<%BANNER%>

Interactions of Porphyromonas gingivalis with Human Endothelial Cells

Permanent Link: http://ufdc.ufl.edu/UFE0018840/00001

Material Information

Title: Interactions of Porphyromonas gingivalis with Human Endothelial Cells
Physical Description: 1 online resource (212 p.)
Language: english
Creator: Totten, Kristen L
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: Molecular Cell Biology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Over the past several years, evidence has been accumulating that strongly suggests a connection between specific periodontal pathogens and cardiovascular disease (CVD). One such pathogen, Porphyromonas gingivalis, has been isolated directly from diseased vascular tissue and has been shown in animal models to be directly involved in pathological changes associated with CVD. In addition, P. gingivalis has been shown to invade human cardiovascular endothelial cells. The aim of our study was to evaluate the ability of P. gingivalis to affect human endothelial cell monolayer destruction and/or cell death by using in-vitro models focused on the presence of the beta-catenin protein of the endothelial adherens junctions and cell death markers. First, the secreted protein fractions (SPF) of P. gingivalis wild-type strains W83 and 381 were tested using human umbilical vein endothelial cells (HUVEC) for their ability to alter the endothelial adherens junction integrity. The results demonstrated that HUVEC monolayers were disrupted and the ability of the endothelial cells to remain adhered to the coverslip was hindered after treatment with SPF of either wild-type strain 381 or W83. Next, live cells of a triple gingipain knock out mutant, CW501, a fimA deficient mutant, YPF1, and P. gingivalis wild-type strains 381 and ATCC 33277 were examined using an invasion assay at an MOI of 100 for their ability to induce HUVE cell death. In addition, live, whole cell P. gingivalis strains 381, ATCC 33277 and W83 caused HUVEC monolayer disruption and cellular detachment after 20 hours whereas the triple gingipain mutant CW501 and fimA- YPF1 did not. As determined by electron microscopy, numerous 381 cells were observed inside the HUVECs after 5 and 20 hours in the invasion assay, whereas the CW501 cells were only seen in very low numbers, primarily at the cell periphery. Flow cytometry analysis of HUVECs after twenty four hours of exposure to wild-types 381 and ATCC 33277 demonstrated approximately 30% cell death, whereas exposure to YPF1 and CW501 resulted in only 12% cell death. When P. gingivalis internalization was inhibited with cytochalasin D, HUVEC death was not observed with the wild-type strains 381 and ATCC 33277. These data indicate that endothelial cell internalization of P. gingivalis and P. gingivalis gingipains play a significant role in the disruption of the adherens junctions as well as facilitating cell death in human endothelial cells.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Kristen L Totten.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Progulske-Fox, Ann.
Local: Co-adviser: Dunn, William A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0018840:00001

Permanent Link: http://ufdc.ufl.edu/UFE0018840/00001

Material Information

Title: Interactions of Porphyromonas gingivalis with Human Endothelial Cells
Physical Description: 1 online resource (212 p.)
Language: english
Creator: Totten, Kristen L
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: Molecular Cell Biology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Over the past several years, evidence has been accumulating that strongly suggests a connection between specific periodontal pathogens and cardiovascular disease (CVD). One such pathogen, Porphyromonas gingivalis, has been isolated directly from diseased vascular tissue and has been shown in animal models to be directly involved in pathological changes associated with CVD. In addition, P. gingivalis has been shown to invade human cardiovascular endothelial cells. The aim of our study was to evaluate the ability of P. gingivalis to affect human endothelial cell monolayer destruction and/or cell death by using in-vitro models focused on the presence of the beta-catenin protein of the endothelial adherens junctions and cell death markers. First, the secreted protein fractions (SPF) of P. gingivalis wild-type strains W83 and 381 were tested using human umbilical vein endothelial cells (HUVEC) for their ability to alter the endothelial adherens junction integrity. The results demonstrated that HUVEC monolayers were disrupted and the ability of the endothelial cells to remain adhered to the coverslip was hindered after treatment with SPF of either wild-type strain 381 or W83. Next, live cells of a triple gingipain knock out mutant, CW501, a fimA deficient mutant, YPF1, and P. gingivalis wild-type strains 381 and ATCC 33277 were examined using an invasion assay at an MOI of 100 for their ability to induce HUVE cell death. In addition, live, whole cell P. gingivalis strains 381, ATCC 33277 and W83 caused HUVEC monolayer disruption and cellular detachment after 20 hours whereas the triple gingipain mutant CW501 and fimA- YPF1 did not. As determined by electron microscopy, numerous 381 cells were observed inside the HUVECs after 5 and 20 hours in the invasion assay, whereas the CW501 cells were only seen in very low numbers, primarily at the cell periphery. Flow cytometry analysis of HUVECs after twenty four hours of exposure to wild-types 381 and ATCC 33277 demonstrated approximately 30% cell death, whereas exposure to YPF1 and CW501 resulted in only 12% cell death. When P. gingivalis internalization was inhibited with cytochalasin D, HUVEC death was not observed with the wild-type strains 381 and ATCC 33277. These data indicate that endothelial cell internalization of P. gingivalis and P. gingivalis gingipains play a significant role in the disruption of the adherens junctions as well as facilitating cell death in human endothelial cells.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Kristen L Totten.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Progulske-Fox, Ann.
Local: Co-adviser: Dunn, William A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0018840:00001


This item has the following downloads:


Full Text

PAGE 1

1 INTERACTIONS OF Porphyromonas gingivalis WITH HUMAN ENDOTHELIAL CELLS By KRISTEN LYNN TOTTEN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Kristen Lynn Totten

PAGE 3

3 ACKNOWLEDGMENTS I acknowledge my father, Donald L Totten, wh ose support over the years has guided and facilitated me to excel in my education efforts. My father and my stepmother, Reba, have been in my corner from the beginning of my graduate career through to its ces sation. Their love and support have meant more to me than any words could ever say. I thank my mother Babs Dinsmoor, and my stepfa ther David, for their love and support. I am truly blessed to have two such wonde rful sets of people in my life. In addition to my family, I acknowledge my mentors Dr. Ann Progulske-Fox and Dr. William Dunn Jr. for their advice and guidance thr oughout my graduate career. Their scientific knowledge has been very instrumental to my growth and success. My path through graduate school has been one of the most fulf illing experiences of my life thus far. I know that I am destined to succeed in my career goals armed with the knowledge and techniques acquired during my time at the University of Florida.

PAGE 4

4 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................3 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES................................................................................................................ .........8 ABSTRACT....................................................................................................................... ............12 CHAPTER 1 INTRODUCTION..................................................................................................................14 Dental Plaque.................................................................................................................. ........14 Porphyromonas gingivalis and Host Interactions...................................................................15 Cardiovascular Disease (CVD)...............................................................................................17 Periodontal Disease is Correlated with Cardiovascular Disease............................................18 P. gingivalis Virulence Factors...............................................................................................22 Capsule........................................................................................................................ ....22 Outer Membrane Vesicles (OMVs).................................................................................23 Toxic Metabolic Products................................................................................................23 Gingipains..................................................................................................................... ...24 Fimbriae....................................................................................................................... ....28 Epithelial and Endothelial Cells.............................................................................................33 Cell-Cell Adhesion Proteins............................................................................................35 Cell-Matrix Adhesion Proteins........................................................................................39 P. gingivalis Invasion of Epithelial and Endothelial Cells.....................................................42 P. gingivalis Persistence in Epithelia l and Endothelial Cells.................................................47 Apoptosis of Epithelia l and Endothelial Cells........................................................................48 Chapter Summary................................................................................................................ ...52 Objectives..................................................................................................................... ..........54 2 MATERIALS AND METHODS...........................................................................................55 Bacterial Strains and Growth Conditions...............................................................................55 P. gingivalis Subculture...................................................................................................55 Cell Culture Conditions...................................................................................................55 Bacterial Preparations......................................................................................................... ....56 P. gingivalis Lysate.........................................................................................................56 Secreted Protein Factor....................................................................................................56 Microassay Procedure for Proteins..................................................................................57 Enzymatic Assay.............................................................................................................57 Inhibitors..................................................................................................................... .....58 Microscopy..................................................................................................................... ........58 Antibodies..................................................................................................................... ...58

PAGE 5

5 Immunofluorescence Microscopy...................................................................................58 Transmission Electron Microscopy.................................................................................59 Cell Death Assay............................................................................................................... .....60 Quantification of Endothelial Cells by Flow Cytometry.................................................60 Antibiotic Protection Assay.............................................................................................61 Inhibitors..................................................................................................................... .....61 Caspase 3 Activity Assay................................................................................................61 Statistical Analysis........................................................................................................... .......62 Transepithelial Resistance..................................................................................................... .62 3 RESULTS........................................................................................................................ .......66 Monolayer Integrity............................................................................................................ ....66 Effects of P. gingivalis Lysate on the Cellular Dist ribution of the Junctional Proteins in HuH7 and HUVE Cells..............................................................................66 Effects of Heat and Protease Inhibitors on the Cellular Distribution of the Cadherin and -catenin Proteins in HuH7 Cells Treated with P. gingivalis W83 Lysates .........68 Proteinase Inhibitor Treatment of P. gingivalis W83 on HuH7 Adherens Junctional Complexes....................................................................................................................69 Effects of P. gingivalis Secreted Protein Fraction ( SPF) on the Adherens Junctional Complex Proteins of HUVECs....................................................................................70 Effects of P. gingivalis on Confluent HUVEC Monolayers...........................................72 Effects of Gingipain Mutants on the Junctional Complex Protein -catenin in Confluent HUVEC Monolayers...................................................................................74 Effects of Adherence Mutants of P. gingivalis from Strains W83 or ATCC 33277 on Confluent HUVEC Monolayers..............................................................................78 Effects of Invasion/Pe rsistence Mutants of P. gingivalis from Strains W83 or ATCC 33277 on Confluent HUVEC Monolayers.......................................................79 Effects of Additional P. gingivalis Mutants from Strain W83 on Confluent HUVEC Monolayers..................................................................................................................81 Effects of Additional P. gingivalis Mutants at Higher MOIs..........................................82 Invasion....................................................................................................................... ............83 Cell Death..................................................................................................................... ..........86 Cell Death of HUVECs after Exposure to P. gingivalis .................................................86 Inhibition of P. gingivalis Internalization of HUVEC....................................................89 P. gingivalis Secreted Protein Fraction (SPF) and HUVE cell death..............................89 HUVEC Exposure to P. gingivalis Results in Caspase 3 Activity..................................90 Transepithelial Resistance..................................................................................................... .91 4 DISCUSSION..................................................................................................................... ..156 Epithelial Barrier Function...................................................................................................156 Monolayer Integrity............................................................................................................ ..158 P. gingivalis W83 and 381 SPF and W83 Lysate Adversely Affect Endothelial and Epithelial Adherens Proteins......................................................................................158 P. gingivalis Gingipain Mutants Demonstrat e Different Effects on HUVE Cell Monolayers................................................................................................................162

PAGE 6

6 Effects of P. gingivalis on Adhesion and Invasion of HUVEC Monolayers................165 Cell Death..................................................................................................................... ........174 LIST OF REFERENCES.............................................................................................................182 BIOGRAPHICAL SKETCH.......................................................................................................212

PAGE 7

7 LIST OF TABLES Table page 2-1 Bacterial strains and plasmids us ed and constructed in this study.....................................65 3-1 Comparison of wild-type P. gingivalis strains and their effects on the HUVEC monolayer...................................................................................................................... ....95 3-2 Comparison of P. gingivalis gingipain mutants and their effects on the HUVEC monolayer...................................................................................................................... ....95 3-3 Comparison of P. gingivalis adhesion mutants and their effects on the HUVEC monolayer...................................................................................................................... ....96 3-4 Comparison of P. gingivalis invasion mutants and their effects on the HUVEC monolayer...................................................................................................................... ....97 3-5 P. gingivalis persistence mutant and its effects on the HUVEC monolayer.....................98 3-6 Comparison of other mutants of P. gingivalis and their effects on the HUVEC monolayer...................................................................................................................... ....99

PAGE 8

8 LIST OF FIGURES Figure page 2-1 Flow cytometry scatter plot of HUVEC s showing live cells and dead cells. (A) Twelve hour scatter plot of untreated HUVECs and (B) scatter plot of HUVECs............64 3-1 Effects of P. gingivalis W83 lysate on the HuH7 junctional proteins. HuH7 cells incubated for 24 hours with anti-Pan-cadherin (A) anti-catenin (B), anti-occludin....100 3-2 Effects of P. gingivalis W83 lysates on HUVEC junctiona l proteins. (A) untreated control incubated with anti-catenin, (B) HUVEC cells treated with 0.05....................101 3-3 Effects of temperature and TLCK treat ment on the proteolytic activity of live P. gingivalis and the P.g lysate. HuH7 cell monolayers were treated with the P.g W83..101 3-4 Effects of temperature and TLCK treat ment on the proteinase activity of live P. gingivalis and P.g lysate on HuH7 cells. All cells were in cubated with an antibody...102 3-5 Effects of varying c oncentrations of SPF from P. gingivalis W83 and 381 on HUVE cell adhesion. HUVECs were treated with the secreted protein factor (SPFs)...............103 3-6 Effects of varying c oncentrations of SPF from P. gingivalis W83 and 381 on HUVE cell adhesion. HUVEC cells were treated with the secreted protein factor....................104 3-7 Junctional complexes of untreated H UVEC monolayers remain intact through 20 hours. HUVEC monolayers were stained with antibodies against the junctional..........105 3-8 Junctional labeling of HUVEC mono layers after treatment with the P. gingivalis strain W83 at an MOI of 100. The untre ated HUVEC monolayer as shown in (A).......106 3-9 Junctional labeling of HUVEC mono layers after treatment with the P. gingivalis strain 381 at an MOI of 100. The untreat ed HUVEC monolayer as shown in (A) was.107 3-10 Junctional labeling of HUVEC mono layers after treatment with the P. gingivalis strain ATCC 33277 at an MOI of 100. The untreated HUVEC monolayer...................108 3-11 Junctional labeling of HUVEC mono layers after treatment with the P. gingivalis strain AJW4 at an MOI of 100. The untreated HUVEC monolayer as..........................109 3-12 Junctional labeling of HUVEC mono layers after treatment with the P. gingivalis wild-type strain W83 at an MO I of 1000. The untreated HUVEC.................................110 3-13 Junctional labeling of HUVEC mono layers after treatment with the P. gingivalis strain 381 at an MOI of 1000. The untreated HUVEC monolayer as.............................111 3-14 Junctional labeling of HUVEC mono layers after treatment with the P. gingivalis strain ATCC 33277 at an MOI of 1000. The untreated HUVEC...................................112

PAGE 9

9 3-15 Junctional labeling of HUVEC mono layers after treatment with the P. gingivalis strain AJW4 at an MOI of 1000. The untreated HUVEC monolayer as........................113 3-16 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis MT10 ( rgpA) at an MOI of 100. The HUVEC monolayers were labeled with........................114 3-17 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis MT10W ( rgpA-, kgp) at an MOI of 100. The HUVEC monolayers were labeled........115 3-18 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis G102 ( rgpB) at an MOI of 100. The HUVEC monolayers were labeled with........................116 3-19 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis G102W ( rgpB-, kgp) at an MOI of 100. The HUVEC monolayers were labeled with...............117 3-20 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis YPP2 ( kgp) at an MOI of 100. The HUVEC monolayers were labeled with..........................118 3-21 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis CW401 ( rgpA-, rgpB) at an MOI of 100. The HUVE C monolayers were labeled....................119 3-22 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis CW501 ( rgpA-, rgpB-, kgp) at an MOI of 100. The HUVEC monolayers were........................120 3-23 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis MT10 ( rgpA) at an MOI of 1000. The HUVEC monolayers were labeled with......................121 3-24 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis MT10W ( rgpA-, kgp) at an MOI of 1000. The HUVEC monolayers were labeled......122 3-25 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis G-102 ( rgpB) at an MOI of 1000. The HUVEC monolayers were labeled with......................123 3-26 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis G102W ( rgpB-, kgp) at an MOI of 1000. The HUVE C monolayers were labeled.....................124 3-27 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis YPP2 ( kgp) at an MOI of 1000. The HUVEC monolayers were labeled with........................125 3-28 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis CW401 ( rgpA-,rgpB) at an MOI of 1000. The HUVE C monolayers were labeled...................126 3-29 Junctional labeling of HUVEC mo nolayers after treatment with P. gingivalis CW501 ( rgpA-, rgpB-, kgp) at an MOI of 1000. The HUVEC monolayers were......................127 3-30 Junctional labeling of HUVEC mono layers after treatment with a P. gingivalis adhesion mutant PG1683 (conserved hypothetical protein) at an MOI of 100. The......128

PAGE 10

10 3-31 Junctional labeling of HUVEC mono layers after treatment with a P. gingivalis adhesion mutant PG0242 (conserved hypothetical protein) at an MOI of 100. The......129 3-32 Junctional labeling of HUVEC mono layers after treatment with a P. gingivalis adhesion and invasion mutant PG1118 ( clpB) at an MOI of 100. The HUVEC...........130 3-33 Junctional labeling of HUVEC mono layers after treatment with a P. gingivalis adhesion mutant YPF1 ( fimA) at an MOI of 100. The HUVEC monolayers were.......131 3-34 Junctional labeling of HUVEC mono layers after treatment with a P. gingivalis invasion mutant PG0717 (putative lipoprotei n) at an MOI of 100. The HUVEC..........132 3-35 Junctional labeling of HUVEC mono layers after treatment with a P. gingivalis invasion mutant PG1286 (ferritin) at an MOI of 100. The HUVEC monolayers were..133 3-36 Junctional labeling of HUVEC mono layers after treatment with a P. gingivalis invasion mutant YPEP ( pepO -) at an MOI of 100. The HUVEC monolayers were......134 3-37 Junctional labeling of HUVEC mono layers after treatment with a P. gingivalis putative cysteine peptidase mutant PG1788 at an MOI of 100. The HUVEC...............135 3-38 Junctional labeling of HUVEC mono layers after treatment with a P. gingivalis putative secretion activator protein muta nt PG0293 at an MOI of 100. The HUVEC...136 3-39 Junctional labeling of HUVEC mono layers after treatment with a P. gingivalis conserved hypothetical pr otein mutant PG0686 at an MOI of 100. The HUVEC.........137 3-40 Junctional labeling of HUVEC mono layers after treatment with a P. gingivalis adhesion mutant YPF1 ( fimA) at an MOI of 1000. The HUVEC monolayers were.....138 3-41 Junctional labeling of HUVEC mono layers after treatment with a P. gingivalis invasion mutant YPEP ( pepO) at an MOI of 1000. The HUVEC monolayers were....139 3-42 Junctional labeling of HUVEC mono layers after treatment with a P. gingivalis conserved hypothetical pr otein mutant PG0686 at an MOI of 1000. The HUVEC.......140 3-43 Comparisons of the HUVEC monolayers af ter treatment with th e gingipain mutants and their parent strains at an MO I of 100 for 20 hours. HUVECs were.........................141 3-44 Comparisons of the HUVEC m onolayers after treatment with P. gingivalis adhesion mutants and their parent strains at an MOI of 100 for 20 hours. The.............................142 3-45 Comparisons of the HUVEC m onolayers after treatment with P. gingivalis invasion mutants and their parent strains at an MOI of 100 for 20 hours. The.............................143 3-46 Comparisons of the HUVEC monolay ers after treatment with additional P. gingivalis mutants and their parent strain at an MOI of 100 for 20 hours. The..............................144

PAGE 11

11 3-47 Transmission electron microscopy of HUVEC monolayers in ternalized with P. gingivalis 381 at an MOI of 100 and stained fo r acid phosphatase. (A) Untreated........145 3-48 Transmission electron microscopy of HUVEC monolayers in ternalized with P. gingivalis CW501 ( rgpA-, rgpB-, kgp) at an MOI of 100 and stained for acid..............146 3-49 Transmission electron microscopy of HUVEC monolayers in ternalized with P. gingivalis wild-type 381, CW501 ( rgpA-, rgpB-, kgp), and a conserved hypothetical..147 3-50 Effects of P. gingivalis on HUVE cell death. HUVEC monolayers were treated for 30 min, 2.5, 12, and 24 hours with P. gingivalis strains 381, ATCC 33277,..................148 3-51 P. gingivalis interactions with HUVEC monolay ers induces HUVE cell death after antibiotic treatement. HUVEC monolaye rs were treated for 2.5, 12, and 24.................149 3-52 Inhibition of P. gingivalis internalization of HUVECs with cytochalasin D inhibits HUVE cell death in the presence of antibiotics. HUVEC monolayers were..................150 3-53 Inhibition of P. gingivalis internalization of HUVEC with cytochalasin D inhibits HUVE cell death. HUVEC monolayer s were pretreated with 5 g/ml..........................151 3-54 Cell death of HUVE cells after treatment with live P. gingivalis 381 with and without cytochalasin D, 0.02 g/ml of the 381 SPF and the CW501 (gingipain-null)....152 3-55 HUVEC monolayers treated with P. gingivalis 381 and ATCC 33277 exhibit caspase 3 activity. HUVEC monolayers were treated for 2.5, 12, and 24 hours with.................153 3-56 Membrane resistance of HuH7 monolayers after apical exposure to P. gin givalis strains W83, 381, and the gingipain-null mutant, CW501, at an MOI of 100. HuH7.....154 3-57 Membrane resistance of HuH7 monolayers after baso lateral exposure to P. gin givalis strains W83, 381, and the gingipainnull mutant, CW501, at an MOI of........................155

PAGE 12

12 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy INTERACTIONS OF Porphyromonas gingivalis WITH HUMAN ENDOTHELIAL CELLS By Kristen Lynn Totten August 2007 Chair: Ann Progulske-Fox Cochair: William Dunn Jr. Major Department: Medical Science Over the past several years, evidence has been accumulating that strongly suggests a connection between specific peri odontal pathogens and cardiovascul ar disease (CVD). One such pathogen, Porphyromonas gingivalis, has been isolated directly from diseased vascular tissue and has been shown in animal models to be dire ctly involved in pathological changes associated with CVD. In addition, P. gingivalis has been shown to invade human cardiovascular endothelial cells. The aim of our st udy was to evaluate the ability of P. gingivalis to affect human endothelial cell monolayer destruction a nd/or cell death by using in-vitro models focused on the presence of the -catenin protein of the endothelial adherens junctions and cell death markers. First, the secreted protein fractions (SPF) of P. gingivalis wild-type st rains W83 and 381 were tested using human umbilic al vein endothelial cells (HUVE C) for their ab ility to alter the endothelial adherens juncti on integrity. The results demo nstrated that HUVEC monolayers were disrupted and the ability of the endothelial cells to remain adhered to the coverslip was hindered after treatment with SPF of either wild-type st rain 381 or W83. Ne xt, live cells of a triple gingipain knock out mutant, CW501, a fimA deficient mutant, YPF1, and P. gingivalis wild-type strains 381 and ATCC 33277 were examined using an invasion a ssay at an MOI of 100

PAGE 13

13 for their ability to induce HUVE cell death. In addition, live, whole cell P. gingivalis strains 381, ATCC 33277 and W83 caused HUVEC monolayer di sruption and cellular detachment after 20 hours whereas the triple gingipain mutant CW501 and fimAYPF1 did not. As determined by electron microscopy, numerous 381 cells were ob served inside the HUVECs after 5 and 20 hours in the invasion assay, whereas the CW501 cells were only seen in very low numbers, primarily at the cell periphery. Flow cytometry analysis of HUVECs after twenty four hours of exposure to wild-types 381 and ATCC 33277 de monstrated approximately 30% cell death, whereas exposure to YPF1 and CW501 resulted in only 13% cell death. When P. gingivalis internalization was inhibited with cytochalasin D, HUVEC death wa s not observed with the wild-type strains 381 and ATCC 33277. These data indicate th at endothelial cell in ternalization of P. gingivalis and P. gingivalis gingipains play a significant role in the disr uption of the adherens junctions as well as facilitating cell death in human endothelial cells.

PAGE 14

14 CHAPTER 1 INTRODUCTION Dental Plaque The formation of dental plaque on oral surf aces is a highly complex process. The hard tissues of the oral cavity (tee th) accrue plaque in the form of multi-component biofilms. Contrary to early observations, de ntal plaque does not exist as a homogeneous or standardized structure. In actuality, plaque ex ists in various forms depending on its location in the oral cavity (Rosan and Lamont, 2000). Greater than 700 bacter ial species have been identified in the oral cavity. Typically, gram-positive organisms such as Streptococcus gordonii, Actinomyces and other related species are found at the gingival level of the tooth wh ere they coexist in a balanced, symbiotic relationship with the host (Liebana et al. 2004; Rosan and Lamont, 2000; Marsh, 1994; Kolenbrander and London, 1993). If this balan ce is disturbed, the e quilibrium between the host gingival tissue and the plaque is lost. As the plaque thic kens, the biofilm becomes more diversified and forms areas with decreased oxyge n reduction potential between the supragingival and subgingival regions. In time, the gram-positi ve organisms are followed by a series of gramnegative anaerobes which situate themselves w ithin these newly developed, lower oxygen areas (Socransky et al. 2000). One such gram-negative anaer obe to inhabit the subgingival dental plaque biofilm is Porphyromonas gingivalis. P. gingivalis is capable of attach ing to a number of oral bacteria including S. gordonii and Fusobacterium nucleatum (Kolenbrander and London, 1993; Lamont et al. 1992) and in-vivo studies using healthy volunteers have demonstrated that P. gingivalis locates predominately on the heavily st reptococcal-laden areas of the subgingival dental plaque (Slots and Gibbons, 1978). The col onization of multiple bacterial species within periodontal pockets leads to the de struction of host gingival tissu e, thus leading to periodontal disease (Nakamura et al. 1999). The migration of P. gingivalis into the dental biofilm is a vital

PAGE 15

15 process in the conversion from a commensal pl aque to a pathogenic unit (Listgarten, 1999; Whittaker et al. 1996). Porphyromonas gingivalis and Host Interactions Periodontal diseases are categori zed as a collection of infecti ons that ultimately lead to gingival inflammation, periodontal tissue destruction, alveolar bone loss, and eventually, tooth exfoliation (Socransky and Haffajee, 1992). Periodontal disease affect s about 23% of adults ages 65-74, and 14% of adults between the ages of 45-54 (US Public Health Service, 2000). In the United States, it is estimated that fifty milli on people have some form of periodontal disease (Cutler et al. 1995). There are two major types of periodontal dis eases: gingivitis and periodontitis. Gingivitis is an acute infection involving the coronal pl aque, and periodontitis is a chronic infection involving the subgingival plaque (Castillo and Alvarez, 2004; Liebana et al. 2004). Dilations of the vasculature and an increase of phagoctyic ce lls are common host tissue responses to growing and multiplying bacteria during acute infections such as gingivitis. Tissue infections of this kind usually stimulate swift, unsyste matic inflammatory responses to the invading bacteria and bacterial products (Olsen and Dahlen, 2004; Dahlen et al. 1982). The viable and dividing bacteria within the tissue repr esent a vital stage of an acute infection. Symptoms commonly associated with these types of infections are te nderness and pain. The th reat of spreading is brought on by the host response to polymorphonucl ear leukocytes (PMNs) and tissue destruction (Dahlen et al. 1982). Characteristically, anaerobic bacteria cooperate with multiple bacteria to promote acute infections. Anaerobic bacteria equipped with polysaccharide capsules, collagenases and other proteolytic enzymes that degrade tissues pose a significant threat. Char acteristics of chronic infections include a slower ti ssue response and lymphocyte domina ted tissue restructuring and

PAGE 16

16 repair (Olsen and Dahlen, 2004; Dahlen et al. 1982). In healthy individuals, the subgingival plaque is located in the virtual space of the gingival sulcus and mainta ins a very low rate of colonizing bacteria (Haffajee et al. 2003; Socransky et al. 1998). Periodontitis is characterized by plaque induced chronic inflammation and lesi on formation, leading to the formation of periodontal pockets and oral bone loss. Peri odontitis can progress to several different manifestations including progres sive, chronic or aggressive, a nd may also be localized or generalized (Slots, 2000; Slots a nd Jorgensen, 2000). Both gingiv itis and periodontitis display gingival tissue inflammation in response to bacteria laden plaque Unlike periodontitis, gingivitis is devoid of the detr imental effects of tissue destru ction and can be eradicated by plaque treatments and proper dental hea lth (Castillo and Alvarez, 2004; Liebana et al. 2004). The oral pathogen, Porphyromonas gingivalis is a vital microbe involved in initiating chronic periodontitis (CP). P. gingivalis is a highly adapted, black-pigmented, asaccharolytic, anaerobic, rod shaped bacterium that is equipped with an array of rec ognized virulence factors capable of causing disease (Genco CA, 1999; Cutler et al. 1995; Genco, 1992a; Genco, RJ 1992c). Examples of these vi rulence factors include fimb riae and lectin-type adhesion molecules, lipopolysaccharide (LPS), a capsula r polysaccharide (CPS) (or K antigen), hemolysins and hemagglutinins (Progulske-Fox et al. 1993), toxic metabolic products, outer membrane vesicles (blebs), and cystei ne proteinases (gingipains) (Aduse-Opoku et al. 2006; Cutler et al. 1995). The host defense system mounts a counterattack to the challeng e of this microbes surfaceassociated and extracellular virulence factors through both the innate and acquired immunities. The balance between the net e quilibrium of the body against the development of destructive disease is dependent on the dynamics of this host-microbe interaction. A successful host

PAGE 17

17 response manages the challenge w ith little inflammation to the gi ngiva and without any lasting tissue destruction. Alternatively, an aggravated host response, such as the response induced by Porphyromonas gingivalis causes an increase in the amount of inflammation and can ultimately lead to permanent destruction of host tissue (Curtis et al. 1999a). P. gingivalis has also been shown to cross the barrier of the oral mucosal tissue and enter the bloodstream after eating, tooth brushing, and flo ssing and this entry into the bloodstream is hypothesized to be crucial in causing systemic infections (Darveau et al. 1995). For example, P. gingivalis invasion of the endothe lial cells of the arterial walls c ould initiate persistent injury leading to adverse cardiovascular ev ents (Deshpande and Khan, 1999; Dorn et al. 1999; Deshpande et al. 1998b; Deshpande et al. 1998a). Cardiovascular Disease (CVD) The development of atherosclerosis is minimized by high plasma levels of the antioxidant/anti-inflammatory, high-de nsity lipoproteins (HDLs) (Castelli et al. 1986; Gordon et al. 1977). In the artery wall, HDL promotes the efflux of cholesterol, binds lipopolysaccharides, protects erythrocytes from de veloping procoagulant activity, s timulates endothelial migration, inhibits endothelial cell synthesi s of the platelet activating fact or, and inhibits endothelial cell adhesion molecules and monocyte ch emotactic protein (MCP-1) (Zhang et al. 2003; Barter et al. 2002; Barter and Rye, 1996; Sugatani et al. 1996; Epand et al. 1994; Murugesan et al. 1994; Levine et al. 1993; Yui et al. 1988; Fleisher et al. 1982). HDLs are antithrombotic and likely to alter endothelial functions by stimulating the expression of endothelial nitric oxide (NO) (Viswambharan et al. 2004; O'Connell and Genest, 2001; Zeiher and Schachinger, 1994). Nitric oxide (NO) mediates chemical anti-atherogenic properties such as blood pressure by acting as a vasodilator or vasorelaxer (Knowles et al. 2000). Early atheroscle rotic events include the

PAGE 18

18 enrollment and build up of monocytes and lymp hocytes, and the coll ection of oxidized lowdensity lipoproteins (LDLs) to the arterial wall (Hegele, 1999). The monocyt es can traverse into the intima where they can replicate and differe ntiate into lipid and cholesterol accruing macrophages or foam cells (Ross, 1993). LDL th en initiates the transp ort of cholesterol and phospholipids into the mammalian cells (Navab et al. 2000). To circumvent this process, macrophages release a 34-kDa glycoprotein, apolipoprotein (apoE), which helps prompt cholesterol efflux to HDL (Lusis, 2000). Apolipoprotein is a structural constituent of LDL, VLDL (very low-density lipoprotein), and some HDLs (Krimbou et al. 1997; Mahley, 1988). Plasma a poE is manufactured mainly by the liver but also by a variety of organ associated macrophages (Williams et al. 1985; Zannis et al. 1985; Driscoll and Getz, 1984). Hypercholestero lemia and scattered atherosclerotic lesions occur from apoE deficiencies (apoE -/-), in humans and gene altered mice (Plump et al. 1992; Zhang et al. 1992; Ghiselli et al. 1981). Interestingly, Li et al. reported that heterozygous apolipoprotein E (apoE) +/mice that received weekly intravenous injections of P. gingivalis into the systemic circulation resulted in elevated lipid deposits and developed rapid atherosclerosis when compared to sham injected mice (Li et al. 2002). Oral infection with P. gingivalis has also been shown to increase the pr ogression of early atherosclerosis within apolipoprotein E-null mice (Lalla et al. 2003). Jain et al. reported that rabbits with experimentally induced periodontit is that were exposed to a hi gh fat diet developed elevated levels of vascular lipid deposits compared to healthy controls fed the same high-fat diet (Jain et al. 2003). Periodontal Disease is Correlated with Cardiovascular Disease In the western world, CVD is by far the primar y cause of death. Un til recently, the most common identified risk factors a ssociated with this disease were smoking, high blood pressure,

PAGE 19

19 diabetes, and obesity. Interes tingly, these common risk factor s for CVD could not explain many coronary related deaths. According to the Fr amingham Heart Study risk scores, approximately 15% of women and 25% of men in the lowest two ri sk factor quintiles of th is study died of CVD (Leaverton et al. 1987). Given this data, it is possible that CVD may be initiated by additional stimuli, possibly poor dental health (Leaverton et al. 1987). In fact, over th e past several years, scientists have accumulated data which strong ly associates a connection between periodontal pathogens and CVD (Mattila et al. 1989). Furthermore, it has been reported that the presence of periodontal pathogens in the host ar e not restricted to the periodontal tissues and in fact, can act systemically (Beck et al. 1996). Entry into the circulation could provide this b acterium with direct access to the cells that make up the vessel walls (Carroll and Sebor, 1980; Silver et al. 1977; Sconyers et al. 1973). A report released in 1989 by Mattila et al showed that patients with myocardial infarctions had an increased level of poor dental health co mpared to the cont rol population (Mattila et al. 1989). Interesting information using the murine macrophage cell line J774 A1 was published by Kuramitsu et al., which reported that foam cells (an impor tant indicator of CVD) were induced by the outer membrane vesicles (blebs) of P. gingivalis as well as P. gingivalis alone, a response likely mediated by LPS (Kuramitsu et al. 2001). In addition, P. gingivalis has been identified directly from vascular tissue in the diseased state. A 1998 publication from a study by Haraszthy et al reported that P. gingivalis had been identified, using specific PCR primers, in atheromatous plaques (Haraszthy et al. 2000). In addition, in 1999, Porphyromonas gingivalis was detected in atherosclerotic plaques using immunost aining techniques (Chiu, 1999). Cavrini et al identified P. gingivalis via PCR and fluorescence in situ hybridization, in athero matous plaques of two atherosclerotic patients, suggesting that P. gingivalis may be metabolically active within these

PAGE 20

20 plaques (Cavrini et al. 2005). After incubating endothelial cells with atherosclerotic plaque homogenates, Kozarov et al reported the presence of both A. actinomycetemcomitans and P. gingivalis (Kozarov et al. 2005). Since only viable bacteria are capable of invading host cells (Lamont et al. 1995; Meyer et al. 1991), this study confir med the presence of live P. gingivalis within diseased atherosclerotic sites (Kozarov et al. 2005). Moreover, Desvarieux et al. conducted an Oral Infections and Vascular Disease Epidemiology Study (INVEST) which was designed to evaluate a possible link between periodontal disease and the development of myocardial infarctions, atherosclerosis, car diovascular disease a nd stroke frequencies (Desvarieux et al. 2005). After adjusting for common cardiovascular disease risk factors such as age, gender, smoking, diabetes and high bl ood pressure, this group reported a direct correlation between the burden of periodontal pathogens, specifically, P. gingivalis and the thickness of the intima-media of the carotid artery (Desvarieux et al. 2005). Ulcerations of the gingiva and changes within the local vasculature caused by P. gingivalis infections can lead to seve re bacteremias and increased levels of inflammation. (Slade et al. 2003) P. gingivalis induced local inflammation within arteri al walls may lead to the acceleration of atherosclerotic events (Gibson et al. 2006). Elevated levels of C-reactive protein (CRP), markers for inflammation and CVD, have al so been shown to be associated with P. gingivalis infections (Teragawa et al. 2004). In the study conducted by Grossi et al serum and carotid endarterectomy samples taken from 16 subjects who presented periodontal disease showed an association between attachment loss of greater than or equal to 4 mm and CVD history (Grossi et al. 1994). This study also determined that the seru m CRP levels were considerably elevated in 75% of these subjects (Glurich et al. 2002; Grossi et al. 1994). In addition, CVD has been shown to be associated with serum antibodies to P. gingivalis (Pussinen et al. 2003). Finally, as

PAGE 21

21 mentioned above, studies using animal models have established accelerated atherosclerotic events following continual injections of P. gingivalis (Lalla et al. 2003; Li et al. 2002) and P. gingivalis orally challenged homozygous ApoE-/mice were vulnerable to accelerated atherosclerosis (Gibson et al. 2004; Lalla et al. 2003). This collection of scientific evidence validates the conclusion that a str ong correlation exists between CVD and P. gingivalis associated periodontal disease. P. gingivalis may induce cardiovascular events by several different mechanisms. First, P. gingivalis endotoxins such as LPS, may antagonize m onocytes and produce elevated numbers of chemokines and cytokines systemically, leading to increased lipid induction, endothelial cell proliferation, and the luring of PMNs, causing vascular wall thickening (Liebana et al. 2004). An increase in the amount of proinflammatory cytoki nes could also result in an increase of foam cells. In murine macrophages, the uptake of LDL is elevated when exposed to P. gingivalis (Qi et al. 2003) and this increased uptak e of LDL has also been show n to induce the formation of foam cells in HUVECs (Kuramitsu et al. 2003). In fact, the ch emoattractant MCP-1 from HUVECs was increased 5-fold after exposure to P. gingivalis strain 381 (Kuramitsu et al. 2001). Second, the activation of complement by endotoxin s and C reactive protein (CRP) leads to an influx of PMNs, and platelet aggregation induced by monocytes, which can lead to increased inflammation associated with cardiovascular disease events (Kinane, 1998). Specifically, P. gingivalis Arg-gingipains (Rgps) can increase the cal cium concentration in human platelets and thereby, cause their aggreg ation (Imamura and Travis, et al ., 2003). P. gingivalis can also induce platelet aggregation by expressing collagenlike platelet aggregati on-associated proteins (PAAP) (Beck and Offenbacher, 2001). Aggreg ation of platelets is a consequence of P. gingivalis present on the platelet surface. These a ggregating platelets may play a part in

PAGE 22

22 thrombus formation (Beck and Offenbacher, 2001; Beck et al. 2000). Also, as mentioned above, the existence of P. gingivalis DNA within atheromatous plaques provides at least circumstantial evidence of P. gingivalis involvement in CVD events (Garcia et al. 2001; Haraszthy et al. 2000). P. gingivalis may also be directly involved in the formation of atherosclerotic plaques. During fibrous plaque development, lipid di scharge and vessel narrowing may be induced by atheroma cells experiencing apoptosis, necrosis, and mineralization (Ross, 1999). Ischemia and myocardial infarctions are initia lly mediated by plaque fissures, and later by thrombosis of the vessel (Lowe, 1998; McGill, 1968). Matrix metalloproteinases (MMPs) induced by P. gingivalis may contribute to these plaque fractu res, or ruptures, leading to the P. gingivalis induced degradation of human atheroma samples in vitro (Kuramitsu et al. 2001). P. gingivalis Virulence Factors Capsule P. gingivalis contains virulence factors such as polysaccharide capsules (K antigen), outer membrane vesicles (blebs), toxic metabolic produ cts, hemagglutinins, gingipains, and fimbriae. The polysaccharide capsules, characteristic of some strains of P. gingivalis, evade the host antimicrobial defenses by playi ng a role in inhibiting the attach ment of complement, creation of pustules, and evasion from phagocytes (Brook, 1987; Dahlen and Nygren, 1982). Encapsulated P. gingivalis contain an important virulence factor, K antigen, which is associated with serum resistance, phagocyte resistance, and the recr uitment of specific antibodies necessary for opsonization and complement-media ted killing (van Winkelhoff et al. 1993). Therefore, capsules provide protection from the host defense system (Olsen and Dahlen, 2004)

PAGE 23

23 Outer Membrane Vesicles (OMVs) During normal growth, many gram negative bacteria including P. gingivalis shed outer membrane vesicles (OMV) into the culture me dia (Beveridge and Kadur ugamuwa, 1996; Grenier and Mayrand, 1987). Components of the bacterial cell surface including L PS, proteinases, and adhesins are contained within OMVs (Qi et al ., 2003) and OMVs have been shown to express amounts of proteolytic activities equal to that of whole cell P. gingivalis (Grenier and Mayrand, 1987). OMVs are not only capable of aiding the a ttachment of parent st rains to host cells, but are also capable of attaching to host cells directly (Grenier and Mayrand, 1987). During periodontitis, P. gingivalis and its OMVs may be secreted into the circulation through transient bacteremias that can occur after chewing f ood, tooth brushing, and/or flossing (Miyakawa et al. 2004; Carroll and Sebor, 1980). The presence of LDL with the heat stable OMVs or LPS of P. gingivalis 381 have been shown to induce foam cells (an essential CVD characteristic) in murine macrophages (Kuramitsu et al. 2001). Toxic Metabolic Products Catalase, peroxidase and super oxide dismutase (SOD) all play a role in neutralizing toxic oxygen metabolites and are expressed by bot h aerobic and anaerobi c bacteria (Amano et al. 1998; Amano et al. 1988). P. gingivalis is known to express SOD in higher amounts than other anaerobic, gram-negative rods, such as Porphyromonas levii, Fusobacterium nucleatum and Prevotella species including intermedia, denticola, me laninogencia, and loescheii (Amano et al. 1998; Amano et al. 1988). An increased level of SOD activity would defend P. gingivalis against the O2 from prolonged air exposure and the neutr ophil created bacter icidal superoxide anion (O2-) (Amano et al. 1998; Amano et al. 1992). Because a gene-inactivated sod mutant rapidly lost viability after O2 exposure and the mRNA sod levels were increased in response to elevated temperatures characteri stic of periodontal pockets, it ha s been proposed that SOD is a

PAGE 24

24 contributing virulence factor in the pathogenicity of P. gingivalis (Mayrand and Holt, 1988). Thus, increased SOD activity in-vivo may provide P. gingivalis with protection against the high levels of superoxide induced by neutrophils within the irritated periodontal pockets (Amano et al. 1994). Gingipains Gingipains are cysteine protei nases and have been implicat ed in the activation of host proenzymes, degradation of host tissues, and the neutralization of the host immune response (Kuramitsu et al. 1995). The generation of nutritiona l requirements, host immune protein degradation, host tissue adhesion, and the unearthing of cell cryptit opes are all examples of the virulence traits that have b een proposed or attributed to P. gingivalis gingipains (Gibbons et al. 1990; Mayrand and Holt, 1988). For example, gingi pains are responsible for the degradation of several host proteins including collag en, fibrinogen, and fibronectin (Abe et al. 1998; Okamoto et al. 1998; Kadowaki et al. 1994), as well as the cytokines in terleukin-8 (IL-8), IL-6, and TNF(Oido-Mori et al. 2001; Banbula et al. 1999; Calkins et al. 1998). Gingipains are also implicated in the degradation of immunoglobulin s and the complement factors C3 and C5 (Abe et al. 1998; Kadowaki et al. 1994; Wingrove et al. 1992), interruption of polymorphonuclear leukocyte (PMN) bactericidal activity (Abe et al. 1998; Nakayama et al. 1995; Kadowaki et al. 1994) and as a potent inducer of human umbili cal vein endothelial cell, HUVEC, and human fibroblast cell death (Baba et al. 2002; Baba et al. 2001). These gingipains also likely function in the progression of vascular pe rmeability, cytokine a nd cytokine receptor inactivation, platelet aggregation, inhibition of blood coagulation, are cytotoxic to host cells, and are important for growth. The ability of P. gingivalis to reproduce and survive with in the periodontal pockets is also partially dependent on the gingipains (Chen et al. 2001b). P. gingivalis also possesses an

PAGE 25

25 endopeptidase, PepO. PepO has considerable homology to the endot helin-converting enzyme (ECE-1), a NEP family member, and has been s hown to affect the kallikrein/kinin cascade by cleaving bradykinin (Awano et al. 1999). The P. gingivalis gingipains include RgpA, RgpB, and Kgp. Rgp and Kgp cleave natural and synthetic substrates following arginine and lysine residues, respectively (Awano et al. 1999). Two related genes encode the arginine-sp ecific gingipain while on ly one gene encodes the lysine-specific gingipain. Two forms of gingipains are produced, a membrane-associated form accounting for approximately 80% of the Rgp and Kgp activity, and secreted forms. Through the use of specific inhibitors, Potempa et al determined that 85% of the bacterial trypsin-like activity was attributed to the gingipains and the concen tration of Rgp is 3-fold higher than Kgp (Potempa et al. 1997). The gingipain genes rgpA and kgp encode a propeptide including both catalytic a nd an adhesion/hemagglutinin domain, which appears to be postranslationally processed. RgpB encodes a propeptide and contai ns the catalytic domain but lacks the hemagglutinin/adhesion domain (Kuramitsu, 1998; Potempa et al. 1995). RgpAcat and RgpB are soluble, single-chain forms of Ar g-gingipains with a mass of 44-50 kDa and are present in the growth media in varying amounts depending on th e strain, age of culture, and media composition (Bedi and Williams, 1994; Kadowaki et al. 1994; Chen et al. 1992; Nishikata and Yoshimura, 1991; Fujimura and Nakamura, 1990; Tsutsui et al. 1987). RgpA also exists as a non-covalent 95 kDa complex containing an adhe sin/hemagglutinin domain and catalytic domain (HRgpA) (Rangarajan et al. 1997; Pike et al. 1996; Pike et al. 1994). Most strains of P. gingivalis maintain RgpB on the cell surface, which is denoted as membrane-type RgpB (mt-RgpB) with masses of 70-90 kDa, and secrete low levels of soluble RgpB (Curtis et al. 1999a; Curtis et al. 1999b). Kgp exists either as a singl e-chain or as a complex containing

PAGE 26

26 non-covalently associated catal ytic domain and adhesion/he magglutinin domains (Fujimura et al. 1998; Pike et al. 1996; Ciborowski et al. 1994; Fujimura et al. 1993; Scott et al. 1993). The activity of Kgp occurs mainly on the bacter ial cell surface and may exist as a complex of fully processed multi-domains (Bhogal et al. 1997). The membrane associated forms are proposed to be primarily accountable for the virule nce associated with this bacterium (Rajapakse et al. 2002; Slakeski et al. 1998). Both Rgp and Kgp are assumed to play a pivot al role in modulating host immune defenses, acquisition of nutrients, and tissue invasion (Genco et al. 1999; Genco CA, 1999; Lamont and Jenkinson, 1998). These enzymes can cause pot entially damaging activities by activating the kallikrein/kinin cascade, degrading host proteina se inhibitors, inactivating complement proteins C3 and C5, inactivating fibrinog ens clotting ability, modifying the antimicrobial activity of neutrophils, and degrading bacteric idal proteins, immunoglobins, and iron transporting proteins. All of these propert ies indicate that the gingipains of P. gingivalis can account for many of the clinical manifestations s een in periodontitis (Curtis et al. 1999a; Curtis et al. 1999b; Mayrand and Holt, 1988). In addition, these enzymes can stimulate serotonin secreting platelets, which could potentially give rise to cardiovascular im pediments because an increase in serotonin leads to an imbalance between the nitric oxide (NO) /serotonin levels, thus promoting endothelial dysfunction (Curtis et al. 1999a; Curtis et al. 1999b). The structure of the rgp gene is highly conserved sin ce cloning and sequencing of rgp from strains HG66, W83, W50, 381, a nd ATCC 33277 displayed very little coding sequence variations (Mikolajczyk-Pawlinska et al. 1998; Allaker et al. 1997). The organization of the genes suggests that either rgpA gave rise to rgpB by gene duplication minus the adhesin/hemagglutinin domain or that rgpB was copied and then introduced with an

PAGE 27

27 adhesin/hemagglutinin domain (Potempa et al. 2003). It has been dete rmined that widespread proteolytic processing of the emerging translatio n products is responsible for the production of the mature enzymes (Pavloff et al. 1997; Pike et al. 1996; Pavloff et al. 1995; Pike et al. 1994). The proteolytic activities of Rgp and Kgp are necessary for the creation of the 43 kDa and 47 kDa proteins from the precu rsor Rgp and Kgp proteins (Abe et al. 2004). Evidence for this was provided by the use of the rgpA-, rgpB-, kgpand the rgpA-, kgp-, hagAmutants as well as the combined use of Rgp and Kgp i nhibitors which completely inhibited the coaggregation ability of P. gingivalis (Abe et al. 2004). Not only does Arg-gingipain contribute to their own maturation via autoproteolysis, they are also responsible for the accurate and well organized maturation of Kgp (Potempa et al. 2003; Kadowaki et al. 1998; Okamoto et al. 1996). Kgp has been shown to be important for hemoglobin adsorption through the processing of the 19-kDa hemoglobin recepto r protein (HBr) encoded by the hagA, kgp and rgpA genes of P. gingivalis and hence, Kgp is likely very important as an energy source for P. gingivalis (Nakayama et al. 1998; Han et al. 1996; Okamoto et al. 1996; Pavloff et al. 1995). Differences in amino acid binding specificities at prime sites exis t between RgpB and HRgpA. The crystal structure of RgpB and HRgp A revealed identical ac tive sites with the exception of four additional amino acid s ubstitutions within the HRgpA. Ally et al. suggested it was the four additional active substitutions that were responsible for the observed differences between the binding specificities of RgpB a nd HRgpA, rather than the presence of the adhesin/hemagglutinin domains in HRgpA (Ally et al. 2003). It was also reported that the RgpAcat domain was unable to mirror HRgpAs bindi ng to fibrinogen, fibronectin, and laminin suggesting that the extra adhesin subunits of HRgpA were responsible for the observed binding to the above proteins (Ally et al. 2003; Pike et al. 1996). Also, when HRgpA, RgpAcat, and

PAGE 28

28 RgpB were tested for the ability to degrade fi brinogen, all three were able to cleave the A -chain but HRgpA, followed by RgpAcat and RgpB degraded the B -chain most proficiently. The extra adhesin domains of HRgpA likely contributed to the observed differences in the degradation of the B -chain of fibrinogen. Since both forms of R gpA were more efficient than RgpB, the active site differences may play a significant role (Ally et al. 2003). Gingipains share regions of homologous sequenc es with each other and have also been reported to have sequence homology with several P. gingivalis hemagglutinin activity genes (Potempa et al. 1995; Progulske-Fox et al. 1993). The proteolytic act ivities of Rgp and Kgp are necessary for hemagglutination (Shi et al., 1999) and the colonization and invasion of host tissues by P. gingivalis is at least partially dependent on RgpA and Kgp adhesin/hemagglutinin activity (Kelly et al. 1997). The requirement of RgpA and Kgp proteolytic activities for adhesion/hemagglutination are likely because th ey process the proproteins involved. RgpA and Kgp participate in the producti on of the 15 kDa adhesin/hema gglutinin domain, HA2, also referred to as the hemoglobin receptor (HbR), which participates in the binding of hemoglobin and the acquisition of heme from erythrocytes (Kelly et al. 1997). The activity of Kgp is vital for the discharge of HbR from the Ha gA protein and progingipains (Okamoto et al. 1998). The downstream processing of HA2 (HbR) is accomplishe d with the contribution of Kgp, since both HA2 processing and pigmentation we re severely hindered in the kgpmutant (Shi et al. 1999; Nakayama et al. 1998). The binding of hemoglobin also makes Kgp a significant player in the acquisition of iron (Travis et al. 1997). Fimbriae Fimbriae play a major role in P. gingivalis attachment and invasion (Njoroge et al ., 1997). Fimbriae also demonstrate a variet y of biological activit ies such as stimula ting the production of

PAGE 29

29 cytokines and bone resorption (Hamada et al. 1998; Yoshimura et al. 1984). Fimbriae were first identified in 1984 by Yoshimura et al. using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). There are currently two known types of fimbriae, major, discovered by Yoshimura et al and minor fimbriae, discove red independently by Ogawa et al in 1995 and Hamada et al. in 1996. The major subunit fimbrillin was determined to be 43 kDa in size and was recognized as the bacteriums firs t step in disease induction and development (Hamada et al. 1998; Yoshimura et al. 1984). The major fimbriae are composed of a subunit protein, fimbrillin (FimA), and function by mediating the bacteria l adhesion and the colonization of host tissue (Amano, 2003; Lamont and Yilmaz, 2002). The filamentous components of major fimbriae are up to 3 m long and 5nm in width, and are located on the cell surface. They are able to specifically bind to, enter, and activate a number of host cells including human endothelial and epithelial cells, periphe ral blood monocytes, fibroblasts, peri otoneal macrophages, THP-1 cells and spleen cells (Lamont and Yilmaz, 2002; Deshpande et al. 1998b). The minor fimbriae were determined to be 67 kDa and were uncovered by use of a fimA (major fimbria-deficient) mutant of wild-t ype ATCC 33277 and shown to possess short fimbrialike additions (Umemoto and Hamada, 2003; Hamada et al. 1996; Kokeguchi et al. 1994). These minor fimbriae range in length from 0.1 m to 0.5 m, width, ca. 6.5nm, and are distinct from the long, major fimbriae, both ge netically and antig enically (Hamada et al. 1996; Ogawa et al. 1995; Yoshimura et al. 1984). In vitro preparations of broken cells showed that the short fimbriae were located in both the membrane fractions and the soluble fractions (Park et al. 2005). The minor fimbriae are composed primarily of the subunit protein (Mfa1), which encodes the mfa1 gene. The Mfa1 molecule has been shown by Yoshimura et al ., (1989), to be

PAGE 30

30 equivalent to the 75 kDa outer membrane protein as detailed in several reports and Pg-II (a 72 kDa cell surface protein) in wild-type ATCC 33277 (Ogawa et al. 1995). Several researchers have proposed the possibi lity that more than one mechanism for fimA regulation exists and that both environmental cu es and endogenous factors are involved in its expression (Nishikawa et al. 2004; Xie et al. 2004). Inflamed subgingival pockets have a characteristically higher temperature (39oC) than normal gingival tissu e. As mentioned above, high temperatures influence fimA regulation by causing a decrease in both fimA mRNA and protein expression, while a lower temperature of (34oC), produced higher fimA transcriptional activity (Xie et al. 1997; Amano et al. 1994). Studies using standard growth conditions and the anti-fimbrillin antibody showed th at FimA from wild-type AT CC 33277 was present in both the prefimbrillin and mature fimbrillin protein forms and was produced throughout the growth cycle (Nishikawa et al. 2004). The maturation of the fimbrial 43kDa and 75kDa proteins were determined to be dependent on the precise cleavages of the amino (NH-2) terminal Arg46-Ala47 bond and the Arg49-Ala50 bond of the precursor proteins, respectively (Hamada et al. 1994; Ogawa et al. 1994; Watanabe et al. 1992; Lee et al. 1991; Dickinson et al. 1988). Using electron microscopy, Kadowaki et al showed that the Kgp-null muta nt and wild-type ATCC 33277 had numerous cell surface associated curly fimbriae a nd the Rgp-null mutant revealed little, if any, fimbriation, which suggest s that fimbriation of P. gingivalis is regulated by Rgp proprotein processing but not by Kgp (Kadowaki et al. 1998). In fact, the pr efimbrillin 45-kDa protein vanished and a 43-kDa protein emerged after trea tment with Rgp and this action was inhibited when the Rgp inhibitors leupeptin and EDTA were used, suggesting adhesion of P. gingivalis to host tissue is partially co ntrolled by Rgp (Kadowaki et al. 1998). To further c onfirm this theory,

PAGE 31

31 western blotting was conducted to compare the F imA levels of the 381 parent strain and G102, rgpAgingipain mutant. The G102 mutant is lik ely deficient in fimbrial expression (Tokuda et al. 1996) because the results revealed that the mu tant contained lower FimA levels than the 381 parent strain. The expressed FimA in the G 102 mutant was appropriate ly processed in size, however the mutant expressed lower fimA mRNA levels than the parent strain, thus suggesting a defect in the fimA gene transcription due to the lack of RgpA (Tokuda et al. 1996). The G102 ( rgpA) mutant also had an increased level of extracellular vesicles. Cultures of P. gingivalis grown in low levels of iron have been shown to induce such increases in surface blebs. The increase of these protease enriched extracellula r vesicles may be an effect of the amino acid deprivation caused from low iron stress (McKee et al. 1986). To further characterize major and minor fimbri ae with respect to bacterial virulence, the oral cavities of rats were inoculated with isogenic mutants of P. gingivalis ATCC 33277 including MPG1, ( fimA -), MPG67, ( mfa1 -) and MPG4167 ( fimA-, mfa1) (Umemoto and Hamada, 2003). The ATCC 33277 parent strain adhe red by auto-aggregation in clumps to the ATCC CCL17 (KBs) originally thought to be oral epithelial cells and now known as Hela cells, and the MPG67, which contained numerous long fi mbriae on its surface, formed even larger clumps than did the parent strain. The MPG1 mutant showed no apparent auto-aggregation and a decrease in adherence to singl e cells, while MPG4167, which lacked both fimbrial structures, lost adhesive abilities entirely. Th is data suggests that adherence of P. gingivalis to epithelial cells is partially dependent on the production of both the major and minor fimbriae. Invasion studies revealed that all three fimbriae-deficient mutants had a re duction in invasion in excess of eight times that of ATCC 33277. The results by Umemoto and Hamada also demonstrated that MPG1 caused a larger amount of oral bone loss as compared MPG 67 but considerably less than

PAGE 32

32 the periodontal bone loss induced by the wild-typ e strain. MPG4167s ab ility to induce oral bone loss was markedly suppressed, suggesting the expression of both major and minor fimbriae are necessary for the virulence associated with P. gingivalis in this model (Umemoto and Hamada, 2003). In agreement with MPG1, DPG3, a fimA(fimbrillin) mutant of strain 381, exhibited no apparent autoaggregation, lower binding to gr am-positive bacteria, a nd reduced binding to epithelial cells (Tokuda et al. 1996). From this data, fimbriae appear to also be required for autoaggregation. Since reduced autoaggrega tion would be mirrored by reduced adhesion, the fimbriae may be a vital participant in colonization in vivo (Tokuda et al. 1996; Genco et al. 1994). Also, another fimAmutant of ATCC 33277, YPF1, is unable to express the FimA protein due to an insertional inac tivation of the fimA gene (Yilmaz et al. 2003; Tokuda et al. 1996; Genco et al. 1994). Data from assays using YPF1 and gingival epithelial cells (GECs) revealed a 10-fold decrease in internalization, yet st ill retained some invasion potential and maintained non-fimbrial dependent interactions with GECs (Yilmaz et al. 2003). The levels of FimA were quickly decreas ed upon the interaction of P. gingivalis with GECs, suggesting that internalized wild-type P. gingivalis may be only sparingly fimbriated (Wang et al. 2002). Extended periods of treatment, e.g., 24 hours, rev ealed both the wild-type parent strain and the YPF1 mutant colocalized with paxillin and FAK, a common ou tcome of internalization, suggesting that the presence of fimbriae was not required (Yilmaz et al. 2003). Taken together, these data establish that th e FimA structural subunit protein contains specific peptide domains which are responsible for much of the a dherence activity of P. gingivalis (Lamont and Jenkinson, 2000). Wild-type P. gingivalis is capable of activating paxillin, a major player in the integrin-asso ciated cell signaling pathway, whereas the YPF1

PAGE 33

33 mutant was unable to induce this pathway (Yilmaz et al. 2002). Fimbriae-independent mechanisms of invasion may be important for P. gingivalis in addition to or when the fimbriae are downregulated (Yilmaz et al. 2002). Epithelial and Endothelial Cells The cells found lining body cavities, such as stomach, intestines, and skin are known as epithelial cells. Epithelial cells are polarized and their plasma membrane contains two separate regions, apical and basolateral that are made up of different sorti ng and transport proteins. Also, epithelial cells contain multiple specialized regions known as cell junctions, within their plasma membranes (Bazzoni and Dejana, 2004). These ce ll junctions function to seal off body cavities by acting as a barrier and restric ting the transport of plasma-membr ane molecules from the apical to the basolateral surfaces. In addition to thei r barrier function, cell junc tions provide rigidity and strength to the tissues. Ex amples of these epithelial cell junctions are tight junctions, adherens junctions, desmosomes a nd gap junctions, which function to maintain the integrity of the epithelium (Bazzoni and Dejana, 2004; Stevens et al. 2000). Tight junc tions are composed of both intracellular, zonula occludens (ZO1, ZO-2, ZO-3), and transmembrane molecules occludin and claudin (Bazzoni and Dejana, 2004; Morita et al. 1999; Furuse et al. 1993). The adherens junctions are composed of cadherins and catenins (Bazzoni a nd Dejana, 2004; Angst et al. 2001a; Angst et al. 2001b; Aberle et al. 1997; Takeichi, 1993). Although different molecules form the tight junctions and adherens junctions, both junctio ns function to promote homophilic interactions and form a pericellu lar zipper-like pattern along the cell boundary (Tsukita et al. 2001; Chitaev and Troyanovsky, 1998). Blood vessels primarily function to partition blood from underlying tissues. One of the main cellular components of blood vessels are endothelial cells. This sheet-like layer of endothelial cells lines the lumen of blood vessels and functions as a barrier, thereby controlling

PAGE 34

34 the influx of cells and blood proteins into the walls of the vasculatur e (Dejana, 2004; Lodish et al. 2000) This endothelial barrier function is also carried out by tight junctions and adherens junctions, which function to maintain the integr ity of the endothelium by defending the vessels from fluctuations in cell permeability and increases of inflammation and platelet aggregation, or thrombotic events (Stevens et al. 2000). Unlike epithelial cells where the tight junctions are primarily located on the apical su rface along the interce llular cleft, endothelia l cells contain tight junctions that are more integrated within the adherens junctions (Lodish et al. 2000). These endothelial junctional complexe s are highly defined but unlike the junctional complexes of epithelial cells, they not as organized and rigi d. Another difference between endothelial and epithelial cells is that huma n endothelial cells lack the ju nctional structures known as desmosomes (Lodish et al. 2000; Dejana et al. 1995). The intricate complex of transmembrane adhe sion proteins that make up the adherens junctions are known as cadherins. Cadherins function to maintain different aspects of vascular homeostasis by mediating cell-cell a dhesions and anchorage to the actin cytoskeleton via specific linkages to defined intracellular molecules, catenins (Dejana et al. 1999a). In addition, the proteins of the adherens junctions can modul ate endothelial cell growth, positioning, motility, and apoptosis through the transmission of intr acellular signals (Lampugnani and Dejana, 1997). This intracellular signaling can be direct by employing growth-factor r eceptors or signaling proteins, or indirect, by anchor ing and maintaining transcripti on factors at the cell membrane, thereby decreasing their transloc ation to the nucleus (Matter and Balda, 2003; Wheelock and Johnson, 2003; Braga, 2002; Bazzoni et al. 1999). Ultimately, adherens junctions play a role in regulating permeability of plasma solutes. They also regulate the opening and closing of their

PAGE 35

35 cell-to-cell linkages and in turn control the e fflux and influx of leukocytes into areas of inflammation (Muller, 2003; Johnson-Leger et al. 2000). Cell-Cell Adhesion Proteins Cell-cell adhesion is predomin ately arbitrated by the Ca2+ dependent, homophilic binding, classical cadherin family cont aining E-, N-, and Pcadherin as well as other members (Angst et al. 2001a; Ranscht, 1994). Endothe lial cells found within all vessels expre ss a specific cadherin called VE cadherin. VE-cadherin first becomes expressed during development when the cells become committed to become endothelial cells (Breier et al. 1996; Dejana et al. 1995). The cadherins consists of a highly conserved cy toplasmic domain and group of repeating Ca2+ binding subdomains (Kemler, 1992). The cadhe rin cytoplasmic domain connects to the cytoskeleton by interacting with plakoglobin, -catenin, -catenin, and p120 (Reynolds et al. 1994; Kemler, 1993; Kemler, 1992; Ozawa et al. 1989). Cadherin-mediated cell adhesion is dependent on the catenin family of cytoplasmi c polypeptides, since aggregation assays showed the removal of the catenin binding site within th e cytoplasmic tail of ca dherins eliminates the functionality of th e cadherins (Ozawa et al. 1989; Nagafuchi and Takeichi, 1988). In addition to the classical cadherins, desmosome-associated cadhe rins are a group of cadherins associated with tight junctions that form intracellular connecti ons with intermediate filaments instead of components of the actin cytoskeleton (Hynes, 1999). Widespread sequence similarities exist between -catenin and plakoglobin ( catenin) as well as between -catenin and the actin binding protein, vinculin (Butz et al. 1992; Herrenknecht et al. 1991; Nagafuchi et al. 1991). Plakoglobin communicates with the classical and desmosomal cadherins, while the -catenin interacts primarily with the cl assical cadherins of the adhere ns junctions (Gumbiner, 1995; Gumbiner and Yamada, 1995).

PAGE 36

36 Classical cadherins participate in the de velopment and protection of tissues during gastrulation, neurulation and organogenesis, and are important in the organization of Xenopus embryos (Aberle et al. 1996; Redies and Takeichi, 1996; Ta keichi, 1995). Many routes of epithelial development, growth and phenotype are influenced by cadherin mediated adhesion (Marrs et al. 1995). The formation of invasive and metast atic tumors is associated with the loss of E-cadherin, thus implying that E-cadherin is crucial for the development and maintenance of the epithelia (Bracke et al. 1996; Larue et al. 1994). Intestin al cells containing overexpressed Eor N-cadherin displayed alterations in thei r proliferation, migrati on, and observed apoptosis (Hermiston et al. 1996; Hermiston and Gordon, 1995a ; Hermiston and Gordon, 1995b). Through contact-induced inhibition, cadherins have also been sugge sted to be responsible for reduced cell growth. This reduction in cell grow th occurs through cell-cy cle arrest at the G1 phase and is at least partiall y due to a decrease in the level of cyclin-D1 (Venkiteswaran et al. 2002; Gottardi et al. 2001; Mueller et al. 2000; St Croix et al. 1998). Cadherin could limit cell growth by occupying -catenin at the cell membrane thereby limiting -catenin translocation to the nucleus where it would ultimate ly upregulate the transcription of cyclin D1 (van de Wetering et al. 2002; Gottardi et al. 2001; Ben-Ze'ev and Geiger, 1998) Collectively, cadherins are involved in signaling pathways that control adhesion and may individually launch regulation signals for central cellular proc esses such as proliferation, mi gration, cell differentiation, and apoptosis (Bracke et al. 1996; Marrs et al. 1995; Larue et al. 1994; McCrea et al. 1991). P. gingivalis has been shown to exhibit a highly elev ated amount of proteolytic activity for the proteins of the epithelial junctions. For example, P. gingivalis gingipains have been shown to degrade E-cadherin in experi ments using MDCK cells (Katz et al. 2002; Katz et al. 2000). Loss of cell adhesion favoring bact erial invasion and inf ection of the underlying tissues could be

PAGE 37

37 a consequence of the loss of E-cadheri n molecules from epithelial cells (Katz et al. 2002; Gottardi et al. 2001). In addition, the loss of E-cadheri n induces cell proliferation by prompting a change in gene expression through -catenin-LEF/TCF upregulation (Sadot et al. 1998; Simcha et al. 1998). P. gingivalis has also been shown to exhibit a hi ghly elevated amount of proteolytic activity for the proteins of the endothelial cell ju nctions. Sheets et al found that the gingipains downregulated the endothelial intercellular juncti onal cadherin in bovine coronary artery endothelial cells (BCAEC) and human microvascular endothelia l cells (HMVEC) leading to increased vascular permeability and apoptosis (Sheets et al. 2005) This increased permeability is caused by a disruption in the endothelial junctional barrier, specifically the severing of the linkage between cadherin and -catenin within the adherens j unctions, and has also been shown to increase inflammation by allowing the migration of leukocytes (Yun et al. 2005; Hordijk, 2003; Schenkel et al. 2002; Muller et al. 1993) A transient bacteremia and subsequent systemic dissemination of P. gingivalis after dental treatments or flossing could potentially make endothelial cells prospective targets of P. gingivalis -catenin is a proto-oncogene and vertebrate homolog to armadillo in Drosophila (McCrea et al. 1991). In addition to directly linking cadherin to the actin cytoskeleton, -catenin is also a signal transduction molecule th at is known to mediate signa ling through the Wnt signaling pathway (Moon et al., 2002). Wnts are a family of highly conserved signaling molecules that are essential for the maintenance of adult tissues (Goichberg et al., 2001). Normally, when the Wnt ligand is absent, the cytosolic -catenin (not bound to cadherin at the membrane) forms a complex with the adenomatous polyposis coli (APC ) protein and is phosphorylated and targeted for degradation by an assembly of proteins, ensuring low -catenin levels. In contrast, when the

PAGE 38

38 Wnt ligand is present, Wnt signaling occurs th rough its receptor frizzled and inhibits the degradation of -catenin by inhibiting its phosphorylation. This inhibition of -catenin degradation allows -catenin to build up in the cytosol wh ere it can then enter the nucleus and interact with the transcription factor T-cell factor (TCF) or le ukocyte enhancing factor (LEF) and induce the expression of target genes (Liebner et al. 2006; Gumbiner, 2005; Bazzoni and Dejana, 2004). This Wnt signaling enables cells within multicellular organisms to communicate with each other in order to organize a significa nt array of cellular processes such as cell performance and migration during morphogenesis, cell propagation versus delineation, and cell survival versus apoptosis (Logan and Nusse, 2004). -catenin is stabilized and ma intained at the cell membrane when it is bound to cadherin. Both the Wnt/Wg pathway and the cadherin-mediated cell adhesion complex compete for the availability of -catenin. Overexpression of cadherins represses the signaling associated with catenin suggesting that the signaling actions of -catenin are physiologically controlled by cadherins (Heasman et al. 1994). If there is a mutation or d ecrease in cadherin expression, the pool of free -catenin may increase, thereby increasing -catenin signaling, possibly leading to endothelial cell growth a nd transformation (Gottardi et al. 2001; St Croix et al. 1998). For example, if the Xenopus embryo overexpresse s cadherin, the cytosolic and/or nuclear -catenin pools are directed to the plasma membrane, thereby causing a parade of developmental anomalies due to the lack of nuclear communication with -catenin (Fagotto and Gumbiner, 1996; Heasman et al. 1994). Mutations of -catenin have been linked with malignant cell transformations and truncation of -catenin in a carcinoma cell line has been shown to induce cadherin dysfunction (Polakis, 2000; Kawanishi et al. 1994). An embryonic lethal phenotype caused by an endothelial-specific -catenin deletion in mice and the phenotype of an

PAGE 39

39 adenomatous polyposis coli (APC) mutant in zebrafish demonstrates that -catenin signaling participates in endothelial tran sformations into other cell linea ges such as mesenchymal cells during development (Liebner et al. 2004; Cattelino et al. 2003; Hurlstone et al. 2003). In summary, -catenin can be found bound to cadherins at the cell membrane where it acts to physically link cadherin to the actin cytoskeleton (Juliano, 2002). -catenin is also responsible for intracellular signaling. For exam ple, cell growth and apoptosis are controlled through the signaling at these inte rcellular contacts. Adherens junction proteins also regulate vascular permeability by controlling the passage of solutes and circulating cells. Some signals may be necessary for endothelial stability while others may primarily function to maintain cellular permeability or the passage of leukocytes through the endothelial junc tions. It is also possible that the junctional prot eins indirectly modulate cell f unction by communicating directly with growth factor receptors (Liebner et al. 2006; Logan and Nusse, 2004). Collectively, the proteins of the intercellular ad herens junctions are responsible for cell-to-cell adhesion and intracellular signaling. Cell-Matrix Adhesion Proteins Animal epithelia and organized assemblies of cells, e.g., muscle tissue, are either surrounded by or underlined with co mponents of the extracellular ma trix, such as collagen fibers, proteoglycans, and adhesive matrix proteins. Functions of the extracellular matrix (ECM) include the organization and coordi nation of tissues, routes for migrating cells, and stimulation of cell growth, proliferation, a nd gene expression through classi c signal-transduction pathways (Haynes and Webb, 1992). Members of the adhesi on receptor family include: selectins, the immunoglobulin cell adhesion superfamily (IgCAMs) a nd integrins. The selectin family, L-, E-, and P-selectin, are composed of lectin-like adhesion receptors that bind carbohydrate moieties on

PAGE 40

40 mucin-like CAMs and negotiate calcium-dependent, heterotypi c cell-cell interactions. Pselectins are physiologically impor tant during the inflammatory pr ocess because they participate in the tightly regulated adheren ce of leukocytes and platelets to endothelial cells (Springer, 1995). Selectins, along with their glycoprotein counterreceptors, regulate the trafficking of leukocytes between the bloodstream and tissues and are involved in the activation of 2 integrins (Hu et al. 2000; Hartwell and Wagner, 1999; Ve stweber and Blanks, 1999; Lorenzon et al. 1998). IgCAMs are adhesive receptors th at play very important role s in development (Murase and Schuman, 1999). Some examples of IgCAMs include NCAM, ICAM, and PECAM-1. NCAMs function as calcium-independent, homotypic, cel l-cell adhesion receptors, while ICAM functions by directly linking specific memb rane receptors to the actin cy toskeleton (Crossin and Krushel, 2000; Bretscher, 1999). PECAM-1 is a homotypic receptor found on platelets, endothelial cells, and various leukocytes, and engage s in the development of endothe lial junctions and the release of leukocytes (Aplin et al. 1998; Newman, 1997). PECAM-1 al so has the abil ity to recruit phosphatases that potentially coul d counteract the tyrosine kinase s recruited by other cell surface receptors (Newman, 1999). The ligation of PECAM-1 can also stimulate the v 3 integrin demonstrating this IgCAM is also proficient in si gnal transduction. The P. gingivalis major fimbria interacts with the host cell and causes the induction of expression of vascular cell adhe sion molecule 1 (VCAM-1), inte rcellular adhesion molecule-1 (ICAM-1) and Eand Pselectins (Khlgatian et al. 2002; Nassar et al. 2002). Confocal microscopy and FACS analysis determined that P. gingivalis A7436 infection of HUVECs induced VCAM-1, ICAM-1, Pand Eselectin expression (Khlgatian et al. 2002). Evidence to support the involvement of P. gingivalis invasion with adhesion molecule expression was

PAGE 41

41 obtained with the use of cytochalasin D, an actin polymerization inhibitor. P. gingivalis invasion of HUVEC was prevented with the addi tion of 1ug/ml cytochalasin D (Deshpande et al. 1998b). Furthermore, pretreatment of HUVECs with cytochalasin D, preceding infection with P. gingivalis resulted in a considerable reduction of ICAM-1 and VCAM-1 expression. The stimulation of these cell adhesion molecules was later determined to be a direct outcome of major fimbriae-mediated attachment (Khlgatian et al. 2002). Further evidence singling out the importance of bacterial attachment was c onfirmed when the use of DPG3, a nonadhering fimA deficient strain, and antibodies against the fimbrillin peptide, failed to yield stimulation of these specific adhesion molecules (Khlgatian et al. 2002). Integrins are chiefly involved in cell-matrix interactions and bind to fibronectin, laminin, collagen and various other matrix proteins to form intricate netw orks called focal adhesions or focal contacts. Collagens are composed of many diverse proteins and are also the most profuse structural components making up the extracellula r matrix within all tissues (Gumbiner and Yamada, 1995). Integrins are heterodimers com posed of noncovalently associated subunits ( and ) and contain a large extracellular domain, short cytoplasmic domain, and a single membrane spanning region that undergo c onformational domain changes or subunit rearrangements during the ligand bi nding process (Hynes, 1999; Aplin et al. 1998). The organization of the cytoskeleton, signal transduc tion, and cell motility are examples of integrin functions carried out, in part, by the cytoplasmic domains of the and subunits (Juliano, 2002). Integrins have also been shown to directly activate aspects of intr acellular signaling. The gathering of integrins bound to groups of radicals, ions, or mo lecules (ligands ), can direct various structural and signaling constituents and stimulate essential signal transduction cascades

PAGE 42

42 such as the mitogen-activated MAP kinases (Apl in and Juliano, 1999; Giancotti and Ruoslahti, 1999; Aplin et al. 1998). Previous reports have shown that P. gingivalis has the ability to direct their entry into both endothelial and epithelial cells by utiliz ing the host cells signa ling pathways (Yilmaz et al. 2002; Deshpande et al. 1998b; Lamont et al. 1995). Periodontal ti ssues infected with periodontal pathogens are repaired and maintained in part by th e interactions between the ECM proteins and the host cells (Kapila et al. 1998; Rautemaa and Meri, 1996). Signals overseeing cell function, migration, and anchorage abilit ies are provided by the adhesion interaction between tissue cells and the proteoglycans a nd/or the integrins of (ECM) proteins (Kapila et al. 1998). Integrin mediated adherence of GEC is commonly associated with temporary, low levels of cytoskeletal rearrangements (Young et al. 1992). The invasion dependent trigger necessary for the reorganization of microtubules as s een for other bacteria like urophathogenic E. coli, is supplied by the activation of inte grin and it is possible that P. gingivalis is also equipped with this mechanism (Guignot et al. 2001). P. gingivalis Invasion of Epithelial and Endothelial Cells In response to bacterial challenge, macrophages, neutrophils and dendri tic cells engulf, or phagocytose bacteria and internalize them into pha gosomes which later transport the bacteria to the lysosome for degradation (Stossel et al. 1999). Several highly pa thogenic pathogens are capable of circumventing this process, thus leading to disease even ts (Pieters, 2001). The concept of bacteria possessing th e ability to invade gingival ti ssue was first proposed by Goadby in the early twentieth century (Goadby, 1907). Bacterial uptake is now known to occur by the utilization of various pathways depending on the b acterial strain and cell type involved (Lamont and Yilmaz, 2002; Lamont and Jenkinson, 2000). Bacterial invasion of host cells is likely stimulated by specific signal transduction path ways given that recent reports have provided

PAGE 43

43 evidence to support the involvement of protein kinases (Collaco et al. 1995; Rosenshine et al. 1994; Rosenshine et al. 1992). Other investigators have reported that a kinase-mediated phosphorylation of a specific 68-kDa protein (p p68) in mouse peritoneal macrophages was inducible by P. gingivalis fimbriae and this inducible activity was repressed after the use of staurosporine, a specific inhibitor to protein kinase C (PKC) (Murakami et al. 1994). P. gingivalis utilizes its major fimbriae, FimA, fo r adherence to KB cells and GECs and appear to also require gingip ain activity for optimal invasion (Lamont and Yilmaz, 2002; Chen et al. 2001b; Lamont et al. 1995), since treatment of these cells with a combination of various protease inhibitors: aprotinin, le upeptin, pepstatin and benzamidin e, drastically reduced invasion (Lamont et al. 1995). KB and GEC cells accomplish bacterial uptake through phagocytosis, allowing the bacteria to be e ither restricted to a membra ne bound vacuole or free in the cytoplasm (Houalet-Jeanne et al. 2001; Njoroge et al. 1997; Sandros et al. 1996; Sandros et al. 1993). In addition, Legionella pneumophilia and Brucella abortus utilize lipid rafts for entrance into the host cells and it has been suggested that P. gingivalis also utilizes this mechanism (Belanger et al. 2006; Tsuda et al. 2005; Duncan et al. 2004; Watarai, 2004; Duncan et al. 2002) Also, several bacteria, for example Yersinia enterocolitica, Salmonella choleraseuis, Actinobacillus actinomycetemcomitans and Shigella flexneri, depend on the action of microfilaments for invasion (Finlay and Falkow, 1997; Sreenivasan et al. 1993; Finlay and Falkow, 1989), while the action of both the micr ofilaments and microtubu les are required for Neisseria gonorrhoeae Citrobacter freundii and Haemophilus influenzae invasion (Oelschlaeger et al. 1993; Donnenberg et al. 1990; Richardson and Sadoff, 1988). Cytoskeletal rearrangements were also determined to be important for the invasion GEC by P. gingivalis since nocodazole, a microtubule assembly inhibitor, an d cytochalasin D, an actin polymerization

PAGE 44

44 inhibitor, severely retard ed invasion (Houalet-Jeanne et al. 2001; Lamont et al. 1995). Further evidence to support this invasi on inhibition was provided by elec tron microscopy, which showed a lack of internalized P. gingivalis present within the nocodazole and cytochalasin D treated GECs (Houalet-Jeanne et al. 2001; Lamont et al. 1995). Deshpande et al and Dorn et al have demonstrated that P. gingivalis is also able to actively invade and replicate within endotheli al cells (ECs). Usi ng an MOI of 1:100 of P. gingivalis strains A7436 and 381 and bovine aortic endothelial cells (BAEC), Deshpande et al determined that the greatest invasion was observed after 2 hours with no additional increase in invasion reported up to four hours (Deshpande et al. 1998b; Deshpande et al. 1998a; Njoroge et al. 1997). Scanning electron microsc opy showed the attachment of P. gingivalis wild-type strain A7436 to BAEC was mirrore d by alterations in the normal architecture of the EC surface and the presence of protruding microvilli from the EC engulfing the attached bacteria. This EC engulfment is likely the stimulus for the preliminar y formation of a cytoplasmic vacuole. In fact, BAEC, human umbilical endotheli al cells (HUVEC), and fetal bovi ne heart endothelial cells (FBHEC), all appeared to contain P. gingivalis A7436 within apparent cytoplasmic vacuoles (Deshpande et al. 1998b; Deshpande et al. 1998a; Njoroge et al. 1997). Further studies investigating the metabolic requireme nts for invasion showed that the de novo protein synthesis inhibitors, chloramphenicol and cycloheximide, a nd nalidixic acid and rifampin, inhibitors of bacterial DNA and RNA synthesis, were una ble to reduce the invasion of BAEC by P. gingivalis (Deshpande et al. 1998b; Deshpande et al. 1998a; Njoroge et al. 1997). However, in agreement with data from GECs, invasion was si gnificantly reduced by the use of cytochalasin D and nocodazole (Lamont et al. 1995). These results indicate that bacterial and BAEC protein synthesis and bacterial DNA and RNA synthesis are not required for P. gingivalis invasion of

PAGE 45

45 endothelial cells. However, th e role of actin polymerization and microtubule formation are important for invasion in both cell types (Deshpande et al. 1998b; Lamont et al. 1995). In addition to cytoskeletal reorga nization, protein phosphor ylation is also likel y involved since the use of the broad spectrum protein kinase inhibitor, staurosporine, inhibited P. gingivalis invasion of ECs (Deshpande et al. 1998b; Lamont et al. 1995). Sodium azide an inhibitor of cytochrome oxidase and proton motive forces, also repressed invasion providing evidence for the involvement of both P. gingivalis and EC energy metabolism (Deshpande et al. 1998b; Deshpande et al. 1998a). It appears that adherence to and invasion of ECs is also partially dependent on the expression of the major fimbriae (Dorn et al. 2000; Dorn et al. 1999; Deshpande et al. 1998b; Deshpande et al. 1998a). The fimA DPG3 mutant was unable to a dhere to BAEC or FBHEC or induce changes to the normal endothelial cell surf ace architecture, providing further evidence to support the pivotal role for major fimbriae in invasion of endotheli al cells (Deshpande et al. 1998b; Deshpande et al. 1998a; Njoroge et al. 1997). The major fimbriae may be responsible for generating a cascade of events leading to the long, microvilli characteristic of cytoskeletal reorganization of the ECs (Deshpande et al. 1998b; Deshpande et al. 1998a). Bacterial attachment to heart and aortic endothelial cells (H CAEC) most likely requires the action of additional adhesins because the major fimbriae alone are not sufficient for optimal invasion (Dorn et al. 2000; Dorn et al. 1999; Deshpande et al. 1998a). Like GECs, cytoskeletal rearrangements are necessary for P. gingivalis invasion of the HCAEC, although once internalized, the bacteria appear to be localized within multimembranous, autophagasome like vacuoles instead of within the cytoplas m as observed with epithelial cells (Dorn et al. 2001; Dorn et al. 1999).

PAGE 46

46 Autophagosomes are multimembranous vacuoles formed by invaginations of the rough endoplasmic reticulum (RER) that are devoi d of ribosomes (Dunn, 1994). Specifically, autophagosomes are discrete stru ctures involved in autophagy, a cellular response to nutrient deprivation whereby the organelles and cytoso lic components are gathered and degraded by lysosomes (Dunn, 1990). Under normal conditions, the autophagosome fuses with a lysosome, forming a hydrolase filled autolysosome capable of degrading cellular co ntents. Early after P. gingivalis invasion, the vac uoles containing P. gingivalis contain Rab5 and HsAtg7, both markers of early endosomes. Subsequent to this, P. gingivalis traffics to a late autophagosome deficient of cathepsin L, but containing both the lysosomal protein LGP120, and the endoplasmic reticulum protein, BIP (Dorn et al. 2001). Other pathogens evad e the host immune system through a similar process. For example, pathogens such as Legionella pneumophilia and Brucella abortus localize and replicate within vacuoles containing proteins of the endoplasmic reticulum (Pizarro-Cerda et al. 1998b; Pizarro-Cerda et al. 1998a). Like L. pneumophilia and B. abortus the autophagic like vacuoles containing P. gingivalis do not mature into a autolysosome containing lysosomal hydrolases (Dorn et al. 2001; Pizarro-Cerda et al. 1998b; Pizarro-Cerda et al. 1998a; Swanson and Isberg, 1995). P. gingivalis appears to have refined its survival within endothelial cells by adapting a mechanism that enables its evasion from lysosomes of the endocytic pathway, and tra fficking to the autophagocytic pathway (Dorn et al. 2001). Bacterial entry into the host cell would provide a protected niche and a generous supply of amino acids for metabolic energy, as well as es sential nutrients such as iron, which is required for P. gingivalis growth and survival (O'Brien-Simpson et al. 2000). Bacterial invasion of HCAEC could result in chronic injury to endothe lial layer of the arteri al wall, thus either promoting the initiation or the progressi on of atherosclerotic development (Dorn et al. 1999)

PAGE 47

47 P. gingivalis Persistence in Epithelial and Endothelial Cells P. gingivalis is able to replicate and persist with in a number of epithelial and endothelial cell lines (Dorn et al. 2000; Dorn et al. 1999; Deshpande et al. 1998a). It undergoes quick and efficient internalization (15 minutes ) within GECs. Invasion of GEC by P. gingivalis ATCC 33277 was determined to be optimal when P. gingivalis was grown to the mid-log phase, followed by the early stationary phase and the late stationary phase (Lamont et al. 1995). High numbers of the active an d viable intracellular P. gingivalis accumulated perinuclearly (HoualetJeanne et al. 2001; Belton et al. 1999). In GECs, this organism is not restricted to a vacuole and is capable of replicating in tracellularly, which suggests that P. gingivalis has adapted a mechanism for survival within these cells (Lamont and Jenkinson, 1998; Lamont et al. 1995). In fact, it has been reported that P. gingivalis remains viable within these cells with no appearance of necrosis or apopt otic cell death for extended pe riods (Lamont and Yilmaz, 2002; Houalet-Jeanne et al. 2001; Belton et al. 1999). In addition, P. gingivalis can be found both within single-membrane vacuoles or free in the cytosol of KB cells. However, KB cells have been shown to undergo cell death after exposure to P. gingivalis (Wang et al ., 1999; Chen et al., 2001) but, Madianos et al reported that P. gingivalis is capable of surviving within KB cells for up to eight days (Madianos et al. 1996). The fact that P. gingivalis survives for extended periods of time within these epithe lial cells suggests that the host cells are capable of adapting to this bacterial challenge of P. gingivalis and/or P. gingivalis modulates its intracellular behavior so as not to profoundly affect its host cell. This persisting in ternal survival of P. gingivalis within these cells provides a possible expl anation for the dormancy and aggravating characteristics associated with periodontal disease (Weinberg et al. 1997). P. gingivalis may, in fact, be capable of regulating the expression of both genes and proteins in relation to their existing epithelial cell surroundings (Zhang et al. 2005; Nelson et al. 2003).

PAGE 48

48 Many pathogens have also acquired mechanisms for survival within endothelial cells. Brucella abortus and Legionella pneumophilia have been shown to tra ffic intracellularly so to achieve entry into multimembranous vacuoles that resembles autophagosomes. These bacteria are capable of replica ting and persisting within these au tophagosome like vacuoles (Dunn, 1994; (Meresse et al. 1999); Pizarro-Cerda et al., 1998a; Pizarro-Cerda et al ., 1998b). In Human Coronary Artery Endothe lial cells (HCAE) cells, P. gingivalis has also been shown to be contained within vacuoles resembling autopha gosomes and has been shown to stimulate autophagy (Dorn et al. 1999). The stimulation of autopha gy may promote the survival of P. gingivalis by supplying a pool of free amino acids, which could be consumed for their own metabolism or to inhibit host prot ein synthesis (Sinai and Joiner, 1997). In addition to promoting cell survival, autophagy has been shown to be important for cell death (Codogno and Meijer, 2005). Therefore, the stimulation of autophagy may be a requirement for cellular homeostasis since it is essential for cell survival as well as cell damage and cell death (Codogno and Meijer, 2005; Shintani and Klionsky, 2004) Apoptosis of Epithelial and Endothelial Cells In vitro studies have shown that the gingipains of P. gingivalis are capable of inducing cell death in several cell types including epithelial ce lls, fibroblasts, and e ndothelial cells (Baba et al. 2001; Graves et al. 2001; Wang et al. 1999; Johansson and Kalfas, 1998; Graves and Jiang, 1995; Morioka et al. 1993; Shah et al. 1992). A physiological t ype of cell death called programmed cell death, or apoptosis, has numerou s biochemical episodes and characteristic, physical cellular outcomes. Periodo ntal lesions contain these apopt otic cells and cells containing the apoptotic regulating molecule p53, (Gamonal et al. 2001; Koulouri et al. 1999; Sawa et al. 1999; Tonetti et al. 1998) mirrored by a reduction in cells positive for the apoptotic suppression molecule, Bcl-2 (Sawa et al. 1999; Tonetti et al. 1998). Chen et al reported that KB cells

PAGE 49

49 treated with gingipain-active W 83 extracts underwent apoptosis s ubsequent to cleavage of the neural cadherin (N-ca dherin) and integrin 1 induced loss of adhesion to the culture surfaces (Chen et al. 2001a). W83 extract pretreated w ith the Rgp and Kgp inhibitor, N -p-tosyl-L-lysine chloromethylketone (TLCK) failed to demonstrat e any considerable rounding or detachment of the KB cells in contrast to cells treated with the extract in the absence of the inhibitor (Chen et al. 2001a; Abaibou et al. 2000; Pike et al. 1996; Fletcher et al. 1994). In addition to gingipains, P. gingivalis promotes MMP secretion from macrophages which can promote detachment from the cell surface and ma y subsequently induce anoikis (Grayson et al. 2003; Kuramitsu, 1998; DeCarlo and Harber, 1997; Arribas et al. 1996; Ding et al. 1995; Ehlers and Riordan, 1991). The P. gingivalis induced MMPs are capable of both activating cadherin and cleaving a number of cell surface proteins includ ing growth factors and cytokine precursors (Grayson et al. 2003; Kuramitsu, 1998; Arribas et al. 1996). In contrast to the apoptosis observed with gingipains and MMPs, many investigators have reported that epithelial cells with internalized P. gingivalis experience no apparent negative side effects (Nakhjiri et al. 2001; Katz et al. 2000; Fives-Taylor et al. 1999; Madianos et al. 1996). Therefore, P. gingivalis may also act to inhibit ce ll death. The interaction of P. gingivalis with integrin may actually alter the natu rally occurring rate of apoptosis. Support for this theory was provided by the reports from Yilmaz et al ., which showed that the effects of the apoptosis inducer, campothecin, were blocked by P. gingivalis invasion of GEC (Yilmaz et al. 2002; Nakhjiri et al. 2001). These short-lived, primary gingival ep ithelial cells were able to withstand a 24 hour exposure to high levels of P. gingivalis (Nakhjiri et al. 2001; Belton et al. 1999). In fact, the levels of the anti-a poptotic Bcl-2 mRNA and protein were upregulated and BAX, a characteristic early step in the apopt otic pathway, was downregulated after P. gingivalis invasion

PAGE 50

50 of GEC (Yilmaz et al. 2002; Nakhjiri et al. 2001). Other investigat ors have shown the swift cytosolic to mitochondrial translocation of a pr o-apoptotic molecule, BAX, after the loss of integrin adhesion in primary mouse mammary cells isolated from pregnant ICR mice (Gilmore et al. 2000). Taken together, the upregulation of Bcl-2 and the down regulation of Bax likely inhibits apoptosis in P. gingivalis infected GEC, thus promoti ng the life of the cell (Nakhjiri et al. 2001). The role of P. gingivalis in endothelial cell apoptosis may also directly involve the gingipains. Endothelial cells of connective tissue and all vess el types continually express integrin 1 (Mechtersheimer et al. 1994; Hormia et al. 1990). It has been suggested that the P. gingivalis gingipains participate in e ndothelial apoptosis/anoikis by disrupting integrin signaling through the rapid degrad ation of integrin 1 (Aoudjit and Vuori, 2001; Matter and Ruoslahti, 2001; Oguey et al. 2000; Lin et al. 1997). The P. gingivalis gingipains have also been shown to directly cleave the adherens j unction protein, VE-cadherin (Sheets et al. 2005). As mentioned previously, VE-cadherin is conne cted to the actin cytoskeleton through the direct interactions with -catenin, plakoglobi n and p120 (Dejana et al. 1999b). VE-cadherin is specific to endothelial cells and functions by forming an intracellular seal within the junctional complex. Thus, the cleavage of CAMS by the P. gingivalis gingipains may induce apoptosis and hence cause damage to the tissues by disrupting the inte grity of the endothelial monolayer. In addition, endothelial cells maintain vasc ular homeostasis by participat ing in a number of vascular processes such as the regulation of vascular permeability, angiogenesis and blood pressure (Rubanyi, 1993). The disturbance of the co mmon endothelial functi ons may modify the endothelial state from healthy to diseased and as a result lead to an influx of inflammatory molecules characteristic of atherosclerosis (Takahashi et al. 2006). Thus, in addition to the

PAGE 51

51 reported gingipain associated damage to the tissues within the pe riodontal pocket (Curtis et al. 1999a; Curtis et al. 1999b), gingipain activity with endotheli al cells may have a significant role in causing the tissue damage associated with cardiovascular disease (Sheets et al. 2005). Recent studies by Sheets et al have shown that Human micr ovascular endothelial cells (HMVECs) and bovine coronary artery endothelial cells (BCAEC) treated with gingipain-active extracts undergo apoptosis after rounding and detaching from the culture surface (Sheets et al. 2005). However, these endotheli al cells exhibited minimal r ounding and detachment from the culture dish and minimal apoptosis in the presence of TLCK or the recA defective mutant which has considerably less gingipain activity comp ared to its parent strain W83 (Sheets et al. 2005; Abaibou et al. 2000). Interestingly, the HMVEC were mo re sensitive to the W83 extract than were the BCAECs, indicating that either the TLCK was more toxic to these cells than the BCAECs or these cells were more stressed during culturing (Sheets et al. 2005). The endothelial cell responses to stimuli may be ligand and/or cell-type specific (Marschang et al. 2006; Takahashi et al. 2006). For example, venous and arte rial endothelial cells have been reported to have diverse gene expression patterns and therefor e likely respond differently to stimuli (Takahashi et al. 2006). In addition, BCAECs treated with the gingipain-active W83 extrac ts have been reported to have an increased level of caspase 3 activity (Sheets et al. 2005). This increase in caspase 3 activity was strongly inhibited in the presence of TLCK, suggesting th at this activity was induced by the P. gingivalis gingipains (Sheets et al. 2005). Thus, these resu lts suggest that the gingipains are directly responsib le for the observed rounding, detachment, and subsequent cell death of BCAEC endothelial cells. Taken together, this evidence combined with the accelerated atheroma development induced by P. gingivalis invasion of aortic tissu e within the ApoE -/-

PAGE 52

52 mouse, suggests that the gingipa ins may stimulate endothelial dys function, leading to apoptosis within the cells of the periodontal pocket and the cardiovascular system (Sheets et al. 2005; Gibson et al. 2004). Chapter Summary P. gingivalis is equipped with an ar ray of recognized virulence factors capable of causing disease (Cutler et al. 1995; Genco, 1992a; Genco, 1992b). Examples of these virulence factors include fimbriae and lectin-type adhesions, a polysaccharide capsule, hemolysins and hemagglutinins, toxic metabolic products, outer membrane vesicles (b lebs), and cysteine proteinases (gingipains) (Cutler et al. 1995). The LPS from P. gingivalis can cause macrophages, fibroblasts, and monocytes to release cytokines such as IL-1, TNF, and/or Creactive proteins (CRP) (Darveau et al. 1998; Valtonen, 1991). P. gingivalis expresses cysteine proteinases or gingipains, called Rgp and Kgp, as well as PepO, an endopeptidase that participates in the degradation of prot eins, peptides, and glycopeptides (Ansai et al. 2003). P. gingivalis gingipains Rgp and Kgp cleave natural and s ynthetic substrates fo llowing arginine and lysine residues, re spectively (Awano et al. 1999). Two related gene s encode the argininespecific gingipain while only one gene encodes the lysine-specifi c gingipain. A catalytic and hemagglutinin domain and a propeptide are encoded in rgpA and kgp (Genco et al. 1999; Genco CA, 1999). Both Rgp and Kgp likely play a pivo tal role in modulating host immune defenses, acquisition of nutrients, and tissue invasion (Genco et al. 1999; Lamont and Jenkinson, 1998). These enzymes can cause potentially damaging activities by degradi ng host cell adhesion molecules and proteinase inhib itors, modifying the antimicrobial activity of neutrophils, and degrading bactericidal proteins immunoglobins, and iron transpor ting proteins. All of these properties signify that the gingipains of P. gingivalis can account for much of the clinical

PAGE 53

53 manifestations seen in periodontitis such as elev ated levels of gingival crevicular fluid, bleeding, and oral bone loss (Curtis et al. 1999a; Mayrand and Holt, 1988). In addition, these enzymes can stimulate serotonin secreting platelets, which could potentially give ri se to cardiovascular impediments because an increase in serotonin le ads to an imbalance between the nitric oxide (NO)/serotonin levels, thus prom oting endothelial dysfunction (Curtis et al. 1999a). Gingipain induced cleavage and /or degradation of adhesi on molecules may provoke detrimental damage to the endothelial monolayer. Gingipain extracts have been shown to i nduce the detachment of Human Microvascular Endothelial Cells (HMVEC) from the monolayer in and this detachment was inhibited by the presence of the cysteine proteinase inhibitor TCLK (Kobayashi-Sakamoto et al. 2006). Previous data have shown that gingipains can faci litate the cleavage of a number of cell surface proteins such as growth factors, cytokine precursors and recept ors, and cell adhesion molecules, including integrin 1 and cadherins, by proteolytica lly activating host MMPs (Herren et al. 1998; Arribas et al. 1996; McGeehan et al. 1994). The cleavage of th e cell adhesion molecules would allow the cells to detach from one anot her, causing disruption of the monolayer (Garton et al. 2003; Herren et al. 1998; Arribas et al. 1996; McGeehan et al. 1994; Ehlers and Riordan, 1991). Gingipain active extracts have also been shown to induce apoptosis in KB epithelial cells and the endothelial cell lines HMVEC and BCAEC (Kobayashi-Sakamoto et al. 2006; Sheets et al. 2005; Chen et al. 2001b). Therefore, the gingipains may participate in vascular tissue damage by degrading a number of adhesion mol ecules, leading to increased endothelial cell detachment and cell death (Sheets et al. 2005; Bazzoni and Dejana, 2001; Dejana et al. 2001).

PAGE 54

54 Objectives P. gingivalis has been identified directly from di seased tissue and support for a role of infectious bacteria, e.g., P. gingivalis with the initiation of athe rosclerotic plaques has been provided (Garcia et al. 2001; Haraszthy et al. 2000). Specifically, the proteolytic ac tivities of the P. gingivalis gingipains and the inducti on of inflammatory molecules by the gingipains have been suggested to play a significant role in the development of atheroscle rotic events. However, definitive proof of the direct involvement of P. gingivalis in the development and progression of CVD is still lacking. The concept of whether the gingipains are primarily acting from the exterior of the cell or perhaps, instead from the interior of the cell, has not been confirmed. In fact, P. gingivalis gingipains may have multiple and different functions depending on their location. For example, the intrace llular bacteria may either promot e cell survival or induce cell death, while extracellular bacteria may degrade CAMs in order to pe netrate the tissue, infiltrate the vasculature and cause disease. This project examined the interactions of P. gingivalis with endothelial cells. The first objective was to investigate the effects of P. gingivalis gingipains on the prot eins of the adherens endothelial junctions of human endothelial cells us ing wild-type strains of P. gingivalis and gingipain knock-out mutants. The second objecti ve was to examine the level of importance of the concept of adhesion versus invasion of P. gingivalis in causing the disruption of the integrity of the endothelial cell monolayer. The third objectiv e was to determine if cell death is stimulated or hindered by P. gingivalis and whether P. gingivalis is likely contributing to these effects from the exterior of the cell or from within the cell.

PAGE 55

55 CHAPTER 2 MATERIALS AND METHODS Bacterial Strains and Growth Conditions P. gingivalis Subculture P. gingivalis strains (see Table 1 for the list of strains used) were grown on tryptic soy agar (Difco Laboratories, Detroit, MI) supplemen ted with 0.5% yeast extr act (Fisher Chemicals, Fair Lawn, NJ), 5% sheep blood (Quad Five, Ryegate, MT), vitamin K (5 g/ml), and hemin (5 g/ml). Overnight cultures were grown in trypt ic soy broth (Difco) supplemented with 0.5% yeast extract, haemin (5 g/mL), and vitamin K (5 g/mL) under anaerobic conditions [37C in a Coy anaerobic chamber (Ann Arbor, M I) with an atmosphere of 5% CO2, 10% H2, and 85% N2]. Cell Culture Conditions Human umbilical vein endothelial cells (HUVEC, Cambrex, Baltimore, MD) and Human hepatic epithelial cells (HuH7, Michael Kilberg, University of Fl orida) were used for this study. The HUVEC were maintained using Microvascul ar Endothelial Growth Medium-2 (EGM-2; Clonetics, Inc., San Diego, CA), which contains endothelial cell basal medium-2 supplemented with hydrocortisone, human recomb inant fibroblast growth factor 10% fetal bovine serum (FBS) (Invitrogen, Carlsbad, CA), ascorbic acid, hum an recombinant epidermal growth factor, gentamicin (Sigma, St. Louis, MO), recombinant insulin growth factor-1, and amphotericin B as well as 10,000 units/ml of streptomycin and 10,000 units/ml of penicillin G, and 200 mM Lglutamine. The HuH7 cells were maintained in Eagles minimum essential medium (Mediatech, Herndon, VA) supplemented with 10% FBS (Inv itrogen) 10,000 units/ml of streptomycin, 10,000 units/ml of penicillin G, and 200 mM L-glutamine (Sigma). Cells were cultured in 75cm2 flasks at 37C in a 5% CO2 humidified atmosphere.

PAGE 56

56 Bacterial Preparations P. gingivalis Lysate For the preparation of P. gingivalis lysates, P. gingivalis W83 was inoculated into 5 mls of Blood Heart Infusion Broth (BHI) (Difco) and a llowed to grow at 37C in a Coy anaerobic chamber (Ann Arbor, MI) with an atmosphere of 5% CO2, 10% H2, and 85% N2, for 5 to 7 days until the BHI broth was very opaque. The 5 mls of the bacterial broth was then transferred to a larger volume (1 liter) of BHI broth and incubated anaerobically for an additional 3 to 5 days. When the bacterial solution was very opaque, the liter of bacterial cells was centrifuged using a Beckman J2-21 centrifuge with a JA-10 rotor at 11,315 X g at 4C for 30 minutes to 1 hour. The resulting cell pellet was resu spended in 5 mls of 50 mM Tris, pH 7.5 and sonicated for 20 minutes to lyse the bacterial cells. The lysed bacterial cells were then concentrated using a Centriprep 10,000 MWCO (Millipore Corp., Bedfor d, MA) and a Sorvall RC5B centrifuge with a HB-4 rotor at 16,320 X g for 30 min. The final volume was approximately 6 mls. Secreted Protein Factor P. gingivalis bacterial strains W83 and 381 were i noculated into 0.5 mls of Blood Heart Infusion Broth (BHI) (Difco) and allowed to gr ow at 37C in a Coy anaerobic chamber (Ann Arbor, MI) with an atmosphere of 5% CO2, 10% H2, and 85% N2 for 2 days. Next, 0.5 mls of each bacterial culture was transferred to 100 ml of BHI and allowed to grow overnight. Following growth, 100 ml of the bacterial cultures were transferred to 1 l iter of BHI and allowed to grow for 24-48 hours. The bacterial cells were spun down using a Beckman J2-21 centrifuge with a JA-10 rotor at 11,315 X g at 4C for 30 minutes to 1 hour and the supernatant was collected and filtered through a 0.2 micron f ilter (Whatman, Mobile, AL). Then, 36.1g ammonium sulfate (Sigma-Aldrich) was added to each 100 mls of filtered supernatant and allowed to dissolve by stirring in a 4C cold r oom overnight. The supernatant was then spun

PAGE 57

57 down using an Eppendorf Centrifuge 5801 with an A4-62 rotor at 3,166 X g and 4C for 30 minutes, the resulting supernatant was decanted, and the pellet was collected and dissolved in 50 mM Tris (Fisher) pH 7.5. This solution was then dialyzed against 50mM Tris, pH 7.5, for 48 hours exchanging with fresh Tris buffer pH 7.5, twice, followed by concentration using a Centriprep 10,000 MWCO (Millipore Corp., Bedfor d, MA) and a Sorvall RC5B centrifuge with a HB-4 rotor at 16,320 X g to a final volume of 1 ml. Microassay Procedure for Proteins The Bio Rad protein assay dye reagent concentr ate was used for the colorimetric detection and quantitation of total protein. Each test tube contained 800 l of the sample. Then, 200 l of the dye concentrate reagent was ad ded to each tube and the tube s were vortexed for 30 seconds. The test tubes were then incubated at room te mperature for five minutes and the absorbance was measured at 595 nm using a Spectronic 10 01 spectrophotometer (Milton Roy Company, Rochester, NY). A standard curve of BS A at concentrations ranging from 5 to 20 g/ml was used to determine protein concentration. The protein solutions were assayed in duplicate. The data was interpreted using Sigma Plot 2000 (Systat, Point Richmond, CA). Enzymatic Assay Arginine-specific cysteine proteinase activity was determined using 5 mM Na -benzolyDL-arginine p -nitroanilide (BAPNA) (Sigma, St. L ouis MO) in 125 mM Tris-HCL, pH 8.0, containing 12.5 mM L-cy steine and 12.5 mM CaCl2 in a total volume of 1 ml at room temperature (Shoji et al. 2002; Chen et al. 2001). After the sample was added, the absorbance (A405) was continuously recorded using a Spectronic 1001 spec trophotometer (Milton Roy Company, Rochester, NY). The proteinase activ ity of the secreted protein fractions was determined by the increase in absorbance/min/ml culture. Kinetic analysis was interpreted using Sigma Plot 2000 (Systat, Point Richmond, CA).

PAGE 58

58 Inhibitors To determine if the effects observed with the cell extract of P. gingivalis W83 were due to the gingipains, HuH7 cells were incubated for twel ve and twenty four hours at 37oC in the presence of a 1 mM concentration of the cysteine protease inhibitor Na-tosyl-L-lysine chloromethyl ketone (TLCK) (Sigma), a 2mM co ncentration of a serine protease inhibitor PMSF (Sigma), and a 2mM concentra tion of a carboxypeptidase, seri ne and cysteine proteinase inhibitor (PIC) (Sigma). Microscopy Antibodies Mouse anti-catenin (Zymed Laboratires Inc ., South San Francisco, CA), rabbit polyclonal anti-catenin (Santa Cruz Biotech Inc., Sa nta Cruz, California), anti-pan cadherin (Sigma-Aldrich), and P. gingivalis rHagB polyclonal (Kohler, J. Song, H.) antibody were used in this study. The purified rHagB was obtained an d prepared as described previously (Song et al ., 2005). The P. gingivalis rHagB polyclonal antisera, A7986, was raised against the purified rHagB protein (Strategic Biosolutions, Newar k, DE). Pre-immune rabbit serum was obtained prior to the first immunization and used as a negative control. Fluorescein isothiocyanate (FITC)-conjugated goat anti-mous e antibody (Sigma-Aldrich) and tetramethyl rhodamine isothiocyanate (TRITC)-conjuga ted goat anti-rabbit antibody (S igma-Aldrich) were used as secondary antibodies. Immunofluorescence Microscopy HUVECs were allowed to grow to confluency on glass coverslips in six-well tissue culture plates (Corning Inc, Corning, NY). The HUVECs were then washed with antibiotic-free EGM2 media three times prior to infection with the P. gingivalis strains listed, at various lengths of time, at a multiplicity of infec tion (MOI) of approximately 100. The media was removed from

PAGE 59

59 each well and the cells were then fixed in 4% paraformaldehyde in PBS for 30 minutes at room temperature. This was followed by washing th ree times with PBS and quenching in 50 mM NH4Cl and 0.2% Triton X 100-PBS for 10 minutes at room temperat ure. The quenching solution was removed and the HUVECs were again wash ed three times with PBS. The primary antibodies, diluted 1:50 for host proteins or 1:200 for bacteria in PBS with 0.5% normal goat serum and 0.2% Triton X-100, were applied at room temperature for 2 hours. The HUVECs were then washed three times with PBS for 5 minut es each. The bacteria were detected with TRITC-conjugated goat anti-rabb it secondary antibody while the host protein markers were detected using FITC-conjugated goat anti-mouse antibody or TRITC-conjugated goat anti-rabbit, depending on labeling techniques. The secondary antibodies, diluted 1:20 0 dilution in PBS with 5% normal goat serum and 0.2% Triton X-100, were applied for 1 hour at room temperature. The HUVECs were then washed three times with PBS before mounting with Vectashield with DAPI (H-1200, (Vector Labs, Burlingame, CA) onto glass microscope slides. A Zeiss Axiophot fluorescence photomicroscope with a Cool Snap (Photometrics) camera and IP Lab imagining software (Scanalytics Inc., Rockville, MD ) was used to view the images. All immunofluorescence experiments were repeated at least twice. Transmission Electron Microscopy HUVECs were allowed to grow to confluency on ACLAR (Kingsley RE) layered in 6-well dishes with EGM-2 supplemented media. Afte r a 5 and 20 hour incubation of the HUVECs in antibiotic free EGM-2 media with bacteria, the cells were fixed in 2% glutaraldehyde in 0.1M sodium cacodylate (Sigma-Aldrich, St. Louis, MO ) buffer, pH 7.4, with 1 mM calcium chloride (Sigma-Aldrich), for 1 hour. Next, the cells we re rinsed 3 times for 5-10 minutes each in 0.1M sodium cacodylate buffer and then postfixed with 1% osmium tetroxide (Electron Microscopy Sciences, Fort Washington, PA) in 0.1M sodium cacodylate for 1 hour. They were rinsed in

PAGE 60

60 deionized H2O three times for 10 minutes each and dehydrat ed in a graded etha nol series (50, 70, 95, 100%) for 10 minutes each, followed by dehydration in 100% acetone for 15 minutes. The cells were then infiltrated with Spurrs resi n (Spurr, A.R.) and acetone 1:1 for 1 hour, followed by 100% resin overnight. The samples were cured in flat embedding molds overnight at 60C. Then, 60-80 nm sections were obtained using Fo rmvar-coated copper grid s, and were stained with 2% uranyl acetate (Sig ma-Aldrich) and 0.1% lead citr ate (Venable JH) or for acid phosphatase. Samples were examined at 60 KV on JEOL 100CX electron microscope. Cell Death Assay Quantification of Endothelial Cells by Flow Cytometry Endothelial cells were seeded and allowed to gr ow to confluency in 6 well dishes. All of the wells were washed twice with antibiotic free EGM-2 media prio r to the addition of P. gingivalis After the indicated incubations, the antibiotic-free media plus free-floating P. gingivalis were transferred using a pipette, to 1.5 ml microcentrifuge tubes (USA Scientific, Ocala, FL) and centrifuged using a Fisher Microc entrifuge model 59A at 5800 X g for 5 minutes. 0.5ml of 0.25% trypsin-EDTA was a dded to each well and the dishes were then placed back into the 37C incubator for 5 minutes. The supern atant from the 1.5 ml centrifuge tubes was decanted and the trypsin-EDTA/cell mixture fr om each well was added to the corresponding tubes and centrifuged for 5 minutes at 5800 X g. The supernatant was decanted and the cell pellet was resuspended in 300 l PBS prior to transfer to a 5 ml polystyrene round bottom tube (Falcon, Bedford, MA). Propidium iodide (Sigma-Aldrich) at 1 g/ml final concentration was then added to each tube. Flow cytometry was performed using a FACSort flow cytometer (BD Biosciences, San Jose, CA). The instrument illu minates individual cells in flow using an argonion laser emitting 15 milliwatts of 488 nm of light Forward and side light scatter and orange fluorescence (585 nm +/21 nm) for each cell was measured. Data for 10,000 particles were

PAGE 61

61 collected per sample. Cell Quest software (Ver sion 3.3; BD Biosciences) was used for data collection and subsequent analysis Intact cells were defined based on their forward and side scatter and their orange fluores cence (propidium iodide emissi on) was displayed. Dead cells (orange fluorescence above backgr ound) were counted and expresse d as a percentage of total cells in each sample (Fig. 2-1). Antibiotic Protection Assay Approximately 105 HUVECs were seeded per well in 6well tissue culture plates (Corning Inc, Corning, NY) followed by washing three tim es with antibiotic-fre e EGM-2 media. The wells were then infected with an overnight liquid culture of 107 bacteria (MOI 100) in 1.0 ml of 37oC EGM-2 with 200 g ml-1, metronidazole (Sigma) and 300 g ml-1, gentamicin (Sigma) antibiotics. After 0.5, 2.5, 12, and/or 24 hours of incubation, the media were collected and centrifuged as described above. Cell onl y wells were used as controls. Inhibitors To determine the effects of the actin polymer ization inhibitor Cytochalasin D (Sigma) on bacterial invasion, HUVECs were preincubate d for 0.5 hrs with cytochalasin D [5 g ml-1 in dimethly sulfoxide (DMSO)] prior to the additio n of the overnight culture preparations of P. gingivalis strains. Caspase 3 Activity Assay The use of both apoptotic and non-apoptotic cells were required for this assay. Confluent monolayers of HUVEC were treat ed for 2.5, 12, and 24 hours with P. gingivalis wild-type strains 381 and ATCC 33277 (MOI 100) with and without an tibiotics. An inducer of apoptosis, staurosporine (STS) at a concentration of 0.5 m, was used as a positive control. At each timepoint, the media were collected on ice, and the adherent cells were scraped from each well and also transferred into 1.5 ml micr ocentrifuge tubes on ice. Afte r centrifugation of the tubes at

PAGE 62

62 1500g for 5 minutes, the medium was decanted a nd the pellets carefully resuspended in 100 l of sterile filtered cell lysis buffer (1X) (10 mM Tris-HCL, 10 mM NaH2PO4/NaPO4, pH 7.5, 130 mM NaCl, 1% TritonX-100, 10 mM NaPPi (sodium pyrophosphate). Following 30 minutes on ice, the lysed cell solution was centr ifuged at 1500g for 10 minutes at 4oC. The liquid was then separated from the pellet and co llected on ice into fresh 500 l microcentrifuge tubes. A 96 well plate was used to conduct the Ca spase 3 protease assay. A 20 M final concentration of AcDEVD-AMC Fluorogenic Substrate (BD Biosci ences Cat. No. 66081U) was added to 170 l of freshly made Protease Assay Buffer (1X): 20 mM HEPES (pH 7.5), 10% glycerol, 2 mM DTT plus 20 l of each sample. The amount of AMC liberated from A-DEVD -AMC was measured using a spectrofluorometer (Perkin-Elmer Vi ctor 1420-011, Wallac, Turku, Finland) with an excitation wavelength of 380 nm a nd a 430-460 nm emission wavelength. Statistical Analysis Significant differences between the means (+/standard deviations) for the cell death, transepithelial resistance, and caspase 3 activity assays were determined using a two-tailed Students t test. Differences in P values of <0.05 were considered significant. Transepithelial Resistance TER was measured across HuH7 monolayers grown on standard 24 well cell culture transwell 6.5 mm, 3.0 m pore size inserts (Corning Incorp orated, Acton, MA) and collagen coated Corning transwell-COL inserts (Corning Inco rporated ). Transepithelial resistance (TER) was measured using an EndOhm chamber (World Precision Instrument, Inc., Sarasota, FL) and the EVOM electrical resistance system (World Precision Instruments, New Haven, CT). Cellfree transwell inserts were used to obtain baseli ne levels. The baseline monolayer resistance was measured on the third day after s eeding. The tight-junctioned HuH7 monolayers were incubated at 37oC in EMEM medium supplemented with 10,000 units/ml of streptomycin and 10,000

PAGE 63

63 units/ml of penicillin G, 5% FBS and 2 mM glutamine. Freshly harvested P. gingivalis at an MOI of 100 in 0.4 ml antibiotic-free EMEM me dium were added to the upper chamber. Antibiotic-free supplemented EMEM medium (1.0 ml ) was then added into the lower chamber. The TER of the HuH7 monolayers was measured at each tim e point. The resistance from each well was corrected for the resistan ce of a cell-free insert, yielding the resistance of the monolayer expressed in Ohms.

PAGE 64

64 Figure 2-1. Flow cytometry scat ter plot of HUVECs showing liv e cells and dead cells. (A) Twelve hour scatter plot of untreated HUVECs and (B) scatter plot of HUVECs exposed to strain 381 at an MOI of 100 for 12 hours. Live and dead cells are labeled in each plot. Dead cells Live cells Dead cells Live cells A B

PAGE 65

65 Table 2-1. Bacterial strains and plasmids used and constructed in this study Wild-type Strains Mutated gene Source/reference 381 None SUNY-Buffalob W83 None SUNY-Buffalob ATCC 33277a None Lamont, RJ AJW4 None B.G. Loos Mutant strain designation Mutated gene Parent strain MT10 rgpA 381 Kuramitsu, HK MT10W rgpA, kgp 381 Kuramitsu, HK G102 rgpB 381 Kuramitsu, HK G102W rgpB, kgp 381 Kuramitsu, HK YPP2 kgp ATCC 33277 Park et al., 1998 CW401 rgpA, rgpB 381 Kuramitsu, HK CW501 rgpA rgpB, kgp 381 Kuramitsu, HK YPF1 fimA ATCC 33277 Love RM, 2000 YPEP pepO ATCC 33277 Park et al., 2004 TIGR ID Gene name Parent strain PG1788 Putative cysteine peptidase W83 Sheila Walters PG0293 Putative secretion activator protein W83 Sheila Walters PG0686 Conserved hypothetical protein W83 Paulo Rodrigues PG1286 Ferritin W83 Paulo Rodrigues PG1118 clpB W83 Ann ProgulskeFox PG0242 Conserved hypothetical protein W83 Paulo Rodrigues PG0717 Putative lipoprotein W83 Paulo Rodrigues PG1683 Conserved hypothetical protein W83 Paulo Rodrigues aAbbreviations: ATCC = American Type Culture Collection bSUNY-Buffalo = State University of New York, Buffalo, NY, USA

PAGE 66

66 CHAPTER 3 RESULTS Monolayer Integrity Effects of P. gingivalis Lysate on the Cellular Distributio n of the Junctional Proteins in HuH7 and HUVE Cells Initial experiments were performed using the Human Hepatic Epithelial Cell line (HuH7), and Human Umbilical Vein Endothelial Cells (H UVEC) to observe whether the lysate (cell extract) of P. gingivalis W83 had the ability to cause the alte ration of the location of the proteins within the adherens junctional complex from the cell border to a more cytoplasmic location, subsequently disrupting integrity of the monolayer. Monolayer inte grity is defined as a layer of continuous, tightly apposed cells. Confluent monolayers of both cell types were treated with varying concentrations of the P. gingivalis W83 lysate and the effect s on the integrity of the intact monolayer were compar ed to a negative control ( E. coli lysate) strain using immunofluorescence microscopy (IMF). HuH7 cells were treated with 0.2 and 0.4 mg/ml concentrations of the W83 lysate and 0.4 mg/ml of the E. coli lysate (Fig. 3-1). The HuH7 cells were labeled with antibodies ag ainst two proteins located in the adherens junctional complex, cadherin and -catenin. The results from the untreated HuH7 cells showed that the majority of these proteins were concentrated at the cell surf ace, between adjacent cells (arrows) (Fig. 3-1). -catenin could also be seen sparsely scatte red throughout the cytoplas m. After 24 hours of incubation, the W83 lysate, but not the E. coli lysate, was able to affect the distribu tion of these proteins within these cells (F ig. 3-1). Treatment of HuH7 cell monolayers with 0.2 mg/ml of the W83 lysate resulted in a loss of monolayer integrity, specifically a loss of cadherin and -catenin from the cell surface. Concurrently, both of thes e proteins were in punctate areas throughout the cytoplasm and -catenin was also detected in the proxim ity of the nucleus (Fig. 3-1). Exposing

PAGE 67

67 HuH7 monolayers to a 0.4 mg/ml concentration of the W83 lysate for 24 hours resulted in no detectable cells remaining on the co verslip. Thus, exposure of the HuH7 cells to the W83 lysate but not the E. coli lysate resulted in an al teration of location of two adherens junction proteins, cadherin and -catenin, and disruption of th e monolayers integrity. We next sought to examine if the P. gingivalis W83 lysate was capable of stimulating similar effects on the endothelial cell line, HUVEC. For these experiments, confluent HUVEC monolayers were treated with 0.2 mg/ml and 0.4 mg/ml of the W83 lysate for 24 hours. As with the HuH7 cells, HUVECs treated with 0.4 mg/ml of the W83 lysate resulted in no detectable cells remaining on the coverslip. Exposure of 0.2 mg/ml of the W83 lysate resu lted in the alteration of the cadherin and -catenin proteins from the cell surface to punctate areas within the cytoplasm. However, very few cells remained a dhered to the coverslip (d ata not shown). Given that the HUVECs were very sensitive to the W83 lysate, lower concentrations (0.05 and 0.1 mg/ml) of the W83 lysate were also examined. The 0.05 mg/ml concentration of the W83 lysate was unable to alter the location of cadherin (data not shown) and -catenin from the cell surface, demonstrating that this concentration was not high enough to disrupt the integrity of the HUVEC monolayer. However, exposure of 0.1 mg/ml of the W83 lysate to the HUVEC monolayers resulted in alteration of the location of the cadherin (data not shown) and -catenin proteins (arrows), shifting these proteins from a cell surface location to puncta te areas within the cytoplasm (Fig. 3-2). In addition, there was a re duction in the amount of cells adhering to the coverslip compared to the untreated ce lls (Fig. 3-2). This data suggests P. gingivalis contains proteins that are capable of affecting the cell-cell adhesion pr oteins cadherin and -catenin and the cell-matrix adhesion protein, integrin, in both epithe lial and endothelial cells.

PAGE 68

68 Effects of Heat and Protease Inhibitors on the Cellular Distribution of the Cadherin and catenin Proteins in HuH7 Cells Treated with P. gingivalis W83 Lysates P. gingivalis is an asaccharolytic anaerobe that e xpresses arginine (Rgp) and lysine (Kgp) cysteine proteinases (gingipains) A primary function of these proteinases is the breakdown of proteins and peptides within its environment, which provide this bacterium with energy and carbon. As discussed in chapter 1, these proteinases have also been shown to be major factors influencing the virulence of this pathogen (OBrien-Simpson et al. 2003). The loss of the HuH7 and HUVEC monolayer integrities observed after exposure of these inta ct monolayers to the W83 lysate suggests that proteins from P. gingivalis are responsible for this disruptive effect. The following experiments were done to determine the identity of the active component(s). We first sought to establish if temper ature affected the activities of the P. gingivalis lysate. The W83 lysate was thus heated for 20 minutes at 60oC (Yumoto et al. 2005; Nassar et al. 2002) to determine if heat treatment had any inhi bitory effects on the effi cacy of the W83 lysate to stimulate epithelial monolayer disruption and the alteration of the location of the adherens junction proteins in HuH7 cells. Confluent HuH7 monolayers treated with the heated W83 lysate resulted in no detectable differences in the localization of cadherin (Fig. 3-3) and -catenin (Fig. 3-4) proteins in these cells compared with the untreated cells. Thus, the failure of the heattreated W83 lysate to disrupt the junctional complexes is likely due to the denaturation of specific proteins associated with the activity. We next sought to determine whether the prot eins involved in stimul ating endothelial and epithelial monolayer disruption were cysteine prot einases (e.g., gingipains). To this end, specific proteinase inhibitors were tested to determine their inhibitory e ffects on the active proteins in the W83 lysates. Three proteinase i nhibitors including PMSF, a serine proteinase inhibitor, TLCK, a cysteine proteinase inhibitor, and PIC, a cocktail of inhibitors of carboxypeptidases, serine and

PAGE 69

69 cysteine proteinases, were tested. Of these inhibitors, TLCK was the only one shown previously to efficiently block P. gingivalis cysteine proteinase activity (Hintermann et al. 2002). After 22 hours of incubation, neither PMSF nor PIC at a con centration of 2 mM inhi bited the destructive effects of the 0.4 mg/ml W83 lysate on the HuH7 monolayers since there we re no detectable cells remaining after treatment. However, TLCK at a concentration of 1 mM in hibited the effects of the W83 lysate on the cell-matrix adhesions by preventing the detachment of cells from the monolayer (Figs. 3-3, 3-4). The results from the TLCK treated HuH7 cells were similar to control cells in that the cell monolayer was almost intact. However, unlike the control cells, the distribution of cadhe rin (Fig. 3-3) and -catenin (Fig. 3-4) changed from a cell surface location to punctate areas within the cytoplasm. The redistribution of cadherin and -catenin ultimately led to the disruption of the a dherens junctions as observed by the loss of the continuous monolayer. The reduced capacity of the W83 lysate in the presence of TLCK to disrupt the cell monolayer suggests that the active component(s) re sponsible for disrupting cell-matrix and cellcell adhesion are cystei ne proteinases. Proteinase Inhibitor Treatment of P. gingivalis W83 on HuH7 Adherens Junctional Complexes The previous experiments using the P. gingivalis W83 lysate suggested that the cysteine proteinases (gingipains) were mo st likely responsible for disrupti ng junctional proteins since the cysteine proteinase inhibitor TLCK was capable of inhibiting this activity. The next step was to determine whether the effects of the gingipains were also mediated from within the cell. Since strain W83 is known to adhere to and invade hos t cells, we determined if the gingipains from live, intact P. gingivalis W83 were capable of stimulating the HuH7 monolayer disruption we observed with the W83 lysate. After strain W83 was co-cultured with confluent HuH7 monolayers at an MOI of 1000 for 24 hours, there were no detectable HuH7 cells remaining on

PAGE 70

70 the coverslip. In order to determine if the cyst eine proteinase inhibito r, TLCK, would suppress these effects, live W83 cells were co-cultured with confluent HuH7 monolayers and 1 mM TLCK (Figs. 3-3, 3-4). After 24 hours, many HuH7 cells remained on the glass coverslip. However, no cadherin or -catenin could be detected at the cell me mbranes of the remaining cells (Figs. 3-3, 3-4). Taken together, these resu lts suggest that the disruption of the epit helial and endothelial cell-cell and cell-matrix adhesions and subsequent loss of monolay er integrity are stimulated by the cysteine proteinases (gingi pains) associated with live P. gingivalis W83 and the W83 lysate. These data do not prove but sugge st that the gingipains may act both from within the epithelial and endothelial cells as well as fr om the cells exterior. Treatm ent with TLCK was capable of inhibiting the P. gingivalis cysteine proteinases Rgp and Kgp fr om stimulating the detachment of HuH7 cells, but was unable to inhibit th e alteration of the location of -catenin from the cell surface to within the HuH7 cytoplasm. Since TLCK has been shown to inhibit Kgp activity more than Rgp activity (Pike et al. 1994), it is possible that the activ ity of Rgp is largely accountable for the observed alteration of location of cadherin and -catenin within the epithelial and endothelial monolayers examined. Effects of P. gingivalis Secreted Protein Fraction (SPF) on the Adherens Junctional Complex Proteins of HUVECs Experiments using the cysteine proteinase in hibitor TLCK suggested that the gingipains from both live P. gingivalis W83 and the W83 lysate are likely responsible for the disruption of the HuH7 monolayer, specifica lly the junctional pr oteins cadherin and -catenin. Since the gingipains are known to exist in both cell membra ne associated and secretory forms (Kadowaki and Yamamoto, 2003; Potempa et al. 2003; Kadowaki et al. 2000; Potempa et al. 1998; Potempa et al. 1997; Travis et al. 1997), we next sought to determine if P. gingivalis surface as well as secreted proteins were capable of stimulating endothe lial monolayer disruption.

PAGE 71

71 To investigate this, the activities of the P. gingivalis soluble surface and/or secreted proteins (SPF) on HUVEC monolayers were studie d to determine their levels of proteolytic activity relative to the junctiona l complexes. The SPF was prepar ed using ammonium sulfate to precipitate out the secreted and cell surface associated proteins from P. gingivalis The protein concentration of the SPF from P. gingivalis was determined to be 23 mg/ml for strain W83 and 0.2 mg/ml for strain 381. Confluent HUVEC mo nolayers were then treated with various concentrations of these SPFs for 24 hours. Twenty-two g/ml of the W83 SPF and 1 g/ml of the 381 SPF resulted in a loss of both -catenin (Fig. 3-5) and cadhe rin (Fig. 3-6) from the cell surface of HUVECs. Approximately half of th e HUVECs treated with the SPFs remained adhered to the coverslip compared with the untreated cells. The results showed that exposure of the HUVECs to 0.02 g/ml of the 381 SPF and 0.2 g/ml of the W83 SPF resulted in an altered distribution of cadherin and -catenin (Fig. 3-5, 3-6). Howeve r, a considerably greater number of cells remained on the coverslip in comparis on to the number of cells exposed to the higher concentrations of the SPFs (22 g/ml of the W83 SPF and 1 g/ml of the 381 SPF). In addition, the disruption in the HUVEC monolayer was more pronounced after treatment with the 381 SPF (0.02 g/ml) than with the W83 SPF (0.2 g/ml) (Fig. 3-5; 3-6). One explanation for this difference is that the level of arginine protease activity of the strain 381 SPF, measured as activity per g of protein using an enzymatic assay with 5 mM Na -benzoly-DL-arginine p nitroanilide (BAPNA) was approxima tely two-fold greater than that of the W83 SPF. This data indicates that the deleterious effect s to the monolayers associated with P. gingivalis are at least in part due to surface associated and/or secreted proteins, (e.g., gingipains), since the SPFs alone were capable of inducing such monolayer disruption.

PAGE 72

72 Effects of P. gingivalis on Confluent HUVEC Monolayers We have shown that the surface associated /secreted proteins (e.g., gingipains) from P. gingivalis W83 affect the HUVE cells so as to a lter the distribution of both cadherin and catenin, as well as to stimulate disruption of the epithe lial and endothelial monolayers. We next sought to test whether live P. gingivalis was capable of inducing th e same effects to the HUVEC monolayer as observed with the P. gingivalis lysate and SPF. IMF was used to detect the loss of or changes in the cellular loca tion of the junctional protein, -catenin, within the confluent HUVEC monolayer after treatment with live P. gingivalis wild-type strains W83, 381, ATCC 33277, and AJW4 (Table 3-1). Data published by ot hers previously reported that strains W83, 381 and ATCC 33277 are all, with different effici encies, capable of adhering to, invading and persisting within a number of cell types includi ng human coronary arte ry endothelial cells, HCAEC, gingival epithelial ce lls, GECs, and HUVECs (Yilmaz et al. 2003; Dorn et al. 1999; Deshpande et al. 1998). In contrast, AJW4 has been s hown to adhere to but not invade or persist within HCAEC or HUVECs (Dorn et al. 2000). These data are summarized in table 3-1. In our study, HUVEC monolayer s co-cultured with these strains at an MOI of 100 appeared unaffected after 1 hour (data not shown) and 5 hours of incubation (Figs. 3-8 3-11). The majority of -catenin was concentrated at the cell surface, between adjacent cells (white arrows) (Figs. 3-8 3-11). Yet, at both times, P. gingivalis W83, 381 and ATCC 33277 appeared to be located on th e cell surface as well as interna lized with the HUVECs (yellow arrows). In contrast, the HUVE C monolayers exposed to AJW4 for 1 and 5 hours showed very few P. gingivalis attached to the monolayer. After 10 hour s of co-culture w ith strains W83 and 381, there was an observable loss of -catenin at the cell surface (whi te arrows) (Figs. 3-8, 3-9). Strains AJW4 and ATCC 33277 did not stimulat e the HUVECs to alter the localization of -

PAGE 73

73 catenin within the conf luent monolayers until 15 and 20 hours, respectively (Figs. 3-10, 3-11). Treatment of HUVEC monolayers fo r 20 hours with all four strains resulted in disruption to the monolayers as evident by th e loss of cells and less -catenin (white arrows ) at the cell surface (Figs. 3-8 3-11). Except for AJW4, each strain of P. gingivalis was internally localized or associated with the HUVEC surface at each of the time points tested (yellow arrows) (Figs. 3-8, 3-11). Strain AJW4 did not appe ar to be internalized or asso ciated with the HUVE cell surface until 10 hours of exposure. Increasing the MOIs from 100 to 1000 caused more dramatic effects on the HUVEC monolayer. HUVEC monolayers expo sed to strain W83 at an MOI of 1000 remained intact and retained -catenin at the cell surface through 5 hours of co-culture (Fig. 3-12). However, 5 hours of co-culture of HUVECs with strains 381, ATCC 33277 and AJW4 at this concentration resulted in a deleterious effect on the monolayers (Figs. 3-13 3-15). At this time, notably reduced amounts of -catenin were detected at the cell surf ace (white arrows). After 10 hours of exposure to all four strains, the HUVEC monol ayers were markedly disrupted and no longer intact. In addition, a higher number of each strain of P. gingivalis including AJW4 (yellow arrows) were intracellularly localized and/or asso ciated with the cell surf ace at this time point, relative to Time 0 (Figs. 3-12 315). Treatment for 20 hours with all four strains at an MOI of 1000 resulted in considerable disruption to the monolayer, causing the ma jority of the HUVECs to elongate and/or detach (Figs. 3-12 -3-15). At this time point, high numbers of all four strains of P. gingivalis, including AJW4, were de tected both associated w ith the HUVEC surfaces and internalized within the cytopl asm (yellow arrows) (Figs. 312 3-15). In addition, HUVEC monolayers treated for 20 hours with each P. gingivalis strain resulted in a lack of junctional complexes and -catenin was no longer detected at the cell surface.

PAGE 74

74 These data presented here indicate that 10 hours of treatment with the P. gingivalis strains W83 and 381 at an MOI of 100, which have been shown to adhere, invade and persist within a number of different cell type s including HUVEC (Table 3-1), is sufficient for inducing HUVEC monolayer disruption and the al teration of the location of -catenin from the cell surfaces, to a location within the cytoplasm (Figs. 3-8, 3-9). The results with strain ATCC 33277 at an MOI of 100, which has also been shown to adhere, invade and persist within HUVEC, showed that this bacterium was also capable of inducing HUVE C monolayer disruption, but only after 20 hours of co-culture (Fig. 3-10). Interestingly, AJ W4 at an MOI of 100, which has been shown to adhere to but not invade or persist within a number of diffe rent cell types including HUVEC, was capable of inducing HUVEC monol ayer disruption after 15 hours of co-culture (Fig. 3-11). Since invasion efficiencies are characteri stically determined after 2.5 hours of P. gingivalis exposure with the cell of interest, it is possi ble that the extended interaction times of P. gingivalis with the HUVE cells were sufficient for AJW4 in vasion. Barring this pos sibility, these data suggest that internalized P. gingivalis play a role in stimula ting HUVEC monolayer disruption and that the gingipains associ ated with the internalized P. gingivalis are perhaps inducing their deleterious effects from within these cells However, since the results with the P. gingivalis SPFs showed monolayer disrupti on, we cant rule out the poss ibility that the gingipains interacting with the HUVE ce lls exterior and/or gingipa ins associated with the P. gingivalis that adhered to the cell surface also participate in this activity. Effects of Gingipain Mutants on the Junctional Complex Protein -catenin in Confluent HUVEC Monolayers The data using specific proteina se inhibitors indicated that the cysteine proteinases, (gingipains) from P. gingivalis are likely responsible for the observed HUVEC monolayer disruption. The goal of the next set of experiments was to determ ine which of the gingipains is

PAGE 75

75 most active in HUVEC monolayer disruption. Gingipain knock-out (KO) mutants of P. gingivalis were used in this set of experiments All gingipain mutants (Table 3-2) were constructed from P. gingivalis strain 381 (Chen and Kuramitsu, 1999) except for YPP2 ( kgp), which was constructed in strain ATCC 33277 (Park and Lamont, 1998). Confluent HUVEC monolayers were co-culture d with live gingipain mutants including MT10 ( rgpA), MT10W ( rgpA-, kgp), G102 ( rgpB), G102W ( rgpB-, kgp), CW401 ( rgpA-, rgpB), CW501 ( rgpA-, rgpB-, kgp) and YPP2 ( kgp) at an MOI of 100 (Figs. 3-16 3-22). Through 10 hours, the P. gingivalis gingipain mutants were detect ed at the cell surface and/or within the cell (yellow arrows). However, unlike the monolayers exposed to strain 381 (Fig. 39), the monolayers exposed to each P. gingivalis gingipain mutant remained intact at this time point. Furthermore, the HUVEC mo nolayers remained intact with -catenin at the cell surface (white arrows) even after 15 hour s of co-culture with G102 ( rgpB) (Fig. 3-18), CW401 ( rgpA-, rgpB) (Fig. 3-21) and CW501 ( rgpA-, rgpB-, kgp) (Fig. 3-22). In contrast, HUVEC monolayers exposed to MT10 ( rgpA) (Fig. 3-16), MT10W ( rgpA-, kgp) (Fig. 3-17) and G102W ( rgpB-, kgp) (Fig. 3-19) displayed a loss of -catenin staining at the cell surface (white arrows) and a disruption of the confluen t monolayers, suggesting some cells within the monolayers had detached. The HUVEC monolayers co-culture d with MT10, MT10W, G102, G102W, and YPP2 for 20 hours (Figs. 3-16 3-20) resulted in numerous P. gingivalis (yellow arrows) attached to and localized within the cells and a lack of -catenin staining at the cell surface (white arrows) (Fig. 3-43). In addition, these monolayers were no longer continuous since some cells had broken free and detached from the monolayer. In contrast, the HUVEC monolayers exposed to CW401 ( rgpA-, rgpB) and CW501 ( rgpA-, rgpB-, kgp) for 20 hours resulted in only a few P. gingivalis attached to or internalized with the cells (Figs. 3-21 3-22) and no alteration of

PAGE 76

76 location of the junctional protein, -catenin. Thus, the HUVEC monolayers treated with CW401 and CW501 maintained monolayer integrity and rese mbled the untreated cells at each time point. A comparison of the HUVEC monolayers treated fo r 20 hours with all of the gingipain mutants and their parent strains can be viewed in figure 3-43. To summarize these results, the HUVE cell monol ayers exposed to the parent strains 381 and ATCC 33277 resulted in obvious monolayer disruption after 20 hours of exposure. HUVE cell monolayers exposed to the gingipain mu tants containing RgpB (MT10W) or RgpA (G102W) only, resulted in partial monolayer disrup tion after 20 hours. In addition, the gingipain mutants containing Kgp and RgpB (MT10) or K gp and RgpA (G102) also resulted in partial monolayer disruption after 20 hours of exposure. In contrast, HUVEC monolayers exposed to the gingipain-null (CW501) and R gp-null (CW401) mutants resulted in no monolayer disruption and resembled the untreated cell m onolayers. Interestingly, the gi ngipain mutant containing both RgpA and RgpB but lacking Kgp resulted in only minimal monolayer disruption after 20 hours of exposure. These data suggest that to some extent, each gingipain is involved in stimulating HUVE cell monolayer disruption. However, it is possible that other factors (e.g., LPS) associated with P. gingivalis are involved in the disrupt ion of the HUVEC monolayer. Next, particularly high concentr ations (MOIs of 1000) of the P. gingivalis gingipain mutants were used to determine if increas ing the concentration would disrupt the HUVEC monolayers more quickly. Similar to the result s observed with the pare nt strains 381 and ATCC 33277, the HUVEC monolayers co-cultured with MT10 ( rgpA), MT-10W ( rgpA-, kgp), G102 ( rgpB), G102W ( rgpB-, kgp), and YPP2 ( kgp) at an MOI of 1000 for 5 hours resulted in partial disruption to the monolayer (Figs. 3-23 3-27). However, at this time, HUVEC monolayers treated with P. gingivalis CW401 ( rgpA-, rgpB) and CW501 ( rgpA-, rgpB-, kgp) remained

PAGE 77

77 intact (Figs. 3-28, 3-29). In addition, all of the gingipain mutants except for CW501 were observed, in varying degrees, in cl usters or aggregates, and appear ed to be attached to the cell surface and/or internalized with in the HUVEC cytoplasm at this time point (yellow arrows) (Figs. 3-23 3-28). HUVEC monolayers exposed to MT10, MT-10W, G102, G102W, and YPP2 for 10, 15, and 20 hours resulted many P. gingivalis attached to the cell surface and/or internalized within th e cell, the loss of -catenin at the cell membrane (white arrows) and a loss of adherent cells (Figs. 3-23 3-27). In contrast, monolay ers co-cultured with CW401 and CW501 remained intact through 15 hours with minimal P. gingivalis attached to or internalized within the HUVECs (Figs. 3-28, 3-29). Af ter 20 hours, HUVEC monolayers treated with CW401 ( rgpA-, rgpB) (Fig. 3-28) resulted in minimal m onolayer disruption and a nominal loss of -catenin from the cell surface (white arrow). However, the monolayers co-cultured with CW501 ( rgpA-, rgpBkgp) (Fig. 3-29) for 20 hours maintained monolayer integrity, as well as -catenin at the cell surf ace (white arrow) and were comparab le to the untreated HUVEC. This data suggests that Kgp is capable of inducing HUVE cell monolayer disruption but only at a high MOI. Taken together, these results es tablish that the gingipains are responsible for the disruption of the HUVEC monolayers. The result that the HUVEC monolayers co-cultured with CW401 ( rgpA-, rgpB) at an MOI 1000 for 20 hours were only mi nimally disrupted suggests that Rgp, but not Kgp, is essential for this activity. Mo reover, the lack of HU VEC monolayer disruption observed after treatment with CW501 further suppor ts the involvement of gingipains. However, it cannot be ruled out that the mi nimal or lack of monolayer disr uption observed af ter co-culture with the CW401 and CW501 respectively, may be due to the decreased invasi on abilities of these

PAGE 78

78 mutants and that some factor other than the gi ngipains may also play a role in monolayer disruption. Effects of Adherence Mutants of P. gingivalis from Strains W83 or ATCC 33277 on Confluent HUVEC Monolayers Data from the above experiments strongly establ ish that the gingipains are required for the disruption of the HUVEC monolayer and the loss of -catenin from the cell surface. We next sought to study whether adherence of P. gingivalis to host cells was also a requirement of P. gingivalis to effect HUVEC monolayer disr uption. To investigate this, P. gingivalis mutants with defective adherence abilities were examin ed. Adhesion mutants (Table 3-3) constructed from the wild-type strain W83, including PG1683 (conserved hypotheti cal protein), PG0242 (conserved hypothetical protein) and PG1118 (clpB), in additi on to an adhesion mutant constructed from the wild-t ype strain ATCC 33277, YPF1 ( fimA) (Love et al. 2000; Xie et al. 2000) were used for these studies. In work done by Love et al. and others in our laboratory, these mutants were tested for their adhesion abiliti es with endothelial (HCA EC) and/or epithelial (GEC or KB) cells. The conserved hypothe tical proteins PG0242 and PG1683 displayed 11.7 fold and 2.1 fold decreases in adhesion to HCAEC, respectively. The clpB mutant, PG1118, and the fimA mutant, YPF1, showed 2.8 fold and 10-fold decreases in adhesion to GECs, respectively. HUVEC monolayers co-cultured for 10 hours with all four adhesion mutants at an MOI of 100 resulted in very few P. gingivalis attached to or localized within the monol ayers (yellow arrows) (Figs. 3-30 3-33). These HUVEC mo nolayers remained visibly intact with -catenin concentrated at the cell surface (white arrows). At 15 hours, the attachment and internalization of YPF1 (Fig. 3-33) and PG0242 (Fig. 3-31) with HUVECs was minimal (yellow arrows). In contrast, numerous PG1118 (Fig. 3-32) and PG1683 (Fig. 3-30) could be detected attached to

PAGE 79

79 and internalized within the HUVEC monolayer afte r 15 hours (yellow arrows). Interestingly, the HUVEC monolayers co-cultured with each of thes e mutants for 15 hours remained intact with catenin (white arrows) at the ce ll surface (Figs. 3-30 3-33). A comparison of the HUVEC monolayers treat ed for 20 hours with all four adhesion mutants (Table 3-3) and their parent strains can be viewed in figure 3-44. At 20 hours, monolayers treated with YPF1 (F ig. 3-33) resulted in fewer P. gingivalis (yellow arrows) attached to or internalized within the HUVECs than the monolayers treated with PG1683 (Fig. 330), PG0242 (Fig. 3-31) and PG1118 (Fig. 3-32). However, HUVEC monolayers co-cultured with all of the aforementioned P. gingivalis mutants, except for PG 0242 (Fig. 3-31) retained catenin staining at the cell surfac e and generally intact monolayers (white arrows). The integrity of HUVEC monolayers co-cultured with PG0242 was disrupted and the concentration of catenin (white arrow) shifte d from the cell surface to sp arsely distribute throughout the cytoplasm (Fig. 3-31). The results with PG 1118, PG1683 and YPF1 suggest that adhesion of P. gingivalis to the HUVEC monolayer is necessary for P. gingivalis to cause monolayer disruption. However, the HUVEC monolayer disruption observed after co-culture with PG0242 refutes this and suggests that adhesion alone is not directly correlated to monolayer disruption. Effects of Invasion/Persistence Mutants of P. gingivalis from Strains W83 or ATCC 33277 on Confluent HUVEC Monolayers To determine the importance of P. gingivalis invasion of HUVECs to the disruption of the monolayers, several P. gingivalis mutants with decreased invasive capabilities (Table 3-4) were tested for their ability to effect disruption of the monolayers. YPEP ( pepO), PG1118 ( clpB), PG0717 (putative lipoprotein) and PG1286 (ferritin) were co-cultured at an MOI of 100 with confluent HUVEC monolayers to determine their effects on -catenin localization and monolayer disruption. The YPEP ( pepO) mutant resulted in a 25% decrease in invasion of

PAGE 80

80 GECs compared to its pare nt strain ATCC 33277 (Park et al. 2004; Park and Lamont, 1998). Work done by others in our laboratory reported that PG1286 and PG0717 showed a 2.0 and 2.7fold decrease in invasion of human coronary artery endothelial cells (HCAECs), respectively, while PG1118 had a 15-fold decrease in invasion of HCAECs (unpublished data). The ferritin mutant PG1286 was also determined to be a persis tence mutant since it demonstrated a 33-fold decrease in persistence co mpared to its parent strain W83 through 48 hours. HUVEC monolayers co-cultured with the P. gingivalis mutants PG1286 (Fig. 3-35) and PG1118 (Fig. 3-32) for 5 hour s resulted in very few P. gingivalis (yellow arrows) attached to or localized within the intact monolayer. In contrast, HUVEC monolayers co-cultured with the mutants PG0717 (Fig. 3-34) and YPEP ( pepO) (Fig. 3-36) resulted in numerous P. gingivalis (yellow arrows) attached to the cell surface and/or internalized, although the monolayers remained intact. HUVEC monolayers treated wi th PG1286 (Fig. 3-35) did not show numerous P. gingivalis (yellow arrows) attached to and/or locali zed within the cells until after 15 hours of treatment. In addition, the P. gingivalis clpBmutant, PG1118, resulted in only minimal attachment and invasion of the HUVEC monol ayers through 20 hours of treatment (yellow arrows) (Fig. 3-32). However, the HUVEC monolayers co-culture d with each of these mutants for 20 hours remained intact with -catenin at the cell surface (wh ite arrows) (Figs. 3-32, 3-34 3-36). A comparison of the HUVEC monolayers treated for 20 hours with all four invasion mutants and their parent strains can be viewed in figure 3-45. After 20 hours of exposure, there were no detectable disruptions of the HUVEC monolayers and -catenin remained at the cell surface. These results with the invasion mutants thus suggest that adhesion of P. gingivalis to the HUVEC monolayers is not as instrumental in causing monolayer disruption, as is P.

PAGE 81

81 gingivalis invasion. In addition, these data support the hypothesis that the P. gingivalis gingipains act from within the HUVE ce ll as well as from the outside. Effects of Additional P. gingivalis Mutants from Strain W83 on Confluent HUVEC Monolayers Additional mutants (Table 3-6) constructed from P. gingivalis W83 were examined at an MOI of 100 to determine their effects on the HUVE C monolayer. These mutants showed either no difference in invasion of HCAEC through 2.5 hours, PG0293 (putative secretion activator protein) and PG0686 (conserved hypothetical protein), or in creased invasion of HCAEC, PG1788 (putative cysteine peptidase). After 5 hours, HUVEC monolayers treated with all three mutants resulted in numerous P. gingivalis attached to and localized within the cells (yellow arrows), yet the monolayers remained intact (Figs. 3-37 3-39). At 10 hours, HUVEC monolayers exposed to PG0293 resulted in monolayer disruption and a decrease of -catenin at the cell surface (white arrow) (Fig. 3-38). However, HUVEC monolayers treated with PG1788 which resulted in a 2-fold increase in invasion of HCAECs compared to its parent strain W83 also demonstrated numerous P. gingivalis (yellow arrows) attached to and internalized within the cells yet remained intact at 5 and 10 hours of exposure. However, at 15 hours there was a decrease of -catenin at the cell surface (w hite arrow) as well as a disrupted monolayer (Fig. 337). In contrast, HUVEC monolayers treated for up to 20 hours with PG0686, which showed no difference in invasion compared to its parent strain W83, resulted in P. gingivalis (yellow arrows) attached to and/or intern alized within th e cells but maintained an intact monolayer with -catenin at the cell surface (w hite arrow) (Fig. 3-39; 3-46 ). A comparison of the HUVEC monolayers treated for 20 hours with these additi onal mutants and their parent strain can be viewed in figure 3-46. The results with PG1788 and PG0293 suggest that th ese proteins are not necessary for monolayer disruption. In cont rast, the lack of HUVE C monolayer disruption

PAGE 82

82 observed with PG0686, which adheres and invades as well as its wild-t ype parental strain, indicates that this conserved hypothetical protein is critical for monolayer disruption. Effects of Additional P. gingivalis Mutants at Higher MOIs The data reported above sugge sts that adhesion and invasion are involved in one possible mechanism that P. gingivalis utilizes to induce HUVEC monolayer disruption and -catenin relocalization. At an MOI of 100, the adhesion mutant, YPF1 ( fimA) (Table 3-3), the invasion mutant, YPEP ( pepO) (Table 3-4), and PG0686 (Table 3-6) did not cause a disruption of the HUVEC monolayers. Therefore, MOIs of 1000 we re examined to observe if increasing the concentration of P. gingivalis could induce monolayer disr uption. Treatment of HUVEC monolayers with YPF1, YPEP and PG 0686 for 5 hours resulted in numerous P. gingivalis (yellow arrows) attached to and internalized with in the cells (Figs. 3-40 3-42). However, the HUVEC monolayers remained intact after treatment for 5 hours with YPF1 ( fimA) (Fig. 3-40) and PG0686 (Fig. 3-42) and -catenin remained concentrated at the cell surface (white arrows). In contrast, HUVEC monolayer s treated with YPEP ( pepO) at this time point were partially disrupted and maintained minimal -catenin (white arrow) at th e cell surface (Fig. 3-41). Additional treatments for 15 hours with all three P. gingivalis mutants demonstrated that the integrity of the HUVEC monolayer was altered in that -catenin was no longer visualized at the cell surface. Also, numerous P. gin givalis (yellow arrows) were observed attached to the HUVE cell surface and internali zed within the cytoplasm (Figs. 3-40 3-42). After 20 hours, the HUVEC monolayers treated with YPEP a nd YPF1 were no longer intact and -catenin was no longer associated with the cell surface. These monolayers were severely disrupted suggesting that numerous cells had become detached. P. gingivalis (yellow arrows) was also observed attached to the HUVE cell surface as well as internalized with the cytoplasm (Figs. 3-40, 3-41).

PAGE 83

83 Similarly, the HUVEC monolayers treated w ith PG0686 for 20 hours also demonstrated numerous P. gingivalis (yellow arrow) attached to the ce ll surface and internalized within the cells (Fig. 3-42). Most significantly, howev er, the HUVEC monolayer was only partially disrupted. The effects observed with YPF1 ( fimA) and YPEP ( pepO) at an MOI of 1000 after 20 hours were comparable to their parent stra in ATCC 33277 (Fig. 3-40, 3-41). These results suggest that both adhesion and invasion impaired P. gingivalis mutants at high enough concentrations (MOI of 1000) are ca pable of altering the location of -catenin from the HUVE cell surface to within the cytoplasm, as well as inducing monolayer disruption. However, HUVEC monolayers co-cultured with PG0686 for 20 hours remained mostly intact, although catenin did not remain concentrat ed at the cell surface (Fig. 3-42) The effects observed with the PG0686 conserved hypothetical protein mutant at an MOI of 1000 further suggest that this protein is critical for P. gingivalis to induce HUVEC monolayer disruption. Invasion The purpose of the following experiments was to determine the effects on the ultrastructure of HUVECs exposed to strain 381 containing gingipains, compar ed to strain CW501, lacking gingipains. To investigate th is, confluent HUVEC monolayers were co-cultured with CW501 (gingipain-null) mutant and its parent stra in 381, at an MOI of 100 for 5 and 20 hours. Transmission electron microscopy (TEM) was used to observe the interactions of P. gingivalis with HUVECs. The untreated HUVECs dem onstrated normal morphology, in that they displayed an intact cell membrane and nucleus as well as a diffuse distribution of organelles (Figs. 3-47, 3-48). Exposu re of HUVEC monolayers to P. gingivalis CW501 for 20 hours was comparable to the untreated cells with the ex ception of the presence of a few empty vacuoles

PAGE 84

84 (Fig. 3-48). Also, very few CW501 bacteria were observed at the cell surface (white arrows). In contrast, co-culture of HUVEC monolayers with strain 381 demonstrated considerably different results. After 5 and 20 hours, many P. gingivalis were observed at the HUVEC surface (white arrows) and internalized within these cells (yellow arrows) (Fig. 3-47). Strain 381 appeared to be primarily localized within vacuoles, possibly autophagi c vacuoles, in the HUVEC cytoplasm (yellow arrows) (Fig. 3-47). Thus, P. gingivalis 381 invasion of HUVECs resulted in morphological alterations in the host cell, specifi cally the appearance of vacuoles containing P. gingivalis (yellow arrows) In contrast, with the exception of a few additional empty vacuoles, there was no apparent disr uption of HUVEC morphology and no internalization of CW501 through 20 hours of co-culture (Fig. 3-48). Ta ken together, the morphological alterations observed with 381 and not with CW501 suggest that P. gingivalis gingipains play a role in this activity. Increasing the MOI of both P. gingivalis 381 and CW501 to 1:1000 had an affect on the morphology of the HUVECs (Fig. 3-49). HUVECs co-cultured with strain 381 for 20 hours showed numerous P. gingivalis 381 located around the periphery (white arrows) of the cells. However, most of the bacteria were internalized within cytoplasmic vacuoles (yellow arrows) (Fig. 3-49). Compared to the untreated cells, th e HUVECs co-cultured with strain 381 exhibited cellular distortions including the presence of numerous empty vacuoles (black arrows) and vacuoles containing P. gingivalis (yellow arrows), as well as invaginations of the cell membrane. In contrast, HUVECs co-c ultured with CW501, demonstrated very few P. gingivalis located around the cell periphery (white ar rows) or internalized with th e cells (yellow arrows) (Fig. 349). With the exception of an increased number of empty vacuoles (black arrows) within the HUVECs co-cultured with CW501, the cellula r morphology resembled the untreated cells

PAGE 85

85 through 20 hours of co-cultu re (Fig. 3-49). These results suggest that the gingip ains are likely involved in effecting invaginations of the cell membrane since th e increased concentration (MOI of 1000) of the gingipain-null mutant, CW501, did not have the same eff ect. However, this concentration of CW501 did result in the appearance of additional vacuoles, most of which were devoid of P. gingivalis In addition, since CW501 does not invade, we cannot rule out that P. gingivalis invasion is a requirement for the obs erved alterations in HUVEC morphology. In previous experiments, the conserved hypothetical protein mutant PG0686 showed no difference in adhesion or invasion abilities compared to its pare nt strain and demonstrated no disruption to the HUVEC monolayer at an MOI of 100 through 20 hours. Therefore, this mutant was also tested to determine its effects on the HUVEC morphology. After 20 hours of co-culture of HUVEC monolayers and PG0686 at an MOI of 1000, numerous P. gingivalis were observed both around the cell periphery (white arrows), as well as internal ized (yellow arro ws) within the cells (Fig. 3-49). Elevated numbers of PG0686 we re localized within mu ltiple vacuoles (yellow arrows) throughout the HUVECs cytoplasm (Fig. 349). However, the cell membrane of the PG0686 treated HUVECs remained intact, not invagina ted like the cells treate d with strain 381. In addition, with the exception of incr eased cytoplasmic vacuoles containing P. gingivalis (yellow arrows), the HUVECs exposed to PG0686 for 20 hours appeared to resemble the untreated cells (Fig. 3-49). These results show that P. gingivalis 381 and PG0686 are capable of internalizing and persisting within the HUVECs for extended times. These strains appear to be primarily localized within intracellular vacuoles in the cytoplasm. In addition, prev ious experiments suggested that P. gingivalis invasion may be necessary for monolaye r disruption since CW501 and the invasion mutants of P. gingivalis, YPEP, PG1118, PG0717, and PG1286, dem onstrated no disruption of

PAGE 86

86 the HUVEC monolayer and very little internaliza tion within the cell. Taken together, these results suggest that the P. gingivalis gingipains may be acting upon the HUVEC monolayer from within the cell. However, HUVEC monolayers tr eated with the conserved hypothetical protein mutant PG0686 at an MOI of 100, which showed no difference in adhe sion or invasion when compared to its parent strain W83, also resulted in a lack of monolayer disruption. This data indicates that the gingipains an d the PG0686 protein are both crit ical for inducing disruption of the HUVEC monolayers. Cell Death Cell Death of HUVECs after Exposure to P. gingivalis The previous experiments demonstrated that (1) exposure of HUVEC monolayers to certain P. gingivalis strains caused disruption of the monolay ers and cell detachment and (2) that the lysate and SPF of P. gingivalis induced epithelial and endothelial cell detachment from monolayers. These data sugge st that a soluble protein or proteins secreted from P. gingivalis (e.g., gingipains) are responsible fo r this observed activity. Thus, we next wanted to examine the fate of the detached cells from the HUVEC monolayer. We sought to determine if P. gingivalis was capable of inducing HUVE cell death a nd if so, was this death a result of P. gingivalis components acting from outside or from within the cells. To investigate this, HUVEC monolayers were co-cultured with different strains of P. gingivalis at an MOI of 100 for 24 hours Included among the P. gingivalis strains tested were mutants CW501 (gingipain-null) and YPF1 ( fimA). Both of these mutants showed reduced abilities to attach to the HUVECs and were una ble to induce HUVEC monolay er disruption at an MOI of 100. In addition, their parent strains, 381 and ATCC 33277, respectively, were also used in these experiments. Cell death was assessed using propidium iodide, which stains the DNA of leaky or permeable cells, and quantified using flow cytometry.

PAGE 87

87 The results through 2.5 hours revealed no signifi cant differences in cell death of the HUVECs co-cultured with any of the P. gingivalis strains compared to the untreated cells (Fig. 3-50). However, by 12 hours, HUVECs treated with ATCC 33277 and 381 resulted in a 12% and 13% (p<0.05) increase in cell death, respec tively. HUVECs exposed to CW501 and YPF1 resulted in approximately 5% cell death, a leve l approximating that of the untreated cells. Exposure of HUVEC monolayers to strains AT CC 33277 and 381 for 24 hours resulted in an 18% and 21% increase in cell death, respectively (F ig. 3-50). The increase in cell death after 24 hours of HUVEC exposure to strains 381 and ATCC 33277 was significant (p<0.001) and (p<0.01) respectively, relative to the 13% cell death observed after exposure to strains CW501 and YPF1, and the 11% cell death of untreated cells (Fig. 3-50). The resu lts with the gingipain negative mutant, CW501, suggest that the gi ngipains are involved in inducing HUVEC cell death. In addition, the results with CW501 and YPF1, which are known to have decreased adhesion and invasion abilities, lik ely due to their lack of the major attachment protein FimA, suggest that P. gingivalis adhesion and/or subsequent inva sion of HUVEC cell monolayers are also necessary for P. gingivalis stimulated cell death. Given that adherence of P. gingivalis to HUVECs has been shown to be necessary for P. gingivalis invasion, we next wanted to determine if invasion was a requirement for P. gingivalis induced cell death. Internalized P. gingivalis were examined for their ability to induce HUVE cell death levels as reported with the previous assay. For these experiments, HUVEC monolayers were again treated with the gi ngipain-null mutant CW501, the major fimbriae (FimA) mutant YPF1, and their parent strain s 381 and ATCC 33277, respec tively, at an MOI of 100. After 1.5 hours of treatment, the bacteria l cells were removed and the HUVECs were washed and subjected to antibiotics (refer to materials and methods) in order to kill any

PAGE 88

88 remaining extracellular P. gingivalis Cells were then incubated in the presence of the antibiotics for the remainder of the 2.5, 12, and 24-hour time poi nts. Consistent with the results obtained previously, there were no stat istical differences in cell deat h between the untreated HUVEC monolayers and any of th e monolayers exposed to P. gingivalis for 2.5 hours (Fig. 3-51). However, at 12 hours of treatment with 381 and ATCC 33277, 24% a nd 23% cell death was observed, an increase of 8% (p<0.05) and 7%, resp ectively, compared to th e untreated cells (Fig. 3-51). HUVEC monolayers treated with CW501 and YPF1 followed the same pattern of cell death (approximately 16%) as the untreated ce lls through 24 hours. In contrast, HUVECs exposed to 381 and ATCC 33277 for 24 hours re sulted in approximately 36% cell death (p<0.001) relative to the 16% deat h observed with the untreated cel ls and the two mutant strains (Fig. 3-51). This 20% increase in cell deat h observed with strains 381 and ATCC 33277 is consistent with those levels reported in the pr evious assay. These resu lts suggest that the cell death observed at 24 hours of co-culture is most likely stimulated by the internalized P. gingivalis After repeating this experiment with stra in 381 and fresh HUVE cells, there was an insignificant difference between the levels of cel l death observed after exposure to strain 381 with and without antibiotics (Fig. 3-54). Th ese results suggest that the interaction of P. gingivalis 381 with the HUVECs before antibiotic prot ection was not sufficient to induce/cause cell death equal to that observed with pulsed P. gingivalis co-culture. In addition to the above data, HUVECs were exposed to strain CW501 at an MOI of 1000 to investigate whether the increas ed concentration could affect cell death. Compared to the untreated cells, there was no signi ficant difference in the numbers of dead cells upon co-culture with strain CW501 at an MOI of 1000 for 2.5 a nd 12 hours (Fig. 3-54). However, 24 hours of

PAGE 89

89 exposure of HUVECs to CW501 at this concentrat ion resulted in a 6% increase in cell death compared to the untreated cells (p<0.001) (Fig. 3-54). These results suggest that P. gingivalis may be equipped with mechanisms in addition to the gingipains for inducing HUVE cell death since the gingipain-null mutant, CW501, caused ce ll death albeit only af ter 24 hours of exposure at an MOI of 1000. Inhibition of P. gingivalis Internalization of HUVEC The next sets of experiments were desi gned to determine if the internalized P. gingivalis were responsible for inducing the observ ed HUVE cell death. To test this, P. gingivalis invasion of HUVECs was inhibited us ing a known inhibitor of P. gingivalis internalizati on, cytochalasin D. HUVECs were pretreated with 5 g/ml cytochalasin D for 30 minutes prior to exposure to P. gingivalis 381 and ATCC 33277 at an MOI of 100. Si nce it is possible that HUVE cells may react adversely in response to the presence of antibio tics, this experiment was conducted with and without antibiotics. The results from thes e experiments showed th ere were no significant differences in the levels of ce ll death observed between the untr eated cells and cytochalasin D treated cells exposed to strains 381 and ATCC 33277 through 24 hours (Fig. 3-52, 3-53). Thus, exposure to cytochalasin D completely inhibited the P. gingivalis induced cell death (Fig. 3-52, 3-53). HUVE cells treated with antibiotics produ ced the same results as those treated without antibiotics, suggesting th at the presence of antibiotics did not cause any adverse affects on the cells. Unless cell death requires factors inhibited by cytochalasin D, these data establish that inhibition of P. gingivalis internalization suppresse d its ability to induce cell death of HUVECs. P. gingivalis Secreted Protein Fraction (SPF) and HUVE cell death As described earlier, proteins secreted from P. gingivalis (SPFs) are capable of disrupting HUVEC monolayers. Thus the following experi ments were designed to determine if the P. gingivalis SPF was also capable of stimulating HUVEC death. A 0.02 g/ml concentration of

PAGE 90

90 the 381 SPF was used for these experiments because this concentration was shown in previous experiments to cause disruption of the HUVEC monolayer proportional to that of an MOI of 100. Exposure of HUVE cells to 0.02 g/ml of the 381 SPF at 2.5 and 12 hours resulted in cell death levels similar to the untreated cells. However, 24 hours of HUVECs exposure to 0.02 g/ml of the 381 SPF resulted in a 3% increase in cell death compared to the untrea ted cells (p<0.05) (Fig. 3-54). Thus, 0.02 g/ml of the 381 SPF was not able to i nduce the same level of cell death as observed with the live strain 381. These results suggest that the gingip ains are involved in HUVE cell death but that invasion of P. gingivalis is a requirement for P. gingivalis to induce optimal levels of cell death. However, since gingip ain inhibitors were not used to confirm this, it is possible that some other component of the SPF is responsible for this activity. Thus, these data suggest that P. gingivalis is inducing its activity from within the HUVE cell. HUVEC Exposure to P. gingivalis Results in Caspase 3 Activity The previous experiments were done using propidium iodide and flow cytometry to quantitate total HUVE cell death. In order to better characterize the cell death o ccurring within the HUVEC monolayers, caspase 3 activity, a ma rker for apoptosis, was quantitated. Staurosporine (STS), an apoptosis inducer that functions by inhibiting protein kinase C (PKC), at a concentration of 0.5 m, was used as a positive control. Compared to the untreated cells, no significant differences were observed when HUVEC monolayers were exposed to P. gingivalis 381 or ATCC 33277 at an MOI of 100 with or wi thout antibiotic treatm ent for 2.5 hours (Fig. 355). The STS-treated HUVEC control showed a 6.7-fold and 6.2-fold increase (p<0.001) in caspase 3 activity, respectively, compared to the untreated cells at 12 and 24 hours. After 12 and 24 hours, there were no significant increases in caspase 3 activity in HUVECs treated with 381 and ATCC 33277 in the presence of antibiotics. In contrast, afte r 12 hours, caspase 3 activity of HUVECs exposed to strains 381 and ATCC 33277 w ithout antibiotics increased approximately

PAGE 91

91 25% (p <0.05) compared to the untreated cells After 24 hours, HUVECs exposed to strains 381 and ATCC 33277 without antibiotic s resulted in a 2.7-fold (p< 0.001) and 2.4-fold (p<0.01) increase in caspase 3 activity respectively, compar ed to the untreated cells Therefore, it appears that the cell death observed in the flow cytome try experiments was coincident with caspaserelated apoptosis. Given the results presented here with the HUVE cells co-cultured with strains 381 and ATCC 33277 at an MOI of 100 with and without antibiotic exposure (t o kill any remaining extracellular P. gingivalis ), it is possible that this endothelial cell undergoes more than one type of cell death. The results reported using propidium iodide and flow cytometry showed that there were significant increases in cell death compar ed to the untreated cells when HUVECs were exposed to strains 381 and ATCC 33277 with and without antibiotic treatment. However, only the HUVECs treated with strains 381 and AT CC 33277 without antibiotics caused a significant increase in caspase 3 activity compared to the untreated cells. Therefore, the P. gingivalis internalized within the cells be fore antibiotic exposure may induce a caspase independent form of cell death while the constant exposure of P. gingivalis to HUVECs appears to stimulate caspase 3 dependent apoptosis. Transepithelial Resistance Bacterial pathogens must first surpass the epithe lial barrier, the hosts first line of defense, before they are able to penetrate deep er tissues and cause disease (Kazmierczak et al. 2001). The paracellular permeability or the transepithelial resistance ( TER) of epithelia l cell monolayers has been shown to be modified by several pathog ens (Sears, 2000). Thus, the evaluation of the TER of endothelial cells after incubation with P. gingivalis would provide additional information concerning the ability of this pathogen to alter the endothelial cell barrier.

PAGE 92

92 These experiments were designed to determin e whether the effects of adherent/invasive strains of P. gingivalis 381 and W83, on TER were different than the effects caused by the non adhesive/invasive gingipain-null mutant, CW 501. Initial experiments using HUVE cell monolayers resulted in inconsiste nt TER readings. To determin e if the HUVECs were producing tight junctions, IMF experime nts using HUVEC monolayers and an antibody against the tight junction protein occludin were conducted. Thes e IMF experiments resulted in no apparent occludin staining, suggesting no tight junctions were present. Af ter determining the endothelial HUVEC were not suitable cells on which to conduc t TER experiments, an epithelial cell line, HuH7, which displayed occludin staining, was utilized as an a lternative cell type (Fig 3-1). HuH7 cells were allowed to seed on Corning transwell filters w ith a pore size of 0.3 m. The HuH7 monolayers were then trea ted for various times with P. gingivalis strains 381, W83, and CW501 at an MOI of 100 from e ither the apical (Fig. 3-56) or basolateral side (Fig. 3-57). The TER of the cell free insert (b aseline TER) was subtracted from each time point to determine the true level of the TER. Apical exposure of HuH7 cell monolayers to all three P. gingivalis strains for 12 hours resulted in no significant differences in the TER of the HuH7 cells relative to the untreated cells. In contrast, at 24 hours, the TER increas ed after apical exposure of the HuH7 cell monolayers to each of the P. gingivalis strains. However, only the 43% increase in the TER of the HuH7 monolayers apically exposed to strain W83 wa s significant (p<0.05) rela tive to the untreated HuH7 cells (Fig. 3-56). At 48 hours, th ere was a slight decr ease in the TER levels of the apically exposed HuH7 cell monolayers treated with strains W 83 and CW501 compared to the untreated cells. This decrease in TER was followed by an additional decrease in TERs after 72 hours (Fig. 3-56). However, these observed decreases in the TERs with strains W83 and CW501 were not

PAGE 93

93 statistically significant. Similarly, af ter 48 hours of apical exposure of HuH7 cell monolayers to strain 381, there was a moderate, but not signifi cant, decrease in the TER relative to the untreated cells. After 72 hours, in contrast to the TER results with strains W83 and CW501, apical exposure of HuH7 cells to strain 381, resulted in a sharp 61% decrease in the TER. This decrease was highly signif icant (p<0.01) compared with the untreated HuH7 cells (Fig. 3-56). Basolateral exposure of HuH7 cell monolayers to P. gingivalis W83, 381 and CW501 at an MOI of 100 resulted in no significant differen ces in TERs through 3 hours. Although the TERs of the basolateral exposed HuH7 monolayers with W83 were slight ly lower than with strains 381 and CW501, they were not significant, compared to the untreated cells (F ig. 3-57). After 12 and 24 hours, basolateral exposed HuH7 cell monolayers to W83 and CW501 had a TER similar to the untreated cells. There was a slight, though non-significant increas e in the TER of the monolayers exposed to P. gingivalis 381. After 48 and 72 hours, basolateral exposure of HuH7 cell monolayers to all three P. gingivalis strains resulted in decreased TERs (Fig. 3-57). However, only after 72 hours of basolateral exposure of HuH7 cells with strains 381 and W83, but not CW501, were the observed decreases in TER significant. Basolateral exposure of HuH7 cell monolayers to strains W83 and 381 resulted in 40% (p<0.01) and 52% (p<0.001) decreases in TER respectively, compared to th e untreated cells (Fig. 3-57). These data indicate that P. gingivalis disrupts the epithelial barrier function. Unexpectedly, the HuH7 basolateral surfaces appear to be mo re vulnerable to this deterioration than the apical surface. The effects on the TERs of ba solateral exposure of HuH7 cell monolayers to P. gingivalis strains suggest that an intimate interaction between the bacterium and the HuH7 cells is not required for the observed e ffects, since the membrane is located between the basolateral side a nd the bacteria. Furthermore, these results suggest that the

PAGE 94

94 contributions of bacterial soluble factors, perh aps gingipains, which are able to disseminate through the membrane pores to elicit these TER e ffects, are the active component. The lack of significant differences in the TERs of apical or basolateral exposed HuH7 cell monolayers to strain CW501 (gingipain-null) further suggest s that the gingipains ar e responsible for the epithelial and endothelial cell monolayer disrupti on observed with IMF. The drop in TER after 48 and 72 hours of apical (Fig. 3-56) a nd basolateral (Fig. 3-57) treated HuH7 cell monolayers could be the result of HuH7 cell death caused by P. gingivalis exposure.

PAGE 95

95 Table 3-1. Comparison of wild-type P. gingivalis strains and their effects on the HUVEC monolayer. Wildtype strain Adhesion Invasion PersistenceCelltype Assayed Monolayer disruption (HUVEC) MOI 100 381 Yes Yes Yes HUVEC GEC HCAEC KB Yes 10 hours W83 Yes Yes Yes HUVEC GEC HCAEC KB Yes 10 hours ATCC 33277 Yes Yes Yes HUVEC GEC HCAEC KB Yes 20 hours AJW4 Yes No No HUVEC HCAEC KB Yes 15 hours Table 3-2. Comparison of P. gingivalis gingipain mutants and th eir effects on the HUVEC monolayer. Strain Mutated genes Parent strain Monolayer disruption (HUVEC) MOI 100 MT10 rgpA 381 Yes 15 hours MT10W rgpA, kgp 381 Yes 15 hours G102 rgpB 381 Yes 20 hours G102W rgpB, kgp 381 Yes 15 hours YPP2 kgp ATCC 33277 Yes 20 hours CW401 rgpA, rgpB 381 No CW501 rgpA, rgpB, kgp 381 No

PAGE 96

96 Table 3-3. Comparison of P. gingivalis adhesion mutants and th eir effects on the HUVEC monolayer. TR ID /mutant notation Gene name Parent strain Adhesion Celltype assayed Monolayer disruption (HUVEC) MOI 100 Comments PG0242b Conserved hypothetical protein W83 11.7-fold decrease HCAEC Yes 20 hours 81% similar to tetrapyrrole methylases. Likely involved in protoporphyrin and protohaem metabolism. PG1118c clpB W83 4.9-fold decrease 2.1-fold decrease 2.8-fold decrease KB HCAEC GEC Noa Chaperone belonging to the AAA+ protein superfamily. Generally drives the assembly and disassembly of protein cplxs by ATPdep remodeling of protein substrate. ClpB mutants of L. monocytogenes and S. typhimurium were found to have decreased virulence. PG1683b Conserved Hypothetical protein W83 2.1-fold decrease HCAEC Noa In an operon with PG1682. Shows 68% homology with amylases. amlylases are involved in coaggregation of P.g with other oral bacteria. YPF1 d Major fimbriae protein FimA ATCC 33277 2-fold decrease GEC Noa FimA is the major fimbrial protein of P.g ., primary function is attachment. a No observed monolayer disruption after 20 hours. b Paulo Rodrigues, unpublished data. c Lihui Yuon, unpublished data. d Love et al. 2000; Xie et al. 2000.

PAGE 97

97 Table 3-4. Comparison of P. gingivalis invasion mutants and th eir effects on the HUVEC monolayer. TGIR ID /mutant notation Gene name Parent strain Invasion (2.5h) Cell-type assayed Monolayer disruption (HUVEC) MOI 100 Comments PG0717b Putative lipoprotein W83 2.7-fold decrease HCAEC Noa No homology PG1286b Ferritin W83 2.0-fold decrease HCAEC Noa One of the intracellular iron storage proteins and may also contribute to the protection of organisms against oxidative stresses. PG1118c clpB W83 51.4-fold decrease 15-fold decrease 197-fold decrease KB HCAEC GEC Noa Chaperone which belongs to the AAA+ protein superfamily. Generally drives the assembly and disassembly of protein cplxs by ATP-dep remodeling of protein substrate. ClpB mutants of L. monocytogenes and S. typhimurium were found to have decreased virulence. YPEP d Endopeptidase PepO ATCC 33277 25-fold decrease GEC Noa Homology to the endothelin converting enzyme ECE-1 which converts Big endothelin in the potent vasoconstrictor ET-1. a No observed monolayer disruption after 20 hours. b Paulo Rodrigues, unpublished data. c Lihui Yuon, unpublished data. d Park et al., 2004; Park and Lamont, 1998.

PAGE 98

98 Table 3-5. P. gingivalis persistence mutant and its effects on the HUVEC monolayer. TIGR ID /mutant notation Gene name Parent strain Persistence (24h) Persistence (48h) Celltype assayed Monolayer disruption HUVEC MOI 100 Comments PG1286b Ferritin W83 10-fold decrease 33.3-fold decrease HCAEC Noa One of the intracellular iron storage proteins and may also contribute to the protection of organisms against oxidative stresses. a No observed monolayer disruption after 20 hours. b Paulo Rodrigues, unpublished data.

PAGE 99

99 Table 3-6. Comparison of other mutants of P. gingivalis and their effects on the HUVEC monolayer. Other Mutants Gene Name Parent strain Invasion (2.5 h) Persistence (48 h) Cell-type assayed Monolayer disruption (HUVEC) MOI 100 Comments PG1788c Putative cysteine peptidase W83 2-fold increase Not tested HCAEC Yes 15 hours Proteases play important roles in nutriet acquisition, tissue invasion and modulation of host immune defense. PG0293c Putative secretion activator protein W83 No difference Not tested HCAEC Yes 10 hours Homologous to B. melitensis secretion activator protein. Secreted proteins may play an important role in virulence. PG0686b Conserved hypothetical protein W83 No difference No difference HCAEC Noa Located in an operon with two genes involved in the metabolism of succinylCoA. This gene could be important in acid neutralization of the autophagosomes. a No observed monolayer disruption after 20 hours. b Paulo Rodrigues, unpublished data. cSheila Walters, unpublished data.

PAGE 100

100 A B C D E F Figure 3-1. Effects of P. gingivalis W83 lysate on the HuH7 junctional proteins. HuH7 cells incubated for 24 hours with anti-Pan-cadherin (A) anti-catenin (B), anti-occludin (C) untreated controls and 0.4 mg/ml E. coli lysate (negative control) (D). HuH7 cells incubated with anti-Pan-cadherin (red) or anti-catenin (red) after treatment with 0.2 mg/ml of the P.g lysate for 24 hours (E & F, respectively). Arrows indicate the cellula r junctions. Magn ification: A-E 340x.

PAGE 101

101 Figure 3-2. Effects of P. gingivalis W83 lysates on HUVEC junctiona l proteins. (A) untreated control incubated with anti-catenin, (B) HUVEC cells tr eated with 0.05 mg/ml of the P.g W83 lysate and (C) 0.1 mg/ml of the P.g .W83 lysate for 24 hours. Arrows indicate the cellula r junctions (A). Magnification: A-C 440x. A B C D Figure 3-3. Effects of temperat ure and TLCK treatment on the proteolytic activity of live P. gingivalis and the P.g lysate. HuH7 cell monolayers were treated with the P.g W83 lysate or live P.g. and incubated with an antibo dy against Pan-cadherin. (A) untreated HuH7 control, (B) HuH7 cells treated with 0.4 mg/ml P.g W83 lysate that had been heated for 20 minutes at 60oC, (C) HuH7 cells treated with W83 lysate incubated with 1 mM of the cysteine prot einase inhibitor TLCK for 24 hours. (D) HuH7 cells incubated with anti-Pancadherin after treatment with live P. gingivalis W83 (MOI of 1000) + 1 mM TLCK for 24 hours. Arrows indicating cell junctions (A, B) and the absence of cell junctio n staining (C). Magnification: A-D 340x. A B C

PAGE 102

102 A B C D Figure 3-4. Effects of temperat ure and TLCK treatment on the proteinase activity of live P. gingivalis and P.g lysate on HuH7 cells. All cells were incubated with an antibody against -catenin. (A) untreated HuH7 control, (B) HuH7 cells treated with 0.4 mg/ml P.g. W83 lysate that had been heated for 20 minutes at 60oC, (C) HuH7 cells treated with 0.4 mg/ml W83 lysate incubated with 1 mM of the cysteine proteinase inhibitor TLCK for 24 hours. (D) HuH7 cells incubated with anti-Pan-cadherin after treatment with live P. gingivalis W83 (MOI of 1000) + 1 mM TLCK for 24 hours. Arrows indicate cell junctions (A, B) and the absence of cell junction staining (C). Magnification: A-D 340x.

PAGE 103

103 A B C D E F Figure 3-5. Effects of varying concentrations of SPF from P. gingivalis W83 and 381 on HUVE cell adhesion. HUVECs were treated with the secreted protein factor (SPFs) of P. gingivalis W83 and 381 for 24 hours and stained using antibodies against -catenin (red). (A) untrea ted cells, (B) 0.2 g/ml of the P. gingivalis W83 SPF, (C) 2 g/ml of the P.g W83 SPF, (D) 22 g/ml of the P.g. W83 SPF, (E) 0.02 g/ml of the P.g. 381 SPF, and (F) 1 g/ml of the P.g 381 SPF. Arrows indicate cell junctions (A, B, E) and a lack of cell junction stai ning (C, E). Magnification: A-F 440x.

PAGE 104

104 A B C D E Figure 3-6. Effects of varying concentrations of SPF from P. gingivalis W83 and 381 on HUVE cell adhesion. HUVEC cells were treated wi th the secreted protein factor (SPF) from strains W83 and 381 for 24 hours a nd stained using antibodies against Pancadherin (red). (A) untreated cells, (B) 0.2 g/ml of the P. gingivalis W83 SPF, (C) 22 g /ml of the P.g W83 SPF, (D) 0.02 g/ml of the P.g 381 SPF, and (E) 1 g/ml of the P.g. 381 SPF. Arrows indicate cell juncti ons (A) and a lack of cell junction staining (B, D). Ma gnification: A-F 440x.

PAGE 105

105 A B C D Figure 3-7. Junctional complexes of untreated HUVEC monolayers remain intact through 20 hours. HUVEC monolayers were stained with antibodie s against the junctional protein -catenin (green) and DAPI (blue) (nuc lear labeling) after (A) 5 hours, (B) 10 hours, (C) 15 hours, and (D) 20 hours. Arro ws indicate the junctions between the cells. Magnification: A-C, 440x; D, 580x.

PAGE 106

106 A B C D E Figure 3-8. Junctional labeling of HUVEC mo nolayers after treatment with the P. gingivalis strain W83 at an MOI of 100. The untre ated HUVEC monolayer as shown in (A) was labeled with antibodies ag ainst the junctional protein -catenin (green), antibodies against P.g. HagB (red), DAPI, a nuclear stai n (blue) and viewed by IMF. HUVEC monolayers treated with P.g. for 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arro ws indicate the junc tions between the cells. Magnification: A-C, 440x; D-E, 580x.

PAGE 107

107 A B C D E Figure 3-9. Junctional labeling of HUVEC mo nolayers after treatment with the P. gingivalis strain 381 at an MOI of 100. The untreat ed HUVEC monolayer as shown in (A) was labeled with antibodies against the junctional protein -catenin (green ), antibodies against P.g. HagB (red), DAPI, a nuclear stain (blue) and viewed by IMF. HUVEC monolayers treated with P.g. for 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arrows indicate the junc tions between the cells. Magnification: A, 440x; B-C 480x; D, 580x; E, 440x. C.10hours

PAGE 108

108 A B C D E Figure 3-10. Junctional labeling of HUVEC monolayers after treatment with the P. gingivalis strain ATCC 33277 at an MOI of 100. Th e untreated HUVEC monolayer as shown in (A) was labeled with antibodies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), DAPI, a nuclear stain (blue) and viewed by IMF. HUVEC monolayers treated with P.g. for 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arro ws indicate the junc tions between the cells. Magnification: A, 500x; B, 340x; C, 440x, D, 340x; -E, 440x.

PAGE 109

109 A B C D E Figure 3-11. Junctional labeling of HUVEC monolayers after treatment with the P. gingivalis strain AJW4 at an MOI of 100. The untre ated HUVEC monolayer as shown in (A) was labeled with antibodies ag ainst the junctional protein -catenin (green), antibodies against P.g. HagB (red), DAPI, a nuclear stain (blue) and viewed by IMF. HUVEC monolayers treated with P.g. for 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arro ws indicate the junc tions between the cells. Magnification: A, 440x; B-C, 480x; D-E, 500x.

PAGE 110

110 A B C D E Figure 3-12. Junctional labeling of HUVEC monolayers after treatment with the P. gingivalis wild-type strain W83 at an MOI of 1000. The untreated HUVEC monolayer as shown in (A) was labeled with antib odies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), DAPI, a nuclear stain (blue) and viewed by IMF. HUVEC monolayers treated with P.g. for 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E ). Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White a rrows indicate the junctions between the cells. Magnification: A-B, 580x; C-E, 440x.

PAGE 111

111 A B C D E Figure 3-13. Junctional labeling of HUVEC monolayers after treatment with the P. gingivalis strain 381 at an MOI of 1000. The untre ated HUVEC monolayer as shown in (A) was labeled with antibodies ag ainst the junctional protein -catenin (green), antibodies against P.g. HagB (red), DAPI, a nuclear stain (blue) and viewed by IMF. HUVEC monolayers treated with P.g. for 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arro ws indicate the junc tions between the cells. Magnification: A; 440x; B, 580x; C, 440x; D-E, 580x.

PAGE 112

112 A B C D E Figure 3-14. Junctional labeling of HUVEC monolayers after treatment with the P. gingivalis strain ATCC 33277 at an MOI of 1000. The untreated HUVEC monolayer as shown in (A) was labeled with antib odies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), DAPI, a nuclear stain (blue) and viewed by IMF. HUVEC monolayers treated with P.g. for 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E ). Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White a rrows indicate the junctions between the cells. Magnification: A, 580x; B-C, 440x; D, 580x; E, 340x.

PAGE 113

113 A B C D E Figure 3-15. Junctional labeling of HUVEC monolayers after treatment with the P. gingivalis strain AJW4 at an MOI of 1000. The unt reated HUVEC monolayer as shown in (A) was labeled with antibodies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), DAPI, a nuclear stain (blue) and viewed by IMF. HUVEC monolayers treated with P.g. for 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arro ws indicate the junc tions between the cells. Magnification: A, 500x; B-E, 440x.

PAGE 114

114 A B C D E F Figure 3-16. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis MT10 ( rgpA) at an MOI of 100. The HUVEC monol ayers were labeled with antibodies against the junc tional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and obs erved by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-typ e parent strain 381 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or local ized within the HUVECs. White arrows indicate the junctions between th e cells. Magnification: A, 580x; B, 500x; C, 440x; D-E, 500x; F, 580x. B. W83

PAGE 115

115 A B C D E F Figure 3-17. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis MT10W ( rgpA-, kgp) at an MOI of 100. The HUVE C monolayers were labeled with antibodies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stai n (blue) and observed by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treate d with the wild-type parent strain 381 at an MOI of 100 for 20 hours (positive co ntrol) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arrows indica te the junctions between the cells. Magnification: A, 500x; B-C, 480x; D-F, 500x.

PAGE 116

116 A B C D E F Figure 3-18. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis G102 ( rgpB) at an MOI of 100. The HUVEC monol ayers were labeled with antibodies against the junc tional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and obs erved by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-typ e parent strain 381 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or local ized within the HUVECs. White arrows indicate the junctions betwee n the cells. Magnification: A-F, 580x.

PAGE 117

117 A B C D E F Figure 3-19. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis G102W ( rgpB-, kgp) at an MOI of 100. The HUVE C monolayers were labeled with antibodies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stai n (blue) and observed by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treate d with the wild-type parent strain 381 at an MOI of 100 for 20 hours (positive co ntrol) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arrows indica te the junctions between the cells. Magnification: A, 580x; B-D 440x; E, 500x; F, 440x.

PAGE 118

118 A B C D E F Figure 3-20. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis YPP2 ( kgp) at an MOI of 100. The HUVEC monolay ers were labeled with antibodies against the junc tional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and obs erved by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild -type parent strain ATCC 33277 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or lo calized within the HUVECs. White arrows indicate the junctions betw een the cells. Magnification: A-D, 580x; E-F, 480x.

PAGE 119

119 A B C D E F Figure 3-21. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis CW401 ( rgpA-, rgpB) at an MOI of 100. The HUVEC monolayers were labeled with antibodies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stai n (blue) and observed by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treate d with the wild-type parent strain 381 at an MOI of 100 for 20 hours (positive co ntrol) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arrows indica te the junctions between the cells. Magnification: A-B 440x; C, 480x; D, 580x; E-F, 480x.

PAGE 120

120 A B C D E F Figure 3-22. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis CW501 ( rgpA-, rgpB-, kgp) at an MOI of 100. The HUVEC monolayers were labeled with antibodies against the junctional protein -catenin (green ), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-type parent strain 381 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arrows indicat e the junctions between the cells. Magnification: A, 720x; B, 500x; C, 440x; D, 580x; E-F, 480x.

PAGE 121

121 A B C D E F Figure 3-23. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis MT10 ( rgpA) at an MOI of 1000. The HUVEC monol ayers were labeled with antibodies against the junc tional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and obs erved by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-typ e parent strain 381 at an MOI of 1000 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or local ized within the HUVECs. White arrows indicate the junctions between th e cells. Magnification: A-D, 440x; E-F, 580x.

PAGE 122

122 A B C D E F Figure 3-24. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis MT10W ( rgpA-, kgp) at an MOI of 1000. The HUVEC monolayers were labeled with antibodies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stai n (blue) and observed by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treate d with the wild-type parent strain 381 at an MOI of 1000 for 20 hours (pos itive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arrows indicat e the junctions between the cells. Magnification: A; 720x; B, 440x; C, 580x; D, 500x, E-F 580x.

PAGE 123

123 A B C D E F Figure 3-25. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis G-102 ( rgpB) at an MOI of 1000. The HUVEC monol ayers were labeled with antibodies against the junc tional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and obs erved by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-typ e parent strain 381 at an MOI of 1000 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or local ized within the HUVECs. White arrows indicate the junctions between the cells. Magnification: A; 720x, B-C, 440x; D-F, 500x.

PAGE 124

124 A B C D E F Figure 3-26. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis G102W ( rgpB-, kgp) at an MOI of 1000. The HUVEC monolayers were labeled with antibodies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stai n (blue) and observed by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treate d with the wild-type parent strain 381 at an MOI of 1000 for 20 hours (pos itive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arrows indicat e the junctions between the cells. Magnification: A, 720x; B, 440x; C, 580x; D-E, 440x; F, 580x.

PAGE 125

125 A B C D E F Figure 3-27. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis YPP2 ( kgp) at an MOI of 1000. The HUVEC monol ayers were labeled with antibodies against the junc tional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and obs erved by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild -type parent strain ATCC 33277 at an MOI of 1000 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or lo calized within the HUVECs. White arrows indicate the junctions betw een the cells. Magnification: A-C, 440x; D, 580x, E, 500x; F, 440x.

PAGE 126

126 A B C D E F Figure 3-28. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis CW401 ( rgpA-,rgpB) at an MOI of 1000. The HUVEC monolayers were labeled with antibodies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stai n (blue) and observed by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treate d with the wild-type parent strain 381 at an MOI of 1000 for 20 hours (pos itive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arrows indicat e the junctions between the cells. Magnification: A-C, 440x; D-E, 500x; F, 580x. CW401

PAGE 127

127 A B C D E F Figure 3-29. Junctional labeling of HUVE C monolayers after treatment with P. gingivalis CW501 ( rgpA-, rgpB-, kgp) at an MOI of 1000. The HUVEC monolayers were labeled with antibodies against the junctional protein -catenin (green ), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-type parent strain 381 at an MOI of 1000 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arrows indicat e the junctions between the cells. Magnification: A, 580x; B-D, 440x; E-F, 580x.

PAGE 128

128 A B C D E F Figure 3-30. Junctional labeling of HUVEC monolayers after treatment with a P. gingivalis adhesion mutant PG1683 (conserved hypothetical protein) at an MOI of 100. The HUVEC monolayers were labeled with anti bodies against the junctional protein catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hour s (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolay er (A) and a HUVEC monolayer treated with the wild-type parent strain W83 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White a rrows indicate the junctions between the cells. Magnification: A-C, 500x; D, 580x; E, 480x; F, 720x.

PAGE 129

129 A B C D E F Figure 3-31. Junctional labeling of HUVEC monolayers after treatment with a P. gingivalis adhesion mutant PG0242 (conserved hypothetical protein) at an MOI of 100. The HUVEC monolayers were labeled with anti bodies against the junctional protein catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hour s (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolay er (A) and a HUVEC monolayer treated with the wild-type parent strain W83 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White a rrows indicate the junctions between the cells. Magnification: A, 720x; B-C, 480x; D, 500x; E-F, 440x.

PAGE 130

130 A B C D E F Figure 3-32. Junctional labeling of HUVEC monolayers after treatment with a P. gingivalis adhesion and invasion mutant PG1118 ( clpB) at an MOI of 100. The HUVEC monolayers were labeled with antibo dies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hours (B), 10 hour s (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-type parent stra in W83 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arro ws indicate the junc tions between the cells. Magnification: A, 720x; B, 500x; C, 440x; D, 500x; E, 480x; F, 720x.

PAGE 131

131 A B C D E F Figure 3-33. Junctional labeling of HUVEC monolayers after treatment with a P. gingivalis adhesion mutant YPF1 ( fimA) at an MOI of 100. The HUVEC monolayers were labeled with antibodies against the junctional protein -catenin (green ), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-type parent strain ATCC 33277 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arrows indicat e the junctions between the cells. Magnification: A, 440x; B-C, 500x; D, 440x; E-F, 580x.

PAGE 132

132 A B C D E F Figure 3-34. Junctional labeling of HUVEC monolayers after treatment with a P. gingivalis invasion mutant PG0717 (putative lipoprot ein) at an MOI of 100. The HUVEC monolayers were labeled with antibo dies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hours (B), 10 hour s (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-type parent stra in W83 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arro ws indicate the junc tions between the cells. Magnification: A, 580x; B, 440x; C-E, 500x; F, 440x.

PAGE 133

133 A B C D E F Figure 3-35. Junctional labeling of HUVEC monolayers after treatment with a P. gingivalis invasion mutant PG1286 (ferritin) at an MOI of 100. The HUVEC monolayers were labeled with antibodies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nucle ar stain (blue) and observed by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVE C monolayer treated with the wildtype parent strain W83 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arro ws indicate the junc tions between the cells. Magnification: A, 580x; B, 440x; C-E, 500x; F, 580x.

PAGE 134

134 A B C D E F Figure 3-36. Junctional labeling of HUVEC monolayers after treatment with a P. gingivalis invasion mutant YPEP ( pepO -) at an MOI of 100. The HUVEC monolayers were labeled with antibodies against the junctional protein -catenin (green ), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-type parent strain ATCC 33277 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arrows indicat e the junctions between the cells. Magnification: A, 500x; B, 440x; C, 580x; D, 500x; E, 440x; F, 580x.

PAGE 135

135 A B C D E F Figure 3-37. Junctional labeling of HUVEC monolayers after treatment with a P. gingivalis putative cysteine peptidase mutant PG1788 at an MOI of 100. The HUVEC monolayers were labeled with antibo dies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hours (B), 10 hour s (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-type parent stra in W83 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arro ws indicate the junc tions between the cells. Magnification: A, 720x; B-D, 480x; E-F, 500x.

PAGE 136

136 A B C D E F Figure 3-38. Junctional labeling of HUVEC monolayers after treatment with a P. gingivalis putative secretion activator protein mu tant PG0293 at an MOI of 100. The HUVEC monolayers were labeled with antibo dies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hours (B), 10 hour s (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-type parent stra in W83 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arro ws indicate the junc tions between the cells. Magnification: A, 720x; B, 440x; C-E, 500x; F, 580x.

PAGE 137

137 A B C D E F Figure 3-39. Junctional labeling of HUVEC monolayers after treatment with a P. gingivalis conserved hypothetical pr otein mutant PG0686 at an MOI of 100. The HUVEC monolayers were labeled with antibo dies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hours (B), 10 hour s (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-type parent stra in W83 at an MOI of 100 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arro ws indicate the junc tions between the cells. Magnification: A, 580x; B-C, 440x; D, 580x; E, 440x; F, 500x.

PAGE 138

138 A B C D E F Figure 3-40. Junctional labeling of HUVEC monolayers after treatment with a P. gingivalis adhesion mutant YPF1 ( fimA) at an MOI of 1000. Th e HUVEC monolayers were labeled with antibodies against the junctional protein -catenin (green ), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-type parent strain ATCC 33277 at an MOI of 1000 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arrows indicat e the junctions between the cells. Magnification: A-B, 580x; C-D, 500x; E, 580x; F, 440x. C. 10 hour E. 20 hour E. 20 hour E. 20 hour

PAGE 139

139 A B C D E Figure 3-41. Junctional labeling of HUVEC monolayers after treatment with a P. gingivalis invasion mutant YPEP ( pepO) at an MOI of 1000. The HUVEC monolayers were labeled with antibodies against the junctional protein -catenin (green ), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hours (B), 10 hours (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-type parent strain ATCC 33277 at an MOI of 1000 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVECs. White arrows indicat e the junctions between the cells. Magnification: A-E, 440x.

PAGE 140

140 A B C D E F Figure 3-42. Junctional labeling of HUVEC monolayers after treatment with a P. gingivalis conserved hypothetical pr otein mutant PG0686 at an MOI of 1000. The HUVEC monolayers were labeled with antibo dies against the junctional protein -catenin (green), antibodies against P.g. HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF after 5 hours (B), 10 hour s (C), 15 hours (D), and 20 hours (E). An untreated HUVEC monolayer (A) and a HUVEC monolayer treated with the wild-type parent strain W 83 at an MOI of 1000 for 20 hours (positive control) (F) are shown for comparison. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVE Cs. White arrows indicate the junctions between the cells. Magnification: A, 500x; B-F 440x.

PAGE 141

141 Figure 3-43. Comparisons of the HUVEC monolayers after treatment with the gingipain mutants and their parent strains at an MOI of 100 for 20 hours. HUVECs were labeled with anti-catenin (green), anti P. gingivalis HagB (red), and DAPI, a nuclear stain (blue) and observed by IMF. (A) Untreated control, (B) 381, (C) MT10 ( rgpA), (D) MT10W ( rgpA-, kgp), (E) G102 ( rgpB), (F) G102W ( rgpB-, kgp) (G) YPP2 ( kgp) ATCC 33277 mutant, (H) CW401 (rgpA-, rgpB), (I) CW501 ( rgpA-, rgpB-, kgp), and (J) ATCC 33277. Yellow arrows indicate P. gingivalis either attached to or localized within the HUVE Cs and white arrows indica te the junctions between the cells. Magnification: A-B, 440x; C-F, 500x, G-I, 480x; J, 580x. B A H I G E D C F J

PAGE 142

142 A B C D E F G Figure 3-44. Comparisons of the HUVEC monolayers after treatment with P. gingivalis adhesion mutants and their parent strains at an MOI of 100 for 20 hours. The HUVEC monolayers were labeled with anti-catenin (green), anti. P. gingivalis HagB (red), and DAPI, a nuclear stain (blue) and obser ved by IMF. (A) Untreated control, (B) YPF1 ( fimA), (C) PG1683 (conserved hypothe tical protein), (D) PG1118 ( clpB -), (E) PG0242 (conserved hypothetical prot ein), (F) W83 and (G) ATCC 33277. Magnification: A, 440x; B, 500x; C-D, 480x; E-F, 440x; G, 580x.

PAGE 143

143 A B C D E F G Figure 3-45. Comparisons of the HUVEC monolayers after treatment with P. gingivalis invasion mutants and their parent strains at an MOI of 100 for 20 hours. The HUVEC monolayers were labeled with anti-catenin (green), anti. P. gingivalis HagB (red), and DAPI, a nuclear stain (blue) and obser ved by IMF. (A) Untreated control, (B) PG1118 ( clpB -), (C) PG0717 (putative lipoprote in), (D) PG1286 (ferritin) (E) YPEP ( pepO), (F) W83 and (G) ATCC 33277. Magnification: A, 440x; B, 480x; C-D, 500x; E-F, 440x; G, 580x.

PAGE 144

144 A B C D E Figure 3-46. Comparisons of the HUVEC mono layers after treatm ent with additional P. gingivalis mutants and their parent strain at an MOI of 100 for 20 hours. The HUVEC monolayers were labeled with anti-catenin (green), anti. P. gingivalis HagB (red), and DAPI, a nuclear stain (blu e) and observed by IMF. (A) Untreated control, (B) PG0293 (putative secretion activator protein), (C) PG1788 (putative cysteine peptidase), (D) PG0686 (conserve d hypothetical protei n) and (E) W83. Magnification: A, 440x; B-C, 500x; D-E, 440x.

PAGE 145

145 A B C Figure 3-47. Transmission electron microscopy of HUVEC monolayers internalized with P. gingivalis 381 at an MOI of 100 and stained for acid phosphatase. (A) Untreated HUVEC monolayer. (B) HUVEC monol ayer after treatment with P. gingivalis strain 381 for 5 hours and (C) 20 hours. Black arrows indicate empty vacuoles, white arrows designate P. gingivalis and yellow arrows indicate internalized P. gingivalis within a cytoplasmic vacuole. 5 m 5 m 3 m

PAGE 146

146 A B C Figure 3-48. Transmission electron microscopy of HUVEC monolayers internalized with P. gingivalis CW501 ( rgpA-, rgpB-, kgp) at an MOI of 100 and stained for acid phosphatase. (A) Untreated HUVEC monol ayer. (B) HUVEC monolayer after treatment with P. gingivalis CW501 for 5 hours, and (C) 20 hours. Black arrows indicate empty vacuoles a nd white arrow designates P. gingivalis 5 m 5 m 5 m

PAGE 147

147 A B C D Figure 3-49. Transmission electron microsc opy of HUVEC monolayer s internalized with P. gingivalis wild-type 381, CW501 ( rgpA-, rgpB-, kgp), and a conserved hypothetical protein mutant PG0686 at an MOI of 1000 and stained with uranyl acetate (A) Untreated HUVEC monolayer. (B) HUVEC monolayer after treatment with P. gingivalis 381 for 20 hours, (C) incubated w ith the gingipain null mutant CW501 for 20 hours (C) and PG0686 for 20 hours. Black arrows indicate empty vacuoles, white arrows designate P. gingivalis, yellow arrows indicate internalized P. gingivalis contained within a cytoplasmic vacuole. 5 m 10 m 5 m 5 m

PAGE 148

148 Figure 3-50. Effects of P. gingivalis on HUVE cell death. HUVEC mo nolayers were treated for 30 min, 2.5, 12, and 24 hours with P. gingivalis strains 381, ATCC 33277, CW501 and YPF1 ( fimA) at an MOI of 100. HUVECs were stained with propidium iodide and quantified for cell death using flow cy tometry. P-values are represented as follows: (*p<0.05), (**p<0.01), and (***p<0.001). 0 5 10 15 20 25 30 35 40 30 m2.5 h12 h24 hTime% Cell Death 33277 381 FimACW501 Control ** *** (h)

PAGE 149

149 0 5 10 15 20 25 30 35 40 45 0102030Time% Cell Death 381 33277 FimACW501 Control Figure 3-51. P. gingivalis interactions with HUVEC monolay ers induces HUVE cell death after antibiotic treatement. HUVEC monolaye rs were treated for 2.5, 12, and 24 hours with P. gingivalis 381, ATCC 33277, CW501 and YPF1 ( fimA) at an MOI of 100 with the addition of the antibiotics metronidazole 200 g ml-1 and gentamicin 300 g ml-1 after 1.5 hours of co-culture. HUVECs were stained with propidium iodide and quantified for cell death using flow cy tometry. P-values are represented as follows: (*p<0.05) and (***p<0.001). *** *** (h)

PAGE 150

150 Figure 3-52. Inhibition of P. gingivalis internalization of HUVECs w ith cytochalasin D inhibits HUVE cell death in the presence of antibiotics. HUVEC monolayers were pretreated with 5 g/ml cytochalasin D for 0.5 hours prior to exposure to P. gingivalis HUVEC monolayers were treated with P. gingivalis 381, or ATCC 33277 at an MOI of 100 for 1.5 hours at whic h time the unattached bacteria were washed off and the cells were then inc ubated for the remainder of the 2.5, 12, and 24 hour time points with fresh EGM-2 media plus the antibiotics metronidazole 200 g ml-1 and gentamicin 300 g ml-1. HUVEC were stained with propidium iodide and quantified for cell death using flow cytometry. There are no statistical differences between any of the time-points tested. 0 5 10 15 20 25 30 35 40 2.51224Time% Cell Death cells only cytoD cells only CytoD 381 CytoD 33277 (h)

PAGE 151

151 Figure 3-53. Inhibition of P. gingivalis internalization of HUVEC with cytochalasin D inhibits HUVE cell death. HUVEC monolayer s were pretreated with 5 g/ml cytochalasin D for 0.5 hours prior to exposure to P. gingivalis HUVEC monolayers were then co-cultured with P. gingivalis 381 or ATCC 33277 at an MOI of 100 for 1.5 hours at which time the unattached bacteria we re washed off and the cells were then incubated for the remainder of the 2.5, 12, and 24 hour time points with fresh EGM2 media. HUVEC were staine d with propidium iodide and quantified for cell death using flow cytometry. There are no statisti cal differences between any of the timepoints tested. 0 5 10 15 20 25 30 35 40 0102030Time% Cell Death cells only cytoD cells only CytoD 381 CytoD 33277 (h)

PAGE 152

152 0 5 10 15 20 25 30 35 2.51224 Time (h)% Cell Death Untreated control 0.02ug/ml 381 SPF 0.02ug/ml 381 SPF cytoD 381 MOI 100 381 MOI 100 AB 381 MOI 100 cytoD CW501 MOI 1000 Figure 3-54. Cell death of HUVE cells after treatment with live P. gingivalis 381 with and without cytochalasin D, 0.02 g/ml of the 381 SPF and the CW501 (gingipain-null) mutant. HUVEC monolayers were incuba ted for 2.5, 12, and 24 hours with 381 at an MOI of 100 with and without antibiotics (metronidazole 200 g ml-1 and gentamicin 300 g ml-1) and P. gingivalis CW501 at an MOI of 1000. HUVEC monolayers were also exposed to 0.02 g/ml of the 381 SPF. In addition, HUVEC monolayers were pretreated with 5 g/ml cytochalasin D for 0.5 hours prior to exposure to P. gingivalis 381 at an MOI of 100 and 0.02 g/ml of the 381 SPF for 1.5 hours. Cells were then washed and incubated for the remainder of the 2.5, 12, and 24 hour time points with fresh EGM-2. HUVECs were staine d with propidium iodide and quantified for cell death using fl ow cytometry. P-values are represented as follows: (*p<0.05) and (***p<0.001). *** *** *** *** ***

PAGE 153

153 0 200000 400000 600000 800000 1000000 1200000 2.51224 Time (hr)Caspase 3 Activity Untreated cells ATCC 33277 + AB ATCC 33277 381 + AB 381 Staurosporine (STS) Figure 3-55. HUVEC monolayers treated with P. gingivalis 381 and ATCC 33277 exhibit caspase 3 activity. HUVEC monolayers we re treated for 2.5, 12, and 24 hours with P. gingivalis 381 and ATCC 33277 at an MOI of 100 with and without the antibiotics (ABs) (metronidazole 200 g ml-1 and gentamicin 300 g ml-1). An apoptosis inducer, Staurosporine (S TS), at a concentration of 0.5 m was used as a positive control. P-values are represented as follows: (*p<0.05), (**p<0.01), and (***p<0.001). *** *** *** ** ***

PAGE 154

154 0 20 40 60 80 100 120 140 160 180 200 0.511.522.5312244872 Time (hr)Resistance control W83 381 CW501 Figure 3-56. Membrane resistance of HuH7 monolayers after apical exposure to P. gin givalis strains W83, 381, and the gingipain-null mutant, CW501, at an MOI of 100. HuH7 monolayers were seeded on collagen coat ed Corning transwell-COL inserts. Transepithelial Resistance (TER) was meas ured using an EndOhm chamber and the EVOM electrical resistance system. Cell-fr ee transwell inserts were used to obtain baseline levels which were then subtr acted from each time point measured. Pvalues are represented as fo llows: (*p<0.05) and (**p<0.01). ** *(Ohms)

PAGE 155

155 0 20 40 60 80 100 120 140 160 180 200 0.511.522.5312244872 Time (hr)Resistance control W83 381 CW501 Figure 3-57. Membrane resistance of HuH7 monolayers after ba solateral exposure to P. gin givalis strains W83, 381, and the gingipa in-null mutant, CW501, at an MOI of 100. HuH7 monolayers were seeded on collage n coated Corning transwell-COL inserts. Transepithelial Resistance (TER ) was measured using an EndOhm chamber and the EVOM electrical resistance system. Cell-free transwell inserts were used to obtain baseline levels which were then subt racted from each time point measured. P-values are represented as fo llows: (**p<0.01) and (***p<0.001). *** **(Ohms)

PAGE 156

156 CHAPTER 4 DISCUSSION P. gingivalis has the ability to invade, persist and replicate in many endothelial and epithelial cell lines includi ng primary cultures of huma n endothelial cells (Dorn et al. 1999; Lamont et al. 1995). Endothelial cells infected with P. gingivalis display a number of phenotypic traits including elevated levels of coagulation and secretion of inflammatory cytokines (Ross, 1993). Vascular modifications and gingival ulcera tions stimulated by the local inflammation caused by P. gingivalis may promote the occurrence and severity of septic bacterial infections after injury to junctional barriers of the gingival tissue. The persistent challenge with P. gingivalis could obstruct the proliferati on and adjustment of cell-matrix adhesion and impede the remodeling of the cell-matrix (Yilmaz et al. 2003). These interactions likely play a key role in the de velopment of periodontal disease by P. gingivalis by providing the cells with signals for regulating cell functi on, cell migration and adherence (Nakamura et al. 1999). The integrity of the oral mucosal basement membrane, specifically the gingival sulcus, is crucial for the prevention of the introduction of oral bacteria into the bloodstream (Daly et al. 1997). In addition, similar interactions may also be of significance in th e endothelial cell layer of cardiovascular tissues. Epithelial Barrier Function The junctional complexes of the cell-cell a nd cell-matrix associations initiate the development of the strong barrier function of the polarized epithelium. Epithelial cells are held together by both tight and adhere ns junctions. Tight junctions function in the paracellular pathway by creating a major barrier to macromol ecule, fluid, electrolyt e and pathogen diffusion (Gumbiner, 1993; Gumbiner and McCrea, 1993). Tight junctions are located on the apical portion of the cell and form a periphery between the surfaces of the ap ical and basolateral

PAGE 157

157 membranes. They are essential for the constr uction and preservation of cell surface polarity because they create a barrier against the diffusi on of proteins and/or lipids in the plasma membrane (van Meer et al. 1986). Tight junctions mainta in this barrier function by communicating with the structured cy toskeletal actin filaments (Fanning et al. 1999). The epithelial structure is also depe ndent on the integrin mediated cel l-matrix adhesion, the ability of the epithelial cell layer to attach to the basement membrane (Balkovetz and Katz, 2003). Bacterial pathogens must first surpass the epithelial barrier, the hosts first line of defense, before they are able to penetrate deeper tissues and cause disease (Kazmierczak et al. 2001). The paracellular permeability or the transepithelial resistance (TER) of epithelial cell monolayers has been shown to be modified by several pathogens (Sears, 2000). A number of bacteria can infiltrate the host tissues and/or secr ete toxins into the host via the paracellular route by surpassing the cell-cell junctio ns, leaving the subepithelial c onnective tissue susceptible to bacterial invasion (Balkovetz and Katz, 2003). The evaluation of the TERs of HuH7 cell monolayers after incubation with P. gingivalis provided information concerning the ability of this pathogen to a lter the epithelial cell barrier. The HuH7 basolateral surfaces appeared to be more vulnerable to this result than the apical surface. The effects on the TERs of basolateral exposure of HuH7 cell monolayers to P. gingivalis strains suggest that an intimate interaction between the bacterium and the HuH7 cells is not required for the observed effects, since the membrane is located between the basolateral side and the bacteria. Furthermore, these resu lts suggest that the cont ributions of bacterial soluble factors, perhaps gingipain s, which are able to disseminate through the membrane pores to elicit these TER effect, ar e the active component. The lack of significant differences in the TERs of apical or basolateral exposed HuH7 cell monolayers to strain CW 501 (gingipain-null) suggests

PAGE 158

158 that the gingipains are intimately involved in this activity. The dr op in TER after 48 and 72 hours of apical and ba solateral treated HuH7 cell monolayers could be the result of disruption of the HuH7 cell junctions or cell death caused by P. gingivalis exposure. Monolayer Integrity P. gingivalis W83 and 381 SPF and W83 Lysate Adversely Affect Endothelial and Epithelial Adherens Proteins An array of events such as differentiation and survival are controlled by the interactions of cells with adjacent cells and with the EC M (Bazzoni and Dejana, 2004; Juliano, 2002). Intercellular adherens junctions hold these adjoining epithelial and endothelial cells together. Adherens junctions are primarily composed of cadherins and cat enins, which together provide stability by anchoring the cells to the actin cytoskeleton (Bazzoni and Dejana, 2004). These cellcell adherens junctions are also essential for the proper organization of vessels during angiogenesis and are involved in the preservation of specific cell properties including contact inhibition of cell growth and the permeability to in flammatory cells or other solutes (Bazzoni and Dejana, 2004; Juliano, 2002). In addition to ce ll-cell adhesion, the adherence of cells to the ECM is crucial for proper organization, development, and function of cells (Wu et al. 2001) Substances responsible for increasing vasc ular endothelial permeability disturb the protective function of the intracellular junctions (Carmeliet et al. 1999; Dejana et al. 1999). In order to gain access to the underlying connective ti ssues, bacterial pathogens must first infringe upon these intracellular adherens junctions (Wu et al. 2001). There are multiple mechanisms by which bacterial infections can o ccur including direct trauma and epithelial and endothelial cell damage (Wu et al. 2001). Thus, the loss of cell adhesi on and subsequent infection of the underlying tissues could be a consequence of th e loss of junctional pr otein molecules from

PAGE 159

159 within the intrace llular junctions of epithelial and endothelial cells (Katz et al. 2002; Gottardi et al. 2001). The results using the lysate of P. gingivalis W83 on both the HuH7 epithelial and HUVEC endothelial cell lines s uggest that proteins from P. gingivalis are capable of disrupting the integrity of the ce ll monolayers. The HuH7 and HUVEC untreated control cells demonstrated that the majority of the -catenin and cadherin proteins were concentrated at the cell surfaces. catenin could also be seen sparsely punctate throughout the cytoplasm. In contrast, the HuH7 and HUVE cell monolayers exposed to the P. gingivalis W83 lysate displayed monolayer disruption and resulted in a loss of cadherin from the ce ll surface concomitant with an alteration in the localization of -catenin from the cell surface to cytoplas mic structures in the proximity of the nucleus. In addition, several cells had detached from the monolayer compared to the untreated control. Thus, the loss of -catenin and cadherin from the cell surface appeared to considerably reduce these cells ability to maintain their inte rcellular contacts, leading to cell detachment. Treatment of the HuH7 cell monolayers with heat-killed W 83 lysate completely inhibited the active component(s) within this lysate from disrupting the HuH7 monolayer. Since lipopolysaccharide, LPS, is partic ularly heat stable, it is like ly LPS does not participate in disrupting the cell monolayer (Kuramitsu et al. 2003). This suggest s that other protein component(s) from P. gingivalis are responsible for this activity. Exposure of HuH7 cell monolayers to either live P. gingivalis W83 or the W83 lysate in the presence of the cysteine proteinase inhibitor, Na-tosyl-L-lysine chloromethyl ketone (TLCK), resulted in a reduction of monolayer disr uption and few detached cells. Treatment with TLCK inhibited the P. gingivalis cysteine proteinases from causing detachment of the HuH7 cells but was unable to fully inhibit the alte ration of the location of cadherin or -catenin from the cell

PAGE 160

160 surface to within the HuH7 cytoplasm. Other proteinase inhi bitors such as PMSF and PIC had no observed inhibitory effect on the W83 lysate. Gi ven that the PIC cocktail contains a cysteine proteinase inhibitor, the inability of PIC to inhibit the activity of the W83 lysate could be because the concentration used wa s not adequate to inhibit the cy steine proteinase activity. Since TLCK has been shown to i nhibit Kgp activity more significantly than Rgp activity (Pike et al. 1994), it is possible that Kgp is more actively involved in a ffecting cell-matrix adhesions, whereas the activity of Rgp is largely accountable for the observed alteration of location of cadherin and -catenin within the epithelial and endot helial monolayers examined. The data establish that the P. gingivalis cysteine proteinases, gingipain s, are the active components effecting both cell-cell and cell-matrix adhesion in both epithelial and endothelial cells. The adhesive properties of endot helial cells are interrupted upon treatment with secreted protein fractions (SPFs). Confluent HUVE C monolayers treated with W83 and 381 SPFs resulted in a loss of cadherin and -catenin at the ce ll surface. The disr uption in the HUVEC monolayer was more pronounced af ter treatment with the 381 SPF than with the W83 SPF. One explanation for this difference is that the level of P. gingivalis 381 Rgp enzymatic activity was 2fold higher than that of strain W83. HUVECs exposed to the SPFs of both strains yielded similar results concerning the location of -catenin and cadherin distributi on as those observed with the W83 lysate. However, a much lower concentratio n of the SPF was needed to yield results equal to those observed with the W83 lysate, suggestin g that the majority of the damaging effects associated with P. gingivalis are surface associated and/or secreted. The results of this study establish that th e proteolytically active extracellular protein fractions from P. gingivalis W83 and 381 were capable of stim ulating epithelial and endothelial cell rounding as well as the detachme nt of cells from the culture dish and from each other. Thus,

PAGE 161

161 P. gingivalis appears to exhibit copious amount of proteo lytic activity that affects the location of the cadherin and -catenin proteins of both epithelial and endothelial junctions. Furthermore, the data suggest that the active component(s) respon sible for cell-cell and cell-matrix adhesions are cysteine proteinases, gingipains. The collection of data using HUVEC and HuH7 cells suggests that periodontal tissue destruction characteristic of chronic periodontitis could theref ore be a consequence of the P. gingivalis surface associated and/or secreted proteins, e.g., gingipains. Data from other laboratories support this conc lusion. For example, oral keratinocytes exposed to P. gingivalis displayed proteolysis of adherens junction proteins (e.g., catenins) a nd this result is likely due to the gingipains since TLCK was capable of inhibiting this activity (Hintermann et al. 2002). Katz et al provided further evidence to support th e involvement of the gingipains in the proteolysis of the proteins of the adherens junc tions. This group also showed that incubation of P. gingivalis with MDCK cell monolayers and imm unoprecipitated E-cadherin caused the degradation of E-cadherin and pr eincubation with the Rgp and Kgp specific inhibitors leupeptin and acetyl-Leu-Val-Lys-aldehyde respec tively, abrogated this activity (Katz et al. 2002). The loss of cell adhesion and subseque nt infection of the underlying tissues could be a consequence of the loss of E-cadherin molecules from epithelial cells (Katz et al. 2002; Gottardi et al. 2001). In addition, Sheets et al reported that the ging ipain active extracts we re responsible for the cleavage of N and VE-cadherin and integrin 1 and the loss of cell adhesion of bovine coronary artery endothelia l cells (Sheets et al., 2005). In agreement with Hintermann et al. this activity could be inhibited or considerably delayed by pr e-incubation of these cells with TLCK, further suggesting the direct involvemen t of the gingipains (Sheets et al. 2005; Hintermann et al. 2002).

PAGE 162

162 In addition to cell surface a nd/or secreted proteins of P. gingivalis, live P. gingivalis could also induce disruption of the epith elial and endothelial monolayers. Four different wild type strains of P. gingivalis including, W83, 381, ATCC 33277 and AJW4 were tested for their abilities to disrupt the junctional protein, -catenin, and thus, the endot helial monolayer. Each strain of P. gingivalis except for AJW4, which has wild-t ype levels of adherence but a significantly reduced ability to invade, has been shown to adhere, invade and persist within HUVE cells. Interestingly, by 10 hours of exposur e, all four strains were detected both associated with the cell surface a nd internalized within the cyt oplasm of the HUVECs. Of these strains, W83 and 381 demonstrated the most activity on the HUVEC endothelial monolayers followed by AJW4 and ATCC 33277, respec tively. Our data suggests that P. gingivalis is working from within the HUVECs since the result s with AJW4 was shown to adhere to but not to invade the HUVE cells (Dorn et al ., 2000). Invasion studies are ch aracteristically determined after the cell line of interest is exposed to th e desired strain for 2.5 hour s. Therefore, it is possible that the extended interaction times of HUVECs with AJW4 were adequate for its invasion of HUVEC monolayers. These data suggest that ad hesion and invasion of live P. gingivalis may also play a role in th e disruption of the cell monolayer s and re-localization of the -catenin and cadherin proteins. Given that adhesion and invasi on efficiencies were strain dependent, there could be a connect ion between the ability of one strain to adhere and invade host cells and its ability to br eakdown host proteins (Hintermann et al. 2002). P. gingivalis Gingipain Mutants Demonstrate Differe nt Effects on HUVE Cell Monolayers The proteolysis of signaling molecules that re gulate adhesion, surv ival, differentiation, proliferation, and migration are likely accomplished by P. gingivalis gingipains (Hintermann et al. 2002). Gingipains are capable of activating the host cell me talloproteinases to breakdown

PAGE 163

163 components of their extracellular matrix in addition to degradi ng cell adhesion molecules from within the periodon tal tissue (Katz et al. 2000; Kuramitsu, 1998). Thus the relocation of the cadherins and catenins by P. gingivalis may potentiate the cellula r and subsequent tissue invasion characteristic of periodontal disease. The Rgp-null mutant CW401, expressing only K gp, was unable to alter the location of catenin from the HUVE cell surface and induce monolayer disruption. This result is consistent with the W83 lysate and TLCK experiments whic h showed a lack of junctional complexes but minimal cell detachment. Since TLCK has been shown to inhibit Kgp more than Rgp (Pike et al ., 1994) this result suggests that Rgp is more e ffective in causing the di sruption of the cell-cell contacts and subsequent disruption to the HUVEC monolayers than is Kgp. HUVEC monolayers co-cultured with the gingip ain mutants that express RgpB, MT10 ( rgpA) and MT10W ( rgpA-, kgp) and the mutant that espresses RgpA, G102W ( rgpB-, kgp), resulted in monolayer disruption and a loss of -catenin staining at the cell surface after 15 hours. HUVECs exposed to the P. gingivalis strain that expresses only RgpA and Kgp, G102 ( rgpB), also resulted in monolayer disruption and a loss of -catenin staining at the cell surface but only at 20 hours of exposure. These data indicate that both RgpA and RgpB are active in HUVEC monolayer disruption. However, the gingipain mutant G102W that expresses only RgpA caused a slightly greater disruption of the HUVEC mono layer after 15 hours than MT10W that espresses only RgpB. Therefore, it is likel y that RgpA is more active th an RgpB in this regard. It is important to note that Rgp is involved in the proteolytic activation of the major P. gingivalis fimbrial protein, FimA, that has been shown to be im portant for the appropriate attachment and invasion of P. gingivalis of host cells (Kadowaki et al. 2000; Potempa et al. 2000; Kadowaki et al. 1998; Xie et al. 1997; Tokuda et al. 1996). In addition, Arg-gingipains

PAGE 164

164 have a role in a number of pathways includi ng contributing to their own post-translational processing, expression of the mature 75-kDa cell surface protein, profimbrillin, and pro-kgp. (Kadowaki et al. 2000; Potempa et al. 2000; Kadowaki et al. 1998; Xie et al. 1997; Tokuda et al. 1996). Thus, the loss of Rgp results in decreased attachment and invasion of P. gingivalis to host cells due to the improper processing of FimA (Kadowaki et al. 2000; Potempa et al. 2000; Kadowaki et al. 1998; Xie et al. 1997; Tokuda et al. 1996). In addition to its lack of fimbriae, the decrea se in attachment and invasion of the gingipainnull mutant, CW501, to HUVEC monolayers is likely a result of its lack of gingipains. The lack of monolayer disruption or redistribution of -catenin observed with HUVECs co-cultured with CW501 provides additional evidence that the gingipains play a role in this activity. In addition, exposure of HUVECs to the Rgp-null mutant, CW401, which is also devoid of fimbriae, suggests that the Arg-gingipain s may disrupt the endothelial HUVEC monolayers. These data also suggest that there may be c ooperation between the gingipains. It is possible that Kgp acts as a tertiary effector and may work in concert with RgpA and/or RgpB since HUVEC monolayers exposed to the gingipain mutant s containing Kgp in addition to RgpA or RgpB, showed a higher level of monolayer disruption than the mutants expressing RgpA or RgpB alone. In contrast, HUVEC monolayers exposed to the YPP2 ( kgp) mutant, that retains both RgpA and RgpB activity, only caused disruption to the monolayer after 20 hours of tr eatment, similar to its parent strain. A possible explanation for these results is a difference between the wild-type strains. YPP2 was constructed in ATCC 33277, a different pare ntal strain than the rest of the gingipain mutants, which were constructed in P. gingivalis 381. In support of this explanation, HUVEC monolayers treated with strain 381 at an MOI of 100 induced monolayer disruption after 10 hours of co-culture, while strain ATCC 33277 at the same MOI, did not induce HUVEC

PAGE 165

165 monolayer disruption until 20 hours of co-culture. These observed differences could be a result of differentially regulated gingipa in expression between strains and/ or from diverse experimental procedures such as the constr uction of the mutants, single versus double crossover (Tokuda et al. 1998). The variation among strains may also be the consequence of a number of factors such as lower metabolic or enzymatic activity of the bacteria. Effects of P. gingivalis on Adhesion and Invasion of HUVEC Monolayers P gingivalis coaggregation with other bacteria as well as adherence to and colonization of host tissues is mediated by adhesion factors su ch as hemagglutinins, vesicles and fimbriae (Hamada et al. 1998). The results with HUVEC monolay ers exposed to the adhesion mutant YPF1 ( fimA) showed minimal adhesion of this mutant to the cell monolayers and no monolayer disruption through 20 hours of co-culture. The adhesion mutants PG1683 (conserved hypothetical protein) and PG1118 ( clpB) showed adhesion to th e HUVEC monolayers after 10 and 15 hours of co-culture, resp ectively. However, the monolay ers remained intact after treatment with each of these adhesion mutants. This data suggests that adhesion alone may not sufficient for monolayer disruption. In contrast, although PG0242 (conserved hypothetical protein) demonstrated minimal attachment to the HUVEC monolayer through 15 hours of coculture, the attachment of PG0242 to HUVEC m onolayers after 20 hours of co-culture was markedly increased. In addition to the in creased attachment, PG0242 also demonstrated HUVEC monolayer disruption, sugg esting that adhesion plays a role in this activity. PG1683 has significant homology with -amylases, which participate in the coaggregation of P. gingivalis with other oral bacteria and in the attachment of P. gingivalis to epithelial cells (Belanger et al. 2006; Ellen et al. 1997; Kamaguchi et al. 1994). ClpB is a heat shock protein that participates in many in fections induced by bacterial pa thogens by acting as a chaperone,

PAGE 166

166 inhibiting the aggregation of proteins and aidi ng in the proper folding of proteins (Weibezahn et al. 2005; Goulhen et al. 2003; Lu and McBride, 1994). The adhesion of PG1118 ( clpB) to the HUVEC monolayer suggests that th is protein is important for P. gingivalis to be able to regulate and control its own aggregation a nd the binding of specific proteins in response to environmental stress. The YPF1 mutant, which lacks the major fimb rial protein important for adhesion, FimA, showed a low level of attachment to the HUVEC monolayer and no disruption of the cell monolayer after 20 hours of co-culture. In contrast, HUVEC monolayer s treated with the PG0242 (conserved hypothetical protein) muta nt for 20 hours resulted in numerous P. gingivalis attached to the cell monolayer and monolayer disruption. A blas t search revealed that PG0242 has 81% similarity with tetrapy rrole methylases and is likely involved in the metabolism of protoporphyrin and protohaem. Work done in our laboratory found that the loss of this protein, which would likely decrease the ability of this P. gingivalis mutant to utilize haem, resulted in decreased adhesion and invasion of P. gingivalis of HCAECs. The availability of hemin or iron has been shown to regulate several fa ctors connected to the virulence of P. gingivalis including growth, survival and the pres ence of gingipains (Kesavalu et al. 2003) P. gingivalis grown under iron limited conditions has been shown to increase the production of outer membrane vesicles (OMVs), LPS and gingi pain expression (Kesavalu et al. 2003). Therefore, the reduced ability of P. gingivalis to breakdown iron storage molecu les would likely result in the upregulation of other virulence factors, wh ich may account for the monolayer disruption observed with PG0242. Taken together, these data suggest that adhesion of P. gingivalis to host cells is not sufficient but facilitates P. gingivalis to stimulate the alteration of location of -catenin as well as

PAGE 167

167 monolayer disruption. However, since the adhe sion mutant PG0242 was capable of altering the location of -catenin and disrupting the HUVEC monolayer, it is possible that attachment is not an absolute requirement for P. gingivalis to induce monolayer di sruption. There may be additional mechanisms that P. gingivalis utilizes to stimulate monolayer disruption. As mentioned previously, P. gingivalis has been shown to inva de a number of cell types including epithelial and endothelial cells (Dorn et al. 1999; Lamont et al. 1995). The ability of P. gingivalis to internalize within vari ous cell types may prolong the su rvival of this bacterium by providing a protective niche from the host i mmune system within a nutritionally abundant environment (Dorn et al. 1999; Lamont et al. 1995). Invasion of endothelial cells by P. gingivalis has been shown to be required for the stimulation of various inflammatory IgCAMs such as ICAM-1, VCAM-1 and E-selectin by P. gingivalis (Khlgatian et al. 2002). In addition, invasive, but not non-invasive P. gingivalis has been shown to activate the endothelium leading to atherosclerotic events in apoE -/mice by accelerating the local inflammatory responses within the aortic arch (Chou et al. 2005; Gibson et al. 2004). Therefore, the alteration of the junctional proteins cadherin and catenin by invasive P. gingivalis would affect the integrity of the tissue, thereby enabling P. gingivalis entry into the vasculature (Chen et al. 2001b). P. gingivalis invasion of HUVEC monolayers was determ ined to be instrumental in the induction of monolayer disruption. HUVECs co-cu ltured with mutants with reduced invasive capabilities, PG1118 ( clpB), PG0717 (putative lipoprotein) PG1286 (ferritin) and YPEP ( pepO) appeared to be attached to the HUVEC surface after 20 hours of exposure. However, none of these invasion mutants had any effects suggesting that inva sion is necessary for HUVEC monolayer disruption.

PAGE 168

168 P. gingivalis PepO is an endopeptidase with significant homology to the mammalian endothelin converting enzyme ECE-1, which is re sponsible for converting big-endothelin into the most potent vasoconstrictor, endothelin-1 (ET-1) (Ansai et al. 2003; Ansai et al. 2002). PepO has been suggested to be a key player in the interaction of P. gingivalis with host cell membranes during invasion (Ansai et al. 2003; Ansai et al. 2002). Thus, the decreased monolayer disruptive abi lities of the invasion pepOmutant may be due to the inability of this mutant to appropriately interact with the HUVEC membrane. In addition, since the pepOmutant retains the FimA protein, the decreased adhesion and invasion obser ved with this mutant is not due to the la ck of fimbriae (Ansai et al. 2003; Ansai et al. 2002) and suggests that this endopeptidase is necessary for P. gingivalis to cause monolayer disruption. The decreased invasion of the lipoprotein mutant, PG0717, is pe rhaps due to the decreased ability of this P. gingivalis mutant to assemble onto lipid rafts. Lipid rafts are present on the mammalian cell surface and are important for the transport of cholesterol, signal transduction and invasion (Giacona et al. 2004). Lipid rafts are utilized by other bacterial species such as Legionella pneumophilia and Brucella abortus for entrance into the host cells (Duncan et al. 2004; Watarai, 2004; Duncan et al. 2002). Recently, P. gingivalis has also been demonstrated to utilize lipid rafts for internalization into host cells (Belanger et al. 2006; Tsuda et al. 2005). Thus, the presence of this lipoprotein appears to be important for P. gingivalis to induce HUVEC monolayer disruption. P. gingivalis 1286 is an intracellular protein encoding a ferritin gene that participates in iron storage and may also play a role in defending P. gingivalis against oxidative stress (Andrews et al. 1998). In addition, the heat shock prot ein (HSP), ClpB, is an immuno-dominant antigen widely expressed by P. gingivalis that helps the bacterial cell to adapt to environmental

PAGE 169

169 stressors such as elevated temperatures (Shelburne et al. 2005; Goulhen et al. 2003). The lack of HUVEC monolayer disrupti on and re-localization of -catenin observed with the P. gingivalis invasion mutants PG1286 and PG1118 ( clpB) may be due to a reduced ability of these mutants to properly process and/or secr ete proteins responsible for this effect in response to environmental stress. Collectively, these data with the inva sion mutants suggests that adhesion of P. gingivalis to the HUVEC monolayers is not as fundamenta l in causing monolayer disruption, as is P. gingivalis invasion. In fact, the effects of internalized bact eria may be characteristically different than external bacteria. The proteins of the j unctional complexes (cadherins and integrins) may be degraded by the OMVs secreted by the internalized P. gingivalis Since OMVs contain hemagglutinins, they may be able to escape th e vacuole through the li pid membrane into the cytosol and act on the juncti onal proteins directly (Qi et al ., 2003; Beveridge and Kadurugamuwa, 1996; Grenier and Mayrand, 1987). These data support the hypothesis that the gingipains of live P. gingivalis induce the majority of their ac tivity from within the HUVE cells, and less from the outside. HUVEC monolayers exposed to mutants of P. gingivalis that displayed no difference in adhesion or invasion of HCAECs compared to the parent st rain W83, PG0293 (putative secretion activator protein), or an incr ease in invasion, PG1788 ( putative cysteine peptidase), resulted in the alteration of location of -catenin and monolayer disruption. A blast search revealed that PG0293 displays homology to the se cretion activator protein of Brucella melitensis and may contribute to the virulence of P. gingivalis The ability of the PG0293 mutant to cause monolayer disruption suggests that this protein does not play a crucial role in P. gingivalis ability to cause monolayer disruption. PG1788 is a puta tive cysteine peptidase that likely plays

PAGE 170

170 important roles in the acquisi tion of nutrients, tissue invasi on and host immune defense modulation (Lamont and Jenkinson, 1998) The ability of this mu tant to disrupt the HUVEC monolayer and alter the location of -catenin suggests that this protein is also not essential for this activity. In contrast, HUVEC monolayers exposed to PG0686 (conserved hypothetical protein), which also showed no difference in adhesion or invasion of HCAECs compared to the parent strain W83, resulted in no monolay er disruption even though numerous P. gingivalis were attached to or internalized within the HUVE Cs. The increased attachment of PG0686 to HUVECs is in agreement with reports that P. gingivalis invasion of epithelial and endothelial cells coincides with an up-regulation of the gene corresponding to PG0686 (Belanger et al. 2006; Hosogi and Duncan, 2005). A blast search revealed that PG0686 has significant homology to hemerythrin, an oxygen transporting protein found in many eukaryotic invertebrates (Isaza et al ., 2006) and to a PAS/PAC sensor protein, wh ich senses oxygen in many varieties of prokaryotes including E. coli and Rhizobium meliloti (Buck et al ., 2001). Shelburne et al ., reported that the expression of the P. gingivalis gingipains rgpA and rgpB were upregulated in response to oxidative stress (Shelburne et al. 2005). It is possible th at the loss of the 0686 gene disrupts the ability of P. gingivalis to sense and respond to envi ronmental stress e.g., oxygen, which could result in the failure of P. gingivalis to upregulate rgpA and rgpB This possibility could account for the decreased ability of PG 0686 to induce HUVEC monolayer disruption, due to its inability to upregulate the gingipains rgpA and rgpB The data from mutants PG0293 and PG1788 suggests that invasion of HUVECs is necessary for P. gingivalis to fully induce the redistribution of -catenin and monolayer disruption. However, invasion of HUVECs by the PG0686 mutant did not induce monolayer disr uption or the alteration of location of -catenin.

PAGE 171

171 This data suggests that P. gingivalis invasion of host cells also requires PG0686 to induce the disruption to the HUVEC monolayers. To summarize the conclusions from this colle ction of data, adhesion and invasion together are a mechanism that P. gingivalis utilizes to induce HUVEC monolayer disruption and catenin re-localization from th e cell surface to cytoplasmic. At an MOI of 100, the adhesion mutant YPF1 ( fimA), the invasion mutant YPEP ( pepO), and the PG0686 mutant do not cause a disruption of the HUVEC monolayers. Ho wever, co-culture with all three P. gingivalis mutants at an MOI of 1000 demonstrated th at the integrity of the HUVEC monolayer was altered in that -catenin was no longer visualized at the cell surface. In addition, numerous P. gin givalis were observed attached to the HUVE cell surface and in ternalized within the HUVEC cytoplasm. The effects observed with the adhesion mutant YPF1 and the invasion mutant YPEP were comparable to their parent strain ATCC 33277. These results suggest that both adhesion and invasion impaired P. gingivalis mutants at high enough concen trations (MOI of 1000) are capable of altering the location of -catenin from the HUVE cell surf ace to within the cytoplasm, as well as inducing HUVEC monol ayer disruption. However, HUVEC monolayers co-cultured with PG0686 at this concen tration for 20 hours remained mostly intact although -catenin did not remain concentrated at th e cell surface. The effects obs erved with the PG0686 conserved hypothetical protein mutant further suggest that this protein is critical for P. gingivalis to induce HUVEC monolayer disruption. Ta ken together, these data su ggest that HUVEC monolayer disruption may be a consequence of P. gingivalis internalization. These data do not prove but suggest that the gi ngipains of live P. gingivalis may act primarily from within the endothelial cells, and less from the cells exterior.

PAGE 172

172 In vitro experiments have shown that P. gingivalis has the ability to invade, persist, and replicate in many endothelial (EC) and epithelial cell lines including primary cultures of human endothelial cells (Dorn et al. 1999; Lamont et al. 1995). The ability of P. gingivalis to invade and persist establishes that this bacterium is equipped with a mechanism capable of evading host defenses, ultimately leading to pathogenicity (F inlay and Falkow, 1997). This data suggests that invasion, following adhesion is more important for P. gingivalis derived disruption of the HUVEC monolayer, than is adhesion alone. Dorn et al. showed that invasive P. gingivalis localizes inside autophagic vacuoles within HCAECs (Dorn et al. 2001). Results from this study us ing transmission electron microscopy (TEM) showed that HUVEC monolayers exposed to the invasive P. gingivalis 381 and PG0686 at an MOI of 100 resulted in th e localization of thes e strains within intracellular vacuoles, possibly autophagic vacuoles. In contrast, HUV EC monolayers treated with the non-invasive strain, CW501, at the same MOI resulted in very few P. gingivalis at the HUVE cell surface. In addition, the morphology of the HUVE cell was very similar to the untreate d control cells. In addition to its lack of gingipain activity, CW 501 likely does not invade because it does not correctly process and express FimA, a protein that facilitates P. gingivalis adhesion and invasion (Kadowaki et al. 2000; Potempa et al. 2000; Kadowaki et al. 1998; Xie et al. 1997; Tokuda et al. 1996). Altogether, the results with the gingipain-null mutant, CW501, suggest that P. gingivalis gingipains and/or fimbriae are nece ssary for both adhesion and invasion. Increasing the MOI of strains 381, PG0686 and CW501 to 1000 affected the morphology of the HUVECs. HUVECs co-cultured w ith strains 381 and PG0686 showed numerous P. gingivalis located around the periphery of the HUVE Cs, as well as inte rnalized within intracellular vacuoles. Compared to the untreated control cells, the HUVECs co-cultured with

PAGE 173

173 strain 381 exhibited gross cellula r distortions, including an inva ginated cell membrane and the presence of numerous empty intracellu lar vacuoles and vacuoles containing P. gingivalis High numbers of PG0686 were localized within mu ltiple intracellular vacuoles throughout the HUVECs. However, the HUVE cell membrane rema ined intact in contrast to the HUVECs treated with strain 381. HUVECs expos ed to CW501 demonstrated very few P. gingivalis located around the cell periphery or internalized within the HUVECs. With the exception of an increased number of empty v acuoles within the HUVECs co-c ultured with CW501, the cellular morphology resembled the untreated control cells. These data suggest th at the gingipains are likely involved in affecting the HUVEC morphology (invagination of the cell membrane) since the increased concentration (MOI of 1000) of th e gingipain-null mutant did not have the same effect on the membrane morphology as did stra in 381. However, incubation of the HUVECs with CW501 at this concentration did result in the appearance of additional intracellular vacuoles, most of which were devoid of P. gingivalis These results support my previous data that the disruption of the HUVEC monolayer caus ed by exposure to strain 381 but not to CW501 is the result of the P. gingivalis gingipains. In addition to the gingipains, P. gingivalis invasion of HUVECs may al so be necessary or a second mechanism for this bacteria to cause mo nolayer disruption since CW501 and the invasion mutants of P. gingivalis, YPEP, PG1118, PG0717, and PG1286, demonstrated very little internalization with in the HUVECs and no monolayer disruption. However, HUVEC monolayers treated with the c onserved hypothetical protein muta nt PG0686 at an MOI of 100, which showed no difference in ad hesion or invasion when compared to its parent strain W83, also resulted in a lack of HU VEC monolayer disruption. Taken together, the results from the

PAGE 174

174 IMF and TEM experiments suggest that the gi ngipains and the 0686 protein in addition to P. gingivalis invasion are critical fo r the disruption of the HUVEC monolayers. Cell Death Our previous experiments revealed that exposure of HUVEC m onolayers to certain P. gingivalis strains caused disruption of the monolayers and cell detachment. We have also shown that the lysate and SPF of P. gingivalis caused epithelial and e ndothelial cell monolayer disruption and the detachment of cells from thes e monolayers. This collection of data suggests that a secreted protei n or proteins from P. gingivalis (e.g., gingipains) are responsible for this observed activity. Using flow cytometry, the detached cells fr om the intact HUVEC monolayers exposed to P. gingivalis 381 and ATCC 33277 were determined to be dead, suggesting th at these strains were capable of inducing HUVE cell death. In contrast, the P. gingivalis mutant strains, CW501 and YPF1, were unable to indu ce such cell death and instead produced cell death levels comparable to the untreated cells. In addition to its lack of gingi pains, it is plausible that CW501 was unable to induce cell death because of its lack of fimbriae and subsequent reduced ability to attach to and invade the HUVECs. Interes tingly, YPF1 does contain the gingipains (Love et al. 2000). However, like CW501, this mutant was unable to induce HUVE cell death. As mentioned previously, adhe rence and invasion of host cel ls have been shown to be hindered by the lack of th e FimA protein (Hamada et al. 1994). However, it is possible that both mutant strains CW501 and YPF1 could utili ze a different mechanism for attachment since other non-fimbrial adhesions asso ciated with the outer membrane have been isolated from P. gingivalis (Lepine et al. 1996; Lamont et al. 1994; Grenier a nd Mayrand, 1987; Okuda et al. 1986; Boyd and McBride, 1984). The existence of other adhesions is anticipated due to the findings that the fimA mutant, DPG3, retained the ability to adhere to normal human gingival

PAGE 175

175 epithelial cells, NHGEC, though 50% less, and to GEC, 65% less (Weinberg et al. 1997; Hamada et al. 1994). In addition to the adhesin do mains of RgpA and Kgp, adherence of P. gingivalis to host cells has been shown to be mediat ed by the hemagglutinins, specifically, Hag A and Hag B, which are large surface proteins with considerable homology to the adhesin domains of the gingipains (Song et al. 2005; Ally et al. 2003; Grenier et al. 2003; Chen et al. 2001a; Dorn et al. 2000; Progulske-Fox et al. 1999; Progulske-Fox et al. 1993). P. gingivalis also possesses a minor fimbrial protein (M fa1), which has been shown to be highly immunogenic, and directly partic ipates in the coadhesion with Streptococcus gordonii (Park et al. 2005; Hiramine et al. 2003). Taken together, the reduce d adhesion and invasion ability of CW501 and YPF1 suggests that P. gingivalis adhesion and/or subseque nt invasion of HUVE cell monolayers as well as P. gingivalis gingipains are necessary for P. gingivalis induced cell death. Given that adherence of P. gingivalis to HUVECs has been shown to be necessary for P. gingivalis invasion, we determined if invasion was a requirement for P. gingivalis induced HUVE cell death. HUVEC monolayers were exposed to P. gingivalis 381, ATCC 33277, YPF1 and CW501 for 1.5 hours before being subjected to the antibiotic protection assay. Thus, internalized P. gingivalis were examined for their ability to induce HUVE cell death. HUVEC control cells treated with antibiotics produced th e same results as the untreated cells, suggesting that the presence of antibiotics did not cause any adverse affects on the cells. HUVECs exposed to CW501 and YPF1 followed the same pattern of cell death (approximately 16%) as the untreated HUVECs through 24 hours. However, consistent with the previous data, HUVECs exposed to both strains 381 and ATCC 33277 for 24 hours showed significantly higher levels of cell death relative to the untreated control cells and the two mutant strains. This data indicates

PAGE 176

176 that the internalized P. gingivalis are capable of stimulating th e HUVE cell death observed at 24 hours of co-culture. One aspect of this experiment was the obs ervation of 16% HUVE cell death at 2.5 hours with the untreated cells. This level of cell d eath was approximately 3 times the untreated control cell death level reported previously, and this level remain ed consistent through 24 hours. The HUVECs used in this experiment were older, passaged 10-14 times compared to the 4-8 passages of cells used with the pulsed P. gingivalis cell death assay. Since it is impossible to precisely standardize every experiment, an explanation for the discrepancy between the continuous exposure assay and the antibiotic protection assay could be experimenter error or a difference in the cell culturing t echniques. In addition, bacteria that were passaged fewer times were more active in cell death compared to cultu res from multiple passages. Because invasion is dependent on both bacterial and cellular me tabolism, discrepancies between different experimental results are most lik ely a consequence of the age of the bacterial and/or endothelial cells. We have shown that the proteins secreted from P. gingivalis (SPFs) are capable of disrupting the HUVEC monolay er. Therefore, the P. gingivalis 381 SPF was tested for its ability to stimulate HUVE cell death. Interestingly, HUVE Cs treated with the 381 SPF resulted in a significant increase in cell death compared to the untreated cont rol cells, but only after 24 hours of co-culture. In addition, th e exposure of HUVECs to CW501 at an MOI of 1000 also resulted in a significant increase in cell death compared to the untreated cells after 24 hours. However, the levels of cell death observed with CW501 and the 381 SPF were dramatically less than the cell death observed with the live strain 381 after the same times of exposure. The induction of HUVE cell death after exposure to the 381 SPF suggests that the surface and/or secreted proteins

PAGE 177

177 of P. gingivalis e.g., gingipains, play a role in this regard. Conversely, it is likely that P. gingivalis is equipped with additional mechanisms other than the gingipains for inducing cell death since the gingipain-null mutant CW 501 was able to induce HUVE cell death. Typically, bacteria are capable of inva ding host cells by evoking reassembly of cytoskeletal components interceded by actin polymerization (Rosenshine and Finlay, 1993). This is likely true for P. gingivalis because invasion of this pa thogen into HCAEC and MDDC cells was profoundly repressed when cytochalasin D, an inhibitor of actin polymerization, was added to the experiment (Jotwani and Cutler, 2004; Dorn et al. 1999). To further determine the importance of invasion on HUVE cell death, we decided to suppress P. gingivalis invasion of HUVECs. In this study, inhibiting the invasion of strains 381 and ATCC 33277 into HUVECs with cytochalasin D, completely inhibited HUVE cell death observed with these strains in pr evious experiments. Overall, P. gingivalis 381 and ATCC 33277 were capable of inducing HUVE cell death. This cell death was likely the result of P. gingivalis components, e.g., gingipains, acting from w ithin the HUVE cells instead of from the outside since the inhibition of P. gingivalis internalization appeared to suppress its ability to induce cell death of HUVECs. Thus, this data provides further support for the hypothesis that P. gingivalis gingipains and invasion of HUVECs are both important for P. gingivalis to induce cell death. The flow cytometry results previously reporte d using propidium iodide showed that there were significant increases in cell death compar ed to the untreated cells when HUVECs were exposed to strains 381 and ATCC 33277 with antib iotics (to kill any remaining extracellular bacteria) and without an tibiotic treatment. However, the quantification of the activation of caspase 3 activity showed only the HUVEC mo nolayers exposed to strain 381 and ATCC 33277

PAGE 178

178 without antibiotics had significan t increases in caspase 3 activ ity relative to the untreated HUVEC control. One explanation for this result is that there is a difference in the number of intracellular bacteria present at each time point. Other bacterial species such as A. actinomycetemcomitans and Salmonella spp ., can replicate intracellularly and multiple researchers have proposed that P. gingivalis is capable of this as well (Lamont et al. 1995; Blix et al. 1992; Finlay and Falkow, 1989). In fact a 90 minute incubation with GEC with antibiotics, which destroys the ex terior bacteria, produced a noticea ble increase in the number of internal bacteria af ter four hours (Lamont et al. 1995). Given that HUVE cells co-cultured with strains 381 and ATCC 33277 with and without antib iotic exposure yielded different caspase 3 activity levels, it is possible that this endothelial cell und ergoes more than one type of cell death in response to P. gingivalis exposure levels. Therefore, the P. gingivalis internalized within the HUVE cells treated with antibioti cs may induce a caspase independent form of HUVE cell death while the constant exposure of P. gingivalis to HUVE cells appears to stimulate caspase 3 dependent apoptosis. In summary, this study has provided evidence that the P. gingivalis conserved hypothetical protein, 0686, is crucial for this bacterium to induce HUVEC monol ayer disruption. In addition, the P. gingivalis gingipains are also important in th e disruption of the HUVEC monolayer and the re-localization of -catenin from the cell surface as well as in HUVE cell death. The P. gingivalis gingipains are responsible for the breakdown of the prot eins (e.g., catenins and VEcadherin) within endothelia l (HMVEC, BAEC and HUVEC) and (catenins and E-cadherin) within epithelial (KB, MDCK, GEC and HuH7) cell-cell junctional complexes (Chen et al. 2001a; Katz et al. 2000; Wang et al. 1999). The loss of cell-cell a dhesion due to the disruption of the HUVEC monolayer and th e adherens junction protein, -catenin, as well as the loss of

PAGE 179

179 cell-matrix adhesions (detachment of cells), may play a role in P. gingivalis induced cell death. The HUVE cell death induced by the internalized P. gingivalis after the administration of antibiotics might be directly linked to the cleavag e of the junctional proteins. The cleavage of the adherens junction proteins a nd subsequent cellular detachment raises the possibility that a unique form of apoptosis called anoikis, which has previously been suggested to be important in pathological progressions such as CVD, is induced by P. gingivalis gingipains (Michel, 2003; Frisch and Francis, 1994). In addition, Sheets et al also suggested that the gingipains may induce cell death via anoikis, si nce N-cadherin, vascular endothe lial cadherin (VE-cadherin) and integrin 1, were cleaved in bovine co ronary artery endothelial cells (BCAEC) and human microvascular endothelial cells (HMVEC), causing detachment from the culture surface (Sheets et al. 2005). Both anoikis and apoptosis can be induced through the lo ss of integrin (cellmatrix) signaling (Michel, 2003; Frisch and Franci s, 1994). As previously mentioned, integrins are members of non-covalently linked transmembran e, heterodimeric glycopr oteins. In addition to regulating attachment to th e ECM, integrins are also inducer s of signal transduction pathways that are responsible for cell transformation, differentiation, migration, proliferation, and inflammatory responses (Schoenwaelder a nd Burridge, 1999; Sheppa rd, 1996; Haynes and Webb, 1992). The increased tissue permeability i nduced by damage to the adherens junctions and extracellular matrix could potentially contribu te to systemic infections by allowing the entry of P. gingivalis into the vascular system (Chen et al. 2001b). Results from the caspase 3 activity assay also demonstrated the possibility that P. gingivalis is capable of utilizing at least two different mechanisms for initiating HUVE cell death. Flow cytometry experiments using pr opidium iodide and HUVECs co-cultured with P. gingivalis with (to kill any remaining extracellular bact eria) and without antib iotics resulted in

PAGE 180

180 significantly higher levels of cell death compared to the control cel ls. In contrast, the caspase 3 activity levels of the HUVECs co-cultured with P. gingivalis with and without antibiotics were different. Although the levels of caspase 3 activity for the HUVECs co-cultured with P. gingivalis and antibiotics were higher than the untreated cells, the difference level was insignificant. Only the continuous exposure of P. gingivalis with HUVECs yielded significantly higher levels of caspase 3 activity compared to the untreated cells. Since the antibiotics would have killed any extracellular P. gingivalis present within the wells, the death associated with increased caspase 3 levels of the HUVECs co-cultured with P. gingivalis and antibiotics must be due to the internalized P. gingivalis. This data suggests P. gingivalis located external to the cells may initiate one form of cell death while the internalized P. gingivalis initiate another. Therefore, in addition to the caspase 3 depende nt apoptosis observed after continuous exposure of P. gingivalis with HUVECs, P. gingivalis may also play a role in caspase independent apoptosis. The results from the PG0686 mutant suggest ed that this prot ein was crucial for P. gingivalis to induce HUVEC monolayer disruption. Future directions for this project could be to analyze and characterize the PG0686 protein. Also, it would be inte resting to determine if this mutant is capable of inducing HUVE cell death comparable to wild-type P. gingivalis or to the gingipain-null mutant, CW501. In addition, the us e of caspase inhibitors such as z-VAD-FMK (general caspase inhibitor), z-YVAD-FMK (caspa se-1 and -4 inhibitor), z-D(OMe)-E(OMe)VD(OMe)-CH2F (caspase-3 inhibitor) and z-VEID-FMK (caspase-6 inhibitor) would be beneficial in determining if P. gingivalis is capable of inducing casp ase independent apoptosis in HUVE cells. Finally, further characterization of the effects of the gingipains on the HUVECs

PAGE 181

181 would be greatly enhanc ed by the production of a P. gingivalis mutant that expresses the gingipains, adheres, but does not invade. The results reported here also prov ide support for the involvement of P. gingivalis in the initiation of cardiovascular dise ase. Alterations within the vessel wall caused from an inflammatory response to chronic infections are essential for the production of adhesion molecules, proinflammatory cytokines, chemokine s, inflammatory cell transmigration, and the formation of atheromas (Giacona et al. 2004). The development of inflammatory diseases such as periodontitis and CVD may depend on the severity of the inflammatory response to insult or injury. Myocardial infarctions may develop from the presence of severe inflammation within the coronary arteries (Kuramitsu et al. 2003). Atherosclerosis, which is likely a continual process, occurs as a consequence of arteri al cell injury, increased leukocyte adhesion, and is arbitrated by inflammation (Ross, 1999). Several pathogens ha ve been detected in human atheromas and suggested to be involved in the pathog enesis of atherosclerosis including Helicobacter pylori, Cytomegalovirus, Herpes simplex virus, St reptococcus sanguinis, Chalmydia pneumoniae, and P. gingivalis (Haraszthy et al. 2000; Chiu et al. 1997). Interestingly, PCR experiments identified at least one periodont al pathogen in 44% of the atheromas from the biopsied carotid endarterectomy specimens tested an d 26% of the observed pathogens was P. gingivalis (Haraszthy et al. 2000). Therefore, infection of cor onary cells with oral pathogens may aggravate the pathogenesis of car diovascular disease (CVD) (Dorn et al. 1999). The confirmation of this hypothesis would be significa nt in that the damage to the vessel walls associated with atherosclerotic plaque formati on could be minimized a nd/or possibly corrected by the direct control of pe riodontal pathogens through immu nological or antibacterial mechanisms (Desvarieux et al. 2005).

PAGE 182

182 LIST OF REFERENCES Abaibou, H., Ma, Q., Olango, G.J., Potempa, J., Tr avis, J. and Fletcher, H.M. (2000) Unaltered expression of the major protease genes in a non-virulent recA-defective mutant of Porphyromonas gingivalis W83. Oral Microbiol Immunol 15 : 40-47. Abe, N., Kadowaki, T., Okamoto, K., Nakayama K., Ohishi, M. and Yamamoto, K. (1998) Biochemical and functional pr operties of lysine-specific cysteine proteinase (Lysgingipain) as a virulence factor of Porphyromonas gingivalis in periodontal disease. J Biochem (Tokyo) 123 : 305-312. Abe, N., Baba, A., Takii, R., Naka yama, K., Kamaguchi, A., Shibata, Y. et al (2004) Roles of Argand Lys-gingipains in coaggregation of Porphyromonas gingivalis : identification of its responsible molecules in translation products of rgpA, kgp, and hagA genes. Biol Chem 385 : 1041-1047. Aberle, H., Schwartz, H. and Kemler, R. (1996) Cadherin-catenin complex: protein interactions and their implications for cadherin function. J Cell Biochem 61 : 514-523. Aberle, H., Bauer, A., Stappert, J ., Kispert, A. and Kemler, R. (1997) beta-catenin is a target for the ubiquitin-proteasome pathway. Embo J 16 : 3797-3804. Aduse-Opoku, J., Slaney, J.M., Hashim, A., Gallagher, A., Gallagher, R.P., Rangarajan, M. et al (2006) Identification and characterization of th e capsular polysaccharide (K-antigen) locus of Porphyromonas gingivalis Infect Immun 74 : 449-460. Allaker, R.P., Aduse-Opoku, J., Batten, J.E. and Cu rtis, M.A. (1997) Natura l variation within the principal arginine-specific protease gene, prpR1, of Porphyromonas gingivalis Oral Microbiol Immunol 12 : 298-302. Ally, N., Whisstock, J.C., Siepra wska-Lupa, M., Potempa, J., Le Bonniec, B.F., Travis, J. and Pike, R.N. (2003) Characterization of the speci ficity of arginine-s pecific gingipains from Porphyromonas gingivalis reveals active site differences between different forms of the enzymes. Biochemistry 42 : 11693-11700. Amano, A. (2003) Molecu lar interaction of Porphyromonas gingivalis with host cells: implication for the microbial pat hogenesis of periodontal disease. J Periodontol 74 : 90-96. Amano, A., Ishimoto, T., Tamagawa, H. and Sh izukuishi, S. (1992) Ro le of superoxide dismutase in resistance of Porphyromonas gingivalis to killing by polymorphonuclear leukocytes. Infect Immun 60 : 712-714. Amano, A., Sharma, A., Sojar, H.T., Kuramits u, H.K. and Genco, R.J. (1994) Effects of temperature stress on expression of fimbriae and superoxide dismutase by Porphyromonas gingivalis Infect Immun 62 : 4682-4685.

PAGE 183

183 Amano, A., Tamagawa, H., Takagaki, M., Muraka mi, Y., Shizukuishi, S. and Tsunemitsu, A. (1988) Relationship between enzyme activit ies involved in oxygen metabolism and oxygen tolerance in black-pigmented Bacteroides. J Dent Res 67 : 1196-1199. Amano, A., Shizukuishi, S., Horie, H., Kimura, S., Morisaki, I. and Ha mada, S. (1998) Binding of Porphyromonas gingivalis fimbriae to proline-rich glyc oproteins in parotid saliva via a domain shared by major salivary components. Infect Immun 66 : 2072-2077. Andrews, J.M., Oglesbee, M. and Lairmore, M. D. (1998) The effect of the cellular stress response on human T-lymphotropic virus type I envelope protein expression. J Gen Virol 79 ( Pt 12) : 2905-2908. Angst, B.D., Marcozzi, C. and Magee, A.I. (2001a) The cadherin superfamily. J Cell Sci 114 : 625-626. Angst, B.D., Marcozzi, C. and Magee, A.I. (200 1b) The cadherin superfamily: diversity in form and function. J Cell Sci 114 : 629-641. Ansai, T., Yu, W., Urnowey, S., Barik, S. and Ta kehara, T. (2003) Construction of a pepO genedeficient mutant of Porphyromonas gingivalis : potential role of endopeptidase O in the invasion of host cells. Oral Microbiol Immunol 18 : 398-400. Ansai, T., Yamamoto, E., Awano, S., Yu, W., Turn er, A.J. and Takehara, T. (2002) Effects of periodontopathic bacteria on the expression of endothelin-1 in gingiva l epithelial cells in adult periodontitis. Clin Sci (Lond) 103 Suppl 48 : 327S-331S. Aoudjit, F. and Vuori, K. (2001) Matrix att achment regulates Fas-induced apoptosis in endothelial cells: a role for c-flip and implications for anoikis. J Cell Biol 152 : 633-643. Aplin, A.E. and Juliano, R.L. (1999) Integrin a nd cytoskeletal regulati on of growth factor signaling to the MAP kinase pathway. J Cell Sci 112 ( Pt 5) : 695-706. Aplin, A.E., Howe, A., Alahari, S.K. and Juli ano, R.L. (1998) Signal transduction and signal modulation by cell adhesion receptors: the ro le of integrins, cadherins, immunoglobulincell adhesion molecules, and selectins. Pharmacol Rev 50 : 197-263. Arribas, J., Coodly, L., Vollmer, P., Kishimot o, T.K., Rose-John, S. and Massague, J. (1996) Diverse cell surface protein ectodomains are sh ed by a system sensitive to metalloprotease inhibitors. J Biol Chem 271: 11376-11382. Awano, S., Ansai, T., Mochizuki, H., Yu, W., Tanzaw a, K., Turner, A.J. and Takehara, T. (1999) Sequencing, expression and bioche mical characterization of the Porphyromonas gingivalis pepO gene encoding a protein homologous to human endothelin-converting enzyme. FEBS Lett 460 : 139-144. Baba, A., Kadowaki, T., Asao, T. and Yamamot o, K. (2002) Roles for Argand Lys-gingipains in the disruption of cytokine responses and lo ss of viability of huma n endothelial cells by Porphyromonas gingivalis infection. Biol Chem 383 : 1223-1230.

PAGE 184

184 Baba, A., Abe, N., Kadowaki, T., Nakanishi, H ., Ohishi, M., Asao, T. and Yamamoto, K. (2001) Arg-gingipain is responsible for the degrad ation of cell adhesion molecules of human gingival fibroblasts and their death induced by Porphyromonas gingivalis Biol Chem 382 : 817-824. Balkovetz, D.F. and Katz, J. (2003) Bacterial invasion by a paracellula r route: divide and conquer. Microbes Infect 5 : 613-619. Banbula, A., Bugno, M., Kuster, A., Heinrich, P.C., Travis, J. and Potempa, J. (1999) Rapid and efficient inactivation of IL-6 gingipains, ly sineand arginine-speci fic proteinases from Porphyromonas gingivalis Biochem Biophys Res Commun 261 : 598-602. Barter, P.J. and Rye, K.A. (1996) High density lipoproteins and coronary heart disease. Atherosclerosis 121 : 1-12. Barter, P.J., Baker, P.W. and Rye, K.A. ( 2002) Effect of high-density lipoproteins on the expression of adhesion molecu les in endothelial cells. Curr Opin Lipidol 13 : 285-288. Bazzoni, G. and Dejana, E. (2001) Pores in the sieve and channels in the wall: control of paracellular permeability by junctional proteins in endothelial cells. Microcirculation 8 : 143-152. Bazzoni, G. and Dejana, E. (2004) Endothelial cell-to-cell juncti ons: molecular organization and role in vascular homeostasis. Physiol Rev 84 : 869-901. Bazzoni, G., Dejana, E. and La mpugnani, M.G. (1999) Endothelia l adhesion molecules in the development of the vascular tree : the garden of forking paths. Curr Opin Cell Biol 11 : 573-581. Beck, J., Garcia, R., Heiss, G., Vokonas, P.S. and Offenbacher, S. (1996) Periodontal disease and cardiovascular disease. J Periodontol 67 : 1123-1137. Beck, J.D. and Offenbacher, S. (2001) The association between pe riodontal diseases and cardiovascular diseases: a st ate-of-the-science review. Ann Periodontol 6 : 9-15. Beck, J.D., Slade, G. and Offenbacher, S. ( 2000) Oral disease, cardi ovascular disease and systemic inflammation. Periodontol 2000 23 : 110-120. Buck, V., Quinn, J., Soto Pino, T., Martin, H., Saldanha, J., Makino, K., Morgan, B.A., and Millar, Jonathan B.A. (2001) Peroxide Sens ors for the Fission Yeast Stress-activated Mitogen-activated Protein Kinase Pathway. Molecular Biology of the Cell 12 : 407-419. Bedi, G.S. and Williams, T. (1994) Purification and characterization of a collagen-degrading protease from Porphyromonas gingivalis J Biol Chem 269 : 599-606. Belanger, M., Rodrigues, P.H ., Dunn, W.A., Jr. and Progulsk e-Fox, A. (2006) Autophagy: a highway for Porphyromonas gingivalis in endothelial cells. Autophagy 2 : 165-170.

PAGE 185

185 Belton, C.M., Izutsu, K.T., Goodwin, P.C., Park, Y. and Lamont, R.J. (1999) Fluorescence image analysis of the association between Porphyromonas gingivalis and gingival epithelial cells. Cell Microbiol 1 : 215-223. Ben-Ze'ev, A. and Geiger, B. (1998) Differential molecular interactions of beta-catenin and plakoglobin in adhesi on, signaling and cancer. Curr Opin Cell Biol 10 : 629-639. Beveridge, T.J. and Kadurugamuwa, J.L. (1996) Pe riplasm, periplasmic spaces, and their relation to bacterial wall structure: novel secretion of selected periplasmic proteins from Pseudomonas aeruginosa Microb Drug Resist 2 : 1-8. Bhogal, P.S., Slakeski, N. and Reynolds, E.C. (1997) A cell-associated protein complex of Porphyromonas gingivalis W50 composed of Argand Ly s-specific cysteine proteinases and adhesins. Microbiology 143 ( Pt 7) : 2485-2495. Blix, I.J., Hars, R., Preus, H.R. and Helgeland, K. (1992) Entrance of Actinobacillus actinomycetemcomitans into HEp-2 cells in vitro. J Periodontol 63 : 723-728. Boyd, J. and McBride, B.C. (1984) Fractionati on of hemagglutinating and bacterial binding adhesins of Bacteroides gingivalis Infect Immun 45 : 403-409. Bracke, M.E., Van Roy, F.M. and Mareel, M.M. (1996) The E-cadherin/ catenin complex in invasion and metastasis. Curr Top Microbiol Immunol 213 ( Pt 1) : 123-161. Braga, V.M. (2002) Cell-cell adhesion and signalling. Curr Opin Cell Biol 14 : 546-556. Breier, G., Breviario, F., Caveda, L., Berthier, R., Schnurch, H., Gotsch, U. et al (1996) Molecular cloning and expression of murine va scular endothelial-cadherin in early stage development of cardiovascular system. Blood 87 : 630-641. Bretscher, A. (1999) Regulation of cortical stru cture by the ezrin-radixinmoesin protein family. Curr Opin Cell Biol 11 : 109-116. Brook, I. (1987) Role of encapsulated anaerob ic bacteria in syne rgistic infections. Crit Rev Microbiol 14 : 171-193. Butz, S., Stappert, J., Weissig, H. and Kemler, R. (1992) Plakoglobin and beta-catenin: distinct but closely related. Science 257 : 1142-1144. Calkins, C.C., Platt, K., Potempa, J. and Travis J. (1998) Inactivation of tumor necrosis factoralpha by proteinases (gingipains) from the periodontal pathogen, Porphyromonas gingivalis Implications of immune evasion. J Biol Chem 273 : 6611-6614. Carmeliet, P., Lampugnani, M.G., Moons, L ., Breviario, F., Comp ernolle, V., Bono, F. et al (1999) Targeted deficiency or cytosolic tr uncation of the VE-cadherin gene in mice impairs VEGF-mediated endothelial survival and angiogenesis. Cell 98 : 147-157.

PAGE 186

186 Carroll, G.C. and Sebor, R.J. (1980) Dental flossing and its relationship to tr ansient bacteremia. J Periodontol 51 : 691-692. Castelli, W.P., Garrison, R.J., Wilson, P.W., Ab bott, R.D., Kalousdian, S. and Kannel, W.B. (1986) Incidence of coronary heart diseas e and lipoprotein cholesterol levels. The Framingham Study. JAMA 256 : 2835-2838. Castillo, A.M. and Alvarez, M. (2004) Periodon tal diseases: microbiol ogical considerations. Orofacial infections 9 : 75-91. Cattelino, A., Liebner, S., Gallini, R., Zanetti, A., Balconi, G., Corsi, A. et al (2003) The conditional inactivation of the beta-catenin ge ne in endothelial cells causes a defective vascular pattern and increased vascular fragility. J Cell Biol 162 : 1111-1122. Cavrini, F., Sambri, V., Moter, A., Se rvidio, D., Marangoni, A., Montebugnoli, L. et al (2005) Molecular detection of Treponema denticola and Porphyromonas gingivalis in carotid and aortic atheromatous plaques by FISH: report of two cases. J Med Microbiol 54 : 93-96. Chen, S., Guttridge, D.C., You, Z., Zhang, Z., Fribley, A., Mayo, M.W. et al (2001a) Wnt-1 signaling inhibits apoptosis by ac tivating beta-catenin /T cell factor-med iated transcription. J Cell Biol 152 : 87-96. Chen, T., Nakayama, K., Belliveau, L. and Duncan, M.J. (2001b) Porphyromonas gingivalis gingipains and adhesion to epithelial cells. Infect Immun 69 : 3048-3056. Chen, W. and Kuramitsu, H.K. (1999) Molecular mechanism for the spontaneous generation of pigmentless Porphyromonas gingivalis mutants. Infect Immun 67 : 4926-4930. Chen, Z., Casiano, C.A. and Fletcher, H.M. (2001c) Protease-active extracellular protein preparations from Porphyromonas gingivalis W83 induce N-cadherin proteolysis, loss of cell adhesion, and apoptosis in human epithelial cells. J Periodontol 72 : 641-650. Chen, Z., Potempa, J., Polanowski, A., Wikstrom M. and Travis, J. (1992) Purification and characterization of a 50-kDa cystei ne proteinase (gingipain) from Porphyromonas gingivalis J Biol Chem 267 : 18896-18901. Chitaev, N.A. and Troyanovsky, S.M. (1998) Adhe sive but not lateral E-cadherin complexes require calcium and catenins for their formation. J Cell Biol 142 : 837-846. Chiu, B. (1999) Multiple infections in carotid atherosclerotic plaques. Am Heart J 138 : S534536. Chiu, B., Viira, E., Tucker, W. and Fong, I.W. (1997) Chlamydia pneumonia e, cytomegalovirus and herpes simplex virus in atherosclerosi s of the carotid artery. Circulation 96 : 21442148.

PAGE 187

187 Chou, H.H., Yumoto, H., Davey, M., Takahashi, Y., Miyamoto, T., Gibson, F.C., 3rd and Genco, C.A. (2005) Porphyromonas gingivalis fimbria-dependent acti vation of inflammatory genes in human aortic endothelial cells. Infect Immun 73 : 5367-5378. Ciborowski, P., Nishikata, M., Allen, R.D. and Lantz, M.S. (1994) Purification and characterization of two forms of a high-molecu lar-weight cysteine pr oteinase (porphypain) from Porphyromonas gingivalis J Bacteriol 176 : 4549-4557. Codogno, P. and Meijer, A.J. (2005) Autophagy and signaling: their ro le in cell survival and cell death. Cell Death Differ 12 Suppl 2 : 1509-1518. Collaco, C., Dyer, R.B., Doan, R., He rzog, N.K. and Niesel, D.W. (1995) Shigella flexneri -HeLa cell interactions: a putative role for host cell protein kinases. FEMS Immunol Med Microbiol 10 : 93-100. Crossin, K.L. and Krushel, L.A. (2000) Cellular signaling by neural cell adhesion molecules of the immunoglobulin superfamily. Dev Dyn 218 : 260-279. Curtis, M.A., Kuramitsu, H.K., Lantz, M., Macrina, F.L., Nakayama, K., Potempa, J. et al (1999a) Molecular genetics and nomenclature of proteases of Porphyromonas gingivalis J Periodontal Res 34 : 464-472. Curtis, M.A., Thickett, A., Slaney, J.M., Rangarajan, M., Aduse-Opoku, J., Shepherd, P. et al (1999b) Variable carbohydrate modi fications to the catalytic ch ains of the RgpA and RgpB proteases of Porphyromonas gingivalis W50. Infect Immun 67 : 3816-3823. Cutler, C.W., Kalmar, J.R. and Genco, C.A. (1995) Pathogenic strategies of the oral anaerobe, Porphyromonas gingivalis Trends Microbiol 3 : 45-51. Dahlen, G. and Nygren, H. (1982) An electron mi croscopic study of surface polysaccharides in Bacteroides Microbios 35 : 119-129. Dahlen, G., Fabricius, L., Heyden, G., Holm, S.E. and Moller, A.J. (1982) Apical periodontitis induced by selected bacterial strains in root canals of immunized and nonimmunized monkeys. Scand J Dent Res 90 : 207-216. Daly, C., Mitchell, D., Grossberg, D., Highfield, J. and Stewart, D. (1997) Bacteraemia caused by periodontal probing. Aust Dent J 42 : 77-80. Darveau, R.P., Belton, C.M., Reife, R.A. and Lam ont, R.J. (1998) Local chemokine paralysis, a novel pathogenic mechanism for Porphyromonas gingivalis Infect Immun 66 : 1660-1665. Darveau, R.P., Cunningham, M.D., Bailey, T., Seachord, C., Ratcliffe K., Bainbridge, B. et al (1995) Ability of bacteria associated with chronic inflammatory disease to stimulate Eselectin expression and promote neutrophil adhesion. Infect Immun 63 : 1311-1317. DeCarlo, A.A. and Harber, G.J. (1997) Hemaggl utinin activity and hete rogeneity of related Porphyromonas gingivalis proteinases. Oral Microbiol Immunol 12 : 47-56.

PAGE 188

188 Dejana, E. (2004) Endothelial cell -cell junctions: happy together. Nat Rev Mol Cell Biol 5 : 261270. Dejana, E., Corada, M. and Lampugnani, M.G. (1995) Endot helial cell-tocell junctions. Faseb J 9 : 910-918. Dejana, E., Bazzoni, G. and Lampugnani, M.G. (1999a) The role of endothelial cell-to-cell junctions in vascular morphogenesis. Thromb Haemost 82 : 755-761. Dejana, E., Bazzoni, G. and Lampugnani, M.G. (1999b) Vascular endoth elial (VE)-cadherin: only an intercellular glue? Exp Cell Res 252 : 13-19. Dejana, E., Spagnuolo, R. and Bazzoni, G. (2001) In terendothelial junctions and their role in the control of angiogenesis, vascular permeability and leukocyte transmigration. Thromb Haemost 86 : 308-315. Deshpande, R.G. and Khan, M.B. (1999) Purifica tion and characterization of hemolysin from Porphyromonas gingivalis A7436. FEMS Microbiol Lett 176 : 387-394. Deshpande, R.G., Khan, M.B. and Genco, C.A. ( 1998a) Invasion of aortic and heart endothelial cells by Porphyromonas gingivalis Infect Immun 66 : 5337-5343. Deshpande, R.G., Khan, M. and Genco, C.A. (1998b) Invasion strategies of the oral pathogen Porphyromonas gingivalis : implications for cardiovascular disease. Invasion Metastasis 18 : 57-69. Desvarieux, M., Demmer, R.T., Rundek, T., BodenAlbala, B., Jacobs, D.R., Jr., Sacco, R.L. and Papapanou, P.N. (2005) Periodontal microbiota and carotid intima-me dia thickness: the Oral Infections and Vascular Disease Epidemiology Study (INVEST). Circulation 111 : 576-582. Dickinson, D.P., Kubiniec, M.A., Yoshimura, F. and Genco, R.J. (1988) Molecular cloning and sequencing of the gene encoding the fimbrial subunit protein of Bacteroides gingivalis J Bacteriol 170 : 1658-1665. Ding, Y., Uitto, V.J., Firth, J., Salo, T., Haapasal o, M., Konttinen, Y.T. and Sorsa, T. (1995) Modulation of host matrix metalloproteinases by bacterial virulence factors relevant in human periodontal diseases. Oral Dis 1 : 279-286. Donnenberg, M.S., Donohue-Rolfe, A. and Keusc h, G.T. (1990) A comparison of HEp-2 cell invasion by enteropathogenic and enteroinvasive Escherichia coli FEMS Microbiol Lett 57 : 83-86. Dorn, B.R., Dunn, W.A., Jr. and Progulske-Fox, A. (1999) Invasi on of human coronary artery cells by periodontal pathogens. Infect Immun 67 : 5792-5798. Dorn, B.R., Dunn, W.A., Jr. a nd Progulske-Fox, A. (2001) Porphyromonas gingivalis traffics to autophagosomes in human corona ry artery endothelial cells. Infect Immun 69 : 5698-5708.

PAGE 189

189 Dorn, B.R., Burks, J.N., Seifert, K.N. and Pr ogulske-Fox, A. (2000) Invasion of endothelial and epithelial cells by strains of Porphyromonas gingivalis FEMS Microbiol Lett 187 : 139144. Driscoll, D.M. and Getz, G.S. (1984) Extr ahepatic synthesis of apolipoprotein E. J Lipid Res 25 : 1368-1379. Duncan, M.J., Shin, J.S. and Abraham, S.N. (20 02) Microbial entry thro ugh caveolae: variations on a theme. Cell Microbiol 4 : 783-791. Duncan, M.J., Li, G., Shin, J.S., Carson, J.L. and Abraham, S.N. (2004) Bacterial penetration of bladder epithelium through lipid rafts. J Biol Chem 279 : 18944-18951. Dunn, W.A., Jr. (1990) Studies on the mechanisms of autophagy: formation of the autophagic vacuole. J Cell Biol 110 : 1923-1933. Dunn, W.A., Jr. (1994) Autophagy and related mechanisms of lysosome-mediated protein degradation. Trends Cell Biol 4 : 139-143. Ehlers, M.R. and Riordan, J.F. (1991) Membrane proteins with soluble counterparts: role of proteolysis in the release of transmembrane proteins. Biochemistry 30 : 10065-10074. Ellen, R.P., Lepine, G. and Nghiem, P.M. (1997) In vitro models that support adhesion specificity in biofilms of oral bacteria. Adv Dent Res 11 : 33-42. Epand, R.M., Stafford, A., Leon, B., Lock, P.E., Ty tler, E.M., Segrest, J.P. and Anantharamaiah, G.M. (1994) HDL and apolipoprotein A-I protect erythrocytes agains t the generation of procoagulant activity. Arterioscler Thromb 14 : 1775-1783. Fagotto, F. and Gumbiner, B.M. (199 6) Cell contact-dep endent signaling. Dev Biol 180 : 445454. Fanning, A.S., Mitic, L.L. and Anderson, J.M. (1999) Transmembrane proteins in the tight junction barrier. J Am Soc Nephrol 10 : 1337-1345. Finlay, B.B. and Falkow, S. (1989) Common themes in microbial pathogenicity. Microbiol Rev 53 : 210-230. Finlay, B.B. and Falkow, S. (1997) Common themes in microbi al pathogenicity revisited. Microbiol Mol Biol Rev 61: 136-169. Fives-Taylor, P.M., Meyer, D.H., Mintz, K.P. and Brissette, C. (1999) Virulence factors of Actinobacillus acti nomycetemcomitans Periodontol 2000 20 : 136-167. Fleisher, L.N., Tall, A.R., Witte, L.D., Miller, R.W. and Cannon, P.J. (1982) Stimulation of arterial endothelial cell prostacyclin synthesis by high density lipoproteins. J Biol Chem 257 : 6653-6655.

PAGE 190

190 Fletcher, H.M., Schenkein, H.A. and Macrina, F. L. (1994) Cloning and characterization of a new protease gene (prtH) from Porphyromonas gingivalis Infect Immun 62 : 4279-4286. Frisch, S.M. and Francis, H. (1994) Disruption of epithelial cell-matri x interactions induces apoptosis. J Cell Biol 124 : 619-626. Fujimura, S. and Nakamura, T. (1990) Purifica tion and characterization of a 43-kDa protease of Bacteroides gingivalis Oral Microbiol Immunol 5 : 360-362. Fujimura, S., Shibata, Y. and Nakamura, T. (199 3) Purification and partial characterization of a lysine-specific protease of Porphyromonas gingivalis FEMS Microbiol Lett 113 : 133137. Fujimura, S., Hirai, K., Shibata, Y., Nakaya ma, K. and Nakamura, T. (1998) Comparative properties of envelope-associated argini ne-gingipains and lysine-gingipain of Porphyromonas gingivalis FEMS Microbiol Lett 163 : 173-179. Furuse, M., Hirase, T., Itoh, M., Nagafuchi, A., Yonemura, S. and Tsukita, S. (1993) Occludin: a novel integral membrane protein localizing at tight junctions. J Cell Biol 123 : 1777-1788. Gamonal, J., Bascones, A., Acevedo, A., Blanco, E. and Silva, A. (2001) Apoptosis in chronic adult periodontitis analyzed by in si tu DNA breaks, electron microscopy, and immunohistochemistry. J Periodontol 72 : 517-525. Garcia, R.I., Henshaw, M.M. and Krall, E.A. (2001) Relationship between periodontal disease and systemic health. Periodontol 2000 25 : 21-36. Garton, K.J., Gough, P.J., Philalay, J., Wille P.T., Blobel, C.P., Whitehead, R.H. et al (2003) Stimulated shedding of vascular cell adhesion molecule 1 (VCAM-1) is mediated by tumor necrosis factor-alpha-con verting enzyme (ADAM 17). J Biol Chem 278 : 37459-37464. Genco, C.A., Odusanya, B.M. and Brown, G. (1994) Binding and accumulation of hemin in Porphyromonas gingivalis are induced by hemin. Infect Immun 62 : 2885-2892. Genco, C.A., Potempa, J., Mikolajczyk-Pawlinska, J. and Travis, J. (1999) Role of gingipains R in the pathogenesis of Porphyromonas gingivalis -mediated periodontal disease. Clin Infect Dis 28 : 456-465. Genco CA, P.J., Mikolajczk-pawlinska J, Trav is J (1999) Role of Gingipains R in the Pathogenesis of Porphyromonas gingivalis -Mediated Periodontal Disease. Clinical Infectious Diseases 28 : 456-465. Genco, R.J. (1992a) Host responses in periodontal diseases: current concepts. J Periodontol 63 : 338-355. Genco, R.J. (1992b) New horizons in peri odontics. The challenge for diagnosis. N Y State Dent J 58 : 32-36.

PAGE 191

191 Genco, R.J. (1992c) Host responses in periodontal disease: current concepts. Journal of Periodontal Research 63 : 338-355. Ghiselli, G., Schaefer, E.J., Gascon, P. and Breser, H.B., Jr. (1981) Type III hyperlipoproteinemia associated with apolipoprotein E deficiency. Science 214 : 12391241. Giacona, M.B., Papapanou, P.N., Lamster, I.B., R ong, L.L., D'Agati, V.D., Schmidt, A.M. and Lalla, E. (2004) Porphyromonas gingivalis induces its uptake by human macrophages and promotes foam cell formation in vitro. FEMS Microbiol Lett 241 : 95-101. Giancotti, F.G. and Ruoslahti, E. (1999) Integrin signaling. Science 285 : 1028-1032. Gibbons, R.J., Hay, D.I., Childs, W.C., 3rd and Da vis, G. (1990) Role of cryptic receptors (cryptitopes) in bacterial adhesion to oral surfaces. Arch Oral Biol 35 Suppl : 107S-114S. Gibson, F.C., 3rd, Yumoto, H., Takahashi, Y ., Chou, H.H. and Genco, C.A. (2006) Innate immune signaling and Porphyromonas gingivalis -accelerated atherosclerosis. J Dent Res 85 : 106-121. Gibson, F.C., 3rd, Hong, C., Chou, H.H., Yumoto, H., Chen, J., Lien, E. et al (2004) Innate immune recognition of invasive bacteria accelerates atherosc lerosis in apolipoprotein Edeficient mice. Circulation 109 : 2801-2806. Gilmore, A.P., Metcalfe, A.D., Romer, L.H. and Streuli, C.H. (2000) Integrin-mediated survival signals regulate the apoptotic function of Bax through its conformation and subcellular localization. J Cell Biol 149 : 431-446. Glurich, I., Grossi, S., Albini, B., Ho, A., Shah, R., Zeid, M. et al (2002) Systemic inflammation in cardiovascular and periodontal disease: comparative study. Clin Diagn Lab Immunol 9 : 425-432. Goadby, K. (1907) Erasmus Wilson' s lecture on pyorrhea alveolaris. Lancet i : 633. Gordon, T., Castelli, W.P., Hjortland, M.C., Kannel, W.B. and Dawber, T.R. (1977) High density lipoprotein as a protective factor ag ainst coronary heart disease. The Framingham Study. Am J Med 62 : 707-714. Gottardi, C.J., Wong, E. and Gumbiner, B.M. (2001) E-cadherin suppresses cellular transformation by inhibiting beta-catenin sign aling in an adhesion-independent manner. J Cell Biol 153 : 1049-1060. Goulhen, F., Grenier, D. and Mayrand, D. (2003) Oral microbial heat-shock proteins and their potential contributi ons to infections. Crit Rev Oral Biol Med 14 : 399-412. Graves, D.T. and Jiang, Y. (1995) Chemokine s, a family of chemotactic cytokines. Crit Rev Oral Biol Med 6 : 109-118.

PAGE 192

192 Graves, D.T., Oskoui, M., Volejnikova, S., Naguib, G., Cai, S., Desta, T. et al (2001) Tumor necrosis factor modulates fibroblast apoptosis PMN recruitment, and osteoclast formation in response to P. gingivalis infection. J Dent Res 80 : 1875-1879. Grayson, R., Douglas, C.W., Heath, J., Rawlinson, A. and Evans, G.S. (2003) Activation of human matrix metalloproteinase 2 by gingival crevic ular fluid and Porphyromonas gingivalis J Clin Periodontol 30 : 542-550. Grenier, D. and Mayrand, D. (1987) Functional characterization of ex tracellular vesicles produced by Bacteroides gingivalis Infect Immun 55 : 111-117. Grenier, D., Roy, S., Chandad, F., Plamondon, P., Yoshioka, M., Nakayama K. and Mayrand, D. (2003) Effect of inactivation of the Argand/ or Lys-gingipain gene on selected virulence and physiological properties of Porphyromonas gingivalis Infect Immun 71 : 4742-4748. Grossi, S.G., Zambon, J.J., Ho, A.W., Ko ch, G., Dunford, R.G., Machtei, E.E. et al (1994) Assessment of risk for periodontal disease. I. Risk indicators for attachment loss. J Periodontol 65 : 260-267. Guignot, J., Bernet-Camard, M.F., Pous, C., Plan con, L., Le Bouguenec, C. and Servin, A.L. (2001) Polarized entry of uropat hogenic Afa/Dr diffusely adhering Escherichia coli strain IH11128 into human epithelial cells: eviden ce for alpha5beta1 integrin recognition and subsequent internalization through a pathway involving caveolae and dynamic unstable microtubules. Infect Immun 69 : 1856-1868. Gumbiner, B.M. (1993) Breaking th rough the tight junction barrier. J Cell Biol 123 : 1631-1633. Gumbiner, B.M. (1995) Signal tr ansduction of beta-catenin. Curr Opin Cell Biol 7 : 634-640. Gumbiner, B.M. (2005) Regulation of cadhe rin-mediated adhesion in morphogenesis. Nat Rev Mol Cell Biol 6 : 622-634. Gumbiner, B.M. and McCrea, P.D. (1993) Catenins as mediators of the cytoplasmic functions of cadherins. J Cell Sci Suppl 17 : 155-158. Gumbiner, B.M. and Yamada, K.M. (1995) Cellto-cell contact and ex tracellular matrix. Curr Opin Cell Biol 7 : 615-618. Haffajee, A.D., Arguello, E.I., Ximenez-Fyvie, L.A. and Socransky, S.S. (2003) Controlling the plaque biofilm. Int Dent J 53 Suppl 3 : 191-199. Hamada, N., Sojar, H.T., Cho, M.I. and Genco, R. J. (1996) Isolation and characterization of a minor fimbria from Porphyromonas gingivalis Infect Immun 64 : 4788-4794. Hamada, N., Watanabe, K., Sasakawa, C., Yosh ikawa, M., Yoshimura, F. and Umemoto, T. (1994a) Construction and characteri zation of a fimA mutant of Porphyromonas gingivalis Infect Immun 62 : 1696-1704.

PAGE 193

193 Hamada, S., Amano, A., Kimura, S., Nakagawa, I ., Kawabata, S. and Morisaki, I. (1998) The importance of fimbriae in the virulen ce and ecology of some oral bacteria. Oral Microbiol Immunol 13 : 129-138. Hamada, S., Fujiwara, T., Morishima, S., Takahash i, I., Nakagawa, I., Kimura, S. and Ogawa, T. (1994b) Molecular and immunological char acterization of the fimbriae of Porphyromonas gingivalis. Microbiol Immunol 38 : 921-930. Han, N., Whitlock, J. and Progulske-Fox, A. (1996) The hemagglutinin gene A (hagA) of Porphyromonas gingivalis 381 contains four large, contiguous, direct repeats. Infect Immun 64 : 4000-4007. Haraszthy, V.I., Zambon, J.J., Tr evisan, M., Zeid, M. and Genco, R.J. (2000) Identification of periodontal pathogens in atheromatous plaques. J Periodontol 71 : 1554-1560. Hartwell, D.W. and Wagner, D.D. (1999) New disc overies with mice mutant in endothelial and platelet selectins. Thromb Haemost 82 : 850-857. Haynes, W.G. and Webb, D.J. (1992) Endothelin: a long-acti ng local constrictor hormone. Br J Hosp Med 47 : 340-349. Heasman, J., Crawford, A., Goldstone, K., Garner-Hamrick, P., Gumbiner, B., McCrea, P. et al (1994) Overexpression of cadherins and undere xpression of beta-caten in inhibit dorsal mesoderm induction in early Xenopus embryos. Cell 79 : 791-803. Hegele, R.A. (1999) Paraoxonase genes and disease. Ann Med 31 : 217-224. Hermiston, M.L. and Gordon, J.I. (1995a) In vivo analysis of cadherin function in the mouse intestinal epithelium: essen tial roles in adhesion, maintenance of differentiation, and regulation of programmed cell death. J Cell Biol 129 : 489-506. Hermiston, M.L. and Gordon, J.I. (1995b) Inflamma tory bowel disease and adenomas in mice expressing a dominant negative N-cadherin. Science 270 : 1203-1207. Hermiston, M.L., Wong, M.H. and Gordon, J.I. ( 1996) Forced expression of E-cadherin in the mouse intestinal epithelium slows cell migration and provides evidence for nonautonomous regulation of cell fa te in a self-renewing system. Genes Dev 10 : 985-996. Herren, B., Levkau, B., Raines, E.W. and Ross, R. (1998) Cleavage of beta-catenin and plakoglobin and shedding of VE-cadherin durin g endothelial apoptosis: evidence for a role for caspases and metalloproteinases. Mol Biol Cell 9 : 1589-1601. Herrenknecht, K., Ozawa, M., Eckerskorn, C., Lotts peich, F., Lenter, M. and Kemler, R. (1991) The uvomorulin-anchorage protein alpha catenin is a vinculin homologue. Proc Natl Acad Sci U S A 88 : 9156-9160.

PAGE 194

194 Hintermann, E., Haake, S.K., Christen, U., Shar abi, A. and Quaranta, V. (2002) Discrete proteolysis of focal contact and adherens junction components in Porphyromonas gingivalis -infected oral keratinocytes: a strategy for cell adhesion and migration disabling. Infect Immun 70 : 5846-5856. Hiramine, H., Watanabe, K., Hama da, N. and Umemoto, T. (2003) Porphyromonas gingivalis 67-kDa fimbriae induced cytoki ne production and osteoclast differentiation utilizing TLR2. FEMS Microbiol Lett 229 : 49-55. Hordijk, P. (2003) Endothelial signal ing in leukocyte transmigration. Cell Biochem Biophys 38 : 305-322. Hormia, M., Ylanne, J. and Virtanen, I. (1990) Expression of integrins in human gingiva. J Dent Res 69 : 1817-1823. Hosogi, Y. and Duncan, M.J. (2005) Gene expression in Porphyromonas gingivalis after contact with human epithelial cells. Infect Immun 73 : 2327-2335. Houalet-Jeanne, S., Pellen-Mussi, P., Tricot-D oleux, S., Apiou, J. and Bonnaure-Mallet, M. (2001) Assessment of interna lization and viability of Porphyromonas gingivalis in KB epithelial cells by c onfocal microscopy. Infect Immun 69 : 7146-7151. Hu, Y., Kiely, J.M., Szente, B.E., Rosenzweig, A. and Gimbrone, M.A., Jr. (2000) E-selectindependent signaling via the mitogen-activated protein kinase pa thway in vascular endothelial cells. J Immunol 165 : 2142-2148. Hurlstone, A.F., Haramis, A.P., Wienholds, E., Begthel, H., Korving, J., Van Eeden, F. et al (2003) The Wnt/beta-catenin pathway regulates cardiac valve formation. Nature 425 : 633637. Hynes, R.O. (1999) Cell adhe sion: old and new questions. Trends Cell Biol 9 : M33-37. Isaza, C.E., Silaghi-Dumitrescu, R., Iyer, R.B. Krutz, D.M. Jr., and Chan, M.K. (2006) Structural Basis for O2 Sensing by the Hemerythrin-like Domain of a bacterial Chemotaxis Protein: Substrate Tunnel and Fluxional N Terminus. Biochemistry 45 : 9023-9031. Jain, A., Batista, E.L., Jr., Serhan, C., Stah l, G.L. and Van Dyke, T.E. (2003) Role for periodontitis in the progre ssion of lipid deposition in an animal model. Infect Immun 71 : 6012-6018. Johansson, A. and Kalfas, S. (1998) Characterizati on of the proteinase-depe ndent cytotoxicity of Porphyromonas gingivalis Eur J Oral Sci 106 : 863-871. Johnson-Leger, C., Aurrand-Lions, M. and Imhof, B.A. (2000) The parting of the endothelium: miracle, or simply a junctional affair? J Cell Sci 113 ( Pt 6) : 921-933.

PAGE 195

195 Jotwani, R. and Cutler, C.W. (2004) Fimbriated Porphyromonas gingivalis is more efficient than fimbria-deficient P. gingivalis in entering human dendritic ce lls in vitro and induces an inflammatory Th1 effector response. Infect Immun 72 : 1725-1732. Juliano, R.L. (2002) Signal transduction by ce ll adhesion receptors a nd the cytoskeleton: functions of integrins, cadherins, selectin s, and immunoglobulin-superfamily members. Annu Rev Pharmacol Toxicol 42 : 283-323. Kadowaki, T. and Yamamoto, K. (2003) Suppression of virulence of Porphyromonas gingivalis by potent inhibitors specific for gingipains. Curr Protein Pept Sci 4 : 451-458. Kadowaki, T., Yoneda, M., Okamoto, K., Maeda, K. and Yamamoto, K. (1994) Purification and characterization of a novel argini ne-specific cysteine proteina se (argingipain) involved in the pathogenesis of periodontal diseas e from the culture supernatant of Porphyromonas gingivalis J Biol Chem 269 : 21371-21378. Kadowaki, T., Nakayama, K., Yoshimura, F., Ok amoto, K., Abe, N. and Yamamoto, K. (1998) Arg-gingipain acts as a major processing en zyme for various cell surface proteins in Porphyromonas gingivalis J Biol Chem 273 : 29072-29076. Kadowaki, T., Nakayama, K., Okamoto, K., Abe, N., Baba, A., Shi, Y. et al (2000) Porphyromonas gingivalis proteinases as virulence de terminants in progression of periodontal diseases. J Biochem (Tokyo) 128 : 153-159. Kamaguchi, A., Baba, H., Hoshi, M. and I nomata, K. (1994) Coaggregation between Porphyromonas gingivalis and Streptococci mutans. Microbiol Immunol 38 : 457-460. Kapila, Y.L., Lancero, H. and Johnson, P.W. (199 8) The response of periodontal ligament cells to fibronectin. J Periodontol 69 : 1008-1019. Katz, J., Sambandam, V., Wu, J. H., Michalek, S.M. and Balkovetz D.F. (2000) Characterization of Porphyromonas gingivalis -induced degradation of epithe lial cell junctional complexes. Infect Immun 68 : 1441-1449. Katz, J., Yang, Q.B., Zhang, P., Potempa, J., Tr avis, J., Michalek, S.M. and Balkovetz, D.F. (2002) Hydrolysis of epithe lial junctional proteins by Porphyromonas gingivalis gingipains. Infect Immun 70 : 2512-2518. Kawanishi, T., Kawanishi, M., Ohata, H., Toyoda K., Takahashi, M., Momose, K. and Hayashi, Y. (1994) The relationship between spontaneous calcium oscillations and cell proliferation in cultured smooth muscle cells. Jpn J Pharmacol 65 : 59-62. Kazmierczak, B.I., Mostov, K. and Engel, J.N. (2 001) Interaction of bact erial pathogens with polarized epithelium. Annu Rev Microbiol 55 : 407-435. Kelly, C.G., Booth, V., Kendal, H., Slaney, J. M., Curtis, M.A. and Lehner, T. (1997) The relationship between colonizati on and haemagglutination inhi biting and B cell epitopes of Porphyromonas gingivalis Clin Exp Immunol 110 : 285-291.

PAGE 196

196 Kemler, R. (1992) Classical cadherins. Semin Cell Biol 3 : 149-155. Kemler, R. (1993) From cadherins to catenins: cy toplasmic protein interac tions and regulation of cell adhesion. Trends Genet 9 : 317-321. Kesavalu, L., Holt, S.C. and Ebersole, J.L. (2003) In vitro envir onmental regulation of Porphyromonas gingivalis growth and virulence. Oral Microbiol Immunol 18 : 226-233. Khlgatian, M., Nassar, H., Chou, H.H., Gibson, F.C., 3rd and Genco, C.A. (2002) Fimbriadependent activation of cell adhe sion molecule expression in Porphyromonas gingivalis infected endothelial cells. Infect Immun 70 : 257-267. Kinane, D.F. (1998) Periodontal dise ases' contributions to cardiovascul ar disease: an overview of potential mechanisms. Ann Periodontol 3 : 142-150. Knowles, J.W., Reddick, R.L., Jennette, J.C., Shes ely, E.G., Smithies, O. and Maeda, N. (2000) Enhanced atherosclerosis and kidney dys function in eNOS(-/-)Apoe(-/-) mice are ameliorated by enalapril treatment. J Clin Invest 105 : 451-458. Kobayashi-Sakamoto, M., Hirose, K., Nishikat a, M., Isogai, E. and Chiba, I. (2006) Osteoprotegerin protects e ndothelial cells against apopt otic cell death induced by Porphyromonas gingivalis cysteine proteinases. FEMS Microbiol Lett 264 : 238-245. Kokeguchi, S., Miyamoto, M., Ohyama, H ., Hongyo, H., Takigawa, M., Kurihara, H. et al (1994) Biochemical properties of the major outer membrane proteins of Porphyromonas gingivalis. Microbios 77 : 247-252. Kolenbrander, P.E. and London, J. (1993) Adhe re today, here tomorrow: oral bacterial adherence. J Bacteriol 175 : 3247-3252. Koulouri, O., Lappin, D.F., Radvar, M. and Kinane D.F. (1999) Cell division, synthetic capacity and apoptosis in periodont al lesions analysed by in situ hybridisation and immunohistochemistry. J Clin Periodontol 26 : 552-559. Kozarov, E.V., Dorn, B.R., Shelburne, C.E., Du nn, W.A., Jr. and Pr ogulske-Fox, A. (2005) Human atherosclerotic plaque contains viable invasive Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis Arterioscler Thromb Vasc Biol 25 : e17-18. Krimbou, L., Tremblay, M., Davignon, J. and C ohn, J.S. (1997) Characterization of human plasma apolipoprotein E-contai ning lipoproteins in the high de nsity lipoprotein size range: focus on pre-beta1-LpE, pre-beta2-LpE, and alpha-LpE. J Lipid Res 38 : 35-48. Kuramitsu, H.K. (1998) Proteases of Porphyromonas gingivalis : what don't they do? Oral Microbiol Immunol 13 : 263-270. Kuramitsu, H.K., Yoneda, M. and Madden, T. (1995) Proteases and collagenases of Porphyromonas gingivalis Adv Dent Res 9 : 37-40.

PAGE 197

197 Kuramitsu, H.K., Kang, I.C. and Qi, M. (2003) Interactions of Porphyromonas gingivalis with host cells: implications for cardiovascular diseases. J Periodontol 74 : 85-89. Kuramitsu, H.K., Qi, M., Kang, I.C. and Chen, W. (2001) Role for peri odontal bacteria in cardiovascular diseases. Ann Periodontol 6 : 41-47. Lalla, E., Lamster, I.B., Hofmann, M.A., Bucciarelli, L., Jerud, A.P., Tucker, S. et al (2003) Oral infection with a periodontal pathoge n accelerates early atherosclerosis in apolipoprotein E-null mice. Arterioscler Thromb Vasc Biol 23 : 1405-1411. Lamont, R.J. and Jenkinson, H.F. (1998) Life be low the gum line: pathogenic mechanisms of Porphyromonas gingivalis Microbiol Mol Biol Rev 62 : 1244-1263. Lamont, R.J. and Jenkinson, H.F. (2000) Subgingival colonization by Porphyromonas gingivalis Oral Microbiol Immunol 15 : 341-349. Lamont, R.J. and Yilmaz, O. (2002) In or out: the invasiveness of oral bacteria. Periodontol 2000 30 : 61-69. Lamont, R.J., Hersey, S.G. and Rosan, B. ( 1992) Characterization of the adherence of Porphyromonas gingivalis to oral streptococci. Oral Microbiol Immunol 7 : 193-197. Lamont, R.J., Hsiao, G.W. and Gil, S. (1994) Identification of a molecule of Porphyromonas gingivalis that binds to Streptococcus gordonii Microb Pathog 17 : 355-360. Lamont, R.J., Chan, A., Belton, C.M., Izutsu, K.T., Vasel, D. and Weinberg, A. (1995) Porphyromonas gingivalis invasion of gingival epithelial cells. Infect Immun 63 : 38783885. Lampugnani, M.G. and Dejana, E. (1997) Interendot helial junctions: stru cture, signalling and functional roles. Curr Opin Cell Biol 9 : 674-682. Larue, L., Ohsugi, M., Hirchenhain, J. and Ke mler, R. (1994) E-cadherin null mutant embryos fail to form a trophectoderm epithelium. Proc Natl Acad Sci U S A 91 : 8263-8267. Leaverton, P.E., Sorlie, P.D., Kleinman, J.C ., Dannenberg, A.L., Ingster-Moore, L., Kannel, W.B. and Cornoni-Huntley, J.C. (1987) Repres entativeness of the Framingham risk model for coronary heart disease mortality: a comparison with a national cohort study. J Chronic Dis 40 : 775-784. Lee, J.Y., Sojar, H.T., Bedi, G.S. and Genco, R.J. (1991) Porphyromonas ( Bacteroides ) gingivalis fimbrillin: size, amino-terminal se quence, and antigenic heterogeneity. Infect Immun 59 : 383-389. Lepine, G., Ellen, R.P. and Progulske-Fox, A. (1996) Construction and preliminary characterization of three hemagglutinin mutants of Porphyromonas gingivalis Infect Immun 64 : 1467-1472.

PAGE 198

198 Levine, D.M., Parker, T.S., Donnelly, T.M., Wals h, A. and Rubin, A.L. (1993) In vivo protection against endotoxin by plasma high density lipoprotein. Proc Natl Acad Sci U S A 90 : 12040-12044. Li, L., Messas, E., Batista, E.L., Jr., Levine, R.A. and Amar, S. (2002) Porphyromonas gingivalis infection accelerat es the progression of atherosclerosis in a heterozygous apolipoprotein E-defi cient murine model. Circulation 105 : 861-867. Liebana, J., Castillo, A.M. and Alvarez, M. (2004) Periodontal di seases: microbiological considerations. Med Oral Patol Oral Cir Bucal 9 Suppl : 82-91; 75-82. Liebner, S., Cavallaro, U. and Dejana, E. (2006) The multiple languages of endothelial cell-tocell communication. Arterioscler Thromb Vasc Biol 26 : 1431-1438. Liebner, S., Cattelino, A., Gallini, R., Rudini, N ., Iurlaro, M., Piccolo, S. and Dejana, E. (2004) Beta-catenin is required for endothelial-mes enchymal transformation during heart cushion development in the mouse. J Cell Biol 166 : 359-367. Lin, T.H., Aplin, A.E., Shen, Y., Chen, Q., Schaller, M., Romer, L. et al (1997) Integrinmediated activation of MAP kinase is indepe ndent of FAK: evidence for dual integrin signaling pathways in fibroblasts. J Cell Biol 136 : 1385-1395. Listgarten, M.A. (1999) Microorg anisms and dental implants. J Periodontol 70 : 220-222. Lodish, H., Berk, A., Zipursky, S., Matsudaria, P., Baltimore, D. and Darnell, J. (2000) Molecular Cell Biology New York, New York: W.H. Freeman and Company. Logan, C.Y. and Nusse, R. (2004) The Wnt sign aling pathway in development and disease. Annu Rev Cell Dev Biol 20 : 781-810. Lorenzon, P., Vecile, E., Nardon, E., Ferrero, E., Harlan, J.M., Tedesco, F. and Dobrina, A. (1998) Endothelial cell Eand P-selectin and vascular cell adhesion molecule-1 function as signaling receptors. J Cell Biol 142 : 1381-1391. Love, R.M., McMillan, M.D., Park, Y. and Je nkinson, H.F. (2000) Coinvasion of dentinal tubules by Porphyromonas gingivalis and Streptococcus gordonii depends upon binding specificity of streptococcal antigen I/II adhesin. Infect Immun 68 : 1359-1365. Lowe, G.D. (1998) Etiopathogene sis of cardiovascular disease: hemostasis, thrombosis, and vascular medicine. Ann Periodontol 3 : 121-126. Lu, B. and McBride, B.C. (1994) Stress response of Porphyromonas gingivalis Oral Microbiol Immunol 9 : 166-173. Lusis, A.J. (2000) Atherosclerosis. Nature 407 : 233-241.

PAGE 199

199 Madianos, P.N., Papapanou, P.N., Nannmark, U., Dahlen, G. and Sandros, J. (1996) Porphyromonas gingivalis FDC381 multiplies and persists within human oral epithelial cells in vitro. Infect Immun 64 : 660-664. Mahley, R.W. (1988) Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science 240 : 622-630. Marrs, J.A., Andersson-Fisone, C., Jeong, M.C ., Cohen-Gould, L., Zurzolo, C., Nabi, I.R. et al (1995) Plasticity in epithe lial cell phenotype: modulation by expression of different cadherin cell adhesion molecules. J Cell Biol 129 : 507-519. Marschang, P., Gotsch, C., Kirchmair, R., Kase r, S., Kahler, C.M. and Patsch, J.R. (2006) Postprandial, but not postabsorptive low-dens ity lipoproteins increas e the expression of intercellular adhesion molecule-1 in human aortic e ndothelial cells. Atherosclerosis 186 : 101-106. Marsh, P.D. (1994) Microbial ecolog y of dental plaque and its signi ficance in health and disease. Adv Dent Res 8 : 263-271. Matter, K. and Balda, M.S. (2003) Si gnalling to and from tight junctions. Nat Rev Mol Cell Biol 4 : 225-236. Matter, M.L. and Ruoslahti, E. (2001) A si gnaling pathway from the alpha5beta1 and alpha(v)beta3 integrins that elevates bcl-2 transcription. J Biol Chem 276 : 27757-27763. Mattila, K.J., Nieminen, M.S., Valtonen, V.V., Rasi, V.P., Kesaniemi, Y.A., Syrjala, S.L. et al (1989) Association between dental hea lth and acute myocardial infarction. Bmj 298 : 779781. Mayrand, D. and Holt, S.C. (1988) Biol ogy of asaccharolytic black-pigmented Bacteroides species. Microbiol Rev 52 : 134-152. McCrea, P.D., Turck, C.W. and Gumbiner, B. (1991) A homolog of the armadillo protein in Drosophila (plakoglobin) asso ciated with E-cadherin. Science 254 : 1359-1361. McGeehan, G.M., Becherer, J.D., Bast, R.C., Jr., Boyer, C.M., Champion, B., Connolly, K.M. et al (1994) Regulation of tumour necrosis fact or-alpha processing by a metalloproteinase inhibitor. Nature. 370 : 558-561. McGill, H.C., Jr. (1968) Fatty streaks in the coronary arteries and aorta. Lab Invest 18 : 560-564. McKee, A.S., McDermid, A.S., Baskerville, A., Dowsett, A.B., Ellwood, D.C. and Marsh, P.D. (1986) Effect of hemin on the physiology and virulence of Bacteroides gingivalis W50. Infect Immun 52 : 349-355.

PAGE 200

200 Mechtersheimer, G., Barth, T., Quentmeier, A. a nd Moller, P. (1994) Differential expression of beta 1, beta 3, and beta 4 integrin subunits in nonneoplastic neural cells of the peripheral and autonomic nervous system and in tumors derived from these cells. Lab Invest 70 : 740752. Meresse, S., Steele-Mortimer, O., Moreno, E., Desj ardins, M., Finlay, B. and Gorvel, J.P. (1999) Controlling the maturation of pathogen-containing vacuoles : a matter of life and death. Nat Cell Biol 1 : E183-188. Meyer, D.H., Sreenivasan, P.K. and Fives-Taylor P.M. (1991) Evidence for invasion of a human oral cell line by Actinobacillus actinomycetemcomitans Infect Immun 59 : 2719-2726. Michel, J.B. (2003) Anoikis in the cardiova scular system: known and unknown extracellular mediators. Arterioscler Thromb Vasc Biol 23 : 2146-2154. Mikolajczyk-Pawlinska, J., Kordula, T., Pavloff, N., Pemberton, P. A., Chen, W.C., Travis, J. and Potempa, J. (1998) Ge netic variation of Porphyromonas gingivalis genes encoding gingipains, cysteine proteinases with arginine or lysine specificity. Biol Chem 379 : 205211. Miyakawa, H., Honma, K., Qi, M. and Kuramitsu, H. (2004) Interaction of Porphyromonas gingivalis with low-density lipoproteins: implicat ions for a role for periodontitis in atherosclerosis. J Periodontal Res 39 : 1-9. Morioka, M., Hinode, D., Nagata, A., Hayashi, H., Ichimiya, S., Ueda, M. et al. (1993) Cytotoxicity of Porphyromonas gingivalis toward cultured human gingival fibroblasts. Oral Microbiol Immunol 8 : 203-207. Morita, K., Furuse, M., Fujimoto, K. and Tsukita S. (1999) Claudin multigene family encoding four-transmembrane domain protein co mponents of tight junction strands. Proc Natl Acad Sci U S A 96 : 511-516. Mueller, S., Cadenas, E. and Schonthal, A.H. (2000) p21WAF1 regulates anchorage-independent growth of HCT116 colon carcinoma cells via E-cadherin expression. Cancer Res 60 : 156163. Muller, W.A. (2003) Leukocyte-endothelial-cell inte ractions in leukocyte transmigration and the inflammatory response. Trends Immunol 24 : 327-334. Muller, W.A., Weigl, S.A., Deng, X. and Phill ips, D.M. (1993) PECAM-1 is required for transendothelial migr ation of leukocytes. J Exp Med 178 : 449-460. Murakami, Y., Hanazawa, S., Watanabe, A., Na ganuma, K., Iwasaka, H., Kawakami, K. and Kitano, S. (1994) Porphyromonas gingivalis fimbriae induce a 68-kilodalton phosphorylated protein in macrophages. Infect Immun 62 : 5242-5246. Murase, S. and Schuman, E.M. (1999) The role of cell adhesion molecules in synaptic plasticity and memory. Curr Opin Cell Biol 11 : 549-553.

PAGE 201

201 Murugesan, G., Sa, G. and Fox, P.L. (1994) High-d ensity lipoprotein stimulates endothelial cell movement by a mechanism distinct from basic fibroblast growth factor. Circ Res 74 : 1149-1156. Nagafuchi, A. and Takeichi, M. (1988) Cell bi nding function of E-cadhe rin is regulated by the cytoplasmic domain. Embo J 7 : 3679-3684. Nagafuchi, A., Takeichi, M. and Tsukita, S. (1991) The 102 kd cadherinassociated protein: similarity to vinculin and posttran scriptional regulation of expression. Cell 65 : 849-857. Nakamura, T., Amano, A., Nakagawa, I. and Hama da, S. (1999) Specific interactions between Porphyromonas gingivalis fimbriae and human extracellular matrix proteins. FEMS Microbiol Lett 175 : 267-272. Nakayama, K., Kadowaki, T., Okamoto, K. and Yamamoto, K. (1995) Construction and characterization of arginine-spe cific cysteine proteinase (Arg -gingipain)-deficient mutants of Porphyromonas gingivalis Evidence for significant contri bution of Arg-gingipain to virulence. J Biol Chem 270 : 23619-23626. Nakayama, K., Ratnayake, D.B., Tsukuba, T., Kadowaki, T., Yamamoto, K. and Fujimura, S. (1998) Haemoglobin receptor protein is intragen ically encoded by the cysteine proteinaseencoding genes and the haema gglutinin-encoding gene of Porphyromonas gingivalis Mol Microbiol 27 : 51-61. Nakhjiri, S.F., Park, Y., Yilmaz, O., Chung, W.O., Watanabe, K., El-Sabaeny, A. et al (2001) Inhibition of epithe lial cell apoptosis by Porphyromonas gingivalis FEMS Microbiol Lett 200 : 145-149. Nassar, H., Chou, H.H., Khlgatian, M., Gibson, F.C., 3rd, Van Dyke, T.E. and Genco, C.A. (2002) Role for fimbriae and lysine-specific cy steine proteinase gingi pain K in expression of interleukin-8 and monocyte chemoattractant protein in Porphyromonas gingivalis infected endothelial cells. Infect Immun 70 : 268-276. Navab, M., Hama, S.Y., Cooke, C.J., Anantharamaiah, G.M., Chaddha, M., Jin, L. et al (2000) Normal high density lipoprotein inhibits thr ee steps in the formation of mildly oxidized low density lipoprotein: step 1. J Lipid Res 41 : 1481-1494. Nelson, K.E., Fleischmann, R.D., DeBoy, R.T ., Paulsen, I.T., Fouts, D.E., Eisen, J.A. et al (2003) Complete genome sequence of the oral pathogenic Bacterium Porphyromonas gingivalis strain W83. J Bacteriol 185 : 5591-5601. Newman, P.J. (1997) Th e biology of PECAM-1. J Clin Invest 100 : S25-29. Newman, P.J. (1999) Switched at bi rth: a new family for PECAM-1. J Clin Invest 103 : 5-9. Nishikata, M. and Yoshimura, F. (1991) Characterization of Porphyromonas ( bacteroides ) gingivalis hemagglutinin as a protease. Biochem Biophys Res Commun 178 : 336-342.

PAGE 202

202 Nishikawa, K., Yoshimura, F. and Duncan, M.J. (2004) A regulation cascade controls expression of Porphyromonas gingivalis fimbriae via the FimR response regulator. Mol Microbiol 54 : 546-560. Njoroge, T., Genco, R.J., Sojar, H.T., Hamada, N. and Genco, C.A. (1997) A role for fimbriae in Porphyromonas gingivalis invasion of oral epithelial cells. Infect Immun 65 : 1980-1984. O'Brien-Simpson, N.M., Paolini, R.A. and Reynolds, E.C. (2000) RgpA-Kgp peptide-based immunogens provide protection against Porphyromonas gingivalis challenge in a murine lesion model. Infect Immun 68 : 4055-4063. O'Connell, B.J. and Genest, J., Jr. (2001) Highdensity lipoproteins and endothelial function. Circulation 104 : 1978-1983. Oelschlaeger, T.A., Guerry, P. and Kopec ko, D.J. (1993) Unusual microtubule-dependent endocytosis mechanisms triggered by Campyl obacter jejuni and Citrobacter freundii. Proc Natl Acad Sci U S A 90 : 6884-6888. Ogawa, T., Mori, H., Yasuda, K. and Ha segawa, M. (1994) Molecular cloning and characterization of the genes encoding the im munoreactive major cell-surface proteins of Porphyromonas gingivalis FEMS Microbiol Lett 120 : 23-30. Ogawa, T., Yasuda, K., Yamada, K., Mori, H., Ochiai, K. and Hasegawa, M. (1995) Immunochemical characterisation and epitope mapping of a novel fimbrial protein (Pg-II fimbria) of Porphyromonas gingivalis FEMS Immunol Med Microbiol 11 : 247-255. Oguey, D., George, P.W. and Ruegg, C. (2000) Di sruption of integrin-d ependent adhesion and survival of endothelial cells by recombinant adenovirus expressing isolated beta integrin cytoplasmic domains. Gene Ther 7 : 1292-1303. Oido-Mori, M., Rezzonico, R., Wang, P.L., Kowa shi, Y., Dayer, J.M., Baehni, P.C. and Chizzolini, C. (2001) Porphyromonas gingivalis gingipain-R enhances interleukin-8 but decreases gamma interferon-inducible protei n 10 production by human gingival fibroblasts in response to T-cell contact. Infect Immun 69 : 4493-4501. Okamoto, K., Kadowaki, T., Nakayama, K. a nd Yamamoto, K. (1996) Cloning and sequencing of the gene encoding a novel lysine-specifi c cysteine proteinase (Lys-gingipain) in Porphyromonas gingivalis : structural relationship with the arginine-specific cysteine proteinase (Arg-gingipain). J Biochem (Tokyo) 120 : 398-406. Okamoto, K., Nakayama, K., Kadowaki, T., Abe, N., Ratnayake, D.B. and Yamamoto, K. (1998) Involvement of a lysine-specific cysteine pr oteinase in hemoglobin adsorption and heme accumulation by Porphyromonas gingivalis J Biol Chem 273 : 21225-21231. Okuda, K., Yamamoto, A., Naito, Y., Takazoe, I., Slots, J. and Genco, R.J. (1986) Purification and properties of hemagglutinin from culture supernatant of Bacteroides gingivalis Infect Immun 54 : 659-665.

PAGE 203

203 Olsen, I. and Dahlen, G. (2004) Salient virulen ce factors in anaerobic b acteria, with emphasis on their importance in e ndodontic infections. Endodontic Topics 9 : 15-26. Ozawa, M., Baribault, H. and Kemler, R. (1989) The cytoplasmic domain of the cell adhesion molecule uvomorulin associates with three i ndependent proteins stru cturally related in different species. Embo J 8 : 1711-1717. Park, Y. and Lamont, R.J. (1998) Cont act-dependent protein secretion in Porphyromonas gingivalis Infect Immun 66 : 4777-4782. Park, Y., Yilmaz, O., Jung, I.Y. and La mont, R.J. (2004) Identification of Porphyromonas gingivalis genes specifically expressed in hum an gingival epithelial cells by using differential display reve rse transcription-PCR. Infect Immun 72 : 3752-3758. Park, Y., Simionato, M.R., Sekiya, K ., Murakami, Y., James, D., Chen, W. et al (2005) Short fimbriae of Porphyromonas gingivalis and their role in coadhesion with Streptococcus gordonii. Infect Immun 73 : 3983-3989. Pavloff, N., Potempa, J., Pike, R.N., Prochazka, V., Kiefer, M.C., Travis, J. and Barr, P.J. (1995) Molecular cloning and structural characteriza tion of the Arg-gingipa in proteinase of Porphyromonas gingivalis Biosynthesis as a protei nase-adhesin polyprotein. J Biol Chem 270 : 1007-1010. Pavloff, N., Pemberton, P.A., Potempa, J ., Chen, W.C., Pike, R.N., Prochazka, V. et al (1997) Molecular cloning and characterization of Porphyromonas gingivalis lysine-specific gingipain. A new member of an emerging fa mily of pathogenic bacterial cysteine proteinases. J Biol Chem 272 : 1595-1600. Pieters, J. (2001) Evasion of host cell de fense mechanisms by pathogenic bacteria. Curr Opin Immunol 13 : 37-44. Pike, R., McGraw, W., Potempa, J. and Travis J. (1994) Lysineand argininespecific proteinases from Porphyromonas gingivalis Isolation, characteri zation, and evidence for the existence of complexes with hemagglutinins. J. Biol. Chem 269 : 406-411. Pike, R.N., Potempa, J., McGraw, W., Coetzer, T.H. and Travis, J. (1996) Characterization of the binding activities of protei nase-adhesin complexes from Porphyromonas gingivalis J Bacteriol 178 : 2876-2882. Pizarro-Cerda, J., Moreno, E., Sa nguedolce, V., Mege, J.L. and Go rvel, J.P. (1998a) Virulent Brucella abortus prevents lysosome fusion and is di stributed within autophagosome-like compartments. Infect Immun 66 : 2387-2392. Pizarro-Cerda, J., Meresse, S., Parton, R.G., van der Goot, G., Sola-Landa, A., Lopez-Goni, I. et al (1998b) Brucella abortus transits through the autophagic pathway and replicates in the endoplasmic reticulum of nonprofessional phagocytes. Infect Immun 66 : 5711-5724.

PAGE 204

204 Plump, A.S., Smith, J.D., Hayek, T., Aalto-S etala, K., Walsh, A., Verstuyft, J.G. et al (1992) Severe hypercholesterolemia and atherosclerosi s in apolipoprotein E-deficient mice created by homologous recombination in ES cells. Cell 71 : 343-353. Polakis, P. (2000) Wnt signaling and cancer. Genes Dev 14 : 1837-1851. Potempa, J., Pike, R. and Travis, J. (1995) The mu ltiple forms of trypsin-like activity present in various strains of Porphyromonas gingivalis are due to the presence of either Arggingipain or Lys-gingipain. Infect Immun 63 : 1176-1182. Potempa, J., Pike, R. and Travis, J. (1997) Titra tion and mapping of the active site of cysteine proteinases from Porphyromonas gingivalis (gingipains) using peptidyl chloromethanes. Biol Chem 378 : 223-230. Potempa, J., Banbula, A. and Travis, J. (2000) Role of bacterial proteinases in matrix destruction and modulation of host responses. Periodontol 2000 24 : 153-192. Potempa, J., Sroka, A., Imamura, T. and Trav is, J. (2003) Gingipains, the major cysteine proteinases and virulence factors of Porphyromonas gingivalis : structure, function and assembly of multidomain protein complexes. Curr Protein Pept Sci 4 : 397-407. Potempa, J., Mikolajczyk-Pawlinska, J., Brassell D., Nelson, D., Thogersen, I.B., Enghild, J.J. and Travis, J. (1998) Comparative properties of two cysteine protei nases (gingipains R), the products of two related but individual genes of Porphyromonas gingivalis J Biol Chem 273 : 21648-21657. Progulske-Fox, A., Rao, V., Han, N ., Lepine, G., Witlock, J. and Lantz, M. (1993) Molecular characterization of hemagglutinin ge nes of periodontopathic bacteria. J Periodontal Res 28 : 473-474. Progulske-Fox, A., Kozarov, E., Dorn, B., Dunn, W., Jr., Burks, J. and Wu, Y. (1999) Porphyromonas gingivalis virulence factors a nd invasion of cells of the cardiovascular system. J Periodontal Res 34 : 393-399. Pussinen, P.J., Jousilahti, P., Alfthan, G., Palo suo, T., Asikainen, S. and Salomaa, V. (2003) Antibodies to periodontal pat hogens are associated with coronary heart disease. Arterioscler Thromb Vasc Biol 23 : 1250-1254. Qi, M., Miyakawa, H. and Kuramitsu, H.K. (2003) Porphyromonas gingivalis induces murine macrophage foam cell formation. Microb Pathog 35: 259-267. Rajapakse, P.S., O'Brien-Simpson, N.M., Slakes ki, N., Hoffmann, B. and Reynolds, E.C. (2002) Immunization with the RgpA-Kgp pr oteinase-adhesin complexes of Porphyromonas gingivalis protects against periodontal bone lo ss in the rat periodontitis model. Infect Immun 70 : 2480-2486.

PAGE 205

205 Rangarajan, M., Aduse-Opoku, J., Slaney, J.M ., Young, K.A. and Curtis, M.A. (1997) The prpR1 and prR2 arginine-s pecific protease genes of Porphyromonas gingivalis W50 produce five biochemically distinct enzymes. Mol Microbiol 23 : 955-965. Ranscht, B. (1994) Cadherins and catenins: interactions and f unctions in embryonic development. Curr Opin Cell Biol 6 : 740-746. Rautemaa, R. and Meri, S. (1996) Protection of gingival epithelium against complementmediated damage by strong expression of the me mbrane attack complex inhibitor protectin (CD59). J Dent Res 75 : 568-574. Redies, C. and Takeichi, M. ( 1996) Cadherins in the developi ng central nervous system: an adhesive code for segmental and functional subdivisions. Dev Biol 180 : 413-423. Reynolds, A.B., Daniel, J., Mc Crea, P.D., Wheelock, M.J., Wu, J. and Zhang, Z. (1994) Identification of a new catenin : the tyrosine kinase substr ate p120cas associates with Ecadherin complexes. Mol Cell Biol 14 : 8333-8342. Richardson, W.P. and Sadoff, J.C. (1988) Induced engulfment of Neisseria gonorrhoeae by tissue culture cells. Infect Immun 56 : 2512-2514. Rosan, B. and Lamont, R.J. ( 2000) Dental plaque formation. Microbes Infect 2 : 1599-1607. Rosenshine, I. and Finlay, B.B. (1993) Salmonella's deceiving signals. J Pediatr Gastroenterol Nutr 17 : 453-454. Rosenshine, I., Duronio, V. and Finlay, B.B. (1992) Tyrosine pr otein kinase inhibitors block invasin-promoted bacterial uptake by epithelial cells. Infect Immun 60 : 2211-2217. Rosenshine, I., Ruschkowski, S., Foubi ster, V. and Finlay, B.B. (1994) Salmonella typhimurium invasion of epithelial cells: role of induced host cell ty rosine protein phosphorylation. Infect Immun 62 : 4969-4974. Ross, R. (1993) The pathogenesis of athe rosclerosis: a pers pective for the 1990s. Nature 362 : 801-809. Ross, R. (1999) Atherosclerosis is an inflammatory disease. Am Heart J 138 : S419-420. Rubanyi, G.M. (1993) The role of endothelium in cardiovascular homeostasis and diseases. J Cardiovasc Pharmacol 22 Suppl 4 : S1-14. Sadot, E., Simcha, I., Shtutman, M., Ben-Ze'ev, A. and Geiger, B. (1998) Inhibition of betacatenin-mediated transactiva tion by cadherin derivatives. Proc Natl Acad Sci U S A 95 : 15339-15344. Sandros, J., Papapanou, P. and Dahlen, G. (1993) Porphyromonas gingivalis invades oral epithelial cells in vitro. J Periodontal Res 28 : 219-226.

PAGE 206

206 Sandros, J., Madianos, P.N. and Papapanou, P.N. (1996) Cellular events concurrent with Porphyromonas gingivalis invasion of oral epithelium in vitro. Eur J Oral Sci 104 : 363371. Sawa, T., Nishimura, F., Ohyama, H., Takahashi, K., Takashiba, S. and Murayama, Y. (1999) In vitro induction of activation-i nduced cell death in lymphocytes from chronic periodontal lesions by exogenous Fas ligand. Infect Immun 67 : 1450-1454. Schenkel, A.R., Mamdouh, Z., Chen, X., Liebman, R.M. and Muller, W.A. (2002) CD99 plays a major role in the migration of monoc ytes through endot helial junctions. Nat Immunol 3 : 143-150. Schoenwaelder, S.M. and Burridge, K. (1999) Bi directional signaling between the cytoskeleton and integrins. Curr Opin Cell Biol 11 : 274-286. Sconyers, J.R., Crawford, J.J. and Moriarty, J.D. (1973) Relationship of bacteremia to toothbrushing in patients with periodontitis. J Am Dent Assoc 87 : 616-622. Scott, C.F., Whitaker, E.J., Hammond, B.F. and Colman, R.W. (1993) Purification and characterization of a potent 70-kDa thiol lysyl-proteinase (Lys-gingivain) from Porphyromonas gingivalis that cleaves kininogens and fibrinogen. J Biol Chem 268 : 7935-7942. Sears, C.L. (2000) Molecular phys iology and pathophysiology of tight junctions V. assault of the tight junction by enteric pathogens. Am J Physiol Gastrointest Liver Physiol 279 : G11291134. Shah, H.N., Gharbia, S.E. and O'Toole, C.M. (1 992) Assessment of the re lative cytotoxicity of Porphyromonas gingivalis cells, products, and components on human epithelial cell lines. J Periodontol 63 : 44-51. Sheets, S.M., Potempa, J., Travis, J., Casiano, C.A. and Fletcher, H.M. (2005) Gingipains from Porphyromonas gingivalis W83 induce cell adhesion molecule cleavage and apoptosis in endothelial cells. Infect Immun 73 : 1543-1552. Shelburne, C.E., Gleason, R.M., Coulter, W.A ., Lantz, M.S. and Lopatin, D.E. (2005) Differential display analysis of Porphyromonas gingivalis gene activation response to heat and oxidative stress. Oral Microbiol Immunol 20 : 233-238. Sheppard, D. (1996) Epithelial integrins. Bioessays 18 : 655-660. Shi, Y., Ratnayake, D.B., Okamoto, K., Abe, N., Yamamoto, K. a nd Nakayama, K. (1999) Genetic analyses of proteolysis, hem oglobin binding, and he magglutination of Porphyromonas gingivalis Construction of mutants with a combination of rgpA, rgpB, kgp, and hagA. J Biol Chem 274 : 17955-17960. Shintani, T. and Klionsky, D.J. (2004) Autophagy in health a nd disease: a double-edged sword. Science 306 : 990-995.

PAGE 207

207 Shoji, M., Ratnayake, D.B., Shi, Y., Kadowaki, T., Yamamoto, K., Yoshimura, F. et al (2002) Construction and characterizati on of a nonpigmented mutant of Porphyromonas gingivalis : cell surface polysaccharide as an anchorage for gingipains. Microbiology 148 : 1183-1191. Silver, J.G., Martin, A.W. and McBride, B.C. (1977) Experimental transient bacteraemias in human subjects with varying degrees of pl aque accumulation and gingival inflammation. J Clin Periodontol 4 : 92-99. Simcha, I., Shtutman, M., Salomon, D., Zhurins ky, J., Sadot, E., Geiger, B. and Ben-Ze'ev, A. (1998) Differential nuclear transl ocation and transactivation po tential of beta-catenin and plakoglobin. J Cell Biol 141 : 1433-1448. Sinai, A.P. and Joiner, K.A. (1997) Safe ha ven: the cell biology of nonfusogenic pathogen vacuoles. Annu Rev Microbiol 51 : 415-462. Slade, G.D., Ghezzi, E.M., Heiss, G., Beck, J.D., Riche, E. and Offenbacher, S. (2003) Relationship between periodontal disease a nd C-reactive protein among adults in the Atherosclerosis Risk in Communities study. Arch Intern Med 163 : 1172-1179. Slakeski, N., Bhogal, P.S., O'Brien-Simpson, N. M. and Reynolds, E.C. (1998) Characterization of a second cell-associated Arg-sp ecific cysteine proteinase of Porphyromonas gingivalis and identification of an adhesin-binding motif i nvolved in association of the prtR and prtK proteinases and adhesins into large complexes. Microbiology 144 ( Pt 6) : 1583-1592. Slots, J. (2000) Primer for an timicrobial periodontal therapy. J Periodontal Res 35 : 108-114. Slots, J. and Gibbons, R.J. (1978) Attachment of Bacteroides melaninogenicus subsp. asaccharolyticus to oral surfaces and its possible role in colonization of the mouth and of periodontal pockets. Infect Immun 19 : 254-264. Slots, J. and Jorgensen, M.G. (2000) Effici ent antimicrobial treatment in periodontal maintenance care. J Am Dent Assoc 131 : 1293-1304. Socransky, S.S. and Haffajee, A.D. (1992) The bacterial etiology of destructive periodontal disease: current concepts. J Periodontol 63 : 322-331. Socransky, S.S., Haffajee, A.D., Smith, C. and Duff, G.W. (2000) Microb iological parameters associated with IL-1 gene polymorphisms in periodontitis patients. J Clin Periodontol 27 : 810-818. Socransky, S.S., Haffajee, A.D., Cugini, M.A., Smith, C. and Kent, R.L., Jr. (1998) Microbial complexes in subgingival plaque. J Clin Periodontol 25 : 134-144. Song, H., Belanger, M., Whitlock, J., Kozarov, E. and Progulske-Fox, A. (2005) Hemagglutinin B is involved in the adherence of Porphyromonas gingivalis to human coronary artery endothelial cells. Infect Immun 73 : 7267-7273.

PAGE 208

208 Springer, T.A. (1995) Traffic signals on endothelium for lymphoc yte recirculation and leukocyte emigration. Annu Rev Physiol 57 : 827-872. Sreenivasan, P.K., Meyer, D.H. and Fives-Tayl or, P.M. (1993) Requirements for invasion of epithelial cells by Actinobacillus actinomycetemcomitans Infect Immun 61 : 1239-1245. St Croix, B., Sheehan, C., Rak, J. W., Florenes, V.A., Slingerland, J.M. and Kerbel, R.S. (1998) E-Cadherin-dependent growth suppression is mediated by the cyclin-dependent kinase inhibitor p27(KIP1). J Cell Biol 142 : 557-571. Stevens, T., Garcia, J.G., Shasby, D.M., Bhatta charya, J. and Malik, A.B. (2000) Mechanisms regulating endothelial cell barrier function. Am J Physiol Lung Cell Mol Physiol 279 : L419-422. Stossel, T.P., Hartwig, J.H., Janmey, P.A. a nd Kwiatkowski, D.J. (1999) Cell crawling two decades after Abercrombie. Biochem Soc Symp 65 : 267-280. Sugatani, J., Miwa, M., Komiyama Y. and Ito, S. (1996) High-den sity lipoprotein inhibits the synthesis of platelet-activating factor in human vascular endothelial cells. J Lipid Mediat Cell Signal 13 : 73-88. Swanson, M.S. and Isberg, R.R. (1995) Association of Legionella pneumophila with the macrophage endoplasmic reticulum. Infect Immun 63 : 3609-3620. Takahashi, Y., Davey, M., Yumoto, H., Gibs on, F.C., 3rd and Genco, C.A. (2006) Fimbriadependent activation of pro-inflammatory molecules in Porphyromonas gingivalis infected human aortic endothelial cells. Cell Microbiol 8 : 738-757. Takeichi, M. (1993) Cadherins in cancer: im plications for invasion and metastasis. Curr Opin Cell Biol 5 : 806-811. Takeichi, M. (1995) Morphogeneti c roles of classic cadherins. Curr Opin Cell Biol 7 : 619-627. Teragawa, H., Fukuda, Y., Matsuda, K., Ueda, K., Higashi, Y., Oshima, T. et al (2004) Relation between C reactive protein concentrations and coronary microvascular endothelial function. Heart 90 : 750-754. Tokuda, M., Duncan, M., Cho, M.I. and Kuramitsu, H.K. (1996) Role of Porphyromonas gingivalis protease activity in colonization of oral surfaces. Infect Immun 64 : 4067-4073. Tokuda, M., Karunakaran, T., Duncan, M., Hamada, N. and Kuramitsu, H. (1998) Role of Arggingipain A in virulence of Porphyromonas gingivalis Infect Immun 66 : 1159-1166. Tonetti, M.S., Cortellini, D. and Lang, N.P. (1998) In situ detection of apoptosis at sites of chronic bacterially induced inflammation in human gingiva. Infect Immun 66 : 5190-5195. Travis, J., Pike, R., Imamura, T. and Potempa, J. (1997) Porphyromonas gingivalis proteinases as virulence factors in the development of periodontitis. J Periodontal Res 32 : 120-125.

PAGE 209

209 Tsuda, K., Amano, A., Umebayashi, K., Inaba, H., Nakagawa, I., Nakanishi, Y. and Yoshimori, T. (2005) Molecular dissecti on of internalization of Porphyromonas gingivalis by cells using fluorescent beads coated with bacterial membrane vesicle. Cell Struct Funct 30 : 8191. Tsukita, S., Furuse, M. and Itoh, M. (2001) Multifunctional strands in tight junctions. Nat Rev Mol Cell Biol 2 : 285-293. Tsutsui, H., Kinouchi, T., Wakano, Y. and Ohnish i, Y. (1987) Purificati on and characterization of a protease from Bacteroides gingivalis 381. Infect Immun 55 : 420-427. Umemoto, T. and Hamada, N. (2003) Characteri zation of biologically active cell surface components of a periodontal pathogen. The roles of major and minor fimbriae of Porphyromonas gingivalis J Periodontol 74 : 119-122. United States Public Health Service. (2000). Periodontal Disease. Retrieved June 15th, 2007, from http://www.hhs.gov./pharmacy/pp/Dhtspresent/. Valtonen, V.V. (1991) Infection as a risk f actor for infarction and atherosclerosis. Ann Med 23 : 539-543. van de Wetering, M., de Lau, W. and Clever s, H. (2002) WNT signa ling and lymphocyte development. Cell 109 Suppl : S13-19. van Meer, G., Gumbiner, B. and Simons, K. (1986) The tight juncti on does not allow lipid molecules to diffuse from one epithelial cell to the next. Nature 322 : 639-641. van Winkelhoff, A.J., Appelmelk, B.J., Kippuw, N. and de Graaff, J. (1993) K-antigens in Porphyromonas gingivalis are associated with virulence. Oral Microbiol Immunol 8 : 259265. Venkiteswaran, K., Xiao, K., Summers, S., Calkin s, C.C., Vincent, P.A., Pumiglia, K. and Kowalczyk, A.P. (2002) Regula tion of endothelial barrier function and growth by VEcadherin, plakoglobin, and beta-catenin. Am J Physiol Cell Physiol 283 : C811-821. Vestweber, D. and Blanks, J.E. (1999) Mechanisms that regulate the function of the selectins and their ligands. Physiol Rev 79 : 181-213. Viswambharan, H., Ming, X.F., Zhu, S., Hubsch, A., Lerch, P., Vergeres, G. et al (2004) Reconstituted high-density lipop rotein inhibits thrombin-induc ed endothelial tissue factor expression through inhibition of RhoA and s timulation of phosphatidylinositol 3-kinase but not Akt/endothelial nitric oxide synthase. Circ Res 94 : 918-925. Wang, C.Y., Guttridge, D.C., Mayo, M.W. and Ba ldwin, A.S., Jr. (1999) NF-kappaB induces expression of the Bcl-2 homologue A1/Bfl-1 to preferentially suppress chemotherapyinduced apoptosis. Mol Cell Biol 19 : 5923-5929.

PAGE 210

210 Wang, J., Dudley, D. and Wang, X.L. (2002) Haplotype-speci fic effects on endothelial NO synthase promoter efficiency: modifiable by cigarette smoking. Arterioscler Thromb Vasc Biol 22 : e1-4. Watanabe, K., Takasawa, T., Yoshimura, F., O zeki, M., Kawanami, M. and Kato, H. (1992) Molecular cloning and expression of a major surface protein (the 75-kDa protein) of Porphyromonas ( Bacteroides ) gingivalis in Escherichia coli FEMS Microbiol Lett 71 : 47-55. Watarai, M. (2004) [Internalizati on and intracellular replication of Brucella in macrophages]. Nippon Saikingaku Zasshi 59 : 465-471. Weibezahn, J., Schlieker, C., Tessarz, P., Mogk, A. and Bukau, B. (2005) Novel insights into the mechanism of chaperone-assi sted protein disaggregation. Biol Chem 386 : 739-744. Weinberg, A., Belton, C.M., Park, Y. and La mont, R.J. (1997) Role of fimbriae in Porphyromonas gingivalis invasion of gingival epithelial cells. Infect Immun 65 : 313-316. Wheelock, M.J. and Johnson, K.R. (2003) Cadherin-mediated cellular signaling. Curr Opin Cell Biol 15 : 509-514. Whittaker, C.J., Klier, C.M. and Kolenbrander, P.E. (1996) Mechanisms of adhesion by oral bacteria. Annu Rev Microbiol 50 : 513-552. Williams, D.L., Dawson, P.A., Newman, T.C. and Rudel, L.L. (1985) Apolipoprotein E synthesis in peripheral ti ssues of nonhuman primates. J Biol Chem 260 : 2444-2451. Wingrove, J.A., DiScipio, R.G., Chen, Z., Potempa, J., Travis, J. and Hugli, T.E. (1992) Activation of complement components C3 and C5 by a cysteine proteinase (gingipain-1) from Porphyromonas ( Bacteroides ) gingivalis J Biol Chem 267 : 18902-18907. Wu, J.H., Billings, B.J. and Balkovetz, D.F. ( 2001) Hepatocyte growth factor alters renal epithelial cell susceptibi lity to uropathogenic Escherichia coli J Am Soc Nephrol 12: 2543-2553. Xie, H., Cai, S. and Lamont, R.J. (1997) Enviro nmental regulation of fimb rial gene expression in Porphyromonas gingivalis Infect Immun 65 : 2265-2271. Xie, H., Kozlova, N. and Lamont, R.J. (2004) Porphyromonas gingivalis genes involved in fimA regulation. Infect Immun 72 : 651-658. Xie, H., Chung, W.O., Park, Y. and La mont, R.J. (2000) Regulation of the Porphyromonas gingivalis fimA (Fimbrillin) gene. Infect Immun 68 : 6574-6579. Yilmaz, O., Watanabe, K. and Lamont, R.J. (2 002) Involvement of integrins in fimbriaemediated binding and invasion by Porphyromonas gingivalis Cell Microbiol 4 : 305-314.

PAGE 211

211 Yilmaz, O., Young, P.A., Lamont, R.J. and Kenny, G.E. (2003) Gingi val epithe lial cell signalling and cytoskeletal responses to Porphyromonas gingivalis invasion. Microbiology 149 : 2417-2426. Yoshimura, F., Takahashi, K., Nodasaka, Y. and Suzuki, T. (1984) Purification and characterization of a novel type of fimbriae from the oral anaerobe Bacteroides gingivalis J Bacteriol 160 : 949-957. Young, V.B., Falkow, S. and Schoolnik, G. K. (1992) The invasin protein of Yersinia enterocolitica : internalization of invasi n-bearing bacteria by eukar yotic cells is associated with reorganization of the cytoskeleton. J Cell Biol 116 : 197-207. Yui, Y., Aoyama, T., Morishita, H., Takahashi, M., Takatsu, Y. and Kawai, C. (1988) Serum prostacyclin stabiliz ing factor is identical to ap olipoprotein A-I (Apo A-I). A novel function of Apo A-I. J Clin Invest 82 : 803-807. Yumoto, H., Chou, H.H., Takahashi, Y., Davey, M., Gibson, F.C., 3rd and Genco, C.A. (2005) Sensitization of human aortic endothelial cells to lipopolys accharide via regulation of Tolllike receptor 4 by bacterial fimbria-dependent invasion. Infect Immun 73 : 8050-8059. Yun, P.L., Decarlo, A.A., Chapple, C.C. and H unter, N. (2005) Functional implication of the hydrolysis of platelet endothe lial cell adhesion molecule 1 (CD31) by gingipains of Porphyromonas gingivalis for the pathology of periodontal disease. Infect Immun 73 : 1386-1398. Zannis, V.I., Cole, F.S., Jackson, C.L., Kurnit, D.M. and Karathanasis, S.K. (1985) Distribution of apolipoprotein A-I, C-II, C-III, and E mR NA in fetal human tissues. Time-dependent induction of apolipoprotein E mRNA by cultures of hu man monocyte-macrophages. Biochemistry 24 : 4450-4455. Zeiher, A.M. and Schachinger, V. (1994) Corona ry endothelial vasodilato r dysfunction: clinical relevance and therapeutic implications. Z Kardiol 83 Suppl 4 : 7-14. Zhang, S.H., Reddick, R.L., Piedrahita, J.A. and Maeda, N. (1992) Spontaneous hypercholesterolemia and arterial lesions in mice lacking apolipoprotein E. Science 258 : 468-471. Zhang, Y., Zanotti, I., Reilly, M.P., Glick, J.M. Rothblat, G.H. and Rader, D.J. (2003) Overexpression of apolipoprotei n A-I promotes reverse transport of cholesterol from macrophages to feces in vivo. Circulation. 108 : 661-663. Zhang, Y., Wang, T., Chen, W., Yilmaz, O., Park, Y., Jung, I.Y. et al (2005) Differential protein expression by Porphyromonas gingivalis in response to secreted epithelial cell components. Proteomics 5 : 198-211.

PAGE 212

212 BIOGRAPHICAL SKETCH Kristen L Totten was born in Riverside, California on July 17th 1974. Kristen has had the opportunity of living in many interesting places throughout her life. When she was very young, her family moved to Hong Kong. At the age of th ree her family moved back to the United States and settled for three years in Columbia, South Ca rolina, before taking up residence in Marietta, Georgia. Kristen graduated from George Wa lton High School in June of 1992. That following August, she enrolled at the University of Alabam a in Tuscaloosa, Alabama. Kristen graduated four years later in May of 1996 with a BS in Biol ogy. She decided to take a few years off before entering graduate school. During those years she held several jobs in the hospitality industry before settling down in Gainesville, Florida with a technician position in the pharmacology laboratory of Dr. David Silverman at the University of Florida. Kristen was a technician in Dr. Silvermans lab for two years be fore she entered the Interdisciplinary Program (IDP) at the University of Florida in August 2000. Kristen deci ded to pursue a project and a joint mentorship with Dr. Ann Progulske-Fox in Oral Biology an d Dr. William Dunn Jr. in Cell Biology. Upon graduation, she plans on pursui ng a career with the government.