<%BANNER%>

Ilomastat as an Agent for Amending the Pro-Inflammatory Cytokine Response of Endocervical Cells

Permanent Link: http://ufdc.ufl.edu/UFE0023856/00001

Material Information

Title: Ilomastat as an Agent for Amending the Pro-Inflammatory Cytokine Response of Endocervical Cells
Physical Description: 1 online resource (41 p.)
Language: english
Creator: Mayor-Lynn, Kathleen
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: cytokines, ilomastat, infection, inflammation, premature, preterm, progesterone
Clinical Investigation (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The rate of preterm births in the United States is a growing public health problem with significant consequences for families and high costs to society. It is commonly believed that the pathways leading to the two most common reasons for preterm birth, preterm labor (PTL) and preterm premature rupture of membranes (PPROM) are inflammatory in nature. This inflammatory response is thought to be initiated by microorganisms present in the genital tract. The objective of this project is to use our established endocervical cell culture as a model for the lower genital tract to evaluate the inflammatory response of endocervical cells to bacterial stimulus and test an anti-inflammatory substance as a potential agent for blunting the inflammatory response that is believed to lead to PTL and PPROM. Using an established endocervical cell culture model, we provoked an inflammatory response by adding various concentrations of a bacterial lysate and TNF-alpha to a multi-well plate containing the immortalized cells. After incubation, the supernatant and prepared cells were tested for concentrations of various inflammatory cytokines and matrix metalloproteinases (MMPs). After assessing which concentration produces a useful inflammatory response, an anti-inflammatory substance, ilomastat, was added to a second multi-well plate, and followed by bacterial lysate after hour incubation. The same procedures were used to measure inflammatory cytokines and MMPs, after the addition of Ilomastat. The experiments were repeated using TNF-alpha as a stimulus after pretreatment with varying concentrations of progesterone. Real-time PCR was performed with a positive control to assess for the presence of progesterone receptors on these cells. We demonstrated the cell line's ability to produce and secrete IL-6, IL-8, TNF-alpha, and pMMP-9. Similarly we demonstrated the cell line's ability to respond to E. coli lysate and TNF-alpha stimuli with increased production and secretion the cytokines and protease. When any treatment with ilomastat was compared to no treatment we found a trend toward decreased production of IL-6 and IL-8 at baseline, but no difference in the amount secreted. A significant increase in the amount of IL-6 or IL-8 produced and secreted when stimulated with E. coli after pretreatment was shown. We found no pro or active MMP-2 and no active MMP-9 when zymogram gels were run. More pro-MMP-9 was secreted when the cells were stimulated with E. coli after ilomastat pretreatment. Pretreatment of the cells with progesterone resulted in a significant decrease in the production of IL-8, compared with no treatment, both with and without stimulation by TNF-alpha. However, the opposite was demonstrated with IL-8 secreted by the cells. Similarly, we found an increase in the IL-6 both produced and secreted by the cells that were pretreated with progesterone, both with TNF-alpha stimulation and without. We found no measurable concentration of progesterone receptor RNA in the END1/E6E7 cells. Preterm birth is a major public health problem. This project allowed us to create a testing model for evaluating substances to potentially reduce the inflammatory response thought to lead to PTL and PPROM. It also tested two anti-inflammatory substances for potential amelioration of this inflammatory response.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Kathleen Mayor-Lynn.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Garvan, Cynthia W.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0023856:00001

Permanent Link: http://ufdc.ufl.edu/UFE0023856/00001

Material Information

Title: Ilomastat as an Agent for Amending the Pro-Inflammatory Cytokine Response of Endocervical Cells
Physical Description: 1 online resource (41 p.)
Language: english
Creator: Mayor-Lynn, Kathleen
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: cytokines, ilomastat, infection, inflammation, premature, preterm, progesterone
Clinical Investigation (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The rate of preterm births in the United States is a growing public health problem with significant consequences for families and high costs to society. It is commonly believed that the pathways leading to the two most common reasons for preterm birth, preterm labor (PTL) and preterm premature rupture of membranes (PPROM) are inflammatory in nature. This inflammatory response is thought to be initiated by microorganisms present in the genital tract. The objective of this project is to use our established endocervical cell culture as a model for the lower genital tract to evaluate the inflammatory response of endocervical cells to bacterial stimulus and test an anti-inflammatory substance as a potential agent for blunting the inflammatory response that is believed to lead to PTL and PPROM. Using an established endocervical cell culture model, we provoked an inflammatory response by adding various concentrations of a bacterial lysate and TNF-alpha to a multi-well plate containing the immortalized cells. After incubation, the supernatant and prepared cells were tested for concentrations of various inflammatory cytokines and matrix metalloproteinases (MMPs). After assessing which concentration produces a useful inflammatory response, an anti-inflammatory substance, ilomastat, was added to a second multi-well plate, and followed by bacterial lysate after hour incubation. The same procedures were used to measure inflammatory cytokines and MMPs, after the addition of Ilomastat. The experiments were repeated using TNF-alpha as a stimulus after pretreatment with varying concentrations of progesterone. Real-time PCR was performed with a positive control to assess for the presence of progesterone receptors on these cells. We demonstrated the cell line's ability to produce and secrete IL-6, IL-8, TNF-alpha, and pMMP-9. Similarly we demonstrated the cell line's ability to respond to E. coli lysate and TNF-alpha stimuli with increased production and secretion the cytokines and protease. When any treatment with ilomastat was compared to no treatment we found a trend toward decreased production of IL-6 and IL-8 at baseline, but no difference in the amount secreted. A significant increase in the amount of IL-6 or IL-8 produced and secreted when stimulated with E. coli after pretreatment was shown. We found no pro or active MMP-2 and no active MMP-9 when zymogram gels were run. More pro-MMP-9 was secreted when the cells were stimulated with E. coli after ilomastat pretreatment. Pretreatment of the cells with progesterone resulted in a significant decrease in the production of IL-8, compared with no treatment, both with and without stimulation by TNF-alpha. However, the opposite was demonstrated with IL-8 secreted by the cells. Similarly, we found an increase in the IL-6 both produced and secreted by the cells that were pretreated with progesterone, both with TNF-alpha stimulation and without. We found no measurable concentration of progesterone receptor RNA in the END1/E6E7 cells. Preterm birth is a major public health problem. This project allowed us to create a testing model for evaluating substances to potentially reduce the inflammatory response thought to lead to PTL and PPROM. It also tested two anti-inflammatory substances for potential amelioration of this inflammatory response.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Kathleen Mayor-Lynn.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Garvan, Cynthia W.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0023856:00001


This item has the following downloads:


Full Text

PAGE 1

ILOMASTAT AS AN AGENT FOR AMENDING THE PRO-INFLAMMATORY CYTOKINE RESP ONSE OF ENDOCERVICAL CELLS By KATHLEEN A. MAYOR-LYNN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008 1

PAGE 2

2008 Kathleen A. Mayor-Lynn 2

PAGE 3

To Andrew, who supported me during this long pr ocess; and to my parents, who gave me the opportunity to strive for all my hearts desire. 3

PAGE 4

ACKNOWLEDGMENTS I thank my husband, parents and Connor and Addison for their support and understanding through this long process. I thank Rodney Edwards for getting me started and encouraging me to continue in his absence. I also thank the PhDs in Drs. Schultz a nd Cheginis laboratories for their help, explanations, and many meetings, speci fically, I thank Olajompo Moloye and Tannaz Toloubeydokhti. I also thank Dr. Schultz and Dr. Garvan for helping me find the meaning in a mess. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES ...........................................................................................................................6 LIST OF FIGURES .........................................................................................................................7 ABSTRACT .....................................................................................................................................8 CHAPTER 1 INTRODUCTION................................................................................................................. .10 Background .............................................................................................................................10 Statement of Problem .............................................................................................................13 Specific Aims..........................................................................................................................14 2 MATERIALS AND METHODS...........................................................................................18 Culturing of END1/E6E7 Cells ..............................................................................................18 Preparation of Bacterial Lysate ..............................................................................................19 Establishing Optimum E. coli Concentration .........................................................................19 Enzyme-Linked Immunosorbent Assay .................................................................................19 Addition of Ilomastat ..............................................................................................................20 Measurement of Metalloproteinases .......................................................................................20 Addition of Progesterone4.......................................................................................................21 RNA PCR for Progesterone Receptors ...................................................................................22 Statistical Analyses .................................................................................................................22 3 RESULTS...................................................................................................................... .........24 4 DISCUSSION................................................................................................................... ......32 Cell Line Responsiveness and Cytokine Production ..............................................................32 Limitations in Use of E. coli as a Stimulus ............................................................................33 Ilomastat .................................................................................................................................34 Progesterone ...........................................................................................................................35 Metalloproteinases ..................................................................................................................35 Conclusions .............................................................................................................................36 LIST OF REFERENCES ...............................................................................................................39 BIOGRAPHICAL SKETCH .........................................................................................................41 5

PAGE 6

LIST OF TABLES Table page 3-1 Background production of inflamma tory cytokines and proteases ....................................25 3-2 IL-6, IL-8, and TNF-alpha produced in response to E. coli lysate stimulus ......................26 3-3 IL-6 and IL8 produced in response to TNF-alpha stimulus ...............................................26 3-4 Cytokine response with varying conc entrations of ilomastat pretreatment .......................27 3-5 Difference in cytokine response after pretreatment with ilomastat ...................................27 3-6 Protease response to E. coli stimulus after pretre atment with ilomastat ............................28 3-7 Difference in cytokine response af ter pretreatment with progesterone .............................28 6

PAGE 7

LIST OF FIGURES Figure page 1-2 Endocervical canal dilati on on transvaginal ultrasound....................................................17 3-1 IL-6 produced in response to E. coli lysate stimulus .........................................................29 3-2 IL-8 produced in response to E. coli lysate stimulus .........................................................29 3-3 TNF-alpha produced in response to E. coli lysate stimulus ...............................................30 3-4 Effect of TNF-alpha on pMMP -9 synthesis on END1/E6E7 cells ....................................30 3-5 Effect of ilomastat and E. coli lysate on pMMP9 synthesis..............................................31 4-1 Revised model for inflammatory cascade responsible for PTL and PPROM ...................38 7

PAGE 8

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science ILOMASTAT AS AN AGENT FOR AMENDING THE PRO-INFLAMMATORY CYTOKINE RESP ONSE OF ENDOCERVICAL CELLS By Kathleen A. Mayor-Lynn December 2008 Chair: Cynthia Garvan Major: Medical SciencesClinical Investigation The rate of preterm births in the United Stat es is a growing public health problem with significant consequences for families and high cost s to society. It is commonly believed that the pathways leading to the two most common reas ons for preterm birth, preterm labor (PTL) and preterm premature rupture of membranes ( PPROM) are inflammatory in nature. This inflammatory response is thought to be initiated by microorganisms present in the genital tract. The objective of this project is to use our established endocervical cell culture as a model for the lower genital tract to evaluate the inflammatory response of endocervical cells to bacterial stimulus and test an anti-inflammatory s ubstance as a potential agent for blunting the inflammatory response that is believ ed to lead to PTL and PPROM. Using an established endocervical cell cult ure model, we provoked an inflammatory response by adding various concentrations of a bacterial lysate and TN F-alpha to a multi-well plate containing the immortalized cells. After incubation, the supern atant and prepared cells were tested for concentrations of various inflamma tory cytokines and matrix metalloproteinases (MMPs). After assessing which c oncentration produces a useful in flammatory response, an antiinflammatory substance, ilomastat, was added to a second multi-well plate, and followed by bacterial lysate after hour incubation. The same procedures were used to measure inflammatory 8

PAGE 9

cytokines and MMPs, after the a ddition of Ilomastat. The experi ments were repeated using TNFalpha as a stimulus after pretreatment with va rying concentrations of progesterone. Real-time PCR was performed with a positive control to as sess for the presence of progesterone receptors on these cells. We demonstrated the cell lin es ability to produce and secr ete IL-6, IL-8, TNF-alpha, and pMMP-9. Similarly we demonstrated th e cell lines ability to respond to E. coli lysate and TNFalpha stimuli with increased production and secretion the cyto kines and protease. When any treatment with ilomastat was compared to no treatment we found a trend toward decreased production of IL-6 and IL-8 at baseline, but no difference in the amount secreted. A significant increase in the amount of IL-6 or IL-8 pr oduced and secreted when stimulated with E. coli after pretreatment was shown. We found no pro or active MMP-2 and no active MMP-9 when zymogram gels were run. More pro-MMP-9 was s ecreted when the cells were stimulated with E. coli after ilomastat pretreatment. Pretreatment of the cells with progesterone resulted in a significant decrease in the production of IL-8, compared with no treatment, both with and without stimulation by TNF-alpha. However, the opposite was demonstrated with IL-8 secreted by the cells. Similarly, we found an increase in th e IL-6 both produced and secreted by the cells that were pretreated with proge sterone, both with TNF-alpha stim ulation and without. We found no measurable concentration of progesterone receptor RNA in the END1/E6E7 cells. Preterm birth is a major public health problem. This project allowed us to create a testing model for evaluating substances to potentially re duce the inflammatory response thought to lead to PTL and PPROM. It also tested two anti-infl ammatory substances for potential amelioration of this inflammatory response. 9

PAGE 10

CHAPTER 1 INTRODUCTION Background The rate of preterm births in the United Stat es is a growing public health problem with significant consequences for families and high costs to society. Despite new treatments, such as 17-alpha-hyproxyprogesterone caproate (17P) which has been shown to decrease the relative risk of recurrent preterm delivery by 33%, the rate continues to rise. 1,17 The percentage of all births occurring prior to 37 weeks has risen by 33% since comparable national data on gestational age became available in 1981. 1 It increased in 2004 to 12.5% from 12.3% the previous year. 1 Prematurity is responsible for the majority of l ong term disability and is the most common cause of perinatal mortality in infa nts without congenital anomalies. 1 In spite of a good deal of effort during the past 40 years, nearly all of the impr ovements in treating this pregnancy complication have been due to advances in neonatology. 1 Much of the difficulty in preventing pret erm labor (PTL) and preterm delivery (PTD) stems from the likelihood that PTL is probabl y the final common pathway of a number of pregnancy complications involving the health of th e cervix, fetus, fetal membranes, placenta, and myometrium. A common component of many of these conditions is inflammation at the maternal-fetal interfacemediated by proinflammatory cytokines. 2 Although various mechanisms may be responsible for preterm delivery; microorganisms ascending the female genital tract and leading to the production of va rious inflammatory mediators ar e believed to be its principal cause 3 (Figure 1-1). Infection is clearly identifiable in over 30% of cases. 4 Bacteria release phospholipases, which initiate the formation of ar achidonic acid, the basis for prostaglandin production. These prostaglandins cause uterine contractions a nd PTL and/or result in the production of matrix metalloprot einases (MMPs) that degrade the fetal membranes and cause 10

PAGE 11

preterm premature rupture of membranes (PPROM). Bacteria also releas e endotoxins, which can cause cells to release pro-inflammatory cy tokines. The primary cytokines involved are interleukins (IL) 1 6 and 8, and tumor n ecrosis factor alpha (TNF-alpha). The proinflammatory cytokines, in turn, stimulate th e expression of enzymes in the prostaglandin biosynthetic pathway, an increase in MMPs, a nd recruitment of neutrophils and macrophages. An extensive body of literature exists that show an association between preterm delivery and elevated levels of pro-inflammatory cytokine s. Levels of pro-inflammatory cytokines are elevated in symptomatic and asymptomatic women who ultimately delivery preterm. Lower genital tract bacteria also produce enzymes wh ich may weaken the protective cervical mucus, favoring ascent of bacteria. 5 The use of antimicrobial agents for th e prevention of preterm delivery, although a reasonable and attractive option, has not been proven effective. Over a decade of trials applying antimicrobial interventions to prevent infection-mediated pret erm delivery has had disappointing results. 6 The use of 17P, while seeming to be a promis ing therapy for PTL, has a major hurdle in its acceptance as a cure-all for preterm birth. The mechanism of action for 17P is less well understood than for all other prophylactic measur es, including those that failed to be proven effective. 7 One possibility is that 17P increases endogenous substances th at act as natural tocolytics to maintain uter ine quiescence, mediated by both genomic and nongenomic mechanisms. 8,9 Progesterone derivatives also inhibit the activation of immune receptors that normally mediate the inflammatory responses in the uterus, cervix, and placenta. 10 In contrast, in vitro data demonstrate 17P could affect interleukin production from whole blood. 11 If a substance that has a plausible mechanism of acti on for the prevention of preterm births could be shown to be effective, the rate of acceptan ce and use among health care providers could be 11

PAGE 12

increased above that for progesterone. The poten tial for reduction in perinatal morbidity and mortality would be far-reaching. 12 The immortalized endocervical cell line End1/E6E7 is known to produce multiple proinflammatory cytokines, including IL 8, pros taglandin E2 (PGE2), and IL-6, among others. 13 Therefore, these cells provide an opportunity to test the stimulation and inhibition of cytokines that are involved in PTL and PPROM in the cell type that lies at the junction of the upper and lower genital tracts. Ilomastat, also known as galardin and GM -6001, is a synthetic, broad-spectrum MMP inhibitor. 14 Ilomastat inhibits the actions of gelatinases and other proteinases that have been implicated in PPROM (Figure 1-1). Ilomasta t also antagonizes the action of TNF-alpha converting enzyme (TACE). TACE, a metalloprotei nase closely related to MMPs, cleaves cellassociated TNF-alpha to its soluble, active form. 15 As a result, ilomastat may inhibit the proinflammatory pathways that have been implicated in preterm delivery, both due to PTL and PPROM. Ilomastat prevents adverse effects cau sed by pro-inflammatory response in multiple animal models. No evidence of cellular toxicity was shown in a cell culture experiment utilizing human fibroblasts. 16 Ilomastat has also been used topica lly on the arms of human volunteers, without evidence of t oxicity, in a study ev aluating would healing. 17 In this project, we used an endocervical cel l model (End1/E6E7) to examine the feasibility of using ilomastat, a synthetic inhibitor of both pro-inflammatory cytokines and MMPs, to ameliorate the pro-inflammatory response of endo cervical cells. Several stimulators of the proinflammatory response are thought to be major inflammatory mediator in the process of preterm delivery. Two of the stimulators, a lysate of Esch erichia coli and TNF-alpha, were used with the End1/E6E7 cell cultures. Stimul ating the cells with these substances should demonstrate a 12

PAGE 13

consistent increase in the cyt okines and MMPs that we hypothesize play major roles in PPROM and PTL. Statement of Problem Preterm birth is a major public health problem with vast consequences. First proposed in the 1970s, large randomized controlled trials have shown 17P to be effective in reducing the rate of recurrent preterm delivery in singleton pregnancies, but only by a 33% reduction in the relative risk. 18 The mechanism of action of 17P is poorly understood. The recommended dose is empiric, the metabolic pathways are not well defi ned especially in pregnant women, and the fetal exposure has not been quantified. The actions of progesterone on the pregnant myometrium include relaxation of the myomet rial smooth muscle, blocking of the action of oxytocin, and inhibition of the formation of gap junctions. 18 In sheep, goats, and some other mammals, a decrease in plasma progesterone and an increas e in circulating estroge n precede the onset of labor. 18 Although no such alteration in the ratio of pl asma estrogen or progesterone precedes the onset of labor in primates, there is evidence that local changes in the progesterone level or the ratio of progesterone to estrogen in the placenta decidua, or fetal membranes may be important in the initiation of labor in humans. 18 In addition, administration of progesterone antagonists in women at term results in an incr eased rate of spontaneous labor. 18 Several other studies indicate that progesterone exerts immuno modulation. Progesterone receptors found on lymphocytes, and the activation of such receptors interferes with interleukin se cretion through the progesteroneinduced blocking factor. 19 We sought to better understand the mechanism of progesterones action in decreasing the rate of spontaneous preterm birth as well as potentially find another therapy for decreasing the in cidence of preterm delivery. 13

PAGE 14

Specific Aims 1. Develop an endocervical cell culture model (E nd 1/E6E7) of the inflammatory processes related to spontaneous preterm delivery. To accomplish this we a. Measured the background pr oduction of inflammatory cytokines and proteases. Specifically we measured IL-6, IL -8, TNF-alpha, MMP-2 and MMP-9. b. Evaluated the effects on the endocervical cells of a bacterial lysate made from E. coli by measuring the production of down stre am inflammatory cytokines, IL-6, IL-8, and TNF-alpha, and two key proteas es, MMP-2 and MMP-9, in response to the inflammatory challenge. c. Evaluated the effects on the endocervical cells of a key pro-inflammatory cytokine, TNF-alpha, on the regulation of the inflammatory cytokines, IL-6 and IL-8, and proteases, MMP-2 and MMP-9. The reason for choosing endocervical cells for this model was relate d to the ascent of bacteria from the lower genital tract causing the inflammatory reaction within the decidua, fetal membranes, etc. The endocervical epithelium is a primary site of plasma and T cell localization in the lower genital tract and th erefore plays an important role in mucosal immune defense of the lower genital tract. As is shown in Figure 1-2, even minimal endocervical canal dilation has been associated with increased risk of preterm labor. We hypothesized that the cells lining that canal ar e involved in protecting the upper genital tract from ascending colonization and infection from th e bacteria present in the lower genital tract. The purpose of characterizing the cytokine produ ction of this model was to have a simple way of manipulating the environment to incr ease production of inflammatory mediators and to subsequently try to reduce that response. 2. Evaluate substances to ameliorate the pr o-inflammatory response the cells produce in response to stimulus. To accomplish this, we a. Measured the degree by which the amount s of IL-6, IL-8, MMP-2 and MMP-9 secreted from cells in response to the E.col i lysate were affected by pretreatment with ilomastat. 14

PAGE 15

b. Measured the degree by which the amounts of IL-6, IL-8 secreted from cells in response to the TNF-alpha were affected by pretreatment with progesterone. The completion of these aims could lead to useful findings to reduce preterm delivery. If a substance can be identified that reduces the inflammation th ought to lead to the two most common causes of preterm delivery, preterm labor and PPROM, the potential for reducing the cost to families and society as a whole as well as reducing infant morbidity due to prematurity is immense. Ilomastat is a unique substance in that its mechanism of action may block both pathways that are thought to be responsible for preterm deliver y. We are unsure if our findings will be generalizable to other fields of medicine in pursuit of treatment and/or prevention of other diseases caused by inflammation due to the multiplicity of cytokines involved in other inflammatory responses in the human body. Future investigators would need to individualize studies based on the appropriate cytokines of interest to thei r field of inquiry. 15

PAGE 16

Proposed Site of Ilomastat Action Figure 1-1. Model for biochemical cascades involved with preterm labor with proposed site of ilomastats action. Neutrophils (N ); Macrophages (M ) 16

PAGE 17

Figure 1-2. Endocervical canal di lation on transvaginal ultrasound 17

PAGE 18

CHAPTER 2 MATERIALS AND METHODS Culturing of END1/E6E7 Cells We obtained the immortalized endocervical ce ll line End1/E6E7 from the American Type Culture Collection. This line was established in 1996 from norm al epithelium taken from one premenopausal woman undergoing a hysterectomy for endometriosis. 20 These cells are known to produce multiple pro-inflammatory cytokine s, including IL-6, IL-8, among others. 16 Cells were grown in Falcon T-75 flasks (Fishe r Scientific) to a concentration of 8.4 x 10 6 cells/mL and maintained in Keratinocyte seru m-free medium (K-sfm) supplemented with bovine pituitary extract at a concentration of 0.05 mg/m L, recombinant epidermal growth factor at 0.1 mg/mL, 1 Molar CaCl 2 at 0.4 mM, and penicillin/streptomycin at 1X (Invitrogen). We did not add hydrocortisone to our medium as described previously, 16 because we intended to use our own immunosuppressant drugs during the experiments and wanted the cells to produce cytokines at baseline. The medium was changed every fourth day a nd cells were passaged to new flasks or plates every seventh day. The cells tested negative for mycoplasma using Takara PCR Mycoplasma Detection Kit at the initiation of re search, and were visually inspected (grossly and microscopically) for evidence of fungal or bact erial infection at every medium change and passage. At 60-90% confluence, cells were harvested by trypsin/EDTA, seeded into 24-well culture plates, grown to initial conflu ence, washed with PBS, and placed in keratinocyte serum-free medium. 18

PAGE 19

Preparation of Bacterial Lysate E. coli colonies from a stock bacterial cultur e line were grown on agar plates for 48 hours. These then were lysed using sonication. A sample of lysate was plated on agar for 48 hours to verify that all bacteria had been killed. An additional sample was tested to determine the concentration of protein using a commercia lly available Protein Assay kit (Biorad). Establishing Optimum E. coli Concentration At 60-90% confluence, the medium was rem oved and replaced with fresh keratinocyte serum-free medium containing 0, 25, 50, 100, 250, and 500 g/mL of E. coli bacterial lysate and incubated for 24 hours. These amounts were chosen based on prior informal laboratory work. At 24 hours the levels of Il-6, Il-8, and TNF-alpha cytokines were meas ured in detergent extracts of End1/E6E7 cells to determine which concentration of E. coli produced a measurable amount of cytokines. Enzyme-Linked Immunosorbent Assay We used commercially available enzyme-linked immunosorbent assay (ELISA) kits (Biosource International, CA) for human IL-6, IL-8, and TNFalpha. The protocols followed were as described in the manual provided by Bios ource; briefly, the samples and standards were added to wells, and incubated along with a bio tinylated anti-human IL-6, IL-8, or TNF-alpha respectively. After aspirating and washing the co ntents of the wells, streptavidin-conjugated horseradish peroxidase was added and incubated. The wells were aspirated and washed again, and chromogen was added. Finally, a stop solution ended the chromogen reaction when the color had developed to approximately 1.5 to 2.0 OD. The plates were then read using an ELx800 Universal Microplate Reader (Fisher) at 450 nm Absorbance of samples was compared with the standard curve. 19

PAGE 20

Addition of Ilomastat Cells were grown in Falcon 24-well plates (Fis her Scientific) to a concentration of 0.2 x 10 6 cells/mL (approximately 1 week), using k-sfm with supplements as above. When the cultures reached 60 to 80% confluence, they were incubated with 0, 5, 15, 25, 35, and 50 g/mL of Ilomastat for 1 hour, then challenged with the predetermined concentration of E. coli lysate for 24 hours. The concentrations of Ilomastat were chosen based on prior informal laboratory work. All cells were growing in a total of 1mL of broth per we ll, containing the calculated concentrations of Ilomastat and the remainder of the liquid made up of the previously described k-sfm with supplements. The amount of E. coli lysate added was 81 ug of our stock, which gave a final concentration of 250 ug/mL. Each experime nt condition occurred in replicates of four. Supernatants were collected and frozen, and ELISA performed as above to determine the concentration of cytokine produced and released by the cells. The cells remaining on the plate were washed to remove any remaining supernat ant, then lysed using Triton-X (Fisher) and the cell lysate samples collected a nd frozen, and ELISA performed to determine the concentrations of cytokine within the cells. Measurement of Metalloproteinases Levels of pro and activated forms of the two gelatinas es, MMP-2 and MMP-9 were measured in the supernatants of th e cells stimulated with TNF-alpha, E. coli lysate and those pretreated with ilomastat then stimulated with E. coli lysate. MMP-2 and MMP-9 were measured using quantitative gelatin zymography as describe d previously by Ladwig GP et al from our institution. 21 Briefly, conditioned medium or cells were incubated with sample buffer (63mM Tris-HCl, pH 6.8, 10% glycerol, 2% sodium dodecyl sulfate (SDS )) and 0.0025% bromophenol blue for 10 minutes at room temperature. Then, 15 l of the equilibrated samples were loaded into a 15-well pre-cast gelatin zymogram gel. Af ter electrophoresis, the gels were immersed in 20

PAGE 21

renaturing buffer (2.7% Triton X-10 0, w/v) on a rotary shaker for 30 minutes at 37 degrees C, then placed in developing buffer ( 50 mM Tris-HCl, 200 mM NaCl, 5 mM CaCl 2 0.2% Brij 35, w/v) on a rotary shaker at 37 degrees C for 24 hours to allow the MMPs to digest the gelatin substrate. After digestion, the gels were stai ned with Coomassie Rapid Stain, photographed with a digital camera and relative pixal density of each band measured using the Kodak Digital Science image analysis software (Kodak Co, Ea stman, NY). Levels of pro and activated MMP-2 and MMP-9 in the samples were calculated from standard curves generated from recombinant pro and activated MMP-2 and MMP-9 a dded to the gelatin zymograms. Addition of Progesterone 4 Cells were grown in Falcon 24-well plates (Fis her Scientific) to a concentration of 0.2 x 10 6 cells/mL (approximately 1 week), using k-sfm with supplements as above. When the cultures reached 60 to 80% confluence, th ey were incubated with 0, 10 -6 10 -7 10 -8 mL Progesterone 4 (P 4 ) for 1 hour, then challenged with 0 or 25ng/mL TNF-alpha for 24 hours. The concentrations of TNF-alpha were chosen based on prior informal laboratory work. The concentrations of P 4 were chosen based on prior laboratory ex perience. We chose to use the P 4 which we had available in the laboratory as the initial substrate with a plan to obtain the clinically used 17-OHP if we were able to demonstrate a difference. All cells were growing in a total of 1mL of broth per well, containing the calculated concentrations of P 4 and the remainder of the liquid made up of the previously described k-sfm with supplements. The amount of TNF-alpha (recombinant from human) added was 2.5 l of 10 ng/mL obtained from Invitrogen, which gave a final concentration of 25 ng/mL. Each experiment condition occurred in replicates of four. Supernatants were collected and ELISA performed as above to determine the concentration of cytokine produced and released by the cells. The cells remaining on the plate were washed to remove any remaining supernatant, then 21

PAGE 22

lysed using Triton-X (Fisher) and the cell lysate samples collected and ELISA performed to determine the concentrations of cytokine within the cells. RNA PCR for Progesterone Receptors Medium was removed from cells contained in eight of the 24 wells on the standard 24-well plate, then washed with PBS. RNA extraction was performed using Trizol reagent (Invitrogen, Carlsbad, CA). Complimentary DNA was generated using 2 g of total RNA using Taqman reverse transcription reagent. The newly synthesized cDNA was used for PCR performed in 96well optical reaction plates using cDNA equi valent to 100 ng RNA in a volume of 50 l reaction containing 1x Taqman Universal Master Mix, op timized concentrations of FAM-labeled probe and specific forward and reverse primer PG R selected from Assay on Demand (Applied Biosystems). The results were analyzed using a comparative method and values normalized to the 18S rRNA expression and converted into fo ld-change based on a doubling of PCR product in each PCR cycle. The PCR was run with a positiv e control of myometrial tissue concurrently. Statistical Analyses Power analysis was conducted. Allowing for a 10% variability in cell cultures and utilizing an of 0.05, replicates of 4 in each experime nt will give a power of 0.80 to detect a three-fold increase in cytokine or MMP concentrations from baseline. Data were checked for possible out of range values using graphs and summary statistics. Means and standard deviations were computed for all amounts of cytokines secreted into the medium and produced by the cell but not secr eted. Amounts of cytokines were computed according to secretion type, cytokine type, and presence/absence of E. coli. For the progesterone and ilomostat data, te sts of hypotheses were conducted to determine if average amounts of cytokine produced and secreted were different between the zero concentration group and concentrations above zero. T-tests were used assuming unequal 22

PAGE 23

variances in groups and the Satterthwaite method for calculating degrees of freedom. A p-value of 0.1 was used as the level of significance fo r all hypothesis testing. SA S software (version 9.1, Cary, N.C.) was used for all analysis. 23

PAGE 24

CHAPTER 3 RESULTS Similar amounts of IL-6 were secreted, 5.325 pg/mL compared with what remained within the cell, 9.370 pg/mL. In contrast, the highest level of secretion was for IL-8 (46.903 pg/mL), with only 3.751 pg/mL remaining within the cell (Figure 3-1). After E. coli stimulation (Figures 3-1, 3-2, 3-3, 34), the cells responded with an increase in production of IL-6 and IL-8 for each concentration of E. coli. For IL-6, more was secreted into the medium (69.299 ug/mL to 242.648 ug/mL) than contained (18.99 ug/mL to 43.404 ug/mL). Similarly, we observed 40 to 50 times more IL-8 secreted into the medium compared with what was produced and not secreted from the cell. The cells showed a trend toward an increase in TNF-alpha secretion into the medium. No significant cha nge in the amount of TNF-alpha contained within the cells was found. After TNF-alpha stimulation (Figure 3-3), the cells produced a significant increase in IL6 (21.070 pg/mL vs. 9.370 pg/mL at baseline) but a decrease in the amount secreted into the medium (3.410 pg/mL vs. 5.325 pg/mL). Similarly, th e cells demonstrated an increase in IL-8 both produced (13.149 pg/mL) and secreted (98.668 pg/mL). Pretreatment of cells with ilomastat prior to the E. coli stimulus resulted in no significant differences when different concentrations were compared (Fi gure 3-4). When any amount of ilomastat was compared to no treatment (Figur e 3-5) the cells demonstrated a significant decrease in baseline production of IL-6 ( 22.182 pg/mL to 18.995 pg/mL), but no difference in the amount secreted into the medium. Similarl y, we found a significant increase in both IL-6 production (2.109 pg/mL to 20.954 pg/mL) and s ecretion (84.997 pg/mL to 219.7 pg/mL) when stimulated with E. coli after pretreatment with ilomastat. The cells produced significantly less IL-8 at baseline, 2.617 pg/mL compared to 5.542 pg/mL, but not when stimulated with E. coli 24

PAGE 25

after pretreatment with ilomastat. When zymogr am gels were run for pro and active forms of MMP-2 and MMP-9 no measurable amounts of the pr o or active forms of MMP-2 or active form of MMP-9 were detected. More pMMP-9 was secreted when E. coli was added to the cells for all concentrations of ilomast at (Figure 3-2 and 3-6). Pretreatment of cells with progesterone prior to the TNF-alpha stimulus resulted in an increase in the IL-6 both produced and secreted by the cells when they were pretreated with progesterone (Figure 3-7). This result was seen both with and without stimulation by TNF-alpha. Similarly, the cells produced le ss IL-8, 3.668 pg/mL, when pretreated with any amount of progesterone compared with no treatment, 8.824 pg/ mL. This outcome was seen both with and without stimulation by TNF-alpha. However, the opposite effect was demonstrated when analysis was done on the IL-8 secreted by the cells. The inconsistent results led us to test the cells for progesterone receptors. We found no measurable concentration of progesterone 4 receptors in the RNA of the END1/E6E7 cells. Table 3-1. Background production of inflammatory cytokines and proteases Number of observations Mean (standard deviation) IL-6 (pg/mL) Secreted into medium 12 5.325 (1.735) Produced, not secreted 12 9.370 (3.640) IL-8 (pg/mL) Secreted into medium 12 46.903 (10.357) Produced, not secreted 12 3.751 (0.906) TNF-alpha (pg/mL) Secreted into medium 16 0.376 (0.332) Produced, not secreted 12 3.027 (0.842) MMP-9 (pg/mL) Secreted into medium 7 6590.32 (2277.569) 25

PAGE 26

Table 3-2. IL-6, IL-8, and TNF-alpha produced in response to E. coli lysate stimulus E. coli concentration 25 ug/mL 50 ug/mL 100 ug/mL 250 ug/mL 500 ug/mL IL -6 (pg/mL) Secreted into medium 69.299 (18.361) 118.125 (46.016) 181.773 (64.202) 145.012 (74.176) 242.648 (89.704) Produced, not secreted 18.999 (9.993) 21.049 (10.474) 28.444 (8.538) 34.155 (8.025) 43.404 (9.816) IL-8 (pg/mL) Secreted into medium 1867.282 (355.511) 1920.657 (406.273) 1890.321 (416.393) 2262.975 (300.536) 2668.498 (532.571) Produced, not secreted 43.147 (13.516) 42.479 (4.614) 47.953 (8.594) 40.767 (5.543) 55.662 (11.539) TNF-alpha (pg/mL) Secreted into medium 1.813 (0.492) 1.880 (0.571) 1.956 (0.659) 1.664 (0.364) 2.090 (0.983) Produced, not secreted 6.438 (1.143) 5.745 (2.240) 4.407 (0.272) 5.379 (0.856) 6.471 (1.164) Results reported as mean (standard deviation) Table 3-3. IL-6 and IL8 produced in response to TNF-alpha stimulus Number of observations Mean (standard deviation) P value compared to baseline secretion IL-6 (pg/mL) Secreted into medium 4 3.410 (1.940) P < 0.0001 Produced, not secreted 4 21.070 (1.211) P < 0.0001 IL-8 (pg/mL) Secreted into medium 4 98.668 (15.588) P < 0.0001 Produced, not secreted 3 13.149 (0.173) P < 0.0001 26

PAGE 27

Table 3-4. Cytokine response with varying concentrations of ilomastat pretreatment Ilomastat concentration 5 ug/mL 15 ug/mL 25 ug/mL 35 ug/mL 50 ug/mL IL6 (pg/mL) E. coli = no Produced, not secreted 8.597 (0.073) 6.655 (0.216) 8.494 (0.072) 8.597 (1.086) 8.034 (0.289) E. coli = no Secreted into medium 3.954 (0.663) 3.098 (1.115) 3.870 (0.317) 3.57 (0.989) 3.977 (0.75) E. coli = yes Produced, not secreted 16.703 (0.737) 15.664 (1.469) 16.444 (1.104) 14.471 (1.539) 15.760 (0.209) E. coli = yes Secreted into medium 61.339 (6.15) 41.545 (15.215) 50.330 (8.309) 38.439 (27.720) 54.25 (16.64) IL 8 (pg/mL) E. coli = no Produced, not secreted 4.081 (0.577) 2.847 (0.392) 5.028 (0.380) 3.261 (0.583) 4.354 (0.192) E. coli = no Secreted into medium 31.211 (11.344) 39.376 (8.878) 38.817 (5.239) 39.977 (1.689) 42.163 (8.735) E. coli = yes Produced, not secreted 33.925 (3.064) 33.544 (4.685) 34.816 (1.803) 34.438 (5.588) 33.925 (3.424) E. coli = yes Secreted into medium 1097.17 (224.635) 956.914 (340.986) 1274.96 (52.655) 1206.32 (110.607) 1020.25 (395.262) Results reported as mean (standard deviation) Table 3-5. Difference in cy tokine response after pret reatment with ilomastat Ilomastat concentration = 0 Ilomastat concentration > 0 p-value IL6 (pg/mL) E. coli = no Produced, not secreted 22.182 (1.223) 18.995 (2.084) 0.08 E. coli = no Secreted into medium 9.870 (1.668) 9.860 (1.619) 0.99 E. coli = yes Produced, not secreted 2.109 (1.906) 20.954 (14.712) 0.03 E. coli = yes Secreted into medium 84.997 (8.169) 219.7 (72.315) 0.0002 IL 8 (pg/mL) E. coli = no Produced, not secreted 5.542 (0) 2.617 (1.849) 0.02 E. coli = no Secreted into medium 41.935 (1.043) 34.459 (13.64) 0.12 E. coli = yes Produced, not secreted 28.233 (4.040) 34.615 (19.196) 0.37 E. coli = yes Secreted into medium 892.07 (131) 695.97 (103.2) 0.25 Results reported as mean (standard deviation) 27

PAGE 28

Table 3-6. Protease response to E. coli stimulus after pretre atment with ilomastat Ilomastat concentration 0 ug/mL 5 ug/mL 15 ug/mL 25 ug/mL 35 ug/mL p M M P 9 E. coli = no Secreted into medium 7907.25 (2115.729) 4076 (567.447) 2464.75 (1828.12) 2287.5 (329.224) 3390 (822.549) E. coli = yes Secreted into medium 9577.5 (862.575) 6797 (481.045) 10859 (860.420) 11622 (1188.953) 15068.75 (3228.343) Results reported as integrated density of the bands (standard deviation) Table 3-7. Difference in cytokine respons e after pretreatment with progesterone Progesterone concentration = 0 Progesterone concentration > 0 p-value IL6 (pg/mL) TNF-alpha no Produced, not secreted 22.25 (0.4699) 28.57 (1.3884) 0.0014 TNF-alpha = no Secreted into medium 1.155 (0.181) 3.937 (2.083) 0.31 TNF-alpha = yes Produced, not secreted 21.07 (1.211) 25.184 (4.026) 0.007 TNF-alpha = yes Secreted into medium 3.41 (1.135) 4.043 (1.753) 0.44 IL 8 (pg/mL) TNF-alpha = no Produced, not secreted 8.824 (0.432) 3.668 (0.852) 0.0002 TNF-alpha = no Secreted into medium 9.329 (1.150) 457.74 (64.15) 0.007 TNF-alpha = yes Produced, not secreted 13.149 (0.173) 7.367 (1.410) < 0.0001 TNF-alpha = yes Secreted into medium 98.668 (15.588) 168.49 (586.08) 0.69 Results reported as mean (standard deviation) 28

PAGE 29

0 50 100 150 200 250 300 350 2550100250500E coli Concentration (ug/ml)IL-6 (pg/m l) IL-6 secreted IL-6 produced Figure 3-1. IL-6 produced in response to E. coli lysate stimulus 0 500 1000 1500 2000 2500 3000 3500 2550100250500E coli Concentration ( u g /ml ) IL -8 (pg/m l) IL-8 secreted IL-8 produced Figure 3-2. IL-8 produced in response to E. coli lysate stimulus 29

PAGE 30

-1 0 1 2 3 4 5 6 7 8 2550100250500E coli Concentration (ug/ml)TN F-alpha (pg/m l ) TNf-alpha secreted TNF-alpha produced Figure 3-3. TNF-alpha produced in response to E. coli lysate stimulus 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 0 1 10 100 1000TNF alpha conc (ng/ml)Intensity of the Band pMMP 9 Figure 3-4. Effect of TNF-alpha on pMMP-9 synthesis on END1/E6E7 cells. 1 ng/mL p<0.05; 10 ng/mL and 100 ng/mL p<0.001 comp ared to non-treated cells 30

PAGE 31

0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 0 ug/ml 5 ug/ml 15 ug/ml 25 ug/ml 35 ug/ml ilomastat concentrationAverage intensity of the bands Media only with E coil lysat e Figure 3-5. Effect of ilomastat and E. coli lysate on pMMP9 synthesis 31

PAGE 32

CHAPTER 4 DISCUSSION Cell Line Responsiveness and Cytokine Production The END1/E6E7 cells were chosen based on Fichorovas and Andersons hypothesis that the two epithelial cell ty pes (endocervical and ectocervical/vag inal) produce distinct profiles of cytokines and other immunological mediators. The different profiles are thought to be due to their different requirements for attraction and support of immune cell populations, tolerance of microorganisms, and maintenance of epithelial integrity. 20 Columnar epithelial cells of the human endocervix produce more cy tokines than do stratified squa mous epithelial cell of the ectocervix and vagina. The endocervical epithe lium may be more active because, being only a single layer thick, it is more vulnerable to pa thogen invasion and tissue injury. Cytokineactivated epithelial cells can shed inter-cellular a dhesion molecule-1 (ICAM1) in response to TNF-alpha into the cervico-vaginal environm ent. This molecule can block adhesion of leukocytes and therefore preven t extensive tissue damage in response to the invasion. ICAM1 can also bind to the target lymphocyte functi on associated antigen-1 (LFA-1) and deliver an activating signal without adhesion. The combination of these actions may have an important function in regulation of in flammation at this site. 20 Fichorova, et al., established th at this endocervical cell line expresses characteristics of simple columnar epithelium and produces several substances without stimulation, including IL-8 and Prostaglandin-E2. 13 They also established that the cells react to stimulation with TNF-alpha by up-regulating several cytoki nes including IL-8 and IL-6. 20 We found a dose response for the secretion of pMMP-9 up to a concentration of 100 ng/mL for TNF-alpha. One limitation of our study is the inability to also demonstrate a dose response for TNF-alpha for IL-6 and IL-8. During the course of the data analysis, la boratory personnel change d and the data from 32

PAGE 33

experiments involving varying co ncentrations of TNF-alpha on the production of down-stream cytokines was lost. In our study we also establis hed that the cells react to stimulation with a bacterial lysate, E. coli with increased secreti on of IL-6 and IL-8. The amounts of IL-6 and IL-8 produced but not secreted did not increase. This suggests that processing and secretion of these cytokines occur concordantly. In the initial studies by Fichor ova, et al., TNF-alpha was not spontaneously released into culture medium by the cell line, it was detected in low concentrations after lysis of cells with Triton X-100. 13 We were able to duplicate those resu lts in our study. In c ontrast, Fichorova did not demonstrate IL-6 and IL-8 in epithelial cell lysates at baseline, s uggesting that processing and secretion of these cytoki nes are concordant events. 13 In our study we we re able to detect both cytokines in the cell at baseline, although at significantly lower amounts. Limitations in Use of E. coli as a Stimulus Toll-like Receptor 4 (TLR4) is required for recognition of lipopolysaccaride (LPS), the most pro-inflammatory cell wall component of gram-negative bacteria. 23 Fichorova, et al., found that the cervical and vaginal epithelial cells expressed mRNA for TLR1, -2, -3, -5, and -6, but failed to express TLR4 and MD2, two essentia l components of the receptor complex for responses to LPS. However, they were capable of responding to whole gram-negative bacteria and bacterial lysates, as demonstrat ed by nuclear factor kappa beta (NF) activation and proinflammatory cytokine production. 23 This finding suggests that the lower female genital tract responds to gram-negative bacterial components in the absence of endot oxins recognition and TLR4 mediated signaling. Based on the lack of TLR4 we chose to use a lysate of our culture grown E. coli as the main stimulant throughout this st udy. This was problematic, as we encountered great variation in 33

PAGE 34

amounts of cytokine produced depending on whic h batch of the lysate was used. The initial experiment was replicated on four different occa sions. We were unable to interpret the complete set of results due to the large variation in amounts of cytokine produced on each experiment day. When the data was analyzed individually, we de termined that the data points from the two experiments that occurred with the same cell cu lture passage and same batch of lysate were comparable. If we had not found such variation there would have been more data points to analyze and potentially significant results wh ere none were found under these conditions. We still found variation, even with limiting the analysis to two expe riment days and repeating the ELISA on questionable results under the same cond itions. Another potential explanation for the variances we found would be mistakes made dur ing the processing of the ELISA plates and inconsistencies in laborator y techniques among personnel. Ilomastat IL-6 did not show a significant decrease at baseline when pretreated with ilomastat, but increased when stimulated with E. coli Similarly, IL-8 production significantly decreased at baseline but no significa nt difference in secretion at baseline or secretion and production when challenged was demonstrated. One pot ential explanation of this fi nding is that the overall model is not effective for representing the inflammati on associated with PTL and PPROM due to the absence of other cells and factor s that would be pres ent in-vivo (i.e., fibroblasts, macrophages, neutrophils, other mediators, a nd chemoattractants). Secondly, as mentioned above, these cells have already been shown to lack LPS receptors which may play a role in inflammation produced by endotoxins in vivo. Additionally, the flaws in e xperimental methods may also play a role.If we had been able to analyze more data point s the results obtained may have been different. Finally, ilomastat may work on substances that are not produced by the END1/E6E7 cell line. 34

PAGE 35

Progesterone Progesterone is the only substan ce so far to show a reduction in the rate of preterm birth from all causes. However, the mechanism of action is still not clear. We used progesterone in an attempt to hypothesize another po tential mechanism for its action. Our preliminary results of the cytokine production for the cells that received progesterone pretreatment were unexpected and difficult to interpret. Since we were unable to dem onstrate a consistent effect (3-7), we attempted to prove the presence or absence of progesterone receptors on the endocervical cells. On Realtime RNA PCR no progesterone receptor R NA was identified. There are two possible explanations: 1) the receptors may have been lost during the processing of the cells, or 2) they may not be present in significant amounts in viv o. This explanation would be another potential argument for why this model is not adequate for testing mechanisms for and prevention of preterm delivery. Metalloproteinases We found an increase in pMMP-9 secreted in response to ilomastat when the cells were stimulated with E. coli. We could argue that one possibl e explanation for this is that E. coli lead to TNF-alpha release by the cells, leading to an increase in secretion of pMMP-9. Ilomastat may have decreased the response we would have dem onstrated in its absence. However, this is unlikely to be a plausible explanation since no perceived difference between the amounts of pMMP-9 secreted in thos e cells with or without E. coli when they were not pretreated with ilomastat was identified. Anothe r potential explanation is that ilomastat inhibits TNF-alpha production and the conversion of pro to activ e MMP-9. Additionally, in the absence of E. coli lysate ilomastat down-regulated th e production of pMMP-9. It is possi ble that there is a decrease in pMMP-9 at 5 g/mL of ilomastat but not at higher c oncentrations. This may be explained by ilomastat inhibiting activation which would then l ead to a build up of the pro form of MMP-9. 35

PAGE 36

Conclusions We were able to develop an endocervical cell culture model (End 1/E6E7) of the inflammation. We succeeded in measuring the b ackground production of inflammatory cytokines and pMMP-9. We demonstrated a dose response to E. coli through increased IL-6, IL-8, TNFalpha, and pMMP-9. We were able to demonstrat e an increase in production in IL-6, IL-8, and pMMP-9 in response to TNF-alpha stimulus We demonstrated a dose response through increased pMMP-9 secretion, but failed to show a dose response for the cytokines due to lost data. The hypothesis that the cells lin ing the endocervical canal ar e involved in protecting the upper genital tract from ascending colonization and infection from the bacteria present in the lower genital tract is still valid. It is more likel y that this model was not adequate for testing the attenuation of inflammatory response produced by a stimulus. We successfully used an endocervical cell cultu re model to elicit inflammatory conditions that we hypothesize to be present in vivo. We evaluated substances to ameliorate the proinflammatory response the cells produced in re sponse to stimulus. Ilomastat and progesterone did not show a definitive amelioration of the inflammation. While several explanations may account for why ilomastat did not yield a consistent decrease in inflammation, we were able to show that these cells do not contain progesterone receptors. Whether this is a property of these cells in vivo or if the recept ors were lost in culture proces sing remains to be elucidated. With the information we obtained from th ese experiments we revised our original inflammatory pathway (Figure 4-1). Bacteria may act on the cervical stro ma and fibroblasts, as well as endocervical epit helial cells. The respons e to the inflammatory stimulation leads to increased production of TNF-alpha by both cel l groups. The TNF-alpha subsequently acts through paracrine as well as autocrine effects to further increase the produ ction of cytokines and 36

PAGE 37

proteases. Those substances can then act on the extracellular matrix/fetal membranes as well as the endometrium/myometrium to lead to PPROM and PTL. If progesterone works on the stromal cells to decrease production of TNF-alpha, then th is action may result in decreased production of the cytokines and proteases that lead preterm delivery. Further research in a lab capable of performing larger sample preparations under identical conditions may be of benefit in further evaluation of a cell model of inflammation as it relates to the prevention of preterm delivery. 37

PAGE 38

Endocervical cells TLR4 Cervical Stroma Fibroblasts TLR4 + LP S Other TNF-alpha By Progesterone B y LPS Paracrine Effects Autocrine Effects MMPs TNF-alpha By Progesterone By LPS Extracellular Matrix/Fetal Membranes PPROM Endometrium Myometrium IL-6, IL-8, MMP9 Bacteria Figure 4-1. Revised model for inflammatory cascade responsible for PTL and PPROM PTL 38

PAGE 39

LIST OF REFERENCES 1. Martin JA, Hamilton BE, Sutton P, et al. Natl Vital Stat Rep 2006;55(1):1-101. 2. Peltier MR. Immunology of term and pret erm labor. Reprod Biol Endocrinol 2003;1:122. 3. Gomez R, Romero R, Edwin SS, David C. Pathogenesis of preterm labor and preterm premature rupture of membranes associated w ith intraamniotic infect ion. Infect Dis Clin North Am 1997;11(1):135-76. 4. Krohn MA, Hillier SL, Lee ML, Rabe LK, Es chenbach DA.Vaginal Bacteroides species are associated with an increased rate of preterm delivery among women in preterm labor. J Infect Dis. 1991;1 64 ( 1 ): 88-93 5. Rizzo G, Capponi A, Vlachopoulou A, Angelini E, Grassi C, Romanini C. The diagnostic value of interleukin-8 and fetal fibronectin concentrations in cervical secretions in patients with preterm labor and intact membranes. J Perinat Med 1997;25:461-8. 6. McGregor JA, French JI, Jones W, Millia gan K, McKinney PJ, Patterson E, et al. Bacterial Vaginosis is associated with pr ematurity and vaginal fluid mucinase and sialidase: results of a contro lled trial of topical clindamycin cream. Am J Obstet Gynecol 1994;170:1048-60. 7. Vidaeff AC, Ramin SM. From Concept to Practice: The recent history of preterm delivery prevention. part II: subclinical inf ection and hormonal effects. Am J Perinat 2006;23(2):75-84. 8. Mesiano S. Myometrial pr ogesterone responsiveness and the control of human parturition. J Soc Gynecol Invest 2004;11:193-202. 9. Sexton, DJ, OReilly MW, Friel AM, Morrison JJ. Functional effects of 17alphahydroxyprogesterone caproate ( 17P) on human myometrial cont ractility in vitro. Reprod Biol Endocrinol 2004:2:80. 10. Elovitz A, Mrinalini C. Can medroxyproge sterone acetate alter toll-like receptor expression in a mouse model of intrauteri ne inflammation? Am J Obstet Gynecol 2005;193:1149-55. 11. Amory J, Lawler R, Shields L. Hydroxyproge sterone caproate and progesterone increase tumor necrosis factor-alpha production in lipopolysaccharide stimulated whole blood from non-pregnant women. J Perinat Med 2005;33:506-9. 12. Ness A, Dias T, Damus K, Burd I Berghella V. Impact of the recent randomized trials on the use of progesterone to prevent preterm birth: a 2005 follow-up survey. Am J Obstet Gynecol. 2006;195(4):1174-9. 39

PAGE 40

13. Fichorova RN, Anderson DJ. Differential e xpression of immunobiol ogical mediators by immortalized human cervical a nd vaginal epithelial cells. Biol Reprod. 1999;60(2):50814. 14. Galardy RE, Cassabonne ME, Giese C, et al. Lo w molecular weight inhibitors in corneal ulceration. Ann N Y Acad Sci. 1994;732:315-23. 15. Sotozono C, He J, Tei M, Honma Y, Kinoshita S. E ffect of metalloprot einase inhibitor on corneal cytokine expression after alkali injury. Invest Ophthalmol Vis Sci 1999;40(10):2430-4. 16. Daniels JT, Cambrey AD, Occleston NL, et al. Matrix metallopr oteinase inhibition modulates fibroblast-mediated matrix cont raction and collagen production in vitro. Invest Ophthalmol Vis Sci. 2003;44(3):1104-10. 17. Agren MS, Mirastschijski U, Karlsmark T, Saarialho-Kere UK. Topical synthetic inhibitor of matrix metalloproteinases delays epidermal regeneration of human wounds. Exp Dermatol. 2001;10(5):337-48. 18. Meis PJ, Klebanoff M, Thom E, Dombro wski, MP, Sibai B, Moawad AH, et al. Prevention of Recurrent Preterm Deliver y by 17 Alpha-Hydroxyprogesterone Caproate. N Engl J Med 2003;348:2379-85. 19. Szekeres-Bartho J, Barakonyi A, Par G, Polgar B, Palkovics T, Szereday L. Progesterone as an immunomodulatory molecule. Int Immunopharmacol. 2001; 1 ( 6 ): 1037-48 20. Fichorova RN Rheinwald JG Anderson DJ. Generation of papillomavirus-immortalized cell lines from normal human ectocervical, e ndocervical, and vaginal epithelium that maintain expression of tissue-sp ecific differentiation proteins. Biol Reprod. 1997;57(4):847-55. 21. Ladwig GP, Robson MC, Liu R, Kuhn MA, Mu ir DF, Schultz GS. Ratios of activated matrix metalloproteinase-9 to tissue inhib itor of matrix metall oproteinase-1 in wound fluids are inversely correlated w ith healing of pressure ulcers. Wound Repair Regen. 2002;10(1):26-37. 22. Petrini JR, Callaghan WM, Klebanoff M, et al. Estimated effect of 17 alphahydroxyprogesterone caproate on preterm birth in the United States. Obstet Gynecol. 2005;105(2):267-72. 23. Fichorova RN, Cronin AO, Lien E, Anderson DJ, Ingalls RR. Response to Neisseria gonorrhoeae by Cervicovaginal Epithelial Cell s Occurs in the Absence of Toll-Like Receptor 4-Mediumted Signaling. J of Immun 2002;168:2424-32. 40

PAGE 41

BIOGRAPHICAL SKETCH Kathleen Mayor-Lynn received her Bachelor of Science degree from the University of Florida in 1996. She then moved on to earn her Do ctor of Medicine degree from the University of Miami School of Medicine in 2002. Her obstetrics and gynecology residency training occurred at Jackson Memorial Hospital in Mi ami, FL, from 2002 to 2006. She is currently in subspecialty fellowship training in maternal-fetal medicine at the University of Florida with an anticipated completion date of June 2009. Du ring the course of her fellowship, she also participated in the Advanced Postgraduate Prog ram in Clinical Investig ation, where she earned a masters degree in Medical Sciences with a concen tration in Clinical Investigation in December 2008. 41