-P appears to result in increased immunostaining intensity of actin filaments and an increase in the size of ILK-1, Mig-2, migfilin, and kindlin-1 focal adhesions. A similar result was obtained with paxillin (data no shown).  58  Figure 7: Migfilin and Actin Immunofluorescent Double Staining. HaCaT keratinocyte spread for 16 hours (overnight) with double staining of migfilin (Alexa Fluor® 488; Green) and actin (Rhodamine; Orange/Red) in the presence of 10 ng/mol TGF-P. In the zoomed-in view, filamentous actin can be seen as cables terminating at areas of migfilin focal adhesions (edge of the cell) and focal contacts (underneath the cell).  Figure 8: Kindlin-1 and Actin Immunofluorescent Double Staining. HaCaT keratinocyte spread for 16 hours (overnight) with double staining of kindlin-1 (Alexa Fluor® 488; Green) and actin (Rhodamine; Orange) in the presence of 10 ng/mol TGF-p. In the zoomed-in view, filamentous actin can be seen as cables terminating at areas of kindlin-1 focal adhesions (edge of the cell) and focal contacts (underneath the cell).  59  4.3  R T - P C R Confirmation of Kindlin-1 Knockdown via siRNA-mediated Gene Silencing of K I N D 1  Kindlin-1 siRNA transfected HaCaT cells were harvested 48 hours after adding siRNA complexes and examined for kindlin-1 expression. Actin was used as an internal control. Kindlin-1 mRNA was reduced in lysates prepared from HaCaT cells transfected with kindlin-1 siRNA as compared to lysates prepared from cells transfected with an irrelevant siRNA (negative control, N-RNA). RT-PCR was performed and agarose gel electrophoresis of the P C R products was run to evaluate the reduction of kindlin-1 mRNA (Fig. 9). Kindlin-1 Dilution  #  ft>  >  Actin  <V »N"  o N*  >V  -  IR +  IR-  Kindlin-1 +  Kindlin-1-  Control  Figure 9: RT-PCR for the Evaluation of Kindlin-1 mRNA Reduction by siRNA Mediated Gene Silencing of KIND 1. HaCaT cells were transfected by siRNA complexes and cultured in 1% FBS D M E M medium with 10 ng/mol TGF-P (Kindlin-1+) or control (no TGF-P, Kindlin-1-) for 24 hours. Cellular mRNA was prepared from cell lysates and reverse transcribed. 4 ul of a series of RT dilutions were used as a template for PCR analysis. The partial PCR products (20 ul) were electrophoresed on agarose gels, stained with ethidium bromide and analyzed with NIH Image software. The optical density from serial dilutions resulted in a curve for each sample, which was used to calibrate the knockdown gene ratio (data not shown). Kindlin-1 mRNA from transfected cells (Kindlin-1+ and Kindlin-1-) was down approximately 70% (using NIH Image software) compared to kindlin-1 mRNA from irrelevant siRNA with or without TGF-P (IR+ and IR-, respectively). The knockdown ratio was greatest in the presence of TGF-p.  60 4.4.1  Effect of Kindlin-1 siRNA Transfection on the Expression of Actin, ILK-1, Mig-2, Migfilin, and Kindlin-1 in HaCaT Keratinocytes  Immunofluorescent staining of kindlin-1 siRNA transfected HaCaT cells spread over night (Fig. 10) demonstrates a qualitative decrease in expression of actin. This is likely the result of significantly impaired cell spreading (Fig. 11 and 13). The only other group to investigate the effect of kindlin-1 siRNA transfection on keratinocyte behaviour (Kloeker et al., 2004) demonstrated a delay in cell spreading. However, this group did not examine the effect of kindlin-1 siRNA transfection on the cytoskeletal organization of actin filaments per se. Similarly, kindlin-1 siRNA transfection of HaCaT cells resulted in a qualitative decrease in expression of Mig-2, migfilin, and kindlin-1 as seen by the presence of fewer focal adhesions. Interestingly, there is less cytoplasmic staining and greater enrichment of Mig-2 and kindlin-1 at the membrane periphery in kindlin-1 siRNA transfected cells than control (Fig. 12: E,F,K,L). It is important to note however, that all cells were grown in the presence of TGF-P, which is reported to induce expression of kindlin-1 (Kloeker et al., 2004). Kloeker et al. (2004, Fig. 9A in the article) demonstrated that kindlin-1 siRNA transfection of HaCaT cells resulted in reduced expression of kindlin-1 (kindlerin) expression and delayed cell spreading (Fig. 9C in the article). However, in this study, kindlin-1 was still expressed at sites of focal adhesions (Kloeker et al., 2004, Fig. 9B in the article). The same group transfected cells for approximately 44 hours. In the present study, kindlin-1 siRNA transfection resulted in significant knockdown of kindlin-1 (Fig. 9) and yet kindlin-1 was still expressed in HaCaT cells (Fig. 10, K,L) in agreement with the findings of Kloeker et al. (2004). However, the number of cells attached on coverslips  61 on which they were seeded after transfection, as well as the proportion of cells spread was significantly decreased as a result of the transfection at both 48 and 72 hours cumulative elapsed time of spreading after kindlin-1 siRNA transfection (Fig. 11 and 12, respectively). The following results are in contrast to the finding by Kloeker et al. (2004) that 'adhesion was not affected by kindlerin (kindlin-1) knockdown when tested 30 min after replating on fibronectin or laminin (data not shown)'. In figure 11a, after 48 hours of kindlin-1 siRNA transfection followed by seeding of cells on coverslips and 24 hours of HaCaT cells spreading in the presence of TGF-P, the mean number of cells present in a microscopic field was 30 ± 2 for transfected and 42 ± 2 for control (p<0.0001). In the absence of TGF-p the difference was 25 ± 1 and 32 ± 2 for control (p = 0.0021). Similarly, the percentage of cells spread (relative to the total number attached, Fig. 1 lb) in the presence of TGF-p was 61.2 ± 1.6% for transfected and 75.6 ± 1.1% for control (p<0.0001). In the absence of TGF-p the difference was 52.8 ± 2.3% for transfected cells and 69.2 ± 1.4% for control (pO.0001). Thus, the effect of kindlin-1 siRNA transfection was significant up to at least 72 hours after transfection began. In a similar experiment, HaCaT cells were seeded upon transfection such that the time elapsed for transfection (48 hours) was equal to the time allowed for cell spreading on coverslips (Fig. 12). In figure 12a, after 48 hours of kindlin-1 siRNA transfection and concomitant spreading in the presence of TGF-P, the mean number of cells present in a microscopic field was 28 ± 1 for transfected and 36 + 1 for control (pO.0001). In the absence of TGF-P the difference was 22 ± 1 and 29 ± 1 for control (p = 0.0021). Similarly, the percentage of cells spread (relative to the total number attached, Fig. 12b) in the presence of TGF-P was 38.1 ± 2.5% for transfected cells and 64.6 ± 2% for control  62 (pO.0001). In the absence of TGF-p the difference was 36.4 ± 2.8% for transfected and 57.9 + 1.4% for control (p<0.0001).  . *.  IP  ^  x f ' ./ \  A  •"*  B  H  ^^^^  • I  , it  •V ''"'v"V. ;' f •• -. ?  1'  c  ».  i  - ,  • -A: " ;• •„  J  D *#»  E  F  •'..>  ^  '  L  Figure 10: Immunolocalization of Filamentous Actin (A,B) and Focal Adhesion Proteins ILK-1 (C,D), Mig-2 (E,F), Migfilin (G,H), Paxillin (I,J), and Kindlin-1 ( K , L ) in Kindlin-1 siRNA Transfected (A,C,E,G,I,K) HaCaT Keratinocytes in the Presence of 10 ng/mol TGF-p. Controls (B,D,F,H,J,L) were transfected with BLOCK-iT™ Fluorescent Oligo (F-RNA). Cells were transfected for 48 hours, trypsinized, seeded, and then allowed to spread for 16 hours (overnight). Kindlin-1 siRNA mediated gene silencing of kindlin-1 expression appears to result in decreased expression in the number of actin filaments and an decrease in the size of ILK-1, Mig-2, migfilin, paxillin, and kindlin-1 focal adhesions.  63  b)  a)  TGF-B+  TGF-B-  TGF-B+  TGF-B-  Figure 11a) and b): Spreading of HaCaT Keratinocytes over 24 Hours. Cells were transfected with kindlin-1 siRNA or F-RNA (control) for 48 hours, trypsinized, and then allowed to spread for 24 hours on glass coverslips in the presence of 10 ng/mol TGF-P. The total number of cells attached (a) and the percentage of spread cells relative to the total number attached (b) was calculated. One typical experiment out of two is shown. Error bars represent standard error of the mean (s.e.m) from 3 randomly selected fields from each of a total of seven wells (n = 21). Statistical comparison between treatment groups was made using using A N O V A and the Student's t test for comparison between kindlin-1 transfected cells (SiRNA+) and negative control (N-RNA).  64  b)  a)  TGF-B+  TGF-B-  TGF-B+  TGF-B-  Figure 12a) and b): Spreading of HaCaT keratinocytes over 48 hours. Cells were transfected with kindlin-1 siRNA or F-RNA (control) for 48 hours and allowed to spread during this period of transfection on glass coverslips in the presence of 10 ng/mol TGF-P. The total number of cells attached (a) and the percentage of spread cells relative to the total number attached (b) was calculated. One typical experiment out of two is shown. Error bars represent standard error of the mean (s.e.m) from 3 randomly selected fields from each of a total of seven wells (n = 21). Statistical comparison between treatment groups was made using using A N O V A and the Student's t test for comparison between kindlin-1 transfected cells (SiRNA+) and negative control (N-RNA).  65 4.4.2  Effect of Kindlin-1 siRNA Transfection on the Expression of Actin, ILK-1, Mig-2, Migfilin, and Kindlin-1 in HaCaT Cells Spread on Type I Collagen, Laminin 10/11 and Fibronectin During wound healing, the cutaneous basement membrane is broken down and  keratinocytes are stimulated to migrate over a provisional E C M (Grinnell, 1992). Herein is the first report of immunofluorescent staining of ILK-1 and Mig-2 in kindlin-1 siRNA transfected HaCaT cells spread over four hours on type I collagen (20 ug/ml), laminin 10/11 (5 ug/ml), and bovine fibronectin (FN; 20 ug/ml) (Fig. 13).  Control non-  transfected cells (Fig. 13: B,D,F,H,J,L) are generally more spread than their transfected counterparts (Fig. 13: A,C,E,G,I,K) as a result of kindlin-1 knockdown. Most notable differences in the expression of ILK-1 are seen on laminin 10/11 in which transfected cells (C) are less spread and have fewer focal adhesions than in control (D). As well, ILK-1 appears more effective in forming focal adhesions in the presence of FN compared to type I collagen and laminin 10/11. Similarly, there is more effective spreading and greater expression of Mig-2 in control (L) compared to transfected (K) cells on fibronectin. In contrast to ILK-1 Mig-2 forms focal adhesions more effectively in the presence of type I collagen and F N compared to laminin 10/11. Similarly, however, both ILK-1 and Mig-2 are not very effective at forming focal adhesions in the presence of laminin 10/11. Figure 14 shows immunofluorescent staining of migfilin (A-F) and paxillin (G-L) in kindlin-1 siRNA transfected HaCaT cells spread over night on type I collagen, laminin 10/11, and F N . Again, there are qualitative differences in spreading effectiveness with transfected cells appearing less spread than controls. There is less expression of migfilin in transfected (A,C) than control cells (B,D) for type I collagen and laminin 10/11  66 matrices. In the case of paxillin, there is less expression in transfected (G,I,K) than control (H,J,L) for all three matrices, type I collagen, laminin 10/11, and fibronectin. In comparison to ILK-1 and Mig-2, however, migfilin and paxillin appear able to form focal adhesions on all three matrices. In the case of kindlin-1 expression (Fig. 15), kindlin-1 did not localize to focal adhesions significantly in kindlin-1 siRNA transfected or control (N-RNA) cells. This might simply reflect the possibility that kindlin-1 assembles in to focal adhesions downstream of other F A proteins (i.e. paxillin, ILK-1, Mig-2, and migfilin).  6 7  f  %  l l  p  c  E  B  G  H  D  I  J  i  ^  F  L  Figure 13: Immunolocalization of ILK-1 (A-F) and Mig-2 ( G - L ) in Kindlin-1 siRNA  Transfected (A,C,E,G,I,K) HaCaT Keratinocytes. Controls (B,D,F,H,J,L) were transfected with Stealth™ R N A i negative control (N-RNA). Cells were transfected for 48 hours, trypsinized, seeded, and then allowed to spread for 4 hours on 20 ug/ml type I collagen (A,B>G,H), 5 pg/ml laminin 10/11 (C,D,I,J), and 20 ug/ml bovine fibronectin ( E , F , K , L ) . Kindlin-1 siRNA mediated gene silencing of kindlin-1 expression does not appear to have a measurable effect on the expression of ILK-1, Mig-2, migfilin, paxillin, and kindlin-1 focal adhesions.  68  Figure 14: Immunolocalization of Migfilin (A-F) and Paxillin (G-L) in Kindlin-1 siRNA Transfected (A,C,E,G,I,K) HaCaT Keratinocytes. Controls (B,D,F,H>J?L) were transfected with Stealth™ R N A i negative control (N-RNA). Cells were transfected for 48 hours, trypsinized, seeded, and then allowed to spread for 4 hours on 20 ug/ml type I collagen (A,B>G,H), 5 ug/ml laminin 10/11 (C,D,I,J), and 20 ug/ml bovine fibronectin (E,F,K,L).  69  r •  $  B  A •  * ,  1^  m  E Figure 15: Immunolocalization of Kindlin-1 (A-F) in Kindlin-1 siRNA Transfected HaCaT keratinocytes. Controls (B,C,F) were transfected with Stealth™ R N A i negative control (N-RNA). Cells were transfected for 48 hours, trypsinized, seeded, and then allowed to spread for 4 hours on 20 ug/ml type I collagen (A,B), 5 ug/ml laminin 10/11 (C,D), and 20 ug/ml bovine fibronectin (E,F).  70 4.4.3  Effect of Kindlin-1 siRNA Transfection on HaCaT Cell Spreading on Type I Collagen, Laminin 10/11 and Fibronectin  As mentioned earlier, kindlin-1 siRNA transfection qualitatively resulted in a decrease in the ability of kindlin-1 siRNA cells to spread as compared with control cells in the presence of the majority of matrices investigated. As a result an effort was made to quantify this phenomenon by examining the proportion of cells that were spread at 45 minutes, 60 or 90 minutes (Figures 16, 17, and 18), and four hours (Fig. 19) from seeding on type I collagen, laminin 10/11, and bovine fibronectin matrices. The first time point, 45 minutes, was chosen based on approximately 50% of cells considered spread. 60 minutes or 90 minutes was chosen as the second time point based on approximately 75% of cells considered spread. Cells were considered spread at these two time points if they exhibited lamellae, visualized under a light microscope. The third time point, four hours (Fig. 19), was chosen based on previous findings in our laboratory that non-transfected HaCaT cells were maximally spread at this time point when spread on the aforementioned matrices (data not shown). 96-well plates are made of polystyrene and several authors have demonstrated differences in cell attachment and spreading on different substrata (i.e. glass versus polystyrene; Kleinman et al., 1981; Reuveny et al., 1984; Varani et al., 1985). Therefore, 45-, 60-, and 90-minute spreading on matrix-coated 96-well plates can not be considered a continuum of time relative to four hour spreading on matrix-coated glass coverslips. The results show that in the case of type I collagen (Fig. 16), there was a significant decrease in the proportion of HaCaT cells spread as a result of kindlin-1 siRNA transfection with the greatest difference (approximately 10%) occurring at one  71  hour (59.7 ± 1.9% versus 70.9 ± 1.4%; p<0.0001). The difference was not significant at 45 minutes (ns). After four hours spreading on type I collagen-coated glass coverslips, there was a small, but statistically significant difference (Fig. 19, p=0.0179). When spread on laminin 10/11 (Fig. 17), kindlin-1 siRNA transfection had the greatest effect on HaCaT cell spreading (greater than a 40% decrease at 45 minutes). This difference was greatest at 45 minutes (Fig. 17; 36 ± 2.6% versus 60.5 ± 0.9%; pO.OOOl) with lesser, but statistically significant differences also occurring at one hour (Fig. 17; 54.2 ± 1.1% versus 61.8 1.5%; p=0.0007) and four hours (Fig. 19; 35.2 ± 1.1% versus 47.5 ± 1.4%; p<0.0001). Finally, looking at spreading on bovine fibronectin (Fig. 18), kindlin-1 siRNA had greatest effect (11% to 12% reduction in cell spreading) at 45 minutes (17.3 ± 0.7% versus 29 ± 1.3%; pO.OOOl) and 90 minutes (53.2 ± 1.5% versus 64.2 ± 1%; pO.OOOl). There was no statistically significant difference (ns) at four hours spreading on fibronectin-coated glass coverslips (Fig. 19).  72  Type I Collagen  0.75  •a  •  siRNA  •  N-RNA  H  ^  sJ  on  5  wi * ;  U o  0.5  .2 « Q.JS  a, 0.25 H  45  60 Time - Minutes •  Figure 16: Effect of Kindlin-1 siRNA Transfection on Short-term HaCaT Cell Spreading. Cells were transfected with kindlin-1 siRNA or negative control (N-RNA) for 48, trypsinized, seeded, and then allowed to spread for 45 and 60 minutes on 20 ug/ml type I collagen. 45- and 60- minute spreading was performed in 96-well plates. HaCaT cells were approximately 50% and 75% spread (as evidenced by the presence of lamellae visualized under a light microscope) for 45- and 60-minute time points, respectively. The total percentage (proportion) of spread cells (ratio of spread to total number attached) was calculated. One typical experiment out of two is shown. Error bars represent standard error of the mean (s.e.m) from four randomly selected fields in triplicate (n = 12 per treatment group). Statistical comparison between treatment groups was made using A N O V A and the Student's t-test (ns, non-significant; ***p=0.0001) for comparison between kindlin-1 transfected cells (siRNA) and negative control (N-RNA).  73  Laminin 10/11  •  siRNA  N-RNA  0.75  H  Figure 17: Effect of Kindlin-1 siRNA Transfection on Short-term HaCaT Cell Spreading. Cells were transfected with kindlin-1 siRNA or negative control (N-RNA) for 48, trypsinized, seeded, and then allowed to spread for 45 and 60 minutes on 5 ug/ml laminin 10/11. 45- and 60-minute spreading was performed in 96-well plates. HaCaT cells were approximately 50% and 75% spread (as evidenced by the presence of lamellae visualized under a light microscope) for 45- and 60-minute time points, respectively. The total percentage (proportion) of spread cells (ratio of spread to total number attached) was calculated. One typical experiment out of two is shown. Error bars represent standard error of the mean (s.e.m) from four randomly selected fields in triplicate (n = 12 per treatment group). Statistical comparison between treatment groups was made using A N O V A and the Student's t-test (***p<0.0001; +++P-0.0007) for comparison between kindlin-1 transfected cells (siRNA) and negative control (N-RNA).  74  Figure 18: Effect of Kindlin-1 siRNA Transfection on Short-term HaCaT Cell Spreading. Cells were transfected with kindlin-1 siRNA or negative control (N-RNA) for 48, trypsinized, seeded, and then allowed to spread for 45 and 90 minutes on 20 ug/ml bovine fibronectin. 45- and 90-minute spreading was performed in 96-well plates. HaCaT cells were approximately 50% and 75% spread (as evidenced by the presence of lamellae visualized under a light microscope) for 45- and 90-minute time points, respectively. The total percentage (proportion) of spread cells (ratio of spread to total number attached) was calculated. One typical experiment out of two is shown. Error bars represent standard error of the mean (s.e.m) from four randomly selected fields in triplicate (n = 12 per treatment group). Statistical comparison between treatment groups was made using A N O V A and the Student's t-test (***p<0.0001) for comparison between kindlin-1 transfected cells (siRNA) and negative control (N-RNA).  75  Figure 19: Effect of Kindlin-1 siRNA Transfection on Short-term HaCaT Cell Spreading. Cells were transfected with kindlin-1 siRNA or negative control (N-RNA) for 48, trypsinized, seeded, and then allowed to spread for 4 hours on 20 ug/ml type I collagen-, 5 ug/ml laminin 10/11-, and 20 ug/ml bovine fibronectin-coated glass coverslips. Maximum spreading of non-transfected HaCaT cells was determined to occur at 4 hours based on previous experiments (data not shown). The total percentage (proportion) of spread cells (ratio of spread to total number attached) was calculated. One typical experiment out of two is shown. Error bars represent standard error of the mean (s.e.m) from four randomly selected fields in triplicate (n = 12 per treatment group). Statistical comparison between treatment groups was made using A N O V A and the Student's t-test (ns, non-significant; *p<0.01; ***p<0.0001) for comparison between kindlin-1 transfected cells (siRNA) and negative control (N-RNA).  76 4.4  Effect of Kindlin-1 siRNA Transfection on HaCaT Cell Proliferation and Migration  One group found that depletion of ILK-1 significantly impaired cell spreading and migration (Vouret-Craviari V et al., 2004), while another has shown decreased cell spreading after depletion of kindlerin (Kloeker et al., 2004). Thus, we sought to examine how suppression of kindlin-1 affects HaCaT keratinocyte cell proliferation (Fig. 20) and migration (Fig. 21). SiRNA-mediated gene silencing of kindlin-1 resulted in a significant decrease in HaCaT cell proliferation at 24, 48, and 72 hours after kindlin-1 siRNA transfection, as seen by a significantly lower number of viable cells. Notably, cells approached confluence shortly after 48 hours, which likely explains the sudden decline in cell proliferation beyond this time point (Fig. 20). The greatest difference in proliferation between kindlin-1 siRNA transfected (1.54 ± 0.04) and control (2.52 ± 0.16; pO.OOOl, normalized ratio relative to 6 hour time point) occurred at 48 hours. This 48-hour time point agrees with the minimal siRNA transfection time utilized by Kloeker et al. (2004). The number of viable cells was calculated as the normalized ratio of the number viable cells at 12, 24, 48, and 72 hours to the number of viable cells at 6 hours. The significant difference seen at 24 hours (1.42 ± 0.02 versus 1.73 ± 0.4; pO.OOOl) suggests the effects of kindlin-1 siRNA transfection begin to take effect at this time point. Figures 21a) and b) demonstrate how kindlin-1 siRNA transfection significantly inhibited HaCaT cell migration following scratch-wounding of confluent cell layers (73.4 ± 4.9% versus 29.5 ± 3.1%>; pO.OOOl). This result represents greater than a two-fold difference in migration between kindlin-1 siRNA transfected and control cells (N-RNA).  77  0.75 4 °-5 H 0  —I  12  1  24  1  1  36 48 Time Point - Hours -  1  1  60  72  Figure 20: Proliferation of Kindlin-1 siRNA Transfected HaCaT Keratinocytes using the Promega CellTiter 96® Non-radioactive Cell Proliferation Assay. Cells were seeded on 96-well plates in D M E M and allowed to attach for 3 hours and FCS added to a final concentration of 10%. HaCaT cell proliferation was terminated at 6, 9, 12, 24, 48, and 72 hours from the time of transfection (seeding) and the number of viable cells measured by absorbance at 595 nm. Wells became confluent shortly after 48 hours of transfection. The number of viable cells was calculated as the normalized ratio of the number viable cells at 12, 24, 48, and 72 hours to the number of viable cells at 6 hours (3 hours after seeding, for attachment). Data represent means ± s.e.m of three experiments (n = 12) and were analyzed using A N O V A and the Student's t test (***p<0.0001) for comparison between kindlin-1 transfected cells (SiRNA+) and negative control (N-RNA).  78  Figure 21a) Effect of Kindlin-1 siRNA Transfection on HaCaT keratinocyte Migration. HaCaT Cells were transfected with kindlin-1 siRNA and grown to confluence in Ca+2free E M E M containing 1% FBS. Confluent HaCaT cell layers were scratch-wounded and treated with 1 ng/ml HB-EGF, allowed to migrate for 24 hours, fixed, stained with crystal violet and photographed. Treatment groups include non-transfected (A), negative control (B, N-RNA), lipofectamine control (C), and kindlin-1 siRNA transfected cells (D). Outlines of the original wound margins are marked with dashed lines. The number of migrated cells within the wounds (8 standardized areas per sample) was counted from the digitized images. Combined results (mean ± s.e.m) from three separate migration experiments are shown in Figure 21b). The migration of non-transfected cells (A) was set as 100%. Statistical comparison between treatment groups was made using A N O V A and the Student's t test (***p<0.0001) for comparison between kindlin-1 transfected cells (siRNA) and negative control (N-RNA).  79  CHAPTER V Discussion 5.1  Immunolocalization of ILK-1, Mig-2, Migflin, Paxillin, and Kindlin-1 in Normal and Wounded Oral Mucosa Since Kindler's first report (Kindler, 1954), more than 100 cases have been  described (Ashton, 2004a). The most salient clinical features of Kindler syndrome (KS) that have been reported include skin fragility and photosensivity and early and rapidly progressive periodontitis (Ashton et al., 2004a; Siegel et al., 2003; Wiebe et al., 2003). On an ultrastructural level, several authors have found defects at the level of the basement membrane including disruptions in and reduplication of the lamina densa (Shimizu et al., 1997), cleft formation in the lamina lucida (Yasukawa et al., 2002; Ashton et al., 2004a,b), focal areas of exaggerated apoptosis (Lanschuetzer et al., 2003), and deposition of type VII collagen found abnormally deep in the connective tissue (CT) stroma (Wiebe and Larjava, 1999). Recently, two groups (Siegel et al., 2003; Ashton et al., 2004b) demonstrated localization of kindlin-1 in basal cells of the dermis and along the dermo-epidermal junction of normal skin with broken or absent staining in patients with KS. There have been no reports of the immunolocalization of kindlin-1 in oral mucosa. The present study is the first report of the localization of kindlin-1 in oral gingival, palatal, and junctional epithelium (JE). Results from VIP peroxidase stainings indicate kindlin-1 localizes primarily (high, +++) to the basement membrane zone (BMZ) with lower (+) staining intensity between basal and suprabasal cells (Fig. 3 and 4). This was confirmed in immunoflourescent stainings (Fig. 5 A-F), the result of which was more specific with kindlin-1 localizing to the BMZ. Kindlin-1 did not localize to blood vessels.  80 In comparison, the homologue of kindlin-1, Mig-2 demonstrated a staining pattern opposite to that of kindlin-1, localizing strongly to intercellular spaces in suprabasal layers, subepithelial connective tissue, and blood vessels. The dramatic contrast in immunolocalization between kindlin-1 and Mig-2 suggests although they have significant homology (62%, a.a.- based), there are likely functional differences between these two focal adhesion proteins. I L K has been shown to function primarily by direct binding to the N-terminus of UNC-112 in C. elegans (Mackinnon et al., 2002). Given the intimate association anticipated between ILK-1 and Mig-2, it was not a surprise to see ILK-1 localize in a manner much like Mig-2. ILK-1 was absent at the B M Z and basal cells, but strongly localized to intercellular spaces suprabasally and in connective tissue and blood vessels. Surprisingly, the N-terminus of kindlin-1 lacks homology to the ILK binding site of UNC-112, which supports the suggestion of a different function for kindlin-1 compares to it's homologue, Mig-2. Interestingly, Tu Y and Wu C (personal communication) claim that kindlin-1, like Mig-2, possesses migfilin-binding activity. Migfilin immunolocalized strongly to the B M Z and blood vessels with moderate staining intensity between suprabasal cells. The staining pattern for migfilin supports the possibility of colocalization of migfilin with kindlin-1, which is also supported by the fact that actin filaments terminate at sites of their focal adhesion (Siegel et al., 2003; Kloeker et al., 2004; Fig. 11 of present study). The presence of migfilin between cells agrees with the finding that migfilin localizes not only to cell-ECM adhesions, but also cell-cell adhesions (Gkretsi et al., 2004; Fig. 9 of the present study). Paxillin localized specifically to the B M Z and blood vessels, which confirms the findings of Yuminamochi et al. (2003). The localization of migfilin, paxillin, and kindlin-  81 1 to the B M Z suggests these focal adhesion proteins interact with each other. Tu Y and Wu C (personal communication) claim that kindlin-1, like Mig-2, possesses migfilin binding activity and suggested that migfilin might link kindlin-1 to filamin-containing actin filaments (Wu et al., 2004). Further studies looking at interactions between paxillin, migfilin, and kindlin-1 are clearly indicated. Many proteins have been identified in focal adhesions, particularly at their cytoplasmic face. Some of these proteins have predominantly a structural role, whereas others are involved in signal transduction. Given the complexity of the composition and architecture of focal adhesions, the present study only sought to investigate a few of the proteins that make up a focal adhesion. During wounding of epithelium, several changes occur including the creation of a 'wound matrix' consisting of blood clot (fibrin, platelet plug, fibronectin, inflammatory cells, etc.), damaged basal lamina, and the proteins and glycoproteins of the subepithelial connective tissue (Graber H-G et al., 1999). Adhesive interactions between a cell and its surrounding E C M regulate  its morphology, migratory properties,  growth, and  differentiation. The most notable differences in the immunolocalization of ILK-1, Mig-2, migfilin, paxillin, and kindlin-1 of wounded oral palatal epithelium at day "3" postwounding was a decrease in the expression of migfilin and an increase in the expression of paxillin at the B M Z of the leading wound edge (Fig. 5). Migfilin still maintained strong staining between cells in the suprabasal layers as in non-wounded epithelium. This finding lends support to the findings of Gkretsi et al. (2004), that migfilin localizes to both cell-cell adherens junctions (via P-catenin) and to cell-ECM adhesions via recruitment to Mig-2 (Tu et al., 2003). The authors (Gkretsi et al., 2004) proposed a  82 model 'in which Mig-2 competes with the components of adherens junctions for interacting with migfilin.' In their model, cells with both cell-cell and cell-ECM adhesions have two pools of migfilin, one which binds Mig-2 and the other which associates with components of adherens junctions. As noted by the authors, 'crosstalking' between cell-cell and cell-ECM adhesions is important in various biological processes including epithelial-mesenchymal transition, wound healing, and cancer metastatsis. Interestingly, when comparing non-wounded to day "3" wounded oral mucosa, the pattern of immunolocalization of migfilin changes so as to resemble that of Mig-2 again providing evidence in support of the model proposed by Gretsi et al. (2004). During wound healing, most of the components of the basement membrane zone including type IV and VII collagens, laminin-1 and heparan sulfate proteoglycan are missing underneath migrating keratinocytes (Larjava et al., 1993; Oksala et al., 1995; Hakkinen et al., 2000). The absence of type IV collagen (Fig. 5) was confirmed in the present study. Wiebe and Larjava (1999) revealed normal expression of basement membrane zone components bullous pemphigoid antigens 1 and 2, type IV collagen, laminins-1 and - 5 , and integrins a3(31 and a6p4, with breaks in the lamina lucida including discontinuities in localization of laminin-1 and - 5 , P4 integrin and abnormally deep deposition of type VII collagen in subepithelial CT. Wiebe and Larjava (1999) and Shimizu et al., (1997) claim the discontinuities do not appear to result from a defect at the level of the hemidesmosome. It is possible then that a loss of function of kindlin-1 leads to defects in the normal organization of the basement membrane (as depicted in Fig. 22) and ultimately the integrity of the skin, oral mucosa, and junctional epithelium.  83 Hormia et al. (2001) looked at the characteristics of the 'dento-epithelial junction', namely those of the 'tooth facing (TF) cells' (i.e. those of the IBL), comparing them to the 'connective tissue facing (CTF) cells' (i.e. those of the EBL). Results from immunolocalization showed that the integrin a6pM, hemidesomosomal components BP180, BP230, and HD1 antigen, and laminin-5 colocalize in the TF cells, whereas type IV collagen, laminin 10/11, type VII collagen, which are normally found in the B M , are all absent. The authors (Hormia et al., 2001) concluded that 'exclusive expression' of laminin 5 in IBL indicates that 'the IBL is not a basal lamina by definition but a simple extracellular matrix with no network structure.' In 2003, Wiebe and company looked at 31 patients (18 with Kindler syndrome and 13 without) from rural Panama to determine the extent to which periodontal disease is associated with Kindler syndrome. The major finding of this study was that individuals with Kindler syndrome develop periodontitis at an earlier age and the disease progresses rapidly. The present study is the first report on the localization of focal adhesion proteins ILK-1, Mig-2, migfilin, paxillin, and kindlin-1 in junctional epithelium (JE, Fig. 4). The results indicate there are no gross differences with that of palatal or gingival epithelium. It is important to note, however, that expression of ILK-1, Mig-2, migfilin, paxillin, and kindlin-1 differed in different parts of the gingival keratinized epithelium, namely, that part proximal to the tooth (free gingiva) and that gingiva distant from the tooth (data not shown). Expression of these focal adhesions decreased and in some instances was absent in gingival epithelium proximal to the tooth, but returned to intensities resembling those of palatal epithelium in areas distant from the tooth and in the JE. Interestingly, migfilin, paxillin, and kindlin-1 all localized to cells of the IBL zone as they did in the B M Z of  84 palatal epithelium and that of the EBL. This finding lends further support to a possible interaction between these focal adhesion proteins. Given the limited scope of this thesis, this finding was not explored further. Nevertheless, their localization to both the internal (tooth-side) and the external (gingival CT-side) basal laminae in junctional epithelium, suggests migfilin, paxillin, and kindlin-1 serve a function in maintaining the structural integrity of the tooth attachment apparatus, the periodontium. Unlike other mammalian appendages, such as feathers, hair, and fingernails, the tooth is a transmucosal organ. It is interposed between the soft (gingiva and periodontal ligament) and hard (cementum, enamel, and alveolar bone) tissues of the dentition underscoring its importance in maintaining homeostasis and defense against a constantly changing oral environment (for reviews, see Schroeder, 1996; Schroeder and Listgarten, 1997; Bosshardt and Lang, 2004). Given the presence of aggressive periodontitis in patients with Kindler syndrome (Wiebe et al., 2003), it is possible that a mutation leading to loss of kindlin-1 function could result in a weakened the union between the junctional epithelium and the underlying connective tissue and/or the tooth. Perhaps loss of kindlin1 function results in defective assembly of focal adhesion proteins (ILK-1, Mig-2, paxillin, and migfilin) necessary for mainting cell-ECM adhesion. Alternatively, loss of kindlin-1 function could lead to a defect in the organization of the B M , namely in areas intervening hemidesmosomes.  A further  look at the ultrastructrural findings of  Lanschuetzer et al. (2003) shows defects in the B M between normally appearing hemidesmosomes, with apparent 'dilatations' in areas of the lamina lucida between the hemidesmosomes. Thus, it is possible that the dermal and oral epithelium of Kindler patients still have some degree of adhesion of their epithelium to their underlying  85 connective tissue (via intact hemidesmosomes), but that loss of kindlin-1 function results in altered focal adhesion assembly and organization of non-hemidesmosomal components of the B M . This might explain why loss of kindlin-1 function is not as deleterious as loss of hemidesmosomes (Pulkinnen and Uitto, 1999). On the other hand, at the IBL of the JE, which is clearly unique in structure compared to the B M of the JE (EBL) and that of other oral mucosa and skin (Hormia et al., 2001), it is possible that loss of kindlin-1 function is sufficient to weaken the adhesion of the directly attached to the tooth (DAT, Bosshard and Lang, 2005; Salonen et al., 1989) cells such that the JE is rendered more susceptible to periodontal disease. Future studies in our laboratory will examine whether there are differences in the expression of these focal adhesion proteins in periodontallydiseased JE and Kindler JE.  86  Normal Mucosa  Kindler Mucosa  laniiniiTSl Focal widening  Lamina Lucida  Type VII Collagen  Drawn by:  Giorgio M. Petricca  Figure 2 2 : Visual Summary of the Ultrastructural and Molecular Features of Kindler Syndrome in Comparison to Normal Mucosa. Main histological and ultrastructural abnormalities as summarized in Table 2. The above ultrastructural features are taken from the findings of Wiebe and Larjava (1999) and Lanschuetzer et al. (2003) and include extensive reduplication of the lamina densa, focal apoptosis of basal keratinocytes, focal widening of the lamina lucida, breaks in the localization of laminin-1 and - 5 , and abnormal deposition of type VII collagen deep into connective tissue. The molecular properties are derived from the present study and Kloeker et al. (2004) and Wu C and Tu Y (personal communication) and include loss of kindlin-1 function (lighter yellow circle) with altered cytoskeletal actin organization (dotted black lines inside basal keratinocytes). Note: this drawing is simplified and does not include all structural components.  87 5.2  Immunolocalization of ILK-1, Mig-2, Migflin, Paxillin, and Kindlin-1 in HaCaT cells and the effect of Kindlin-1 siRNA Transfection  TGF-P signaling plays a role in carcinogenesis, autoimmunity, angiogenesis, and wound healing (Kim et al., 2005). Degranulation of platelets at sites of injury releases a bolus of TGF-P 1, resulting in an increase in the production of E C M by inducing various collagen gene promoters (Flanders, 2004). Topical application of TGF-P improves healing, even in radiation-impaired wounds (Bernstein et al., 1991). Kloeker et al. (2004) demonstrated TGF-P stimulation resulted in a marked induction of kindlin-1 R N A with western blotting demonstrating a corresponding increase in protein abundance. The same authors also demonstrated a change in the cytoplasmic staining pattern of actin, which was confirmed in the present study (Fig. 6). As noted by others (Kloeker et al., 2004) and in the present study, in non TGF-P treated control cells, the actin cytoskeleton network organized as a network of filaments circumscribing each cell colony. However, in the presence of TGF-p, actin filaments reorganized to establish 'actin arrays that are typical of fibroblastic cells.' Some authors have demonstrated that filamentous actin terminates at sites of both migfilin (Tu et al., 2003) and kindlin-1 (Siegel et al., 2003; Kloeker et al., 2004). These findings were confirmed as shown in figures 7 and 8, respectively. In the present study, TGF-P stimulation generally resulted in less diffuse cytoplasmic with increased staining at sites of focal adhesions suggesting a reorganization of the focal adhesions as a result of increased cell spreading induced by TGF-P stimulation. As noted by Kloeker et al., (2004), TGF-P resulted in an enrichment of kindlin-1 along the confines of the cell membrane (Fig. 6; K,L). The author (Kloeker et al., 2004) also noted 'a notable increase in kindlerin expression in a more diffuse cytoplasmic staining  88 pattern', which was also frequently noted in the present study. Interestingly, the authors found peak levels of kindlerin (also known as kindlin-1) mRNA and protein occurred at 6 and 48 hours, respectively, after treatment with the cytokine (Kloeker et al., 2004). Thus, it is possible the effect of TGF-P stimulation seen in the present study are underestimated since HaCaT cell spreading in the presence of TGF-P was allowed for only 16 hours. The gene responsible for KS has been designated as kindlerin by one group (Jobard et al., 2003) and KIND1 by another (Siegel et al., 2003). Since the discovery of the K I N D l  gene several authors have used siRNA-mediated gene silencing and  immunolocalization technique in attempt to elucidate the mechanism, which might explain the clinical outcomes Kindler's syndrome at a molecular level. Siegel et al. (2003) proposed that 'the loss of cytoskeletal-ECM adhesion noted in nematodes carrying mutations in UNC-112 would seem to be analogous to the adhesive defect observed in the skin of patients with Kindler syndrome.' Kindlin-1, the protein expressed by K I N D l , demonstrates closest homology to UNC-112 (Siegel et al., 2003), which in turn has closest homology with Mig-2 in humans. Kindlin-1 has 62% a.a identity with Mig-2 (Tu et al., 2003). UNC-112 binds to a molecule with potential adapter and signaling functions, PAT-4, the nematode homolog of ILK-1 (Mackinnon et al., 2002). Given the complex molecular composition of and interactions at focal contacts and adhesions alike (Zamir and Geiger, 2001a,b; Zaidel-Bar, 2004), the identification of binding partners of kindlin-1 and its homologues is key to understanding the function of kindlin-1. It remains to be seen whether the main function of kindlin-1 is primarily that of structural tethering (i.e of basal cells to the basement membrane) or whether it is that of regulation and/or recruitment of other molecules mediating actin-ECM adhesion. In the present study  89 siRNA mediated knockdown of K I N D l (Fig. 9) resulted in a qualitative decrease in localization of actin, Mig-2, migfilin, and kindlin-1 (Fig. 10). It is important to note however, that all cells were grown in the presence of TGF-P, which is reported to induce expression of kindlin-1 (Kloeker et al., 2004). Kloeker et al. (2004, Fig. 9A in the article) demonstrated that kindlin-1 siRNA transfection of HaCaT cells resulted in reduced expression of kindlin-1 (kindlerin) expression after 30 minutes spreading on fibronectin-coated coverslips. However, in this study and that of Kloeker et al. (2004, Fig. 9B), kindlin-1 was still expressed. The authors also demonstrated delayed cell spreading as a result of the transfection by measuring the cross-sectional area of the same cells (Kloeker et al., 2004, Fig. 9C in the article). In the present study kindlin-1 siRNA transfection of HaCaT cells resulted in a significant reduction in attachment (approximately 20%, p<0.0001) and spreading (approximately 40%; pO.0001) after 48 hours spreading (48 hours elapsed transfection time, Fig. 11). After 24 hours spreading, 72 hours elapsed transfection (Fig. 12) there was little difference in the reduction in attachment remained relatively constant while the effect of transfection on spreading decreased from approximately 40% to approximately 20%, in the presence of TGF-P (p<0.0001). The following results are in contrast to the finding by Kloeker et al. (2004) that 'adhesion was not affected by kindlerin (kindlin-1) knockdown when tested 30 min after replating on fibronectin or laminin (data not shown)'. The decrease in effect of kindlin-1 siRNA transfection from 48 to 72 hours elapsed transfection time suggests the cells are recovering from the transfection, generating new kindlin-1 mRNA, and thus new kindlin-1 protein.  90 5.3  Kindlin-1- and Integrin-mediated ECM-cell Signaling in HaCaT Cell Spreading on Type I Collagen, Laminin  10/11,  and Fibronectin  Alpha-beta heterodimeric integrins mediate dynamic adhesive cell-cell and cellextracellular matrix (ECM) interactions. A central feature of these receptors is their ability to change rapidly and reversibly their adhesive functions by modulating their ligand-binding affinity. In the case of an epithelial wound, only protein and glycoproteins of the E C M are known to be ligands of integrins (Graber et al., 1999). For the sake of simplicity, three of the most common integrin-ECM ligands found in an epithelial wound, type I collagen, laminin 10/11, and fibronectin were chosen for study. Epithelial integrins known to bind type I collagen include a2pi and a 3 p i , while those known to bind fibronectin (FN) include a 3 p i , a 5 p i , a V p i , aVp3, aVp5 (Graber et al., 1999). Integrins shown to bind laminin 10/11 include oc2pl, a 3 p i , a 6 p i , a 2 p i , and a6p4 (Kikkawa et al., 2000; Pouliot et al., 2000). Pouliot et al. (2002) has shown in in vitro cell adhesion assays, that laminin 10/11 is a potent adhesive substrate for keratinocytes mediated by a 3 p i and cc6p4 integrin. The authors also found that laminin 10/11 stimulated keratinocyte migration in an in vitro wound healing assay similar to the one employed in the present study. Integrins serve as the link between focal adhesion proteins and the E C M . In attempt to provide insight into which integrins mediate such interactions for ILK-1, Mig-2, migfililin, paxillin, and kindlin-1, kindlin-1 siRNA transfected HaCaT keratinocytes were spread for four hours on type I collagen-, laminin 10/11-, and fibronectin-coated glass coverslips. In comparison to overnight spreading (Fig. 10) whereby cells produced their own matrices, cells spread on the aforementioned matrices (Fig. 13, 14, and 15) did not express focal adhesions as well (as seen by smaller focal  91 adhesions by immunofluorescence). However, this is likely reflective of difference in duration of spreading, 16 hours versus four hours. Most notable differences in the expression of ILK-1 are seen on laminin 10/11 in which transfected cells are less spread and have fewer focal adhesions than in control. The disruption of the integrity of the basement membrane as a result of genetic defects in a3 and p i integrin and kindlin-1 suggests these proteins are important structures in the assembly of the basement membrane. Kloeker et al. (2004) found indirect evidence of a possible interaction between kindlin-1 and cytoplasmic tails of p i A and P3 integrin domains using recombinant structural mimics of these integrins. However, the P3 integrin is not expressed by keratinocytes during wound healing (Hakkinen et al., 2000; Santoro and Gaudino, 2005) nor is there evidence that it is expressed significantly in keratinocytes (Hynes, 2002). I L K was described in 1995 as a Ser/Thr kinase that binds to the cytoplasmic tails of p i , P2 and p3-integrin subunits (Hannigan et al., 1996). A decrease in ILK-1 expression and cell spreading from knockdown of kindlin-1 (Fig. 13) suggests kindlin-1 might interact with the p i integrin, since knocking down kindlin-1 might alter integrin-mediated ECM-cell signaling via ILK-1. Although there have been no reports of interactions between Mig-2 and integrins domains per se, functional studies using siRNA have led to the conclusion that Mig-2 recruits migflin to cell-ECM adhesions enabling migfilin to function as an anchor for filamin-containing actin filaments (Tu et al., 2003). The same group found that Mig-2 is able to interact with ILK-1 (Wu and Tu, personal communication) suggesting Mig-2 is functionally coupled to ILK-1 and thus integrins. UNC-112, the C. elegans homologue of Mig-2 has been shown to colocalize with P-integrin in muscle cell membrane in C. elegans (Rogalski et al., 2000). There was more  92 effective spreading and greater expression of Mig-2 in control compared to transfected cells spread on fibronectin (Fig. 13). Since the interaction between Mig-2 and integrins appears to be indirect and via functional coupling to ILK-1 (Tu et al., 2003), the effects of kindlin-1 siRNA transfection on integrin-mediated cell spreading is likely through interactions with ILK-1. Again, this lends further support to the possibility that kindlin-1 interacts with the J31 cytoplasmic domain. The effect of siRNA transfection on migfilin expression was such that transfected cells appeared less spread than controls and expression of migfilin was decreased in transfected cells spread on type I collagen and laminin 10/11 matrices (Fig. 14). Again, little is know about the interactions between migfilin and integrins, however, the present understanding is that migfilin localizes to cell-ECM adhesions via it's interaction with Mig-2 (Tu et al., 2003) and to cell-cell adhesion via it's interaction with P-catenin in adherens junctions (Gkretsi et al., 2004). Therefore, migfilin's interaction with integrins is likely indirect via Mig-2 and ultimately ILK-1. In the case of paxillin, there is less expression in transfected than control for all three matrices, type I collagen, laminin 10/11, and fibronectin (Fig. 14). A principal function for paxillin is in the integration and dissemination of signals from integrins and growth factor receptors to effect efficient cellular migration (Brown and Turner, 2003). Direct associations of paxillin with p i integrin, ot4- and a9-subunit cytoplasmic tails have been summarized (Brown and Turner, 2004). As well, paxillin interacts directly with the actin binding actopaxin family members as well as I L K (Brown and Turner, 2004). Nikolopoulos and Turner (2001) found that I L K binding to the paxillin LD1 motif regulates I L K localization to focal adhesions. As suggested in Figure 2 (Introduction) and in accordance with the  93 aforementioned summary of summary of the literature on ILK-1, Mig-2, migfilin, paxillin, and kindlin-1, kindlin-1 likely is assembled into focal adhesions down stream of it's potential binding partners. This might explain why kindlin-1 does not appear to significantly localize to focal adhesions when kindlin-1 siRNA transfected cells were spread for only 4 hours on laminin 10/11 and fibronectin (Fig. 17) in comparison to 16 hours HaCaT cell spreading in serum (Fig. 10). The known interactions between integrin cytoplasmic domains and the matrices utilized for cell spreading in the present study as well as the findings of Kloeker et al. (2004) implicates the p i integrin domain as being responsible for kindlin-1 mediated ECM-cell adhesions. Moreover, the finding that the effect of kindlin-1 knockdown was greatest on HaCaT cells spread on laminin 10/11 (up to 40%, Fig. 17) supports the histological findings of kindlin-1 localization to the B M Z (Fig. 3, 4, and 5). Future investigation involving colocalization studies between kindlin-1 and various integrins including pi-containing integrins are indicated.  5.4  Knockdown of Kindlin-1 decreases HaCaT Cell Spreading, Proliferation, and Migration  Kloeker et al. (2004) found that short-term adhesion was not affected by kindlin-1 (kindlerin) knockdown after 30 minutes on fibronectin or laminin. They suggested it was unlikely kindlin-1 is required for 'integrin-ligand interaction.' They also concluded based on the lack of evidence that kindlin-1 does not lead to integrin activation, that kindlin-1 is involved in cell spreading at some point time 'after integrin-ligand binding.' The present study has shown that kindlin-1 is important in HaCaT cell attachment (Fig. 11 and 12),  94 however, attachment was assessed at 24 and 48 hours spreading and cells were spread in flasks, which means they created their own E C M . In light of the ultrastructural finding of finding of focal apoptosis in the epithelial basal layer (Lanschuetzer et al., 2003) combined it is possible that the observed decreased in HaCaT cell attachment from kindlin-1 knockdown is due to kindlin-1 siRNA induced apopotosis of HaCaT cells. In order to futher test Kloeker et al's (2004) hypothesis of kindlin-l's 'after integrin-ligand binding', kindlin-1 transfected HaCaT cells were spread on type I collagen, laminin 10/11, and fibronectin for 45 minutes to four hours and the effect of transfection measured (Figures 16, 17, 18, and 19). Cells were still able to attach at 45 minutes confirming the findings of Kloeker et al. (2004). However, knockdown of kindlin-1 resulted in a significant decrease in cell spreading on the difference matrices. The greatest effects on cell spreading were at one hour on type I collagen (Fig. 16), and 45 minutes for both laminin 10/11 (Fig. 17) and fibronectin (Fig. 20). The differences were still significant at four hours for type I collagen and laminin 10/11. The present results agree with the findings of others who demonstrated that a 'reduction of kindlerin (kindlin-1) by siRNA resulted in delayed cell spreading' on fibronectin (Kloeker et al., 2004). These findings suggest kindlin-1, like it's homologue, Mig-2, and its potential binding partner, migfilin, acts downstream of the integrin in mediating interactions between the actin cytoskeleton and the E C M . Kindlin-1 might be recruited by paxillin to sites of ECM-cell adhesion and like migfilin, serve as an anchor for actin filamanents of the cytoskeleton. Alternatively, like it's homologue, Mig-2, kindlin-1 might recruit migfilin to focal adhesions. In either case, paxillin is much like a conductor in an orchestra consisting of numerous focal adhesion proteins. ILK-1 on the other hand seems  95 more like a facilitator, enabling interactions between focal adhesion proteins and paxillin, with focal adhesion proteins being linked to the actin cytoskeleton. A rendition of the proposed interaction between Paxillin, ILK-1, Mig-2, migfilin, and kindlin-1 is portrayed in Figure 2 of the introduction. A key role for the adhesion molecules seems to be to organize membraneproximal cytoskeletal structures that then serve as scaffolds for signaling cascades. The precise compositions of the cytoskeletal scaffolds organized by various adhesion receptors will no doubt differ. This would then provide considerable biochemical and biological diversity in terms of the qualitative and quantitative impacts on signaling cascades. The challenge is to be able to dissect these detailed molecular interconnections. The exact composition of a given focal adhesion will ultimately control cellular behaviors such as adhesion, migration, proliferation, and differentiation (Wozniak et al., 2004). One group found that depletion of ILK-1 significantly impaired cell spreading and migration (Vouret-Craviari V et al., 2004), while another has shown decreased cell spreading after depletion of kindlerin (Kloeker et al., 2004). In the present study, siRNA-mediated gene silencing of kindlin-1 resulted in a significant decrease in HaCaT cell proliferation at 24, 48, and 72 hours after kindlin-1 siRNA transfection (Fig. 20). The sudden decline in cell proliferation seen after 48 hours is likely explained by the fact that cells became confluent at approximately 54 hours. The greatest difference (approximately 1.6 fold) in proliferation between kindlin-1 siRNA transfected and control cells occurred at 48 hours. The significant difference seen at 24 hours (approximately 1.2 fold) suggests the effects of kindlin-1 siRNA transfection begin to take effect at this time point. As mentioned earlier, it is possible that loss of kindlin-1 function results in apoptosis (Lanschuetzer et  96 al., 2003), which, in part, might explain the significant difference in viable HaCaT cells (Fig. 20). Kindlin-1 siRNA transfection also significantly inhibited HaCaT cell migration following scratch-wounding of confluent cell layers by a difference of greater than twofold (Fig. 21). The use of the HaCaT cells in culture mimicks the study of a cell in a wound as cells grown in culture have no contact with neighboring cells as found in intact skin and oral mucosa. In the present study kindlin-1 was found to specifically immunolocalize to the B M Z of the migrating wound tip of 3 day-old wounds, which was continuous with the adjacent B M Z of non-wounded epithelium (Fig. 5). Damage to the epithelium activates basal cells; after 24 to 48 hours, the cells enter into 'horizontal migratory activity in conjunction with mitotic cell division (Grinell et al., 1990; Graber et al., 1999). Periodontal epithelium plays a critical role in the protection, destruction and repair of human periodontium. For optimal repair to occur, epithelium migrates and covers the wound surface to prevent infection and protect the underlying connective tissue (Larjava et al., 1996). Periodontal disease then mimics a state of chronic wounding. Periodontal diseases affect over half the adults in the U.S. (Tanner et al., 2005). In a homogeneous group of subjects in Bocas del Toro, Panama, patients with Kindler syndrome (KS) have periodontitis with an early onset (teenage years) and with its rate of progression rapid, resembling aggressive periodontitis, compared to non-Kindler individuals of the same geographic and ethnic origin (Wiebe et al., 2003). Results of the present study suggest that a deleterious mutation in kindlin-1 renders a subject susceptible to periodontal disease and skin fragility and photosensivity by weakening of the adhesion between basal epithelial cells and the underlying basement membrane. As  97 described by Larjava et al. (1996), 'in the course of periodontal disease the epithelial attachment to the tooth surface is lost and the epithelium proliferates and extends pseudorete ridges deep into the inflamed connective tissue. Both scenarios, repair and destruction, involve active epithelial migration either in the wound provisional matrix or in the inflamed connective tissue matrix, respectively.' However, the main clinical presentation of Kindler syndrome, skin fragility and photosensitivity and aggressive periodontitis does not suggest that loss of function of kindlin-1 leads to a significant compromise in wound healing per se. Rather, the findings of the present study combined with findings of altered organization of basement membrane components including extensive reduplication of the lamina densa, focal widening of the lamina lucida, and focal apoptosis  of basal keratinocytes (Lanschuetzer et al., 2003) as well as  discontinuities in laminin 1 and 5 and abnormal deposition of type VII collagen (Wiebe and Larjava, 1999) in Kindler suggest kindlin-1 is necessary for proper organization of components of the basement membrane in skin and oral mucosa. A summary of the ultrastructural findings of Kindler syndrome is depicted in Fig. 22. Kindlin-1 is involved in the focal adhesions responsible for linking the keratinocyte to its E C M , that of the basement membrane, such that loss of function of kindlin-1 results in altered ECM-cell signaling, which in turn is crucial for cell adhesion, spreading, migration, and proliferation.  98 5.5  Limitations of the Study  Some of the limitations of the present study include the use of a kindlin-1 monoclonal antibody, which seemed to produce variable immunostaining results. Ideally, further investigation in to the creation of a more specific antibody for kindlin-1 will be necessary to corroborate the findings of this study. As well, the conclusions of the present study are based primarily on the use of the HaCaT cell, an immortalized human keratinocyte cell line. The genotype of such a cell line might not accurately reflect that of normal human basal keratinocytes. Finally, there is a need for confirmation of the in vitro data presented in this study by the creation of a kindlin-1 knockout in an animal in conjunction with the study of wound healing in this animal model.  99  CHAPTER VI Conclusions 1. Paxillin, migfilin, and kindlin-1 localized to the basement membrane zone (BMZ) of wounded and non-wounded normal human oral epithelium suggesting a molecular interaction exists between these focal adhesion proteins.  2. Paxillin, migfilin, and kindlin-1 localized to the internal basal lamina (IBL) zone of junctional epithelium (JE) suggesting a possible role for these focal adhesion proteins in attachment of the JE to a tooth.  3. siRNA mediated knockdown of kindlin-1 did not affect early (45 minutes to 4 hours) HaCaT cell attachment, but did decrease HaCaT cell spreading and the expression of ILK-1, Mig-2, migfilin, paxillin, and kindlin-1 on type I collagen, laminin 10/11, and fibronectin. This confirms the findings of Kloeker et al., (2004).  4. siRNA  mediated knockdown of kindlin-1 decreased  HaCaT  spreading, proliferation, and migration in the presence of serum.  keratinocyte  100  CHAPTER VII Future Directions 1. The findings of the present study combined with findings of Lanschuetzer et al. (2003) and Wiebe and Larjava (1999) suggest kindlin-1 is necessary for proper organization of components of the basement membrane in skin and oral mucosa.  2. Given the shared immunolocalization to the B M Z and IBL zone of paxillin, migfilin, and kindlin-1, combined with the observation termination  of actin filament  into sites of migfilin and kindlin-1 focal adhesions,  further  investigation into a possible interaction between these two F A proteins is necessary.  3. Kindlin-1 is not likely involved with initial cell attachment mediated by integrins, but is likely involved in ECM-cell signaling downstream from the pi-integrin cytoplasmic tail. Further investigation examining a possible interaction between kindlin-1 and it's spatial organization with the pi-integrin is warranted.  4. In light of the observed decrease of HaCaT cell migration and proliferation as a result of kindlin-1 knockdown in vitro, corroboration of such findings in vivo using a kindlin-1 knockout animal model might shed further light into kindlin-1 's role in wound healing.  5. Considering the ultrastructural finding in Kindler epidermis of possible focal apoptosis (Lanschuetzer et al., 2003) along with the present observation of decreased HaCaT cell proliferation from kindlin-1 knockdown, it might prove useful to assay whether kindlin-1 knockdown induces apoptosis in keratinocytes.  101 BIBLIOGRAPHY 1.  Akazawa H , Kudph S, Mochizuki N , Takekoshi N , Takano H , Nagai T et al. (2004). A novel L I M protein Cal promotes cardiac differentiation by association with CSX/NKX2-5. J Cell Biol 164: 395-405.  2.  Altschul SF, Gish W, Miller W, Meyers EW, Lipman DJ (1990). Basic local alignment search tool. JMol Biol 215(3): 403-410.  3.  Ashton GHS (2004a). Kindler Syndrome. Clin Exp Derm 29: 116-121.  4.  Ashton GHS, McLean WHI, South AP, Oyama N , Smith FJD, Al-Suwaid R et al. (2004b). Recurrent mutations in kindlin-1, a novel keratinocyte focal contact protein, in the autosomal recessive skin fragility and photosensitivity disorder, Kindler syndrome. J Invest Dermatol 122: 78-83.  5.  Bach I (2000). The L I M domain: regulation by association. Mech Dev 91: 5-17.  6.  Bernstein EF, Harisiadis L, Salomon G, Norton J, Sollberg S, Uitto J, Glatstein E, Glass J, Talbot, Russo A (1991). Transforming growth factor-beta improves healing of radiation-impaired wounds. J Invest Dermatol 97: 430-434.  7.  Block ER, Matela A R , SundarRaj N , Iszkula ER, Klarlund J K (2004). Wounding induces motility in sheets of corneal epithelial cells through loss of spatial constraints: Role of heparin-binding epidermal growth factor-like growth factor signaling. J Biol Chem 279: 24307-24312.  8.  Bosshardt D D , Lang N P (2005). The junctional epithelium: from health to disease. JDent Res 84(1): 9-20.  9.  Boukamp P, Petrussevska RT, Breitkreutz D, Hornung J, Markham A , Fusenig N E (1988). Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte line. J Cell Biol 106: 761-771.  10.  Brown M C , Turner C E (2004). Paxillin: Adapting to change. Physiol Rev 1315-1339.  11.  Burridge K , Chrzanowska-Wodnicka M (1996). Focal adhesions, contractility, and signaling. Annu Rev Cell Dev Biol 12: 463-518.  12.  Calderwood D A , Shattil SJ, Ginsberg M H (2000). Integrins and actin filaments: reciprocal regulation of cell adhesion and signaling. J Biol Chem 215: 2260722610.  84:  102 13.  Calderwood D A , Yan B , de Pereda J M , Alvarez B G , Fujioka Y , Liddington RC et al. (2002). The phosphotyrosine binding-like domain of talin activates integrins. J Biol Chem 277: 21749-21758.  14.  Calderwood D A , Zent R, Grant R, Rees DJ, Hynes RO, Ginsberg M H (1999). The Talin head domain binds to integrin beta subunit cytoplasmic tails and regulates integrin activation. J Biol Chem 21 A: 28071-28074.  15.  Chishti A H , K i m A C , Marfatia S M , Lutchman M , Hanspal M , Jindal H et al. (1998). The F E R M domain: a unique module involved in the linkage of cytoplasmic proteins to the membrane. Trends Biochem Sci 23: 281-282.  16.  Chrzanowska-Wodnicka M , Burridge K (1996). Rho-stimulated contractility drives the formation of stress fibers and focal adhesions. J Cell Biol 133: 1403-1415.  17.  Critchley D R (2000). Focal adhesions-the cytoskeletal connection. Curr Opin Cell Biol 12: 133-139.  18.  Danielsen A J , Maihle N J (2002). The EGF/ErbB receptor family and apoptosis. Growth Factors 20: 1-15.  19.  Dedhar S (2002). Cell-substrate interactions and signaling through ILK. Curr Opin Cell Biol 12:250-256.  20.  Elbashir S M , Harborth J, Weber K , Tuschl T (2002). Analysis of gene function in somatic mammalian cells using small interfering RNAs. Methods 26: 199-213.  21.  Ellis PD, Hadfield K M , Pascall JC, Brown K D (2001). Heparin-binding epidermalgrowth-factor-like growth factor gene expression is induced by scrape-wounding epithelial cell monolayers: involvement of mitogen-activated protein kinase cascades. Biochem J354: 99-106.  22.  Ferletta M , Ekblom P (1999). Identification of laminin-10/11 as a strong cell adhesive complex for a normal and a malignant human epithelial cell line. J Cell Sci 112: 1-10.  23.  Ffrench-Constant, C (1995). Alternative splicing of fibronectin—many different proteins but few different functions. Exp Cell Res 221: 261-271.  24.  Flanders K C (2004). Smad3 as a mediator of the fibrotic response. Int J Exp Pathol 85: 47-64.  25.  Forman A B , Prendiville JS, Esterly N B , Hebert A A , Duvic M , Horiguchi Y et al. (1989). Kindler syndrome: report of two cases and review of the literature. Pediatr Dermatol 6: 91-101.  103 26.  Fuchs E (1990). Epidermal differentiation: the base essentials. J Cell Biol 111: 2807-2814.  27.  Fujimoto N , Yeh S, Kang H Y , Inui S, Chang H C , Mizokami A et al. (1999). Cloning and characterization of androgen receptor coactivator, ARA55, in human prostate. J Biol Chem 21 A: 8316-8321.  28.  Gaidano G, Bergui L , Schena M , Gaboli M , Cremona O, Marchisio PC et al. (1990). Integrin distribution and cytoskeleton organization in normal and malignant monocytes. Leukemia 4: 682-7.  29.  Geiger B, Bershadsky A , Pankov R, Yamada K M (2001). Transmembrane crosstalk between the extracellular matrix-cytoskeleton crosstalk. Nat Rev Mol Cell Biol 2: 793-805.  30.  Giancotti FG, Ruoslahti E (1999). Integrin signaling. Science 285: 1028-1032.  31.  Gkretsi V , Zhang Y , Tu Y , Chen K , Stolz DB, Yang Y et al. (2004). Physical and functional association of migfilin with cell-cell adhesions. J Cell Sci 118: 697-710.  32.  Glenney JR Jr, Zokas L (1989). Novel tyrosine kinase substrates from Rous sarcoma virus-transformed cells are present in the membrane skeleton. J Cell Biol 108: 2401-2408.  33.  Goldfinger L E , Hopkinson SB, deHart GW, Collawn S, Couchman JR, Jones JCR (1999). The a3 laminin subunit, a6b4 and a3bl integrin coordinately regulate wound healing in cultured epithelial cells in the skin. J Cell Sci 112: 2615-2629.  34.  Graber H - G , Conrads G, Wilharm J, Lampert F (1999). Role of interactions between integrins and extracellular matrix components in healthy epithelial tissue and establishment of a long junctional epithelium during periodontal wound healing: a review. JPerio 70: 1511-1522.  35.  Grinell F (1990). The activated keratinocyte: upregulation of cell adhesion and migration during wound healing. J Trauma 30 (Suppl. 12): S144-S149.  36.  Grinnell F (1992). Wound repair, keratinocyte activation and integrin modulation. J Cell Sci 101: 1-5.  37.  Hagel M , George E L , K i m A , Tamimi R, Opitz SL, Turner C E et al. (2002). The adaptor protein paxillin is essential for normal development in the mouse and is a critical transducer of fibronectin signaling. Mol Cell Biol 22: 901-915.  38.  Hagmann J, Grob M , Welman A , van Willigen G, Burger M M (1998). Recruitment of the L I M protein hic-5 to focal contacts of human platelets. J Cell Sci 111: 21812188.  104 39.  Hakkinen L , Koivisto L , Larjava H (2002). A n improved method for culture of epidermal keratinoctyes from newborn mouse skin. Methods Cell Sci 23: 189-196.  40.  Hannigan GE, Leung-Hagesteijn C, Fitz-Gibbon L, Coppolino M G , Radeva G et al. (1996). Regulation of cell adhesion and anchorage-dependent growth by a new betal-integrin-linked protein kinase. Nature 379: 91-96.  41.  Hannon GJ (2002). R N A interference. Nature 418: 244-251.  42.  Hintner H , Stingl G, Schuler G, Fritsch P, Stanley J, Katz S et al. (1981). Immunofluorescencemapping of antigenic determinants within the dermalepidermal junction in the mechanobullous diseases. J Invest Dermatol 76: 113.  43.  Hormia M , Owaribe K , Virtanen I (2001). The dento-epithelial junction: cell adhesion by type I hemidesmosomes in the absence of a true basal lamina. J Perio 72: 788-797.  44.  Hotchin N A , Hall A (1995). The assembly of integrin adhesion complexes requires both extracellular matrix and intracellular rho/rac GTPases. J Cell Biol 131: 18571865.  45.  Hovnanian A , Blanchet-Bardon C, de Prost Y (1989). Poikiloderma of Theresa Kindler: report of a case with ultra-structural study, and review of the literature. Pediatr Dermatol 6:82-90.  46.  Hudson DF, Morrison C, Ruchaud S, Earnshaw W C (2002). Reverse genetics of essential genes in tissue-culture cells: 'dead cells talking'. Trends Cell Biol 12: 281287.  47.  Hudson L G , McCawley L J (1998). Contributions of the epidermal growth factor receptor to keratinocyte motility. Microsc Res Tech 43: 444-455.  48.  Huttenlocher A , Sandborg R, Horwitz A F (1995). Adhesion in cell migration. Curr Opin Cell Biol 7: 697-706.  49.  Hynes RO (2002). Integrins: bi-directional, allosteric signaling machines. Cell 110: 673-687.  50.  Hynes, RO (1990). Fibronectins. Springer-Verlag New York Inc., New York. 544 pp.  51.  Itoh T, Takenawa T (2002). Phosphoinositide-binding domains: functional units for temporal and spatial regulation of intracellular signaling. Cell Signal 14: 733-743.  52.  Jamora C, Fuchs E (2002). Intercellular adhesion, signalling and the cytoskeleton. Nat Cell Biol A: 101-108.  105  53.  Jobard F, Bouadjar B , Caux F, Hadj-Rabia S, Has C, Matsuda F et al. (2003). Identification of mutations in a new gene encoding a F E R M family protein with a pleckstrin homology domain in Kindler syndrome. Hum Mol Genet 12: 925-935.  54.  Jones P H , Watt F M . Separation of human epidermal stem cells from transit amplifying cells on the basis of differences in integrin function and expression. Cell 23: 713-724.  55.  Kikkawa Y , Sanzen N , Sekiguchi K (1998). Isolation and characterization of laminin-10/11 secreted by human lung carcinoma cells: laminin-10/11 mediates cell adhesion through integrin alpha-3 -beta-1. J Biol Chem 273: 15854-15859.  56.  Kikkawa Y , Sanzen N , Fujiwara H , Sonnenberg A , Sekiguchi K (2000). Integrin binding specificity of laminin-10/11: laminin 10/11 are recognized by a3(31, a 6 p i and a6pi integrins. J Cell Sci 113: 869-876.  57.  K i m IY, K i m M M , K i m S-J (2005). Transforming growth factor-P: biology and clinical relevance. J Biochem Mol Biol 38(1): 1-8.  58.  Kindler T (1954). Congenital poikiloderma with traumatic bulla formation and progressive cutaneous atrophy. Br J Dermatol 66: 104-111.  59.  Kleinman H K , Klebe RJ, Martin G R (1981). Role of collagenous matrices in the adhesion and growth of cells. J Cell Biol 88: 473-485  60.  Kloeker S, Major M B , Calderwood D A , Ginsberg M H , Jones D A , Beckerle M C (2004). The kindler syndrome protein is regulated by transforming growth factor-B and involved in integrin-mediated adhesion. J Biol Chem 279: 6824-6833.  61.  Koivisto L , Alavian K , Hakkinen L , Pelech S, McCulloch C A , Larjava H et al. (2003). Glycogen synthase kinase-3 regulates formation of long lamellipodia in human keratinocytes. J Cell Sci 116: 3749-3760.  62.  Koivisto L , Jiang G, Hakkinen L et al. (2005). HaCaT keratinocyte migration is dependent on epidermal growth factor receptor signaling and glycogen synthase kinase-3a. Personal Communication/Manuscript  63.  Lanschuetzer C M , Muss W H , Emberger M , Pohla-Gubo G, Klausegger A , Bauer JW et al. (2003). Characteristic immunohistochemical and ultrastructural findings indicate that Kindler's syndrome is an apoptotic skin disorder. J Cutan Pathol 30: 553-560.  64.  Larjava H , Peltonen J, Akiyama SK, Yamada SS, Gralnick HR, Uitto J et al. (1990). Novel function of p i integrins in keratinocyte cell-cell interactions. J Cell Biol 110: 803-815.  106  65.  Larjava H , Salo T, Haapasalmi K , Kramer R H , Heino J (1993). Expression of integrins and basement membrane components by wound keratinocytes. J Clin Invest 92: 1425-1435.  66.  Larjava H , Haapasalmi K , Salo T, Wiebe C, Uitto V J (1996). Keratinocyte integrins in wound healing and chronic inflammation of the human periodontium. Oral Dis 2(1): 77-86.  67.  Laukaitis C M , DJ Webb, Donais K , Horwitz A F (2000). Differential dynamics of {alpha}5 integrin, paxillin, and {alpha}-actinin. J Cell Biol 150: 627-642.  68.  Leung-Hagesteijn C, Mahendra A , Naruszewicz I, Hannigan G E (2001). Modulation of integrin signaling transduction by I L K A P , a protein phosphatase 2C assocating with the integrin-linked kinase, ILK1. EMBO J 20: 2160-2170.  69.  L i F, Zhang Y , Wu C (1999). Integrin-linked kinase is localized to cell-matrix focal adhesions but not cell-cell adhesion sites and the focal adhesion localization of integrin-linked kinase is regulated by PINCH binding A N K repeats. J Cell Sci 112: 4589-4599.  70.  L i S, Harrison D, Carbonetto S, Fassler R, Smyth N , Edgar D et al. (2002). Matrix assembly, regulation and survival functions of laminin and its receptors in embryonic stem cell differentiation. J Cell Biol 157: 1279-1290.  71.  Lipsky BP, Beals CR, Staunton D E (1998). Leupaxin is a novel L I M domain protein that forms a complex with P Y K 2 . J Biol Chem 273: 11709-11713.  72.  Listgarten M A (1972a). Ultrastructure of the dento-gingival junction after gingivectomy. J Periodontal Res 7: 151-160.  73.  Liu S, Calderwood D A , Ginsberg M H (2000). Integrin cytoplasmic domain-binding proteins. J Cell Sci 113: 3563-3571.  74.  Mackinnon A C , Qadota H , Norman K R , Moerman D G , Williams B D (2002). C. elegans PAT-4/ILK functions as an adaptor protein within integrin adhesion complexes. Curr Biol 12:787-797.  75.  Maffuci T, Falasca M (2001). Specificity in pleckstrin homology (PH) domain membrane targeting: a role for a phosphoinositide-protein co-operative mechanism. FEBS Letters 506: 173-179.  76.  Magklara A , Brown TJ, Diamandis EP (2002). Characterization of androgen receptor and nuclear receptor co-regulator expression in human breast cancer cell lines exhibiting differential regulation kallikreins 2 and 3. Int J Cancer 100: 507514.  107 77.  Marikovsky M , Breuing K , Liu P Y , Eriksson E, Higashiyama S, Farber P et al. (1993). Appearance of heparin-binding EGF-like growth factor in wound fluid as a response to injury. Proc Natl Acad Sci USA 90: 3889-3893.  78.  McCarthy DW, Downing M T , Brigstock DR, Luquette M H , Brown K D , Abad M S et al. (1996). Production of heparin-binding epidermal growth factor-like growth factor (HB-EGF) at sites of thermal injury in pediatric patients. J Invest Dermatol 106: 49-56.  79.  Miyamoto S, Akiyama SK, Yamada K M (1995). Synergistic roles for receptor occupancy and aggregation in integrin transmembrane function. Science 267: 883885.  80.  Nei M (1987). Molecular evolutionary genetics. Columbia University Press, New York.  81.  Nguyen BP, Ryan M C , G i l SG, Carter W G (2000). Deposition of laminin 5 in epidermal wounds regulates integrin signaling and adhesion. Cell Adhes Commun 12: 554-562.  82.  Nikolopoulos SN, Turner C E (2001). Integrin-linked kinase (ILK) binding to paxillin LD1 motif regulates ILK localization to focal adhesions. J Biol Chem 276: 23499-23505.  83.  Oksala O, Salo T, Tammi R, Ha'kkinen L , Jalkanen M , Inki P, Larjava H (1995). Expression of proteoglycans and hyaluronan during wound healing. J Histochem Cytochem 43: 125-135.  84.  Patrizi A , Pauluzzi P, Neri I, Trevisan G, De Giorgi L B , Pasquinelli G (1996). Kindler syndrome: report of a case with ultrastructural study and review of the literature. PedDermatol 13: 397^102.  85.  Pearson M A , Reczek D, Bretscher A , Karplus P A (2000). Structure of the E R M protein moesin reveals the F E R M domain fold masked by an extended actin binding tail domain. Cell 101: 259-270.  86.  Pouliot N , Saunders N A , Kaur P (2002). Laminin 10/11: an alternative adhesive ligand for epidermal keratinocytes with a functional role in promoting proliferation and migration. Exp Dermatol 11(5): 387-97.  87.  Pouliot N , Connolly L M , Moritz R L , Simpson RJ, Burgess A W (2000). Colon cancer cells adhesion and spreading on autocrine laminin-10 is mediated by multiple integrin receptors and modulated by EGF receptor stimulation. Exp Cell Res 261: 360-371.  88.  Pulkinnen L , Uitto J (1999). Mutation analysis and molecular genetics of epidermolysis bullosa. Matrix Biol 18(1): 29-42.  108  89.  Reuveny S, Mizrahi A , Shahar A , Kotler M (1984). Role of substrata in determining the growth topology of transformed and nontransformed cells in culture. Cell Biol Int Rep 8(7): 539-49.  90.  Rogalski T M , Mullen GP, Gilbert M M , Williams BD, Moerman D G (2000). The UNC-112 gene in Caenorhabditis elegans encodes a novel component of cell-matrix adhesion structures required for integrin localization in the muscle cell membrane. J Cell Biol 150: 253-264.  91.  Rottner K, Hall A , Small JV (1999). Interplay between rac and rho in the control of substrate contact dynamics. Curr Biol 9: 640-648.  92.  Rottner K , Hall A , Small JV (1999). Interplay between the Rac and Rho in the control of substrate contact dynamics. Curr Biol 9: 64-648.  93.  Sakai T, L i S, Docheva D, Grashoff C, Sakai K , Kostka G et al. (2003). Integrinlinked kinase (ILK) is required for polarizing the epiblast, cell adhesion and controlling actin accumulation. Genes Dev 17: 926-940.  94.  Salonen JI, Kautsky M B , Dale B A (1989). Changes in cell phenotype during regeneration of junctional epithelium of human gingiva in vitro. J Periodontal Res 24:370-377.  95.  Santoro M M , Gaudino G (2005). Cellular and molecular facets of keratinocyte reepithelization during wound healing. Exp Cell Res 304(1): 274-86.  96.  Sastry SK, Burridge K (2000). Focal adhesions: a nexus for intracellular signaling and cytoskeletal dynamics. Exp Cell Res 261: 25-36.  97.  Schroeder H E (1996). The junctional epithelium: origin, structure, and significance. A Review. Acta Med Dent Helv 1: 155-167.  98.  Schroeder H E , Listgarten M (1997). The gingival tissues: the architecture of periodontal protection. Periodontal 2000 13: 91-120.  99.  Sethuraman G, Fassihi H , Ashton G H , Bansal A , Kabra M , Sharma V K et al. (2005). A n Indian child with Kindler syndrome resulting from a new homozygous nonsense mutation (C468X) in the KIND1 gene. Clin Exp Dermatol 30: 286-288.  100. Shimizu H , Sato M , Ban M , Kitajima Y , Ishizaki S, Harada T et al. (1997). Immunohistochemical, ultrastructural, and molecular features of Kindler syndrome distinguish it from dystrophic epidermolysis bullosa. Arch Dermatol 133: 11111117.  109 101. Siegel D H , Ashton G H , Penagos H G , Lee JV, Feiler HS, Wilhelmsen K C et al. (2003). Loss of Kindlin-1, a human homolog of the Caenorhabditis elegans actinextracellular matrix linker protein UNC-112, causes Kindler syndrome. Am J Hum Genet73: 174-187. 102. Stossel TP, Condeelis J, Cooley L, Hartwig JH, Noegel A , Schleicher M et al. (2001). Filamins as integrators of cell mechanics and signaling. Nat Rev Mol Cell Biol 2: 138-145. 103. Takafuta T, Saeki M , Fujimoto TT, Fujimura K , Shapiro SS et al. (2003). A new member of the L I M protein family binds to filamin B and localizes at stress fibers. J BiolChem21%: 12175-12181. 104. Takeuchi K , Kawashima A , Nagafuchi A , Tsukita S (1994). Structural diversity of band 4.1 superfamily members. J Cell Sci 107: 1921-1928. 105. Tanner A C , Kent R Jr., Van Dyke T, Sonis ST, Murray L A (2005). Clinical and other risk indicators for early periodontitis in adults. JPerio 76(4): 573-581. 106. Tu Y , L i F, Goicoechea S, Wu C (1999). The LIM-only protein PINCH directly interacts with integrin-linked kinase and is recruited to integrin-rich sites in spreading cells. Mol Cell Biol 19: 2425-2434. 107. Tu Y , L i F, Huang Y , Zhang Z, Hua Y , Wu C (2001). A new focal adhesion protein that interacts with integrin-linked kinase and regulates cell adhesion and spreading. J Cell Biol 153: 585-598. 108. Tu Y , Wu S, Shi X , Chen K , Wu C (2003). Migfilin and Mig-2 link focal adhesions to filamin and the actin cytoskeleton and function in cell shape modulation. Cell 13: 37-47. 109. Turner C E (2000). Paxillin and focal adhesion signaling. Nat Cell Biol 2: E231E236. 110. Turner C E , Glenney JR Jr, Burridge K (1990). Paxillin: a new vinculin-binding protein present in focal adhesions. J Cell Biol 111:1059-1068. 111. Turner CE, Kramarcy N , Sealock R, Burridge K (1991). Localization of paxillin, a focal adhesion protein, to smooth muscle dense plaques, and the myotendinous and neuromuscular junctions of skeletal muscle. Exp Cell Res 192: 651-655. 112. Varani J, Dame M , Rediske J, Beals TF, Hillegas W (1985). Substrate-dependent differences in growth and biological properties of fibroblasts and epithelial cells grown in microcarrier culture. Biol Stand 13(1): 67-76.  110 113. Vespa A , Darmon A J , Turner CE, D'Souza SJ, Dagnino L (2003). Ca+2 -dependant localization of integrin-linked kinase to cell junctions in differentiating keratinocytes. J Biol Chem 13: 11528-11535. 114. Vouret-Craviari V , Boulter E, Grail D, Matthews C, Van Obberghen-Schilling E (2004). ILK is required for the assembly of matrix-forming adhesions and capillary morphogenesis in endothelial cells. J Cell Sci 117: 4559-4569. 115. Weary PE, Manley WF, Graham GF (1971). Heriditary acrokeratotic poikiloderma. Arch Derm 103: 409-422. 116. Weinstein EJ, Bourner M , Head R, Zakeri H , Bauer C, Mazzarella R et al. (2003). URP1: a member of a novel family of P H and F E R M domain-containing membrane-associated proteins is significantly over-expressed in lung and colon carcinomas. Biochim Biophys Acta 1637: 207-216. 117. White SJ, McLean WHI (2005). Kindler surprise: mutation in a novel actinassociated protein cause Kindler syndrome. J Derm Sci 38: 169-175. 118. Wick M , Burger C, Brusselbach S, Lucibello FC, Muller R (1994). Identification of serum-inducible genes: different patterns of gene regulation during GO S and GI S progression. J Cell Sci 107 Pt 1: 227-239 (erratum 107 [pt 3]: precedi). 119. Wiebe C B , Penagos H , Luong N , Slots J, Epstein E Jr et al. (2003). Clinical and microbiologic study of periodontitis associated with Kindler syndrome. J Perio 74: 25-31. 120. Wiebe, C B , Larjava, HS (1999). Abnormal deposition of type VII collagen in Kindler syndrome. Arch Dermatol Res 291: 6-13. 121. Wienke D, MacFadyen JR, Isacke C M (2003). Identification and characterization of the endocytic transmembrane glycoprotein Endo 180 as a novel collagen receptor. Mol Biol Cell 14: 3592-3604. 122. Woods A , Oh ES, Couchman JR (1998). Syndecan proteoglycans and cell adhesion. Matrix Biol 17: 461-463. 123. Wozniak M A , Modzelewska K , Kwong L , Keely PJ (2004). Focal adhesion regulation of cell behaviour. Bioch et Biophys Acta 1692: 103-119. 124. Wu C (2001). ILK interactions. J Cell Sci 114: 2549-2550. 125. Wu C (2005). Migfilin and its binding partners: from cell biology to human diseases. J Cell Sci 118: 659-664.  Ill 126. Wu C, Dedhar S (2001). Integrin-linked kinase (ILK) and its interactors: a new paradigm for the coupling of extracellular matrix to actin cytoskeleton and signaling complexes. J Cell Biol 155: 505-510. 127. Wu C, Keightley SY, Leung-Hagesteijn C, Radeva G, Coppolino M , Goicoechea S et al. (1998). Integrin-linked protein kinase regulates fibronectin matrix assembly, E-cadherin expression, and tumorigenicity. J Biol Chem 273: 528-536. 128. Yasukawa K , Sato-Matsumura K C , McMillan J, Tsuchiya K , Shimizu H (2002). Exclusion of COL7A1 mutation in Kindler syndrome. J Am Acad Dermatol 46: 447-450. 129. Yuan W and Varga J (2001). Transforming growth factor repression of matrix metalloproteinase-1 in dermal fibroblasts involves Smad3. J Biol Chem 276: 3850238510. 130. Yuminamochi T, Yutaka Yatomi Y , Osada M , Ohmori T, Ishii Y , Nakazawa K , Hosogaya S, Ozaki Y (2003). Expression of the L I M Proteins Paxillin and Hic-5 in Human Tissues. J Histochem and Cytochem 51: 513-522. 131. Zaidel-Bar R, Ballestrem C, Kam Z, Geiger B (2003). Early molecular events in the assembly of matrix adhesions at the leading edge of migrating cells. J Cell Sci 116: 4605-4613. 132. Zaidel-Bar R, Cohen M , Addadi L, Geiger B (2004). Hiearchal assembly of cellmatrix adhesion complexes. Biochem Soc Transae 32: 416-420. 133. Zamir E, Geiger B (2001a). Components of cell-matrix adhesions. J Cell Sci 114: 3577-3579. 134. Zamir E, Geiger B (2001b). Molecular complexity and dynamics of cell-matrix adhesions. J Cell Sci 114: 3583-3590. 135. Zamir E, Katz B Z , Aota S, Yamada K M , Geiger B, Kam Z (1999). Molecular diversity of cell-matrix adhesions. J Cell Sci 112: 1655-1669. 136. Zhang J, Zhang L X , Meltzer PS, Barrett JC, Trent J M (2000). Molecular cloning of human hic-5, a potential regulator involved in signal transduction and cellular senescence. Mol Carcinog 27: 177-183. 137. Zipkin ID, Kindt R M , Kenyon CJ (1997). Role of a new Rho family member in cell migration and axon guidance in C. elegans. Cell 90: 883-894.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0092142/manifest

Comment

Related Items