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

TABLE OF CONTENTS 1.2.1 Characteristics of the Hepatic Stellate Cells and their 1.2.2 Activation of Hepatic Stellate Cells during Liver Injury 3 1.5 Zinc Finger Transcription Factor Sn

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

TABLE OF CONTENTS

1.2.1 Characteristics of the Hepatic Stellate Cells and their

1.2.2 Activation of Hepatic Stellate Cells during Liver Injury 3

1.5 Zinc Finger Transcription Factor Snai1 10

2.2 Treatment with Recombinant Human TGF-1 14

2.3 Reverse Transcription and Quantitative PCR 14

2.6 Analysis of Gap Junction Intercellular Communication 17

2.8 Snai1 and Connexin 43 siRNAs Transfection 19

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

3.6.2 Nuclear Extracts of HSCs Bind to the Snai1 Consensus

Publication: “TGF-1 down-regulates connexin 43 expression

and gap junction intercellular communication in rat hepatic

stellate cells”

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

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

LIST OF TABLES

Table 2.1 Sequences of siRNAs used for transfection 20

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 4

of HSCs

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 7

Figure 1.7 Downstream targets of Snai1 genes and their association 11

with different physiological processes

Figure 3.1 Exogenous rhTGF-1 suppressed Cx43 mRNA in HSCs 22

Figure 3.2 Exogenous rhTGF-1 down-regulated Cx43 expression 24

in HSCs

Figure 3.3 Exogenous rhTGF-1 affected the phosphorylation of 25

Cx43 in HSCs

Figure 3.4 rhTGF-1 induced the phosphorylation of Cx43 at 26

serine 368 via the PKC pathway

Figure 3.5 Distribution of non-phosphorylated Cx43 and 28

pCx43 S368 in HSC-2 cells as visualized by immunofluorescence staining

Figure 3.6 FRAP analysis of gap junction intercellular 31

communication in HSC-2 cells

Figure 3.7 Exogenous addition of rhTGF-1 up-regulated 32

Snai1 transcript in HSCs

Figure 3.8 Analysis of the transcript and protein levels of Cx43 and 33

Snai1 after rhTGF-1 treatment

Figure 3.9 Using Snai1 siRNAs transfection to study correlation 34

between Snai1 and Cx43 expression in HSCs

Figure 3.10 Binding of Snai1 to the potential Snai1 recognition 36

sequence (CAGGTG) in the rat Cx43 promoter

Figure 3.11 Exogenous rhTGF-1 or Snai1 siRNAs transfection 37

affected the binding of Snai1 to its consensus sequence

in the rat Cx43 promoter

Figure 3.12 rhTGF-1 decreased the proliferation of HSC-2 cells as 39

assessed by the expression of the proliferation marker PCNA and cell number

Figure 3.13 Effect of Snai1 siRNA on rhTGF-1-dependent 40

regulation of HSC proliferation

Figure 3.14 Cx43 siRNAs transfection decreased the Cx43 transcript 41

level in HSC-2 cells

Figure 3.15 Cx43 siRNA transfection decreased the proliferation of 42

HSC-2 cells as assessed by cell number

Figure 3.16 Cx43 siRNA transfection decreased the proliferation of 43

HSC-2 cells as assessed by the expression of the proliferation marker PCNA

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 drug-induced 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 life-threatening 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

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 storing 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)

A-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

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-

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

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

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

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 hexameric 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 http://en.wikipedia.org/wiki/Connexin

homo-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

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

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)

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)

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

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

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 CO2 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)

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 Na2HPO4.7H2O and 48g KH2PO4

2.3 Reverse Transcription and Quantitative PCR

in 1L water The pH was adjusted

to 7.4

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 MgCl2 (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

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

2.4 SDS-PAGE and Western Blot

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

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 SDS-

polyacrylamide 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

The blots were blocked in blocking solution containing 5% non-fat milk in Tris-buffered saline-Tween buffer (TBS-T) The composition of 10X TBS-T buffer is as follows: 24g Tris, 80g NaCl and 1% Tween 20 in 1L water The Cx43 (sc-9059, Santa Cruz Biotechnology, USA), phosphorylated Cx43 at serine 368 (pCx43 Ser368, 3511S, Cell Signaling Technology, USA), PCNA (Ab29, Abcam, UK) and β-actin (A2228, Sigma, USA) primary antibodies were applied at a dilution of 1:1000, 1:750, 1:5000 and 1:7500, respectively in blocking solution After three washes of 5 min each in TBS-T, the appropriate secondary antibody conjugated with horse radish peroxidase (Santa Cruz Biotechnology, USA) was added at a dilution of 1:2000 in blocking solution After three washes of 5 min each in TBS-T, the membrane was developed using ECL Plus reagents (RPN2132, GE Healthcare, UK) The incubation with pCx43 antibody was performed overnight at 4°C All other incubations were carried out for 1 hour at room temperature Semi-quantitative densitometric analysis of Western blots was performed using the ImageJ software (W Rasband, NIH; http://rsb.info.nih.gov/ij/)

2.5 Immunofluorescence Staining

One day before staining, HSC-2 cells were seeded on glass cover slips

in 24-well cell culture dish such that they will be 50-60% confluence the next day The cells were fixed with 4% paraformaldehyde at room temperature for

10 min After three times washing of 5 min with PBS, the cells were permeabilized for 1 hour at 37°C with blocking solution containing

0.1% Triton X-100, 10% horse serum and 89.9% PBS The primary antibody was pipetted directly into the blocking solution and further incubated for

1 hour at 37°C After three times washing of 5 min with PBS, the secondary antibody was added in the desired dilution in the blocking solution and incubated for 1 hour at 37°C After three times washing of 5 min with PBS, the cover slips were transferred face-down to a microscopic slide with 4 µl DAPI (4'-6-Diamidino-2-phenylindole) using a clean forcep

The primary antibodies Cx43 (C-6219, Sigma, USA) and phosphorylated Cx43 at serine 368 (pCx43 Ser368, 3511S, Cell Signaling Technology, USA) were applied at a dilution of 1:50 The secondary antibodies, anti-rabbit Alexa 488 and anti-rabbit Alexa 555 (Invitrogen, USA), were used at a dilution of 1:200 Images were taken using the LEICA RMB-

DM epifluorescence microscope (LEICA, Germany)

2.6 Analysis of Gap Junction Intercellular Communication

HSC-2 cells were cultured in a 60 mm cell culture dish overnight At 90% cell confluence, rhTGF-β1 (final concentration 10 ng/ml) or carbenoxolone (final concentration 40 µM) was added and incubated for

6 hours For the control treatment, only PBS was given to the cells Alternatively, BIM I (final concentration 5 µM) was added 30 min before the addition of rhTGF-β1 After a brief rinse in PBS, cells were incubated in D-MEM without phenol red, containing 5,6-carboxyfluorescein diacetate (Research Organics, USA) at a final concentration of 50 µg/ml and incubated

in a 37°C humidified incubator for 30 min The cells were then rinsed twice

with PBS and D-MEM without phenol red was added before proceeding with the Fluorescence Recovery After Photobleaching (FRAP) assay using the FRAP application included in the software package of the LEICA TCS SP2 equipped with the DM6000 (LEICA, Germany) A 63X immersed objective (Leica HCX APO L U-V-I 63x/0.90 Water UV) was used The argon laser at

488 nm was used for excitation and the fluorescence signal was captured between 500 and 535 nm The conditions were as follows: 5 pre-bleach scans

at 10% laser power, 40 bleach scans at 100% laser power followed by 60 bleach scans at 15 seconds intervals During the bleaching period, the Zoom mode was used to bleach a single cell (target cell) defined in a region of interest (ROI) All data were corrected for photobleaching during post-bleach acquisition using the whole scanned area The time constant of recovery, tau (τ), was estimated by fitting the corrected experimental data (OriginPro 7 SR4, OriginLab USA) to the following function: F(t)=F0 +(F∞ −F0 )( 1−e−t/τ), with F(t) being the corrected fluorescence intensity and F∞

post-τ

/1

=

k

being the asymptotic value of the fluorescence intensity The transfer constant (k) was calculated from and normalized by dividing by the number of cells in contact with the target cell The fluorescence recovery for each cell was about 50%

2.7 Electrophoretic Mobility Shift Assay

Based on the rat Cx43 gene (NW_001084790), a biotinylated stranded oligonucleotide probe containing the Snai1 consensus sequence (underlined), 5’-TGCTCAACCCAGTCAGGTGATGCCTGAACAAA-3’,

was synthesized (Research Biolabs, Singapore) In the mutated stranded oligonucleotide probe, the Snai1 consensus sequence was changed to

double-CAGGAA

2.8 Snai1 and Connexin 43 siRNAs Transfection

Nuclear protein extract was obtained using the NE-PER Nuclear and Cytoplasmic Extraction kit (Pierce, USA) The electrophoretic mobility shift assay was performed using reagents from the Snai1 kit according to its protocol (AY1398, Panomics, USA) Briefly, 5 µg nuclear protein extract was incubated in a reaction mixture consisting of 1 µl poly d(I-C), 2 µl 5X binding buffer and nuclease-free water for 5 min before the addition of 1 µl probe (stock 20 nM) The total reaction volume was 10 µl For the competition assay, 2 µl unlabeled probe (stock 2 µΜ) was added 5 min prior to the addition of the labeled probe The reaction was incubated in a thermocycler at 15°C for 30 min The samples were separated in a 6% non-denaturing polyacrylamide gel (Invitrogen) at 120 V for 50 min in a 4˚C cold room and

transferred onto a nylon membrane

The membrane was blocked in 1X blocking buffer for 15 min with gentle shaking This was followed by the addition of 20 µl Streptavidin-conjugated horse radish peroxidase enzyme and a further incubation of

15 min The bands were visualized by exposing the membrane to X-ray film after incubation with the substrate for 5 min All incubations were carried out

at room temperature

The cells were seeded on the day of transfection 1-2 X 106 HSC-2 cells were seeded in 100 mm cell culture dishes and incubated at 37°C The

siRNA was added at a final concentration of 10 nM to 1 ml of D-MEM without serum, followed by 120 µl of HiPerfect transfection reagent (Qiagen, Germany) and incubated for 10 min The siRNA/transfection reagent solution was then added drop-wise to the cells and incubated for 24 or 48 hours As mock control, only HiPerfect reagent was added to the cells All siRNAs used were purchased from Qiagen (Germany) The sequences were depicted in table 2.1

Table 2.1 Sequences of siRNAs used for transfection

2.9 Cell Counting

For cell counting experiments, the medium was aspirated and the cells were washed once with PBS and detached using trypsin/EDTA Complete detachment of the cells was observed under the light microscope and D-MEM was added to quench the reaction Following centrifugation at 800 rpm for

4 min and subsequent aspiration of the solution, the cell pellet was resuspended in 1 ml D-MEM containing 0.05% ethylenediaminetetraacetic acid and ready for counting The cells were counted using the forward scatter function of the GUAVA PCA-96 (Guava Technologies, CA, USA)

2.10 Statistical Analysis

All quantitative results were presented as mean ± SD Experimental

data were analyzed using two-tailed Student’s t-test assuming equal variances

and One-Way ANOVA with Scheffé’s Post-Hoc test where applicable The criterion for data significance is a P-value < 0.05 The P-values presented in

the figure legends are based on the Student’s t-test, unless otherwise stated

10 hours Real-time PCR data showed that 1 ng/ml and 10 ng/ml rhTGF-β1 led to a 30% and 45% decrease of Cx43 transcripts, respectively in the HSC-2 cells (Fig 3.1)

Fig 3.1 Exogenous rhTGF-β1 suppressed Cx43 mRNA in HSCs

HSC-2 and 10 days in vitro activated primary HSCs (pHSCs) were treated

with 1 ng/ml and 10 ng/ml rhTGF-β1 for 10 hours prior to Cx43 mRNA analysis The mRNA expression of Cx43 was obtained by quantitative real-time PCR and the data were analyzed as fold change relative to the control (PBS-treated) The data represent the mean ± SD of three independent experiments (* P < 0.05, ** P < 0.005)

Besides studying the TGF-β effects on an established HSC cell line, primary HSCs were also subjected to rhTGF-β1 treatment In this case, primary HSCs were isolated from rat livers and cultured for 10 days in tissue culture flasks Previous studies have shown that cultivation of primary HSCs

on plastic induces the activation of these cells in vitro (Friedman et al., 1989),

a phenotype similar to that of the HSC-2 cell line As can be seen from Figure 3.1, 1 ng/ml and 10 ng/ml rhTGF-β1 caused the Cx43 mRNA to

decrease by 8% and 10%, respectively in 10 days in vitro activated primary

HSCs (pHSCs)

In addition, the ability of rhTGF-β1 to induce changes of Cx43 protein level was investigated The Western blot data of HSC-2 and pHSCs for Cx43 were analyzed and quantified following rhTGF-β1 treatment The results showed that 1 ng/ml and 10 ng/ml rhTGF-β1 reduced the Cx43 expression by 20% and 30% in HSC-2, as well as 23% and 39% in pHSCs, respectively (Fig 3.2)

A

B

Fig 3.2 Exogenous rhTGF-β1 down-regulated Cx43 expression in HSCs

(A) HSC-2 cells and 10 days in vitro activated primary HSCs (pHSCs) were

treated with 1 ng/ml and 10 ng/ml rhTGF-β1 for 24 hours before protein analysis Ten micrograms total protein was applied for Cx43 analysis in Western blot The β-actin expression was shown as the loading control

A representative blot for each cell source is shown (B) Quantification of Western blots depicting Cx43 expression in HSC-2 cells and pHSCs after rhTGF-β1 treatment The band intensities were estimated using the ImageJ software and normalized against β-actin The graph represents the expression

of Cx43 in the treatments relative to control The data represent the average ±

SD of two to three independent experiments (* P < 0.05, ** P < 0.005)

3.2 Effect of Exogenous rhTGF- β1 on the Phosphorylation of Cx43 in HSCs

To determine the impact of exogenous rhTGF-β1 treatment on the

phosphorylation of Cx43, the HSC-2 cells were subjected to rhTGF-β1 for

6 hours An increase in the phosphorylation of Cx43 at the serine 368 residue

(pCx43 S368) was observed (Fig 3.3) More specifically, there was an

increase in the proportion of pCx43 S368 in the total (decreasing) pool of

Cx43 The authenticity of the pCx43 band was validated by its disappearance

after λ-phosphatase treatment (Fig 3.3)

Fig 3.3 Exogenous rhTGF-β1 affected the phosphorylation of Cx43 in

HSCs

HSC-2 cells were treated with 10 ng/ml rhTGF-β1 for 6 hours before the cells

were harvested for total cell lysate Ten and forty micrograms total protein

was applied for the Western blot analysis of the expression of Cx43 and Cx43

phosphorylated at the serine residue 368 (pCx43 S368), respectively

A representative blot is shown for one of three independent experiments The

expression of pCx43 S368 was normalized against Cx43 using the ImageJ

software The graph shows an increase in the pCx43 S368 expression in

rhTGF-β1-treated samples in comparison to PBS-treated control The data

represent the average ± SD of three independent experiments (** P < 0.005)

λ-phosphatase treatment eradicates the pCx43 S368 band, thereby confirming

the specificity of the antibody used to detect pCx43 S368

3.3 The PKC Pathway is implicated in the rhTGF- β1 Induction of Cx43 Phosphorylation at Serine 368 in HSCs

Several studies have shown that the phosphorylation of Cx43 at serine

368 may be mediated by protein kinase C (PKC) (Husoy et al., 2001; Solan et al., 2003) Consequently, experiments were performed to investigate whether

the PKC pathway is involved in the increase of the pCx43 S368 caused by rhTGF-β1 Indeed, pre-treatment of the HSC-2 cells with a PKC inhibitor, BIM I, followed by rhTGF-β1 reduced the phosphorylation of Cx43 at serine

of two independent experiments is shown The band intensities were estimated using the ImageJ software The numbers represent the intensity of the bands relative to control

3.4 Distribution of Cx43 and pCx43 S368 in the HSCs

Earlier, Musil and Goodenough (1991) demonstrated that unphosphorylated Cx43 was transported to the plasma membrane, which was then followed by the transient phosphorylation of Cx43 and the subsequent assembly of functional gap junctions Following the observation that rhTGF-β1 can induce the phosphorylation of Cx43 in the HSCs, the question arose as to whether this would affect the trafficking of Cx43 to the membrane For this reason, immunofluorescence staining of Cx43 and pCx43 S368 was used to study the cellular distribution of the non-phosphorylated Cx43 and pCx43 S368 in HSC-2 cells As expected in view of its role in forming gap junctions at the membrane, Cx43 was, to a great extent, distributed along the membrane (Fig 3.5, top panel, arrows) On the other hand, pCx43 S368 showed a more diffused or spotted staining in the cytoplasm with some membrane localization (Fig 3.5, middle panel, arrows)

Fig 3.5 Distribution of non-phosphorylated Cx43 and pCx43 S368 in HSC-2 cells as visualized by immunofluorescence staining

HSC-2 cells on cover slips were stained with Cx43 and pCx43 S368 antibodies for immunofluorescence studies Cx43 was mostly localized in the membrane (top panel), while pCx43 S368 showed some membrane and, for the most part, cytosolic staining (middle panel) The bottom image showed the cells stained with the secondary antibody alone (control)

3.5 Effect of rhTGF- β1 on the Gap Junction Intercellular Communication between HSCs

Cx43 is the major gap junction protein expressed in the HSCs and has

been shown to form functional gap junctions (Fischer et al., 2005) The FRAP technique (Abbaci et al., 2007) was the method of choice to analyze the

gap-GJIC between HSCs In order to show that this technique was properly executed, we illustrated that there was no spontaneous recovery of fluorescence in an isolated bleached cell (Fig 3.6A, arrow), whereas a contacting cell recovered about 50% of its fluorescence (Fig 3.6B, arrow) This result confirmed that we were indeed measuring the transfer of dye from unbleached cells to a bleached cell via gap junctions and not a recovery of the fluorescence signal in the bleached cell as such Figure 3.6C is a representative graph depicting the recovery of fluorescence in a single bleached cell over

15 minutes after normalization for laser-induced photobleaching The red line illustrated the normalized experimental data given in the graph fitted to the function The fitted function was used to estimate the normalized transfer rate (k) of the dye

Carbenoxolone is an established GJIC inhibitor (Doll et al., 1968) In

our case, carbenoxolone reduced the dye transfer rate (k) to almost 50% (Fig 3.6D), serving as a positive control for the reliability of the gap-FRAP technique to measure changes in GJIC Consequently, the transfer rate of the fluorescence dye 5,6-carboxyfluorescein diacetate was found to be significantly lower in rhTGF-β1-treated HSCs (Fig 3.6D), implying that GJIC was decreased in these cells in comparison to PBS-treated HSCs (control)

Furthermore, the rhTGF-β1-induced down-regulation of GJIC was attenuated when the cells were treated with BIM I, a PKC inhibitor, prior to the addition

of rhTGF-β1 (Fig 3.6D)