IMPERIAL COLLEGE

The roles of Prostanoid EP receptors in the

control of contractions of human myometrium

Shankari Arulkumaran

Institute of Reproductive and Developmental Biology

MD (Res)

2

TABLE OF CONTENTS

Declaration 4

Acknowledgements 5

Publications 6

Poster presentations 7

Abbreviations 8

Table of Figures 11

Abstract 13

1.1 Foreword 15

1.2 Clinical Aspects of Preterm Birth 21

1.3 The Anatomy of the Uterus and Cervix 47

1.4 The Endocrinology and Biochemistry of Labour 49

1.5 Myometrial Contractility 50

1.6 Hypothesis and Objectives 73

2.0 Materials and Methods 81

2.1 Contractility Studies 86

2.2 Drug Assessment 91

2.3 Data Analysis 93

3.1 Contractility in upper versus lower segment

myometrium

95

3.2 Contractility in prostaglandin stimulated upper and

lower segment myometrium

97

3.3 Discussion of contractility in upper and lower

segment myometrium

99

4.1 The effect of PGE2 on spontaneously contracting

lower segment myometrium.

101

4.2 The effect of PGF2α on spontaneously contracting

lower segment myometrium.

104

4.3 The effect of acetyl salicylic acid (ASA) on

spontaneously contracting lower segment myometrium

and PGE2 induced contractions

105

3

4.4 Discussion of the effects of PGE2, PGF2α and ASA

on contractility

111

5.1 The effect of an EP1 antagonist, ZD6416 upon

spontaneous and PGE2 stimulated contractions

113

5.2 The effect of an EP3 antagonist, L798106 upon

spontaneous and PGE2 stimulated contractions

119

5.3 Discussion of the role of EP1 and EP3 in human

lower segment myometrial contractility in relation to

their respective receptor antagonists

125

6.1 The effect of FPA, a PGF2α antagonist upon

spontaneous and PGF2α stimulated contractions

129

6.2 Discussion of the effect of FPA, a novel PGF2α

antagonist on contractility

133

7.1 The effect of combining the agonists, OT and PGE2

on spontaneously contracting lower segment

myometrium

135

7.2 The effect of combining the antagonists of OT and

PGE2 on spontaneously contracting lower segment

myometrium

138

7.3 Discussing the effect of combining the agonists, OT

and PGE2 and their respective antagonists on

spontaneously contracting lower segment myometrium

140

8.1 Discussion 142

References 154

4

DECLARATION

I declare that this thesis is my own work and is based on research that was

undertaken by me at Institute of Reproductive and Developmental Biology,

Imperial College Heathcare Trust from September 2005-2008. To the best

of knowledge, this thesis contains no material previously published or

written by another person, except where due reference has been made in the

text.

5

ACKNOWLEDGEMENTS

I would like to sincerely thank all the women at Queen Charlotte’s Hospital

who selflessly volunteered to partake in this project. This project would not

have been possible without their understanding of the bigger picture and

willingness to help other women and babies in need.

I have personal gratitude to my supervisor, Professor Bennett, for giving me

this opportunity. His unfailing patience and guidance has led not only to my

completion of this MD but he has been instrumental in my career

progression. I have him to blame for keeping me in Obstetrics and that is the

true definition of a mentor.

I am grateful to all the staff on the Queen Charlotte’s labour ward; past and

present, as well as my colleagues in the laborotary. In particular, to Yooni

Lee for her patience in trying to teach a clinician, basic science and to

Mandeep Kandola for her collaboration on this project.

I would like to thank Ian for his support, understanding and love throughout

my good and bad days.

I am most grateful to my parents, Arul and Gaytri for their love and support.

It is their solid foundation that has got me here and high standards that will

continue to spur me on. This MD is dedicated to them.

6

PUBLICATIONS

Arulkumaran S, Kandola MK, Hoffman B, Hanyaloglu AC, Johnson MR,

Bennett PR. The Roles of Prostaglandin EP 1 and 3 Receptors in the Control

of Human Myometrial Contractility. J Clin Endocrinol Metab. 2011.

Arulkumaran S, Bennett PR. Pathophysiology of Preterm birth in Fetal and

Neonatal Physiology. Polin, Fox, Abman (eds.) 4th

edition 2011.

Kandola MK, Almazidi M, Samarage M, Khanjani S, Arulkumaran S,

Bennett PR. The Role of EP2 and EP4 in Human Myometrium.

Reproductive Sciences 2010; (17): 81A-81A.

Taher SE, Chandiramani M, Soutter P, Arulkumaran S, Eliahoo J, Teoh T,

Hassan S, McIndoe A, Shennan A, Bennett PR. Ultrasound-indicated

cerclage for prevention of preterm delivery in primiparas following

treatment of cervical intraepithelial neoplasia. Ultrasound Obstet Gynecol

2009; 34(S1):144.

Arulkumaran S, Chollet A, Bennett PR. The role of PGF2 on myometrial

contractility studied with a selective FP antagonist. Reproductive Sciences

2008; (15): 3A-57A.

Arulkumaran S, Kandola M, McArthur Wilson RJ, Brooks DP, Bennett PR.

EP3 versus EP1 receptor antagonism in human term myometrial

contractility. Reproductive Sciences 2008; (15): 3A-57A.

Arulkumaran S, Kandola M, Wilson RJ, Brooks DP, Bennett PR. The Role

of Prostaglandin E2 Receptors in Human Myometrium During Pregnancy.

Reproductive Sciences 2007; (14): 227.

7

POSTER PRESENTATIONS

The role of PGF2 on myometrial contractility studied with a selective FP

antagonist.

Arulkumaran S, Chollet A, Bennett PR.

Society for Gynecologic Investigation International Conference, San Diego,

California, United States of America. March 2008.

EP3 versus EP1 receptor antagonism in human term myometrial

contractility.

Arulkumaran S, Kandola M, McArthur Wilson RJ, Brooks DP, Bennett PR.

Society for Gynecologic Investigation International Conference, San Diego,

California, United States of America. March 2008.

The role of prostaglandin E2 receptors in human myometrium during

pregnancy.

Arulkumaran S, Kandola M, Wilson RJ, Brooks DP, Bennett PR.

Society for Gynecologic Investigation International Conference, Las Vegas,

Nevada, United States of America. March 2007.

8

ABBREVIATIONS

AA arachidonic acid

ADH anti-diuretic hormone

ANOVA one-way analysis of variance

ASA acetyl salicylic acid

ASF/SF2 alternative splicing factor/splicing factor 2

ATP adenosine triphosphate

BkCa large conductance calcium-activated potassium (channel)

BMI body mass index

BV bacterial vaginosis

Ca Cl2 calcium chloride

Ca2+

calcium ion

cAMP cyclic adenosine monophosphate

CAP contraction associated protein

CI confidence interval

COX Cyclooxygenase

CPI-17 myosin phosphatase inhibitor phosphoprotein

CRH corticotrophin releasing hormone

CRP C-reactive protein

CX-43 connexin -43

DAPI 4',6-diamidino-2-phenylindole fluorescent stain

EP prostaglandin E receptor

fFN fetal fibronectin

Fig. Figure

FP prostaglandin F receptor

FPA prostaglandin F antagonist (Serono)

G protein guanine triphosphate-binding protein

GAPDH Glyceraldehyde 3-phosphate dehydrogenase

GPCRs G-protein coupled receptors

GSK Glaxo-Smith Kline

GTN Glyceryl trinitrate

GTP guanosine triphosphate

GW796679X OT anatgonist (GSK)

Gαi / Gαq / Gαs heterotrimeric G protein subunit

IL-6 interleukin -6

IL-8 interleukin -8

InsP3 inositol 1, 4, 5,-triphosphate

IVH intraventricular haemorrhage

KCL potassium chloride

9

kDa kilo Dalton

KH2PO3 monopotassium phosphate

Ki dissociation constant

L- labour minus

L+ labour plus

L798106 EP3 antagonist (GSK)

LLETZ laser loop excision of the transformation zone

LPS Lipopolysaccharide

LS lower segment

MgSO4 Magnesium sulphate

MLCK myosin light chain kinase

MLCP myosin phophotase

MMP matrix metalloproteinase

MRC Medical Research Council

mRNA messenger ribonucleic acid

NaHCO3 sodium bicarbonate

NaCl sodium chloride

NF-κB nuclear factor kappa beta

No. Number

NSAID non-steroidal anti-inflammatory drug

OT oxytocin

OTR oxytocin receptor

p Probability

pc personal computer

PG prostaglandin

PGD2 prostaglandin D2

PGE2 prostaglandin E2

PGF2α prostaglandin F2 alpha

PGG2 hydroperoxy endoperoxide

PGH2 hydroxyl endoperoxide

PGHS prostaglandin G/H synthase

PGI2 Prostacylcin

pH power' of hydrogen ion

PKA cAMP dependent protein kinase

PPROM preterm premature rupture of membranes

PR Progesterone

PUFA unbalanced polyunsaturated fatty acids

RCOG Royal College of Obstetricians and Gynaecologists

RDS respiratory distress syndrome

RNA ribonucleic acid

10

RR relative risk

RU486 Mifepristone

SDS-PAGE sodium dodecyl sulphate polyacrylamide gel

electrophoresis

SNP single nucleotide polymorphism

SP-A surfactant protein A

STD sexually transmitted disease

TXA2 Thromboxane A2

TXB2 Thromboxane B2

TNFα tumour necrosis factor alpha

US upper segment

ZD6416 EP1 antagonist (GSK) oC degree Celcius

β Beta

γ Gamma

11

TABLE OF FIGURES

Chapter 1

Page

Fig. 1.1 Cerebral palsy rates according to gestational age,

infant gender and plurality 16

Fig. 1.2 Neonatal survival by gestational age and improvement

in survival by week 22

Fig. 1.3 Epidemiological and environmental risk factors for preterm

labour 25

Fig. 1.4 Chorioamnionitis 28

Fig. 1.5 Placental abruption 30

Fig. 1.6 Cervical length measurement 36

Fig. 1.7 Anatomy of the non-pregnant uterus 49

Fig. 1.8 Factors influencing preterm labour 51

Fig. 1.9 Pathway for action of placental corticotrophin releasing

hormone on parturition. 54

Fig. 1.10 Biosynthetic pathway for the formation of eicosanoids derived

from arachidonic acid 58

Fig. 1.11 Signaling pathways for EP1 and EP2/4 67

Fig. 1.12 Signaling pathways for EP3 isoforms 69

Fig. 1.13 Summary of the literature findings with regards to myometrial

expression and changes of EP1 and EP3 72

Fig. 1.14 Labour associated change of EP1 receptor protein expression

in human myometrium at term 75

Fig. 1.15 Labour associated change of EP3 receptor protein expression

in human myometrium at term 76

Fig. 1.16 EP1 receptor expression in matched LS and US segments of

term L+ and L- myometrium 78

Fig.1.17 EP3 receptor expression in matched LS and US segments of

term L+ and L- myometrium 79

Chapter 2

Fig. 2.1 Myometrial sample taken from the lower segment of human

myometrium at elective caesarean sections at term 82

Fig. 2.2 Myometrial sample taken from the upper segment of human

myometrium at elective caesarean sections at term 84

Fig. 2.3 A cervical punch biopsy 85

Fig. 2.4 The Myograph 800MS 87

Fig. 2.5 The effect of stretch on myometrial contractility 89

Fig. 2.6 A real time image of spontaneous contractions 90

Fig. 2.7 An illustration of the parameters used to calculate the data 94

12

Chapter 3

Fig. 3.1 Spontaneous contractility in upper versus lower segment

myometrium 96

Fig. 3.2 Contractility in prostaglandin stimulated upper and lower

segment myometrium 98

Chapter 4

Fig. 4.1 A pictorial representation of increasing concentrations of

PGE2 on lower segment myometrium 101

Fig. 4.2 The effect of PGE2 on spontaneously contracting lower

segment myometrium 103

Fig. 4.3 A pictorial representation of increasing concentrations of

PGF2α on lower segment myometrium 104

Fig. 4.4 The effect of PGF2α on spontaneously contracting lower

segment myometrium 106

Fig. 4.5 The effect of ASA on spontaneously contracting lower

segment myometrium 108

Fig. 4.6 The effect of ASA on PGE2 induced contractions 110

Chapter 5

Fig. 5.1 The effect of increasing concentrations of ZD6416 on

spontaneous contractions 114

Fig. 5.2 The effect of pre-incubating LS myometrium with ZD6416 on

spontaneous contractions 116

Fig. 5.3 The effect of ZD6416 on PGE2 stimulated contractions 118

Fig. 5.4 The effect of increasing concentrations of L798106 on

spontaneous contractions 120

Fig. 5.5 The effect of pre-incubating LS myometrium with L798106

on spontaneous contractions 122

Fig. 5.6 The effect of L798106 on PGE2 stimulated contractions 124

Chapter 6

Fig. 6.1 The effect of increasing concentrations of FPA on

spontaneous contractions 130

Fig. 6.2 The effect of FPA on PGF2α stimulated contractions 132

Chapter 7

Fig. 7.1 The effect of combining OT and PGE2 on spontaneous

contractions 137

Fig. 7.2 The effect of combining GW796679X and L798106 on

spontaneous contractions 139

13

ABSTRACT

The primary function of the uterus during pregnancy is to harbour the

growing fetus in a quiescent environment. Upon maturation of the fetus, the

uterus generates forceful contractions during labour. Prostaglandins are

central in the parturition process. PGE2 in particular, is produced in large

quantities by fetal membranes and possible roles include cervical ripening

and the stimulation of uterine contractions.

PGE2 exhibits a wide spectrum of physiological actions depending on the

distribution and subtypes of EP receptors. EP1 and EP3 mediate

contractions, but there is no consensus about their relative importance.

Preliminary experiments were undertaken to examine the possible

expression patterns of EP1 and EP3. The expression data showed there to be

no significant changes between the localisation in upper or lower segment

biopsies using both RNA and protein samples. However, there was a labour

associated change seen in EP3 isoform expression at mRNA level together

with changes in the protein expression of EP3 in the lower segment using

immunofluorescence. Given that the expression studies suggest that it is the

EP3 receptor, and not the EP1 receptor, that is most important in

myometrial contractility, I established an in vitro tissue bath to conduct

functional studies examining contractility in upper and lower segment

myometrial biopsies.

14

My data demonstrated that stretch of term human myometrium results in

spontaneous contractions. There was no difference between upper and lower

segment myometrium, so subsequent experiments were conducted on lower

segment biopsies. The addition of acetyl salicylic acid (a non-selective COX

inhibitor) did not completely stop spontaneous contractility but reduced the

total work done significantly. This could be reversed by add back of PGE2.

Prostaglandins are therefore significant but not crucial for spontaneous

contractility. Both PGE2 and PGF2α increased the total work done

significantly compared to spontaneous contractions; E2 more so than F2α.

The PGF2α antagonist did not inhibit spontaneous contractions, but did

inhibit PGF2α induced contractions. The EP1 antagonist did not inhibit

spontaneous contractions but EP3 antagonist did. These results suggest that

an EP3 antagonist is a better candidate as a new tocolytic compound

compared to FP or EP1 antagonist.

15

1.1 Foreword

For the most part of the 20th

century, preterm birth, defined as birth at less

than 37 completed weeks of gestation, was viewed as an unpredictable and

inevitable fact of life. Medical efforts thus focused on reducing the

consequences of prematurity, rather than preventing its occurrence. This

approach resulted in improved neonatal outcomes, but preterm birth remains

costly both emotionally and financially to the affected infants and their

families, and adds an economic burden to society.

The EPIPAGE study (Larroque B et al 2008) concluded that one in ten

children born before 33 weeks gestation had cerebral palsy by the age of

five years and 40% had some kind of disability. Fig 1.1 illustrates the

prevalence of cerebral palsy decreasing with increasing gestational age (i.e.

20% at <27 weeks, 12% at 27 to 28 weeks, 8% at 29 to 30 weeks, 7% at 31

weeks, and 4% at 32 weeks).

16

No.

Proportion With Cerebral

Palsy, % P

Total 1954 8.2

Gestational age at

birth

<.001

24–25 wk 62 19.4

26 wk 82 22.0

27 wk 146 12.3

28 wk 191 11.0

29 wk 196 8.2

30 wk 315 8.3

31 wk 424 6.8

32 wk 538 4.4

Infant gender .10

Male 1018 9.2

Female 936 7.1

Plurality .35

Singletons 1339 8.2

Twins 529 8.9

Triplets and

quadruplets

86 4.4

Fig 1.1 Cerebral Palsy Rates According to Gestational Age, Infant

Gender and Plurality. Cerebral palsy was slightly but not significantly

more frequent among boys than girls. There was however, no

difference in cerebral palsy rates according to plurality (Pierre-Yves

Ancel 2006).

17

Despite diseases of pregnancy and the perinatal period contributing

considerably to the burden of ill health worldwide, there is an apparent lack

of initiative by pharmaceutical companies in this field. Fisk et al (2008)

found just 17 obstetric drugs being developed, compared with 660 for

cardiovascular disease and 34 for amyotrophic lateral sclerosis. Only 9 of

the 17 are completely new, and interestingly, these are mostly oxytocin

antagonists. The authors claim pregnant women are still being treated with

drugs belonging to a bygone age – magnesium sulphate, nifedipine and

hydralazine, and that the market must therefore be modified to allow

obstetric therapeutics to modernise.

A lack of detailed information on the mechanisms of preterm labour has

hindered the development of novel treatments and preventative measures.

Therefore, there is a huge unmet need for effective tocolysis. Tocolytics are

medications used to suppress premature labor (from the Greek tokos,

childbirth, and lytic, capable of dissolving). The use of a tocolytic drug is

associated with a prolongation of pregnancy for up to one week, but with no

significant effect on preterm birth or on perinatal or neonatal morbidity.

However, if the few days are put to good use, such as completing a course

of corticosteroids or in utero transfer, tocolysis should be considered.

Indomethacin, a non-selective prostaglandin G/H synthase (PGHS)

inhibitor, has been used widely in the prevention of preterm birth. King JF

et al (2005) assessed the effects on maternal, fetal and neonatal outcomes of

18

COX inhibitors administered as a tocolytic agent to women in preterm

labour when compared with (i) placebo or no intervention and (ii) other

tocolytics. The review included data from 13 trials with a total of 713

women. Indomethacin was used in 10 trials. When compared with placebo,

COX inhibition (Indomethacin) resulted in a reduction in birth before 37

weeks' gestation (one trial, 36 women), an increase in gestational age and

birthweight (two trials, 67 women). A comparison of non-selective COX

inhibitors versus any COX-2 inhibitor (two trials, 54 women) did not

demonstrate any differences in maternal or neonatal outcomes. Due to the

small numbers, all estimates were thought to be imprecise. However, it has

been its adverse effects, in particular effects on the fetal kidneys and the

ductus arteriosus, as well as associations with an increased risk of intra-

ventricular haemorrhage (IVH) and necrotizing enterocolitis in the neonate

that have limited the use of Indomethacin (Seubert De et al 1999).

Nimesulide, a PGHS-2 selective non-steroidal anti-inflammatory drug

(NSAID) has been used in women at high risk of preterm delivery but is

also associated with serious fetal side effects (Sawdy RJ, Lye S et al 2003).

Rofecoxib, a PGHS-2 specific inhibitor is associated with few fetal side

effects but does not reduce the incidence of preterm delivery and is

associated with an increased risk of late preterm birth (Groom KM et al

2005). The authors give a number of possible explanations for the

disappointing efficacy of prostaglandin synthesis inhibitors including the

possibility that the inhibition of COX-2 may activate feed-forward

19

mechanisms, therefore increasing the availability of arachidonic acid or

directly increasing COX-2 expression. Furthermore, it may be that once the

COX-2 pathway is inhibited, prostaglandins are synthesised via another

pathway, such as COX-1. Finally, it is possible that the inhibition of

prostaglandin synthesis alone is not sufficient to stop the inflammatory

cascade and therefore the process leading to labour continues via other

mechanisms.

Drugs which inhibit the various PG receptors are now available raising the

possibility that specific inhibition of PG receptors might be a strategy to

prevent preterm birth. PG receptors are coupled via G proteins to the second

messengers cAMP, inositol triphosphate or intracellular calcium release. PG

receptors have been classified pharmacologically into five main subtypes

(DP, EP, FP, IP, and TP) one for each of the major classes of PGs i.e. PGD,

PGE, PGF, prostacyclin and thromboxane respectively. Inhibition of FP has

been proposed as a strategy to prevent preterm delivery. FP antagonists have

been shown to inhibit contractions in isolated sheep and human

myometrium (Olson DM et al 2003) and to open BKCa channels (Doheny

HC et al 2007). FP antagonists also delay RU486 (Mifepristone) or

lipopolysaccharide (LPS) induced labour in mice, rat and sheep (Olson DM

et al 2003, Chollet A et al 2007). However PGE2 is a major PG product of

fetal membranes and decidua and is thought to act to induce labour partly

through diffusion to the myometrium. PGE2 binding to myometrium has

been shown to be 8-10 fold higher than PGF2α (Hoffman HJ et al 1984).

20

PGE2 binds to its myometrial receptor with higher affinity than does PGF2α

(Giannopoulos G et al 1985) to its receptor, and PGE2 has been shown to be

8-10 fold more potent than PGF2α in stimulating myometrial contractions

(Martin RL et al 2002). Targeting the PGE receptor alone, or in combination

with other receptor agonists/ antagonists, may therefore be a more effective

strategy for preventing or delaying preterm delivery.

21

1.2 Clinical Aspects of Preterm Birth

Epidemiology

Preterm birth refers to birth before 37 completed weeks of gestation. It

accounts for 6-10% of all births in Western countries and more than two-

thirds of all perinatal deaths. Preterm birth may result from preterm labour,

spontaneous or induced, or it may be by planned Caesarean section due to

maternal or fetal complications (Murphy DJ 2007). It is estimated that up to

30-40% of all cases are induced or elective and the remaining 60-70% occur

spontaneously (Khan KS et al 2007). Epidemiological studies of preterm

labour vary in terms of categorization, but generally, labour at <24 weeks is

considered pre-viable, <28 weeks is extremely preterm, 28-31 weeks very

preterm and 32-36 weeks mildly preterm (Moutquin JM 2003). Gross

outcomes after 34 weeks are generally considered to be as good as those

after 37 weeks gestation (Fig 1.2), though the risk of minor morbidities

remains (Seubert DE et al 1999). It is the 3-7% of preterm births that occur

before 34 weeks gestation that account for around 75% of neonatal mortality

and 50% of long term neurological impairment in children (Khan KS et al

2007).

22

Gestational

age

(weeks)

Approximate

survival

(%)

Approx

improvement in

survival per week

(%)

21 0 -

22 Rare -

23 25 25

24 50 25

25 70 10

26 80 6

27 86 5

28 91 3

29 94 1

30 95 1

31 96 1

32 97 1

33 98 1

34 99 1

35 99+ <1

36 99+ <1

Fig 1.2 Neonatal survival by gestational age and improvement in

survival by week. Gross outcomes after 34 weeks are generally

considered to be as good as those after 37 weeks gestation

(Seubert DE et al 1999).

23

Risk factors for preterm labour include race, age, socio-economic status and

body mass index (BMI). In the United Kingdom, the risk of preterm birth is

6% in white Europeans but 10% in Africans or Afro Caribbeans (Murphy

DJ 2007). Similarly, African-American women are over-represented in rates

of pre-term birth in the United States. A poorer socio-economic status in

such groups and other minority groups add environmental causes to the

plausible genetic causes of preterm birth (i.e. polymorphisms in tumour

necrosis factor alpha, TNF-α, a pro-inflammatory cytokine) (Macones GA et

al 2004, Engel SA et al 2005). Other environmental causes include nutrition,

substance abuse, and psychosocial factors (Murphy DJ 2007).

A strong association with preterm birth in women with a low BMI, poor

weight gain or excess weight gain and obesity was reported in eight primary

studies of 122,647 women (Dietz PM et al 2006). Alcohol consumption,

smoking and cocaine exposure all increase the risk of preterm birth. Dose-

response effects have been described in alcohol consumption in various

European studies, as well as increased risks from low to moderate and heavy

smoking (Albertsen et al 2004, Burguet et al 2004). Studies from the United

States confirm increased risks with cocaine exposure and other psychiatric

and substance use disorders (Bada HS et al 2005, Kelly RH et al 2002).

Maternal stress may increase susceptibility to preterm labour. Maternal

stress may act via a neuroendocrine pathway that activates the maternal-

placental-fetal endocrine systems that promote parturition via an immune

and/or inflammatory pathway (Wadhwa et al 2001). The complex

24

interactions between maternal demographics, environmental exposures and

genetic susceptibility of the mother and fetus require a coordinated research

approach that brings together the disciplines of epidemiology, laboratory

based science and clinical trials to offer the best hope of significant

advances in understanding preterm birth (Murphy DJ 2007) (Fig 1.3).

25

Maternal Demographic

African-American / Hispanic race

Low BMI/ excessive weight gain/ loss

Young maternal age

Obstetric

Previous early pregnancy loss

Previous preterm birth

Short inter-pregnancy interval (<12 months)

Medical

Procedures – LLETZ, amniocentesis

Fetal Male gender

Multi-fetal pregnancy

Assisted conception

Paternal Older paternal age

Environmental Infection

Bacterial vaginosis/ STD’s

Peridontal infection

Socioeconomic/ Psychosexual

Social inequality

Physical violence

Marital status – single/ cohabitation

Stressful/ traumatic life events/ depression

Substance use/ toxins

Excess alcohol - >3 drinks/ day

Smoking – particularly moderate/ heavy

Cocaine

Pollutants – sulphur dioxide/ particulate matter

Nutrition

Elevated homocysteine/ suboptimal B-12, B-6

Unbalanced polyunsaturated fatty acids (PUFA)

Genetic TNF-a pro-inflammatory cytokine – polymorphisms

Factor VII/XII – polymorphisms

Fig 1.3 Epidemiological and environmental risk factors for preterm

labour (Murphy DJ 2007).

26

Causes of Preterm Labour

Common associations with the onset of preterm labour in the human are

intrauterine infection, placental abruption, cervical incompetence, premature

membrane rupture, multiple pregnancy, intrauterine death, fetal and uterine

abnormalities and chorioamnionitis. However in many cases, there is no

discernable underlying pathology (Blanks AM et al 2007). Whilst current

antenatal screening relies on overt clinical and environmental signs and

symptoms, the clustering of preterm labour within families and recurrence

in susceptible women presents the case for investigating a possible

underlying genetic predisposition (Orsi NM et al 2007).

Infection

There is a strong correlation between infection within the uterus and the

onset of spontaneous preterm labour. Infection has the potential to activate

all of the biochemical pathways ultimately leading to cervical ripening and

uterine contractions; even more so in a cervix that may be shortened or

weakened. The bacterial species identified in the majority of cases of

congenital infections is usually identical to that in the maternal lower genital

tract. Following twin preterm delivery, chorioamnionitis is more common

and severe in the presenting twin than in the second twin. These factors all

suggest that ascending infection from the lower genital tract is the

commonest mechanism for chorioamnionitis (von Dadelszen P 2003).

27

There is also a recent hypothesis surrounding chronic periodontal infection

acting as reservoirs for bacterial products and/ or inflammatory mediators

that play a role in the development of preterm labour. Moderate to severe

periodontal disease identified in the antenatal period, is associated with an

increased risk of preterm delivery, as is periodontal disease progression

(Offenbacher S et al 2006). Several causative pathogens have been

identified, including the prevotella and pophyromomas species (Urban E et

al 2006).

Chorioamnionitis

Chorioamnionitis usually starts with overgrowth of potential pathogens in

the maternal lower genital tract. This is followed by the presence of the

organisms in the uterine cavity, particularly in the decidua of the lower

segment. A localized inflammatory reaction then occurs, leading to

deciduitis, chorionitis and extension through the amnion into the amniotic

cavity (Fig 1.4), before finally infection of the fetus itself, by aspiration and

swallowing of infected amniotic fluid (Romero R et al 1988). The most

common microbial isolates from the amniotic cavity of women in preterm

labour are ureaplasma, urealiticum, fusibacteria and mycoplasma hominis.

Bacterial vaginosis and trichomonas vaginalis are the two most studied

organisms associated with preterm birth (Carey JC et al 2000). However,

screening and treatment of bacterial vaginosis does not seem to influence

the increased risk of preterm birth (Klebanoff MA et al 2005).

28

Fig 1.4 Chorioamnionitis is defined by the overgrowth of potential pathogens in the

maternal lower genital tract followed by the presence of the organisms in the

uterine cavity, particularly in the decidua of the lower segment. A localized

inflammatory reaction then occurs, leading to deciduitis, chorionitis and extension

through the amnion into the amniotic cavity before finally infection of the fetus itself

occurs via aspiration and swallowing of infected amniotic fluid.

29

Abruption

Placental abruption (Fig 1.5) may lead to the onset of preterm labour. This is

thought to be through release of thrombin which stimulates myometrial

contractions by protease activated receptors. This may explain the clinical

impression that preterm labour associated with chorioamnionitis is often

rapid whereas that associated with placental abruption is less so, because in

placental abruption there is no pre-ripening of the uterine cervix.

Generation of thrombin may also play a role in preterm labour associated

with chorioamnionitis when it is released as a consequence of decidual

haemorrhage (Nath CA et al 2007).

30

Fig 1.5 Placental abruption is a complication of pregnancy, wherein the placenta

has separated from the uterus of the mother. It is the most common pathological

cause of late pregnancy bleeding. In humans, it refers to the abnormal separation

after 20 weeks of gestation and prior to birth.

31

Cervical Incompetence

The cervix maintains the fetus in situ during pregnancy and dilates during

labour to allow delivery of the baby. This distal component of the uterus is

primarily composed of connective tissue, of which the majority is collagen

arranged in fibres and embedded in proteoglycans. There is a small amount

of smooth muscle (10%) and fibroblast cells (Thomson A et al 2005). Prior

to and in early labour, the biochemical makeup of the cervix changes,

making it softer (i.e. ripens) and allowing for dilatation. These biochemical

changes include a decrease in collagen associated with an increase in

cervical prostaglandins, pro-inflammatory cytokines, leucocytes, cell

adhesion molecules and nitric oxide synthase production (Norman JE 2007).

Classical cervical incompetence may be diagnosed in women who have an

inherent weakness in the cervix or those who have a weakened cervix due to

previous medical or surgical treatments. Such treatments include mechanical

cervical dilatation at the time of a termination of pregnancy or surgical

evacuation of products of conception. During treatment for cervical

neoplasia, destruction of cervical tissue can occur due to laser treatment or

biopsies, and in the case of cervical carcinoma, the entire cervix can be

removed during a trachelectomy. It is now recognized that ‘cervical

competence’ is a continuum. In a woman with a shortened or damaged

cervix, bacteria may ascend into the lower segment of the uterus, stimulate

an inflammatory response, which itself leads to further cervical shortening

and dilation and so to a ‘vicious cycle’ ending with cervical dilatation and

32

then labour. The same syndrome may be seen where the cervix is of normal

length but the microbiological load is high. Variations in microbiological

load may lead to delivery at different gestational ages in different

pregnancies in the same woman.

Multiple pregnancies

The rate of preterm labour is high in clinical conditions such as multiple

pregnancy and polyhydramnios, which are associated with uterine stretch.

Over distension of the uterus leads to premature up regulation of contraction

associated proteins which mediate cervical ripening. Studies in sheep have

shown that stretch increases myometrial the expression of oxytocin

receptors (OTR) and COX-2 (Wu WX et al 1999); whilst in human models,

stretch of myometrial and amnion epithelial cells have been shown to

increase the expression of COX-2 and PGE2 synthesis (Mohan AR et al

2007). In addition, multiple pregnancies are associated with multiple

placentae, and therefore an earlier rise in placental CRH concentrations in

the circulation and subsequently higher rates of preterm birth (Warren WB

et al 1990).

Genetics

The most consistent risk factor in identifying preterm labour is a previous

pregnancy resulting in preterm birth. Most studies quote a risk of 15-18%

with one previous preterm birth and 28-32% with two previous preterm

deliveries (Hoffman HJ et al 1984, Carr-Hill RA et al 1985). This has been

33

suggested to be evidence for a genetic basis to preterm labour. However,

this is problematic, as preterm labour is a complex syndrome involving

various aetiologies and is masked by multiple confounding environmental

risk factors. To date, a single gene determinant for preterm labour inherited

in a conventional Mendelian manner has not been identified. Instead, it is

generally accepted that susceptibility to preterm labour is governed by

multigene effects (Orsi NM et al 2007). These effects would regulate

inflammatory, uteroplacental, endocrine, uterine contraction/ remodeling

and metabolic pathways (Berhman RE et al 2007).

Matrix metalloproteinases (MMPs) are proteolytic enzymes that cause

extracellular matrix degradation and are associated with tissue destruction

and remodeling. In preterm labour, activation is a cytokine-mediated

inflammatory process, resulting in the rupture of fetal membranes and

cervical ripening/ dilation. Single nucleotide polymorphisms (SNPs) which

may regulate MMP expression and activity have been studied. At present, a

vast number of polymorphisms have been identified in these proteases and

their tissue inhibitors (Orsi NM et al 2007, Fujimoto T et al 2002). In

addition to SNPs, complementary analytical approaches such as microarray

technology and chip arrays have now all been harnessed into determining

the susceptibility to preterm labour (Esplin MS et al 2005). Once these

genetic polymorphisms can be clearly related to physiological effects,

individual predispositions to preterm labour may enable a more targeted

approach for preventative care.

34

Prediction of Preterm Labour

As precise disease mechanisms behind preterm birth are not so well

understood, clinical predictive tests and preventative treatments remain

limited. However, if women who are at high risk of spontaneous preterm

birth in early pregnancy can be identified, they may be targeted for more

intensive antenatal care and prophylactic interventions. A history of

previous preterm birth, second trimester loss or induced abortion is a non-

invasive, simple, cost-effective method to identify women who may benefit

from more and intensive screening. Further assessment can be made by the

use of ultrasonographic cervical length measurement, cervical-vaginal fetal

fibronectin screening and the detection of uterine contractions.

Fetal fibronectin

Fetal fibronectin (fFN) is an extracellular matrix found in the decidua

basalis next to the placental intervillous space. It helps to attach the fetal

membranes to the uterine decidua and when mechanical or inflammatory

changes occur, is leaked into the cervico-vaginal fluid. It can be measured

quantitatively in cervico-vaginal secretions, obtained from the posterior

fornix when performing an internal examination with a speculum. A

commercial kit for use on-site has been produced, which gives a positive

result when concentrations are > 50ng/ml.

fFN is useful in women in threatened preterm labour with not only strong,

painful contractions, but also in those who are asymptomatic, where a

35

positive test is considered to reflect the consequences of intrauterine

infection (Shennan A et al 2006). At present, it is widely accepted as a good

test for excluding preterm birth in those presenting with threatened preterm

labour due to its strong negative predictive value: 0.5% deliver within 7-10

days after a negative test (Honest H et al 2002). Furthermore, a positive

result is associated with the greatest odds ratio for preterm delivery before

35 weeks gestation (Goldenburg RL et al 2001).

Cervical length measurement

Ultrasonographic cervical length measurement is a reproducible method of

assessing the cervix, with little inter- and intra-observer error (Fig 1.6). A

transvaginal probe is used to obtain the length between the internal and

external os. The risk of preterm delivery increases exponentially with

decreasing length, from <1% at 30mm to 80% at 5mm (Heath VC et al

1988). In women with symptoms of preterm labour, it is estimated that 50%

of women with a cervical length of less than 15mm deliver within 7 days

(Nicolaides KH et al 2004).

36

A.

B.

Fig.1.6 In ultrasonographic cervical length measurements, a transvaginal

probe is used to obtain the length between the internal and external os.

Cervical shortening is demonstrated above by A) a normal cervical canal

with a closed internal os and B) funneling and shortening of the cervix.

37

Prevention of preterm labour

Cerclage

If preterm labour does occur secondary to cervical incompetence, inserting a

suture around the cervix is commonly used to increase mechanical strength

to the cervix and thereby prevents premature opening in a subsequent

pregnancy. There are two approaches to the cerclage; abdominal and

vaginal. In the vaginal approach, the McDonald or Shirodkar methods are

employed. The McDonald approach involves a suture or tape placed around

the cervix at the junction between the vagina and the cervix. In the

Shirodkar suture, the supra-cervical vagina is incised and the bladder

reflected upwards, allowing the suture to be placed at the cervico-isthmic

junction. The vaginal epithelium is then repaired (Norman JE 2007).

The most common indication for an abdominal cerclage is a failed vaginal

cerclage. It requires a transverse incision in the lower abdomen and the

peritoneum over the cervix and bladder to be deflected down, in order to

place a suture above the cardinal and uterosacral ligaments. A caesarean

section is required to deliver the baby and the suture is often not removed,

unlike in the vaginal cerclage, where removal at 37-38 weeks gestation

enables vaginal delivery of the baby. Abdominal cervical cerclage may also

be placed laparoscopically, although published data is currently limited to

isolated case reports. Although cervical cerclage is widely used in clinical

practice, there is still limited evidence on its efficacy, whether it is

38

beneficial once cervical shortening has occurred and at what length to

intervene (Simcox R et al 2007). An international multicentre trial

conducted by the Medical Research Council (MRC) and Royal College of

Obstetricians and Gynaecologists (RCOG) recruited 1292 women with a

history of early delivery or cervical surgery. The findings showed 25

women needed to be treated to prevent one preterm delivery with no benefit

to the neonate. However, in women with three or more second trimester

losses, a cervical suture halved the incidence of preterm delivery prior to 33

weeks gestation, and in these cases a cervical cerclage was recommended

(MRC/RCOG 1984).

Progesterone

Progesterone is a pro-pregnancy hormone and withdrawal in many animal

models, have shown to lead to an onset of labour. In humans, a functional

withdrawal is more likely but supplementation of progesterone may

nevertheless be beneficial in preventing preterm labour. Progesterone has a

negative effect on many contraction associated proteins and on

prostaglandin activity (Patel FA et al 1999). In addition, it decreases the

uterine sensitivity to oxytocin (Grazzini E et al 1998). Placebo controlled

studies using 100mg progesterone or placebo as a vaginal suppository

between 24 and 34 weeks, found a significant reduction in preterm delivery

rates at <37 weeks and <34 weeks in high risk populations (da Fonseca EB

et al 2003). A study which also examined neonatal outcomes showed a

reduction in the incidence of birth weight <2500g, necrotizing enterocolitis,

39

intraventricular haemorrhage and the need for supplemental oxygen, but not

perinatal death or respiratory distress syndrome (Meis PJ 2003). Studies of

the clinical use of progesterone are ongoing. On the available evidence, it is

estimated that one in seven high risk women need to be treated with

progesterone prophylactically to prevent one case of spontaneous preterm

birth before 34 weeks gestation (Khan KS et al 2007).

Management

The management of preterm labour may be prophylactic to prevent the onset

of preterm labour, as discussed above, or reactive once preterm labour has

started. The primary aims of such measures are for the antenatal

administration of corticosteroids to the mother and for in utero transfer from

a peripheral unit to a hospital with neonatal intensive care facilities where

the outcome for the preterm neonate can be improved significantly. It is

therefore essential that a diagnosis of preterm labour should not be

overlooked. It is usual to define the onset of labour at term as being when

regular uterine contractions lead to cervical change or dilatation, but an

initial diagnosis of preterm labour is usually based upon contractions alone.

The majority of women diagnosed as being in preterm labour based upon

contractions alone will go on to deliver at term without intervention.

Therapy can be more accurately targeted if it is limited to women who are

either fibronectin positive or have a short cervix.

40

Currently used tocolytics

Tocolytics may delay delivery for short periods of time in a minority of

women. Worldwide the recommendation for their use is varied and the

preferred pharmacological agent differs significantly. They include non

steroidal anti-inflammatory drugs (NSAIDS), β-adrenergic agonists,

oxytocin antagonists, calcium channel blockers, magnesium sulphate and

nitric oxide donors.

β-mimetics, such as Ritodrine and Salbutamol have been used widely as

tocolytics. They stimulate β2-adrenergic receptors in smooth muscle, which

act via cAMP to reduce sensitivity and absolute levels of intracellular

calcium, causing myometrial relaxation (Groom KM 2007). They may be

successful in delaying preterm delivery for a 48 hour period, but have no

significant impact on improving neonatal morbidity or mortality.

Sympathomimetics are associated with significant and severe maternal side

effects. These include chest pain, dyspnoea, tachycardia, palpitations,

tremor and hypokalaemia (Giles W et al 2007).

NSAIDS such as Indomethacin, appear to be able to delay delivery for 7-10

days and reduce rates of delivery <37 weeks gestation (Keirse M 1995).

They inhibit the cyclo-oxygenase (COX) enzyme responsible for the

conversion of arachidonic acid to prostaglandins. Their use is limited

however, by fetal side effects. Premature closure of the ductus arteriosus can

occur in up to 10-50% of fetuses exposed to indomethacin and is more

41

prevalent in later gestations (>32 weeks) and if treatment occurs for

>48hours (Macones GA et al 2001). Other concerns include risks of

necrotizing enterocolitis, intraventricular haemorrhage and effects on the

fetal kidneys (Norton ME et al 1993, Kirshon B et al 1991). Selective COX-

2 inhibitors, such as Rofecoxib have been trialed, but unsuccessfully.

Rofecoxib has reversible negative side effects to the fetus, but does not

reduce preterm delivery rates <30 weeks and surprisingly, was shown to

increase deliveries < 37 weeks gestation (Groom KM et al 2005).

Magnesium sulphate, often used in the management of pre-eclampsia, is

widely used as a tocolytic in North America. It acts as a calcium antagonist

at the neuromuscular junction (Groom KM 2007). However, randomised

placebo controlled trials of magnesium sulphate have shown no significant

short term delay of delivery, increase in birth weight or difference in

perinatal mortality when compared to placebo (Lyell DJ et al 2007). With

limited access to more efficacious tocolytics with fewer side effects, it

remains the most commonly used tocolytic in the United States.

Nitric oxide donors, such as in the use of GTN patches are popular due to

the ease of administration and low side effect profile. However, they have

yet to be proven efficacious in randomised controlled trial and fail on an

intention-to-treat basis (Duckitt K et al 2002, Bisits A et al 2004).

42

Atosiban marketed as an oxytocin antagonist, is in fact a competitive

oxytocin and vasopressin antagonist. Oxytocin receptor blockade prevents

oxytocin induced contractions as well as a reduction in extracellular calcium

influx and release of calcium from intracellular stores and therefore an

inhibition of contractility (Groom KM 2007). Administration of Atosiban

results in a dose dependent inhibition of uterine contractility and oxytocin

mediated prostaglandin release. In pregnant women, Atosiban is 46-48%

plasma protein bound and appears to have limited cross-over into the fetal

circulation.

The randomised controlled trials comparing Atosiban to beta mimetics show

similar efficacy, but a superior safety profile in terms of maternal adverse

outcomes (Marsal K 2001, Moutquin JM et al 2001). There is still some

uncertainty over the preferred use of Atosiban over Nifedipine (a calcium

channel blocker) for tocolysis. In the absence of a direct comparison

between the two tocolytics, adjusted indirect comparisons with Nifedipine

have shown to significantly reduce rates of respiratory distress syndrome

(RDS) in the neonate but do not significantly decrease time to delivery

(Coomaraswamy A et al 2003).

Nifedipine is a calcium channel blocker originally developed almost fifty

years ago as a treatment for angina pectoris. It is now used widely for the

treatment of coronary heart disease, hypertension and chronic stable angina.

The effect of calcium channel blockers in relaxing the myometrium has also

43

been known for several years. Calcium channel blockers exert their effect

by binding to L-type channels and reducing intracellular levels of calcium.

This blocks the transmembrane influx of calcium ions into muscle cells. By

binding to intracellular calcium binding proteins, Nifedipine is therefore,

able to block the flow of extracellular calcium into myometrial cells and in

this way relaxes the myometrial smooth muscle.

There have been no trials comparing Nifedipine with placebo, but it has

been compared against a beta sympathomimetic, Ritodrine in a number of

randomized controlled trials. Nifedipine is preferable as it has a better side

effect profile and demonstrated a reduction in the number of preterm births

within seven days of starting treatment (King et al 2003). The RCOG

support the use of Nifedipine for tocolysis, but it is not licensed for use in

United Kingdom and there is no standardized administration dose or regime.

Corticosteroids

Corticosteroids in the management of preterm labour can be administered as

a single course of either betamethasone or dexamethasone. If given,

between 24 and 34 weeks gestation and up to 7 days before preterm

delivery, it has a significant effect upon neonatal morbidity and mortality.

This is principally due to a significant reduction in respiratory distress

syndrome (RDS) and intraventricular haemorrhage (IVH) rates. Antenatal

corticosteroids have a receptor mediator effect on all of the components of

the surfactant system in type-2 pneumocytes. They also however have

44

effects on the structural development of the lungs, lead to accelerated

maturation of the fetal intestine, have effects upon the myocardium and on

catecholamine responsiveness. This may explain the reduced incidences of

necrotising enterocolitis and intraventricular haemorrhage seen in extremely

preterm infants that appear to be independent of the effect upon RDS

(Roberts D et al 2006). The long term consequences of recurrent exposure to

high dose steroids, suggesting adverse effects on development and

behaviour, has lead to a policy of limiting administration of antenatal

corticosteroids to a single course if possible.

Antibiotics

The administration of antibiotics to prevent preterm labour is variable in

clinical practice. There have been many studies of different antibiotics and

routes of administration, but the results vary according to the pre-existing

risk of the population. Research has focused on bacterial vaginosis (BV),

where the normal vaginal lactobacilli are replaced by a mixed flora with

high concentrations of anaerobic bacteria. Other organisms known to

increase the risk of adverse outcomes are Trichomonas vaginalis and

Mycoplasma hominis (Leitich H et al 2003). The effects of metronidazole,

erythromycin and clindomycin have been shown in some studies to improve

success in symptomatic women, women at high risk and those with a

positive fFN (Hauth JC et al 1995, Lamont RF et al 2003). However, other

studies have shown an increase in rate of delivery before 37 weeks gestation

with the use of metronidazole; perhaps, due to the eradication of normal

45

flora and hence the proliferation of harmful bacteria (Klebanoff MA et al

2001). A recent meta-analysis looking at prophylactic antibiotics for the

prevention of preterm birth in women at risk included 17 trials and found no

reduction in preterm birth irrespective of the criteria used to assess risk (i.e.

abnormal vaginal flora, previous preterm birth or positive fetal fibronectin

status). This was regardless of the antimicrobial agent used or the

gestational age at time of treatment (Simcox R et al 2007). Similarly, the

ORACLE II study showed no evidence of benefit of antibiotics in

established preterm labour. A later analysis of the longer term outcomes of

the ORACLE II babies shows that antibiotic use increases the risk of

cerebral palsy (Kenyon S et al 2008).

46

Preterm prelabour rupture of membranes

Preterm prelabour rupture of membranes (PPROM) occurs in approximately

2% of all pregnancies and accounts for up to one third of preterm deliveries.

The consequence of PPROM is as many as 50% cases deliver within a

week, 75% within two weeks and 85% within a month. As with preterm

labour, postnatal survival following PPROM is directly related to

gestational age at delivery and birth weight. The retention of amniotic fluid

within the uterus is associated with a better outcome. The management of

PPROM continues to be controversial, however many obstetricians will

favour conservative management in preterm premature rupture of

membranes before 34 weeks and would induce labour relatively early in

women whose membrane rupture occurs subsequent to 34 weeks.

Conservative management should include clinical surveillance for signs of

chorioamnionitis including regular recording of maternal temperature and

heart rate and cardiotocography. A rising white cell count or a rising CRP

level may indicate the development of chorioamnionitis, though neither of

these is highly specific. Lower genital tract swabs are routinely taken and

positive cultures for potential pathogens are useful in directing antibiotic

therapy for both the mother and the preterm neonate. The potential benefits

of tocolytic drugs do not apply in the majority of cases of PPROM and

neither do the few case reports of amnio infusion.

47

1.3 The Anatomy of the Uterus and Cervix

The uterus is the organ of gestation, receiving the fertilised egg in its cavity,

retaining and supporting it during the development of the fetus and

becoming the principal agent in its expulsion at the time of parturition. It is

situated in the pelvis between the bladder and rectum; it is retained in

position by the round and broad ligaments on each side and the cardinal and

uterosacral ligaments below (Fig 1.7). The non-gravid uterus measures

about 7cm in length, 4cm in breadth at the fundus and almost 2cm in

thickness.

The uterus consists of two parts: (1) the corpus, with its broad upper

extremity, the fundus and (2) the cervix, which is partly above the vagina

and partly in the vagina. The division between the corpus and the cervix is

indicated externally by a slight constriction, and by the reflection of the

peritoneum from the anterior surface of the uterus on to the bladder, and

internally by a narrowing of the canal, the internal os.

The uterus is a muscular organ which increases in fibrous and elastic tissue

during pregnancy, together with the stretching and hypertrophy of

myometrial cells. These smooth muscle cells form functional contractile

bundles, covered in connective tissue and interspersed with an extensive

vasculature. In addition, there is a varied population of leucocytes that

secrete cytokines, including macrophages, neutrophils, large granular

48

lymphocytes and other descendents of T-cells as reviewed by Terzidou

(2007). Single myocytes may produce force in any direction thereby

allowing the uterus to generate a multidirectional force as a whole. During

active labour, the fundal region contracts to expedite delivery, whilst the

lower segment relaxes to allow passage of the fetal head. A fundo- cervico

gradient has been described with regards to contractile receptor proteins

such as oxytocin and connexin-43 (Fuchs AR et al 1984, Sparey C et al

1999).

The cervix comprises the lower, cylindrical portion of the uterus, containing

collagen fibrils, elastic connective tissue, blood vessels and some muscle

fibres. Macrophages and neutrophils infiltrate the cervix at term inducing

cytokines and also producing enzymes capable of digesting extracellular

matrix proteins, necessary for cervical ripening.

49

Fig 1.7 Diagram demonstrating the anatomy of the non-pregnant uterus.

50

1.4 The Endocrinology and Biochemistry of Labour

There is a challenge in predicting, diagnosing and managing preterm labour,

because despite extensive research, the mechanisms that control the length

of human pregnancy and signal the onset of labour have not yet been fully

determined. Throughout pregnancy the uterine cervix needs to remain firm

and closed whilst the body of the uterus grows by hypertrophy and

hyperplasia whilst remaining quiescent. During pregnancy, ‘pro-pregnancy’

factors such as progesterone and prostacyclin inhibit myometrial

contractility. Parturition involves a synchronisation of changes in the

structure of the cervix, myometrial activity manifesting as painful uterine

contractions and ruptured fetal membranes leading to cervical dilation and

effacement. At the end of the pregnancy, ‘pro-labour’ factors begin to

mediate remodeling of the cervix and the uterus is stimulated to begin co-

coordinated contractions (Terzidou V 2007). It has been suggested that

labour results from the activation of a ‘cassette of contraction-associated

proteins’ (CAPs), which include gap junction proteins, oxytocin and

prostanoid receptors, enzymes for prostaglandin synthesis and cell signaling

proteins (Lye et al 1994). However, the term ‘labour associated proteins’

may be more appropriate since these are also responsible for activating fetal

membrane prostaglandin and cytokine production, as well as cervical

remodeling and ripening (Fig 1.8).

51

Choriodecidual bacterial colonisation

(endotoxins and exotoxins)

Fetal tissue response Maternal response

Fetus Chorioamnion and placenta Decidua

Increased Decreased chorionic prostaglandin

Corticotrophin- dehydrogenase Increased cytokines

releasing hormone and chemokines

Increased adrenal Increased prostaglandins Neutrophil infiltration

Cortisol production

Increased metalloproteases

Myometrial Chorioamnion weakening

contractions and rupture Cervical weakening

Preterm delivery

Fig 1.8 Factors influencing preterm labour.

52

Inflammation

Inflammation is now considered to be integral to the pathophysiology

behind preterm labour. Microarray and suppression subtractive

hybridisation studies have demonstrated an up regulation of a vast number

of pro-inflammatory genes in fetal membranes, myometrium and the cervix

with the onset of labour (Charpigny G et al 2003). This is demonstrated by

an influx of inflammatory cells into the uterus and increased expression of

pro-inflammatory cytokines (Young A et al 2002). The transcription factor,

nuclear factor-kappa B (NF-κB) which is activated within the uterus at the

time of labour, regulates a number of inflammatory factors, including IL-6,

IL-8 and COX-2 (which leads to prostaglandin synthesis) (Allport VC et al

2003). At term, this inflammatory-type response may be triggered by the

maturing fetal lung which produces increased surfactant lipid and protein

(i.e. surfactant protein A, SP-A). This is associated with the activation of

macrophages in the amniotic fluid and an increased inflammatory response

in the uterus (Condon JC et al 2004).

Corticotrophin-releasing hormone

Corticotrophin-releasing hormone (CRH) is a peptide produced by the

hypothalamus in response to stress. CRH is also expressed abundantly in the

placenta and in maternal serum CRH has been shown to rise in the second

half of pregnancy, peak during labour and rapidly decline postpartum. In

patients who have gone on to deliver prematurely, elevated levels of CRH in

53

the maternal circulation have been identified as early as 18 weeks of

gestation (McLean M et al 1995). The exponential curve depicting the CRH

increase is shifted to the left in women who subsequently deliver preterm

and to the right in women who deliver post dates. This suggests that CRH

production is linked to a placental clock (Fig 1.9) which determines the

length of gestation (Smith R, Van Helden D et al 2007). The implications of

this clinically could mean that maternal plasma CRH concentrations could

help indentify women at high risk of preterm delivery and CRH antagonists

may be useful in preventing preterm labour (Smith R, Nicholson RC et al

2007).

54

Maternal Stress Fetal Stress

Cortisol

Angiotensin II +

Norepinephrine membranes enhanced fetal

ADH placenta lung maturity

Acetylcholine decidua

CRH

+

PG

Contractions Rupture of Cervical change

membranes

Fig 1.9 Pathway for action of placental corticotrophin releasing hormone on

parturition.

55

Progesterone

Progesterone is thought to act during pregnancy to inhibit labour-associated

gene expression and inhibit contractions. Unlike most mammalian

counterparts, human circulating progesterone levels and progesterone

receptors (PR) in the myometrium do not fall with the onset of labour.

However, clinically the administration of a PR antagonist RU486 can be

used to ripen the cervix and induce labour. In the absence of a demonstrable

fall in progesterone concentrations, it may be possible that a ‘functional

progesterone withdrawal’ precedes the onset of labour. This could be due to

changes in the different PR levels or isoforms (Mesiano S et al 2002),

specifically a reduction in myometrial responsiveness to progesterone by an

increased expression of PR-A relative to PR-B (Pieber D et al 2001). It may

also be due to the interaction between NF-κB and PR or changes in PR co-

activator and co-repressor expression.

NF-κB

NF-κB is a cytokine-inducible transcription factor activated in response to

infection and pro-inflammatory cytokines which regulates many pro-

inflammatory and labour associated genes (Baeuerle PA et al 1996). In most

cells, NF-κB exists in an inactive form in the cytoplasm, bound to an

inhibitory protein, such as IκBα. Three distinct NF-κB signaling pathways

have been identified to date (Terzidou V 2007). Labour is associated with an

increase in the baseline level of NF-κB DNA binding and transcriptional

56

activity in the human amnion and because NF-κB down regulates

progesterone receptor function, it seems to be related to a ‘functional

progesterone withdrawal’ in the human myometrium (Chapman NR et al

2004).

Prostaglandins and cyclo-oxygenases

It is now established that prostaglandins (PGs) play a central and essential

role in the biochemistry of human labour

(Albertsen K et al 2004).

Prostaglandins of the E and F series stimulate uterine contractions

throughout pregnancy and are used clinically to induce first and second

trimester abortion and labour at term. There is increased synthesis of PGs by

intrauterine tissues, particularly the fetal membranes, at the time of labour

which is associated with a rise in the concentrations of PGs and PG

metabolites in amniotic fluid and maternal plasma (Slater DM et al 1995,

1999).

Prostaglandins belong to the eicosanoid family of lipid mediators and are

derived from arachidonic acid (AA) found in membrane phospholipids of

the cell. The prostaglandin biosynthetic pathway (Fig 1.10) can be divided

into three main stages. The initial step involves the mobilization of AA from

membrane phospholipids by the action of phopholipase enzymes. In the

second stage, the free AA is converted to an unstable prostaglandin

57

intermediate, in a two-step reaction catalysed by the cyclo-oxygenase

(COX) enzymes. This reaction involves the conversion of AA to

hydroperoxy endoperoxide (PGG2), which in turn undergoes reduction to

form hydroxyl endoperoxide (PGH2). Since no pathway exists for the re-

conversion PGH2 to AA, the COX enzymes mediate the committing step of

prostaglandin biosynthesis. Finally, the PGH2 intermediate is converted to

the terminal prostaglandins PGE2, PGF2α, PGD2, prostacyclin (PGI2) and

thromboxane by their respective synthase enzymes (Kniss DA 1999).

In addition to mediating the committing step, the COX enzymes are also

thought to catalyse the principle rate-limiting step of prostaglandin

biosynthesis (Slater DM et al 1995). The COX enzymes exist as two main

isoforms: the constitutively expressed COX-1 and the inducible COX-2

isoenzyme. In human labour, there is substantial evidence for the greater

importance of the COX-2 enzyme rather than COX-1. Amnion is a major

source of prostaglandins and displays a marked increase in the synthesis of

PGE2 with labour onset and stretch. This is associated with the selective

induction of COX-2 expression (Slater DM et al 1999). Similarly, a labour

associated increase in COX activity in the chorion is attributed to the

upregulation of COX-2, but not COX-1 expression. Upregulation of COX-2

expression in lower segment myometrium has been shown to occur with

advancing gestational age and was maximal just prior to labour at term,

suggesting a role for myometrial COX-2 in the early phases of the

58

parturition cascade (Romero R et al 1988).

Arachidonic Acid

Epoxygenases Lipoxygenases

Cyclogenase reaction COX-1

COX-2

HETES Epoxides

Leukotrienes

PGG2

Peroxidase reaction COX-1

COX-2

PGI synthase PGH2 TXA synthase

PGI2 TXA2

TXB2

PGD PGF PGF

synthase synthase isomerase

PGD2 PGF2α PGE2

Fig 1.10 Biosynthetic pathway for the formation of eicosanoids derived from

arachidonic acid.

59

In relation to the role of prostaglandins in preterm labour, they are products

of the inflammatory process and also play a key role in the mechanisms of

infection associated premature labour. Infection increases prostaglandin

production by amnion, chorion and decidua through the activity of bacterial

products, cytokines, growth factors and other inflammatory mediators

(Gomez R et al 1995, Kelly RW 2002). Concentrations of PGE2 and PGF2α

in amniotic fluid are significantly higher in women with preterm labour who

have confirmed chorioamnionitis than women with preterm labour and no

evidence of infection (Romero R et al 1988). In addition, higher levels of

prostaglandins are produced from the amnion of patients with histologic

chorioamnionitis compared to amnion from patients without documented

chorioamnionitis (Lopez Bernal A et al 1987).

Interleukin 8

IL-8 acts a chemokine and is produced by human endometrium, chorio-

decidua and myometrium. IL-8 plays an important role in cervical ripening,

probably aids in the formation of the lower segment of the uterus in late

pregnancy and mediates the influx of inflammatory mediators, specifically

leukocytes, seen in the myometrium during labour (Mohan AR et al 2004).

60

Oxytocin

Oxytocin (OT) is a nonapeptide produced by the posterior pituitary, known

to have a potent contractile activity on the pregnant uterus. However, it does

not seem to be essential for normal parturition as there is no increase in

circulating oxytocin in the mother or fetus with the onset or during labour.

Moreover, normal parturition has been observed in cases of clinical pituitary

gland dysfunction (Phelan JP et al 1978). However, the expression of the

oxytocin receptor (OTR) does increase in the pregnant uterus. An up-

regulation by two orders of magnitude has been demonstrated leading to a

strong increase in sensitivity towards OT (Soloff MS 1975). Comparable

increases in OTR mRNA concentrations in the myometrium are associated

with this up-regulation of OT- binding sites (Jeng YL et al 2003). Post

delivery, myometrial OT-binding sites rapidly decrease, whereas OTR

expression remains high in the mammary gland throughout lactation (Soloff

MS et al 1979).

In many animals, OTR is regulated by changes in steroid hormone

concentrations (Soloff MS 1975). In humans, the lack of marked changes in

oestrogen concentrations or progesterone withdrawal suggests that OTR has

a more complex regulation (Gimpl G et al 2001). Further relationships in the

OTR system have been demonstrated, in particular with prostaglandins.

Treatment with naproxen, a prostaglandin-synthesis inhibitor, resulted in a

dramatic decrease in OTR in the rat uterus at term. Administration of PGF2α

61

restored the increase in OTR, implying the possibility of a positive feedback

loop between prostaglandins and OTR (Chan WY 1980).

62

1.5 Myometrial Contractility

Irrespective of the aetiology of preterm labour, the underlying mechanism

by which the myometrium contracts is thought to be the same. During

pregnancy, the uterus maintains a quiescent environment, whereas at the

point of fetal maturity, it generates forceful contractions, observed to

originate from the fundus to expel the conceptus, whilst the lower segment

remains relaxed in order to allow the fetal head to pass.

The basic functional unit of the myometrium is the smooth muscle cell,

which is in turn, arranged into bundles of approximately 300+/- 100µm in

diameter. Each bundle is surrounded by connective tissue interspersed with

microvasculature. The bundles are further organized into fasiculi, which are

covered by a dense collagen matrix and the major vasculature of the

myometrium (Young RC et al 1999). This therefore enables the uterus to

generate an expulsive force dependent on the orientation of these fibres

relative to the cervix. Uterine contractions are then initiated by the

propagation of action potentials within these defined vectors of muscle

bundles (Blanks AM et al 2007). The intracellular basis for contraction is via

the cyclical attachment and detachment of cross-bridges between myosin

and actin filaments, triggered via activation of the regulatory 20kDa chain

of myosin molecules. Myosin light chain is activated by phosphorylation by

myosin light chain kinase, which in turn, is activated by calmodulin (Word

RA et al 1993). It is through the actions of calmodulin, that during an action

63

potential, the calcium influx can be connected with a contraction; a process

known as excitation-contraction coupling (Blanks AM et al 2007). Once the

uterus has been activated, the myometrium is susceptible to stimulation by a

number of different agonists, primarily prostaglandins and oxytocin.

Myometrial cells maintain ionic gradients across the plasma membrane and

with the change of membrane permeability to calcium, sodium, chloride and

potassium, generate an action potential. In human myometrium, the

depolarising current is mainly mediated by calcium (via L-type calcium

channels) and repolarisation is mainly carried by potassium (Shmygol A et

al 2007). The process of contractility in myometrial smooth muscle is ATP-

dependent and is proportional to the activity ratio of myosin light chain

kinase (MLCK) to myosin phosphotase (MLCP). It is by altering the

activity ratio of MLCK to MLCP that oxytocin and PGF2α increase the

sensitivity of the myometrium to calcium (Woodcock NA et al 2004, 2006).

This sensitisation occurs after contraction has been initiated and is mediated

by a decrease in MLCP activity, which in turn is controlled by several

processes such as the activation of G-protein rhoA or the activation of CPI-

17 by protein kinase C (Kitazawa T et al 2000).

In order to propagate action potentials, myometrial cells are connected by

connexins (gap junction proteins) through which ions and some cellular

metabolites can pass. In the myometrium, an increase in the number of gap

junctions has been observed in both term and preterm labour, as well as

64

improved intracellular conductance and metabolic communication (Garfield

RE et al 1988, Sims SM et al 1982). Connexin-43 (CX-43) appears to be the

most important in gap junction formation as it is present in all gap junction

plaques and a delay in parturition has been observed in CX-43 knockout

mice (Kilarski WM et al 2001, Döring B et al 2006). Furthermore, in human

myometrium, the expression of CX-43 was found to be the greatest at the

fundus implicating a stronger contractile potential to the upper segment

(Sparey C et al 1999).

Physiological stimulation once myometrial activation has occurred is

principally via prostaglandins and oxytocin. Oxytocin increases contraction

frequency, duration and force by increasing the frequency of depolarisation,

sensitising the contractile machinery to calcium and prolonging the duration

of the action potential plateau (Nakao K et al 1997, Woodcock NA et al

2004). G-protein coupled oxytocin receptors release G-protein subunits

(from Gαi) which leads to the activation of phospholipase C, which is

responsible for the mobilisation of calcium from sarcoplasmic reticular

stores via inositol 1, 4, 5,-triphosphate (InsP3). This is sufficient to open L-

type calcium channels and allow full depolarisation (Phaneuf S et al 1996).

65

Prostaglandin receptors

Prostaglandins were originally thought to interact directly with the cell

membrane, thereby altering its characteristics and modulating cell

physiology. However, their different activity profiles and generation of

second messengers, principally cAMP, phosphatidylinositol turnover and

Ca2+

, suggests that they interact with discrete receptors and that different

receptors exist for each PG. Coleman et al (1994) used specific antagonists

to further characterize these receptors pharmacologically. They classified

the receptors as one each for PGs, D2, F2α, I2 and TXA2, as DP, FP, IP and

TP and four for PGE2 as EP1-4. The first receptor that was isolated and

cloned was the TP receptor. This is a G protein linked member of the

rhodopsin family of receptors with seven transmembrane domains.

Subsequently, homolgy screening using mouse cDNA libraries led to the

identification of the structures of all eight receptors, each encoded by a

different gene (Narumiya S et al 1999).

More recently, several splice variants have been identified for human EP3

(eight isoforms), sheep and bovine FP (two isoforms), rat EP1 (two

isoforms) and human TP (two isoforms) (Pierce KL et al 1998). They vary

largely in their cytoplasmic C-termonal regions where coupling with G

protein occurs. The amino acid sequences have been fully described for rat

and sheep. There is only 20% homology between amino acid sequences of

various receptors within a species, but 79-89% homology exists between

like receptors of different species (Olson DM 2003).

66

The EP receptor family is composed of four distinct subtypes; EP2 and EP4

stimulate adenylate cyclase and cAMP production via the stimulatory

subunit Gs, resulting in muscle relaxation (Fig. 1.11), whereas EP1 and EP3

mediate muscle contraction through increased phosphinositol turnover and

calcium mobilisation, or decrease adenylate cyclase and cAMP levels,

respectively (Coleman RA et al 1994).

67

Figure 1.11 Signaling pathways for EP1 and EP2/4. EP1 is coupled to Gq and

upon PGE2 binding there is a signal cascade that results in activation of

phospholipase C (PLC) and mobilisation of intracellular calcium (Ca2+

) stores. EP2

and EP4 are coupled with Gs. PGE2 binding results in activation of adenylyl

cyclase (AC) and cyclic adenomonophosphate (cAMP). This leads to activation of

protein kinase A (PKA) and subsequent phosphorylation (P) of CREB, which leads

to up-regulation of the cyclic AMP response element (CRE) (Kandola M 2008).

68

The EP3 receptor is unique in that eight different receptor isoforms have

been identified, encoded by nine transcripts: EP3-I(1a), EP3-I(1b), EP3-II,

EP3-III, EP3-IV, EP3-V, EP3-VI, EP3-c and EP3-f generated by alternative

splicing at the carboxy terminal tail (Schmid et al 1995, Kotani et al 1997,

2000). As the carboxy terminal end of the receptor is important for

mediating G protein coupling, changes in this region appear to alter the

coupling affinity. EP3 is mainly coupled to Gi and hence towards a

contractile phenotype. However, splice variants EP3-II and EP3-IV may

couple Gs and/or Gq leading to different cellular responses (Namba et al

1993, Kotani et al 1995) (Fig. 1.12).

69

Figure 1.12. Signaling pathways for EP3 isoforms. A. EP3 is coupled to Gq and

upon PGE2 binding there is a signal cascade that results in activation of

phospholipase C (PLC) and mobilisation of intracellular calcium (Ca2+

) stores. B.

EP3 is coupled to Gi. PGE2 binding causes inhibition of adenylyl cyclase (AC) and

therefore mobilisation of intracellular Ca2+

. C. EP3 is coupled with Gs. PGE2

binding results in activation of AC and cyclic adenomonophosphate (cAMP). This

leads to activation of protein kinase A (PKA) and subsequent phosphorylation (P)

of CREB, which leads to up-regulation of the cyclic AMP response element (CRE)

(Kandola M 2008).

70

PGF2α similar to PGE2 is a potent stimulator of myometrial tissue

(Parkington HC et al 1999), but there is no consensus on how. The FP

receptor is coupled to Gαq class of G-proteins and therefore stimulation of

PLC should release InsP3 (Carrasco MP et al 1996). However this has not

been consistently demonstrated in freshly isolated myometrial strips from

pregnant patients (Schery MP et al 1988). What has been demonstrated

however is the calcium sensitising effect of PGF2α is rho-kinase mediated.

This is preceded by depolarisation and results in a large increase in

intracellular calcium (Woodcock NA et al 2006). This effect may aid

delivery by producing a strong contraction, whilst increasing the recovery

period between contractions to reduce the risk of hypoxia and fetal distress

(Parkington HC et al 1999).

There is little consistency in the literature concerning the myometrial

expression of the two ‘contractile’ receptors EP1 or EP3 (Fig 1.13). Astle et

al (2007) found that EP1 expression is upregulated in human lower segment

myometrium at the time of labour but found no change in upper segment

myometrium. Conversely, Smith GC et al (2001) found that EP1 expression

was greatest in baboon upper segment myometrium and found no changes

with labour. Grigsby PL et al (2006) found that human myometrial EP1

expression increases with gestational age in both upper and lower segment

but found no regional differences. Ma X et al (1999) found no expression of

EP1 at all in sheep myometrium.

71

With respect to EP3, Astle et al (2007) found greater expression in upper

segment, but mostly in vascular smooth muscle cells not myocytes.

Moreover, down regulation of EP3 with labour was observed. Smith GC et

al (2001) found EP3 expression to be greater in upper segment myometrium

with no labour associated changes. Grigsby PL et al (2006) found no

regional differences and no changes in association with labour. Leonhardt

A et al (2003) found that there was no expression of EP3 in human pregnant

myometrium but did find expression of EP1 and so concluded that PGE

leads to contractions via EP1. Ma X et al (1999) found a labour associated

increase in EP3 expression in sheep myometrium.

72

Fig 1.13 Summary of the literature findings with regards to myometrial expression

and changes of the ‘contractile’ PG receptors, EP-1 and EP-3 with regards to their

presence in the upper segment (US) and lower segment (LS) and changes with

gestation and labour.

Author Myometrial

sample

EP

receptor

Presence

in US

Presence

in LS

Changes

with

gestation

Changes

with

labour

Astle Human 1 + + ↑ LS

3 ++ + ↓

Smith Baboon 1 ++ + Nil

3 ++ + Nil

Grigsby Sheep 1 + + ↑

3 + + Nil

Ma Sheep 1 - -

3 + + ↑

Leonhardt Human 1 + +

3 - -

73

1.6 Hypothesis and Objectives

Currently, there are no effective tocolytics free of side-effects. PGE2

receptors represent potential targets, but which EP should be targeted would

depend upon expression and functional data. EP1 and EP3 are said to

mediate contractions, but there is no consensus about their relative

importance. There is little consistency in the data regarding EP expression

in myometrium during pregnancy. To examine the roles of EP1 and EP3 in

causing myometrial contractility, we undertook expression studies and

contractility studies. Mandeep Kandola, a scientist in our group, examined

the expression of EP receptors by Western blotting, immunofluoresence and

QRT-PCR in the upper and lower segment myometrial samples that I had

obtained from the pregnant, non-labouring women, at term, at elective

Caesarean sections and from labouring women at emergency Caesarean

sections.

Expression of EP-R protein on the cell surface

In order to study the changes that occur between quiescence and active

labour, as defined by contractions, immunohistochemistry combined with

immunofluorescence is often used to decipher the expression of proteins at a

particular time point. Immunohistochemistry refers to the process of

detecting antigens (e.g. proteins) in cells of a tissue section by exploiting the

principle of antibodies binding specifically to antigens in biological tissues.

Immunofluroscence uses the specificity of antibodies to their antigen to

74

target fluorescent dyes to specific biomolecule targets within a cell, and

therefore allows visualisation of the distribution of the target molecule

through the sample. The advantage of this method is that it is quantitative as

opposed to qualitative. Therefore, the higher the immunofluroscence seen,

the more receptor is present. To quantitatively assess the presence of

proteins, Western blotting is also used.

Using these two techniques we were able to assess PGE2 receptor protein

expression in the upper and lower segment myometrium. In the lower

segment, EP1 expression in the lower segment (LS) was less than that in the

upper segment (US). There were no significant differences between labour

plus (L+) and labour minus (L-) samples (Fig. 1.14). Similarly, there were

no significant changes observed in EP3 expression in both US and LS, as

well as L+ and L- myometrial biopsies (Fig. 1.15).

75

Figure 1.14. No labour associated change of EP1 receptor

protein expression in human myometrium at term. EP1 receptor

expression was compared in matched lower segment (LS) and

upper (US) segments of term laboring (L+) and term non-labouring

(L-) myometrium. Whole cell extracts were subjected to Western

blotting and blots were scanned for densitometric analysis. Values

were normalised using the housekeeping gene GAPDH. Results

are the mean of 3-6 samples and not significant (p>0.05).

Representative Western blots are shown.

76

Figure 1.15. No labour associated change of EP3 receptor

protein expression in human myometrium at term. EP3 receptor

expression was compared in matched lower (LS) and upper (US)

segments of term laboring (L+) and term non-labouring (L-)

myometrium. Whole cell extracts were subjected to Western blotting

and blots were scanned for densitometric analysis. Values were

normalised using the housekeeping gene GAPDH. Results are the

mean of 3-6 samples and the differences were not significant

(p>0.05). Representative western blots are shown.

77

Both the EP1 and EP3 receptors were demonstrated in all the sample groups

using immunofluorescence. Co-localisation with the smooth muscle alpha-

actin stain confirmed their presence in the smooth muscle. In the upper

segment, EP1 receptor protein was not significantly affected by labour

status in the US. There was however, significant change in the labouring

lower segment versus the non-labour samples (p=0.05) (Fig.1.16). Similarly

the EP3 receptor increased significantly in the labouring LS myometrium

(p=0.02) (Fig. 1.17).

78

Figure 1.16. EP1 receptors are significantly upregulated in the lower

segment of the labouring myometrium. EP1 receptor expression was

measured in matched lower (LS) and upper (US) segments of term labouring

(L+) and non-labouring (L-) myometrium. Tissue was paraffin-embedded and (A)

the specificity of the antibody determined using immunohistochemistry with

polyclonal EP1 antibodies. Replacement of anti-EP1 antibodies with pre-immune

IgG and saturation of anti-EP1 antibodies with blocking peptides were used as

negative controls. (B) Immunofluorescent analysis for EP1 was visualized with

red-fluorescent Alexa Fluor 594 dye and α-actin was visualized with green

fluorescent Alexa Fluor 488 dye. Nuclei were stained with DAPI. (C) Intensity of

receptor expression was analysed by measuring levels of fluorescence and

applying statistical analyses. These graphs represent the intensity of expression

of each PGE2 receptor in the cytosol of myocytes (n=3).

79

Figure 1.17. EP3 receptors are significantly upregulated in the lower

segment of the labouring myometrium. EP3 receptor expression was

measured in matched lower (LS) and upper (US) segments of term labouring

(L+) and non-labouring (L-) myometrium. Tissue was paraffin-embedded and (A)

the specificity of the antibody determined using immunohistochemistry with

polyclonal EP3 antibodies. Replacement of anti-EP3 antibodies with pre-immune

IgG and saturation of anti-EP3 antibodies with blocking peptides were used as

negative controls. (B) Immunofluorescent analysis for EP3 was visualized with

red-fluorescent Alexa Fluor 594 dye and α-actin was visualized with green

fluorescent Alexa Fluor 488 dye. Nuclei were stained with DAPI. (C) Intensity of

receptor expression was analysed by measuring levels of fluorescence and

applying statistical analyses. These graphs represent the intensity of expression

of each PGE2 receptor in the cytosol of myocytes (n=3).

80

A fundo- cervico gradient has been described with regards to contractile

receptor proteins such as oxytocin and connexin-43 (Fuchs AR et al 1984,

Sparey C et al 1999). However, there is no clear consensus between EP1

and EP3 localisation in labouring and non-labouring myometrium, as

illustrated by Fig.1.13.

In summary, the expression data showed there to be no significant changes

between the localisation in upper or lower segment biopsies using both

RNA and protein samples. However, there was a labour associated change

seen with EP1 and EP3 in the lower segment using immunofluorescence.

Furthermore, there were changes in the intracellular localisation of the

receptors; EP1 showed a significant increase in the nuclei of labouring

upper and lower segment myometrial biopsies. EP3, however, was primarily

in the cytoplasm. To demonstrate the significance of this, I established an in

vitro tissue bath to conduct functional studies examining contractility in

upper and lower segment myometrial biopsies.

81

2.0 Materials and Methods

The research was conducted at the Institute for Reproductive and

Developmental Biology, Imperial College at the Hammersmith Hospital

site. Collection of tissue was approved by the local Ethics Committee and

written informed consent was given by all subjects. Experiments were

performed on human myometrial biopsies taken at elective caesarean

section at term. All caesarean sections were conducted at Queen Charlotte’s

Hospital and samples stored in at least 10mls of Eagle’s medium. In each

case, a single lower segment biopsy of about 2 x 2 x 1cm was taken from

the upper margin of the lower segment incision (Fig 2.1). In the laboratory,

this was divided into 8 strips approximately 1 x 0.5 x 0.5cm (Fig 2.2) and

mounted within the 8 individual baths in the myograph.

82

Fig 2.1 Myometrial sample taken from the lower segment of human

myometrium at elective caesarean sections at term. Only women having

elective caesarean sections for breech or a previous caesarean section were

consented. Following the delivery of the fetus and placenta, myometrial

tissue would be cut from the upper edge of the uterine scar. This did not

cause additional bleeding or add significant time to the operation.

83

In order to sample upper segment myometrial tissue, four biopsies

measuring 0.5 x 0.5 x 0.5cm (Fig 2.2), were taken by inserting a

cervical punch biopsy (Fig 2.3) through the lower segment incision

and obtaining tissue from the fundus. Limitations to this methodology

include a degree of damage to the upper segment myometrium by the

cervical punch biopsy apparatus as well as some endometrial lining

being included in the biopsy. Pre-labour samples were taken at

elective caesarean section for indications such as breech presentation,

previous caesarean section or maternal request. In-labour samples

were taken from women with spontaneous onset of labour who

reached at least 4cm cervical dilatation who had not received either

prostaglandin or oxytocin. All samples were used within three hours

of collection.

84

Fig 2.2 Strips of myometrial tissue; lower segment (on the left) divided

from a larger sample, as illustrated in Fig 2.1, and upper segment (on

the right), obtained using the cervical punch biopsy.

85

Fig 2.3 A cervical punch biopsy used for obtaining upper segment myometrium at

caesarean sections at term.

86

2.1 Contractility Studies

Contractility studies were undertaken using a Myograph 800MS (Danish

Myo Technology) (Fig.2.4), which can be used to study small myometrial

samples of about 0.5cm in length. The Myograph 800MS with smaller organ

baths (capacity 4mls) was ideal for the smaller upper segment biopsies.

However, smaller baths are also more sensitive to changes in temperature,

pH, volume and oxygenation. For example, concentrations of substrates

that were added to the bath had to be diluted into smaller volumes so as to

not upset the bath by huge changes in total volume.

Myometrial samples were mounted between metal clasps attached to

isotonic tension transducers in individual organ baths filled with 4ml of

modified Krebs solution; the concentrate was kindly provided by Glaxo

Smith Kline, Harlow, and made up in the lab to the following specifications:

NaCl 119mmol/l, CaCl2 16mmol/l, NaHCO3 25mmol/l, KCl 14.7mmol/l,

KH2PO3 1.18mmol/l, MgSO4 1.17mmol/l, Glucose 5.5mmol/l to a pH 7.4 at

a temperature of 37oC. This was then perfused with a 95% O2/ 5% CO2

ratio gas mixture. Transducers were calibrated once a month by balancing

known 2 g weights on inverted scales that applied force onto the

transducers. This was used as a reference in the Chart software version 5.0

and later upgraded to 5.2 (PowerLab, ADI Instruments).

87

Fig 2.4 The Myograph 800MS with smaller organ baths (capacity 4mls) which was ideal for the smaller upper segment biopsies. The muscle strip was placed between two clamps and dissection forceps were used to attach the strips in order to reduce handling of the myometrium and minimise damage. The two clamps, when pulled apart, recorded the tension produced by the myometrial strip through a transducer. This was connected directly to the computer software which processed the data.

88

Preliminary experiments showed that to measure both spontaneous and

stimulated contractions, myometrial samples needed to be placed under 4g

of tension (Fig 2.5). The choice of setting the initial tension to 4g was based

on the tension that the most number of strips responded to (i.e. 80%). As the

experiment progressed, the tension would settle to 2g. This is consistent

with current published data. A baseline period of one hour was observed and

experiments were only conducted on strips that contracted spontaneously

within 90 min of the set-up and had achieved a minimum peak tension of 1g

(Fig 2.6). The average life span for the tissue strip was for a total of 90

minutes following the initial 60 minutes of incubation. Most experiments

were conducted on spontaneous contractions, unless otherwise stated. The

contractions had been stable for a minimum of 15 minutes before additional

agents were added in 15 minute intervals. All the data obtained; rate per

contraction, peak tension per contraction, duration per contraction, work

done per contraction and total work done was recorded using the Chart

software and processed by a PC based Excel program.

89

Tension (g) No. of strips contracting at 1hr (%)

0 0

2 25

4 80

6 50

8 40

Fig 2.5 Preliminary experiments showed that stretch is essential for

spontaneous contractility in myometrial strips. When using the Myograph

800MS, myometrial samples needed to be under a minimum of 4g

tension to achieve the maximum of strips to contract.

90

(g)

(mins)

60 70 80 90

2

4

6

8

10

12

Fig 2.6 A real time image of spontaneous contractions resulting from the

stretch of a non-labouring, lower segment myometrial strip after an hour.

Co

ntr

acti

on

forc

e (g

)

Time (s)

91

2.2 Drug Assessment

The first series of experiments compared upper segment with lower segment

myometrium. The initial step was to compare spontaneous contractions, in

both the upper and lower segment. The next experiment explored the effect

of increasing concentrations of PGE2 on spontaneous contractions.

Cumulative concentrations of PGE2 (10-10

, 10-9

, 10-8

, 10-7

, 10-6

mol/l) were

added to the organ baths at 15 minute intervals. The effect of increasing the

concentration of PGE2 on myometrial contractility was assessed by

calculating an average of the preceding 15 minute interval as a ratio of the

average in the first 15 minutes of spontaneous contractions.

A novel OT antagonist, GW796679X with a molecular weight (MW) of 484

and a dissociation constant (Ki) of 1.5nM was provided by Glaxo Smith

Kline, and compared against equimolar concentrations of a known OT

antagonist, Atosiban. Cumulative concentrations of OT (10-10

,10-9

,10-8

,10-7

,

10-6

mol/l Syntocinon) were then added to the organ baths at 15 minute

intervals. Similar to the first design and in following experiments, the effect

of the antagonists and/or agonists on myometrial contractility was assessed

by calculating an average of the preceding 15 minute interval as a ratio of

the average in the first 15 minutes of spontaneous contractions.

Other compounds that were tested include acetyl-salicylic acid (ASA)

(Sigma-Aldrich, Poole, UK), PGF2α, a novel FP antagonist (FPA) donated

92

by Serono and two novel EP1 and EP3 antagonists donated by Glaxo Smith

Kline (GSK). FPA has a Ki of 6nM for the prostaglandin F receptor (FP)

and is 10-100 folds selective for FP compared with other prostanoid

receptors.

The EP1 (ZD6416) and EP3 (L798106) antagonists are both small

molecules. ZD6416 shows fpKi for EP1 8.4 +/-0.6 and for EP3 6.3 +/-0.4.

L798106 shows fpKi for EP1 6.0+/-0.4 and for EP3 7.9 +/-0.4. Schild

analysis has shown that each is a competitive antagonist devoid of agonist

activity up to 10µM. The effective concentrations of each against their

respective receptors are therefore 2 orders of magnitude apart. A

concentration of 100nM of each antagonist was calculated to ensure

complete and specific inhibition of the relevant receptor.

93

2.3 Data Analysis

Data was calculated by taking the average of each 15 minute interval and

expressing it as a ratio of the average in the first 15 minutes of spontaneous

contractions. In the case, where there were no spontaneous contractions

after the baseline one hour, due to the pre-incubation of an antagonist, the

data was not normalized and will be stated as such.

The first parameter analysed was rate per contraction. This looked at the

number of contractions that occurred within the set time period in seconds

(in the majority of cases, 900s = 15 minutes). The second parameter was the

peak tension, i.e. the maximum tension achieved by each strip in grams. The

third parameter was duration per contraction. This illustrated how long a

contraction lasted in seconds. The final parameters were the average work

done by each contraction and the total work done by all the contractions

within a set time frame. This was a calculation of the area under the curve

[height (gram) x width (second) x ½]. Statistical significance was

determined by ANOVA, where the p value was less than 0.05 (Fig 2.7).

94

Duration per contraction (s)

Peak tension (g)

Work done per contraction (g.s)

Rate per contraction = Number of contractions within a time framePeak tension (g) = Maximum tension achieved by the contractionDuration per contraction (s) = Time take per contraction from start to finishWork done per contraction (g x s) = Energy generated by each contraction (area under the curve)Total work done = Energy generated by all the contractions within the specified time frame (900s)

Fig 2.7 An illustration of the parameters used to calculate the results: rate

per contraction, peak tension per contraction, duration per contraction, work

done per contraction and total work done.

95

3.1 Contractility in upper versus lower segment myometrium

Two series of experiments were undertaken to compare contractility

between upper and lower segment myometrium. In the first series matched

(i.e. from the same patient) upper and lower segment biopsies from term

non-laboured myometrium were established in the myograph and the

absolute contractility parameters were compared (Fig 3.1). There were no

differences between lower and upper segment samples in the duration taken

for contractions to begin, or in any of the contraction parameters: rate per

contraction, peak tension per contraction, the duration per contraction, work

done per contraction and therefore in the total work done (p=0.149) at 90

minutes.

96

A B

C D

E

Fig 3.1 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from upper and lower segment, non-labouring myometrium at term over 90

mins. No significant difference in any of the parameters was demonstrated

between matched upper and lower segment human myometrium (n=8) as

demonstrated by the total work done at 90 mins (p=0.149).

97

3.2 Contractility in prostaglandin stimulated upper and lower

segment myometrium

In the second series of experiments comparing upper and lower segment

myometrium, PGE2 was added at concentrations of 10-10

, 10-9

, 10-8

, 10-7

and

10-6

at 15 minute intervals, after the baseline period of spontaneous

contractions had been established (Fig 3.2). The effect of PGE2 upon

contractility was expressed as a ratio to spontaneous contracting

myometrium.

PGE2 caused a dose dependent increase in contractility in both upper and

lower segment myometrium at concentrations above 10-8

(p=0.042) versus

spontaneously contracting myometrium. However, there was no difference

in the total work done between PGE2 stimulated upper and lower segment

myometrium (p=0.177).

98

A B

C D

E

Fig 3.2 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from upper and lower segment, non-labouring myometrium at term with

increasing concentrations of PGE2 added over 90 mins. No significant difference in

any of the parameters was demonstrated between matched upper and lower

segment human myometrium (n=6) as demonstrated by the total work done at

concentrations of 10-6

(p=0.177). However, PGE2 caused a dose dependent

increase in contractility in both upper and lower segment myometrium at

concentrations above 10-8

(p=0.042) versus spontaneously contracting

myometrium.

99

3.3 Discussion of contractility in upper and lower segment

myometrium

The body of the uterus comprises of two parts, the uterine corpus (upper

segment) and the isthmus (lower segment). A concept of functional

differences has been proposed both physiologically and at a protein

expression level. Various studies have shown differences in mRNA

expression between upper and lower segment, leading to criticism that

experiments on the lower segment may not truly represent the uterine body.

A fundal to cervical gradient has been described with regards to

prostaglandin receptors by Wikland et al (1983), to oxytocin receptors by

Fuchs et al (1984) and splicing factors SF2/ ASF by Pollard et al (2000).

This study was initiated in the first instance to identify, if any, a difference

between upper and lower segment, term, myometrial tissue. This study and

that of Lucas et al (2000) are the only studies to have directly compared

upper and lower segment myometrium from the same patient. Some studies

have suggested that PGE2 might contract the upper but have no effect, or

relax the lower segment (Gordon-Wright et al 1980). It is possible that

differences between studies may be due to the differences in the site from

which the myometrial biopsy is taken. If a lower segment myometrial

sample is taken from very close to the cervix then there may be more

cervical than myometrial smooth muscle cells present and the effect of

PGE2 may be different.

100

The first set of data demonstrated the effect of PGE2 on both upper and

lower segment myometrium, and agrees with similar data from Lucas at el

(2000). There were no differences in spontaneous contractility between

upper and lower segment myometrium (p=0.149). Furthermore, PGE2

consistently caused contractions of upper and lower segment tissue and

again there was no difference between the corpus and isthmus (p=0.177). In

addition, our own expression studies of EP1 and EP3 expression also show

no difference in upper versus lower segment myometrium. Recent evidence

using trans-abdominal magneto-myographic recording has demonstrated the

synchronisation of electrical burst activities, which are essential for

effective uterine contractions, are identified not only at the fundus but over

the entire uterus as well (Eswaran et al 2005). Furthermore, no specific pace

maker activity seems to have been identified over any specific area of the

uterus. This would correlate with anatomical findings by Lucas et al (2000)

and Word et al (1993) that demonstrate similar proportions of smooth

muscle in the uterine muscle in both upper (65.4%) segment and lower

(69.8%) segment myometrium. Given the ease to obtain lower segment

myometrium and my findings, subsequent experiments used only lower

segment myometrium.

101

4.1 The effect of PGE2 on spontaneously contracting lower

segment myometrium.

Fig 4.1 illustrates the effect of PGE2 on spontaneously contracting human

lower segment uterine tissue taken in real time from the myograph.

pge2+ep3,1

Channel 3 (

g)

4

6

8

10

12

14

16

pge-1

0

pge-9

pge-8

pge-7

pge-6

1:06:40 1:15:00 1:23:20 1:31:40 1:40:00 1:48:20 1:56:40 2:05:00 2:13:20 2:21:40 2:30:001 2 3 4 5

03/11/2005 4:16:33.906

PGE-10 PGE-8 PGE-7 PGE-6PGE-9

(g)

Time (s)

Fig 4.1 A pictorial representation of increasing concentrations of PGE2 on an

isolated myometrial strip taken from the lower segment of non-labouring

myometrium at 15 min intervals.

Increasing concentrations of PGE2 over time (s)

Contr

action f

orc

e (

g)

102

PGE2 added at concentrations of 10-10

, 10-9

, 10-8

, 10-7

, 10-6

at 15 minute

intervals, after the baseline period of spontaneous contractions had been

established caused a dose dependent increase in contractility in

concentrations above 10-8

versus spontaneously contracting myometrium

(p=0.042). The effect of PGE2 upon contractility was expressed as a ratio to

spontaneous contractility in the preceding 15 minutes and then as a

percentage of maximally stimulated contractility (Fig 4.2).

103

A B

C D

E

Fig 4.2 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from lower segment, non-labouring myometrium at term with increasing

concentrations of PGE2 added over 75 mins. PGE2 caused a dose dependent

increase in contractility in lower segment myometrium at concentrations over 10-8

(p=0.042) versus spontaneously contracting myometrium as demonstrated by the

total work done (n=7).

104

4.2 The effect of PGF2α on spontaneously contracting lower

segment myometrium.

Fig 4.3 illustrates the effect of PGF2α on spontaneously contracting human

lower segment uterine tissue taken in real time from the myograph.

pgf2

Channel 1 (

g)

0

5

10

15

20

25

pgf-

10

pgf-

9

pgf-

8

pgf-

7

pgf-

6

1:06:40 1:15:00 1:23:20 1:31:40 1:40:00 1:48:20 1:56:40 2:05:00 2:13:20 2:21:40 2:30:00 2:38:201 2 3 4 5

02/11/2005 14:01:12.003

PGF-10 PGF-9 PGF-8 PGF-7 PGF-6

(g)

Time (s)

Fig 4.3 A pictorial representation of increasing concentrations of PGF2α on

an isolated myometrial strip taken from the lower segment of non-labouring

myometrium at 15 min intervals.

Increasing concentrations of PGF2 α over time (s)

Contr

action f

orc

e (

g)

105

There was a statistically significant increase in the total work done when

PGF2α at a concentration of 10-7

was added to spontaneously contracting

lower segment myometrium (p=0.014), (Fig 4.4). It appears to be less potent

than PGE2 as a higher concentration of PGF2α is required to significantly

increase the total work done by spontaneously contracting myometrium.

106

A B

C D

E

Fig 4.4 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from lower segment, non-labouring myometrium at term with increasing

concentrations of PGF2α added over 75 mins. PGF2α caused a dose dependent

increase in contractility in lower segment myometrium at concentrations over 10-7

(p=0.014) versus spontaneously contracting myometrium as demonstrated by the

total work done (n=10).

107

4.3 The effect of acetyl salicylic acid (ASA) on spontaneously

contracting lower segment myometrium and PGE2 induced

contractions

Pre-incubation of lower segment human myometrial strips with ASA, an

inhibitor of both COX-1 and COX-2, for one hour prior to the experiment

led to a 3-fold reduction in peak tension and therefore overall work done,

from 15 minutes into the experiment (t15) (p=0.022). Due to the design of

the experiment (i.e. pre-incubation of the myometrium), the data could not

be normalised and is represented as such (Fig 4.5).

108

A B

C D

E

Fig 4.5 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from lower segment, non-labouring myometrium at term with a pre-incubation

of 1μl of ASA for 60 mins prior to the experiment commencing (n= 10). This led to a

3-fold reduction in peak tension and therefore a significant reduction in the total

work done from 15 minutes into the experiment (p=0.022).

109

In the next experiment, PGE2 was added in increasing concentrations from

10-10

to 10-6

to myometrium pre-treated with ASA, and the effects were

compared with spontaneously contracting myometrium and ASA pre-treated

myometrium (Fig 4.6). The addition of PGE2 returned peak tension and

overall work done per contraction to control values.

As above, inhibition of contractility was again demonstrated by ASA, as the

total work done by the controls was significantly greater (p=0.007). The

total work done by the ASA pre-treated strips that had the addition of PGE2,

was significantly greater than the strips with ASA alone (p=0.001) but not

the controls (p=0.311).

110

A B

C D

E

Fig 4.6 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from lower segment, non-labouring myometrium at term with increasing

concentrations of PGE2 added to pre-incubated strips of myometrium of ASA for an

hour, against strips pre-incubated with ASA alone and against controls (n=7).

There was statistical significance in the reduction of total work done by the strips

with ASA alone in comparison to the controls (p=0.007) and those strips that had

increasing concentrations of PGE2 (p=0.001). There was no statistical difference

between the controls and the ASA pre-incubated strips with the addition of PGE2

(p=0.311).

111

4.4 Discussion of the effects of PGE2, PGF2α and ASA on

contractility.

The second set of experiments quantified the effect of known agonists,

PGE2 and PGF2α. There was a statistically significant increase in the total

work done by the myometrial strips following the addition of either agent.

This was observed at concentrations above 10-8

(p=0.042) and 10-7

(p=0.014) for PGE2 and PGF2α respectively. This confirms results of

previous studies which demonstrated that PGE2 binds to myometrium 8-10

fold more than PGF2α (Hoffman HJ et al 1984). Furthermore, Giannopoulos

G et al (1985) have demonstrated that PGE2 binds to its myometrial receptor

with a higher affinity than PGF2α, and Martin RL et al (1995) have shown

PGE2 to be 8-10 fold more potent than PGF2α in stimulating myometrial

contractions.

Incubation of strips with acetyl salicylic acid (ASA) an inhibitor of both

COX-1 and COX-2 for one hour prior to the experiment led to a 3-fold

reduction in peak tension and therefore overall work done, from 15 minutes

into the experiment (p=0.022).

Furthermore, the addition of PGE2 (from concentrations of 10-10

) to pre-

incubated myometrium containing ASA, returned peak tension and overall

work done per contraction to normal values. The total work done by the

ASA pre-treated strips that had the addition of PGE2, was significantly

112

greater than the strips with ASA alone (p=0.01) but not the controls

(p=0.311). ASA (as in the previous experiment) again demonstrated

inhibition of contractility against spontaneously contracting myometrium

(p=0.07). These data suggest that stretch and synthesis of prostaglandins is

essential for spontaneous contractility in human myometrial strips and that

PGE2 is more important in stimulating contractions than PGF2α.

Having established that the synthesis of PGs is key for spontaneous

contractility, further experiments were conducted to discover a method of

blocking the effects. An established approach has been to block the pathway

in the synthesis of PGs, essentially by blocking the enzyme pathways (i.e.

COX), but NSAIDs (e.g. Rofecoxib, Indomethacin) are associated with

harmful side effects for both mother and fetus (Groom KM et al 2005,

Sawdy RJ, Lye S et al 2003). An alternative approach would be to allow the

production of PGs, but to specifically block the receptors instead.

113

5.1 The effect of an EP1 antagonist, ZD6416 upon spontaneous

and PGE2 stimulated contractions.

Three series of experiments were undertaken to assess the effects of the EP1

antagonist ZD6416 upon spontaneous and PGE2 stimulated contractions. In

the first series, ZD6416 was added at increasing concentrations of 10, 100

and 1000nM (i.e. the optimal concentration and one order of magnitude

either side) to spontaneously contracting myometrial strips at 15 minute

intervals (Fig 5.1). The EP1 antagonist had no effect upon any individual

aspect of contractility or upon total work done at any of the above

concentrations (p=0.117, p=0.103, p=0119) respectively.

114

A B

C D

E

Fig 5.1 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from lower segment, non-labouring myometrium at term with increasing

concentrations of ZD6416 added over 45 mins (n=6). No significant difference in

any of the parameters was demonstrated as illustrated by the total work done at

the following concentrations, 10nM, 100nM or 1000nM (p=0.117, p=0.103, p=0119)

respectively.

115

In the second series of experiments, ZD6416 at 100nM was added to the

Kreb’s solution at the start of the experiment (Fig 5.2). Pre-incubation with

ZD6416 at 100nM for an hour did not inhibit any aspect of contractility nor

the total work done (p=0.104). Due to the nature of the experiment (i.e. pre-

incubation of the myometrium), the data could not be normalised and is

represented as such.

116

A B

C D

E

Fig 5.2 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from lower segment, non-labouring myometrium at term with a pre-incubation

of 100nM of ZD6416 for 60 mins prior to the experiment commencing (n= 14). Pre-

incubation with ZD6416 did not inhibit any aspect of contractility including the total

work done up to 45 mins following the pre-incubation period (p=0.104).

117

In the third series of experiments involving ZD6416, 100nM was added to

the Kreb’s solution after the baseline period of spontaneous contractions had

been established. Following which, at fifteen minute intervals, increasing

concentrations of PGE2 was added at concentrations from 10-10

to 10-6

(Fig

5.3).

As seen previously PGE2 caused a dose dependent increase in contractility

at concentrations above 10-8

(p=0.048), but there was no inhibitory effect of

ZD6416 at this concentration of PGE2 (p=0.899).

118

A B

C D

E

Fig 5.3 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from lower segment, non-labouring myometrium at term with increasing

concentrations of PGE2 added over 75 mins to spontaneously contracting

myometrial strips and to strips pre-incubated with a 100nM of ZD6416 (n=7). PGE2

caused a dose dependent increase in contractility in lower segment myometrium at

concentrations over 10-8

(p=0.048) versus spontaneously contracting myometrium

as demonstrated by the total work done. There was no inhibitory effect of ZD6416

at this concentration of PGE2 (p=0.899).

119

5.2 The effect of an EP3 antagonist, L798106 upon spontaneous

and PGE2 stimulated contractions.

An identical series of experiments were undertaken to assess the effects of

the EP3 antagonist L798106 upon spontaneous and PGE2 stimulated

contractions. In the first series, L798106 was added at increasing

concentrations at 10nM, 100nM and 1000nM to spontaneously contracting

myometrial strips at 15 minute intervals (Fig 5.4). A significant decrease in

contractility at concentrations of 100nM (p=0.029) and 1000nM (p=0.020)

was demonstrated.

120

A B

C D

E

Fig 5.4 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from lower segment, non-labouring myometrium at term with increasing

concentrations of L798106 added over 45 mins (n=7). A significant decrease in the

total work done was demonstrated at concentrations of 100nM (p=0.029) and

1000nM (p=0.020) respectively.

121

In the second series (Fig 5.5), pre-incubation of the lower segment

myometrium with L798106 at 100nM caused a significant inhibition, but

not a complete abolition, of spontaneous contractility (p=0.040) from thirty

minutes into the onset of spontaneous contractions (t30).

122

A B

C D

E

Fig 5.5 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from lower segment, non-labouring myometrium at term with a pre-incubation

of 100nM of L798106 for 60 mins prior to the experiment commencing (n= 10). Pre-

incubation with L798106 caused a significant inhibition, but not a complete

abolition, of spontaneous contractility from thirty minutes into the onset of

spontaneous contractions (p=0.040).

123

In the third series of experiments involving L798106, 100nM of the EP3

antagonist was added to the Kreb’s solution after the baseline period of

spontaneous contractions had been established. Following which, at fifteen

minute intervals, increasing concentrations of PGE2 was added at

concentrations from 10-10

to 10-6

(Fig 5.6).

At a concentration on 100nM, L798106 completely inhibited the dose

dependent increase in contractility caused by PGE2 from 10-9

(p=0.028).

124

A B

C D

E

Fig 5.6 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D) work done

per contraction and (E) total work done in isolated myometrial strips taken from lower

segment, non-labouring myometrium at term with increasing concentrations of PGE2 added

over 75 mins to spontaneously contracting myometrial strips and to strips pre-incubated

with a 100nM of L798106 (n=8). As previously demonstrated, PGE2 caused a dose

dependent increase in contractility in lower segment myometrium at concentrations over

10-8

. The myometrial strips pre-incubated with L798106 significantly reduced the increase

in total work done caused by PGE2 from concentrations of 10-9

(p=0.028).

125

5.3 Discussion of the role of EP1 and EP3 in human lower

segment myometrial contractility in relation to their respective

receptor antagonists.

There are four PGE2 receptor subtypes (EP1, EP2, EP3, and EP4), each

encoded by a separate gene (Astle et al 2005). EP1 and EP3 are excitatory

(EP1 is coupled to Gαq and EP3 to Gαi) whereas EP2 and EP4 relax

myometrial smooth muscle (both couple to Gαs) (Bastien L et al 1994). In

the contractility experiments that I conducted, I concentrated on two novel

EP1 and EP3 receptor antagonists, ZD6416 and L798106 respectively.

Three series of experiments were undertaken to assess the effects of EP1

antagonist ZD6416 upon spontaneous and PGE2 stimulated contractions. In

the first series, ZD6416 added at increasing concentrations of 10, 100 and

1000nM (i.e. the optimal concentration and one order of magnitude either

side) to human, lower segment myometrial strips at fifteen minute intervals.

The EP1 antagonist had no effect upon any individual aspect of contractility

or upon total work done at any of the above concentrations (p=0.117,

p=0.103, p=0119 respectively).

In the second series of experiments involving ZD6416, 100nM was added to

the Kreb’s solution at the start of the experiment. Pre-incubation of the

myometrial strips with ZD6416 at 100nM did not inhibit any aspect of

contractility including the total work done (p=0.104).

126

In the third series of experiments, 100nM of ZD6416 was added to the

Kreb’s solution following the establishment of spontaneous contractions.

Increasing concentrations of PGE2 was added from 10-10

to 10-6

at 15 minute

intervals. As seen previously PGE2 caused a dose dependent increase in

contractility at concentrations above 10-8

, but there was no inhibitory effect

of ZD6416 at this concentration of PGE2 (p=0.899).

An identical series of experiments were undertaken to assess the effects of

the EP3 antagonist L798106 on spontaneous and PGE2 stimulated

contractions. L798106 was added at increasing concentrations of 10, 100

and 1000nM (i.e. the optimal concentration and one order of magnitude

either side) to spontaneously contracting myometrial strips at fifteen minute

intervals. Unlike the EP1 antagonist, a significant decrease in contractility

was demonstrated at concentrations of 100nM and 1000nM (p=0.029,

p=0.020) respectively.

In the second series of experiments involving L798106, pre-incubation of

the myometrium at 100nM caused a significant inhibition, but not complete

abolition, of spontaneous contractility (p=0.040), thirty minutes from the

onset of spontaneous contractions.

In the third series of experiments involving the EP3 antagonist, following

the onset of spontaneous contractions, increasing concentrations of PGE2,

were added to myometrium pre-incubated with a 100nM of L798106. The

127

EP3 antagonist successfully inhibited the dose dependent increase in

contractility caused by PGE2 from concentrations of 10-9

(p=0.028).

Two conclusions can therefore, be drawn from the preceding. Firstly, as the

reduction in contractility caused by ASA (COX inhibitor) was more

significant than that caused by either of the PGE2 receptor antagonists (EP1

or EP3), it is likely that PGE2 plays a vital but not exclusive role in

myometrial contractility.

Secondly, given that L798106 demonstrated a significant reduction in

myometrial contractility in comparison to ZD6416 (which demonstrated

none); it is likely that the stimulating effect of PGE2 acts predominantly via

the EP3 receptor. This provides evidence that the antagonism of the EP3

receptor (rather than the EP1 receptor) may prove a suitable strategy in

pharmacological tocolysis.

The contractility data correlates with our expression studies on EP1 and EP3

receptors in non-labouring, term, lower segment myometrium

(unpublished). It was demonstrated that EP1 becomes more nuclear

localised in association with labour. This data is also consistent with that of

Grigsby et al (2006), who found no regional differences in expression of

EP1 at term or changes in association with labour and also found that

expression of EP1 was localized to the nucleus in association with labour.

Astle et al (2007) also showed that EP1 is nuclear localized, but they

128

identified a significant increase in EP1 expression at mRNA level in the

lower segment in association with labour. However this increase, which was

not demonstrated at a protein level was only of the order of 25% and no

statistical correction was made for multiple testing. There is considerable

inter-subject variation in expression levels of EP and we have concluded

that, taken together, the available data suggests that there are no regional

differences or significant changes in EP1 expression with the onset of

labour, but that at the time of labour the receptor is localised to the nucleus.

This suggests that the function of EP1 at term may be less to mediate

intercellular signaling and more to mediate the nuclear effects of

endogenous, intracellular PGE2. EP1 is therefore unlikely to be the principal

receptor which mediates PGE2 stimulated myometrial contractions.

Furthermore, at a protein level it was demonstrated that EP3 expression

remains cytosolic in myometrium with the onset of labour, and like Grigsby

et al (2006), but unlike Astle et al (2007) or Leonhardt et al (2003) we found

significant expression in myocytes. We have concluded that the available

data suggests that there are no regional differences or significant changes in

EP3 expression with the onset of labour, but that EP3 is probably principally

contractile in its action and remains cytosolic or on the cell membrane. EP3

is therefore able to be liganded by extracellular, exogenous PGE2 and is

likely to be the receptor which mediates PGE2 stimulated myometrial

contractions.

129

6.1 The effect of FPA, a PGF2α antagonist upon spontaneous

and PGF2α stimulated contractions.

Increasing concentrations (100-1000nM) of a novel and selective FP

antagonist (FPA) were added to isolated, lower segment myometrial strips.

In comparison to the non-treated controls, FPA did not reduce the total work

done by spontaneous contractions, even at the highest concentration of

1000nM (10-6

) (p=0.50).

130

A B

C D

E

Fig 6.1 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from lower segment, non-labouring myometrium at term with increasing

concentrations of FPA added over 45 mins (n=6). No significant difference in any of

the parameters was demonstrated as illustrated by the total work done even at the

highest concentration of 1000nM (10-6

) (p=0.5).

131

In the next experiment, 10-8

PGF2 was added to spontaneously contracting

strips of lower segment myometrium, following which increasing

concentrations of FPA (10-8

to 10-5

) were added at 15 minute intervals. The

total work done by the myometrial strips which had the cumulative addition

of FPA is decreased five-fold (Fig 6.2). This is in comparison to the PGF2

pre-treated strips alone (p=0.044).

132

A B

C D

E

Fig 6.2 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from lower segment, non-labouring myometrium at term with increasing

concentrations of FPA (10-8

to 10-5

) added over 45 mins following an initial addition

of 10-8

PGF2 (n=12). Increasing concentrations of FPA, significantly decreased the

total work done by strips treated with PGF2 in comparison the PGF2 - treated

strips alone and at concentrations of 10-5

, total work done was reduced up to five-

fold (p=0.044).

133

6.2 Discussion of the effect of FPA, a novel PGF2α antagonist on

contractility.

The inhibition of FP has been proposed as a strategy to prevent preterm

delivery. FP antagonists have been shown to inhibit contractions in isolated

sheep and human myometrium (Olson DM et al 2003) and to open BKCa

channels (Doheny HC et al 2007). FP antagonists also delay RU486 or LPS

induced labour in mice, rat and sheep (Olson DM et al 2003, Chollet A et al

2007). Therefore, I attempted to demonstrate the inhibition of human

myometrial contractility with a novel and selective FP antagonist (FPA).

Pre-incubation of the baths with concentrations up to 1uM (10-6

) did not

affect the total work done by spontaneous contractions in comparison to the

non-treated controls (p=0.50).

However, in myometrium pre-treated with a concentration of 10-8

of PGF2α,

increasing concentrations of the FPA (from 10-7

) decreased the contractility

by five-fold. This was in comparison to PGF2α pre- treated myometrial

strips alone (p=0.044).

This demonstrates that FPA is able to block FP receptors, but does not

inhibit spontaneous contractions. It is likely therefore, that PGF2α has no

role in spontaneous myometrial contractility. This is in keeping with data

previously published; i.e. PGE2 binds to myometrium 8-10 times more than

PGF2α (Hoffman HJ et al 1984), PGE2 is 8-10 times more potent than PGF2α

in stimulating myometrial contractions (Martin et al 1995) and PGE2 binds

134

to its myometrial receptor with higher affinity than does PGF2α to its

receptor (Giannopoulos G et al 1985).

135

7.1 The effect of combining the agonists, OT and PGE2 on

spontaneously contracting lower segment myometrium.

Having successfully demonstrated the effects of the EP1 and EP3

antagonist, GSK were eager to investigate the effects of a novel OT

antagonist, GW796679X in combination with the aforementioned EP3

antagonist, L798106 on spontaneous contractions. The first step was to

investigate the effect of combining OT and PGE2 on spontaneously

contracting lower segment myometrium.

To determine if OT and PGE2 combined produce a synergistic effect, first a

concentration of 10-8

OT was added to spontaneously contracting

myometrium. At fifteen minute intervals, increasing concentrations of

PGE2 from 10-10

to 10-7

were added to the organ baths. These strips

containing both agonists were compared against myometrium with either

agonist alone or spontaneously contracting controls (Fig 7.1).

OT alone at a concentration of 10-8

increased contractility significantly in

comparison to spontaneously contracting controls (p=0.001). A similar

effect was demonstrated by PGE2 alone at a concentration of 10-9

(p=0.03).

Furthermore, in combination, the PGE2 and OT significantly increased the

total work done by the lower segment myometrium (p=0.035).

136

However, the combination of the two agonists caused a significant increase

in total work done in comparison to the myometrium with the OT alone

(p=0.041) but not by the PGE2 alone (p=0.32).

137

A B

C D

E

Fig 7.1 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from lower segment, non-labouring myometrium at term with increasing

concentrations of PGE2 added over 60 mins following an initial addition of 10-8

of

OT; compared against PGE2 alone, OT alone and spontaneously contracting

myometrium (n=8). Both the agonists individually, 10-8

OT (p=0.001), 10-9

PGE2

(p=0.03) and in combination (p= 0.035) significantly increased the total work done.

Furthermore, the combined effect of the agonists on total work done was

statistically greater than the strips with OT alone (p=0.041) but not PGE2 alone

(p=0.32).

138

7.2 The effect of combining the antagonists of OT and PGE2 on

spontaneously contracting lower segment myometrium.

To determine if there is a synergistic effect when combining antagonists,

equimolar concentrations the OT antagonist, GW796679X and the EP3

antagonist L798106, lower segment myometrial strips were pre-incubated

with either the EP3 antagonist alone, the OT antagonist alone or a bath with

both the OT and the EP3 antagonists in combination against control strips of

spontaneously contracting myometrium.

There was a statistically significant inhibition of contractility in the

myometrial strips with L798106 alone at 30mins (p=0.029), but no

inhibition by GW796679X alone (p=0.547), nor by a combination of both

the antagonists (p=0.683).

A synergistic effect of combining both an EP3 and an OT antagonist was

not demonstrated in this experiment.

139

A B

C D

E

Fig 7.2 (A) Contraction rate, (B) peak tension, (C) duration per contraction, (D)

work done per contraction and (E) total work done in isolated myometrial strips

taken from lower segment, non-labouring myometrium at term pre-incubated with

equimolar concentrations of L798106, GW796679X and of the two antagonists

together for a period of 60 mins (n=10). There was a statistically significant

reduction in total work done by L798106 alone at 30mins (p=0.029), but no

inhibition by GW796679X alone (p=0.547), nor a combination of the two

antagonists (p=0.683).

140

7.3 Discussing the effect of combining the agonists, OT and

PGE2 and their respective antagonists on spontaneously

contracting lower segment myometrium.

Oxytocin is known to have a potent contractile activity on the pregnant

uterus. In many animal models, OTR is regulated by changes in steroid

hormone concentrations (Soloff MS 1975). In humans, the lack of marked

changes in oestrogen concentrations or progesterone withdrawal suggests

that OTR has a more complex regulation (Gimpl G et al 2001). Further

relationships in the OTR system have been demonstrated, in particular with

prostaglandins. Treatment with Naproxen, a prostaglandin-synthesis

inhibitor, resulted in a dramatic decrease in OTR in the rat uterus at term.

Administration of PGF2α restored the increase in OTR, implying the

possibility of a positive feed- back loop between prostaglandins and OTR

(Chan WY 1980).

In the final series of experiments, 10-8

OT increased total work done by

sponataneously contracting lower segment myometrium significantly

(p=0.001) as did 10-9

PGE2 (p=0.03). Unsurprisingly, a combination of OT

and PGE2 demonstrated an increase in total work done on spontaneous

contractions (p= 0.035). This combination of OT and PGE2 also increased

the total work done significantly over myometrial strips with 10-8

OT alone

(p= 0.041). This effect however, was not demonstrated in relation to

myometrial strips with PGE2 alone (p=0.32). This experiment therefore,

demonstrates that whilst PGE2 is a potent stimulator of spontaneous

141

contractions (more so than OT), no synergistic effect between PGE2 and OT

was observed.

The next step was to determine if a synergistic effect could be demonstrated

between an OT antagonist and a PGE2 antagonist. Control strips of lower

segment myometrium were set up in parallel with strips pre-incubated with

either the EP3 antagonist alone, the OT antagonist alone and a bath with

equimolar concentrations of both the OT antagonist and the EP3 antagonists

in combination. There was a statistically significant inhibition of the

myometrial strips by the EP3 antagonist alone at 30mins (p=0.029), but no

inhibition by the OT antagonist alone (p=0.547), nor the combination of

both antagonists (p=0.683). A synergistic effect of combining both an EP3

and an OT antagonist was not demonstrated in this experiment.

Given that 1) PGE2 is more powerful than OT in stimulating contractions

and 2) the OT antagonist is not effective as the EP3 antagonist in inhibitng

spontaneous contractions, these set of experiments support the argument in

favour of a new tocolytic.

142

8.1 Discussion

The primary function of the uterus during gestation is to harbour the

growing fetus in a largely quiescent environment. Upon maturation of the

fetus, the uterus must remodel itself sufficiently to generate forceful

contractions during labour. During preterm delivery, the process of

remodelling of the myometrium occurs early due to a number of causes,

although the underlying basis of myometrial contraction remains the same.

What is still unknown is to what degree various signalling factors and

hormones affect myometrial contractility and certainly myometrial

contractility in preterm labour. Due to the multi-factorial origin of preterm

labour, treatment options generally target the inhibition of uterine

contractility, rather than the underlying mechanisms of labour.

Tocolytics, used in Europe for the last 30 years, are the backbone of

treatment for the prevention of preterm labour. Tocolytic drugs are indicated

to prolong pregnancy in women with acute risk of preterm birth; in some

cases they are also administered for prophylaxis. The main rationale for use

of these drugs is to delay delivery for at least 48 hours in order to allow

treatment effect of corticosteroids or transfer of the pregnant woman to a

specialised high-risk obstetric unit. Both actions have been show to

markedly reduce neonatal morbidity and mortality (Husslein P et al 2003).

As most randomised trials of tocolytic agents lack power to provide definite

answers, and use delay in delivery as a surrogate endpoint for implied

143

improvements in neonatal outcome related to the administration of steroids,

systemic reviews and meta-analyses have been conducted to more

adequately study drug efficacy and safety. Several Cochrane reviews have

been published that compare different tocolytic agents to each other or to

placebo for women in preterm labour (King JF et al 2003, 2005,

Anotayanonth S et al 2004, Whitworth M et al 2008).

Many trials included maintenance treatment if and after contractions

stopped. Some trials excluded women with ruptured membranes, but in

others they were included. The most frequently evaluated agent was

Ritodrine, which has predominantly β2- receptor effects; relaxing muscles

in the uterus, arterioles and bronchi. Other tocolytics evaluated in these

trials included Isoxuprine, Terbutaline, Magnesium sulphate, Indomethacin,

Nifedipine and Atosiban. Overall tocolytics were associated with a

reduction in the odds of delivery within 24 hours (OR 0.47; 95% CI 0.29-

0.77), 48 hours (OR 0.57; 95% CI 0.38-0.83) and seven days (OR 0.60; 95%

CI 0.38-0.95). For beta agonists, Indomethacin, Nifedipine and Atosiban

these effects were statistically significant, but not for Magnesium sulphate.

However, there was no statistically significant reduction in births before 30

weeks (OR 1.33; 95% CI 0.53-3.33), before 32 weeks (OR 0.81; 95% CI

0.61-1.07) or 37 weeks of gestation (OR 0.17; 95% CI 0.02-1.62) (RCOG

2006). The choice of tocolytic agent depends mainly on the availability

following regulatory approval in the country, the drug efficacy and

144

effectiveness, the fetal and maternal safety profiles and the cost of

treatment.

Beta-mimetics have until the recent past, been the most commonly used

tocolytic agent. It is certainly the most thoroughly evaluated, but as their

mechanism of action is organ non-specific, they notoriously lead to

numerous adverse events in multiple organ systems, which are largely

cardiovascular in origin. Common adverse effects when beta-agonists are

compared with no treatment or placebo include: palpitations (48% beta-

agonists vs. 5% control), tremor (39% vs. 4%), nausea (20% vs. 12%),

headache (23% vs. 6%) and chest pain (10% vs. 1%) (Gyetvai K et al 2009).

Rare but serious and potentially life threatening adverse effects have been

reported following the use of beta-mimetics and there have been a small

number of maternal deaths associated with the use of these drugs.

Pulmonary oedema is a well documented complication, often associated

with aggressive intravenous hydration.

For the other tocolytic drugs, fewer types of adverse effects were reported,

and they occurred less frequently (Gyetvai K et al 2009). With Magnesium

sulphate, 7% discontinued treatment compared with none on placebo and

none of the women allocated Indomethacin discontinued treatment

compared to the control. For Atosiban, reported side effects were nausea

(11% Atosiban vs. 5% placebo), vomiting (3% vs. 4%), headache (5% vs.

7%), chest pain (1% vs. 4%) and dyspnoea (0.4% vs. 3%).

145

Systemic reviews of Nifedipine versus beta-mimetics have come to similar

conclusions (Oei SG et al 1999, Tsatsaris V et al 2001). In both, there has

been insufficient evidence for any conclusion, in particular with regards to

the effect on neonatal mortality. Nifedipine, is however, associated with a

greater chance of delaying labour for more than 48 hours (RR 1.13; 95% CI

1.01-1.267), a lower risk of respiratory distress syndrome and admission to

the neonatal unit. In addition, there were fewer maternal adverse effects

with Nifedipine in comparison to Ritodrine (16% vs. 45%). Nifedipine has

the advantage of oral administration and is economically sounder.

Trials comparing Indomethacin with beta-agonists have shown not much

difference in delay to delivery, but the former have fewer maternal side-

effects. However, the risk of premature closure of the ductus, renal and

cerebral vasoconstriction and necrotising enterocolitis associated with a

high dose and prolonged exposure of Indomethacin has led to concern and

reluctance of its use in the clinical arena (Eronen M et al 1991, Norton ME

et al 1993).

Atosiban has been compared with at least three different beta-agonists;

Ritodrine, Salbutamol and Terbutaline. The delay in delivery was not

statistically significant at 48 hours (88% Atosiban vs. 89% beta-mimetics)

(RR 0.99%; 95% CI 0.94-1.04) or at 7 days (80% vs. 77%) (RR 1.03%;

95% CI 0.95 – 1.11) (The Worldwide Atosiban vs. Beta-agonists Study

Group 2001). The side effect profile of Atosiban is significantly better than

146

beta-agonists, including chest pain (1% Atosiban vs. 5% beta-agonists),

palpitations (2% vs. 16%), tachycardia (6% vs. 76%), hypotension (3% vs.

6%), dyspnoea (0.3% vs. 7%), nausea (12% vs. 16%), vomiting (7% vs.

22%) and headache (10% vs. 19%). The financial implications of using

Atosiban are significantly greater however (£498.30 Atosiban vs. £0.15

Nifedipine), assuming complete 48 hour treatment regardless of in-utero

transfer (Siassakos D et al 2009).

Therefore, if a decision is made to use a tocolytic agent, Ritodrine is no

longer seems to be the drug of choice, given the most serious side effect is

pulmonary oedema (Tan TC et al 2006). Atosiban does not affect the

maternal heart rate or blood pressure (Goodwin TM et al 1994) and can be

used safely in women with cardiac disease and diabetic women. When

compared with placebo in a randomised trial (Romero R et al 2000),

Atosiban resulted in lower birth weight and an increase in infant deaths. In

this trial, there was an imbalance in randomisation that resulted in more

women under 26 weeks gestation and fewer over 32 weeks assigned to the

Atosiban group. Considering the significantly higher morbidity and

mortality of babies born under 26 weeks, compared with those born over 26

or 32 weeks, this imbalance can be argued to invalidate the findings of the

study (Word RA et al 2003, 2005). Other randomised trials without such

problems have not replicated these findings (Valenzuela GJ et al 2000).

With regards to the long term follow-up of infants of mothers who received

147

Atosiban, it has been reported that developmental outcomes at 6, 12 and 24

months are the same when compared to placebo (Papatsonis DN et al 2005).

Nifedipine, unlike Atosiban, is not licensed for use as a tocolytic in the

United Kingdom but given its relatively safe side-effect profile, its similar

efficacy to Atosiban in delaying labour and definite cost effectiveness, it

used in most obstetric units around the country. However, Nifedipine is

contraindicated in women with cardiac disease and should be used in

caution in women with diabetes. Cardiac failure has been reported with the

use of Nifedipine (van Geijn HP et al 2005, Tan TC 2006) and in cases of

maternal hypertension, where Nifedipine has been administered, there has

been an association with a redistribution of fetal blood flow (Vojcek L et al

1988). Brown MA et al (2002) have concluded that signs of fetal distress

during fetal heart rate monitoring under these circumstances, have increased

the rate of caesarean sections It is unknown whether these estimates can be

applied to Nifedipine when used for tocolysis.

In summary, it seems further evidence is required regarding the effects of

both Nifedipine and Atosiban on more substantive outcomes such as

neonatal mortality and morbidity, and on safety and long-term outcome for

the child. It is not therefore unreasonable to consider the need for a new

tocolytic.

148

Prostaglandins have long been postulated to play a role in the parturition

process, mainly as their synthesis is dramatically increased in association

with labour (Challis JR et al 2002, Olsen DM et al 2003). PGE2 in

particular, is produced in large quantities by fetal membranes and possible

roles include cervical ripening and the stimulation of uterine contractions

(Challis JR et al 2000). Therefore, these desired effects lend them to use in

obstetric practice.

PGE2 exhibits a wide spectrum of physiological actions depending on the

distribution and subtypes of EP receptors present (Breyer RM et al 2001).

All four PGE2 receptors EP1-4 are present in human myometrial tissue

during pregnancy, with distinct cellular patterns of expression being

observed. Expression of EP1-4 receptor proteins are localised to vascular

smooth muscle, vascular endothelium, glandular epithelium and decidual

components of myometrial tissue. However, there is no consensus on

whether there is a definite up-regulation of excitatory EP1 and EP3 or a

down-regulation of EP2 or EP4 in labour.

Functionally, EP1 and EP3 couple to G proteins, Gqα or Giα, and promote

calcium influx or the inhibition of adenylate cylase respectively (Breyer RM

et al 2001, Bos CL et al 2004). The stimulation of EP1 and EP3 would,

therefore, result in the contraction of smooth muscle or an increase in

inflammatory outputs. By contrast, EP2 and EP4 couple to Gsα to increase

adenylyl cyclase activity and intracellular cAMP levels, leading to smooth

149

muscle relaxation and repressed expression of many inflammatory cytokines

(Regan JW 2003).

The only consistency in the literature is that both contractile and relaxatory

EP receptors are simultaneously expressed in myometrial tissue. Thus,

pharmacological studies using isolated myometrial strips and EP receptor

agonists and antagonists is a more systematic approach to an understanding

of the mechanisms by which PGE2 exerts its effects. This may lead to the

pharmacological isolation of pro-labour actions of PGE2 from pro-

pregnancy actions and therefore to the improved management of obstetric

patients from current management.

In summary, our data demonstrates that the stretch of term human

myometrium results in spontaneous contractions. The addition of ASA (a

non-selective COX inhibitor) did not completely stop spontaneous

contractility but reduced the total work done by up to five-fold. This

indicates that prostaglandins play a significant role but not exclusive role in

spontaneous contractility. The PGF2α antagonist did not have an inhibitory

effect on spontaneous contractions, but only those contractions induced by

the addition of PGF2α prior to the experiment. The EP1 antagonist similar to

the PGF2α antagonist did not prove to be efficacious in inhibiting

spontaneous contractions. The only significant reduction in total work done

on spontaneously contracting lower segment myometrium was by the EP3

antagonist.

150

Schild analysis has shown that each is a competitive antagonist devoid of

agonist activity up to 10µM. The effective concentrations of each antagonist

against their respective receptors were two orders of magnitude apart. A

concentration of 100nM of each antagonist was calculated to ensure

complete and specific inhibition of the relevant receptor. The EP3 receptor

antagonist proved to cause significant reduction in spontaneous and PGE2

induced contractility in term, non-labouring myometrium. The EP1 receptor

antagonist, however, was not effective. These results therefore support the

use of the EP3 antagonist in further clinical trials with regards to a new

tocolytic compound.

The advantages of targeting the EP3 antagonist alone, centre in that it would

be more specific than knocking out PGE2 in general, and therefore

potentially have fewer side effects. EP3 has been described in the literature

as having other specific roles in the body. Inflammatory breast cancer (IBC)

is the most aggressive subtype of locally advances breast cancer with the

worst prognosis and shortest overall survival of any variant of this type of

disease (Dawood S et al 2008). Current observations indicate that COX-2,

EP3 and EP4 are all upregulated in IBC cells. Robertson FM et al (2010)

have suggested that agents directly or indirectly targeting the signalling

pathways of PGE2, EP3 or EP4 receptors may be important in inhibiting the

aggressive phenotype of IBC.

151

However, in the gastrointestinal tract, PGE2 appears to play an important

role in the modulating the mucosal integrity and various functions of the

alimentary tract. In the duodenum, acid perfusion produces damage in

animals lacking EP3 receptors (Takeuchi K 2010). As with all receptors in

the body, there are very few that are exclusive to one organ system in the

body. Ultimately one would have to work out a concentration of EP3

antagonist to effectively knock off myometrial contractility whilst avoiding

significant adverse effects globally in the body.

Whilst there has been considerable literature concerning EP receptor

function, expression and regulation in myometrium, there is little

information on key issues such as cellular distribution of EP receptor

expression and regulation in the uterine cervix. Several lines of evidence

support a pivotal role for PGE2 in the regulation of the cervical ripening

process. Firstly, PGE2 has been used clinically for at least 25 years to induce

softening of the cervix successfully at term human pregnancy (Hughes EG

et al 2001). Furthermore, COX-2 expression and PGE2 synthesis has been

found to increase at the time of parturition in the rat and sheep cervix

(Mitchell MD et al 1978, Dong YL et al 1996). In addition, PGE2 increases

human cervical collagenolytic activity and GAG synthesis in the rat cervix

(Cabrol D et al 1987), induces vasodilatation of human cervical arteries

(Allen J et al 1988) and thus promotes subsequent oedema and leukocyte

infiltration (Kelly RW 2002).

152

Schmitz T et al (2006) demonstrated the presence of all four EP receptor

proteins in all tissue components of the ovine cervix: the endothelium and

smooth muscle of the cervical blood vessels, the epithelium of the cervical

canal, the circular and longitudinal muscle layers and the stroma. The

localisation of the four EP receptors in blood vessels emphasises

components of the vascular system as possible targets for PGE2. EP1 and

EP3 receptors induce vasoconstriction, whereas EP2 and EP4 receptors

cause vasodilatation (Wright DH et al 2001). Furthermore, the effects of

steroid hormones on EP receptor expression in the cervix, demonstrate that

whilst progesterone has no effect on any of the four EP receptors, estradiol

significantly decreased EP1 and EP3 receptor protein expression in the

blood vessel media of the cervix (Schmitz T et al 2006).

Estradiol, by decreasing EP1 and EP3 receptor expression, would facilitate

EP2 and EP4 receptor dependent vasodilator effects of PGE2 and therefore,

promote cervical ripening. In other studies, EP4 receptor agonists have been

shown to increase pregnant guinea-pig cervical compliance (Kanayama N et

al 2004). In summary, by altering the balance of positive and negative

influences on the cervix, reducing EP1 or EP3 receptor expression, may as a

result increase the EP2 or EP4 receptor expression and cause cervical

ripening. The regulation of PG synthesis and action occurring in the

myometrium and cervix in association with labour appear to differ markedly

between the two tissues, indicating tissue- specific roles for prostaglandins.

153

It may be that the classical approach of using tocolytics, with the sole

purpose of arresting uterine contractions for the treatment of preterm labour

is not the correct approach. Preterm labour should be viewed as a syndrome

(Romero R et al 2006), where a number of pathological processes including

intrauterine infection/ inflammation, uterine ischaemia, uterine over-

distension, abnormal allograft reaction, allergy, cervical insufficiency and

hormonal disorders (progesterone related and corticotrophin-releasing factor

related) may provide a trigger which stimulates the complex cascade

culminating in uterine contractility, cervical dilatation and activation of the

membranes/ decidua all of which initiate labour. Therefore, a multi-factorial

approach in the diagnosis and management of preterm labour would appear

to be the best solution in the treatment of preterm birth.

154

References

Albertsen K, Andersen AM, Olsen J et al. Alcohol consumption during

pregnancy and the risk of preterm delivery. Am J Epidemiol 2004; 159: 155-

161.

Allen J, Hansen V, Maigaard S, Anderson KE, Forman A. Effects of some

neurotransmitters and prostanoids on isolated human intracervical arteries.

AJOG 1998; 158: 637-641.

Allport VC, Pieber D, Slater DM et al. Human labour is associated with

nuclear factor-kappa B activity which mediates cyclo-oxygenase-2

expression and is involved with the ‘functional progesterone withdrawal’.

Mol Hum Reprod 2001; 7: 581-586.

Anotayanonth S, Subhedar NV, Garner P, Neilson JP, Harigopal S.

Betamimetics for inhibiting preterm labour. Cochrane Database Syst Rev

2004; (4): CD004352.

Astle S, Newton R, Thornton S, Vatish M, Slater DM. Expression and

regulation of prostaglandin E synthase isoforms in human myometrium with

labour. Mol Hum Reprod 2007; 13(1): 69-75.

155

Astle S, Thornton S, Slater DM. Identification and localization of

prostaglandin E2 receptors in upper and lower segment human myometrium

during pregnancy. Mol Hum Reprod 2005; 11(4): 279-87.

Bada HS, Das A, Bauer CR et al. Low birth weight and preterm births:

etiologic fraction attributable to prenatal drug exposure. J Perinatol 2005;

25: 631-637.

Baeuerle PA, Baltimore PD. NF-kappaB: ten years after. Cell 1996; 87: 13-

20.

Bastien L, Sawyer N, Grygorczyk R, Metters KM, Adam M. Cloning,

functional expression, and characterization of the human prostaglandin E2

receptor EP2 subtype. J Biol Chem 1994; 269(16): 11873-7.

Berhman RE, Butler AS (eds.). Role of gene-environment interactions in

preterm birth in Preterm birth: causes, consequences and prevention.

Washington D.C. The National Academies Press, 2007; 207-228.

Bisits A, Madsen G, Knox M et al. The randomized tocolysis trial (RNOTT)

for the treatment of preterm labor. Am J Obstet Gynecol 2004; 191: 683-

690.

Blanks AM, Shmygol A, Thornton S. Myometrial function in pregnancy.

Clinical Obstetrics and Gynaecology 2007; 21(5) : 807-819.

156

Bos CL, Richel DJ, Ritsema T, Peppelenbosch MP, Versteeg HH. Indicated

labor induction with vaginal prostaglandin E2 increases the risk of

caesarean section even in multiparous women with no previous caesarean

section. J Perinat Med 2004; 32: 31-36.

Breyer RM, Bagdassarian CK, Myers SA, Breyer MD. Prostanoid receptors:

subtypes and and signalling. Annu Rev Pharamcol Toxicol 2001; 41: 661-

690.

Brown MA, Buddle ML, Farrell T, Davis GK. Efficacy and safety of

nifedipine tablets for the acute treatment of severe hypertension in

pregnancy. Am J Obstet Gynecol 2002; 187(4):1046-50.

Burguet A, Kaminski M, Abraham-Lerat L et al. EPIPAGE Study Group.

The complex relationship between smoking in pregnancy and very preterm

delivery. Results of the epipage study. BJOG 2004; 111: 258-265.

Cabrol D, Dubois P, Sedbon E, Dallot E, Legagneux J, Amichot G, Cedard

L, Sureau C. Prostaglandin E2-induced changes in the distribution of

glycosaminoglycans in the isolated rat uterine cervix. Eur J Obstet Gynecol

Reprod Biol 1987; 26(4):359-65.

Carey JC, Klebanoff MA, Hauth JC et al. Metronidazole to prevent preterm

delivery in pregnant women with asymptomatic bacterial vaginosis.

157

National Institute of Child Health and Human Development Network of

Maternal-Fetal Medicine Units. N Engl J Med 2000; 342: 534-540.

Carrasco MP, Phaneuf S, Asbóth G, López Bernal A. Fluprostenol activates

phospholipase C and Ca2+ mobilization in human myometrial cells. J Clin

Endocrinol Metab 1996; 81(6):2104-10.

Carr-Hill RA, Hall MH. The repetition of spontaneous preterm labour.

BJOG 1985; 92: 921-928.

Challis JRG. Mechanism of parturition and preterm labor. Obstet Gynecol

Surv 2000; 55(10): 650-60.

Challis JRG, Matthews SG, Gibb W, Lye SJ. Endocrine and paracrine

regulation of birth at term and preterm. Endocr Rev 2000; 21(5): 514-50.

Challis JR, Sloboda DM, Alfaidy N, Lye SJ, Gibb W, Patel FA, Whittle

WL, Newnham JP. Prostaglandins and mechanisms of preterm birth.

Reproduction 2002; 124(1): 1-17.

Chan WY. The separate uterotonic and prostaglandin-releasing actions of

oxytocin. Evidence and comparison with angiotensin and metacholine in the

isolated rat uterus. J Pharmcol Exp Ther 1980; 213: 575-579.

158

Chapman NR, Europe-Finner GN, Robson SC. Expression and

deoxyribonucleic acid-binding activity of the nuclear factor kappaB family

in the human myometrium during pregnancy and labour. J Clin Endocrinol

Metab 2004; 89: 5683-5693.

Charpigny G, Leroy MJ, Breuiller-Fouche M et al. A functional genomic

study to identify differential gene expression in the preterm and term human

myometrium. Biol Reprod 2003; 68: 2289-2296.

Chollet A, Tos EG, Cirillo R. Tocolytic effect of a selective FP receptor

antagonist in rodent models reveals an innovative approach to the treatment

of preterm labor. BMC Pregnancy Childbirth 2007; 7(1): S16.

Condon JC, Jeyasuria P, Faust JM et al. Surfactant protein secreted by the

maturing mouse fetal lung acts as a hormone that signals the initiation of

parturition. Proc Natl Acad Sci USA 2004; 101: 4978-4983.

Coleman RA, Smith WL, Narumiya S. International Union of

Pharmacology classification of prostanoid recpetors: properties, distribution

and structure of the receptors and their subtypes. Pharmacol Rev 1994; 46:

205-229.

159

Coomarasamy A, Knox EM, Gee H et al. Effectiveness of nifedipine versus

atosiban for tocolysis in preterm labour: a meta-analysis with indirect

comparison of randomized trials. BJOG 2003; 110: 1045-1049.

da Fonseca EB, Bittar RE, Carvalho MH et al. Prophylactic administration

of progesterone by vaginal suppository to reduce the incidence of

spontaneous preterm birth in women at increased risk: a randomized

placebo-controlled double-blind study. Am J Obstet Gynecol 2003; 188:

419-424.

Dawood S, Ueno NT, Cristofanilli M. The medical treatment of

inflammatory breast cancer [review]. Semin Oncol 2008; 35: 64-71.

Defea K. Beta-arrestins and heterotrimeric G-proteins: collaborators and

competitors in signal transduction. Br J Pharmacol 2008; 153 (1): S298-

309.

Dietz PM, Callaghan WM, Cogswell ME et al. Combined effects of

prepregnancy body mass index and weight gain during pregnancy on the

risk of preterm delivery. Epidemiology 2006; 17: 170-177.

Doheny HC, O'Reilly MJ, Sexton DJ, Morrison JJ. THG113.31, a specific

PGF2alpha receptor antagonist, induces human myometrial relaxation and

BKCa channel activation. Reprod Biol Endocrinol 2007; 5: 10.

160

Dong YL, Gangula PR, Fang L, Yallampallai C. Differential expression of

cyclooxegenase-1 and -2 proteins in rat uterus and cervix during the estrous

cycle, pregnancy, labor and in myometrial cells. Prostaglandins Leukot

Essnt Fatty Acids 1996; 70: 431-439.

Döring B, Shynlova O, Tsui P, Eckardt D, Janssen-Bienhold U, Hofmann F,

Feil S, Feil R, Lye SJ, Willecke K. Ablation of connexin43 in uterine

smooth muscle cells of the mouse causes delayed parturition. J Cell Sci

2006; 119(9): 1715-22.

Duckitt K, Thornton S. Nitric oxide donors for the treatment of preterm

labour. Cochrane Database Syst Rev 2002; (3).

Engel SA, Erichsen HC, Savitz DA et al. Risk of spontaneous preterm birth

is associated with common pro-inflammatory cytokine polymorphisms.

Epidemiology 2005; 16: 469-477.

Eronen M, Pesonen E, Kurki T, Ylikorkala O. The effects of indomethacin

and a beta-sympathomimetic agent on the fetal ductus arteriosus during

treatment of preterm labor: a randomised double-blind study. Am J Obstet

Gynecol 1991; 164: 141-6.

161

Esplin MS, Varner MW. Genetic factors in preterm birth – the future. BJOG

2005; 112(1): 97-102.

Final report of the Medical Research Council/ Royal College of

Obstetricians and Gynaecologists multicentre randomized trial of cervical

cerclage. MRC/RCOG Working Party on Cervical Cerclage. Br J Obstet

Gynaecol 1984; 91: 731-735.

Fisk NM, Atun R. Market failure and the poverty of new drugs in maternal

health. PLoS Med 2008; 5(1):e22.

Friel AM, O'Reilly MW, Sexton DJ, Morrison JJ. Specific PGF(2alpha)

receptor (FP) antagonism and human uterine contractility in vitro. BJOG

2005; 112(8): 1034-42.

Fuchs AR, Fuchs F. Endocrinology of human parturition: a review. Br J

Obstet Gynaecol 1984; (10): 948-67.

Fuchs AR, Fuchs F, Husslein P, Soloff MS. Oxytocin receptors in the

human uterus during pregnancy and parturition. Am J Obstet Gynecol 1984;

150(6): 734-41.

Fujimoto T, Parry S, Urbanek M et al. A single nucleotide polymorphism in

the matrix metalloproteinase-1 (MMP-1) promoter influences amnion cell

162

MMP-1 expression and risk for preterm premature rupture of the fetal

membranes. J Biol Chem 2002; 277: 6296-6302.

Garfield RE, Blennerhassett MG, Miller SM. Control of myometrial

contractility: role and regulation of gap junctions. Oxf Rev Reprod Biol.

1988; 10: 436-90.

Giannopoulos G, Jackson K, Kredentser J, Tulchinsky D. Prostaglandin E

and F2 alpha receptors in human myometrium during the menstrual cycle

and in pregnancy and labor. Am J Obstet Gynecol 1985; 153(8): 904-10.

Giles W, Bisits A. The present and future of tocolysis. Clinical Obstetrics

and Gynaecology 2007; 21(5) : 857-868.

Gimpl G, Fahrenholz F. The oxytocin receptor system: structure, function

and regulation. Physiol Rev 2001; 81: 629-683.

Goldenberg RL, Iams JD, Mercer BM et al. The Preterm Prediction Study:

toward a multiple marker test for spontaneous preterm birth. Am J Obstet

Gynecol 2001; 185: 643-651.

163

Gomez R, Ghezzi F, Romero R, Muñoz H, Tolosa JE, Rojas I. Premature

labor and intra-amniotic infection. Clinical aspects and role of the cytokines

in diagnosis and pathophysiology. Clin Perinatol 1995; 22(2):281-342.

Goodwin TM, Paul R, Silver H, Spellacy W, Parsons M, Chez R, Hayashi

R, Valenzuela G, Creasy GW, Merriman R. The effect of the oxytocin

antagonist atosiban on preterm uterine activity in the human. Am J Obstet

Gynecol 1994; 170(2): 474-8.

Gordon-Wright AP, Elder MG. Effect of prostaglandin E2 and its

metabolites on lower segment myometrium in vitro. Eur J Obstet Gynecol

Reprod Biol 1980; 10(5): 297-302.

Grazzini E, Guillon G, Mouillac B et al. Inhibition of oxytocin receptor

function by direct binding of progesterone. Nature 1998; 392: 509-512.

Grigsby PL, Sooranna SR, Brockman DE, Johnson MR & Myatt L.

Localization and expression of prostaglandin E2 receptors in human

placenta and corresponding fetal membranes with labor. Am J Obstet

Gynecol 2006; 195: 260-269.

Groom KM. Pharmacological prevention of prematurity. Clin Obstet and

Gynaecol 2007; 21(5): 843-856.

164

Groom KM, Shennan AH, Jones BA et al. TOCOX- a randomized, double-

blind, placebo-controlled trial of rofecoxib (a COX-2-specific prostaglandin

inhibitor) for the prevention of preterm delivery in women at high risk.

BJOG 2005; 112: 725-730.

Gyetvai K, Hannah ME, Hodnett ED, Ohlsson A. Tocolytics for preterm

labour: a systematic review. Obstet Gynecol 1999; 94: 869-77.

Hauth JC, Goldenberg RL, Andrews WW et al. Reduced incidence of

preterm delivery with metronidazole and erythromycin in women with

bacterial vaginosis. N Eng J Med 1995; 333: 1732-1736.

Heath VC, Southall TR, Souka AP, et al. Cervical length at 23 weeks of

gestation: relation to demographic characteristics and previous obstetric

history. Ultrasound Obstet Gynecol 1998; 12: 304-311.

Heath VC, Southall TR, Souka AP, et al. Cervical length at 23 weeks of

gestation: prediction of spontaneous preterm delivery. Ultrasound Obstet

Gynecol 1998; 12: 312-317.

Hirst JJ, Parkington HC, Young IR, Palliser HK, Peri KG, Olson DM. Delay

of preterm birth in sheep by THG113.31, a prostaglandin F2alpha receptor

antagonist. Am J Obstet Gynecol 2005; 193(1): 256-66.

165

Hoffman HJ, Bakketeig LS. Risk factors associated with the occurance or

preterm birth. Clin Obstet Gynecol 1984; 27: 539-552.

Honest H, Bachmann LM, Gupta JK et al. Accuracy of cervicovaginal fetal

fibronectin test in predicting risk of spontaneous preterm birth: systematic

review. BMJ 2002; 325: 301.

Hughes EG, Kelly AJ, Kavanagh J. Dinoprostone vaginal insert for cervical

ripening and labor induction: a meta-analysis. Obstet Gynecol 2001; 97:

847-855.

Husslein P, Quartarolo JP. Review of clinical experience with atosiban and

the Tractocile Efficacy Assessment Survey in Europe (TREASURE) study

protocol. Int J Clin Pract 2003; 57(2): 121-7.

Jeng YL, Soloff SL, Anderson GD et al. Regulation of oxytocin receptor

expression in cultured human myometrial cells by fetal bovine serum and

lysophospholipids. Endocrinology 2003; 144: 61-68.

Jin J, Guang F, Ashby, B. Constitutive activity of human prostaglandin E

receptor EP3 isoforms. Br J Pharmacol 1997; 121: 317-323.

166

Kanayama N, Matsuura H, Inada C, Yamaguchi S, Mori T, Hayashi R,

Yamada N. Pharmacological profile of new cervical ripening agent, APS-

999 in guinea pig. Program of 51st SGI meeting, Houston, Tx, USA 2004.

Abstract 16.

Kandola MK. EP receptor signalling pathways. Phd. upgrade, Imperial

College, London 2008.

Keirse M. Indomethacin tocolysis in preterm labour in Enkin MK, Kierse

MJNC, Renfrew MJ, Neilson JP (eds). Pregnancy childbirth module.

Oxford: Updat Software, 1995.

Kelly RH, Russo J, Holt VL et al. Psychiatric and substance abuse disorders

as risk factors for low birth weight and preterm delivery. Obstet Gynecol

2002; 100: 297-304.

Kelly RW. Inflammatory mediators and cervical ripening. J Reprod

Immunol 2002; 57: 217-224.

Kenyon S, Pike K, Jones DR, Brocklehurst P, Marlow N, Salt A, Taylor DJ.

Childhood outcomes after prescription of antibiotics to pregnant women

with spontaneous preterm labour: 7-year follow-up of the ORACLE II trial.

Lancet 2008; 372(9646): 1319-27.

167

Khan KS, Honest H. Risk Screening for spontaneous preterm labour.

Obstetrics and Gynaecology 2007; 21(5) : 821-830.

Kilarski WM, Rothery S, Roomans GM, Ulmsten U, Rezapour M,

Stevenson S, Coppen SR, Dupont E, Severs NJ. Multiple connexins

localized to individual gap-junctional plaques in human myometrial smooth

muscle. Microsc Res Tech 2001; 54(2): 114-22.

King JF, Flenady V, Cole S, Thornton S. Cyclo-oxygenase (COX) inhibitors

for treating preterm labour. Cochrane Database Syst Rev 2005; (2):

CD001992.

King JF, Flenady VJ, Papatsonis DN, Dekker GA, Carbonne B. Calcium

channel blockers for inhibiting preterm labour. Cochrane Database Syst Rev

2003; (1): CD002255.

Kirshon B, Moise Jr. KJ, Mari G et al. Long-term indomethacin treatment

decreases fetal urine output and results in oligohydramnios. Am J Perinatol

1991; 8: 86-88.

Kitazawa T, Yamada Y, Iwano H, Yokota H, Yuasa A, Taneike T. Smooth

muscle layer-dependent distribution of 5-hydroxytryptamine(7) receptor in

the porcine myometrium. Br J Pharmacol 2000; 130(1): 79-89.

168

Klebanoff MA, Carey JC, Hauth JC et al. Failure of metronidazole to

prevent preterm delivery among pregnant women with asymptomatic

Trichomonas vaginalis infection. N Eng J Med 2001; 345: 487-493.

Klebanoff MA, Hillier SL, Nugent RP et al. is bacterial vaginosis a stronger

risk factor for preterm birth when it is diagnosed earlier in gestation? Am J

Obstet Gynecol 2005; 192: 470-477.

Kniss DA. Cyclooxygenases in reproductive medicine and biology. J Soc

Gynecol Investig 1999; 6(6):285-92.

Kotani M, Tanaka I, Ogawa Y, Usui T, Mori K, Ichikawa A, Narumiya S,

Yoshimi T, Nakao K. Molecular cloning and expression of multiple

isoforms of human prostaglandin E receptor EP3 subtype generated by

alternative messenger RNA splicing: multiple second messenger systems

and tissue-specific distributions. Mol Pharmacol 1995; 48(5): 869-79.

Kotani M, Tanaka I, Ogawa Y, Usui T, Tamura N, Mori K, Narumiya S,

Yoshimi T, Nakao K. Structural organization of the human prostaglandin

EP3 receptor subtype gene (PTGER3). Genomics 1997; 40(3): 425-34.

169

Kotani M, Tanaka I, Ogawa Y, Suganami T, Matsumoto T, Muro S,

Yamamoto Y, Sugawara A, Yoshimasa Y, Sagawa N, et al. Multiple signal

transduction pathways through two prostaglandin E receptor EP3 subtype

isoforms expressed in human uterus. J Clin Endocrinol Metab 2000; 85

4315-4322.

Lamont RF, Duncan SL, Mandal D et al. Intravaginal clindamycin to reduce

preterm birth in women with abnormal genital tract flora. Obstet Gynecol

2003; 101: 516-522.

Larroque B, Ancel PY, Marret S, Marchand L, André M, Arnaud C, Pierrat

V, Rozé JC, Messer J, Thiriez G, Burguet A, Picaud JC, Bréart G, Kaminski

M; EPIPAGE Study group. Neurodevelopmental disabilities and special

care of 5-year-old children born before 33 weeks of gestation (the EPIPAGE

study): a longitudinal cohort study. Lancet 2008; 371(9615):813-20.

Leitich H, Bodner-Adler B, Brunbauer M, Kaider A, Egarter C, Husslein P.

Bacterial vaginosis as a risk factor for preterm delivery: A meta-analysis.

Am J Obstset Gynecol 2003; 189: 139-147.

Leonhardt A, Glaser A, Wegmann M, Hackenberg R, Nüsing RM.

Expression of prostanoid receptors in human lower segment pregnant

myometrium. Prostaglandins Leukot Essent Fatty Acids 2003; 69(5): 307-

13.

170

López Bernal A, Hansell DJ, Cañete Soler R, Keeling JW, Turnbull AC.

Prostaglandins, chorioamnionitis and preterm labour. Br J Obstet Gynaecol

1987; 94(12):1156-8.

Luckas MJ, Wray S. A comparison of the contractile properties of human

myometrium obtained from the upper and lower uterine segments. BJOG

2000; 107(10):1309-11.

Lye S. The initiation and inhibition of labour – toward a molecular

understanding. Semin Reprod Endocrinol 1994; 12: 284-294.

Lyell DJ, Pullen K, Campbell L, Ching S, Druzin ML, Chitkara U, Burrs D,

Caughey AB, El-Sayed YY. Magnesium sulfate compared with nifedipine

for acute tocolysis of preterm labor: a randomized controlled trial. Obstet

Gynecol 2007; 110(1): 61-7.

Ma X, Wu WX, Nathanielsz PW. Differential regulation of prostaglandin

EP and FP receptors in pregnant sheep myometrium and endometrium

during spontaneous term labor. Biol Reprod 1999; 61(5): 1281-6.

Macones GA, Marder SJ, Clothier B et al. The controversy surrounding

indomethacin for tocolysis. Am J Obstet Gynecol 2001; 184: 264-272.

171

Macones GA, Parry W, Elkousy M et al. A polymorphism in the promoter

region of TNF and bacterial vaginosis preliminary evidence of gene-

environment interaction in the etiology of spontaneous preterm birth. Am J

Obstet Gynecol 2004; 190: 1504-1508.

Marsal K. The European Atosiban Study Group. The oxytocin antagonist

atosiban versus the β-agonist terbutaline in the treatment of preterm labour.

A randomized, double-blind, controlled study. Acta Obstet Gynecol Scand

2001; 80: 413-422.

Martin RL, Whittle WL, Holloway AC, Gyomorey S, Gibb W, Lye S,

Challis JR. Ontogeny and regulation of ovine placental prostaglandin E2

synthase. Biol Reprod 2002; 67(3): 868-73.

McLean M, Bisits A, Davies J et al. A placental clock controlling the length

of human pregnancy. Nat Med 1995; 1: 460-463.

McCoshen JA, Johnson KA, Dubin NH & Ghodgaonkar RB. Prostaglandin

E2 release on the fetal and maternal sides of the amnion and chorion-

decidua before and after term labor. Am J Obstet Gynecol 1987; 156: 173-

178.

172

Meis PJ, Klebanoff M, Thom E et al. Prevention of recurrent preterm

delivery by 17 alpha-hydroxyprogesterone caproate. N Eng J Med 2003;

348: 2379-2385.

Mesiano S, Chan EC, Fitter JT et al. Progesterone withdrawal and estrogen

activation in human parturition are coordinated by progesterone receptor A

expression in the myometrium. J Clin Endocrinol Metab 2002; 87: 2924-

2930.

Mitchell MD, Flint AP (1978). Use of meclofenamic acid to investigate the

role of prostaglandin biosynthesis during induced parturition in sheep. J

Endocrinol; 76: 101-109.

Milligan G, Kostenis E. Heterotrimeric G-proteins: a short-history. Br J

Pharmacol 2006 ; 147 (1): S46-55.

Mohan AR, Loudon JA, Bennett PR. Molecular and biochemical

mechanisms of preterm labour. Semin Fetal Neonatl Med 2004; 9: 437-444.

Mohan AR, Soorana SR, Lindstrom TM et al. The effect of mechanical

stretch on cyclooxygenase type 2 and activator protein-1 and nuclear factor-

kappaB activity in human amnion cells. Endocrinology 2007; 148: 1850-

1857.

173

Moutquin JM. Classification and heterogeneity of preterm birth. BJOG

2003; 110:20 : 30-33.

Moutquin JM, Cabrol D, Fisk NM et al. Effectiveness and the safety of the

oxytocin antagonist atosiban versus beta-adrenergic agonists in the

treatment of preterm labour. BJOG 2001; 108: 133-142.

Muglia LJ, Katz M. The enigma of spontaneous preterm birth. N Eng J Med

2010; 326(6): 529-535.

Murphy DJ. Epidemiology and environmental factors in preterm labour.

Clinical Obstetrics and Gynaecology 2007; 21:5 : 771-789.

Nakao K, Inoue Y, Okabe K, Kawarabayashi T, Kitamura K. Oxytocin

enhances action potentials in pregnant human myometrium--a study with

microelectrodes. Am J Obstet Gynecol 1997; 177(1): 222-8.

Namba T, Sugimoto Y, Negishi M, Irie A, Ushikubi F, Kakizuka A, Ito S,

Ichikawa A, Narumiya S. Alternative splicing of C-terminal tail of

prostaglandin E receptor subtype EP3 determines G-protein specificity.

Nature 1993; 365(6442): 166-70.

Narumiya S, Sugimoto Y, Ushikubi F. Prostanoid receptors: structures,

properties, and functions. Physiol Rev. 1999; 79(4):1193-226.

174

Nath CA, Ananth CV, Smulian JC, Shen-Schwarz S, Kaminsky L; New

Jersey-Placental Abruption Study Investigators. Histologic evidence of

inflammation and risk of placental abruption. Am J Obstet Gynecol 2007;

197(3): 319.e1-6.

Nicolaides KH, Tsoi E, To MS. Sonographic measurement of cervical

length and preterm delivery in Critchley HO, Bennett PR & Thornton S

(eds.). Preterm birth. RCOG Press 2004: 124-139.

Norman JE. Cervical function and prematurity. Clinical Obstetrics and

Gynaecology 2007; 21:5 : 791-806.

Norton ME, Merrill J, Cooper BA et al. Neonatal complications after the

administration of indomethacin for preterm labor. N Eng J Med 1993; 329:

1602-1607.

Oei SG, Mol BW, de Kleine MJ, Brolmann HA. Nifedipine versus ritodrine

for suppression of preterm labour: a meta-analysis. Acta Obstet Gynecol

Scand 1999; 78: 783-8.

Offenbacher S, Boggess KA, Mutha AP et al. Progressive periodontal

disease and risk of very preterm delivery. Obstet Gynecol 2006; 107: 29-36.

175

Orsi NM, Gopichandran N, Simpson NAB. Genetics of preterm labour.

Clinical Obstetrics and Gynaecology 2007; 21:5 : 757-772.

Olson DM. The role of prostaglandins in the initiation of parturition in Best

Practice & Research Clinical Obstetrics and Gynaecology 2003; 17(5):

717-730.

Olson DM, Zaragoza DB, Shallow MC, Cook JL, Mitchell BF, Grigsby P,

Hirst J. Myometrial activation and preterm labour: evidence supporting a

role for the prostaglandin F receptor--a review. Placenta 2003; 24(A): S47-

54.

Palliser HK, Hirst JJ, Ooi GT, Rice GE, Dellios NL, Escalona RM,

Parkington HC, Young IR. Prostaglandin e and f receptor expression and

myometrial sensitivity at labor onset in the sheep. Biol Reprod 2005;

72(4):937-43.

Papatsonis DN, Carbonne B, Dekker GA, Flenady V, King JF. Severe

hypotension and fetal death due to tocolysis with nifedipine. BJOG 2005;

112(11): 1582-3.

176

Parkington HC, Tonta MA, Brennecke SP, Coleman HA. Contractile

activity, membrane potential, and cytoplasmic calcium in human uterine

smooth muscle in the third trimester of pregnancy and during labor. Am J

Obstet Gynecol 1999; 181(6): 1445-51.

Parkington HC, Tonta MA, Davies NK, Brennecke SP, Coleman HA.

Hyperpolarization and slowing of the rate of contraction in human uterus in

pregnancy by prostaglandins E2 and f2alpha: involvement of the Na+ pump.

J Physiol 1999; 514 (1): 229-43.

Patel FA, Clifton VL, Chwalisz K et al. Steroid regulation of prostaglandin

dehydrogenase activity and expression in human term placenta and chorio-

decidua in relation to labour. J Clin Endocrinol Metab 1999; 84: 291-299.

Phaneuf S, Carrasco MP, Europe-Finner GN, Hamilton CH, López Bernal

A. Multiple G proteins and phospholipase C isoforms in human myometrial

cells: implication for oxytocin action. J Clin Endocrinol Metab 1996;

81(6):2098-103.

Phelan JP, Guay AT, Newman C. Diabetes insipidus in pregnancy: a case

review. Am J Obstet Gynecol 1978; 130: 365-366.

177

Pieber D, Allport VC, Bennett PR. Progesterone receptor isoform A inhibits

isoform B-mediated transactivation in human amnion. Eur J Pharmacol

2001; 427: 7-11.

Pierce KL, Regan JW. Prostanoid receptor heterogeneity through alternative

mRNA splicing. Life Sci. 1998; 62(17-18): 1479-83.

Pierre-Yves Ancel et al. Cerebral palsy among very preterm children in

relation to gestational age and neonatal ultrasound abnormalities: The

EPIPAGE Cohort study. Pediatrics 2006; 3(7): 828-835.

RCOG Guideline Compendium. Tocolytic drugs for women in preterm

labour. RCOG Press 2006: 91-98.

Regan JW. EP2 and EP4 prostanoid receptor signaling. Life Sci 2003; 74(2-

3): 143-53.

Roberts D, Dalziel S. Antenatal corticosteroids for accelerating fetal lung

maturation for women at risk of preterm birth. Cochrane Database Syst Rev

2006; 3: CD004454.

178

Robertson FM, Simeone AM, Lucci A, McMurray JS, Ghosh S, Cristoanilli

M. Differential regulation of the aggressive phenotype of inflammatory

breast cancer cells by prostanoid receptors EP3 and EP4. Cancer 2010; 116

(11): 2806-14.

Romero R, Espinoza J, Kusanovic JP, Gotsch F, Hassan S, Erez O,

Chaiworapongsa T, Mazor M. The preterm parturition syndrome.

BJOG 2006; 113(3): 17-42.

Romero R, Mazor M. Infection and preterm labor. Clin Obstet Gynecol

1988; 31: 553-584.

Romero R, Sibai BM, Sanchez-Ramos L, Valenzuela GJ, Veille JC, Tabor

B, et al. An oxytocin receptor antagonist (atosiban) in the treatment of

preterm labor; a randomised double-blind, placebo-controlled trial woth

tocolyitc rescue. Am J Obstet Gynecol 2000; 182: 1173-83.

Sawdy RJ, Lye S, Fisk NM, Bennett PR. A double-blind randomized study

of fetal side effects during and after the short-term maternal administration

of indomethacin, sulindac, and nimesulide for the treatment of preterm

labor. Am J Obstet Gynecol 2003; 188(4):1046-51.

179

Sawdy RJ, Sullivan MH, Bennett PR. The effects of non-steroidal anti-

inflammatory compounds on human myometrial contractility. Eur J Obstet

Gynecol Reprod Biol 2003; 109(1):33-40.

Schmid A, Thierauch KH, Schleuning WD, Dinter H. Splice variants of the

human EP3 receptor for prostaglandin E2. Eur J Biochem 1995; 228(1): 23-

30.

Schmitz T, Levine BA, Nathanielsz PW (2006). Localization and steroid

regulation of prostaglandin E2 receptor protein expression in ovine cervix.

Reproduction; 131: 743-750.

Schrey MP, Cornford PA, Read AM, Steer PJ. A role for phosphoinositide

hydrolysis in human uterine smooth muscle during parturition. Am J Obstet

Gynecol 1988; 159(4):964-70.

Seubert DE, Stetzer BP, Wolfe HM et al. Delivery of the marginally preterm

infant: what are the minor morbidities? Am J Obstet Gynecol 1999; 181:

1087-1091.

180

Shennan A, Crawshaw S, Briley A, Hawken J, Seed P, Jones G, Poston L. A

randomised controlled trial of metronidazole for the prevention of preterm

birth in women positive for cervicovaginal fetal fibronectin: the PREMET

Study. BJOG 2006; 113(1): 65-74.

Shmygol A, Blanks AM, Bru-Mercier G, Gullam JE, Thornton S. Control of

uterine Ca2+ by membrane voltage: toward understanding the excitation-

contraction coupling in human myometrium. Ann N Y Acad Sci 2007;

1101:97-109.

Simcox R, Shennan A. Cervical cerclage in the prevention of preterm birth.

Clinical Obstetrics and Gynaecology 2007; 21(5) : 831-842.

Simcox R, Sin WA, Seed PT et al. Prophylactic antibiotics for the

prevention of preterm birth in women at risk: A meta-analysis. Aust NZ J

Obstet Gynaecol 2007; 47: 368-377.

Sims SM, Daniel EE, Garfield RE. Improved electrical coupling in uterine

smooth muscle is associated with increased numbers of gap junctions at

parturition. J Gen Physiol 1982; 80(3): 353-75.

181

Sissakos D, O’Brien K, Draycott T. Healthcare evaluation of the use of

atosiban and fibronectin for the management of pre-term labour. J Obstet

Gynae 2009; 29(6): 507-11.

Skinner KA & Challis JR. Changes in the synthesis and metabolism of

prostaglandins by human fetal membranes and decidua at labor. Am J

Obstet Gynecol 1985; 151: 519-523.

Slater DM, Astle S, Woodcock N, Chivers JE, de Wit NC, Thornton S,

Vatish M & Newton R. Anti-inflammatory and relaxatory effects of

prostaglandin E2 in myometrial smooth muscle. Mol Hum Reprod 2006; 12:

89-97.

Slater DM, Berger L, Newton R et al. Expression of cyclooxygenase types 1

and 2 in human fetal membranes at term. Am J Obstet Gynecol 1995; 172:

77-82.

Slater DM, Dennes W, Sawdy R et al. Expression of cyclo-oxygenase types-

1 and -2 in human fetal membranes throughout pregnancy. J Mol

Endocrinol 1999; 22: 125-130.

Smith GC, Wu WX, Nathanielsz PW. Effects of gestational age and labor

on the expression of prostanoid receptor genes in pregnant baboon cervix.

Prostaglandins Other Lipid Mediat 2001; 63(4):153-63.

182

Smith GC, Wu WX, Nathanielsz PW. Expression of prostanoid receptor

genes in baboon chorion and decidua during pregnancy and parturition. J

Endocrinol 2001; 168(2):263-72.

Smith R, Nicholson RC. Corticotrophin releasing hormone and the timing of

birth. Front Biosci 2007; 12: 912-918.

Smith R, Van Helden D, Hirst J et al. Pathological interactions with the

timing of birth and uterine activation. Aust NZ J Obstet Gynaecol 2007; 47:

430-437.

Soloff MS. Uterine receptor for oxytocin: effects of estrogen. Biochem

Biophys Res Commun 1975; 65: 205-212.

Soloff MS, Alexandrova M, Fernstrom MJ. Oxytocin receptors: triggers for

parturition and lactation? Science 1979; 204: 1313-1315.

Sparey C, Robson SC, Bailey J, Lyall F, Europe-Finner GN. The differential

expression of myometrial connexin-43, cyclooxygenase-1 and -2, and Gs

alpha proteins in the upper and lower segments of the human uterus during

pregnancy and labor. J Clin Endocrinol Metab 1999; 84(5): 1705-10.

183

Takeuchi K. Prostaglandin EP receptors and their roles in mucosal

protection and ulcer healing in the gastrointestinal tract. Adv Clin Chem

2010; 51: 121-44.

Tan TC, Devendra K, Tan LK, Tan HK. Tocolytic treatment for the

management of preterm labour: a systemic review. Singapore Med J 2006;

47(5):361-6.

Terzidou V. Biochemical and endocrinological preparation for parturition.

Clinical Obstetrics and Gynaecology 2007; 21(5) : 729-756.

The World-wide Atosiban versus Beta-agonists Study Group. Effectiveness

and safety of the oxytocin antagonist atosiban versus beta-adrenergic

agonists in the treatment of preterm labor. BJOG 2001; 108: 133-42.

Thomson A, Norman J. Biology of preterm labour. Preterm labour –

managing risk in clinical practice. Cambridge University Press 2005; 26-75.

Tsatsaris V, Papatsonis D, Goffinet F, Dekker G, Carbonne B. Tocolysis

with nifedipine or beta-adrenergic agonists: a meta-analysis. Obstet Gynecol

2001; 97: 840-7.

Urban E, Radnai M, Novak T et al. Distribution of anaerobic bacteria

among pregnant periodontitis patients. Anaerobe 2006; 12: 52-57.

184

Valenzuela GJ, Sanchez-Ramos L, Romero R, Silver HM, et al.

Maintenance treatment of preterm labor with the oxytocin antagonist

atosiban. The Atosiban PTL-098 Study Group. Am J Obstet Gynecol 2000;

182: 1184-90.

von Dadelszen P, Kives S, Delisle MF, Wilson RD, Joy R, Ainsworth L, Al

Kharusi L, Oskamp M, Barrett JF, Ryan G, Farine D, Seaward PG. The

association between early membrane rupture, latency, clinical

chorioamnionitis, neonatal infection, and adverse perinatal outcomes in twin

pregnancies complicated by preterm prelabour rupture of membranes. Twin

Res 2003; 6(4): 257-62.

van Geijn HP, Lenglet JE, Bolte AC. Nifedipine trials: effectiveness and

safety aspects.

BJOG 2005; 112(1): 79-83.

Vojcek L, Princzkel E, Lampé LG, Turnbull AC, Kovács T. Effects of

calcium channel-blocker tokolysis on the foetal circulation. Acta Physiol

Hung 1988; 71(4): 517-28.

Wadhwa PD, Culhane JF, Raugh V et al. Stress, infection and preterm birth:

a biobehavioural perspective. Pediatr Perinat Epidemiol 2001; 15:2: 17-29.

185

Warren WB, Goland RS, Wardlaw SL, Stark RI, Fox HE, Conwell IM.

Elevated maternal plasma corticotropin releasing hormone levels in twin

gestation. J Perinat Med 1990;18(1):39-44.

Whitworth M, Quenby S. Prophylactic oral betamimetics for preventing

preterm labour in singleton pregnancies. Cochrane Database Syst Rev 2008;

(1): CD006395.

Wing DA, Goharkhay N, Hanna M, Naidu YM, Kovacs BW & Felix JC.

EP3-2 receptor mRNA expression is reduced and EP3-6 receptor mRNA

expression is increased in gravid human myometrium. J Soc Gynecol

Investig 2003; 10: 124-129.

Wiqvist N, Bryman I, Lindblom B, Norstrom A & Wikland M. The role of

prostaglandins for the coordination of myometrial forces during labour. Acta

Physiol Hung 1985; 65: 313-322.

Woodcock NA, Taylor CW, Thornton S. Effect of an oxytocin receptor

antagonist and rho kinase inhibitor on the [Ca++] sensitivity of human

myometrium. Am J Obstet Gynecol 2004; 190(1):222-8.

Woodcock NA, Taylor CW, Thornton S. Prostaglandin F2alpha increases

the sensitivity of the contractile proteins to Ca2+ in human myometrium.

Am J Obstet Gynecol 2006; 195(5):1404-6.

186

Word RA, Landrum CP, Timmons BC, Young SG, Mahendroo MS.

Transgene insertion on mouse chromosome 6 impairs function of the uterine

cervix and causes failure of parturition. Biol Reprod 2005; 73(5): 1046-56.

Word RA, Stull JT, Casey ML et al. Contractile elements and myosin light

chain phosphorylation in myometrial tissue from nonpregnant and pregnant

women. J Clin Invest 1993; 92: 29-37.

Wright DH, Abran D, Bhattacharya M, Hou X, Baernier SG et al.

Prostanoid receptor: ontogeny and implications in vascular physiology. Am

J Physiol 2001; 281: 1343-1360.

Wu WX, Ma XH, Yoshizato T et al. Differential expression of myometrial

oxytocin receptor and prostaglandin H synthase 2, but not estrogen receptor

alpha and heat shock protein 90 messenger ribonucleic acid in the gravid

horn and nongravid horn in sheep during betamethasone-induced labor.

Endocrinology 1999; 140: 5712-5718.

Young A, Thompson AJ, Ledingham M et al. Immunolocalisation of

proinflammatory cytokines in myometrium, cervix and fetal membranes

during human parturition at term. Biol Reprod 2002; 66: 445-449.

Young RC, Hession RO. Three-dimensional structure of the smooth muscle

in the term-pregnant human uterus. Obstet Gynecol 1999; 93: 94-99.