characterisation of caspase-14 in the human placenta · characterisation of caspase-14 in the human...

i

Characterisation of

Caspase-14 in the Human

Placenta Evidence for trophoblast-specific inhibition of

differentiation by caspase-14

Lloyd John White (BSc (Hons))

12/09/2008

This thesis is presented for the degree of

Doctor of Philosophy

at

The University of Western Australia

School of Anatomy and Human Biology

2008

ii

"...the only people for me are the mad ones, the ones who are mad to live,

mad to talk,

mad to be saved,

desirous of everything at the same time,

the ones that never yawn or say a commonplace thing, but burn, burn, burn

like fabulous yellow roman candles exploding like spiders across the stars..."

Jack Kerouac, On the Road, Part 1, Ch. 1

iii

Table of Contents

Table of Contents ............................................................................................... iii

List of Figures and Tables .................................................................................viii

Abbreviations ................................................................................................... xvii

Abstract ............................................................................................................ xix

Acknowledgements .......................................................................................... xxi

Declaration ....................................................................................................... xxii

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

Introduction ......................................................................................................... 1

1.1 Placenta ........................................................................................................ 1

1.1.1 Extravillous Trophoblast .................................................................................. 4

1.1.2 Hormones ......................................................................................................... 5

1.1.3 Endothelia ........................................................................................................ 5

1.1.3.1 Reactive Oxygen Species ........................................................................... 7

1.1.3.2 Endothelial Nitric Oxide Synthase ............................................................. 7

1.1.3.3 Vascular Endothelial Growth Factor .......................................................... 7

1.1.4 Preeclampsia .................................................................................................... 9

1.1.5 In Vitro Trophoblast Models .......................................................................... 10

1.2 Apoptosis .............................................................................................................. 10

1.2.1 Placental Apoptosis ........................................................................................ 11

1.3 Caspases ................................................................................................................ 12

1.4 Caspase-14 ............................................................................................................ 15

1.4.1 Caspase-14 functional pathways .................................................................... 16

1.4.1.1 Filaggrin ................................................................................................... 18

1.4.1.2 p57KIP2

...................................................................................................... 19

1.4.1.3 Krüppel-like Factor 4 ............................................................................... 19

1.5 Canonical Wnt signalling ...................................................................................... 20

1.5.1 E-cadherin and β-catenin interactions ............................................................ 21

iv

1.5.2 Secreted Frizzled-Related Protein-4 (sFRP4) ................................................ 23

1.6 Experimental Design ............................................................................................. 24

1.6.1 Aims ............................................................................................................... 24

1.6.2 Hypotheses ..................................................................................................... 24

Chapter 2 .......................................................................................................... 25

Materials and Methods ...................................................................................... 25

2.1 BeWo cell line ....................................................................................................... 25

2.2 Human Placenta Extracts ...................................................................................... 25

2.3 Assessment of Apoptosis ...................................................................................... 26

2.3.1 JC-1 ................................................................................................................ 27

2.3.2 DNA extraction from BeWo Cell Culture ..................................................... 27

2.3.3 Explant Culture Model of Apoptosis ............................................................. 28

2.3.4 DNA Extraction from Placenta ...................................................................... 29

2.3.5 3‟-End Labelling ............................................................................................ 30

2.4 Assessment of RNA Expression ........................................................................... 31

2.4.1 RNA extractions ............................................................................................. 31

2.4.1.1 RNA extraction from human placental explants ...................................... 31

2.4.1.2 RNA extraction from the cell culture (BeWo) model .............................. 32

2.4.2 DNAse Treatment and Reverse Transcription of Total RNA ........................ 32

2.4.3 Post-PCR Clean-up ........................................................................................ 33

2.4.4 Dilution of Primers ......................................................................................... 33

2.4.5 Real Time Quantitative PCR .......................................................................... 34

2.5 Assessment of Protein Expression and Localisation ............................................. 36

2.5.1 Immunocytochemistry.................................................................................... 36

2.5.2 Immunohistochemistry ................................................................................... 36

2.5.3 Extraction of Protein ...................................................................................... 37

2.5.4 Bradford Assay for Protein Quantitation ....................................................... 38

v

2.5.5 Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS-PAGE)

................................................................................................................................. 38

2.5.6 Immunoblotting .............................................................................................. 40

2.6 Stimulation of the Endothelial Pathway ................................................................ 41

2.7 RNA Interference .................................................................................................. 41

2.7.1 Optimisation of siRNA .................................................................................. 42

2.7.2 Silencing of Caspase-14 ................................................................................. 42

2.8 Statistical Analysis ................................................................................................ 43

Chapter 3 .......................................................................................................... 45

Placental Gene Expression across Human Gestation and in Preeclampsia ..... 45

3.1 Background ........................................................................................................... 45

3.2 Techniques ............................................................................................................ 46

3.2.1 Real Time Polymerase Chain Reaction.......................................................... 46

3.2.2 Immunohistochemistry ................................................................................... 46

3.2.3 Explant Culture Model ................................................................................... 46

3.3 Hormonal Function of the Placenta....................................................................... 47

3.3.1 Beta-human chorionic gonadotrophin (β-hCG) ............................................. 47

3.3.2 Human Placental Lactogen (hPL) .................................................................. 49

3.4 Caspase-14 in the Human Placenta ....................................................................... 50

3.4.1 Profilaggrin .................................................................................................... 53

3.4.2 Krüppel Like Factor 4 (KLF4) ....................................................................... 54

3.4.3 Cytokeratin-18................................................................................................ 55

3.5 Secreted Frizzled-Related Protein 4 ...................................................................... 56

3.6 Endothelial Function of the Placenta .................................................................... 58

3.6.1 Endothelial Nitric Oxide Synthase (eNOS) ................................................... 58

3.6.2 Vascular Endothelial Growth Factor (VEGF)................................................ 60

3.7 Limitations ............................................................................................................ 61

Chapter 4 .......................................................................................................... 63

vi

BeWo Model Establishment and Validation ...................................................... 63

4.1 Background ........................................................................................................... 63

4.2 Validation of BeWo Cells ..................................................................................... 64

4.3 BeWo Apoptosis ................................................................................................... 65

4.3.1 Staurosporine ................................................................................................. 65

4.3.2 JC-1 ................................................................................................................ 66

4.3.3 DNA fragmentation ........................................................................................ 67

4.3.4 Caspases ......................................................................................................... 68

4.3.5 Conclusions .................................................................................................... 70

4.4 BeWo Differentiation ............................................................................................ 70

4.4.1 Confirmation of Biochemical Differentiation ................................................ 70

4.4.2 Confirmation of Morphological Differentiation ............................................ 72

4.4.3 Conclusions .................................................................................................... 74

4.5 Real Time Reverse Transcriptase Polymerase Chain Reaction ............................ 74

Chapter 5 .......................................................................................................... 76

Caspase-14 in the Trophoblast ......................................................................... 76

5.1 Background ........................................................................................................... 76

5.2 Caspase-14 in Apoptosis ....................................................................................... 76

5.3 Caspase-14 in BeWo Cell Differentiation............................................................. 78

5.3.1 Profilaggrin .................................................................................................... 80

5.3.2 Krüppel-like Factor 4 (KLF4) ........................................................................ 81

5.3.3 Cytokeratin-18................................................................................................ 84

5.3.4 Secreted Frizzled Related Protein 4 (sFRP4) ................................................. 84

5.4 Discussion ............................................................................................................. 85

Chapter 6 .......................................................................................................... 89

Endothelial Function of the Human Trophoblast ............................................... 89

6.1 Background ........................................................................................................... 89

6.2 Endothelial Biomarkers in BeWo Differentiation................................................. 90

vii

6.3 Treatment of BeWo Cell Line with Endothelial Pathway Compounds ................ 92

6.3.1 Effect of NO pathway agonists on BeWo differentiation .............................. 92

6.3.2 Endothelial Nitric Oxide Synthase (eNOS) ................................................... 94

6.3.3 Vascular Endothelial Growth Factor A (VEGF-A) ....................................... 97

6.3.4 Effect of NO pathway mediators on Caspase-14 expression ......................... 99

6.3.5 Discussion .................................................................................................... 103

Chapter 7 ........................................................................................................ 104

RNA Interference of Caspase-14 in BeWo ..................................................... 104

7.1 Background ......................................................................................................... 104

7.2 Validation of RNA Interference .......................................................................... 104

7.2.1 Knockdown of Gene Transcripts ................................................................. 105

7.2.2 Knockdown of Protein ................................................................................. 106

7.3 Caspase-14 Silencing .......................................................................................... 106

7.3.1 Optimisation ................................................................................................. 106

7.3.2 Confirmation of caspase-14 suppression following Forskolin treatment .... 108

7.3.2 Effect of caspase-14 siRNA on BeWo cells ................................................ 112

7.3.3 Effect of caspase-14 siRNA on differentiating BeWo cells......................... 118

7.3.4 Discussion .................................................................................................... 124

Chapter 8 ........................................................................................................ 128

Discussion ...................................................................................................... 128

8.1 Caspase-14 in the Human Trophoblast ............................................................... 128

8.2 Endothelial Properties of the Trophoblast and Caspase-14 ................................ 130

8.3 Concluding Remarks ........................................................................................... 131

Chapter 9- Bibliography .................................................................................. 133

Appendix I ....................................................................................................... 141

Published Articles ........................................................................................... 141

Caspase-14: A New Player in Cytotrophoblast Differentiation ................................ 141

viii

The expression of secreted Frizzled Related Protein 4 (sFRP4) in the Primate

Placenta ..................................................................................................................... 155

Appendix II ...................................................................................................... 167

Melt Curve Analysis of Real Time PCR Products ........................................... 167

List of Figures and Tables

Figure 1.1 The formation of the fetal-maternal blood barrier. With blastocyst

implantation the outer trophoblast layers form lacunae, which fill with maternal blood.

Chorionic villi invade these lacunae, inducing the formation of fetal capillaries within

their mesenchymal cores. Nutrients, gases, hormones and wastes can then pass through

the maternal-fetal barrier. .................................................................................................. 2

Figure 1.2 Comparison in morphology of human chorionic villi between the first

trimester and term. First trimester placentae characterised by large villi, thick

trophoblast and few, centrally located fetal vessels. Term placentae characterised by

many small villi, thin trophoblast and many trophoblast-associated fetal vessels. Circles

denote chorionic villi; arrows indicate trophoblast; arrowheads indicate fetal blood

vessels. .............................................................................................................................. 3

Figure 1.3 Uterine invasion by fetal trophoblasts. Fetal anchoring villi composed of

extravillous trophoblasts invade the endometrial decidua, attaching chorionic villi to the

maternal tissues. ................................................................................................................ 4

Figure 1.4 Endothelial Pathway. A typical endothelial pathway leading to dilation of

blood vessels and control of blood flow by the action of nitric oxide. Stimulation of the

pathway causes increased intracellular calcium, resulting in the eNOS-mediated

cleavage of L-arginine into nitric oxide. NO induces vasodilatation, reduced thrombosis

and reduced neutrophil adhesion. Vasodilatation can be inhibited by Reactive Oxygen

Species (ROS), which in turn are inhibited by Superoxide Dismutase (SOD) (Adapted

from Lassila 2000). ........................................................................................................... 6

Figure 1.5 Vascular endothelial growth factor (VEGF) signalling pathways leading to

angiogenesis and lymphangiogenesis. VEGF-A and -B bind the VEGFR-1 and -2 to

initiate angiogenesis. VEGF-A binding to VEGFR-2 also stimulates the NO pathway to

ix

initiate vasodilation. VEGF-C and -D bind VEGFR-3 to initiate lymph vessel formation

(Adapted from Roy et al. 2006). ....................................................................................... 8

Figure 1.6 The formation and release of apoptotic syncytial knots from the villous

syncytiotrophoblast. ........................................................................................................ 11

Figure 1.7 Cleavage and activation of caspases. Inactive caspases consist of an inactive

prodomain and long and short subunits. Activation occurs through procaspase cleavage

to form heterodimers, then aggregation into activated heterotetramers. Effector caspases

possess either a caspase recruiting domain (CARD) or a death effector domain (DED)

(Adapted from Bleackley et al. 2001). ............................................................................ 13

Figure 1.8 Schematic of the classical extrinsic and intrinsic apoptotic cascades. External

apoptotic stimuli cause Fas to bind its receptor and recruit and activate caspase-8. This

may directly or indirectly activate caspase-3 and subsequent DNA fragmentation and

apoptosis. The indirect route involves Bid cleavage stimulating mitochondrial

cytochrome c release and formation of the apoptosome. This recruits and activates

caspase-9, which in turn activates caspase-3. Intrinsic apoptotic stimuli lead to Bax, Bad

or Bim-induced mitochondrial cytochrome-c release. Again, the apoptosome is formed

and the cascade progresses as per the extrinsic pathway (Adapted from Matsuno et al.

2003). .............................................................................................................................. 14

Figure 1.9 Profilaggrin. Sequential production and processing of profilaggrin into its

highly phosphorylated protein, dephosphorylation and cleavage by caspase-14 into its

component filaggrin repeats, and final degradation into individual amino acids forming

the Natural Moisturising Factor of the corneocytes. ....................................................... 18

Figure 1.10 β-catenin signalling pathways. β-catenin is a key component in both the

maintenance of cell adhesion complexes (A) and the canonical Wnt pathway (B and C).

In adherens junctions it acts as a mediary between extracellular signalling and the actin

cytoskeleton. This competes for β-catenin with canonical Wnt pathway, where Wnt

ligand binding of Frizzled receptors leads to stabilisation of β-catenin, its nuclear

translocation, and subsequent transcription of cell cycle genes (B). Antagonism of Wnt

binding by sFRPs results in the formation of the GSKβ3/Axin/APC-complex, which

dephosphorylates and degrades β-catenin (C). ............................................................... 21

Table 2.1 Primer Accession Numbers, Sequences and Sizes. ........................................ 34

Table 2.2 Real-Time PCR Cycling Conditions. .............................................................. 35

x

Table 2.3 Antibodies used for protein detection. IHC= Immunohistochmistry; IF=

Immunofluorescence; IB= Immunoblotting. ................................................................... 37

Table 2.4 Details of caspase-14 siRNAs ......................................................................... 43

Figure 3.1 RT-PCR for caspase-14 (204bp) in the first trimester human placenta during

apoptosis or its inhibition by superoxide dismutase. ...................................................... 47

Figure 3.2 The expression of β-hCG mRNA in the human placenta (*=P<0.05 vs. First

Trimester). ....................................................................................................................... 48

Figure 3.3 The change in β-hCG mRNA expression throughout gestation (R=-0.63). .. 48

Figure 3.4 Localisation of A) hCG and B) hPL in the first trimester human placenta

(CTB=cytotrophoblasts; STB=syncytiotrophoblast). ..................................................... 49

Figure 3.5 The expression of hPL mRNA in the human placenta (*=P<0.05 vs. First

Trimester). ....................................................................................................................... 49

Figure 3.6 The change in hPL mRNA expression throughout gestation (R=0.62). ........ 50

Figure 3.7 The expression of caspase-14 mRNA in the human placenta (*=P<0.05 vs.

First Trimester)................................................................................................................ 51

Figure 3.8 The change in caspase-14 mRNA expression throughout gestation (R=0.13).

......................................................................................................................................... 52

Figure 3.9 Immunohistochemistry for caspase-14 in the chorionic villi of first trimester

and term human placentae. Caspase-14 is expressed exclusively in the trophoblast bi-

layer, particularly in the first trimester. ........................................................................... 52

Figure 3.10 The expression of filaggrin mRNA in the human placenta. ........................ 53

Figure 3.11 Filaggrin mRNA expression compared between the human placenta and

epidermis (*=P<0.05 vs. 1st Trimester, Term and Preeclampsia). .................................. 53

Figure 3.12 The expression of KLF4 mRNA in the human placenta (*=P<0.05 vs. First

Trimester; †=P<0.05 vs. First Trimester and Term). ...................................................... 54

Figure 3.13 The change in KLF4 mRNA expression throughout gestation (R=0.15). ... 55

Figure 3.14 The expression of cytokeratin-18 mRNA in the human placenta. ............... 55

Figure 3.15 The change in cytokeratin 18 mRNA expression throughout gestation (R=-

0.27). ............................................................................................................................... 56

Figure 3.16 The expression of sFRP4 mRNA in the human placenta (P>0.05). ............ 57

xi

Figure 3.17 The change in sFRP4 mRNA expression throughout gestation (R=0.22). .. 57

Figure 3.18 Immunohistochemistry for sFRP4 in the chorionic villi of first trimester and

term human placentae. sFRP4 is localised extensively to the syncytio-trophoblast

throughout gestation. ....................................................................................................... 58

Figure 3.19 The expression of eNOS mRNA in the human placenta (*=P<0.05 vs.

Term). .............................................................................................................................. 59

Figure 3.20 The change in eNOS mRNA expression throughout gestation (R=-0.5). ... 59

Figure 3.21 The expression of VEGF-A mRNA in the human placenta. ....................... 61

Figure 3.22 The change in VEGF-A mRNA expression throughout gestation (R=0.1). 61

Figure 4.1 Paraffin embedded BeWo cells stained with A) haematoxylin and eosin to

show morphology, and B) hCG primary antibody (brown). ........................................... 64

Figure 4.2 Cytokeratin-18 mRNA expression in BeWo cells. ........................................ 64

Figure 4.3 Staurosporine dose-response curve (* = P<0.05 vs. Control; † = P<0.05 vs.

Control and 0.1µM)......................................................................................................... 66

Figure 4.4 Effect of 6 hours of A) DMSO and B) 1µM Staurosporine treatment on the

morphology of BeWo cells. ............................................................................................ 66

Figure 4.5 Quantification of BeWo apoptosis by JC-1 analysis after 1 and 3 hours of

Staurosporine (STS) treatment (* = P<0.05 vs. vehicle and 1µM STS; † = P<0.05 vs.

vehicle and FCCP). ......................................................................................................... 67

Figure 4.6 3‟-end labelling. A) DNA fragmentation following treatment of BeWo cells

with DMSO (i and iii), or 1μM Staurosporine (ii and iv) for 3 or 6 hours. B) Graphical

representation of incorporated radiolabel in each group (* = P<0.05 vs. Control). ........ 68

Figure 4.7 Caspase-8 mRNA expression in Staurosporine-treated BeWo cells (* =

P<0.05 v. Control). .......................................................................................................... 69

Figure 4.8 RT-PCR for caspase-3 in Staurosporine-treated BeWo. ............................... 69

Figure 4.9 A) Expression of β-hCG mRNA in BeWo cells following Forskolin

treatment. B) Western blot for β-hCG protein in Forskolin treated BeWo cells, and C)

Quantitation of corresponding Western Blot (B) (* = P<0.05 v. Control). .................... 71

Figure 4.10 Expression of hPL mRNA in BeWo cells following Forskolin treatment (*

= P<0.05 v. Control). ...................................................................................................... 72

xii

Figure 4.11 Expression of E-cadherin mRNA in BeWo cells following Forskolin

treatment (* = P<0.05 v. Control). .................................................................................. 73

Figure 4.12 Confirmation of morphological differentiation. Immunofluorescence for E-

cadherin in 48 h (A−C) control and (D−F) Forskolin treated BeWo cells. A) and D)

show Hoechst labelled nuclei; B) and E) show E-cadherin staining; C) and F) show

overlays. Arrows indicated membrane-associated E-cadherin. Arrowheads indicate

diffuse E-cadherin. Original magnifications: 400x ......................................................... 73

Figure 4.13 Expression of caspase-8 mRNA in BeWo cells following Forskolin

treatment (*=P<0.05 v. Control). .................................................................................... 74

Figure 4.14 Expression of Ki67 mRNA in BeWo cells following Forskolin treatment

(*=P<0.05 v. Control). .................................................................................................... 75

Figure 5.1 A) Expression of caspase-14 mRNA in BeWo cells following Staurosporine

treatment. B) Western blot of procaspase-14 protein in BeWo cells following

Staurosporine treatment. C) Quantitation of the corresponding Western blot (B). Note

there is no significant difference between any of the groups. ......................................... 77

Figure 5.2 A) Expression of caspase-14 mRNA in BeWo cells following Forskolin

treatment. B) Western blot of procaspase-14 protein and internal control GAPDH in

BeWo cells following Forskolin treatment; and C) quantitation of the corresponding

Western blot (B) (* = P<0.05 v. Control). ...................................................................... 79

Figure 5.3 Expression of Filaggrin mRNA in BeWo cells following Forskolin treatment.

The HaCaT cell line (epidermal) was used as a positive control, and is plotted on the

secondary vertical axis. ................................................................................................... 80

Figure 5.4 A) Expression of KLF4 mRNA in BeWo cells following Forskolin treatment

(*=P<0.05 v. Control). B) Western blot analysis of KLF4 and internal control GAPDH

in BeWo cells following Forskolin treatment, and C) quantification of the corresponding

Western blot (B) (*=P<0.05 v. Control). ........................................................................ 82

Figure 5.5 A) Expression of cytokeratin-18 mRNA in BeWo cells following Forskolin

treatment (*=P<0.05 v. Control). B) Western blot analysis of cytokeratin-18 and

internal control GAPDH in BeWo cells following Forskolin treatment, and C)

quantification of the corresponding Western blot (B) (*=P<0.05 v. Control). ............... 83

Figure 5.6 Expression of sFRP4 mRNA in BeWo cells following Forskolin treatment. 85

xiii

Figure 6.1 Expression of eNOS mRNA in BeWo cells following Forskolin treatment

(*=P<0.05 v. Control). .................................................................................................... 91

Figure 6.2 Expression of VEGF-A mRNA in BeWo cells following Forskolin treatment

(*=P<0.05 v. Control). .................................................................................................... 91

Figure 6.3 Effect of Calcium treatment on β-hCG mRNA expression in Forskolin

mediated BeWo differentiation. L-NAME acts as an inhibitor of the NO pathway. ...... 93

Figure 6.4 Effect of VEGF treatment on β-hCG mRNA expression in Forskolin


Figure 6.5 Effect of Bradykinin treatment on β-hCG mRNA expression in Forskolin


Figure 6.6 Expression of eNOS mRNA in Forskolin treated BeWo cells following

treatment with nitric oxide pathway modulators (* = P<0.05 vs. 0h and 24h). .............. 95

Figure 6.7 Effect of Calcium treatment on eNOS mRNA expression in Forskolin


Figure 6.8 Effect of VEGF treatment on eNOS mRNA expression in Forskolin mediated

BeWo differentiation (* = P<0.05 vs. Control and VEGF+L-NAME). L-NAME acts as

an inhibitor of the NO pathway. ...................................................................................... 96

Figure 6.9 Effect of Bradykinin treatment on eNOS mRNA expression in Forskolin

mediated BeWo differentiation (* = P<0.05 vs. Control and VEGF+L-NAME). L-

NAME acts as an inhibitor of the NO pathway. ............................................................. 96

Figure 6.10 Effect of Calcium treatment on VEGF-A mRNA expression in Forskolin

mediated BeWo differentiation (* = P<0.05 vs. Control). L-NAME acts as an inhibitor

of the NO pathway. ......................................................................................................... 97

Figure 6.11 Effect of VEGF treatment on VEGF-A mRNA expression in Forskolin

mediated BeWo differentiation (* = P<0.05 vs. Control and VEGF; † = P<0.05 vs.

Control). L-NAME acts as an inhibitor of the NO pathway. .......................................... 98

Figure 6.12 Effect of Bradykinin treatment on VEGF-A mRNA expression in Forskolin

mediated BeWo differentiation (* = P<0.05 vs. Control). L-NAME acts as an inhibitor

of the NO pathway. ......................................................................................................... 98

xiv

Figure 6.13 Effect of the NO pathway inhibitor L-NAME on caspase-14 mRNA

expression in Forskolin mediated BeWo differentiation (* = P<0.05 vs. Control). No

data was obtained for 48h L-NAME. .............................................................................. 99

Figure 6.14 Effect of calcium treatment on caspase-14 mRNA expression in Forskolin

mediated BeWo differentiation (* = P<0.05 vs. Control; † = P<0.05 vs. both Control

and Ca2+

). L-NAME acts as an inhibitor of the NO pathway. ...................................... 100

Figure 6.15 Effect of VEGF-A treatment on caspase-14 mRNA expression in Forskolin

mediated BeWo differentiation (* = P<0.05 vs. Control; † = P<0.05 vs. both Control

and VEGF). L-NAME acts as an inhibitor of the NO pathway. ................................... 101

Figure 6.16 Effect of bradykinin treatment on caspase-14 mRNA expression in

Forskolin mediated BeWo differentiation (* = P<0.05 vs. Control; † = P<0.05 vs. both

Control and Bradykinin). L-NAME acts as an inhibitor of the NO pathway. .............. 102

Figure 6.17 Expression of Caspase-14 mRNA in response to nitric oxide pathway

modulators in BeWo cells after 72 hours of Forskolin treatment (*=P<0.05 v. Control;

†=P<0.05 v. Control and treatment). ............................................................................. 102

Figure 7.1 Silencing of GAPDH mRNA at various concentrations of siRNA (*=P<0.05

vs. untreated). ................................................................................................................ 105

Figure 7.2 Silencing of GAPDH protein following 24 and 48 hours of siRNA treatment

(* = P<0.05 v. Lipo) (i = 24h Control; ii = 24h GAPDH siRNA; iii = 48h Control; iv =

48h GAPDH siRNA). ................................................................................................... 106

Figure 7.3 Testing of 3 sets of 100nM caspase-14 siRNA in the BeWo cell line

(*=P<0.05 vs. scrambled siRNA). Set #1 was found to be most effective in silencing

caspase-14 mRNA expression. ..................................................................................... 107

Figure 7.4 Knock down of caspase-14 mRNA at various concentrations of siRNA

(*=P<0.05 vs. untreated). .............................................................................................. 108

Figure 7.5 Dose-response curve for caspase-14 siRNA in the BeWo cell line (R=-0.88).

....................................................................................................................................... 108

Figure 7.6 A) Expression of caspase-14 mRNA in BeWo cells following treatment with

mock or caspase-14 siRNA. B) Western blot of caspase-14 protein in BeWo cells after

treatment with siRNA. „Mock‟ stands for treatment with mock siRNA; „C14‟ stands for

treatment with caspase-14 siRNA. C) Quantitation of corresponding Western Blot (B)

(* = P<0.05 vs. Mock siRNA). ..................................................................................... 109

xv

Figure 7.7 A) Expression of caspase-14 mRNA in differentiating BeWo cells following

treatment with mock or caspase-14 siRNA. B) Western blot of caspase-14 protein in

differentiating BeWo cells after treatment with siRNA. „Mock‟ stands for treatment

with mock siRNA; „C14‟ stands for treatment with caspase-14 siRNA. C) Quantitation

of corresponding Western Blot (B) (* = P<0.05 vs. Mock siRNA). ............................. 111

Figure 7.8 A) Expression of β-hCG mRNA in BeWo cells following treatment with

mock or caspase-14 siRNA. B) Western blot of β-hCG protein in BeWo cells after



(* = P<0.05 vs. Mock siRNA). ..................................................................................... 113

Figure 7.9 Expression of hPL mRNA in BeWo cells following treatment with mock or

caspase-14 siRNA. ........................................................................................................ 114

Figure 7.10 Expression of eNOS mRNA in BeWo cells following treatment with mock

or caspase-14 siRNA (* = P<0.05 v. mock siRNA). .................................................... 114

Figure 7.11 A) Expression of KLF4 mRNA in BeWo cells following treatment with

mock or caspase-14 siRNA. B) Western blot of KLF4 protein in BeWo cells after



(* = P<0.05 vs. Mock siRNA). ..................................................................................... 115

Figure 7.12 A) Expression of cytokeratin-18 mRNA in BeWo cells following treatment

with mock or caspase-14 siRNA. B) Western blot of cytokeratin-18 protein in BeWo

cells after treatment with siRNA. „Mock‟ stands for treatment with mock siRNA; „C14‟

stands for treatment with caspase-14 siRNA. C) Quantitation of corresponding Western

Blot (B) (* = P<0.05 vs. Mock siRNA). ....................................................................... 116

Figure 7.13 Expression of E-cadherin mRNA in BeWo cells following treatment with

mock or caspase-14 siRNA. .......................................................................................... 117

Figure 7.14 A) Expression of β-hCG mRNA in differentiating BeWo cells following

treatment with mock or caspase-14 siRNA. B) Western blot of β-hCG protein in




xvi

Figure 7.15 Expression of hPL mRNA in differentiating BeWo cells following

treatment with mock or caspase-14 siRNA. .................................................................. 120

Figure 7.16 Expression of eNOS mRNA in differentiating BeWo cells following

treatment with mock or caspase-14 siRNA (* = P<0.05 v. mock siRNA). .................. 120

Figure 7.17 A) Expression of KLF4 mRNA in differentiating BeWo cells following

treatment with mock or caspase-14 siRNA. B) Western blot of KLF4 protein in




Figure 7.18 A) Expression of cytokeratin-18 mRNA in differentiating BeWo cells

following treatment with mock or caspase-14 siRNA. B) Western blot of cytokeratin-18

protein in differentiating BeWo cells after treatment with siRNA. „Mock‟ stands for

treatment with mock siRNA; „C14‟ stands for treatment with caspase-14 siRNA. C)

Quantitation of corresponding Western Blot (B) (* = P<0.05 vs. Mock siRNA)......... 122

Figure 7.19 Expression of E-cadherin mRNA in differentiating BeWo cells following

treatment with mock or caspase-14 siRNA (* = P<0.05 v. mock siRNA). .................. 123

Table 7.1 A summary of the effects of caspase-14 siRNA treatment on the BeWo cell

line. Arrows indicate the direction of any significant change in mRNA or protein

expression (P<0.05). Dashes indicate no significant change (P>0.05). NA indicates no

available data. “Ck-18” stands for “cytokeratin-18”. .................................................... 125

xvii

Abbreviations

AP-1 Activator Protein 1

APC Adenomatous Polyposis Coli

APS ammonium persulfate

Bk Bradykinin

BSA Bovine Serum Albumin

CARD Caspase Recruitment Domain

cdk Cyclin-dependent kinase

CO2 Carbon dioxide

Da Dalton

DED Death Effector Domain

dH2O Distilled water

DISC Death Inducing Signalling Complex

DMSO Dimethyl Sulfoxide

DNA Deoxyribonucleic Acid

eNOS Endothelial nitric oxide synthase

GSK3β Glycogen Synthase Kinase 3 Beta

hCG Human chorionic gonadotrophin

JNK c-Jun

k kilo

KLF4 Kruppel-like Factor 4

LEF Lymphoid Enhancer Factor

L-NAME N (G)-nitro-L- arginine methyl ester

M Molar

MAPK Mitogen Activating protein Kinase

mRNA Messenger Ribonucleic Acid

PBS Phosphate Buffered Solution

PCR Polymerase Chain Reaction

RIPA Radioimmunoprecipitation

RNA Ribonucleic Acid

rpm Revolutions per minute

RT Reverse Transcriptase

sFlt1 soluble fms-like tyrosine kinase 1

xviii

sFRP4 Secreted Frizzled-related Protein 4

SOD Sulfoxide Dismutase

TBS-T Tris Buffered Saline (0.01% Tween-20)

TCF T-Cell Factor

VEGF Vascular Endothelial Growth Factor

μ micro

xix

Abstract

The placenta forms a barrier regulating the transfer of gases, nutrients and wastes

between the mother and the developing conceptus, and also produces hormones

affecting both the fetus and the mother. This barrier is formed by the differentiation of

the outer layer of the blastocyst- the trophoblast- to facilitate implantation and

subsequent invasion of the uterus. The trophoblast consists of an underlying

proliferative pool of cytotrophoblasts, which differentiate to replenish the overlying

continuous, multi-nucleated syncytiotrophoblast that forms the barrier between the

mother and fetus. Moreover, the location of the syncytiotrophoblast directly in contact

with the maternal circulation suggests an endothelial role for the trophoblast regulating

blood flow, thrombosis and immune cell adhesion.

Disruption to the function of the human trophoblast may result in preeclampsia,

a maternally manifested disorder of pregnancy characterised by hypertension and

proteinurea. Blood flow to preeclamptic placentae is reduced and the cytotrophoblast

pool is diminished; however the exact cause (or causes) remains elusive. Many potential

causes are hypothesised, including endothelial damage, premature remodelling of

maternal spiral arteries, increased oxidative stress and impaired trophoblast

differentiation and apoptosis.

Caspase-14 is an unusual caspase in that it is not involved in apoptosis.

Furthermore, it possesses a limited, predominantly epithelial, tissue distribution. In the

epidermis, caspase-14 is expressed in the apical differentiating layers. Here it cleaves

profilaggrin to stabilise intracellular keratin intermediate filaments, and indirectly

provides natural hydration and UV protection to the corneocytes. Thus, caspase-14 is

vital to the maintenance of the barrier function of the skin.

No studies have examined the role of caspase 14 in the placenta. In light of its

function in keratinocyte differentiation and barrier formation, it is hypothesised that

caspase-14 is conserved in barrier formation, including the human trophoblast.

Concordantly, this thesis investigates the role of caspase-14 in the trophoblast,

particularly in relation to differentiation and syncytialisation. This was achieved by

analysing first trimester, term and preeclamptic human trophoblast samples for several

differentiation-associated genes; the manipulation of the cytotrophoblast-like BeWo cell

line to undergo apoptosis and differentiation; and RNA Interference to elucidate the role

of caspase-14 during BeWo differentiation.

The expression of caspase-14 in the human trophoblast across gestation, together

with genes relating to hormone production and endothelial function, was confirmed by

xx

the use of Real Time PCR and Western blot analysis. Furthermore, the pattern of

expression of caspase-14 and genes involved in hormonal and endothelial functions are

tightly regulated by trophoblast differentiation as analysed by controlled manipulation

of the BeWo cell line. However, profilaggrin is not expressed in the human trophoblast,

indicating that other proteins are likely to be targeted by caspase-14 in the human

trophoblast.

Suppression of caspase-14 was undertaken using RNA Interference. The

synthesis of the hormone human chorionic gonadotrophin (hCG) and the key mediator

of endothelial function eNOS was further enhanced by the absence of caspase-14 during

differentiation. As differentiation-associated genes were elevated in the absence of

caspase-14, this implies that caspase-14 suppresses biochemical trophoblast

differentiation.

The cytoskeletal keratin network was also examined following RNA

Interference. The synthesis of cytokeratin 18 was significantly enhanced after caspase-

14 suppression during BeWo differentiation, linking caspase-14 with keratin

homeostasis. Therefore caspase-14 suppresses trophoblast differentiation, potentially

through modulation of the cytoskeletal keratin filament network. The precise

mechanism remains to be elucidated, however the identification of pathways regulated

by caspase-14 advances our knowledge of trophoblast differentiation and potential

causes of disorders of pregnancy.

In summary, caspase-14 appears to be involved in the suppression of

differentiation in the human trophoblast. As disorders of pregnancy such as

preeclampsia often feature disturbed differentiation and a diminished cytotrophoblast

pool, a greater understanding of caspase-14 biology in the human placenta could lead

potential therapies for various disorders of pregnancy.

xxi

Acknowledgements

I would like to thank my supervisors Professor Arun (Dharma) Dharmarajan and Doctor

Adrian Charles for their vital input into this thesis. Without their ideas and direction this

work would not have been possible.

This thesis would not have been possible without funding from the Raine

Foundation, the School of Anatomy and Human Biology for my stipend for 2 years the

University of Western Australia for my University Postgraduate Award (UPA) for 18

months. Furthermore, I would sincerely like to thank the WA Freemasons, especially

Jean and Peter Stokes, and John Leslie and Dorise Barron, for providing considerable

personal financial assistance, and Eric and Yvonne Phillips for sponsoring me through

the scholarship application process.

To the mothers who generously donated their placenta‟s to be used in the

generation of this thesis, my sincerest thanks. Also thanks to the staff at the Marie

Stopes Clinic in Midland WA for allowing me access to their facilities.

Four cDNA samples from both term and preeclamptic human placentae were

provided by Professor Robert Friis of the University of Bern, Switzerland.

The primary antibody raised against caspase-14 was a generous donation from

Dr. Wim Declercq of the University of Ghent, Belgium.

Special thanks go to Gopi Krishna Kolluru for the establishment of the Griess

and DAF assays, from which the cDNA was obtained for the bulk of Chapter 6

(Trophoblast Endothelial Function).

Considerable assistance in the lab was provided by Greg Cozens, Peter Mark

and the remainder of the Cancer Biology/Apoptosis Lab (especially Simon Mahoney)

co-ordinated by Dharma.

On a more personal note I wish to thank Tash Commons for her enduring love

and support throughout this process. For seeing me through and keeping me calm.

xxii

Declaration

The work presented within this thesis was completed between March 2005 and

September 2008 in the School of Anatomy and Human Biology at the University of

Western Australia. I hereby declare that all work presented is entirely my own, unless

explicitly stated otherwise. All contributions by others are formally disclosed and duly

acknowledged.

Lloyd J. White

12th

September 2008

Chapter 1- Introduction

1

Chapter 1

Introduction

1.1 Placenta

The placenta is a transient organ primarily responsible for the exchange of nutrients,

gases and wastes between the mother and developing fetus. It is formed by the

differentiation of the outer layer of the blastocyst after implantation and invasion into

the mother‟s uterus. The outer trophoblast layer is responsible for this invasion, and

develops to form a continuous barrier that facilitates the exchange of gases, nutrients,

hormones and wastes between the mother and fetus.

Chorionic villi protrude into the maternal tissue to form a bi-directional barrier

with a large surface area for exchange between the maternal and fetal circulations

(Figure 1.1). These villi are covered by cells of the two layers of the trophoblast, the

inner cytotrophoblast and overlying syncytiotrophoblast. The cytotrophoblast provides

the proliferative pool of cells to fuse with and replenish the overlying

syncytiotrophoblast, which is appositional with the maternal circulation. The

syncytiotrophoblast is a continuous multinucleated syncytium which structurally and

functionally separates the maternal circulation from the fetal tissues. As well as forming

a functional barrier, it also fulfils hormonal and endothelial functions and protects the

fetus against maternal pathogens (Reviewed in Gude et al. 2004).

The fetal placental vasculature starts growing from Day 21 in a relatively

hypoxic environment, and continues throughout gestation. At the commencement of this

process, primary villi are composed of vast numbers of cytotrophoblasts covered with a

thick layer of multinucleated syncytiotrophoblast. A core of connective tissue descends,

folding the trophoblast layers into secondary villi which protrude into the space between

the fetal and maternal tissues. Masses of endothelial precursor cells then invade this

stromal core to form primitive blood vessels in what are now referred to as tertiary villi.

By the 28th

day of gestation a lumen is visible within the core. By 32 days new vessels

are sprouting from the distal ends of the existing vessels, and by 6 weeks a basement

membrane forms around the villous vessels. Development of this vascular system is

seen until week 12, from whence the capillaries coil, bulge, form sinusoids and protrude

toward the trophoblast layer. Here they form a membrane separating the placental


2

vasculature from the trophoblast and facilitating the exchange between the fetal and

maternal circulatory systems.

Figure 1.1 The formation of the fetal-maternal blood barrier. With blastocyst

implantation the outer trophoblast layers form lacunae, which fill with maternal blood.

Chorionic villi invade these lacunae, inducing the formation of fetal capillaries within

their mesenchymal cores. Nutrients, gases, hormones and wastes can then pass through

the maternal-fetal barrier.

In the first trimester the embryo receives its nutrition through trophoblast

phagocytosis of endometrial glandular secretions, including glycogen and glycoproteins.

After 12 weeks, when the uterine spiral arteries are opened, nutrition is gained from

maternal blood. The exchange of gases between the mother and fetus is further

facilitated through fetal haemoglobin having a greater affinity for oxygen and a lower


3

affinity for CO2 than maternal haemoglobin, favouring the transfer of oxygen to the

fetus and CO2 to the mother (Reviewed in Gude et al. 2004).

Figure 1.2 Comparison in morphology of human chorionic villi between the first

trimester and term. First trimester placentae characterised by large villi, thick

trophoblast and few, centrally located fetal vessels. Term placentae characterised by

many small villi, thin trophoblast and many trophoblast-associated fetal vessels. Circles

denote chorionic villi; arrows indicate trophoblast; arrowheads indicate fetal blood

vessels.

In accordance with the changing demands, morphological changes accompany

the maturation of the placenta across gestation (Figure 1.2). Chorionic villi become

progressively smaller and more numerous to increase surface area, while the

cytotrophoblast is reduced due to its continual incorporation into the

syncytiotrophoblast. Accordingly, the proliferative potential of the placenta decreases

over time. Additionally, the villous blood vessels become more closely associated with

the trophoblast basement membrane, aiding the exchange of materials across the barrier.

Accompanying the increased surface area, nucleus extrusion from the syncytium

symptomatic of cell death increases (Huppertz et al. 1998).


4

1.1.1 Extravillous Trophoblast

During early gestation, there are two sub-sets of cytotrophoblasts, both with their own

discrete differentiation pathways (James et al. 2005). Classical villous differentiation

leads to fusion of one of these cytotrophoblast populations to form the typical

trophoblast bi-layer, with the syncytiotrophoblast in direct apposition with the maternal

blood (Figure 1.1). Other population however are able to differentiate to invade the

maternal tissues and anchor the chorionic villi to the uterus (Figure 1.3) (James et al.

2005). These are known as extravillous trophoblasts.

Figure 1.3 Uterine invasion by fetal trophoblasts. Fetal anchoring villi composed of

extravillous trophoblasts invade the endometrial decidua, attaching chorionic villi to the

maternal tissues.

Optimal trophoblast invasion is dependent upon a hypoxic environment (James

et al. 2005). To create this hypoxic environment, a sub-set of extravillous trophoblasts

differentiate and invade further to occlude the uterine spiral arteries, where they restrict

the maternal blood entering sites of exchange (Huppertz et al. 2005; James et al. 2005;

Huppertz et al. 2006). These endovascular trophoblasts also remodel the spiral arteries

to provide ample blood to the fetus following the unblocking of the uterine spiral


5

arteries (Huppertz et al. 2005). Indeed, inadequate invasion or spiral artery remodelling

due to poor regulation of oxygen tension may lead to preeclampsia or intrauterine

growth retardation through inappropriate perfusion of oxygenated blood to the villi

(Huppertz et al. 2005).

Due to the invasive and hypoxia-control functions of the extravillous

trophoblast, particularly during the first trimester of gestation, a considerable

endothelial phenotype is evident, reinforcing the need for endothelial functions of the

trophoblast. Moreover, the villous cytotrophoblasts must also possess endothelial

features as it is also in direct contact with the maternal blood stream.

1.1.2 Hormones

The villous trophoblast is responsible for the production and secretion of a variety of

hormones into the maternal circulation. Two of the major hormones produced by the

trophoblast are human chorionic gonadotrophin (hCG) and human placental lactogen

(hPL) (Kliman et al. 1986). hCG consists of an alpha-subunit that is identical to that of

luteinising hormone (LH), follicle stimulating hormone (FSH) and thyroid stimulating

hormone (TSH). It is the beta subunit of hCG however that confers hormonal

specificity. During the first trimester, hCG is involved in the maintenance of pregnancy

and the continual maintenance of the syncytiotrophoblast (Shi et al. 1993; Cronier et al.

1994; Yang et al. 2003). Indeed it is the most widely used marker of pregnancy. hCG

stimulates the corpus luteum, preventing its involution and maintaining the luteal

production of progesterone and oestrogen, preventing the onset of menses. From 12

weeks of gestation however, hCG production is greatly diminished as production of

oestrogen and progesterone by the trophoblast is escalated. Thus hCG is no longer

necessary for survival after the first trimester.

While hPL is produced in the trophoblast, its target organs and effects are far

reaching. Along with regulating fetal development, hPL is responsible for increasing

maternal metabolism to support the development of the fetus, and in preparing the

breast for lactation. In line with the increasing maternal metabolism, production of hPL

increases with advancing gestation.

1.1.3 Endothelia

Typically, endothelial cells line the lumen of blood vessels and react to and modulate

blood pressure, cytokines and immune responses within the vessel and the tissue it


6

supplies. The primary pathway of endothelial function involves the production and

secretion of nitric oxide (NO) to elicit a vasodilatory response in the underlying smooth

muscle (Reviewed in Moncada et al. 1991; Lassila 2000).

Figure 1.4 Endothelial Pathway. A typical endothelial pathway leading to dilation of

blood vessels and control of blood flow by the action of nitric oxide. Stimulation of the

pathway causes increased intracellular calcium, resulting in the eNOS-mediated

cleavage of L-arginine into nitric oxide. NO induces vasodilatation, reduced thrombosis

and reduced neutrophil adhesion. Vasodilatation can be inhibited by Reactive Oxygen

Species (ROS), which in turn are inhibited by Superoxide Dismutase (SOD) (Adapted

from Lassila 2000).

Figure 1.4 outlines the basic endothelial pathway, whereby the lumenal surface

of the endothelial cell may come into direct contact with a variety of chemical (e.g.

bradykinin; VEGF) or physical stimuli (e.g. shear stress). Upon such stimulation,

intracellular calcium (Ca2+

) concentrations are elevated, leading to endothelial nitric

oxide synthase (eNOS) mediated cleavage of L-arginine into L-citrulline and the

physiologically active nitric oxide (NO) gas. NO is released in a paracrine manner into

the lumen or underlying smooth muscle cells. In smooth muscle, NO activates guanylate

cyclise (GC), catalysing the production of cyclic guanosine 3‟-5‟-monophosphate

(cGMP), leading to smooth muscle relaxation and consequent vasodilation. This results

in increased blood flow to the target tissue. Additional consequences of the endothelial

NO pathway within the placenta include decreased platelet aggregation and decreased

neutrophil adhesion (Eis et al. 1995), as well as the suppression of hCG production

(Myat et al. 1996). Thus the regulation of endothelial pathways in the placenta must be

tightly regulated to mediate vasotone, immune functions and hormone production.


7

1.1.3.1 Reactive Oxygen Species

NO has a very short half life (0.1sec) in vivo as it is rapidly scavenged by the superoxide

anion and other reactive oxygen species (ROS) (Figure 1.4) and inhibited by

oxyhemoglobin. Therefore the action of NO is local. ROS thereby inhibit vasodilatation

through the scavenging of NO. An imbalance in the NO/ROS system may lead to

oxidative stress. However, ROS themselves are scavenged by a variety of enzymes such

as superoxide dismutase (SOD), a free radical species responsible for the scavenging

and conversion of the superoxide anion (O2-) into H2O2 for removal from the system,

reducing oxidative stress and suppressing subsequent cell death (Sikkema et al. 2001).

1.1.3.2 Endothelial Nitric Oxide Synthase

Nitric oxide production is catalysed by the Nitric Oxide Synthase (NOS) family of

proteases, which consists of the neural (nNOS), inducible (iNOS) and endothelial

(eNOS) members. As a vital component of the vasodilatory NO pathway, eNOS is

restricted to endothelial cells. As the trophoblast layer of the placenta is in direct contact

with the maternal blood space, it is reasonable to hypothesise an endothelial role for the

syncytiotrophoblast. Indeed, eNOS expression and NO activity are increased with the

differentiation of villous cytotrophoblasts (Conrad et al. 1993; Eis et al. 1995;

Rossmanith et al. 1999; Orange et al. 2003), supporting a considerable endothelial role

for the syncytiotrophoblast in the human placenta. The low perfusion pressure within

the placenta may cause thrombosis, however a trophoblast eNOS/NO system could

conceivably suppress this by preventing platelet and leukocyte adhesion and

aggregation (Eis et al. 1995). Although terminal fetal capillaries in chorionic villi do not

express eNOS (Orange et al. 2003), syncytial eNOS/NO production could conceivably

interact with the fetal capillaries to regulate fetal blood flow into discrete chorionic villi.

Therefore the syncytiotrophoblast possesses a significant endothelial-like function to

regulate perfusion and gas exchange between maternal and fetal tissues.

1.1.3.3 Vascular Endothelial Growth Factor

Interestingly, vascular endothelial growth factor (VEGF) is also expressed in the human

trophoblast, particularly during early gestation when oxygen tension control is

paramount (Jia et al. 2006). VEGF is a 6-member family of secreted multi-functional

pro-angiogenic modulators of the eNOS/NO pathway capable of inducing mitogenesis,

endothelial chemotaxis, induction of extracellular matrix degradation, increasing

vascular permeability, and vasodilation (Zygmunt et al. 2003; Roy et al. 2006). VEGF-


8

A, -B and -E are involved in the formation of new blood vessels, while VEGF-C and -D

control the formation of lymph vessels (Figure 1.5) (Meyer et al. 1999; Roy et al. 2006).

VEGF-F is derived from snake venom (Suto et al. 2005).

Figure 1.5 Vascular endothelial growth factor (VEGF) signalling pathways leading

to angiogenesis and lymphangiogenesis. VEGF-A and -B bind the VEGFR-1 and -2

to initiate angiogenesis. VEGF-A binding to VEGFR-2 also stimulates the NO pathway

to initiate vasodilation. VEGF-C and -D bind VEGFR-3 to initiate lymph vessel

formation (Adapted from Roy et al. 2006).

At least 3 distinct transmembrane VEGF receptors (VEGFR) modulate the

signalling of the VEGF isoforms. The VEGF members bind to the VEGFR-1, -2, and -3

in a specific pattern to exert their effects (Roy et al. 2006). VEGFR-1 and especially

VEGFR-2 are responsible for the pro-angiogenic actions of VEGF-A and -B, while

VEGFR-3 regulates lymphangiogenesis via its binding of VEGF-C and -D (Figure 1.5)

(Zygmunt et al. 2003; Roy et al. 2006). Furthermore, VEGFR-1 and VEGFR-2 can

cross-talk via homo- and hetero-dimers to modulate the signalling properties of VEGF-

A (Huang et al. 2001). Ligand binding of VEGFR-1 may be antagonised by its secreted

form soluble fms-like tyrosine kinase 1 (sFlt-1), which has also been implicated in the

pathogenesis of preeclampsia (Maynard et al. 2003).


9

VEGF-A consists of at least 6 separate transcripts which differ in the splicing of

exons 6 and 7. They increase endothelial permeability via an NO-dependent cGMP

pathway by binding to transmembrane VEGFR-2. VEGF-A may also promote

vasodilation through the induction of eNOS and subsequent production of NO (Roy et

al. 2006). Both VEGFR-1 and VEGFR-2 are expressed on the trophoblast and placental

endothelia of both first trimester and term placentae, and in multiple trophoblast cell

lines (Charnock-Jones et al. 1994; Zygmunt et al. 2003), indicating significant

endothelial-like properties of the trophoblast. Treatment with VEGF leads to increased

production of both eNOS and NO in the trophoblast BeWo cell line resulting in the

suppression of proliferation (Cha et al. 2001). Furthermore, VEGF-A expression in the

placenta may also be regulated by oxygen tension, growth factors, cytokines and

placental hormones (Zygmunt et al. 2003). Consequently, VEGF-A is the primary focus

of VEGF biology in the current study.

1.1.4 Preeclampsia

Preeclampsia is a maternally manifested disorder of pregnancy diagnosed by maternal

hypertension and proteinuria, occurring in around 5% of pregnancies and is a significant

cause of maternal and fetal morbidity and mortality. It is well established that maternal

blood flow to the placenta is reduced in preeclampsia, however the exact cause of this is

unknown (Fox et al. 2007).

Several theories have been proposed for preeclampsia, including endothelial

damage, premature opening of uterine spiral arteries, increased oxidative stress

(Poranen et al. 1996; Takagi et al. 2004), reduced extravillous trophoblast invasion of

the uterine spiral arteries (Kadyrov et al. 2006), and increased trophoblast apoptosis and

syncytial shedding (Ishihara et al. 2002; Gupta et al. 2005). It appears that these

processes compound to manifest in preeclampsia (Huppertz et al. 2003; Huppertz et al.

2006).

Characteristic histological features include cytotrophoblast hyperplasia and

thickening of the trophoblast basement membrane, the severity of these appearing

related to the severity or preeclampsia (Fox et al. 2007). Furthermore, reduced

trophoblast secretion and localised loss of syncytial microvilli are also observed. The

placental changes seen in typical early onset preeclampsia include villous ischemia and

thrombosis. The maternal spiral arteries also remain muscularised, leading to atherosis

and the deposition of fibrin in the decidua. This reflects a failure in spiral artery


10

remodelling. These same changes may also be seen in intrauterine growth restriction

and some still births.

1.1.5 In Vitro Trophoblast Models

The examination of signalling pathways within the trophoblast can be conducted

through the use of primary cultures and cell lines. Upon their isolation and purification,

primary trophoblast cells undergo spontaneous differentiation, inducing the production

of hCG and subsequent fusion of cytotrophoblasts to form syncytial over the course of

several days (Kliman et al. 1986; Coutifaris et al. 1991).

A number of trophoblast cell lines also exist, the most common of these being

BeWo and JEG-3. However they do differ slightly in their in vitro behaviour, so the

selection of which cell line to use must be done carefully and with specific aims in mind

(Al-Nasiry et al. 2006). For instance, BeWo cells can be induced to undergo

intercellular fusion reminiscent of the syncytiotrophoblast and possess endothelial-like

properties, whereas JEG-3 do not (Coutifaris et al. 1991; Kiss et al. 1998; Cha et al.

2001; Al-Nasiry et al. 2006). Additionally, whereas primary trophoblast cultures

undergo spontaneous differentiation in vitro, the BeWo cell line requires the addition of

the cyclic adenosine monophosphate (cAMP) pathway agonist Forskolin in order to

initiate differentiation. In this way it is possible to finely manipulate the differentiation

pathway to assess its regulation.

The BeWo cell line was derived from a tumour of the cytotrophoblast known as

a choriocarcinoma (Pattillo et al. 1968). Many aspects of trophoblastic growth,

differentiation, metabolism and hormone production are observed in this cell line

(Pattillo et al. 1968). As such it is most representative of in vitro trophoblast

differentiation, and the focus of much research presented in this thesis.

1.2 Apoptosis

In order to ensure the correct activity and functioning of a tissue, mechanisms must be

in place to remove old, damaged or unneeded cells. This occurs by the active process of

programmed cell death, otherwise known as apoptosis (Kerr et al. 1972). A variety of

cellular and molecular events characterise apoptosis including mitochondrial membrane

depolarisation, chromatin condensation, DNA cleavage, phosphatidylserine efflux and

apoptotic vesicle formation (Kerr et al. 1972). This is a rapid process, however unlike

necrosis, occurs without inflammation and disruption to surrounding cells.


11

1.2.1 Placental Apoptosis

As in other tissues, the role of apoptosis in the placenta is to remove redundant nuclei

and cytoplasm. Trophoblast apoptosis has been extensively examined and reviewed

(Huppertz et al. 2006). While classical apoptosis occurs within hours, trophoblast

apoptosis occurs over a 3-4 week period, commencing in the cytotrophoblast leading to

the donation of its contents into the overlying syncytium (Huppertz et al. 1998; Smith

2000). Apoptosis is completed via the aggregation of syncytial nuclei into structures

known as syncytial knots, prior to being shed into the maternal circulation (Figure 1.6)

(Huppertz et al. 1998). Syncytial knots appear as dense multinuclear aggregations

underlying the apical membrane of the syncytiotrophoblast (Figure 1.6). Additionally,

trophoblast apoptosis is significantly increased across gestation, indicating a role in the

normal development and maturation of the placenta (Smith et al. 1997).

The extreme delay in the apoptotic mechanism may be due to the combinatory

effects of up-regulation of anti-apoptotic proteins and down-regulation of pro-apoptotic

proteins suppressing the effector stages of apoptosis and preventing the premature

extrusion of syncytial nuclei (Huppertz et al. 1998; Marzioni et al. 1998; Huppertz et al.

1999; Levy et al. 2000; Ratts et al. 2000; Yusuf et al. 2002). Co-ordinately,

incorporation of anti-apoptotic proteins into the syncytium may also focally inhibit

further apoptotic progression in order to maintain syncytial integrity (Huppertz et al.

1998; Marzioni et al. 1998; Shiozaki et al. 2003).

Figure 1.6 The formation and release of apoptotic syncytial knots from the

villous syncytiotrophoblast.


12

Pro-apoptotic proteins, including some caspases, are up-regulated prior to and

during fusion of the cytotrophoblast with the overlying syncytium (Huppertz et al.

1999). For instance, caspase-3 cleaves proteins involved in cell-cell contact, leading to

intercellular fusion (Brancolini et al. 1997; Getsios et al. 2000; Steinhusen et al. 2001).

Thus proteins involved in apoptosis are also involved in placental differentiation

indicating a link between these two cellular processes (Huppertz et al. 1998; Yusuf et al.

2002; Black et al. 2004). Additionally, terminal differentiation of epithelia, myoblasts,

osteoblasts, lens fibres and spermatids also involve apoptotic mechanisms, adding

credence to this hypothesis (Ishizaki et al. 1998; Gandarillas 2000; Fernando et al. 2002;

Arama et al. 2003; Mogi et al. 2003).

Furthermore, the cytoskeletal trophoblast marker cytokeratin-18 is cleaved

during trophoblast apoptosis, resulting in the destabilisation of the cell architecture and

apoptotic vesicle formation(Leers et al. 1999; Kadyrov et al. 2001). As a key

intermediate filament protein in the trophoblast, the expression, structure and

localisation of cytokeratin 18 remains plastic to modulate the cytoskeletal arrangement

in response to a multitude of cell processes. In particular, trophoblast differentiation

requires extensive cytoskeletal rearrangement to prepare and exact syncytialisation.

Thus cytokeratin-18 is presumably involved in trophoblast differentiation as well as

apoptosis.

1.3 Caspases

Most work examining the relationship between apoptosis and differentiation has

focused on the expression patterns of a family of apoptotic proteins known as caspases.

These proteins are responsible for the commencement and completion of apoptosis

following stimulation by either extrinsic or intrinsic pathways (De Falco et al. 2004).

Caspases are initially synthesised as enzymatically inactive proforms, the prodomain of

which renders the protein inactive. Upon activation the prodomain is enzymatically

removed, and the remainder cleaved into a long and short subunit. The long and short

subunit aggregate into a heterodimer, before two of these align to form an activated

heterotetramer (Figure 1.7).

Caspases are classified into 2 sub-families according to their structure and

function. Initiator caspases (2, 8, 9 and 10) contain a long prodomain and are activated

upon stimulation from either an internal or external source. Within the prodomain of

initiator caspases may be a caspase recruitment domain (CARD) (caspases 2 and 9) or a


13

death effector domain (DED) (caspases 8 and 10) (Figure 1.7). As their name suggests,

CARDs are responsible for the aggregation of caspases, while DED mediate the

recruitment of other proapoptotic proteins, resulting in the transduction of death signals

(Weber et al. 2001). By contrast, effector caspases (3, 6 and 7) contain a short

prodomain and are activated by initiator caspases. Effector caspases are responsible for

the enzymatic cleavage and degradation of cytoskeletal and nuclear proteins leading to

the apoptotic morphology and formation of apoptotic vesicles.

Figure 1.7 Cleavage and activation of caspases. Inactive caspases consist of an

inactive prodomain and long and short subunits. Activation occurs through procaspase

cleavage to form heterodimers, then aggregation into activated heterotetramers. Effector

caspases possess either a caspase recruiting domain (CARD) or a death effector domain

(DED) (Adapted from Bleackley et al. 2001).

Activation of initiator caspases does not necessarily result in apoptosis as anti-

apoptotic members of the Bcl-2 family, amongst others, can prevent effector caspase

activation (Levy et al. 2000). Only once effector caspases are activated is the apoptotic

process irreversible.


14

Figure 1.8 Schematic of the classical extrinsic and intrinsic apoptotic cascades.

External apoptotic stimuli cause Fas to bind its receptor and recruit and activate

caspase-8. This may directly or indirectly activate caspase-3 and subsequent DNA

fragmentation and apoptosis. The indirect route involves Bid cleavage stimulating

mitochondrial cytochrome c release and formation of the apoptosome. This recruits and

activates caspase-9, which in turn activates caspase-3. Intrinsic apoptotic stimuli lead to

Bax, Bad or Bim-induced mitochondrial cytochrome-c release. Again, the apoptosome

is formed and the cascade progresses as per the extrinsic pathway (Adapted from

Matsuno et al. 2003).

The extrinsic apoptotic pathway (Figure 1.8) involves the binding of Fas-ligand

to its receptor (Fas) causing Fas-receptor aggregation leading to the formation of a

death-inducing signal complex (DISC) (Aschkenazi et al. 2002; De Falco et al. 2004).

Procaspase-8 is recruited to this complex and activated before cleaving procaspase-3,


15

which in turn cleaves multiple targets within the cell leading to its destruction. This

system has been implicated in the maintenance of immune privilege. As the trophoblast

provides an immune privileged barrier, this pathway may be important in placental

apoptosis (Aschkenazi et al. 2002; Jerzak et al. 2002).

Upon internal cell stress, the intrinsic apoptotic pathway (Figure 1.8) is

stimulated whereby pro-apoptotic members of the Bcl-2 family (Bax, Bad, Bim)

associate with the mitochondria, modulating the rapid release of cytochrome-c into the

cytoplasm (De Falco et al. 2004; Goldstein et al. 2005). Apaf-1 protein binds

cytochrome c, along with dATP and ATP, to form the apoptosome complex.

Procaspase-9 is recruited to the apoptosome, whereupon it is activated. In turn, this

activated caspase-9 initiates procaspase-3 cleavage. Activated caspase-3 stimulates the

fragmentation of genomic DNA and leads to cellular degradation and apoptosis.

1.4 Caspase-14

The 29.5kDa caspase-14 protein contains the conserved active QACRG domain

characteristic of caspases (Ahmad et al. 1998; Hu et al. 1998; Van de Craen et al. 1998).

Caspase-14 possesses an abnormally short prodomain characteristic of effector

caspases, however cannot be activated by other caspases (Hu et al. 1998; Chien et al.

2002). Additionally, mutations of caspase-14 are rarely seen in tumours suggesting

strict regulation of caspase-14 expression (Yoo et al. 2007).

Caspase-14 can be cleaved into large and small subunits (Van de Craen et al.

1998; Chien et al. 2002), although cleavage alone is insufficient to cause activation

(Mikolajczyk et al. 2004). The mode of cleavage and activation remain unknown,

however it is not activated by other caspases or pro-apoptotic members of the Bcl-2

family (Hu et al. 1998; Mikolajczyk et al. 2004). Thus caspase-14 appears to function

independently of the classical apoptotic pathways (Hu et al. 1998). The in vivo

activation mechanism remains elusive, however filaggrin has been identified as a target

substrate of caspase-14 in the skin (Denecker et al. 2007), while other targets are likely

to possess the motif W/Y-X-X-D (Park et al. 2006).

No pro-apoptotic role has been determined for caspase-14, however it is

activated in epidermal differentiation in a process similar to apoptosis (Eckhart et al.

2000; Lippens et al. 2000; Fischer et al. 2004). Caspase-14 is the only caspase family

member up-regulated during stratum corneum formation, indicating that it is not

involved in classical apoptotic pathways (Eckhart et al. 2000; Lippens et al. 2000; Rendl


16

et al. 2002; Chaturvedi et al. 2006). Processing of caspase-14 coincides with nucleus

extrusion from corneocytes, indicating a possible role for caspase-14 in barrier

formation (Eckhart et al. 2000; Lippens et al. 2000; Rendl et al. 2002; Fischer et al.

2005). Non-cornifying epithelia do not express caspase-14, further supporting this

hypothesis (Kuechle et al. 2001). Moreover, increased caspase-14 activity is essential

for cornification following barrier disruption (Demerjian et al. 2007). In addition,

caspase-14 may also prevent against UVB radiation damage to the stratum corneum

(Denecker et al. 2007).

As caspase-14 has a likely role in barrier formation and maintenance in the skin,

it may also be involved in placental differentiation as cells of the trophoblast fuse to

form a protective barrier. Indeed, caspase-14 has been identified in the trophoblast of

both first trimester and term placentae (Kam et al. 2005). The formation of syncytia is

limited to very few human tissues such as giant cells, skeletal muscle and osteoclasts,

thus it is an attractive hypothesis that caspase-14 plays an integral role in

syncytialisation throughout the body, including the placenta. Indeed, caspase-14 is

restricted to barrier-forming tissues such as the skin and choroid plexus (Lippens et al.

2003). Accordingly, I examined the expression and activation of caspase-14 during the

development of the human placenta.

Several factors suggest a potential role for caspase-14 outside of apoptosis in the

trophoblast. Firstly, despite apoptosis only occurring in less than 1% of trophoblastic

cells at any one time (Smith 2000), caspase-14 is abundant throughout the trophoblast

(Kam et al. 2005). Furthermore, there are higher levels of caspase-14 in the human first

trimester trophoblast than in comparable apoptotic trophoblast (Kam et al. 2005).

Therefore, the presence of caspase-14 within the differentiating trophoblast, coupled

with the lack of caspase-14 activation during apoptosis, suggest a potential role for

caspase-14 in terminal differentiation leading to placental barrier formation (Kam et al.

2005).

1.4.1 Caspase-14 functional pathways

Psoriasis is a chronic pathological condition of the epidermis likely caused by an

autoimmune attack in response to infection, stress or hormonal changes, resulting in the

formation of dry, non-cornified, scaly lesions (Marieb 1998). Caspase-14 expression is

reduced in psoriatic skin associated with aberrant keratinocyte differentiation (Lippens

et al. 2004; Hsu et al. 2005; Walsh et al. 2005). However upon treatment with either


17

Vitamin D3 or green tea polyphenols such as epigallocatechin gallate (EGCG), caspase-

14 expression is elevated in concert with p57KIP2

, leading to the restoration of the

normal skin phenotype (Lippens et al. 2004; Hsu et al. 2005).

Psoriasis leads to the disruption of various Mitogen-Activated Protein Kinase

(MAPK) transcription factor pathways (Hsu et al. 2007). It is through the p38 and c-Jun

N-terminal Kinase (JNK) MAPK transcription factor pathways that green tea

polyphenols (GTP) induce the restoration of caspase-14 in psoriatic keratinocytes (Hsu

et al. 2007). Furthermore, topical application of GTPs reduced the symptoms of flaky

skin mice and led to effective caspase-14 processing and activation (Hsu et al. 2007).

Transcription factors of the activator protein-1 (AP-1) family and nuclear factor

kappa-B (NFκB) bind to the proximal caspase-14 promoter upstream of exon-1 to

activate the expression of caspase-14 under basal conditions (Ballaun et al. 2008). The

AP-1 protein complexes are comprised of members of the Jun and Fos subfamilies

forming heterodimers, which bind genomic DNA to activate the transcription of the

target genes (Karin et al. 1997). Several discrete AP-1 factors including JunB, c-jun,

JunD, fra-1 and fra-2 bind to the AP-1 binding site within the caspase-14 promoter,

while caspase-14 is reduced in keratinocytes lacking JunB (Ballaun et al. 2008). These

data indicate a necessity for AP-1 transcription factors in the induction and basal

production of caspase-14.

In terms of the placenta all AP-1 transcription factors, with the notable

exceptions of c-Jun and JunD, are absent from the villous trophoblast throughout

gestation, while most AP-1 members are expressed in the extravillous trophoblasts

(Bamberger et al. 2004). This suggests that AP-1 transcription factors may regulate

extravillous trophoblast proliferation and differentiation. In the chorionic villi,

cytotrophoblasts and fetal endothelia express the mitogenic c-Jun, while

syncytiotrophoblasts express JunD (Bamberger et al. 2004). The syncytiotrophoblast

produced hormone hPL is a target of AP-1 transcription factors, thus JunD may

modulate hPL expression (Peters et al. 2000). Intriguingly, hCG suppresses the action of

NF-κB and AP-1 transcription factors (Manna et al. 2000). This may be a feedback

mechanism by which differentiation inhibits further proliferation, and may also explain

the absence of most AP-1 members from the trophoblast.

The recently described caspase-14-/-

mouse model has provided significant

insight into the direct function of caspase-14 in epidermal differentiation and function

(Demerjian et al. 2007; Denecker et al. 2007). The skin of caspase-14-/-

mice appears

shiny and leathery, suggesting an alteration to the stratum corneum (Denecker et al.


18

2007). As this mouse model lacks the gene encoding caspase-14, it represents a

powerful tool in the elucidation of the functional properties of caspase-14.

1.4.1.1 Filaggrin

Profilaggrin is a highly phosphorylated insoluble protein produced by a 3-exon gene,

and contains multiple filaggrin-repeat sequences separated by short linker peptides

(Figure 1.9) (Presland et al. 1992; Rawlings et al. 2004). It is vital for the correct

differentiation of apical keratinocytes so is therefore expressed predominantly in the

stratum granulosum where it self-aggregates to form keratohyalin bundles (K-bundles).

Filaggrin is responsible for recruiting and coupling keratin intermediate filaments into

dense macrofibrils, a feature characteristic of the stratum corneum (Presland et al.

1992).

Figure 1.9 Profilaggrin. Sequential production and processing of profilaggrin into its

highly phosphorylated protein, dephosphorylation and cleavage by caspase-14 into its

component filaggrin repeats, and final degradation into individual amino acids forming

the Natural Moisturising Factor of the corneocytes.

In the upper cornified layers, filaggrin is further degraded into its individual

amino acids, which are a major component of Natural Moisturising Factor (NMF)

(Rawlings et al. 2004). This acts as a „moisture sponge‟, soaking up water to maintain

the skins‟ natural hydration and regulating water flux and retention (Rawlings et al.

2004). Thus filaggrin is crucial for the maintenance of the skins‟ barrier function.


19

Recent evidence points towards caspase-14 as a protease responsible for

dephosphorylation and processing of profilaggrin into its functional constituent filaggrin

repeats upon cornification (Denecker et al. 2007). Accordingly, the skin of caspase-14-/-

mice is dehydrated and displays increased water loss (Denecker et al. 2007).

1.4.1.2 p57KIP2

The cell cycle regulator p57KIP2

is a cyclin-dependent kinase inhibitor important for G1-

phase cell cycle withdrawal and differentiation. p57KIP2

binds and inhibits complexes of

cyclin E-cdk2, cyclin D2-cdk4, and cyclin A-cdk2, thereby suppressing proliferation

(Lee et al. 1995; Matsuoka et al. 1995). In the human placenta, p57KIP2

is present in the

nuclei of both villous and extravillous cytotrophoblasts, but only rarely in the

syncytiotrophoblast (Chilosi et al. 1998), indicating a role in initiation of cell cycle

withdrawal rather than maintenance of the terminally differentiated phenotype. Given

its role in differentiation, and a potential role in the inhibition of preeclampsia

(Kanayama et al. 2002), p57KIP2

may also be involved in the formation of the

syncytiotrophoblast. Indeed, p57KIP2

is able to induce caspase-14 expression in psoriatic

lesions (Hsu et al. 2005).

The elucidation of p57KIP2

signalling in the trophoblast is complicated by its

complete absence in gestational trophoblastic diseases and choriocarcinoma cell lines

such as BeWo (Morrish et al. 2007), which is used extensively in this thesis. However,

transfection of this cell line with p57KIP2

slows proliferation and invasion, while

enhancing syncytialisation and even inducing the production of hCG (Morrish et al.

2007). Thus p57KIP2

may be involved in the induction of trophoblast differentiation.

In the epidermis, p57KIP2

is associated with the p38 pathway, while caspase-14 is

associated with both the p38 and c-Jun N-terminal Kinase (JNK) MAPK pathways (Hsu

et al. 2007). Thus while p57KIP2

may be a regulator of caspase-14 expression in the

epidermis via p38 (Hsu et al. 2005; Hsu et al. 2007), caspase-14 is also associated with

the JNK MAPK pathway (Hsu et al. 2007), indicating that caspase-14 is regulated by

multiple cellular signalling pathways.

1.4.1.3 Krüppel-like Factor 4

Keratinocyte differentiation and barrier formation is dependent on the expression of the

Krüppel-Like Factor 4 (KLF4) transcription factor (Segre et al. 1999; Jaubert et al.

2003; Dai et al. 2004). KLF4 regulates the transcription of cyclin-D2 by binding to its

promoter region, leading to cell cycle arrest at the G1/S-phase boundary


20

(Klaewsongkram et al. 2007). KLF4-/-

mice present a similar, yet lethal, phenotype to

that observed in the absence of caspase-14. Loss of KLF4 causes mortality of newborn

mice through trans-epidermal water loss as a result of impaired formation of cornified

envelopes of the stratum corneum within 15 hours of birth (Segre et al. 1999). However,

unlike caspase-14-/-

mice, profilaggrin and its functional subunits are normally

processed in the absence of KLF4, indicating non-overlapping roles in hydration of the

epidermis (Segre et al. 1999).

It has been noted that caspase-14 contains a potential binding site for KLF4

(Eckhart et al. 2000), and given the role of both factors in keratinocyte differentiation,

and that of KLF4 in the maintenance of the human placenta (Blanchon et al. 2001;

Blanchon et al. 2006) it is attractive to speculate a synergy between the two in

trophoblast barrier formation. Furthermore, KLF4 contains a protein sequence (YHCD)

consistent with the proposed caspase-14 binding motif (Park et al. 2006). Accordingly,

the expression of KLF4 will be examined upon differentiation of the BeWo cell line.

1.5 Canonical Wnt signalling

Cellular proliferation is controlled and regulated by a variety of cellular pathways, one

of which involves the 17-member Wingless (Wnt) family of extracellular ligands. Under

non-proliferative conditions, β-catenin is phosphorylated by the GSKβ3, Axin and APC

complex, leading to its cytosolic inactivation and degradation (Polakis 1999). Wnt

ligands bind Frizzled receptors in the plasma membrane resulting in the stabilization of

cytosolic β-catenin, leading to its nuclear translocation and the initiation of proliferation

through the activation of transcription factors (Figure 1.10) (Huelsken et al. 2002;

Saldanha et al. 2004). This is achieved by activation of the lymphoid enhancer factor

(LEF) and T-cell factor (TCF) transcription factors by β-catenin, leading to activation of

cell cycle genes such as Cyclin D1 and c-myc, stimulating cellular proliferation

(Shtutman et al. 1999; Eberhart et al. 2001; Stockinger et al. 2001). Consequently,

antagonism of the canonical Wnt pathway may be required in preparation for terminal

differentiation.

Most Wnt ligands and Frizzled receptors are present in human trophoblast with

variable expression according to gestational age and trophoblast sub-type, culminating

in a complex Wnt signalling repertoire in the human trophoblast. (Tulac et al. 2003;

Sonderegger et al. 2007). Nuclear translocation of β-catenin is evident in proliferative

cytotrophoblasts but not terminally differentiated syncytiotrophoblast (Eberhart et al.


21

2001). Thus, Wnt pathways are active in the proliferative cytotrophoblasts, and

suppression of Wnt signalling may lead to trophoblast differentiation and adoption of

the terminal phenotype. Furthermore, deletion of various Wnt ligands or receptors

results in aberrant placental angiogenesis and vascularisation (Monkley et al. 1996;

Ishikawa et al. 2001). Therefore Wnt signalling is important not only in trophoblast

proliferation, but also early vasculogenesis and endothelial functions.

Figure 1.10 β-catenin signalling pathways. β-catenin is a key component in both the

maintenance of cell adhesion complexes (A) and the canonical Wnt pathway (B and C).

In adherens junctions it acts as a mediary between extracellular signalling and the actin

cytoskeleton. This competes for β-catenin with canonical Wnt pathway, where Wnt

ligand binding of Frizzled receptors leads to stabilisation of β-catenin, its nuclear

translocation, and subsequent transcription of cell cycle genes (B). Antagonism of Wnt

binding by sFRPs results in the formation of the GSKβ3/Axin/APC-complex, which

dephosphorylates and degrades β-catenin (C).

1.5.1 E-cadherin and β-catenin interactions

Cell-cell communication is maintained through the formation of adherens junctions

between adjacent cells. A key component of epithelial and trophoblast adherens


22

junctions is the transmembrane protein E-cadherin (Figure 1.10). As E-cadherin

expression is restricted to the plasma membrane it acts as an indicator of the cellular

surface area to nucleus ratio. With intercellular fusion the surface area to nuclei ratio of

the cell is decreased (i.e. more nuclei, less surface area), thereby reducing the

expression of E-cadherin (Coutifaris et al. 1991; Al-Nasiry et al. 2006). Thus E-

cadherin expression can be used as a marker for cytotrophoblast fusion, as reduced E-

cadherin expression indicates cellular fusion.

Adherens junctions are typically composed of a complex of transmembrane

cadherins, and intracellular α- and β-catenin. In epithelia and trophoblasts, the

cytoplasmic tail of the transmembrane E-cadherin forms complexes with α- and β-

catenin, the former of which creates dynamic links with the actin cytoskeleton (Figure

1.10) (Saldanha et al. 2004; Drees et al. 2005; Yamada et al. 2005). Indeed, monomeric

α-catenin preferentially binds to adherens junctions, while dimeric α-catenin

preferentially binds actin (Drees et al. 2005). Therefore the linkage between adherens

junctions and the cytoskeleton is dynamic and constantly remodelling.

One function of E-cadherin is to sequester β-catenin at the plasma membrane,

preventing its translocation to the nucleus and down-regulating its mitogenic

transcriptional activity (St Croix et al. 1998; Orsulic et al. 1999; Stockinger et al. 2001).

Alternatively, unsequestered β-catenin can be degraded in the cytosol in association

with the GSKβ3/Axin/APC-complex. However, the binding of Wnt ligands to their

Frizzled receptors stabilises β-catenin, promoting its translocation to the nucleus and its

activation of transcription factors, leading to cellular proliferation (Figure 1.10)

(Brembeck et al. 2006).

Complicating matters, phosphorylation of the E-cadherin cytoplasmic domain by

Casein Kinase II and GSKβ3 promotes its binding to β-catenin, strengthening cell

adhesion complexes (Lickert et al. 2000). Thus GSKβ3 has the joint role of

phosphorylating both E-cadherin and β-catenin; stabilising cell-cell communication and

inhibiting proliferation, respectively. Furthermore, upon severing adherens junctions

during apoptosis or differentiation, the cytoplasmic tail of E-cadherin may render the β-

catenin inactive (Steinhusen et al. 2001). Thus whilst β-catenin is bound to E-cadherin

cell growth is suppressed, however severing these bonds results in cell cycle

progression.

The function of β-catenin may not be determined purely by direct competition

between adhesion and transcriptional proteins, but rather through different

conformations of β-catenin selectively binding either E-cadherin or the TCF


23

transcription factor following Wnt signalling, coupled with the phosphorylation status

of E-cadherin (Gottardi et al. 2004). That is, the conformational and phosphorylation

status of β-catenin mediates whether it is involved in adhesion or transcriptional

regulation. Thus, when required, cells can separate the adhesive and Wnt signalling

functions of β-catenin.

1.5.2 Secreted Frizzled-Related Protein-4 (sFRP4)

The Wnt ligand family is crucial for regulating signalling during development and

initiating proliferation in a wide variety of cell types. However, antagonism of Wnt

signalling can occur at many isolated steps in the cascade. The secreted Frizzled-related

protein (sFRP) family antagonise Wnt signalling by either directly binding Wnt ligands

(Wawrzak et al. 2007) or competitively binding their Frizzled receptors. sFRP4 shares

significant homology with the extracellular domain of Frizzled proteins (Wolf et al.

1997), and as with the other sFRPs it antagonizes Wnt-signalling by competitively

binding Wnts and their Frizzled receptors in the plasma membrane (Figure 1.10)

(Rattner et al. 1997; Wong et al. 2002; He et al. 2005). This leads to the phosphorylation

and subsequent degradation of cytosolic β-catenin, initiating cell cycle withdrawal and

increasing the incidence of apoptosis and differentiation (Schumann et al. 2000; He et

al. 2005). Thus disruption to sFRP4 may be involved in uncontrolled proliferation and

tumourigenesis (He et al. 2005).

Analysis of the rat placenta reveals not only the presence of sFRP4, but also its

inhibition of the Wnt pathway reducing proliferation and subsequent placental growth

(Hewitt et al. 2006). Furthermore, as sFRP4 is involved in cell-cycle withdrawal by

antagonising Wnt signalling, it may therefore also induce differentiation and apoptosis,

or if over-expressed lead to placental complications such as Intrauterine Growth

Restriction (IUGR). Additionally, β-catenin is translocated to the nucleus of

proliferative cytotrophoblasts but not syncytiotrophoblasts (Eberhart et al. 2001).

Therefore, sFRP4 may well be a candidate suppressor of Wnt activity. Thus sFRP4

expression may enhance differentiation and even apoptosis in the trophoblast.


24

1.6 Experimental Design

1.6.1 Aims

This thesis aims to examine the expression patterns of hormonal, differentiation and

endothelial related genes in trophoblast from human placentae of different gestational

ages. Of particular emphasis is the expression and function of caspase-14 in the human

placenta, particularly during trophoblast differentiation. I propose characterising the

signalling mechanisms and interactions of caspase-14, and thereby its role during

trophoblast differentiation.

1.6.2 Hypotheses

It is hypothesised that caspase-14 expression will not be altered in response to

trophoblast apoptosis. Furthermore it is expected that caspase-14 will be up regulated

with trophoblast differentiation. Known and putative binding partners and targets of

caspase-14 activity are hypothesised to be expressed in the human trophoblast.

Moreover expression of these proteins is expected to be disrupted in the absence of

caspase-14.

I also postulate a considerable endothelium-like function of the human

trophoblast, particularly in response to differentiation. Markers of endothelial function

will be examined during trophoblast differentiation, across gestation and in

preeclampsia, and correlated with changes in caspase 14. It is hypothesised that such

pathways will be dependent upon trophoblast differentiation consistent with an

endothelium-like function of the syncytiotrophoblast.

Chapter 2- Materials and Methods

25

Chapter 2

Materials and Methods

2.1 BeWo cell line

BeWo cells were obtained from the American Type Culture Collection (ATCC) and

maintained in growth media containing 5% fetal bovine serum (FBS) and 0.5%

penicillin/streptomysin in Hams F12K. For treatment, cells were seeded at a density of

2.5x105 cells/ml and grown for 24 hours at 37ºC and 5% CO2. After this time, growth

media was replaced with treatment media to stimulate either apoptosis or differentiation,

and cells incubated for designated time periods. Apoptosis in BeWo cells was

stimulated by the addition of 1μM Staurosporine (Cayman Chemicals, Cat. No. 81590)

for 1, 3 or 6 hours, while biochemical and morphological differentiation were initiated

by the addition of 10μM Forskolin (LC Laboratories, Cat. No. F-9929) for 24, 48, 72

and 96 hours. Controls were maintained by the addition of an equivalent volume of

DMSO at the commencement of the treatment period.

To ensure adequate frozen stocks of low passage BeWo, cells were diluted to

1x106 cells/ml, and 600µl added to a cryovial containing 300µl FBS and 100µl DMSO

and placed in a freezing unit containing isopropanol. The freezing unit was placed in a -

80°C freezer overnight to ensure the stable freezing of cells, before the cryovials were

placed in liquid nitrogen.

2.2 Human Placenta Extracts

Normal first trimester placentae were obtained with clearance from the Marie Stopes

Clinic, Midland and the informed consent of the mother following surgical termination.

Immediately after delivery samples were fixed in 10% Formaldehyde, snap frozen in

liquid nitrogen, or incubated in chilled (4°C) serum free media (100ml Minimum

Essential Medium (MEM), 100mg Bovine Serum Albumin (BSA), 500μl

penicillin/streptomycin, 29.2mg L-glutamine) with or without 75U/ml Superoxide

Dismutase (SOD) (Sigma, Cat. No. S9636).

Extracts from term human placentae were obtained after approved ethical

clearance from King Edward Memorial Hospital, the University of Western Australia,

and the informed consent of the mother following childbirth. Samples were immediately


26

fixed in 4% Formaldehyde, or snap frozen in liquid nitrogen for further analysis in the

laboratory.

All preeclamptic placental samples were clinically diagnosed according to the

maternal presentation of pregnancy-related hypertension and proteinurea. One such

sample was obtained from King Edward Memorial Hospital, Western Australia with

informed consent of the mother and the relevant ethical committees. Additional

preeclamptic placental samples were obtained by Dr. Henning Schneider of the

Department of Obstetrics and Gynaecology, University Women‟s Hospital, Berne,

Switzerland. Upon extraction of RNA and conversion into cDNA (Section 2.5), samples

were shipped to the School of Anatomy and Human Biology, University of Western

Australia for further analysis.

2.3 Assessment of Apoptosis

Staurosporine is a known inducer of apoptosis in a variety of tissues in vitro, including

the BeWo cell line (Das et al. 2004; Poliseno et al. 2004; Goldstein et al. 2005). BeWo

cells were incubated in the presence of 1μM Staurosporine for 1, 3 or 6 hours (Das et al.

2004), after which samples were analysed for changes in gene transcription and protein

expression during apoptosis.

Apoptosis of the BeWo cell line was determined via 3‟-end labelling and

measurement of mitochondrial membrane depolarisation using the JC-1 (5,5‟,6,6‟-

tetrachloro-1,1‟,3,3‟-tetraethylbenzimidazolcarbo-cyanine iodide) technique, while

apoptosis in the human placenta was quantified by 3‟-end labelling only. At the onset of

apoptosis, the mitochondrial membrane is rapidly depolarised (Das et al. 2004).

Accompanying this, JC-1 disaggregates to form a monomer and the ratio of aggregated

to monomeric JC-1 gives a quantitative representation of the extent of mitochondrial

membrane permeability (Smiley et al. 1991; Zamzami et al. 2000). This enables the

identification and quantification of apoptosis at the various time points following

Staurosporine treatment. A hallmark of apoptosis is the degradation of DNA into 185

base pair fragments. These fragments can be quantified and radiolabelled for

visualisation of the characteristic laddering effect.

The expressions of various mRNA transcripts were examined by reverse

transcriptase polymerase chain reaction (RT-PCR), while protein levels were assessed

by Western Blot.


27

2.3.1 JC-1

At the onset of apoptosis, the mitochondrial membrane is rapidly depolarised (Das et al.

2004). Accompanying this, JC-1 incorporates into mitochondria and forms monomers

that fluoresce green (520nm), whereas at high membrane potentials it forms J-

aggregates which fluoresce red (590nm) (Smiley et al. 1991; Zamzami et al. 2000).

Therefore the ratio of aggregated versus monomeric JC-1 gives a quantitative

representation of mitochondrial membrane permeability such that a low red: green ratio

is indicative of apoptosis (Smiley et al. 1991; Zamzami et al. 2000).

Cells grown in 96-well plates were treated with 1μM Staurosporine for 1 or 3

hours, media removed and 50μl of 2.5mM JC-1 (Molecular Probes T-3168) staining

solution diluted 1:75 in Hams F12-K added and incubated for 60 minutes. JC-1 was not

conducted after 6 hours of Staurosporine treatment as insufficient cells remain attached

to the flask. Positive controls were conducted by the addition of 50μM carbonyl cyanide

p-(trifluoromethoxy) phenylhydrazone (FCCP) to the JC-1 culture medium to prevent

aggregation of dye. Thus an apoptotic response is simulated by the high proportion of

monomeric (red) dye. JC-1 was removed and cells washed with 200μl Phosphate

Buffered Solution (PBS, pH 7.4) containing 5% BSA for 5 minutes at 37ºC and 5%

CO2. After removal of this, 100μl PBS was added and the plate analysed using a

FluoStar fluorescent plate reader at 520nm (green) and 590nm (red). Raw values at

590nm were divided by the raw values of the corresponding sample at 520nm.

2.3.2 DNA extraction from BeWo Cell Culture

To assess apoptosis, DNA was extracted from BeWo cells following either 3 or 6 hours

incubation with 1μM Staurosporine using DNA Homogenising Buffer (0.1M NaCl,

0.01M ethylene diamine tetraacetic acid (EDTA, pH 8.0), 0.3M Tris-HCl (pH 8.0) and

0.2M sucrose). Assessment was not undertaken after 1 hour of treatment as apoptosis

had not been initiated at this time point (as assessed by the JC-1 technique). Following

the centrifugation of the treatment media, 500μl of DNA Homogenising Buffer was

added to each well and cells dislodged using a cell scraper. To lyse the cells, 31.2μl of

10% SDS was added, mixed and incubated at 64ºC for 30 minutes. Denaturation of

proteins was performed by the addition of 87.5μl of 8M Potassium Acetate to the

samples, which were subsequently incubated on ice for 1 hour. Samples were next

centrifuged for 10 minutes at 14000rpm and 4ºC. To the supernatant, 2μl of 500μg/ml

RNase (Promega, Cat. No. M4261) was added and incubated at 37ºC for 1 hour. Next,


28

an equal volume of phenol:chloroform:isoamylalcohol (250μl each) was added, mixed

and centrifuged for 5minutes at 6000rpm. The upper phase was collected and an equal

volume (500μl) of chloroform:isoamylalcohol added prior to further centrifugation at

6000rpm for 5 min at 4ºC. Again, the upper phase was collected and 0.1 volumes of 3M

Sodium Acetate (50µl) and 2.5 volumes of cold 100% ethanol (1250μl) added and

incubated overnight at -80ºC.

The next day each sample was spun at 4°C and 14000rpm for 30 minutes and the

supernatant removed. The resultant pellet was washed thoroughly with 100μl of 70%

ethanol and centrifuged at 14000rpm for 3 minutes. The supernatant was removed, and

the pellet allowed to air dry completely in a fume-hood for at least 1 hour. Finally the

pellet was resuspended in 20μl of dH2O and incubated at 37ºC for 15 minutes. Prior to

storage at -20ºC, DNA quality and quantity was assessed by spectrophotometry. Briefly,

2μl of sample was diluted in 48μl dH2O and added to the cuvette. DNA quantity was

calculated by multiplying the Absorbance at a wavelength of 260nm by the dilution

factor (25) and DNA factor (50).

2.3.3 Explant Culture Model of Apoptosis

Changes in gene transcription and protein expression during first trimester placental

apoptosis in vivo was assessed using an explant culture model of apoptosis (Rao et al.

2000; Charles et al. 2005; Kam et al. 2005). Within this model, explants are incubated

with superoxide dismutase (SOD) to suppress apoptosis. Free radicals accumulate

within the tissue leading to oxidative stress and apoptosis. Superoxide dismutase is a

free radical species responsible for the scavenging and conversion of the superoxide

anion (O2-) into H2O2 for removal from the cell, reducing oxidative stress and

suppressing apoptosis (Sikkema et al. 2001). By scavenging free radicals SOD

suppresses apoptosis, thus providing a model to assess the molecular mechanisms

involved in placental apoptosis.

First trimester placental samples were obtained from the Marie Stopes Clinic in

Midland, Western Australia under ethical approval and informed maternal consent.

Incubation media for the explant culture model consisted of 100ml Minimum Essential

Medium (MEM, Gibco), 100mg BSA, 500μl penicillin/streptomycin and 29.2mg

powdered L-glutamine, sterile filtered. The presence of chorionic villi was detected

upon delivery of the placenta, and divided 3 ways into 2ml incubation media (control),

2ml incubation media supplemented with 75U/ml SOD, and 2ml of 1% Formaldehyde


29

for confirmation of chorionic villi. SOD-model samples were immediately stored on wet

ice until arrival at the School of Anatomy and Human Biology and The University of

Western Australia, Crawley, WA after approximately 2 hours.

Upon delivery to the laboratory, Control and SOD-treated samples were

incubated with gentle mixing for 4 hours at 37ºC and 5% CO2. After 4 hours incubation,

each explant was divided evenly into 3 homogenising tubes, 1 each for DNA, RNA and

protein extractions, and stored at -80ºC.

The degree of apoptosis was determined via 3-prime end labelling. Alterations in

the molecular fingerprint during human placental apoptosis were determined by

comparing expressions between the SOD-treated samples and those undergoing

spontaneous apoptosis.

2.3.4 DNA Extraction from Placenta

The extraction of DNA from first trimester, term and preeclamptic human placentae was

performed with minor variation to that performed in the BeWo cell culture model

(Section 2.2.2). To the tissue in a homogenising tube, 1ml of DNA Homogenisation

Buffer was added, and the tissue homogenised using a Kinematica AG Polytron PT2100

homogeniser. Cell lysis was performed by the addition of 62.5μl of 10% SDS, and

incubated for 30 minutes at 65ºC. Denaturation of proteins was conducted by addition

of 175μl of 8M Potassium Acetate and incubation on wet ice for 1 hour. Denatured

proteins were pelleted by centrifugation for 10 minutes at 14000rpm. To the supernatant

an equal volume (500μl) each of phenol and chloroform:isoamylalcohol was added,

mixed and centrifuged for 5 minutes at 6000rpm. The upper phase was collected and an

equal volume of chloroform:isoamylalcohol added prior to further centrifugation at

6000rpm for 5minutes. Again, the upper phase was collected (400μl) and 2.5 volumes

of cold 100% ethanol (1ml) added and incubated overnight at -80ºC.

Following centrifuged at maximum speed for 30minutes, the supernatant was

removed and the pellet resuspended in 100μl of 1x TE buffer. Additionally, 2μl of

RNase (500μg/ml) was added, the sample was vortexed and quickly spun prior to

incubation at 37ºC for 1 hour. An equal volume of phenol : chloroform:isoamylalcohol

(125μl each) was added the sample, which was then vortexed and centrifuged at

6000rpm for 5minutes. The upper phase was collected (200μl) and an equal volume of

chloroform:isoamylalcohol added before vortexing and centrifugation for 5mins at

6000rpm. The upper phase was collected (200μl) and 0.1 volume of 3M Sodium


30

Acetate (20μl) and 2.5 volumes of cold 100% ethanol (500μl) were added, and the

sample incubated overnight at -80ºC.

Samples were centrifuged at maximum speed for 30minutes and the supernatant

carefully removed and 100μl of cold 80% ethanol added and incubated on ice for 5mins,

followed by centrifugation for 5minutes at maximum speed. The supernatant was again

removed carefully, and the pellet allowed to air dry for up to 2 hours until completely

dry. The final DNA pellet was resuspended in 50μl of dH2O. A 2μl sample of the final

DNA was diluted 1:25 in distilled H2O and analysed by spectrophotometry. The

resultant DNA concentration as determined by the spectrophotometer was multiplied by

the dilution factor (25) and the DNA factor (50) to gain a final DNA concentration. The

DNA was stored at -20ºC until analysed by 3‟-end labelling as outlined in Section 2.2.3.

2.3.5 3’-End Labelling

A major characteristic of late stage apoptosis is fragmentation between nucleosomal

bundles of genomic DNA, resulting in multiples of approximately 185 base pairs. Upon

gel electrophoresis and autoradiography, these bundles produce a typical DNA

fragmentation ladder. Internucleosomal fragmentation is examined by using terminal

transferase to bind one molecule of radioactively-labelled 2´,3´-Dideoxyadenosine 5´-

Triphosphate (ddATP) to the 3‟ ends of double and single stranded DNA. Following gel

electrophoresis and autoradiography, the total amount of incorporated radiolabel can be

quantitated and used to estimate the degree of apoptotic DNA fragmentation (Tilly et al.

1993; Drake et al. 2003).

500ng of total DNA was co-incubated with [α32

P]-ddATP (50µCi; 3.4 pmol/µl),

terminal tranferase and cobalt chloride (CoCl2) at 37°C for 1 hour. The reaction was

terminated by the addition of 5µl 0.25M EDTA. Following this the carrier Transfer

RNA (tRNA) buffered in Ammonium Acetate and 100% Ethanol was added and

precipitated at -70°C for an hour. After centrifuging and decanting, the precipitate was

resuspended in 1x TE, pH 8.0 and once more co-incubated with Ammonium Acetate

and 100% Ethanol for an hour at -70°C. The supernatant was again discarded and the

pellet allowed to dry for 90 minutes prior to re-suspension in 1x TE, pH 8.0 and storage

overnight at -20°C. Samples (20µl) were loaded together with DNA loading buffer into

a 2% agarose gel and run at 60V for 3 hours in TAE buffer. Gels were dried for 2 hours

in a slab-dryer before being wrapped in plastic film, exposed to X-ray film in a

darkroom and placed in a light-impervious cassette and incubated at -70°C for 6 days.


31

In a darkroom the X-ray film was removed from the cassette and incubated in

Developing Solution for 1 minute before rinsing in H2O. The film was transferred to

Fixing Solution for 10 seconds and air dried for several hours.

Meanwhile, the gel was allowed to thaw for 1 hour before each lane of the gel

was carefully excised and placed in emulsion agent. Vials were placed in a scintillation

counter and the radiation content of each sample was quantified. The extent of DNA

fragmentation was compared between control BeWo cells and those treated with 1μM

Staurosporine. This provided an estimate of the extent of apoptosis induced by

Staurosporine.

2.4 Assessment of RNA Expression

To examine the expressions of gene transcripts within each experimental model, total

RNA was extracted, treated with DNase to remove any double stranded DNA, reverse

transcribed to produce cDNA, and any incurred contaminations removed. In this way a

known concentration (1µg) of mRNA was transcribed into cDNA. This cDNA was then

analysed for the expression of specific genes using Real Time Quantitative Polymerase

Chain Reaction (qPCR) and standardised against the expression of the internal control

L19 to assess the relative mRNA quantities of specific genes in each treatment group.

2.4.1 RNA extractions

The protocol for extracting RNA differed slightly between the cell culture and tissue

explant models, and the variations are outlined below.

2.4.1.1 RNA extraction from human placental explants

To each sample from both the SOD and tissue sample models, 1ml of TriReagent

(Sigma, Cat. No. T9424) was added and the sample homogenised and incubated at room

temperature for 5mins. To this, 200μl of chloroform was added and vigorously shaken

for 15 seconds prior to incubation at room temperature for 10mins. Following

centrifugation at 12000g for 15 minutes at 4ºC, the upper phase was transferred to a

fresh tube, to which 500μl isopropanol was added. This was vortexed and incubated for

10mins at room temperature. Samples were centrifuged at 12000g for 8mins at 4ºC,

then the supernatant removed and the pellet washed with 1ml of 70% ethanol. This was

centrifuged at 7600g for 5 minutes at 4ºC and the supernatant removed. The pellet was

allowed to air dry for up to 5 minutes, ensuring the pellet did not dry completely. The


32

final RNA pellet was dissolved in 50μl of RNase-free diethylpyrocarbonate-treated H2O

(DEPC H2O). A 2μl sample of RNA was diluted 1:50 and its quantity and quality

analysed with the ultraviolet spectrophotometer. The final concentration was determined

by multiplying the 260nm wavelength value by the dilution factor (50) and RNA factor

(40). Good quality RNAs were taken to be samples with a

260nm:280nmspectrophotometry ratio in excess of 1.6.

2.4.1.2 RNA extraction from the cell culture (BeWo) model

Total RNA was extracted from BeWo cells using the TriReagent method. Briefly, the

supernatant was collected and centrifuged to gather any suspended cells, while the

adherent cells were treated with 1ml TriReagent for 5 minutes at room temperature.

Next, 200μl chloroform was added to extracts and incubated for 10 minutes following

vigorous mixing. Samples were then centrifuged, the upper aqueous phase collected,

and 500μl isopropanol added and incubated overnight at 4ºC. Following pelleting, the

supernatant was removed the pellet washed with 1ml 75% ethanol. The sample was

again spun, the supernatant decanted and the pellet air dried for 5 minutes. Next the

pellet was dissolved in 20μl DEPC H2O, 2μl of which was diluted 1:50 in dH2O and the

quantity and quality of the RNA determined under UV-light using the

spectrophotometer. Total RNA concentration (μg/ml) was determined by multiplying

the absorbance at 260nm by the pathlength factor (40) and dilution factor (50).

Additionally, 5μl of RNA was run on a 2% agarose gel to look for the presence of

distinct 28s, 18s and 5s bands indicating high quality. The remainder of the RNA was

stored at -80ºC for later treatment.

2.4.2 DNAse Treatment and Reverse Transcription of Total RNA

2µg of RNA was made up to 7µl with DEPC H2O. To this 1μl of 10x DNase buffer and

2μl of RQ1 DNase (Promega,Cat. No. M6101) was added and incubated for 30 minutes

at 37ºC. To inactivate this reaction, 1μl DNase stop solution was added and incubated at

65°C for 10 minutes. 1µg of RNA (5.5µl) was removed and made up to 13.5µl with

DEPC H2O prior to addition of 0.5μl of random primers were added and incubated for 5

minutes at 70ºC. After chilling, 5μl of 5x reaction buffer, 1.3μl 10mM Deoxynucleotide

Triphosphates (dNTPs) (Promega, Cat. No. U1330), 1μl Moloney Murine Leukemia

Virus (MMLV) enzyme (Promega, Cat. No. M3682) and 3.7μl DEPC H2O were added.

Samples were reverse transcribed by heating to 25ºC for 10 minutes, 55ºC for 50


33

minutes and 70ºC for 15 minutes. Samples were then stored at -20ºC until required for

PCR.

2.4.3 Post-PCR Clean-up

To remove excess primers, dNTPs, enzyme and buffering salts from the resultant

cDNA, samples were processed using the UltraClean™ PCR Clean-up kit from MoBio

Laboratories, Inc. (Cat No 12500). Briefly, each PCR reaction was treated with a 5-

times volume (125µl) of SpinBind (Guanidine HCl/Isopropanol), mixed well and

transferred to a spin filter unit and centrifuged at 13000rpm for 20 seconds. The flow-

through was discarded, the filter basket returned to the tube and 300µl of SpinClean

(<80% Ethanol solution) added to the filter. The spin column was then spun twice at

13000rpm for 20 seconds, with the flow-through removed after each spin. The filter

basket was then removed to a clean 1.5ml tube and 50µl of Elution Buffer (10mM Tris,

pH 8.0) added directly to the filter. The spin column was centrifuged at 13000rpm for

30 seconds, the filter basket disposed of, with the resultant flow-through (cDNA) stored

at -20°C.

2.4.4 Dilution of Primers

Primers used for RT and Real Time PCR (Table 2.1) were diluted to a working

concentration of 5μM. Briefly, primer stocks were resuspended in 200μl of 1x TE

Buffer and allowed to incubate at room temperature for 5mins. 5μl of diluted primer

was further diluted in 195μl DEPC H2O. This was vortexed and the primer

concentration quantified by UV-spectrophotometry. The stock concentration (μg/ml) of

primer was determined by multiplying the absorbance at a wavelength of 260nm by the

dilution factor (40) and the ssDNA factor (33). To convert this to μM, the concentration

was divided by the Molecular Weight of the primer, and multiplied by 1000. To dilute

this to a 5μM working solution, 5μl of primer stock was diluted in a volume (μl) of

dH2O equivalent to the stock concentration (μM), minus 5. Thus, if the stock primer

concentration was 300μM, the volume of DEPC H2O required to dilute 5μl of the

primer to 5μM would be 295μl.


34

Primer Accession Number Primer Sequence Size

Caspase-14 NM_012114 Forward5‟-tgcacgtttattccacggta-3‟

Reverse 5‟-tgctttggatttcagggttc-3‟ 204bp

Caspase-3 NM_004346 Forward5‟-cccatttctccatacgcact-3‟

Reverse 5‟-tgacagccagtgagacttgg-3‟ 358bp

Caspase-8 NM_001228 Forward 5‟-aagcaaacctcggggatact-3‟

Reverse 5‟-ggggcttgatctcaaaatga-3‟ 164bp

cytokeratin-18 NM_000224 Forward 5‟-cacagtctgctgaggttgga-3‟

Reverse 5‟-gagctgctccatctgtaggg-3‟ 164bp

E-cadherin NM_004360 Forward 5‟-tgcccagaaaatgaaaaagg-3‟

Reverse 5‟-gtgtatgtggcaatgcgttc-3‟ 200bp

eNOS NM_000603 Forward 5‟-accctcaccgctacaacatc-3‟

Reverse 5‟-gctcattctccaggtgcttc-3‟ 198bp

Filaggrin NM_002016 Forward 5‟-ggcaaatcctgaagaatcca-3‟

Reverse 5‟-tgctttctgtgcttgtgtcc-3‟ 187bp

GAPDH NM_002046 Forward 5‟-cagaacatcatccctgcatccact-3‟

Reverse 5‟-gttgctgttgaagtcacaggagac-3‟ 185bp

β-hCG NM_001009050 Forward 5‟-gcaccaaggatggagatgtt-3‟

Reverse 5‟-gcacagatggtggtgttgac-3‟ 173bp

hPL NM_001317 Forward 5‟-ccgttatccaggctttttga-3‟

Reverse 5‟-tggagggtgtcggaatagag-3‟ 175bp

Ki67 NM_002417 Forward 5‟-agtcagacccagtggacacc-3‟

Reverse 5‟-tgctgccggttaagttctct-3‟ 225bp

KLF4 NM_004235 Forward 5‟-cccacacaggtgagaaacct-3‟

Reverse 5‟-atgtgtaaggcgaggtggtc-3‟ 169bp

L-19 NM_000981 Forward 5‟-ctgaaggtcaaagggaatgtg-3‟

Reverse 5‟-ggacagagtcttgatgatctc-3‟ 194bp

sFRP4 NM_003014 Forward 5‟-gcctgggacagcctatgtaa-3‟

Reverse 5‟-tctgtaccaaagggcaaacc-3‟ 160bp

VEGF-A NM_001025366 Forward 5‟-cccactgaggagtccaacat-3‟

Reverse 5‟-tttcttgcgctttcgttttt-3‟ 186bp

Table 2.1 Primer Accession Numbers, Sequences and Sizes.

2.4.5 Real Time Quantitative PCR

Real Time Quantitative PCR was conducted in a mixture of the following components:

5μl SybrIQ (BioRad, Cat. No. 170-8882); 1μl forward primer; 1μl reverse primer; 2μl

DEPC H2O; and 1μl cDNA. This mixture was denatured, annealed and extended for at

least 40 cycles thereof following the conditions outlined for each primer set in Table

2.2. Each annealed primer-cDNA set denatures into single strands within a small

temperature range according to its length, sequence and GC content (Ririe et al. 1997).


35

Product Start Denature Annealing Extension Cycles

Caspase-14 95ºC; 3 mins 95ºC; 1 secs 60ºC; 15 secs 72ºC; 5 secs 40

Caspase-3 94°C; 10 mins 94C; 5 secs 58°C; 15 secs 72°C; 15 secs 35

Caspase-8 95ºC; 3 mins 95ºC; 0 secs 60ºC; 15 secs 72ºC; 5 secs 40

cytokeratin-18 95ºC; 5 mins 95ºC; 1 sec 60ºC; 15 secs 72ºC; 5 secs 40

E-cadherin 95ºC; 3 mins 95ºC; 0 secs 60ºC; 15 secs 72ºC; 5 secs 40

eNOS 95ºC; 3 mins 95ºC; 1 secs 60ºC; 15 secs 72ºC; 5 secs 40

Filaggrin 95ºC; 3 mins 95ºC; 1 secs 60ºC; 15 secs 72ºC; 5 secs 50

GAPDH 95ºC; 3 mins 95ºC; 1 secs 60ºC; 15 secs 72ºC; 5 secs 40

β-hCG 95ºC; 3 mins 95ºC; 1 secs 60ºC; 15 secs 72ºC; 5 secs 40

hPL 95ºC; 5 mins 95ºC; 1 sec 60ºC; 15 secs 72ºC; 5 secs 40

Ki67 95ºC; 3 mins 95ºC; 0 secs 60ºC; 15 secs 72ºC; 5 secs 40

KLF4 95ºC; 3 mins 95ºC; 0 secs 60ºC; 15 secs 72ºC; 5 secs 40

L-19 95ºC; 3 mins 95ºC; 1 secs 51ºC; 15 secs 72ºC; 10 secs 40

sFRP4 95ºC; 3 mins 95ºC; 1 secs 60ºC; 15 secs 72ºC; 5 secs 45

VEGF-A 95°C; 5 mins 95°C; 1sec 58°C; 15 secs 72°C; 5 secs 45

Table 2.2 Real-Time PCR Cycling Conditions.

Melt curves were constructed for each primer set by heating the reaction in a

0.5ºC stepwise fashion from 72ºC up to 99ºC to ensure the exclusive melting of the

desired product (See Appendix 1). The fluorescence of the sample is plotted against the

temperature, with a typical spike in fluorescence occurring at the temperature at which

the product denatures. Accordingly, the desired product can be identified, as well as any

contaminants such as primer-to-primer binding. Thus the quality of the reaction and the

final product can be controlled and only „clean‟ reactions used for quantification

purposes.

To ensure accurate quantitation of the specific product, a post-PCR sample was

run on a 1.5% agarose gel containing 200ng/ml Ethidium Bromide to ensure the correct

sized product was being amplified. This band was then excised and the total product

extracted from the gel using a QIAquick® Gel Extraction Kit (Qiagen, Australia, Cat.

No. 28704). The resultant solution was then serially diluted 1:10 10 times in DEPC

H2O. These standards were given arbitrary units such that Standard #1 was 10 times

greater than Standard #2, and so on. These were then cycled as per Real Time RT-PCR

to construct a Standard Curve (R2>0.99), with the experimental samples assigned a

value depending on where the fit on the standard curve. By standardising these results

against the human ribosomal L19 internal control, the amount of product in each sample

is quantifiable.


36

2.5 Assessment of Protein Expression and Localisation

2.5.1 Immunocytochemistry

Protein expression and its subcellular localisation can be determined through the use of

immunocytochemistry. BeWo cells were grown on coverslips and treated with 20μM

Forskolin for 24, 48, 72 and 96 hours. At each time point coverslips were fixed and

stained for the expression of E-cadherin protein.

Briefly, cells were fixed with 4% paraformaldehyde in 1x PBS containing 0.6%

Triton X-100 for 15 minutes. Non-specific binding was then blocked with 1% goat

serum in PBS containing 1% BSA (blocking buffer) for 60 minutes. Next, the E-

cadherin (Sigma, USA, Cat. No. U3254) primary antibody was diluted 1:500 in

blocking buffer and added to the cells for 60 minutes. After washing with blocking

buffer, cells were treated with an antibody mixture containing a FITC-conjugated

Alexa488 (Sigma, USA, Cat. No. F2012) secondary antibody (1:500), TRITC- labelled

Phalloidin (Sigma, USA, Cat. No. P1951) (1:500) actin stain and Hoechst nuclear dye

(Sigma, USA, Cat. No. B2261) (1:500) for 90 minutes at room temperature. Coverslips

were then placed on slides using PVA and imaged using a fluorescent microscope at

Cell Central (ANHB, UWA).

2.5.2 Immunohistochemistry

Chorionic villous samples were excised from first trimester and term placentae and

fixed in 4% formaldehyde in 0.1M PBS, pH 7.3 for 4 hrs and washed overnight in cold

0.1M PBS. Samples were gradually dehydrated in increasing concentrations of ethanol

(70%, 90%, and 100%) for 2 hours each step. The 100% ethanol step was repeated 3

times to ensure complete dehydration. Samples were next washed twice in toluene for 1

hour each time, before undergoing paraffin wax embedding twice for 1 hour each time.

Finally, blocks were vacuumed for 30 minutes to ensure complete immersion in

paraffin.

Embedded samples were cut uniformly at a thickness of 5µm and mounted on

glass slides that had been pre-salinated in 2% (v/v) aminopropyltriethoxysaline (Sigma,

USA) in acetone for 5 minutes prior to rinsing in double distilled water and drying

overnight at 37°C. Following mounting, samples were air-dried at 45°C for at least 24

hours prior to staining.

Slides were incubated at 65°C for 15 minutes and de-waxed by washing twice,

for 5 minutes each time, with toluene. Slides were rehydrated in descending ethanol


37

concentrations from 100% to 0% ethanol in dH2O. Antigen retrieval was performed

using microwave treatment in 10mM sodium citrate, pH 7.0 for 5 minutes (maximum

temperature, 95°C). Slides were cooled for 20 minutes at room temperature and rinse 3

times in Tris Buffered Saline (TBS-T) composed of 1L dH2O, 12.1g Tris, 9g NaCl, 1ml

Tween-20, pH 7.5 for 5 minutes each prior to immunostaining.

Non-specific antigen binding was blocked by incubating slides for 1 hour in

20% goat serum diluted in TBS-T. Slides were washed three times for 5 minutes each,

and incubated with primary antibodies (Table 2.3) diluted 1:100 in TBS-T overnight at

4°C. Following this, slides were again washed 3 times with TBS-T prior to an hour‟s

incubation with a biotinylated goat anti-rabbit IgG secondary antibody (Dako). Slides

were washed as before and treated with an avidin-biotin-complex (Dako) coupled to

horseradish peroxidase. Slides were cover-slipped and sealed with a commercial nail

varnish.

Target Source Species Use Size (kDa) Dilution

Caspase-14 Wim Declercq, Ghent,

Belgium

rabbit IB 29; 20; 18.5 1:2500

Caspase-14 BD, 611511 mouse IB, IHC 36 1:1000

β-hCG Dako, A0231 rabbit IB, IHC 28 1:6000

hPL Dako, A0137 rabbit IHC N/A 1:5000

E-cadherin Sigma, U3254 rat IF 120; 84 1:500

GAPDH Invitrogen, 39-8600 mouse IB 42 1:5000

Cytokeratin-18 Calbiochem, AP1021 rabbit IB 44 1:5000

KLF4 Zymed Labs, 42-4100 rabbit IB 58 1:5000

Hoechst Sigma, B2261 NA IF N/A 1:500

FITC IgG Sigma, F2012 goat IF N/A 1:125

Table 2.3 Antibodies used for protein detection. IHC= Immunohistochmistry; IF=

Immunofluorescence; IB= Immunoblotting.

2.5.3 Extraction of Protein

Protein was extracted from both BeWo cells and human placental samples using

Radioimmunoprecipitation (RIPA) Buffer consisting of 150mM NaCl, 50mM Tris-HCl,

pH 7.5, 1% Triton X-100, 0.5% Sodium Deoxycholate, 0.1% SDS and 0.1mM

Phenylmethyl-sulfonyl fluoride (PMSF) in dH2O. This was made up fresh for every

extraction by combining 50mg Sodium Deoxycholate, 1.5ml 1M NaCl, 100μl of 10%

SDS, 500μl 1M Tris-HCl (pH 7.5) and 100μl Triton X-100 and dissolving in 10ml of

dH2O. Immediately prior to use, 10μl of 100mM PMSF was added. Upon the addition


38

of RIPA Buffer, placental samples were homogenised and incubated on ice for 30

minutes. Samples were then centrifuged and the supernatant stored at -80°C.

From the BeWo cell line the media was removed and spun at maximum speed

for 3minutes. 1ml of 1M PBS was added to the well and the cells dislodged using a cell

scraper. The supernatant was discarded from the spun vial, and the scraped cells in PBS

were added and spun at maximum speed for 3 minutes. All the supernatant was

removed thoroughly and 50μl of RIPA Buffer added and vortexed. The sample was

incubated on ice for 30 minutes, after which the sample was again vortexed and spun at

maximum speed for 5 minutes at 4ºC, and the supernatant collected and stored at -80ºC.

2.5.4 Bradford Assay for Protein Quantitation

Protein standards were made by serial dilution of Acetylated Bovine Serum Albumin

(BSA) (Promega, USA; Cat. No R3961) to create standards ranging from 500 to

100μg/ml protein. The 500μg/ml standard (Standard A) was created by diluting 50μl of

BSA in 950μl of 0.01M PBS. 760μl of Standard A was diluted with 190μl 0.01M PBS

to create Standard B (400μg/ml). A 300μg/ml Standard C was made by diluting 750μl

of Standard B in 250μl 0.01M PBS. This was diluted further by adding 300μl of 0.01M

PBS to 600μl of Standard C to make a 200μg/ml Standard D. 400μl of Standard D was

diluted in an equal volume of 0.01M PBS to make the 100μg/ml Standard E.

Protein samples were diluted 1:20 in 0.01M PBS and 10μl transferred into glass

tubes. In duplicate, 10μl of each Standard was transferred to glass tubes, to which 200μl

of protein dye was added, vortexed and incubate for 5 minutes at room temperature. The

colour of the protein samples was checked against the Standards to ensure their colour

would fit within the standard curve. 0.01M PBS was used as a zero value. Standards

were analysed by spectrophotometry using visible light and a standard curve produced.

Samples were examined by adding 50μl to the cuvette and analysing with the

spectrophotometer, and the concentration of protein (μg/ml) extrapolated from the

standard curve.

2.5.5 Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS-PAGE)

SDS-PAGE is the process by which protein samples are separated into fragments

according to their molecular weight. This is achieved through loading protein into an

acrylamide gel and running a current through the gel. Proteins travel through the gel

according to their molecular weight, charge and conformation. Once sufficiently


39

separated, proteins are transferred from the gel onto a nitrocellulose membrane. This

enables the identification of specific proteins via Immunoblotting.

Glass plates with 1mm spacers and their accompanying smaller glass plate were

first cleaned with 70% ethanol before being aligned and placed into their apparatus. A

12% Separating Gel consisting of 3.35ml dH2O, 2.5ml 1.5M Tris-HCl (pH 8.8), 4ml

300% Acrylamide-bis, 100μl of 10% SDS, 10μl of N,N,N',N'-

Tetramethylethylenediamine (TEMED) and 100μl of 10% ammonium persulfate (APS)

was made up in a 15ml tube. The Separating Gel was carefully transferred to the space

between the glass plates to a level 2.5cm below the top of the smaller plate. In order to

ensure a smooth upper edge of the separating gel, 100μl of dH2O was slowly added to

either side of the space, and the gel allowed to set for 15 minutes.

A Stacking Gel consisting of 3.05ml dH2O, 1.25ml 0.5M Tris-HCl (pH 6.8),

650μl of 30% Acrylamide-bis, 50μl 10% SDS, 5μl TEMED and 25μl 10% APS was

made up in a 15ml tube. The dH2O atop the Separating Gel was removed by blotting,

and the Stacking Gel gently poured on top and the comb placed into the top whilst

ensuring there were no air bubbles. The Stacking Gel was allowed to set for 20 minutes.

Gels were placed into the electrophoresis tank with the smaller plates facing each other,

and the reservoir between them filled with Electrode Buffer consisting of 25mM Tris-

HCl, 200mM glycine and 0.1% SDS. Electrode Buffer was also added to the

electrophoresis tank to a depth of 2cm.

In duplicate, 30µg of each protein sample was taken and made up to an equal

volume between samples with 0.01M PBS. An equal volume of Loading Buffer

consisting of 4% SDS, 2% β-mercaptoethanol, 20% v/v glycerol, 250mM Tris-HCl (pH

6.8) and 0.006% bromophenolblue was added to the protein. The mixture was incubated

at 95ºC for 5 minutes in order to denature the proteins. After carefully removing the

comb to ensure there were no air bubbles, samples were added to each well. To the first

well on each gel, 5μl of SeeBlue Plus 2 molecular weight standard (Invitrogen, Cat. No.

LC5925) was added. Proteins were loaded so that the two gels were an identical mirror

image of one another. This ensured that samples on both gels ended up with the same

loading pattern.

After loading the protein, the electrodes were inserted into the apparatus and the

electrical source, with the black electrode connected to the negative terminal. The

apparatus was set to a voltage of 100V until the samples reached the top of the

Separating Gel. The voltage was then increased to 130V and run until the 6kDa

molecular weight band reached the bottom of the Separating Gel. Once finished, the


40

gels were removed from the electrophoresis tank, and the glass plates gently prised apart

to remove the gel.

In preparation for the transfer of proteins to a Hybond-C supported

nitrocellulose membrane, 4 pieces of blotting paper and 2 nitrocellulose membranes

(one for each gel) were cut to the required size and, along with 6 thin pads, soaked in

Transfer Buffer (800ml dH2O, 3.03g Tris, 14.4g Glycine, 200ml Methanol). The

transfer cassettes were set up with one pad, then a piece of blotting paper, the separating

gel, the nitrocellulose membrane, more blotting paper and 2 more pads, with each

component layered consecutively upon the black side of the rack. The contents were

kept wet throughout, and any air bubbles were removed from between the gel and

membrane with a rolling pin. Carefully, the racks were closed and inserted into a

transfer apparatus with the black side to the back. A magnetic flea was placed into the

bottom of the electrophoresis tank and the transfer apparatus and an ice brick placed in

over the top. The tank was then filled with Transfer Buffer, the electrodes attached and

run at 100V for 75 minutes.

Following the transfer, the gel was discarded and the nitrocellulose membrane

rinsed quickly in TBS-T prior to being briefly covered with Ponceau S (2% Ponceau S,

2%Trichloroacetic acid (TCA)) to reveal all transferred proteins. The remainder of the

Ponceau S was returned to the bottle, and the excess removed by repeated washes of

dH2O. The membrane was cut into strips containing the desired range of protein sizes

for Immunoblotting. The remaining Ponceau S was removed from the membrane with

TBS-T for 5 minutes, and the membranes stored in sealed plastic bags at 4ºC until

required for Immunoblotting.

2.5.6 Immunoblotting

In order to prevent non-specific antigen binding, membranes were blocked in 5% instant

milk powder (Diploma) in TBS-T for 30 minutes at room temperature. Membranes were

then incubated in 10ml of appropriately diluted primary antibody overnight at 4°Cwhile

shaking (Table 2.3). All samples were standardised against either β-actin or GAPDH as

an internal loading control.

Membranes were washed with TBS-T three times for 10 minutes each to remove

any unbound antibody. Next, they were incubated for 90 minutes in 10ml of primary

antibody-specific HRP-conjugated secondary antibody diluted in TBS-T. After

incubation, excess liquid was removed completely with blotting paper, and 1ml each of


41

the two SuperSignal West Pico Chemiluminescent Substrates (ECL) (Pierce, Cat. No.

34080) were combined and placed onto the membrane and incubated for 5 minutes.

Excess liquid was again removed completely with blotting paper, and the membrane

bagged and imaged on the ImageStation (Kodak, IS2000MM).

Following imaging, antibodies were removed from the membranes using

Stripping Buffer composed of 1% β-mercaptoethanol in TBS-T for 30 minutes.

Membranes were then re-probed with different antibodies to assess protein expression

from the same samples.

For the sake of uniformity and readability of results, the order of the gels lanes

was rearranged using Adobe Photoshop CS2. However it must be stressed that each lane

shown within each Figure was run on the same gel.

2.6 Stimulation of the Endothelial Pathway

Analysis of the production of the known vasodilator Nitric Oxide (NO) was conducted

by stimulating BeWo cells with an array of stimulators of the NO endothelial pathway.

Calcium Chloride (10µM; Sigma, Cat No. C1016), Vascular Endothelial Growth Factor

Type A (VEGF-A) (10µg/ml; Sigma, Cat No. V7259), Bradykinin (10µM; Sigma, Cat

No. B3259), either with or without the NO inhibitor L-NAME (1mM; Sigma, Cat No.

N5751) were added to the media of BeWo cells that had been incubated for 0, 24, 48 or

72 hours with either Forskolin to initiate differentiation or DMSO as a control. Fifteen

minutes was allowed for the chemicals to take effect, and then RNA was extracted as

described in Section 2.4.1.

2.7 RNA Interference

The ability to selectively and transiently silence individual gene products is a technique

developed within the past decade, earning its inventors (A. Fire and C. Mello) the Nobel

Prize in Physiology or Medicine in 2006. Gene products are selectively silenced through

the cleavage of their respective mRNA, preventing translation into the functional

protein (Fire et al. 1998). Upon the introduction of gene-specific double stranded RNA

(dsRNA) to a cell the dsRNA binds to the Dicer protein complex, which cleaves it into

21-23bp fragments known as small interfering RNA (siRNA). These siRNA are then

bound by the RNA-induced silencing complex (RISC), the sense strand discarded while

the antisense strand guides the complex to complimentary mRNA. The RISC cleaves

this mRNA and marks it for degradation, preventing its translation into protein and thus


42

effectively silencing the gene. For in vitro silencing of genes in mammalian cells, pre-

fabricated synthetic siRNA are generally used, thereby bypassing the Dicer protein

complex step in the process.

2.7.1 Optimisation of siRNA

To assess the efficiency of RNAi in the BeWo cell line, siRNA oligonucleotides for

GAPDH (Invitrogen, Cat. No. 12935-140) were used as a positive control. Titration of

the GAPDH siRNA was conducted at concentrations of 25, 50, 100 and 200nM, paired

against RNAi Negative Control Duplexes (Invitrogen, Cat. No. 12935-300).

Oligonucleotides were delivered into the cells using the transfection agent

Lipofectamine2000 (Invitrogen, Cat. No. 11668). Briefly, 1.25x105 cells/ml BeWo cells

were added to a 24-well plate and incubated overnight at 37°C and 5% CO2. Aliquots of

the siRNA/Lipofectamine2000 complex were made by combining siRNA with twice the

volume of Lipofectamine2000 and diluted in OptiMEM (Gibco, Cat. No. 31985).

Aliquots were incubated at room temperature for 15 minutes prior to addition to cells in

a dropwise manner. Plates were incubated at 37°C and 5% CO2 for 24 hours, at which

time media was refreshed. Following a further 24 hours of incubation, RNA and protein

were extracted for confirmation of gene silencing as outlined in Section 2.4 and Section

2.5, respectively.

2.7.2 Silencing of Caspase-14

24 hours after plating BeWo cells into 24-well plates, cells were treated with 100nM of

one of three sets of discrete annealed 25bp caspase-14 siRNA (Invitrogen, Cat.

No.1299003) (Table 2.4). Media was replaced for fresh growth media 24 hours later.

RNA and protein were extracted from cultures 48 hours after the initial treatment with

the siRNA.

siRNA were delivered into BeWo cells using Lipofectamine2000 transfection

reagent. Briefly, for each well 4µl of Lipofectamine2000 was diluted in 21µl OptiMEM

and allowed to settle for 10 minutes. For each well, 5µl of either caspase-14 or

scrambled siRNA were added to 20µl of OptiMEM, and 25µl of diluted

Lipofectamine2000 mixed with this for each well. These conjugates were allowed to

rest for 15 minutes before 50µl was added to each well in a drop-wise manner. Plates

were gently rocked for 30 seconds and incubated at 37°C and 5% CO2.


43

To assess the influence of caspase-14 on the differentiation of BeWo cells, 24

hours after silencing commenced, 20µM Forskolin was added. Differentiation was

allowed to progress for 24, 48 and 72 hours, after which RNA and protein were

extracted and treated as outlined in Sections 2.4 and 2.5.

Name Description Sequence Exon

Caspase-14.1 HSS118993 (sense) ggcccuaauacugugugucaccaaa 2

HSS118993 (antisense) uuuggugacacacaguauuagggcc 3

Caspase-14.2 HSS118994 (sense) ccuggaugcucuggaacacauguuu 3

HSS118994 (antisense) aaacauguguuccagagcauccagg 5

Caspase-14.3 HSS118995 (sense) gggagauggucaagcuggagaaucu 6

HSS118995 (antisense) agauucuccagcuugaccaucuccc 7

Table 2.4 Details of caspase-14 siRNAs

2.8 Statistical Analysis

Data obtained for gene expression in the human placenta between First Trimester (n=8),

Term Placenta (n=12) and Preeclamptic samples (n=5) was submitted to Analysis of

Variance (ANOVA) in Microsoft Excel. Statistical significance was achieved at P<0.05.

Linear Regressions (Pearson‟s correlation) were used when determining the relationship

between gene expression and gestational age of the sample. The correlation between

gene expression and gestational age was considered small if R was between 0.1 and

0.29; medium if between 0.3 and 0.49; or strong if greater than 0.5. Cohen‟s d was also

calculated to determine the effect size. A d value in excess of 0.2 was considered a

small; 0.5 a medium; and an effect size greater than 0.8 was considered large (i.e. good).

Data for Staurosporine and Forskolin-treated BeWo cell gene expression

(Chapters 4 and 5) were obtained from 4 independent samples within each group, and

submitted to One Way ANOVA‟s as a test of statistical significance.

Data obtained for BeWo cells treated with an array of endothelial pathway

mediators (Chapter 6) were obtained from 4 independent samples within each group,

and subjected to One Way ANOVA‟s with Bonferroni‟s correction. A P-value below

0.05 was considered significant.

Furthermore, all data obtained from the RNA Interference experiments (Chapter

7) were obtained from 8 independent samples in each group, and submitted to one-way

ANOVA‟s to determine statistical significance, where P<0.05 was deemed significant.


44

All bar graphs presented within this thesis show the Mean and Standard Error of

the Mean.

Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia

45

Chapter 3

Placental Gene Expression across Human Gestation and in

Preeclampsia

3.1 Background

The human placenta is a dynamic organ which adapts according to the local

environment and the changing needs of the developing fetus. Consequently, both

genotypic and phenotypic changes occur in the placenta and its constituent tissues,

particularly the trophoblast, to meet these requirements. These changes may affect the

signalling pathways modulating the multiple functions of the placenta, including

hormonal, differentiation, apoptotic and endothelial pathways. Gene transcription

profile alterations were examined between the first and third trimesters of normal

human placentae. Genes examined encompassed a wide variety of placental functions,

and also novel genes as yet unexamined in the human placenta. This therefore gave us

an insight as to the dynamic alterations undertaken by the placenta during gestation and

in disease as it is continually remodelled to provide the best environment for the growth

of the fetus.

Data presented in Section 3.5 regarding sFRP4 expression and localisation in the

human placenta have been accepted for publication by the journal „Reproductive

BioMedicine Online, and is awaiting assignment to a specific issue (see

http://www.rbmonline.com/4DCGI/Article/Detail?38%091%09=%203274%09 ).

Aim: To examine alterations in the gene expression profile of first trimester and term

human placentae, and how these related to preeclampsia.

Hypothesis: Genes involved in hormonal, endothelial and differentiation pathways will

be altered during gestational development, and disrupted in preeclampsia.

http://www.rbmonline.com/4DCGI/Article/Detail?38%091%09=%203274%09


46

3.2 Techniques

3.2.1 Real Time Polymerase Chain Reaction

Gene transcripts were quantified and validated against the internal standard L19 for

each experimental group using Real Time PCR. Using this technique gene expression of

β-hCG, eNOS and sFRP4 were examined. Details of the methods used are given in

Section 2.4.

3.2.2 Immunohistochemistry

To assess the presence and cellular localisation of the various proteins of interest, term

human placental samples were examined by immunohistochemistry as outlined in

Section 2.5.2. Briefly, samples were fixed with formaldehyde and embedded within

paraffin blocks. Following sectioning and mounting onto glass slides, samples were

stained with primary and secondary antibodies to identify each specific protein.

3.2.3 Explant Culture Model

To study apoptosis in the first trimester human placenta following delivery, explants

were treated as outlined in Section 2.3.3. Briefly, human first trimester placental

samples were incubated with the free radical scavenger superoxide dismutase (SOD) to

suppress apoptosis. Those incubated without SOD undergo spontaneous apoptosis. Thus

the SOD model allows the investigator to assess the role of apoptosis in the human

placenta (Charles et al. 2005; Kam et al. 2005).

The expression of caspase-14 mRNA in the SOD model of placental apoptosis

was determined using RT-PCR (Figure 3.1). Fresh and serum-free-incubated placentae

expressed caspase-14 mRNA, however first trimester placentae incubated for 4 hours in

the presence of SOD were void of the transcript. This suggests that caspase-14 is up

regulated by oxidative stress in the placenta. Thus caspase-14 appears to be involved in

the regulation of apoptosis in the first trimester placenta.


47

Figure 3.1 RT-PCR for caspase-14 (204bp) in the first trimester human placenta

during apoptosis or its inhibition by superoxide dismutase.

3.3 Hormonal Function of the Placenta

3.3.1 Beta-human chorionic gonadotrophin (β-hCG)

As outlined in Section 1.1.2 β-hCG is produced and secreted by the differentiated

villous trophoblast upon the formation of the maternal fetal barrier. The production of

β-hCG by the syncytiotrophoblast is much greater during the first trimester than at term

as it is involved in the maintenance of pregnancy and the auto/paracrine signalling for

differentiation (Shi et al. 1993; Cronier et al. 1994). Consequently, the expression of β-

hCG mRNA and protein were examined and compared between human first trimester,

term and preeclamptic placentae. As expected, β-hCG mRNA levels were substantially

higher in first trimester human placental samples than in all other experimental groups

(P<0.05) (Figure 3.2). Strong effect sizes (d>> 0.8) were noted between first trimester

placentae and the other sample groups. Accordingly, a strong Pearson‟s correlation

(R=0.63) was observed between β-hCG mRNA expression and gestational age (Figure

3.3). Therefore β-hCG expression is diminished across gestation.

It has been suggested that β-hCG may be reduced in placentae complicated with

preeclampsia (Fox et al. 2007), however no difference was noted between preeclamptic

and normal term samples (P>0.05). This may be due to the age of the placentas being

examined, as β-hCG expression is greatly reduced from the second trimester onward

when progesterone production increases to maintain the pregnancy. Due to the nature of

preeclampsia as a mid-late onset disease, early gestation preeclamptic placentae are

difficult to obtain. Indeed, the sampled preeclamptic placentae were aged between 31


48

and 38 weeks, well into the third trimester of gestation. Therefore, baseline β-hCG

mRNA levels were already low in comparison with the first trimester.

Figure 3.2 The expression of β-hCG mRNA in the human placenta (*=P<0.05 vs.

First Trimester).

Figure 3.3 The change in β-hCG mRNA expression throughout gestation (R=-

0.63).

As shown in Figure 3.4A, hCG is strongly expressed in the human first trimester

placenta. Corresponding with the known pattern, staining appears particularly intense in

the syncytiotrophoblast, from where it is released into the maternal circulation.

However, some mild expression of β-hCG is also evident in the underlying

cytotrophoblast. This suggests that β-hCG may be synthesised in the cytotrophoblast

and transported to the syncytium upon fusion.

0

5

10

15

20

25

30

35

40

First Trimester Term Preeclampsia

hC

G /

L1

9

Beta-hCG mRNA in the Human Placenta

* *

0

20

40

60

80

0 5 10 15 20 25 30 35 40 45

Ge

ne

Exp

ress

ion

Gestational Age (weeks)

Beta-hCG mRNA across human gestation


49

Figure 3.4 Localisation of A) hCG and B) hPL in the first trimester human

placenta (CTB=cytotrophoblasts; STB=syncytiotrophoblast).

3.3.2 Human Placental Lactogen (hPL)

Human placental lactogen (hPL) is another placentally derived hormone involved in the

maintenance of pregnancy. It is responsible for increasing the maternal metabolism to

support the development of the fetus and in the preparation of the breast for lactation.

As such, its production by the placenta increases across gestation. Accordingly, hPL

mRNA levels were examined in first trimester, term and preeclamptic placental samples

to confirm these previous findings.

Figure 3.5 The expression of hPL mRNA in the human placenta (*=P<0.05 vs.

First Trimester).

Transcriptional regulation of hPL within the placenta was markedly increased at

term compared with the first trimester (P<0.05) (Figure 3.5), and its expression is

strongly correlated with gestational age (R=0.62) (Figure 3.6). This is consistent with

0

1

2

3

4

5


hP

L /

L19

hPL mRNA in the Human Placenta

* *


50

previously published data and confirms the trophoblasticity of my placental samples.

Interestingly, hPL was not differentially expressed between term and preeclamptic

human placentae (P>0.05).

As seen using immunohistochemistry, hPL was exclusively expressed in the

syncytiotrophoblast of the first trimester human placenta (Figure 3.4B). Not only does

this confirm its production in the human trophoblast, it also indicates that hPL is

produced exclusively by the syncytiotrophoblast. However its transcriptional and

translational regulation remains poorly understood.

Figure 3.6 The change in hPL mRNA expression throughout gestation (R=0.62).

3.4 Caspase-14 in the Human Placenta

No significant differences in caspase-14 mRNA expression were noted between the first

trimester and term placenta (P>0.05) (Figure 3.7). Additionally, caspase-14 mRNA was

unchanged between term and preeclamptic placentae (P>0.05), however levels were

statistically elevated in preeclampsia compared with the first trimester (P<0.05).

Furthermore, a small effect size was noted between first trimester and term placentae

(d=-0.36), and a large effect size was noted between first trimester and preeclamptic

placental samples (d=-1.1), indicating potential relationships between these groups and

caspase-14 mRNA expression. However, the preeclamptic samples were obtained from

placentae with ages closely matched to term. Therefore the comparison in caspase-14

expression between control and preeclamptic placentae should be made against term

samples.

0

1

2

3

4

5

6

7

0 5 10 15 20 25 30 35 40 45

hP

L /

L19


hPL mRNA across human gestation


51

When caspase-14 mRNA expression was plotted against gestational age

however, only a mild correlation was noted between the level of expression and the

maturity of the tissue (R=0.13) (Figure 3.8). This indicates that caspase-14 gene

expression is tightly regulated throughout gestation.

Figure 3.7 The expression of caspase-14 mRNA in the human placenta (*=P<0.05

vs. First Trimester).

The slight increase in caspase-14 mRNA across gestation (Figure 3.8) may

represent an alteration in the gene profile coinciding with placental maturation. This

may be a consequence of the reduction in the pool of proliferative cytotrophoblasts

observed across gestation (i.e. increased syncytiotrophoblast:cytotrophoblast ratio).

Thus, as the syncytiotrophoblast represents an ever increasing proportion of the

trophoblast across gestation, the observed rise in caspase-14 mRNA may reflect a

syncytium-specific expression of caspase-14.

Alternatively, caspase-14 may take a more prominent role in the differentiation

and syncytialisation of trophoblasts as gestation progresses. During the first trimester

the placenta is concerned with establishing new villi, whereas by term villi are already

formed and functional. Increased caspase-14 across gestation may be a feature of the

functional specialisation of the villi. Thus the more focused nature of the differentiation

at term results in increased amounts of syncytialisation-specific genes. This supports the

hypothesis that caspase-14 is a differentiation and syncytialisation-specific protease.

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

1st Trimester Term Preeclampsia

Cas

pas

e-1

4 /

L1

9

Caspase-14 mRNA in the Human Placenta

*


52

Figure 3.8 The change in caspase-14 mRNA expression throughout gestation

(R=0.13).

Immunohistochemistry for caspase-14 was conducted on samples of the human

placenta at first trimester and term. Caspase-14 was found to be expressed exclusively

in the trophoblast layers of the chorionic villi, particularly in the first trimester (Figure

3.9). It is expressed in both the proliferative cytotrophoblast and differentiated

syncytiotrophoblast. The exclusive expression of caspase-14 in the trophoblast indicates

that the mRNA data obtained from the human placenta is representative of the

trophoblast, and also confirms the barrier-specific expression of caspase-14.

Figure 3.9 Immunohistochemistry for caspase-14 in the chorionic villi of first

trimester and term human placentae. Caspase-14 is expressed exclusively in the

trophoblast bi-layer, particularly in the first trimester.

0

0.5

1

1.5

2

2.5

0 5 10 15 20 25 30 35 40 45

Cas

pas

e-1

4 /

L1

9


Caspase-14 mRNA across human gestation


53

Figure 3.10 The expression of filaggrin mRNA in the human placenta.

3.4.1 Profilaggrin

As profilaggrin is a target of caspase-14 activity in the epidermis, it is hypothesised that

it will also be a target in the placenta. However, analysis reveals that profilaggrin

mRNA is not consistently transcribed by the human placenta across gestation and in

preeclampsia (Figure 3.10). Indeed no first trimester samples, only one third of term and

one fifth of preeclamptic placentae were found to express profilaggrin mRNA. Even so,

the levels relative to that observed in the human epidermis were minute (Figure 3.11);

indicating that profilaggrin and its functional filaggrin repeats are not expressed in the

human placenta. Therefore, caspase-14 must have tissue-specific activity.

Figure 3.11 Filaggrin mRNA expression compared between the human placenta

and epidermis (*=P<0.05 vs. 1st Trimester, Term and Preeclampsia).

0

0.004

0.008

0.012

0.016

1st Trimester Term Preeclampsia

Fila

ggri

n /

L1

9

Filaggrin mRNA in the Human Placenta

0

30

60

90

120

150

180

210

1st Trimester Term Preeclampsia skin

Fila

ggri

n /

L1

9

Comparison of Placental versus Epidermal Filaggrin mRNA

*


54

3.4.2 Krüppel Like Factor 4 (KLF4)

KLF4 is a crucial transcription factor in human keratinocyte differentiation (Segre et al.

1999), and is also involved in the maintenance of the placenta across gestation through

its proposed synthesis of pregnancy-specific glycoproteins (Blanchon et al. 2006).

Furthermore, as discussed in Section 1.4.1.3 caspase-14 is a potential binding partner of

KLF4, leading to the investigation of its expression in the human placenta across

gestation.

The expression of KLF4 mRNA was noted throughout gestation, with

significantly increased expression in term compared to first trimester placentae (P<0.05)

(Figure 3.12). This resulted in a small correlation between KLF4 mRNA expression and

gestational age (R=0.15) (Figure 3.13). This suggests that KLF4 transcriptional activity

may be increased with advancing maturation of the trophoblast. The slight increase in

KLF4 expression across gestation is consistent with the elevated caspase-14

transcription (Figure 3.8). This supports a synergism between caspase-14 and KLF4 in

the human trophoblast.

Interestingly, KLF4 mRNA was drastically increased in preeclampsia (P<0.05).

KLF4 regulates the cell cycle by binding to the cyclin D2 promoter, thereby inducing

G1/S-phase arrest (Klaewsongkram et al. 2007). Its elevated expression in preeclampsia

indicates enhancement of KLF4-modulated cell cycle suppression and suggests a

decrease in proliferation as part of the preeclampsia aetiology (Gupta et al. 2005). Thus

proliferation may be reduced in preeclampsia as a result of KLF4 disruption.

Figure 3.12 The expression of KLF4 mRNA in the human placenta (*=P<0.05 vs.

First Trimester; †=P<0.05 vs. First Trimester and Term).

0

50

100

150

200

250

300

350


KLF

4 /

L1

9

KLF4 mRNA in the Human Placenta

*

†


55

Figure 3.13 The change in KLF4 mRNA expression throughout gestation

(R=0.15).

3.4.3 Cytokeratin-18

In the epidermis, caspase-14 indirectly regulates the cytokeratin intermediate filament

architecture. Accordingly, the expression of cytokeratin-18 was examined to confirm

the presence of keratin within the human placenta, and to examine any changes in

expression across gestation or in preeclampsia. Cytokeratin-18 mRNA was detected in

all first trimester, term and preeclamptic placentae; however no alteration in expression

was noted between groups (P>0.05) (Figure 3.14).

Figure 3.14 The expression of cytokeratin-18 mRNA in the human placenta.

0

100

200

300

400

500

0 5 10 15 20 25 30 35 40 45

KLF

4 /

L1

9

gestational age (weeks)

KLF4 mRNA across human gestation

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04


ck1

8 /

L1

9

Cytokeratin-18 mRNA in the Human Placenta


56

Figure 3.15 The change in cytokeratin 18 mRNA expression throughout gestation

(R=-0.27).

Cleavage of cytokeratin-18 is a frequently used marker of trophoblast apoptosis,

so it is no great surprise that it is observed in the placenta. Indeed it may be considered a

trophoblast marker, further verifying the placenta samples used in this thesis as being of

a trophoblast origin.

A small decrease in cytokeratin-18 transcription was observed across gestation

(R=-0.27) (Figure 3.15), however as no statistical difference was found between first

trimester and term samples this effect was not deemed to be significant.

3.5 Secreted Frizzled-Related Protein 4

As discussed in Section 1.5.2, secreted Frizzled-related protein 4 (sFRP4) is an

antagonist of canonical Wnt signalling, suppressing cellular proliferation by

competitively binding both Wnt ligands and Frizzled receptors, thereby enhancing the

degradation of cytosolic β-catenin (Schumann et al. 2000; He et al. 2005; Wawrzak et

al. 2007). Expression of sFRP4 has been examined in the rat placenta (Hewitt et al.

2006), but as there are distinct morphological differences between the rat and human

placenta I assessed the regulation of sFRP4 in the context of the human trophoblast.

Real Time quantitative PCR of mRNA transcripts in first trimester, term and

preeclamptic placentae reveal no significant differences in sFRP4 transcription between

any of these groups (P>0.05) (Figure 3.16), suggesting no alteration in sFRP4 regulation

of Wnt signalling between placentae across gestation or in preeclampsia. Indeed, only a

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0 5 10 15 20 25 30 35 40 45

ck1

8 /

L1

9

gestational age

Cytokeratin-18 mRNA across gestation


57

weak positive correlation was noted between sFRP4 expression and gestational age

(R=0.22) (Figure 3.17).

Figure 3.16 The expression of sFRP4 mRNA in the human placenta (P>0.05).

Figure 3.17 The change in sFRP4 mRNA expression throughout gestation

(R=0.22).

Immunohistochemical examination of the human placenta reveals extensive

sFRP4 expression in the syncytiotrophoblast, but not the proliferative cytotrophoblast in

both first trimester and term chorionic villus samples (Figure 3.18).

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

1st Trimester Term PE

sFR

P4

/ L

19

sFRP4 mRNA in the Human Placenta

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 5 10 15 20 25 30 35 40 45

sFR

P4

/ L

19


sFRP4 mRNA across human gestation


58

Figure 3.18 Immunohistochemistry for sFRP4 in the chorionic villi of first

trimester and term human placentae. sFRP4 is localised extensively to the

syncytio-trophoblast throughout gestation.

3.6 Endothelial Function of the Placenta

The balance of vessel constriction and dilatation in endothelial tissues is regulated

through the production of nitric oxide (NO). Functional NO production is modulated by

the enzymatic action of nitric oxide synthase (NOS). Several subtypes of NOS exist,

and it is endothelial NOS (eNOS) which specifically regulates production of functional

NO in vascular endothelia.

VEGF is an important upstream modulator of angiogenesis and NO production.

Several isoforms of VEGF exist (A-D), each of which contains multiple transcript

variants (VEGF-A has 7 variants). As this study is only interested in the gross

implications of VEGF rather than the complex arrangement of VEGF signalling,

primers specific only to the first three transcript variants of VEGF-A were used to

assess the general pattern of VEGF transcription in the human placenta.

3.6.1 Endothelial Nitric Oxide Synthase (eNOS)

The expression of eNOS mRNA was identified in the human placenta and its levels

monitored in both first trimester and term placentae in an attempt to characterise its

expression in trophoblast maturation. It was found that eNOS mRNA is significantly

higher in first trimester samples than at term (P<0.05; d=1.0) (Figure3.19),

corresponding with the greater need for endothelial regulation in the developing


59

placenta to control invasion and the hypoxic environment beneficial to early placental

development. In support of this, a strong relationship was noted between the eNOS

mRNA levels and the gestational age of the samples (R = 0.5) (Figure 3.20). Thus

eNOS is a modulator of trophoblastic endothelial function, especially during the first

trimester.

Figure 3.19 The expression of eNOS mRNA in the human placenta (*=P<0.05 vs.

Term).

Figure 3.20 The change in eNOS mRNA expression throughout gestation (R=-

0.5).

These data however conflict with the findings of Rossmanith et al. (1999), who

found that both transcription and translation of eNOS increases across gestation. This

discrepancy may be explained by differences in methodology between the studies. In

0

2

4

6

8

10

12


eN

OS

/L1

9

eNOS mRNA in the Human Placenta

* *

0

5

10

15

20

25

0 5 10 15 20 25 30 35 40 45

eN

OS

/ L1

9


eNOS mRNA across human gestation


60

their study, Rossmanith et al. (1999) used semi-quantitative RT-PCR to analyse eNOS

mRNA expression, whereas quantitative Real Time PCR was used in the present study.

Furthermore, protein expression was analysed by immunocytochemistry which, while

useful for examining sub-cellular localisation, is not necessarily a reliable quantitative

measure of protein expression. Alternately, the noted differences may be attributable to

the tissues analysed. In their study, Rossmanith et al. (1999) used a combination of in

vitro primary trophoblast cultures and perifusion methods, whereas the present data

represent snap frozen samples of chorionic villi.

Preeclampsia is a maternally manifested disorder of pregnancy potentially

attributable to disrupted trophoblast differentiation and endothelial function. Due to its

role in endothelial regulation, levels of eNOS mRNA were examined in the chorionic

villi of preeclamptic placentae (Figure 3.19). In conflict with previous studies

(Rossmanith et al. 1999; Orange et al. 2003), the transcription of eNOS was

significantly elevated in placentae complicated with preeclampsia compared with

normal term placentae (P<0.05). Thus the eNOS:NO pathway is severely disrupted in

preeclampsia.

Logic suggests that an increase in eNOS would lead to an elevation in the

cleavage of L-arginine to form NO, thus increasing NO levels and eliciting extensive

vasodilation, which in turn reduces blood pressure. However, a major manifestation of

preeclampsia is maternal hypertension, indicating a high degree of vasoconstriction.

This paradox may be explained by high levels of reactive oxygen species (ROS)

(Takagi et al. 2004), and/or reduced activity of free radical scavengers like SOD

(Poranen et al. 1996), leading to increased scavenging of NO before it can elicit

vasodilation. Alternatively there may be a deficiency in L-arginine in preeclampsia

(Kim et al. 2006), thereby negating the effects of increased eNOS and minimising the

amount of NO produced, leading to maternal hypertension. Therefore elevated eNOS in

preeclampsia may be a function of either disrupted blood pressure regulation, or

increased oxidative stress manifesting in maternal hypertension.

3.6.2 Vascular Endothelial Growth Factor (VEGF)

VEGF-A mRNA is expressed in the human placenta (Figure 3.21), however no

significant alterations in its expression were noted either across gestation or with

preeclampsia (P>0.05). Consequently, only a small correlation (R=0.1) was noted for

VEGF-A mRNA expression across gestation (Figure 3.22). When considered alongside


61

eNOS, whose expression is significantly higher during early gestation, this indicates

that placentally derived VEGF-A is not the sole regulator of the eNOS pathway in the

trophoblast. Thus other upstream mediators of eNOS activity are also involved in NO

production in the placenta.

Figure 3.21 The expression of VEGF-A mRNA in the human placenta.

Figure 3.22 The change in VEGF-A mRNA expression throughout gestation

(R=0.1).

3.7 Limitations

As this study used whole chorionic villi as samples of human placentae there is the

potential for contamination by cells associated with the trophoblast, such as

erythrocytes, Hofbauer and stromal cells. Additionally, as samples were selected

0

0.05

0.1

0.15

0.2

0.25

0.3


VEG

F-A

/ L

19

VEGF-A mRNA in the Human Placenta

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

0 5 10 15 20 25 30 35 40 45

VEG

F-A

/ L

19


VEGF-A mRNA across human gestation


62

macroscopically some may contain traces of extravillous trophoblast or maternal

decidua. This could result in samples, and thus data, being non-representative of the

villous trophoblast. To control for this, concurrent histological samples from each

placenta were examined, and only those samples with a villous trophoblast phenotype

were included in the study, thus eliminating this potential source of contamination.

Furthermore, the expression of trophoblast-specific products such as hCG and hPL in

each of the samples strongly indicates that the results presented in this chapter are

representative of the human villous trophoblast.

Chapter 4- BeWo Model Establishment and Validation

63

Chapter 4

BeWo Model Establishment and Validation

4.1 Background

In the human placenta the trophoblast forms a barrier between the maternal and fetal

circulation to provide an immune privileged conduit for the passage of gases, nutrients,

hormones and wastes. This barrier forms through the fusion of cytotrophoblasts to form

a multinucleated syncytiotrophoblast, which mediates blastocyst implantation, invasion

and eventually the fetal-maternal blood barrier. Co-ordinately, underlying

cytotrophoblasts replenish the barrier by fusing with the overlying syncytiotrophoblast

in the act of terminal differentiation.

Old nuclei aggregate at the apex of the syncytium, where they form syncytial

knots in a process reminiscent of late stage apoptosis, thus completing the trophoblast

life cycle. Syncytial knots are extruded into the maternal circulation and removed from

the system without any maternal immune response. Therefore in this system, apoptosis

completes the process of terminal differentiation suggesting the two processes are

intrinsically linked. Due to the central function of the trophoblast, an appreciation of

barrier formation and maintenance is important for understanding mechanisms of

maternal or fetal disorders of the trophoblast, particularly preeclampsia.

Parts of this Chapter crucial to the validation of BeWo cells as a model for

trophoblast apoptosis and differentiation were published in the peer reviewed journal

„Reproductive BioMedicine Online‟ (Appendix 1) (White et al. 2007).

Aim: To validate the BeWo cell line as a reliable model of human trophoblast apoptosis

and differentiation.

Hypothesis 1: Treatment of the BeWo cell line with Staurosporine will induce

apoptosis.

Hypothesis 2: Forskolin will successfully induce biochemical and morphological

differentiation of the BeWo cell line comparable to in vivo trophoblast differentiation.


64

4.2 Validation of BeWo Cells

The authenticity of BeWo cells as a human placental trophoblast cell line was analysed.

Immunohistochemical staining with the trophoblast marker human chorionic

gonadotrophin (hCG) revealed its presence in BeWo cells indicating the trophoblastic

origins of this cell line (Figure 4.1). Staining for hCG was not ubiquitous as it is only

expressed upon the onset of biochemical differentiation. Therefore BeWo cells produce

hCG in the same manner as the human trophoblast.

Figure 4.1 Paraffin embedded BeWo cells stained with A) haematoxylin and eosin

to show morphology, and B) hCG primary antibody (brown).

Figure 4.2 Cytokeratin-18 mRNA expression in BeWo cells.

The transcription of the trophoblast marker cytokeratin-18 was also examined in

the BeWo cell line. As seen in Figure 4.2, cytokeratin-18 is present in all BeWo samples

at each time point. Together with the hCG immunohistochemical analysis, this indicates

0

2

4

6

8

10

12

24h 48h 72h 96h

ck1

8 /

L1

9

Time

Cytokeratin-18 mRNA in the BeWo cell line


65

that the BeWo cell line provides a representative in vitro model for the examination of

trophoblast behaviour.

4.3 BeWo Apoptosis

As discussed in Section 1.5, apoptosis is central to the maintenance and renewal of a

mature and functional tissue. Disruption to the apoptosis pathway is a feature of many

tumours. In the human trophoblast, apoptosis is responsible for regulating the health of

the tissue and the associated mother and fetus. Disruption to this process is a probably

cause of some of the features of preeclampsia and intrauterine growth restriction. As

such, the understanding of the molecular mechanisms underlying trophoblast apoptosis

is vital to the development of measures to prevent or limit disruption to the apoptotic

machinery.

4.3.1 Staurosporine

Apoptosis in the BeWo cell line can be induced through treatment with the protein

kinase inhibitor Staurosporine (Das et al. 2004). As protein kinases regulate many

intracellular signalling pathways, their inhibition disrupts normal cellular function and

metabolism, promoting cellular apoptosis.

To validate Staurosporine as an inducer of apoptosis in the BeWo cell line, and

to optimise its concentration for use in later experiments, cells were treated with various

doses of Staurosporine (5, 1, 0.5, 0.1 μM). After incubation of cultures at 37ºC and 5%

CO2 for 3 and 6 hours, cell viability was determined by trypan exclusion using a

haemocytometer. It was determined that Staurosporine is indeed a potent initiator of

BeWo cell apoptosis, with maximal cell death occurring at a concentration of 1μM

Staurosporine (Figure 4.3). Morphologically, BeWo cells became rounded, with

considerable membrane blebbing indicative of apoptosis (Figure 4.4) (Kerr et al. 1972).

Accordingly, a dose of 1µM Staurosporine was used throughout the experiments as a

model of inducing apoptosis.


66

Figure 4.3 Staurosporine dose-response curve (* = P<0.05 vs. Control; † = P<0.05

vs. Control and 0.1µM).

Figure 4.4 Effect of 6 hours of A) DMSO and B) 1µM Staurosporine treatment on

the morphology of BeWo cells.

4.3.2 JC-1

JC-1 provides an accurate measure of the initiation stage of apoptosis by analysing the

rapid depolarisation of the mitochondrial membrane (Smiley et al. 1991; Zamzami et al.

2000). When the mitochondrial membrane is depolarised the JC-1 dye (5,5′,6,6′-

tetrachloro-1,1′,3,3′-tetraethyl-benzimidazolyl-carbocyanine iodide) is identifiable as a

green fluorescent monomer, whereas a polarised membrane causes JC-1 aggregation

and red fluorescence. Therefore the ratio of aggregated to monomeric JC-1 gives a

quantitative representation of the mitochondrial membrane polarity. Increased

mitochondrial membrane depolarisation as shown by a preponderance of green

fluorescence corresponds with the onset of apoptosis. Positive controls were conducted

by the addition of 50μM FCCP (carbonyl cyanide p-(trifluoromethoxy)

phenylhydrazone) to prevent aggregation of the JC-1 dye.

0

100

200

300

400

500

Control 0.1 0.5 1 5

viab

le c

ells

/ul

STS concentration (μM)

Staurosporine dose-response curve

3 hours

6 hours

*

† ††


67

After 1 hour incubation with 1μM Staurosporine, BeWo cells exhibited

significantly increased mitochondrial membrane depolarisation compared with media

alone (P<0.05), however this was not significant when compared with DMSO-treated

control samples (P<0.1) (Figure 4.5A). Following 3 hours incubation with

Staurosporine, apoptosis was successfully initiated in comparison with media and

DMSO controls (P<0.001) (Figure4.5B). These results validate this model of in vitro

trophoblast apoptosis by confirming the initiation of apoptosis in BeWo cells exposed to

Staurosporine.

Figure 4.5 Quantification of BeWo apoptosis by JC-1 analysis after 1 and 3 hours

of Staurosporine (STS) treatment (* = P<0.05 vs. vehicle and 1µM STS; † = P<0.05

vs. vehicle and FCCP).

4.3.3 DNA fragmentation

A major characteristic of late stage apoptosis is internucleosomal DNA fragmentation,

which upon electrophoresis and autoradiography, produce a typical DNA fragmentation

ladder (Tilly et al. 1993). Terminal transferase is used to bind radioactively-labelled

ddATP to the 3‟ ends of fragmented DNA and the total amount of incorporated

radiolabel can be quantified and used to estimate the extent of apoptotic DNA

fragmentation (Tilly et al. 1993; Drake et al. 2003). This technique was conducted on

Staurosporine treated BeWo cells to determine the presence and extent of apoptosis in

this model.

In contrast with the JC-1 data, 3 hours Staurosporine incubation did not result in

the appearance of DNA laddering. However following 6 hours of treatment, the amount

0

0.4

0.8

1.2

1.6

2

1h 3h

RFU

(5

90

nm

)/R

FU(5

35

nm

)

Time

JC-1 Analysis of Staurosporine treated BeWo cells

vehicle (DMSO)

1uM STS

FCCP* *

†


68

of labelled fragmented DNA was statistically significant (P<0.05) (Figure 4.6).

Additionally, the presence of the typical DNA laddering is indicative of apoptosis. Thus

Staurosporine treatment is able to induce apoptosis.

It is worth noting that mitochondrial membrane depolarisation, which is

indicative of the initiation of apoptosis, was significantly increased 3 hours after the

addition of Staurosporine; while DNA fragmentation, which is one of the final stages of

apoptosis, was not significantly induced until 6 hours after treatment. This indicates a 3

hour delay between the functional initiation of apoptosis and the degradation of nuclear

matter.

Figure 4.6 3‟-end labelling. A) DNA fragmentation following treatment of BeWo

cells with DMSO (i and iii), or 1μM Staurosporine (ii and iv) for 3 or 6 hours. B)

Graphical representation of incorporated radiolabel in each group (* = P<0.05 vs.

Control).

4.3.4 Caspases

The expression of mRNA for the initiator caspase-8 in apoptotic BeWo cells was

quantified using Real Time PCR. Interestingly, caspase-8 exhibited a time-dependant

reduction in mRNA expression following the induction of apoptosis (Figure 4.7).

Transcription was significantly reduced after both 3 and 6 hours of Staurosporine

treatment (P<0.05). Caspase-8 is a key initiator of the classical apoptosis cascade and

one would suspect that its expression would increase in response to apoptosis, however

as the apoptotic response to Staurosporine is so rapid there is little time to instigate

transcription. Thus caspase-8 protein may be stored in preparation for the apoptosis,

whereupon it is activated and initiates apoptotic cleavage. Under these circumstances,

existing caspase-8 mRNA may be degraded as part of the apoptotic cascade.


69

Figure 4.7 Caspase-8 mRNA expression in Staurosporine-treated BeWo cells (* =

P<0.05 v. Control).

Staurosporine treated BeWo cultures were examined by RT-PCR for the

presence of the key apoptotic protease caspase-3. Interestingly, caspase-3 mRNA was

present in control samples at all time points, however it was greatly diminished after 1

hour incubation with Staurosporine, and absent after 3 and 6 hours of Staurosporine

treatment (Figure 4.8). These data correspond with the reduced caspase-8 mRNA

discussed above.

Figure 4.8 RT-PCR for caspase-3 in Staurosporine-treated BeWo.

0

0.05

0.1

0.15

0.2

1h 3h 6h

casp

ase

-8 /

L1

9

Time

Caspase-8 mRNA in apoptotic BeWo

Control

Staurosporine

*

*


70

4.3.5 Conclusions

Treatment of the BeWo cell line with 1μM Staurosporine effectively initiates and

executes classical apoptosis within 6 hours of treatment as shown through both JC-1 and

3‟-end labelling. Therefore Staurosporine is an efficient stimulator of apoptosis in the

current model. The reduction in pro-apoptotic caspase mRNA in response to apoptosis

suggests the existence of intracellular stores of caspases in preparation for the

stimulation of the apoptosis cascade.

4.4 BeWo Differentiation

The biochemical and morphological differentiation of BeWo cultures was induced by

the addition of the adenyl cyclase promoter Forskolin (Wice et al. 1990; Lyden et al.

1993). The dosage of Forskolin to be used in subsequent experiments was determined

by analysing the morphology of the cells when exposed to various concentrations in

culture. Concentrations ranging from 1μM to 100μM were used, with commencement of

fusion in the 100μM Forskolin group within 48 hours. By 72 hours, syncytia were

observed in both 10μM and 100μM cultures, however sheets of syncytia were detaching

from the 100μM plate. By 96 hours, only few cells remained attached in the 100μM

Forskolin culture, extensive syncytialisation was evident at 10μM, and some

differentiation had occurred at 1μM. In line with these observations, a final

concentration of 20μM Forskolin was selected for future experiments, as the selected

time points for this study require initiation of morphological differentiation prior to 72

hours incubation, with extensive syncytia evident at 96 hours. It was postulated that

20μM Forskolin would provide a workable balance between these two requirements.

4.4.1 Confirmation of Biochemical Differentiation

Differentiation was confirmed by the quantification of β-hCG, which is actively

produced and secreted by the syncytiotrophoblast following differentiation. As assessed

by Real Time RT-PCR, the expression of β-hCG mRNA was elevated in Forskolin-

treated BeWo relative to time-matched controls within 24 hours, and this was observed

throughout the study (Figure 4.9A). As transcription of hCG precedes the release of the

functional hormone in BeWo cells by 12 hours (Kudo et al. 2004), quantification of β-

hCG mRNA is a more sensitive marker for biochemical differentiation than hormone

secretion itself .


71

Figure 4.9 A) Expression of β-hCG mRNA in BeWo cells following Forskolin

treatment. B) Western blot for β-hCG protein in Forskolin treated BeWo cells, and

C) Quantitation of corresponding Western Blot (B) (* = P<0.05 v. Control).

Nevertheless, production of hCG protein was investigated for Forskolin treated

BeWo cells as described in Section 2.5.6. hCG is significantly increased at each time

point throughout Forskolin treatment (P<0.01) (Figure 4.9B and C). This confirms the

0

1

2

3

4

5

24h 48h 72h 96h

hC

G /

L1

9

Time

Beta-hCG mRNA in Forskolin-treated BeWo

Control

Forskolin*

**

*

A

0

0.2

0.4

0.6

0.8

1

1.2

24h 48h 72h

hC

G /

GA

PD

H

β-hCG protein in Forskolin-treated BeWo

Control

Forskolin

**

*

C


72

mRNA data and further proves that Forskolin treatment of BeWo cells induces

biochemical differentiation in a manner consistent with the human trophoblast.

Contrary to previously published data (Wice et al. 1990), hPL mRNA was

significantly reduced in Forskolin treated BeWo cells across the time course of the

experiment (Figure 4.10). hPL is a major placental hormone produced by the

trophoblast, particularly during late gestation, allowing for increased maternal

metabolism to feed the developing fetus. Thus it would be predicted that hPL would be

increased with trophoblast differentiation, however this was not the case. These data

suggest that undifferentiated cytotrophoblasts are responsible for the transcription of

hPL in the human placenta.

Figure 4.10 Expression of hPL mRNA in BeWo cells following Forskolin

treatment (* = P<0.05 v. Control).

4.4.2 Confirmation of Morphological Differentiation

To confirm morphological differentiation E-cadherin mRNA was also quantified.

Briefly, E-cadherin is involved in cellular adhesion, so following cellular fusion E-

cadherin is decreased due to the increased volume:surface area ratio (Coutifaris et al.

1991). Accordingly, E-cadherin mRNA was decreased in Forskolin-treated BeWo from

72 hours treatment with 20μM Forskolin (Figure 4.11), indicating the fusion of BeWo

cells and the formation of syncytia indicative of morphological differentiation.

Immunofluorescence for E-cadherin was conducted, as outlined in Section 2.5.1,

on control and Forskolin-treated BeWo cultures in order to confirm the process of

morphological differentiation (i.e. syncytialisation). As E-cadherin is plasma

0

2

4

6

8

10

12

14

24h 48h 72h 96h

hP

L /L

19

Time

hPL mRNA in Forskolin-treated BeWo

Control

Forskolin*

***


73

membrane-associated protein vital for intercellular attachment in all epithelial tissues

including the trophoblast, the fusion of adjacent cells results in diminished E-cadherin at

the plasma membrane. Additionally, syncytialisation leads to multiple nuclei enclosed

within a single plasma membrane.

Figure 4.11 Expression of E-cadherin mRNA in BeWo cells following Forskolin

treatment (* = P<0.05 v. Control).

Figure 4.12 Confirmation of morphological differentiation. Immunofluorescence

for E-cadherin in 48 h (A−C) control and (D−F) Forskolin treated BeWo cells. A)

and D) show Hoechst labelled nuclei; B) and E) show E-cadherin staining; C) and

F) show overlays. Arrows indicated membrane-associated E-cadherin.

Arrowheads indicate diffuse E-cadherin. Original magnifications: 400x

0

1

2

3

4

24h 48h 72h 96h

E-ca

dh

eri

n /

L1

9

Time

E-cadherin mRNA in Forskolin-treated BeWo

Control

Forskolin

* *


74

After 48 hours incubation, an intense staining of E-cadherin was observed

associated with the plasma membrane of control BeWo cells, however in Forskolin

treated cells E-cadherin staining was much more diffuse and no longer confined to the

plasma membrane (Figure 4.12). This indicates that morphological differentiation is

initiated within 48 hours of Forskolin treatment, resulting in the formation of

multinuclear syncytia.

4.4.3 Conclusions

Forskolin is an effective inducer of both biochemical and morphological differentiation

in the BeWo cell line. Biochemical differentiation was successfully initiated within 24

hours of treatment as shown by the transcription of β-hCG, while morphological

differentiation (i.e. fusion) commenced by 48 hours as seen by immunofluorescence for

E-cadherin. Consequently, treatment of the BeWo cell line is an effective model for the

examination of trophoblast differentiation, with the switch between biochemical and

morphological differentiation occurring between 48 and 72 hours after the addition of

Forskolin.

Figure 4.13 Expression of caspase-8 mRNA in BeWo cells following Forskolin

treatment (*=P<0.05 v. Control).

4.5 Real Time Reverse Transcriptase Polymerase Chain Reaction

Interestingly, the initiator caspase-8 showed reduced mRNA expression after 48 hours

of Forskolin treatment (P<0.05) (Figure 4.13). This suggests that early apoptotic

proteases are not involved in the process of differentiation. However, caspase-8 mRNA

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

24h 48h 72h 96h

Cas

pas

e-8

/ L

19

Time

Caspase-8 mRNA in Forskolin-treated BeWo

Control

Forskolin* *


75

was also reduced upon the induction of trophoblast apoptosis and it was postulated that

this may be attributable to dormant intracellular caspase stores, which upon apoptotic

stimulation are depleted. If this is the case, diminished caspase-8 mRNA after 48 and 72

hours of Forskolin treatment may indicate the induction of initiator caspase cascades

during morphological differentiation of the trophoblast. This correlates with other

studies that have suggested a key role for the initiator caspase-8 in the induction of

trophoblast syncytialisation (Huppertz et al. 1999; Black et al. 2004).

The extent of proliferation within a culture can be determined through the

expression of the mitotic marker Ki67. Real Time RT-PCR revealed no change in Ki67

mRNA between control and differentiating cultures, suggesting that the proliferative

state of the differentiating BeWo cultures was unchanged (Figure 4.14). This is

somewhat surprising as in order for differentiation to occur cells must exit the cell

cycle, decreasing the expression of Ki67. This data therefore indicates that although the

cells are undergoing differentiation, the mitotic state is independent of BeWo

differentiation. Alternatively, cells may already have achieved confluence, resulting in

contact-inhibition of the cell cycle. This could explain the unaltered expression of Ki67

mRNA between control and differentiating cells.

Figure 4.14 Expression of Ki67 mRNA in BeWo cells following Forskolin


0

0.4

0.8

1.2

1.6

2

24h 48h 72h 96h

Ki6

7 /

L1

9

Time

Ki67 mRNA in Forskolin-treated BeWo

Control

Forskolin

Chapter 5- Caspase-14 in the Trophoblast

76

Chapter 5

Caspase-14 in the Trophoblast

5.1 Background

Apoptosis is regulated by a family of proteases known as caspases at both the initiation

and execution stages. Proteases are classified into this family by the possession of the

QACRG sequence. Caspase-14 however is an atypical caspase in that no pro-apoptotic

function has been identified. It is however activated in terminal keratinocyte

differentiation (Eckhart et al. 2000; Lippens et al. 2000; Rendl et al. 2002; Fischer et al.

2004) through the cleavage of profilaggrin and subsequent intermediate filament

stabilisation and corneocyte hydration (Denecker et al. 2007). Additionally, caspase-14

is not associated with apoptosis in the human placenta (Kam et al. 2005), however its

complete function within the trophoblast is yet to be elucidated. As such I investigated

the expression profile of caspase-14 in classical apoptosis and both biochemical and

morphological differentiation of the human cytotrophoblast using the BeWo cell line as

a model for trophoblast function. I hypothesise that caspase-14 possesses a conserved

role in barrier formation through involvement in terminal differentiation and fusion of

the cytotrophoblast.

The majority of this Chapter pertaining to caspase-14 expression and potential

interactions in the BeWo cell line were published in the peer-reviewed journal

„Reproductive BioMedicine Online‟ (Appendix 1) (White et al. 2007).

Aim: To assess caspase-14 expression during trophoblast apoptosis and differentiation.

Hypothesis: Caspase-14 expression will be unaltered with BeWo apoptosis, but

differentially regulated during BeWo differentiation

5.2 Caspase-14 in Apoptosis

BeWo cells express caspase-14 mRNA following 1, 3 and 6 hours of 1μM

Staurosporine treatment (apoptosis). Levels of caspase-14 mRNA remained stable

throughout apoptosis of the BeWo cell line as examined by Real Time PCR (P>0.05)


77

(Figure 5.1A), indicating that it is either not transcriptionally altered during trophoblast

apoptosis or that there is steady turnover of its transcript.

Figure 5.1 A) Expression of caspase-14 mRNA in BeWo cells following

Staurosporine treatment. B) Western blot of procaspase-14 protein in BeWo cells

following Staurosporine treatment. C) Quantitation of the corresponding Western

blot (B). Note there is no significant difference between any of the groups.

Similarly, Western blot analysis of Staurosporine-treated BeWo cells revealed

no statistical alteration in the expression of procaspase-14 at any time (P>0.05) (Figure

0

150

300

450

600

750

900

1h 3h 6h

casp

ase

-14

/ L

19

Time

Caspase-14 mRNA in Staurosporine treated BeWo

Control

Staurosporine

A

B

C


78

5.1B and C). These data reinforce the hypothesis that caspase-14 acts independently of

classical apoptosis pathways (Eckhart et al. 2000; Lippens et al. 2000; Chien et al. 2002;

Rendl et al. 2002; Lippens et al. 2003; Fischer et al. 2004).

5.3 Caspase-14 in BeWo Cell Differentiation

Altered expression patterns for caspase-14 were observed upon differentiation of the

BeWo cell line. Caspase-14 mRNA levels were significantly reduced after 24, 72 and

96 hours of Forskolin treatment compared with time matched controls, as analysed by

Real Time PCR (P<0.05) (Figure 5.2A). This confirms differential transcription of

caspase-14 upon the differentiation of BeWo cells, and suggests that caspase-14 is

involved in suppressing trophoblast differentiation. Interestingly, no significant

difference was noted in caspase-14 mRNA after 48 hours of Forskolin treatment. This

time point coincides with the switch from biochemical to morphological differentiation

and suggests that caspase-14 is not transcriptionally regulated at the onset of

syncytialisation.

Intriguingly caspase-14 mRNA gradually increased in its expression in control

cultures across the course of the experiment (Figure 5.2A). As the cultures were initially

seeded at quite a high density (2.5 x 105 cells/ml), this may reflect changes in caspase-

14 regulation in response to cell senescence brought about by contact inhibition of the

cell cycle. Regulation of caspase-14 synthesis by the proliferative potential of the

trophoblast is an interesting prospect which requires further investigation.

Interestingly, when looking at Forskolin treated BeWo cells in isolation,

caspase-14 mRNA was significantly increased after 48 hours compared with all other

time points (Figure 5.2A). This immediately precedes E-cadherin realignment

associated with syncytialisation (Section 4.4), suggesting increased transcription of

caspase-14 at the initiation of morphological differentiation of the BeWo cell line.

The full length procaspase-14 protein was identified in the BeWo cell line using

Western blot analysis. Procaspase-14 is significantly reduced after 48and 72 hours of

Forskolin treatment relative to age matched controls (P<0.05) (Figure 5.2B and C).

Thus expression of caspase-14 is reduced in late biochemical differentiation and during

morphological differentiation of the BeWo cell line.


79

Figure 5.2 A) Expression of caspase-14 mRNA in BeWo cells following Forskolin

treatment. B) Western blot of procaspase-14 protein and internal control GAPDH

in BeWo cells following Forskolin treatment; and C) quantitation of the

corresponding Western blot (B) (* = P<0.05 v. Control).

Caspase activation and subsequent function, requires its cleavage into large

(P18) and small (P10) subunits. Indeed, in positive control samples of the human

epidermis the cleaved subunits, particularly the P18 fragment, were evident (Figure

0

50

100

150

200

250

300

350

24h 48h 72h 96h

casp

ase

-14

/ L

19

Time

Caspase-14 mRNA in Forskolin treated BeWo cells

Control

Forskolin

** *

A

0

0.2

0.4

0.6

0.8

24h 48h 72h

casp

ase

-14

/ G

AP

DH

Time

Caspase-14 protein in Forskolin treated BeWo

Control

Forskolin*

*

C


80

5.2B). However no cleaved caspase-14 was detected in BeWo cells, either in controls or

after Forskolin treatment. Thus, caspase-14 is not cleaved in BeWo cells, and by proxy

the human trophoblast. This then indicates that caspase-14 is not activated in the human

trophoblast. The presence, and subsequent function, of caspase-14 in the trophoblast

therefore requires further investigation.

5.3.1 Profilaggrin

As discussed in Section 1.7, profilaggrin is cleaved and dephosphorylated by caspase-14

into its functional filaggrin repeats, which organise the keratin cytoskeleton of the

stratum corneum. Furthermore, filaggrin degrades into its component amino acids,

which provide natural moisturising factor (NMF) and possible UV-B protection to the

stratum corneum.

Figure 5.3 Expression of Filaggrin mRNA in BeWo cells following Forskolin

treatment. The HaCaT cell line (epidermal) was used as a positive control, and is

plotted on the secondary vertical axis.

To assess whether filaggrin is present and important in human trophoblast

biology, profilaggrin transcription was assessed in Forskolin-treated BeWo cells by

Real Time PCR. Profilaggrin was identified in Forskolin-treated cultures, but not in

controls (Figure 5.3). However, not all differentiated samples were positive for

profilaggrin mRNA, suggesting differential expression in the trophoblast. Additionally,

detectable mRNA levels were dramatically lower in BeWo samples than positive

keratinocyte controls by a magnitude of 100,000 fold (Figure 5.3), indicating that

filaggrin is not a major factor in the human trophoblast.

0

5

10

15

20

25

30

35

40

45

0

0.0001

0.0002

0.0003

0.0004

0.0005

0.0006

24h 48h 72h 96h

fila

ggri

n /

L1

9

fila

ggri

n /

L1

9

Time

Filaggrin mRNA in Forskolin treated BeWo

Control

Forskolin

HaCaT


81

Western analysis was attempted in both human placental and BeWo samples;

however I was unable to get the Profilaggrin antibody to consistently work, even in

protein extracted from human keratinocytes. Therefore analysis of functional

profilaggrin expression is not presented in the current study. Indeed, as profilaggrin

mRNA was not expressed, it is highly likely that the protein would also be undetectable.

5.3.2 Krüppel-like Factor 4 (KLF4)

Keratinocyte differentiation and barrier formation is dependent on the expression of the

Krüppel-Like Factor 4 (KLF4) transcription factor (Segre et al. 1999; Jaubert et al.

2003). It has been noted that caspase-14 contains a potential binding site for KLF4

(Eckhart et al. 2000), and given the role of both in keratinocyte differentiation and the

expression of KLF4 in the human placenta (Blanchon et al. 2001; Blanchon et al. 2006),

there may be synergy between the two in trophoblast barrier formation. Accordingly,

the expression of KLF4 mRNA was examined upon differentiation of the BeWo cell

line (Figure 5.4A). Interestingly, KLF4 mRNA expression was significantly elevated

upon Forskolin treatment at all times (P<0.05); an expression pattern opposite that for

caspase-14.

Validation by Western blot analysis reveals that KLF4 protein expression is

greatly reduced after 48 hours of Forskolin treatment (P<0.05) (Figure 5.4B and C). No

other changes in KLF4 expression were noted during BeWo cell differentiation. The

decrease in KLF4 protein is in conflict to data obtained for its mRNA, which was

increased during differentiation at all time points (Figure 5.4A). It either implies that

while KLF4 transcription is enhanced it is not being translated into protein at nearly the

same rate, or that KLF4 protein turnover is elevated. Indeed, after 48 hours KLF4

mRNA was greatly increased, whereas the protein was significantly decreased. The

mechanism responsible for this remains elusive; however the mRNA and protein data in

combination suggest a role for KLF4 in trophoblast differentiation.


82

Figure 5.4 A) Expression of KLF4 mRNA in BeWo cells following Forskolin

treatment (*=P<0.05 v. Control). B) Western blot analysis of KLF4 and internal

control GAPDH in BeWo cells following Forskolin treatment, and C)

quantification of the corresponding Western blot (B) (*=P<0.05 v. Control).

0

10

20

30

40

50

60

70

80

24h 48h 72h 96h

KLF

4 /

L1

9

Time

KLF4 mRNA in Forskolin treated BeWo

Control

Forskolin

*

*

* *

A

0

0.2

0.4

0.6

0.8

1

1.2

1.4

24h 48h 72h

KLF

4 /

GA

PD

H

Time

KLF4 protein in Forskolin treated BeWo

Control

Forskolin

*

C


83

Figure 5.5 A) Expression of cytokeratin-18 mRNA in BeWo cells following

Forskolin treatment (*=P<0.05 v. Control). B) Western blot analysis of

cytokeratin-18 and internal control GAPDH in BeWo cells following Forskolin

treatment, and C) quantification of the corresponding Western blot (B) (*=P<0.05

v. Control).

0

3

6

9

12

15

24h 48h 72h 96h

ck1

8 /

L1

9

Time

Cytokeratin-18 mRNA in Forskolin treated BeWo

Control

Forskolin

*

A

0

0.4

0.8

1.2

1.6

24h 48h 72h

ck1

8 /

GA

PD

H

Time

Cytokeratin-18 protein in Forskolin treated BeWo

Control

Forskolin

*

C


84

5.3.3 Cytokeratin-18

In the epidermis, caspase-14 is involved in keratin modulation during terminal

differentiation through its processing of profilaggrin into functional repeat units, which

bind keratin into dense bundles. As with keratinocytes, the trophoblast undergoes

significant morphological alterations during terminal differentiation; fusing with the

overlying syncytium and drastically changing its cellular architecture as it is

incorporated into the multinucleated syncytiotrophoblast. Presumably then, the

intermediate filament networks within cytotrophoblasts undergo rearrangement in

preparation for morphological differentiation. Consequently, the expression of the

intermediate filament protein and trophoblast marker cytokeratin-18 was examined in

response to differentiation of the BeWo cell line.

Cytokeratin-18 was consistently expressed at all times and treatments (Figure

5.5A), however after 48 hours of Forskolin treatment its transcription was significantly

elevated relative to time-matched controls (P<0.05). Intriguingly, this occurs

immediately prior to the onset of morphological differentiation, indicating that

cytokeratin-18 synthesis in BeWo cells is increased by the need for rearrangement of

the cytoskeleton at the commencement of cell-cell fusion.

At the protein level, cytokeratin-18 expression is stable throughout biochemical

differentiation; however by the onset of morphological differentiation at 72 hours

cytokeratin-18 expression is significantly increased (Figure 5.5B and C). This further

suggests that cytokeratin-18 is involved in the cell fusion process. Indeed the elevated

mRNA levels after 48 hours of Forskolin treatment (Figure 5.5A) immediately precede

the increase in protein levels (Figure 5.5C). Therefore the increased transcription of

cytokeratin-18 follows through into increased translation with morphological

differentiation. Consequently, cytokeratin-18 modulates morphological differentiation.

5.3.4 Secreted Frizzled Related Protein 4 (sFRP4)

In the human placenta samples, sFRP4 mRNA was found to be variably present

throughout gestation (Section 3.5). When examined in the BeWo cell line however, very

little if any sFRP4 mRNA is present in either Forskolin treated or time-matched controls

(Figure 5.6). Indeed, only one of six Forskolin-treated BeWo samples at any time

exhibited sFRP4 mRNA. While it was more frequently expressed in control cultures,

only a maximum of two out of six samples at any time point were positive for sFRP4


85

mRNA. This strongly indicates that the BeWo cell line is not a producer of the Wnt

pathway antagonist sFRP4.

Figure 5.6 Expression of sFRP4 mRNA in BeWo cells following Forskolin

treatment.

5.4 Discussion

Caspase-14 has a limited expression profile in adults; however it is involved in human

keratinocyte differentiation (Eckhart et al. 2000; Lippens et al. 2000; Chien et al. 2002;

Rendl et al. 2002; Fischer et al. 2004). In particular, it is up-regulated in the

proliferative stratum granulosum, with expression increasing in apical layers coinciding

with the onset of terminal differentiation and cornification. As its expression pattern

appears to be limited to barrier forming tissues we hypothesised that caspase-14 is

involved in barrier formation of the trophoblast in addition to the epidermis.

To date, there is no evidence to link caspase-14 with classical apoptosis

pathways, however it is suggested that terminal differentiation leading to intercellular

fusion or barrier formation shares common signalling pathways with apoptosis (Ishizaki

et al. 1998; Gandarillas 2000; Fernando et al. 2002; Arama et al. 2003; Lippens et al.

2003; Mogi et al. 2003; Chaturvedi et al. 2006). Due to the homology of caspase-14

with pro-apoptotic proteases, it may represent a conserved link between apoptosis and

differentiation, with particular reference to barrier formation (Lippens et al. 2003). To

test this hypothesis, the cytotrophoblast-like BeWo cell line was used as an in vitro

model of trophoblast function, and induced to undergo both apoptosis and

differentiation. Both caspase-14 mRNA and protein levels remained stable throughout

0

0.05

0.1

0.15

0.2

0.25

24h 48h 72h 96h

sFR

P4

/ L

19

Time

sFRP4 mRNA in Forskolin treated BeWo

Control

Forskolin


86

classical apoptosis (Figure 5.1A-C), indicating a role for caspase-14 independent of

classical apoptosis. Indeed, our lab has previously demonstrated caspase-14 to be

expressed in the human placenta in a pattern unrelated to apoptosis (Kam et al. 2005).

Therefore these results signify that caspase-14 is not involved in human trophoblast

apoptosis.

When BeWo cells were induced to undergo differentiation by the addition of

Forskolin, both caspase-14 mRNA and protein were consistently decreased (Figure

5.2A-C). This indicates a role for caspase-14 in both biochemical and morphological

differentiation of BeWo cells. While the extent and mechanisms of this remain to be

determined, it is likely that caspase-14 suppresses trophoblast differentiation.

Terminal differentiation of keratinocytes results in the formation of a barrier

between the internal and external environments that is essential for survival. The

involvement of caspase-14 in the formation of this barrier and that between mother and

fetus, coupled with its limited expression pattern indicate a conserved function for

caspase-14 in barrier formation. Caspase-14 is expressed in several cell types affiliated

with functional barriers, further supporting a role in barrier formation (Eckhart et al.

2000; Lippens et al. 2003).

Interestingly, control cultures showed an increase in caspase-14 mRNA

throughout the experiment (Figure 5.2A). Concomitantly, caspase-14 protein was also

increased after each successive time point (Figure 5.2B and C). This may signify a

function for caspase-14 in cell senescence as the cultures were confluent and therefore

exhibiting cell contact inhibition at these times. Indeed caspase-14 was significantly

reduced during BeWo cell differentiation, a process requiring cell-cycle withdrawal.

This then may represent a hitherto unknown role for caspase-14 in suppression of the

cell cycle.

While the expression of caspase-14 mRNA and protein during early trophoblast

differentiation is confirmed, the functional properties of this protease remain

undetermined. While one may assume that there is a reason caspase-14 is down

regulated with differentiation, the role it plays therein remains elusive. Indeed, as no

cleaved caspase-14 subunits were observed in any samples, there is little evidence that

caspase-14 is activated in the trophoblast.

Profilaggrin has recently been identified as a substrate of caspase-14 in the

epidermis; however Real Time PCR analysis of both BeWo and human placentae fails

to identify this protein in the trophoblast (Figure 5.3). This indicates that caspase-14 has

tissue-specific targets, the nature of which remains to be determined. A functional role


87

for caspase-14 in barrier formation cannot be confirmed without conducting substrate

cleavage and activity assays; however such assays are at present commercially

unavailable, adding to the difficulty of exploring its biological significance.

The putative binding partner of caspase-14, KLF4 (Park et al. 2006) was

observed in the BeWo cell line, however despite a trophoblast expression pattern similar

to that of caspase-14, there is no evidence for interaction between the two. Interestingly,

KLF4 mRNA was significantly increased throughout the experiment in differentiating

BeWo cells (Figure 5.4A). As KLF4 activity as a transcription factor directly relates to

Cyclin D2 suppression, elevated KLF4 during differentiation indicates an involvement

in cell cycle withdrawal in preparation for and maintenance of differentiation. However

protein levels were reduced at the 48 hour time point (Figure 5.4B and C) casting doubt

over the function of KLF4 in the human trophoblast.

Cytokeratin-18 mRNA was elevated after 48 hours of Forskolin treatment

(Figure 5.5A), a feature that was seen translated into the protein after 72 hours (Figure

5.5B and C). As BeWo cells commence fusion between 48 and 72 hours after the

addition of Forskolin, the observed rise in cytokeratin 18 mRNA immediately precedes

the onset of fusion, while elevated protein was observed in fusing cells. This supports a

significant rearrangement of the intermediate filament cytoskeleton in preparation for,

and during, intercellular fusion in the trophoblast. Thus, cytokeratin-18 may modulate

morphological differentiation events. As the keratin intermediate filament architecture is

indirectly affected by caspase-14 activity in keratinocytes, it is attractive to postulate

that cytokeratin-18 and its arrangement is a downstream target of caspase-14 in the

morphological differentiation of the trophoblast, thereby suggesting a conserved role for

caspase-14 in barrier formation.

It is interesting to note that although the canonical Wnt pathway antagonist

sFRP4 is present in the human trophoblast throughout gestation (Section 3.5), it is

absent from the BeWo cell line both under control and differentiation conditions (Figure

5.6). As sFRP4 is a secreted protein, this suggests that the trophoblast itself does not

synthesise sFRP4, but rather internalises it. Therefore, while sFRP4 may not be

produced by the trophoblast, it does react to its presence. Further investigation into

sFRP4 regulation in the trophoblast must be undertaken to determine its precise function

in the placenta.

To further assess the properties of caspase-14 in trophoblast differentiation,

functional studies into its significance in the trophoblast must be undertaken.

Additionally, the properties of the rising caspase-14 expression levels in senescent


88

cultures must be evaluated to determine whether it is involved in cell contact inhibition

of proliferation.

In conclusion caspase-14 is down regulated during trophoblast differentiation as

represented by the BeWo cell line, suggesting a conserved role in human barrier

formation. The exact nature of its role remains uncharacterised, although examination of

caspase-14 cleavage, activity and interactions should elucidate its full role. Further

investigation into the formation of the maternal-fetal barrier has important implications

for the understanding of diseases in which differentiation may be disrupted, such as

Preeclampsia and Intrauterine Growth Restriction (IUGR).

Chapter 6- Endothelial Function of the Human Trophoblast

89

Chapter 6

Endothelial Function of the Human Trophoblast

6.1 Background

The exchange of materials between the mother and fetus is mediated by the highly

differentiated, multinucleated syncytiotrophoblast that covers the placental chorionic

villi. Due to its location and function it is likely that the trophoblast mediates

endothelial functions such as the control of vasotone, platelet aggregation and

subsequent thrombosis, and immune cell adhesion. Despite this, the endothelial

potential of the trophoblast has received very little attention.

Caspase-14 is associated with epidermal maturation and barrier formation

(Lippens et al. 2003; Demerjian et al. 2007). An important function of the

syncytiotrophoblast is the formation of a barrier against the maternal circulation.

Therefore caspase-14 may be an important modulator of both syncytiotrophoblast

formation and preparation for endothelial functions.

Impaired placental blood flow and excessive fibrin deposition is a significant

barrier problem. As described in Section 1.1.3, endothelial cells respond to a wide

variety of stimuli such as VEGF, bradykinin or shear stress to initiate Nitric Oxide (NO)

release into the cell and underlying smooth muscle to modulate luminal blood pressure,

platelet aggregation and cell adhesion. NO is produced by endothelial nitric oxide

synthase (eNOS) degrading L-arginine in response to intra- and extra-cellular

stimulation. Impaired placental blood flow and fibrin deposition may lead to impaired

barrier formation and/or function resulting in the onset of preeclampsia.

Complete differentiation of the BeWo cell line occurs over a period of 96 hours

when treated with 20µM Forskolin, with biochemical differentiation occurring up to 48

hours, and morphological differentiation (i.e. fusion) occurring from 72 hours. The

endothelial functions of the human trophoblast were investigated throughout

differentiation by treating the BeWo cell line with an array of endothelial pathway

stimulants (VEGF, Bradykinin, Ca2+

), both with and without the presence of the eNOS

inhibitor L-NAME. Levels of β-hCG, eNOS and caspase-14 mRNA were examined

using Real Time PCR to investigate the potential for endothelial-like properties of the

villous trophoblast, and the influence of the NO pathway on trophoblast differentiation.


90

Aim 1: To investigate the endothelial-like functions of the human trophoblast.

Aim 2: Assess the influence of BeWo differentiation on the NO pathway.

Hypothesis 1: The human trophoblast will respond to endothelial stimulants in

an endothelial-like manner.

Hypothesis 2: Differentiation will modulate the endothelial properties of the

BeWo cell line, as only the mature syncytiotrophoblast is in direct contact with the

maternal blood.

6.2 Endothelial Biomarkers in BeWo Differentiation

To assess any potential endothelial role of the human trophoblast the BeWo cell line

was treated with Forskolin to induce differentiation, and the mRNA expression levels of

eNOS and VEGF-A were examined using Real Time PCR standardised against L19.

Due to the preponderance of VEGF-A transcript variants, the primers used to assess the

transcriptional expression of VEGF-A in this study only recognise the transcript

variants 1, 2, and 3. Therefore other variants of the VEGF-A gene (4-7) are not detected

within these results.

Expression of the eNOS transcript was detected in all cultures, with considerably

elevated levels present in those undergoing biochemical differentiation (P<0.05) (Figure

6.1). This implies that the NO pathway is an important regulator of trophoblast

differentiation, particularly in the preparation for morphological differentiation.

Importantly, eNOS mRNA is elevated in fully differentiated BeWo cultures (P<0.05),

indicating an ongoing endothelial function for the mature syncytiotrophoblast.

As with eNOS, VEGF-A mRNA was expressed throughout the experiment.

However its expression pattern was markedly dissimilar to that of eNOS. After 24hours

of Forskolin treatment, BeWo cells exhibit significantly reduced levels of VEGF-A

(P<0.05) (Figure 6.2); however by 48 hours VEGF-A transcription is elevated relative

to controls (P<0.05). This indicates a biphasic switch in VEGF-A expression between

the instigation of differentiation and the preparation for fusion. At 72 hours transcript

levels had stabilised between controls and differentiating BeWo cells, however by 96

hours VEGF-A mRNA expression was again significantly reduced in Forskolin treated

BeWo cell cultures (P<0.05). This suggests that auto/paracrine VEGF-A-mediated

signalling is down-regulated in the syncytiotrophoblast.


91

Figure 6.1 Expression of eNOS mRNA in BeWo cells following Forskolin


Figure 6.2 Expression of VEGF-A mRNA in BeWo cells following Forskolin


Interestingly, while the relative levels of eNOS mRNA gradually increased in

control cultures across the experiment, VEGF-A mRNA in the control cultures

fluctuated throughout the course of the experiment. The mechanism behind this

fluctuation in seemingly homeostatic cultures remains elusive.

In summary, eNOS production is associated with the initiation of trophoblast

differentiation and the subsequent maintenance of the mature phenotype. Interestingly,

VEGF-A production does not appear to be required during early biochemical

differentiation, but is amplified in preparation for morphological differentiation.

Senescent syncytia however show a diminished need for VEGF-A production. This

0

0.08

0.16

0.24

0.32

0.4

24h 48h 72h 96h

eN

OS

/ L1

9

Time

eNOS mRNA in Forskolin treated BeWo

Control

Forskolin

**

*

0

3

6

9

12

15

18

24h 48h 72h 96h

VEG

F-A

/ L

19

Time

VEGF-A mRNA in Forskolin treated BeWo

Control

Forskolin

*

*

*


92

could be attributed to a reduced requirement by syncytia for VEGF-A mediated

endothelial stimulation.

6.3 Treatment of BeWo Cell Line with Endothelial Pathway

Compounds

The BeWo cell line was treated with 20µM of Forskolin every 24 hours for up to 96

hours in order to induce biochemical and morphological differentiation. At the

commencement of Forskolin treatment and after every subsequent 24 hours, cells were

treated with Ca2+

, VEGF, or Bradykinin either with or without the eNOS inhibitor L-

NAME. Following incubation with the various endothelial pathway compounds for 20

minutes RNA was extracted, reverse transcribed and the resultant cDNA purified as per

Section 2.5.

6.3.1 Effect of NO pathway agonists on BeWo differentiation

In support of previous data (Section 4.4.1), β-hCG mRNA was elevated in all BeWo

cultures after Forskolin treatment. Similarly, treatment with exogenous calcium ion in

conjunction with Forskolin results in increased β-hCG transcription compared with time

zero samples (P<0.05) (Figure 6.3). β-hCG synthesis appears to be increased in

response to calcium after 48 hours of Forskolin treatment, however statistical

significance was not achieved (P>0.05). Therefore exogenous calcium does not

modulate hCG production during BeWo differentiation.

Treatment of biochemically differentiating BeWo cultures with VEGF does not

affect the transcription of β-hCG at any time point (P>0.05) (Figure 6.4). Additionally,

the inhibition of the NO pathway by L-NAME did not elicit any change in β-hCG

production compared with both controls and VEGF treated cultures. Therefore the

VEGF-mediated NO pathway does not affect hormone production in the BeWo cell

line.


93

Figure 6.3 Effect of Calcium treatment on β-hCG mRNA expression in Forskolin

mediated BeWo differentiation. L-NAME acts as an inhibitor of the NO pathway.

Figure 6.4 Effect of VEGF treatment on β-hCG mRNA expression in Forskolin


Bradykinin appears to increase the transcription of β-hCG particularly around

morphological differentiation, however these data do not achieve significance against

controls and L-NAME treated samples (P>0.05) (Figure 6.5). It would be interesting to

note whether an increased number of samples would result in statistical significance

being reached as there is a trend toward elevated hCG production in response to the

bradykinin-mediated NO pathway.

0

4

8

12

16

20

0h 24h 48h 72h

hC

G /

L1

9

Time

β-hCG mRNA in Ca2+ treated differentiating BeWo

Control

Ca

Ca+L-NAME

0

2

4

6

8

10

12

14

0h 24h 48h 72h

hC

G /

L1

9

Time

β-hCG mRNA in VEGF-A treated differentiating BeWo

Control

VEGF

VEGF+L-NAME


94

Figure 6.5 Effect of Bradykinin treatment on β-hCG mRNA expression in

Forskolin mediated BeWo differentiation. L-NAME acts as an inhibitor of the NO

pathway.

When taken as a whole however, data indicate that stimulation of the BeWo cell

line with NO pathway agonists during differentiation has no impact on biochemical

differentiation. Thus exogenous stimulation of the endothelial nitric oxide pathway

neither enhances nor suppresses BeWo differentiation.

6.3.2 Endothelial Nitric Oxide Synthase (eNOS)

Assessment of alteration to the endothelial pathway was examined by looking at eNOS,

which is immediately responsible for the cleavage of L-arginine and subsequent

production of NO. eNOS mRNA expression was not altered by the addition of

endothelial pathway modulators between the initiation of differentiation and 24 hours

(P>0.05) (Figure 6.6). However in agreement with previous data (Figure 6.1) by 48

hours several groups including controls exhibited elevated eNOS in comparison with

both 0 and 24 hours of Forskolin treatment (P<0.05).

Calcium treatment did not affect eNOS transcription during early biochemical

differentiation (Figure 6.7). Similarly, no statistically significant change was observed

in eNOS mRNA following calcium treatment of BeWo cultures differentiated for either

48 or 72 hours (P>0.05). Therefore exogenous calcium treatment does not affect eNOS

transcription in the BeWo cell line.

0

5

10

15

20

25

0h 24h 48h 72h

hC

G /

L1

9

Time

β-hCG mRNA in Bradykinin treated differentiating BeWo

Control

Bk

Bk+L-NAME


95

Figure 6.6 Expression of eNOS mRNA in Forskolin treated BeWo cells following

treatment with nitric oxide pathway modulators (* = P<0.05 vs. 0h and 24h).

Figure 6.7 Effect of Calcium treatment on eNOS mRNA expression in Forskolin


Addition of VEGF and/or L-NAME to untreated BeWo does not affect the

transcription of eNOS (Figure 6.8). However eNOS synthesis is enhanced following

VEGF treatment after 24 hours of differentiation (P<0.05), and is suppressed by co-

incubation with the eNOS inhibitor L-NAME (P<0.05). Interestingly, the addition of

VEGF to 72 hour differentiating BeWo cultures significantly reduced the expression of

eNOS mRNA (P<0.05), which was further reduced when incubated with L-NAME

(P<0.05). Cumulatively these data indicate that VEGF influences the

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Control Ca VEGF Bk L-NAME Ca + L-NAME

VEGF + L-NAME

Bk + L-NAME

eN

OS

/L1

9

Treatment

Influence of endothelial function on eNOS mRNA in Forskolin treated BeWo

0h

24h

48h

72h

*

*

**

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0h 24h 48h 72h

eN

OS

/ L1

9

Time

eNOS mRNA in Ca2+ treated differentiating BeWo

Control

Ca

Ca + L-NAME


96

production of eNOS in biochemically differentiating BeWo. This suggests a role for the

endothelial VEGF-eNOS axis in the differentiating human trophoblast.

Figure 6.8 Effect of VEGF treatment on eNOS mRNA expression in Forskolin


NAME acts as an inhibitor of the NO pathway.

Figure 6.9 Effect of Bradykinin treatment on eNOS mRNA expression in Forskolin


NAME acts as an inhibitor of the NO pathway.

The addition of bradykinin to BeWo cultures incubated with Forskolin for 0 and

24 hours does not affect the transcription of eNOS (Figure 6.9). After 48 hours of

Forskolin treatment bradykinin significantly increased levels of eNOS mRNA above

those seen in control and L-NAME treated samples (P<0.05). 72 hours after the addition

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0h 24h 48h 72h

eN

OS

/ L1

9

Time

eNOS mRNA in VEGF treated differentiating BeWo

Control

VEGF

VEGF + L-NAME

*

*

0

0.1

0.2

0.3

0.4

0.5

0.6

0h 24h 48h 72h

Time

eNOS mRNA in Bradykinin treated differentiating BeWo

Control

Bk

Bk + L-NAME

*


97

of Forskolin, eNOS transcription by BeWo cells was no longer affected by the addition

of bradykinin. These data indicate that bradykinin is a modulator of eNOS transcription

in BeWo cells preparing to undergo morphological differentiation. Thus of those

stimulants analysed, only bradykinin-mediated endothelial stimulation is important in

differentiating trophoblasts.

6.3.3 Vascular Endothelial Growth Factor A (VEGF-A)

As VEGF is a major component of the angiogenic and endothelial pathways, its

transcription in the BeWo cell line was examined in response to a variety of endothelial

modulators. No notable alterations in VEGF-A mRNA were noted with calcium

treatment during biochemical differentiation, however VEGF-A levels were reduced

with the addition of calcium during morphological differentiation (P<0.05) (Figure

6.10). This suggests that calcium, a downstream component of the NO pathway,

suppresses the production of VEGF-A by the syncytiotrophoblast.

Figure 6.10 Effect of Calcium treatment on VEGF-A mRNA expression in

Forskolin mediated BeWo differentiation (* = P<0.05 vs. Control). L-NAME acts as

an inhibitor of the NO pathway.

The addition of L-NAME in conjunction with exogenous VEGF completely

abrogated the expression of VEGF-A mRNA (P<0.05) (Figure 6.11). This is an

intriguing response, especially as VEGF-A mRNA expression is again dampened under

these conditions after 48 and 72 hours of Forskolin treatment. That VEGF-A mRNA

levels remain stable in VEGF treated BeWo cells except after 72 hours of Forskolin

0

0.0005

0.001

0.0015

0.002

0.0025

0.003

0.0035

0h 24h 48h 72h

VEG

F-A

/ L

19

Time

VEGF-A mRNA in Ca2+ treated differentiating BeWo

Control

Ca

Ca + L-NAME

* *


98

treatment is another curious result. As exogenous VEGF treatment did not suppress

VEGF-A transcription it is possible that L-NAME itself targets VEGF-A production.

Reduced expression of VEGF-A mRNA in response to VEGF treatment after 72

hours of differentiation indicates that VEGF is auto-suppressive in the differentiated

BeWo cell line. That is, high levels of VEGF down-regulate the need for VEGF-A

transcription. The effects of exogenous VEGF on VEGF mediated regulation of

trophoblast differentiation remain an interesting field for further examination.

Figure 6.11 Effect of VEGF treatment on VEGF-A mRNA expression in Forskolin

mediated BeWo differentiation (* = P<0.05 vs. Control and VEGF; † = P<0.05 vs.

Control). L-NAME acts as an inhibitor of the NO pathway.

Figure 6.12 Effect of Bradykinin treatment on VEGF-A mRNA expression in

Forskolin mediated BeWo differentiation (* = P<0.05 vs. Control). L-NAME acts as

an inhibitor of the NO pathway.

0

0.0005

0.001

0.0015

0.002

0.0025

0.003

0.0035

0.004

0h 24h 48h 72h

VEG

F-A

/ L

19

Time

VEGF-A mRNA in VEGF treated differentiating BeWo

Control

VEGF

VEGF + L-NAME† †

*

0

0.0005

0.001

0.0015

0.002

0.0025

0.003

0.0035

0h 24h 48h 72h

VEG

F-A

/ L

19

Time

VEGF-A mRNA in Bradykinin treated differentiating BeWo

Control

Bk

Bk + L-NAME


99

Addition of bradykinin to Forskolin treated BeWo cultures did not significantly

affect the expression of VEGF-A mRNA at any time point (P>0.05) (Figure 6.12). This

indicates that bradykinin does not affect the expression of VEGF-A mRNA in the

BeWo cell line, either during homeostasis, biochemical differentiation or morphological

differentiation. Therefore bradykinin acts independently of the VEGF-A pathway in the

human trophoblast.

6.3.4 Effect of NO pathway mediators on Caspase-14 expression

Throughout the course of the experiment L-NAME initiated a significant decrease in

caspase-14 mRNA expression (P<0.05) (Figure 6.13), suggesting that the NO pathway

may play a role in the regulation of caspase-14 expression. Furthermore, caspase-14

mRNA levels are elevated after both 48 and 72 hours of Forskolin treatment relative to

0 and 24 hours, indicating that caspase-14 synthesis is increased with cell fusion.

Therefore caspase-14 is associated with both endothelial and differentiation pathways.

Indeed these processes may be inextricably linked.

Figure 6.13 Effect of the NO pathway inhibitor L-NAME on caspase-14 mRNA

expression in Forskolin mediated BeWo differentiation (* = P<0.05 vs. Control).

No data was obtained for 48h L-NAME.

When treated with calcium the expression profile of caspase-14 mRNA is

altered according to the differentiation state of the culture (Figure 6.14). Prior to the

initiation of BeWo differentiation, and during the biochemical phase of differentiation

(24 and 48 hours) no changes in the expression of caspase-14 mRNA were noted

0

0.1

0.2

0.3

0.4

0.5

0h 24h 48h 72h

casp

ase

-14

/ L

19

Time

Caspase-14 mRNA in L-NAME treated differentiating BeWo

Control

L-NAME* **


100

between controls, Calcium treated and Calcium and L-NAME co-incubated cells

(P>0.05). However, in morphologically differentiated BeWo cells (72 hours) treatment

with Calcium significantly reduced the expression of caspase-14 mRNA (P<0.05).

Interestingly, inhibition of the NO pathways by the addition of L-NAME further

reduced the expression of caspase-14 mRNA (P<0.05). The reduced caspase-14 mRNA

expression in response to L-NAME treatment suggests that inhibition of the endothelial

NO pathway also inhibits caspase-14 production in morphologically mature BeWo

cells. Thus caspase-14 expression is regulated by the endothelium-like properties of the

syncytiotrophoblast.

Figure 6.14 Effect of calcium treatment on caspase-14 mRNA expression in

Forskolin mediated BeWo differentiation (* = P<0.05 vs. Control; † = P<0.05 vs.

both Control and Ca2+). L-NAME acts as an inhibitor of the NO pathway.

Treatment of BeWo cells with exogenous VEGF does not significantly affect the

transcription of caspase-14 in the first 24 hours of differentiation (P>0.05) (Figure

6.15). However after 48 and 72 hours of Forskolin treatment, VEGF elicited a response

in caspase-14 transcription reminiscent of that obtained for 72 hour differentiated BeWo

cells following calcium treatment (Figure 6.14). For instance, after 48 hours, co-

incubation with VEGF and L-NAME significantly reduced caspase-14 mRNA levels

relative to both VEGF and Forskolin alone (P<0.05). Furthermore in morphologically

differentiating BeWo cells, VEGF suppressed caspase-14 transcription, while co-

treatment with VEGF and L-NAME decreased the quantity of caspase-14 mRNA even

further (P<0.05). Therefore VEGF inhibits caspase-14 transcription in syncytializing

BeWo cells. Moreover when added in conjunction with the NO inhibitor L-NAME,

0

0.1

0.2

0.3

0.4

0.5

0h 24h 48h 72h

casp

ase

-14

/ L

19

Time

Caspase-14 mRNA in Ca2+ treated differentiating BeWo

Control

Ca

Ca + L-NAME†

*


101

VEGF further decreases caspase-14 mRNA. As both VEGF and L-NAME suppress

caspase-14 production independently, this suggests that caspase-14 is affected by

multiple intracellular signalling pathways.

As with VEGF treatment, the addition of bradykinin to BeWo cultures did not

affect the transcription of caspase-14 in homeostatic or 24 hour Forskolin treated cells

(Figure 6.16). However an identical pattern of expression to that observed for VEGF

treatment was observed when 48 and 72 hour differentiating BeWo cultures were

treated with exogenous bradykinin. That is, L-NAME suppressed caspase-14 mRNA

relative to both bradykinin and control cultures at 48 and 72 hours, and bradykinin

inhibited caspase-14 transcription in fusing BeWo cultures. This indicates that caspase-

14 production by the differentiated trophoblast is inhibited by VEGF, bradykinin and L-

NAME, potentially via separate signalling pathways.

Figure 6.15 Effect of VEGF-A treatment on caspase-14 mRNA expression in


both Control and VEGF). L-NAME acts as an inhibitor of the NO pathway.

0

0.1

0.2

0.3

0.4

0.5

0h 24h 48h 72h

casp

ase

-14

/ L

19

Time

Caspase-14 mRNA in VEGF treated differentiating BeWo

Control

VEGF

VEGF + L-NAME*

†

†


102

Figure 6.16 Effect of bradykinin treatment on caspase-14 mRNA expression in


both Control and Bradykinin). L-NAME acts as an inhibitor of the NO pathway.

Figure 6.17 Expression of Caspase-14 mRNA in response to nitric oxide pathway

modulators in BeWo cells after 72 hours of Forskolin treatment (*=P<0.05 v.

Control; †=P<0.05 v. Control and treatment).

By 72 hours and the onset of morphological differentiation a clear pattern of

caspase-14 expression has emerged (Figure 6.17). Treatment with NO pathway agonists

leads to a dramatic reduction in caspase-14 gene transcription (P<0.05). Moreover,

transcription was further reduced by co-incubation with L-NAME (P<0.05). This

indicates that NO has an inhibitory effect upon caspase-14 transcription during

morphological differentiation. Furthermore, inhibition of NO production diminishes

caspase-14 transcription. As both NO agonism and antagonism suppress caspase-14

0

0.1

0.2

0.3

0.4

0.5

0h 24h 48h 72h

casp

ase

-14

/ L

19

Time

Caspase-14 mRNA in Bradykinin treated differentiating BeWo

Control

Bk

Bk + L-NAME

†

†*

0

0.05

0.1

0.15

0.2

0.25

0.3

Controls Ca VEGF Bk

casp

ase

-14

/ L

19

Treatment

Effect of endothelial modulators on caspase-14 mRNA in 72 hour Forskolin treated BeWo

control

w/ L-NAME* †††


103

production independently, their mode of action must therefore be independent of one

another. Thus caspase-14 is affected by multiple intracellular endothelial signalling

pathways.

Interestingly, suppression of caspase-14 transcription at the onset of

morphological differentiation (72 hours) was greatest in response to the upstream

stimulators of NO production (VEGF and bradykinin) compared with the downstream

regulator (calcium) (Figure 6.17). This indicates that the inhibition of caspase-14

transcription by upstream NO-modulators is greater than that induced by downstream

components. Thus calcium, VEGF and bradykinin initiate additional pathways outside

the classical eNOS/NO axis. This means caspase-14 transcription is inhibited by both

NO-dependent and independent pathways.

6.3.5 Discussion

The small sample size (n=4) and short incubation period (20 minutes) may be

contributing to the overall lack of statistical significance observed. 20 minutes is a very

short time period of incubation prior to RNA sampling as it typically takes longer than

this for transcription to occur (Hargrove et al. 1991). Indeed it can take up to 16 hours

for complete transcription of certain long genes (Tennyson et al. 1995). Therefore

extension of the incubation period may affect the results.

However, this study aimed to investigate the immediate effect of endothelial

modulators on gene transcription, rather than the secondary or even tertiary effects that

a longer incubation period may have provided. Accordingly a very short incubation

period was selected to observe immediate effects of the NO pathway. Indeed, significant

changes in transcript levels were found in this study, indicating that the time period

allowed was indeed sufficient to elicit a transcriptional response in the BeWo cell line.

This experiment was set up as a pilot study to explore a potential endothelial-

like role for caspase-14, as no other function for this protease in the trophoblast has

been identified. While the incubation period was very short, significant changes in

caspase-14 mRNA levels indicate that it is affected by modulators of the eNOS/NO

pathway. A much broader study is however required to fully investigate the mechanisms

of the endothelial-like role of the trophoblast, and the part caspase-14 plays therein.

Chapter 7- RNA Interference of Caspase-14 in BeWo

104

Chapter 7

RNA Interference of Caspase-14 in BeWo

7.1 Background

As outlined in Section 2.10, treatment of in vitro cell cultures with small interfering

RNAs (siRNA) results in the specific degradation of the target mRNA and protein in a

process referred to as RNA Interference (RNAi). In this way, one is able to investigate

the function of a protein by analysing what happens when it is suppressed.

As has already shown differential expression of caspase-14 during trophoblast

differentiation as represented by the BeWo cell line, this chapter aims to address

whether caspase-14 influences biochemical and morphological differentiation. Caspase-

14 expression was suppressed using RNAi technologies, whereby the natural

transcriptional modulation (Dicer) system within the cell is hijacked to result in the

specific suppression of protein activity.

The cellular reaction to caspase-14 suppression was observed throughout the

BeWo differentiation time course (up to 72 hours). Concomitantly, the effect of

caspase-14 siRNA on control BeWo cells was examined over the same time period,

providing information as to the effects of caspase-14 in homeostasis. As such the

function of caspase-14 in both control and differentiating BeWo cells could be

extrapolated from data obtained for hormonal, endothelial, cytoskeletal and cell cycle

control pathway markers.

Aim: To investigate the function of caspase-14 in normally cycling and differentiating

BeWo cells.

Hypothesis: Suppression of caspase-14 will result in enhanced morphological

differentiation of the BeWo cell line.

7.2 Validation of RNA Interference

Our confidence in the validity of gene silencing depends on the successful knockdown

of gene expression both at the transcriptional and translational levels. As the siRNA

interfere directly with the mRNA to degrade gene transcripts, mRNA expression was


105

examined by Real Time PCR. Additionally, as the protein is the functional form of the

gene, knockdown of the protein was also examined by Immunoblotting.

As described in Section 2.7, serial titrations were conducted using

Lipofectamine2000 as the transfection agent and siRNA targeting the positive control

gene GAPDH to determine the ideal transfection conditions. The siRNA concentrations

used were 25, 50, 100 and 200nM as suggested by the manufacturer. In each case, the

volume of Lipofectamine2000 used was twice that as for the siRNA.

7.2.1 Knockdown of Gene Transcripts

To confirm successful mRNA suppression, the expression of the genes against which

the siRNAs were directed was quantified by Real Time PCR and compared against non-

transfected controls. Treatment of BeWo cells with GAPDH-specific siRNA resulted in

a broad reduction in GAPDH mRNA expression (Figure 7.1), with those treated with

25, 50, 100 or 200nM of siRNA all exhibiting a statistically significant reduction in a

dose-dependent manner (P<0.05). The extent of mRNA knockdown was greatest in the

100 and 200nM groups, however considerable cell death was observed qualitatively in

the 200nM siRNA group, suggesting cytotoxicity at this concentration. Therefore a

GAPDH siRNA concentration of 100nM was considered optimal for the BeWo cell

line.

Figure 7.1 Silencing of GAPDH mRNA at various concentrations of siRNA

(*=P<0.05 vs. untreated).

0

0.2

0.4

0.6

0.8

1

GA

PD

H /

L1

9

Optimisation of GAPDH siRNA in the BeWo cell line

0nM

25nM

50nM

100nM

200nM

***

*


106

7.2.2 Knockdown of Protein

While it is useful to examine gene knockdown at the transcript level, it is also important

to analyse knockdown at the protein level. Moreover, it is the GAPDH protein that is

the active component responsible for the genes activity and function.

Protein was extracted from BeWo cells cultured with and without siRNA

targeting GAPDH, as outlined in Section 2.9.3. SDS-PAGE and Immunoblotting were

conducted on 10µg samples of protein as described in Section 2.9.4-2.9.6 to quantify the

expression and relative suppression of GAPDH, as standardised against β-actin.

Figure 7.2 Silencing of GAPDH protein following 24 and 48 hours of siRNA

treatment (* = P<0.05 v. Lipo) (i = 24h Control; ii = 24h GAPDH siRNA; iii = 48h

Control; iv = 48h GAPDH siRNA).

Treatment of the BeWo cell line with anti-GAPDH siRNA after 24 hours did not

affect the protein expression of GAPDH; however protein levels were halved 48 hours

after treatment (Figure 7.2). 70% knockdown of protein is widely acknowledged as

being efficient, however as GAPDH is ubiquitous within all cells and vital for energy

production and metabolism, excessive knockdown would lead to severe cell stress and

apoptosis. To avoid this situation, and considering our analysis of GAPDH being purely

for technique validation, we consider 50% knockdown of GAPDH to be sufficient for

identifying silencing of this gene.

7.3 Caspase-14 Silencing

7.3.1 Optimisation

Three commercially-designed unique siRNAs directed against caspase-14 were

purchased from Invitrogen (Cat. No. 1299003) (Table 2.4). In order to determine which


107

set of annealed siRNA oligonucleatides to use for the remainder of the experiments,

each siRNA was incubated in the BeWo cell line for 48 hours at a concentration of

100nM. This concentration was selected as it was most efficient in silencing GAPDH in

BeWo cells. As shown in Figure 7.3 each set of unique siRNA reduced the mRNA

expression of caspase-14 relative to scrambled siRNA, with set #1 achieving the

greatest suppression. Accordingly, caspase-14 siRNA set number 1 was used throughout

the remainder of the study.

Figure 7.3 Testing of 3 sets of 100nM caspase-14 siRNA in the BeWo cell line

(*=P<0.05 vs. scrambled siRNA). Set #1 was found to be most effective in

silencing caspase-14 mRNA expression.

Caspase-14 siRNA were delivered into BeWo cells at concentrations of 50, 100

and 200nM to examine their silencing efficiency and cytotoxicity. The expression of

caspase-14 mRNA was significantly reduced relative to those treated with scrambled

siRNA for each siRNA concentration (P<0.05) (Figure 7.4). Indeed, at 200nM there

was a 95% suppression of caspase-14 mRNA expression. A strong exponential dose-

response curve was noted for the siRNA (R=-0.88) (Figure 7.5), further indicating the

efficiency of the transfection. However, as with the GAPDH siRNA extensive

cytotoxicity was observed qualitatively in the 200nM treatment group, so for the

remainder of the study caspase-14 siRNA was delivered into the BeWo cell line at a

concentration of 100nM.

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

Neg C14 #1 C14 #2 C14 #3

casp

ase

-14

/ G

AP

DH

mR

NA

Validation of caspase-14 siRNAs in the BeWo cell line

***


108

Figure 7.4 Knock down of caspase-14 mRNA at various concentrations of siRNA

(*=P<0.05 vs. untreated).

Figure 7.5 Dose-response curve for caspase-14 siRNA in the BeWo cell line (R=-

0.88).

7.3.2 Confirmation of caspase-14 suppression following Forskolin treatment

To analyse the function of caspase-14 in the human trophoblast, the BeWo cell model of

trophoblast function and differentiation was utilised. As described in Section 4.4, this

model requires the addition of 20µM of the adenyl cyclase promoter Forskolin to the

culture medium to induce the onset of biochemical and morphological differentiation

(Wice et al. 1990; Lyden et al. 1993), with DMSO-treated cells used as a normal cycling

control. To ensure that neither DMSO nor Forskolin adversely affected the outcome of

RNAi in BeWo cells, suppression of caspase-14 by its specific siRNA was compared

with those treated with a non-specific Negative control siRNA. Confirmation of

0

0.005

0.01

0.015

0.02

0.025

0.03

Neg 50nM 100nM 200nM

casp

ase

-14

/ G

AP

DH

Optimisation of caspase-14 siRNA in the BeWo cell line

*

**

-58%-83%

-95%

0

0.006

0.012

0.018

0.024

0.03

0 50 100 150 200

casp

ase-

14

/ G

AP

DH

mR

NA

siRNA concentration (nM)

Caspase-14 siRNA dose-response curve

100%

-58%

-83%-95%


109

suppression of both caspase-14 mRNA and protein after each time point was analysed

by Real Time PCR and Western blot analysis, respectively.

Figure 7.6 A) Expression of caspase-14 mRNA in BeWo cells following treatment

with mock or caspase-14 siRNA. B) Western blot of caspase-14 protein in BeWo

cells after treatment with siRNA. „Mock‟ stands for treatment with mock siRNA;

„C14‟ stands for treatment with caspase-14 siRNA. C) Quantitation of

corresponding Western Blot (B) (* = P<0.05 vs. Mock siRNA).

0

0.01

0.02

0.03

0.04

0.05

24h 48h

casp

ase

-14

/ L

19

Time

Caspase-14 mRNA in BeWo cells after caspase-14 RNAi

Mock siRNA

Caspase-14 siRNA* *

-55% -78%

A

0

0.2

0.4

0.6

0.8

24h 48h 72h

casp

ase

-14

/ G

AP

DH

Time

Caspase-14 protein in Control BeWo after RNAi

Mock siRNA

Caspase-14 siRNA*

*

-60%

-85%

-60%

C


110

Real Time PCR revealed that caspase-14 mRNA was significantly reduced in

DMSO treated BeWo cells at all times compared with those treated with the Mock

siRNA (P<0.05) (Figures 7.6A). Thus caspase-14 is effectively suppressed in the BeWo

cell line following treatment with transcript-specific siRNA.

The successful suppression of caspase-14 protein production was quantified

using Western blot analysis. Caspase-14 levels were significantly reduced after both 48

and 72 hours of treatment with siRNA in control BeWo cells (P<0.05) (Figure 7.6B and

C). Furthermore, caspase-14 protein was suppressed to a level approaching significance

after 24 hours of treatment (P=0.06). Therefore caspase-14 is successfully suppressed

for the duration of the experiment in non-differentiating BeWo cells.

BeWo cells treated with Forskolin in conjunction with siRNA also had reduced

expression of caspase-14 mRNA after both 24 and 48 hours of treatment (P<0.05)

(Figure 7.7A). Additionally, as with control cells, caspase-14 protein levels were

approaching significance 24 hours after the addition of Forskolin in caspase-14 siRNA

treated BeWo cells (P=0.09) (Figure 7.7B and C). However caspase-14 protein was

effectively suppressed 48 and 72 hours after the induction of differentiation (P<0.05).

This indicates that the addition of Forskolin does not interfere with the efficacy of

caspase-14 siRNA in the BeWo cell line. Therefore, caspase-14 is significantly

suppressed at both the mRNA and protein levels in both control and differentiating

BeWo cells.

Importantly, Mock siRNA treated BeWo cells exhibited an identical response in

caspase-14 expression in response to Forskolin as non-transfected cells (Figure 5.2 A-

C). That is, caspase-14 mRNA and protein were significantly reduced in differentiating

BeWo cells. Furthermore, all such data concerning Forskolin treatment obtained in this

Chapter strongly correlate with data presented in Chapters 4 and 5. Therefore

transfection does not affect the response of BeWo cells to differentiation.


111

Figure 7.7 A) Expression of caspase-14 mRNA in differentiating BeWo cells

following treatment with mock or caspase-14 siRNA. B) Western blot of caspase-

14 protein in differentiating BeWo cells after treatment with siRNA. „Mock‟ stands

for treatment with mock siRNA; „C14‟ stands for treatment with caspase-14

siRNA. C) Quantitation of corresponding Western Blot (B) (* = P<0.05 vs. Mock

siRNA).

The addition of DMSO or Forskolin occurred 16 hours after treatment with

siRNA. Therefore the successful suppression of the caspase-14 protein requires longer

0

0.005

0.01

0.015

0.02

0.025

24h 48h

casp

ase

-14

/ L

19

Time

Caspase-14 mRNA in Differentiating BeWo cells after caspase-14 RNAi

Mock siRNA

Caspase-14 siRNA**

-61%-71%

A

0

0.1

0.2

0.3

0.4

0.5

24h 48h 72h

casp

ase

-14

/ G

AP

DH

Time

Caspase-14 protein in Differentiating BeWo after RNAi

Mock siRNA

Caspase-14 siRNA*

*

-41%

-61%

-56%

C


112

than 40 hours of siRNA treatment in this model; and indicating a considerable half-life

of the caspase-14 protein within the BeWo cell line, and by proxy the human

trophoblast. However, suppression of caspase-14 is achievable, making RNAi a

powerful tool with which to investigate the role of caspase-14 in the human trophoblast.

7.3.2 Effect of caspase-14 siRNA on BeWo cells

Sixteen hours after treatment with caspase-14 siRNA, 2µl/ml of DMSO was added to

BeWo cultures as a control for Forskolin treated samples (discussed in Section 7.3.3). In

this way, the effect of caspase-14 siRNA on normally cycling BeWo cells could be

assessed alongside that on differentiating BeWo cells.

Following 24 and 48 hours of culture in the absence of caspase-14, the quantity

of β-hCG mRNA was elevated (Figure 7.8A). Therefore transcription of β-hCG is, in

part, negatively regulated by the activity of caspase-14. This indicates that caspase-14 is

biologically involved in the regulation of placental hormone production in proliferative

BeWo cells, and may act as a switch for repressing hCG production in the

cytotrophoblast.

While β-hCG mRNA is affected by caspase-14 suppression, its protein remained

stable during the first 48 hours of culture (P>0.05) (Figure 7.8B and C). However after

72 hours of incubation, β-hCG protein was significantly elevated in control BeWo cells

after caspase-14 suppression. Due to the high seeding rate at the commencement of the

treatment, after 72 hours in vitro BeWo cells were confluent. As BeWo cells grow as

monolayers, confluence induces contact inhibition of the cell cycle. That β-hCG protein

was increased in BeWo cells at this point after caspase-14 knock-down suggests that

caspase-14 suppresses hCG production after cell cycle withdrawal in the trophoblast.

Despite the suppression of caspase-14 in the BeWo cell line, hPL mRNA

remained unchanged in normally cycling BeWo cells after both 24 and 48 hours of

incubation (Figure 7.9). This indicates that unlike β-hCG, hPL production is not

regulated by the presence of caspase-14. Therefore caspase-14 is not a pan-hormone

suppressor of the trophoblast, but rather a specific regulator of β-hCG synthesis.


113

Figure 7.8 A) Expression of β-hCG mRNA in BeWo cells following treatment with

mock or caspase-14 siRNA. B) Western blot of β-hCG protein in BeWo cells after

treatment with siRNA. „Mock‟ stands for treatment with mock siRNA; „C14‟

stands for treatment with caspase-14 siRNA. C) Quantitation of corresponding

Western Blot (B) (* = P<0.05 vs. Mock siRNA).

0

0.05

0.1

0.15

0.2

0.25

24h 48h / 10

hC

G /

L1

9

Time

β-hCG mRNA in Control BeWo after RNAi

Mock siRNA

Caspase-14 siRNA

*

*

+43%

+392%

A

0

0.05

0.1

0.15

0.2

0.25

24h 48h 72h

hC

G /

GA

PD

H

Time

β-hCG protein in Control BeWo after Caspase-14 RNAi

Mock siRNA

Caspase-14 siRNA

*

+176%

C


114

Figure 7.9 Expression of hPL mRNA in BeWo cells following treatment with mock

or caspase-14 siRNA.

Suppression of caspase-14 results in increased transcription of the key NO

catalyst eNOS in BeWo cells after both 24 and 48 hours of treatment (P<0.05) (Figure

7.10). Therefore caspase-14 suppresses the synthesis of eNOS in cytotrophoblasts. This

data suggests that caspase-14 is involved in the inhibition of endothelial-like properties

of the human trophoblast.

Figure 7.10 Expression of eNOS mRNA in BeWo cells following treatment with

mock or caspase-14 siRNA (* = P<0.05 v. mock siRNA).

KLF4 acts as a transcription factor to suppress the synthesis of cyclin D2,

thereby inducing cell cycle arrest at the G1/S phase boundary. Transcription of KLF4 is

increased in BeWo cells after both 24 and 48 hours of treatment with caspase-14

specific siRNA (P<0.05) (Figure 7.11A). Given this elevation, it suggests that caspase-

0

0.002

0.004

0.006

0.008

0.01

24h 48h

hP

L /

L19

Time

hPL mRNA in Control BeWo after RNAi

Mock siRNA

Caspase-14 siRNA

0

0.03

0.06

0.09

0.12

0.15

24h 48h

eN

OS

/ L1

9

Time

eNOS mRNA in Control BeWo after RNAi

Mock siRNA

Caspase-14 siRNA

*

*+63%

+74%


115

14 inhibits KLF4-mediated cell cycle suppression in BeWo cells. Under this hypothesis,

this would suggest that caspase-14 agonises proliferation of the human trophoblast.

Figure 7.11 A) Expression of KLF4 mRNA in BeWo cells following treatment with

mock or caspase-14 siRNA. B) Western blot of KLF4 protein in BeWo cells after

treatment with siRNA. „Mock‟ stands for treatment with mock siRNA; „C14‟

stands for treatment with caspase-14 siRNA. C) Quantitation of corresponding

Western Blot (B) (* = P<0.05 vs. Mock siRNA).

0

0.2

0.4

0.6

0.8

1

24h 48h

KLF

4 /

L1

9

Time

KLF4 mRNA in Control BeWo after RNAi

Mock siRNA

Caspase-14 siRNA+46%

+104%

*

*

A

0

0.2

0.4

0.6

0.8

1

24h 48h 72h

KLF

4 /

GA

PD

H

Time

KLF4 protein in Control BeWo after Caspase-14 RNAi

Mock siRNA

Caspase-14 siRNA

-34%

*

C


116

However KLF4 translation was not found to be significantly altered after either

24 or 72 hours of treatment, but rather reduced after 48 hours (P<0.05) (Figure 7.11B

and C). This is in conflict with the mRNA data, indicating post-transcriptional

regulation of KLF4 activity in the BeWo cell line.

Figure 7.12 A) Expression of cytokeratin-18 mRNA in BeWo cells following

treatment with mock or caspase-14 siRNA. B) Western blot of cytokeratin-18

protein in BeWo cells after treatment with siRNA. „Mock‟ stands for treatment

with mock siRNA; „C14‟ stands for treatment with caspase-14 siRNA. C)

Quantitation of corresponding Western Blot (B) (* = P<0.05 vs. Mock siRNA).

0

0.04

0.08

0.12

0.16

0.2

24h 48h

ck1

8 /

L1

9

Time

Cytokeratin-18 mRNA in Control BeWo after RNAi

Mock siRNA

Caspase-14 siRNA*

*

+39%

+160%

A

-2E-15

0.3

0.6

0.9

1.2

1.5

1.8

24h 48h 72h

ck1

8 /

GA

PD

H

Time

Cytokeratin-18 protein in Control BeWo after RNAi

Mock siRNA

Caspase-14 siRNA

*

+54% +46%

C


117

By suppressing caspase-14, the transcription of cytokeratin-18 was significantly

increased after 24 and 48 hours (P<0.05) (Figure 7.12A). This indicates that caspase-14

is involved in inhibiting the synthesis of cytokeratin-18 in the BeWo cell line. No

increase was found at the protein level until 72 hours after the initiation of treatment

(P<0.05), however after 48 hours it was approaching significance (P=0.056) (Figure

7.12B and C). This suggests that cytokeratin-18 translation requires a long time period

for completion. Alternatively, post-transcriptional mechanisms may inhibit the

translation of the mRNA until such time as intermediate filament and cytoskeletal

modulation is required.

Figure 7.13 Expression of E-cadherin mRNA in BeWo cells following treatment

with mock or caspase-14 siRNA.

E-cadherin is a vital component of adherens junctions in the plasma membrane,

regulating intercellular communication and attachment. Morphological differentiation of

the trophoblast also induces changes in the expression pattern of E-cadherin, reducing

its expression as the surface area:volume ratio decreases. No alteration in E-cadherin

mRNA expression was found in BeWo cells following the suppression of caspase-14

(P>0.05) (Figure 7.13). This indicates that caspase-14 does not influence adherens

junction-mediated cell-cell communication, nor does it modulate morphological

differentiation in the BeWo cell line and by proxy the human trophoblast.

0

0.2

0.4

0.6

0.8

1

24h 48h

E-ca

d /

L1

9

Time

E-cadherin mRNA in Control BeWo after RNAi

Mock siRNA

Caspase-14 siRNA


118

7.3.3 Effect of caspase-14 siRNA on differentiating BeWo cells

Sixteen hours after treatment with caspase-14 siRNA, 20µM of Forskolin was added to

BeWo cultures to induce differentiation. From these data I propose to deduce a role for

caspase-14 in trophoblast differentiation.

In support of data for control BeWo cells (Figure 7.8), β-hCG mRNA was

increased following the suppression of caspase-14 24 hours after the addition of

Forskolin (Figure 7.14A), confirming a role for caspase-14 in the regulation of hCG

synthesis in homeostatic and differentiating trophoblast. As hCG production is greatly

elevated during differentiation, β-hCG mRNA not being increased after 48 hours of

differentiation suggests that there may be a threshold limit to its production. This may

explain why there is a smaller increase in β-hCG after siRNA in Forskolin treated

BeWo compared to DMSO treated controls (Figure 7.14 versus Figure 7.8).

hCG protein production was increased 48 and 72 hours after Forskolin treatment

(P<0.05) (Figure 7.14B and C). However, β-hCG protein levels were significantly

reduced 24 hours after Forskolin treatment, representing an interesting regulatory

mechanism of hCG production by caspase-14 in the biochemically differentiating

trophoblast. Indeed, this may show that caspase-14 does not directly affect hCG

production.

In accordance with data for control BeWo cells (Figure 7.9), hPL mRNA

remained unchanged in differentiating BeWo cells after both 24 and 48 hours of

incubation (P<0.05), (Figure 7.15), although after 48 hours this was approaching

statistical significance (P=0.06). Therefore capase-14 suppression does not affect hPL

synthesis in the BeWo cell line, indicating that caspase-14 specifically targets hCG

homeostasis rather than hPL.


119

Figure 7.14 A) Expression of β-hCG mRNA in differentiating BeWo cells following

treatment with mock or caspase-14 siRNA. B) Western blot of β-hCG protein in

differentiating BeWo cells after treatment with siRNA. „Mock‟ stands for

treatment with mock siRNA; „C14‟ stands for treatment with caspase-14 siRNA.

C) Quantitation of corresponding Western Blot (B) (* = P<0.05 vs. Mock siRNA).

0

0.6

1.2

1.8

2.4

3

24h 48h / 10

hC

G /

L1

9

Time

β-hCG mRNA in Differentiating BeWo after RNAi

Mock siRNA

Caspase-14 siRNA

*

+26%

+47%

A

0

0.4

0.8

1.2

1.6

2

24h 48h 72h

hC

G /

GA

PD

H

Time

β-hCG protein in Differentiating BeWo after RNAi

Mock siRNA

Caspase-14 siRNA

** *

-26%+24% +51%

C


120

Figure 7.15 Expression of hPL mRNA in differentiating BeWo cells following

treatment with mock or caspase-14 siRNA.

Figure 7.16 Expression of eNOS mRNA in differentiating BeWo cells following

treatment with mock or caspase-14 siRNA (* = P<0.05 v. mock siRNA).

Increased eNOS mRNA levels were observed after 24 hours of BeWo

differentiation in the absence of caspase-14 (P<0.05) (Figure 7.16). Therefore caspase-

14 suppresses the synthesis of eNOS in early biochemical differentiation of the BeWo

cell line. However by late stage biochemical differentiation (48 hours), eNOS synthesis

was unchanged compared with those treated with Mock siRNA (P>0.05). This indicates

that caspase-14 regulation of eNOS is restricted to early BeWo differentiation. However

the most notable alterations in caspase-14 synthesis by endothelial modulators were

observed during morphological differentiation (72 hours) (Figure 6.17). Therefore

examination of eNOS modulation by caspase-14 is required for the 72 hour time point

in order to confirm or deny a tight synergy between the two pathways.

0

0.0005

0.001

0.0015

0.002

0.0025

0.003

24h 48h

hP

L /

L19

Time

hPL mRNA in Differentiating BeWo after RNAi

Mock siRNA

Caspase-14 siRNA

0

0.08

0.16

0.24

0.32

0.4

24h 48h

eN

OS

/ L1

9

Time

eNOS mRNA in Differentiating BeWo after RNAi

Mock siRNA

Caspase-14 siRNA

+65%

*


121

Figure 7.17 A) Expression of KLF4 mRNA in differentiating BeWo cells following

treatment with mock or caspase-14 siRNA. B) Western blot of KLF4 protein in

differentiating BeWo cells after treatment with siRNA. „Mock‟ stands for

treatment with mock siRNA; „C14‟ stands for treatment with caspase-14 siRNA.

C) Quantitation of corresponding Western Blot (B) (* = P<0.05 vs. Mock siRNA).

0

1

2

3

4

5

24h 48h

KLF

4 /

L1

9

Time

KLF4 mRNA in Differentiating BeWo after RNAi

Mock siRNA

Caspase-14 siRNA

A

0

0.2

0.4

0.6

0.8

1

1.2

1.4

24h 48h 72h

KLF

4 /

GA

PD

H

Time

KLF4 protein in Differentiating BeWo after RNAi

Mock siRNA

Caspase-14 siRNA*

-86%

C


122

Figure 7.18 A) Expression of cytokeratin-18 mRNA in differentiating BeWo cells

following treatment with mock or caspase-14 siRNA. B) Western blot of

cytokeratin-18 protein in differentiating BeWo cells after treatment with siRNA.

„Mock‟ stands for treatment with mock siRNA; „C14‟ stands for treatment with

caspase-14 siRNA. C) Quantitation of corresponding Western Blot (B) (* = P<0.05

vs. Mock siRNA).

0

0.05

0.1

0.15

0.2

24h 48h

ck1

8 /

L1

9

Time

Cytokeratin-18 mRNA in Differentiating BeWo after RNAi

Mock siRNA

Caspase-14 siRNA

+43%

+65%

*

A

0

0.5

1

1.5

2

24h 48h 72h

ck1

8 /

GA

PD

H

Time

Cytokeratin-18 protein in Differentiating BeWo after RNAi

Mock siRNA

Caspase-14 siRNA

* *

-33% +30%

-24%

C


123

Figure 7.19 Expression of E-cadherin mRNA in differentiating BeWo cells

following treatment with mock or caspase-14 siRNA (* = P<0.05 v. mock siRNA).

Unlike the other mRNAs presented thus far, KLF4 in differentiating BeWo cells

was not affected by the loss of caspase-14 (P>0.05) (Figure 7.17A). However,

expression of its protein was significantly reduced following 24 hours of Forskolin

treatment (P<0.05) (Figure 7.17B and C). Therefore caspase-14 only influences KLF4

during early BeWo cell differentiation. As KLF4 suppresses proliferation, reduced

protein levels indicate that differentiating BeWo cells may be escaping cell cycle

withdrawal in the absence of caspase-14, thereby enhancing proliferation. This however

is in conflict with data pertaining to differentiation, which suggest that differentiation is

enhanced in the absence of caspase-14 (Figures 7.8; 7.10; 7.14; 7.16). Therefore an

interesting mechanism exists in the interaction between caspase-14 and KLF4.

After 24 hours of Forskolin treatment, cytokeratin-18 mRNA levels were

increased in response to Forskolin in the absence of caspase-14 (P<0.05) (Figure

7.18A). Furthermore, after 48 hours of differentiation cytokeratin-18 expression was

approaching differentiation (P=0.06). Therefore, the lack of caspase-14 in

differentiation also leads to the increased transcription of key cytoskeletal components.

Interestingly, the cytokeratin-18 protein was significantly reduced at this time point

(P<0.05) (Figure 7.18B and C). Moreover, cytokeratin-18 is significantly elevated 48

hours after Forskolin treatment in BeWo cells deficient in caspase-14 indicating a

period of lag between transcription and translation lasting approximately 24 hours.

However, by the time that morphological differentiation is initiated (72 hours)

expression of cytokeratin-18 was unchanged (P>0.05) (Figure 7.18B and C), suggesting

that caspase-14 does not influence cytokeratin-18 production in fusing BeWo cells.

0

0.1

0.2

0.3

0.4

0.5

24h 48h

E-ca

d /

L1

9

Time

E-cadherin mRNA in Differentiating BeWo after RNAi

Mock siRNA

Caspase-14 siRNA

*

+57%


124

Therefore caspase-14 modulates cytokeratin-18 production during biochemical

differentiation of BeWo cells in preparation for cytoskeletal rearrangements

accompanying morphological differentiation.

Suppression of caspase-14 in differentiating BeWo resulted in an increase in E-

cadherin mRNA expression after 24 hours (P<0.05) (Figure 7.19). By 48 hours

however, no change in E-cadherin synthesis was observed after the inhibition of

caspase-14, suggesting that caspase-14 is affecting BeWo cell attachment and

communication during early biochemical differentiation.

7.3.4 Discussion

Human chorionic gonadotrophin (hCG) is produced upon implantation and

differentiation of the trophoblast, and is widely used as the earliest marker of

pregnancy. When treated with Forskolin the BeWo cell line undergoes differentiation,

initiating the production of hCG. Following suppression of caspase-14, synthesis and

production of hCG was generally increased (Table 7.1), suggesting that caspase-14 is

involved in the regulation of hCG.

The possibility that caspase-14 suppresses trophoblast hormone production as a

whole was assessed by analysing the effect of caspase-14 suppression on the regulation

of hPL transcription. While hPL mRNA is significantly reduced in differentiating BeWo

cells, inhibition of caspase-14 does not affect hPL transcription (Table 7.1). Therefore

caspase-14 does not regulate trophoblast hormone production as a general event, but

rather targets the hCG pathway.

Endothelial Nitric Oxide Synthase (eNOS) is the key catalyst of Nitric Oxide

(NO) production in endothelial cells. As discussed in Chapter 6, BeWo cells possess

endothelial-like properties, including the up-regulation of eNOS during differentiating.

Intriguingly, caspase-14 transcription is mediated by endothelial pathway agonists

during morphological differentiation of BeWo cells (Section 6.3.4), indicating a

relationship between caspase-14 and endothelial regulation. Upon the suppression of

caspase-14, eNOS mRNA was increased both in control and early differentiating BeWo

cells (Table 7.1). This indicates that caspase-14 restricts the endothelial-like properties

of the trophoblast, perhaps serving as a checkpoint against premature adoption of the

endothelial-like phenotype. These data also confirm the findings of Section 6.3.4, that

there exists a symbiosis between endothelial function and caspase-14 biology.


125

Control Differentiation

Gene Product 24h 48h 72h Gene Product 24h 48h 72h

β-hCG mRNA ↑ ↑ NA

β-hCG mRNA ↑ - NA

Protein - - ↑ Protein ↓ ↑ ↑

hPL mRNA - - NA hPL mRNA - - NA

eNOS mRNA ↑ ↑ NA eNOS mRNA ↑ - NA

KLF4 mRNA ↑ ↑ NA

KLF4 mRNA - - NA

Protein - ↑ - Protein ↓ - -

Ck-18 mRNA ↑ ↑ NA

Ck-18 mRNA ↑ - NA

Protein - - ↑ Protein ↓ ↑ -

E-cadherin mRNA - - NA E-cadherin mRNA ↑ - NA

Table 7.1 A summary of the effects of caspase-14 siRNA treatment on the BeWo

cell line. Arrows indicate the direction of any significant change in mRNA or

protein expression (P<0.05). Dashes indicate no significant change (P>0.05). NA

indicates no available data. “Ck-18” stands for “cytokeratin-18”.

KLF4 is a transcription factor involved in suppression of Cyclin D2, thereby

initiating cell cycle withdrawal. It has been postulated that KLF4 may contain a binding

recognition sequence for caspase-14. Consequently the expression of KLF4 was

examined following treatment of BeWo cells with caspase-14 siRNA. Knock-down of

caspase-14 in BeWo cells induced the transcription of KLF4 in non-differentiating

BeWo cells (Table 7.1). However, this did not translate into increased protein levels,

rather diminishing its levels after 48 hours of RNAi. During differentiation, KLF4

mRNA remained unchanged following caspase-14 siRNA treatment, however during

early differentiation (24 hours) its protein was significantly reduced. Together these data

suggest translational regulation of KLF4 in the absence of caspase-14, and an

interesting symbiosis between the two genes in the trophoblast. As KLF4 is an

antagonist of the cell cycle, this indicates that during early differentiation caspase-14

suppresses cell cycle suppression. That is, caspase-14 promotes proliferation. Therefore,

upon treatment with caspase-14 siRNA, BeWo cells undergo KLF4-mediated cell cycle

withdrawal, thereby promoting differentiation as seen by increased production of β-hCG

(Table 7.1). Ergo caspase-14 suppresses trophoblast differentiation.

In the epidermis caspase-14 is concerned with cleaving and activating

profilaggrin during terminal differentiation, culminating in stabilisation of the keratin

cytoarchitecture. Subsequently the regulation of cytokeratin-18, the most abundant

keratin in the trophoblast, following the suppression of caspase-14 in BeWo cells was

assessed to investigate whether caspase-14 possess a conserved role in keratin


126

homeostasis in barrier formation. Cytokeratin-18 transcription was elevated in both

control and early differentiating BeWo cells after caspase-14 knockdown (Table 7.1).

Furthermore, an interesting expression pattern was noted for its protein expression.

Cytokeratin-18 was reduced during early biochemical differentiation, yet increased after

48 hours of Forskolin treatment. Therefore an intriguing symbiosis exists between

caspase-14 and cytokeratin-18 during biochemical differentiation of BeWo cells in

preparation for the cytoskeletal rearrangements accompanying morphological

differentiation.

E-cadherin is a marker of BeWo cell morphological differentiation, in that its

expression is diminished in fusing or fused cells due to the reduced surface area-to-

volume ratio (Section 4.4.2). Accordingly, increased transcription of E-cadherin in early

stage differentiating BeWo cells indicates that cell-cell fusion is accelerated in response

to caspase-14 suppression. That no alteration in E-cadherin mRNA expression was

noted after 48 hours indicates that this transcriptional regulation of E-cadherin by

caspase-14 is transient and perhaps over-ridden by further factors involved in

trophoblast differentiation. Furthermore, caspase-14 did not affect E-cadherin mRNA

expression in control BeWo cells, indicating that caspase-14 is not a direct suppressor of

cell-cell fusion. Although increased transcription during early-stage differentiation

suggests that caspase-14 is involved, albeit as a secondary checkpoint of morphological

differentiation.

On the face of it, caspase-14 is a promiscuous molecule. Its suppression

seemingly affects a multitude of signalling pathways including hormonal (hCG),

endothelial (eNOS), cell cycle (KLF4), cytoskeletal (cytokeratin-18) and attachment

and intercellular communication (E-cadherin). However, each of these proteins are also

implicated in trophoblast differentiation. Synthesis and production of β-hCG, eNOS and

KLF4 is significantly increased throughout BeWo cell differentiation (Chapters 4 and

5). Moreover E-cadherin is disrupted and reduced upon morphological differentiation of

BeWo cells (Section 4.4.2), and cytokeratin-18 synthesis and production is increased

across the crucial period between 48 and 72 hours when morphological differentiation is

induced (Section 5.4.3). Therefore each of those products found to be influenced by

caspase-14 expression is somehow involved in BeWo cell differentiation.

That caspase-14 is significantly reduced across BeWo cell differentiation

indicates that it is somehow involved in differentiation, likely in a suppressive capacity

(Chapter 5). Indeed data presented in this Chapter strongly indicate that caspase-14 does

indeed suppress BeWo cell differentiation both in the biochemical (24-48 hours) and


127

morphological (48-72 hours) phases. This suggests as intriguing function for caspase-14

as a checkpoint for adequate BeWo cell differentiation. Future studies into the question

of whether caspase-14 regulates the extent of differentiation and acts as an essential

checkpoint to prevent uncontrolled differentiation may provide further insight into its

function in the human placenta.

The target of caspase-14 activity in the epidermis- profilaggrin- is not present in

either the human placenta or BeWo cell line (Sections 3.4.1 and 5.4.1). Therefore other

targets must exist for caspase-14 in the human placenta. Despite the significant

advances in the understanding of caspase-14 biology that have been made in the current

study, its immediate target(s) in the trophoblast remain elusive.

Chapter 8- Discussion

128

Chapter 8

Discussion

Caspase-14 is expressed in remarkably few tissues, most of which form a barrier

between two privileged environments indicating a conserved role in barrier formation

(Lippens et al. 2003). Its characterisation has been best explored in the epidermis where

it indirectly modulates corneocyte intermediate filament stability, hydration, water

balance and UV-B protection during keratinocyte differentiation (Eckhart et al. 2000;

Rendl et al. 2002; Fischer et al. 2004; Denecker et al. 2007), however very little

attention has been granted to caspase-14 in the human trophoblast. As such this thesis is

dedicated to the investigation of caspase-14 expression and function in the human

trophoblast. It was hypothesised that caspase-14 represents a conserved link in

differentiation and barrier formation between the epidermis and trophoblast.

8.1 Caspase-14 in the Human Trophoblast

Caspase-14 is consistently expressed in the human trophoblast throughout gestation

(Figures 3.7-3.9). Moreover, caspase-14 is transcriptionally regulated upon apoptotic

stimulation of chorionic villous explants, indicating a role in trophoblast apoptosis

(Figure 3.1).

To date, no evidence suggests that caspase-14 is involved in classical apoptosis

pathways, however it is suggested that terminal differentiation leading to intercellular

fusion or barrier formation shares common signalling pathways with apoptosis (Ishizaki

et al. 1998; Gandarillas 2000; Fernando et al. 2002; Arama et al. 2003; Mogi et al. 2003;

Chaturvedi et al. 2006). Due to the homology of caspase-14 with pro-apoptotic

proteases, it may represent a conserved link between apoptosis and differentiation, with

particular reference to barrier formation (Lippens et al. 2003).

The apoptotic index is enhanced in the trophoblast of preeclamptic placentae

(Leung et al. 2001; Austgulen et al. 2004), so one would then infer that the expression

of apoptotic molecules would be increased with preeclampsia. However, caspase-14

expression is not affected by the development of preeclampsia (Figure 3.7). Therefore

caspase-14 is not involved in the biology of apoptosis in the human trophoblast. To test


129

this hypothesis, the cytotrophoblast-like BeWo cell line was used as an in vitro model of

trophoblast function, and induced to undergo both apoptosis and differentiation.

Treatment of the BeWo cell line with 1μM Staurosporine successfully induces

apoptosis (Section 4.3); however this does not affect caspase-14 expression (Figure 5.1

A-C). Caspases are typically involved in either the initiation or execution of classical

apoptotic pathways, so the unaltered transcription of caspase-14 during apoptosis

indicates a non-apoptotic function for caspase-14. Moreover, the transcription of the key

apoptotic protease caspase-3 is greatly reduced after Staurosporine treatment (Figure

4.8). This suggests that stores of mRNA for apoptotic caspases are translated upon

stimulation of apoptotic cascades. Therefore as it was transcriptionally unaltered,

caspase-14 is not involved in classical trophoblast apoptosis.

Forskolin is an effective inducer of both biochemical and morphological

differentiation in the BeWo cell line (Section 4.4). While my original hypothesis stated

that caspase-14 expression would be increased during trophoblast differentiation,

induction of BeWo cell differentiation results in reduced caspase-14 expression at both

the transcriptional and translational levels (Figure 5.2 A-C). As the direction of the

change is the opposite of that observed in the epidermis (Eckhart et al. 2000; Lippens et

al. 2000; Rendl et al. 2002) the mechanism of caspase-14 activation and subsequent

function is likely to be different in the trophoblast. Indeed, the target of caspase-14

activity in the epidermis, profilaggrin, is absent from both the human villous trophoblast

and BeWo cell line (Figures 3.10 and 5.3), confirming a disparity in caspase-14 function

between the two tissues.

The reduced expression of caspase-14 throughout BeWo cell differentiation

suggests a role suppressing trophoblast differentiation. Analysis of this prospect was

conducted by looking at the effect of caspase-14 suppression on hCG, which is

commonly used as a marker of trophoblast differentiation and pregnancy. The

production of hCG is elevated following caspase-14 suppression (Figures 7.8 and 7.14),

confirming a role for caspase-14 in trophoblast differentiation inhibition.

The cell cycle inhibitor KLF4 contains a binding motif consistent with that

identified for caspase-14 (Park et al. 2006). Furthermore, KLF4 is involved in the

maintenance of the human placenta (Blanchon et al. 2001; Blanchon et al. 2006), while

mouse knock-out models of caspase-14 and KLF4 display remarkably similar epidermal

phenotypes (Segre et al. 1999; Demerjian et al. 2007; Denecker et al. 2007). However,

the pattern of KLF4 mRNA expression in BeWo cells is the opposite of that for

caspase-14, increasing significantly throughout differentiation (Figure 5.4 A). As KLF4


130

is a cell cycle suppressor, this indicates that KLF4 is involved in the cell cycle

withdrawal required for differentiation, suggesting an antagonism between KLF4 and

caspase-14 whereby KLF4 promotes while caspase-14 suppresses trophoblast

differentiation.

Indeed, suppression of caspase-14 by RNAi results in increased KLF4 mRNA

expression in line with the enhanced differentiation indicated by hCG production (Table

7.1). However despite all of this there is no concrete evidence for direct interaction

between the two, so further examination of the potential relationship between caspase-

14 and KLF4 is required.

Cytokeratin-18 mRNA and protein were elevated after 48 and 72 hours of

Forskolin treatment (Figure 5.5 A-C), correlating with the onset of morphological

BeWo cell differentiation. Thus elevated cytokeratin-18 parallels with the onset of

fusion. This supports a significant rearrangement of the intermediate filament

cytoskeleton in preparation for intercellular fusion of the trophoblast.

That cytokeratin-18 is further increased following caspase-14 suppression

(Figures 7.12 and 7.18) indicates that morphological differentiation is accelerated in the

absence of caspase-14. This provides further evidence for a role for caspase-14 in

inhibiting trophoblast differentiation. Furthermore, as the cytokeratin architecture is

indirectly affected by caspase-14 activity in keratinocytes (Denecker et al. 2007), it is

attractive to postulate that cytokeratin-18 arrangement is a downstream target of

caspase-14 during the morphological differentiation of the trophoblast.

8.2 Endothelial Properties of the Trophoblast and Caspase-14

As a secondary series of experiments, the effect of differentiation on the potential

endothelial-like properties of the trophoblast, and the part played by regulators of the

Nitric Oxide (NO) pathway on differentiation itself was analysed. Synthesis of the

catalyst for NO production eNOS is increased across the course of BeWo cell

differentiation (Figure 6.1). Concomitantly, synthesis of the up-stream NO pathway

agonist VEGF-A remains unchanged during early biochemical differentiation, but is

amplified in preparation for morphological differentiation (Figure 6.2). Thus trophoblast

differentiation affects the endothelial NO pathway. This indicates that the trophoblast,

particularly the differentiated phenotype, modulates the endothelial-like properties of

the trophoblast.


131

While differentiation amplifies eNOS and VEGF production, stimulation with

NO pathway agonists does not impact on the synthesis of hCG (Section 6.3.1),

suggesting that an endothelial phenotype does not impact on the rate or success of

BeWo cell differentiation. Therefore differentiation initiates endothelial-like properties

of the trophoblast, but stimulation of the NO pathway does not affect differentiation.

While caspase-14 production is reduced in differentiating BeWo cells (Figure

5.2 A-C), it is further reduced in response to endothelial agonists in morphological

differentiating BeWo cells (Section 6.3.4). Interestingly however, suppression of the

NO pathway also decreases caspase-14 synthesis throughout differentiation (Figure

6.13). Therefore a complex signalling network exists between caspase-14 and the

endothelial properties of the trophoblast.

8.3 Concluding Remarks

In summary caspase-14 suppresses trophoblast differentiation, potentially by promoting

proliferation, while co-ordinately curbing the endothelial function of the trophoblast.

Indeed, endothelial-like properties are in themselves a function of terminally

differentiated trophoblast, so caspase-14 may indeed be solely involved in the regulation

of differentiation.

Microarray technology could be used in conjunction with RNAi to reveal genes

and pathways directly affected by caspase-14 in the differentiating trophoblast. hCG

inhibits the activation of AP-1 transcription factors (Manna et al. 2000), the latter of

which also regulate the transcription of caspase-14 (Ballaun et al. 2008). Indeed the pro-

mitotic c-Jun is restricted to the proliferative cytotrophoblast, while the quiescence-

maintaining JunD is confined to the syncytium (Bamberger et al. 2004). Thus c-Jun may

stimulate and JunD inhibit caspase-14 transcription in the trophoblast. Further studies

involving RNAi against c-Jun and JunD may reveal the mechanisms controlling

caspase-14-mediated differentiation inhibition in the trophoblast. Thereby the molecular

triggers dictating caspase-14 activity may be revealed.

To date, much of the research concerning caspase-14 has been conducted using a

strain of knockout mouse. It would be interesting to look at the morphology, cell, and

molecular biology of the knockout mouse placenta as a means of examining the role of

caspase-14 in the trophoblast. These could then be compared against data for the wild

type mouse placenta, and data obtained from human placentae, BeWo cells and primary


132

cell cultures. This would provide another excellent resource for the study of caspase-14

in the context of trophoblast biology.

While caspase-14 synthesis is unaffected in preeclampsia, the morphology of the

trophoblast is frequently disrupted. Consequently, inappropriate trophoblast

differentiation is a potential cause of the aetiology of preeclampsia. Further

investigation into the formation of the maternal-fetal barrier has important implications

for the understanding of diseases in which differentiation may be disrupted, especially

Preeclampsia and Intrauterine Growth Restriction (IUGR).

Chapter 9- Bibliography

133


Ahmad, M., S. M. Srinivasula, et al. (1998). "Identification and characterization of

murine caspase-14, a new member of the caspase family." Cancer Res 58(22):

5201-5.

Al-Nasiry, S., B. Spitz, et al. (2006). "Differential effects of inducers of syncytialization

and apoptosis on BeWo and JEG-3 choriocarcinoma cells." Human

Reproduction 21(1): 193-201.

Arama, E., J. Agapite, et al. (2003). "Caspase activity and a specific cytochrome C are

required for sperm differentiation in Drosophila." Developmental Cell 4(5): 687-

97.

Aschkenazi, S., S. Straszewski, et al. (2002). "Differential regulation and function of the

Fas/Fas ligand system in human trophoblast cells." Biol Reprod 66(6): 1853-61.

Austgulen, R., C. V. Isaksen, et al. (2004). "Pre-eclampsia: associated with increased

syncytial apoptosis when the infant is small-for-gestational-age." J Reprod

Immunol 61(1): 39-50.

Ballaun, C., S. Karner, et al. (2008). "Transcription of the caspase-14 gene in human

epidermal keratinocytes requires AP-1 and NFkappaB." Biochem Biophys Res

Commun.

Bamberger, A. M., C. M. Bamberger, et al. (2004). "Expression pattern of the activating

protein-1 family of transcription factors in the human placenta." Mol Hum

Reprod 10(4): 223-8.

Black, S., M. Kadyrov, et al. (2004). "Syncytial fusion of human trophoblast depends on

caspase 8." Cell Death and Differentiation 11(1): 90-8.

Blanchon, L., J. L. Bocco, et al. (2001). "Co-localization of KLF6 and KLF4 with

pregnancy-specific glycoproteins during human placenta development."

Mechanisms of Development 105(1-2): 185-9.

Blanchon, L., R. Nores, et al. (2006). "Activation of the human pregnancy-specific

glycoprotein PSG-5 promoter by KLF4 and Sp1." Biochemical and Biophysical

Research Communications 343(3): 745-53.

Bleackley, R. C. and J. A. Heibein (2001). "Enzymatic control of apoptosis." Nat Prod

Rep 18(4): 431-40.

Bradford, M. M. (1976). "A rapid and sensitive method for the quantitation of

microgram quantities of protein utilizing the principle of protein-dye binding."

Analytical Biochemistry 72: 248-54.

Brancolini, C., D. Lazarevic, et al. (1997). "Dismantling cell-cell contacts during

apoptosis is coupled to a caspase-dependent proteolytic cleavage of beta-

catenin." J Cell Biol 139(3): 759-71.

Brembeck, F. H., M. Rosario, et al. (2006). "Balancing cell adhesion and Wnt signaling,

the key role of beta-catenin." Curr Opin Genet Dev 16(1): 51-9.

Cha, M. S., M. J. Lee, et al. (2001). "Endogenous production of nitric oxide by vascular

endothelial growth factor down-regulates proliferation of choriocarcinoma

cells." Biochem Biophys Res Commun 282(4): 1061-6.

Charles, A. K., S. Hisheh, et al. (2005). "The expression of apoptosis related genes in

the first trimester human placenta using a short term in vitro model." Apoptosis

10(1): 135-40.

Charnock-Jones, D. S., A. M. Sharkey, et al. (1994). "Vascular endothelial growth

factor receptor localization and activation in human trophoblast and

choriocarcinoma cells." Biol Reprod 51(3): 524-30.


134

Chaturvedi, V., L. A. Sitailo, et al. (2006). "Defining the caspase-containing apoptotic

machinery contributing to cornification in human epidermal equivalents."

Experimental Dermatology 15(1): 14-22.

Chien, A. J., R. B. Presland, et al. (2002). "Processing of native caspase-14 occurs at an

atypical cleavage site in normal epidermal differentiation." Biochemical and

Biophysical Research Communications 296(4): 911-7.

Chilosi, M., E. Piazzola, et al. (1998). "Differential expression of p57kip2, a maternally

imprinted cdk inhibitor, in normal human placenta and gestational trophoblastic

disease." Lab Invest 78(3): 269-76.

Chomczynski, P. and N. Sacchi (1987). "Single-step method of RNA isolation by acid

guanidinium thiocyanate-phenol-chloroform extraction." Analytical

Biochemistry 162(1): 156-9.

Conrad, K. P., M. Vill, et al. (1993). "Expression of nitric oxide synthase by

syncytiotrophoblast in human placental villi." Faseb J 7(13): 1269-76.

Coutifaris, C., L. C. Kao, et al. (1991). "E-cadherin expression during the differentiation

of human trophoblasts." Development 113(3): 767-77.

Cronier, L., B. Bastide, et al. (1994). "Gap junctional communication during human

trophoblast differentiation: influence of human chorionic gonadotropin."

Endocrinology 135(1): 402-8.

Dai, X. and J. A. Segre (2004). "Transcriptional control of epidermal specification and

differentiation." Curr Opin Genet Dev 14(5): 485-91.

Das, M., B. Xu, et al. (2004). "Phosphatidylserine efflux and intercellular fusion in a

BeWo model of human villous cytotrophoblast." Placenta 25(5): 396-407.

De Falco, M., V. Fedele, et al. (2004). "Immunohistochemical distribution of proteins

belonging to the receptor-mediated and the mitochondrial apoptotic pathways in

human placenta during gestation." Cell Tissue Res 318(3): 599-608.

Demerjian, M., J. P. Hachem, et al. (2007). "Acute Modulations in Permeability Barrier

Function Regulate Epidermal Cornification. Role of Caspase-14 and the

Protease-Activated Receptor Type 2." Am J Pathol.

Denecker, G., E. Hoste, et al. (2007). "Caspase-14 protects against epidermal UVB

photodamage and water loss." Nat Cell Biol 9(6): 666-74.

Drake, J. M., R. R. Friis, et al. (2003). "The role of sFRP4, a secreted frizzled-related

protein, in ovulation." Apoptosis 8(4): 389-97.

Drees, F., S. Pokutta, et al. (2005). "Alpha-catenin is a molecular switch that binds E-

cadherin-beta-catenin and regulates actin-filament assembly." Cell 123(5): 903-

15.

Eberhart, C. G. and P. Argani (2001). "Wnt signaling in human development: beta-

catenin nuclear translocation in fetal lung, kidney, placenta, capillaries, adrenal,

and cartilage." Pediatr Dev Pathol 4(4): 351-7.

Eckhart, L., J. Ban, et al. (2000). "Caspase-14: analysis of gene structure and mRNA

expression during keratinocyte differentiation." Biochemical and Biophysical

Research Communications 277(3): 655-9.

Eckhart, L., W. Declercq, et al. (2000). "Terminal differentiation of human

keratinocytes and stratum corneum formation is associated with caspase-14

activation." Journal of Investigative Dermatology 115(6): 1148-51.

Eis, A. L., D. E. Brockman, et al. (1995). "Immunohistochemical localization of

endothelial nitric oxide synthase in human villous and extravillous trophoblast

populations and expression during syncytiotrophoblast formation in vitro."

Placenta 16(2): 113-26.

Fernando, P., J. F. Kelly, et al. (2002). "Caspase 3 activity is required for skeletal

muscle differentiation." Proceedings of the National Academy of Science of the

USA 99(17): 11025-30.


135

Fire, A., S. Xu, et al. (1998). "Potent and specific genetic interference by double-

stranded RNA in Caenorhabditis elegans." Nature 391(6669): 806-11.

Fischer, H., H. Rossiter, et al. (2005). "Caspase-14 but not caspase-3 is processed

during the development of fetal mouse epidermis." Differentiation 73(8): 406-

13.

Fischer, H., M. Stichenwirth, et al. (2004). "Stratum corneum-derived caspase-14 is

catalytically active." FEBS Letters 577(3): 446-50.

Fox, H. and N. J. Sebire (2007). Pathology of the Placenta, Saunders.

Gandarillas, A. (2000). "Epidermal differentiation, apoptosis, and senescence: common

pathways?" Experimental Gerontology 35(1): 53-62.

Getsios, S., G. T. Chen, et al. (2000). "Regulation of beta-catenin mRNA and protein

levels in human villous cytotrophoblasts undergoing aggregation and fusion in

vitro: correlation with E-cadherin expression." J Reprod Fertil 119(1): 59-68.

Goldstein, J. C., C. Munoz-Pinedo, et al. (2005). "Cytochrome c is released in a single

step during apoptosis." Cell Death Differ 12: 453-462.

Gottardi, C. J. and B. M. Gumbiner (2004). "Distinct molecular forms of beta-catenin

are targeted to adhesive or transcriptional complexes." J Cell Biol 167(2): 339-

49.

Gude, N. M., C. T. Roberts, et al. (2004). "Growth and function of the normal human

placenta." Thromb Res 114(5-6): 397-407.

Gupta, A. K., C. Rusterholz, et al. (2005). "A comparative study of the effect of three

different syncytiotrophoblast micro-particles preparations on endothelial cells."

Placenta 26(1): 59-66.

Hargrove, J. L., M. G. Hulsey, et al. (1991). "The kinetics of mammalian gene

expression." Bioessays 13(12): 667-74.

He, B., A. Y. Lee, et al. (2005). "Secreted frizzled-related protein 4 is silenced by

hypermethylation and induces apoptosis in beta-catenin-deficient human

mesothelioma cells." Cancer Res 65(3): 743-8.

Hewitt, D. P., P. J. Mark, et al. (2006). "Placental expression of secreted frizzled related

protein-4 in the rat and the impact of glucocorticoid-induced fetal and placental

growth restriction." Biol Reprod 75(1): 75-81.

Hsu, S., D. Dickinson, et al. (2007). "Green tea polyphenol induces caspase 14 in

epidermal keratinocytes via MAPK pathways and reduces psoriasiform lesions

in the flaky skin mouse model." Exp Dermatol 16(8): 678-84.

Hsu, S., T. Yamamoto, et al. (2005). "Green tea polyphenol-induced epidermal

keratinocyte differentiation is associated with coordinated expression of

p57/KIP2 and caspase 14." J Pharmacol Exp Ther 312(3): 884-90.

Hu, S., S. J. Snipas, et al. (1998). "Caspase-14 is a novel developmentally regulated

protease." J Biol Chem 273(45): 29648-53.

Huang, K., C. Andersson, et al. (2001). "Signaling properties of VEGF receptor-1 and -

2 homo- and heterodimers." Int J Biochem Cell Biol 33(4): 315-24.

Huelsken, J. and J. Behrens (2002). "The Wnt signalling pathway." J Cell Sci 115(Pt

21): 3977-8.

Huppertz, B., H. G. Frank, et al. (1998). "Villous cytotrophoblast regulation of the

syncytial apoptotic cascade in the human placenta." Histochemistry and Cell

Biology 110(5): 495-508.

Huppertz, B., H. G. Frank, et al. (1999). "Apoptosis cascade progresses during turnover

of human trophoblast: analysis of villous cytotrophoblast and syncytial

fragments in vitro." Laboratory Investigation 79(12): 1687-702.

Huppertz, B., D. Hemmings, et al. (2005). "Extravillous trophoblast apoptosis--a

workshop report." Placenta 26 Suppl A: S46-8.


136

Huppertz, B., M. Kadyrov, et al. (2006). "Apoptosis and its role in the trophoblast." Am

J Obstet Gynecol 195(1): 29-39.

Huppertz, B., J. Kingdom, et al. (2003). "Hypoxia favours necrotic versus apoptotic

shedding of placental syncytiotrophoblast into the maternal circulation."

Placenta 24(2-3): 181-90.

Ishihara, N., H. Matsuo, et al. (2002). "Increased apoptosis in the syncytiotrophoblast in

human term placentas complicated by either preeclampsia or intrauterine growth

retardation." Am J Obstet Gynecol 186(1): 158-66.

Ishikawa, T., Y. Tamai, et al. (2001). "Mouse Wnt receptor gene Fzd5 is essential for

yolk sac and placental angiogenesis." Development 128(1): 25-33.

Ishizaki, Y., M. D. Jacobson, et al. (1998). "A role for caspases in lens fiber

differentiation." Journal of Cell Biology 140(1): 153-8.

James, J. L., P. R. Stone, et al. (2005). "Cytotrophoblast differentiation in the first

trimester of pregnancy: evidence for separate progenitors of extravillous

trophoblasts and syncytiotrophoblast." Reproduction 130(1): 95-103.

James, J. L., P. R. Stone, et al. (2005). "The regulation of trophoblast differentiation by

oxygen in the first trimester of pregnancy." Hum Reprod Update.

Jaubert, J., J. Cheng, et al. (2003). "Ectopic expression of kruppel like factor 4 (Klf4)

accelerates formation of the epidermal permeability barrier." Development

130(12): 2767-77.

Jerzak, M. and P. Bischof (2002). "Apoptosis in the first trimester human placenta: the

role in maintaining immune privilege at the maternal-foetal interface and in the

trophoblast remodelling." Eur J Obstet Gynecol Reprod Biol 100(2): 138-42.

Jia, J. J., Z. N. Wang, et al. (2006). "[Expression of vascular endothelial growth factor

in human placental trophoblasts and angiogenesis in the chorionic villi]." Nan

Fang Yi Ke Da Xue Xue Bao 26(4): 505-8.

Kadyrov, M., P. Kaufmann, et al. (2001). "Expression of a cytokeratin 18 neo-epitope is

a specific marker for trophoblast apoptosis in human placenta." Placenta 22(1):

44-8.

Kadyrov, M., J. C. Kingdom, et al. (2006). "Divergent trophoblast invasion and

apoptosis in placental bed spiral arteries from pregnancies complicated by

maternal anemia and early-onset preeclampsia/intrauterine growth restriction."

Am J Obstet Gynecol 194(2): 557-63.

Kam, D., A. K. Charles, et al. (2005). "Caspase-14 expression in the human placenta."

Reproductive BioMedicine Online 10(7): 236-243.

Kanayama, N., K. Takahashi, et al. (2002). "Deficiency in p57Kip2 expression induces

preeclampsia-like symptoms in mice." Mol Hum Reprod 8(12): 1129-35.

Karin, M., Z. Liu, et al. (1997). "AP-1 function and regulation." Curr Opin Cell Biol

9(2): 240-6.

Kerr, J. F., A. H. Wyllie, et al. (1972). "Apoptosis: a basic biological phenomenon with

wide-ranging implications in tissue kinetics." British Journal of Cancer 26(4):

239-57.

Kim, Y. J., H. S. Park, et al. (2006). "Reduced L-arginine level and decreased placental

eNOS activity in preeclampsia." Placenta 27(4-5): 438-44.

Kiss, H., C. Schneeberger, et al. (1998). "Expression of endothelial (type III) nitric

oxide synthase in cytotrophoblastic cell lines: regulation by hypoxia and

inflammatory cytokines." Placenta 19(8): 603-11.

Klaewsongkram, J., Y. Yang, et al. (2007). "Kruppel-like factor 4 regulates B cell

number and activation-induced B cell proliferation." J Immunol 179(7): 4679-

84.


137

Kliman, H. J., J. E. Nestler, et al. (1986). "Purification, characterization, and in vitro

differentiation of cytotrophoblasts from human term placentae." Endocrinology

118(4): 1567-82.

Kudo, Y., C. A. Boyd, et al. (2004). "An analysis using DNA microarray of the time

course of gene expression during syncytialization of a human placental cell line

(BeWo)." Placenta 25(6): 479-88.

Kuechle, M. K., H. M. Predd, et al. (2001). "Caspase-14, a keratinocyte specific

caspase: mRNA splice variants and expression pattern in embryonic and adult

mouse." Cell Death Differ 8(8): 868-70.

Lassila, M. (2000). Cyclosporine A-induced hypertension and nephrotoxicity in

spontaneously hypertensive rats on high-sodium diet. Faculty of Medicine,

Institute of Biomedicine, Department of Pharmacology and Toxicology.

Helsinki, University of Helsinki. PhD: 70.

Lee, M. H., I. Reynisdottir, et al. (1995). "Cloning of p57KIP2, a cyclin-dependent

kinase inhibitor with unique domain structure and tissue distribution." Genes

Dev 9(6): 639-49.

Leers, M. P., W. Kolgen, et al. (1999). "Immunocytochemical detection and mapping of

a cytokeratin 18 neo-epitope exposed during early apoptosis." J Pathol 187(5):

567-72.

Leung, D. N., S. C. Smith, et al. (2001). "Increased placental apoptosis in pregnancies

complicated by preeclampsia." Am J Obstet Gynecol 184(6): 1249-50.

Levy, R. and D. M. Nelson (2000). "To be, or not to be, that is the question. Apoptosis

in human trophoblast." Placenta 21(1): 1-13.

Levy, R., S. D. Smith, et al. (2000). "Apoptosis in human cultured trophoblasts is

enhanced by hypoxia and diminished by epidermal growth factor." Am J Physiol

Cell Physiol 278(5): C982-8.

Lickert, H., A. Bauer, et al. (2000). "Casein kinase II phosphorylation of E-cadherin

increases E-cadherin/beta-catenin interaction and strengthens cell-cell adhesion."

J Biol Chem 275(7): 5090-5.

Lippens, S., M. Kockx, et al. (2004). "Vitamin D3 induces caspase-14 expression in

psoriatic lesions and enhances caspase-14 processing in organotypic skin

cultures." Am J Pathol 165(3): 833-41.

Lippens, S., M. Kockx, et al. (2000). "Epidermal differentiation does not involve the

pro-apoptotic executioner caspases, but is associated with caspase-14 induction

and processing." Cell Death and Differentiation 7(12): 1218-24.

Lippens, S., C. VandenBroecke, et al. (2003). "Caspase-14 is expressed in the

epidermis, the choroid plexus, the retinal pigment epithelium and thymic

Hassall's bodies." Cell Death and Differentiation 10(2): 257-9.

Lyden, T. W., A. K. Ng, et al. (1993). "Modulation of phosphatidylserine epitope

expression by BeWo cells during forskolin treatment." Placenta 14(2): 177-86.

Manna, S. K., A. Mukhopadhyay, et al. (2000). "Human chorionic gonadotropin

suppresses activation of nuclear transcription factor-kappa B and activator

protein-1 induced by tumor necrosis factor." J Biol Chem 275(18): 13307-14.

Marieb, E. N. (1998). Human Anatomy and Physiology, Benjamin/Cummings Science

Publishing.

Marzioni, D., J. Muhlhauser, et al. (1998). "BCL-2 expression in the human placenta

and its correlation with fibrin deposits." Hum Reprod 13(6): 1717-22.

Matsuno, H., Y. Tanaka, et al. (2003). "Biopathways representation and simulation on

hybrid functional Petri net." In Silico Biol 3(3): 389-404.

Matsuoka, S., M. C. Edwards, et al. (1995). "p57KIP2, a structurally distinct member of

the p21CIP1 Cdk inhibitor family, is a candidate tumor suppressor gene." Genes

Dev 9(6): 650-62.


138

Maynard, S. E., J. Y. Min, et al. (2003). "Excess placental soluble fms-like tyrosine

kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and

proteinuria in preeclampsia." J Clin Invest 111(5): 649-58.

Meyer, M., M. Clauss, et al. (1999). "A novel vascular endothelial growth factor

encoded by Orf virus, VEGF-E, mediates angiogenesis via signalling through

VEGFR-2 (KDR) but not VEGFR-1 (Flt-1) receptor tyrosine kinases." EMBO

18(2): 363-374.

Mikolajczyk, J., F. L. Scott, et al. (2004). "Activation and substrate specificity of

caspase-14." Biochemistry 43(32): 10560-9.

Mogi, M. and A. Togari (2003). "Activation of caspases is required for osteoblastic

differentiation." Journal of Biological Chemistry 278(48): 47477-82.

Moncada, S., R. M. Palmer, et al. (1991). "Nitric oxide: physiology, pathophysiology,

and pharmacology." Pharmacol Rev 43(2): 109-42.

Monkley, S. J., S. J. Delaney, et al. (1996). "Targeted disruption of the Wnt2 gene

results in placentation defects." Development 122(11): 3343-53.

Morrish, D. W., Y. Kudo, et al. (2007). "Growth factors and trophoblast differentiation-

-workshop report." Placenta 28 Suppl A: S121-4.

Myat, T., Y. Hitsumoto, et al. (1996). "Attenuation of human chorionic gonadotropin

release by nitric oxide in choriocarcinoma cell lines." J Endocrinol 150(2): 243-

53.

Orange, S. J., D. Painter, et al. (2003). "Placental endothelial nitric oxide synthase

localization and expression in normal human pregnancy and pre-eclampsia."

Clin Exp Pharmacol Physiol 30(5-6): 376-81.

Orsulic, S., O. Huber, et al. (1999). "E-cadherin binding prevents beta-catenin nuclear

localization and beta-catenin/LEF-1-mediated transactivation." J Cell Sci 112 (

Pt 8): 1237-45.

Park, K., M. K. Kuechle, et al. (2006). "Expression and characterization of

constitutively active human caspase-14." Biochemical and Biophysical Research

Communications 347(4): 941-8.

Pattillo, R. A., G. O. Gey, et al. (1968). "Human hormone production in vitro." Science

159(822): 1467-9.

Peters, T. J., B. M. Chapman, et al. (2000). "Placental lactogen-I gene activation in

differentiating trophoblast cells: extrinsic and intrinsic regulation involving

mitogen-activated protein kinase signaling pathways." J Endocrinol 165(2): 443-

56.

Polakis, P. (1999). "The oncogenic activation of beta-catenin." Curr Opin Genet Dev

9(1): 15-21.

Poliseno, L., L. Bianchi, et al. (2004). "Bcl2-low-expressing MCF7 cells undergo

necrosis rather than apoptosis upon staurosporine treatment." Biochem J 379(Pt

3): 823-32.

Poranen, A. K., U. Ekblad, et al. (1996). "Lipid peroxidation and antioxidants in normal

and pre-eclamptic pregnancies." Placenta 17(7): 401-5.

Presland, R. B., P. V. Haydock, et al. (1992). "Characterization of the human epidermal

profilaggrin gene. Genomic organization and identification of an S-100-like

calcium binding domain at the amino terminus." J Biol Chem 267(33): 23772-

81.

Rao, R. M., A. M. Dharmarajan, et al. (2000). "Cloning and characterization of an

apoptosis-associated gene in the human placenta." Apoptosis 5(1): 53-60.

Rattner, A., J. C. Hsieh, et al. (1997). "A family of secreted proteins contains homology

to the cysteine-rich ligand-binding domain of frizzled receptors." Proc Natl Acad

Sci U S A 94(7): 2859-63.


139

Ratts, V. S., X. J. Tao, et al. (2000). "Expression of BCL-2, BAX and BAK in the

trophoblast layer of the term human placenta: a unique model of apoptosis

within a syncytium." Placenta 21(4): 361-6.

Rawlings, A. V. and C. R. Harding (2004). "Moisturization and skin barrier function."

Dermatol Ther 17 Suppl 1: 43-8.

Rendl, M., J. Ban, et al. (2002). "Caspase-14 expression by epidermal keratinocytes is

regulated by retinoids in a differentiation-associated manner." Journal of

Investigative Dermatology 119(5): 1150-5.

Ririe, K. M., R. P. Rasmussen, et al. (1997). "Product differentiation by analysis of

DNA melting curves during the polymerase chain reaction." Anal Biochem

245(2): 154-60.

Rossmanith, W. G., U. Hoffmeister, et al. (1999). "Expression and functional analysis

of endothelial nitric oxide synthase (eNOS) in human placenta." Mol Hum

Reprod 5(5): 487-94.

Roy, H., S. Bhardwaj, et al. (2006). "Biology of vascular endothelial growth factors."

FEBS Lett 580(12): 2879-87.

Saldanha, G., V. Ghura, et al. (2004). "Nuclear beta-catenin in basal cell carcinoma

correlates with increased proliferation." Br J Dermatol 151(1): 157-64.

Schumann, H., J. Holtz, et al. (2000). "Expression of secreted frizzled related proteins 3

and 4 in human ventricular myocardium correlates with apoptosis related gene

expression." Cardiovasc Res 45(3): 720-8.

Segre, J. A., C. Bauer, et al. (1999). "Klf4 is a transcription factor required for

establishing the barrier function of the skin." Nature Genetics 22(4): 356-60.

Shi, Q. J., Z. M. Lei, et al. (1993). "Novel role of human chorionic gonadotropin in

differentiation of human cytotrophoblasts." Endocrinology 132(3): 1387-95.

Shiozaki, A., K. Kataoka, et al. (2003). "Survivin inhibits apoptosis in

cytotrophoblasts." Placenta 24(1): 65-76.

Shtutman, M., J. Zhurinsky, et al. (1999). "The cyclin D1 gene is a target of the beta-

catenin/LEF-1 pathway." Proc Natl Acad Sci U S A 96(10): 5522-7.

Sikkema, J. M., B. B. van Rijn, et al. (2001). "Placental superoxide is increased in pre-

eclampsia." Placenta 22(4): 304-8.

Smiley, S. T., M. Reers, et al. (1991). "Intracellular heterogeneity in mitochondrial

membrane potentials revealed by a J-aggregate-forming lipophilic cation JC-1."

Proceedings of the National Academy of Science of the USA 88(9): 3671-5.

Smith, S. C. (2000). "Re: Apoptotic and proliferative activities in first trimester

placentae (Placenta (1999), 20, 223-227, C. C. W. Chan, T. T. Lao and A. N. Y.

Cheung)." Placenta 21(2-3): 286-8.

Smith, S. C., P. N. Baker, et al. (1997). "Placental apoptosis in normal human

pregnancy." Am J Obstet Gynecol 177(1): 57-65.

Sonderegger, S., H. Husslein, et al. (2007). "Complex expression pattern of Wnt ligands

and frizzled receptors in human placenta and its trophoblast subtypes." Placenta

28 Suppl A: S97-102.

St Croix, B., C. Sheehan, et al. (1998). "E-Cadherin-dependent growth suppression is

mediated by the cyclin-dependent kinase inhibitor p27(KIP1)." J Cell Biol

142(2): 557-71.

Steinhusen, U., J. Weiske, et al. (2001). "Cleavage and shedding of E-cadherin after

induction of apoptosis." J Biol Chem 276(7): 4972-80.

Stockinger, A., A. Eger, et al. (2001). "E-cadherin regulates cell growth by modulating

proliferation-dependent beta-catenin transcriptional activity." J Cell Biol 154(6):

1185-96.

Suto, K., Y. Yamazaki, et al. (2005). "Crystal structures of novel vascular endothelial

growth factors (VEGF) from snake venoms: insight into selective binding to


140

kinase insert domain-containing receptor but not to fms-like tyrosine kinase-1." J

Biol Chem 280(3): 2126-2131.

Takagi, Y., T. Nikaido, et al. (2004). "Levels of oxidative stress and redox-related

molecules in the placenta in preeclampsia and fetal growth restriction."

Virchows Arch 444(1): 49-55.

Tennyson, C. N., H. J. Klamut, et al. (1995). "The human dystrophin gene requires 16

hours to be transcribed and is cotranscriptionally spliced." Nat Genet 9(2): 184-

90.

Tilly, J. L. and A. J. Hsueh (1993). "Microscale autoradiographic method for the

qualitative and quantitative analysis of apoptotic DNA fragmentation." Journal

of Cell Physiology 154(3): 519-26.

Tulac, S., N. R. Nayak, et al. (2003). "Identification, characterization, and regulation of

the canonical Wnt signaling pathway in human endometrium." J Clin Endocrinol

Metab 88(8): 3860-6.

Van de Craen, M., G. Van Loo, et al. (1998). "Identification of a new caspase

homologue: caspase-14." Cell Death Differ 5(10): 838-46.

Walsh, D. S., J. L. Borke, et al. (2005). "Psoriasis is characterized by altered epidermal

expression of caspase 14, a novel regulator of keratinocyte terminal

differentiation and barrier formation." J Dermatol Sci 37(1): 61-3.

Wawrzak, D., M. Metioui, et al. (2007). "Wnt3a binds to several sFRPs in the

nanomolar range." Biochem Biophys Res Commun 357(4): 1119-23.

Weber, C. H. and C. Vincenz (2001). "The death domain superfamily: a tale of two

interfaces?" Trends Biochem Sci 26(8): 475-81.

White, L., A. Dharmarajan, et al. (2007). "Caspase-14: a new player in cytotrophoblast

differentiation." Reprod Biomed Online 14(3): 300-7.

Wice, B., D. Menton, et al. (1990). "Modulators of cyclic AMP metabolism induce

syncytiotrophoblast formation in vitro." Experimental Cell Research 186(2):

306-16.

Wolf, V., G. Ke, et al. (1997). "DDC-4, an apoptosis-associated gene, is a secreted

frizzled relative." FEBS Lett 417(3): 385-9.

Wong, S. C., S. F. Lo, et al. (2002). "Expression of frizzled-related protein and Wnt-

signalling molecules in invasive human breast tumours." J Pathol 196(2): 145-

53.

Yamada, S., S. Pokutta, et al. (2005). "Deconstructing the cadherin-catenin-actin

complex." Cell 123(5): 889-901.

Yang, M., Z. M. Lei, et al. (2003). "The central role of human chorionic gonadotropin in

the formation of human placental syncytium." Endocrinology 144(3): 1108-20.

Yoo, N. J., Y. H. Soung, et al. (2007). "Mutational analysis of caspase-14 gene in

common carcinomas." Pathology 39(3): 330-3.

Yusuf, K., S. D. Smith, et al. (2002). "Trophoblast differentiation modulates the activity

of caspases in primary cultures of term human trophoblasts." Pediatric Research

52(3): 411-5.

Zamzami, N., D. Metivier, et al. (2000). "Quantitation of mitochondrial transmembrane

potential in cells and in isolated mitochondria." Methods in Enzymology 322:

208-13.

Zygmunt, M., F. Herr, et al. (2003). "Angiogenesis and vasculogenesis in pregnancy."

Eur J Obstet Gynecol Reprod Biol 110 Suppl 1: S10-8.

Appendix I

141

Appendix I

Published Articles

Online - Vol 14. No 3. 2007 300-307 Reproductive BioMedicine Online; www.rbmonline.com/Article/2592 on web 15

January 2007

300 © 2007 Published by Reproductive Healthcare Ltd, Duck End Farm, Dry Drayton, Cambridge CB3 8DB, UK

Caspase-14: A New Player in Cytotrophoblast Differentiation

L. Whitea, A. Dharmarajan

a,b, A. Charles

c

a School of Anatomy and Human Biology, Faculty of Life and Physical Sciences, The

University of Western Australia, 35 Stirling Hwy Crawley, Perth, Western Australia

6009. b

Correspondence: Prof. A. Dharmarajan, School of Anatomy and Human Biology,

M309, The University of Western Australia, 35 Stirling Highway, CRAWLEY, Western

Australia, Australia, 6009; Tel: +61 8 64882981; e-mail: [email protected] c King Edward Memorial Hospital for Women (KEMH), 374 Bagot Rd, Subiaco,

Western Australia 6008.

Declaration

The authors would like to declare that all following information and data are the sole

work of the authors. The data herein is unpublished and not under consideration by any

other journal. All named authors have seen, commented upon, and authorised the

submission of the final manuscript draft. All relevant ethical guidelines have been

strictly adhered to.

Summary for Layreaders

The human placenta is responsible for the exchange of nutrients, gas and wastes

between the mother and the fetal circulations. Caspase-14 is a member of the cell death

(apoptosis) inducing family of caspase proteins; however it does not appear to be

involved in cell death. It is found in only a relatively few tissues of the body, most

notable the skin and the placenta. It is involved in the outer layer of skin development

which forms a barrier between the skin and the external environment. The outer layer of

the placenta is formed by trophoblast cells which differentiate, and fuse to become the

multinucleated syncytiotrophoblast, and these form a functional barrier between the

placenta and the maternal blood. We have shown previously that caspase 14 is present

in the trophoblast and so we propose that caspase-14 has a conserved role in cellular

differentiation and the fusion of trophoblast cells. BeWo is a cell line derived from

trophoblast tumour cells which can be induced to behave like trophoblast cells. Using

the BeWo cell line as a model of trophoblast formation we show in this paper that

mailto:[email protected]

Appendix I

142

caspase-14 is not involved in apoptosis of trophoblast cells. It does appear to be

involved in early trophoblast differentiation however, suggesting a role in barrier

formation. Therefore, caspase-14 may be involved in the pathways leading to the

formation of a functional barrier in the skin and placenta.

Abstract

The human placenta is responsible for the exchange of nutrients, gas and wastes through

the trophoblast maternal-fetal barrier, which is formed by the fusion of villous

cytotrophoblasts to form the continuous multinucleated syncytiotrophoblast separating

the maternal and fetal circulations. Caspase-14 is a seemingly non-apoptotic caspase

involved in keratinocyte differentiation and cornification. We propose that caspase-14

has a conserved role in cellular differentiation and a role in differentiation and fusion in

the trophoblast. The human choriocarcinoma BeWo cell line was treated with

Staurosporine and Forskolin to induce apoptosis and differentiation respectively.

Staurosporine initiated apoptosis within 3 hours of treatment, while apoptosis was

completed following 6 hours treatment. Caspase-14 gene and protein expression was

unchanged throughout this process. During BeWo differentiation, caspase-14 mRNA

was elevated after 48 hours Forskolin treatment, while its protein was increased after 24

hours. Therefore caspase-14 is upregulated during trophoblast differentiation as

represented by the BeWo cell line. Moreover, caspase-14 may interact with other

signalling molecules to facilitate differentiation. This new data confirms the potential

for the BeWo cell line in the functional dissection of this unusual caspase and its

prospective role in trophoblast differentiation.

Keywords

Caspase-14, Placenta, Trophoblast, BeWo, Differentiation, Apoptosis

Abbreviations

STS- Staurosporine; RT-PCR- Reverse Transcriptase Polymerase Chain Reaction;

hCG- human chorionic gonadotrophin; mRNA- messenger ribonucleic acid; IUGR-

intrauterine growth restriction; KLF4- Kruppel-like factor 4

Introduction

The human placenta is a multi-functional organ responsible for exchange of nutrients,

gas and wastes between mother and fetus. The trophoblast is responsible for this

exchange by forming a barrier between the maternal and fetal circulations through the

fusion of villous cytotrophoblasts to form the continuous multinucleated

syncytiotrophoblast, which structurally and functionally separates maternal and fetal

circulations.

As the link between maternal and fetal circulations, the syncytiotrophoblast

facilitates many cellular processes including absorptive, secretory, immune and

endothelial functions. Due to its versatility, an appreciation of barrier formation is

important for understanding mechanisms of maternal or fetal disorders of the

trophoblast, including preeclampsia and intrauterine growth restriction (IUGR).

Appendix I

143

To ensure the correct activity and functioning of all tissues, old, damaged or

unneeded cells are removed by apoptosis (Kerr et al. 1972). This is an active and rapid

process, however unlike necrosis is programmed and occurs without inflammation and

disruption to surrounding cells. Thus apoptosis acts as a necessary safe-guard against

aberrant growth, function and tumour formation within a tissue. Apoptosis can be

initiated by either intrinsic or extrinsic forces, and the cellular pathways governing this

process typically involve the caspase family of pro-apoptotic proteases.

Once formed, the syncytiotrophoblast is maintained by continual incorporation

of cytotrophoblasts into the syncytium, with old nuclei extruded from the apical surface

as syncytial knots. This process is considered a modified form of apoptosis (Huppertz et

al. 1998). Some pro-apoptotic caspases are up-regulated prior to and during fusion of

the cytotrophoblast with the overlying syncytium, indicating a link between apoptosis

and terminal trophoblast differentiation (Huppertz et al. 1998; Huppertz et al. 1999;

Yusuf et al. 2002; Black et al. 2004). Additionally, terminal differentiation of epithelia,

myoblasts, osteoblasts, lens fibres and spermatids also involve apoptotic mechanisms

(Ishizaki et al. 1998; Gandarillas 2000; Fernando et al. 2002; Arama et al. 2003; Mogi

et al. 2003), further indicating a conserved link between apoptosis and terminal

differentiation.

Caspase-14 is involved in terminal keratinocyte differentiation and subsequent

barrier formation; however unlike other caspases no classical apoptotic function for

caspase-14 has yet been determined (Eckhart et al. 2000; Lippens et al. 2000; Chien et

al. 2002; Rendl et al. 2002; Lippens et al. 2003; Fischer et al. 2004). We have recently

confirmed the presence of caspase-14 in the human placenta, however its function

within the trophoblast has not yet been elucidated (Kam et al. 2005). As such we

investigated whether caspase-14 is involved in either apoptosis or differentiation of the

cytotrophoblast and consequent barrier formation, using the BeWo cell line as a model

for trophoblast function. We hypothesise that caspase-14 possesses a conserved role in

barrier formation through involvement in terminal differentiation and fusion of the

cytotrophoblast.

Materials and Methods

The cytotrophoblastic BeWo cell line is a well characterised in vitro model for the

examination and delineation of a variety of trophoblast functions, and can be

manipulated to undergo either apoptosis or functional differentiation (Pattillo et al.

1968; Lyden et al. 1993; Das et al. 2004). BeWo cells were obtained from the American

Type Culture Collection (ATCC) and maintained in T-75 flasks containing Hams F12-K

Nutrient Mixture (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine

serum (FBS) and 0.5% penicillin-streptomycin. Cells were passaged by treatment with

trypsin for 5 minutes, and experiments conducted between passage 5 and 15. For

extractions, 5x105

cells were plated onto 6-well plates and allowed to incubate overnight

prior to treatment at Time Zero. Apoptosis was initiated through treatment with 1μM

Staurosporine (STS) (Sigma, St. Louis, USA) for 0, 1, 3 or 6 hours (Das et al. 2004).

Differentiation was induced by treatment with 20μM Forskolin (LC Laboratories,

Boston, USA) for up to 96 hours for extractions, and immunocytochemistry for E-

Appendix I

144

cadherin conducted every 24 hours (Wice et al. 1990; Coutifaris et al. 1991; Al-Nasiry

et al. 2006).

JC-1

Initiation of apoptosis was determined using the JC-1 technique (Smiley et al. 1991;

Zamzami et al. 2000). At the onset of apoptosis the mitochondrial membrane is rapidly

depolarised (Das et al. 2004). When the mitochondrial membrane is polarised the JC-1

dye (5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethyl-benzimidazolyl-carbocyanine iodide)

aggregates and fluoresces red. Upon depolarisation, JC-1 forms a green fluorescent

monomer, so the ratio of aggregated to monomeric JC-1 gives a quantitative

representation of the extent of mitochondrial membrane permeability. Briefly, cells

were grown at 1x105c/ml in 96-well plates for 24 hours. After either 1 or 3 hours of

1μM STS treatment, media was removed and 50μl of 2.5mM JC-1 (Molecular Probes T-

3168) dye diluted 1:75 in Hams F12-K added and incubated for 60 minutes. Positive

controls were conducted by the addition of 50μM FCCP (carbonyl cyanide p-

(trifluoromethoxy) phenylhydrazone). JC-1 was removed and cells washed with 5%

bovine serum albumin (BSA) in PBS. Plates were analysed using a FluoStar fluorescent

plate reader at 520nm (green) and 590nm (red). Raw values at 590nm were divided by

the raw values of the corresponding well at 520nm.

3‟-end Labelling of DNA

A major characteristic of late stage apoptosis is internucleosomal DNA fragmentation,

which upon electrophoresis and autoradiography, produce a typical DNA fragmentation

ladder (Tilly et al. 1993). Terminal transferase is used to bind radioactively-labelled

ddATP to the 3‟ ends of fragmented DNA. The total amount of incorporated radiolabel

can be quantitated and used to estimate the extent of apoptotic DNA fragmentation

(Tilly et al. 1993; Drake et al. 2003). To assess the presence and extent of apoptosis,

DNA was extracted from BeWo cells following 0, 3 or 6 hours incubation with 1μM

STS using DNA homogenisation buffer. 0.5μg of DNA was labelled with radioactive

[α32

P]-ddATP (Amersham Biosciences, Buckinghamshire, UK) and analysed by gel

electrophoresis and autoradiography. The extent of DNA fragmentation was compared

between control BeWo cells and those treated with 1μM STS.

Immunofluorescence

BeWo cells were grown on coverslips and treated with 20μM Forskolin for 0, 24, 48, 72

and 96 hours, with media changed every 24 hours. At each time point coverslips were

fixed and stained for the expression of E-cadherin, a key component of adherens

junctions. As E-cadherin is localised to the plasma membrane, staining enables

visualisation of syncytium-incorporated nuclei, and therefore the extent of cell fusion.

Briefly, cells were fixed with 4% paraformaldehyde in 1x PBS containing 0.6% Triton

X-100 for 20 minutes. Unspecific binding was then blocked with 10% goat serum in

Wash Buffer (1x PBS containing 1% BSA and 0.1% Triton X-100) for 60 minutes.

Coverslips were incubated overnight at 4ºC with anti-mouse E-cadherin (Sigma,

U3254) primary antibody (1:500), prior to addition of the secondary antibody solution

containing FITC-conjugated goat anti-mouse Alexa488 (Sigma, F2012) secondary

antibody (1:125) and Hoechst (Sigma, B2261) nuclear dye (1:500) overnight at 4ºC.

Coverslips were mounted onto slides and imaged using a fluorescent microscope at Cell

Central (UWA).

Appendix I

145

Quantitative Real Time Reverse Transcriptase Polymerase Chain Reaction (RT-PCR)

RNA was extracted using the TriReagent method (Chomczynski et al. 1987). Briefly,

supernatant debris and adherent cells were collected and treated with 1ml TriReagent

(MRC Inc, Cincinnati, USA) and chloroform added. Samples were spun, the upper

aqueous phase collected and isopropanol added. After overnight incubation at 4ºC,

samples were spun, washed with 75% ethanol and resuspended in RNase-free H2O.

RNA (2.1μg) was treated with DNase (Promega, Madison, USA) before undergoing

reverse transcription (RT) by heating to 25ºC for 10 minutes, 55ºC for 50 minutes and

70ºC for 15 minutes. Following this, cDNA were stored at -20ºC until required.

Five microlitres (μl) SybrIQ (BioRad, Hercules, USA) mixed with 1μl of each

primer, 2μl dH2O and 1μl cDNA was used for all quantitative Real Time RT-PCR

(Polymerase Chain Reaction) reactions conducted. Primers used were: Caspase-14L-

tgcacgtttattccacggta and Caspase-14R- tgctttggatttcagggttc (204bp); Caspase-8L-

aagcaaacctcggggatact and Caspase-8R- ggggcttgatctcaaaatga (164bp); Ki67L-

agtcagacccagtggacacc and Ki67R- tgctgccggttaagttctct (225bp); β-hCGL-

gcaccaaggatggagatgtt and β-hCGR- gcacagatggtggtgttgac (173bp); E-cadherinL-

tgcccagaaaatgaaaaagg and E-cadherinR- gtgtatgtggcaatgcgttc (200bp); KLF4L-

cccacacaggtgagaaacct and KLF4R- atgtgtaaggcgaggtggtc (169bp); and L-19L-

ctgaaggtcaaagggaatgtg and L-19R-ggacagagtcttgatgatctc (194bp). Each reaction was

denatured at 95ºC for 3 minutes and subsequently cycled 40 times at 95ºC for 1sec,

60ºC for 15secs and 72ºC for 5secs. Melt analysis was then conducted by raising the

temperature from 72ºC to 99ºC at 0.5ºC intervals for 3mins to ensure accumulation of

the desired product. All PCR data obtained was standardised against the expression of

L19.

Western Blotting

Protein was extracted from cells using Radioimmunoprecipitation (RIPA) buffer

containing 150mM NaCl, 50mM Tris-HCl, pH 7.5, 1% Triton X-100, 0.5% sodium

deoxycholate, 0.1% SDS, and 0.1mM Phenylmethylsulfonyl fluoride (PMSF). The

amount of protein was estimated using the Bradford assay (Bradford 1976). Thirty

micrograms of protein was denatured by heating to 95ºC for 5mins prior to separating

on a 12% SDS-polyacrylamide gel. This was then transferred to a nitrocellulose

membrane and stained with Ponceau. Membranes were blocked with 5% milk in 0.05%

TBS-T and incubated with anti-mouse caspase-14 primary antibody (1:333) (BD

Biosciences, Cat. No. 611511) in TBS-T overnight at 4ºC. Membranes were washed 3

times in TBS-T and incubated for 90mins with a goat anti-mouse IgG Alkaline

Phosphatase-conjugate (1:500) (Sigma, Cat. No. A3688), washed a further 3 times with

TBS-T and developed with an Alkaline Phosphatase Conjugate Substrate Kit (Bio-Rad,

Hercules, USA) as per manufacturers instructions. Membranes were imaged using the

Kodak ImageStation. Statistical significance for all experiments was determined using

2-tailed t-tests using Microsoft Excel, with statistical significance established at P<0.05.

All experiments stated in this paper were repeated (n≥6), and the results obtained were

reproducible and strengthened the data.

Results

Cytotrophoblast Apoptosis

Appendix I

146

One of the earliest events of apoptosis is mitochondrial membrane depolarisation. JC-1

provides an accurate measure of apoptosis by analysing the mitochondrial membrane

polarity, with depolarisation corresponding with apoptosis. Within 3 hours 1μM

Staurosporine treatment, mitochondrial membrane permeability had significantly

increased compared with controls (P<0.001) (Figure 1A), indicating the onset of

apoptosis.

To confirm completion of apoptosis, 3‟-end labelling was conducted on

Staurosporine treated BeWo. Three hour Staurosporine treatment did not induce late-

stage apoptosis as determined by DNA fragmentation; however after 6 hours classical

DNA fragmentation laddering was evident, and late stage apoptosis had been induced

(Figure 1B and C). Thus 1μM Staurosporine was sufficient to stimulate and complete

apoptosis in BeWo cells.

Caspase-14 mRNA levels remained stable throughout BeWo apoptosis (Figure

2), indicating that it is not transcriptionally altered during trophoblast apoptosis. This

reinforces the belief that caspase-14 acts independently of classical apoptosis pathways

(Eckhart et al. 2000; Lippens et al. 2000; Chien et al. 2002; Rendl et al. 2002; Lippens

et al. 2003; Fischer et al. 2004).

Cytotrophoblast Differentiation

The biochemical and morphological differentiation of the BeWo cells was induced by

the addition of the adenyl cyclase promoter Forskolin. Biochemical differentiation was

confirmed by Real Time PCR quantitation of β-hCG (beta- human chorionic

gonadotrophin) mRNA, the protein of which is actively secreted by the

syncytiotrophoblast with differentiation. Levels of β-hCG mRNA were elevated in

Forskolin-treated BeWo relative to time-matched controls within 24 hours, a feature

evident throughout the experiment (Figure 3A).

To confirm morphological differentiation, E-cadherin mRNA was quantified by

Real Time PCR. E-cadherin is a key protein in the formation and maintenance of

adherens junctions. With cellular fusion, the surface area is decreased in relation to the

number of nuclei, meaning that there will be a decreased amount of E-cadherin per

nucleus. Accordingly, E-cadherin mRNA levels decrease from 72 hours after treatment

(Figure 3B), indicating morphological differentiation.

Additionally, immunofluorescence for E-cadherin was conducted on control and

Forskolin-treated BeWo. After 48 hours incubation, E-cadherin strongly stained the

plasma membrane of control cultures (Figure 4). In Forskolin-treated cultures however,

E-cadherin staining was much more diffuse and no longer confined to the plasma

membrane. This indicates that initiation of morphological differentiation is occurring

from 48 hours Forskolin treatment, resulting in the formation of multinuclear syncytia.

After 48 hours of Forskolin treatment the level of caspase-14 mRNA is elevated

relative to 24 hours as assessed by Real Time PCR (Figure 5A). Indeed, the elevation in

caspase-14 precedes E-cadherin changes (intercellular fusion) and is associated with β-

hCG production, suggesting a role in the initiation of cytotrophoblast differentiation.

Western blotting revealed an increase in the expression of the full length

procaspase-14 after 24 hour hours of treatment with Forskolin; however no significant

changes in its expression were noted throughout the remainder of the experiment

Appendix I

147

(Figure 5B and C). Thus caspase-14 translation is transiently increased at the onset of

differentiation.

Caspase activation requires cleavage into large and small subunits; however the

antibody used in this study only detects full length caspase-14. Thus immunoblotting

using an antibody sensitive to the caspase-14 subunits is required to fully investigate the

expression of caspase-14.

It has been previously shown that activity of the apoptosis-initiating protease

caspase-8 is required for successful syncytialisation (Black et al. 2004). Surprisingly,

caspase-8 transcription was reduced from 48 hours of Forskolin treatment (Figure 6A).

This suggests that gene expression of early apoptosis markers is not involved in the

onset of differentiation; further indicating a role for caspase-14 independent of classical

apoptotic pathways.

In order to control for artificial results caused by proliferation, the extent of

proliferation was determined through the expression of the mitotic marker Ki67, which

is active from late G1 through to M phase of the cell cycle. As seen in Figure 6B, there

is no change in Ki67 transcription between control and differentiating cultures at any

time. The stable mitotic state from 48 hours after Forskolin treatment may be explained

by confluence of the cultures resulting in cell contact-inhibition of the cell cycle leading

to cell senescence. Prior to this however, Ki67 mRNA levels were elevated in both

control and Forskolin-treated cultures, suggesting that confluence and contact inhibition

had not yet been obtained. However, the stable expression of Ki67 between controls and

treated cells at 24 hours indicates that the onset of differentiation occurred regardless of

the mitotic state of the culture.

Keratinocyte differentiation and barrier formation is dependent on the

expression of the Kruppel-Like Factor 4 (KLF4) transcription factor (Segre et al. 1999;

Jaubert et al. 2003). It has been noted that caspase-14 contains a potential binding site

for KLF4 (Eckhart et al. 2000), and given the role of both in keratinocyte

differentiation, and the expression of KLF4 in the human placenta (Blanchon et al.

2001; Blanchon et al. 2006) it is attractive to speculate a synergy between the two in

trophoblast barrier formation. Accordingly, the expression of KLF4 was examined upon

differentiation of the BeWo cell line (Figure 6C). Interestingly, KLF4 transcription was

significantly upregulated at all time points following forskolin treatment (P<0.05); an

expression pattern replicating that of caspase-14. These data suggest a role for KLF4 in

trophoblast differentiation, possibly in conjunction with caspase-14.

Discussion

The BeWo cell line is derived from a human choriocarcinoma (Pattillo et al. 1968), thus

results described in this paper are not necessarily reflective of physiological trophoblast

differentiation. The BeWo cell line is however a widely used and well defined model

for trophoblastic function and activity, allowing the delineation of cytotrophoblast

differentiation and apoptosis (Wice et al. 1990; Lyden et al. 1993; Das et al. 2004).

Caspase-14 has a limited expression profile in adults; however it is involved in

human keratinocyte differentiation (Eckhart et al. 2000; Lippens et al. 2000; Chien et

al. 2002; Rendl et al. 2002; Fischer et al. 2004). In particular, it is upregulated in the

proliferative stratum granulosum, with expression increasing in apical layers coinciding

Appendix I

148

with the onset of terminal differentiation and cornification. Thus we hypothesised that

caspase-14 is involved in barrier formation of the trophoblast in addition to the

epidermis.

To date, caspase-14 has not been found to be involved in classical apoptosis

pathways. However it is suggested that in select tissues, particularly those involving

cellular fusion or barrier formation, terminal differentiation and apoptosis share

common signalling pathways (Ishizaki et al. 1998; Gandarillas 2000; Fernando et al.

2002; Arama et al. 2003; Lippens et al. 2003; Mogi et al. 2003; Chaturvedi et al. 2006).

Due to the homology of caspase-14 with pro-apoptotic proteases, it may represent an

indirect link between apoptosis and differentiation, with particular reference to barrier

formation (Lippens et al. 2003). To test this hypothesis, the cytotrophoblastic BeWo

cell line was used as an in vitro model of cytotrophoblast function, and induced to

undergo both apoptosis and differentiation. Caspase-14 mRNA levels remained stable

throughout classical apoptosis (Figure 5A), indicating a role for caspase-14 in the

cytotrophoblast independent of classical apoptosis. Indeed, our lab has previously

demonstrated caspase-14 to be expressed in the human placenta at term and the first

trimester in a pattern unrelated to free radical-mediated apoptosis (Kam et al. 2005).

These results therefore signify that caspase-14 is not involved in apoptosis within the

human villous trophoblast.

When BeWo cells were induced to undergo differentiation caspase-14 protein

expression was substantially increased after 24 hours treatment (Figure 5C). Terminal

differentiation as characterised by E-cadherin translocation from the plasma membrane,

did not begin until 48 hours after the induction of differentiation (Figure 4), so the

increase in procaspase-14 indicates that it is translationally upregulated at the initiation

of trophoblast differentiation, suggesting a role in biochemical differentiation of the

cytotrophoblast. Caspase-14 mRNA levels were also increased from 24 to 48 hours after

Forskolin treatment, further indicating a role in the biochemical differentiation of the

trophoblast (Figure 5A). The discrepancy between the increased transcription and the

stable protein expression from 48 hours onward could be attributed to an increased

turnover of procaspase-14, or an unknown post-transcriptional control mechanism.

Control cultures showed unusual changes in that there was an ongoing rise in

caspase-14 mRNA throughout the experiment. There was a concomitant rise in the

procaspase-14 protein levels from 72 hours in control cultures (Figure 5B and C). This

may represent a hitherto unknown role for caspase-14 in cell senescence as the cultures

were confluent at these times. Indeed in these control cultures an unknown double band

was detected by immunoblotting (Figure 5B), supporting potential modifications to

caspase-14 under senescent conditions. These modifications may however be due to in

vitro culture anomalies.

Terminal differentiation of keratinocytes results in the formation of a barrier

between the internal body and external environment which is essential for survival. The

involvement of caspase-14 in the formation of this barrier and that between mother and

fetus, coupled with its limited expression pattern indicate a conserved function for

caspase-14 in barrier formation. Caspase-14 is expressed in several cell types affiliated

with functional barriers, further supporting a role in barrier formation (Eckhart et al.

2000; Lippens et al. 2003).

Appendix I

149

While we have confirmed the expression of caspase-14 mRNA and protein

during early trophoblast differentiation, the functional properties of this protease remain

undetermined. Although caspase-14 is presumably purposely upregulated, there is no

evidence that it is active in this system. A functional role in barrier formation cannot be

confirmed without conducting substrate cleavage and activity assays; however no in

vivo human caspase-14 substrates have yet been identified, adding to the difficulty of

exploring its biological significance.

Despite the difficulty in examining caspase-14 activity, there may be a couple of

transcription factors containing potential caspase-14 binding sites (Eckhart et al. 2000).

One of these is Kruppel-Like Factor 4 (KLF4), which is essential for proper barrier

formation in the epidermis (Segre et al. 1999; Jaubert et al. 2003). Interestingly, KLF4

is also believed to be involved in the maintenance of the human placenta (Blanchon et

al. 2001; Blanchon et al. 2006). As such, we have identified KLF4 mRNA in the

cytotrophoblastic BeWo cell line, with a similar expression pattern to caspase-14

(Figures 5A and 6C). Furthermore, KLF4 contains a putative cleavage site for caspase-

14 (YHCD) (Park et al. 2006), suggesting an intriguing synergy between the two in

trophoblast barrier formation.

Caspase-14 is transcriptionally and translationally upregulated during early

trophoblast differentiation. To further assess the properties of caspase-14 in trophoblast

differentiation, additional analysis using antibodies against its cleaved subunits must be

undertaken to investigate potential activation. Additionally, the properties of the rising

caspase-14 expression levels in senescent cultures must be evaluated to determine

whether it is somehow involved in cell contact inhibition of proliferation. Also, work is

also required to assess the activity of this protease during this process, and the protein

interactions thereof.

In conclusion caspase-14 is upregulated with trophoblast differentiation,

suggesting a conserved role in human barrier formation. The exact nature of its purpose

remains unknown; however examination of caspase-14 cleavage, activity and

interactions should elucidate its full role. Further investigation into the formation of the

maternal-fetal barrier has important implications for the understanding of diseases in

which differentiation may be disrupted, such as Preeclampsia and Intrauterine Growth

Restriction (IUGR).

References

Al-Nasiry, S, B Spitz, M Hanssens, et al. (2006) Differential effects of inducers of

syncytialization and apoptosis on BeWo and JEG-3 choriocarcinoma cells.

Human Reproduction 21, 193-201.

Arama, E, J Agapite and H Steller (2003) Caspase activity and a specific cytochrome C

are required for sperm differentiation in Drosophila. Developmental Cell 4, 687-

97.

Black, S, M Kadyrov, P Kaufmann, et al. (2004) Syncytial fusion of human trophoblast

depends on caspase 8. Cell Death and Differentiation 11, 90-8.

Blanchon, L, JL Bocco, D Gallot, et al. (2001) Co-localization of KLF6 and KLF4 with

pregnancy-specific glycoproteins during human placenta development.

Mechanisms of Development 105, 185-9.

Appendix I

150

Blanchon, L, R Nores, D Gallot, et al. (2006) Activation of the human pregnancy-

specific glycoprotein PSG-5 promoter by KLF4 and Sp1. Biochemical and

Biophysical Research Communications 343, 745-53.

Bradford, MM (1976) A rapid and sensitive method for the quantitation of microgram

quantities of protein utilizing the principle of protein-dye binding. Analytical

Biochemistry 72, 248-54.

Chaturvedi, V, LA Sitailo, B Bodner, et al. (2006) Defining the caspase-containing

apoptotic machinery contributing to cornification in human epidermal

equivalents. Experimental Dermatology 15, 14-22.

Chien, AJ, RB Presland and MK Kuechle (2002) Processing of native caspase-14 occurs

at an atypical cleavage site in normal epidermal differentiation. Biochemical and


Chomczynski, P and N Sacchi (1987) Single-step method of RNA isolation by acid

guanidinium thiocyanate-phenol-chloroform extraction. Analytical Biochemistry

162, 156-9.

Coutifaris, C, LC Kao, HM Sehdev, et al. (1991) E-cadherin expression during the

differentiation of human trophoblasts. Development 113, 767-77.

Das, M, B Xu, L Lin, et al. (2004) Phosphatidylserine efflux and intercellular fusion in

a BeWo model of human villous cytotrophoblast. Placenta 25, 396-407.

Drake, JM, RR Friis and AM Dharmarajan (2003) The role of sFRP4, a secreted

frizzled-related protein, in ovulation. Apoptosis 8, 389-97.

Eckhart, L, J Ban, H Fischer, et al. (2000) Caspase-14: analysis of gene structure and

mRNA expression during keratinocyte differentiation. Biochemical and


Eckhart, L, W Declercq, J Ban, et al. (2000) Terminal differentiation of human

keratinocytes and stratum corneum formation is associated with caspase-14

activation. Journal of Investigative Dermatology 115, 1148-51.

Fernando, P, JF Kelly, K Balazsi, et al. (2002) Caspase 3 activity is required for skeletal

muscle differentiation. Proceedings of the National Academy of Science of the

USA 99, 11025-30.

Fischer, H, M Stichenwirth, M Dockal, et al. (2004) Stratum corneum-derived caspase-

14 is catalytically active. FEBS Letters 577, 446-50.

Gandarillas, A (2000) Epidermal differentiation, apoptosis, and senescence: common

pathways? Experimental Gerontology 35, 53-62.

Huppertz, B, HG Frank, JC Kingdom, et al. (1998) Villous cytotrophoblast regulation

of the syncytial apoptotic cascade in the human placenta. Histochemistry and

Cell Biology 110, 495-508.

Huppertz, B, HG Frank, F Reister, et al. (1999) Apoptosis cascade progresses during

turnover of human trophoblast: analysis of villous cytotrophoblast and syncytial

fragments in vitro. Laboratory Investigation 79, 1687-702.

Ishizaki, Y, MD Jacobson and MC Raff (1998) A role for caspases in lens fiber

differentiation. Journal of Cell Biology 140, 153-8.

Jaubert, J, J Cheng and JA Segre (2003) Ectopic expression of kruppel like factor 4

(Klf4) accelerates formation of the epidermal permeability barrier. Development

130, 2767-77.

Appendix I

151

Kam, D, AK Charles and AM Dharmarajan (2005) Caspase-14 expression in the human

placenta. Reproductive BioMedicine Online 10, 236-243.

Kerr, JF, AH Wyllie and AR Currie (1972) Apoptosis: a basic biological phenomenon

with wide-ranging implications in tissue kinetics. British Journal of Cancer 26,

239-57.

Lippens, S, M Kockx, M Knaapen, et al. (2000) Epidermal differentiation does not

involve the pro-apoptotic executioner caspases, but is associated with caspase-14

induction and processing. Cell Death and Differentiation 7, 1218-24.

Lippens, S, C VandenBroecke, E Van Damme, et al. (2003) Caspase-14 is expressed in

the epidermis, the choroid plexus, the retinal pigment epithelium and thymic

Hassall's bodies. Cell Death and Differentiation 10, 257-9.

Lyden, TW, AK Ng and NS Rote (1993) Modulation of phosphatidylserine epitope

expression by BeWo cells during forskolin treatment. Placenta 14, 177-86.

Mogi, M and A Togari (2003) Activation of caspases is required for osteoblastic

differentiation. Journal of Biological Chemistry 278, 47477-82.

Park, K, MK Kuechle, Y Choe, et al. (2006) Expression and characterization of

constitutively active human caspase-14. Biochemical and Biophysical Research

Communications 347, 941-8.

Pattillo, RA, GO Gey, E Delfs, et al. (1968) Human hormone production in vitro.

Science 159, 1467-9.

Rendl, M, J Ban, P Mrass, et al. (2002) Caspase-14 expression by epidermal

keratinocytes is regulated by retinoids in a differentiation-associated manner.

Journal of Investigative Dermatology 119, 1150-5.

Segre, JA, C Bauer and E Fuchs (1999) Klf4 is a transcription factor required for

establishing the barrier function of the skin. Nature Genetics 22, 356-60.

Smiley, ST, M Reers, C Mottola-Hartshorn, et al. (1991) Intracellular heterogeneity in

mitochondrial membrane potentials revealed by a J-aggregate-forming lipophilic

cation JC-1. Proceedings of the National Academy of Science of the USA 88,

3671-5.

Tilly, JL and AJ Hsueh (1993) Microscale autoradiographic method for the qualitative

and quantitative analysis of apoptotic DNA fragmentation. Journal of Cell

Physiology 154, 519-26.

Wice, B, D Menton, H Geuze, et al. (1990) Modulators of cyclic AMP metabolism

induce syncytiotrophoblast formation in vitro. Experimental Cell Research 186,

306-16.

Yusuf, K, SD Smith, Y Sadovsky, et al. (2002) Trophoblast differentiation modulates

the activity of caspases in primary cultures of term human trophoblasts.

Pediatric Research 52, 411-5.

Zamzami, N, D Metivier and G Kroemer (2000) Quantitation of mitochondrial

transmembrane potential in cells and in isolated mitochondria. Methods in

Enzymology 322, 208-13.

Acknowledgements

Appendix I

152

This study was supported by grants from the Raine Medical Research Foundation,

Perth, Western Australia. We thank Ms Sue Hisheh and Mr. Greg Cozens for their

technical assistance.

Figures and Legends

Figure 1. Confirmation of BeWo Apoptosis. (A) JC-1 data showing a significant

increase in apoptosis following 3 hours treatment with STS compared with vehicle

controls (* = P<0.001). FCCP is a positive control. (B) 3‟-end labelling showing the

200bp fragments indicative of apoptosis. i) shows 3h control; ii) shows 3h STS; iii)

shows 6h control; and iv) shows 6h STS. (C) Quantitation of 6h DNA fragmentation.

Figure 2. Caspase-14 mRNA during BeWo Apoptosis. Expression of caspase-14

mRNA in BeWo cells 1, 3 and 6 hours after STS treatment.

Figure 3. Confirmation of BeWo Differentiation. Graphical representation of

quantified (A) β-hCG and (B) E-cadherin mRNA in Forskolin-treated BeWo cells,

standardised against L19 expression (* = P<0.05).

Appendix I

153

Figure 4. Confirmation of Morphological Differentiation. Immunofluorescence for

E-cadherin in 48 hour (A-C) Control and (D-F) Forskolin treated BeWo cells. A and D

show Hoechst labelled nuclei; B and E show E-cadherin staining; C and F show

overlays. Arrows indicated membrane-associated E-cadherin. Arrowheads indicate

diffuse E-cadherin. Original Magnifications: x400

Figure 5. Caspase-14 Expression during BeWo Differentiation. (A) Real Time PCR

analysis of caspase-14 mRNA following Forskolin treatment. (B) Western blot for

procaspase-14 following Forskolin treatment. (C) Graph of procaspase-14 expression

during differentiation standardised against β-actin (* = P<0.05).

Appendix I

154

Figure 6. Transcriptional Regulation during BeWo Differentiation. Graphical

representation of (A) caspase-8; (B) Ki67; and (C) KLF4 mRNA expression following

treatment of BeWo cells with Forskolin, standardised against L19 (* = P<0.05).

Received 12 September 2006; refereed 5 October 2006; accepted 6

December 2006.

307

Article - Caspase-14 and cytotrophoblast differentiation - L White et al.

RBMOnline®

Appendix I

155

The expression of secreted Frizzled Related Protein 4 (sFRP4) in the Primate

Placenta

White L1 , Suganthini Ganapathipillai

2, Friis R

2, Dharmarajan A

1, Charles A

3

1School of Anatomy and Human Biology. The Faculty of Life and Physical Sciences,

The University of Western Australia, 35 Stirling Highway, CRAWLEY 6009, Western

Australia. 2Department of Clinical Research, Faculty of Medicine, University of Berne,

Tiefenaustrasse 120, 3004, Berne. 3School of Women and Infants Health, University of Western Australia. School of

Women's and Infants' Health, King Edward Memorial Hospital, 374 Bagot Road

Subiaco WA 6008.

Abstract

BACKGROUND: Secreted Frizzed-Related Protein 4 (sFRP4) blocks the Wnt

signalling pathway by competitively binding Wnt ligands (Frizzled receptors). This

pathway is important during development and oncogenesis. It is however complex with

a large number of interacting proteins, isoforms and receptors. The Wnt signalling

pathway has a role in human placental development and implantation, particularly in the

trophoblast. Humans and macaque monkeys exhibit a similar remodelling of the

decidual spiral arteries. METHODS: We have examined the expression of sFRP4 in

human and macaque placentas of different gestational ages by immunohistochemistry,

in situ hybridisation, real time PCR, and Western blotting. RESULTS: We demonstrate

that sFRP4 is expressed predominantly in the villous syncytiotrophoblast and the

invasive intermediate cytotrophoblast, and in the amnion. CONCLUSIONS: These

observational studies suggest that sFRP4 has a role in placental development and

implantation, and may be an important factor in the development of the decidual

fibrinoid zone, and in trophoblast apoptosis and a band of apoptosis in the underlying

decidua deep to the trophoblast.

Summary for Lay-Readers

sFRP4 is a protein that appears to have an important role in the reproductive tissues. It

appears to be involved in cell processes related to programmed cell death and also to

function of cells lining and forming blood vessels. The human and other primate

placentas have a similar structure, but are different from most other animals. The outer

layer of the placenta, called the trophoblast has many functions, including blood vessel

lining function. We have examined human and monkey placentas to examine the

expression of sFRP4, and found it in the trophoblast and also in a zone under the

placenta where the placenta implants into the uterus.

Keywords

Placenta; secreted Frizzled-related Protein 4 (sFRP4); Apoptosis; Primate

Introduction

Appendix I

156

The placenta is responsible for the exchange of nutrients, gases and wastes between the

mother and developing fetus. There are large differences in the structure of the placenta

between different species (Carter, 2007) and the human and primate placenta share a

haemochorial structure, where the outer trophoblast layer invades the maternal tissues

and forms a continuous barrier between the villi and the surrounding maternal blood.

The chorionic villi are covered by the two layers of trophoblast, the inner

cytotrophoblast which gives rise to the overlying syncytiotrophoblast, which forms a

continuous multinucleated syncytium and facilitates exchange of gases and nutrients

between mother and fetus, as well as various hormonal and endothelial functions

(Reviewed in (Gude, 2004). In addition to the villous trophoblast described above,

extravillous trophoblast migrates into the uterus and anchors the placenta to the

maternal tissues (James, 2005), (although this invasion is deeper in the human than in

the macaque) and also occludes the uterine spiral arteries to facilitate the hypoxic

environment suited for placental invasion( reviewed in (Huppertz, 2005)). Inadequate

invasion or spiral artery remodelling may lead to preeclampsia or intrauterine growth

retardation. The trophoblast functions as an endothelium in the superficial portions of

the remodelled spiral arteries (Enders, 1991), as well as lining the maternal blood space

of the placenta and covering the villi. The maturation of the cytotrophoblast to become

syncytiotrophoblast appears to involve some elements of the apoptotic pathway (Black,

2004) .

sFRP4 is a member of a family of secreted proteins sharing homology with the

extracellular domain of Frizzled proteins (Wolf, 1997). It antagonizes Wnt-signalling by

competitively binding Wnts and their receptors in the plasma membrane (Rattner, 1997;

Wong, 2002; He, 2005) . This leads to phosphorylation and subsequent degradation of

cytosolic β-catenin, reducing the rate of cellular proliferation and increasing the

incidence of apoptosis (Schumann, 2000; He, 2005). Disruption to sFRP4 may therefore

be involved in uncontrolled proliferation and tumourigenesis (He, 2005). Several studies

have shown physiological expression of sFRP4 in the human endometrial cycle (Abu-

Jawdeh, 1999; Tulac, 2003), as well as in the ovary and corpus luteum, the human

prostate, and mammary gland (Lacher, 2003) where the gene was first identified

(Bielke, 1997) . SFRP4 is found to be down regulated in some tumours such as

endometrial sarcoma (Hrzenjak, 2004) . Endothelia also express sFRP4 suggesting an

important role in vascular development.

A recent study has demonstrated the complex number of Wnt pathway ligands

and receptors in the placenta, with 14 of the known Wnt ligands and 8 of the 10 Frizzled

receptors expressed in the human placenta, suggesting the Wnt pathway is important in

placental development (Sonderegger, 2007). A previous study showed increased

expression of sFRP4 associated with reduced placental growth in normal placentas in

the rat. Increased sFRP4 was also seen following glucocorticoid induced growth

restriction and this appeared to be mediated through alterations in the Wnt pathway

(Hewitt, 2006). The expression pattern in the human or primate placenta has not yet

been described.

Human placental studies are usually limited to the examination of the placenta

without the accompanying uterus, and examination of the implantation site during early

pregnancy is limited to abnormal pregnancies. The availability of such material from the

Appendix I

157

macaque monkey allowed the examination of the underlying uterine tissues including

decidua and spiral arteries at different gestational ages.

Because sFRP4 expression has been frequently observed in conjunction with

developmental, physiological apoptosis in reproductive tissues (Guo, 1998; Drake,

2003; Lacher, 2003) and particularly in rat placental development (as above), we

reasoned that investigation of sFRP4 expression at the functional interface of placenta

and decidua ought to be informative because of the importance of apoptosis in this

developmental process. Apoptosis is also involved with the extrusion of nuclei from the

syncytiotrophoblast. Furthermore, because a better understanding of apoptosis during

the placental invasion may shed light on implantation, we wanted to examine a primate

system to be able to examine the deeper parts of the implantation site, which are not

readily available on routine human samples. Hence, we studied a set of formaldehyde-

fixed, paraffin-embedded placental implantation samples from pregnant cynomolgus

monkeys (Macaca fascicularis) ranging from 22 days gestation until 165 days (term).

Methods

Immunohistochemistry

As soon as possible after delivery, samples of first trimester (12 weeks gestation age)

and term human placentas (samples taken as part of approved study reviewed by KEMH

ethics committee) were either snap frozen or fixed in formalin for later analysis.

Placental cDNA samples from mothers with preeclampsia were provided from a

previous study. Immunohistochemistry was performed on 2 cases each of first trimester

and term placenta using the method outlined below.

Animals

Primate samples were obtained from cynomolgus macaques (Macaca fascicularis) at

different stages after implantation as described previously . Macaque samples were the

generous gift of Dr. Alen Enders, Davis, California received as paraffin blocks of tissue

which had been fixed with 4% formaldehyde in 0.1 M phosphate buffer, pH 7.3 for 4

hrs and washed overnight in cold 0.1 M phosphate buffer prior to embedding .

Immunohistochemical detection was done as previously described . Rabbit

affinity-purified anti-sFRP4 serum was prepared in our laboratory as previously

described . An antibody specific for Cytokeratin 18 was obtained from Santa Cruz

Biotechnology (Santa Cruz, CA., USA). For detection of activated (cleaved) Caspase-3,

antibody specific for the cleaved Caspase-3 (asp175) was obtained from Cell Signaling

Technology (Beverly, MA.). All antibodies were used at a dilution of 1:100 in Tris

buffered saline (Tris 50 mM, NaCl 150 mM, pH 7.5) containing 5% bovine serum

albumin (Sigma). Antigen re-capture was performed on all tissue sections using

microwave treatment in 10 mM sodium citrate, pH 7.0 for 5 minutes at reduced energy

(maximum temperature, 95°C). Antibody treatment was overnight at 4ºC, followed by

incubation with a second antibody labelled with biotin (Dako) and directed against

rabbit IgG. Detection was performed using avidin-biotin-complex (Dako) coupled to

horseradish peroxidase. Di-amino benzidine was used as substrate.

Terminal Deoxynucleotidyl Transferase-Mediated dUTP Nick End-Labeling (TUNEL)

The TUNEL Reaction was performed as previously described using BODIPY-dUTP

(Molecular Probes, Eugene, Oregon). The fluorescent reaction product was visualized

Appendix I

158

directly in the fluorescence microscope. For the TUNEL, optimization of the proteinase

K pre-treatment proved crucial and was done with 5 ug/ml recombinant proteinease K

(Roche Diagnostics) in Tris (20 mM, pH 7.8) buffer containing 5 mM ethylene diamine

tetra-acetic acid (EDTA) at 30°C for 15 minutes.

Preparation of Macaque cDNA

These experiments used tissue samples from the macaque monkeys preserved for

approximately 10 years as paraffin blocks . Our initial effort was to isolate RNA from

thick sections obtained from these blocks, to destroy contaminating DNA, and to

prepare cDNA allowing us a semi-quantitative estimate of expression at different times

of gestation. To do this, cDNA was prepared using 2 first-strand primers together in the

same tube of a reverse transcriptase reaction (see below): primers for sFRP4 and for

glyceraldehyde phosphate dehydrogenase (GAPdH), as a house-keeping gene standard.

Total RNA was prepared from paraffin embedded tissue samples using the “Absolutely

RNA FFPE Kit” of Stratagene (http://www.stratagene.com). This kit employs d-

limonene for de-paraffinization, followed by alcohol washes, and proteinase K digestion

of the fixed tissue to release nucleic acids. DNA and RNA were solubilized in a

guanidine thiocyanate-containing buffer, filtered and specifically bound to the silica-

based fiber matrix of a micro-column. DNA is then removed by an exhaustive on-

column digestion with DNAse I. The immobilized RNA is washed and finally eluted

from the column in 30 ul aqueous buffer. The yields from 5 sections of approximately

20 µ each was approximately 4 ug according to the Nanodrop instrument (Nanodrop

Technologies).

Semi-Quantitative Reverse Transcriptase Polymerase Chain Reaction

Semi-quantitative polymerase chain reactions (sQ PCR) were carried out by firstly

performing a reaction using Transcriptor Reverse Transcriptase (Roche Diagnostics)

with 1 ug total RNA and two primers simultaneously in each reaction: anti-sense for

sFRP4: (5‟-CTT GTC CTG AAT TGT TCT CTG CTG-3‟) and anti-sense for

glyceraldehyde phosphate dehydrogenase (GAPdH: (5‟-GGG CCA TCC ACA GTC

TTC T-3‟)). Subsequently, the sQ PCR was performed with Taq DNA Polymerase

(Roche Diagnostics) using separate aliquots of the reverse transcriptase reaction in

individual reactions. The primers for sFRP4 were: sense (5‟-AAG CCC TGA TCG

GTG CAA GTG-3‟) and anti-sense (5‟- CTA AGC AAT TTT CAA GAA GCA TCA

TCC-3‟). For GAPdH, the primers were: sense (5‟-TTC ACC ACC ATG GAG AAG

GC-3‟) and anti-sense (5‟-GGC ATG GAC TGT GGT CAT GAG-3‟). Following the

reaction, the products were analysed by electrophoresis on agarose gels, scanned and

quantified by densitometry after ethidium bromide staining.

Quantitative Real Time Polymerase Chain Reaction

To quantitate the relative expression of sFRP4 mRNA between normal first trimester,

normal term, and those complicated with preeclampsia, cDNA samples were prepared

using random primers and analysed by Real Time RT-PCR using a Corbett Rotorgene

Light Cycler, and data standardized against the endogenous control L19. Reactions were

made up with 5μl SybrIQ (BioRad, Hercules, USA), 1μl of each primer, 2μl dH2O and

1μl cDNA. The sFRP4-specific primers used were: sFRP4L- 5‟-gcctgggacagcctatgtaa-

3‟ and sFRP4R- 5‟-tctgtaccaaagggcaaacc-3‟ (160bp); and for the L-19 housekeeping

gene, L-19L- ctgaaggtcaaagggaatgtg and L-19R-ggacagagtcttgatgatctc (194bp). The

Appendix I

159

cycling conditions for sFRP4 were denaturation at 95ºC for 5 minutes, and subsequent

cycling (45 times) at 95ºC for 1sec, 60ºC for 15secs and 72ºC for 10secs. Melt analysis

was then conducted by raising the temperature from 72ºC to 95ºC at 0.5ºC intervals

over 5mins to ensure amplification of the desired product. All PCR data obtained was

standardised against the expression of L19, which following denaturation for 3 minutes

at 95ºC, was cycled 45 times at 95ºC for 1sec, 51ºC for 15secs and 72ºC for 10secs.

Results

Macaque Placenta samples

sQ-PCR analysis of macaque placentas demonstrates that sFRP4 mRNA is produced

throughout gestation, gradually increasing in amount from 22 days (Fig 1).

Immunohistochemistry was therefore performed to demonstrate if sFRP4 protein

expression can be detected, and in association with which cells. Serial tissue sections of

22 day gestation were stained with anti-cytokeratin 18 antibody specific for

trophoblastic cells and also for glandular epithelium of the endometrium and anti-sFRP4

antibody. The result shows a widespread, but modest sFRP4-staining of endometrial

cells, with more intense staining marking the invading trophoblastic cells lining small

blood vessels. At 30 days of gestation the intravascular trophoblast is stained positively

for sFRP4 by immunohistochemistry (Fig. 2A, 2B) and in situ hybridisation (Fig 2C)

and the nuclei of the intravascular trophoblast appear to be extruded (Fig. 2B). Strong

sFRP4 staining of the intravascular trophoblast is also seen at 45 days gestation, (Fig

3A) and to a lesser extent in the trophoblastic shell (Fig 3B).

As the deeper parts of the implantation site are available for examination in the

samples from the monkey (unlike the human apart from rare hysterectomy specimens),

the zone of apoptosis and future fibrinoid change was examined. Fig. 4 illustrates the

region of apoptosis where placental villi have point of contact to the endometrium at 30

days gestation. This corresponded with the staining pattern with antibody specific for

activated (cleaved) caspase-3, a marker for apoptosis. The blood vessels occluded with

trophoblast that also showed sFRP4 staining also show increased TUNEL staining (Fig.

4C).

Human Placenta samples

The trophoblast of both first trimester and term human placentas express sFRP4 as

shown by immunohistochemistry. It is uniformly present on the first trimester chorionic

villi, largely confined to the syncytiotrophoblast and not the underlying cytotrophoblast

(Fig. 5). Within the decidua the infiltrating cytotrophoblast cells are positive for sFRP4,

while some maternal vessels contain a few sFRP4-positive trophoblast cells. Within the

mesenchymal core of the villi are scattered positive cells, which appear to be

macrophages (Hofbauer cells). In accordance with the anti-proliferative role of sFRP4,

the large proliferative areas of intermediate trophoblast in the first trimester do not

produce sFRP4. In the term villi again the syncytiotrophoblast is positive (Fig 6a). The

smaller blood vessels within the villi are also negative. Larger fetal vessels in the

chorionic plate show some sFRP4-positive staining of the endothelial cells and also of

some of the smooth muscle (Fig 6b). In the term villi the staining is less uniform, but

again confined to the syncytiotrophoblast. It appears stronger at sites of syncytial knots.

The amnion cells also shows cytoplasmic positivity (Figure 6c).

Appendix I

160

Real time RT-PCR demonstrates mRNA for sFRP4 is present in samples of first

trimester, term and preeclamptic human placentas. There appear to be no changes in

sFRP4 mRNA levels over pregnancy, supporting the macaque data presented above.

Conclusions

Previous studies have shown that sFRP4 has anti-proliferative, pro-apoptotic and also

pro-angiogenic properties (Guo, 1998; Drake, 2003; Lacher, 2003) . The consistent

expression of sFRP4 in three species (rat, macaque and human) in the trophoblast of the

placenta suggests that it has an important, evolutionary conserved function.

There appears to be an important role for sFRP4 during implantation with

invasion of the intermediate trophoblast into the decidua and the consequent

remodelling of the maternal spiral arteries. Expression of sFRP4 is seen throughout

gestation and strongly in the invasive intermediate trophoblast. The zone of apoptosis

preceding the invasive, sFRP4-secreting intermediate trophoblast suggests that sFRP4 is

a mediator of the zone of apoptosis. The zone appears to correlate with the fibrinoid

zone seen in the placental bed later in pregnancy, which helps form the cleavage plane

when the placenta separates at the time of delivery. The function of sFRP4 in

angiogenesis also suggests there is a role in angiogenesis associated with placental

development. Most interesting, was the intense expression of sFRP4 by trophoblast

elements occluding maternal blood vessels. These occluding elements showed clear

extrusion of “apoptotic” nuclei out of trophoblast. Clusters of apoptotic trophoblast cells

were also apparent inside the maternal vessels as shown with TUNEL assays.

The present investigation has been performed with placenta taken from two

primates, the macaque and human. The detailed relationships between fetal tissues of

placenta and maternal uterine tissues can vary greatly according to the species, but in

animals such as macaques and humans, a hemochorial placenta is formed (Mossman,

1987). Study of sFRP4 expression at the functional interface of placenta and decidua

ought to be informative because of the importance of apoptosis in decidualization.

Furthermore, a better understanding of the process of apoptosis during placental

invasion/decidualization of the endometrium may shed light on the underlying causes of

preeclampsia. These findings are also of interest with the observations by our group of

increased expression in the rat placenta where increased sFRP4 is associated with

reduced growth (Hewitt, 2006), as this may be due to effects of sFRP4 on cell growth

and vascular development. The anti-angiogenic role of sFRP4 may be a mechanism for

the reduction in placental size and subsequent growth restriction.

In the placenta sFRP4 was seen in some of the endothelial cells and smooth

muscle of the larger fetal vessels in the chorionic plate, but not in the smaller,

intravillous blood vessels. The reason for this is unclear. The presence of sFRP4 in the

amnion is also of interest as embryological the amnion is continuous with the fetal

epidermis and previous observations have shown sFRP4 is expressed in the human

epidermis (unpublished data).

The high level expression of sFRP4 in the syncytiotrophoblast rather than the

villous cytotrophoblast fits with the protein being secreted since the syncytiotrophoblast

is directly exposed to the maternal blood. It is interesting that the staining appears to be

stronger in focal areas of the trophoblast associated with syncytial knots (see Fig. 2),

Appendix I

161

which is the site of trophoblastic apoptosis and is likely to be the source of trophoblastic

material in the maternal blood stream. This may indicate a systemic maternal role for

sFRP4. The maintenance of the syncytiotrophoblast throughout pregnancy is

completely dependent on the continuous fusion of cytotrophoblast cells with this

multinucleated layer. In preeclampsia, not only is apoptosis increased, but it seems as if

the whole turnover of villous trophoblast from proliferation, via fusion, to apoptosis is

enhanced. The increase in apoptosis found in preeclampsia may not be a sign of

damage, but rather of higher traffic or turnover of villous trophoblast due to an adaptive

process .

Our findings show that apoptosis is a definite part of the invasive trophoblast

program. It may be required for reversing the maternal vessel occlusion once they have

swelled to larger diameters, allowing the flow of maternal blood through the intervillous

space. The expression of sFRP4, however, could not be correlated to such apoptosis.

While expression is high in trophoblast forming the lining of maternal blood vessels,

only a small proportion of such cells enter apoptosis.

The role of sFRP4 in human placental development and formation of the decidua

remains uncertain but these observational studies suggest a role for this protein which is

known to be important in development of other reproductive organs, the rat placenta and

in angiogenesis. Its role amid the confusing number of Wnt ligands and Fz receptors

also remains a challenge.

References

Abu-Jawdeh, G., N. Comella, et al. 1999. "Differential expression of frpHE: a novel

human stromal protein of the secreted frizzled gene family, during the

endometrial cycle and malignancy." Lab Invest 79(4): 439-47.

Bielke, W., G. Ke, et al. 1997. "Apoptosis in the rat mammary gland and ventral

prostate: detection of cell death-associated genes using a coincident-expression

cloning approach." Cell Death Differ 4(2): 114-24.

Black, S., M. Kadyrov, et al. 2004. "Syncytial fusion of human trophoblast depends on

caspase 8." Cell Death Differ 11(1): 90-8.

Carter, A. M. 2007. "Animal models of human placentation--a review." Placenta 28

Suppl A: S41-7.

Drake, J. M., R. R. Friis, et al. 2003. "The role of sFRP4, a secreted frizzled-related

protein, in ovulation." Apoptosis 8(4): 389-97.

Enders, A. C. and B. F. King 1991. "Early stages of trophoblastic invasion of the

maternal vascular system during implantation in the macaque and baboon." Am

J Anat 192(4): 329-46.

Gude, N. M., C. T. Roberts, et al. 2004. "Growth and function of the normal human

placenta." Thromb Res 114(5-6): 397-407.

Guo, K., V. Wolf, et al. 1998. "Apoptosis-associated gene expression in the corpus

luteum of the rat." Biol Reprod 58(3): 739-46.

He, B., A. Y. Lee, et al. 2005. "Secreted frizzled-related protein 4 is silenced by

hypermethylation and induces apoptosis in beta-catenin-deficient human

mesothelioma cells." Cancer Res 65(3): 743-8.

Appendix I

162

Hewitt, D. P., P. J. Mark, et al. 2006. "Placental expression of secreted frizzled related

protein-4 in the rat and the impact of glucocorticoid-induced fetal and placental

growth restriction." Biol Reprod 75(1): 75-81.

Hrzenjak, A., M. Tippl, et al. 2004. "Inverse correlation of secreted frizzled-related

protein 4 and beta-catenin expression in endometrial stromal sarcomas." J Pathol

204(1): 19-27.

Huppertz, B., D. Hemmings, et al. 2005. "Extravillous trophoblast apoptosis--a

workshop report." Placenta 26 Suppl A: S46-8.

Huppertz, B., M. Kadyrov, et al. 2006. "Apoptosis and its role in the trophoblast." Am J

Obstet Gynecol 195(1): 29-39.

James, J. L., P. R. Stone, et al. 2005. "Cytotrophoblast differentiation in the first

trimester of pregnancy: evidence for separate progenitors of extravillous

trophoblasts and syncytiotrophoblast." Reproduction 130(1): 95-103.

Lacher, M. D., A. Siegenthaler, et al. 2003. "Role of DDC-4/sFRP-4, a secreted

frizzled-related protein, at the onset of apoptosis in mammary involution." Cell

Death Differ 10(5): 528-38.

Mossman, H. W. 1987. Vertebrate Fetal Membranes. New Brunswick (NJ), Rutgers

University Press.

Rattner, A., J. C. Hsieh, et al. 1997. "A family of secreted proteins contains homology

to the cysteine-rich ligand-binding domain of frizzled receptors." Proc Natl Acad

Sci U S A 94(7): 2859-63.

Schumann, H., J. Holtz, et al. 2000. "Expression of secreted frizzled related proteins 3

and 4 in human ventricular myocardium correlates with apoptosis related gene

expression." Cardiovasc Res 45(3): 720-8.

Sonderegger, S., H. Husslein, et al. 2007. "Complex expression pattern of Wnt ligands

and frizzled receptors in human placenta and its trophoblast subtypes." Placenta

28 Suppl A: S97-102.

Tulac, S., N. R. Nayak, et al. 2003. "Identification, characterization, and regulation of

the canonical Wnt signaling pathway in human endometrium." J Clin Endocrinol

Metab 88(8): 3860-6.

Wolf, V., G. Ke, et al. 1997. "DDC-4, an apoptosis-associated gene, is a secreted

frizzled relative." FEBS Lett 417(3): 385-9.

Wong, S. C., S. F. Lo, et al. 2002. "Expression of frizzled-related protein and Wnt-

signalling molecules in invasive human breast tumours." J Pathol 196(2): 145-

53.

Acknowledgments

We thank Dr. Alen Enders (Davis, California) for his generous gift of formaldehyde-

fixed paraffin blocks of macaque implantation material and Dr. Henning Schneider

(Bern, Switzerland) for making available -80ºC frozen placenta tissue from normal and

preeclamptic patients. The authors thank our colleagues, Profs. Anne-Catherine Andres

and Andrew Ziemiecki, Department of Clinical Research, Bern, for many valuable

discussions and for their help in producing the affinity-purified rabbit antibody against

sFRP4. We thank Ms. Susanne Saurer, Department of Clinical Research, Bern, for her

valuable technical assistance. This research project has been supported by grants from

Appendix I

163

the Stiftung für Klinisch Experimentelle Tumorforschung and the Swiss National

Science Foundation (to RRF), and from the Australian Research Council, the Raine

Foundation (AC & LW), UWA Post graduate studentship (LW), and NH & MRC,

Australia (to AMD).

Figures and Legends

Figure 1: (A). Macaque placenta. Expression of sFRP4 was investigated using semi-

quantitative PCR for both sFRP4 and GAPdH. The results of two typical PCR reactions

are shown for samples at 1) 22 days gestation, 2) 30, 3) 40, 4) 45, 5) 50, 6) 56, 7) 129,

and 8) 165 days. The “O” track is a blank PCR reaction without template. The Hind III

digested Lamda phage size marker for DNA is shown in track “M”. B Densitometry

was performed on independent reaction runs (utilizing the same RT product) in order to

obtain a semi-quantitative estimate of sFRP4 expression. The relative intensity of

sFRP4 reaction product for each sample in arbitrary units was normalized against the

yield from the GAPdH reaction obtained with the same RT product. The results are

summarized in B. The samples in both A and B were as follows: Significance was

p<0.01 for samples 5 (50 days) and 7 (129 days).

Appendix I

164

Figure 2: Parallel, serial sections from 30 days gestation (KE 44 as in Fig. 5) were

stained with affinity-purified rabbit antibody specific for sFRP4 (A, 5X objective and B;

20X objective) or with an in situ hybridisation procedure using DIG-labeled antisense

riboprobe specific for sFRP4 (C; 20X objective). The box in A indicates the region

centered in the other 2 micrographs. Arrows designate the strongly sFRP4-expressing

vascular syncytiotrophoblast. “V” denotes the villi, “Gl” endometrial gland and “TS”

denotes the trophoblastic shell.

Figure 3: Sections from 45 days gestation (A; 20X objective) and 56 days gestation (B;

20X objective) illustrate the high level of sFRP4 expression in syncytiotrophoblast

which have invaded the vascular system (arrows). The sFRP4-specific staining of

trophoblast in the trophoblastic shell seen in B is recognizably weaker than with the

intravascular trophoblast.

Appendix I

165

Figure 4: Sections from 30 days gestation are shown which have

been exposed to a TUNEL reaction to demonstrate apoptotic nuclei with fragmented

DNA. All were photographed with the 20X objective. A shows individual positive cells

in villi, while B illustrates the “zone of destruction” and C apoptotic cells in an

occluded vessel.

Figure 5: First trimester human placenta, immunohistochemistry (x40 objective) with

sFRP4 showing positive syncytiotrophoblast, with negative villous intermediate

cytotrophoblast, and some positive stromal cells.

Appendix I

166

Figure 6: Term human placenta, immunohistochemistry (x40 objective) with sFRP4

showing in A positive syncytiotrophoblast covering the villi. The blood vessels and

stroma are negative. In B there is positive staining of some endothelial cells in the large

fetal vessels of the chorionic plate and in some of the adjacent muscle fibres. In C there

is some staining of the amnion.

Appendix II

167

Appendix II

Melt Curve Analysis of Real Time PCR Products

Figure 1. Melt Curve analysis of caspase-14 mRNA in Forskolin-treated BeWo samples

(22-03-06) using Real Time PCR.

Figure 2. Melt Curve analysis of caspase-8 mRNA in Forskolin-treated BeWo samples


Figure 3. Melt Curve analysis of cytokeratin-18 mRNA in all placental samples (22-06-

07) using Real Time PCR.

Appendix II

168

Figure 4. Melt Curve analysis of E-cadherin mRNA in Forskolin-treated BeWo samples


Figure 5. Melt Curve analysis of eNOS in first trimester and term placental samples (20-

04-07) using Real Time PCR.

Figure 6. Melt Curve analysis of Filaggrin mRNA in all placental samples (26-07-07)

using Real Time PCR.

Appendix II

169

Figure 7. Melt Curve analysis of GAPDH in RNAi samples (13-04-07) using Real Time

PCR.

Figure 8. Melt curve analysis of β-hCG mRNA in all placental samples (18-04-07)

using Real Time quantitative Polymerase Chain Reaction (PCR).

Figure 9. Melt Curve analysis of hPL mRNA in all placental samples (18-06-07) using

Real Time PCR.

Appendix II

170

Figure 10. Melt Curve analysis of Ki67 mRNA in Forskolin-treated BeWo samples (03-

05-06) using Real Time PCR.

Figure 11. Melt Curve analysis of KLF4 mRNA in Forskolin-treated BeWo samples


Figure 12. Melt Curve analysis of L19 in term placental samples (17-04-07) using Real

Time PCR.

Appendix II

171

Figure 13. Melt Curve analysis of sFRP4 mRNA in Breast Tunour samples (26-05-07)


Figure 14. Melt Curve analysis of VEGF-A mRNA in all placental samples (20-11-07)


characterisation of caspase-14 in the human placenta · characterisation of caspase-14 in the human...

Documents