<%BANNER%>

Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2011-12-31.

Permanent Link: http://ufdc.ufl.edu/UFE0041316/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2011-12-31.
Physical Description: Book
Language: english
Creator: Riewe, Sally
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: 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

Statement of Responsibility: by Sally Riewe.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Lamont, Richard J.
Electronic Access: INACCESSIBLE UNTIL 2011-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041316:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041316/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2011-12-31.
Physical Description: Book
Language: english
Creator: Riewe, Sally
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: 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

Statement of Responsibility: by Sally Riewe.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Lamont, Richard J.
Electronic Access: INACCESSIBLE UNTIL 2011-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041316:00001


This item has the following downloads:


Full Text

PAGE 1

1 INTERACTIONS OF P orphyromonas gingivalis WITH HUMAN PLACENTAL CELLS AND THE IMPLICATIONS FOR PRETERM BIRTH By SALLY DIANA RIEWE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Sally Riewe

PAGE 3

3 To my family

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my family for all of their support, m y parents for all of the effort they have made in providing me opportunities m y sister for her enthusiasm. I would like to thank Bret for his patience and support I would like to thank Kallie for staying up for many late nights of homework. I appreciate the support from my advisory committee. I would like to thank Dr. Lamont for his guidance and for giving me this opportunity. I would like to thank Dr. Brown for all of the things he has taught me over the past few years. I wou ld also like t o thank Dr. Handfield for his knowledge and assistance and Dr. Peck for helping me to get the most out of this experience. I would also like to acknowledge Dr. Jeff Mans for his advice regarding microarrays I would like to thank everyone i n the Lamont lab for working with me. I would especially like to thank Brian for always being willing to help me and Karen for all of her assistance. I would like to thank Brittany for cheering me up when I needed it. I would like to thank Joyce Conners for all of the time she has spent aiding me. I would like to acknowledge the program directors Rolf Renne and Richard Snyder. I am grateful the department of oral biology for all of the resources and motivation it has provided. I would also like to say that I am grateful for the grants NIDCR DE11111, DE16715.

PAGE 5

5 TABLE OF CONTENTS page 1 ACKNOWLEDGMENTS .................................................................................................................... 4 LIST OF FIGURES .............................................................................................................................. 7 ABSTRACT .......................................................................................................................................... 9 1 INTRODUCTION .......................................................................................................................... 11 Preterm Birth ............................................................................................................................... 11 The Oral Cavity ........................................................................................................................... 12 Effects on Pregnancy .................................................................................................................. 13 Experimental Model .................................................................................................................... 15 2 MICROARRAY ............................................................................................................................. 16 Introduction ................................................................................................................................. 16 Methods ....................................................................................................................................... 16 Results .......................................................................................................................................... 18 Discussion .................................................................................................................................... 19 3 EFFECTOR MOLECULES OF Porphyromonas gingivalis ....................................................... 29 Introduction ................................................................................................................................. 29 Methods ....................................................................................................................................... 30 Confocal Laser Scanning Microscopy................................................................................ 30 Fluorescence Assays ............................................................................................................ 31 Results .......................................................................................................................................... 32 Confocal Laser Scanning Microscopy................................................................................ 32 Fluorescence Assays ............................................................................................................ 32 Discussion .................................................................................................................................... 33 4 PATHWAYS .................................................................................................................................. 40 Introduction ................................................................................................................................. 40 Methods ....................................................................................................................................... 41 Experimental Conditions ..................................................................................................... 41 Preparation of Lysates ......................................................................................................... 41 Western Bl otting .................................................................................................................. 42 Image Analysis ..................................................................................................................... 42 Results .......................................................................................................................................... 43 Discussion .................................................................................................................................... 44 5 CYTOKINES .................................................................................................................................. 54

PAGE 6

6 Introduction ................................................................................................................................. 54 Methods ....................................................................................................................................... 55 Results .......................................................................................................................................... 56 Discussion .................................................................................................................................... 57 6 CONCLUSIONS ............................................................................................................................ 62 LIST OF REFERENCES ................................................................................................................... 67 BIOGRAPHICAL SKETCH ............................................................................................................. 72

PAGE 7

7 LIST OF FIGURES Figure page 2 1 Divergence of HTR 8/SVneo cells co-cultured with P. gingivalis, F. nucleatum or no bacteria. ........................................................................................................................... 25 2 2 Pathway express figure showing the impact of P. gingivalis 33277 int eraction with HTR 8/SVneo cells ................................................................................................................ 27 2 3 Pathway express figure showing the impact of P. gingivalis 33277 interaction with HTR 8/SVneo cells ................................................................................................................ 28 3 1 Confocal image showing P. gingivalis after 2 hour co-culture with HTR 8/SVneo cells. ..................................................................................................................................... 35 3 2 Confocal image showing xz -slice show the x-z projection of an optical slice through the trophoblasts to clarify that Pg bacteria can be found within the cells10 ...................... 35 3 3 Confocal image showing fimA mutant P. gingivalis after 2 hour co-cu lture with HTR 8/SVneo cells ................................................................................................................ 36 3 4 Confocal image shows HTR 8/SVneo cells not exposed to bacteria. ............................. 37 3 5 Total fluorescence at 490nm of AlexaFluor labeled P. gingivalis after 2 hour infection of HTR 8/SVneo cells. ......................................................................................... 38 3 6 Total fluorescence at 490nm of AlexaFluor labeled P. gingivalis attached to HTR 8/SVneo cells after 2 hour co-culture. ................................................................................ 39 4 1 Western immunoblots of HTR 8/SVneo cells infected with wild type P. gingivalis, fimA mutant P. gingivalis and uninfected controls ............................................................. 46 4 2 actin band intensity. ......................................................................................................................... 46 4 3 Results of densitometric analyses of the ratio of phosphoactin band intensity. ............................................................................................................... 47 4 4 Results of densitometric analyses of the ratio of phospho-p38 band intensity to total p38 band intensity. ................................................................................................................. 47 4 5 Mean ( the standard deviation) of phosphorylated p38 over total p38 for two sets of blots. ...................................................................................................................................... 48 4 6 Western immunoblots of HTR 8/SVneo cells infected with wild type P. gingivalis, fimA mutant P. gingivalis and uninfected controls. ............................................................ 49

PAGE 8

8 4 7 Graph showing the results of densitometric analyses of the ratio of total MEK3 band actin band intensity. ........................................................................................ 49 4 8 Graph showing the results of densitometric analyses of the ratio of phospho-MEK3 actin band intensity. .............................................................................. 50 4 9 Graph showing the results of densitometric analyses of the ratio of phospho-MEK3 band intensity to total MEK3 ba nd intensity. ...................................................................... 50 4 10 Mean ( the standard deviation) of phosphorylated MEK3 over total MEK3 for two sets of blots. ............................................................................................................................ 51 4 11 Western immunoblots of HTR 8/SVneo cells infected with wild type P. gingivalis, fimA mutant P. gingivalis and uninfected controls ............................................................. 52 4 12 actin band intens ity. .................................................................................................................................. 52 4 13 Mean ( the standard error) of transcription factor Max for three sets of blots. ................ 53 5 1 Graph depicting measured concentration of IL 1beta present in cell culture supernanent. .......................................................................................................................... 59 5 2 Graph depicting measured concentration of IL 6 present in cell culture supernanent. ... 60 5 3 Graph depicting measured concentration of IL 8 present in HTR 8/ SVNEOcell culture supernanent. .............................................................................................................. 61 A 1 Pathways most affected by co -culture with P. gingivalis relative to the uninfected control (p<0.001). ................................................................................................................... 65

PAGE 9

9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science INTERACTIONS OF PORP HYROMONAS GINGIVALIS WITH HUMAN PLACENTAL CELLS AND THE IMPLIC ATIONS FOR PRETERM B IRTH By Sally Diana Riewe December 2009 Chair: Richard J. Lamont Major: Medical Sciences Objective: Preterm birth is the leading cause of infant morbidity and mortality and affects 12.5% of births in the United States. Epidemiological studies have indicated periodontal disease to be a risk factor for preterm birth. Mechanistically, the cause could be oral -hematogenous spread of Porphyr omonas gingivalis to the gestational tissues inducing inflammation. He nce, the objective was to examine the interactions between P. gingivalis and placental tissues. Methods: Human placental trophoblasts ( HTR 8/ SVneo cell line) were co -cultured with P. gingivalis (or a control of no bacteria) for 2 hours at 37C at a multiplicity of infection of 200 bacteria per cell. For microarrays t otal RNA was extracted, purified, quantified, reverse transcribed, and used to probe Affymetrix HG -U133 Plus 2.0 human m icroarrays. Differential expression patterns were analyzed with bio informatics, statistical, and gene ontology tools and databases. Images of infected cells were taken with the confocal laser scanning microscope. Western blotting was used to analyze t he phosphorylation state of factor involved the p38 MAPK pathway. ELISAs were performed on culture supernatants and lysates to verify cytokine expression

PAGE 10

10 Results: Among the pathways and factors up-regulated at the mRNA level by exposure to P. gingivalis were pathways for cytokine signaling Confocal images showed attachment of P. gingivalis to the HTR 8/SVNeo cells. Western blots showed activation of the p38 MAPK pathway. ELISA s showed expression of cytokines IL 8 and IL 6 were increased after exposure to P. gingivalis relative to the control. Conclusions: Exposure of HTR 8/SVneo human trophoblast cells to P gingivalis leads to the up-regulation and increased production of inflammatory cytokines and induction of pathways leading to inflammation.

PAGE 11

11 CHAPTER 1 INTRODUCTION Preterm Birth Preterm birth is defined as birth occurring before the 37th week of gestation. It is the leading cause of i nfant morbidity and mortality and affects 12.5% of births in the United States. Preterm babies are at increased risk of neurological, respiratory and gastrointestinal problems (Goldenberg et al. 2008) Preterm birth is either induced or spontaneous. Labor may be induced, or a caesarean section performed if it becomes necessary for the health of the mother or baby. However the majority of preterm births occur following spontaneous preterm labor with or without preterm premature rupture of the membranes Spontaneous preterm labor has many causes with infection being present about 50% of the time (Michalowicz & Durand, 2007) R outes of uterine infection include ascending from the vagina and cer vix through the genital tract h ematogen ous dissemination from the maternal bloodstream through the placenta retrograde seeding from the peritoneal cavity, and accidental introduction during invasive procedures (Romero et al. 2006) The placenta functions as an interface between mother and fetus, allowing for the exchange of nutrients and waste products. The placenta is formed from the trophoblast layer of the bl astocyst which forms 4 days after fertilization The blastocyst is a thin -walled, hollow, fluid filled structure made up of four componen ts. These components are a glycoprotein membrane, the trophoblast which forms the placenta, the embryoblast which will form the embryo, and a fluid filled cavity. (Blackburn, 2007) A normal pregnancy lasts 40 weeks. Parturition can be divi ded into four phases, quiescence, activation of the fetal hypothalamic pituitary adrenal (HPA) axis, stimulation by corticotropin releasing hormone (CRH), and uterine contraction. CRH plays a major role in

PAGE 12

12 determi ning the length of gestation. (Behrman & Butler, 2007) (McLean & Smith, 1999) HPA activation and increased expression of CRH lead to progesterone withdrawal and estrogen activation. This leads to cervical ripening, uterine contractility and eventually labor. In that CRH is secreted from the placenta a bnormalities of placental CRH may be involved in the pathogenesis of preterm labor (McLean & Smith, 1999) Glucocorticoids stimulate production of CRH in the placenta, despite being inhibitors of CRH in the hypothalamus. The lack of ner ves in the placenta implicates circulating cytokines as a likely cause of increased CRH production. The Oral Cavity The oral cavity is populated by over seven hundred species of bacteria (Thomas & Nakaishi, 2006) These bacteria range from beneficial commensals to opportunistic pathogens, to overt pathogens. Beneficial commensal bacteria such as Streptococcus sanguinis and Streptococcus gordonii can supply nutrients, regulate epithelial development, and contribute to the maturation of the immune system. Beneficial bacteria he lp to control pathogenic bacteria by competing with them for limited nutrients and space. Opportunistic pathogens such as Fusobacterium nucleatum may function as commensal bacteria until environmental conditions allow them to overcome host defenses and initiate disease (Handfield et al. 2008) An overtly pathogenic species of oral bacteria is Porphyromonas gingivalis P. gingivalis is a non-motile, Gram -negative anaerobic asacch a rolytic cocc o bacillus found in mature biofilms primarily in the subgingival crevice P. gingivalis is a late colonizer of the oral cavity, attaching to other bacteria such as Streptococcus oralis, S. sanguinis and S gordonii P. gingivalis can bind to other late colonizers, such as F nucleatum whic h may provide stability and aid in nutrient acquisition (Lamont & Jenkinson, 2000) Study of P. gingivalis has revealed an array of virulence factors. Virulence factors of P. gingivalis include proteolytic enzymes,

PAGE 13

13 hemagglutinins and fimbriae. P. gingivalis produces at least five hemagglutinating molecules (Lamont & Jenkinson, 1998) Hemagglutinins target the cells towards hemin which is a necessary source of iron. Fimbrilin is the monomeric subunit of the major fimbriae and is e ncoded by the fimA gene (Lamont & Jenkinson, 1998) P. gingivalis is the major etiolog ic agent of periodontal disease (Lamont & Jenkinson, 1998) Periodontal disease is a bacterially induced chronic inflammatory condition consisting of the diseases gingivitis and periodontitis. Gingivitis is reversible inflammation of the gingiva l tissues around the teeth. Periodontitis results in destruction of the tissues and structures supporting the teeth including the alveolar bone, periodontal ligament and connective tissue, eventually leading to the exfoliation of the teeth. It occurs in 5% to 40% of pregnant women (Boggess et al. 2005, Lieff et al. 2004) Effects on Pregnancy Adverse pregnancy outcomes including preterm birth have been linked to periodontal disease A review by Boggess and Edelstein concluded that maternal periodontal disease in a risk factor for preterm birth (Bogge ss & Edelstein, 2006) Pitiphat et al. evaluated periodontitis in relation to preterm birth among a cohort of medically insured, middle -class women. The authors concluded that periodontitis was an independent risk factor for adverse pregnancy outcomes among the women studied (Pitiphat et al. 2008) Animal models have been used to study infection by P. gingivalis of pregnant rodents. Lin et al found that P. gingivalis infection enhanced fetal growth res triction in mice (Lin et al. 2003) P. gingivalis was shown to in vade both maternal and fetal tissues by Belanger et al (Belanger et al. 2008) Collins et al. examined the effects of a non-disseminating, localized subcutaneous P. gingivalis infection in the golden hamste r and concluded that their data su gg est

PAGE 14

14 that infection by Gram negative periodontal pathogens can result in adverse pregnancy outcomes (Collins et al. 1994) A possible mechanism for periodontal disease to cause preterm birth would be P. gingivalis entering the m aternal blood steam and traveling to the placenta where it would induce inflamma tion. P. gingivalis has been isolated from the blood following scaling and root planning (Lafaurie et al. 2007) P. gingivalis has also been isolated from the human placenta A study examining periopathogenic microorganisms in placentas of women with pre eclampsia found that five of the 16 placental samples from women with preeclampsia and two of the placental samples from healthy control group of women were positive for P. gingivalis with the number of bacterial cells per sample ranging from zero to 3,100 (Barak et al. 2007) Additionally, Leon et al studied women with either preterm premature rupture of the membranes or premature labor and found that eight of 26 study participants had microbial invasion of the amniotic cavity by P. gingivalis as detected by PCR (Leon et al. 2007) There is epidemiological evidence linking periodontal disease with preterm birth. A review by Xiong et al. concluded that the studies they reviewed suggest ed that periodontal disease was a risk factor for preterm birth (Xiong et al. 2006) A study published in the Journal of th e American Dental Association looked at 1313 pregnant women recruited in Birmingham, Alabama and after adjusting for smoking, race, maternal age, and parity, found an association between the presence of severe or general periodontal disease at 21 to 24 wee ks gestation and preterm birth (Jeffcoat et al. 2001) A 5 year prospective study published in 2001 by Off enba cher et al. reported clinical data from 812 deli veries from a cohort study. A fter adjusting for race, parity, and baby gender, they concluded that the data provided evidence that maternal periodontal disease and incident

PAGE 15

15 progression are significant co ntributors to risk for preterm delivery (Offenbacher et al. 2001) Looking at markers of periodontal infection and preterm birth, Jarjoura et al. also concluded that periodontitis is independently associated with preterm birth (Jarjoura et al. 2005) In 2002 Lopez et al. published the result s of a study in which they investigated whether treatment of periodontal disease reduced that risk of preterm low birth weight. There data showed that the women who did not receive treatment for periodontal disease experienced a more than three -fold increase in the risk for preterm birth. (Lopez et al. 2002) Experimental Model Graham et al. transfected first trimester human trophoblast cells, designated HTR 8/SVn eo with a plasmid containing the gene for the simian virus 40 (SV40) large T antigen (Tag). The result was the HTR 8/ SVneo cell line. These cells are phenotypically similar to the ir parent strain but able to be cultured in vitro for prolonged periods of time. Unlike choriocarcinoma cells, t hey respond to the migration and invasion signals of TGF -beta in a manner similar to primary cells (Graham et al. 1993) These characteristics make them particularly suitable for studies of placental gene expression which involve cell cultur e for extended periods of time and thus a good model for this study.

PAGE 16

16 CHAP TER 2 MICROARRAY Introduction Specific Aim 1 was to perform microarray analysis on HTR 8/SV n eo cells exposed to P. gingivalis Interactions between a pathogen and its host cell are complex and multifaceted. Microarrays are useful in studying these inte ractions because they allow gene expression to be studied on a global scale (Handfield et al., 2008) Through microarray analysis genes from pathogens may be identified that are virulence factors involved in pathoge nicity (Kato -Maeda et al. 2001) The effect of these virulence genes on the host cell can be studied by monitoring the response of the host cell such as the differential regulation of various pathways, after infection (Cummings & Relman, 2000) For this study, m icroarrays were performed to analyze the effec s t of P. gingivalis on HTR 8/SV n eo cells. The effects of F. nucleatum were also analyzed to identify which effects were specific to P. gingivalis F. n ucleatum was chosen as a control becaus e like P. gingivalis it is a gram -negative an a erobe found in subgingival biofilms However, unlike P. gingivalis F. nucleatum is an opportunistic commensal found in both the presence and absence of disease It was hypothesized that P. gingivalis would alter gene expression in the HTR -8/SV n eo cells in a manner that could be hostile to the fetus and thus induce preterm labor. Methods HTR 8/SVneo cells were grown to 90% confluency in T 25 flasks in RPMI 1640 media (Sigma) supplemented with 5% fetal bovine serum at 37C in the presence of 5% CO2. P. gingivalis ATCC 33277 and F. nucleatum ATCC 25586 w ere grown anaerobic ally at 37C in trypticase -soy broth medium supplemented with yeast extract (1mg/ml), hemin (5u g/ml), and m enadione (1ug/ml). Bacteria were harvested in the log phase c ells were co -cultured with P.

PAGE 17

17 gingivalis F. nucleatum or a control of no bacteria. Co -cultures were carried out in quadruplicate. The cells were lysed with Trizol (Invitrogen, Carlsbad, CA ) prior to RNA extraction. Total RNA was extracted from cells, D NAse treated, purified, and quantified. cDNA was synthesized according to a standardized protocol purified and used as a template for labeled cRNA synthesis. In vitro transcription of cRNA was performed to incorporate biotinylated nucleotides by using a B ioArray high-yeild RNA transcript labeling kit ( Enzo Life Sciences, Farmingdale NY) cRNA was subsequently fragmented and hybrid ized on Affymetrix HG -U133 Plus 2.0 (Affymetrix) Each sample was studied in parallel, and the samples were not pooled. The microarrays were hybridized for 16 hours at 45C and then scanned with an Affymetrix GeneChip 3000 scanner. To begin analysis, an expression filter was applied by removing probe sets that were flagged as absent on all arrays. The signal intensities of the remaining probe sets were ranked according to the coefficient of variation and the 50% of the data with the greatest coefficient of variation was then normalized to a mean of 0 and a standard deviation of 1 H ierarchical cluster analysis was performed on the variancenormalized data set and viewed with GeneCluster and TreeView ( Michael Eisen, Stanford University) as shown in Figure 2 1. Following the initial expression filter a supervised analysis was performed to investigate differences in gene regulation among experimental conditions using BRB Array Tools (R. Simon and A. Peng Lam, National Cancer Institute, Rockville, MD). The purpose of t his supervised analysis was to identify genes differentially expressed among the treatment classes: cells co cultured with P. gin givalis cells co -cultured with F. nucleatum or cells not exposed to bacteria. The ability of gene identification to predict treatment class was assessed by a l eave -one out

PAGE 18

18 cross -validation u sing several prediction models including compound covariate predictor, nearest -neighbor predictor, and s upport vector machine predictor. Monte Carlo simulations were also used with 2000 permutations of the data set. Differential expression patterns were analyzed with bio -informatics, statistical, and gene ontology tools and databases Results The heat map, Figure 2 1, shows the 2045 genes whose regulation was significantly altered at p<0.001 among the three classes The figures in Appendix A show the pathways most affected The most affected pathways in cells infected with P. gingivalis relative to uninfected cells include the cell cycle, MAPK signaling, ap optosis, cell adhesion molecules cytokine cytokine receptor signaling, and TGF -beta signaling. Pathway Express (available at http://vortex.cs.wayne.edu/projects.htm ) was used to populate gene ontology trees using genes found to be up or down regulated by a t least 1.25 times relative to their expression in the uninfectecd control cells at p<0.01. Genes that were not differentially expressed by a factor of at least 1. 25 were filtered out because their inclusion may have lead to falsely hig h confidence in the effect of microbial challenge on the pathway. Among the pathways found to be differentially regulated Pathway Express analysis revealed key factors that were involved in this differential regulation In the cell cycle and apoptosis pathways these include TNFR and CycD (up regulated), CycE (downregulated), and caspase 3/7 and CDK4 /6 (not differentially regulated). In the TGF -beta and cell adhesion molecule pathways important factors include that were down -regulated in cells co -cultured with P. gingiv a lis relative to cells not exposed to bacteria include MHC 1, CLDN, OCLN, and CDH1/2 Factors that were up regulated include Smad2 and Smad3

PAGE 19

19 Cytokines and f actors relating to MAPK that were up regulated by HTR 8 cells exposed to P. gingivalis relative to cells not exposed to bacteria include p38, MKK3, and transcription factor Max of the mit ogen activated kinase pathway as well as c ytokines IL -1beta and IL 8 (Figure 2 2) Cy tokines Il 1beta and IL 8, as well as MKK3 were also upregulated in the cells co -cultured with P. gingivalis relative to the cells co -cultured with F. nucleatum (Figure 2 3) Surprisingly, the macrophage attracting CXCL2 was found to be down regulated by exposure to P. gingivalis Down regulation of CXCL2 appears to be in contrast to the upregulation of pro inflammatory cytokines esp ecially chemoattractant IL 8. T his down regulation may be a n attempt by P. gingivalis to evade host immune responses. Discussion Study objectives commonly encountered in microarray experiments are class comparison, class prediction, and class discovery (Simon et al. 2003) In class comparison the classes are predefined and they are defined independently of the expression profiles. Class prediction also involves predefined classes but the classes are based on gene expression. Class discovery does not involve predefined classes (Simon et al., 2003) Analysis can be considered supervised if distinctions are made among the samples based on predefined class labels or unsupervised if no information about sample grouping is used (S imon, 2003) Cluster analysis, an unsupervised method of analysis, is useful in class discovery. Supervised methods are appropriate for class prediction studies, although over -fitting the predictor can be a major limitation. Over -fitting results from fitting the model to random variations in the data that do not represent true relationships. (Simon et al., 2003) A class prediction models performance can be measured by using a validation set of data that is independent from the training set used to create the model. Cross -validation uses data efficiently as only a small number of specimens are used for the validation set (Simon, 2008) In

PAGE 20

20 leave -one out cross -validation one specimen is left out at a time and classified by the predictor built by all of the remaining specimens. It is important that all components of class prediction be cro ss -validated including selection of informative genes, computation of weights for selected informative genes, and creation of a prediction rule. (Simon et al., 2003) Cross -validation is not sufficient for assessing the significance of a class result. To a ssess significance, permutation methods can be used to calculate the probability of producing a cross validated error rate as small as observed given no association between class membership and expression profiles (Simon et al., 2003) A more practical method than examining every possible permutation is with a Monte Carlo method. This involves examining randomly selected permutations and using the proportion that have the same or fewer misclassifications to estimate the achieved significance level (Radmacher et al. 2002) For the microarrays performed in this study, it was hypothesized that P. gingivalis would alter gene expression in the HTR 8/SVneo cells in a manner that could be hostile to the fetus and induce preterm labor. The results of the microarray revealed that exposure of HTR 8/SVneo cells to P. gingivalis caused, in addition to the up regulat ion of the p38 MAP kinase pathway, an up -regulation of pro -inflammatory cytokines. This finding is significant because inflammation and production of cytokines are known to b e associated with preterm birth (Romero et al., 2006, Keelan et al. 2003) There have been previous studies usi ng microarrays to evaluate cellular host response s to P. gingivalis One such study by Handfield et al. used transcriptional profiling bioinfor matics, statistical and ontology tools to explore genes and pathways modulated upon interaction of P. gingivali s or the Gram -negative periodontal pathogen Aggregatibacter actinomycetemcomitans with human immortalized gingival keratinocytes A s predicted from the clinical differences in

PAGE 21

21 the bacteria l species, transcriptional response in the HIGK cells differed greatly (Handfield et al. 2005) Serum and glucocorticoid-inducible kinase (SGK) is related to Akt and negatively regulates transcription factor FOXO3A which participates in apoptosis. Consistent with its antiapoptotic phenotype, P. gingivalis was found Handfield et al. to activate SG K Another factor that plays a role in cell survival and proliferation found to be up regulate d was cM YC cMYC can repress transcription of GADD45A, a pro apoptotic transcription factor (Handfield et al. 2005) The same study showed that fimbriae -deficient mutant YPF1 had a transcriptional pattern strikingly divergent from the parental strain, indicating a significant role for fimbriae in gene expression (Handfield et al., 2005) Another study evaluating the role of fimbriae fo und that human aortic endothelial cells infected with P. gingivalis strain 381 showed up -regulation of 68 genes including IL 8 and IL 6. However in cells infected with the isogenic fimA mutant DPG3 only four of the 68 genes were upregulated. The authors concluded that fimbriae -mediated invasion up regulates inflammatory gene expression in human aortic endothelial cells. (Chou et al. 2005) Milward et al. looked at the effects of P. gingivalis and F. nucleatum on gene expression in o ral epith elial cells. They found that F. nucleatum induced a greater number of gene expression changes than did P. gingivalis They also found many cytokines to be upregulated which is consistent with the find ings of the current study. Cytokines found to be upregulated after exposure to P. gingivalis in both the Milward study and the current study were IL 8, IL 1alpha, and IL 1beta (Milward et al. 2007)

PAGE 22

22 In another study Zhou and Amar examined gene expression profiles of human macrophages exposed to P. gingivalis or purified LPS or FimA from P. gingivalis They concluded that P. gingivalis as well as its LPS and FimA, induce NF kB -containing genes through TLR2 o r TLR7 MyD88 p38 MAPK pathway (Zhou & Amar, 2007) The current study found that the p 38 MAPK pathway is also upregulated by P. gingivalis in the HTR 8/SVneo cells. A microarray study by Ohno et al. was done in bone marrow derived mouse ST2 cells that showed some pre -osteoblastic characteristics The stu dy looked at global cellular gene expression profiles in the ST2 cells after exposure to P. gingivalis The investigators concluded that P. gingivalis infection does induce gene expression of a wide variety of proinflammatory proteins in stromal cells (Ohno et al. 2006) This was signifi cant to their research because inflammation causes alveolar bone resorption and relates to the current study as well, by showing that P. gingivalis promotes inflammation in another cell type In a subsequent study, the authors investigated the signaling pathways involved in those proinflammatory responses. Unfortunately, p38 MAPK was found to be spontaneously phosphorylated in the unstimulated cells so the effect of P. gingivalis infection could not be evaluated (Ohno et al. 2008) In the current study pathways that were differentially regulated in cells co -cultured in P. gingivalis relative to control cells include MAPK signaling, apoptosis, cell cycle adhesion molecules, and TGF -beta signalin g. A recent publication by Inaba et al. found that P. gingivalis invasion of HTR 8/SVneo cells leads to G1 arrest and apoptosis. The apoptosis pathway is programmed cell death characterized by nuclear condensation, cell shrinkage, and DNA fragmentation. Caspases are central regulators of apoptosis. Activation of TNFR leads to the activation of caspase 8 and caspse 10, eventually leading to apoptosis. The microarray results in

PAGE 23

23 the present study show an upregulation of TNFR but no differential regulation of TNF, caspase 3, caspase 7 caspase 9 caspase 10, or Bcl 2 Inaba et al. also found Bcl 2 expression unchanged, however, the authors did find increa sed expression of caspase 3/7 and caspase 9 after challenge with P. gingivalis in a fluorescent substrate assay. (Inaba et al. 2009) In the cell cycle, the primary G 1/ S checkpoint controls the transition from the gap phase (G1) to the DNA synthesis phase (S). Inaba et al fou nd that cells challenged with P. gingivalis showed decreased levels of cyclin D CDK4, CD K6, and increased levels of cyclin E for 24 to 72 hours (Inaba et al., 2009) The microa rray data from the current study showed HTR 8 cells co cultured with P. gingi valis increased expression of CycD decreased expression of CycE, and d id not differentially express CDK4 or CDK6. Possible reasons for this discrepency include that Inaba et al. used densiometric analysis of immunoblots while the current study used microarray ana lysis, and that Inaba et al. used time points of 24, 48, and 72 hours while the array data was collected from cells 2 hours after initial exposure to P. gingivalis Another group of molecules shown to be affected by P. gingivalis in the array data was Cell adhesion molecules (CAMs) Many cell adhesion m olecules were down regulated in the cells exposed to P. gingivalis including M HC 1, CL DN, OCLN, and CDH1/2. A possible cause for this woul d be that the cells were down regulating these molecules to fight invasion by P. gingivalis Another consideration is that immune molecules are already perturbed in trying to balance destroying pathogens with not rejecting the healthy fetal or maternal tissues. The transforming growth factor beta ( TGF -beta ) signaling pathway functions in a wide ra nge of biological systems, playing a role in cell growth, differentiation, and development. Signal initiation involves oligomerization of receptor kinases and phosphorylation of cytoplasmic signaling molecules Smad2 and Smad3 The cells exposed to P. gin givalis in the

PAGE 24

24 current microar ray study show an up regulation of TGF -beta receptor type 1, and Smad2/3, but no differential regulation of TG F beta. Interestingly, p15 is also up -regulated which leads to G1 arrest. The mitogen activated protein kinases (MA PK) are involved in many host cell signaling pathways including mitotic response to growth factors, cytokine responses, cytoskeletal reorganization, and stress responses. F actors relating to MAPK that were up regulated by HTR 8 cells exposed to P. gingivalis relative to cells not exposed to bacteria include p38, MKK3, and transcription factor Max (Figure 2 2) A review of the literature indicated the importance of cytokines and inflammation in preterm birth (Goldenberg et al., 2008) D ue to its association with cytokine production and inflammation, the p38 pathway of the MAPK pathways was chosen for further analysis later in the study.

PAGE 25

25 Figure 2 1. Divergence of HTR 8/SVneo cells co -cultured with P. gingivalis, F. nucleatum or no bacteria. T his heat map and was constructed from 2045 probe sets differentially expressed between experimental classes at the significance level of P< 0.001. Probe set signal intensities were variance normalized, mean centered across samples, and subjected to hierarchical cluster analysis. Average linkage clustering by un -centered correlation was performed for genes and samples. The degree of similari ty between the transcriptional profiles of each sample is expressed by Pearsons correlation coefficient distance metric which is a measure of dependence obtained by dividing the covariance by the product of the standard deviations according to the adja cent scale. The expression state of each data point is represented as standard deviations from the mean expression level for that gene in all samples. Red indicates a relative increase, green indicates a relative decrease, and black indicates no relative change of mRNA transcripts for a given genome.

PAGE 26

26

PAGE 27

27 F igure 2 2 Pathway express figure showing the impact of P. gingivalis 33277 interaction with HTR 8/SVneo cells relative to HTR 8/SVneo cells not exposed to bacteria, on the p38 MAPK pathway and cytokine -cytokine receptor interactions Red terms including p38, MKK3, IL 8 and IL 1beta are transcriptionally up regulated. T he blue term is down regulated and t erms in green were not significa ntly modulated compared to baseline conditions Modulated genes were significant at the P<0.01 threshold and were changed by at least a 1.25 fol d magnitude.

PAGE 28

28 Figure 2 3. Pathway express figure showing the impact of P. gingivalis interaction with HTR 8/SVneo cells relative to HTR 8/SVneo cells co -cultured with F. nucleatum upon the p38 MAPK pathway and cytokine -cytokine receptor interactions. Red terms including are transcriptionally up-regulated. The blue term is down regulated and terms in green were not significantly modulated compared to baseline conditions. Modulated genes were significant at the P<0.01 threshold and were changed by at least a 1.25 fold magnitude.

PAGE 29

29 CHAPTER 3 EFFECTOR MOLECULES O F Porphyromonas g ingivalis Introduction Specific Aim 2 was to i dentify effector molecules of P. gingivalis Host -microbe interactions are inherently complex. In the previous chapter, cellular responses of HTR 8/SVneo cells to P. gingivalis were evaluated on a global scale using human DNA microarrays Study of P. gingivalis has revealed an array of virulence factors including proteolytic enzymes, hemagglutinins and fimbriae. In this chapter an individual virulence factor of P. gingivalis the major fimbriae, is examined to elucidate its affect on the interaction of P. gingivalis with HTR 8/SVneo cells Fimbrilin is the monomeric subunit of the major fimbriae and is encoded by the fimA gene (Lamont & Jenkinson, 1998) The fimA gene is regulated by environmental factors. Expression increases as hemin concentration increases and as temper ature decreases from 39 to 34C (Xie et al. 1997) P. gingivalis appears to express high amount of fimbrilin to facilitate adherence and invasion in the early stages of colonization and repress fimbrilin later to reduce the host immune response. (Xie & Lamont, 1999) The fimA gene is r egulated at both the transcriptional and posttranscriptional level (Xie et al. 2004) At the transcriptional level, a 70 like promoter region caries out basal level transcription while cis acting regulatory elements are required for maximal transcription of the fimA gene. 70 recognized sequences expressed include 10, 35, and UP elements. AT rich upstream regulator sequences are necessary for full expression of fimA (Xie & Lamont, 1999) Trans acting components fimbrilin arginine ( Rgp ), and lysine ( Kgp ) bind to the upstream region of the fimA promoter and are necessary for maximal fimA transcription (Xie et al. 2000)

PAGE 30

30 Fimbriae mediate attachment to cells and invasion of previo usly studied epithelial cells. To examine the effects of fimbriae on P. gingivalis infection of HTR 8/S Vneo cells a fimbriaedeficient mutant of P. gingivalis 33277, designated YPF1, was utilized. YPF1 was created by homologous recombination between a P. gingivalis ATCC 33277 chromosomal DNA and a suicide plasmid carrying an internal fragment of the fimA gene The f ragment of the fimA gene in the suicide plasmid was 0.65 kb in len gth, cut from restriction sites Pst I to Hinc II, downstream from the transcriptional regulatory regions and start site (Dickinson et al. 1988, Love et al. 2000) As a result of this insertional inactivation of the fimA gene the YPF1 strain is unable to produce FimA protein and lacks the major fimbriae (Love et al., 2000) In the current study t wo different assays were performed to elucidate the differences between the wild type and fimA mutant bacteria when infecting the cells. The first assay, in which formalin fixing was done subsequent to the co-culture, showed the total amount of bacteri a including those attached to the cells and those that had invaded the cells. In the second assay, the cells were formalin fixed before co -culture with bacteria to prevent bacterial invasion of the cells. Thus the second assay compared only the ability of the bacteria to attach to the cells. It was hypothesized that P. gingivalis would be able to infect HRT 8 /SVneo cells. Furthermore, since fimbriae were necessary for P. gingivalis infection of other host cell types, it was expected that the fi m briae -deficient mutant YPF1 would be less effective at infecting HTR 8/SVneo cells Methods Confocal Laser Scanning Microscopy HTR 8/SVneo c ells passage 106, were seeded onto glass cover -slips that had been acid washed and coated with poly L -lysine. C ells were cultured in RPMI 1640 media (Sigma),

PAGE 31

31 supplemented with 5% fetal bovine serum, at 37 C in 5% carbon dioxide. P. gingivalis 33277 and YPF 1 fimA mutant were grown anaerobic ally at 37C in trypticase -soy medium supplemented with yeast extract (1mg/ml), hemin (5ug/ml), and menadione (1ug/ml). Erythromycin (15ug/ml) was added to media for YPF 1. Cells were co -cultured with bacteria harvested in the log phase at a multiplicity of infection of 200 bacteria per cell for 2 hours at 37C in the pres ence of 5% CO2, fixed in 10% formalin for 30 minutes, permea bilized in 0.1% Triton -X 100 for 5 minutes, and blocked in 1% BSA. They were then exposed to anti 33277 antibody from rabbits for 45 minutes rinsed, and exposed to Alexa -F luor 488 (Invitrogen) anti rabbit IgG Texas Red phal l oi din stain (Invitrogen) and DRAQ5 DNA probe. Images were acquired using a spinning disk confocal system including a CSU10 Yokagowa confocal scan head, a Roper Cascade II EMCCD 512b camera, a Leica DMIRB microscope, and the open source software package Micro -manager ( http://www.micro manager.org/ ). Analysis of images was done using Imaris 5.0.1 (Bi tplane AG; Zurich, Switzerland) software. Fluorescence As says HTR 8/SVneo cells were cultured in RPMI 1640 media (Sigma), supplemented with 5% fetal bovine serum, at 37C in 5% carbon dioxide. P. gingivalis 33277 and YPF 1 fimA mutant were grown anaerobic ally at 37C in trypticase -soy medium supplemented with ye ast extract (1mg/ml), hemin (5ug/ml), and menadione (1ug/ml). Erythromycin (15ug/ml) was added to media for YPF 1. Cells were plated onto black 96 -well plates and co -cultured with bacteria harvested in the log phase at a multiplicity of infection of 200 bacteria per cell for 2 hours at 37C in the presence of 5% CO2 Cells were then fixed in 10% formalin for 30 minutes, perm eabilized with 0.1% Triton -X 100 for 5 minutes, blocked in 1% BSA overnight at 4 C, exposed to anti 33277

PAGE 32

3 2 antibody, wa shed, and exposed to Alexa F luor 488 anti rabbit. The procedure was subsequently performed with the formaldehyde fixing prior to infection. Fluorescence was read by Victor3 Wallac 1420 Multi label Detection S ystem (Perkin Elmer ). Statistical analysis was done by averaging the results of each well (n=8) in a treatment group and determi ning the standard deviation. A t test appropriate for two small samples with different standard deviations was used to verify significance (McClave & Sincich, 2009) Results Confocal Laser Scanning Microscopy Cells were visualized by a confocal laser scanning microscope. Images show that the HTR 8/SVneo cells were infected with P. gingivalis As shown in F igure 3 -1 bacterial cells appear to be inside the HTR 8/SVneo cells when view from above To clarify this, the cells were also viewed from the side The X Z projection shown in Figure 3 2 also indicates that P. gingivalis were inside the cells. The cells co -cultured with the fimA mutant appear to be infected with fewer bacteria that the cells co -cultured with the wild type bacteria. Fluorescence Assays Results from the confocal microscope w ere confirmed through fluorescence assays. P. gingivalis was able to interact with HTR 8/SVn eo cells in the 2 hour co-culture. The fimA mutant was far less able to interact with the cells that the wild type bacteria. In fection of HTR 8/SVneo cells in both assays was lower in the cells co-cultured with the fimA mutant than it was in cells co -cultured with wild type P. gingivalis The assay comparing the difference in total bacterial fluorescence is similar to the assay comparing the difference in fluorescence of bacterial attachment in that both show a n approximately 5x decrease in fluorescence between wild type P. gingivalis and the fimA mutant

PAGE 33

33 Discussion Colonization of the subgingival crevice by P. gingivalis requires attachment to antecedent bacteria via fimbriae (Xie et al., 2000) Fimbriae are an important virulence factor of P. gingivali s functioning in both adhesion and invasion. F imbrilin is the monomeric subunit of the major fimbriae and is encoded by the fimA gene (Lamont & Jenkinson, 1998) Regulation of the fimA gene is affected by environmental factors such as hemin concentration and temperature (Xie & Lamont, 1999) The fimbriae -deficient mutant YPF 1 was used to elucidate the role fimA play s in attachment and invasion of HTR 8/SVneo cells. It was hypothesized based on t he ability of P. gingivalis to infect and invade epithelial cells, that P. gingivalis would als o be able to infect and possible invade HRT 8 /SVneo cells. Furthermore, since fimbriae were necessary for P. gingivalis infection of other host cell types, it was expected that the fi m briae -deficient mutant YPF1 would be less effective at attaching to, infecting or invading HTR 8/SVneo cells Results from the spinning -disk confocal microscope show that P. gingivalis was able to infect HTR 8/SVneo cells Confocal images appear to show P. gingivalis surrounded by cellular actin in both X Y and X Z projections. This provides strong, while not entirely conclusive, evidence that P. gingivalis were able to successfully invade the HTR 8/SVneo cells. Additional results from the confocal microscope show far fewer P. gingivalis present in images of HTR 8/SVneo cells co -cultured with the fimA mutant bacteria than in the images of the cells co cultured with wild type P. gingivalis This leads to the conclusion that the fimA mutant was less effective at infecting the HTR 8/SVneo cells Florescence assays performed on 96 -well plates and read using the Victor3 Wallac 1420 Multi label Detection System were consistent with the results obtained by the confocal microscope. Furthermore, the florescence assays quantified the difference in the a mount of wild

PAGE 34

34 type P. gingivalis and fimA mutant P. gingivalis A second fluorescence assay was performed comparing infection of live HTR 8/SVneo cells to the infection of HTR 8/SVneo cells fixed in formalin prior to infection Thi s second assay Figure 3 6 measures only the ability of the bacteria to attach to the surface of the HTR 8/SVneo cells. If the difference in total interaction shown in the first assay, Figure 3 5, resulted in large part from interaction with the living HTR 8/SVneo cells then the second assay would not show a similar decrease between the wild type P. gingi valis and the fimA mutant. The decrease in attachment shown in Figure 36 leads to the conclusion that the fimA mutant s decreased ability to interact wi th HTR 8/SVneo cells is due, either entirely or at leaset in large part, to its decrea s ed ability to at tach to HTR 8/SVneo cells.

PAGE 35

35 F igure 3 1 Confocal image showing P. gingivalis after 2 hour co -culture with HTR 8/SVneo cells P. gingivalis bacteria, marked with Alexa Fluor 488, are shown in green. The actin of HTR 8/SVneo cells stained with Texas Red p ha l loiden, is shown in red. Figure 3 2. Confocal image showing a 20m slice of the X Z projection of P. gingivalis after 2 hour co -culture with HTR 8/SVneo cells P. gingivalis bacteria, marked with Alexa Fluor 488, are shown in green. The actin of HTR 8/S Vneo cells, stained with Texas Red phalloiden, is shown in red.

PAGE 36

36 F igure 3 3 Confocal image showing fimA mutant P. gingivalis after 2 hour co-culture with HTR 8/SVneo cells P. gingivalis bacteria, marked with Alexa Fluor 488, are shown in green. The actin of HTR 8/SVneo cells stained with Texas Red p ha l loiden, is shown in red. The cellular nucleus, stai ned with DRAQ5, is shown in blue.

PAGE 37

37 F igure 3 4 Confocal image shows HTR 8/SVneo cells not exposed to bacteria. The actin of HTR 8/SVneo cells stained with Texas Red p hal l oiden, is shown in red.

PAGE 38

38 F igure 3 5 Total fluorescence at 490nm of AlexaFluor labeled P. gingivalis after 2 hour infection of HTR 8/SVneo cells After infection, cells were fixed in formal in and permeabilized with Triton X 100. Asteric k s indicate a significa nt difference (p<0.0 05) between the indicated cells and cells infected with wild type P. gingivalis 0 2000 4000 6000 8000 10000 12000 14000 16000 Wt FimA Ctrl *

PAGE 39

39 F igure 3 6 Total fluorescence at 490nm of AlexaFluor labeled P. gingivalis attached to HTR 8/SVneo cells after 2 hour co-culture. HTR 8/SVneo cells were fixed with formalin before exposure to bacteria and permeabilized with Triton X 100 after co-culture. Asteric k s indicate a significant difference (p<0.025) between the indicated cells and the cells infected with wild type P. gingivalis 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 Wt FimA Ctrl

PAGE 40

40 CHAPTER 4 PATHWAY S Introduction Specific Aim 3 was to characterize pathways affected by P. gingivalis infection that lead to inflammation. The microarray study showed that several pathways were differentially regulated in cells infected with P. gingivalis relative to uninfected controls. Of these pathways the p38 MAPK pathway appeared as a likely candidate to be a pathway that lead to inflammation as a result of P. gingivalis infection. The mitogen activated protein kinases (MAPK) are involved in many host cell signaling pathways including mitotic response to growth factors, cytokine responses, cytoskeletal reorganization, and stress responses. p38 MAP kinase is a serine threonine kinase activated by phospho rylation of tyrosine and threonine residues by MAP kinase kinase (M E K) 3 and 6, as shown in Fig. 2.2 The p38 pathway leads to cytokine production and inflammation (Schindler et al. 2007). The p38 MAPK pathway has been shown to play a pivotal role in inflammatory cytokine and chemokine gene regulation at both the transcriptional and post tran scriptional levels (Patil & Kirkwood, 2007) Phosphor ylation of p38 and the resulting phosphorylation of downstream intermediates activates gene transcription and stabilizes the AU ric h element (ARE) -containing mRNA (Patil & Kirkwood, 2007) Transcription factor Max, a basic helix -loophelix zipper pro tein that forms a sequence specific DNA binding complex with Myc, is important for the regulation of cell development and proliferation. Transcription factor Max and t he p38 pathway were shown to be up -regulated i n the microarray ana lysis However, a simple increase in production of mRNA shown in the array is not adequate to show that the p38 pathway is more active in cells infected with P. gingivalis MKK3 and p38 are activated by phosphorylation. Western blots were performed to

PAGE 41

41 d etermine phosphorylation status of the p38 pathway in cells that had been infected with P. gingivalis Based on the array data, it was hypothesized that the p38 pathway was upregulated in cells exposed to P. gingivalis Based on previous experiments evaluating the activity of YPF1, it was expected that the fimbr i ae -def icient mutant would be less effective at activating the p38 pathway. Methods Experimental Conditions HTR 8/SVneo cells passage 126, were grow n to 90% confluency in T 25 flasks in RPMI 1640 media (Sigma) supplemented with 5% fetal bovine serum at 37C in the presence of 5% CO2. P. gingivalis wild type and YPF 1 fimA mutant were grown anaerobic ally at 37C in trypticase -soy broth medium supplemented with yeast extract (1mg/ml), hemin (5ug/ml), and menadione (1ug/ml). Erythromycin (15ug/ml) was added to media for YPF -1. Bacteria were harvested in the log phase were co -cultured with HTR 8/SVneo cells at a multiplicity of infection of 200 bacteria per cell at 37C in the presence of 5% CO2. Time points were 5, 20, and 60 minutes. These shorter periods of co-culture, relative to the co -culture times for the microarray, were chosen because the cellular response being tested was phosphorylation. Preparation of Lysates After co -culture cells were washed twice with ice -cold Tris Buffered Saline (TBS 20mM Tris HCl, 150mM NaCl) and lysed in RIPA buffer (50 mM Tris; 150 mM NaCl; 1mM EDTA; 1% Triton -X 100; 5% Protease Inhibitor Cocktail, 20x concentrate from Sigma; 20 mM PMSF; 200 mM Na3VO4; 200 mM NaF) for 30 minutes at 4C. The soluble fraction was collected by centrifugation at 13000 rpm f or 10 minutes at 4C.

PAGE 42

42 Western Blotting Cell extract was den atured in sample buffer (625 mM Tris -HCl, pH 6.8; 25% glycerol; 2% SDS; 0.01% bromophenol blue; 5% beta mercaptoethanol, added immediately prior to use ), resolved by SDS -PAGE, and electro transferred to a nitrocellulose membrane (Bio -Rad). The membrane was blocked with 5% bovine serum albumin (Santa Cruz) in TBS T (20mM Tris HCl, 150mM NaCl, 0.10% Tween 20) for 1 hour at room temperature and then incubated in primary antibody overnight at 4C. Primary antibodies were from Santa Cruz, diluted 1:200, and were antibodies to either p38, phospho -p38, MEK 3, phosphoMEK3/6, or Max. After washing the membrane was incubated in biotin linked bovine anti rabbit immunoglobulin (Santa Cruz), diluted 1:250, at room temperature for 1 hour. After washing, the membrane was then incubated in streptavidin peroxidase polymer (SIGMA) diluted 1:10,000 in TBS T at room temperature for 1 hour. The membrane was then washed five times in TBS T and once in TBS w ithout Tween 20. Bands were visualized by Pierce ECL Western Blotting Substrate (ThermoScientific, Rockford, IL). The membrane was then stripped in Restore Western Blot Stripping Buffer (ThermoScientific, Rockford, IL) re blocked in 5% BSA for 1 hour at room temperature and actin (Cell Signal) o vernight at 4C Biotin linked secondary antibody, streptavidin peroxidase, and visualization by substrate were repeated by the same methods. Analysis Blot i mages were analyzed and quantitated using ImageJ software (written by Wayne Rasband at the U.S. National Institutes of Health http://rsb.info.nih.gov/ij/ ) following the method outlined at http://www.lukemiller.org/journal/2007/08/quantifying -western -blots -without.html Values referred to as areas by ImageJ, ob tained from this meth od represent ed both the size and the intensity of the bands on the blots. These areas were then standardized with the first band of

PAGE 43

43 each blot, the sample from cells exposed to wild type P. gingivalis for 5 min, set equal to a value of 1.0 This resulted in 10 sets of relative values, one each of total p38, phos pho -p38, total MEK3, phoshoMEK3, and Max as well as five blots actin Each of the five blots was then divided by the values of actin blot. The data were further analyzed by div actin adjusted values of phosphop38 and phospho-MEK3 by the values of total p38 and total MEK3, respectively. Results Graphs were generated showing the ratio of total p38 to ac t in (Figure 4 2 ) and phospho p38 to actin (Figure 4 3 ). A third graph (Figure 4 4) was generated showing the ratio of phospho-p38 to total p38. Figure 4 5 shows the ratio of phospho-p38 to total p38 for two sets of blots. At the 20 and 60 minute time points the values from samples from cells exposed to the wild P. gingivalis are significantly higher than the sample from the uninfected control cells (p<0.05) M E K3 is an activator of p38. Graphs were generated to show total MEK3 adjusted for actin (Figur e 4 7 ), phosphoMEK adjusted for actin (Figure 4 8 ), and ac tin ad j usted phosphor -MEK3 over actin ad j usted total MEK3. In one set of blots, all three time points cells infect ed with wild -type P. gingivalis produced a greater ratio of phosphorylated M E K3 that the uninfected control cells and c ells infected with the fimA mutant showed an intermediate value at all three time points (Figure 4 9 ). However, when results of multiple repetitions of the experiment were looked at together no real difference among the values could be determined (Figure 4 10) Max is a transcription factor whose production is upregulated by phosphorylated p 38. Figure 4 13 shows that c ells infected with wild type P. gingivalis produced more Max than

PAGE 44

44 uninfected controls with cells infected with the fimA mutant showing and intermediate value at all three time points. However, none of these differences were stat istically significant. Discussion The effect of P. gingivalis on MAPK pathways in HTR 8/SVneo cells has not been thoroughly examined in previous studies; however there have been studies on the effects of P. gingivalis on gingival epithelial cells (GECs). GECs frequently encounter P. gi ngivalis in vivo Although u nable to detect p38 MAPK activation in either control or bacterially stimulated cells, results from a study by Watanabe et al suggest the P. gin givalis can selectively target components of the MAP kinase pathway (Watanab e et al. 2001) Another study found that infection by P. gingivalis accelerates progression of gingival epithelial cells throug h the cell cycle and that this acceleration was inhib ited in fimbriae deficient YPF1 (Kuboniwa et al. 2008) A study tha t was done using HTR 8/SVneo cells examined the effects of cadmium on MAPK signaling pathways. The authors concluded that cadmium, as well as hydrogen peroxide, trigge r the activation of MAPK signal ing pathways in HTR 8/ SVneo cells (Valbonesi et al. 2008) A study focusing on the p38 MAPK pathway in human choriodecidual explants found that the p38 pathway was involved in causing preterm labor through prostaglandin synthesis in the presence of infection and inflammation (Shoji et al. 2007) In osteoblasts, p38 was found to regulate IL 1beta -stimulated IL 6 expression (Patil et al. 2004) Th e current study examined the effect s of co -culture with P. gingivalis on the p38 MAP kinase pathway in HTR 8/SVneo cells The p38 MAPK pathway has been shown to play a pivotal role in inflammatory cytokine and chemokine gene regulation at both the transcriptional and post transcriptional levels (Patil & Kirkwood, 2007) The p38 MAP kina se is a serine threonine kinase activated by phosphor ylation of tyrosine and threonine residues by MAP kinase

PAGE 45

45 kinase (MEK) 3 and 6. The time point s used in these experiments were 5, 20, and 60 minutes. These shorter periods of co-culture were chosen because the cellular response being tested was phosphorylation. Based on the array data, it was hypothesized that the p38 pathway was upregulated in cells exposed to P. gingivalis Results of this study showed that co -culture with P. gingivalis did increase the phosphorylation of p38 and of M E K3. Additionally, co -culture with the fimbriae deficient mutant YPF1 resulted in a diminished effect on phosphorylation of p38 and M E K3. However, the only results that were statistically significant at any re asonable p -value were the differences in p38 phosphorylation relative to total p38 between cell co cultured with wild type P. gingi valis and those cells not exposed to any bacteria at 5 and 20 minutes (Figure 4 5). Therefore, t hese re sults cannot be said to definitively confirm the hypothesis that the pathway of p38 MAP kinase is activated by infection with P. gingivalis and to a lesser extent, with infection by the fimA mutant in HTR 8 /SVneo cells However, the results do indicate that this is a possibility and may be shown to be true with further study. This finding would be signific ant because the p38 pathway leads to the production of cytokines that have also been associated with preterm birth (Patil & Kirkwood, 2007)

PAGE 46

46 Lane 1 2 3 4 5 6 7 8 9 p38 phospho-p38 actin of p38 actin of phospho-p38 Figure 4 1. Western i mmunoblots of HTR 8/SVneo cells infected with wild type P. gingivalis, fimA mutant P. gingivalis and uninfected controls for 5, 20, or 60 minutes. The blots were probed with antibodies to p38 and phospho -p38 and then stripped and reprobed actin. Figure 4 2. R actin band intensity. 0 0.2 0.4 0.6 0.8 1 1.2 Wt 5min FimA 5min Ctrl 5min Wt 20min FimA 20min Ctrl 20min Wt 60min FimA 60min Ctrl 60minRelative Intensity Lane 1 P. gingivalis for 5min 2 FimA mutant for 5min 3 No bacteria for 5 min 4 P. gingivalis for 20min 5 FimA mutant for 20 min 6 No bacteria for 20 min 7 P. gingivalis for 60min 8 FimA mutant for 60 min 9 No bacteria for 60 min

PAGE 47

47 Figure 4 3. Re sults of densitometric analyses of the ratio of phosphoactin band intensity. Figure 4 4. R esults of densitometric analyses of the ratio of phospho-p38 band intensity to total p38 band intensity. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Wt 5min FimA 5min Ctrl 5min Wt 20min FimA 20min Ctrl 20min Wt 60min FimA 60min Ctrl 60minRelative Intensity 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Wt 5min FimA 5min Ctrl 5min Wt 20min FimA 20min Ctrl 20min Wt 60min FimA 60min Ctrl 60minRelative Intensity

PAGE 48

48 Figure 4 5. Mean ( the standard deviation) of phosphorylated p38 over total p38 for two sets of blots. Astericks indicate a significant difference (p<0.05) between cells infected with wild type P. gingivalis and uninfected control cells. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Wt 5min FimA 5min Ctrl 5min Wt 20min FimA 20min Ctrl 20min Wt 60min FimA 60min Ctrl 60minRelative Intensity* *

PAGE 49

49 Lane 1 2 3 4 5 6 7 8 9 MKK3 phospho-MKK3 actin of MKK3 actin of phospho-MKK3 Figure 4 6 Western immunoblots of HTR 8/SVneo cells infected with wild type P. gingivalis, fimA mutant P. gingivalis and uninfected controls for 5, 20, or 60 minutes. The blots were probed with antibodies to total M E K3 and phospho-M E K3 and then stripped actin Figure 4 7 Graph showing the results of densitometric analyses of the ratio of total M E K3 band actin band intensity. 0 0.2 0.4 0.6 0.8 1 1.2 Wt 5min FimA 5min Ctrl 5min Wt 20min FimA 20min Ctrl 20min Wt 60min FimA 60min Ctrl 60minRelative Intensity Lane 1 P. gingivalis for 5min 2 FimA mutant for 5min 3 No bacteria for 5 min 4 P. gingivalis for 20min 5 FimA mutant for 20 min 6 No bacteria for 20 min 7 P. gingivalis for 60min 8 FimA mutant for 60 min 9 No bacteria for 60 min

PAGE 50

50 Figure 4 8 Graph showing the results of densitometric analyses of the ratio of phospho-M E K3 actin band intensity. Figure 4 9 Graph showing the results of densitometric analyses of the ratio of phospho-M E K3 band intensity to total M E K3 band intensity. 0 0.2 0.4 0.6 0.8 1 1.2 Wt 5min FimA 5min Ctrl 5min Wt 20min FimA 20min Ctrl 20min Wt 60min FimA 60min Ctrl 60minRelative Intensity 0 0.2 0.4 0.6 0.8 1 1.2 Wt 5min FimA 5min Ctrl 5min Wt 20min FimA 20min Ctrl 20min Wt 60min FimA 60min Ctrl 60minRelative Intensity

PAGE 51

51 Figure 4 10. Mean ( the standard deviation ) of phosphorylated MEK3 over total MEK3 for two sets of blots. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Wt 5min FimA 5min Ctrl 5min Wt 20min FimA 20min Ctrl 20min Wt 60min FimA 60min Ctrl 60minRelative Intensity

PAGE 52

52 Lane 1 2 3 4 5 6 7 8 9 Max actin of Max Figure 4 11. Western immunoblots of HTR 8/SVneo cells infected with wild type P. gingivalis, fimA mutant P. gingivalis and uninfected controls for 5, 20, or 60 minutes. The blots were probed with antibodies to transcription factor Max and then stripped and actin. Figure 4 1 2 R actin band intensity. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Wt 5min FimA 5min Ctrl 5min Wt 20min FimA 20min Ctrl 20min Wt 60min FimA 60min Ctrl 60minRelative Intensity Lane 1 P. gingivalis for 5min 2 FimA mutant for 5min 3 No bacteria for 5 min 4 P. gingivalis for 20min 5 FimA mutant for 20 min 6 No bacteria for 20 min 7 P. gingivalis for 60min 8 FimA mutant for 60 min 9 No bacteria for 60 min

PAGE 53

53 Figure 4 13. Me an ( the standard error) of transcription factor Max for three sets of blots. 0 0.5 1 1.5 2 2.5 3 Wt 5min FimA 5min Ctrl 5min Wt 20min FimA 20min Ctrl 20min Wt 60min FimA 60min Ctrl 60minRelative Intensity

PAGE 54

54 CHAPTER 5 CYTOKINES Introduction Cytokines are small proteins that function in cellular signaling. Cytokines have been found to be related to preterm birth. Figueroa et al found that concentrations of IL 1alpha, IL 6, and IL 8 were significantly higher in the amniotic fluid of women who delivered within seven days of the amniocentesis at which their samples were taken than those that deliv ered more than seven days later (Figueroa et al. 2005) IL 1 can be found in elevated concentrations in the amniotic fluid of women experiencing preterm labor IL 1beta was fo u nd to be the key subtype (Hill, 2000) and the elevated IL 1 activity in fluid with infection was more consistent with IL 1beta than IL 1alpha (Hill, 2000, Mitchell et al. 1993) Steinborn et al found increased IL 1beta production in women delivering preterm compared to women delivering at term and not in labor (Steinborn et al. 1996) Animal studies also indicate IL 1beta as a trigger of preterm birt h. Romero et al showed that IL 1 stimulated preterm delivery in mice (Romero et al. 1991) A study by Sadowsky et al found that intra amniotic infusions of IL 1beta in pregnant rhesus monkeys induced pre term labor in all cases (n = 5) (Sadowsky et al. 2006) IL 6 is a multipotent pro inflammatory cytokine that functions in B cell stimulation as well as promoting inflammation. Human tropho blast s are known to produce IL 6 (Saji et al. 2000) IL 1 and IL 6 have been shown to stimulate CRH from human placental trophoblasts in vitro (Petraglia et al. 1990) IL 6 has been found to be elevated in the amniotic f luid of women within seven days of delivery (Figueroa et al., 2005) Women experiencing preterm labor also have elevated amniotic fluid and serum concentrations of chemokines. C hemokines are chemoattractant cytokines that induce

PAGE 55

55 chemotaxis in nearby cells by setting up a concentration gradient leading neutrophils to the site of infection. IL 8 was the first chemokine to be characterized. IL 8 has been found to be elevated in the amniotic fluid of women experiencing preterm labor (Hill, 2000) Dowd et al also fou nd that IL 8 levels in the cervical secretions of women who deliver preterm were significantly higher than th ose who did not deliver preterm (Dowd et al. 2001) Based on the array data, it was hypothesized that production of cytokines IL 8 and IL 1beta would increased in cells infected with P. gingivalis Pr evious experiments conducted using YPF1 and the HTR 8/SVneo cells lead to the prediction that YPF1 would be less effective at increasing cytokine production. The microarray data indicated that cytokines IL 8 and IL 1 were upregulated in HTR 8/SVneo cells infected with P. gingivalis However, cytokines can be regulated post tran s criptionally as well as at the mRNA level. ELISAs were performed to confirm increased production of cytokines. The use of ELISA also allowed for additional cytokines and multipl e time points to be evaluated. Methods H TR 8/SVneo cells were grown to 90% confluency in T 25 flasks in RPMI 1640 media (Sigma), supplemented with 5% fetal bovine serum, at 37 Celsius in 5% carbon dioxide. P. gingivalis 33277 and YPF 1 fimA mutant wer e grown anaerobic ally at 37C in trypticase -soy medium supplemented with yeast extract (1mg/ml), hemin (5ug/ml), and menadione (1ug/ml). Erythromycin (15ug/ml) was added to media for YPF 1. Cells were co -cultured with bacteria harvested in the log p hase at a multiplicity of infection of 200 bacteria per cell for 2 hours at 37C in the presence of 5% CO2. Afte r 2 hours the supernatant from 2 hour flasks was collected and cell lysis buffer (Sigma) was then used to solubilize cells. Supernatant and lysate samples were frozen at 20C until use. The r emaining

PAGE 56

56 flasks were washed at 2 hours with PBS and given fresh media. Samples for 6 hour and 24 hour time points were collected in the same manner as the 2 hour time point. Enzyme -linked immunosorbent assay ( ELIS A) was used to quantify production of cytokines IL 8, IL 6, and IL 1beta Cell culture supernatant was measured for IL 6 and IL 8 and cell lysates were measured for IL 1beta. Commercially available r eagents and kits (R&D Systems) were used for ELISAs in accordance with the manufacturers instructions. Results The results of the ELISA measuring the HTR 8/SVneo cell lysates, Fig ure 5 1 show the measured concentr ation of IL1beta. There is a significant increase in the concentration of IL 1beta in cells co -cultured with wild type P. gingivalis rel ative to the cells not exposed to bacteria (p>0.005) at all three time points, which include 2, 6, and 24 hours. There is also a significant decrease (p>0.005) in the concentration of IL 1beta in lysates from cells exposed to the f imA mutant P. gingivalis relative to the wild type P. gingivalis at all three time points. Surprisingly, at the 24 hour time point the concentration of IL 1beta is lower in cells exposed to the fimA mutant than in cells exposed to no bacteria. ELISA analysis of the cell supernatant also showed an increase i n cytokine production after exposure to P. gingivalis Figure 5 2 shows the concentration of IL 6 was significantly lower (p<0.0005) in the uninfected control cell supernatant than in the supernatant from cells infected with wild type P. gingivalis at 6 and 24 hours. At 24 hours, the supernatant from cells infected with the f imA mutant P. gingivalis had a significantly lower (p<0.0005) concentration of IL 6 than the cells infected with wild type P. gingivalis and a si gnificantly higher (p<0.0005) concentration than cells no exposed to bacteria. T he concentration of IL 8 as shown in Figure 5 3, was significantly lower in the supernatants from cells not exposed to bacteria than those of the cells co -cultured with the wild

PAGE 57

57 type P. gingivalis at 6 h ours (p<0.001) and 24 hours (p<0.005). At 24 hours, t he concentration of IL 8 in the supernatant from the cells co -cultured with the fimA mutant was significantly lower than the wild type (p<0.001) and slightly, but significantly (p<0.01) higher than the concentration from cells not exposed to bacteria. Discussion The effects on cytokine production in trophoblast cells of various species of pathogenic bacteria have been studied. Griesinger et al studied the effects of Escherichi a coli Bacteroides fragilis Mycoplasma hominis Staphylococ cus aureus and Streptococcus agalactiae all of which have been identified in intrauterine infections. They found that these pathogenic microorganisms induced a dose and time -dependent release of IL 1beta, IL 6, and IL 8 (Griesinger, Saleh et al. 2001). The hypothesis being tested in this study was that infection of HTR 8/SVneo cells with P. gingivalis wo uld increase production of cytokines IL 8 and IL 1beta. Results of ELISA confirmed the up regulation of IL 1beta and IL 8 at the two hour time point when the array samples were collected. However, the use of ELISA allowed for the inclusion of additional time points. The increase in cytokine production was much greater at 24 hours than 2 hours in for both cytokines. Interestingly, the finding that P. gingivalis infection results in increased production of IL 8 is in contrast to previous studies that have found P. gingivalis inhibits production of IL 8 by human gingival epithelial cells (Darveau et al. 1998) Additional cytokines, TNF 6, were also evaluated using ELISA. TNF not produced in a measurable concentration by any of the cells at any time point. IL 6, however, was shown to be increased in the 6 and 24 hour time points. In addition to the cells not exposed to bacteria and the cells exposed to wild type P. gingivalis cytokine concentrations produced from cells infected with YPF1 were measured and found to be lower than cell infected with wild

PAGE 58

58 type P. gingivalis These results are impor tant bec ause the cytokines IL 1beta, IL 6, and IL 8 have all been attributed to preterm birth.

PAGE 59

59 Figure 5 1. Graph depicting measured concentration of IL 1beta present in cell culture supernanent. The horizontal axis shows time, in hours, following i nitial exposure to bacteria. The vertical axis shows the concentration, in pg/ml, of IL 1beta. The blue bars show data from cells infected with wild type P. gingivalis ATCC 33277. The red bars show data from cells infected with a fimA mutant. The green bars show data from cells that were not exposed to bacteria. Asteric k s indicate a significant difference (p<0.005) between the indicated cells and the cel l s infected with wild -type P. gingivalis at the same time point 0 100 200 300 400 500 600 700 800 900 2 6 24Concentration (pg/mL)Time (hours) Wt FimA Ctrl* *

PAGE 60

60 Figure 5 2. Graph depicting measured concentration of IL 6 present in cell culture supernanent. The horizontal axis shows time, in hours, following initial exposure to bacteria. The vertical axis shows the concentration, in pg/ml, of IL 6. The blue bars show data from cells infect ed with wild type P. gingivalis ATCC 33277. The red bars show data from cells infected with a fimA mutant. The green bars show data from cells that were not exposed to bacteria. Asteric k s indicate a significant difference (p<0.00 0 5) between the indicate d cells and the cells infected with wild type P. gingivalis at the same time point 0 500 1000 1500 2000 2500 3000 2 6 24Concentration ( pg/mL )Time (hours) Wt FimA Ctrl*

PAGE 61

61 Figure 5 3 Graph depic ting measured concentration of IL 8 present in HTR 8/SVNEO cell culture supernanent The horizontal axis shows time, in hours, following initial exposure to bacteria. The vertical axis shows the concentration, in pg/ml, of IL 8. The blue bars show data from cells infected with wild type P. gingivalis ATCC 33277. The red bars show data from cells infected with a fimA mutant. The green bars show data from cells that were not exposed to bacteria. Asteric k s indicate a significant difference ( **p<0.001; *p<0 005) between the indicated cells and the cells infected with wild -type P. gingival is at the same time point 0 200 400 600 800 1000 1200 1400 1600 1800 2 6 24Concentration (pg/mL)Time (hours) Wt FimA Ctrl* **

PAGE 62

62 CHAPTER 6 CONCLUSIONS Preterm birth, the leading cause of infant mortality, has increased in the Unite d States over the past 20 years (Michalowicz & Durand, 2007) Maternal periodontal disease is associated with preterm birth but the pathogenesis is not fully understood. The mechanism proposed in this study was P. gingivalis traveling from the mouth to the placenta via the mothers blood stream induced inflammation in the fetal tissues This is reasonable to suggest because P. gingivalis and other oral bacteria have been found in fetal tissues (Barak et al., 2007) This study examined the effects P. gingivalis ATCC 33277 has on HTR 8/SVneo human placental cells in vitro Cells were exposed to bacteria at a multiplicity of infec tion of 200 at 37C in the presence of 5% CO2. The e ffects of bacterial exposure on cells were measured by microarray, confocal laser scanning microscopy, fluorescence assay western blot, and ELISA. The results of the microarray reveal ed that exposure of HTR 8/SVneo cells to P. gingivalis caused the differential regulation of the p38 MAP kinase pathway and pro inflammatory cytokines. This finding is significant because inflammation is known to be associated with preterm birth (Romero et al. 1998) Images from the confocal laser scanning microscope show that P. gingivalis was able to infect HTR 8/SVneo cells The fimbriae -deficient mutant YPF1 was less effective at infecting HTR 8/SVneo cells This indicates that the major fimbriae function in infection of the HTR 8/SVneo cells Fluorescence assays comparing infection of live HTR 8/SVneo cells to HTR 8/SVneo cells fixed in formalin prior to infection showed similar results. This indicates that the YPF1 strain was less able to attach to the HTR 8/SVneo cells thus implyin g a role for fimbriae in attachment to HTR 8/SVneo cells

PAGE 63

63 Western blots were performed to study the effect of P. gingivalis infection on the p38 MAP kinase pathway in HTR 8/SVneo cells The p38 MAP kinase is a serine -threonine kinase activated by phospho r ylation of tyrosine and threonine residues by MAP kinase kinase (MEK) 3 and 6. Based on the array data, it was hypothesized that the p38 pathway was upregulated in cells exposed to P. gingivalis Results of this study showed that infection with P. gingi valis increased the phosphorylation of p38 and of MKK3. Additionally, infection with the fimbriae deficient mutant YPF1 resulted in a diminished effect on phosphorylation of p38 and MKK3. These results confirm the hypothesis that the proinflammatory path way of p38 MAP kinase is activated by infection with P. gingivalis and to a lesser extent, with infection by YPF1. This finding is significant because the p38 MAPK pathway has been shown to play a pivotal role in inflammatory cytokine and chemokine gene regulation at both the transcriptional and post transcriptional levels (Patil & Kirkwood, 2007) ELISA was used to further test t he hypothesis that infection of HTR 8/SVneo cells with P. gingivalis would increase production of specific cytokines. Results of ELISA confirmed the upregulation of IL 1beta and IL 8 at the two hour time point when the array samples were collected. Additionally, t he increase in cytokine production by cells infected with P. gingivalis relative to uninfected c ontrols was much gre ater at 24 hours than 2 hours for both cytokines. Additional cytokines, TNF 6, were also evaluated using ELISA. TNF not produced in a measurable concentration by any of the cells at any time point. IL 6, however, was shown to be increased in the 6 and 24 hour time points. In addition to the cells not exposed to bacteria and the cells exposed to wild type P. gingivalis cytokine concentrations produced from cells infected with YPF1 were measured and determined to be lo wer than that of the cells

PAGE 64

64 infected with wild -type P. gingivalis These results are important because the cytokines IL 1beta, IL 6, and IL 8 have all been attributed to preterm birth. Taken together, the results of this study demonstrate that P. gingivali s found in previous studies to be present in some placentas, has the ability to induce an inflammatory response in cultured human placental cells. Given the role that inflammation is understood to play in preterm parturition, this suggests a role for P. gingivalis in the cause of preterm birth. The results of this study suggest that benefits may be obtained from further study. Additional studies may involve other clinical models such as animal models or the use of primary cells in vitro Various m ethods may also be used to enhance understanding of the HTR 8/SVneo pathways P. gingivalis affects For example, study into what HTR 8/SVneo cell surface receptor s are used by P. gingivalis Of particular interest, the role of the TGF -beta pathway in lea ding to G1 arrest in the cell cycle or to the p38 MAPK pathway. In the future, more research may be done to further elucidate the role P. gingivalis plays in preterm birth.

PAGE 65

65 APPENDI X P<0.05 pathways Pg vs. Neg P<0.001 genes (HTR8 cells) Figure A 1. Pathways most affected by co -culture with P. gingivalis relative to the uninfected control (p<0.001).

PAGE 66

66 P<0.05 pathways Pg vs. Fuso P<0.001 genes (HTR8 cells) Figure A 2. Pathways most affected by co -culture with P. gingivalis relative to cells co -cultured with F. nucleatum (p <0.001).

PAGE 67

67 LIST OF REFERENCES Barak, S., O. Oettinger -Barak, E. E. Machtei, H. Sprecher & G. Ohel, (2007) Evidence of periopathogenic microorganisms in placentas of women with preeclampsia. J Periodontol 78: 670676. Behrman, R. E. & A. S. Butler, (2007) Preterm Birth Causes, Consequences, and Prevention. In Washington, D.C.: The National Academies Press, pp. 772. Belanger, M., L. Reyes, K. von Deneen, M. K. Reinhard, A. Progulske -Fox & M. B. Brown, (2008) Colonization of maternal and f etal tissues by Porphyromonas gingivalis is strain dependent in a rodent animal model. Am J Obstet Gynecol 199: 86 e8187. Blackburn, S. T., (2007) Maternal, Fetal, and Neonatal Physiology Elsevier Health Sciences. Boggess, K. A. & B. L. Edelstein, (2006) Oral health in women during preconception and pregnancy: implications for birth outcomes and infant oral health. Matern Child Health J 10: S169174. Boggess, K. A., K. Moss, P. Madianos, A. P. Murtha, J. Beck & S. Offenbacher, (2005) Fetal immune respo nse to oral pathogens and risk of preterm birth. Am J Obstet Gynecol 193 : 11211126. Chou, H. H., H. Yumoto, M. Davey, Y. Takahashi, T. Miyamoto, F. C. Gibson, 3rd & C. A. Genco, (2005) Porphyromonas gingivalis fimbria -dependent activation of inflammatory genes in human aortic endothelial cells. Infect Immun 73: 53675378. Collins, J. G., H. W. Windley, 3rd, R. R. Arnold & S. Offenbacher, (1994) Effects of a Porphyromonas gingivalis infection on inflammatory mediator response and pregnancy outcome in hams ters. Infect Immun 62: 43564361. Cummings, C. A. & D. A. Relman, (2000) Using DNA microarrays to study host -microbe interactions. Emerg Infect Dis 6 : 513 525. Darveau, R. P., C. M. Belton, R. A. Reife & R. J. Lamont, (1998) Local chemokine paralysis, a novel pathogenic mechanism for Porphyromonas gingivalis. Infect Immun 66 : 16601665. Dickinson, D. P., M. A. Kubiniec, F. Yoshimura & R. J. Genco, (1988) Molecular cloning and sequencing of the gene encoding the fimbrial subunit protein of Bacteroides gingivalis. J Bacteriol 170: 16581665. Dowd, J., N. Laham, G. Rice, S. Brennecke & M. Permezel, (2001) Elevated interleukin8 concentra tions in cervical secretions are associated with preterm labour. Gynecol Obstet Invest 51: 165168.

PAGE 68

68 Figueroa, R., D. Garry, A. Elimian, K. Patel, P. B. Sehgal & N. Tejani, (2005) Evaluation of amniotic fluid cytokines in preterm labor and intact membranes J Matern Fetal Neonatal Med 18: 241247. Goldenberg, R. L., J. F. Culhane, J. D. Iams & R. Romero, (2008) Epidemiology and causes of preterm birth. Lancet 371: 75 84. Graham, C. H., T. S. Hawley, R. G. Hawley, J. R. MacDougall, R. S. Kerbel, N. Khoo & P. K. Lala, (1993) Establishment and characterization of first trimester human trophoblast cells with extended lifespan. Exp Cell Res 206: 204211. Handfield, M., H. V. Baker & R. J. Lamont, (2008) Beyond good and evil in the oral cavity: insights into host -microbe relationships derived from transcriptional profiling of gingival cells. J Dent Res 87: 203223. Handfield, M., J. J. Mans, G. Zheng, M. C. Lopez, S. Mao, A. Progulsk e -Fox, G. Narasimhan, H. V. Baker & R. J. Lamont, (2005) Distinct transcriptional profiles characterize oral epithelium -microbiota interactions. Cell Microbiol 7 : 811823. Hill, J. A., (2000) Cytokines in Human Reproduction. In. : John Wiley & Sons, Inc., pp. 341. Inaba, H., M. Kuboniwa, B. Bainbridge, O. Yilmaz, J. Katz, K. T. Shiverick, A. Amano & R. J. Lamont, (2009) Porphyromonas gingivalis invades human trophoblasts and inhibits proliferation by inducing G1 arrest and apoptosis. Cell Microbiol 11: 1517 1532. Jarjoura, K., P. C. Devine, A. Perez Delboy, M. Herrera -Abreu, M. D'Alton & P. N. Papapanou, (2005) Markers of periodontal infection and preterm birth. Am J Obstet Gynecol 192: 513519. Jeffcoat, M. K., N. C. Geurs, M. S. Reddy, S. P. Cliver, R. L. Goldenberg & J. C. Hauth, (2001) Periodontal infection and preterm birth: results of a prospective study. J Am Dent Assoc 132: 875880. Kato -Maeda, M., Q. Gao & P. M. Small, (2001) Microarray analysis of pathogens and their interaction with hosts. Cell Microbiol 3 : 713719. Keelan, J. A., M. Blumenstein, R. J. Helliwell, T. A. Sato, K. W. Marvin & M. D. Mitchell, (2003) Cytokines, prostaglandins and parturition-a review. Placenta 24 Suppl A : S33 46. Kuboniwa, M., Y. Hasegawa, S. Mao, S. Shizukuishi, A. Amano, R. J. Lamont & O. Yilmaz, (2008) P. gingivalis accelerates gingival epithelial cell progression through the cell cycle. Microbes Infect 10: 122128.

PAGE 69

69 Lafaurie, G. I., I. Mayorga Fayad, M. F. Torres, D. M. Castillo, M. R. Aya, A. Baron & P. A. Hur tado, (2007) Periodontopathic microorganisms in peripheric blood after scaling and root planing. J Clin Periodontol 34: 873879. Lamont, R. J. & H. F. Jenkinson, (1998) Life below the gum line: pathogenic mechanisms of Porphyromonas gingivalis. Microbiol Mol Biol Rev 62: 12441263. Lamont, R. J. & H. F. Jenkinson, (2000) Subgingival colonization by Porphyromonas gingivalis. Oral Microbiol Immunol 15: 341349. Leon, R., N. Silva, A. Ovalle, A. Chaparro, A. Ahumada, M. Gajardo, M. Martinez & J. Gamonal, (2 007) Detection of Porphyromonas gingivalis in the amniotic fluid in pregnant women with a diagnosis of threatened premature labor. J Periodontol 78: 12491255. Lieff, S., K. A. Boggess, A. P. Murtha, H. Jared, P. N. Madianos, K. Moss, J. Beck & S. Offenbacher, (2004) The oral conditions and pregnancy study: periodontal status of a cohort of pregnant women. J Periodontol 75: 116126. Lin, D., M. A. Smith, C. Champagne, J. Elter, J. Beck & S. Offenbacher, (2003) Porphyromonas gingivalis infection during pregnancy increases maternal tumor necrosis factor alpha, suppresses maternal interleukin 10, and enhances fetal growth restriction and resorption in mice. Infect Immun 71: 51565162. Lopez, N. J., P. C. Smith & J. Gutierrez, (2002) Higher risk of preterm birth and low birth weight in women with periodontal disease. J Dent Res 81: 58 63. Love, R. M., M. D. McMillan, Y. Park & H. F. Jenkinson, (2000) Coinvasion of dentinal tubules by Porphyromonas gingivalis and Streptococcus gordonii depends upon binding specificity of streptococcal antigen I/II adhesin. Infect Immun 68: 13591365. McClave, J. T. & T. Sincich, (2009) Statistics, p. 834. Pearson Education Inc, Upper Saddle River. McLean, M. & R. Smith, (1999) Corticotropinreleasing Hormone in Human Pregnancy and Parturition. Trends Endocrinol Metab 10: 174178. Michalowicz, B. S. & R. Durand, (2007) Maternal periodontal disease and spontaneous preterm birth. Periodontol 2000 44 : 103112. Mi lward, M. R., I. L. Chapple, H. J. Wright, J. L. Millard, J. B. Matthews & P. R. Cooper, (2007) Differential activation of NF kappaB and gene expression in oral epithelial cells by periodontal pathogens. Clin Exp Immunol 148: 307324. Mitchell, M. D., M. S. Trautman & D. J. Dudley, (1993) Cytokine networking in the placenta. Placenta 14: 249275.

PAGE 70

70 Offenbacher, S., S. Lieff, K. A. Boggess, A. P. Murtha, P. N. Madianos, C. M. Champagne, R. G. McKaig, H. L. Jared, S. M. Mauriello, R. L. Auten, Jr., W. N. Herb ert & J. D. Beck, (2001) Maternal periodontitis and prematurity. Part I: Obstetric outcome of prematurity and growth restriction. Ann Periodontol 6 : 164174. Ohno, T., N. Okahashi, S. Kawai, T. Kato, H. Inaba, Y. Shibata, I. Morisaki, Y. Abiko & A. Amano, (2006) Proinflammatory gene expression in mouse ST2 cell line in response to infection by Porphyromonas gingivalis. Microbes Infect 8 : 10251034. Ohno, T., N. Okahashi, I. Morisaki & A. Amano, (2008) Signaling pathways in osteoblast proinflammatory respo nses to infection by Porphyromonas gingivalis. Oral Microbiol Immunol 23: 96 104. Patil, C., X. Zhu, C. Rossa, Jr., Y. J. Kim & K. L. Kirkwood, (2004) p38 MAPK regulates IL 1beta induced IL 6 expression through mRNA stability in osteoblasts. Immunol Inves t 33: 213233. Patil, C. S. & K. L. Kirkwood, (2007) p38 MAPK signaling in oral -related diseases. J Dent Res 86: 812825. Petraglia, F., G. C. Garuti, B. De Ramundo, S. Angioni, A. R. Genazzani & L. M. Bilezikjian, (1990) Mechanism of action of interleukin1 beta in increasing corticotropinreleasing factor and adrenocorticotropin hormone release from cultured human placental cells. Am J Obstet Gynecol 163: 13071312. Pitiphat, W., K. J. Joshipura, M. W. Gillman, P. L. Williams, C. W. Douglass & J. W. Rich Edwards, (2008) Maternal periodontitis and adverse pregnancy outcomes. Community Dent Oral Epidemiol 36: 3 11. Radmacher, M. D., L. M. McShane & R. Simon, (2002) A p aradigm for class prediction using gene expression profiles. J Comput Biol 9 : 505511. Romero, R., J. Espinoza, L. F. Goncalves, J. P. Kusanovic, L. A. Friel & J. K. Nien, (2006) Inflammation in preterm and term labour and delivery. Semin Fetal Neonatal M ed 11: 317326. Romero, R., R. Gomez, F. Ghezzi, B. H. Yoon, M. Mazor, S. S. Edwin & S. M. Berry, (1998) A fetal systemic inflammatory response is followed by the spontaneous onset of preterm parturition. Am J Obstet Gynecol 179: 186193. Romero, R., M. Mazor & B. Tartakovsky, (1991) Systemic administration of interleukin1 induces preterm parturition in mice. Am J Obstet Gynecol 165: 969971.

PAGE 71

71 Sadowsky, D. W., K. M. Adams, M. G. Gravett, S. S. Witkin & M. J. Novy, (2006) Preterm labor is induced by int raamniotic infusions of interleukin1beta and tumor necrosis factor alpha but not by interleukin 6 or interleukin8 in a nonhuman primate model. Am J Obstet Gynecol 195: 15781589. Saji, F., Y. Samejima, S. Kamiura, K. Sawai, K. Shimoya & T. Kimura, (2000 ) Cytokine production in chorioamnionitis. J Reprod Immunol 47: 185196. Schindler, J. F., J. B. Monahan & W. G. Smith, (2007) p38 pathway kinases as anti inflammatory drug targets. J Dent Res 86: 800811. Shoji, T., S. Yoshida, M. Mitsunari, N. Miyake, S. Tsukihara, T. Iwabe, T. Harada & N. Terakawa, (2007) Involvement of p38 MAP kinase in lipopolysaccharide -induced production of proand anti inflammatory cytokines and prostaglandin E(2) in human choriodecidua. J Reprod Immunol 75: 82 90. Simon, R., (2 003) Diagnostic and prognostic prediction using gene expression profiles in highdimensional microarray data. Br J Cancer 89: 15991604. Simon, R., (2008) Microarray -based expression profiling and informatics. Curr Opin Biotechnol 19: 26 29. Simon, R., M. D. Radmacher, K. Dobbin & L. M. McShane, (2003) Pitfalls in the use of DNA microarray data for diagnostic and prognostic classification. J Natl Cancer Inst 95: 14 18. Steinborn, A., H. Gunes, S. Roddiger & E. Halberstadt, (1996) Elevated pla cental cytokine release, a process associated with preterm labor in the absence of intrauterine infection. Obstet Gynecol 88: 534539. Thomas, J. G. & L. A. Nakaishi, (2006) Managing the complexity of a dynamic biofilm. J Am Dent Assoc 137 Suppl : 10S 15S. Valbonesi, P., L. Ricci, S. Franzellitti, C. Biondi & E. Fabbri, (2008) Effects of cadmium on MAPK signalling pathways and HSP70 expression in a human trophoblast cell line. Placenta 29: 725733. Watanabe, K., O. Yilmaz, S. F. Nakhjiri, C. M. Belton & R J. Lamont, (2001) Association of mitogen activated protein kinase pathways with gingival epithelial cell responses to Porphyromonas gingivalis infection. Infect Immun 69: 67316737. Xie, H., S. Cai & R. J. Lamont, (1997) Environmental regulation of fimb rial gene expression in Porphyromonas gingivalis. Infect Immun 65: 22652271. Xie, H., W. O. Chung, Y. Park & R. J. Lamont, (2000) Regulation of the Porphyromonas gingivalis fimA (Fimbrillin) gene. Infect Immun 68: 65746579.

PAGE 72

72 Xie, H., N. Kozlova & R. J. Lamont, (2004) Porphyromonas gingivalis genes involved in fimA regulation. Infect Immun 72: 651658. Xie, H. & R. J. Lamont, (1999) Promoter architecture of the Porphyromonas gingivalis fimbrillin gene. Infect Immun 67: 32273235. Xiong, X., P. Buekens, W. D. Fraser, J. Beck & S. Offenbacher, (2006) Periodontal disease and adverse pregnancy outcomes: a systematic review. BJOG 113: 135143. Zhou, Q. & S. Amar, (2007) Identification of signaling pathways in macrophage exposed to Porphyromonas gingivalis or to its purified cell wall components. J Immunol 179: 77777790.

PAGE 73

73 BIOGRAPHICAL SKETCH Sally Riewe was raised in Satellite Beach, Florida. She attended Surfside Elementary, DeLaura Jr. High, and Satellite High School. She received a Bachelor of S cience degree in environmental sciences from the University of Florida in 2006 and a Master s of Science degree in medical science from the University of Florida in 2009. She hopes to attend dental school and continu e her education to become a dentist She would like to practice and research dentistry in the state of Florida. She cu rrently lives with her fianc Bret, dog Kallie, cat Lynx, and some birds and fish. She enjoys gardening in the Florida sunshine.