The Effects of Lactobacillus rhamnosus GR-1 on

Cytokines/Chemokines and Prostaglandins in Human Amnion Cells

by

Rebecca Jean Elizabeth Koscik

A thesis submitted in conformity with the requirements

for the degree of Masters of Science

Department of Physiology

University of Toronto

© Copyright by Rebecca Jean Elizabeth Koscik (2011)

ii

The Effects of Lactobacillus rhamnosus GR-1 on

Cytokines/Chemokines and Prostaglandins in Human Amnion Cells

Rebecca Koscik

Masters of Science

Department of Physiology

University of Toronto

2011

Abstract

The incidence of preterm labor has risen over recent decades and preventative treatments are

ineffective. Associated with a 40% increased risk of preterm birth, bacterial vaginosis is

characterized by a decrease in lactobacilli and an increase in pathogenic bacteria in the vaginal

microbiota. Ascent of bacterial products to the intrauterine environment stimulates cytokine and

prostaglandin secretion from invading immune cells and gestational tissues. Probiotic

lactobacilli modulate the immune responses in mouse macrophages and human placental

trophoblast cells. The focus of this thesis was to determine the influence of Lactobacillus

rhamnosus GR-1 (GR-1) on cytokines and prostaglandins which are part of the activated

pathway in infection and/or inflammation-mediated preterm labour. Pro-inflammatory cytokines

were decreased by GR-1. The release of several chemokines and prostaglandin E2 were elevated

by GR-1. It is possible that GR-1 may enhance the host defense barriers of the amnion to

pathogenic bacteria.

iii

“When is it that nature does anything in vain?”

Paraphrased from Sir Isaac Newton

iv

Dedicated to those who guided me with their years of wisdom, intellectual creativity and passion

for academia; to those who insisted on lunch breaks and offered constant encouragement and

reassurance during my midnight trips to the laboratory; to wiped tears, shared happiness and

laughter; and to my family whose love and support has made me capable of overcoming

challenges and of celebrating successes.

v

Table of Contents

Abstract .......................................................................................................................................... ii

Table of Contents ...........................................................................................................................v

List of Abbreviations ................................................................................................................... ix

List of Figures .............................................................................................................................. xii

Chapter 1 ........................................................................................................................................1

General Introduction .....................................................................................................................2

1.1 Human Parturition ............................................................................................................2

1.1.1 Anatomy ..................................................................................................................4

1.1.2 Mechanisms of Human Parturition ......................................................................9

1.2 Preterm Birth ...................................................................................................................12

1.2.1 Epidemiology ........................................................................................................12

1.2.2 Etiology .................................................................................................................13

1.2.3 Current Diagnostic Strategies and Treatments for Preterm Birth .................13

1.3 Infection and Inflammation ............................................................................................15

1.3.1 Vaginal Microbiota ..............................................................................................16

1.3.2 Routes of Intrauterine Infection .........................................................................17

1.3.3 Bacterial Vaginosis...............................................................................................18

1.3.4 Current Diagnostic Strategies and Treatments for Bacterial Vaginosis ........19

1.4 Probiotics ..........................................................................................................................22

1.4.1 Lactobacillus Species as Candidate Probiotics for use During Pregnancy. ....23

1.4.2 The Potential Mechanisms of Probiotics............................................................26

1.5 Cytokines and Chemokines .............................................................................................27

1.5.1 Pro-inflammatory Cytokines ..............................................................................28

1.5.2 CXC and CC Chemokines...................................................................................30

vi

1.5.3 Immune Cell Activating Cytokines ....................................................................31

1.6 Prostaglandins ..................................................................................................................34

1.6.1 Prostaglandin Production ...................................................................................34

1.6.2 The Role of Amnion Produced Prostaglandins in Preterm Labour ................39

1.7 Summary ...........................................................................................................................41

Chapter 2 ......................................................................................................................................42

Hypotheses, Rationale and Specific Aims ..................................................................................43

2.1 Overall Hypothesis ...........................................................................................................43

2.1.1 Rationale ...............................................................................................................43

2.2 Chapter 3: Effect of Lactobacillus rhamnosus GR-1 on Cytokines and

Chemokines in Human Amnion Cells at Term .............................................................44

2.3 Chapter 4: Effect of Lactobacillus rhamnosus GR-1 on Prostaglandin Output

and Prostaglandin Metabolizing Enzymes in Human Amnion Cells at Term ...........45

Chapter 3 ......................................................................................................................................46

Effect of Lactobacillus rhamnosus GR-1 on Cytokines and Chemokines in Human

Amnion Cells at Term .................................................................................................................47

3.1 Introduction ......................................................................................................................47

3.2 Methodology .....................................................................................................................48

3.2.1 Placenta Collection...............................................................................................48

3.2.2 Mixed Amnion Epithelial and Mesenchymal Cell Culture ..............................48

3.2.3 Lactobacilli Preparation ......................................................................................49

3.2.4 Treatment Protocol ..............................................................................................50

3.2.5 Lactate Dehydrogenase Assay ............................................................................50

3.2.6 Cytokine and Chemokine Measurement ...........................................................51

3.2.7 Statistical Analysis ...............................................................................................52

3.3 Results ...............................................................................................................................52

3.4 Comment ...........................................................................................................................57

vii

Chapter 4 ......................................................................................................................................70

Effect of Lactobacillus rhamnosus GR-1 on Prostaglandin Output and Prostaglandin

Metabolizing Enzymes in Human Amnion Cells at Term .......................................................71

4.1 Introduction ......................................................................................................................71

4.2 Methodology .....................................................................................................................72

4.2.1 Placenta Collection...............................................................................................72

4.2.2 Amnion Mixed Epithelial and Mesenchymal Cell Culture ..............................72

4.2.3 Lactobacilli Preparation ......................................................................................73

4.2.4 Treatment Protocol ..............................................................................................73

4.2.5 Lactate Dehydrogenase Assay ............................................................................73

4.2.6 Prostaglandin E2 Measurement ..........................................................................73

4.2.7 Prostaglandin Metabolizing Enzyme Expression Measurement .....................74

4.2.8 Placental Trophoblast Cell Culture ...................................................................75

4.2.9 Statistical Analysis ...............................................................................................76

4.3 Results ...............................................................................................................................76

4.4 Comment ...........................................................................................................................78

Chapter 5 ......................................................................................................................................91

Overall Conclusions and Future Directions ..............................................................................92

5.1 Overall Conclusions .........................................................................................................92

5.1.1 Potential Increases in TNFα and IL-1β Antagonists ........................................92

5.1.2 Recruitment of Immune Cells .............................................................................93

5.1.3 Modulation of Immune Cell Activity .................................................................93

5.1.4 Stimulation of Prostaglandin E2 .........................................................................95

5.1.5 Additional Effects of Prostaglandins on the Amnion .......................................97

5.1.6 Lactobacillus rhamnosus GR-1 Supernatant Metabolite(s) .............................98

viii

5.1.7 Lactobacillus rhamnosus GR-1 and Prevention of Preterm Birth

Associated with Infection and/or Inflammation................................................99

5.1.8 Limitations ..........................................................................................................100

5.2 Future Directions ...........................................................................................................102

References ...................................................................................................................................105

ix

List of Abbreviations

ACTH Adrenocorticotropic Hormone

BSA Bovine Serum Albumin

cAMP cyclic Adenosine Monophosphate

CAP Contraction Associated Protein

CCB Calcium Channel Blocker

cGMP cyclic Guanosine Monophosphate

CRH Corticotropin-Releasing Hormone

DMEM Dulbecco Modified Eagle Medium

ECL Enhanced Chemiluminescence

EIA Enzyme Immuno-Assay

ELISA Enzyme-Linked ImmunoSorbent Assay

ER Estrogen Receptor

FBS Fetal Bovine Serum

G-CSF Granulocyte Colony Stimulating Factor

GM-CSF Granulocyte-Macrophage Colony Stimulating Factor

HPA Hypothalamic-Pituitary-Adrenal

HSD Hydroxysteroid Dehydrogenase

IFNγ Interferon γ

IL Interleukin

x

LDH Lactate Dehydrogenase

LTA Lipoteichoic Acid

LPS Lipopolysaccharide

MAMP Microbe-Associated Molecular Pattern

MCP Monocyte Chemotactic Protein

MgSO4 Magnesium Sulphate

MLCK Myosin Light Chain Kinase

MMP Metalloproteinase

MRS de Man, Rogosa, and Sharpe media

NCS Newborn Calf Serum

NS Not significant

NSAID Non-Steroidal Anti-Inflammatory Drug

PAMP Pathogen-Associated Molecular Pattern

PBS Phosphate Buffered Saline

PBST Phosphate Buffered Saline + Tween

PG Prostaglandin

PGDH 15-Hydroxyprostaglandin Dehydrogenase

PGES Prostaglandin E Synthase

PGHS Prostaglandin endoperoxidase H Synthase

PR Progesterone Receptor

xi

PROM Premature Rupture Of the Membranes

ra receptor antagonist

RANTES Regulated on Activation, Normal T Cell Expressed and Secreted

RIPA RadioImmunoPrecipitation Assay

ROD Relative Optical Density

SEM Standard Error of the Mean

SNP Single Nucleotide Polymorphism

TLR Toll-Like Receptor

TNFα Tumor Necrosis Factor α

xii

List of Figures

Figure 1.1: The layers of the fetal membranes: the amnion and chorion ...................................... 8

Figure 1.2: Potential sites of bacterial infection amongst the intrauterine tissues. ...................... 21

Figure 1.3: Proposed pathway of infection and/or inflammation-mediated preterm birth .......... 33

Figure 1.4: Effects of stimulants on the prostaglandin production pathway in the amnion. ....... 38

Figure 3.1: Experimental design .................................................................................................. 60

Figure 3.2: The effect of treatments on amnion cell toxicity ....................................................... 61

Figure 3.3: Lipoteichoic and lipopolysaccharide dose optimization ........................................... 62

Figure 3.4: Time course responses of Interleukin-6 and Interluekin-8 ........................................ 63

Figure 3.5: Lactobacillus rhamnosus GR-1 dilution optimization .............................................. 64

Figure 3.6: Correlation between ELISA and bioplex assay methodologies ................................ 65

Figure 3.7: The effects of Lactobacillus rhamnosus GR-1 on Tumor Necrosis Factor - α

concentration in human amnion cells ........................................................................................... 66

Figure 3.8: The effects of Lactobacillus rhamnosus GR-1 on Interleukin-6 concentration in

human amnion cells ...................................................................................................................... 67

Figure 3.9: The effect of Lactobacillus rhamnosus GR-1 on chemokine concentrations in human

amnion cells .................................................................................................................................. 68

Figure 3.10: The effects of Lactobacillus rhamnosus GR-1 on immune cell activating cytokines

....................................................................................................................................................... 69

Figure 4.1: Experimental design .................................................................................................. 83

Figure 4.2: The effect of IL-1β, dexamethasone (Dex), lipopolysaccharide (LPS) and de Man,

Rogosa, and Sharpe (MRS) on prostaglandin production in human amnion cells ....................... 84

xiii

Figure 4.3: The effect of Lactobacillus rhamnosus GR-1 on Prostaglandin E2 concentration in

medium from cultured human amnion cells ................................................................................. 85

Figure 4.4: The effects of Lactobacillus rhamnosus GR-1 on PGHS-2 expression in human

amnion cells .................................................................................................................................. 86

Figure 4.5: The effect of Lactobacillus rhamnosus GR-1 on cytosolic PGES expression in

human amnion cells ...................................................................................................................... 87

Figure 4.6: The effect of Lactobacillus rhamnosus GR-1 on microsomal PGES1 expression in

human amnion cells ...................................................................................................................... 88

Figure 4.7: The effects of Lactobacillus rhamnosus GR-1 on microsomal PGES2 expression in

human amnion cells ...................................................................................................................... 89

Figure 4.8: Correlation between PGES expression and PGE2 production in human amnion cells

....................................................................................................................................................... 90

Figure 5.1: The effects of Lactobacillus rhamnosus GR-1 on inflammatory mediators in human

amnion cells ................................................................................................................................ 101

1

Chapter 1

General Introduction

2

Chapter 1

General Introduction

Preterm birth accounts for up to 8% and 13% of all births in Canada and the United States of

America respectively (Goldenberg et al., 2008; Public Health Agency of Canada, 2008). Thirty

percent of all preterm births are associated with infection and/or inflammation. Despite ongoing

studies of the mechanisms underlying infection and/or inflammation-mediated preterm birth, no

effective preventative treatment exists. Preterm birth results in neonatal morbidities that last into

adulthood, cost an estimated $26 billion in medical care annually in the United States of America

and place an emotional and psychological burden on families that is immeasurable by any

monetary means. Probiotics are increasingly being investigated as preventative therapy due to

their ability to modulate immune responses across many tissues and systems. However, the

effects of probiotics during pregnancy are largely uninvestigated. Therefore, the focus of this

thesis was to determine the possibility of Lactobacillus rhamnosus GR-1 as a potential

preventative treatment by measuring its effects on cytokines, chemokines and prostaglandins that

are known to be elevated in association with preterm delivery in human amnion cells.

1.1 Human Parturition

Human parturition occurs as the result of a series of feed-forward paracrine, autocrine and

endocrine events resulting in myometrial contractions, remodeling of the cervix and weakening

and rupture of the fetal membranes, ultimately leading to expulsion of the fetus from the uterus

(Challie et al., 2002). During pregnancy, the uterus and cervix work together to promote a

quiescent state in order to maintain gestation. Throughout gestation, the myometrium transitions

through several phases: quiescence, activation, stimulation and expulsion (Challis et al, 2002).

During these phases, multiple factors are recruited and interact to promote synchronous uterine

activity. These factors include uterine stretch due to fetal growth and biochemical factors

associated with an increase in the activation of the fetal hypothalamic-pituitary-adrenal (HPA)

axis that occurs in late gestation.

A quiescent state is maintained for approximately 95% of gestation by high levels of circulating

progesterone, locally produced prostacyclin, relaxin, nitric oxide and parathyroid hormone-

related protein that maintain myometrial inactivity (Challis et al., 2002). Many of these agents

3

inhibit muscle contractions by activating downstream pathways that increase cytosolic cAMP

and/or cGMP. Consequentially, increases in cAMP and/or cGMP cause intracellular calcium to

be sequestered and the inactivation of myosin light chain kinase (MLCK). Without the activity of

MLCK, myosin is not phosphorylated and the actin-myosin cross-bridges in the myometrial

smooth muscle remain in the relaxed state (Challis et al., 2000). Progesterone inhibits the

actions of estrogen by preventing expression of the estrogen receptor, (ER)-α to also maintain

quiescence (Mesiano et al., 2002).

As gestation progresses the uterine myometrium transitions to an active state. Throughout this

transition, uterine stretch caused by the growing fetus leads to up-regulation of contraction-

associated protein (CAP) genes (Challis et al, 2002). As a result, the myometrium becomes

capable of responding to uterotonins, or stimulants of myometrial contractility.

Uterotonins are molecules that alter the myometrial phenotype to an active state making it

capable of producing synchronous contractions. Oxytocin, prostaglandins and corticotropin-

releasing hormone (CRH) promote myometrial contractility and their concentrations rise

throughout gestation. As gestation progresses, the HPA axis of the fetus becomes increasingly

active resulting in increased adrenal cortisol production. Cortisol promotes the activity of

estrogen and prostaglandins while decreasing the sensitivity of the myometrium to progesterone.

The final phase of expulsion occurs as a result of powerful myometrial co-ordinated contractions

that are capable of expelling the fetus and placenta from the uterus.

As the myometrium progresses through different phenotypic states, other gestation tissues,

namely the cervix and the fetal membranes, undergo alterations in preparation for labour.

During pregnancy the cervix remains closed while the fetal membranes maintain the volume of

the amniotic fluid and provide mechanical strength necessary in pregnancy. As gestation

progresses, prostaglandins increase metalloproteinase (MMP) expression in the amnion. MMPs

degrade collagen molecules that constitute a large proportion of the extracellular matrix of both

the cervix and the fetal membranes. This results in remodeling of the cervix and weakening of

the fetal membranes (Menon et al., 2004; Yoshida et al., 2002).

Therefore, parturition results from the interaction of two interdependent pathways. The first is

mechanical stretch of the uterus and the other is biochemical factors and endocrine activities

4

associated with increased activation of the HPA axis of the fetus. In cases of membrane rupture

or infection, the amnion contributes a large pool of prostaglandins that may result in premature

myometrial activation, cervical ripening and membrane weakening.

1.1.1 Anatomy

1.1.1.1 Uterine Myometrium

The uterine myometrium is the strong, densely packed smooth muscle layer of the uterus that

gives much of the bulk to this organ. In the non-pregnant uterus, this layer contains blood and

lymphatic vessels as well as cholinergic, sympathetic and peptidergic nerve innervations. The

myocytes of the uterus, like other smooth muscle cells, are highly plastic and are capable of

transitioning through several phenotypes as gestation progresses. These include the proliferative,

synthetic and contractile stages leading to preparation for expulsion of the fetus during

parturition (Shynlova et al., 2009). Throughout pregnancy the control of the myometrial

phenotype switches from a global autonomic control system with the loss of cholinergic,

peptidergic and sympathetic nerves from the myometrium to one of humoral or local control by

prostaglandins, oxytocin and steroid hormones (Riemer et al., 1998).

The first phase of uterine growth during pregnancy is due to myometrial hyperplasia and is under

the control of estrogen and progesterone. Estrogen and progesterone simultaneously

hyperpolarize myometrial cells making them less responsive to contractile agonists (Riemer et

al., 1998; Shynlova et al., 2004; Shynlova et al., 2005). Myometrial quiescence is partially

maintained by high levels of oxytocinase and 15-hydroxyprostaglandin dehydrogenase (PGDH)

expression. These proteins metabolize oxytocin and prostaglandins, respectively (Mitchell et al.,

1995). Studies in pregnant rats have characterized the phenotypic changes that occur across

pregnancy (Shynlova et al., 2009). As uterine tension increases, stretch and progesterone act in

concert to induce growth and remodeling resulting in hypertrophy and matrix changes conducive

to the contractile phase of myometrial activity (Shynlova et al., 2009). The switch to the

contractile phenotype is characterized by increased expression of basement membrane proteins

such as fibronectin, laminin β2 and collagen IV within myometrial cells. Near the end of

gestation, uterine growth ceases while fetal growth continues, resulting in biomechanical

distension of the uterus. This distension is necessary, but not sufficient to cause up-regulation of

CAP gene expression as demonstrated by the lack of up-regulation of these genes in the absence

5

of progesterone withdrawal (Shynlova et al., 2009). Therefore, the phenotype of the uterus is

dependent upon both the tension of the myometrium and the endocrine environment that is

altered throughout pregnancy and at the onset of parturition.

1.1.1.2 Fetal Membranes/Amnion

The fetal membranes consist of the chorion and the amnion. The fetal membranes derive from

fetal origins and thus have the same chromosomal sex of the fetus. The amnion begins to form

eight days after conception from the ectodermal cell nest in the dorsal aspects of the zygote

(Calvin et al., 2007). The amniotic cavity rapidly increases in size as amniotic fluid begins to

surround the developing fetus. By 10-12 weeks of gestation, the amnion comes into contact with

the chorion forming the chorioamnion membrane. As the amniotic cavity further expands, the

fetal membranes contact the decidua along the uterine walls essentially sealing the endometrial

cavity by 16 weeks gestation (Calvin et al., 2007). Between the two fetal membranes, a layer of

phospholipids running parallel to the two tissues provides a surfactant-like effect to prevent

shearing of the membranes during movement. By the end of gestation, the fetal membranes

encompass an area of 1000-1200 cm2, resistant to the increased pressure placed upon them by

increased volumes of amniotic fluid and by fetal movements that become more vigorous as

gestation progresses (Bryant-Greenwood, 1998; Myatt et al., 2010).

The amnion has no blood supply or nerve innervation and as such relies on amniotic fluid and

the chorion for nutrient supply and waste transport (Parry et al., 1998). It is a largely an acellular

tissue in comparison to the chorion and provides the tensile strength and elasticity required to

withhold the amniotic fluid (Bryant-Greenwood, 1998). The amnion consists of two major

portions. The first is the placental amnion, defined as the portion of the amnion in direct contact

with the chorion of the placental chorion plate and the second portion consists of the remaining

amnion, termed the membranous amnion (Calvin et al., 2007).

The amnion consists of five layers starting from the visceral side or that in contact with the

amniotic fluid, fetus, and umbilical cord to the outer layer, or that in contact with the chorion: 1)

Epithelium, 2) Basement Membrane, 3) Compact Layer, 4) Fibroblast Layer, and 5) Spongy

Layer or Zona spongiosa (Figure 1.1, page 8; Bourne, 1962; Calvin et al., 2007; Parry et al.,

1998).

6

The epithelial layer of the amnion consists of cuboidal cells that are typically mononuclear with

dense granular cytoplasm with many vacuoles and few small mitochondria (Bourne, 1962). The

cells have a slight convex shape on the most apical surface in contact with the amniotic fluid and

the edges form a brush border of microvilli (Bourne, 1962). Many vacuoles are found in amnion

epithelial cells that are connected by fine channels close to adjacent cell membranes (Bourne,

1962). These tunnels and channels form an intricate network that allows amnion epithelial cells

to communicate with the extracellular matrix (Bourne, 1962). Additionally, both murine and

human amnion epithelial cells contain tight junctions (Kobayashi et al., 2009; Kobayashi et al.,

2010a). Amnion epithelial cells produce MMP-1, -2, -9, and collagens I, III, and IV, as well as

the non-collagenase proteins laminin, fibronectin and nidogen that compose the other amnion

layers (Aplin et al., 1985; Bourne, 1962; Parry et al., 1998). Epithelial cells attach to a basement

membrane composed of a network of collagen III, IV and V that act as a scaffold for the

assembly of laminin, fibronectin, heparin sulphate proteoglycan and nidogen (Bryant-

Greenwood, 1998; Parry et al., 1998). The epithelial cells attach to the basement membrane

through digitations of blunt processes from both the reticular network of the second layer, and

the first layer cells (Bourne, 1962).

The third layer of the amnion is the compact layer, a dense acellular network of reticular fibres.

It is composed of fibronectin, collagen I, III, and IV with low levels of collagens V, VI, and VII

(Bryant-Greenwood, 1998; Parry et al., 1998). Collagens I and II form large parallel bundles

that sustain mechanical integrity and give the amnion much of its tensile strength. The compact

layer is relatively permeable to chemotaxis of macrophages and rarely has antigen presenting

cells present unless they are actively phagocytic (Bourne, 1962).

The thickest layer of the amnion, the fibroblast layer is composed of the second major cell type

constituting the amnion, the mesenchymal cells. This layer is relatively weak compared to the

compact layer and is composed of a loose network of non-collagenous glycoproteins, reticulin,

collagens I, III, and IV, lamninin, nidogen, and fibronectin and scattered mesenchymal cells

(Bourne, 1962; Parry et al., 1998). This layer contains Hofbauer cells, macrophages of placental

origin where they exist as residential cells capable of phagocytosis (Bourne, 1962).

The outer most layer of the amnion, the spongy layer is in direct contact with the chorion. It is

composed of loose bundles of reticulin composed of collagens I, III, and IV as well as hydrated

7

proteoglycans covered by mucin (Bourne, 1962; Parry et al., 1998). This layer is prone to edema

as proteoglycans absorb water causing increases in the thickness of the amnion (Bourne, 1962).

The presence of mucin and the hydrophilic characteristics of proteoglycans allow the amnion to

move along the chorion while absorbing physical stress (Bryant-Greenwood, 1998). This layer

also contains resident Hofbauer cells (Bourne, 1962).

The amnion is responsive to many physiologic factors including cytokines, growth factors,

bacterial endotoxins, and glucocorticoids. It is equipped with receptors necessary to respond to

these factors including cytokine receptors such as CXCR2, bacterial endotoxins, namely toll-like

receptor (TLR) 2, TLR4, the family of prostaglandin receptors EP1, EP1, EP3, EP4, and F, as

well as the glucocorticoid receptor (Kallapur et al., 2009; Sun et al., 1996; Unlugedik et al.,

2010). Progesterone receptors are absent or expressed in very low levels in the amnion as

mRNA for both progesterone receptor (PR)-A and PR-B are undetectable (Merlino et al., 2009).

The amnion expresses natural antimicrobials and innate immune molecules that have anti-

bacterial, anti-viral, and anti-fungal activity. These include human α-defensins 1, 2 and 3,

human β-defensins 1, 2 and 3 as well as elafin and secretory leukocyte protease inhibitor (King

et al., 2007). As important components of the innate immune system, these molecules are part of

the immune defense against uterine infection during menstruation, pregnancy and labour (King

et al., 2007). The amnion produces pro-inflammatory and anti-inflammatory cytokines as well as

chemokines and steroids. However, a major product of the amnion is prostaglandin, in particular

prostaglandin (PG) E2 and to a lesser degree, PGF2α.

The amnion is in an ideal anatomical position to respond to many different molecules allowing

for signaling between the maternal and fetal compartments during pregnancy. Direct contact

between the amnion and the amniotic fluid provides not only a pathway for molecules secreted

from the fetus, but also foreign bacterial products during microbial invasion of the amniotic

cavity to influence the release of cytokines, chemokines and prostaglandins from the epithelial

layer of the amnion. In addition, the amnion relies on the chorion for nutrients and waste

elimination. Consequentially, ascending bacteria and bacterial products may influence the

mesenchymal cells of the amnion in close proximity to the chorion. Therefore, an intimate

relationship between the amnion, fetal environment and chorion exists which differentially

dictate the response and role of the amnion throughout gestation and parturition.

8

Figure 1.1: The layers of the fetal membranes: the amnion and chorion

The inner layer of the fetal membranes, the amnion is an avascular tissue composed of five layers: 1)

Epithelium, 2) Basement membrane, 3) Compact Layer, 4) Fibroblast Layer, 5) Spongy Layer of Zona

Spongiosa.

[The image has been modified from American Journal of Obstetrics and Gynecology, Vol 79. Bourne

GL. Microscopic anatomy of human amnion and chorion, 1070-1073. Copyright Elsevier (1960) and

appears here with the permission of the journal.]

2 1

3

4

5

9

1.1.2 Mechanisms of Human Parturition

1.1.2.1 Uterine Stretch

Stretch induces contractions in many muscle systems including the myocardium and the

myometrium (Riemer et al., 1998). Ultimately, the ability of the uterine myometrium to contract

is regulated by the availability of intracellular calcium. The myometrium consists of smooth

muscle cells that contain actin-myosin bridges. As intracellular calcium levels increase, MLCK

is activated resulting in increased adenosine triphosphatase activity of the myosin head which

becomes phosphorylated and the muscle contracts. Agents that decrease intracellular calcium

levels and/or increase cyclic adenosine monophosphate (cAMP) or cyclic guanosine

monophosphate (cGMP) levels that sequester intracellular calcium, result in relaxation of the

uterine myometrium. In contrast, agents that increase intracellular calcium result in uterine

myometrial contraction.

As gestation progresses, the growth of the fetus causes biomechanical distension of the uterus

leading to both hypertrophy and hyperplasia of the myometrial smooth muscle cells in the

presence of progesterone (Challis et al., 2005). Throughout gestation, circulating progesterone

maintains uterine quiescence through nuclear PR-B expressed on both the decidua and the nuclei

of myometrial cells (Merlino et al., 2007). Progesterone maintains quiescence by restricting

transcription of CAPs genes which translate into proteins involved in myometrial stimulation and

include the oxytocin receptor, progesterone receptors, the gap junction protein Cx43, increasing

expression of PGDH, an enzyme responsible for degradation of PGE2 in the chorion and

myometrium and inducing phosphokinase A signaling that promotes smooth muscle relaxation

by inhibiting phospholipase C activity (Merlino et al., 2007). Unlike other mammalian species,

circulating progesterone levels in the human do not decrease and estrogen levels do not increase

abruptly near the end of gestation (Challis et al., 2001). However, a functional withdrawal of

progesterone accompanied by functional estrogen activation occurs at term. This results in a

conversion of the myometrial phenotype from quiescence to activation. Throughout gestation,

PR-A expression increases. PR-A resides in the nucleus of myometrial cells and thus is ideally

located to inhibit the expression of PR-B (Merlino et al., 2007). As a result, the effects of PR-B

that maintain uterine quiescence until late gestation are withdrawn, including the inhibitory

effects on estrogen receptor (ER)-α expression (Merlino et al., 2007; Mesiano et al., 2002).

10

Circulating estrogen subsequently increases uterine contractility through the up-regulation of

CAP genes as the inhibitory actions of progesterone are functionally withdrawn (Kamel, 2010;

Mesiano et al., 2002). Estrogen, in conjunction with mechanical distension works in concert to

activate the myometrium, thus allowing for high amplitude, high-frequency contractions to occur

in the next phenotypic state of the myometrium.

Up-regulation of CAP genes results in increased expression of three major groups of proteins

necessary for activation and subsequent contraction of the myometrium: 1) Ion channels, 2) Gap

junction proteins, and 3) Contractile agonist receptors (Challis et al., 2000; Challis et al., 2002).

Ion channels determine the resting membrane potential of myocytes and thus determine the

contractile response to contractile agonists. Gap junction proteins, specifically C43, synchronize

longitudinal co-ordinated contractions over the entire uterine myometrium during labor (Merlino

et al., 2007; Riemer et al., 1998). Contractile agonist receptors, including the oxytocin receptor

as well as the prostaglandin receptors F, EP1 and EP3 increase the sensitivity of the myometrium

to oxytocin and prostaglandins that act as uterotonins during myometrial activation. Stretch also

causes increases in myometrial prostaglandin H synthase (PGHS)-2 expression (Riemer et al.,

1998). Experiments in unilateral pregnant rats demonstrate the inability of experimentally

induced stretch to increase the expression of CAP genes in the non-gravid horn, unlike results in

the gravid horn (Ou et al., 1998). This suggests that CAP gene expression is dependent on

stretch as well as the endocrine environment, which is largely governed by the fetal genes and

activation of the HPA axis (Shynlova et al., 2009).

1.1.2.2 Hypothalamic-pituitary-adrenal Axis

A key mechanism for parturition is the endocrine action of increased glucocorticoids and their

influence on the gestational tissues. Across many species, including humans, the fetal HPA axis

matures near the end of gestation. Glucocorticoids are vital in the maturation of organs

necessary for survival outside of the womb and also play a crucial role in determining the length

of gestation (Smith et al., 2007). Experiments in sheep have characterized the mechanisms in the

fetal brain that account for an increase in HPA activity. Initially, CRH mRNA levels increase in

the paraventricular nucleus of the hypothalamus as adrenocorticotrophic hormone (ACTH)

concentrations in the fetal circulation rise (Challis et al., 2000). The fetal sheep adrenal gland

increases expression of enzymes involved in cortisol production as ACTH receptor expression is

11

up-regulated (Holloway et al., 2000). In this species, the net result of the maturing fetal HPA

axis is increased sensitivity of the fetal adrenal gland to ACTH, and greater production of

cortisol (Challis et al., 2001). In primates, similar increases in cortisol occur in late gestation. In

addition to increased adrenal cortisol output in the human fetus, glucocorticoids influence human

gestational tissue (Karalis et al., 1996). In the human placenta, prostaglandins down-regulate the

expression of 11β-hydroxysteroid dehydrogenase (HSD) 2 and increase the expression of 11 β -

HSD1 thereby decreasing conversion of cortisol to cortisone (Alfaidy et al., 2001). Similarly, in

the fetal membranes cortisone is produced through the activity of 11β-HSD1 (Sun et al., 2003;

Myatt et al., 2010). Near the end of term maternal blood concentrations of CRH increase and

CRH-binding protein levels decrease. Altogether, the influence of CRH and glucocorticoids

increase near the end of term and activate downstream cascades that lead to parturition onset.

This activation includes a feed-forward loop initiated by increased glucocorticoid production and

results in increased prostaglandins.

Elevated CRH levels stimulate prostaglandin production by modulating the levels of

prostaglandin metabolizing enzymes expressed in the amnion, chorion and placenta trophoblast

cells. PGHS-2 expression in the fetal membranes and placenta is increased by CRH (Alvi et al.,

1999; Challis et al., 2002). Simultaneously, CRH and cortisol decrease the expression of PGDH

in chorionic trophoblast cells. As a result, prostaglandins act on the fetal membranes to increase

the activity of 11β–HSD1 in chorionic trophoblasts and decrease expression of 11β-HSD2 in

placental trophoblasts (Alfaidy et al., 2001; Challis et al., 2001). The activity of 11β-HSD1

converts inactive cortisone to cortisol whereas activity of 11β-HSD2 inactivates cortisol to

cortisone. Elevated CRH levels promote different effects based on the regionalization of the

cortisol receptor subtypes throughout the gestational tissues (Hillhouse et al., 2002). In the

fundal region, CRH acts through CRH receptor subtype 1 to increase prostaglandin production

through increased PGHS-2 and decreased PGDH (Challis et al., 2000). As a result of varying

receptor expression, the fundal region of the uterus relaxes allowing for regionalization of

activity at term (Cong et al., 2009; Stevens et al., 1998). As well, the surge of CRH in late

pregnancy is associated with increased estrogen resulting in altered progesterone to estrogen

ratios indicative of a role for CRH in myometrial activation (Smith et al., 2009). Elevated

prostaglandins induce expression of MMP-1, 2, and 9 in the amnion that begin to degrade the

collagen present in the epithelium, compact, fibroblast and spongy layers that constitute the

12

tissue. Therefore, increased glucocorticoid levels result in a positive feed-forward loop between

cortisol, CRH and prostaglandins that is necessary for activation of the myometrium, cervical

ripening and membrane weakening.

1.2 Preterm Birth

Both autocrine and paracrine signals mediated by uterine stretch and activation of the HPA axis

interact to promote phenotypic changes in the gestational tissues leading to the onset of labour at

term. However, the phenotypic state of the gestational tissues can be altered earlier in gestation

if similar autocrine and paracrine signals are up-regulated. During infection and/or inflammation

cytokines and prostaglandins levels become elevated that subsequently activate pathways

involved in modifying the phenotype of the myometrium, cervix and fetal membranes.

Ultimately, the ability of these signals to modify and activate downstream pathways indicates a

potential mechanism leading to preterm delivery and highlights potential molecules to be

targeted in preventative treatment of infection and/or inflammation-mediated preterm birth.

1.2.1 Epidemiology

Preterm birth is a major obstetrical concern that occurs in 8% and 13% of all pregnancies in

Canada and the United States of America, respectively (Goldenberg et al., 2008; Public Health

Agency of Canada, 2008). Defined as labour occurring between 20 and 37 weeks gestation,

preterm birth accounts for 75-85% of all neonatal morbidities and mortalities (Challis et al.,

2002). Many of these morbidities, including pulmonary disorders, blindness, deafness, cerebral

palsy, and neurodevelopmental delays linger into adulthood causing poor health later in life. The

cost of caring for preterm infants has a large economic impact on the healthcare system. In 2005,

it was estimated that $26.2 billion in the United States of America was spent on medical care for

preterm neonates (Institute of Medicine Report, 2006). An impact that cannot be measured

through monetary means is the emotional and psychological burden placed upon the families

caring for infants born prematurely.

13

1.2.2 Etiology

The incidence of preterm birth has been on the rise since the 1980s (Challis et al., 2002). Much

of this trend is accounted for by increased use of reproductive assisted techniques and the

associated rise in multi-fetal pregnancies. Other predisposing factors include advanced or young

maternal age, black race, low socioeconomic status of either or both the father and mother, single

parenthood, type of employment, smoking, alcohol and substance abuse as well as carrying a

male fetus (Goldenberg et al., 2003; Meis et al., 1995; Robinson et al., 2001). Certain aspects of

the mother’s medical history also increase the risk of preterm birth. Such medical conditions

include periodontal disease, lung disease, chronic hypertension, proteinuria or bacteriuria, a

history of a previous preterm delivery, uterine malformation or short cervix. Reproductive tract

infection, pelvic infection, vaginal microbiota alterations and infections including bacterial

vaginosis, Neisseria gonorrhoeae, Chlamydia trachomatis, group B Streptococcus, Ureaplasma

urealyticum, and Trichomonas vaginalis as well as obstetrical complications during pregnancy

such as early pregnancy bleeding, placental abruption, placenta praevia, hydramnios,

preeclampsia, premature rupture of the membranes (PROM), cervical surgery or a previously

induced abortion are all associated with increased preterm birth (Goldenberg et al., 1998;

Goldenberg et al., 2003; Meis et al., 1995; Norwitz et al., 1999; Robinson et al., 2001; Romero et

al., 2002). Preterm birth can be categorized into three groups: 1) Idiopathic (50%), 2) Indicated ,

those induced by a physician as more deleterious to prolong to either the health of the fetus or the

mother due to medical complications (20-30%), and 3) Infection and/or inflammation-mediated

(30%). Early preterm deliveries tend to be associated more so with the presence of histologic

chorioamnionitis and inflammation compared to later preterm deliveries (Goldenberg et al.,

2000; Goldenberg et al., 2003; Vogel et al., 2005). However, despite ongoing research, the

etiology of spontaneous preterm birth remains largely unknown.

1.2.3 Current Diagnostic Strategies and Treatments for Preterm Birth

The diagnosis and prevention of preterm birth remains problematic despite ongoing research.

Commonly, pregnant women are informed by their physician of how to recognize signs of labour

including contractions, pelvic pressure, vaginal discharge and back pain (Denney et al., 2008).

However, only one third of women showing signs of preterm labour give birth within 24–48

14

hours (Bocking et al., 1999; Katz et al., 1999). Despite ongoing research, the onset of labour is

only delayed up to 72 hours using tocolytic therapy. This allows for glucocorticoid treatment to

mature fetal lungs; however the other developmental consequences associated with preterm birth

cannot be corrected within this relatively short prolongation of gestation. Therefore, it is

important to develop effective preventative therapy.

Assessment of a woman’s medical history plays an important role in determining the individual

risk she possesses of undergoing preterm delivery. A history of a previous spontaneous preterm

birth is one of the highest predictive indicators of an increased risk of preterm birth (Goldenberg

et al., 2003). The preterm prediction study by the Maternal Fetal Medicine Network, which

included 3000 pregnancies, indicated the three most effective predictive assessments for preterm

delivery: 1) A positive cervical or vaginal fluid fetal fibronectin test, 2) A cervical length of less

than or equal to 25 mm, and 3) Elevated serum levels of α-fetoprotein, alkaline phosphatase and

granulocyte-colony stimulating factor (G-CSF; Goldenberg et al., 2003). As well, amniotic fluid

samples with elevated levels of interleukin (IL)-6, IL-8, or Ureaplasma urealyticum are

associated with high levels of preterm delivery (Vogel et al., 2005). Within maternal serum or

vaginal and cervical fluids many molecules have been identified as potential predictive markers

of preterm birth. These include low levels of ferritin, folate, zinc and high levels of C-reactive

protein, cytokines, MMPs as well as altered levels of activin, relaxin, collagenase, non-

phosphorylated insulin-like growth factor and CRH (Andrews et al., 2003; Norwitz et al., 1999;

Vogel et al., 2005). Recently, single nucleotide polymorphisms (SNPs) in the promoter region of

cytokines that up-regulate their transcription levels have been correlated with an increased risk of

preterm birth (Harper et al., 2011).

It is important for physicians to determine whether or not prolonging gestation may be more

detrimental to the fetus in a hostile intrauterine environment exposed to hypoxia or infection

and/or to the mother as is the case in preeclampsia. In these cases, preterm delivery may be the

most beneficial option compared to allowing gestation to progress. Tocolytics are agents that

ameliorate uterine contractions and thus prevent expulsion of the fetus from the uterine

environment for 24-72 hours (Norwitz et al., 1999). Ethanol was the first effective tocolytic

identified (Fuchs et al., 1981). However, due to severe fetal side effects, its use was abandoned.

Four routine tocolytic treatments are used in the United States of America: 1) β-mimetics, 2)

15

Magnesium sulphate, 3) Non-steroidal anti-inflammatory drugs, and 4) Calcium channel

blockers (Giles et al., 2007; Katz et al., 1999; Meis et al., 2003; Vercauteren et al., 2009).

However, like early tocolytic drugs, many of these options are associated with adverse side

effects, including myocardial ischemia, tachycardia and arrhythmias as well as severe

consequences to the health of the fetus including intraventricular hemorrhage and premature

closure of the ductus arteriosus (Giles et al., 2007; Katz et al., 1999; Vercauteren et al., 2009).

Therefore, the benefits of tocolytic drugs have been questioned and their use abandoned in many

cases for the prevention of preterm delivery.

After the diagnosis of a pregnancy at high risk of undergoing preterm delivery, it is essential to

provide effective preventative treatment if prolongation of gestation is not deleterious to the

health of the fetus or the mother. However, currently no effective preventative treatment exists.

A common method once used to prevent preterm birth was the use of antibiotics. Although

bacterial vaginosis and Trichomonas vaginalis are largely asymptomatic, these alterations to the

healthy vaginal microbiota are associated with an increased risk of preterm birth (Romero et al.,

2002). When these conditions are diagnosed, antibiotics such as metronidazole, erythromycin,

clindamycin and/or ampicillin, are administered in an attempt to replenish the altered microbiota

by eliminating pathogenic bacteria. However, numerous meta-analyses, cohort studies, and

randomized trials indicate that antibiotic treatment is ineffective at decreasing the risk of preterm

birth (Andrews et al., 2003; Carey et al., 2000; Kenyon et al., 2001; Leitich et al., 2003;

McDonald et al., 2007; Okun et al., 2005). In some cases, the use of metronidazole treatment

actually increases the risk of preterm birth (Andrews et al., 2003; Okun et al., 2005). Antibiotics

may be ineffective at decreasing the risk of preterm birth for several reasons: 1) Bacteria may

have already ascended into the upper genital tract, 2) Antibiotics are unable to eradicate biofilms,

3) Antibiotics are unable to inactivate sialidases and, 4) Antibiotics can eradicate bacteria that

are necessary for host defense mechanisms (Reid et al., 2003). Also, prolonged use of antibiotics

has led to increasingly resistant bacterial strains and promotes an alkaline environment in the

vagina that encourages pathogenic bacterial growth (Locksmith et al., 2001).

1.3 Infection and Inflammation

Infection and/or inflammation accounts for 30% of preterm births. Inflammation acts as a first

line of defense against pathogenic threats by recruiting and activating immune cells while

16

forming a physical barrier to stop the spread of infection. In many cases, the signs and

symptoms of infection and/or inflammation during pregnancy are undetectable unless invasive

procedures such as amniocentesis are performed. The prevalence of chorioamnionitis, an

inflammation of the chorioamnion membranes has been found in up to 50% of pregnancies

ending prematurely. Many of the organisms found in the upper genital tract are of vaginal origin

and bacterial vaginosis, an alteration to the vaginal microbiota where lactobacilli levels decrease,

increases the risk of preterm delivery by 1.4 to 3.8 fold (Andrews et al., 2000). Treatment of

bacterial vaginosis with antibiotics is associated with a high incidence of re-occurrence.

Consequentially, current research has shifted to determine the use of candidate probiotic strains

of bacteria including lactobacilli, to reinstate a healthy vaginal microbiota. Many Lactobacillus

strains exert immunomodulatory effects across various systems; however the role Lactobacillus

strains may exert on the inflammatory processes in the gestational tissues that are associated with

infection and/or inflammation preterm delivery remain largely uninvestigated.

1.3.1 Vaginal Microbiota

The fetal vagina is a sterile environment which is first colonized by bacteria endogenous to the

mother’s vaginal birth canal, the skin of caretakers and the infant’s faeces upon birth (Reid et al.,

2011; Spiegel, 1991). It is not until after menarche, when estrogen levels increase that the

childhood vaginal microbiota characterized by intestinal and cutaneous bacterial species

becomes dominated by Lactobacillus species, the hallmark species of a healthy adult vaginal

microbiota (Forsum et al., 2005; Spiegel, 1991). Estrogen levels at puberty promotes the

production of glycogen in the vaginal epithelial cells necessary for the survival of lactobacilli as

it provides a source of glucose through fermentation (Forsum et al., 2005; Spiegel, 1991). The

bi-product of fermentative metabolism, lactic acid creates an acidic vaginal environment with a

pH of 3.8-4.2 which promotes self-growth of Lactobacillus species within a healthy microbiota.

Three or four species, including Lactobacillus crispatus, Lactobacillus iners, Lactobacillus

gasseri, and Lactobacillus jensenii are present in much higher levels in the vaginal microbiota

(Antonio et al., 1999; Burton et al., 2002; Forsum et al., 2005; Lamont et al., 2011; Pavlova et

al., 2002; Song et al., 1999; Vasquez et al., 2002; Yamamoto et al., 2009). As well, over twenty

species of Lactobacillus, including Lactobacillus acidophilus, Lactobacillus fermentum,

Lactobacillus plantarum, Lactobacillus oris, Lactobacillus ruminis, Lactobacillus reuteri, and

17

Lactobacillus rhamnosus have been isolated from the vaginal microbiota (Antonio et al., 1999;

Forsum et al., 2005; Pavlova et al., 2002). Gram-negative bacteria, including Streptococcus

species, Atopobioum vaginae, Prevotella spp. and Gardnerella vaginalis are present in low

levels in healthy vaginal microbiota (Yamamoto et al., 2009). Using 16S RNA gene sequencing

analysis, Gardnerella vaginalis was identified as a core component in the vaginal microbiota of

the entire female population sampled (Hummelen et al., 2010). Therefore, many bacteria that

may become pathogenic if levels are not maintained are present in healthy vaginal flora at

detectable levels.

The vaginal microbiota provides a barrier to exterior influences and actively participates in

adaptive immunity, the portion of the immune system that is specific and non-inherited (Forsum

et al., 2005). However, the vaginal microbiota is susceptible to alterations resulting in bacterial

vaginosis, Vulvovaginal candidiasis, and Trichomonas vaginalis. Accordingly, women with

altered vaginal microbiota are more susceptible to the transmission of sexually transmitted

diseases including HIV and gonorrhoea, pelvic inflammatory disease, urinary tract infections,

post-operative infections, and upper genital tract infections including chorioamnionitis (Antonio

et al., 1999; Deb et al., 2004; Forsum et al., 2005). Pathogenic bacteria are equipped with

molecules, including proteases, mucinases, and sialidases that promote self-growth through

degradation of the cervicovaginal mucous lining. Beyond the vaginal environment, bacterial

collagenases, phospholipases and endotoxins directly or through stimulated production of

downstream molecules result in remodeling of the collagen structuring of the fetal membranes

and cervix and initiate inflammatory cascades in the gestational tissues of the intrauterine

environment (Goldenberg et al., 2000).

1.3.2 Routes of Intrauterine Infection

Microbial invasion of the amniotic cavity is associated with adverse outcomes including PROM

and preterm birth. Intrauterine infection may result from four different routes of bacterial access:

1) Retrograde migration from the abdominal cavity through the fallopian tubes, 2)

Haematogenous spread from the maternal blood through the placenta, 3) Iatrogenic introduction

through invasive procedures, such as amniocentesis, and 4) Ascent from the vagina and cervix

through the fetal membranes (Figure 1.2, page 21; Goldenberg et al., 2000). The latter of these, is

thought to be the most common route for bacteria to reach the uterine cavity. Evidence for this

18

includes the fact that histologic chorioamnionitis is more severe at the location of membrane

rupture than other sites including the chorionic plate or the site of umbilical cord attachment

(Benirschke, 1960; Romero et al., 1989d). As well, microbes found in congenital infections are

similar to those isolated from the maternal vagina (Benirschke, 1960). The most common

microorganisms found in amniotic fluid samples are Ureaplasma urealyticum, Gardnerella

vaginalis, Bacteroides spp. and Mycoplasma hominis (Goldenberg et al., 2000). The

morphotypes of these bacteria are used by the Nugent scoring system to diagnose bacterial

vaginosis (Goldenberg et al., 2000; Nugent et al., 1991). During twin gestations, twins are

commonly oriented one over the other, such that the membranes of only one twin comes come in

contact with the cervical os, while the other twin is located superior to this twin. Histologic

chorioamnionitis is more frequent in the fetal membranes directly over the cervix. Bacteria

present in these membranes are commonly of vaginal origin indicative of their ascent from the

lower genital tract (Benirschke, 1960). Together, this evidence indicates the most common route

of infection of the intrauterine environment is ascending from the lower genital tract.

1.3.3 Bacterial Vaginosis

1.3.3.1 Epidemiology

Bacterial vaginosis accounts for approximately 80% of all diagnosed vaginal microbiota

alterations and is present in 15% - 20% of pregnant women (Hillier et al., 1995; Imseis et al.,

1997; Romero et al., 2004). Bacterial vaginosis is associated with many factors including: single

marital status, African American ethnicity, low socioeconomic status, previous delivery of a low-

birth-weight infant, douching practices, spermicide and antimicrobial use, smoking, absence of

barrier birth control, and sexual activity (Hawes et al., 1996; Hillier et al., 1995; Reid, 2008).

Bacterial vaginosis increases the incidence of pelvic inflammatory disease, post-operative

infections, post-cesarean endometritis, PROM, histologic chorioamnionitis, and subsequent

preterm delivery as the infection may reach the intra-uterine environment (Hawes et al., 1996;

Imseis et al., 1997; MacPhee et al., 2010; Nugent et al., 1991; Romero et al., 2004; Soper, 1993;

Sweet, 1995).

1.3.3.2 Etiology

Bacterial vaginosis is an alteration to the endogenous vaginal microbiota where levels of

Lactobacillus species decrease and anaerobic bacteria begin to dominate. It is unknown if the

19

decrease of lactobacilli precedes the growth of pathogenic bacteria or overgrowth of harmful

bacteria causes a decrease in Lactobacillus; however, the total concentration of bacteria in the

vagina increases by 100 to 1000 times that of the healthy microbiota (Forsum et al., 2005).

Characteristic bacteria that dominate the microbiota in bacterial vaginosis include Gardnerella

vaginalis, Bacteriodes spp., Mycoplasma hominis, Mobiluncus spp., Prevotella spp., and

Atopobium vaginae; many of which reside in low levels in a healthy vaginal microbiota (Falagas

et al., 2007; Hummelen et al., 2010; Nugent et al., 1991). The microbial profiles of women with

bacterial vaginosis show large inter-patient variability and greater diversity when compared to

profiles of women with a healthy vaginal microbiota. In agreement, no singular microbe has been

associated with the etiology of bacterial vaginosis and diagnosis is dependent on symptoms and

signs as well as a proportional analysis of bacterial species present in characteristically healthy

versus altered vaginal microbiota.

1.3.4 Current Diagnostic Strategies and Treatments for Bacterial Vaginosis

Currently, the diagnosis of bacterial vaginosis combines both clinical syndrome observation with

Nugent scoring according to the following four hallmark indicators: 1) A vaginal pH greater than

4.5, 2) An amine fishy odour when vaginal fluid is mixed with potassium chloride, 3) The

presence of clue cells - vaginal epithelial cells covered in Gardnerella vaginalis making them

appear rough edged under a microscope in a vaginal fluid sample, and 4) A Nugent score greater

than 6. Nugent scoring is based on a weighted representation of the bacterial species visualized

by gram-staining that summate to a score from 0 to 10. Bacteria scored include: Lactobacillus

(gram-positive rods), Gardnerella vaginalis (small gram-variable rods), Bacteroides spp. (small

gram-negative rods), and Mycoplasma spp. (curved gram-variable rods; Nugent et al., 1991).

The values 0-3 represent a normal microbiota, with high proportions of Lactobacillus spp. and

few of the other morphotypes; 4-6 represents an intermediate microbiota with higher proportions

of pathogenic morphotypes, and a score of 7-10 represents bacterial vaginosis, with low

lactobacilli morphotypes present. Inaccuracies in diagnosis persist due to the lack of symptoms

in up to an estimated 50% of women with bacterial vaginosis and the acuity required to identify

associated symptoms (Nugent et al., 1991). Other diagnostic techniques once used for bacterial

vaginosis include laboratory cultures positive for Gardnerella vaginalis growth, gas

chromatography for products of pathogenic vaginal bacteria and a proline amniopeptidase test

20

(Nugent et al., 1991). However, finding Gardnerella vaginalis is no longer deemed a definitive

diagnosis. Gram-staining is the most inexpensive, efficient and reproducible; therefore, gram-

staining and Nugent scoring is the method of choice for bacterial vaginosis diagnosis.

As bacterial vaginosis is a disruption to the vaginal eubiosis, treatment has largely been focused

on the use of antibiotics, administered both orally and vaginally. Despite evidence of cure after

initial treatment, a high reoccurrence rate exists in many patients regardless of the route and/or

type of antibiotic administration. Repetitive antibiotic use is avoided as it may encourage the

development of resistant bacterial strains. In addition, antibiotic treatment of bacterial vaginosis

in pregnant women has not shown any decrease in the incidence of preterm birth.

The association of bacterial vaginosis with preterm delivery has led to the study of adjunctive

probiotic and antibacterial treatment, or probiotic supplementation independently. Despite

relatively low endogenous levels in the vaginal microbiota, Lactobacillus rhamnosus GR-1 and

Lactobacillus reuteri RC-14 are the most effective at restoring and maintaining a normal vaginal

microbiota (Reid et al., 2011). These two strains of lactobacilli persist in the vagina after vaginal

insertion and also alter the host immune defenses (Cadieux et al., 2002; Gardiner et al., 2002).

Lactobacillus rhamnosus GR-1 in particular can populate the vagina, reduce pathogenic bacteria

from reaching the vagina from the anal-rectal area, inhibit growth, adhesion and biofilm

formation of gram-negative bacteria as well as inhibit the persistence of Candida albicans

(Cadieux et al., 2002; Reid, et al., 2003; Reid, 2008).

Bacteria associated with bacterial vaginosis are capable of releasing proteases and endotoxins

that free arachidonic acid from phospholipid membranes and subsequently increase

prostaglandin production in the amnion, chorion and decidua. Increased prostaglandins elevate

the risk of preterm labour by inducing collagen remodeling in the fetal membranes. Antibiotic

treatment is ineffective at preventing preterm birth. This may be a consequence of the inability of

antibiotics to alter downstream pathways that have been activated prior to their administration.

There is little research on the role lactobacilli may play at preventing the production of cytokines

and prostaglandins in the amnion, a major contributor to the amniotic fluid cytokine pool and

prostaglandins. However, due to the known immunoregulatory and other beneficial roles of

Lactobacillus species, it is possible that these bacteria may act as a probiotic to decrease the risk

of preterm birth associated with infection and/or inflammation.

21

Figure 1.2: Potential sites of bacterial infection amongst the intrauterine tissues.

Potential routes for microbes to gain access to the intrauterine environment: 1) Retrograde

migration from the abdominal cavity through the fallopian tubes, 2) Haematogenous spread from

the maternal blood through the placenta, 3) Iatrogenic introduction through invasive procedures,

such as amniocentesis, and 4) Ascent from the vagina and cervix through the fetal membranes.

[The image is modified from New England Journal of Medicine. Vol 342.20. Goldenberg RL,

Hauth JC, Andrews WW. Intrauterine infection and preterm delivery. 1500-1507 (2000) and

appears here with the permission of the journal.]

1

2

3

4 4

22

1.4 Probiotics

Probiotics are defined as, “live microorganisms which when administered in adequate amounts

confer a health benefit on the host” (FAO and WHO, 2001). Probiotics have been used

therapeutically and have shown beneficial effects across a wide spectrum of medical conditions.

These effects include modulation of immunity, lowering cholesterol, treatment of rheumatoid

arthritis, prevention of cancer, improvement of lactose intolerance, prevention of diarrhea and

constipation, reduction or prevention of acute dermatitis and treatment of urinary tract infections

(Kligler et al., 2007; Lee do et al., 2011; Ohara et al., 2010; Ojetti et al., 2010; Reid, 1999; So et

al., 2011; Verna et al., 2010). Although the use of lactobacilli as probiotics have shown promise

in restoring vaginal microbiota for the treatment of vaginal infections and microbiota alterations

during pregnancy, there are currently insufficient studies to assess the impact probiotics have on

pregnancies at risk of undergoing preterm birth associated with infection and/or inflammation

(Othman et al., 2007).

Although the exact mechanism by which probiotics exert their beneficial effects remains

unknown, various types of probiotic bacteria are known to benefit their host through competitive

exclusion of bacterial adherence, bacteriocin and butyrate production, maintenance of the

epithelial barrier, modulation of immune responses through NFΚB transcriptional activity

modification and immune cell recruitment and activation as well as increased IgA production

(Ahn et al., 1990; Falagas et al., 2007; Floch, 2010; Jijon et al., 2004; Lara-Villoslada et al.,

2007; Madsen et al., 2001; Moorthy et al., 2009; Reid, 2008; Yan et al., 2002). Pathogenic

bacteria possess pathogen-associated molecular patterns (PAMPs) on their cellular membrane

that are recognized by the body’s immune system through activation of TLRs. Unlike probiotic

bacteria that instead harbor microbe-associated molecular patterns (MAMPs), recognition of

PAMPs initiate an immune response to defend against pathogenic bacterial threats (Servin,

2004). Additionally, the signaling pathway activated by TLR9 has been shown to be necessary

for immunostimulation by bacterial CpG DNA sequences that play a role in the anti-

inflammatory effects exerted by various probiotics (Rachmilewitz et al., 2004). Therefore, an

opportunity exists for probiotics and/or their byproducts to be used as treatment for many

infectious and autoimmune conditions.

23

1.4.1 Lactobacillus Species as Candidate Probiotics for use During Pregnancy.

The criteria for defining a probiotic candidate has been established including: 1) Of no threat to

the host, therefore they are non-carcinogenic, non-pathogenic, and non-invasive, 2) Able to

persist and multiply, 3) Resistant to vaginal microbicides, 4) Capable of co-aggregation and

forming a normal eubiosis, 5) Capable of adherence to cells, 6) Capable of exclusion or

reduction of pathogenic adherence, and 7) Capable of antagonistic growth through the

production of acids, hydrogen peroxide and bacteriocins (Reid, 1999). In addition, when

identifying a potential probiotic strain of bacteria, it is important to consider the commensal

bacterial microbiota as well as the intrinsic resistance of the probiotic to antibiotics (Servin,

2004). In the case of the vaginal microbiota, loss of the endogenous dominant species of a

healthy vagina, lactobacilli, is associated with pathologies including bacterial vaginosis.

Bacterial vaginosis in pregnant women increases risk of preterm delivery; therefore, a strain of

Lactobacillus may exert beneficial effects during pregnancy. In particular, Lactobacillus

rhamnosus GR-1 possesses many of the characteristics required to define it as a probiotic

bacteria for use in the urogenital tract.

1.4.1.1 Safety of lactobacilli administration

Lactobacillus species are safe for use in humans with low intrinsic immunogenicity as they are

neither carcinogenic nor pathogenic (Cadieux et al., 2002; Servin, 2004). In fact, Lactobacillus

rhamnosus GR-1 injected directly into the bladder of human patients failed to induce an infection

(Hagberg et al., 1989). In addition, a meta-analysis of randomized controlled trials involving

lactobacilli and bifidobacterium during pregnancy indicated that either bacterium do not alter the

incidence of caesarean section, birth weight, malformations, or gestational age at birth in

pregnant women (Dugoua et al., 2009). Therefore, the safety of Lactobacilli rhamnosus GR-1 in

humans is established and has a low or negligible chance of threatening the health of a mother

and her fetus throughout pregnancy.

1.4.1.2 Ability of lactobacilli to colonize the vagina

Studies have shown that after instillation, Lactobacillus rhamnosus GR-1 and Lactobacillus

reuteri RC-14 persist in the vaginal environment for 19 days compared to only 5 days for

Lactobacillus rhamnosus GG (Cadieux et al., 2002). Another study showed Lactobacillus

24

rhamnosus GR-1 present in the vaginal microbiota up to 7 weeks after instillation (Reid et al.,

1994). Importantly, Lactobacillus rhamnosus GR-1 and Lactobacillus reuteri RC-14 are capable

of restoring and maintaining a healthy vaginal microbiota (Reid et al., 2001a, Reid et al., 2003).

Passive forces including electrostatic interactions, hydrophobic steric forces and steric

hinderence via lipoteichoic acid (LTA) allow Lactobacillus to interact with the mucosal layer of

urogenital cells (Servin, 2004). Lactobacillus rhamnosus strains have a high capacity for

adherence to urogenital cells compared to other lactobacilli strains (Reid et al., 1987; Servin,

2004). Since adherence to epithelium plays an important role in initiating the pathogenesis

involved with many urogenital tract infections, the ability of lactobacilli to compete, inhibit,

and/or prevent colonization of the urogenital tract by pathogenic bacteria is significant (Servin,

2004).

1.4.1.3 Pathogenic bacterial defense mechanisms of lactobacilli

Lactobacilli promote exclusion or reduction of pathogenic bacteria adherence through the

production and actions of biosurfactants, hydrogen peroxide, lactic acid and bacteriocins.

Together these molecules limit the adherence of pathogenic bacteria to urogenital cells and

minimize their ability to colonize and flourish within the vaginal environment.

Biosurfactants are surface-active compounds that help restore homeostasis by exclusion or

reduction of pathogenic bacteria (Reid et al., 2011). Biosurfactants produced from vaginal

lactobacilli strains are composed of lipids, proteins and carbohydrates that bind to collagen III

and VI as well as fibronectin which are major components of the extracellular matrices of the

vagina and other intrauterine tissues (Howard et al., 2000; Reid et al., 2011). Lactobacilli

rhamnosus GR-1 and/or Lactobacillus reuteri RC-14 effectively disrupt Escherichia coli,

Enterococcus faucalis, Gardnerella vaginalis, Shigella dysenteria, Candida albicans,

Streptrococcuss aureus, and other bacteria biofilms and/or adherence that are associated with

bacterial vaginosis, urinary tract infections and yeast vaginitis (Howard et al., 2000; Moorthy et

al., 2009; Reid et al., 1995; Reid, 2008; Saunders et al., 2007; Velraeds et al., 1996; Velraeds et

al., 1998; Xu et al., 2008).

Hydrogen peroxide is a non-specific oxidizing agent that promotes the production of free

radicals and activates epithelial cell responsiveness to inflammatory stimuli (Voltan et al., 2008).

25

Gardnerella vaginalis is a non-catalase producing bacteria and therefore, hydrogen peroxide

inhibits its growth (MacPhee et al., 2010). Women with hydrogen peroxide-producing

lactobacilli present in their vaginal microbiota have lower incidence of bacterial vaginosis

(Eschenbach et al., 1989; Hillier et al., 1992; Hillier et al., 1993). Although Lactobacillus

rhamnosus GR-1 does not produce much hydrogen peroxide, it produces lactic acid through the

fermentation of glycogen from the vaginal epithelium (Saunders et al., 2007). Bacteriocins have

a narrow killing spectrum in comparison to hydrogen peroxide and lactic acid as they damage the

cytoplasmic membranes of bacteria (Reid et al., 2011; Servin, 2004).

1.4.1.4 Immunomodulatory properties of lactobacilli

Various strains of lactobacilli have been shown to modulate the immune response in immune cell

lines, gut epithelial and placental trophoblast cells. DC429, a strain of Lactobacillus rhamnosus

increased chemotaxis of polymorphonuclear immune cells, activated phagocytosis in the

recruited cells and increased cytokine production in murine air pouch models (Kotzamanidis et

al., 2010). The cytokine profile stimulated by DC429 was characterized by increased production

of IFNγ, IL-5, IL-6, and IL-10 in the gut mucosal lining of mice. The authors of this study

suggested that DC429 helped to maintain a physiologic state of inflammation crucial to the

defense of the gut epithelial layer (Kotzamanidis et al., 2010). Heat-killed Lactobacillus

rhamnosus GG stimulated IL-4, IL-10 and urocortin release from human placental trophoblast

cells (Bloise et al., 2010). Similar to results shown with the supernatant of Lactobacillus

rhamnosus GR-1, heat-killed Lactobacillus rhamnosus GG prevents LPS-stimulated TNFα

release from primary placental trophoblast cultures (Bloise et al., 2010; Yeganegi et al., 2010).

Additionally, Lactobacillus rhamnosus GR-1 supernatant increased IL-10 output in primary

cultures of human placental trophoblast cells (Yeganegi et al., 2010). Interestingly,

Lactobacillus rhamnosus GR-1 supernatant is capable of altering levels of PGDH, PGHS-2 and

G-CSF in a sex-dependent manner (Yeganegi et al., 2009; Yeganegi et al., 2011). These results

suggest that the supernatant of Lactobacillus rhamnosus GR-1 contains metabolites capable of

exerting the beneficial effects regardless of the presence of live or heat-killed bacteria.

Therefore, Lactobacillus rhamnosus strains and/or supernatant metabolites are capable of

modifying both immune cell chemotaxis and activity levels as well as the cytokine and

prostaglandin metabolizing enzymes in a gestational tissue.

26

1.4.2 The Potential Mechanisms of Probiotics

The exact mechanism by which lactobacilli exert their beneficial effects on the host remains

elusive. However, studies have shown that the supernatant of some Lactobacillus species can be

beneficial. After separating the supernatant from Lactobacillus plantarum 10hk2 into protein

and polysaccharide portions and subsequent analysis of fractions based on molecular weight, an

active protein fraction of 8.7 kDa was identified. This protein fraction was capable of inducing

IL-10 production from a murine macrophage cell line and eliminates activation of key molecules

in the NFκB-cytokine transcription pathway (Chon et al., 2010). Similarly, a soluble protein

factor eluted from a gastrointestinal probiotic reduced Tumor Necrosis Factor α (TNFα) and

Interferon γ (IFNγ) secretion from T84 monolayers (Madsen et al., 2001). The active product in

two other lactic acid bacterium family members, Bifidobacterium breve and Streptococcus

thermophilus have active enzymes resistant to digestive enzymes with a molecular mass of < 3

kDa in culture medium capable of inhibiting lipopolysaccharide (LPS)-induced TNFα in

intestinal cell monolayers. This inhibitory effect was preserved after transepithelial transport

across the intestinal cell monolayer (Menard et al., 2004). Additionally, medium collected from

cultured Lactobacillus rhamnosus GG was sufficient in preventing cytokine-induced apoptosis in

intestinal epithelial cells (Yan et al., 2002). Together, these studies indicate the possibility that

probiotic metabolites are sufficient to exert their effects without the presence of live or heat-

killed bacteria. Although no such evidence has been found regarding the active molecule in

Lactobacilli rhamnosus GR-1, it stands that the culture medium from these bacteria may also be

capable of exerting effects independent of the presence of live bacteria.

The experiments in this thesis were designed to further determine the effects of probiotics on

gestational tissues during pregnancy. Taken together, the above evidence identifies Lactobacillus

rhamnosus GR-1 is a safe candidate for use in pregnancy. Lactobacillus rhamnosus GR-1

displays many anti-pathogenic characteristics that may prevent pathogenic bacteria from

dominating the vaginal flora. If access to the gestational tissues is possible, it is more likely that

a small protein byproduct or metabolite would be capable of accessing the intrauterine

environment than live probiotic bacteria. Currently, however, there is little research focused on

the effects of probiotics and their metabolites to access the intrauterine tissues during pregnancy,

or on the effects this interaction may have on the other gestational tissues including the amnion.

27

1.5 Cytokines and Chemokines

Preterm labour associated with infection is accompanied by an inflammatory cascade activated

by invasive bacterial products that stimulate increased secretion of cytokines from invading

immune cells and gestational tissues (Figure 1.3, page 33). Elevated cytokine levels stimulate

prostaglandin production in the amnion, chorion and decidua activating downstream pathways

that result in myometrial contractility, membrane rupture and cervical ripening mimicking

parturition events. Evidently, cytokines and chemokines play crucial roles in the mechanisms

involved in preterm labour associated with infection and/or inflammation. Activated neutrophils

are capable of entering into the compact, mesenchymal and spongy layers of the amnion and may

enter the amniotic cavity between the amnion epithelial cells while stimulating cytokine and

chemokine production in the fetal membranes (Saji et al., 2000). Intra-amniotic infection,

microbial invasion of the intrauterine environment and/or chorioamnionitis have been associated

with elevated levels of LPS, IL-6 and IL-8 as well as IL-1, TNFα, monocyte chemotactic protein

(MCP)-1, Regulated on Activation, Normal T cell Expressed and Secreted (RANTES) and G-

CSF in amniotic fluid compared to women with uncomplicated pregnancies (Andrews et al.,

1995; Coultrip et al., 1994; Esplin et al., 2005; Hillier et al., 1993; Romero et al., 1988b; Romero

et al., 1989a; Romero et al., 1989c). In many cases, however, amniotic fluid is sterile despite

elevated cytokine levels indicating that the presence of bacteria is not necessary for

inflammatory markers to be present (Seong et al., 2006; Smulian et al., 1999). Elevated maternal

serum levels of IL-8 and cervical or amniotic fluid levels of IL-6 are indicative of

chorioamnionitis and subsequent increased risk of preterm delivery (Andrews et al., 1995;

Jacobsson et al., 2005). Cytokine levels are increased in the gestational tissues and bodily fluids

indicative of an ongoing inflammatory response.

The amnion is composed of epithelial and mesenchymal cells that work in concert to produce

cytokines and chemokines. Many cytokines and chemokines are produced and secreted from

primary cultures of human amnion including: IL-2, IL-4, IL-6, IL-8, IL-15, IL-1 receptor

antagonist (IL-1ra), MCP-1, macrophage inflammatory protein-1α, RANTES, and TNFα.

Production and secretion of cytokines and chemokines from the amnion and chorion decreases

when the two portions of the fetal membranes are cultured independently compared to explants

of intact membrane (Menon et al., 2004). This indicates that coordinated interaction exists

between the amnion and chorion in vivo. It is hypothesized that shuttling of mRNA as well as

28

bidirectional communication and cooperation occurs between the amnion and chorion (Zaga et

al., 2004; Zaga-Clavellina et al., 2007).

During infection, bacterial products such as LTA and LPS activate pathogen recognition

receptors that activate the inflammatory cascade. LTA and LPS are wall-component proteins of

gram positive and gram-negative bacteria respectively. LPS activates TLR4 by binding to LPS-

binding protein. TLR4 co-localizes with CD14 resulting in activation of downstream MAP

kinase pathways and transcription of NFΚB genes including IL-1β, TNFα, IL-6 and IL-8. The

amnion is responsive to both bacterial endotoxin and cytokine stimulation as it expresses many

cytokine receptors as well as TLR2 and TLR4. Amnion cells maintained in vitro and as

membrane explants, secrete cytokines and chemokines after stimulation with pro-inflammatory

molecules to levels similar to those found in amniotic fluid samples from women with

pregnancies complicated by infection (Fortunato et al., 1996).

1.5.1 Pro-inflammatory Cytokines

Cytokines are small proteoglycan molecules that are secreted from immune cells and other

tissues that modulate the immune response through up-regulation of other cytokines and

chemokines and direct activation of immune cells and other aspects of immunity. Pro-

inflammatory cytokines are elevated in the amniotic fluid during preterm labour associated with

infection and/or inflammation. Pro-inflammatory cytokines induce the production of other

cytokines and chemokines, growth promotion and inhibition, angiogenesis, cytotoxicity and

other mechanisms to defend against pathogenic threats. The major cascade of pro-inflammatory

cytokines involves the first response cytokines, TNFα and IL-1β as well as IL-6. IL-6 has been

shown to play a more immunoregulatory role than its counterparts in the triad. In the female

reproductive system and throughout pregnancy, TNFα plays a role in follicular development,

ovulation, menstruation and endometrial function regulation (Argiles et al., 1997). However, in

preterm labour associated with infection and/or inflammation TNFα, IL-1β and IL-6 stimulate

prostaglandin production in the amnion. Elevated prostaglandins initiate the onset of parturition

and if levels are increased due to infection, pro-inflammatory cytokines may promote early onset

of labour.

TNFα and IL-1β are the first cytokines found in serum in response to acute infection (Petersen et

al., 2006). In LPS-induced fetal loss in pregnant mice, both TNFα and IL-1β levels are increased

29

(Silver et al., 1994; Silver et al., 1997). Unlike IL-1β levels that become detectable in the third

trimester of pregnancy, TNFα is undetectable in amniotic fluid unless an infection is present

(Romero et al., 1989c). The amnion does not produce IL-1β and its mRNA is not localized to

the amnion. However, IL-1β mRNA is found in both the chorion and decidua and it may be

shuttled and therefore account for secretion found on the fetal side of in vitro membrane explants

(Menon et al., 1995; Thiex et al., 2009; Zaga et al., 2004). IL-1β and TNFα both increase the

availability of phospholipids, the substrate for prostaglandin production, phospholipase protein

levels as well as the expression of PGHS-2, the rate-limiting step in prostaglandin production in

the amnion (Blumenstein et al., 2000; Bry et al., 1992; Pollard et al., 1993; Romero et al., 1989b;

Romero et al., 1989c). As well, IL-1β and TNFα also decrease PGDH expression in chorion

cells resulting in decreased prostaglandin metabolism (Brown et al., 1998; Pomini et al., 1999).

Therefore, pro-inflammatory cytokines cause a net increase in prostaglandin production from the

gestational tissues (Kent et al., 1993; Pollard et al., 1993).

Levels of IL-6 are increased in vaginal fluid at term and preterm labour, in cervical fluid

obtained from women with intrauterine infection and in the amniotic fluid of women with

chorioamnionitis (Imseis et al., 1997). Importantly, IL-6 increases amnion prostaglandin

production in cultured amnion cells to levels similar to those found in the amniotic fluid of

women with intrauterine infections and preterm labour (Kent et al., 1993; Mitchell et al., 1991).

IL-6 is part of the pro-inflammatory cytokine triad alongside TNFα and IL-1β. Despite its pro-

inflammatory initiation of the acute phase response from the liver in response to acute infection,

IL-6 is defined as a pleiotropic immunomodulatory cytokine. In LPS-stimulated IL-6 deficient

knock-out mice, levels of both TNFα and neutrophilia were three times higher than in wild-type

counterparts (Xing et al., 1998). These effects were abolished with treatment of recombinant IL-

6 (Xing et al., 1998). Furthermore, in cancer patients receiving recombinant IL-6 intravenous

therapy, serum levels of both IL-1ra and sTNFα receptor, the antagonists to TNFα and IL-1β,

were increased (Tilg et al., 1994). IL-6 treatment increases levels of the anti-inflammatory

cytokine, IL-10 (Steensberg et al., 2003). Negative feedback is provided by IL-6 against TNFα

and IL-1β which may help maintain and facilitate the decrease and resolution of pro-

inflammatory immune responses (Steensberg et al., 2003).

30

1.5.2 CXC and CC Chemokines

Chemokine release is stimulated by pro-inflammatory molecules from immune cells and other

immune responsive tissues. Classified by patterns of amino acids in their amino terminus,

chemokines are separated into two families: CXC or CC. Chemokines are responsible for

immune cell recruitment, chemotaxis, activation and motility. Levels of chemokines that recruit

immune cells are elevated at term and prior to the onset of labour. However, the levels of many

chemokines increase throughout gestation and in association with infection during pregnancy

and many of their specific roles remain unknown.

IL-8 or CXCL8 is a member of the CX chemokines and plays a role in endothelial cell

proliferation and survival as well as inducing directional motility and activation of neutrophils

and T-lymphocytes (Li et al., 2003). As such, elevated concentrations of IL-8 in amniotic fluid

are associated with microbial infection, increased neutrophil count, membrane rupture, and

cervical ripening (Cherouny et al., 1993; Hsu et al., 1998). IL-8 is constitutively expressed in the

amnion and as mechanical tension increases towards the end of the mid-trimester, the expression

of the stretch-responsive IL-8 gene is increased and subsequently levels of IL-8 in the amniotic

fluid rise (Kendal-Wright, 2007; Romero et al., 1991). It is thought that increased IL-8 levels

associated with intrauterine infection recruit neutrophils into the fetal membranes and placenta

(Cherouny et al., 1993). Unlike TNFα, IL-1β, and IL-6, IL-8 does not stimulate amnion

prostaglandin production. However when the amnion is stimulated with endotoxin, amnion cells

respond with increased IL-8 to levels similar to those found in women with intra-amniotic

infection.

The family of CC chemokines includes MCP-1 and RANTES which recruit immune cells that

express the receptor, CCR5 including monocytes, basophils and eosinophils. MCP-1 is released

from ovarian follicles, endometrium and the choriodecidua as well as the amnion in the third

trimester (Arici et al., 1995; Arici et al., 1997; Denison et al., 1998). MCP-1 levels in amniotic

fluid are increased during preterm labour with chorioamnionitis and with preterm labour with or

without infection (Esplin et al., 2005). RANTES is released by the endometrium, first trimester

trophoblast cells and the amnion (Denison et al., 1998; Hornung et al., 1997; Svinarich et al.,

1996). RANTES levels in amniotic fluid decrease with gestational age and increase with

preterm labour associated with microbial invasion as well as at term (Athayde et al., 1999).

31

RANTES attracts monocytes and lymphocytes and increases immune cell adhesion. CC

chemokines activate immune cells. The important role of CC chemokines is demonstrated by the

inability of MCP-1 deficient knock-out mice to resist microbial infection and an inability to clear

bacteria (Yadav et al., 2010).

1.5.3 Immune Cell Activating Cytokines

Both IFNγ and granulocyte-macrophage colony-stimulating factor (GM-CSF) play important

roles in the adaptive immune system by participating in the activation of recruited immune cells

at the site of infection. IFNγ is an immunomodulatory cytokine crucial in the adaptive immune

system for defense against viral and intracellular bacterial infections. IFNγ increases antigen

presentation through up-regulation of MHC I and MHC II complexes, release of reactive oxygen

species in macrophages and promotes adhesion and binding of leukocytes during migration.

IFNγ plays a role in maternal recognition of pregnancy as it is involved in implantation and

blastocyst attachment (Murphy et al., 2009). In the decidua, IFNγ receptor expression decreases

in both term and preterm placentae (Hanna et al., 2004). IFNγ decreases PGHS-2 expression and

consequentially the production of PGE2 in placental explants in vitro (Hanna et al., 2004). It is

suggested that IFNγ receptor down-regulation occurs at term. With a loss of IFNγ receptor

expression, IFNγ is no longer able to repress PGHS-2 expression in the placenta resulting in an

increase in prostaglandin production (Hanna et al., 2004). IFNγ has not been identified in the

amniotic fluid; however, IFNγ is expressed in amnion tissue and shows no changes as labour

progresses (Veith et al., 1999).

Similarly, GM-CSF promotes proliferation of monocyte, neutrophils and eosinophils, GM-CSF

also activates phagocytosis, chemotaxis and adhesion as well as promoting cell growth and

survival (Gasson, 1991). During early pregnancy, GM-CSF is involved in murine normal

placental growth and differentiation as well as embryonic implantation and cytotrophoblast

differentiation (Bry et al., 1997; Garcia-Lloret et al., 1994; Robertson et al., 1992). Transgenic

mice with overexpression of GM-CSF show high levels of macrophage proliferation and

activation (Metcalf et al., 1988). Levels of GM-CSF increase throughout gestation and women

with PROM show higher levels compared to those with intact membranes during pregnancy (Bry

et al., 1997). GM-CSF is produced in the amnion, trophoblasts and decidua. Both TNFα and IL-

1β stimulate GM-CSF release from amnion cells (Bry et al., 1995). Expression of GM-CSF in

32

the amnion and levels of GM-CSF in the amniotic fluid are elevated during chorioamnionitis and

intrauterine infection respectively (Keelan et al., 2003; Stallmach et al., 1995). It is suggested

that GM-CSF may amplify the immune response in the amnion while increasing host resistance

to infection.

The interaction between cytokines and chemokines plays a crucial role in the immune response

against pathogenic threats. However, during pregnancy an imbalance of TH1 to TH2 cytokines

results in increased risk of preterm delivery through subsequent increases in prostaglandin

production. Infection of the amniotic cavity and the accompanying inflammatory processes are

harmful to the fetus because it may lead to Fetal Inflammatory Response Syndrome associated

with cerebral white matter damage and development of cerebral palsy (Yoon et al., 2003).

Although the fetal membranes act as a barrier to cytokines produced in the decidua, both the

chorion and the amnion are responsive to bacterial endotoxin and cytokine stimulation (Kent et

al., 1994; Menon et al., 2004). Therefore, the contribution of the amnion to the cytokine pool in

response to bacterial endotoxin and cytokines in infection and inflammation may play a major

role in determining the onset of parturition and outcome of labour. By controlling the level of

cytokines and chemokine stimulation during pregnancy, it may be possible to eliminate

activation of downstream pathways leading to preterm delivery. Therefore, this has led to the

investigation of the ability of Lactobacillus rhamnosus GR-1 to modulate the cytokine and

chemokine production from the amnion.

33

Figure 1.3: Proposed pathway of infection and/or inflammation-mediated preterm birth

[The image is taken from New England Journal of Medicine. Vol 342.20. Goldenberg RL, Hauth

JC, Andrews WW. Intrauterine infection and preterm delivery. 1500-1507 (2000) and appears

here with the permission of the journal.]

34

1.6 Prostaglandins

The family of biologically active lipid molecules consists of 20 carbon fatty acids that

categorized into leukotrienes, lipoxins, and prostaglandins (Olson, 2003). Prostaglandins are

produced by all cells of the body and act as local hormones initiating downstream cascades in a

paracrine manner. Prostaglandin production and metabolizing pathways are well characterized

and many of the enzymes involved are mediated by cytokines, bacterial products and endotoxins,

growth factors and glucocorticoids that increase prostaglandin production in states of infection

and/or inflammation. The biosynthesis of prostaglandins begins with the cleavage of the lipid

bilayer membrane either directly by members of the phospholipase A2 family or indirectly

through the actions of phospholipase C (Olson, 2003). As a result, arachidonic acid is freed from

the phospholipid membrane into the cytosol where it is readily converted through varying

enzymatic pathways to form lipoxins, leukotrienes, or prostaglandins. In early gestation,

metabolism leading to lipoxygenase products is favoured; however, as term approaches

metabolism is shifted towards production of prostaglandins dominating prior to the onset of

labour (Figure 1.4, page 38; Mitchell et al., 1995).

1.6.1 Prostaglandin Production

1.6.1.1 Phospholipases

Prostaglandin production begins with the cleavage of arachidonic acid from phospholipid

membranes by phospholipase enzymes. Amnion cell membranes are rich in arachidonate,

therefore cleavage of amnion cell membranes are ideal residues for prostaglandin substrate

(Curbelo et al., 1981). The family of phospholipase A2 enzymes consists of up to 15 isoforms.

Type IV, cytosolic phospholipase A2 has been shown to have a prominent role in the amnion;

however, many isoforms are also expressed in the fetal membranes (Myatt et al., 2010).

Cytosolic phospholipase 2 translocates to the cell membrane in response to agonist stimulation,

including the cytokines, IL-1 and IL-6 as well as bacterial endotoxins (Challis et al., 2002).

Therefore, much like PGHS, the enzyme involved in the rate-limiting step in prostaglandin

production, levels of phospholipase enzymes can be up-regulated in a state of infection and or

inflammation in the amnion.

35

1.6.1.2 Prostaglandin H Synthase

Freed arachidonic acid is converted into PGH2 by two subsequent reactions involving PGHS.

The first reaction is a cyclooxygenase reaction that combines two free oxygen molecules with

arachidonic acid to form PGG2, a relatively unstable intermediate. A peroxidase reaction follows

between PGG2 and PGHS to form PGH2. Two major forms of PGHS, PGHS-1 and PGHS-2, are

present in the amnion, similar to most cell types including various gestational tissues throughout

pregnancy (Hirst et al., 1995; Rose et al., 1990; Teixeira et al., 1994). PGHS enzymes

commonly reside near the membrane and within lipid bodies in various cell types ideally situated

for conversion of freed arachidonic acid substrate (Dvorak et al., 1994; Smith et al., 1981).

PGHS-1 and PGHS-2 share considerable sequence homology and up to 60-65% cDNA sequence

homology despite being the products of two separate genes (Blumenstein et al., 2000; Mitchell et

al., 1995; Xu et al., 1995). Both enzymes are homodimers and have equal affinity and capacity

to convert arachidonic acid to PGG2 and subsequently PGH2 with equal relative efficiency

(Olson, 2003). The major difference in the role of the two enzymes is their differential

expression and regulation. PGHS-2 is inducible as its promoter consists of many binding

elements that can influence its expression pattern: TATA box, two NFκB, NF-IL6, GRE, CRE,

and AP-2, SP-1, and ETS-1 binding sites (Inoue et al., 1995; Tazawa et al., 1994). The

expression of PGHS-2 is regulated and altered by cytokines, oxytocin, mechanical stretch,

glucocorticoids, bacterial endotoxins, growth factors, activators of protein kinase C and factors

that increase intracellular calcium concentrations such as CRH (Bennett et al., 1987;

Blumenstein et al., 2000; Economopoulos et al., 1996; Masferrer et al., 1992; Mohan et al., 2007;

Molnar et al., 1999; Smieja et al., 1993; Zakar et al., 1994; Zakar et al., 1995). Due to the ability

of PGHS-2 mRNA levels, protein expression and activity to be modulated by external stimuli,

conversion of arachidonic acid to PGH2 is a likely candidate for the rate-limiting step in

prostaglandin production. Amnion PGHS-2 mRNA levels, protein expression and activity levels

increase throughout gestation and further at term and with the onset of labour in an NFΚB

dependent manner (Fuentes et al., 1996; Gibb et al., 1996; Hirst et al., 1995; Johnson et al., 2006;

Mijovic et al., 1999; Mitchell et al., 1995; Slater et al., 1998; Teixeira et al., 1994). TNFα, IL-1β

and bacterial endotoxin increase PGHS-2 expression levels up to 80-fold compared to a 2 to 3

fold increase seen in PGHS-1 expression (Blumenstein et al., 2000; Mitchell et al., 1993; Pollard

et al., 1993). Therefore, by altering PGHS-2, cytokines and bacterial products associated with

36

infection and/or inflammation during labour have an ability to largely affect amnion

prostaglandin production.

1.6.1.3 Prostaglandin Synthases

The last step in prostaglandin production is performed by the prostaglandin synthase enzymes.

PGH2 is converted by a member of the prostaglandin synthase enzyme family to either

prostaglandin E2, F2α, D2 and I2. The major prostaglandin product formed in the amnion is

prostaglandin E2 and the minor product PGF2α due to the enzymatic activity of prostaglandin E

synthase (PGES) and prostaglandin F2α synthase, respectively. Three isoforms of PGES are

expressed in the amnion: microsomal (m)PGES1, mPGES2, and cytosolic (c)PGES also known as

mPGES3 (Ackerman et al., 2008). cPGES is constitutively expressed and its activity is thought

to be coupled to PGHS-1 (Han et al., 2002; Myatt et al., 2010). cPGES is localized to the

cytosol, in particular at lipid bodies of amnion epithelium as well as in fibroblasts whereas

mPGES1 is localized only to amnion epithelium membranes and lipid bodies (Meadows et al.,

2003). mPGES1 is the most catalytic efficient form of PGES above both other isoforms (Thoren

et al., 2003). Expression of mPGES1 is induced by various cytokines such as IL-1β, LPS and

other pro-inflammatory stimuli (Forsberg et al., 2000; Jakobsson et al., 1999; Mancini et al.,

2001). As well, mPGES1 is coupled to expression of PGHS-2 and is thought to be regulated

through NFΚB transcriptional activity (Kudo et al., 1999; Thoren et al., 2000). The third isoform

of PGES is mPGES2 which is formed as a golgi-membrane bound protein that translocates to the

cytosol to perform its enzymatic function (Murakami et al., 2003). mPGES2 is coupled to both

PGHS-1 and PGHS-2 and is constitutively expressed (Murakami et al., 2003). No change in

either cPGES or mPGES2 is seen in the fetal membranes at term or preterm with or without

labour; however mPGES1 can be induced by pro-inflammatory cytokines (Meadows et al.,

2003). The exact influence alterations to PGES expression has on labour with and without

infection is unknown (Premyslova et al., 2003).

1.6.1.4 15-Hydroxyprostaglandin Dehydrogenase

PGDH is a catabolic enzyme for PGE2 and PGF2α that plays an important role in minimizing the

spread of prostaglandins produced in the amnion from reaching the other gestational tissues

(Sangha et al., 1994). Unlike the amnion where expression is absent, chorion trophoblast cells

express high levels of PGDH (Cheung et al., 1990). This enables the chorion to act as a

37

metabolic barrier to amnion produced prostaglandins from acting on the other gestational tissues.

Chorionic PGDH activity decreases at term and in idiopathic preterm deliveries allowing amnion

produced prostaglandins to influence the other intrauterine tissues (Sangha et al., 1994; Van Meir

et al., 1996). Cytokines have been shown to decrease the expression of PGDH in the chorion. As

well, inconsistencies of PGDH expression throughout the chorion can be induced during

infection and inflammation (Cheung et al., 1990; Sangha et al., 1994; Van Meir et al., 1996; Van

Meir et al., 1997). Therefore, changes in PGDH throughout the chorion may grant passageway

of amnion produced prostaglandins access to the uterine myometrium and cervix ultimately

leading to activation and ripening respectively and a subsequent increased risk of preterm

delivery (Challis et al., 2002).

38

Figure 1.4: Effects of stimulants on the prostaglandin production pathway in the amnion.

Molecules that stimulate prostaglandin production are in blue font at the top of the diagram. Red

font indicates enzymes involved in prostaglandin production. Hatched lines indicate

metabolizing enzymatic activity and solid lines indicate synthesizing enzyme activity. The major

product of the amnion is PGE2 with a minor product of PGF2α.

Prostaglandin H

Arachidonic Acid

Prostaglandin E2 Prostaglandin F

2α

Phospholipids

Phospholipase A2

Prostaglandin G2

Prostaglandin H Synthase -1 and 2 Cyclooxygenase Reaction +2O2

Peroxidase Reaction -2e-

Cytosolic and microsomal

Prostaglandin E Synthase-1,

2

Prostaglandin F Synthase

Prostaglandin H Synthase-1 and 2

Bacterial Products, LPS, TNFα, IL-1β, IL-6, IL-10, Oxytocin, Glucocorticoids

15- hydroxyprostaglandin dehydrogenase

39

1.6.2 The Role of Amnion Produced Prostaglandins in Preterm Labour

At term, elevated amniotic fluid prostaglandins are paralleled with increased levels of

prostaglandin producing enzymes and activity in amnion epithelial and fibroblast cells. Amnion

produced prostaglandins play a role in fluid or ion balance as well as mediation of

transmembrane ion flow (Challis et al., 2002; Saunders-Kirkwood et al., 1993). Although the

amnion, chorion and decidua produce increasing amounts of prostaglandins throughout gestation,

only the amnion and chorion show increased prostaglandin production prior to the onset of

labour (Olson et al., 1983; Skinner et al., 1985). Increased prostaglandin levels initiate a set of

events necessary for parturition onset and successful completion of labor including: myometrial

activation, cervical ripening, up-regulation of MMPs leading to membrane rupture, fetal

adaptations to labour including decreased breathing and movements, increases in fetal HPA axis

activity and changes in uterine and placental blood flow (Kitterman, 1987; Rankin, 1976; Sastry

et al., 1997; So, 1993; Thorburn, 1992). Elevated levels of prostaglandins can lead to preterm

labour as is demonstrated by administration of prostaglandins in mammalian species that result

in induction of labour or abortion at any stage of gestation (Embrey, 1971).

The two cell types of the amnion, epithelial and mesenchymal cells respond differently to stimuli

when cultured separately in vitro. It is suggested that the two cell types of the amnion exert

paracrine actions on one another to form a regulation loop for prostaglandin production (Whittle

et al., 2001). Mesenchymal cells produced greater per cell quantities of prostaglandins; however

since epithelial cells outnumber amnion fibroblast cells 10 000:1, the overall contribution of

epithelial cells to the prostaglandin pool is greater (Whittle et al., 2000). Many cytokines

including IL-1α, IL-1β, TNFα, IL-6, MIP-1α increase prostaglandin production from the amnion

(Dudley et al., 1996; Furuta et al., 2000; Mitchell et al., 1991; Romero et al., 1989b; Romero et

al., 1989c). Additionally, gram-negative bacteria, LPS and arachidonic acid increase

prostaglandin production in the amnion (Reisenberger et al., 1998; Romero et al., 1988a; Romero

et al., 1988b). In contrast to the typical anti-inflammatory properties of IL-10 across many

tissues, the amnion shows increases prostaglandin production in response to IL-10 treatment

(Economopoulos et al., 1996; Gibb et al., 1990; Mitchell et al., 2004; Potestio et al., 1988). As

well, glucocorticoids stimulated prostaglandin production in amnion mesenchymal cells contrary

to its well-established anti-inflammatory effects in many other tissues (Whittle et al., 2000).

40

In cultured human placenta trophoblast cells, the supernatant of Lactobacillus rhamnosus GR-1

modulates expression levels of enzymes involved in prostaglandin production. Pre-treatment of

Lactobacillus rhamnosus GR-1 prevents LPS-induced PGHS-2 expression in placentae derived

from pregnancies carrying a male fetus (Yeganegi et al., 2009). As well, in placentae derived

from pregnancies carrying a female fetus, Lactobacillus rhamnosus GR-1 supernatant increases

PGDH expression to a higher degree than in placentae derived from pregnancies carrying a male

fetus (Yeganegi et al., 2009). Although the exact mechanism for the decreased expression of

PGDH is unknown, however these results may give a molecular basis for the higher incidence of

preterm labour that occurs in pregnancies carrying a male fetus compared to a female.

Comparatively, another strain, Lactobacillus rhamnosus GG induces PGHS-2 expression in a

time and dose-dependent manner in T84 colon epithelial cells (Korhonen et al., 2004).

Therefore, the effects of Lactobacillus species are dependent on the strain as well as the cell type

and gender of the host tissue. Many stimuli associated with infection and/or inflammation are

capable of stimulating prostaglandin production in the amnion. In addition, molecules that

typically decrease prostaglandin production across tissues, such as cortisol, also up-regulate

prostaglandin production in the amnion. The amnion has been identified as the main site of

prostaglandin production at the end of gestation and the onset of labour (Challis et al., 1997). If

the amnion is stimulated early in gestation by molecules associated with infection and/or

inflammation, events normally associated with term labour may be activated leading to an

increased risk of onset of preterm delivery.

41

1.7 Summary

Bacterial vaginosis, which is associated with increased risk of preterm labour by 40%, is

characterized by the colonization of pathogenic bacteria in the vaginal tract and decreased levels

of endogenous Lactobacillus species (Hillier et al., 1995). Bacterial endotoxins and bacterial

products stimulate cytokine, chemokine and prostaglandin production that may activate

downstream pathways leading to preterm labour (Figure 1.3, page 33). Interfering with the

upstream stimulants of these pathways may allow the prevention of preterm delivery mediated

by infection and/or inflammation. The particular interest of this thesis was on the

immunomodulatory role of Lactobacillus species in the amnion. Lactobacillus species have been

previously shown to modulate the inflammatory response in LPS-stimulated placental

trophoblasts and mouse macrophages and are capable of colonization and restoration of the

vaginal microbiota. However, the effect of lactobacilli on the amnion remains unknown. In

order to help determine its ability to be a potential preventative treatment, this thesis examined

the role of Lactobacillus rhamnosus GR-1 in regulating mediators of inflammation and/or

infection associated with preterm birth release from the human amnion. In particular, the effects

of Lactobacillus on cytokines, chemokines, prostaglandins and prostaglandin metabolizing

enzymes in human amnion were investigated.

42

Chapter 2

Rationale, Hypothesis and Specific Aims

43

Chapter 2

Hypotheses, Rationale and Specific Aims

2.1 Overall Hypothesis

Lactobacillus rhamnosus GR-1 will prevent elevation of cytokines, chemokines and

prostaglandins associated with infection and/or inflammation-mediated preterm birth from the

human amnion.

2.1.1 Rationale

Infection-mediated preterm birth is associated with stimulation of an exaggerated inflammatory

response in both the gestational tissues and recruited immune cells. Bacteria and bacterial

products stimulate cytokines and chemokines from the gestational tissues. During

chorioamnionitis and microbial invasion of the amniotic cavity, the cytokine pool in the amniotic

fluid is characterized by increases in pro-inflammatory cytokines and chemokines. As many

cytokines produced in the decidua are unable to cross the fetal membranes, the amnion is thought

to be the prominent producer of elevated cytokines during an inflammatory response associated

with infection. The amnion also produces prostaglandin E2 (PGE2) that may exert paracrine

actions on both the myometrium and cervix to cause activation and remodeling respectively. As

well, PGE2 exerts paracrine effects on the fetal membranes resulting in collagen degradation and

weakening. Probiotics, namely Lactobacillus species have known immunomodulatory roles in

murine macrophages, monocytic cell lines and human placental trophoblasts which promote an

anti-inflammatory cytokine profile and increased chemokine production. Additionally,

Lactobacillus species modulate prostaglandin-metabolizing enzymes. Therefore, the effects of

Lactobacillus rhamnosus GR-1 on cytokines/chemokines and prostaglandin production from the

amnion was determined in these experiments.

Although numerous strains of lactobacilli have been shown to exert immunomodulatory effects,

Lactobacillus rhamnosus GR-1 was used to treat amnion cells in the current study. Due to its

ability to colonize and replenish the vaginal flora when administered orally and ability to alter

the immune response in human placental trophoblasts and other epithelial cell types,

Lactobacillus rhamnosus GR-1 was assessed in its effects to possibly exert similar effects in

44

human amnion cells. Although the exact route of orally administered probiotics to the vaginal

microbiota is unknown, Lactobacillus species have been shown to be bile and acid resistant and

have been suggested to migrate from the rectum to the vagina after travelling through the

gastrointestinal tract (Reid et al, 2001b).

The intrauterine tissues during healthy pregnancies are sterile; therefore we chose to use treat

amnion cells with the supernatant of Lactobacillus rhamnosus GR-1 as opposed to live or freeze

dried bacteria. It is currently unknown whether endogenous lactobacilli are capable of

interacting with the fetal membranes; however, recent research has identified that small protein

molecules are capable of exerting the effects of other probiotics, including strains of

Lactobacillus. It may be that a small molecule is more capable of accessing the amnion

compared to live or heat killed bacteria.

During bacterial vaginosis, gram-negative bacteria, such as Bacteroides spp. and gram-

positive/gram-variable bacteria, such as Gardnerella vaginalis and Mycoplasma spp. levels are

elevated. Therefore, both LTA, a wall-component of gram-positive bacteria and LPS, a wall-

component of gram-negative bacteria were used to stimulate amnion cells in this study.

Importantly, Lactobacillus rhamnosus GR-1 is a gram-positive bacterium; therefore, LTA

treatment acts as an internal control to separate the effects of Lactobacillus rhamnosus GR-1

metabolites present in the supernatant used from the effects of the characteristic gram-positive

bacterial endotoxin, LTA.

2.2 Chapter 3: Effect of Lactobacillus rhamnosus GR-1 on cytokines and chemokines in human amnion cells at term

Hypothesis:

Lactobacillus rhamnosus GR-1 supernatant will ameliorate increases in LTA and/or LPS-

stimulated pro-inflammatory cytokines and chemokines in human amnion cells in order to

prevent activation of downstream pathways associated with infection and/or inflammation -

mediated preterm birth

Specific Aims:

1. To determine the optimal dose and length of exposure to LTA and LPS on cytokine and

chemokine release in medium collected from cultured human amnion cells at term.

45

2. To determine the effects of Lactobacillus rhamnosus GR-1 on the release of interleukin

(IL)-6, IL-8 and tumor necrosis factor α (TNFα) in medium collected from LPS and LTA

treated human amnion cells at term.

3. To determine the effects of Lactobacillus rhamnosus GR-1 on global cytokine and

chemokine release in medium collected from LTA and LPS treated human amnion cells

at term.

2.3 Chapter 4: Effect of Lactobacillus rhamnosus GR-1 on Prostaglandin Output and Prostaglandin Metabolizing Enzymes in Human Amnion Cells at Term

Hypothesis:

Lactobacillus rhamnosus GR-1 supernatant will ameliorate increases in LTA and/or LPS-

stimulated prostaglandin production in human amnion cells in order to prevent propagation of

downstream pathways associated with infection and/or inflammation -mediated preterm birth

Specific Aims:

1. To determine the effects of Lactobacillus rhamnosus GR-1 on prostaglandin E2

output in LPS and LTA treated human amnion cells at term.

2. To determine expression profiles of prostaglandin metabolic and synthetic

enzymes (Prostaglandin H Synthase-2, cytosolic Prostaglandin E Synthase,

microsomal Prostaglandin E Synthase 1, microsomal Prostaglandin E Synthase 2)

that may account for any changes in PGE2 release caused by LTA, LPS and

Lactobacillus rhamnosus GR-1 in human amnion cells at term.

46

Chapter 3

Effect of Lactobacillus rhamnosus GR-1 on Cytokines and

Chemokines in Human Amnion Cells at Term

47

Chapter 3

Effect of Lactobacillus rhamnosus GR-1 on Cytokines and Chemokines in Human Amnion Cells at Term

3.1 Introduction

Approximately half of all preterm births are idiopathic. However, 30% of preterm births are

associated with infection and/or an activated inflammatory process (Goldenberg et al., 2008;

Hillier et al., 1995). This inflammatory process is characterized by bacterial endotoxin

stimulation of amnion cytokine and chemokine production. Despite ongoing research, current

efforts at prevention are largely ineffective and consequentially, preterm birth remains a major

obstetrical concern.

The presence of bacteria triggers a response by the body’s innate immune system which is

characterized by increased levels of cytokines and chemokines. During pregnancy, intrauterine

infections are associated with increased levels of cytokines, including IL-1β, TNFα, IL-6, IL-8,

RANTES, and MCP-1 in the amniotic fluid (Andrews et al., 1995; Coultrip et al., 1994; Esplin et

al., 2005; Hillier et al., 1993; Romero et al., 1989c). Many cytokines produced in the decidua are

not capable of passing through the membranes. Therefore, the amnion and chorion play a major

role in contributing to the pool of cytokines and chemokines in the amniotic fluid. Elevated

levels of cytokines and chemokines recruit immune cells and increase prostaglandin production

in the amnion, chorion, and decidua (Challis et al., 1997). The net result is activation of

downstream pathways leading to increased metalloproteinase expression and subsequent cervical

ripening and membrane weakening as well as myometrial activation.

Antibiotics, such as metronidazole and clindamycin have been used to prevent infection and/or

inflammation-associated preterm birth. However, in some cases antibiotics may increase the

incidence of preterm labour (Andrews et al., 2003; Okun et al., 2005). Probiotics, defined as,

“live microorganisms which when administered in adequate amounts confer a health benefit on

the host,” are candidates for prevention of preterm labour (FAO and WHO, 2001; Reid, 1999).

Bacterial vaginosis is present in approximately 20% of pregnant women and increases the risk of

undergoing preterm birth by 40% (Hillier et al., 1995). The hallmark characteristic of bacterial

vaginosis is a decline in endogenous Lactobacillus species and a rise in pathogenic bacteria

48

resulting in an alteration to the vaginal microbiota. Due to the ability of Lactobacillus rhamnosus

GR-1 to colonize and restore the vaginal microbiota and exert immunomodulatory effects in

murine macrophages and placental trophoblasts, probiotics stand as a candidate preventative

treatment for infection and/or inflammation-mediated preterm birth (Reid et al., 2003; Yeganegi

et al., 2010; Yeganegi et al., 2011).

The effects of Lactobacillus rhamnosus GR-1 on the human amnion have not been previously

investigated. The amnion is responsive to both bacterial products and cytokines and therefore

plays an important role during infection and/or inflammation. Therefore, Lactobacillus

rhamnosus GR-1 may enable the amnion to resist bacterial endotoxin stimulation of cytokine and

chemokine production that occurs in infection and/or inflammation-mediated preterm birth.

3.2 Methodology

3.2.1 Placenta Collection

All studies were approved by the Faculty of Medicine, University of Toronto and Mount Sinai

Hospital (Toronto, Canada) ethics review boards in accordance with the Canadian Tri-Council

Policy Statements on Human Ethics Reviews (institutional review board #04-0018-U). Placentae

were collected and informed consent was obtained by the Biobank program of the Research

Centre for Woman`s and Infant`s Health at Mount Sinai Hospital (http://biobank.lunenfeld.ca)

from women undergoing elective caesarean sections at term (>37 weeks) at Mount Sinai

Hospital (Toronto, Canada). Inclusion criteria included: 1) No complications during pregnancy,

2) No abnormal antenatal test results (including laboratory results and ultrasounds), 3) An

unremarkable maternal medical history, free of any conditions that may significantly impact

placental pathology (i.e. renal disease, diabetes, hypertension, cardiovascular disease), 4) No

clinical signs of infection or labour 5) Free from any infections during pregnancy including HIV,

Hepatitis B and group B streptococcus. Indications for delivery by caesarean section included:

abnormal presentation of the fetus, cephalopelvic disproportion, or previous caesarean sections.

Patients were not screened for bacterial vaginosis.

3.2.2 Mixed Amnion Epithelial and Mesenchymal Cell Culture

Amnion membranes were stripped from the underlying choriodecidual layer and washed in

saline until the majority of blood was removed. The amnion was suspended in forceps and cut in

49

strips by cutting towards the forceps at 1.5 cm intervals. The amnion was then digested for 45

mins in Dulbecco Modified Eagle Medium (DMEM; GIBCO Invitrogen; Burlington, Ontario,

Canada) + 0.1% trypsin (GIBCO Invitrogen) to obtain amnion epithelial cells in a 37°C

incubator. Tissue was then filtered using a 0.8 nm gauze filter. Filtered DMEM was treated with

filtered newborn calf serum (NCS; Wisent Incorporated; St. Bruno, Quebec, Canada) to inhibit

trypsin activity. Tissue was cut into fragments of approximately 1.0 cm in length and further

digested for 45 mins in DMEM + 0.1% trypsin; filtered and treated with filtered NCS to inhibit

trypsin activity. Epithelial cells from the first and second digestion were pelleted by

centrifugation at 1250 g for 5 mins. Excess medium was discarded and epithelial cells were re-

suspended in culture medium DMEM-F12 (GIBCO Invitrogen) + 10% fetal bovine serum (FBS;

Wisent Incorporated) + 1% antibiotics. The remaining tissue was cut into approximately 0.5 cm

pieces and digested for 1 hr in DMEM + 0.1% collagenase (Roche Diagnostics; Mannheim

Germany) in order to obtain amnion mesenchymal cells. After digestion, mesenchymal cells

were centrifuged for 10 mins at 1500 g. Total mesenchymal cells obtained were re-suspended in

culture medium and mixed with total epithelial cells from previous digestions, despite in vivo

proportions. Mixed amnion cells were counted using a Trypan Blue exclusion assay to ensure

greater than 98% viability. 20 µl of mixed amnion cell culture was diluted in 180 µl of culture

medium. 40 µl of diluted cultured cells were mixed with 40 µl Trypan Blue. Cells were them

viewed under 40x magnification and counted on a hemocytometre. Blue cells representing cell

death were assessed in the viewing field and cultures with dispersed cells displaying greater than

2% cell death were discarded. Mixed amnion epithelial and mesenchymal cells were plated to a

density of 0.8x106 cells per well in a 24-well culture plate.

3.2.3 Lactobacilli Preparation

Lactobacillus rhamnosus GR-1 was grown anaerobically for 48 hours to reach the stationary

growth phase in de Man, Rogosa, and Sharpe media (MRS) by colleagues in Dr. S.O. Kim’s

laboratory at the University of Western Ontario. The culture medium was then collected and

centrifuged at 6000 g for 10 minutes at 4⁰C. Residual bacteria were removed by filtration of the

supernatant through a 0.22 μm pore size filter. Supernatant was aliquoted and frozen at -80⁰C

until used.

50

3.2.4 Treatment Protocol

Cells were cultured for 48 hr at 37°C in 5% CO2 and 95% room air and washed with room

temperature phosphate buffered saline (PBS) without calcium and magnesium. Culture medium

was replaced and cell confluency was reached after an additional 48 hr incubation. Endotoxin

dose, time and Lactobacillus rhamnosus GR-1 supernatant optimization experiments were

performed. Both LTA (Invivogen; San Diego, California, United States of America) and LPS

(Sigma-Aldrich; St Louis, Missouri, United States of America) at doses 0.1, 1, 10, 100, and 1000

ng/ml were tested for a time period of 12 hr after starvation in serum-free conditions. Medium

was collected for LTA (10 ng/ml) and LPS (100 ng/ml) at time 4 hr, 8 hr, 12 hr and 24 hr post-

treatment. The effects of various Lactobacillus rhamnosus GR-1 supernatant dilutions, 1:10,

1:20, 1:50, and 1:100 were assessed at 12 hr post-treatment. After conditions were optimized,

cell cultures were serum starved for 12 hr. Lactobacillus rhamnosus GR-1 supernatant was

thawed and diluted in DMEM at a ratio of 1:20 and incubated for 12 hr. Excess supernatant was

discarded after each experiment. Fresh batches of supernatant were received every two months to

maintain similar levels of enzymatic activity across experiments. Cells were then treated with

LTA (10 ng/ml) or LPS (100 ng/ml) for 12 hr in serum-free DMEM and GR-1 treatment was

washed off and discontinued prior to the addition of the endotoxins. Aliquots of medium (100

µl) were collected and frozen at -80°C (Figure 3.1, page 60).

3.2.5 Lactate Dehydrogenase Assay

Cultured amnion cells were plated to a density of 0.8x106 cells per well in a 24-well culture

plate. After incubation, cells were treated with 1% Triton X (Sangon Limited; Mississauga,

Ontario, Canada), control, GR-1 (1:20 dilution), LTA (10 ng/ml), or LPS (100 ng/ml). Medium

was collected after a 12 hr incubation period and 100 µl was transferred into an optically clear

96-well plate. Reaction mixture was prepared to a ratio of catalyst solution (BioVision; Mountain

View, California, United States of America) 1:4 to dye solution (BioVision) and 100 µl was

immediately added to corresponding wells. The plate was incubated in the dark at room

temperature for 30 mins and then was read using a spectrophotometer at 495 nm with a reference

wavelength of 600 nm. Lactate dehydrogenase (LDH) was measured in collected medium from

each treatment group in relation to Triton X treatment groups representing one hundred percent

LDH release.

51

3.2.6 Cytokine and Chemokine Measurement

3.2.6.1 Enzyme-Linked ImmunoSorbent Assay

Levels of IL-1β, IL-4, IL-6, IL-8, IL-10, TNFα, and GM-CSF were measured in culture media in

commercially available Enzyme-Linked ImmunoSorbet Assay (ELISA) kits (GM-CSF from R

and D Systems, Burlington, Ontario, Canada; All other cytokines from EBioscience; San Diego,

California, United States of America). A 96-well plate was coated with 100 µl/well capture

antibody diluted in coating buffer to a dilution of 1:250. The plates were incubated overnight at

4°C and were then washed with 300 µl of wash buffer (PBS+0.5% Tween 20) 5 times for 1

minute incubation periods each. Between each wash, plates were blotted on absorbent paper.

Assay diluent diluted at a ratio of 1:1 in Millipore filtered water was added to each well at a

volume of 200 µl and incubated at room temperature for 1 hr. After 5 washes, prepared standard

sets in 1:2 dilutions were added to the plate at a volume of 100 µl/well. Samples were added in

1:10 dilutions for IL-6 and IL-8 kits and 1:1 dilution for IL-1β, IL-4, GM-CSF and TNFα kits at

a volume of 100 µl to each well. Plates were incubated overnight at 4°C after which the plates

were washed an additional 5 times. 100 µl to each well of detection antibody diluted to 1:250 in

assay diluent was added followed by an incubation of 1 hr at room temperature. Plates were

washed 5 times and avidin-horseradish peroxidase diluted to 1:250 in assay diluent was added

100 µl to each well and incubated in the dark at room temperature for 30 mins. Plates were

washed 7 times before addition of 100 µl of substrate solution to each well for 15 mins at room

temperature. 50 µl of stop solution (H2SO4) was added to each well before reading the plate in a

spectrophotometer at 450 nm using 570 nm as a reference wavelength. The minimum detection

limits for IL-1β, IL-4, IL-6, IL-8, IL-10, GM-CSF, and TNFα kits were 4.0 pg/ml, 2.0 pg/ml, 2.0

pg/ml, 4.0 pg/ml, 4.0 pg/ml, 7.8 pg/ml, and 2.0 pg/ml respectively.

3.2.6.2 Bioplex Pro Human 27-Cytokine/Chemokine Assay

A Bioplex Pro Human 27-Cytokine/Chemokine Assay (BioRad Laboratories Incorporated;

Mississauga, Ontario, Canada) was performed to measure the levels of the following cytokines

and chemokines with minimum detection limits as indicated: Platelet-Derived Growth Factor

(1.33pg/ml), IL-1ra (2.96 pg/ml), IL-4 (0.21 pg/ml), IL-6 (1.54 pg/ml), IL-8 (1.53 pg/ml), IL-9

(1.58 pg/ml), IL-10 (1.48 pg/ml), IL-12 (2.19 pg/ml), IL-15 (1.75pg/ml), IL-17 (1.36 pg/ml),

Eotaxin (1.51 pg/ml), Fibroblast growth factor (0.66 pg/ml), G-CSF (1.76 pg/ml), GM-CSF (0.84

52

pg/ml), IFNγ (1.44 pg/ml), IFN-inducible Protein (2.51 pg/ml), MCP-1 (1.22 pg/ml), MIP-1α

(1.02 pg/ml), MIP-1β (0.92 pg/ml), RANTES (1.20 pg/ml), TNFα (5.72 pg/ml), Vascular

Endothelial Growth Factor (1.93 pg/ml), IL-1β (1.92 pg/ml), IL-2 (1.1 pg/ml), IL-5 (1.79 pg/ml),

IL-7 (1.67 pg/ml), and IL-13 (2.2 pg/ml). Samples were prepared in a total volume of 100 µl

with 0.5% bovine serum albumin (BSA; Sigma-Aldrich). After addition of BSA, samples were

vortexed and centrifuged at 1000 g for 10 mins and stored on ice. Stock standard solution was

reconstituted and 96 µl were added to 183 µl culture medium to a total volume of 275 µl. 128 µl

of each the reconstituted standard were mixed with 72 µl of standard diluent to prepare standard

1. Serial dilutions of 1:2 were prepared to create standards 2-8. Next, 5 µl of beads were added

to each well in a 50 µl of assay buffer. Excess fluid was aspirated and the plate was washed 2

times with wash buffer. Standards and samples were mixed on a vortex shaker, added to wells at

a volume of 50 µl and incubated on a shaker covered with foil at room temperature at 1000 g for

30 mins. The plate was washed 3 times and 25 µl of detection antibody, prepared to a 1:10

dilution in detection antibody diluent, were added to each well. The plate was then incubated on

a shaker covered with foil at room temperature at 1000 g for 30 mins. The plate was washed 3

times and 50 µl streptavidin-PE, prepared to a 1:100 dilution in assay buffer, were added to each

well. The plate was incubated on a shaker with foil for 10 mins, washed 3 times and 125 µl of

assay buffer were added to each well. The plate was shaken at 1 100 rpm for 30 secs before it

was read in a Bio-Plex reader according to manufacturer`s software.

3.2.7 Statistical Analysis

Data are presented as mean values ± standard error of the mean (SEM) where each n represents

individual amnion cultures from placentas received from patients meeting inclusion and

exclusion criteria as state previously. Statistical significance between groups was determined

using a One-way Anova statistical test if data was normally distributed followed by Student-

Newman-Keuls post-hoc analysis or using a non-parametric Kruskal-Wallis statistical analysis if

data did not have normality or equal variation using STAT32 version 2.0 statistical software.

Statistical significance was accepted at p<0.05.

3.3 Results

There were no significant differences in measured LDH levels between Lactobacillus rhamnosus

GR-1 (1:20) dilution, LTA (10 ng/ml) and LPS (100 ng/ml) and control (DMEM only; Figure

53

3.2, page 61). Triton X treatment, representing maximal cell death confirmed the validity of the

LDH assay (n=5). Therefore, treatments do not cause significant amounts of amnion cell death.

After 12 hr stimulation, all subsequent treatments were performed in serum-free DMEM.

Previous studies have shown that the presence of serum in medium increased basal levels of

prostaglandin output in cultured human amnion cells (Whittle et al., 1994). Thereby eliminating

the presence of serum, any confounding stimulants of either cytokines or prostaglandins was

prevented.

Throughout experiments in this study, amnion cells were treated with an LTA dose of 10 ng/ml

or an LPS dose of 100 ng/ml and medium was collected after 12 hours incubation with or

without pre-treatment of Lactobacillus rhamnosus GR-1 supernatant diluted to a ratio of 1:20

with DMEM. These values were chosen based on IL-6 and IL-8 time course and dose

optimization studies. LPS increased levels of IL-6 to maximal stimulation at 100 ng/ml where

IL-8 peaked at a dose of 10 ng/ml. The concentration of IL-8 was significantly increased by an

LPS dose of 100 ng/ml compared to control (Figure 3.3, page 62). Therefore, an LPS dose of 100

ng/ml was used for subsequent experiments. IL-6 concentrations showed two peaks at an LTA

dose of 10 ng/ml and 1000 ng/ml. However, in contrast, IL-8 concentrations appeared to peak at

an LTA dose of 10 ng/ml which then decreased with higher doses of endotoxin (Figure 3.3, page

63). Therefore, based on preliminary data, a dose of 10 ng/ml was used for LTA treatment.

To assess time course response, medium was collected and analyzed 4, 8, 12 and 24 hours after

LTA (10 ng/ml) or LPS (100 ng/ml) treatment (Figure 3.4, page 63). At 12 hr, both IL-6 and IL-

8 showed significant increases compared to 4 hr control by LPS stimulation. LTA also showed

increased IL-6 and IL-8 at 12 hr compared to 4 hr and 8 hr control. Although LTA significantly

increased IL-6 at 24 hr and LPS increased IL-8 at 24 hr compared to 4 hr and 8 hr controls, the

24 hr control levels of both cytokines appear to be elevated above the other control groups.

Therefore, a time point of 12 hr was chosen due to the elevated basal cytokine concentrations

observed at 24 hr and to the significantly increase in both IL-6 and IL-8 that was stimulated by

LTA and LPS. Additionally, numerous cytokines and chemokines showed peak stimulation at 12

hrs as assessed by Bioplex analysis including: MCP-1, RANTES, and IFNγ (data not shown).

54

Based on analysis of medium collected from amnion cells treated with various dilutions of

Lactobacillus rhamnosus GR-1 supernatant, the greatest differences in IL-6 concentration

compared to control were observed at a Lactobacillus rhamnosus GR-1 supernatant dilution of

1:20 (Figure 3.5, page 64). Lactobacillus rhamnosus GR-1 significantly increased IL-8

concentration compared to control, 1:50 and 1:100 treatments. Therefore, we chose a 1:20

dilution for all experiments. The same dilution of supernatant, 1:20 was used in previous studies

in studies on primary cultures of placental trophoblast and mouse macrophages (Kim et al., 2006;

Yeganegi et al., 2009; Yeganegi et al., 2010; Yeganegi et al., 2011). The lack of response for a

1:10 dilution make indicate a possible dose-dependency or threshold level required for the

unknown metabolite or metabolites in Lactobacillus rhamnosus GR-1 supernatant to exert its

effects on the unknown receptor present on the amnion. In addition, the bell curve shape

obtained suggests that the unknown signalling molecule present in Lactobacillus rhamnosus GR-

1 supernatant requires optimal dosage to exert its effects on target tissues. Competition may arise

if the concentration of this molecule is too high or homodimerization may occur preventing

interaction with the appropriate receptors. However, due to the unknown identity of the

molecule responsible for signalling in probiotic strains, these suggestions are merely speculation.

Two different methods were used to analyze cytokines and chemokines in these studies. To

ensure results were comparable between both methodologies, the medium collected from both a

pregnancy carrying a male fetus and one carrying a female fetus whose levels, as previously

measured by ELISA ,where closest to the preliminary mean were assessed once again in ELISA

and simultaneously in a Bioplex assay for IL-6 (Figure 3.6, page 65). Measured IL-6 levels were

highly correlated between the two methodologies (R2 –value = 0.8728).

Both LTA (10 ng/ml) and LPS (100 ng/ml) caused statistically significant increases in TNFα

concentrations in cultured amnion cell medium compared to control (Figure 3.7, page 66). There

was no significant change in TNFα with treatment of GR-1 (1:20) compared to control. The

increase in TNFα with either LPS or LTA treatment was attenuated when cultured amnion cells

were pre-treated with GR-1 in GR-1+LTA and GR-1+LPS treatment groups. No significant

changes were observed with other treatment groups. IL-1β (N=8), IL-10 (N=8) and IL-4 (N=6)

levels were undetectable across all treatment groups with lowest detectable limits (data not

shown).

55

Both LTA (10 ng/ml) and LPS (100 ng/ml) caused significant increases in IL-6 concentrations in

cultured amnion cell medium compared to control (Figure 3.8, page 67). GR-1 (1:20) caused a

significant increase in IL-6 concentration compared to control. GR-1+LTA and GR-1+LPS

treatments caused a significant increase in IL-6 concentration compared to control. No other

significant changes were observed across all other treatment groups.

Both LTA (10 ng/ml) and LPS (100 ng/ml) caused significant increases in IL-8 concentrations in

cultured amnion cell medium compared to control (Figure 3.9A, page 68). GR-1 (1:20) caused a

significant increase compared to control. GR-1+LTA and GR-1+LPS treatments caused a

significant increase compared to control group. RANTES concentrations were increased by GR-

1 compared to control (Figure 3.9B, page 68). As well, GR-1+LTA significantly increased

RANTES compared to control. GR-1+ LPS treatment significantly increased RANTES

compared to control. No other significant changes were observed between treatment groups.

MCP-1 was increased by GR-1 compared to control (Figure 3.9C, page 68). GR-1+LTA

significantly increased MCP-1 compared to control, LTA and LPS alone. As well, GR-1+LPS

treatment showed significant increases of MCP-1 compared to control, LTA and LPS alone. No

other significant changes were observed between treatment groups.

There was no significant change in IFNγ concentrations in medium collected from cultured

human amnion cells with either LTA (10 ng/ml) or LPs (100 ng/ml; Figure 3.10A, page 69). GR-

1 (1:20) significantly increased IFNγ concentration compared to control, LTA and LPS alone.

GR-1+LTA significantly increased IFNγ concentration compared to control, LTA and LPS

alone. GR-1+LPS significantly increased IFNγ concentration compared to control, LTA and

LPS alone. Neither GR-1 nor LTA and LPS caused significant increases in GM-CSF

concentration compared to control (Figure 3.10B, page 69). GR-1+LTA significantly increased

GM-CSF concentration compared to control, GR-1 and LTA alone. There was no significant

change in GM-CSF concentration with either LPS or GR-1+LPS treatments.

All data were additionally analyzed according to the gender of the fetus carried in the pregnancy

from which tissue was received. For each cytokine measured, the numbers for tissues derived

from a pregnancy carrying a female fetus versus a male fetus were as follows: TNFα (2 female: 3

male), IL-6 (4 female: 9 female), IL-8 (5 female: 6 female), RANTES (3 female: 5 male), MCP-

1 (3 female; 5 male), IFNγ (4 female: 6 male), and GM-CSF (3 female: 5 male). No obvious

56

trends were observed based on the gender of the fetus carried in pregnancies from which amnion

tissue was derived (data not shown).

57

3.4 Comment

The current study focused on the effects of Lactobacillus rhamnosus GR-1 on LTA and LPS

stimulated human amnion cells. In accordance with previous results, we have shown that

Lactobacillus rhamnosus GR-1 attenuated LTA and LPS-stimulated TNFα release from the

amnion similar to experiments performed on human placental trophoblasts and murine

macrophages (Yeganegi et al., 2010; Kim et al., 2006). Additionally, we demonstrated the

ability of Lactobacillus rhamnosus GR-1 to increase chemokine production, namely IL-8, MCP-

1 and RANTES as well as cytokines capable of activating recruited immune cells, IFNγ and GM-

CSF. Together, these results suggest that Lactobacillus rhamnosus GR-1 promotes a strong

chemotactic response in the amnion while minimizing the potential for an exaggerated pro-

inflammatory immune response to enable the amnion to defend against bacterial endotoxin

entrance into the fetal environment.

TNFα is a member of the pro-inflammatory triad alongside IL-1β and IL-6. As expected in

accordance with previous literature, IL-1β was not found above detectable limits in our culture

system. However, Lactobacillus rhamnosus GR-1 attenuated both LTA and LPS stimulated

TNFα production and increased IL-6 production to a similar level of LTA and LPS alone. TNFα

is only detectable in the amniotic fluid in the presence of an infection and/or inflammatory

response and is therefore indicative of an underlying infection (Romero et al., 1989c). The

influence of TNFα on the gestational tissues results in the activation of downstream pathways

leading to infection and/or inflammation-mediated preterm delivery. Alongside IL-1β, TNFα

elevates prostaglandin production in the gestational tissues leading to metalloproteinase

stimulation and subsequent cervical ripening and fetal weakening as collagen is degraded as well

as myometrial activation (Blumenstein et al., 2000; Bry et al., 1992; Pollard et al., 1993; Romero

et al, 1989b; Romero et al., 1989c). These events normally occur in parturition at term and early

activation by uterotonins, such as prostaglandins can lead to preterm delivery. By decreasing the

release of LTA and LPS-stimulated TNFα from the human amnion, Lactobacillus rhamnosus

GR-1 may significantly interfere with activation of infection and/or inflammation-mediated

processes associated with preterm labour in the amnion.

The amnion responds to Lactobacillus rhamnosus GR-1 with increased release of IL-6, IFNγ and

GM-CSF prior to LTA and LPS stimulation. IL-6, IFNγ, and GM-CSF all mediate the extent of

58

the pro-inflammatory response initiated by TNFα and IL-1β. These cytokines increase the

production of the antagonists to TNFα and IL-1β, namely the soluble TNFα receptor and IL-1

receptor antagonist (Aderka et al., 1989; Schindler et al., 1990; Tilg et al., 1994; Xing et al.,

1998). The soluble TNFα receptor binds and neutralizes TNFα. Similarly, IL-1 receptor

antagonist binds to IL-1 receptor preventing attachment of its agonist, IL-1β. Therefore,

Lactobacillus rhamnosus GR-1 may indirectly attenuate the effects of TNFα and IL-1β by

increasing cytokines that may stimulate the production of pro-inflammatory cytokine

antagonists.

The fetal membranes are situated between the amniotic fluid and the maternal gestational tissue.

As such, they act as a defensive barrier for pathogens, bacterial products and endotoxin from

reaching the fetal environment. We have demonstrated that Lactobacillus rhamnosus GR-1

stimulates a cytokine profile in the human amnion that may be capable of recruiting and

activating immune cells to the fetal membranes. Immune cells of monocyte and neutrophil origin

are localized to both the amnion and the chorion and neutrophilia in these tissues increases at

term (Witkin et al., 2011). Our studies show that Lactobacillus rhamnosus GR-1 increases

chemokines of both the CXC and CC family, namely IL-8 and MCP-1 and RANTES. IL-8 is

responsible for the recruitment and activation of neutrophils whereas MCP-1 and RANTES play

a role in the chemotaxis of monocytes. In addition, Lactobacillus rhamnosus GR-1 enhanced

MCP-1 production when challenged with LTA and LPS compared to endotoxin stimulation

alone. Therefore, Lactobacillus rhamnosus GR-1 may permit chemotaxis of immune cells,

especially monocytes to the amnion to allow for increased defense against bacterial endotoxin.

The activation state of monocytes/macrophages is dependent on the immunological signals

present in the microenvironment to which they are recruited. Although classically defined as

taking either an M1 or an M2 phenotypic state of activation, recent research has shown that the

activity of the macrophage is much more continuous. The M1 phenotype is characterized by

strong pro-inflammatory properties that encourage potent bacterial death and clearance (Mehta et

al., 2004; Sierra-Filardi et al., 2011). In comparison, the M2 phenotype encourages attenuated

pro-inflammatory responses and elevated anti-inflammatory responses (Mehta et al., 2004). In

this study Lactobacillus rhamnosus GR-1 increased the production of IFNγ and when pre-treated

prior to LTA and LPS stimulation, increased both IFNγ and GM-CSF release from amnion cells.

59

In conjunction with TNFα, IFNγ and GM-CSF promote the activation of macrophages to an M1

phenotype. However, Lactobacillus rhamnosus GR-1 attenuated LTA and LPS-stimulated TNFα

production in amnion cells. When stimulated with IFNγ in the absence of TNFα, immune cells

are activated to a phenotype characterized by its ability to phagocytose and produce anti-

microbial products with a less severe pro-inflammatory response (Mosser et al., 2008).

Therefore, Lactobacillus rhamnosus GR-1 promotes the amnion to produce cytokines capable of

activating immune cells to defend against pathogenic threats while minimizing the associated

pro-inflammatory response that may be deleterious to the prolongation of gestation.

In summary, Lactobacillus rhamnosus GR-1 increases the ability of the amnion to defend against

bacteria and bacterial products while minimizing the extent of the accompanying pro-

inflammatory response. However, caution must be taken in interpretation of these results for an

exaggerated immune cell response may also be deleterious to the prolongation of gestation if not

restricted. Chorioamnionitis is associated with preterm delivery that is characterized by increased

neutrophilia in the membranes. Therefore, further studies are necessary to determine whether or

not the regulated activation of monocytes/macrophages provided by cytokine and chemokines

up-regulated by Lactobacillus rhamnosus GR-1 in the amnion is conducive to defense against

pathogens as well as the prolongation of gestation. In addition, IL-6 has been shown to stimulate

prostaglandin production in the amnion (Kent et al., 1993; Mitchell et al., 1991). IL-6 is also

elevated in conjunction with chorioamnionitis and at term in amniotic fluid (Imseis et al., 1997).

As mentioned previously, prostaglandin elevation is associated with infection and/or

inflammation-mediated preterm birth. Lactobacillus rhamnosus GR-1 increases IL-6 to a level

comparable to LTA and LPS stimulation alone. Therefore, further experiments are necessary to

determine if a controlled immune cell response and IL-6 stimulation of prostaglandin production

is deleterious.

60

48

ho

urs

4

8 h

ou

rs

12

ho

urs

1

2 h

ou

rs

12

ho

urs

Pri

ma

ry

Cu

ltu

re

Med

ium

Co

llec

tion

Wash

Cu

ltu

re u

nti

l co

nfl

uen

t

(≤ 9

6 h

ou

rs)

Sta

rve

GR

-1 (

1:2

0)

LT

A

(10 n

g/m

l)

LP

S

(100 n

g/m

l)

Fig

ure

3.

1:

Exp

erim

enta

l d

esig

n

Pri

mar

y a

mnio

n c

ult

ure

s w

ere

gro

wn f

or

48 h

ours

and t

hen

was

hed

. A

fter

anoth

er 4

8 h

our

when

cel

ls r

each

ed c

onfl

uen

cy,

cult

ure

s w

ere

seru

m s

tarv

ed f

or

12 h

ours

. C

ells

wer

e pre

-tre

ated

wit

h L

act

obaci

llus

rham

nosu

s G

R-1

super

nat

ant

(GR

-1)

for

12 h

ours

. A

fter

12 h

ours

of

LT

A o

r L

PS

tre

atm

ent,

med

ium

was

coll

ecte

d.

61

Figure 3.2: The effect of treatments on amnion cell toxicity

Lactate dehydrogenase measured in the media of cultured human amnion cells treated with

control (DMEM), Lactobacillus rhamnosus GR-1 supernatant (GR-1, 1:20 dilution), lipoteichoic

acid (LTA, 10 ng/ml), and lipopolysaccharide (LPS, 100 ng/ml). Results are expressed as mean

values ± SEM relative to LDH release by Triton X-treated cell medium. Statistical significance

between groups is denoted with a different letter as assessed by One-way Anova followed by a

Student-Newman-Keuls post-hoc analysis (N=5, p<0.05).

a

b

b b b

0

20

40

60

80

100

120

Triton X control GR-1 LTA LPS

LDH

Lev

el (

%)

62

Figure 3.3: Lipoteichoic and lipopolysaccharide dose optimization

(A) IL-6 concentration (pg/ml; N=4, NS) and (B) IL-8 concentration (pg/ml; N=7, p<0.05) in media from amnion

cell cultures after treatment with control (DMEM), LTA or LPS doses of 0.1, 1, 10, 100, and 1000 ng/ml for 12

hours. Results are expressed as mean values ± SEM. Statistical significance within groups is denoted by different

letters as assessed by Kruskal-Wallis statistical test based on ranks followed by a Dunn’s post-hoc analysis. Circled

points show results at dosages chosen for subsequent experiments.

0

50

100

150

200

250

300

350

400

450

500

control 0.1 1 10 100 1000

IL-6

(p

g/m

l)

LTA

LPS

A

B

a

a

a

a b b

0

500

1000

1500

2000

2500

3000

3500

4000

4500

5000

control 0.1 1 10 100 1000

IL-8

(p

g/m

l)

Endotoxin Dose (ng/ml)

LTA

LPS

63

Figure 3.4: Time course responses of Interleukin-6 and Interluekin-8

Histogram displaying (A) IL-6 concentration (pg/ml) and (B) IL-8 (pg/ml) in media from amnion cell cultures after

treatment with control (DMEM), LTA (10 ng/ml) or LPS (100 ng/ml) for periods of 4, 8, 12, and 24 hours. Results

are expressed as mean values ± SEM. Significance was determined using a Kruskal-Wallis statistical test based on

ranks followed by a Dunn’s post-hoc analysis. * denotes significance to 4h control, + denotes significance to 8h

control, ** denotes significance to 12h control, ^ denotes significance to 4h LTA (N=6, p<0.05).

* ^

* +

** ^ *

0

500

1000

1500

2000

2500

3000

3500

4000

4h 8h 12h 24h

IL-6

(p

g/m

l)

control

LTA

LPS

* +

* +

*

* + ^

^

* + ^

* +

0

500

1000

1500

2000

2500

3000

3500

4000

4500

4h 8h 12h 24h

IL-8

(p

g/m

l)

control

LTA

LPS

A

B

64

Figure 3.5: Lactobacillus rhamnosus GR-1 dilution optimization

(A) IL-6 concentration (pg/ml) and (B) IL-8 concentration (pg/ml) in media from amnion cell cultures

after treatment with control (DMEM), Lactobacillus rhamnosus GR-1 supernatant (GR-1; 1:10, 1:20,

1:50, 1:100 dilutions). Results are expressed as mean values ± SEM. Statistical significance within groups

is denoted with a different letter as assessed by Kruskal-Wallis statistical test based on ranks followed by

a Dunn’s post-hoc analysis (N=3, p<0.05).

0

200

400

600

800

1000

1200

control 1:10 GR-1 1:20 GR-1 1:50 GR-1 1:100 GR-1

IL-6

(p

g/m

l)

a

b,c c

b

b

0

500

1000

1500

2000

2500

3000

3500

4000

4500

control 1:10 GR-1 1:20 GR-1 1:50 GR-1 1:100 GR-1

IL-8

(p

g/m

l)

A

B

65

Figure 3.6: Correlation between ELISA and Bioplex assay methodologies

Correlation between ELISA measurements and Bioplex measurements for Interleukin-6 in

human amnion cultures. (n=2, R2=0.8728).

R² = 0.8728

0

500

1000

1500

2000

2500

3000

3500

0 500 1000 1500 2000 2500 3000 3500 4000 4500

ELIS

A IL

-6 (

pg/

ml)

Bioplex IL-6 (pg/ml)

66

Figure 3.7: The effects of Lactobacillus rhamnosus GR-1 on Tumor Necrosis Factor - α

concentration in human amnion cells

Histogram showing TNFα (pg/ml) concentrations in media from human amnion cell cultures.

Control (DMEM), GR-1 (1:20), LTA (10 ng/ml), LPS (100 ng/ml). Results are expressed as

mean values ± SEM. Statistical significance between groups is denoted with a different letter as

assessed by One-way Anova followed by a Student-Newman-Keuls post-hoc analysis (N=5,

p<0.05).

a a

b

a

c

a

0

50

100

150

200

250

300

350

400

450

500

control GR-1 LTA GR-1+LTA LPS GR-1+LPS

TNFα

(p

g/m

l)

67

Figure 3.8: The effects of Lactobacillus rhamnosus GR-1 on Interleukin-6 concentration in

human amnion cells

Histogram showing IL-6 (pg/ml) concentrations in media from human amnion cell cultures.

Control (DMEM), GR-1 (1:20), LTA (10 ng/ml), LPS (100 ng/ml). Results are expressed as

mean values ± SEM. Statistical significance between groups is denoted with a different letter as

assessed by One-way Anova followed by a Student-Newman-Keuls post-hoc analysis (N=13,

p<0.05).

a

b

b b b

b

0

500

1000

1500

2000

2500

3000

control GR-1 LTA GR-1+LTA LPS GR-1+LPS

IL-6

(p

g/m

l)

68

Figure 3.9: The effect of Lactobacillus rhamnosus GR-1 on chemokine concentrations in

human amnion cells

Histograms showing (A) IL-8 (pg/ml) (B) RANTES (pg/ml) and (C) MCP-1 (pg/ml) concentrations in media from

human amnion cell cultures. Control (DMEM), GR-1 (1:20), LTA (10 ng/ml), LPS (100 ng/ml). Results are

expressed as mean values ± SEM. Statistical significance between groups is denoted with a different letter as

assessed by One-way Anova followed by a Student-Newman-Keuls post-hoc analysis (N=11 for IL-8, N=8 for

RANTES, N=8 for MCP-1, p<0.05).

a

b

b

b

b

b

0

1000

2000

3000

4000

5000

6000

7000

8000

control GR-1 LTA GR-1+LTA LPS GR-1+LPS

IL-8

(p

g/m

l)

a

b

a,b

b

a,b

b

0

20

40

60

80

100

120

140

160

control GR-1 LTA GR-1+LTA LPS GR-1+LPS

RA

NTE

S (p

g/m

l)

a

b

a

b

a

b

0

500

1000

1500

2000

control GR-1 LTA GR-1+LTA LPS GR-1+LPS

MC

P-1

(p

g/m

l)

A

B

C

69

Figure 3.10: The effects of Lactobacillus rhamnosus GR-1 on immune cell activating

cytokines

Histograms showing (A) IFNγ (pg/ml) and (B) GM-CSF (pg/ml) concentrations in media from

human amnion cell cultures. Control (DMEM), GR-1 (1:20), LTA (10 ng/ml), LPS (100 ng/ml).

Results are expressed as mean values ± SEM. Statistical significance between groups is denoted

with a different letter as assessed by One-way Anova followed by a Student-Newman-Keuls

post-hoc analysis (N=10, for IFNγ, N=8 for GM-CSF, p<0.05).

a

b

a

b

a

b

0

5

10

15

20

25

30

35

40

45

50

control GR-1 LTA GR-1+LTA LPS GR-1+LPS

IFN

γ (p

g/m

l)

a a

a,b

b

a,b

a,b

0

20

40

60

80

100

120

140

control GR-1 LTA GR-1+LTA LPS GR-1+LPS

GM

-CSF

(p

g/m

l)

A

B

70

Chapter 4

Effect of Lactobacillus rhamnosus GR-1 on Prostaglandin Output

and Prostaglandin Metabolizing Enzymes in Human Amnion Cells

at Term

71

Chapter 4

Effect of Lactobacillus rhamnosus GR-1 on Prostaglandin Output and Prostaglandin Metabolizing Enzymes in Human Amnion Cells at Term

4.1 Introduction

Preterm delivery is a major obstetrical concern as it accounts for up to 80% of all neonate

morbidities and mortalities. Its incidence has been rising over the past few decades and over

50% of all preterm births remain idiopathic despite advances in understanding its multifactorial

etiology. It is estimated that approximately 30% of all pre-term births are due to an underlying

infection and/or inflammation; however, the prevention of preterm birth by antibiotic treatment

is largely ineffective and in some cases increase its risk (Andrews et al., 2003; Okun et al.,

2005).

Bacterial vaginosis is an alteration to the endogenous vaginal microbiota where Lactobacillus

species decrease and pathogenic anaerobes increase and is known to increase the risk of preterm

birth by 40% (Hillier et al., 1995). Intrauterine infection can result from bacteria and/or bacterial

products gaining access to the intrauterine environment by ascent from the vagina through the

cervix (Goldenberg et al., 2000). The majority of microbes found in amniotic fluid samples

associated with intrauterine infections are of vaginal origin, leading us to study the effect of

lactobacilli on the amnion (Benirschke, 1960).

Infectious and/or inflammatory processes present in the intrauterine tissues are known to activate

a downstream cascade increasing prostaglandin production from the amnion, chorion and

decidua. In a state of infection, the chorionic 15-hydroxyprostaglandin dehydrogenase (PGDH)

metabolic barrier to prostaglandins produced in the amnion is decreased due to bacterial

endotoxin and cytokine release (Cheung et al., 1990; Sangha et al., 1994; Van Meir et al., 1996).

Bacterial endotoxin, bacterial products, cytokines, and chemokines increase net prostaglandin

output by up-regulating both phospholipase and prostaglandin H synthase (PGHS) -2 expression

in amnion cells (Kent et al., 1993; Mitchell et al., 1991; Pollard et al., 1993; Romero et al.,

1988a; Romero et al., 1989a). A net increase of prostaglandin production accompanies infection

and/or inflammation due to decreased metabolism and increased production. Prostaglandins

72

activate downstream pathways that result in a myometrial activation, cervical ripening and

membrane weakening.

Although numerous investigators have studied the role of antibiotics in preventing preterm birth,

their use has been shown to be largely ineffective and in some cases increases the risk of preterm

labour (Andrews et al., 2003; Okun et al., 2005). Probiotics are defined as, “live microorganisms

which when administered at adequate amounts confer a health benefit on the host” and have been

identified as a potential treatment alternative (FAO and WHO, 2001; Reid, 1999). Lactobacillus

rhamnosus GR-1 is capable of colonizing and restoring the microbiota of the vagina and

modulation of prostaglandin production enzymes in the gestational tissues (Reid et al., 2003;

Yeganegi et al., 2009).

In placental trophoblasts, Lactobacillus rhamnosus GR-1 increased expression of the

prostaglandin metabolizing enzyme, PGDH in pregnancies carrying a female but not male fetus

(Yeganegi et al., 2009). Also, Lactobacillus rhamnosus GR-1 reduced the LPS-induced increase

in PGHS-2 in placental trophoblast cells from pregnancies carrying a male but not female fetus

(Yeganegi et al., 2009). One of the major products of the amnion is prostaglandins. In a state of

infection and/inflammation, where metabolic barriers become less effective, the amnion

contribution to prostaglandin elevation may be detrimental to the maintenance of gestation.

Therefore, we hypothesized that Lactobacillus rhamnosus GR-1 will prevent alterations to

prostaglandin production and key prostaglandin metabolizing enzymes in amnion cells in

response to bacterial endotoxin stimulation.

4.2 Methodology

4.2.1 Placenta Collection

Placenta collection was performed as described in Chapter 3 (Page 48)

4.2.2 Amnion Mixed Epithelial and Mesenchymal Cell Culture

Amnion mixed epithelial and mesenchymal cell cultures were performed as described in Chapter

3 (Page 48-49). Cells were plated to a density of 8x106 cells per well in 60 mm culture dishes.

Protein was collected from cell lysates in RIPA lysis buffer + protease inhibitors (Roche

Diagnostics) and frozen at -80°C.

73

4.2.3 Lactobacilli Preparation

Lactobacillus rhamnosus GR-1 was grown as described in Chapter 3 (Page 49).

4.2.4 Treatment Protocol

Treatment was performed as described in Chapter 3 (Page 50; Figure 4.1, page 83).

4.2.5 Lactate Dehydrogenase Assay

LDH assay was performed as described in Chapter 3 (Page 50).

4.2.6 Prostaglandin E2 Measurement

4.2.6.1 Enzyme Immuno-Assay

Levels of PGE2 were measured using commercially available quantitative enzyme immuno-assay

(EIA) kits (Cayman Chemical; Ann Arbor, Michigan, United States of America). EIA buffer was

added to non-specific binding wells and maximum binding wells to a volume of 100 µl and 50 µl

respectively. Standards 2-8 were prepared in 1:2 dilutions in EIA buffer and standard 1 of 1000

pg/ml, and 50 µof each were added to wells. Samples were added in a dilution of 1:10 in EIA

buffer to a final volume of 50 µl to wells. Prostaglandin E2 AchE Tracer was reconstituted in 6

ml EIA buffer and 50 µl were added to all wells except Total Activity and Blank wells.

Prostaglandin E2 Monoclonal Antibody was reconstituted in 6ml EIA Buffer and 50 µl were

added to each well except Total Activity, Non-Specific Binding wells, and blank wells. Plates

were incubated overnight at 4°C. Plates were washed 5 times in 300 µl of Wash Buffer. Ellman’s

reagent colour developing solution was reconstituted in 20 ml of Millipore-filtered water and 200

µl were added to each well to develop colour reaction. Prior to incubation at room temperature

in the dark, 5 µl of Prostaglandin E2 Acetylcholine Esterase Tracer was added to the Total

Activity well. The plate was read when the 405 nm absorbance of the Maximum Binding wells

minus the 450 nm absorbance of the Blank wells was greater than 0.3, a time ranging from 60 to

90 minutes in a spectrophotometer (Tecan Infinite M200). The minimum detection limit for

PGE2 was 80% B/Bo of 15 pg/ml.

74

4.2.7 Prostaglandin Metabolizing Enzyme Expression Measurement

4.2.7.1 Western Blot Analysis

Protein samples in RIPA were thawed and immediately mixed on a vortex shaker before

incubation for 20 mins on ice. This was repeated 3 times followed by centrifugation at 4°C for

30 mins at 1000 g. Supernatant was then collected. Protein concentration was calculated using a

Bio-Rad Protein Assay with BSA serial dilutions as a standard. Samples were frozen at -80°C

until further analysis. Thirty micrograms of protein extract were mixed with specific volumes of

RIPA lysis buffer and 5µl of loading dye to a total volume of 25 µl.

Protein samples were run on a 10% sodium dodecyl sulphate poly-acrylamide gel electrophoresis

alongside a Pageruler pre-stained protein ladder (Fermentas; Burlington, Ontario, Canada). After

electrophoresis, proteins were transferred to a nitrocellulose membrane at 100 V for 90 mins.

Protein transfer was visually assessed using Ponceau S dye followed by 3 washes of 5 mins each

in PBS with 0.1 Tween 20 (PBST, Sigma-Aldrich). Membranes were blocked overnight at 4°C

in a blocking solution consisting of 5% non-fat skim milk powder (BioShop Canada

Incorporated; Burlington, Ontario, Canada) dissolved in PBST.

Membranes were incubated separately on a rocking platform in primary antibody diluted in

blocking solution for 1 hr at room temperature at the following dilutions: β-actin monoclonal

antibody 1:1000 (Sigma-Aldrich), TLR4 monoclonal antibody 1:250 (Cayman Chemicals),

PGHS-2 monoclonal antibody 1:500 (Sigma-Aldrich), PGDH monoclonal antibody 1:500

(Sigma-Aldrich), cPGES polyclonal antibody 1:500 (Cayman Chemicals), mPGES1 polyclonal

antibody 1:500 (Cayman Chemicals), and mPGES2 polyclonal antibody 1: 500 (Cayman

Chemicals). Membranes were washed 5 times for 5 mins in PBST. Secondary anti-mouse IgG

(anti-PGHS-2, anti-β-actin; GE Healthcare; Little Chalfont, Buckinghamshire, United Kingdom)

and anti-rabbit IgG (anti-cPGES, anti-mPGES1, anti-mPGES2; GE Healthcare), horseradish

peroxidase linked whole antibody 1:4000 dilution incubation on a rocking platform was

performed for 1 hr at room temperature at a dilution of 1:4000 for 1 hr. Membranes were

washed 6 times for 5 mins in PBST. Membranes were incubated in enhanced

chemiluminescence (ECL; Perkin Elmer, Waltham Massachusetts, United State of America)

substrate for 1 min or in ECL Advance (Perkin Elmer) for 2 mins and were exposed to Kodak X-

OMAT Blue XB film for 30 secs, 1 mins, 5 mins, 10 mins and 15 mins. Optical densities were

75

measured using Scion Imaging Software. Optical densities of proteins were divided by optical

density of β-actin to standardize for protein loading. These relative densities were then used for

statistical analysis using Scion Image computer analysis software version Beta 4.0.2.

4.2.8 Placental Trophoblast Cell Culture

Placental trophoblast cells were isolated based on previously established primary culture

protocols (Sun et al., 1997). Briefly, placental tissue was separated from fetal membranes and

washed in saline. Approximately 60 g of villous placental tissue was cut and digested in a 37°C

incubator with 0.125% trypsin and 0.02% deoxyribonuclease-I (Sigma-Aldrich) in DMEM for

30 mins. Supernatant was them removed and discarded as tissue was incubated in 0.125% trypsin

in a 37°C incubator two more times for 30 mins. Supernatant was collected after the second and

third digestion and centrifuged at 1700g for 10 mins and resuspended in DMEM. The pooled

supernatant was then filtered through a 200 μm pore size nylon gauze filter. Cells were then

purified using a Percoll gradient (35, 40, 65, 70 increments) and centrifuged for 20 mins at

2500g. Placental trophoblasts were removed from the appropriate Percoll band and resuspended

in culture medium. Cells were plated to a density of 106 in 60 mm culture dishes.

4.2.8.1 Treatment

Cells were incubated for 72 hr in 37°C 5% CO2 and 95% air in order to form a syncytium after

which cells were serum starved for 12 hr. Cells were then treated with Lactobacillus rhamnosus

GR-1 supernatant for 12 hours prior to LPS (100 ng/ml) incubation for 8 hr. Protein was then

collected from cell lysates in RIPA + protease incubator and frozen until use at -80°C.

4.2.8.2 Western Blot Analysis

Proteins were extracted and run on a Western Blot Analysis as previously described (Chapter 4,

page 73). 25 µg of protein was loaded as previously described (Chapter 4, page 73). Primary

antibodies were used to the following dilutions: β-actin monoclonal antibody 1:1000 and PGHS-

2 monoclonal antibody 1:500 (Sigma-Aldrich). Secondary antibodies were used to the following

dilutions: anti-mouse IgG (anti-PGHS-2, anti-β-actin) horseradish peroxidase linked whole

antibody 1:4000 dilution.

76

4.2.9 Statistical Analysis

PGE2 data are presented as average values ± SEM. Protein data are presented as average values ±

SEM relative to control. Statistical significance between groups was determined using a Kruskal-

Wallis statistical test based on ranks followed by a Dunn’s post-hoc analysis using STAT32

version 2.0 statistical software. A non-parametric Kruskal-Wallis statistical analysis was

performed based on the lack of normality and equal variation in the data of Chapter 4 as is

required for parametric One-way Anova statistical analysis. Statistical significance was accepted

at p<0.05.

4.3 Results

Out of the stimulants tested and at the amounts stated, only IL-1β (100 ng/ml) produced a

significant increases in PGE2 release compared to control treatment of DMEM alone in human

amnion cells (Figure 4.2, page 84). No significant differences in PGE2 production between LPS

of 1000 ng/ml, LPS of 100 ng/ml compared to control were observed. Dexamethasone had no

effect on PGE2 production. Both IL-1β and dexamethasone, in mesenchymal cells only, at 100

ng/ml have previously been shown to stimulate prostaglandin production in cultured human

amnion cells at term (Romero et al., 1989b, Whittle et al., 2000). MRS (1:20 dilution), the

medium in which Lactobacillus rhamnosus GR-1 was cultured was tested as a control for

supernatant treatment and did not cause a significant change in PGE2 release.

Placental trophoblast cell cultures were performed to confirm the duplication of previously

established results on LPS stimulation of PGHS-2 expression (data not shown). This was

performed to not only confirm culture technique but also to confirm the ability of LPS to modify

prostaglandin metabolizing and synthesizing enzyme expression. Placental trophoblasts were

cultured and then treated according to the following groups: control (DMEM), GR-1 (1:20

dilution), LPS (100 ng/ml), and GR-1+LPS. Final results were similar to findings by Yeganegi et

al. 2009. Although significance was not assessed due to an n-value of 2, similar trends were

observed as in experiments performed by Yeganegi et al., namely LPS increased PGHS-2

expression that was prevented by pre-treatment with GR-1.

PGE2 was significantly increased by GR-1 (1:20) alone and when pre-treated prior to endotoxin

stimulation with both LTA (10 ng/ml) and LPS (100 ng/ml) compared to control (DMEM)

77

treatment (Figure 4.3, page 85). In addition, GR-1+LTA stimulation caused a significant increase

in PGE2 concentration compared to control and LTA. GR-1+LPS stimulation caused a

significant increase in PGE2 concentration compared to control and LTA as well as a significant

decrease to GR-1. However, without pre-treatment with GR-1, neither LTA nor LPS caused

significant changes in PGE2 concentration compared to control.

To determine whether changes in PGE2 output were due to prostaglandin enzyme expression

modification, PGDH, PGHS-2, cPGES, mPGES1, and mPGES2 levels were measured across all

treatment groups. PGDH was found at very low levels across all treatment groups as was

expected from previously performed experiments indicating little or no PGDH is expressed in

the human amnion (N=6; data not shown). No significant differences in PGHS-2, cPGES or

mPGES1 levels were observed between treatment groups: control (DMEM only), GR-1 (1:20

dilution), LTA (10 ng/ml), GR- 1+LTA, LPS (100 ng/ml), and GR-1+LPS (Figure 4.4, 4.5, 4.6,

pages 86-88). Interestingly, a significant increase in mPGES2 expression was observed between

GR-1 and GR-1+LTA to control treatment groups (Figure 4.7, page 89). No other significant

changes were observed between all other treatment groups.

Expression of the different isoforms of PGES, cPGES, mPGES1, and mPGES2 were correlated

with PGE2 release from human amnion cells (Figure 4.8, page 90). An R2 –value of 0.6321,

0.6249, and 0.8214 exist for cPGES, mPGES1, and mPGES2 with PGE2 respectively.

All data were additionally analyzed according to the gender of the fetus carried in the pregnancy

from which tissue was received. For PGE2 and enzymes measured, the numbers for tissues

derived from a pregnancy carrying a female fetus versus a male fetus were as follows: PGE2 (4

female: 8 male), PGHS-2 (2 female: 4 male), cPGES (2 female: 5 male), mPGES1 (3 female: 5

male), mPGES2 (3 female: 6 male). No obvious trends were observed based on the gender of the

fetus carried in pregnancies from which amnion tissue was derived (data not shown).

78

4.4 Comment

The focus of this study was to determine the influence of Lactobacillus rhamnosus GR-1

supernatant on amnion prostaglandin output and modulation of the enzymes involved in their

metabolism. We have shown that Lactobacillus rhamnosus GR-1 significantly increased PGE2

release in human amnion cells. This response appears to be slightly attenuated in amnion cells

stimulated with either LTA or LPS after Lactobacillus rhamnosus GR-1 pre-treatment compared

to Lactobacillus rhamnosus GR-1 alone. From these results, we suggest that enhanced

prostaglandin production caused by Lactobacillus rhamnosus GR-1 permits the amnion to: 1)

Control the extent of a cytokine-induced immune cell response and 2) Provide a means of

pregnancy termination if the threat of bacterial endotoxin exposure becomes detrimental to the

health of the fetus.

Surprisingly, in our culture system we were unable to demonstrate enhanced prostaglandin

release by LPS stimulation in amnion cells as has been shown previously (Bennett, et al., 1987;

Reisenberger et al., 1998; Romero et al., 1988a; Shon et al., 2002). Consequentially, we

performed several subsequent experiments to verify our results. These included verification of

the viability of our cell culture as well as the activity of LPS and expression of its receptor,

TLR4. As well, we verified the ability to produce and secrete prostaglandins and performed

parallel experiments in a different cell culture, placental trophoblast cells.

Cells release prostaglandins during apoptosis. If a high proportion of cultured cells were in an

apoptotic phenotype at the time our samples were taken, a change in prostaglandin production by

LPS may be masked by elevated levels at baseline prior to stimulation. However, an LDH assay

indicated little cell death compared to those cells treated with Triton X, a compound that

solubilizes cell membranes to release into the cell medium a maximal level of LDH, an enzyme

regularly bound to the mitochondrial membrane. In this assessment, the proportion of LDH in the

medium collected across treatment groups is representative of the proportion of cell death. All

treatment groups including LPS at a dose of 100 ng/ml exhibited low LDH levels indicating little

cell death. Therefore, the lack of prostaglandin production stimulated by LPS was not due to an

elevated baseline release of prostaglandins due to cell death.

One potential cause for the lack of LPS-stimulated prostaglandin release in our cultured human

amnion cells may be a lack of biological activity of our LPS. It is well established that LPS

79

treatment of cultured human amnion cells increases the production of many cytokines. In our

system, we demonstrated that in amnion IL-6, IL-8 and TNFα release were stimulated by LPS at

a dose similar to previous experiments of 100 ng/ml. Therefore, the activity of LPS used in our

experiments shows biological activity capable of stimulating increased release similar to that

previously shown by other investigators.

LPS is a bacterial endotoxin that is recognized by cell membrane receptors of the body known as

TLRs. The main receptor for LPS is TLR4 that is expressed on many immune cells and tissues,

including the amnion. We also determined that TLR4 is expressed in our culture system

verifying the presence of the receptor for LPS on amnion cells present in our cell culture system.

Prostaglandin release from the amnion is stimulated by many factors including bacterial products

such as endotoxins, pro-inflammatory cytokines as well as glucocorticoids (Pollard et al., 1993;

Romero et al., 1988a; Romero et al., 1989b; Whittle et al, 2000; Zaga et al., 2004). Studies have

shown that an LPS dose within a range of 10 ng/ml to 1000 ng/ml stimulated prostaglandin

production and release from human amnion cells cultured with a similar method, incubation

period and treatment time as used in our experiments (Romero et al., 1988a, Thiex et al., 2009).

However, here both 100 ng/ml and 1000 ng/ml of LPS were unable to increase prostaglandin

production significantly compared to control groups. Similarly, dexamethasone at 100 ng/ml was

unable to increase prostaglandin production. It has been previously demonstrated that

mesenchymal amnion cells produce greater amounts of PGE2 compared to epithelial amnion

cells (Whittle et al., 2000). Dexamethasone may not have shown stimulation in mixed amnion

cell cultures in this study if the proportion of epithelial cells were much greater than

mesenchymal cells. The proportion of epithelial and mesenchymal cells obtained from the

culture techniques previously described, were not mixed to a ratio similar to that which exists in

vivo. Therefore a greater proportion of epithelial cells may account for the lack of

dexamethasone stimulation of PGE2 observed. Previous studies indicate that IL-1β increases

prostaglandin production and release in cultured human amnion cells through increases in

phospholipase and PGHS-2 expression at a dose of 100 ng/ml (Blumenstein et al., 2000; Bry et

al., 1992; Pollard et al., 1993; Romero et al., 1989a; Romero et al., 1989b). Similarly, we showed

that IL-1β at a dose of 100 ng/ml increase prostaglandin release from our cultured human amnion

cells (Romero et al., 1989b). Therefore, we have shown that in our system, prostaglandin release

can be stimulated by exogenous factors as was previously reported.

80

Placental trophoblasts are another gestational tissue responsive to LPS. Previously, our

laboratory has demonstrated the ability of LPS to modulate enzymes involved in prostaglandin

metabolism in cultured human placental trophoblast cells (Yeganegi et al., 2009). To verify not

only the activity of LPS by attempting to parallel previously established results, we cultured

placental trophoblast cells to verify culture and treatment technique. Previously, it had been

shown that LPS-stimulated PGHS-2 expression that was minimized by Lactobacillus rhamnosus

GR-1 pre-treatment (Yeganegi et al., 2009). These results were duplicated demonstrating that the

LPS used in amnion cell experiments is biologically active and verified culture techniques.

In human placental trophoblasts, Lactobacillus rhamnosus GR-1 supernatant prevented changes

in PGHS-2 expression caused by LPS in placentae derived from pregnancies carrying a male

fetus (Yeganegi et al., 2009). As well, in placentae derived from pregnancies carrying a female

fetus, Lactobacillus rhamnosus GR-1 supernatant increases PGDH expression to a higher degree

than in placentae derived from pregnancies carrying a male fetus (Yeganegi et al., 2009). In

contrast Lactobacillus rhamnosus GR-1 stimulated PGE2 release from amnion cells in this study.

In the amnion Lactobacillus rhamnosus GR-1 significantly increased mPGES2 expression alone

and when pre-treated prior to LTA stimulation. Together, with the trends observed in the

expression profiles of cPGES and mPGES1, the changes observed in mPGES2 may partly

account for the effects of Lactobacillus rhamnosus GR-1 on human amnion prostaglandin

production. Placental trophoblasts and the amnion have diverse roles throughout pregnancy and

are situated at anatomically distinct locations. Therefore it is not surprising that Lactobacillus

rhamnosus GR-1 exerts different effects on prostaglandin production and prostaglandin

metabolizing and synthesizing enzymes in placental trophoblasts and the amnion based on the

role of each tissue in gestation.

The amnion is the most biologically active producer of prostaglandins amongst the gestational

tissues at the onset of labour (Kinoshitak, 1984). Not surprisingly, numerous stimulants of

prostaglandin release from human amnion cells have been identified. These include but are not

limited to 1) bacterial products and endotoxins including phospholipases, LTA and LPS, 2) Pro-

inflammatory cytokines such as IL-1α, IL-1β, TNFα, and IL-6, 3) Chemokines, such as MIP-1α,

4) Anti-inflammatory cytokines, including IL-10 and IL-1ra, and 5) Glucocorticoids in

mesenchymal cells (Dudley et al., 1996; Furuta et al., 2000; Mitchell et al., 1991; Reisenberger

et al., 1998; Romero et al., 1989b; Romero et al., 1989c). Three of these stimulants stand out as

81

having paradoxically stimulatory roles in the amnion compared to their well-established anti-

inflammatory roles in other tissues: IL-1ra, IL-10, glucocorticoids (Economopoulos et al., 1996;

Gibb et al., 1990; Mitchell et al., 2004; Potestio et al., 1988). We have demonstrated that

Lactobacillus rhamnosus GR-1 can be added to this list of prostaglandin stimulants in cultured

human amnion cells.

The amnion increases prostaglandin production in response to stimulants, such as glucocorticoids

that normally prevent their production in other tissues. Therefore, one might speculate that

prostaglandins may play an important function in the role of the amnion, possibly to enhance the

barrier provided by the membranes against pathogenic bacteria and their products from reaching

the amniotic fluid. It is well-established that elevated levels of prostaglandins stimulate increased

expression of metalloproteinases that cause weakening of the fetal membranes and cervical

ripening ultimately leading to preterm delivery. Therefore, uncontrolled elevated levels of

prostaglandins stimulated by bacterial endotoxins and cytokines may lead to an elevated risk of

preterm birth. However, low or absent prostaglandin levels have been associated with

exaggerated immune responses due to the immunomodulatory role this family of eicosanoids

exert (Goodwin et al., 1983).

Arachidonic acid metabolites, including prostaglandins have significant roles in suppressing or

terminating immune responses (Chizzolini et al., 2009). An established function of PGE2 in

particular, is modulation of cytokine production in macrophages (Nataraj et al., 2001). PGE2 acts

to shift the activated macrophage phenotype from a strong pro-inflammatory response to a more

anti-inflammatory response (Harris et al., 2002). This shift in phenotype is characterized by

prevention of LPS stimulated release of type 1 immune response cytokines such as TNFα and IL-

12 while elevating the release of type two response cytokines such as IL-4 and IL-10 (Betz et al.,

1991, Katamura et al., 1995, Scales et al., 1989, Shinomiya et al., 2001, van der Pouw Kraan et

al., 1995). A similar role of PGE2 on the amnion or any other tissue has not been shown, but this

effect is modulated through EP4 (Nataraj et al., 2001). EP4 is expressed on the amnion; therefore

a similar modulation of cytokine response may occur due to the actions of PGE2 in the amnion. It

is important to consider that a metabolic barrier in the chorion expressing high levels of PGDH

may be maintained allowing the deleterious effects of increased prostaglandin production to be

minimized amongst the other gestational tissues while providing the immunoregulatory effects at

the amnion. Therefore, Lactobacillus rhamnosus GR-1 stimulation of prostaglandins in the

82

amnion tissue may aid in maintaining the anti-microbial defenses properties inherent to this layer

of the fetal membrane.

The amnion is the layer of tissue in closest proximity to the fetal environment. Therefore, the

ability of the amnion to defend against pathogenic threats may be the deciding factor if bacteria

and bacterial endotoxin reach the fetal environment causing developmental damage. The amnion

possesses inherent anti-microbial, anti-bacterial and anti-viral properties that prevent infection

during various obstetrical events. Although elevated levels can lead to activation of downstream

pathways associated with preterm labor, the production of prostaglandins may be crucial in the

defense provided by the amnion against pathogenic threats. Prostaglandins modulate the immune

response by minimizing pro-inflammatory cytokines from recruited immune cells. Therefore,

increased prostaglandin release caused by Lactobacillus rhamnosus GR-1 may contribute to the

overall ability of the amnion to defend against infections by controlling the extent of the

inflammatory response if levels of prostaglandins are also controlled and maintained.

Conversely, increased prostaglandin production by Lactobacillus rhamnosus GR-1 may provide

the amnion with the ability to terminate the pregnancy through increased prostaglandin

production in order to protect the fetus from exposure to bacterial endotoxin given that the threat

of deleterious effects associated with infection may be too great to allow for prolonged gestation.

Further experiments are required to differentiate between these potential effects of Lactobacillus

rhamnosus GR-1 on amnion production of prostaglandins.

83

Fig

ure

4.

1:

Exp

erim

enta

l d

esig

n

Pri

mar

y a

mnio

n c

ult

ure

s w

ere

gro

wn

for

48 h

ou

rs a

nd t

hen

was

hed

. A

fter

anoth

er 4

8 h

our

wh

en

cell

s re

ached

con

fluen

cy,

cult

ure

s w

ere

seru

m s

tarv

ed f

or

12 h

ours

. C

ells

wer

e pre

-tre

ated

wit

h

Lact

obaci

llus

rham

nosu

s G

R-1

super

nat

ant

(GR

-1)

for

12 h

ours

. A

fter

12 h

ours

of

LT

A o

r L

PS

trea

tmen

t, p

rote

in f

rom

cel

l ly

sate

s w

as c

oll

ecte

d.

48

hou

rs

48

ho

urs

1

2 h

ou

rs

12

ho

urs

1

2 h

ou

rs

Pri

ma

ry

Cu

ltu

re

Pro

tein

Co

llec

tion

fro

m C

ell

Ly

sate

s

Wash

Cu

ltu

re u

nti

l co

nfl

uen

t

(≤ 9

6 h

ou

rs)

Sta

rve

GR

-1 (

1:2

0)

LT

A

(10

ng/m

l)

LP

S

(10

0 n

g/m

l)

84

Figure 4.2: The effect of IL-1β, dexamethasone (Dex), lipopolysaccharide (LPS) and de

Man, Rogosa, and Sharpe (MRS) on prostaglandin production in human amnion cells

Histogram showing PGE2 output in media from human amnion cell cultures by various treatments.

Control (DMEM), IL-1β (100 ng/ml), Dexamethasone (100 ng/ml), Lipopolysaccharide (100 and 1000

ng/ml), and de Man, Rogosa, and Sharpe (1:20 dilution). Results are mean values ± SEM. Statistical

significance within groups is denoted with a different letter as assessed by Kruskal-Wallis statistical test

based on ranks followed by a Dunn’s post-hoc analysis (N=4, p<0.05).

a

b

a

a a

a

0

200

400

600

800

1000

1200

1400

1600

control Il-1β Dex LPS100 LPS1000 MRS

PG

E 2 (

pg/

ml)

85

Figure 4.3: The effect of Lactobacillus rhamnosus GR-1 on Prostaglandin E2 concentration

in medium from cultured human amnion cells

Histogram showing PGE2 in media collected from human amnion cell cultures. Control

(DMEM), LTA (10 ng/ml), LPS (100 ng/ml), GR-1 (1:20). Results are expressed as mean

values ± SEM. Statistical significance within groups is denoted with a different letter as assessed

by Kruskal-Wallis statistical test based on ranks followed by a Dunn’s post-hoc analysis (N=12,

p<0.05).

a

b

a

b

a

a,b

0

200

400

600

800

1000

1200

1400

1600

control GR-1 LTA GR-1+LTA LPS GR-1+LPS

PG

E 2 (

pg

/ml)

86

PGHS-2 72 kDa

β-actin 42 kDa

Figure 4.4: The effects of Lactobacillus rhamnosus GR-1 on PGHS-2 expression in human

amnion cells

Histogram showing (A) PGHS-2 expression fold increase relative to control in human amnion

cell cultures. Control (DMEM), LTA (10 ng/ml), LPS (100 ng/ml), GR-1 (1:20). Results are

expressed as mean values ± SEM. Statistical significance assessed using a Kruskal-Wallis

statistical test based on ranks followed by a Dunn’s post-hoc analysis (B) Relative Optical

Density (ROD) of PGHS-2/β-actin western blot (N=6, NS).

A

B

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

control GR-1 LTA GR-1+LTA LPS GR-1+LPS

PG

HS-

2/β

-act

in O

D (

rela

tive

to

co

ntr

ol)

Co

ntr

ol

GR

-1

LTA

GR

-1+

LTA

LPS

GR

-1+L

PS

87

cPGES 23 kDa

β-actin 42 kDa

Figure 4.5: The effect of Lactobacillus rhamnosus GR-1 on cytosolic PGES expression in

human amnion cells

Histogram showing (A) cPGES expression fold increase relative to control in human amnion cell

cultures. Control (DMEM), LTA (10 ng/ml), LPS (100 ng/ml), GR-1 (1:20). Results are

expressed as mean values ± SEM. Statistical significance assessed using a Kruskal-Wallis

statistical test based on ranks followed by a Dunn’s post-hoc analysis. (B) ROD of cPGES/β-

actin western blot (N=7, NS).

0

0.5

1

1.5

2

2.5

control GR-1 LTA GR-1+LTA LPS GR-1+LPS

cPG

ES/β

-act

in O

D (

rela

tive

to

co

ntr

ol)

A

B

Co

ntr

ol

GR

-1

LTA

GR

-1+

LTA

LPS

GR

-1+

LPS

88

mPGES1 16 kDa

β-actin 42 kDa

Figure 4.6: The effect of Lactobacillus rhamnosus GR-1 on microsomal PGES1 expression

in human amnion cells

Histogram showing (A) mPGES1 expression fold increase relative to control in human amnion

cell cultures. Control (DMEM), LTA (10 ng/ml), LPS (100 ng/ml), GR-1 (1:20). Results are

expressed as mean values ± SEM. Statistical significance assessed using a Kruskal-Wallis

statistical test based on ranks followed by a Dunn’s post-hoc analysis. (B) ROD of mPGES1/ β-

actin (N=8, NS).

0

0.2

0.4

0.6

0.8

1

1.2

1.4

control GR-1 LTA GR-1+LTA LPS GR-1+LPS

mP

GES

1/β

-act

in (

OD

rel

ativ

e to

co

ntr

ol)

B

A

Co

ntr

ol

GR

-1

LTA

GR

-1+

LTA

LPS

GR

-1+L

PS

89

mPGES2 33 kDa

β-actin 42 kDa

Figure 4.7: The effects of Lactobacillus rhamnosus GR-1 on microsomal PGES2 expression

in human amnion cells

Histogram showing (A) mPGES2 expression fold increase relative to control in human amnion

cell cultures. Control (DMEM), LTA (10 ng/ml), LPS (100 ng/ml), GR-1 (1:20). Results are

expressed as mean values ± SEM. Statistical significance within groups denoted with a different

letter as assessed by Kruskal-Wallis statistical test based on ranks followed by a Dunn’s post-hoc

analysis. (B) ROD of mPGES2/β-actin western blot (N=9, p<0.05).

a

b

a,b

b

a,b

a,b

0

1

2

3

4

5

6

control GR-1 LTA GR-1+LTA LPS GR-1+LPS

mP

GES

2/β

-act

in (

rela

tive

to

co

ntr

ol)

B

A

Co

ntr

ol

GR

-1

LTA

GR

-1+

LTA

LPS

GR

-1+

LPS

90

Figure 4.8: Correlation between PGES expression and PGE2 production in human amnion

cells

Correlation plot between (A) cPGES, (B) mPGES1 and (C) mPGES2 expression and PGE2

production in human amnion cell cultures (n=4, R2=0.6321 for cPGES, R

2=0.6249 for mPGES1,

and R2=0.8214 for mPGES2).

R² = 0.6321

0

0.5

1

1.5

2

0 200 400 600 800 1000 1200

cPG

ES/β

-act

ion

OD

(r

ela

tive

to

co

ntr

ol)

PGE2 (pg/ml)

R² = 0.6249

0

0.2

0.4

0.6

0.8

1

1.2

0 200 400 600 800 1000 1200

mP

GES

1/β

-act

ion

OD

(r

ela

tive

to

co

ntr

ol)

PGE2 (pg/ml)

R² = 0.8214

0

0.5

1

1.5

2

2.5

3

3.5

4

0 200 400 600 800 1000 1200

mP

GES

2/β

-act

in O

D

(re

lati

ve t

o c

otn

rol)

PGE2 (pg/ml)

A

B

C

91

Chapter 5

Overall Conclusions and Future Directions

92

5.1 Overall Conclusions

Infection and/or inflammation-mediated preterm birth is characterized by increased pro-

inflammatory actions amongst the gestational tissues. In the amnion specifically this includes

induction of a milieu of pro-inflammatory markers that are increased in the amniotic fluid during

intrauterine infection. In vitro studies have shown that many of these markers are stimulated by

LPS, a gram-negative bacterial endotoxin. This work has described the effects of Lactobacillus

rhamnosus GR-1 on LTA and LPS treated amnion cells and the modulation of cytokines,

chemokines, prostaglandins and prostaglandin metabolizing and synthesizing enzymes. The

effects of Lactobacillus rhamnosus GR-1 can be grouped accordingly as molecules that may

cause: 1) Increases in TNFα and IL-1β antagonists, 2) Recruitment of immune cells, 3)

Modulation of immune cell activity and, 4) Stimulation of PGE2.

5.1.1 Potential Increases in TNFα and IL-1β Antagonists

LPS induces TNFα from the amnion and IL-1β from intact fetal membranes. Both TNFα and IL-

1β are strong TH1 pro-inflammatory cytokines that alongside LPS and other bacterial products,

induce transcription of other cytokines through NκFB control as well as increased prostaglandin

production through modification of prostaglandin metabolizing enzymes. We have shown that

IL-1β is not detectable from our culture system as it was not detected above lowest acceptable

limit of the assay. However, similar to previous results shown in placental trophoblast cells and

murine macrophages, Lactobacillus rhamnosus GR-1 minimizes LTA and LPS stimulated TNFα

release from our human amnion culture system. Release of TNFα may also be decreased by the

effects that other cytokines stimulated by Lactobacillus rhamnosus GR-1 cause. Our studies

demonstrated that IL-6, IFNγ, and GM-CSF release from the amnion is increased by

Lactobacillus rhamnosus GR-1. Each of these cytokines has been shown across various systems

to increase the production of the soluble TNFα receptor and IL-1 receptor antagonist, the

antagonists to TNFα and IL-1β respectively (Aderka et al., 1989; Schindler et al., 1990; Tilg et

al., 1994; Xing et al., 1998). Essentially, IL-6, IFNγ and GM-CSF together provide the potential

to minimize TNFα and IL-1β stimulation of prostaglandins in the amnion. Therefore, by up-

regulating IL-6, IFNγ and GM-CSF, Lactobacillus rhamnosus GR-1 in addition to the direct

effects observed by supernatant alone prevents the actions of pro-inflammatory cytokines

93

responsible for initiation of downstream pathways leading to infection and/or inflammation-

mediated preterm birth.

Lactobacillus rhamnosus GR-1 increased IFNγ release from amnion cells compared to control

and endotoxin stimulation. In placental trophoblast cells, IFNγ reduced PGHS-2 expression and

prostaglandin production (Hanna et al., 2004). At term, this protective mechanism is lost as

IFNγ receptor expression decreases. The result is a net increase in prostaglandin production in

placental trophoblasts at term (Hanna et al., 2004). It is unknown if similar effects on expression

of amnion PGHS-2 are caused by IFNγ. However, an increase in IFNγ caused by Lactobacillus

rhamnosus GR-1 release form the amnion may act in a paracrine manner to help maintain low

expression levels of PGHS-2 in placental trophoblasts to help minimize prostaglandin

production. Consequentially, the cytokines elevated by Lactobacillus rhamnosus GR-1 may be

able to interact with the other gestational tissues to promote additional beneficial effects beyond

the amnion.

5.1.2 Recruitment of Immune Cells

The chemokines MCP-1, RANTES, and IL-8 are involved in chemotaxis of both neutrophils and

macrophages. Lactobacillus rhamnosus GR-1 increases production of these chemokines

compared to control and endotoxin treatments. Consequentially, increased chemotaxis may occur

in the amnion with the influence of Lactobacillus rhamnosus GR-1. Histological analysis of the

amnion shows the presence of immune cells of Hofbauer cells of placental origin in the compact

layer and as residential cells amongst the mesenchymal cells (Bourne, 1962). Therefore,

infiltration and activation of immune cells can occur in the amnion.

5.1.3 Modulation of Immune Cell Activity

Macrophages act as a primary defense against invading pathogenic threats, including bacteria.

The microenvironment to which immune cells are recruited dictate their phenotype and activity.

Bacterial toxins, cytokines, and chemokines all influence the activity of immune cells. The M1

macrophage phenotype is induced by TNFα, IFNγ, GM-CSF and LPS resulting in macrophages

that exhibit strong pro-inflammatory properties and encourage potent bacterial death (Mehta et

al., 2004; Sierra-Filardi et al., 2011). In contrast, the M2 macrophage phenotype is stimulated by

94

IL-4 and IL-13 resulting in macrophages that encourage attenuated pro-inflammatory responses

and elevated anti-inflammatory responses (Mehta et al., 2004). Recently, macrophage activation

has been described as more fluid than simply having two polar phenotypes. Studies have shown

that dependent on the cytokines, endotoxins and other stimulants present in the

microenvironment, macrophages can display characteristics along a wide spectrum between the

M1 and M2 phenotypes (Mosser et al., 2008).

Lactobacillus rhamnosus GR-1 increased IFNγ release from the amnion and Lactobacillus

rhamnosus GR-1 prior to LTA or LPS treatment increases GM-CSF compared to control

treatments. Therefore, Lactobacillus rhamnosus GR-1 increased two of the cytokines involved in

activation of monocytes/macrophages to a highly active pro-inflammatory state capable of

defending against intercellular pathogens. However, studies have shown that optimal activation

of an M1 phenotype is dependent on the activities of TNFα and IFNγ. Monocytes/macrophages

influenced by IFNγ without the presence of TNFα show a phenotype capable of phagocytosis

and production of anti-microbial molecules with a less severe pro-inflammatory phenotype

compared to those activated by both IFNγ and TNFα (Mosser et al., 2008). Lactobacillus

rhamnosus GR-1 attenuated LTA and LPS-stimulated TNFα and increased IFNγ release from the

amnion. Therefore, by the effect of Lactobacillus rhamnosus GR-1 in the amnion, it is expected

that immune cells are recruited and activated to a phenotype capable of bacteria and bacterial

product defense accompanied by a less severe pro-inflammatory responses.

Other studies have demonstrated that prostaglandins and other lipid mediators also influence the

activity of recruited immune cells. Lactobacillus rhamnosus GR-1 increased the release of PGE2

from amnion cells. Amongst all prostaglandins, PGE2 shows the greatest effect on modulation of

macrophage activity (Chizzolini et al., 2009). The role of PGE2 on macrophages is described as

immunosuppressive. In particular, PGE2 modifies the activity of macrophages towards a M2

phenotype as strong pro-inflammatory responses are minimized as anti-inflammatory responses

dominate (Harris et al., 2002). This shift in phenotype is characterized by prevention of LPS

stimulated release of type 1 immune response cytokines such as TNFα and IL-12 while elevating

the release of type two response cytokines such as IL-4 and IL-10 (Betz et al., 1991, Katamura et

al., 1995, Scales et al., 1989, Shinomiya et al., 2001, van der Pouw Kraan et al., 1995). Other

95

studies demonstrate an immunomodulatory role for Lactobacillus rhamnosus GR-1 in mouse

macrophages. In particular, LPS-treated mouse macrophages pretreated with supernatant show

increased levels of G-CSF production, an anti-inflammatory cytokine (Kim et al., 2006). G-CSF

is also increased by Lactobacillus rhamnosus GR-1 supernatant in a fetal sex-dependent manner

in LPS-stimulated placental trophoblast cells (Yeganegi et al., 2011). Therefore, Lactobacillus

rhamnosus GR-1 increases many molecules involved in the activation of immune cells that

appear to promote defense against bacteria and their products while minimizing pro-

inflammatory responses.

Therefore, Lactobacillus rhamnosus GR-1 increased chemokines that may recruit immune cells

while also increasing chemokines, cytokines and prostaglandins that influence the activity of

activated macrophages. The recruitment and activated phenotype of immune cells is difficult to

interpret from the given data and future experiments are necessary to determine this. However,

these results suggest it may be possible that Lactobacillus rhamnosus GR-1 causes immune cells

to be recruited and activated to a state capable of defending against pathogens while the pro-

inflammatory actions that accompany this response are minimized.

5.1.4 Stimulation of Prostaglandin E2

The supernatant of Lactobacillus rhamnosus GR-1 increases cytokines and immune cell activity

that have the potential to minimize LPS and pro-inflammatory prostaglandin stimulation. In

parallel, Lactobacillus rhamnosus GR-1 directly enhances prostaglandin production in amnion

cells. This is similar to the effects of Lactobacillus rhamnosus GG demonstrated to increase

PGHS-2 expression in T84 colon epithelial cells (Korhonen et al., 2004). As well, Lactobacillus

rhamnosus GR-1 increased IL-6 in amnion cells. IL-6 has been shown to stimulate prostaglandin

production in the amnion (Kent et al., 1993; Mitchell et al., 1991). These results suggest that

prostaglandins are stimulated in a more controlled manner, independent of LPS and pro-

inflammatory stimulation in the presence of Lactobacillus rhamnosus GR-1. Amnion produced

prostaglandins may exert stimulatory actions on the myometrium, and cause structural changes

to the fetal membranes and cervix in conjunction with preterm birth when produced in

uncontrolled levels. Lactobacillus rhamnosus GR-1 attenuates LPS stimulation of prostaglandins

perhaps minimizing the harmful effects of deregulation caused in infection and/or inflammation-

96

mediated preterm birth. However, prostaglandins may be produced both directly and indirectly

by actions exerted by Lactobacillus rhamnosus GR-1 in the amnion.

Various physiological stimuli have been shown to increase prostaglandin production in the

amnion including glucocorticoids and IL-10 (Economopoulos et al., 1996; Gibb et al., 1990;

Mitchell et al., 2004; Potestio et al., 1988). Both glucocorticoids and IL-10 exert typically anti-

inflammatory actions in a wide variety of other systems. This indicates that the role of amnion

prostaglandins may be crucial to amnion function. The amnion displays anti-bacterial, anti-

fungal, and anti-viral properties. These include the expression of natural antimicrobials and

innate immune molecules including human α-defensins 1, 2 and 3, human β-defensins 1, 2 and 3

as well as elafin and secretory leukocyte protease inhibitor (King et al., 2007). The innate

expression of such a variety of components of the innate immune system suggests a major role of

the amnion is to prevent uterine infection during pregnancy and labour (King et al., 2007).

Prostaglandins are released from macrophages, NK cells, mast cells, basophils and eosinophils as

part of the immediate innate immune response. Prostaglandins act to perpetuate the

inflammatory response by modulating endothelial cells to allow for entrance of additional

immune cells including leukocytes and are capable of degrading membrane phospholipids of

pathogenic bacteria. Therefore a basal level of prostaglandin production stimulated by various

physiological stimuli may help maintain the important anti-microbial role innate to the amnion.

Consequentially, glucocorticoids, IL-10 and Lactobacillus rhamnosus GR-1 stimulation of

prostaglandins may act to enhance the protective mechanism of the amnion to bacterial and

endotoxin exposure to the fetal environment.

During infection and/or inflammation-mediated preterm birth, bacteria, bacterial products and

pro-inflammatory cytokines, TNFα and IL-1β stimulate amnion prostaglandin production. As the

chorionic PGDH barrier is decreased by similar stimulants, elevated prostaglandin levels lead to

myometrial activation and increased metalloproteinase expression and subsequent cervical

remodeling and membrane weakening. We suggest that Lactobacillus rhamnosus GR-1 prevents

this uncontrolled pathway of prostaglandin production in the amnion and replaces it with a more

controlled probiotic and immunomodulatory cytokine induced pathway. This pathway is possibly

characterized by direct stimulation of prostaglandin production by Lactobacillus rhamnosus GR-

97

1 metabolites or through indirect consequences of Lactobacillus rhamnosus GR-1 induced IL-6

production (Figure 5.1, page 101). The abundance of stimulants of prostaglandin production in

the amnion including the paradoxical actions of IL-10, glucocorticoids and Lactobacillus

rhamnosus GR-1, suggest that prostaglandins play an essential role in the innate immune

functions of the amnion.

5.1.5 Additional Effects of Prostaglandins on the Amnion

The amnion is the closest layer of tissue to the fetal environment and as such is an important

barrier for entrance into the fetal environment. Part of the barrier function the amnion plays is

dictated by the permeability of its epithelial layer dictated by the sealing of the intercellular

space by part of the intercellular complex, tight junctions. In states of infection, bacteria,

bacterial products, cytokines and prostaglandins have been shown to alter tight junction integrity,

structure, and function in various different epithelial layers (Bruewer et al., 2003; Lecuit et al.,

2000; McNamara et al., 2001; Nusrat et al., 2001; Prasad et al., 2005; Talavera et al., 2004;

Tanaka et al., 2008). Disruption to the epithelial barrier is a common step in the inflammatory

response and causes initiation of a mucosal immune response that can then stimulate a stronger

pro-inflammatory immune response that may harm instead of protect the host.

Commensual strains of bacteria in the gastrointestinal tract, including lactobacilli can prevent

changes in tight junction function and structure induced by pathogenic bacteria. Lactobacillus

rhamnosus GG, a strain that adheres to the intestinal cell wall, prevents alterations to tight

junction structure and function caused by Escherichia coli O157:H7 in MDCK-I cell line

(Johnson-Henry et al., 2008). Lactobacillus rhamnosus GG is capable of preventing alterations

caused by Escherichia coli and thus may prevent increased permeability in the gut thereby

preventing entrance of the bacteria into the interstitial space. Lactobacillus plantarum can

prevent similar changes caused by Escherichia coli 0124:NM to both tight junction structure and

function in intestinal epithelial cells (Qin et al., 2009). Recently tight junctions have been

identified in both murine and human amnion (Kobayashi et al., 2009; Kobayashi et al., 2010a).

Inflammatory markers such as TNFα, LPS, and PGE2 have been shown to disrupt the amniotic

barrier through induction of apoptosis and alterations to tight junction structure (Kobayashi et al.,

2010b). Therefore, Lactobacillus rhamnosus GR-1 may also exert protective effects on human

98

amnion tight junction structure and function to help maintain their innate role in a state of

infection and/or inflammation-mediated preterm birth.

Lactobacillus rhamnosus GR-1 administered both orally and through vaginal instillation has

been shown to colonize and replenish the vaginal microbiota. As well, when injected directly

into the bladder of patients, this strain of lactobacilli does not result in any pathogenic responses.

Therefore, the use of therapeutic Lactobacillus rhamnosus GR-1 will most likely be safe for use

in pregnant women. In addition, in vitro studies have shown that both primary amnion and

placental trophoblast cultures, Lactobacillus rhamnosus GR-1 attenuated release of the pro-

inflammatory cytokine, TNFα demonstrating a beneficial immunomodulatory role when the

tissues are challenged with either LTA or LPS. This prevention of pro-inflammatory cytokine

release may act in ameliorating the activation of downstream pathways associated with infection

and/or inflammation-mediated preterm delivery. Additionally, in the amnion, Lactobacillus

rhamnosus GR-1 increases the defensive capabilities of the fetal membranes by recruiting and

activating immune cells capable of defending against pathogenic threats at the fetal environment

interface. Therefore, given the well-established safety of lactobacilli, the ability to colonize the

vagina and the beneficial immunomodulatory effects on the gestational tissues, the results of this

thesis have further demonstrated the potential for probiotic use in pregnancy to prevent infection

and/or inflammation-mediated preterm birth.

5.1.6 Lactobacillus rhamnosus GR-1 Supernatant Metabolite(s)

Although the exact signaling mechanism and molecule or molecules responsible for the effects

caused by the supernatant collected from Lactobacillus rhamnosus GR-1 is unknown, studies on

other probiotics have been performed. Previously, probiotic supernatant was shown to exert pro-

inflammatory actions through polysaccharide molecules and anti-inflammatory molecules

extracted from collected media (Chon et al., 2010). As well, probiotics have been shown to

signal through the MAPK and JAK/STAT pathways as JNK and JAK/STAT molecules are

phosphorylated through unknown metabolites present in the supernatant of various probiotics

(Kim et al., 2008; Yeganegi et al., 2010). Interestingly, the current study demonstrates a

threshold dilution of Lactobacillus rhamnosus GR-1 to increase IL-6 and IL-8 concentrations in

human amnion cells of 1:20 compared to 1:50 and 1:100 where no stimulation was shown.

99

Therefore, the metabolite or metabolites present in the Lactobacillus rhamnosus GR-1

supernatant may only exert its effects if it is present above a certain threshold at optimal

concentrations. As suggested by previously performed experiments, the unknown metabolite

may interact with TLR2 or TLR4, or unknown receptors. This may occur either through

antagonistic activity to the ligands of a given receptor by either blocking the ligands interaction

with the receptor of by leading to downstream signaling that may interfere with those activated

by pathogenic ligands such as LTA or LPS or by activating downstream signaling independent of

TLR2 and TLR4 that leads to an anti-inflammatory response. In addition, the current study

demonstrated that a 1:10 dilution of Lactobacillus rhamnosus GR-1 supernatant does not exert

the same effects on IL-6 and IL-8 as 1:20. This may suggest that homodimerization may occur

between the metabolite or competition between various metabolites may occur to block the

ability to activate associated receptors. Therefore, the effects exerted by probiotics are dependent

on both the strain of bacteria and the tissue and cell type of the host as well as the concentration

of metabolites produced by the bacteria.

5.1.7 Lactobacillus rhamnosus GR-1 and Prevention of Preterm Birth Associated with Infection and/or Inflammation

When considering the use of Lactobacillus rhamnosus GR-1 as a preventative treatment, the

effect of its supernatant on the individual gestational tissue must be considered. In placental

trophoblasts, Lactobacillus rhamnosus GR-1 exerts anti-inflammatory effects characterized by

increased IL-10 and G-CSF and decreased LPS-stimulated TNFα release (Yeganegi et al., 2009;

Yeganegi et al., 2010; Yeganegi et al., 2011). This prevention of LPS-stimulated TNFα release

is paralleled in the amnion; however it is accompanied by a strong chemotactic response capable

of modulating immune cell activity and increasing prostaglandin production. Although further

studies are required to determine if the in vivo effects Lactobacillus rhamnosus GR-1 on immune

cell recruitment and activation to the amnion are beneficial or detrimental to the prolongation of

gestation, the effect of this probiotic on up-regulation of PGE2 must be taken into consideration

when determining the route of administration for preventative treatment for infection and/or

inflammation mediated preterm birth. As preventative therapy, the most user-friendly method

would be oral administration possibly as part of a daily supplement for pregnancy women.

Orally administrated probiotics have been shown capable of reaching the vaginal flora and

100

despite the sterility of the amnion in healthy pregnancies, bacteria or their metabolites may be

capable of reaching the fetal membranes in vivo. Therefore, the dose and the potential

interactions with the inner layer of the fetal membranes, the amnion must be taken into

consideration in a more representative model than in vitro cell cultures to determine the best

method of preventative treatment for pregnancy women from undergoing infection and/or

inflammation mediated preterm birth.

5.1.8 Limitations

Several limitations are of importance in regards to the experiments and conclusions presented in

this study. Many experiments performed on the amnion are done on either intact membranes

together with the chorion or on dispersed cells separated into epithelial and mesenchymal cell

types. As the response of the amnion varies dependent on the presence of absence of not only the

other tissue of the membranes, the chorion as well as the partner cell type, epithelial or

mesenchymal cells the results of this study are limited in their in vivo representation. In addition,

cultured cells were not mixed in proportion to in vivo amnion tissue thus over or under

representation of either cell type may be possible.

When intact membranes are tested in a transwell system, shuttling of cytokine mRNA is thought

to occur between the amnion and chorion. Due to the lack of the chorion in the system used in

this study, both IL1β and IL-10 were not detected from mixed amnion cell cultures. IL-1β is a

strong pro-inflammatory cytokine that is thought to play a role in initiation of infection and/or

inflammation mediated preterm birth whereas IL-10 is an anti-inflammatory cytokine that

counteracts many pro-inflammatory actions and is compatible with the TH2 pregnancy cytokine

profile. Therefore, the effects of Lactobacillus rhamnosus GR-1 on IL-1β and IL-10 from intact

membranes cannot be determined using the system of mixed amnion cell cultures used in this

study.

101

Am

nio

tic

Flu

id

Am

nio

n

Lact

ob

aci

llus

rha

mn

osu

s G

R-1

met

abo

lite

(s)

+ LT

A o

r LP

S

Asc

en

din

g b

acte

rial

en

do

toxi

ns

LTA

or

LPS

TNFα

LTA

LPS

Me

mb

ran

e w

eak

enin

g

MM

P-2

, -9

PG

E 2

TNFα

IL-6

IFNγ

GM

-CSF

IL-8

MC

P-1

RA

NT

ES

Rec

ruit

men

t A

ctiv

atio

n

LTA

LPS

LPS

LTA

Me

mb

ran

e w

eak

enin

g

MM

P-2

, -9

PG

E 2

A

B

C

Fig

ure

5.1

: T

he

effe

cts

of

La

ctob

aci

llu

s rh

am

no

sus

GR

-1 o

n h

um

an

am

nio

n c

ells

. (A

) U

ntr

eate

d s

tate

of

infe

ctio

n a

nd/o

r in

flam

mat

ion

. T

NF

α l

evel

s in

crea

se P

GE

2

pro

du

ctio

n l

ead

ing t

o c

oll

agen

deg

rad

atio

n,

mem

bra

ne

wea

ken

ing a

nd

en

tran

ce o

f b

acte

rial

en

do

toxin

to t

he

amn

ioti

c fl

uid

and

th

e on

set

of

pre

term

lab

ou

r. (

B)

Eff

ect

of

La

ctob

aci

llu

s rh

am

no

sus

GR

-1 t

reat

men

t o

n i

mm

un

e ce

lls.

Tre

atm

ent

pri

or

to L

TA

or

LP

S s

tim

ula

tion

res

ult

s in

att

enu

atio

n o

f T

NF

α a

nd

in

crea

se c

hem

ota

xis

and

acti

vat

ion

of

imm

un

e ce

lls

to a

ph

eno

typ

e ca

pab

le o

f d

efen

din

g a

gai

nst

bac

teri

a an

d b

acte

rial

pro

du

cts

wh

ile

min

imiz

ing p

ro-i

nfl

amm

ato

ry r

esp

on

ses.

(C

) E

ffec

t o

f

La

ctob

aci

llu

s rh

am

no

sus

GR

-1 o

n p

rost

agla

nd

ins.

In

crea

sed

PG

E2 p

rod

uct

ion

by L

act

ob

aci

llu

s rh

am

no

sus

GR

-1 m

ay a

ct a

s a

pro

tect

ive

def

ense

agai

nst

pat

hogen

s

reac

hin

g t

he

feta

l en

vir

on

men

t b

y t

erm

inat

ing p

regn

ancy

if

bac

teri

al t

hre

ats

bec

om

e d

etri

men

tal

to t

he

hea

lth

of

the

fetu

s.

Ep

ith

elia

l ce

lls

Mes

ench

ym

al c

ells

Imm

un

e ce

lls

102

5.2 Future Directions

Currently, there is no effective treatment to prevent preterm delivery that occurs in

approximately 10% of all pregnancies. Consequentially, preterm neonates account for 80% of all

neonate morbidities and mortalities. The ultimate goal of the research presented in this thesis was

to aid in developing an effective preventative treatment for infection and/or inflammation

mediated preterm birth. The effects of Lactobacillus rhamnosus GR-1 on cytokine, chemokine

and prostaglandin production in the amnion described in this thesis hopefully bring researchers

one step closer to developing a treatment to prevent preterm birth and its associated

consequences.

The future directions of the current study include further investigation into the effects of

Lactobacillus rhamnosus GR-1 on products of the human amnion. Arachidonic acid is

metabolized into several different end products including leukotrienes, lipoxins and eicosanoids.

Both leukotrienes and lipoxins influence inflammation and immune cell responses. Another

molecular target that may be of interest is the metalloproteinases and TIMPs that are

differentially expressed under LPS stimulation. These enzymes are responsible for degradation

or prevention of degradation of collagen during membrane weakening and cervical remodeling.

They are up-regulated by prostaglandins and are part of the downstream pathways associated

with infection and/or inflammation-mediated preterm labour. Therefore, the effect of

Lactobacillus rhamnosus GR-1 on leukotrienes, lipoxins, MMPs and TIMPs may further

illustrate the effects of probiotic treatment on the immune response from the amnion.

The cytokine and chemokine profile produced in response to Lactobacillus rhamnosus GR-1

described in this thesis suggests an effect on the recruitment and activation of immune cells.

However, many of the cytokines and chemokines mentioned in this study have well-defined roles

amongst the other gestational tissues that have not yet been established in the amnion.

Therefore, before the conclusion of immune cell recruitment and activation can be confirmed,

determining the precise effects of cytokines and chemokines on the amnion is crucial. Of

particular interest are the effects of IFNγ on PGHS-2 expression and prostaglandin production as

well as GM-CSF and IL-6 on production of sTNFα receptor and IL-1ra in the amnion. In

103

addition, it will be important to determine if immune cell recruitment can occur in vivo and if this

reaction provides additional defenses or is deleterious to the prolongation of gestation.

In order to better understand the mechanism of action of the metabolites of Lactobacillus

rhamnosus GR-1, studies identifying the downstream signaling associated with probiotic

treatment should be determined. Previously, Lactobacillus rhamnosus GR-1 was shown to

modify the phosphorylation state of enzymes involved in MAP kinase and NFκB signaling

pathways. Probiotic enhancement of G-CSF in mouse macrophages is caused by a JNK-

dependent mechanism (Kim et al., 2006). Determining the effects of Lactobacillus rhamnosus

GR-1 on common pathways known to modulate cytokine transcription and translation will

provide insight into the signaling pathway.

By determining the effects of Lactobacillus rhamnosus GR-1 on the type of molecules produced

in the amnion that are associated with infection and/or inflammation-mediated preterm birth, a

better understanding of the effects of probiotic treatment on this tissue should be determined.

Since Lactobacillus rhamnosus GR-1 may cause both recruitment and activation of immune cells

and increased prostaglandin production from the amnion the use of its supernatant on pregnancy

may be two-fold. Firstly, by administering lactobacilli in a route that minimizes contact and

interaction with the amnion, pro-inflammatory responses in placental trophoblasts and amnion

may be attenuated and therefore prevent infection and/or inflammation preterm labour.

Secondly, it may be possible that administration of Lactobacillus rhamnosus GR-1 in a route that

allows for direct interaction of the supernatant with the amnion may be used to promote

protection from bacterial endotoxin entrance to the fetal environment for use in idiopathic or

induced labour where prolongation of gestation is deemed deleterious to the health of the fetus or

mother. However, these options may be dependent on the dose effects of Lactobacillus

rhamnosus GR-1 on the amnion which were not assessed in this study. Therefore, various routes

of supernatant administration including oral, vaginal or directly into the cervix may result in

different interaction of supernatant with the amnion and consequentially different effects on

infection and/or inflammation-mediated preterm birth.

104

The goal of the research presented in this thesis was to aid in the development of an effective

preventative treatment for infection and/or inflammation-mediated preterm birth. Hopefully, the

presented work has not only answered apparent questions but also opened routes and areas of

interest for future research. By answering these questions, an effective treatment will help

prevent the devastating consequences of preterm birth and lead to greater overall health and

success of high-risk pregnancies.

105

References

n.a. (2006). Preterm Birth: Causes, Consequences, and Prevention. Institute of Medicine of the

National Academies. Available at: http://www.iom.edu/~/media/Files/Report%20Files/2006/Pre

term-Birth-Causes-Consequences-and prevention/Preterm%20Birth%202006%20Report%20Bri

ef.pdf

n.a. Public Health Agency of Canada. (2008). Canadian Perinatal Health Report - 2008 Edition.

Available at: http://www.phac-aspc.gc.ca/publicat/2008/cphr-rspc/pdf/cphr-rspc08-eng.pdf

n.a. Food and Agriculture Organization of the United Nations (FAO) and World Health

Organization (WHO). (2001). Health and nutritional properties of probiotics in food including

powder milk with live lactic acid bacteria. FAO and WHO Joint Expert Committee Report,

2001. Available at: http://www.who.int/foodsafety/publicatons/fs_management/en/probiotics.pdf

Ackerman, W. E.,4th, Summerfield, T. L., Vandre, D. D., Robinson, J. M., & Kniss, D. A.

(2008). Nuclear factor-kappa B regulates inducible prostaglandin E synthase expression in

human amnion mesenchymal cells. Biology of Reproduction. 78, 68-76.

Aderka, D., Le, J. M., & Vilcek, J. (1989). IL-6 inhibits lipopolysaccharide-induced tumor

necrosis factor production in cultured human monocytes, U937 cells, and in mice. Journal of

Immunology (Baltimore, Md.: 1950). 143, 3517-3523.

Ahn, C., & Stiles, M. E. (1990). Antibacterial activity of lactic acid bacteria isolated from

vacuum-packaged meats. Journal of Applied Bacteriology. 69, 302-310.

106

Alfaidy, N., Xiong, Z. G., Myatt, L., Lye, S. J., MacDonald, J. F., & Challis, J. R. (2001).

Prostaglandin F2alpha potentiates cortisol production by stimulating 11beta-hydroxysteroid

dehydrogenase 1: A novel feedback loop that may contribute to human labor. Journal of Clinical

Endocrinology and Metabolism. 86, 5585-5592.

Alvi, S.A., Brown, N.L., Bennett, P.R., Elder, M.G., Sullivan, M.H. (1999). Corticotrophin-

releasing hormone and platelet-activating factor induce transcription of the type-2 cyclo-

oxygenase gene in human fetal membranes. Molecular Human Reproduction. 5, 476-480.

Andrews, W.W., Hauth J.C., Goldenberg, R.L. (2000). Infection and preterm birth. American

Journal of Perinatology. 17(7), 357-365.

Andrews, W. W., Hauth, J. C., Goldenberg, R. L., Gomez, R., Romero, R., & Cassell, G. H.

(1995). Amniotic fluid interleukin-6: Correlation with upper genital tract microbial colonization

and gestational age in women delivered after spontaneous labor versus indicated delivery.

American Journal of Obstetrics and Gynecology. 173, 606-612.

Andrews, W. W., Sibai, B. M., Thom, E. A., Dudley, D., Ernest, J. M., McNellis, D., Leveno,

K. J., Wapner, R., Moawad, A., O'Sullivan, M. J., Caritis, S. N., Iams, J. D., Langer, O.,

Miodovnik, M., Dombrowski, M., & National Institute of Child Health & Human

Development Maternal-Fetal Medicine Units Network. (2003). Randomized clinical trial of

metronidazole plus erythromycin to prevent spontaneous preterm delivery in fetal fibronectin-

positive women. Obstetrics and Gynecology. 101, 847-855.

107

Antonio, M. A., Hawes, S. E., & Hillier, S. L. (1999). The identification of vaginal

lactobacillus species and the demographic and microbiologic characteristics of women colonized

by these species. Journal of Infectious Diseases. 180, 1950-1956.

Aplin, J. D., Campbell, S., & Allen, T. D. (1985). The extracellular matrix of human amniotic

epithelium: Ultrastructure, composition and deposition. Journal of Cell Science. 79, 119-136.

Argiles, J. M., Carbo, N., & Lopez-Soriano, F. J. (1997). TNF and pregnancy: The paradigm

of a complex interaction. Cytokine & Growth Factor Reviews. 8, 181-188.

Arici, A., MacDonald, P. C., & Casey, M. L. (1995). Regulation of monocyte chemotactic

protein-1 gene expression in human endometrial cells in cultures. Molecular and Cellular

Endocrinology. 107, 189-197.

Arici, A., Oral, E., Bukulmez, O., Buradagunta, S., Bahtiyar, O., & Jones, E. E. (1997).

Monocyte chemotactic protein-1 expression in human preovulatory follicles and ovarian cells.

Journal of Reproductive Immunology. 32, 201-219.

Athayde, N., Romero, R., Maymon, E., Gomez, R., Pacora, P., Araneda, H., & Yoon, B.H.

(1999). A Role for the novel cytokine RANTES in pregnancy and parturition. American Journal

of Obstetrics and Gynecology. 181, 989-994.

Benirschke, K. (1960). Routes and types of infection in the fetus and the newborn. A.M.A.

Journal of Diseases of Children. 99, 714-721.

108

Bennett, P. R., Rose, M. P., Myatt, L., & Elder, M. G. (1987). Preterm labor: Stimulation of

arachidonic acid metabolism in human amnion cells by bacterial products. American Journal of

Obstetrics and Gynecology. 156, 649-655.

Betz, M., Fox, B.S. (1991). Prostaglandin E2 inhibits production of Th1 lymphokines but not of

Th2 lymphokines. Journal of Immunology. 146(1), 108-113.

Bloise, E., Torricelli, M., Novembri, R., Borges, L. E., Carrarelli, P., Reis, F. M., Petraglia,

F. (2010). Heat-killed Lactobacillus rhamnosus GG Modulates Urocortin and Cytokine Release

in Primary Trophoblast Cells. Placenta. 31, 867-872.

Blumenstein, M., Hansen, W. R., Deval, D., & Mitchell, M. D. (2000). Differential regulation

in human amnion epithelial and fibroblast cells of prostaglandin E(2) production and

prostaglandin H synthase-2 mRNA expression by dexamethasone but not tumour necrosis factor-

alpha. Placenta. 21, 210-217.

Bocking, A. D., Challis, J. R., & Korebrits, C. (1999). New approaches to the diagnosis of

preterm labor. American Journal of Obstetrics and Gynecology. 180, S247-8.

Bourne, G. (1962). The foetal membranes. A review of the anatomy of normal amnion and

chorion and some aspects of their function. Postgraduate Medical Journal. 38, 193-201.

Brown, N. L., Alvi, S. A., Elder, M. G., Bennett, P. R., & Sullivan, M. H. (1998). Interleukin-

1beta and bacterial endotoxin change the metabolism of prostaglandins E2 and F2 alpha in intact

term fetal membranes. Placenta. 19, 625-630.

109

Bry, K., & Hallman, M. (1992). Transforming growth factor-beta opposes the stimulatory

effects of interleukin-1 and tumor necrosis factor on amnion cell prostaglandin E2 production:

Implication for preterm labor. American Journal of Obstetrics and Gynecology. 167, 222-226.

Bry, K., Hallman, M., & Lappalainen, U. (1995). Production of granulocyte-macrophage

colony-stimulating factor (GM-CSF) by amnion cells and its regulation by proinflammatory

cytokines. Journal of the Society for Gynecologic Investigation. 2, 167-167.

Bry, K., Hallman, M., Teramo, K., Waffarn, F., & Lappalainen, U. (1997). Granulocyte-

macrophage colony-stimulating factor in amniotic fluid and in airway specimens of newborn

infants. Pediatric Research. 41, 105-109.

Bryant-Greenwood, G. D. (1998). The extracellular matrix of the human fetal membranes:

Structure and function. Placenta. 19, 1-11.

Burton, J.P., Reid, G. (2002). Evaluation of the bacterial vaginal flora of 20 postmenopausal

women by direct (Nugent score) and molecular (polymerase chain reaction and denaturing

gradient gel electrophoresis) techniques. Journal of infectious diseases. 186, 1770-1780.

Cadieux, P., Burton, J., Gardiner, G., Braunstein, I., Bruce, A. W., Kang, C. Y., & Reid, G.

(2002). Lactobacillus strains and vaginal ecology. Journal of the American Medical Association.

287, 1940-1941.

Calvin, S. E., & Oyen, M. L. (2007). Microstructure and mechanics of the chorioamnion

membrane with an emphasis on fracture properties. Annals of the New York Academy of

Sciences. 1101, 166-185.

110

Carey, J. C., Klebanoff, M. A., Hauth, J. C., Hillier, S. L., Thom, E. A., Ernest, J. M.,

Heine, R. P., Nugent, R. P., Fischer, M. L., Leveno, K. J., Wapner, R., & Varner, M. (2000).

Metronidazole to prevent preterm delivery in pregnant women with asymptomatic bacterial

vaginosis. national institute of child health and human development network of maternal-fetal

medicine units. New England Journal of Medicine. 342, 534-540.

Challis, J. R., Bloomfield, F. H., Bocking, A. D., Casciani, V., Chisaka, H., Connor, K.,

Dong, X., Gluckman, P., Harding, J. E., Johnstone, J., Li, W., Lye, S., Okamura, K., &

Premyslova, M. (2005). Fetal signals and parturition. Journal of Obstetrics and Gynaecology

Research. 31, 492-499.

Challis, J. R., Lye, S. J., & Gibb, W. (1997). Prostaglandins and parturition. Annals of the New

York Academy of Sciences. 828, 254-267.

Challis, J. R., Sloboda, D., Matthews, S. G., Holloway, A., Alfaidy, N., Patel, F. A., Whittle,

W., Fraser, M., Moss, T. J., & Newnham, J. (2001). The fetal placental hypothalamic-

pituitary-adrenal (HPA) axis, parturition and post natal health. Molecular and Cellular

Endocrinology. 185, 135-144.

Challis, J. R., Sloboda, D. M., Alfaidy, N., Lye, S. J., Gibb, W., Patel, F. A., Whittle, W. L.,

& Newnham, J. P. (2002). Prostaglandins and mechanisms of preterm birth. Reproduction

(Cambridge, England). 124, 1-17.

Challis, J. R. G., Matthews, S. G., Gibb, W., & Lye, S. J. (2000). Endocrine and paracrine

regulation of birth at term and preterm. Endocrine Reviews. 21, 514-550.

111

Cherouny, P. H., Pankuch, G. A., Romero, R., Botti, J. J., Kuhn, D. C., Demers, L. M., &

Appelbaum, P. C. (1993). Neutrophil attractant/activating peptide-1/interleukin-8: Association

with histologic chorioamnionitis, preterm delivery, and bioactive amniotic fluid leukoattractants.

American Journal of Obstetrics and Gynecology. 169, 1299-1303.

Cheung, P. Y., Walton, J. C., Tai, H. H., Riley, S. C., & Challis, J. R. (1990).

Immunocytochemical distribution and localization of 15-hydroxyprostaglandin dehydrogenase in

human fetal membranes, decidua, and placenta. American Journal of Obstetrics and Gynecology.

163, 1445-1449.

Chizzolini, C., Brembilla, N.C. (2009). Prostaglandin E2: igniting the fire. Immunology and

cell biology. 87(7), 510-511.

Chon, H., Choi, B., Jeong, G., Lee, E., & Lee, S. (2010). Suppression of proinflammatory

cytokine production by specific metabolites of lactobacillus plantarum 10hk2 via inhibiting NF-

kappaB and p38 MAPK expressions. Comparative Immunology, Microbiology and Infectious

Diseases. 33, e41-9.

Cong B., Zhang L., Gao L., Ni X. (2009). Reduced expression of CRH receptor type 1 in upper

segment human myometrium during labour. Reproductive Biology Endocrinology. 7, 43-51.

Coultrip, L. L., Lien, J. M., Gomez, R., Kapernick, P., Khoury, A., & Grossman, J. H.

(1994). The value of amniotic fluid interleukin-6 determination in patients with preterm labor

and intact membranes in the detection of microbial invasion of the amniotic cavity. American

Journal of Obstetrics and Gynecology. 171, 901-911.

112

Curbelo, V., Bejar, R., Benirschke, K., & Gluck, L. (1981). Premature labor. I. prostaglandin

precursors in human placental membranes. Obstetrics and Gynecology. 57, 473-478.

Deb, K., Chaturvedi, M. M., & Jaiswal, Y. K. (2004). Comprehending the role of LPS in

gram-negative bacterial vaginosis: Ogling into the causes of unfulfilled child-wish. Archives of

Gynecology and Obstetrics. 270, 133-146.

Denison, F. C., Kelly, R. W., Calder, A. A., & Riley, S. C. (1998). Cytokine secretion by

human fetal membranes, decidua and placenta at term. Human Reproduction (Oxford, England).

13, 3560-3565.

Denney, J. M., Culhane, J. F., & Goldenberg, R. L. (2008). Prevention of preterm birth.

Women's Health (London, England). 4, 625-638.

Dudley, D. J., Edwin, S. S., & Mitchell, M. D. (1996). Macrophage inflammatory protein-I

alpha regulates prostaglandin E2 and interleukin-6 production by human gestational tissues in

vitro. Journal of the Society for Gynecologic Investigation. 3, 12-16.

Dugoua, J. J., Machado, M., Zhu, X., Chen, X., Koren, G., & Einarson, T. R. (2009).

Probiotic safety in pregnancy: A systematic review and meta-analysis of randomized controlled

trials of Lactobacillus, Bifidobacterium, and Saccharomyces spp. Journal of Obstetrics and

Gynaecology Canada. 31, 542-552.

Dvorak, A. M., Morgan, E. S., Tzizik, D. M., & Weller, P. F. (1994). Prostaglandin

endoperoxide synthase (cyclooxygenase): Ultrastructural localization to nonmembrane-bound

113

cytoplasmic lipid bodies in human eosinophils and 3T3 fibroblasts. International Archives of

Allergy and Immunology. 105, 245-250.

Economopoulos, P., Sun, M., Purgina, B., & Gibb, W. (1996). Glucocorticoids stimulate

prostaglandin H synthase type-2 (PGHS-2) in the fibroblast cells in human amnion cultures.

Molecular and Cellular Endocrinology. 117, 141-147.

Embrey, M. (1971). PGE compounds for induction of labour and abortion. Annals of the New

York Academy of Sciences. 180, 518-523.

Eschenbach, D. A., Davick, P. R., Williams, B. L., Klebanoff, S. J., Young-Smith, K.,

Critchlow, C. M., & Holmes, K. K. (1989). Prevalence of hydrogen peroxide-producing

lactobacillus species in normal women and women with bacterial vaginosis. Journal of Clinical

Microbiology. 27, 251-256.

Esplin, M. S., Romero, R., Chaiworapongsa, T., Kim, Y. M., Edwin, S., Gomez, R., Mazor,

M., & Adashi, E. Y. (2005). Monocyte chemotactic protein-1 is increased in the amniotic fluid

of women who deliver preterm in the presence or absence of intra-amniotic infection. The

Journal of Maternal-Fetal & Neonatal Medicine. 17, 365-373.

Falagas, M. E., Betsi, G. I., & Athanasiou, S. (2007). Probiotics for the treatment of women

with bacterial vaginosis. Clinical Microbiology and Infection. 13, 657-664.

Floch, M. H. (2010). The effect of probiotics on host metabolism: The microbiota and

fermentation. Journal of Clinical Gastroenterology. 44 Suppl 1, S19-21.

114

Forsberg, L., Leeb, L., Thoren, S., Morgenstern, R., & Jakobsson, P. (2000). Human

glutathione dependent prostaglandin E synthase: Gene structure and regulation. FEBS Letters.

471, 78-82.

Forsum, U., Holst, E., Larsson, P. G., Vasquez, A., Jakobsson, T., & Mattsby-Baltzer, I.

(2005). Bacterial vaginosis--a microbiological and immunological enigma. APMIS : Acta

Pathologica, Microbiologica, Et Immunologica Scandinavica. 113, 81-90.

Fortunato, S. J., Menon, R. P., Swan, K. F., & Menon, R. (1996). Inflammatory cytokine

(interleukins 1, 6 and 8 and tumor necrosis factor-alpha) release from cultured human fetal

membranes in response to endotoxic lipopolysaccharide mirrors amniotic fluid concentrations.

American Journal of Obstetrics and Gynecology. 174, 1855-61; discussion 1861-2.

Fuchs, A. R., & Fuchs, F. (1981). Ethanol for prevention of preterm birth. Seminars in

Perinatology. 5, 236-251.

Fuentes, A., Spaziani, E. P., & O'Brien, W. F. (1996). The expression of cyclooxygenase-2

(COX-2) in amnion and decidua following spontaneous labor. Prostaglandins. 52, 261-267.

Furuta, I., Yamada, H., Sagawa, T., & Fujimoto, S. (2000). Effects of inflammatory cytokines

on prostaglandin E(2) production from human amnion cells cultured in serum-free condition.

Gynecologic and Obstetric Investigation. 49, 93-97.

Garcia-Lloret, M. I., Morrish, D. W., Wegmann, T. G., Honore, L., Turner, A. R., &

Guilbert, L. J. (1994). Demonstration of functional cytokine-placental interactions: CSF-1 and

115

GM-CSF stimulate human cytotrophoblast differentiation and peptide hormone secretion.

Experimental Cell Research. 214, 46-54.

Gardiner, G. E., Heinemann, C., Bruce, A. W., Beuerman, D., & Reid, G. (2002).

Persistence of Lactobacillus fermentum RC-14 and Lactobacillus rhamnosus GR-1 but not L.

rhamnosus GG in the human vagina as demonstrated by randomly amplified polymorphic DNA.

Clinical and Diagnostic Laboratory Immunology. 9, 92-96.

Gasson, J. C. (1991). Molecular physiology of granulocyte-macrophage colony-stimulating

factor. Blood. 77, 1131-1145.

Gibb, W., & Lavoie, J. C. (1990). Effects of glucocorticoids on prostaglandin formation by

human amnion. Canadian Journal of Physiology and Pharmacology. 68, 671-676.

Gibb, W., & Sun, M. (1996). Localization of prostaglandin H synthase type 2 protein and

mRNA in term human fetal membranes and decidua. Journal of Endocrinology. 150, 497-503.

Giles, W., & Bisits, A. (2007). Preterm labour. the present and future of tocolysis. Best Practice

& Research.Clinical Obstetrics & Gynaecology. 21, 857-868.

Goldenberg, R. L., Culhane, J. F., Iams, J. D., & Romero, R. (2008). Epidemiology and

causes of preterm birth. Lancet. 371, 75-84.

Goldenberg, R. L., Hauth, J. C., & Andrews, W. W. (2000). Intrauterine infection and preterm

delivery. New England Journal of Medicine. 342, 1500-1507.

116

Goldenberg, R. L., Iams, J. D., Mercer, B. M., Meis, P., Moawad, A., Das, A., Copper, R.,

Johnson, F., & National Institute of Child Health and Human Development Maternal-Fetal

Medicine Units Network. (2003). What we have learned about the predictors of preterm birth.

Seminars in Perinatology. 27, 185-193.

Goldenberg, R. L., & Rouse, D. J. (1998). Prevention of premature birth. New England Journal

of Medicine. 339, 313-320.

Goodwin, J.S., Ceuppens, J. (1983). Regulation of the immune response by prostaglandins.

Journal of Clinical Immunology. 3(4), 295-315.

Hagberg, L.A., Bruce, A.W., Reid, G., Svanborg Eden, C., Lincoln, K., Lidin-Janson G.

(1989). Colonization of the urinary tract with live bacteria from the normal fecal and urethral

flora in patients with recurrent symptomatic urinary tract infections. University of Chicago Press.

194-197.

Han, R., Tsui, S., & Smith, T. J. (2002). Up-regulation of prostaglandin E2 synthesis by

interleukin-1beta in human orbital fibroblasts involves coordinate induction of prostaglandin-

endoperoxide H synthase-2 and glutathione-dependent prostaglandin E2 synthase expression.

Journal of Biological Chemistry. 277, 16355-16364.

Hanna, N., Bonifacio, L., Reddy, P., Hanna, I., Weinberger, B., Murphy, S., Laskin, D., &

Sharma, S. (2004). IFN-gamma-mediated inhibition of COX-2 expression in the placenta from

term and preterm labor pregnancies. American Journal of Reproductive Immunology (New

York, N.Y.: 1989). 51, 311-318.

117

Harper, M., Zheng, S.L., Thom, E., Klebanoff, M.A., Thorp, J. Jr., Sorokin, Y., Varner,

M.W., Iams, J.D., Dinsmoor, M., Mercer, B.M., Rouse, D.J., Ramin, S.M., Anderson, G.D.

(2011) Cytokine gene polymorphisms and length of gestation. Obstetrics and Gynecology.

117(1), 125-130.

Harris, S.G., Padilla, J., Koumas, L., Ray, D., Phipps, R.P. (2002). Prostaglandins as

modulators of immunity. Trends in Immunology. 23(3), 144-150.

Hawes, S. E., Hillier, S. L., Benedetti, J., Stevens, C. E., Koutsky, L. A., Wolner-Hanssen,

P., & Holmes, K. K. (1996). Hydrogen peroxide-producing lactobacilli and acquisition of

vaginal infections. Journal of Infectious Diseases. 174, 1058-1063.

Hillhouse, E.W., Grammatopoulos, D.K. (2002). Role of stress peptides during human

pregnancy and labour. Reproduction. 124, 323-329.

Hillier, S. L., Krohn, M. A., Klebanoff, S. J., & Eschenbach, D. A. (1992). The relationship of

hydrogen peroxide-producing lactobacilli to bacterial vaginosis and genital microflora in

pregnant women. Obstetrics and Gynecology. 79, 369-373.

Hillier, S. L., Krohn, M. A., Rabe, L. K., Klebanoff, S. J., & Eschenbach, D. A. (1993). The

normal vaginal flora, H2O2-producing lactobacilli, and bacterial vaginosis in pregnant women.

Clinical Infectious Diseases. 16, S273-81.

Hillier, S. L., Nugent, R. P., Eschenbach, D. A., Krohn, M. A., Gibbs, R. S., Martin, D. H.,

Cotch, M. F., Edelman, R., Pastorek, J. G.,2nd, & Rao, A. V. (1995). Association between

118

bacterial vaginosis and preterm delivery of a low-birth-weight infant. the vaginal infections and

prematurity study group. New England Journal of Medicine. 333, 1737-1742.

Hillier, S. L., Witkin, S. S., Krohn, M. A., Watts, D. H., Kiviat, N. B., & Eschenbach, D. A.

(1993). The relationship of amniotic fluid cytokines and preterm delivery, amniotic fluid

infection, histologic chorioamnionitis, and chorioamnion infection. Obstetrics and Gynecology.

81, 941-948.

Hirst, J. J., Teixeira, F. J., Zakar, T., & Olson, D. M. (1995). Prostaglandin endoperoxide-H

synthase-1 and -2 messenger ribonucleic acid levels in human amnion with spontaneous labor

onset. Journal of Clinical Endocrinology and Metabolism. 80, 517-523.

Holloway, A.C., Gyomorey S., Challis J.R. (2000). Effects of labor on pituitary expression of

proopiomelanocortin, prohormone convertase (PC)-1, PC-2, and glucocorticoid receptor mRNA

in fetal sheep. Endocrine. 13, 17-23.

Hornung, D., Ryan, I. P., Chao, V. A., Vigne, J. L., Schriock, E. D., & Taylor, R. N. (1997).

Immunolocalization and regulation of the chemokine RANTES in human endometrial and

endometriosis tissues and cells. Journal of Clinical Endocrinology and Metabolism. 82, 1621-

1628.

Howard, J. C., Heinemann, C., Thatcher, B. J., Martin, B., Gan, B. S., & Reid, G. (2000).

Identification of collagen-binding proteins in lactobacillus spp. with surface-enhanced laser

desorption/ionization-time of flight ProteinChip technology. Applied and Environmental

Microbiology. 66, 4396-4400.

119

Hsu, C. D., Meaddough, E., Aversa, K., & Copel, J. A. (1998). The role of amniotic fluid L-

selectin, GRO-alpha, and interleukin-8 in the pathogenesis of intraamniotic infection. American

Journal of Obstetrics and Gynecology. 178, 428-432.

Hummelen, R., Fernandes, A.D., Macklaim, J.M., Dickson, R.J., Chanqalucha, J., Gloor,

G.B., Reid, G. (2010). Deep sequencing of vaginal microbiota of women with HIV. PLoS One.

5, 1-9.

Imseis, H. M., Greig, P. C., Livengood, C. H.,3rd, Shunior, E., Durda, P., & Erikson, M.

(1997). Characterization of the inflammatory cytokines in the vagina during pregnancy and labor

and with bacterial vaginosis. Journal of the Society for Gynecologic Investigation. 4, 90-94.

Inoue, H., Kosaka, T., Miyata, A., Hara, S., Yokoyama, C., Nanayama, T., & Tanabe, T.

(1995). Structure and expression of the human prostaglandin endoperoxide synthase 2 gene.

Advances in Prostaglandin, Thromboxane, and Leukotriene Research. 23, 109-111.

Jacobsson, B., Mattsby-Baltzer, I., & Hagberg, H. (2005). Interleukin-6 and interleukin-8 in

cervical and amniotic fluid: Relationship to microbial invasion of the chorioamniotic

membranes. Journal of Obstetrics and Gynaecology. 112, 719-724.

Jakobsson, P. J., Thoren, S., Morgenstern, R., & Samuelsson, B. (1999). Identification of

human prostaglandin E synthase: A microsomal, glutathione-dependent, inducible enzyme,

constituting a potential novel drug target. Proceedings of the National Academy of Sciences of

the United States of America. 96, 7220-7225.

120

Jijon, H., Backer, J., Diaz, H., Yeung, H., Thiel, D., McKaigney, C., De Simone, C., &

Madsen, K. (2004). DNA from probiotic bacteria modulates murine and human epithelial and

immune function. Gastroenterology. 126, 1358-1373.

Johnson, R. F., Mitchell, C. M., Giles, W. B., Bisits, A., & Zakar, T. (2006). Mechanisms

regulating prostaglandin H2 synthase-2 mRNA level in the amnion and chorion during

pregnancy. Journal of Endocrinology. 188, 603-610.

Kallapur, S. G., Moss, T. J., Auten, R. L.,Jr, Nitsos, I., Pillow, J. J., Kramer, B. W., Maeda,

D. Y., Newnham, J. P., Ikegami, M., & Jobe, A. H. (2009). IL-8 signaling does not mediate

intra-amniotic LPS-induced inflammation and maturation in preterm fetal lamb lung. American

Journal of Physiology.Lung Cellular and Molecular Physiology. 297, L512-9.

Kamel, R. M. (2010). The onset of human parturition. Archives of Gynecology and Obstetrics.

281, 975-982.

Karalis, K., Goodwin, G., & Majzoub, J. A. (1996). Cortisol blockade of progesterone: A

possible molecular mechanism involved in the initiation of human labor. Nature Medicine. 2,

556-560.

Katamura, K., Shintake, N., Yamauchi, Y., Fukui, T., Ohshima, Y., Mayumi, M., Furusho,

K. (1995). Prostaglandin E2 at priming of naïve CD4+ cells inhibits acquisition of an ability to

produce IFN-gamma and IL-2, but not IL-4 and IL-5. Journal of Immunology. 155, 4604-4612.

Katz, V. L., & Farmer, R. M. (1999). Controversies in tocolytic therapy. Clinical Obstetrics

and Gynecology. 42, 802-819.

121

Keelan, J. A., Blumenstein, M., Helliwell, R. J., Sato, T. A., Marvin, K. W., & Mitchell, M.

D. (2003). Cytokines, prostaglandins and parturition--a review. Placenta. 24 Suppl A, S33-46.

Kendal-Wright, C. E. (2007). Stretching, mechanotransduction, and proinflammatory cytokines

in the fetal membranes. Reproductive Sciences. 14, 35-41.

Kent, A. S., Sullivan, M. H., & Elder, M. G. (1994). Transfer of cytokines through human fetal

membranes. Journal of Reproduction and Fertility. 100, 81-84.

Kent, A. S., Sullivan, M. H., Sun, M. Y., Zosmer, A., & Elder, M. G. (1993). Effects of

interleukin-6 and tumor necrosis factor-alpha on prostaglandin production by cultured human

fetal membranes. Prostaglandins. 46, 351-359.

Kent, A. S., Sun, M. Y., Sullivan, M. H., & Elder, M. G. (1993). The effects of interleukins 1

alpha and 1 beta on prostaglandin production by cultured human fetal membranes.

Prostaglandins. 46, 51-59.

Kenyon, S. L., Taylor, D. J., Tarnow-Mordi, W., & ORACLE Collaborative Group. (2001).

Broad-spectrum antibiotics for spontaneous preterm labour: The ORACLE II randomised trial.

ORACLE collaborative group. Lancet. 357, 989-994.

Kim, S. O., Sheikh, H. I., Ha, S. D., Martins, A., & Reid, G. (2006). G-CSF-mediated

inhibition of JNK is a key mechanism for Lactobacillus rhamnosus-induced suppression of TNF

production in macrophages. Cellular Microbiology. 8, 1958-1971.

King, A. E., Kelly, R. W., Sallenave, J. M., Bocking, A. D., & Challis, J. R. (2007). Innate

immune defences in the human uterus during pregnancy. Placenta. 28, 1099-1106.

122

Kitterman, J. A. (1987). Arachidonic acid metabolites and control of breathing in the fetus and

newborn. Seminars in Perinatology. 11, 43-52.

Kligler, E., Hanaway, P., Cohrssen, A. (2007). Probiotics in children. Pediatric Clinics of

North America. 54, 949-967.

Kobayashi, K., Inai, T., Shibata, Y., & Yasui, M. (2009). Dynamic changes in amniotic tight

junctions during pregnancy. Placenta. 30, 840-847.

Kobayashi, K., Kadohira, I., Tanaka, M., Yoshimura, Y., Ikeda, K., & Yasui, M. (2010a).

Expression and distribution of tight junction proteins in human amnion during late pregnancy.

Placenta. 31, 158-162.

Kobayashi, K., Miwa, H., Yasui, M. (2010b). Inflammatory mediators weaken the amniotic

membrane barrier through disruption of tight junctions. Journal of Physiology. 588, 4859-4869.

Korhonen, R., Kosonen, O., Korpela, R., & Moilanen, E. (2004). The expression of COX2

protein induced by Lactobacillus rhamnosus GG, endotoxin and lipoteichoic acid in T84

epithelial cells. Letters in Applied Microbiology. 39, 19-24.

Kotzamanidis, C., Kourelis, A., Litopoulou-Tzanetaki E., Tzanetakis, N., Yiangou, M.

(2010). Evaluation of adhesion capacity, cell surface traits and immunomodulatory activity of

presumptive probiotic Lactobacillus strains. International Journal of Food Microbiology. 140,

154-163.

Kudo, I., & Murakami, M. (1999). Diverse functional coupling of prostanoid biosynthetic

enzymes in various cell types. Advances in Experimental Medicine and Biology. 469, 29-35.

123

Lamont, R., Sobel, J., Akins, R., Hassan, S., Chaiworapongsa, T., Kusanovic, J., &

Romero, R. (2011). The vaginal microbiome: New information about genital tract flora using

molecular based techniques. BJOG. 118, 533-549

Lara-Villoslada, F., Sierra, S., Boza, J., Xaus, J., & Olivares, M. (2007). Beneficial effects of

consumption of a dairy product containing two probiotic strains, Lactobacillus coryniformis

CECT5711 and Lactobacillus gasseri CECT5714 in healthy children. Nutricion Hospitalaria. 22,

496-502.

Lee do, K., Park, S.Y., Jang, S., Baek, E.H., Kim, M.J., Huh, S.M., Choir, K.S., Chung,

M.J., Kim, J.E., Lee, K.O., Ha. N.J. (2011). The combination of mixed lactic acid bacteria and

dietary fiber lowers serum cholesterol levels and fecal harmful enzyme activities in rats.

Archives of Pharmacology Research. 34, 23-29.

Leitich, H., Brunbauer, M., Bodner-Adler, B., Kaider, A., Egarter, C., & Husslein, P.

(2003). Antibiotic treatment of bacterial vaginosis in pregnancy: A meta-analysis. American

Journal of Obstetrics and Gynecology. 188, 752-758.

Li, A., Dubey, S., Varney, M. L., Dave, B. J., & Singh, R. K. (2003). IL-8 directly enhanced

endothelial cell survival, proliferation, and matrix metalloproteinases production and regulated

angiogenesis. Journal of Immunology (Baltimore, Md.: 1950). 170, 3369-3376.

Livengood, C. H. (2009). Bacterial vaginosis: An overview for 2009. Reviews in Obstetrics and

Gynecology. 2, 28-37.

124

Locksmith, G., & Duff, P. (2001). Infection, antibiotics, and preterm delivery. Seminars in

Perinatology. 25, 295-309.

MacPhee, R. A., Hummelen, R., Bisanz, J. E., Miller, W. L., & Reid, G. (2010). Probiotic

strategies for the treatment and prevention of bacterial vaginosis. Expert Opinion on

Pharmacotherapy. 11, 2985-2995.

Madsen, K., Cornish, A., Soper, P., McKaigney, C., Jijon, H., Yachimec, C., Doyle, J.,

Jewell, L., & De Simone, C. (2001). Probiotic bacteria enhance murine and human intestinal

epithelial barrier function. Gastroenterology. 121, 580-591.

Mancini, J. A., Blood, K., Guay, J., Gordon, R., Claveau, D., Chan, C. C., & Riendeau, D.

(2001). Cloning, expression, and up-regulation of inducible rat prostaglandin e synthase during

lipopolysaccharide-induced pyresis and adjuvant-induced arthritis. Journal of Biological

Chemistry. 276, 4469-4475.

Masferrer, J. L., Seibert, K., Zweifel, B., & Needleman, P. (1992). Endogenous

glucocorticoids regulate an inducible cyclooxygenase enzyme. Proceedings of the National

Academy of Sciences of the United States of America. 89, 3917-3921.

McDonald, H. M., Brocklehurst, P., & Gordon, A. (2007). Antibiotics for treating bacterial

vaginosis in pregnancy. Cochrane Database of Systematic Reviews (Online). (1), CD000262.

Meadows, J. W., Eis, A. L., Brockman, D. E., & Myatt, L. (2003). Expression and localization

of prostaglandin E synthase isoforms in human fetal membranes in term and preterm labor.

Journal of Clinical Endocrinology and Metabolism. 88, 433-439.

125

Mehta, A., Brewington, R., Chatterji, M., Zoubine, M., Kinasewitz, G. T., Peer, G. T.,

Chang, A. C., Taylor, F. B.,Jr, & Shnyra, A. (2004). Infection-induced modulation of m1 and

m2 phenotypes in circulating monocytes: Role in immune monitoring and early prognosis of

sepsis. Shock (Augusta, Ga.). 22, 423-430.

Meis, P. J., Klebanoff, M., Thom, E., Dombrowski, M. P., Sibai, B., Moawad, A. H., Spong,

C. Y., Hauth, J. C., Miodovnik, M., Varner, M. W., Leveno, K. J., Caritis, S. N., Iams, J. D.,

Wapner, R. J., Conway, D., O'Sullivan, M. J., Carpenter, M., Mercer, B., Ramin, S. M.,

Thorp, J. M., Peaceman, A. M., Gabbe, S., & National Institute of Child Health and

Human Development Maternal-Fetal Medicine Units Network. (2003). Prevention of

recurrent preterm delivery by 17 alpha-hydroxyprogesterone caproate. New England Journal of

Medicine. 348, 2379-2385.

Meis, P. J., Michielutte, R., Peters, T. J., Wells, H. B., Sands, R. E., Coles, E. C., & Johns,

K. A. (1995). Factors associated with preterm birth in cardiff, wales. II. indicated and

spontaneous preterm birth. American Journal of Obstetrics and Gynecology. 173, 597-602.

Menard, S., Candalh, C., Bambou, J. C., Terpend, K., Cerf-Bensussan, N., & Heyman, M.

(2004). Lactic acid bacteria secrete metabolites retaining anti-inflammatory properties after

intestinal transport. Gut. 53, 821-828.

Menon, R., & Fortunato, S. J. (2004). Fetal membrane inflammatory cytokines: A switching

mechanism between the preterm premature rupture of the membranes and preterm labor

pathways. Journal of Perinatal Medicine. 32, 391-399.

126

Menon, R., Swan, K. F., Lyden, T. W., Rote, N. S., & Fortunato, S. J. (1995). Expression of

inflammatory cytokines (interleukin-1β and interleukin-6) in amniochorionic membranes.

American Journal of Obstetrics and Gynecology. 172, 493-500.

Merlino, A., Welsh, T., Erdonmez, T., Madsen, G., Zakar, T., Smith, R., Mercer, B., &

Mesiano, S. (2009). Nuclear progesterone receptor expression in the human fetal membranes

and decidua at term before and after labor. Reproductive Sciences. 16, 357-363.

Merlino, A. A., Welsh, T. N., Tan, H., Yi, L. J., Cannon, V., Mercer, B. M., & Mesiano, S.

(2007). Nuclear progesterone receptors in the human pregnancy myometrium: Evidence that

parturition involves functional progesterone withdrawal mediated by increased expression of

progesterone receptor-A. Journal of Clinical Endocrinology and Metabolism. 92, 1927-1933.

Mesiano, S., Chan, E. C., Fitter, J. T., Kwek, K., Yeo, G., & Smith, R. (2002). Progesterone

withdrawal and estrogen activation in human parturition are coordinated by progesterone

receptor A expression in the myometrium. Journal of Clinical Endocrinology and Metabolism.

87, 2924-2930.

Metcalf, D., & Moore, J. G. (1988). Divergent disease patterns in granulocyte-macrophage

colony-stimulating factor transgenic mice associated with different transgene insertion sites.

Proceedings of the National Academy of Sciences of the United States of America. 85, 7767-

7771.

Mijovic, J. E., Zakar, T., Angelova, J., & Olson, D. M. (1999). Prostaglandin endoperoxide H

synthase mRNA expression in the human amnion and decidua during pregnancy and in the

amnion at preterm labour. Molecular Human Reproduction. 5, 182-187.

127

Mitchell, B. F., & Wong, S. (1995). Metabolism of oxytocin in human decidua, chorion, and

placenta. Journal of Clinical Endocrinology and Metabolism. 80, 2729-2733.

Mitchell, M. D., Dudley, D. J., Edwin, S. S., & Schiller, S. L. (1991). Interleukin-6 stimulates

prostaglandin production by human amnion and decidual cells. European Journal of

Pharmacology. 192, 189-191.

Mitchell, M. D., Edwin, S. S., Lundin-Schiller, S., Silver, R. M., Smotkin, D., & Trautman,

M. S. (1993). Mechanism of interleukin-1 beta stimulation of human amnion prostaglandin

biosynthesis: Mediation via a novel inducible cyclooxygenase. Placenta. 14, 615-625.

Mitchell, M. D., Romero, R. J., Edwin, S. S., & Trautman, M. S. (1995). Prostaglandins and

parturition. Reproduction, Fertility, and Development. 7, 623-632.

Mitchell, M. D., Simpson, K. L., & Keelan, J. A. (2004). Paradoxical proinflammatory actions

of interleukin-10 in human amnion: Potential roles in term and preterm labour. Journal of

Clinical Endocrinology and Metabolism. 89, 4149-4152.

Mohan, A. R., Sooranna, S. R., Lindstrom, T. M., Johnson, M. R., & Bennett, P. R. (2007).

The effect of mechanical stretch on cyclooxygenase type 2 expression and activator protein-1

and nuclear factor-kappaB activity in human amnion cells. Endocrinology. 148, 1850-1857.

Molnar, M., Rigo, J.,Jr, Romero, R., & Hertelendy, F. (1999). Oxytocin activates mitogen-

activated protein kinase and up-regulates cyclooxygenase-2 and prostaglandin production in

human myometrial cells. American Journal of Obstetrics and Gynecology. 181, 42-49.

128

Moorthy, G., Murali, M. R., & Devaraj, S. N. (2009). Lactobacilli facilitate maintenance of

intestinal membrane integrity during Shigella dysenteriae 1 infection in rats. Nutrition. 25, 350-

358.

Mosser, D.M., Edwards, J.P. (2008). Exploring the full spectrum of macrophage activation.

Nature Reviews: Immunology. 8, 958-969.

Murakami, M., Nakashima, K., Kamei, D., Masuda, S., Ishikawa, Y., Ishii, T., Ohmiya, Y.,

Watanabe, K., & Kudo, I. (2003). Cellular prostaglandin E2 production by membrane-bound

prostaglandin E synthase-2 via both cyclooxygenases-1 and -2. Journal of Biological Chemistry.

278, 37937-37947.

Murphy, S. P., Tayade, C., Ashkar, A. A., Hatta, K., Zhang, J., & Croy, B. A. (2009).

Interferon gamma in successful pregnancies. Biology of Reproduction. 80, 848-859.

Myatt, L., & Sun, K. (2010). Role of fetal membranes in signaling of fetal maturation and

parturition. International Journal of Developmental Biology. 54, 545-553.

Nataraj, C., Thomas, D. W., Tilley, S. L., Nguyen, M. T., Mannon, R., Koller, B.H.,

Coffman, T. M. (2001). Receptors for prostaglandin E(2) that regulate cellular immune

responses in the mouse. Journal of Clinical Investigation. 108(8), 1229-1235.

Norwitz, E. R., Robinson, J. N., & Challis, J. R. (1999). The control of labor. New England

Journal of Medicine. 341, 660-666.

129

Nugent, R. P., Krohn, M. A., & Hillier, S. L. (1991). Reliability of diagnosing bacterial

vaginosis is improved by a standardized method of gram stain interpretation. Journal of Clinical

Microbiology. 29, 297-301.

Ohara, T., Yoshino, K., Kitajima, M. (2010). Possibility of preventing colorectal

carcinogenesis with probiotics. Hepatogastroenterology. 57, 1411-1415.

Ojetti, V., Gigante, G., Gabrielli, M., Ainora, M.E., Mannocci, A., Lauritano, E.C.,

Gasbarrini, G., Gasbarrini, A. (2010). The effect of oral supplementation with Lactobacillus

reuteri or tilactase in lactose intolerant patients: randomized trial. European review for medical

and pharmacological sciences. 14, 163-170.

Okun, N., Gronau, K. A., & Hannah, M. E. (2005). Antibiotics for bacterial vaginosis or

trichomonas vaginalis in pregnancy: A systematic review. Obstetrics and Gynecology. 105, 857-

868.

Olson, D. M. (2003). The role of prostaglandins in the initiation of parturition. Best Practice &

Research. Clinical Obstetrics & Gynaecology. 17, 717-730.

Olson, D. M., Skinner, K., & Challis, J. R. (1983). Prostaglandin output in relation to

parturition by cells dispersed from human intrauterine tissues. Journal of Clinical Endocrinology

and Metabolism. 57, 694-699.

Othman, M., Neilson, J. P., & Alfirevic, Z. (2007). Probiotics for preventing preterm labour.

Cochrane Database of Systematic Reviews (Online). 1, CD005941.

130

Ou, C.W., Chen, Z.Q., Qi, S., Lye, S.J. (1998) Increased expression of the rate myometrial

oxytocin receptor messenger ribonucleic acid during labor requires both mechanical and

hormonal signals. Biology of Reproduction. 59(5), 1055-1061.

Parry, S., & Strauss, J. F.,3rd. (1998). Premature rupture of the fetal membranes. New England

Journal of Medicine. 338, 663-670.

Pavlova, S. I., Kilic, A. O., Kilic, S. S., So, J. S., Nader-Macias, M. E., Simoes, J. A., & Tao,

L. (2002). Genetic diversity of vaginal lactobacilli from women in different countries based on

16S rRNA gene sequences. Journal of Applied Microbiology. 92, 451-459.

Petersen, A. M., & Pedersen, B. K. (2006). The role of IL-6 in mediating the anti-inflammatory

effects of exercise. Journal of Physiology and Pharmacology. 57, 43-51.

Pollard, J. K., & Mitchell, M. D. (1993). Tumor necrosis factor alpha stimulates amnion

prostaglandin biosynthesis primarily via an action on fatty acid cyclooxygenase. Prostaglandins.

46, 499-510.

Pomini, F., Caruso, A., & Challis, J. R. (1999). Interleukin-10 modifies the effects of

interleukin-1beta and tumor necrosis factor-alpha on the activity and expression of prostaglandin

H synthase-2 and the NAD+-dependent 15-hydroxyprostaglandin dehydrogenase in cultured

term human villous trophoblast and chorion trophoblast cells. Journal of Clinical Endocrinology

and Metabolism. 84, 4645-4651.

131

Potestio, F. A., Zakar, T., & Olson, D. M. (1988). Glucocorticoids stimulate prostaglandin

synthesis in human amnion cells by a receptor-mediated mechanism. Journal of Clinical

Endocrinology and Metabolism. 67, 1205-1210.

Premyslova, M., Li, W., Alfaidy, N., Bocking, A. D., Campbell, K., Gibb, W., & Challis, J.

R. (2003). Differential expression and regulation of microsomal prostaglandin E(2) synthase in

human fetal membranes and placenta with infection and in cultured trophoblast cells. Journal of

Clinical Endocrinology and Metabolism. 88, 6040-6047.

Rachmilewitz, D., Katakura, K., Karmeli, F., Hayashi, T., Reinus, C., Rudensky, B., Akira,

S., Takeda, K., Lee, J., Takabayashi, K., & Raz, E. (2004). Toll-like receptor 9 signaling

mediates the anti-inflammatory effects of probiotics in murine experimental colitis.

Gastroenterology. 126, 520-528.

Rankin, J. G. (1976). A role for prostaglandins in the regulation of the placental blood flows.

Prostaglandins. 11, 343-353.

Reid, G. (1999). The scientific basis for probiotic strains of Lactobacillus. Applied and

Environmental Microbiology. 65, 3763-3766.

Reid, G. (2008). Probiotic lactobacilli for urogenital health in women. Journal of Clinical

Gastroenterology. 42, S234-S236.

Reid, G., Beuerman, D., Heinemann, C., & Bruce, A. W. (2001a). Probiotic Lactobacillus

dose required to restore and maintain a normal vaginal flora. FEMS Immunology and Medical

Microbiology. 32, 37-41.

132

Reid, G., & Bocking, A. (2003). The potential for probiotics to prevent bacterial vaginosis and

preterm labor. American Journal of Obstetrics and Gynecology. 189, 1202-1208.

Reid, G., Bruce, A.W., Fraser, N., Heinemann, C., Owen, J., Henning, B. (2001b). Oral

probiotics can resolve urogenital infections. FEMS Immunology and Medical Microbiology. 30,

49-52.

Reid, G., Charbonneau, D., Erb, J., Kochanowski, B., Beuerman, D., Poehner, R., & Bruce,

A. W. (2003). Oral use of Lactobacillus rhamnosus GR-1 and L. fermentum RC-14 significantly

alters vaginal flora: Randomized, placebo-controlled trial in 64 healthy women. FEMS

Immunology and Medical Microbiology. 35, 131-134.

Reid, G., Cook, R. L., & Bruce, A. W. (1987). Examination of strains of lactobacilli for

properties that may influence bacterial interference in the urinary tract. Journal of Urology. 138,

330-335.

Reid, G., Millsap, K., & Bruce, A. W. (1994). Implantation of Lactobacillus casei var

rhamnosus into vagina. Lancet. 344, 1229.

Reid, G., Tieszer, C., & Lam, D. (1995). Influence of lactobacilli on the adhesion of

staphylococcus aureus and candida albicans to fibers and epithelial cells. Journal of Industrial

Microbiology. 15, 248-253.

Reid, G., Younes, J. A., Van der Mei, H. C., Gloor, G. B., Knight, R., & Busscher, H. J.

(2011). Microbiota restoration: Natural and supplemented recovery of human microbial

communities. Nature Reviews. Microbiology. 9, 27-38.

133

Reisenberger, K., Egarter, C., Knofler, M., Schiebel, I., Gregor, H., Hirschl, A. M., Heinze,

G., & Husslein, P. (1998). Cytokine and prostaglandin production by amnion cells in response

to the addition of different bacteria. American Journal of Obstetrics and Gynecology. 178, 50-53.

Riemer, R. K., & Heymann, M. A. (1998). Regulation of uterine smooth muscle function

during gestation. Pediatric Research. 44, 615-627.

Robertson, S. A., Mayrhofer, G., & Seamark, R. F. (1992). Uterine epithelial cells synthesize

granulocyte-macrophage colony-stimulating factor and interleukin-6 in pregnant and non-

pregnant mice. Biology of Reproduction. 46, 1069-1079.

Robinson, J. N., Regan, J. A., & Norwitz, E. R. (2001). The epidemiology of preterm labor.

Seminars in Perinatology. 25, 204-214.

Romero, R., Brody, D. T., Oyaizu, E., Mazor, M., Wu, Y. K., Hobbins, J. C., & Durum, S.

K. (1989a). Infection and labor. III. interleukin-1: A signal for the onset of parturition. American

Journal of Obstetrics and Gynecology. 160, 1117-1123.

Romero, R., Ceska, M., Avila, C., Mazor, M., Behnke, E., & Lindley, I. (1991). Neutrophil

attractant/activating peptide-1/interleukin-8 in term and preterm parturition. American Journal of

Obstetrics and Gynecology. 165, 813-820.

Romero, R., Chaiworapongsa, T., Kuivaniemi, H., & Tromp, G. (2004). Bacterial vaginosis,

the inflammatory response and the risk of preterm birth: A role for genetic epidemiology in the

prevention of preterm birth. American Journal of Obstetrics and Gynecology. 190, 1509-1519.

134

Romero, R., Durum, S., Dinarello, C. A., Oyarzun, E., Hobbins, J. C., & Mitchell, M. D.

(1989b). Interleukin-1 stimulates prostaglandin biosynthesis by human amnion. Prostaglandins.

37, 13-22.

Romero, R., Espinoza, J., Chaiworapongsa, T., & Kalache, K. (2002). Infection and

prematurity and the role of preventive strategies. Seminars in Neonatology. 7, 259-274.

Romero, R., Hobbins, J. C., & Mitchell, M. D. (1988a). Endotoxin stimulates prostaglandin E2

production by human amnion. Obstetrics and Gynecology. 71, 227-228.

Romero, R., Manogue, K. R., Mitchell, M. D., Wu, Y. K., Oyarzun, E., Hobbins, J. C., &

Cerami, A. (1989c). Infection and labor. IV. cachectin-tumor necrosis factor in the amniotic

fluid of women with intraamniotic infection and preterm labor. American Journal of Obstetrics

and Gynecology. 161, 336-341.

Romero, R., Roslansky, P., Oyarzun, E., Wan, M., Emamian, M., Novitsky, T. J., Gould,

M. J., & Hobbins, J. C. (1988b). Labor and infection. II. bacterial endotoxin in amniotic fluid

and its relationship to the onset of preterm labor. American Journal of Obstetrics and

Gynecology. 158, 1044-1049.

Romero, R., Sirtori, M., Oyarzun, E., Avila, C., Mazor, M., Callahan, R., Sabo, V.,

Athanassiadis, A. P., & Hobbins, J. C. (1989d). Infection and labor. V. prevalence,

microbiology, and clinical significance of intraamniotic infection in women with preterm labor

and intact membranes. American Journal of Obstetrics and Gynecology. 161, 817-824.

135

Rose, M. P., Myatt, L., & Elder, M. G. (1990). Pathways of arachidonic acid metabolism in

human amnion cells at term. Prostaglandins, Leukotrienes, and Essential Fatty Acids. 39, 303-

309.

Saji, F., Samejima, Y., Kamiura, S., Sawai, K., Shimoya, K., & Kimura, T. (2000). Cytokine

production in chorioamnionitis. Journal of Reproductive Immunology. 47, 185-196.

Sangha, R. K., Walton, J. C., Ensor, C. M., Tai, H. H., & Challis, J. R. (1994).

Immunohistochemical localization, messenger ribonucleic acid abundance, and activity of 15-

hydroxyprostaglandin dehydrogenase in placenta and fetal membranes during term and preterm

labor. Journal of Clinical Endocrinology and Metabolism. 78, 982-989.

Sastry, B. V., Hemontolor, M. E., Chance, M. B., & Johnson, R. F. (1997). Dual messenger

function for prostaglandin E2 (PGE2) in human placenta. Cellular and Molecular Biology. 43,

417-424.

Saunders, S., Bocking, A., Challis, J., & Reid, G. (2007). Effect of Lactobacillus challenge on

gardnerella vaginalis biofilms. Colloids and Surfaces. B, Biointerfaces. 55, 138-142.

Saunders-Kirkwood, K., Cates, J. A., & Roslyn, J. J. (1993). Prostaglandin E2 stimulates ion

transport in prairie dog gallbladder. Digestive Diseases and Sciences. 38, 167-172.

Scales, W. E., Chensue, S. W., Otterness, I., Kunkel, S. L. (1989). Regulation of monokine

gene expression: prostaglandin E2 suppresses tumor necrosis factor but not interleukin-1 alpha or

beta-mRNA and cell-associated bioactivity. Journal of Leukocyte Biology. 45(5), 416-421.

136

Schindler, R., Mancilla, J., Endres, S., Ghorbani, R., Clark, S. C., & Dinarello, C. A.

(1990). Correlations and interactions in the production of interleukin-6 (IL-6), IL-1, and tumor

necrosis factor (TNF) in human blood mononuclear cells: IL-6 suppresses IL-1 and TNF. Blood.

75, 40-47.

Seong, H., Kang, Y. D., Lee, S. I. E., Shim, S. S., Romero, R., & Yoon, B. H. (2006). Lack of

evidence for microorganisms in most women with clinical chorioamnionitis: A need to revisit the

clinical and microbiologic criteria for one of the most important obstetrical complications.

American Journal of Obstetrics and Gynecology. 195, S73-S73.

Servin, A. L. (2004). Antagonistic activities of lactobacilli and bifidobacteria against microbial

pathogens. FEMS Microbiology Reviews. 28, 405-440.

Shinomiya, S., Naraba, H., Ueno, A., Utsunomiya, I., Maruyama, T., Ohuchida, S.,

Ushikubi, F., Yuki, K., Narumiya, S., Sugimoto, Y., Ichikawa, A., Oh-ishi, S. (2001).

Regulation of TNFalpha and interleukine-10 production by prostaglandins I(2) and E(2): studies

with prostaglandin receptor-deficient mice and prostaglandin E-receptor subtype-selective

synthetic agonists. Biochemical Pharmacology. 61(9), 1153-1160.

Shon, Y.H., Kim, J.H., Nam, K.S. (2002). Effect of Astragali radix extract on

lipopolysaccharide-induced inflammation in human amnion. Biol Pharm Bull. 25, 77-80.

Shynlova, O., Mitchell, J. A., Tsampalieros, A., Langille, B. L., & Lye, S. J. (2004).

Progesterone and gravidity differentially regulate expression of extracellular matrix components

in the pregnant rat myometrium. Biology of Reproduction. 70, 986-992.

137

Shynlova, O., Tsui, P., Dorogin, A., Chow, M., & Lye, S. J. (2005). Expression and

localization of alpha-smooth muscle and gamma-actins in the pregnant rat myometrium. Biology

of Reproduction. 73, 773-780.

Shynlova, O., Tsui, P., Jaffer, S., & Lye, S. J. (2009). Integration of endocrine and mechanical

signals in the regulation of myometrial functions during pregnancy and labour. European Journal

of Obstetrics, Gynecology, and Reproductive Biology. 144 Suppl 1, S2-10.

Sierra-Filardi, E., Puig-Kroger, A., Blanco, F. J., Nieto, C., Bragado, R., Palomero, M. I.,

Bernabeu, C., Vega, M. A., & Corbi, A. L. (2011). Activin A skews macrophage polarization

by promoting a pro-inflammatory phenotype and inhibiting the acquisition of anti-inflammatory

macrophage markers. Blood. epub ahead of print

Silver, R. M., Edwin, S. S., Umar, F., Dudley, D. J., Branch, D. W., & Mitchell, M. D.

(1997). Bacterial lipopolysaccharide-mediated murine fetal death: The role of interleukin-1.

American Journal of Obstetrics and Gynecology. 176, 544-549.

Silver, R. M., Lohner, W. S., Daynes, R. A., Mitchell, M. D., & Branch, D. W. (1994).

Lipopolysaccharide-induced fetal death: The role of tumor-necrosis factor alpha. Biology of

Reproduction. 50, 1108-1112.

Skinner, K. A., & Challis, J. R. (1985). Changes in the synthesis and metabolism of

prostaglandins by human fetal membranes and decidua at labor. American Journal of Obstetrics

and Gynecology. 151, 519-523.

138

Slater, D., Allport, V., & Bennett, P. (1998). Changes in the expression of the type-2 but not

the type-1 cyclo-oxygenase enzyme in chorion-decidua with the onset of labour. British Journal

of Obstetrics and Gynaecology. 105, 745-748.

Smieja, Z., Zakar, T., & Olson, D. M. (1993). Stimulation of cultured amnion cell

prostaglandin endoperoxide H synthase activity by glucocorticoids and phorbol ester. American

Journal of Obstetrics and Gynecology. 169, 653-661.

Smith, R., & Nicholson, R. C. (2007). Corticotrophin releasing hormone and the timing of birth.

Frontiers in Bioscience. 12, 912-918.

Smith, R., Smith, J. I., Shen, X., Engel, P. J., Bowman, M. E., McGrath, S. A., Bisits, A. M.,

McElduff, P., Giles, W. B., & Smith, D. W. (2009). Patterns of plasma corticotropin-releasing

hormone, progesterone, estradiol, and estriol change and the onset of human labor. Journal of

Clinical Endocrinology and Metabolism. 94, 2066-2074.

Smith, W. L., Rollins, T. E., & DeWitt, D. L. (1981). Subcellular localization of prostaglandin

forming enzymes using conventional and monoclonal antibodies. Progress in Lipid Research. 20,

103-110.

Smulian, J. C., Shen-Schwarz, S., Vintzileos, A. M., Lake, M. F., & Ananth, C. V. (1999).

Clinical chorioamnionitis and histologic placental inflammation. Obstetrics and Gynecology. 94,

1000-1005.

139

So, J.S., Song, M.K., Kwon, H.K., Lee, C.G., Chae, C.S., Sahoo, A., Jash, A., Lee, S.H.,

Park, Z.Y., Im, S.H. (2011). Lactobacillus casei enhances type II collagen/glucosamine-

mediated suppression of inflammatory responses in experimental osteoarthritis. 88, 358-366.

So, T. (1993). The role of matrix metalloproteinases for premature rupture of the membranes].

Nippon Sanka Fujinka Gakkai Zasshi. 45, 227-233.

Song, Y. L., Kato, N., Matsumiya, Y., Liu, C. X., Kato, H., & Watanabe, K. (1999).

Identification of and hydrogen peroxide production by fecal and vaginal lactobacilli isolated

from japanese women and newborn infants. Journal of Clinical Microbiology. 37, 3062-3064.

Soper, D. E. (1993). Bacterial vaginosis and postoperative infections. American Journal of

Obstetrics and Gynecology. 169, 467-469.

Spiegel, C. A. (1991). Bacterial vaginosis. Clinical Microbiology Reviews. 4, 485-502.

Stallmach, T., Hebisch, G., Joller, H., Kolditz, P., & Engelmann, M. (1995). Expression

pattern of cytokines in the different compartments of the feto-maternal unit under various

conditions. Reproduction, Fertility, and Development. 7, 1573-1580.

Steele, E. C.,Jr, Kerscher, S., Lyon, M. F., Glenister, P. H., Favor, J., Wang, J., & Church,

R. L. (1997). Identification of a mutation in the MP19 gene, Lim2, in the cataractous mouse

mutant To3. Molecular Vision. 3, 5.

Steensberg, A., Fischer, C. P., & Keller, C. (2003). IL-6 enhances plasma IL-1ra, IL-10, and

cortisol in humans. American Journal of Physiology and Endocrinology Metabolism. 285, 433-7.

140

Stevens, M.Y., Challis, J.R., Lye, S.J. (1998). Corticotropin-releasing hormone receptor

subtype 1 is significantly up-regulated at the time of labor in the human myometrium. Journal of

Clinical Endocrinology and Metabolism. 83, 4107-4115.

Sun, K., Myatt, L. (2003). Enhancement of glucocorticoid-induced 11beta-hydroxysteroid

dehydrogenase type 1 expression by proinflammatory cytokines in cultured human amnion

fibroblasts. Endocrinology. 144, 5568-5677.

Sun M., Ramirez M., Challis J. R., Gibb W. (1996). Immunohistochemical localization of the

glucocorticoid receptor in human fetal membranes and decidua at term and preterm delivery.

Journal of Endocrinology. 149, 243-248.

Svinarich, D. M., Bitonti, O. M., Araneda, H., Romero, R., & Gonik, B. (1996). Induction

and postranslational expression of G-CSF and RANTES in a first trimester trophoblast cell line

by lipopolysaccharide. American Journal of Reproductive Immunology (New York, N.Y.: 1989).

36, 256-259.

Sweet, R. L. (1995). Role of bacterial vaginosis in pelvic inflammatory disease. Clinical

Infectious Diseases. 20, 271-275.

Tazawa, R., Xu, X. M., Wu, K. K., & Wang, L. H. (1994). Characterization of the genomic

structure, chromosomal location and promoter of human prostaglandin H synthase-2 gene.

Biochemical and Biophysical Research Communications. 203, 190-199.

Teixeira, F. J., Zakar, T., Hirst, J. J., Guo, F., Sadowsky, D. W., Machin, G., Demianczuk,

N., Resch, B., & Olson, D. M. (1994). Prostaglandin endoperoxide-H synthase (PGHS) activity

141

and immunoreactive PGHS-1 and PGHS-2 levels in human amnion throughout gestation, at

term, and during labor. Journal of Clinical Endocrinology and Metabolism. 78, 1396-1402.

Thiex, N. W., Chames, M. C., & Loch-Caruso, R. K. (2009). Tissue-specific cytokine release

from human extra-placental membranes stimulated by lipopolysaccharide in a two-compartment

tissue culture system. Reproductive Biology and Endocrinology. 7, 117-127.

Thorburn, G. D. (1992). The placenta, PGE2 and parturition. Early Human Development. 29,

63-73.

Thoren, S., & Jakobsson, P. J. (2000). Coordinate up- and down-regulation of glutathione-

dependent prostaglandin E synthase and cyclooxygenase-2 in A549 cells. inhibition by NS-398

and leukotriene C4. European Journal of Biochemistry. 267, 6428-6434.

Thoren, S., Weinander, R., Saha, S., Jegerschold, C., Pettersson, P. L., Samuelsson, B.,

Hebert, H., Hamberg, M., Morgenstern, R., & Jakobsson, P. J. (2003). Human microsomal

prostaglandin E synthase-1: Purification, functional characterization, and projection structure

determination. Journal of Biological Chemistry. 278, 22199-22209.

Tilg, H., Trehu, E., Atkins, M. B., Dinarello, C. A., & Mier, J. W. (1994). Interleukin-6 (IL-6)

as an anti-inflammatory cytokine: Induction of circulating IL-1 receptor antagonist and soluble

tumor necrosis factor receptor p55. Blood. 83, 113-118.

Tomeczek, L., Reid, G., Cuperus, P. L., McGroarty, J. A., van der Mei, H. C., Bruce, A.

W., Khoury, A. E., & Busscher, H. J. (1992). Correlation between hydrophobicity and

142

resistance to nonoxynol-9 and vancomycin for urogenital isolates of lactobacilli. FEMS

Microbiology Letters. 73, 101-104.

Unlugedik, E., Alfaidy, N., Holloway, A., Lye, S., Bocking, A., Challis, J., & Gibb, W.

(2010). Expression and regulation of prostaglandin receptors in the human placenta and fetal

membranes at term and preterm. Reproduction, Fertility, and Development. 22, 796-807.

van der Pouw Kraan, T.C., Boeije, L.C., Smeenk, R.J., Wijdenes, J., Aarden, L. A. (1995).

Prostaglandin-E2 is a potent inhibitor of human interleukin 12 production. Journal of

Experimental Medicine. 181, 775-779.

Van Meir, C. A., Matthews, S. G., Keirse, M. J., Ramirez, M. M., Bocking, A., & Challis, J.

R. (1997). 15-hydroxyprostaglandin dehydrogenase: Implications in preterm labor with and

without ascending infection. Journal of Clinical Endocrinology and Metabolism. 82, 969-976.

Van Meir, C. A., Sangha, R. K., Walton, J. C., Matthews, S. G., Keirse, M. J., & Challis, J.

R. (1996). Immunoreactive 15-hydroxyprostaglandin dehydrogenase (PGDH) is reduced in fetal

membranes from patients at preterm delivery in the presence of infection. Placenta. 17, 291-297.

Vasquez, A., Jakobsson, T., Ahrne, S., Forsum, U., & Molin, G. (2002). Vaginal

Lactobacillus flora of healthy Swedish women. Journal of Clinical Microbiology. 40, 2746-2749.

Veith, G. L., & Rice, G. E. (1999). Interferon gamma expression during human pregnancy and

in association with labour. Gynecologic and Obstetric Investigation. 48, 163-167.

143

Velraeds, M. M., van de Belt-Gritter, B., van der Mei, H. C., Reid, G., & Busscher, H. J.

(1998). Interference in initial adhesion of uropathogenic bacteria and yeasts to silicone rubber by

a Lactobacillus acidophilus biosurfactant. Journal of Medical Microbiology. 47, 1081-1085.

Velraeds, M. M., van der Mei, H. C., Reid, G., & Busscher, H. J. (1996). Inhibition of initial

adhesion of uropathogenic enterococcus faecalis by biosurfactants from Lactobacillus isolates.

Applied and Environmental Microbiology. 62, 1958-1963.

Vercauteren, M., Palit, S., Soetens, F., Jacquemyn, Y., & Alahuhta, S. (2009).

Anaesthesiological considerations on tocolytic and uterotonic therapy in obstetrics. Acta

Anaesthesiologica Scandinavica. 53, 701-709.

Verna, E.C., Lucak, S. (2010). Use of probiotics in gastrointestinal disorders: what to

recommend? Therapeutic advances in gastroenterology. 3, 307-319.

Vogel, I., Thorsen, P., Curry, A., Sandager, P., & Uldbjerg, N. (2005). Biomarkers for the

prediction of preterm delivery. Acta Obstetricia Et Gynecologica Scandinavica. 84, 516-525.

Voltan, S., Martines, D., Elli, M., Brun, P., Longo, S., Porzionato, A., Macchi, V., D'Inca,

R., Scarpa, M., Palu, G., Sturniolo, G. C., Morelli, L., & Castagliuolo, I. (2008).

Lactobacillus crispatus M247-derived H2O2 acts as a signal transducing molecule activating

peroxisome proliferator activated receptor-gamma in the intestinal mucosa. Gastroenterology.

135, 1216-1227.

Whittle, W.L., Benzie, R.J., Gibb, W. (1994). Influence of culture media on prostaglandin

output by dispersed amnion cells. Prostaglandins Leukot Essent Fatty Acids. 50, 93-96.

144

Whittle, W. L., Gibb, W., & Challis, J. R. (2000). The characterization of human amnion

epithelial and mesenchymal cells: The cellular expression, activity and glucocorticoid regulation

of prostaglandin output. Placenta. 21, 394-401.

Whittle, W. L., Patel, F. A., Alfaidy, N., Holloway, A. C., Fraser, M., Gyomorey, S., Lye, S.

J., Gibb, W., & Challis, J. R. (2001). Glucocorticoid regulation of human and ovine parturition:

The relationship between fetal hypothalamic-pituitary-adrenal axis activation and intrauterine

prostaglandin production. Biology of Reproduction. 64, 1019-1032.

Witkin, S. S., Linhares, I. M., Bongiovanni, A. M., Herway, C., & Skupski, D. (2011).

Unique alterations in infection-induced immune activation during pregnancy. BJOG. 118, 145-

153.

Xing, Z., Gauldie, J., Cox, G., Baumann, H., Jordana, M., Lei, X. F., & Achong, M. K.

(1998). IL-6 is an anti-inflammatory cytokine required for controlling local or systemic acute

inflammatory responses. Journal of Clinical Investigation. 101, 311-320.

Xu, H. Y., Tian, W. H., Wan, C. X., Jia, L. J., Wang, L. Y., Yuan, J., Liu, C. M., Zeng, M.,

& Wei, H. (2008). Antagonistic potential against pathogenic microorganisms and hydrogen

peroxide production of indigenous lactobacilli isolated from vagina of Chinese pregnant women.

Biomedical and Environmental Sciences. 21, 365-371.

Xu, X. M., Hajibeige, A., Tazawa, R., Loose-Mitchell, D., Wang, L. H., & Wu, K. K. (1995).

Characterization of human prostaglandin H synthase genes. Advances in Prostaglandin,

Thromboxane, and Leukotriene Research. 23, 105-107.

145

Yadav, A., Saini, V., & Arora, S. (2010). MCP-1: Chemoattractant with a role beyond

immunity: A review. Clinica Chimica Acta. 411, 1570-1579.

Yamamoto, T., Zhou, X., Williams, C. J., Hochwalt, A., & Forney, L. J. (2009). Bacterial

populations in the vaginas of healthy adolescent women. Journal of Pediatric and Adolescent

Gynecology. 22, 11-18.

Yan, F., & Polk, D. B. (2002). Probiotic bacterium prevents cytokine-induced apoptosis in

intestinal epithelial cells. Journal of Biological Chemistry. 277, 50959-50965.

Yeganegi, M., Leung, C. G., Martins, A., Kim, S. O., Reid, G., Challis, J. R., & Bocking, A.

D. (2010). Lactobacillus rhamnosus GR-1-induced IL-10 production in human placental

trophoblast cells involves activation of JAK/STAT and MAPK pathways. Reproductive

Sciences. 17, 1043-1051.

Yeganegi, M., Leung, C. G., Martins, A., Kim, S. O., Reid, G., Challis, J. R., & Bocking, A.

D. (2011). Lactobacillus rhamnosus GR-1 stimulates colony-stimulating factor 3 (granulocyte)

(CSF3) output in placental trophoblast cells in a fetal sex-dependent manner. Biology of

Reproduction. 84, 18-25.

Yeganegi, M., Watson, C. S., Martins, A., Kim, S. O., Reid, G., Challis, J. R., & Bocking, A.

D. (2009). Effect of Lactobacillus rhamnosus GR-1 supernatant and fetal sex on

lipopolysaccharide-induced cytokine and prostaglandin-regulating enzymes in human placental

trophoblast cells: Implications for treatment of bacterial vaginosis and prevention of preterm

labor. American Journal of Obstetrics and Gynecology. 200, 532.e1-532.e8.

146

Yoon, B. H., Park, C. W., & Chaiworapongsa, T. (2003). Intrauterine infection and the

development of cerebral palsy. BJOG. 110, 124-127.

Yoshida M., Sagawa N., Itoh H., Yura S., Takemura M., Wada Y., Sato T., Ito A., Fujii S.

(2002). Prostaglandin F2α, cytokines and cyclic mechanical stretch augment matrix

metalloproteinase-1 secretion from cultured human uterine cervical fibroblast cells. Molecular

Human Reproduction. 8, 681-687.

Zaga, V., Estrada-Gutierrez, G., Beltran-Montoya, J., Maida-Claros, R., Lopez-Vancell, R.,

& Vadillo-Ortega, F. (2004). Secretions of interleukin-1beta and tumor necrosis factor alpha by

whole fetal membranes depend on initial interactions of amnion or choriodecidua with

lipopolysaccharides or group B streptococci. Biology of Reproduction. 71, 1296-1302.

Zaga-Clavellina, V., Garcia-Lopez, G., Flores-Herrera, H., Espejel-Nunez, A., Flores-

Pliego, A., Soriano-Becerril, D., Maida-Claros, R., Merchant-Larios, H., & Vadillo-Ortega,

F. (2007). In vitro secretion profiles of interleukin (IL)-1beta, IL-6, IL-8, IL-10, and TNF alpha

after selective infection with Escherichia coli in human fetal membranes. Reproductive Biology

and Endocrinology. 5, 46-53.

Zakar, T., Hirst, J. J., Mijovic, J. E., & Olson, D. M. (1995). Glucocorticoids stimulate the

expression of prostaglandin endoperoxide H synthase-2 in amnion cells. Endocrinology. 136,

1610-1619.

Zakar, T., Teixeira, F. J., Hirst, J. J., Guo, F., MacLeod, E. A., & Olson, D. M. (1994).

Regulation of prostaglandin endoperoxide H synthase by glucocorticoids and activators of

protein kinase C in the human amnion. Journal of Reproduction and Fertility. 100, 43-50.