<%BANNER%>

Interactions of Oral Bacteria with Gingival Epithelial Cells

Permanent Link: http://ufdc.ufl.edu/UFE0041817/00001

Material Information

Title: Interactions of Oral Bacteria with Gingival Epithelial Cells
Physical Description: 1 online resource (98 p.)
Language: english
Creator: Dickinson, Brittany
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: apoptosis, biology, confocal, cytokine, disease, gingivitis, infection, invasion, oral, periodontal
Medicine -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The present study investigated the responses of multilayers of primary gingival epithelial cells (GECs) and a common immortalized cell line extensively used in lieu of the primary cells (human immortalized gingival keratinocytes- HIGKs) to infection with oral bacteria associated with health and periodontal disease. GECs were cultured in low calcium conditions (0.06 mM) on permeable membranes and grown in an air-liquid interface into three layers. HIGKs were cultured under two conditions, medium containing .15 mM calcium and 1.2 mM calcium to represent undifferentiated and differentiated cells. They too were cultured on permeable membranes in an air-liquid interface into three layers. The multilayer models were infected with fluorescently labeled Aggregatibacter actinomycetemcomitans (A.a.), Porphyromonas gingivalis (P.g.), Fusobacterium nucleatum (F.n.), or Streptococcus gordonii (S.g.). The epithelial cells were stained with TRITC-phalloidin and bacterial association was determined by confocal microscopy and quantitative image analysis. Barrier function of the epithelial cells was measured by transmembrane epithelial resistance (TER), and induction of apoptosis was determined by Annexin V staining. Culture supernatants were collected and examined for cytokine expression by Luminex. Multilayers exhibited TERs of 175 Ohms x cm2 and TER was not significantly altered by bacterial infection. The undifferentiated HIGK multilayers exhibited an average TER of 140 Ohms x cm2 and TER was disrupted by A.a., F.n., and P.g. The differentiated HIGKs exhibited an average TER of 120 Ohms x cm2 and TER was disrupted after bacterial challenge with A.a., F.n., and P.g. S.g. remained extracellular and didn t exhibit significant movement through the cell layers. P.g. invaded intracellularly (78% of total associated organisms), and showed intercellular movement with 30% of total associated bacteria reaching the middle layer by 24 hours. A.a. remained extracellular but 66% penetrated the second and third cell layers. F.n. effectively penetrated the multilayers with nearly 40% reaching the bottom layer by 24 hours but caused cellular destruction. Apoptosis was induced by S.g., A.a., and F.n. in GECs, undifferentiated HIGKs, and differentiated HIGKs. S.g. stimulated IL-1beta, IL-6, IL-8 and TNF-alpha in GECs. F.n. stimulated IL-1beta and TNF-alpha in GECs and IL-6 and IL-8 in HIGKs. A.a. stimulated IL-1beta, IL-6, IL-8, and TNF-alpha in GECs as well as IL-6 and IL-8 in HIGKs. P.g. stimulated the secretion of IL-10 and IL-12(p40) in HIGKs. In GEC and HIGK multilayer models P.g. is intracellularly invasive but does not induce host cell death or inflammatory cytokine production, consistent with its properties as a stealth pathogen. A.a., S.g., and F.n. induced cytokines and apoptosis although S.g. does not penetrate through the multilayers. F.n. and A.a. are both efficient at permeating the multilayers, though neither one invaded intracellularly in the multilayer models.
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 Brittany Dickinson.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Lamont, Richard J.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041817:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041817/00001

Material Information

Title: Interactions of Oral Bacteria with Gingival Epithelial Cells
Physical Description: 1 online resource (98 p.)
Language: english
Creator: Dickinson, Brittany
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: apoptosis, biology, confocal, cytokine, disease, gingivitis, infection, invasion, oral, periodontal
Medicine -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The present study investigated the responses of multilayers of primary gingival epithelial cells (GECs) and a common immortalized cell line extensively used in lieu of the primary cells (human immortalized gingival keratinocytes- HIGKs) to infection with oral bacteria associated with health and periodontal disease. GECs were cultured in low calcium conditions (0.06 mM) on permeable membranes and grown in an air-liquid interface into three layers. HIGKs were cultured under two conditions, medium containing .15 mM calcium and 1.2 mM calcium to represent undifferentiated and differentiated cells. They too were cultured on permeable membranes in an air-liquid interface into three layers. The multilayer models were infected with fluorescently labeled Aggregatibacter actinomycetemcomitans (A.a.), Porphyromonas gingivalis (P.g.), Fusobacterium nucleatum (F.n.), or Streptococcus gordonii (S.g.). The epithelial cells were stained with TRITC-phalloidin and bacterial association was determined by confocal microscopy and quantitative image analysis. Barrier function of the epithelial cells was measured by transmembrane epithelial resistance (TER), and induction of apoptosis was determined by Annexin V staining. Culture supernatants were collected and examined for cytokine expression by Luminex. Multilayers exhibited TERs of 175 Ohms x cm2 and TER was not significantly altered by bacterial infection. The undifferentiated HIGK multilayers exhibited an average TER of 140 Ohms x cm2 and TER was disrupted by A.a., F.n., and P.g. The differentiated HIGKs exhibited an average TER of 120 Ohms x cm2 and TER was disrupted after bacterial challenge with A.a., F.n., and P.g. S.g. remained extracellular and didn t exhibit significant movement through the cell layers. P.g. invaded intracellularly (78% of total associated organisms), and showed intercellular movement with 30% of total associated bacteria reaching the middle layer by 24 hours. A.a. remained extracellular but 66% penetrated the second and third cell layers. F.n. effectively penetrated the multilayers with nearly 40% reaching the bottom layer by 24 hours but caused cellular destruction. Apoptosis was induced by S.g., A.a., and F.n. in GECs, undifferentiated HIGKs, and differentiated HIGKs. S.g. stimulated IL-1beta, IL-6, IL-8 and TNF-alpha in GECs. F.n. stimulated IL-1beta and TNF-alpha in GECs and IL-6 and IL-8 in HIGKs. A.a. stimulated IL-1beta, IL-6, IL-8, and TNF-alpha in GECs as well as IL-6 and IL-8 in HIGKs. P.g. stimulated the secretion of IL-10 and IL-12(p40) in HIGKs. In GEC and HIGK multilayer models P.g. is intracellularly invasive but does not induce host cell death or inflammatory cytokine production, consistent with its properties as a stealth pathogen. A.a., S.g., and F.n. induced cytokines and apoptosis although S.g. does not penetrate through the multilayers. F.n. and A.a. are both efficient at permeating the multilayers, though neither one invaded intracellularly in the multilayer models.
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 Brittany Dickinson.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Lamont, Richard J.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041817:00001


This item has the following downloads:


Full Text

PAGE 1

INTERACTIONS OF ORAL BACTERIA WITH GINGIVAL EPITHELIAL CELLS By BRITTANY DICKINSON A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORID A IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2010 1

PAGE 2

2010 Brittany Dickinson 2

PAGE 3

To my family who kept me going with their words of encouragement and friends who kept me going with a large glass of wine 3

PAGE 4

ACKNOWLEDGMENTS I would like to thank my mentor, Dr. Ric hard Lamont, for his continuous guidan ce and reassurance throughout my past two years of work. I also owe gratitude to my committee members Dr. Thomas Brown and Dr. Shannon Wallet, for their suggestions and advice through the entire rese arch and writing process. Lastly, I would like to express my appreciation to the members of the Lamont Lab for their continued support. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDG MENTS..................................................................................................4LIST OF FI GURES..........................................................................................................7ABSTRACT ...................................................................................................................10 CHAPTER 1 BACKGRO UND...................................................................................................... 12Periodontal Disease ................................................................................................12Gingival Epit helium.................................................................................................14In Vitro Models of Epithelial Cells ...........................................................................15Gingival Epithel ial Cell s....................................................................................15Three-Dimensiona l Model s...............................................................................16Cytokera tins .....................................................................................................16Transepithelial Electr ical Resi stance................................................................17Bacterial Invasion in Monolay ers............................................................................17Epithelial Cell Respon ses to Ba cteria .....................................................................18Apoptosis in Ep ithelial Cells ....................................................................................19Streptococcus gordon ii (S. gor donii) .......................................................................20Porphyromonas gingivalis (P. gingiv alis)................................................................21Adhesion, Invasion, and Migrat ion...................................................................22Migratio n...........................................................................................................23Post-Invasion Activiti es.....................................................................................23Pathogenic Pr operties ......................................................................................24Aggregatibacter actinomycetemcomitans (A. actinomycete mcomitans).................24Fusobacterium nucleatum (F. nucl eatum)...............................................................262 MATERIALS A ND METHOD S................................................................................27Primary Cell Culture................................................................................................27Immortalized Cell Culture........................................................................................28Three-Dimensional ModelMu ltilayer Me mbranes.................................................28GEC Multil ayers...............................................................................................28HIGK Mult ilayers..............................................................................................29Air/Liquid In terface...........................................................................................30Bacterial Cu lture.....................................................................................................30Experiment s............................................................................................................31Keratin Expr ession ...........................................................................................31Transepithelial Electr ical Resi stance................................................................31Bacterial C hallenge ..........................................................................................31Actin Staining...................................................................................................33 5

PAGE 6

Annexin V St aining...........................................................................................35Luminex A nalysis .............................................................................................363 RESULT S...............................................................................................................39Keratin Expr ession..................................................................................................39Invasion as say........................................................................................................40GEC Multil ayers...............................................................................................41HIGK Mult ilayers..............................................................................................42Transepithelial Electr ical Resi stance......................................................................43Cytokine Secr etion..................................................................................................44Apoptos is................................................................................................................454 DISCU SSION.........................................................................................................83Characterization of the 3-D M odels ........................................................................83Interactions of P. gingivalis, S. gordonii F. nucleatum and A. actinomycetemcomitans with the 3-D Models ......................................................85LIST OF RE FERENCES...............................................................................................92BIOGRAPHICAL SKETCH ............................................................................................98 6

PAGE 7

LIST OF FIGURES Figure page 3-1 Image of HIGK multilayer tested for expression of cytokeratin 13 (A.), cytokeratin 14 (B.), and cyt okeratin 19 (C.)........................................................46 3-2 Results of te sting for keratin expressi on of GECs, undifferentiated HIGKs and differentia ted HIGK s.........................................................................473-3 Example of the pr ocess for bacteri al count ing...................................................483-4 Confocal image 3-D co mpilation of HIGK multila yers........................................493-5 Confocal image 3-D compilation of GE C multilayers (respresentative of 3 cell layers) after bacterial challenge with Aggregatibacter actinomycetemcomitans .....................................................................................503-6 Confocal image 3-D compilation of GEC multilayers (representative of 3 cell layers) after bacterial challenge with Fusobacterium nucleatum ..................513-7 Confocal image 3-D compilation of GEC multilayers (representative of 3 cell layers) after bacterial challenge with Streptococcus gordonii .......................52 3-8 Confocal image 3-D compilation of GEC multilayers (representative of 3 cell layers) after bacterial challenge with Porphyromonas gingivalis ..................53 3-9 Percentage of the total bacte ria present in each of the three (top, middle, or botto m) layers....................................................................................543-10 Percentage of the total bacte ria present in each of the three (top, middle, or botto m) layers....................................................................................553-11 Percentage of the total bacte ria present in each of the three (top, middle, or botto m) layers....................................................................................563-12 Percentage of the total bacte ria present in each of the three (top, middle, or botto m) layers....................................................................................573-13 Percentage of bacteria present after 2 hours of challenge in each (top, middle, or bottom) layer of HIGK mu litilaye rs.....................................................583-14 Percentage of bacteria pres ent after 24 hours of challenge in each (top, middle, or bottom) layer of HIGK mulitilayers cult ured...............................593-15 Confocal image (3-D compilati on) of HIGK multilayers (representative of 3 cell layers) after bacterial challenge with P. gingivalis for 2 hours..............60 7

PAGE 8

3-16 Confocal image (3-D compilati on) of HIGK multilayers (representative of 3 cell layers) after bacterial challenge with A. actinomycetemcomitans .........613-17 Confocal image (3-D compilati on) of HIGK multilayers (representative of 3 cell layers) after bacterial challenge with F. nucleatum for 24 hours...........623-18 Confocal image (3-D compilati on) of HIGK multilayers (respresentative of 3 cell layers) after bacterial challenge with S. gordonii for 24 hours...............633-19 Graphical represent ation of the efficiency of P. gingivalis at invading (internalizing) HIGK multilaye rs..........................................................................643-20 Slice view of confocal im age of HIGK multilayer after bacterial challenge with P. gingivalis for 24 hours.............................................................653-21 Measurement of transepithelial electr ical resistance of GEC multilayers after bacte rial cha llenge...................................................................663-22 Measurement of transepithelial electr ical resistance of HIGK multilayers grown in .15 mM calcium media after bacteri al challenge................673-23 Measurement of transepithelial electr ical resistance of HIGK multilayers cultured in 1.2 mM calc ium after bacterial challenge........................683-24 Secretion of IL-1 by GEC mu ltilaye rs................................................................693-25 Secretion of IL -6 by GEC mult ilayers..................................................................703-26 Secretion of IL -8 by GEC mult ilayers..................................................................713-27 Secretion of IL -10 by GEC mu ltilayers ................................................................723-28 Secretion of IL12(p40) by GEC mu ltilayers ........................................................733-29 Secretion of MCP-1 by GEC multilayers.............................................................743-30 Secretion of TNFby GEC mult ilayers ..............................................................753-31 Secretion of IL -8 by HIGK multilayers cultured in media containing .15 mM calciu m........................................................................................................763-32 Secretion of IL -6 by HIGK multilayers cultured in media containing .15 mM calciu m........................................................................................................773-33 Secretion of IL -10 by HIGK multilayers cultured in media containing .15 mM ca lcium..................................................................................................783-34 Secretion of IL-12(p40) by HIGK multilayers cultured in media containing .15 mM calciu m.................................................................................79 8

PAGE 9

3-35 Detection of the induction of apoptosis in HIGKs (cultured in .15 mM calcium m edia)...................................................................................................803-36 Detection of the induction of apoptosis in HIGKs (cultured in 1.2 mM calcium m edia)...................................................................................................81 3-37 Detection of the induction of apoptosis in GECs.................................................82 9

PAGE 10

Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for t he Degree of Master of Science INTERACTIONS OF ORAL BACTERIA WITH GINGIVAL EPITHELIAL CELLS By Brittany Dickinson May 2010 Chair: Richard Lamont Major: Medi cal Sciences The present study investigated the responses of multilayers of primary gingival epithelial cells (GECs) and a common immortaliz ed cell line extensively used in lieu of the primary cells (human immortalized gingiva l keratinocytesHIGKs) to infection with oral bacteria associated with health and periodont al disease. GECs were cultured in low calcium conditions (0.06 mM) on permeabl e membranes and grown in an air-liquid interface into three layers. HIGKs were cultured under two conditions, medium containing .15 mM calcium and 1.2 mM calc ium to represent undifferentiated and differentiated cells. They too were cultur ed on permeable membranes in an air-liquid interface into three layers. The multilaye r models were infected with fluorescently labeled Aggregatibacter actinomycetemcomitans (A.a .), Porphyromonas gingivalis ( P.g .), Fusobacterium nucleatum (F.n .), or Streptococcus gordonii (S.g .). The epithelial cells were stained with TRITC-phalloidin an d bacterial association was determined by confocal microscopy and quantitative image analys is. Barrier function of the epithelial cells was measured by transmembrane epit helial resistance (TER), and induction of apoptosis was determined by Annexin V staini ng. Culture supernatants were collected and examined for cytokine expression by Luminex. Multilayers exhibited TERs of 175 10

PAGE 11

11 x cm2 and TER was not significantly altered by bacterial infection. The undifferentiated HIGK multilayers exhibited an average TER of 140 x cm2 and TER was disrupted by A.a. F.n., and P.g. The differentiated HIGKs exhi bited an average TER of 120 x cm2 and TER was disrupted after bacterial challenge with A.a ., F.n ., and P.g S.g remained extracellular and didnt exhibit signific ant movement through the cell layers. P.g. invaded intracellularly (78% of total associated organisms), and showed intercellular movement with 30% of total associated bacteri a reaching the middle layer by 24 hours. A.a. remained extracellular but 66% penetra ted the second and third cell layers. F.n. effectively penetrated the multilayers with nearly 40% reaching the bottom layer by 24 hours but caused cellular destruction. Apoptosis was induced by S.g. A.a. and F.n. in GECs, undifferentiated HIGKs, and differentiated HIGKs. S.g. stimulated IL-1 IL-8, IL8 and TNFin GECs. F.n. stimulated IL-1 and TNFin GECs and IL-6 and IL-8 in HIGKs. A.a. stimulated IL-1 IL-6, IL-8, and TNFin GECs as well as IL-6 and IL-8 in HIGKs. P.g. stimulated the secretion of IL-10 and IL-12(p40) in HIGKs. In GEC and HIGK multilayer models P.g. is intracellularly invasive but does not induce host cell death or inflammatory cytokine production, c onsistent with its proper ties as a stealth pathogen. A.a. S.g. and F.n. induced cytokines and apoptosis although S.g. does not penetrate through the multilayers. F.n. and A.a. are both efficient at permeating the multilayers, though neither one invaded intracellularly in the multilayer models.

PAGE 12

CHAPTER 1 BACKGROUND Periodontal Disease The human oral microb iota is a complex ecosystem characterized by the presence of a large number of normal colonizers associated with health, living alongside and thriving with opportunistic and pathogenic species [1]. The complex etiology of oral infectious diseases involves groups of bacteria thriving in biofilms and working in concert with immunological susceptibilities in the host [1]. The presence of over 700 different species, or phylotypes, of bacteri a has been noted in the human oral cavity [23]. In health, it is normal for 20-30 different species to be present at a specific site [3]. The healthy state of the peridontium is a careful balance and disease episodes may ensue from a shift in this balance between bact erial and host factors [4]. At diseased sites there are higher than no rmal numbers of bacteria pres ent, involving upwards of 200 species of bacteria. Microbial colonization patterns result in biofilm formation on all surfaces in the oral cavity. On the hard tissue surfaces of the dentition, the initial colonizers of the plaque biofilm are principally or al streptococci and actinomyces soon followed by gramnegative bacteria such as Fusobacterium nucleatum, and later by gram-negative anaerobes such as Porphyromonas gingivalis. On ce full colonization of the subgingival area has occurred, bacteria may shed from the plaque biof ilm on the tooth and interact with host epithelial cells along with the pl anktonic bacteria already present in the gingival crevicular fluid [2]. Periodontal diseases (ranging from mild to severe) can affect up to 90% of the worlds population [5]. Periodont al diseases, including gingivitis and periodontitis, are a 12

PAGE 13

group of infections that lead to inflammation of the gingiva, destruction of the periodontium, and, if left untreat ed, loss of alveolar bone with eventual exfoliation of the teeth [1, 6-7]. Periodontitis has been defined as inflammation and either aggressive (rapid) or chronic (slower) destruction of the tissues supporting the tooth. Periodontal diseases are serious infections and have been indicated as a main cause of tooth loss in adults [8]. Micro-organisms implicated in periodontit is include Aggregatibacter (localized aggressive periodontitis), Porphyromonas, Treponema, Tannerella, Fusobacterium, and Prevotella. Severe forms of periodontitis are associated with gramnegative bacteria [4]. Many organisms can adhere to epithelial ce lls but only a small subset including Porphyromonas gingivalis, Aggregatibacte r actinomycetemcomitans and Fusobacterium nucleatum can invade intracellularly [9-11]. Commensals and opportunistic commensals have developed a balanced evolutionary relationship with t he host [12]. Commensal microbiota is an integral part of a complex homeostasis mechanism that impedes the activity of pathogenic microorganisms [2]. Human immortalized gingi val keratinocytes (HIGKs) in monolayers have exhibited a general hyporesponsiv eness to commensals which may be advantageous in order to lim it tissue destruction that might occur if a strong proinflammatory response were induced [13]. Periodontal disease has most recently come under the mi croscope not singly due to its localized effects, but also because of its numerous systemic implications. Higher levels of periodontal bacteria have been imp licated in cases of cardiovascular disease and increased incidence of low birthweight pr eterm births [4, 14]. Periodontal disease is 13

PAGE 14

also now considered useful for study as an infection model for cardiovascular disease because active bacterial infections with in flammatory consequences are often present for years and are easily accessible in the subgingiva [14]. Gingival Epithelium Colonization of the oral cavity involves unique biofilms on both hard tissues of the tooth and soft tissues of the epi thelial cells such as the gi ngivae [4]. Oral surfaces are generally coated with a pellicle composed predominantly of salivary molecules but also including serum-derived molecules from gi ngival crevicular fluid (GCF) along with products related to host nutrition and epithelial cell turnover [4]. The tissues surrounding the dentition and lyin g over the alveolar bone are referred to as the gingiva. They are tightly bound to the bone and create a s eal around the tooth structure. The gingival epithel ium is classified into the gingival oral epithelium, the sulcular epithelium, and the junctional epithelium [15]. T he subgingival region between the tooth root and the su lcular epithelium is known as the gingival crevice or sulcus. In periodontitis, the gingival sulcus deepens to the periodontal pocket. The environment of the periodontal pocket is less oxygenated which will favor the growth of the anaerobes that can cause periodontal tissue destruc tion and bone loss [16]. The epithelial layer that lines the subgingival crevice forms the initial interface between the host and microbes in the progression of periodontal disease [2]. Epithelial cells have long been considered a mechanical barrier to infection, protecting the periodontal tissue from toxic or microbiological influences. They are now increasingly reported as sensors that may signal a microbial intrusion to the immune cells by generating and transmitting signa ls between bacteria and the adjacent and underlying immune cells in the per iodontal tissues [1, 4, 17]. 14

PAGE 15

The junctional epithelium fo rms the floor of the ging ival sulcus and extends apically in apposition to the su rface of the enamel to form a seal between the epithelium and the tooth [15]. The cells of the junctional epithelium provide an early barrier to tissue penetration by peri odontal pathogens [4]. They are non-keratinized and are characterized by ovally elongated nuclei, prominent Golgi complexes, extended cisternae of the rough endoplasmic reticulum and relative ly low cytoplasmic filament bundles [15]. The junctional epithelium pres ents with wide intracellular spaces, due in part to the low number and distribution of desmosomes, which interconnect cells mechanically. The oral epithelium has 66 desmosomes per 100 m2 in comparison to the 14 in junctional epithelium. This evidence supports the theory that the junctional epithelium is highly permeable [15]. In Vitro Models of Epithelial Cells Gingival Epithelial Cells Primary gingival epithelial cells isolated fr om gingival tissue obt ained during either third molar extraction or crow n lengthening procedures [6]. They are difficult to obtain, and can only be passaged up to 10 times, which makes them a difficult cell line to use for experimentation. An immortalized gi ngival epithelial cell line (HIGKHuman Immortalized Gingival Keratinocytes) is an established line obtai ned from Dr. Oda, University of Washington, that has long been used as a model for periodontal disease studies. The Human papillomavirus (HPV) was utilized to transform the cells, using viral proteins E6 and E7 from HPV [17]. HPV E6/7 play an important role in the increased cell proliferation by altering the cell-cycle re gulatory factors. E6 can induce ubiquitinmediated degradation of p53 tu mor suppressor protein while E7 binds and inactivates both the Rb suppressor gene and other cellular proteins associated with cell cycle 15

PAGE 16

regulation [18]. Expression of E6 and E7 stops the effects of transforming growth factor-mediated growth arrest and eradicates the halt of G1-phase of the cell cycle by p53. E7 also disconnects the prolifer ation from the progression of differentiation in epithelial cells by mediating DNA synthesis independent ly [17]. These immortalized cells have been used in place of the primar y gingival epithelilal cell s (GECs) because of their dependability and ease of culturing. It is not presently known, however, how comparable the widely used immortalized cell s are to the primary cells. Three-Dimensional Models Multilayer models have been developed for us e in other studies. Two gingival epithelial cell lines originatin g from transgenic mice were established and when grown in multilayers, showed a phenotype characteristic of nonkeratinized sulcular epithelium [19]. Other epithelial cell lines (such as immortalized human cor neal epithelial cells) have been developed and show 3-4 layers of gr owth after 2-3 weeks in an air-liquid interface on similar inserts to th e ones used for this study [20]. Cytokeratins Cytokeratins (keratins) are filament-formi ng proteins of epithel ial cells that are highly differentiation-specific in their ex pression patterns. The 54 human keratins and their genes are divided into three categories: epithelial, hair, and keratin pseudogenes. The epithelial keratins and their genes are numbered K1-28 [21]. Cli nical specimens of the junctional epithelium express cytokeratins (CK) 10, 13, 16, and 19; providing evidence that it is a unique, non-differentiat ed stratified epithelium. CK19 has classically been used as a histological differentiation ma rker for the junctional epithelium and CKs 10, 13 and 16 can be used as cellular markers to distinguish junctio nal epithelium from oral epithelium and sulcular epithelium [15, 17, 22]. In addition, toot h-facing cells of the 16

PAGE 17

junctional epithelium uniquely express cytok eratin-14 [15], while both CK5 and CK14, which form in vivo complexes, are expressed in the basal cells of stratified epithelium [23]. With this knowledge in mind, all cult ured gingival epithelia l cells (GECs) and human immortalized gingival keratinocytes (HIGK s) were characterized by their keratin expression. Transepithelial Electrical Resistance As an early barrier to tissue penetration by bacteria, an import ant feature of the gingiva is the ability to form transepithelial electrical resistance, an indicator for the generation of cell to cell contacts such as ti ght junctions and strength of epithelial barrier function [17]. The junctional epithelium c ontains only a few desmosomes, composed only of desmoglein 3 [24]. The anchoring j unctions connecting junctional epithelium cells are lax, causing widened intercellular spaces [24]. Bacterial Invasion in Monolayers Monolayers of gingival epit helial cells (GECs) have been used as a model for the study of bacterial-host cell interactions. Monolayers of GECs have been shown to be susceptible to rapid intracellular invasion by Porphyromonas gingivalis allowing the bacteria to internalize within 12 minutes of infection then localize in the perinuclear region and remain viable for ex tended periods of time [25]. Aggregatibacter actinomycetemcomitans has also displayed the ability to invade GEC monolayers in small amounts and remain vi able within the cell [26]. Fusobacterium nucleatum has exhibited a capacity to invade intracellula rly in human immortalized gingival epithelial cells, while Streptococcus gordonii is known as an essentially extracellular bacterium [13]. 17

PAGE 18

Epithelial Cell Responses to Bacteria An important aspect of periodontal health is the gingivas defense mechanisms, particularly at the dento-epithel ial level [27]. Epithelial ce lls can produce oxidants and antimicrobial peptides to actively participate in fending off intruding microbes [1]. Host cells distinguish the infecting bacteria and tail or a response while the bacteria attempt to manipulate host cell responses [2].Several fa milies of natural antibiotic peptides or proteins are expressed in the oral epithelium ( -defensins, calprotectin, and adrenomedullin) [28]. -defensins are proposed to play a role in the maintenance of steady state levels of micro-organisms in the oral cavity where they have shown strainspecific variability and are induc ed in the gingival epithelium by only a subset of bacteria and TLR ligands [29]. For instance, the -defensin hBD-3 is produced in the junctional epithelium and is part of the protective barrier function of the epithelium. It is antibacterial, antifungal, and antiviral [28]. Ad renomedullin is produc ed in the epithelium and is antibacterial, mitogenic, vasodilatory, and is an inducible protein [28]. The cells of the junctional epithelium also release cytokines (signaling molecules) as a means of communicating with each other and immune cells needed for bacterial clearance [30]. Junctional epitheliu m is a potent source of TNF, IL-1 and IL-1 suggesting that it plays an important role in the first line of defense [15]. These proinflammatory cytokines can produce fever, induce inflammation, and result in tissue destruction, while anti-inflammatory cytokines usually attenuate the effects of the proinflammatory cytokines [31]. IL-10 is an anti-inflammatory cytokine and a potent activator of B-lymphocytes [31]. MCP-1 is a chemotactant for mono cyte, T-cell, and NKcell recruitment. IL-8 mainly recruits neutroph ils but can also act to recruit monocytes [32]. TNFcan stimulate other proinflammatory m ediators as well as interfere with the 18

PAGE 19

growth, differentiation, and death of cells [33 ]. IL-6, while exhibiti ng pro-inflammatory effects, also has anti-inflammatory properties as it inhibits other pro-inflammatory cytokines [34]. IL-12 (p40) initiate s interferonsecretion and cytolytic activity of NK cells as well as promoting differentiation and proliferation of T helper 1 cells [35]. IL-1 is one of the major cytokines produced at infl amed sites and is involved in the initiation and progression of tissue destruction [36]. Many studies have demonstrated that the host cell core transcriptional responses to F. nucleatum and S. gordonii are extensively more than the response to P. gingivalis and A. actinomycetemcomitans [2]. After monitoring the multilayers response through these cytokines, a more complete vision of the possible host cell response can be built. Apoptosis in Epithelial Cells Apoptosis is physiological cell death that takes place in both health and disease. It plays a role in development and homeostasis normally, but can also be caused by the body to eliminate virus-infected cells, mutated ce lls, or bacteria-infected cells [37]. It is separate from necrosis and is characteriz ed by nuclear chromatin condensation, cytoplasmic shrinking, dilated endoplasmi c reticulum, and membrane blebbing [38]. There are also changes in the composition of the cell membrane during apoptosis. Phosphatidyl serine (PS) is normally f ound in the cytoplasmic side of the plasma membrane, but during early apoptosis it is redi stributed to the outer leaflet. Annexin-V is a phospholipid binding protein wit h an especially high affinity for PS that can be used to detect these changes to the membrane [39]. While there are many pathw ays and signals that can lead to apoptosis, there is just one mechanism that is the ultimate cause of cell death; the activation of a proteolytic cascade of cystein proteases (c aspases) which affect literally all cell 19

PAGE 20

functions [40]. Inappropriate activation of t he pathways that lead to this mechanism can cause or worsen disease [37]. Monolayers of human gingival epithelia l cells have exhibited an apoptotic response to challenge with a biofilm consisting of many different species of oral bacteria [41]. Following challenge with P. gingivalis monolayers of gingival epithelial cells have shown a brief increase in apoptotic molecule s followed by a decrease to balance out apoptosis [42]. Aggregatibacter actinomycetem comitans has shown to induce apoptosis mediated by a cDNA endonuclease in KB cells(a common human oral epidermoid cell line) [43]. Apoptosis has been induced by Fusobacterium nucleatum in PMNs and PBMCs, specifically by bacterial surfac e proteins and signaling proteins [44]. Streptococcus gordonii (S. gordonii) Streptococcus gordonii (of the viridians group) is one of the initial (primary) colonizers of the salivary pellicle. It is a non-invasive, gram-positive facultative anaerobic cocci that excretes hydrogen peroxide as a byproduct of metabolism, and was classified by Socransky as part of the yellow complex [2, 16]. Streptococcus gordonii is associated with oral health and tends to be more prevalent in disease-free areas [16]. The specific strain used in ex perimentation is DLI, a common lab strain. Although colonization of the dent ition by streptococci is one of the first steps in the development of plaque [1], after 4 hours of plaque formation, streptococci make up 6070% of the biofilm [45]. Binding of S. gordonii to the acquired salivary pellicle is mediated by several adhesins: cell wall anchored adhesins, lipoprotein adhesins and anchorless adhesions [46]. For instance, AbpA binds amylase in the salivary pellicle [47], while Ssp polypeptides bind a mucin-like salivary agglutinin of the salivary pellicle. Importantly, Ssp also mediates coadhesion with P. gingivalis [47]. Because of its role as 20

PAGE 21

a primary colonizer and the nece ssity of its presence for furt her colonization by other, more pathogenic oral bacteria (such as P. gingivalis), S. gordonii s infection patterns and host cell respons es are of interest for this project. Porphyromonas gingivalis (P. gingivalis) P. gingivalis is the most frequently det ected species whose pres ence is elevated in periodontitis patients [48]. P. gingivalis is a gram-negative black pigmented strict anaerobic bacterium [49], which is assa crolytic, depending on nitrogenous substrates for energy. In addition, it has an obligate iron requirement for growth, but lacks a siderophore system and therefore uses hemin (f rom blood) to satisfy this requirement. Levels of hemin in the oral cavity tend to fluctuate and bleeding can result from inflammation which may explain why some periodontally diseased si tes are predisposed to P. gingivalis accumulation [4]. P. gingivalis is associated with severe, chronic manifestations of periodontal disease which involve inflammatory tiss ue destruction [1, 26]. It has been categorized as part of the red complex, a group composed of periodontal pathogens [50]. P. gingivalis is usually among the late or secondary col onizers of the oral cavity [4], it can form biofilms with S. gordonii and other oral streptococci with the exception of S. mutans [1]. P. gingivalis can adhere to many ear ly plaque bacteria [4]. This attachment is facilitated by P. gingivalis long fimbriae (FimA) and short fimbriae (Mfa). Other bacteria may aid in the colonization of cells by P. gingivalis by providing attachment sites for interspecies adherence (which lead s to biofilm formation), supplying growth substrates, and reducing oxygen tension to the low levels required for growth and survival of this obligate anaerobe [4]. P. gingivalis can also bind to other, later 21

PAGE 22

colonizers such as Fusobacterium nucleatum where it has exhibited a 2-20 fold increase in invasion efficiency wh en the two are coincubated [4, 51]. Adhesion, Invasion, and Migration Adhesion and intracellular invasion of epithelial cells by P. gingivalis is crucial to establish persistence and is central to t he pathogenesis of periodontitis, especially during the initial stages of infection [49]. Adhesion of P. gingivalis to host cells is multimodal and involves a vari ety of cell-surface and extrac ellular components including fimbriae, proteases, hemagglutinins, and lip opolysaccharides [49]. Among the large array of virulence factors produced by P. gingivalis FimA as well as the cysteine proteinases gingipains contribute to the attachment and invasion of oral epithelial cells via different receptors [49]. P. gingivalis has at least two distinctive fimbriae for adhesion; long and short fimbriae. Long fimbri ae are composed of a fimbrillin monomer subunit (encoded by the fimA gene) and mediate attachment to salivary proline-rich proteins and statherin; early colonizers such as Streptococcus gordonii; epithelial cells, endothelial cells, and fibroblasts; and matrix proteins such as fibronectin and fibrinogen. There is a stable association of fimbrillin wi th its salivary receptor, thus ultimately establishing bacterial adherence to saliva-coated surfaces in the oral cavity [4]. Short fimbriae (composed of Mfa) mediate attachment to other bacteria. Invasion is considered an impor tant virulence factor for P. gingivalis affording protection from the host immune system and contribut ing to tissue damage [6]. P. gingivalis has been shown to invade cells of mu ltilayered pocket epithelium, primary gingival epithelial cells in monolayers, and transformed epithel ial cells in monolayers [6, 52]. While adherence of P. gingivalis to epithelial cells is multimodal, the subsequent invasion is fimbria mediated. In primary gingival epithelial cells, P. gingivalis can induce 22

PAGE 23

membrane invaginations that surround and engulf them but then rapidly localize in the cytoplasm without being first constrained by a membrane-bound vacuole [4]. Intracellula r invasion of P. gingivalis is mediated by FimA interacting with integrin receptors on gingival epithelial cells. Integrin-dependent signa ling along with signaling induced by P. gingivalis secretion of SerB protein (serine phosphatase) results in remodeling of the host microf ilament and microtubule cytoskeleton that is necessary for bacterial engulfment and trafficking to the peri nuclear area. Both bi nding to and entry of P. gingivalis into primary epithelial cells are more efficient than with transformed cells such as KB cells [49]. This invasive pr ocess occurs within 12 minutes with subsequent subversion of host intercellular events and la rge numbers of bacteria localized in the perinuclear region [4, 25]. P. gingivalis can remain resident in the perinuclear area for extended periods without causing host cell death [2]. Migration Following invasion, P. gingivalis have shown to penetrate beneath the superficial cell layer, migrate through the basement membrane, and reach the underlying connective tissue. The bacteria have been show n to internalize within multilayered gingival epithelium, as well as at the junction between the stratified epithelium and the lamina propria [49]. Intracellular P. gingivalis is capable of spreading between host through actin-based intercellular protrusi ons [2, 53]. Its capacity to disseminate intercellularly appears to be acquired relatively late in the intracellular invasion process being detected significantly only after 24 hours [53]. Post-Invasion Activities Gingival epithelial cells containing internalized P. gingivalis exhibit morphological changes such as cell rounding and detac hment, however apoptotic cell death is 23

PAGE 24

suppressed by inhibition of the intrinsic apop totic pathways [49]. It also suppresses the inflammatory response as supported by inhibiti on of IL-8 secretion by gingival epithelial cells, inducing anti-inflammatory cytoki nes, degrading existing cytokines [4]. Pathogenic Properties Virulence factors of P. gingivalis include extracellular proteolytic enzymes, toxic metabolites, and cellular constituents and adherence factors [6]. P. gingivalis LPS has shown to be a poor activator of monocyte pr oduction of IL-1 beta and TNF-alpha [4]. P. gingivalis is a skilled evader of the immune response, it can impinge on PMN recruitment and activity and inhibit neutrophil chemotaxis using low molecular weight fatty acids [4]. At least five hemaggl utinating molecules are produced by P. gingivalis [4]. It also produces multiple proteases that can degrade a number of potentially important substrates in the gingival crevice, including co llagen, fibronectin, fibrinogen, laminin, and keratin [4]. Proteinases secreted by P. gingivalis degrade extracellular matrix proteins, activate MMPs to dysregulate tissue r epair, inactivate plasma proteinase inhibitors, cleave cell surface rece ptors, activate and inactivate complement factors and cytokines, and activate the kallik rein cascade, all contributing to tissue degradation and stifling of host defense mechanisms [4]. These combined factors make P. gingivalis a particularly exciting candidate for this comprehensive study of visualization of attachment, invasion, and various epithelial cell responses. Aggregatibacter actinomycetemcomi tans (A. actinomycetemcomitans) Aggregatibacter actinomycetemcomitans is a gram-negative coccobacillus that is almost exclusively associated with locali zed aggressive periodontitis (LAP). Clinical cases of localized aggressive periodontitis (LAP) involve acute tissue destruction, and are associated with the proapoptotic and proinflammatory bacterium A. 24

PAGE 25

actinomycetemcomitans Yet these cases do not result in a large-scale inflammation or gingival destruction [2, 26]. A. actinomyc etemcomitans is a capnophile, meaning it thrives in carbon dioxide-rich environments, such as the one created in periodontal disease [54]. There are six different serotypes of A. actinomycetemcomitans based on the polysaccharides present on the surface of the organism [55]. Serotype b is more frequently observed in patients with aggressive periodontitis [55]. A. actinomycetemcomitans has been linked to several systemic conditions including septic endocarditis, brain and lung abcesses, cardio vascular diseases, and chronic periodontal diseases [54]. A. actinomycetemcomitans has the ability to invade intracellularly through a very specific mechanism. Initial adhesion to the transferrin receptor of epithelial cells alongside binding to integrins stimulates the invasion process. This primary attachment induces effacement of the microvilli and the bacteria enter through ruffled apertures in the cell membrane. The bacteria are initia lly confined within a host-derived membrane vacuole, which is quickly broken down, releasing them into the cytoplasm. A. actinomycetemcomitans can migrate between adjoining cells through formation of surface membrane protrusions[56]. A. actinomycetemcomitans produces a number of viru lence factors including adhesins, endotoxin [lipopolysaccharide (LPS )], leukotoxin (Lkt), and cytolethal distending toxin (Cdt) [57]. The leukotoxin is heat-labile, and has been associated with its ability to evade host cell defenses of t he periodontal tissues [54]. Type IV Flp fimbriae are responsible for non-specific adherence and biofilm formation on solid 25

PAGE 26

26 surfaces such as tooth structure [16]. Au totransporter proteins Aae and ApiA mediate specific adhesion to epithelial cells [16]. The extracellular matrix protein adhesion A (EmaA), also an autotransporter protein, mediates binding to collagen [16]. A. actinomycetemcomitans enhances its chance of colonization by producing actinobacillin, an antibiotic that is active against both streptococci and Actinomyces, primary colonizers of the tooth surface [58]. A. actinomycetemcomitans has three known individual factors t hat can stimulate bone resorp tion (lipopolysaccharide, proteolysis-sensitive factor and GroEL), as we ll as a number of activities (collagenase, fibroblast cytotoxin) that affe ct connective tissue and the extr acellular matrix [58]. These bone resorptive factors can eventually lead to tooth loss. Fusobacterium nucleatum (F. nucleatum) Fusobacterium nucleatum is a gram-negative, anaerobic, non-spore-forming, spindle-shaped or fusiform r od [59]. It is present in high numbers in supraand subgingival plaque both in health and disease. F. nucleatum is carried in 80% of adults regardless of health or disease [60]. It is an opportunistic commens al, meaning it is normally present during health, but under cert ain conditions, can escape host restraint mechanisms and initiate disease [13]. The bacterium can invade intracellularly and is part of the orange comple x, associated with gingivitis and gingival bleeding [2, 13]. It produces tissue irritants such as butyric acid, proteases and cytokines and has strong adhesive properties because of lectin-like cell wa ll proteins [61]. IL-6 and IL-8 secretion has been observed from cultured HIGK m onolayers following infection with the periodontal pathogen, which may serve to further increase its pathogenicity [1-2]. F. nucleatum has been implicated in several systemic infections such as Lemierres syndrome, aspiration pneumoni a, and liver abscess [62].

PAGE 27

CHAPTER 2 MATERIALS AND METHODS Primary Cell Culture A cryovial containing 2 million primary gingival epithelial cells was taken from freezer stocks of cells acquired previously [63]. The cryovial was rapidly thawed by immersing it in a 37 C waterbath. The cryovial was wiped with 70% ethanol and thawed cells were transferred to a sterile 15 mL centrifuge tube. 10 mL of pre-warmed Basal Growth Medium (BGM) cont aining antibiotic/antimycoti c (penicillin 100 units/mL + streptomycin 100 g/mL + amphotericin B 25 g/mL) (Gibco, Gaithersburg, MD) was added to the centrifuge tube and the two were mixed gently. The cells were centrifuged for 5 minutes at 1000 rpm, the supernatant was aspirated and the cell pellet was resuspended in 1 mL of BGM. Cells were pl ated in one T25 Corning flask with 5 mL of BGM and medium was changed the following morning to elim inate floating dead cells and any residual DMSO. When cells were at 80% confluence, medium was removed from the cells, 2 mL of 0.05% Trypsin/0.53 mM EDTA (CellGro/Medi atech Inc., Manasas, VA) was added to the T25 and incubated for 2 minutes at 37 C (or until the majority of the cells had detached). The suspension was transferred to a 15 mL conical tube, the Trypsin was neutralized with 2 mL of BGM, then centrifuged for 5 minutes at 1000 rpm to pellet the cells. The supernatant was aspirated and the cell pellet was resuspended by gently pipetting up and down with 5 mL of BGM. The cells were then re-seeded into a T75 Corning flask in 15 mL of BGM. 27

PAGE 28

Immortalized Cell Culture The frozen stocks of human immortalized gingival keratinocyte (HIGK) cells described previously were quickthawed in a 37C water bath, suspended in 5mL Keratinocyte Serum-free Medium (K-SFM, Gibco/ Invitrogen, Carlsbad, CA) supplemented with 0.05 mM calcium chloride, 200 mM L-glutamine (G ibco/ Invitrogen), and 1% antibiotic/antimycotic (Gibco/ Invitrogen). The 5 mL cell suspension was then centrifuged to pellet for 5 minutes at 2,000 RPM. The pelleted cells were then resus pended in 5 mL of K-SFM and seeded to one T-25 flask and stored at 37 C in 5% CO2. Media was changed every 2 days until 100% confluence was reached. The confluent flask was then reseeded to one T-75 flask by incubating with 2 mL trypsin for 2 minutes collecting the suspended cells to a 15mL microcentrifuge tube, neutralizin g the trypsin with 2mL of K-SFM, and centrifuging the suspended cells to pellet for 5 minutes at 2,000 RPM. Supernatant was discarded and the cell pellet was resuspended in 5 mL K-SFM and added to the T-75 flask already containing 10 mL K-SFM. Three-Dimensional ModelMultilayer Membranes GEC Multilayers When the T-75 reached 80% confluence, the cells were again trypsinized (as previously described), collected by centri fugation, resuspended and reseeded into an appropriate number of T-75 Corning flasks. 12 12mm membranes can be seeded from 1 T-75 flask, the number of T-75 flasks to re seed into is determined from the number of membranes required for experimentation. When those reached a confluence of 80%, they were trypsinized, collected by centrifugation, and resuspended an appropriate amount on BGM to account for 300 L total volume per well. Polyester (PET) 28

PAGE 29

Membrane Transwell Clear Inse rts (12 mm diameter, 0.4 M pore size) (Tissue Culture Treated, Corning 3450) were placed in 12-well culture dishes (Corning), and cells were seeded onto the membranes at 2 x 105 cells per well in 12-well culture dishes with BGM containing antibiotic/antimycotic (Gibco/ Invi trogen). Medium was also placed in the well under the membrane insert at a volume of 500 L per well. Cells were placed in a 37 C incubator overnight and medi a was changed the following morn ing to eliminate floating dead cells and any residual Trypsin. Medium was changed daily in the amounts of 550 L both below and within the membrane insert until cells r eached 100% confluence. At th is stage, the media was removed permanently from within the memb rane insert, thus beginning air/liquid interface. Cells had access to nutrients fr om media traveling upward through the membrane while being relieved of the pressure media exerted from above. This allowed the cells to grow upward into multilayers. HIGK Multilayers By the same suspension technique previous ly described, cells were collected from flasks, pelleted, and resuspended. The resuspended cells were quantified using the Coulter Cell Counter and live/dead analysis was performed using Trypan blue live/dead staining. Cells were reseeded on to 12mm di ameter collagen-coated membrane inserts (Corning Costar) in the quantity of 2x105 in 500 L of K-SFM. 500 L of K-SFM was also added beneath the membrane inse rt. Media was changed daily in the amounts of 500 L both within and below the membrane inse rt, increasing the amount to 750 L both above and below the membrane insert over the weekend. 29

PAGE 30

Air/Liquid Interface When cells reached 100% confluence on the membranes (time averaging one week), media was removed from within the insert. This allowed for the cells to grow upward into multiple layers simulating tissue layers in vivo. Experiments were performed when approximately 3 layers were present (3 weeks of growth). Bacterial Culture All bacteria were quick-thawed from sto cks of OD= 1 frozen at -80 C. 100 L were streaked out on to their respective agar plat es. Colonies were grown up, restreaked onto new plates, and grown up again to assure purity. From these new plates, individual colonies were selected, gram-s tained to verify identity, and inoculated into broth media. Identity was verified for a second time before experiments by gram-stain. Aggregatibacter actinomycetemcomitans strainVT1169 was cultured in tryptic soy broth and yeast extract. VT1169 is a niladixic acid and rifampin resistant clone derived from the clinical strain SUNY 465 [26, 64]. Porphyromonas gingivalis ( P. gingivalis ) strain 32277 was cultured in tryptic soy broth, s upplemented with yeast extract (1mg/ml), hemin(5 g/ml), and menadione (1 g/ml). Fusobacterium nucleatum strain ATCC25586 was cultured in tryptic soy br oth, supplemented wit h, 2.5% glutamic acid, hemin (5 g/ml), and menadione (1 g/ml). Streptococcus gordonii strain DL1 was cultured in BBL Brain Heart Infusion. All plates were made by adding 20% agar to the respective liquid medium solution for each bacteria, sterilizing via autoclave, and poured into 20mL plates. P. gingivalis species plates also required t he addition of 0.5% sheeps blood (BBL, Becton Dickinson & Co., Sparks, MD) after cooling in a wa terbath, but before pouring. Sg, Fn, and Pg were incubated at 37 C in 5% CO2 anaerobic conditions Aa was incubated at 37 C and in 10% CO2. 30

PAGE 31

Experiments Keratin Expression The GEC and HIGK undifferentiated and differentiated cell lines were characterized to ensure their comparability and to examine relevance to oral epithelium. Keratin expression by the cell lines was te sted for using immunofluorescence (1). Mouse anti-Cytokeratins 1/10, 5/6, 13, 14, and 19 (Zymed Laboratories/Invitrogen) were used as primary antibody and Alexaflu or 555 Goat anti-mouse (Invitrogen) was used as the detection secondary antibody. Transepithelial Electrical Resistance Transepithelial Electrical Resistan ce was measured using the Millicell-ERS (Millipore) to determine how densely packed t he cells were. After 3weeks of growth, membranes were submerged in 1X PBS (4g NaCl, 0.1G KCl, 0.575g Na2HPO4H2O, brought up to 1L with DI water), the electrode components were immersed in the PBS solution and electrical readings were recorded. The tran smembrane epithelial resistance was measured after bacterial challenge to determine whether the challenge disrupted the multilayers. Measurements were taken using the Millicell-ERS (Millipore), as previously described. Using this information in combination with immunohistochemistry data the two cell lines similarities and differences were analyzed for experimental comparability. Bacterial Challenge The optical density of P. gingivalis, F. nuc leatum, and S. gordonii was taken at wavelength 600 using the Eppendorf BioP hotometer and the optical density of A. actinomycetemcomitans was taken at wavelength 495 using the BioRad Smart Spec Plus. Each bacteria type was harvested during log phase (OD .4-.8) and centrifuged to 31

PAGE 32

pellet for 10 minutes at 10,000 RPM (8609 g) using the Biofuge Fresco (Heraeus Instruments). In prim ary GEC monolayers, P. gingivalis in the lag phase has been shown to invade poorly (~2%), whereas inva sion was more efficient fr om the log to late stationary phase (10-15%) [6]. For this reas on, all bacteria for this project were harvested during mid-log phase. It is estima ted that there are 800,000 epithelial cells contained within the 3 layers on each memb rane. Enough bacertia were harvested to infect with a multiplicity of infection (MOI) of 200. After centrifuga tion, the cells were then resuspended in 1mL of 1X PBS and stained using BacLight Red (Invitrogen). A stock solution of Baclight Red was prepared by diluting the dye concentrate in 69 l of methanol. From this stock solution a working solution was prepared by adding 2 L of the stock solution to 18 L of PBS. 1.5 L of the working solution was inoculated into the 1mL of resuspended bacterial cells and in cubated in darkness (as Baclight is lightsensitive) at 37C anaerobically for 30 minutes. Baclight is absorbed by the bacterial cell membrane during this incubation period. The bac teria were then centrifuged to pellet for 10 minutes at 10,000 RPM. T he supernatant was discarded, cells were resuspended in 1mL of 1X PBS and centrifuged to pellet ag ain at 10,000 RPM for 10 minutes. The supernatant was again discarded and pellet wa s resuspended in an amount of K-SFM appropriate to allow for 300 L of resuspended bacteria to be inoculated to each membrane. On the day of bacterial challenge, the epithelial cell multilayers were washed twice and left incubated in antibiotic-free medium fo r at least one hour to eliminate antibiotics from the assay, which may hinder the bacteri al challenge. During bacterial challenge, each bacteria type was added solitarily within the membrane insert in the volume of 32

PAGE 33

300 L. The membranes are then placed back in their normal incubation chamber (while wrapped in foil to prevent the fluorescent Bac light dye from fading ) until the appropriate timepoint of 2, 6, or 24 hours. After the appropriate incubation time t he supernatant was removed from the membrane insert, the membrane was washed with .5 mL of 1X PBS both below and within the insert, then fixed with 4% parafor maldehyde (Electron Microscopy Sciences, 16% solution) in DMSO (Dimet hyl Sulfoxide) (Fisher Biotec h) for 30 minutes while on a Reliable Scietific, Inc. Rocking Shaker. T he paraformaldehyde was suctioned off using sterile glass pipettes, the membranes were washed with warm 1X PBS and stored submerged in PBS in a 4C re frigerator until required for various analysis techniques. Actin Staining The membranes previously described were removed from storage. All work hereafter was performed in the dark to prevent fading of the fluorescent dyes. The PBS was suctioned off from both above and below the membranes using sterile glass pipettes. 300 L of 1% bovine serum albumin (BSA)(Sigma, Fraction V, 96%) in 1X PBS was added to both above and below each membrane; they were then incubated for 30 minutes with rocking. the BSA was suctioned o ff using sterile glass pipettes, then the membranes were washed with 500 L of warm 1X PBS and placed back on the rocker for an additional 15 minutes. After the 15 minute wash period, PBS wa s removed from within the membrane insert and 500 L of Alexa Fluor 635 Phalloidin (Invitrogen) working solution was added. The working solution was prepared as follo ws: 1.5 mL methanol was added to the frozen vial provided by the company to reco nstitute the dye. This stock solution was 33

PAGE 34

then further diluted in 1% BSA in a ratio of 1 part dye and 3 parts 1% BSA. The membranes were then re-covered and placed on a rocker for 30 minutes. After staining, the membranes were wa shed two times with 500L of warm 1X PBS both within an d below the membrane, wit h rocking for 15 minutes after each wash. To mount the membranes on glass slides for analysis, each membrane insert was removed from the well, blotted and placed on an agar plate. Using Protected Disposable Scalpels (stainless steel blade #11) (Bec ton Dickinson Baird-Parker), each membrane was carefully cut from the plastic inser t. The free membranes were then placed on a glass slide, one drop of Vectashield Mount ing Medium (Vector La boratories, H-1000) was added on top, then the coverslip was placed avoiding bubbles on the membrane. Mounted samples were then stored at 4C protected from light. Images of the membranes were collect ed using a Leica DM IRM confocal microscope. The confocal microscope collec ts a series of images (slices) along a specified depth (z-steps) of the specimen. The im ages were analyzed using Micromanager 1.2 (beta), Im age J, and Imaris v6.0.0 software. The Micromanager software was used to separate the different channels from the raw data gathered by the microscope. Imaris software was then used to compile all of the slices from both channels into a 3-dimensional volume view in which the channel intensities can be manipulated as well as the orient ation of the volume. Also f unctions of Imaris are OrthoSlicer and Spot-Counter. The Or tho-Slicer function was used to slice the volume compilation of images into three sections (r epresenting the three ce ll layers). The SpotCounter function was then used to detect and c ount the number of fl uorescent signals in a specified channel and also wit hin a segmented area of the volume. It was used in this 34

PAGE 35

project to count the number of fluorescent signals in the ch annel the bacteria fluoresce at within each of the 3 layers segmented using the Ortho-Slicer function. Annexin V Staining The Apo Target Annexin-V FITC Apoptosis Kit (Invitrogen, PHN1018) was utilized to quantify cells in an apoptotic state. The test wa s run after bacterial challenge to obtain an objective difference in the bacterial species abilities to induce apoptosis. Annexin-V is a 35-36 kDa, calcium-dependent, phospholipid binding protein with high affinity for phosphatidylserine (PS). The Annexin -V binding assay is based on the rapid and selective binding to the PS found in the outer cell membrane beginning early in the process of apoptosis (progr ammed cell death). Viable cells maintain an asymmetric distribution of different phospho lipids between the inner and outer leaflets of the plasma membrane. Choline-containing phospholipids are primarily loca ted on the outer leaflet of viable cells and aminophospholipids (like PS) are found at the cy toplasmic face of viable cells. During apoptosis, the plasma memb rane changes include a redistribution of phosphatidylserine from the cytoplasmic side to the outer leaflet, thus making it accessible for Annex in-V staining. This protocol was modified from the liter ature provided by Invitrogen along with the kit. The Apo Target kit used employs a fluorescent labeled Annexi n-V (Annexin-V FITC) along with propidium i odide (PI) to detect cells undergoing apoptosis. During the early stage of apoptosis, cells begin to display PS on their surfaces. The 10X Annexin-V binding Buffer was d iluted 1:10 in distilled water. The membranes were washed twice with PBS, and then 300 L of PBS was added within and 500 L below the membrane inserts. 5 L of Annexin-V FITC and 10 L of 35

PAGE 36

Propidium Iodide Buffer were added within each membrane insert and incubated for 15 minutes in the dark. 400 L of 1X AnnexinV Binding Buffer was added within each membrane insert and incubated for 30 minutes in the dark. The membranes were then washed with PBS, cut out as described previously on agar plates, mounted on glass slides as described above, and analyzed by confocal microscopy. Luminex Anal ysis MILLIPLEXTM MAP is based on the Lumin ex xMAP technology Luminex uses proprietary techniques to internally colo r-code microsperes with two fluorescent dyes. Through precise concentrations of these dyes 100 distinctly colored bead sets can be created, each of which is coated with a spec ific capture antibody. After an analyte from a test sample is captured by the bead, a bi otinylated detection antibody is introduced. The reaction mixture is then incubated with streptavidin-PE conj ugate, the reporter molecule, to complete the reaction on the surface of each microsphere. The microsperes pass through a laser which ex cites the internal dyes marking the microspere set. A second laser excites PE the fluorescent dye on the reporter molecule. Finally, high speed digital-sign al processors identify each individual microspere and quantify the result of its bioassay based on fluorescent reporter signals. The capability of adding multip le conjugated beads to each sample results in the ability to obtain multiple results from each sample. Supernatant, taken after GE Cs and HIGKs had been stimulated for two, six, and twenty-four hours, was used in the Millipore MILLIPLEXTM Map Kit Human Cytokine/Chemokine custom 7-Plex Mu lti-Cytokine Detection System. The MILLIPLEXTM Human Cytokine/Chemoki ne standard was reconstituted with 250 L 36

PAGE 37

deionized water then serial diluted 1:5 to make six standards. 200 L of MILLIPLEXTM Assay Buffer was pipetted into each well of the MILLIPLEXTM 96-well filter plate (to prewet the filters) then removed by vacuum. 25L of the standards and sample were added to a well of a primed filter bottom 96 well plate. The 7 MILLIPLEXTM Human Antibody-Immobilized Beads (IL-1beta, IL-10, IL-12(p40), IL-6, IL-8, MCP1, and TNFalpha) were delivered in individual vials and thus required sonication for 30 seconds, and vortexing at high speed for one minute using a microbean sonicator bath. They were then mixed together in a volume of 60 l each and brought up to a total volume of 3 ml with MILLIPLEXTM Bead Diluent, after wh ich 25L of the bead solution was added to each well. The filter plate wells were then covered and incubat ed with agitation on a plate shaker in a dark room at 4 C overnight. The vacuum manifold was applied to the bottom of the filter plate and the li quid was removed, afterwhich, 200L of MILLIPLEXTM 1X Wash Buffer (containing 0.05% Proclin) was used to wash the well content two times. 200L of MILLIPLEXT M Wash Buffer was used to suspend the wells contents, then 25L of MILLIPLEXTM Detection Antibody was then added to each well and the filter plate was incubated for 1 hour in the dark at ro om temperature with agitation on a plate shaker. MILLIPLEXTM Streptavidin-Phycoerythrin was diluted 1:12.5 in MILLIPLEXTM Cytokine Assay Buff er. 25L of MILLIPLEXTM StreptavidinPhycoerythrin dilution was added to each well. The filter plate was covered and mixed by vortex at a low speed followed by thirty mi nutes of incubation in a dark room at room temperature on a plate shak er. 25L of MILLIPLEXTM Stop solution was added after which the filter plate was vortexed gently and incubated for five minutes at room temperature in the dark. Vacuum manifold was then applied to the bottom of the filter 37

PAGE 38

plate and liquid was removed. 125L of sheath fluid was then add to each well then mixed by vortex at a low speed and plac ed on a plate shaker for one minute. The Luminex 100 System was used to acquire the results and MILLIPLEXTM Analyst Software (VigeneTec h) was used to analyze the results. 38

PAGE 39

CHAPTER 3 RE SULTS Keratin Expression Epithelial cells express cytoker atins that are displayed on the outer leaflet of the cell membrane. These cytokeratins are expressed differentially and can be tested for to differentiate between cell types. GECs, HIGK s cultured in .15 mM calcium, and HIGKs cultured in 1.2 mM calcium were grown into multilayers for three weeks, until they were each approximately 3 layers (30 m) thi ck. The nature of the multilayers was characterized by testing for the expression of CK-1/10, CK-5/6, CK-13, CK-14, and CK19. An example of what a positive result l ooks like can be seen in Figure 3-1. A negative result showed as a completely black image with no fluorescent signal. CK-19 has classically been used as a histological di fferentiation marker for the junctional epithelium. CK -1/10 is a marker for termi nal differentiation. CK-13 is a marker of junctional epithelium and can be detected to distinguish junctional epithelial cells from oral and sulcular epithelial cells [15, 17, 22]. Cytokeratin-14 is a marker to distinuguish tooth-facing cells of the junctional epitheliu m [15]. CK-5 and CK-14 are expressed in the basal cells of stratified epithelium, and are known as 'basal' keratins [23] The HIGKs cultured in media consisting of .15 mM calcium showed uniform expression of cytokeratin 13 (F igure 3-1A), cytokeratin 14 (F igure 3-1B), and cytokeratin 19 (Figure 3-1C). CK-1/10 and CK-5/6 were not detected in the HIGK multilayers cultured in .15 mM calcium (Figure 3-2). Ther efore, it can be said that the multilayers are representative of the junctional epithe lium and are not terminally differentiated. Multilayers of HIGKs were also cultured in a medium consisting of 1.2 mM calcium in order to induce differentiation. The HIGKs cultured in media containing high 39

PAGE 40

concentrations of calcium showed uniform expr ession of cytokeratin 14, cytokeratin 19, and cytokeratin 5/6 (Figure 3-2). These results signify that t he HIGK multilayers cultured in 1.2 mM calcium are a representative model for the basal and tooth facing cells of the junctional epithelium. The GEC multilayers exhibited strong expression of cytokeratins 19 and 13 as well as moderate expression of cytokeratin 5/6 (F igure 3-2). These results signify that the GEC multilayers are a repres entative model for the basal layer of the junctional epithelium. Invasion assay This experiment determined the ability of or al bacteria to penetrate multilayers of GECs and HIGKs. These two different cell types were used becaus e both cell types have been used as monolayer model systems and this experiment will help to discern how comparable those results are. It had pr eviously been demonstrated that pathogens can invade monolayers; the multilayer model used for this project gives a more comprehensive view of the in vivo bacterial interactions with the gingiva. Multilayers underwent bacterial challenge an d bacterial infiltration of the multilayers as well as intracellular invasion was quantified using immunofluorescence. Images were collected using confocal microscopy (Figures 3-3 th rough 3-8, 3-15 through 3-17), while Imaris software to compile the fluorescent signals into a volume reconstruction, the ortho-slicer function was used to cut the three-dimensiona l image reconstructions into 3 cell layers (Figure 3-3). The Imaris spot counter function was used to set a threshold and count the number of bacteria in each layer (Figure 34). Multiple bacteria including pathogens (P. gingivalis, A. actinomycetemcomitans ), an opportunistic commensal ( F. nucleatum ), and 40

PAGE 41

a commensal ( S. gordonii ) were utilized in order to visu alize the int eractions that may take place with varying pat hogenicities of bacteria. GEC Multilayers Porphyromonas gingivalis was observed as penetrating the cell layers, with 7080% of the total bacteria associated with the layers present in the top layer after 2 hours, but moving between layers and remainin g internalized to have a significant presence, upwards of 30%, in the middle layer after 24-hours (Figures 3-8, 3-11). P. gingivalis was also observed as being intracellular after 2 hours, which continued into the 6-hour and 24-hour timepoint s (Figures 3-19, 3-20). Aggregatibacter actinomycetemcomitans penetrated the multilayers very effectively from early on. After 2 hours, nearly 60% of the tota l bacteria present had already penetrated the top laye r of cells to reach the sec ond and third layer. It is clear that the bacteria continued to spread th rough the 6-hour timepoint to be evenly dispersed between the three layers by t he 24-hour timepoint. There was extensive destruction of the cells obs erved after challenge with A. actinomycetemcomitans (Figure 3-5, 3-10). Fusobacterium nucleatum was visualized after 2 hours with 70-80% of the total bacteria being in the top layer, about 20% in the middle layer, and a negligible amount in the bottom layer. After 6 hours, the number present in the bottom layer had risen to 20-30% of the total bacteria present. After 24 hours, the bacteria were more evenly spread, with 40-50% in the top layer, 10-20% in the middle layer, and 30-40% reaching the bottom layer (Figures 3-6, 3-9). There was marked destruction of the cells noted. 41

PAGE 42

Streptococcus gordonii was the least effective penetrator of the cell layers. After 2 hours, the majority of the bacteria were in the top layer. This trend continued through the 6-hour and 24-hour timepoint s (Figures 3-7, 3-12). HIGK Multilayers Porphyromonas gingivalis did not show substantial pen etration of the cell layers. At the 2 hour timepoint most bacteria resi ded in the top layer (Figures 3-13, 3-15) and only 10% had moved to the second layer by the 24 hour timepoint (Figure 3-14). Aggregatibacter actinom ycetemcomitans moved through the layer equally as effectively as F. nucleatum but perhaps more quickly. At the 2 hour timepoint, 70% were in the top layer, 20% in the middle layer, and 10% in the bottom layer (Figure 313). By the 24 hour timepoint, the presence was roughly the same as F. nucleatum with 40% in the top layer, 40% in the middle la yer, and 20% in the bottom layer (Figures 314, 3-16). The bacteria apparently moved th rough the layers as aggregates, as they were observed in clusters at both timepoints. Fusobacterium nucleatum was effective at penetrating the cell layers. At the 2 hour timepoint, 60-70% of the bacteria were in the top layer, 30-40% had penetrated into the second layer and about 5% had delved into the bottom layer (Figures 3-13, 317). By the 24 hour timepoint, 40% were in t he top layer, 40% in the second layer, and 20% in the bottom layer (Figure 3-14) Cell destruction was also observed. Streptococcus gordonii stayed, for the most part in the top layer throughout the 24 hours. Only negligible amounts of the bacteri um were able to move into the second layer (Figures 3-13, 3-14, 3-18). 42

PAGE 43

Transepithelial Electrical Resistance Transepithelial electrical resistance is a measure of membrane potential and resistance of epithelial cells. Multilayers of GECs, HIGKs cultured in .15 mM calcium, and HIGKs cultured in 1.2 mM calcium were te sted to determine if they were capable of establishing a stable transepithelial electrical resistance. Primary keratinocytes in monolayers have displayed transepithelial electrical resistance values of 110 x cm2. Multilayers of GECs were tested for building a stable transepithelial electrical resist ance, and reported a value of 125-225 (Figure 3-21) indicating that they did build a stable TER. The GEC multilayers were also tested after bacterial challenge. There was no significant increase or decrease in transepithelial electrical resistance measures with any of the bacteria over 24 hours. (Figure 3-21) Immortalized keratinocyte cell lines in monolayers have displayed transepithelial electrical resistances of 160 x cm2, indicating that a stable transepithelial electrical resistance was built [17]. The multilayers for this project were grown to a thickness of approximately three layers (30m).The transepi thelial electrical resistance for HIGK multilayers cultured in medium containing .15 mM calcium tested at 140 x cm2 (Figure 3-22). Human immortalized gingival epithelial cells were cultured for three weeks in a medium containing 1.2 mM calcium to i nduce differentiation. The transepithelial electrical resistance of these different iated multilayers measured between 110-250 x cm2 (Figure 3-23). After bacterial challenge, the transepithelia l electrical resistance was measured to determine if there was a significant effect on the junctions formed between cells. These multilayers were cultured for an equivalent ti me to the control (uninfected) multilayers mentioned above. The HIGKS cu ltured in .15 mM calcium displayed an increase in 43

PAGE 44

transepithelial electrical resistance from 140 to 153.5 x cm2 (p<0.05) after challenge with A. actinomycetemcomitans for 6 hours. There was also an observed increase from 140 x cm2 to 159 x cm2 (p<0.05) after challenge with F. nucleatum for 6 hours. There was an increase in trans epithelial electrical resistance from 140 x cm2 to 149.5 x cm2 (p<0.05) after 24 hours challenge with P. gingivalis .(Figure 3-22) The HIGK multilayers cultured in media containing 1.2 mM calcium were retested for forming a stable transepithe lial resistance after bacterial challenge as well. There was an increase in transepithelial electrical resistance from 114 x cm2 to 135 x cm2 and 134 x cm2 to 153 x cm2 (p<0.05) after challenge with A. actinomycetemcomitans for 6 and 24 hours, respectively. There was also an observed increase from 140 x cm2 to 162.5 x cm2 (p<0.05) after challenge with F. nucleatum for 2 hours. There was an increase in transepi thelial electrical resistance from 114 x cm2 to 137.5 x cm2 (p<0.05) after 6 hours challenge with P. gingivalis.(Figure 3-23). Cytokine Secretion Following 2 hour, 6 hour, and 24 hour cha llenge with bacteria, the multilayers supernatants were collected and analyzed for cyt okine secretion from the cells. This project used Luminex technology to monito r the release of Inte rleukin 1-beta (IL-1 ), Interleukin 6 (IL-6), Interleukin 8 (IL-8), Interleukin 10 (IL-10), Interleukin 12 (IL-12 (p40)), monocyte chemoattracta nts protein 1 (MCP-1), and tu mor necrosis factor alpha (TNF) from HIGKs and GECs after undergoi ng bacterial challenge over a 24-hour period. The experiments were repeated three separate times with each repeat consisting of two biological replicates. The GEC multilayers exhibited an increase in the secretion of IL-1 following challenge with S. gordonii F. nucleatum and A. actinomycetemcomitans (Figure 3-24). 44

PAGE 45

There was an observed increase in IL-6 and IL-8 following challenge with S. gordonii and marginally with A. actinomycetemcomitans (Figures 3-25, 3-26). P. gingivalis did not induce secretion of IL-1 IL-6, or IL-8 (Figures 3-24 through 3-26). There was no observed change in IL-10, IL-12(p40), or MCP-1 (Figures 3-27 though 3-29). TNFdisplayed an increase after bacterial challenge with S. gordonii, and F. nucleatum (Figure 3-30). The HIGK multilayers cultured in media containing .15 mM calcium exhibited an increased secretion of IL-6 and IL-8 following challenge with A. actinomycetemcomitans and F. nucleatum (Figures 3-31, 3-32). P. gingivalis caused a decrease in the secretion of IL-6 (Figures 3-32). The secretion of IL-10 and IL-12(p40) increased following bacterial challenge with P. gingivalis (Figures 3-33, 3-34). Apoptosis After the multilayers underwent bacterial challenge, they were tested for the induction of apoptosis. The Annexin V-FITC kit was used for detection of apoptosis markers and images were collected using co nfocal microscopy. Slidebook software was used to measure the average intensity of the fluorescent signals from each challenged multilayer. For the HIGK multilayers cultured in .15 mM calcium media there was an observed trend of apoptosis being induced afte r 6 hours of bacterial challenge; S. gordonii was the strongest inducer, and A. actinomycetemcomitans and F. nucleatum induced apoptosis as well, but to a lesser extent. P. gingivalis did not induce apoptosis. (Figure 3-35) For the HIGK multilayers cultured in medi a containing 1.2 mM calcium, the same trends were observed as with the multilaye rs cultured in .15 mM calcium media. 45

PAGE 46

Apoptosis was induced by bacterial challenge with S. gordonii and A. actinomycetemcomitans F. nucleatum increased apoptosis but not as strongly as the other two bacteria (Figure 3-36). For the GEC multilayers S. gordonii F. nucleatum and A. actinomycetemcomitans all stimulated the i nduction of apoptosis. P. gingivalis caused an increase in apoptosis at first, but by the 24-hour timepoint, was observed at an intensity equal to the control (Figure 3-37). Figure 3-1. Image of HIGK multilayer cult ured under .15 mM calcium conditions tested for expression of cytokeratin 13 (A.), cytokeratin 14 (B.), and cytokeratin 19 (C.). 46

PAGE 47

Figure 3-2. Results of test ing for keratin expression of GECs, undifferentiated HIGKs and differentiated HIGKs. Green check ma rks indicate a positive result for that cytokeratins expre ssion. Red Xs indicate a negative result for that cytokeratins expression. 47

PAGE 48

Figure 3-3. Example of the process fo r bacterial counting. Confocal image 3-D compilation of HIGK multilayers (repr esentative of 3 cell layers) following bacterial challenge with Aggregatibacter actinomycetemcomitans for 24 hours (A). HIGK cells fluorescently label ed with Texas Red (red channel). Bacteria fluorescently labeled with Baclight Green (green channel). This figure demonstrates the Imaris Ortho Slicer f unction (yellow line) that was used to distinguish the three cell layers (B and C)in the compiled volume view of the confocal image. Volume = 478.5 mm3 48

PAGE 49

Figure 3-4. Confocal image 3-D compilation of HIGK multilayers (representative of 3 cell layers) following bacterial challenge with A. actinomycetemcomitans for 24 hours. Actin fluorescently labeled with Texas Red (red channel). Bacteria fluorescently labeled with Baclight Green(green channel not shown). Spot counter function demonstrated (green dots) counting fluorescent signals from bacteria in the top layer of cells (A.), top and middle layers of cells (B.), and top, middle, and bottom layers of cells (C.). Volume = 478.5 mm3 49

PAGE 50

Figure 3-5. Confocal image 3-D compilation of GEC multilayers (respresentative of 3 cell layers) after bacterial cha llenge with Aggregatibacter actinomycetemcomitans for 24 hours. Red is acti n labeled with Texas Red. Blue is DAPI. Green is A. actinomycetemcomitans labeled with Baclight green. Volume = 478.5 mm3 50

PAGE 51

Figure 3-6. Confocal image 3-D compilation of GEC multilayers (representative of 3 cell layers) after bacterial challenge with Fusobacterium nucleatum for 24 hours. Red is actin labeled with Texas Red. Blue is DAPI. Green is F. nucleatum labeled with Baclight green. Volume = 478.5 mm3 51

PAGE 52

Figure 3-7. Confocal image 3-D compilation of GEC multilayers (representative of 3 cell layers) after bacterial challenge with Streptococcus gordonii for 2 hours. Red is actin labeled with Texas Red. Blue is DAPI. Green is S. gordonii labeled with Baclight green. Volume = 478.5 mm3 52

PAGE 53

Figure 3-8. Confocal image 3-D compilation of GEC multilayers (representative of 3 cell layers) after bacterial challenge with Porphyromonas gingivalis for 2 hours. Red is actin labeled with Texas Red. Blue is DAPI. Green is P. gingivalis labeled with Baclight green. Volume = 478.5 mm3 53

PAGE 54

Figure 3-9. Percentage of the to tal bacteria present in each of the three (top, middle, or bottom) layers. Multilayers of primary gingival epithelial cells underwent bacterial challenge with F. nucleatum for 2, 6, or 24 hours. Both the cells and bacteria were fluorescently labeled, and 3-D images were collected with a confocal microscope. Numbers of to tal bacteria were pulled from those images and percentages for each layer were calculated. The error bars are representative of standard deviation. 54

PAGE 55

Figure 3-10. Percentage of the total bacteria present in each of the three (top, middle, or bottom) layers. Multilayers of prim ary gingival epithelial cells underwent bacterial challenge with A. actinomycetemcomitans for 2, 6, or 24 hours. Both the cells and bacteria were fluorescently labeled, and 3-D images were collected with a confocal microscope. Nu mbers of total bact eria were pulled from those images and per centages for each layer were calculated. Error bars are representative of standard deviation. 55

PAGE 56

Figure 3-11. Percentage of the total bacteria present in each of the three (top, middle, or bottom) layers. Multilayers of prim ary gingival epithelial cells underwent bacterial challenge with P. gingivalis for 2, 6, or 24 hours. Both the cells and bacteria were fluorescently labeled, and 3-D images were collected with a confocal microscope. Numbers of to tal bacteria were pulled from those images and percentages for each layer were calculated. Error bars are representative of standard deviation. 56

PAGE 57

Figure 3-12. Percentage of the total bacteria present in each of the three (top, middle, or bottom) layers. Multilayers of prim ary gingival epithelial cells underwent bacterial challenge with Streptococcus gordonii for 2, 6, or 24 hours. Both the cells and bacteria were fluorescently labeled, and 3-D images were collected with a confocal microscope. Numbers of total bacteria were pulled from those images and percentages for each layer were calculated. Error bars are representative of standard deviation. 57

PAGE 58

Figure 3-13. Percentage of bacteria present after 2 hours of challenge in each (top, middle, or bottom) layer of HIGK mulitila yers cultured in media containing .15 mM calcium. Percentages were calculated as described in previous figures. Graph is an average of 6 samples for each bacterium. Error bars are standard deviation. 58

PAGE 59

Figure 3-14. Percentage of bacteria present after 24 hours of challenge in each (top, middle, or bottom) layer of HIGK mulitila yers cultured in media containing .15 mM calcium. Percentages were calculated as described in previous figures. Graph is an average of 6 samples for each bacterium. Error bars are standard deviation. 59

PAGE 60

Figure 3-15. Confocal image (3-D compilation ) of HIGK multilayers (representative of 3 cell layers) after bacterial challenge with P. gingivalis for 2 hours. Actin fluorescently labeled with Texas Red (red channel). Bacteria fluorescently labeled with Baclight Green (green channel). Volume = 478.5 mm3 60

PAGE 61

Figure 3-16. Confocal image (3-D compilation ) of HIGK multilayers (representative of 3 cell layers) after bacterial challenge with A. actinomycetemcomitans for 24 hours. Actin fluorescently labeled with Texas Red (red channel). Bacteria fluorescently labeled with Baclight Green (green channel). Volume = 478.5 mm3 61

PAGE 62

Figure 3-17. Confocal image (3-D compilation ) of HIGK multilayers (representative of 3 cell layers) after bacterial challenge with F. nucleatum for 24 hours. Actin fluorescently labeled with Texas Red (red channel). Bacteria fluorescently labeled with Baclight Green (green channel). Volume = 478.5 mm3 62

PAGE 63

Figure 3-18. Confocal image (3-D compilation ) of HIGK multilayers (respresentative of 3 cell layers) after bacterial challenge with S. gordonii for 24 hours. Actin fluorescently labeled with Texas Red (red channel). Bacteria fluorescently labeled with Baclight Green (green channel). Volume = 478.5 mm3 63

PAGE 64

Figure 3-19. Graphical represent ation of the efficiency of P. gingivalis at invading (internalizing) HIGK multilayers. Percentage of total bacteria present internally and externally at each timepoint. Data was gathered using Imaris software. Imaris enables a user to zoom and rota te the 3-D compilation in order to discern whether a bacterium is inside the cell. 64

PAGE 65

Figure 3-20. Slice view of confocal image of HIGK multilayer afte r bacterial challenge with P. gingivalis for 24 hours. Actin fluorescent ly labeled with Texas Red (red channel). Bacteria fluorescently label ed with Baclight Green (green channel). Volume sliced through the cent er of top cell layer to di splay internalization of P. gingivalis 65

PAGE 66

Figure 3-21. Measurem ent of transepithelia l electrical resistance of GEC multilayers after bacterial challenge. Control multila yers were inoculated with media void of bacteria. 66

PAGE 67

* Figure 3-22. Measurement of transepithelial electrical resistance of HIGK multilayers grown in .15 mM calcium media after bac terial challenge. Control multilayers were inoculated with media void of bacteria. 67

PAGE 68

* Figure 3-23. Measurement of transepithelial electrical resistance of HIGK multilayers cultured in 1.2 mM calcium after bacteri al challenge. Contro l multilayers were inoculated with media void of bacteria. 68

PAGE 69

* Figure 3-24. Secretion of IL-1 by GEC multilayers. pg/mL of cytokines measured by Luminex, analysis performed us ing Milliplex software. 69

PAGE 70

* Figure 3-25. Secretion of IL -6 by GEC multilayers. pg/mL of cytokines measured by Luminex, analysis performed us ing Milliplex software. 70

PAGE 71

* * Figure 3-26. Secretion of IL -8 by GEC multilayers. pg/mL of cytokines measured by Luminex, analysis performed us ing Milliplex software. 71

PAGE 72

Figure 3-27. Secretion of IL -10 by GEC multilayers. pg/mL of cytokines measured by Luminex, analysis performed us ing Milliplex software. 72

PAGE 73

Figure 3-28. Secretion of IL -12(p40) by GEC multilayers. pg/mL of cytokines measured by Luminex, analysis performed using Milliplex software. 73

PAGE 74

Figure 3-29. Secretion of MC P-1 by GEC multilayers. pg/mL of cytokines measured by Luminex, analysis performed us ing Milliplex software. 74

PAGE 75

* Figure 3-30. Secretion of TNFby GEC multilayers. pg/mL of cytokines measured by Luminex, analysis performed us ing Milliplex software. 75

PAGE 76

* Figure 3-31. Secretion of IL-8 by HIGK multilayers cultured in media containing .15 mM calcium. 76

PAGE 77

* Figure 3-32. Secretion of IL-6 by HIGK multilayers cultured in media containing .15 mM calcium. 77

PAGE 78

* Figure 3-33. Secretion of IL -10 by HIGK multilayers cultured in media containing .15 mM calcium. 78

PAGE 79

* Figure 3-34. Secretion of IL -12(p40) by HIGK multilayers cultured in media containing .15 mM calcium. 79

PAGE 80

Figure 3-35. Detection of the induction of apoptosis in HIGKs (cultured in .15 mM calcium media) after challenge with S. gordonii A. actinomycetemcomitans P. gingivalis and F. nucleatum The Annexin V-FITC kit was used and fluorescent signals were collected usi ng a confocal microscope and Slidebook software. Slidebook calculated the av erage intensity and st andard deviations for each sampling. 80

PAGE 81

Figure 3-36. Detection of the induction of apoptosis in HIGKs (cultured in 1.2 mM calcium media) after challenge with S. gordonii A. actinomycetemcomitans P. gingivalis and F. nucleatum The Annexin V-FITC kit was used and fluorescent signals were collected usi ng a confocal microscope and Slidebook software. Slidebook calculated the av erage intensity and standard deviations for each sampling. 81

PAGE 82

Figure 3-37. Detection of the induction of apoptosis in GECs after challenge with S. gordonii, A. actinomycetemcomitans, P. gingivalis and F. nucleatum The Annexin V-FITC kit was used and fluoresc ent signals were collected using a confocal microscope and Slidebook so ftware. Slidebook calculated the average intensity and standard devia tions for each sampling. 82

PAGE 83

CHAPTER 4 DISCUSSION Past periodontal disease research has been done using either primary gingival epithelial cells (GECs) or human immortalized gingival keratinocytes (HIGKs). Also, this other previous work had been done using monolayer models. This project used GECs and HIGKs side-by-side in order to determine how comparable results from the two cell lines are. Also, a unique multilayer model was applied in order to bridge the gaps between the monolayers used in the past and in vivo interactions. Characterization of the 3-D Models Epithelial keratin expression can be exploi ted to distinguish different cell types. The nature of both the GEC and HIGK multilayers were char acterized by testing for keratin expression. CK-19 has classically been used as a hi stological differentiation marker for the junctional epith elium. Cytokeratin 13 is a ma rker of junctional epithelium and can be used to distinguish junctional epithelial cells from oral and sulcular epithelial cells [15, 17, 22]. CK-13 is expressed by stratified squamous epithelia. CK-1/10 is a marker for terminal differentia tion and is expressed by suprabas al cells of the stratified squamous epithelium. CK-5 and CK-14 are expressed in the basal cells of stratified epithelium, and are known as 'basal' keratins [23]. CK-14 is a marker to distinuguish tooth-facing cells of the juncti onal epithelium, it is express ed by stratifying epithelial cell types [15]. CK-5/6 is expressed during differ entiation of simple and stratified epithelial tissue. The HIGKs cultured in media consisting of .15 mM calcium showed uniform expression of cytokeratin 13, cytokeratin 14, and cytokerat in 19. The expression of these specific cytokines builds a backstory to these cells as bei ng representative of 83

PAGE 84

undifferentiated tooth-facing cells of the junctional epithelium leading us to believe that these multilayers are a good 3D model representing the tissue of the junctional epithelium. The HIGK multilayers cultured in media containing high concentrations of calcium (to induce differentiation) showed uniform expr ession of cytokeratin 14, cytokeratin 19, and cytokeratin 5/6. From t hese results, it can be deduced that the HIGK multilayers cultured in a high concentration of calciu m is a good 3-D model for differentiated junctional epithelium. The GEC multilayers expressed cytokerat ins 19 and 13 strongly and cytokeratin 5/6 moderately. Since CK19 is a marker for the junctional epithelium, CK13 is secreted by the stratified squamous cells of the juncti onal epithelium and CK 5/6 is expressed by basal cells, we can extrapolate that the GEC multilayers are a good model for simulating the junctional epith elium of the gingivae. When characterizing the GECs, the trans epithelial electrical resistance and cytokine expression were tested. Monolayers of primary keratinocytes in the past have presented with an average TER of 110 x cm2 [17]. The multilayers of GECs underwent testing for the building of a stable transepithelial electr ical resistance of 125-225 This value indicates that the GECs are in fact building a stable transepithelial electrical resistance through forming tight junctions. In previous studies, immortalized kerati nocytes cultured in monolayers have been proven to form transepithelial electrical resistance just as epithelial cells do clinically. These monolayers presented with an average TER of 160 x cm2 [17]. The transepithelial electrical resi stance of the HIGKs cultured in multilayers was tested and 84

PAGE 85

displayed comparable values, showing that they are capable of building a stable transepithelial electrical resistance as well. While the TER values fo r both the GECs and HIGKs indicate that each cell line is building a stable transepithelial electrical re sistance, the slight difference in their numbers may reveal that the HIGKs have sli ghtly tighter contacts. This theory is supported in the confocal images taken of the cells which show the GECs to be less tightly packed than the HIGKs. With the except ion of this near-negligible difference, the two cells lines are highly comparable in this respect. The GEC multilayers, and HIGK multilayers in both conditions of calcium are all similarly efficient 3-D models of the juncti onal epithelium. They each form a stable transepithelial electrical resistance. Both cell lines and calcium conditions express cytokeratins representative of the junctional epithelium. It can therefore be extrapolated that the GEC multilayers, undi fferentiated HIGK multilayers (in .15 mM calcium), and the differentiated HIGK multilayers (in 1.2 mM calcium) are all comparable 3-D models upon which to perform experiments to study period ontal disease in the junctional epithelium. Interactions of P. gingivalis S. gordonii F. nucleatum and A. actinomycetemcomitans with the 3-D Models The GEC multilayers and HIGK multilayers grown in a low concentration of calcium both underwent bacterial challenge with Porphyromonas gingivalis Streptococcus gordonii Fusobacterium nucleatum and Aggregatibacter actinomycetemcomitans The invasion and penetration efficiency of each bacteria was studied by capturing images with confocal microscopy. The host cell response was monitored using Luminex to look at what cytokines and chemokines were released by the multilayers. Induction of apoptosis in t he host cells by each bacteria was examined 85

PAGE 86

by Annexin-V staining and confocal microscopy. Also disruption of the cell-cell junctions by each of the four bacteria was monitored by testing t he transepithelial electr ical resistance after bacterial challenge. In the GEC multilayers, A. actinomycetemcomitans and F. nucleatum were the most effective penetrators of the layers, eventually becoming integrated fairly evenly amongst the three layers. They were followed by P. gingivalis, which (by the 24 hour timepoint) had moved 30% into the second layer. The least effective penetrator of the multilayers was S. gordonii which for the most part stayed in the top layer. P. gingivalis was the only bacterium observed intrace llularly. These penetration depths are consistent with the roles of each bacteria in the oral cavity. Commensal bacteria part of the normal microflora of the or al cavity and are characterized by their ability to adhere to tissues (teeth and periodontal tissue) while no t causing disease. They are found mainly in plaque, not really in periodontal tiss ues. As a primary colonizer and commensal bacteria, S. gordonii should not exhibit an efficient penetration of the tissue. As an opportunistic commensal that has been implicated in gingival bleeding, it makes sense that F. nucleatum was effective at penetrating the ce ll layers and caused some visible destruction of cells. Since A. actinomycetemcomitans is associated almost exclusively with localized aggressive periodo ntitis, its presentation as a highly effective penetrator of the cell layers causing cellular destruction is very consistent with its role. P. gingivalis is a member of Socranskys red complex, and is associated with severe, chronic manifestations of periodontal disease. Bacter ia in the red complex are considered the major agents in periodontitis. P. gingivalis has exhibited an ability to internalize within gingival epithelial cells and monolayers, P. gingivalis internalization capacity was 86

PAGE 87

exhibited in high numbers for the multilaye rs (80% of all bacteria present were intracellular), and it showed an ability to penetrate the layers, movi ng into the second layer. Although other studies have implicated P. gingivalis in the destruction of cells at very high MOIs, no cellular destruction was seen in the GEC multilayers by P. gingivalis (MOI= 200) the 24 hour timepoint. It is possibl e that the multilayers are more resilient than monolayers, and can withst and the internalization of bacteria more readily. Another possibility is that 24 hours of bacterial challenge with P. gingivalis is simply not enough time for the cells to react, since the bacteri um has developed an evasion strategy for its own survival [49]. It secretes a nucl eoside diphosphate kinase that hydrolyzes extracellular ATP and prevents apoptosis mediated through P2X7 [65]. In the HIGK multilayers, again, A. actinomycetemcomitans and F. nucleatum were effective at penetrating the la yers. Much less effective was P. gingivalis and even less effective was S. gordonii A. actinomycetemcomitans had already penetrated the layers by the 2-hour timepoint, and by 24 hours had spread pretty evenly through the layers. This observation has an interesting correlation with the clinical observation of A. actinomycetemcomitans particularly quick progression in localized aggressive periodontitis. It was the fastest moving bacte ria in terms of getting a higher percentage of the total bacteria present to th e very bottom of the cell layers. F. nucleatum was also very effective at infiltrating the cell laye rs. At the 2-hour timepoint, about 30% of the bacteria were already into the second layer, yet there were less in the third layer than A. actinomycetemcomitans had (only about 5%). There was al so visually observed cellular destruction in the multilayers. These observations are consistent with F. nucleatum s 87

PAGE 88

role as an opportunistic commensal in So cranskys orange complex. The clinically observed gingival bleeding assoc iated with this group may be attributable to this cellular destruction by F. nucleatum P. gingivalis did not infiltrate the cell layers near ly as effectively as the previous 2 bacteria discussed. Also of note is that a much smaller percentage of the total bacteria in the HIGK multilayers than in the GEC mu ltilayers were able to permeate the second layer (10% in HIGKs compared to 30% in GECs ). In the HIGK mult ilayers, most of the P. gingivalis were in the top layer (from 80-100%). And the majority of bacteria (80-90%) in the second layer were intracellular, indicati ng that they are the re sult of intercellular migration. Infection with P. gingivalis did not cause any observable cell destruction, which is consistent with other researchers findings that the bacterium can prevent cell death in order to preserve its own survival internally [2]. S. gordonii was the least effective bacterium at permeating the HIGK cell layers. With less than 5% of the total bacteria even reaching the second layer, there was hardly any change from the 2-hour timepoint to t he 24-hour timepoint. It could be theorized even that the bacteria that are present in the second layer are only the result of a gap in confluency of the cell layers. S. gordonii is a primary colonize r of the oral cavity abundantly present during health as well as disease. It did not cause any observable damage to the cells which is consistent with its classification as a commensal organism. A. actinomycetemcomitans, F. nucleatum and P. gingivalis also displayed an ability to increase the transepith elial electrical resistance (TER) in the HIGK multilayers in both low calcium and high calcium conditions. This increase in TER is associated with more tight junctions. S. gordonii did not have any significant effect on the TER of HIGK 88

PAGE 89

multilayers in either condition. Also of intere st is that the stable TER in GEC multilayers was unaffected by challenge with any of the four bacteria. GECs have long been thought of as more sensitive than immortalized cell lines. They certainly have shown to be more susceptible to infiltration and invasi on by these bacteria, so it is very unexpected that the ti ght junctions are seemingly una ffected by this infiltration. A complex relationship between cancer and tight junctions has been shown [66], and it is possible that the cancer-causing HPV virus causes some changes to the tight junction proteins of the HIGK cells that make t hem mo re affected by bacterial challenge. Induction of apoptosis followed similar trends in HIGKs cultured in low calcium, HIGKs cultured in high calcium, and GECs. S. gordonii A. actinomycetemcomitans and F. nucleatum all displayed an ability to induce apoptosis in both cell lines and calcium conditions. The exception was P. gingivalis While in both HIGKs cultured in low calcium and HIGKs cultured in high calcium P. gingivalis did not induce apoptosis, the GEC multilayers exhibited an induction of apoptos is during the 2-hour and 6-hour timepoints, but then returned to the same appearance of uninfected cells by the 24-hour timepoint. It appears as if there is a lag time between in tracellular invasion (which takes place in the first 20 minutes [4, 49]) and P. gingivalis self preservation mechanisms being effective at shutting down the host-cell response. After undergoing bacterial challenge, supernat ants were collected from the HIGK multilayers and GEC multilayers and cytokine/chemokine secr etion was measured using Luminex. In the HIGK multilayers t here was an observed trend of IL-6 and IL-8 levels increasing after challenge with A. actinomycetemcomitans. F. nucleatum also elicited an increase in the secretion of IL-6. Also, the secretion of IL-6 decreased after 89

PAGE 90

challenge with P. gingivalis P. gingivalis subsequently stimulated se cretion of IL-10 and IL-12(p40). IL-6 and IL-8 are ma jor pro-inflammatory cytoki nes. These results indicate that A. actinom ycetemcomitans and F. nucleatum elicited an inflammatory response from the immortalized multilayers while P. gingivalis repressed the inflammatory response. In the GEC multilayers, bacterial challenge with A. actinomycetemcomitans and S. gordonii increased the secretion of IL-1, IL-6, and IL-8. F. nucleatum elicited an increase in the production of IL-1 as well. TNFalso showed an increased secretion after challenge with S. gordoni, and F. nucleatum These results (similarly to the results in HIGK multilayers) indicate that these bac teria can elicit an in flammatory response from the GEC multilayers. Taking all of the above results into consi deration, a comprehensive view of the bacterial-host interactions in peri odontal disease can begin to be built. S. gordonii as a commensal bacterium does not permeate multilayers of cells, but does induce apoptosis and may induce a pro-inflammatory response. P. gingivalis as a periodontal pathogen does permeate multilayers of cells, has a high success rate for invasion of epithelial cells, did not induce apoptosis in the immortalized cells, and while apoptosis was initially induced in the primary cell line, the pr ocess was promptly stopped. P. gingivalis also repressed the inflammatory response in HIGK multilayers. F. nucleatum as an opportunistic commensal had a strong effe ct on the multilayers when they were exposed to the bacterium in singularity. It permeated the cell layers fairly easily and efficiently, induced apoptosis, and stimul ated pro-inflammatory cytokines. A. actinomycetemcomitans as a periodontal pathogen primarily responsible for localized aggressive periodontitis had a str ong effect on the multilayers as well. It was by far the 90

PAGE 91

fastest to infiltrate the multila yers and was very effective at di stributing itself fairly evenly between the three layers. It showed an abili ty to induce apoptosis and stimulate proinflammatory cytokines. The induction of inflammatory effects and apoptosis carries implications for the progression of disease. These cellular responses can bring neutrophils and macrophages into the area causing tiss ue destruction and bone loss. 91

PAGE 92

LIST OF REFERENCES 1. Mans, J.J., K. von Lackum, C. Dorsey, S. Willis, et al., The degree of microbiome comp lexity influences the epithelial response to infection. BMC Genomics, 2009. 10: p. 380. 2. Hendrickson, E.L., Q. Xia, T. Wang, R. J. Lamont, et al., Pathway analysis for intracellular Porphyromonas gingivalis using a strain ATCC 33277 specific database. BMC Microbiol, 2009. 9 : p. 185. 3. Aas, J.A., B.J. Paster, L. N. Stokes, I. Olsen, et al., Defining the normal bacterial flora of the oral cavity. J Clin Microbiol, 2005. 43(11): p. 5721-32. 4. Lamont, R.J. and H.F. Jenkinson, Life below the gu m line: Pathogenic mechanisms of Porphyromonas gingivalis. Microbiol Mol Biol Rev, 1998. 62(4): p. 1244-63. 5. Pihlstrom, B.L., B.S. Mi chalowicz, and N.W. Johnson, Periodontal diseases. Lancet, 2005. 366 (9499): p. 1809-20. 6. Lamont, R.J., A. Chan, C.M. Belton, K.T. Izutsu, et al., Porphyromonas gingivalis invasion of gingival epithelial cells. Infect Immun, 1995. 63(10): p. 3878-85. 7. Jain, M., A. Mathur, S. Ku mar, P. Duraiswamy, et al., Oral hygiene and periodontal status among Terapanthi Svet ambar Jain monks in India. Braz Oral Res, 2009. 23(4): p. 370-6. 8. American Dental Association, Health Topics AZ: Periodontal (Gum) Disease [updated 2010; cited 01/2010]; Available from: http://www.ada.org/3063.aspx?currentTab=1 9. Saito, A., S. Inagaki, and K. Ishihara, Differential ability of periodontopathic bacteria to modulate invasion of human gingival epithelial cells by Porphyromonas gingivalis. Microb Pathog, 2009. 47(6): p. 329-33. 10. Meyer, D.H., P.K. Sreeniva san, and P.M. Fives-Taylor, Evidence for invasion of a human oral cell line by Actinobac illus actinomycetemcomitans. Infect Immun, 1991. 59(8): p. 2719-26. 11. Han, Y.W., W. Shi, G.T. H uang, S. Kinder Haake, et al., Interactions between periodontal bacteria and human oral epit helial cells: Fusobacterium nucleatum adheres to and invades epithelial cells. Infect Immun, 2000. 68(6): p. 3140-6. 12. Handfield, M., H.V. Baker, and R.J. Lamont, Beyond good and evil in the oral cavity: insights into host-microbe relationships derived from transcriptional profiling of gingival cells. J Dent Res, 2008. 87(3): p. 203-23. 92

PAGE 93

13. Hasegawa, Y., J.J. Mans, S. Mao, M.C. Lopez, et al., Gingival epithelia l cell transcriptional responses to commensal an d opportunistic oral microbial species. Infect Immun, 2007. 75(5): p. 2540-7. 14. Demmer, R.T., P.N. Papapanou, D.R. Jacobs, Jr., and M. Desvarieux, Evaluating clinical periodontal measures as surrogates for bac terial exposure: The Oral Infections and Vascular Disease Epidemiology Study (INVEST). BMC Med Res Methodol, 2010. 10(1): p. 2. 15. Shimono, M., T. Ishikawa, Y. Enokiya, T. Muramatsu, et al., Biological characteristics of t he junctional epithelium. J Electron Microsc (Tokyo), 2003. 52(6): p. 627-39. 16. Lamont, R.J. and H.F. Jenkinson, Oral Microbiology at a Glance 2010: WileyBlackwell. 17. Groger, S., J. Michel, and J. Meyle, Establishment and characterization of immortalized human gingiva l keratinocyte cell lines. J Periodontal Res, 2008. 43(6): p. 604-14. 18. Feller, L., R.A. Khammissa N.H. Wood, and J. Lemmer, Epithelial maturation and molecular biology of oral HPV. Infect Agent Cancer, 2009. 4 : p. 16. 19. Hatakeyama, S., Y. Ohara-Nemoto, N. Yanai, M. Obinata, et al., Establishment of gingival epithelial cell lines from transgenic mice harbor ing temperature sensitive simian virus 40 large T-antigen gene. J Oral Pathol Med, 2001. 30(5): p. 296-304. 20. Liu, J., G. Song, Z. Wang, B. Huang, et al., Establishment of a corneal epithelial cell line spontaneously derived fr om human limbal cells. Exp Eye Res, 2007. 84(3): p. 599-609. 21. Schweizer, J., P.E. Bowden, P. A. Coulombe, L. Langbein, et al., New consensus nomenclature for mammalian keratins. J Cell Biol, 2006. 174 (2): p. 169-74. 22. Jiang, Q. and D.Y. Li, [Cytokeratin expression in hum an junctional epithelium, oral epithelium and sulcul ar epithelium]. Zhonghua Kou Qiang Yi Xue Za Zhi, 2005. 40(4): p. 298-301. 23. Gusterson, B.A., D.T. Ross, V.J. Heath, and T. Stein, Basal cytokeratins and their relationship to the cellular origin and f unctional classification of breast cancer. Breast Cancer Res, 2005. 7 (4): p. 143-8. 24. Hatakeyama, S., T. Yaegashi, Y. Oikawa, H. Fujiwara, et al., Expression pattern of adhesion molecules in junctional epithelium differs from that in other gingival epithelia. J Periodontal Res, 2006. 41(4): p. 322-8. 93

PAGE 94

25. Belton, C.M., K.T. Izutsu, P.C. Goodwin Y. Park, et al., Fluorescence image analysis of the association between Po rphyromonas gingivalis and gingival epithelial cells. Cell Microbiol, 1999. 1 (3): p. 215-23. 26. Handfield, M., J.J. Mans, G. Zheng, M.C. Lopez, et al., Distinct transcriptional profiles characterize oral epith elium-microbiota interactions. Cell Microbiol, 2005. 7 (6): p. 811-23. 27. Bartold, P.M., L.J. Walsh, and A.S. Narayanan, Molecular and cell biology of the gingiva. Periodontol 2000, 2000. 24: p. 28-55. 28. Dale, B.A. and L.P. Fredericks, Antimicrobial peptides in the oral environment: expression and function in health and disease. Curr Issues Mol Biol, 2005. 7 (2): p. 119-33. 29. Diamond, G., N. Beckloff, and L.K. Ryan, Host defense peptides in the oral cavity and the lung: similarities and differences. J Dent Res, 2008. 87(10): p. 915-27. 30. Gilman A, G.L., Hardman JG, Limbird LE Goodman & Gilman's The Pharmacological Ba sis of Therapeutics 2001, New York: McGraw-Hill. 31. Dinarello, C.A., Proinflammatory cytokines. Chest, 2000. 118 (2): p. 503-8. 32. Hu, C.J., Y.L. Lee, N.Y. Shih, Y.Y. Yang, et al., Reduction of monocyte chemoattractant protein-1 and interleukin-8 levels by ticlopidine in TNF-alpha stimulated human umbilical vein endothelial cells. J Biomed Biotechnol, 2009. 2009 : p. 917837. 33. Dong, J., Y. Gao, Y. Liu, J. Shi, et al., The protective antibodies induced by a novel epitope of human TNF-alpha could suppress the development of collageninduced arthritis. PLoS One, 2010. 5 (1): p. e8920. 34. Naka, T., N. Nishimoto, and T. Kishimoto, The paradigm of IL-6: from basic science to medicine. Arthritis Res, 2002. 4 Suppl 3 : p. S233-42. 35. Collison, L.W. and D.A. Vignali, Interleukin-35: odd one out or part of the family? Immunol Rev, 2008. 226 : p. 248-62. 36. Boch, J.A., N. Wara-a swapati, and P.E. Auron, Interleukin 1 signal transduction-current concepts and relevance to periodontitis. J Dent Res, 2001. 80(2): p. 400-7. 37. Kumar, S., Apoptosis: Biology and Mechanisms. Results and Problems in Cell Differentiation Vol. 23. 1998, Berlin: Springer. 38. Group, A.I. About Apoptosis [updated 1999; cited 01/2010]; Available from: http://www.nih.gov/sigs/aig/Aboutapo.html 94

PAGE 95

39. Atsumi, T., K. Tonosaki, and S. Fujisawa, Induction of early apoptosis and ROSgeneration activity in human gingival fi broblasts (HGF) and hu man submandibular gland carcinom a (HSG) cells treated with curcumin. Arch Oral Biol, 2006. 51(10): p. 913-21. 40. Jacobson, M.D. and N.J. McCarthy, Apoptosis Frontiers in molecular biology 40. 2002, Oxford, OX ; New Yo rk: Oxford University Press. xxiv, 321 p. 41. Guggenheim, B., R. Gmur, J.C. Galicia, P.G. Stathopoulou, et al., In vitro modeling of host-parasite interactions: the 'subgingival' biofilm challe nge of primary human epithelial cells. BMC Microbiol, 2009. 9 (1): p. 280. 42. Yilmaz, O., T. Jungas, P. Verbeke, and D.M. Ojcius, Activation of the phosphatidylinositol 3-kinase/Akt pathway contributes to survival of primary epithelial cells infected with the periodontal pathogen Porphyro monas gingivalis. Infect Immun, 2004. 72(7): p. 3743-51. 43. Kato, S., K. Nakashima, M. Inoue, J. Tomioka, et al., Human epithelial cell death caused by Actinobacillus actinomycetemcomitans infection. J Med Microbiol, 2000. 49(8): p. 739-45. 44. Jewett, A., W.R. Hume, H. Le, T.N. Huynh, et al., Induction of apoptotic cell death in peripheral blood mononuclear and polymorphonuclear cells by an oral bacterium, Fusobacterium nucleatum. Infect Immun, 2000. 68(4): p. 1893-8. 45. Nyvad, B. and M. Kilian, Microbiology of the early colonization of human enamel and root surfaces in vivo. Scand J Dent Res, 1987. 95(5): p. 369-80. 46. Kesimer, M., N. Kilic, R. Mehrotra, D.J. Thornton, et al., Identification of salivary mucin MUC7 binding proteins from Streptococcus gordonii. BMC Microbiol, 2009. 9 : p. 163. 47. Rosan, B. and R.J. Lamont, Dental plaque formation. Microbes Infect, 2000. 2 (13): p. 1599-607. 48. Colombo, A.V., C.M. Silva, A. Haffajee, and A.P. Colombo, Identification of oral bacteria associated with crevicular epithelial cells from chronic periodontitis lesions. J Med Microbiol, 2006. 55(Pt 5): p. 609-15. 49. Andrian, E., D. Grenier, and M. Rouabhia, Porphyromonas gingivalis-epithelial cell interactions in periodontitis. J Dent Res, 2006. 85(5): p. 392-403. 50. Holt, S.C. and J.L. Ebersole, Porphyromonas gingivalis Treponema denticola, and Tannerella forsythia: the "red complex", a prototype polybacterial pathogenic consortium in periodontitis. Periodontol 2000, 2005. 38: p. 72-122. 95

PAGE 96

51. Saito, A., S. Inagaki, R. Kimizuka, K. Okuda, et al., Fusobacterium nucleatum enhances invasion of human gingival ep ithelial and aortic endot helial cells by Porphyromonas gingivalis. FEMS Immunol Med Microbiol, 2008. 54(3): p. 349-55. 52. Duncan, M.J., S. Nakao, Z. Skobe, and H. Xie, Interactions of Porphyromonas gingivalis with epithelial cells. Infect Immun, 1993. 61(5): p. 2260-5. 53. Yilmaz, O., The chronicles of Porphyromonas gingivalis: the microbium, the human oral epitheliu m and their interplay. Microbiology, 2008. 154 (Pt 10): p. 2897903. 54. Vieira, E.M., S.A. Raslan, T.C. Wahasugui, M.J. Avila-Campos, et al., Occurrence of Aggregatibacter actinomycetemcomitans in Brazilian indians from Umutina Reservation, Mato Grosso, Brazil. J Appl Oral Sci, 2009. 17(5): p. 440-5. 55. Shimada, T., N. Sugano, R. Nishihara, K. Suzuki, et al., Differential effects of five Aggregatibacter actinomycetemcomitans strains on gingival epithelial cells. Oral Microbiol Immunol, 2008. 23(6): p. 455-8. 56. Lamont, R.J. and O. Yilmaz, In or out: the invasiv eness of oral bacteria. Periodontol 2000, 2002. 30: p. 61-9. 57. Xynogala, I., A. Volgina, J. M. DiRienzo, and J. Korostoff, Evaluation of the humoral immune response to the cytolethal distending toxin of Aggregatibacter actinomycetemcomitans Y4 in subjects with localized aggressive periodontitis. Oral Microbiol Immunol, 2009. 24(2): p. 116-23. 58. Fives-Taylor, P.M., D.H. Meye r, K.P. Mintz, and C. Brissette, Virulence factors of Actinobacillus actinomycetemcomitans. Periodontol 2000, 1999. 20: p. 136-67. 59. Ji, S., J.E. Shin, Y.S. Kim, J.E. Oh, et al., Toll-like receptor 2 and NALP2 mediate induction of human beta-defensins by fu sobacterium nucleatum in gingival epithelial cells. Infect Immun, 2009. 77(3): p. 1044-52. 60. Papapanou, P.N., A. Sellen, J.L. Wennstrom, and G. Dahlen, An analysis of the subgingival microflora in randomly selected subjects. Oral Microbiol Immunol, 1993. 8 (1): p. 24-9. 61. Bolstad, A.I., H.B. Jensen, and V. Bakken, Taxonomy, biology, and periodontal aspects of Fusobacterium nucleatum. Clin Microbiol Rev, 1996. 9 (1): p. 55-71. 62. Kajiya, T., T. Uemura, M. Kajiya, H. Kaname, et al., Pyogenic liver abscess related to dental disease in an immunocompetent host. Intern Med, 2008. 47(7): p. 675-8. 63. Lamont, R.J., D. Oda, R.E. Persson, and G.R. Persson, Interaction of Porphyromonas gingivalis with gingival epithelial cells main tained in culture. Oral Microbiol Immunol, 1992. 7 (6): p. 364-7. 96

PAGE 97

97 64. Mintz, K.P. and P.M. Fives-Taylor, impA, a gene coding for an inner membrane protein, influences colonial morphology of Actinobacillus actinomycetemcomitans. Infect Immun, 2000. 68(12): p. 6580-6. 65. Yilmaz, O., L. Yao, K. Maeda, T.M. Rose, et al., ATP scavenging by the intracellular pathogen Porphyro monas gingivalis inhibits P2X7-mediated host-cell apoptosis. Cell Microbiol, 2008. 10(4): p. 863-75. 66. Brennan, K., G. Offiah, E.A. McSherry, and A.M. Hopkins, Tight junctions: a barrier to the initiation and progression of breast cancer? J Biomed Biotechnol, 2010. 2010 : p. 460607.

PAGE 98

BIOGRAPHICAL SKETCH Brittany Dickinson is a Florida native, born in Hollywood, FL, and raised in Ft. Lauderdale, FL. She began her college career in August 2001, attending classes at St. Thomas University as a high-school student. In May 2008, she received a Bachelor of Science degree in microbiology and molecular biology from the Univ ersity of Central Florida. She also completed a minor in busi ness administration, which was c ompleted at both the University of Central Flori da and the London School of Economics. Most recently she has received a Master of Medi cal Sciences degree from the University of Florida while concentrating her research in the Oral Biology Department. 98