Study of the effect of transforming growth factor 1 on the gap junction protein connexin 43 in hepatic stellate cells

STUDY OF THE EFFECT OF TRANSFORMING
GROWTH FACTOR β-1 ON THE GAP JUNCTION
PROTEIN CONNEXIN 43 IN HEPATIC STELLATE CELLS

LIM CHIN CHIA MICHELLE
(B.Sc. (Hons)), KING’S COLLEGE, LONDON, UK

A THESIS SUBMITTED
FOR THE DEGREE OF MASTER OF SCIENCE

DEPARTMENT OF BIOLOGICAL SCIENCES
NATIONAL UNIVERSITY OF SINGAPORE
2010


ACKNOWLEDGEMENTS

I would like to express my gratitude to my supervisor Dr. Zhuo Lang
for his guidance, help and patience during the course of my study. I would also
like to thank my co-supervisor Dr. Chan Woon Khiong for his valuable
discussion, advice and help, especially in the writing of my thesis. Special
thanks go to my colleagues Dr. Gunter Maubach and Dr. Zhaobing Ding for
their useful discussion and help throughout my work. I would like to convey
my thanks to my colleagues Dr. Zhiyuan Ke and Nur-Afidah Bte Mohamed
Suhaimi for their help and comradeship in the lab.
I would like to express my deep-felt gratefulness to Professor Jackie
Ying and Noreena Abubakar, directors of the Institute of Bioengineering and
Nanotechnology (IBN), for their encouragement and tremendous support
during my work at IBN. I would like to acknowledge the funding support by
IBN (Biomedical Research Council, Agency for Science, Technology and
Research, Singapore).

I would also like to acknowledge the Department of Biological
Sciences (DBS), National University of Singapore, for providing me the
opportunity to pursue my Master of Science study. Finally, I would like to
thank Reena Devi and Li Xingzuan Priscilla, staff of DBS, for their patience
and help with all my administrative queries.

I


TABLE OF CONTENTS

ACKNOWLEDGEMENTS

I

TABLE OF CONTENTS

II

SUMMARY

V

LIST OF TABLES

VII

LIST OF FIGURES

VIII

CHAPTER 1

1

1 INTRODUCTION

1

1.1 Liver Fibrosis

1

1.2 Hepatic Stellate Cells

2

1.2.1 Characteristics of the Hepatic Stellate Cells and their
Functions in Normal Liver
1.2.2 Activation of Hepatic Stellate Cells during Liver Injury

2
3

1.3 Transforming Growth Factor-1 (TGF-1)

5

1.4 Gap Junction Protein - Connexin 43

6

1.4.1 Family of Connexins

6

1.4.2 Connexin 43 in Hepatic Stellate Cells

9

1.5 Zinc Finger Transcription Factor Snai1

10

1.6 Aim of this Study

11

CHAPTER 2
2 MATERIALS AND METHODS

13
13

2.1 Cell Culture Conditions

13

2.2 Treatment with Recombinant Human TGF-1

14

II


2.3 Reverse Transcription and Quantitative PCR

14

2.4 SDS-PAGE and Western Blot

15

2.5 Immunofluorescence Staining

16

2.6 Analysis of Gap Junction Intercellular Communication

17

2.7 Electrophoretic Mobility Shift Assay

18

2.8 Snai1 and Connexin 43 siRNAs Transfection

19

2.9 Cell Counting

20

2.10 Statistical Analysis

21

CHAPTER 3
3 RESULTS
3.1 Effect of Exogenous rhTGF-1 on Cx43 in HSCs

22
22
22

3.2 Effect of Exogenous rhTGF-1 on the Phosphorylation of Cx43 in
HSCs

25

3.3 The PKC Pathway is implicated in the rhTGF-1 Induction of
Cx43 Phosphorylation at Serine 368 in HSCs

26

3.4 Distribution of Cx43 and pCx43 S368 in the HSCs

27

3.5 Effect of rhTGF-1 on the Gap Junction Intercellular

Communication between HSCs

29

3.6 Evidence for the TGF-1 Down-regulation of Cx43 Expression
via Snai1 in HSCs

32

3.6.1 Inverse Correlation between Snai1 and Cx43 Transcript and
Protein Expression

32

3.6.2 Nuclear Extracts of HSCs Bind to the Snai1 Consensus
Sequence in the Cx43 Promoter
3.7 Regulation of HSC Proliferation

35
38
III


3.7.1 Regulation of HSC Proliferation by rhTGF- 1

38

3.7.2 Regulation of HSC Proliferation by Connexin 43

41

CHAPTER 4
4 DISCUSSION
4.1 rhTGF-1 Regulation of Cx43 in Hepatic Stellate Cells

44
44
44

4.2 From the Perspective of Liver Fibrosis - Relevance of rhTGF-1induced Down-regulation of Cx43 in HSCs

45

4.3 Cell Type-Specific TGF-1 Regulation of Cx43

46

4.4 Possible Regulation of Cx43 by TGF-1 through Snai1

47

4.5 Functional Significance of Cx43 Suppression by TGF-1 in
Hepatic Stellate Cells
4.6 Future Studies

48
50

BIBLIOGRAPHY

52

RESPONSE TO EXAMINERS

69

APPENDIX

74

Publication: “TGF-1 down-regulates connexin 43 expression
and gap junction intercellular communication in rat hepatic
stellate cells”

IV


SUMMARY

The study of cell-cell communication has attracted considerable
research interest because this process has been implicated in many important
aspects of cellular functions, including cell growth and development. This
kind of communication is made possible by the existence of hydrophilic gap
junctions between adjacent cells. The gap junctions connect the cytosol of
neighboring cells, enabling them to exchange information in the form of ions
and small molecules less than 1 kiloDalton in size. These intercellular
junctions are composed of protein subunits called connexins. Different cells
express different connexins and therefore, gap junctions can comprise of
identical or diverse connexins.
Hepatic stellate cells (HSCs) play a significant role during the

pathogenesis of liver fibrosis because they contribute greatly to the
accumulation of extracellular matrix proteins. HSCs can establish a concerted
response by communicating to each other through functional gap junctions
made up of connexin 43 (Cx43) proteins. Other researchers have shown that
Cx43 in the HSCs can be regulated by several pro-inflammatory cytokines and
other molecules.
In this work, we showed that exogenous recombinant human TGF-1
(rhTGF-1), a pro-fibrotic stimulus, suppressed Cx43 mRNA and protein
expression in a rat HSC cell line and in vitro activated primary rat HSCs.
Furthermore, rhTGF-1 increased the phosphorylation of Cx43 at serine 368.
These effects led to a decrease in the gap junction intercellular communication
between the HSCs, as shown by gap-FRAP analysis. We also observed the
V


binding of Snai1, from the nuclear extract of HSCs, to a Snai1 consensus
sequence in the Cx43 promoter. In the same context, Snai1 siRNA transfection
induced the expression of Cx43, suggesting that TGF-1 regulates Cx43 via
Snai1. In addition, we demonstrated that the knockdown of Cx43 by siRNA
transfection slowed down the proliferation of HSCs. These findings shed light
on the following: (1) TGF-1 regulates intercellular communication in the
HSCs by affecting the expression level and the phosphorylation state of Cx43
through Snai1 signaling; and (2) Cx43 is implicated in the TGF-1-mediated
regulation of HSC proliferation.
The results presented in this master thesis have been published (Lim et
al., 2009) and the publication is attached as appendix.

VI

LIST OF TABLES

Table 2.1

Sequences of siRNAs used for transfection

20

VII


LIST OF FIGURES

Figure 1.1

Events that occur during liver fibrosis

3

Figure 1.2

Paracrine factors and cell types involved in the activation
of HSCs

4

Figure 1.3

Involvement of TGF-1 in collagen type I homeostasis

6

Figure 1.4

Overview of the structure and assembly of connexins into
gap junctions

7

Figure 1.5

Life cycle of connexins and their interactions with other
proteins

8

Figure 1.6

The activation of Snail genes by a variety of extracellular
signals

10

Figure 1.7

Downstream targets of Snai1 genes and their association
with different physiological processes

11

Figure 3.1

Exogenous rhTGF-1 suppressed Cx43 mRNA in HSCs

22

Figure 3.2

Exogenous rhTGF-1 down-regulated Cx43 expression
in HSCs

24

Figure 3.3

Exogenous rhTGF-1 affected the phosphorylation of
Cx43 in HSCs

25

Figure 3.4

rhTGF-1 induced the phosphorylation of Cx43 at
serine 368 via the PKC pathway

26

VIII

Figure 3.5

Distribution of non-phosphorylated Cx43 and
pCx43 S368 in HSC-2 cells as visualized by
immunofluorescence staining

28

Figure 3.6

FRAP analysis of gap junction intercellular
communication in HSC-2 cells

31

Figure 3.7

Exogenous addition of rhTGF-1 up-regulated
Snai1 transcript in HSCs

32

Figure 3.8

Analysis of the transcript and protein levels of Cx43 and
Snai1 after rhTGF-1 treatment

33

Figure 3.9

Using Snai1 siRNAs transfection to study correlation
between Snai1 and Cx43 expression in HSCs

34

Figure 3.10

Binding of Snai1 to the potential Snai1 recognition
sequence (CAGGTG) in the rat Cx43 promoter

36

Figure 3.11

Exogenous rhTGF-1 or Snai1 siRNAs transfection
affected the binding of Snai1 to its consensus sequence
in the rat Cx43 promoter

37

Figure 3.12

rhTGF-1 decreased the proliferation of HSC-2 cells as
assessed by the expression of the proliferation marker
PCNA and cell number

39

Figure 3.13

Effect of Snai1 siRNA on rhTGF-1-dependent
regulation of HSC proliferation

40

Figure 3.14

Cx43 siRNAs transfection decreased the Cx43 transcript
level in HSC-2 cells

41

Figure 3.15

Cx43 siRNA transfection decreased the proliferation of
HSC-2 cells as assessed by cell number

42

Figure 3.16

Cx43 siRNA transfection decreased the proliferation of
HSC-2 cells as assessed by the expression of the
proliferation marker PCNA

43

IX

CHAPTER 1

INTRODUCTION

1.1 Liver Fibrosis

Liver fibrosis is the wound healing response to chronic injury of the
liver parenchyma and by itself, is a reversible process. The causes of liver
fibrosis are manifold, including toxins such as alcohol, drugs, autoimmune
diseases and viruses (Hepatitis B and C) (Driessen et al., 1999; Dufour et al.,
1997; Hollinger and Lau, 2006; Ishak et al., 1995; Lee, 1995; Paradis et al.,
1996; Siegmund et al., 2005; Wong et al., 1996). Consequently, the
progression of the fibrotic process in the liver differs, depending on the
etiologies. It can advance fast (weeks or several months) as seen for druginduced injury or hepatitis C virus infection, or in most cases the process is
slow and can take decades to develop due to the regenerative capabilities of
the liver (Friedman, 2008a). Ultimately, liver fibrosis results in lifethreatening conditions, for example portal hypertension, liver failure and
hepatocellular carcinoma.
Essentially, liver fibrosis is characterized by an over-expression of
extracellular matrix (ECM) proteins, particularly the fibrillar collagen types I
and III, which leads to a scaring of the liver (Clement et al., 1986; Yamamoto
et al., 1984). The cellular sources involved in this process are versatile and
have been extended in recent years. In principle, the hepatic stellate cells
(HSCs) are regarded as the main source of ECM proteins, although other cell

1


types like the portal fibroblasts, bone marrow-derived cells as well as
fibroblasts derived from epithelial-mesenchymal transition (EMT) also

contribute considerable amounts (Forbes et al., 2004; Friedman, 2008b;
Kinnman et al., 2003; Kinnman and Housset, 2002; Ramadori and Saile, 2004;
Wells, 2008).

1.2 Hepatic Stellate Cells

1.2.1 Characteristics of the Hepatic Stellate Cells and their Functions in
Normal Liver

Hepatic stellate cells also called Ito cells, fat-storing cells, vitamin Astoring cells, hepatic pericytes or lipocytes were first described by Karl
Wilhelm von Kupffer in 1876 (v. Kupffer, 1876). Comprising approximately
15% of the total cell population in the liver, HSCs are located in the space of
Disse between the hepatocytes and the endothelial cells in the liver (Fig. 1.1).
As the name implies, HSCs are spindle- or star-shaped with elongated nuclei.
In the normal liver, HSCs exist in a quiescent state and act as the main
storage of vitamin A in the liver (80-90%) in the form of retinyl esters in lipid
droplets (Hendriks et al., 1985; Hendriks et al., 1988). HSCs also express
several retinoid-related proteins such as the cellular retinol-binding protein
and retinol palmitate hydrolase, indicating their involvement in retinoid
metabolism (Blaner et al., 1985). In addition, several studies have shown that
HSCs express morphogenic proteins such as epimorphin and pleiotrophin,
suggesting a possible function(s) of HSCs during liver development and

2


regeneration (Asahina et al., 2002; Yoshino et al., 2006). Probably the most
unexpected finding is that numerous researchers have provided evidence to
show that HSCs are professional antigen presenting cells; and there is a mutual
regulation between the HSCs and the hepatic immune response (Maher, 2001;

Winau et al., 2008).

Fig.1.1. Events that occur during liver fibrosis.
A distinctive architectural difference can be seen between a normal liver (A)
and a fibrotic liver (B). The HSC is situated in the space of Disse between the
hepatocytes and the endothelial cells. During liver fibrosis, lymphocytes are
recruited to the hepatic parenchyma; some hepatocytes become apoptotic; and
Kupffer cells become activated. The HSCs also become activated to
myofibroblast-like cells, secreting abundant ECM proteins. The image is taken
from (Bataller and Brenner, 2005).

1.2.2 Activation of Hepatic Stellate Cells during Liver Injury

During fibrogenesis, quiescent HSCs will transdifferentiate into a
proliferative

and

contractile myofibroblast-like phenotype

(Borkham-

3


Kamphorst et al., 2007; Pinzani, 2002; Rockey, 2001; also see Friedman, 1993
for review). This process, also referred to as the activation of HSCs, is
initiated by many paracrine factors that are secreted by all neighboring cell
types, including the membrane lipid degradation products of the hepatocytes;
fibronectin from the endothelial cells; as well as cytokines and reactive

oxidative species from the Kupffer cells, to name a few (Bilzer et al., 2006;
Jarnagin et al., 1994; Novo et al., 2006).

Fig.1.2. Paracrine factors and cell types involved in the activation of
HSCs.
Resident liver cells (red) and infiltrating inflammatory cells (green) interact
extensively with the HSCs via the production of assorted signaling molecules.
TGF-β is a potent fibrogenic factor that is secreted by almost all the cell types
depicted as well as by the HSCs (represented by a yellow cell in the middle),
giving rise to paracrine and autocrine signaling to ensure the perpetual
activation of the HSCs. Image is taken from (Gressner et al. 2007).

The HSCs activation process is sustained by both paracrine and
autocrine signaling involving numerous cytokines (Gressner et al., 2007).
Paracrine stimulation depends on many different cell types in the liver, for
instance the hepatocytes, endothelial cells, platelets and Kupffer cells (Fig.
4


1.2). These cells secrete different cytokines like the transforming growth
factor-β1 (TGF-β1), platelet-derived growth factor (PDGF), basic fibroblast
growth factor (bFGF) and endothelial growth factor (EGF) (Friedman, 2008c).

1.3 Transforming Growth Factor-β1 (TGF-β1)

The TGF-β1 peptide belongs to the TGF-β superfamily of cytokines,
which consists of three isoforms of TGF-β (TGF-β1, β2, β3), bone
morphogenetic proteins, activins and growth differentiation factors. TGF-β
isoforms and their receptors are produced by almost all cell types (Howe et al.,
2003). TGF-β signaling is involved in different functions like cell cycle,

apoptosis, angiogenesis, wound healing, immune regulation and tumor
biology, depending on the context and cell type (Letterio and Roberts, 1998;
Massague, 2000, 2008; Massague et al., 2000; Massague and Chen, 2000;
Pardali and Moustakas, 2007; Rolfe et al., 2007; Serrati et al., 2009; Wan and
Flavell, 2007).
TGF-β1 is one of the best-studied signaling molecules with diverse
effects on the HSCs, including regulation of collagen metabolism, contraction
and proliferation (Hellerbrand et al., 1999; Kato et al., 2004; Kharbanda et al.,
2004; Verrecchia and Mauviel, 2007). In the context of liver fibrosis, TGF-β1
is often regarded as a pro-fibrotic cytokine because of its effect as the most
powerful stimulus of collagen type I production in HSCs by stimulating the
transcription of procollagen genes (Cao et al., 2002; Ponticos et al., 2009;
Tsukada et al., 2005). On the other hand, TGF-β1 represses collagen type I
degradation by the down-regulation of matrix metalloproteinases (MMP) and
5


the up-regulation of tissue inhibitors of matrix metalloproteinases (TIMP)
(Edwards et al., 1987; Knittel et al., 1999; Lechuga et al., 2004; Verrecchia
and Mauviel, 2007). Eventually, TGF-β1 causes the net deposition of collagen
type I to increase, an important factor in the development of fibrotic tissue
(Fig. 1.3).

Fig. 1.3. Involvement of TGF-β1 in collagen type I homeostasis.
TGF-β is known to shift the equilibrium between collagen production and
degradation in the fibrotic tissue. This is accomplished by increasing the
production of collagen by inducing its gene expression. On the other hand, the
degradation of collagen is regulated by suppression of the MMP expression
and an increased availability of TIMPs. Diagram is taken from (Verrecchia
and Mauviel, 2007).

1.4 Gap Junction Protein - Connexin 43

1.4.1 Family of Connexins

Gap junctions are microscopic channels formed between adjacent cells
that allow intercellular communication by means of the exchange of small

6


molecules and ions (cyclic nucleotides, inositol phosphates, Ca2+, K+). Two
neighboring cells contribute a hemi-channel (connexon) each to form a gap
junction channel.

Fig. 1.4. Overview of the structure and assembly of connexins into gap
junctions.
The connexin protein consists of an amino acid chain containing 4
transmembrane helices with the amino and carboxy terminus located
intracellularly (A). Through interactions between the helices, the protein forms
a tightly packed membrane structure (B), which assembles into a homohexameric hemi-channel (connexon, C). The connexon from adjacent cells
form a gap junction, which can exist in either an open or closed state (D).
Picture is taken from />
The connexon itself consists of an assembly of six protein subunits
called connexins (Goodenough et al., 1996), of which more than 20 different
connexins are known to date (Eyre et al., 2006) (Fig. 1.4). A connexon is
denoted as homomeric or heteromeric when it is composed of identical or
different connexins, respectively. Similarly, a gap junction channel can be
either homotypic if it is formed by the same connexon or heterotypic when it
is made up of two connexons with different connexin isotypes (Kumar and

7


Gilula, 1996). The constituent of the gap junction has a direct impact on its
gating properties (Cottrell et al., 2002).

Fig. 1.5. Life cycle of connexins and their interactions with other proteins.
Connexins are synthesized by ribosomes attached to the endoplasmic
reticulum (ER). In general, oligomerization of connexins into connexons
occurs during the transport from the ER to the trans-Golgi network. The
connexons are subsequently moved to the plasma membrane, where they can
either remain as hemichannels or form functional gap junctions with
connexons from adjacent cells. Gap junctions can be degraded by the
lysosomes or proteosomes or recycled to the plasma membrane (dashed
arrow). Connexins associate with many proteins including the cytoskeletal
molecules (red), junctional proteins (blue), kinases (green) and others
(yellow). The figure is adapted from (Dbouk et al., 2009).

Connexins have a dynamic life cycle that not only results in a rapid
turnover time of several hours (Musil et al., 2000), but also involves a great
8


number of interactions with other proteins, such as the cytoskeletal, adheren
junction-associated and tight junction-associated proteins (Fig. 1.5). These
interactions together with the phosphorylation and dephosphorylation events
by different kinases and phosphatases regulate the properties of the connexins
(Giepmans, 2004). Some of the properties affected include transport to the
plasma membrane (Musil and Goodenough, 1991), assembly into gap

junctions (Falk, 2000) and degradation of the connexins (Beardslee et al.,
1998).

1.4.2 Connexin 43 in Hepatic Stellate Cells

In the liver, hepatocytes express connexins 26 and 32 (Cx26 and
Cx32), whereas nonparenchymal cells (endothelial cells, HSCs, oval cells,
Kupffer cells) express connexin 43 (Cx43) (Gonzalez et al., 2002). Cx26 and
Cx32 can form heterotypic gap junctions with each other, but not with Cx43
(Segretain and Falk, 2004). Different liver injury models lead to a decrease in
Cx26 and Cx32 expression (De Maio et al., 2002). In contrast, previous
findings from Fischer and colleagues established that the expression of Cx43
increases in activated HSCs, resulting in a corresponding enhancement in the
gap junction intercellular communication (GJIC) between these cells (Fischer
et al., 2005). Based on these results, the down-regulation of connexins in the
hepatocytes could be interpreted as a self-defense mechanism to prevent the
spreading of tissue injury, whereas the up-regulation of Cx43 in activated
HSCs could play a role in facilitating a concerted action of this cell type
during tissue repair (De Maio et al., 2002).

9


1.5 Zinc Finger Transcription Factor Snai1

Snai1 belongs to the Snail family of zinc finger transcriptional factors,
which down-regulates a number of genes during embryonal development,
morphogenesis, EMT and cancer development (Barrallo-Gimeno and Nieto,
2005; Batlle et al., 2000; Cano et al., 2000; Murray and Gridley, 2006; Nieto,
2002). The expression of Snai1 is induced by FGF, EGF, TGF-β, parathyroid

hormone related peptide and others (Fig. 1.6) (Cho et al., 2007; De Craene et
al., 2005).

Fig. 1.6. The activation of Snail genes by a variety of extracellular signals.
Genes encoding for the Snail family of proteins are regulated by many
extracellular signals. Depicted below each extracellular mediator are the tissue
and the cellular events in which it has been studied. The localization of the
Snail proteins, regulated by several molecules (yellow), also affects their
activity. AMF, autocrine motility factor; BMP, bone morphogenetic protein;
E-cad, E-cadherin; EGF, epidermal growth factor; FGF, fibroblast growth
factor; GSK3, glycogen synthase kinase-3; ILK, integrin-linked kinase; LIV1,
metalloprotease, zinc transporter; MTA3, metastasis-associated protein 3;
PAK1; p21-activated kinase; PTH(rP)R, parathyroid hormone related peptide
receptor; SCF, stem cell factor; TGFβ, transforming growth factor β. The
figure is taken from (Barrallo-Gimeno and Nieto, 2005).

10


The downstream targets of Snai1 are versatile and affect several
physiological processes like proliferation, expression of epithelial and
mesenchymal markers and migration (Fig. 1.7).

Fig. 1.7. Downstream targets of Snai1 genes and their association with
different physiological processes.
Depicted in red and green are the target genes of Snai1 and events which are
down- and up-regulated by Snai1, respectively. BID, Bcl-interacting death
agonist; CDK, cyclin-dependent kinase; DFF, DNA fragmentation factor;
ERKs, extracellular signal-regulated kinases; MMPs, metalloproteinases;
PI3K, phosphoinositide 3-kinase; p21, cyclin-dependent kinase inhibitor; p53,

tumour suppressor; Rb, retinoblastoma; XR11, Xenopus Bcl-xL homologue.
The figure is taken from (Barrallo-Gimeno and Nieto, 2005).

1.6 Aim of this Study

Intercellular communication via gap junctions is important to maintain
tissue homeostasis by regulating several cellular events, for instance
proliferation, apoptosis and even differentiation. Given that liver fibrosis is a
disease whereby such normal processes are deregulated, it is conceivable that
the GJIC in a fibrotic liver may be altered. It is therefore logical to suggest that
under such a circumstance, the regulation of the connexins in the different
hepatic cell types may also be affected.

11


Although some studies have been done on Cx26 and Cx32 in the
hepatocytes and Cx43 in the Kupffer cells (Eugenin et al., 2007; Yamaoka et
al., 2000), little is known about the connexins in HSCs. Greenwel and
colleagues (1993) discovered that HSC cell lines derived from the liver of rats
with cirrhosis express Cx43. This observation was confirmed by another
study, which showed that HSCs isolated from fibrotic livers express higher
level of Cx43 than those from normal livers (Fischer et al., 2005). Fischer and
collaborators also showed the regulation of GJIC upon treatment with different
regulatory molecules and cytokines. More specifically, they demonstrated that
certain

molecules

like

the

pro-inflammatory

interleukin-1β

and

lipopolysaccharide, as well as the pro-fibrogenic endothelin 1 increased the
expression of Cx43 in the HSCs.
Surprisingly, the experimental regime of Fischer and coworkers (2005)
did not include TGF-β1, a powerful pro-fibrogenic stimulus of HSCs during
liver fibrosis. Several studies have linked either TGF-β1 or Cx43 to the wound
healing process of damaged tissue (Crowe et al., 2000; Huang et al., 2002;
Mori et al., 2006; Qiu et al., 2003). As liver fibrosis is the outcome of a
deregulated repair process, an examination of the consequence of TGF-β1
treatment on Cx43 in the HSCs will provide a mechanistic understanding of
the interplay between these participating molecules. In my thesis, I will report
on my investigations of the effect of TGF-β1 on Cx43 expression in the HSCs
and the subsequent functional implications.

12


CHAPTER 2

MATERIALS AND METHODS

2.1 Cell Culture Conditions

Primary HSCs were isolated from male Wistar rats according to the
pronase and collagenase treatment method (Weiskirchen and Gressner, 2005).
The protocol was approved by the Institutional Animal Care and Use
Committee (IACUC) of the Biomedical Research Council of Singapore.
Freshly isolated primary HSCs were seeded in a 25 cm2 uncoated tissue
culture flask (Nunc). The medium was replaced after 24 hours. At this point,
there were approximately 1-2 x 106 primary HSCs attached to the flask. The
cells were passaged 2-3 times until use for experiments. The purity was
assessed by vitamin A autofluorescence one day after isolation.
The (male Wistar rat) cell line HSC-2 was derived in our lab and is
described elsewhere (Maubach et al., 2008). All cells were cultivated in a
humidified 37ºC incubator with 5% circulating CO 2 . High glucose Dulbecco’s
modified Eagle medium (D-MEM) supplemented with 10% fetal bovine
serum, 100 units/ml penicillin and 100 µg/ml streptomycin was used during
cell culture.
10X Trypsin/EDTA solution (0.5/0.2%) in (10X) phosphate-buffered
saline (PBS) without Ca2+ and Mg2+ was purchased from Biochrome
(Germany). All other cell culture reagents were from Invitrogen (CA, USA).

13


2.2 Treatment with Recombinant Human TGF-β1

Twenty four hours prior to treatment, HSCs were seeded in 75 cm2
tissue culture flasks so that they will be 60-70% confluence the next day. For
the experiments, recombinant human TGF-β1 (rhTGF-β1) was added at a final
concentration of 1 ng/ml or 10 ng/ml and incubated for 2, 6, 10, 24 or
30 hours. For the control treatment, only PBS was given to the cells. In some

experiments, HSC-2 cells were treated with bisindolylmaleimide I (BIM I) at a
final concentration of 5 µM for 30 min before the addition of 10 ng/ml
rhTGF-β1.
The rhTGF-β1 was purchased from Biovision (iDNA, Singapore).
BIM I was bought from Merck KGaA (Darmstadt, Germany). The
composition

of

10X

PBS

is

as

follows:

160g

NaCl,

4g

KCl,

53.6g Na 2 HPO 4 .7H 2 O and 48g KH 2 PO 4 in 1L water. The pH was adjusted
to 7.4.

2.3 Reverse Transcription and Quantitative PCR

Total RNA was isolated from cells according to the manufacturer’s
protocol (RNA II kit, Machery-Nagel, Germany). All reagents for reverse
transcription and real-time PCR were from Applied Biosystems (CA, USA).
One microgram of total RNA was reverse transcribed to cDNA in a reaction
mixture containing 5 µl 10X buffer, 11 µl MgCl 2 (stock 25 mM), 10 µl dNTPs
mix (stock 10 mM), 2.5 µl random hexamers (stock 50 µM), 1 µl RNase
inhibitor (stock 20U/µl), 1.25 µl reverse transcriptase (stock 50 U/µl) and

14


made up with nuclease-free water to a total reaction volume of 50 µl. The
reverse transcription reaction conditions were 25°C for 10 min, 48°C for
30 min and 95°C for 5 min. Real-time PCR reactions were performed using
the Fast Real Time PCR System (Applied Biosystems). Three microlitres of
cDNA were used in a PCR reaction mixture of 5 µl 2X Taqman Universal
PCR master mix, 0.5 µl 20X Taqman gene expression assay mix and
1.5 µl nuclease-free water.
The Taqman gene expression assay mix for target genes Cx43 and
Snai1, as well as for the endogenous control β-actin were Rn01433957_m1,
Rn00441533_g1 and 4352341E, respectively. The PCR conditions were 95°C
for 20 sec and 40 cycles of amplification at 95°C for 3 sec and 60°C for
30 sec.

2.4 SDS-PAGE and Western Blot

Cells were lysed in a buffer containing 63 mM Tris-HCl (pH 6.8),
1% sodium dodecyl sulphate (SDS) and protease/phosphatase inhibitor

cocktail (Pierce, USA). After incubation at 95°C for 10 min and centrifugation
at 16,000g for 10 min, the cell lysate was transferred to a clean microtube and
ready for further use. Depending on the protein to be identified, 10-40 µg cell
lysate was separated under reducing conditions in a 4-12% Bis-Tris SDSpolyacrylamide gel (Invitrogen) in 1X MES running buffer until the loading
dye reached the bottom of the gel. Proteins were transferred to nitrocellulose
membranes using the XCell IITM Blot Module (Invitrogen) at 30 V for 1 hour.

15