characterisation of caspase-14 in the human placenta · characterisation of caspase-14 in the human...
TRANSCRIPT
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
mediated BeWo differentiation. L-NAME acts as an inhibitor of the NO pathway. ...... 93
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. ...... 94
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
mediated BeWo differentiation. L-NAME acts as an inhibitor of the NO pathway. ...... 95
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
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). ..................................................................................... 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
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). ..................................................................................... 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
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). ............................. 119
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
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). ............................. 121
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
Chapter 1- Introduction
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
Chapter 1- Introduction
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).
Chapter 1- Introduction
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
Chapter 1- Introduction
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
Chapter 1- Introduction
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.
Chapter 1- Introduction
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-
Chapter 1- Introduction
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).
Chapter 1- Introduction
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
Chapter 1- Introduction
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.
Chapter 1- Introduction
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.
Chapter 1- Introduction
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
Chapter 1- Introduction
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.
Chapter 1- Introduction
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,
Chapter 1- Introduction
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
Chapter 1- Introduction
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
Chapter 1- Introduction
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.
Chapter 1- Introduction
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.
Chapter 1- Introduction
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
Chapter 1- Introduction
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.
Chapter 1- Introduction
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
Chapter 1- Introduction
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
Chapter 1- Introduction
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.
Chapter 1- Introduction
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
Chapter 2- Materials and Methods
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.
Chapter 2- Materials and Methods
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,
Chapter 2- Materials and Methods
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
Chapter 2- Materials and Methods
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
Chapter 2- Materials and Methods
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.
Chapter 2- Materials and Methods
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
Chapter 2- Materials and Methods
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
Chapter 2- Materials and Methods
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.
Chapter 2- Materials and Methods
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).
Chapter 2- Materials and Methods
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.
Chapter 2- Materials and Methods
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
Chapter 2- Materials and Methods
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
Chapter 2- Materials and Methods
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
Chapter 2- Materials and Methods
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
Chapter 2- Materials and Methods
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
Chapter 2- Materials and Methods
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
Chapter 2- Materials and Methods
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.
Chapter 2- Materials and Methods
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.
Chapter 2- Materials and Methods
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.
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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.
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
First Trimester Term Preeclampsia
hP
L /
L19
hPL mRNA in the Human Placenta
* *
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
Gestational Age (weeks)
hPL mRNA across human gestation
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
*
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
Gestational Age (weeks)
Caspase-14 mRNA across human gestation
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
*
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
First Trimester Term Preeclampsia
KLF
4 /
L1
9
KLF4 mRNA in the Human Placenta
*
†
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
First Trimester Term Preeclampsia
ck1
8 /
L1
9
Cytokeratin-18 mRNA in the Human Placenta
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
Gestational Age (weeks)
sFRP4 mRNA across human gestation
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
First Trimester Term Preeclampsia
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
Gestational Age (weeks)
eNOS mRNA across human gestation
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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
First Trimester Term Preeclampsia
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
Gestational Age (weeks)
VEGF-A mRNA across human gestation
Chapter 3- Placental Gene Expression across Human Gestation and in Preeclampsia
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.
Chapter 4- BeWo Model Establishment and Validation
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
Chapter 4- BeWo Model Establishment and Validation
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.
Chapter 4- BeWo Model Establishment and Validation
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
*
† ††
Chapter 4- BeWo Model Establishment and Validation
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* *
†
Chapter 4- BeWo Model Establishment and Validation
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.
Chapter 4- BeWo Model Establishment and Validation
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
*
*
Chapter 4- BeWo Model Establishment and Validation
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 .
Chapter 4- BeWo Model Establishment and Validation
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
Chapter 4- BeWo Model Establishment and Validation
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*
***
Chapter 4- BeWo Model Establishment and Validation
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
* *
Chapter 4- BeWo Model Establishment and Validation
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* *
Chapter 4- BeWo Model Establishment and Validation
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
treatment (*=P<0.05 v. Control).
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)
Chapter 5- Caspase-14 in the Trophoblast
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
Chapter 5- Caspase-14 in the Trophoblast
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.
Chapter 5- Caspase-14 in the Trophoblast
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
Chapter 5- Caspase-14 in the Trophoblast
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
Chapter 5- Caspase-14 in the Trophoblast
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.
Chapter 5- Caspase-14 in the Trophoblast
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
Chapter 5- Caspase-14 in the Trophoblast
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
Chapter 5- Caspase-14 in the Trophoblast
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
Chapter 5- Caspase-14 in the Trophoblast
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
Chapter 5- Caspase-14 in the Trophoblast
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
Chapter 5- Caspase-14 in the Trophoblast
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
Chapter 5- Caspase-14 in the Trophoblast
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.
Chapter 6- Endothelial Function of the Human Trophoblast
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.
Chapter 6- Endothelial Function of the Human Trophoblast
91
Figure 6.1 Expression of eNOS mRNA in BeWo cells following Forskolin
treatment (*=P<0.05 v. Control).
Figure 6.2 Expression of VEGF-A mRNA in BeWo cells following Forskolin
treatment (*=P<0.05 v. Control).
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
*
*
*
Chapter 6- Endothelial Function of the Human Trophoblast
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.
Chapter 6- Endothelial Function of the Human Trophoblast
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
mediated BeWo differentiation. L-NAME acts as an inhibitor of the NO pathway.
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
Chapter 6- Endothelial Function of the Human Trophoblast
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
Chapter 6- Endothelial Function of the Human Trophoblast
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
mediated BeWo differentiation. L-NAME acts as an inhibitor of the NO pathway.
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
Chapter 6- Endothelial Function of the Human Trophoblast
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
mediated BeWo differentiation (* = P<0.05 vs. Control and VEGF+L-NAME). L-
NAME acts as an inhibitor of the NO pathway.
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.
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
*
Chapter 6- Endothelial Function of the Human Trophoblast
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
* *
Chapter 6- Endothelial Function of the Human Trophoblast
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
Chapter 6- Endothelial Function of the Human Trophoblast
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* **
Chapter 6- Endothelial Function of the Human Trophoblast
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†
*
Chapter 6- Endothelial Function of the Human Trophoblast
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
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.
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*
†
†
Chapter 6- Endothelial Function of the Human Trophoblast
102
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.
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* †††
Chapter 6- Endothelial Function of the Human Trophoblast
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
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
***
*
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
***
Chapter 7- RNA Interference of Caspase-14 in BeWo
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%
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
Chapter 7- RNA Interference of Caspase-14 in BeWo
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.
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
Chapter 7- RNA Interference of Caspase-14 in BeWo
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.
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
Chapter 7- RNA Interference of Caspase-14 in BeWo
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%
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
Chapter 7- RNA Interference of Caspase-14 in BeWo
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.
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
Chapter 7- RNA Interference of Caspase-14 in BeWo
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%
*
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
Chapter 7- RNA Interference of Caspase-14 in BeWo
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%
Chapter 7- RNA Interference of Caspase-14 in BeWo
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.
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
Chapter 7- RNA Interference of Caspase-14 in BeWo
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
Chapter 8- Discussion
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
Chapter 8- Discussion
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.
Chapter 8- Discussion
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
Chapter 8- Discussion
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
Chapter 9- Bibliography
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.
Chapter 9- Bibliography
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.
Chapter 9- Bibliography
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.
Chapter 9- Bibliography
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.
Chapter 9- Bibliography
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.
Chapter 9- Bibliography
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.
Chapter 9- Bibliography
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
Chapter 9- Bibliography
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
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
Biophysical Research Communications 296, 911-7.
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
Biophysical Research Communications 277, 655-9.
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
(31-03-06) using Real Time PCR.
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
(22-03-06) using Real Time PCR.
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
(30-06-06) using Real Time PCR.
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)
using Real Time PCR.
Figure 14. Melt Curve analysis of VEGF-A mRNA in all placental samples (20-11-07)
using Real Time PCR.