University of Southern California Dissertations and Theses

Histone deacetylase 4 represses matrix metalloproteinases in myofibroblastic hepatic stellate cells

(USC Thesis Other)

Histone deacetylase 4 represses matrix metalloproteinases in myofibroblastic hepatic stellate cells

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
HISTONE DEACETYLASE 4 REPRESSES MATRIX METALLOPROTEINASES IN
MYOFIBROBLASTIC HEPATIC STELLATE CELLS

by
Lan Qin

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(SYSTEMS BIOLOGY AND DISEASE)

May 2010

Copyright 2010 Lan Qin
ii

Dedication
To my dearest granddad, Mr. Chao Pu, and to my cousin, Cheng Pu, who live in my heart
for ever.
I could not have made it through this journey without the support of my family. Daddy,
Mommy, Xu and little Joyce, you gave me unconditional love when I felt lost or hopeless.
You also provide me a warm harbor when I felt tired. I am so grateful and proud to be in
the family.
To my friends Miao, Jeff, Mingshu, Jingjing, Chunli, Haitao and Xinhua for sharing the
good times with me, helping me through the bad times and for believing in me.
iii

Acknowledgements
I would like to thank my mentor, Dr. Yuan Ping Han, who guided me through the last
five years, for giving me so many learning opportunities and the challenges to become a
better person than when I started.
I would also like to thank my dissertation committee, Dr. Hidekazu Tsukamoto, Dr.
Ebrahim Zandi, for the guidance to follow the project and the great support of my
education.
I also owe thanks to Dr. Alicia McDonough, for the unyielding encouragement and for
providing the inspirational program.
iv

Table of Contents
DEDICATION ii
ACKNOWLEDGEMENTS iii
ABSTRACT viii
Chapter 1: Introduction 1
1.1 The liver and liver diseases 1
1.1.1 An overview 1
1.1.2 Liver fibrosis 4
1.2 Matrix Metalloproteinases 5
1.2.1 An overview 5
1.2.2 Regulations of MMP expressions 7
1.2.3 MMPs in liver fibrosis 8
1.3 Hepatic Stellate Cells 9
1.3.1 An overview 9
1.3.2 Cytokines in HSC activation 10
1.3.3 The reciprocal influence between ECM and HSC 14
1.3.4 Transcriptional controls of HSCs 15
1.3.5 Epigenetic controls of HSCs 16
1.4 Histone acetylation 18
1.4.1 An overview 18
1.4.2 Histone deacetylases 19
1.4.3 Histone deacetylase inhibitors 22

Chapter 2: Results 23
2.1 MMP-9 and MMP-13 are silenced in the fibrotic liver 23
2.1.1 Injury induced MMPs are repressed in the fibrotic liver 23
2.1.2 HSCs are the major source of MMP-9 and MMP-13 in
acute liver injury
27
2.2 MMP-9 and MMP-13 are silenced in myofibroblastic HSCs 29
2.3 Major signal pathways are as active in mHSCs as those in qHSCs 39
2.4 Dual phase kinetics of MMP expression in HSCs 41
2.5 Recruitment of c-Jun to MMP promoters is impaired in mHSCs 43
2.6 Global histone acetylation decreases in HSC trans-differentiation 48
2.7 Histone acetylation on MMP promoter is lower in mHSCs 50
2.8 HDAC4 accumulates during HSC trans-differentiation 53
2.9 HDAC4 suppresses MMP expression 56

v

Chapter 3: Discussion and future direction 59
3.1 MMP silencing in activated HSCs 59
3.2 MMP production in HSC activation and liver fibrosis 62
3.3 Epigenetic regulation in HSC trans-differentiation 66
3.4 Studies of HSCs and the implication in the treatment of liver
diseases
70
3.5 Experimental obstacles of in vitro HSC studies 74
3.6 Future direction 75

Chapter 4: Materials and Methods 78
Bibliography 87

vi

List of Tables
Table 1: Groups of MMPs 7
Table 2: primers for cloning of MMP-9 and MMP-13 promoter from rat
genomic DNA

81
Table 3: Primers for ChIP assay 83
Table 4: Primers for quantitative real-time PCR 84
vii

List of Figures
Figure 1: General domain arrangement of MMPs 6
Figure 2: Histone acetylation and gene expression 19
Figure 3: Induction of liver fibrosis in a mouse model 24
Figure 4: MMPs were induced in acute liver injury but repressed in liver
fibrosis

26
Figure 5: Hepatic stellate cells are the major source of MMPs in liver injury 28
Figure 6: In vitro HSC activation 31
Figure 7: Titration of IL-1  in HSCs cultured in type I collagen 34
Figure 8: Repression of MMPs genes during HSC trans-differentiation 37
Figure 9: RNA polymerase II recruitment was abrogated in myofibroblastic
HSCs
39
Figure 10: The signal pathways essential for MMP expression were intact in
myofibroblastic HSCs
41
Figure 11: Dual phase kinetics of MMP transcription in HSCs 43
Figure 12: Time courses of c-Jun recruitment to rat MMP-9 and MMP-13
promoters in quiescent and myofibroblastic HSCs
45
Figure 13: Global reduction of histone acetylation during HSC
trans-differentiation
49
Figure 14: Impaired histone acetylation in MMP gene promoters in
myofibroblastic HSCs
51
Figure 15: Accumulation of HDAC4 during HSC trans-differentiation 54
Figure 16: Ectopic expression of HdAC4 in quiescent HSCs suppressed the
MMP transcription
57
viii

Abstract
It is well established that matrix metalloproteinases (MMPs) play important roles in
tissue injury, cell differentiation, and cancer metastasis. In liver injury, hepatic stellate
cells (HSCs) express high level of MMPs and undergo trans-differentiation to become
myofibroblast like cells. However, largely unknown is that how MMP genes are
repressed or silenced in tissue fibrosis to favor ECM accumulation. In this study, it was
investigated how MMP genes were progressively repressed in fibrotic liver and
myofibroblastic HSCs. First, it was shown that upon interleukin-1 (IL-1) stimulation, the
major signaling pathways that are essential for MMP expression were as active in
myofibroblastic HSCs as in quiescent HSCs, indicating the repression was attained at
chromatin levels. Indeed, it was found that in myofibroblastic HSCs both MMP-9 and
MMP-13 genes were less accessible for transcriptional factors, which was associated with
impaired histone acetylation and RNA polymerase II recruitment to the promoters.
Further, it was observed that during HSC trans-differentiation, the class II histone
deacetylase HDAC4 accumulated, in accordance with global reduction of histone
acetylation. To demonstrate a causal relationship of HDAC4 elevation in MMP gene
repression, HDAC4 was ectopically expressed in quiescent HSCs, which resulted in
sufficient repression of MMP promoter activities as well as endogenous MMP-9 protein
expression. Thus, a mechanism was uncovered about how MMP genes are epigenetically
silenced in tissue fibrogenesis partially through HDAC4 accumulation.
1

Chapter 1: Introduction
1.1 Liver and liver diseases
1.1.1 An overview
Liver anatomy
As the largest internal organ of vertebrates, the liver is necessary for survival. It is located
in the right upper quadrant of the abdomen, just below the diaphragm. The liver is a
triangular organ with dark red color due to the rich blood flow through it. The liver is
covered entirely by visceral peritoneum except the area where it is connected to the
diaphragm. Visceral peritoneum, a double layered membrane, reduces friction of the liver
against other organs, and it folds back on itself to form the falciform ligament and the left
and right triangular ligaments. The ligaments attach the upper surface of the liver to the
diaphragm and abdominal wall, and the under surface to the stomach and the duodenum.
The liver is divided into four lobes by traditional gross anatomy. On the anterior side, the
falciform ligament divides the liver into a left and a right anatomical lobe. On the visceral
surface, there are additional two lobes called the caudate lobe and the quadrate lobe
which is located below the caudate lobe. Each of these lobes is the network of hepatic
lobules, which looks like a six-sided cylinder. The hepatic lobule is surrounded by
connective tissues and bundles of vessels including the portal vein, hepatic arteries and a
bile duct. In the center of the lobule runs the central vein, which is surrounded by cords
2

of liver cells that radiate out in all directions. And between these cords are thin walled
blood vessels called sinusoids.
Liver blood supplies
There are two blood supplies to the liver: venous blood drained from the spleen and
gastrointestinal tract enters the liver through the portal vein, constituting about 75% of
the total liver blood supply; the hepatic artery carries the other 25% of the blood supply
from aorta. The blood flows through the sinusoids, where nutrient and oxygen exchanges
take place between hepatocytes and the blood. In the sinusoids, the plasma filters into the
space of Disse and contribute to lymph. After sinusoids, the blood empties into the
central vein which coalesces into hepatic veins that leaves the liver and drains into the
inferior vena cava.
Liver functions
The liver is a vital organ that possesses various functions. Due to the large amount of
blood flowing through it, the liver is an important organ for blood circulation and
filtration. It also plays a major function in regulating blood glucose concentration by
storing glucose in the form of glycogen and releasing glucose as needed, through the
process of glycogenesis and glycolysis respectively. Besides, the liver is a critical site for
the synthesis and storage of amino acid, proteins and vitamins. For example, it produces
albumin which is crucial in regulating blood osmotic pressure, and clotting factor that
stops bleeding. In addition, the liver performs an essential role in detoxification as
3

processing ammonia, bilirubin, drugs and toxins. Last but not least, the liver excretes bile
that emulsifies fats required for the digestion.
Cellular components of the liver
The liver is composed of parenchymal and non-parenchymal cells. Hepatocytes,
constituting 60% to 80% of the liver mass, are liver parenchymal cells that perform the
major liver functions. Liver non-parenchymal cells are also of crucial importance such as
participating in innate immune response and storing vitamin A, and they are made up of
sinusoidal endothelial cells, Kupffer cells, blood cells and hepatic stellate cells.
Hepatocytes are arranged into plates separated by sinusoids, which displays fenestrated
endothelial cell lining. The hepatocytes are separated from sinusoids by the space of
Disse. Kupffer cells are specialized macrophage located in the liver, and they are
scattered between endothelial cells. Hepatic stellate cells reside in the space of Disse, and
have been identified with multiple physiological functions as detailed later.
Liver is able to regenerate after injury. Hepatocytes would re-enter the cell cycle, which
is the major contribution to liver regeneration of lost tissue. In addition, hepatic oval cells,
bipotential stem cells, were also reported to differentiate into hepatocyte and thus to help
liver regenerate as well (73).
The essential role of liver in human body makes it significant to study the underlying
mechanism of liver diseases as well as the potential treatment. Unfortunately, so far liver
transplantation has been the only option for patients with irreversible liver failure (6).
4

And this procedure was made imperfect by limited availability of donors and the risk of
postoperative complications. Therefore, researches are still going on to address the
pathological events in liver diseases such as fibrosis and subsequent cirrhosis as well as
liver cancer. And the ultimate goal is to find an effective way to treat liver diseases on the
cellular or even molecular level.
1.1.2 Liver fibrosis
Liver fibrosis is a representation of chronic wound repair after liver injury such as virus
infection and toxic damage. Liver fibrosis may result in liver cirrhosis, which causes
hepatocellular dysfunction and portal hypertension, resulting in the increased intrahepatic
resistance to blood flow. Liver dysfunction is one of the most serious medical problems,
which usually results in death if without efficient treatment. The main causes of liver
fibrosis in industrialized countries include chronic HCV infection, alcohol abuse, and
non-alcoholic steatohepatitis (NASH) (6).
Wound healing proceeds via a series of overlapping events including inflammation, re-
epithelialization, formation of granulation tissue, matrix and tissue remodeling. Failure to
resolve the inflammation can lead to chronic nonhealing wound, while uncontrolled
matrix accumulation leads to excess scar formation and fibrosis.
After acute liver injury, hepatocytes regenerate and replace the necrotic or apoptotic cells.
This process is associated with an inflammatory response and a limited deposition of
extracellular matrix (ECM). If the hepatic injury persists, then eventually the liver
5

regeneration fails, and hepatocytes are substituted with abundant ECM, including fibrillar
collagen (24). In the healthy liver, the collagen is limited to the capsule around the big
vessel and in the portal areas, and type I and type III collagen are the major components
(18). After fibrogenic injury, both type I and type III collagen increase while type I
become more abundant. The accumulation of fibrillar collagen results in the formation of
nodules encapsulated by fibrillar scar matrix (92). This deposition of fibrillar collagen
mainly happens in the space of Disse. As the result, sinusoidal endothelia lose the
original fenestration, which blocks nutrient exchange between blood and hepatocytes,
leading to impaired metabolic function of hepatocytes. As the fibrogenesis proceeds, a
bridging fibrosis connecting portal spaces to central vein forms, which causes
angiogenesis and shunt formation to compensate for the increasing resistance to the blood
flow. In summary, liver fibrosis is the pathological disorder of ECM turnover, governed
by a group of enzymes named matrix metalloproteinases and their inhibitors.
1.2 Matrix Metalloproteinases
1.2.1 An overview
Matrix metalloproteinases (MMPs) are zinc dependent proteinases which play important
roles in tissue remodeling. They were first described by Jerome Gross and Charles
Lapiere in 1962 (34). The researchers put the tadpole fin skin into a collagen matrix and
observed collagenase activity that result in the lysis of the collagen substrate. And the
enzyme was named interstitial collagenase as MMP-1. Later in 1968, Jerome Gross group
6

isolated MMP-1 from human skin (23), and then in 1971 the same group identified that
this enzyme was synthesized as zymogen (37).
Most MMPs are secreted proteins while there are a few membrane anchored forms
termed as “MT-MMP”. MMPs share three common domains including the pro-peptide,
the catalytic domain and the haemopexin-like C-terminal domain which is linked to the
catalytic domain by a flexible hinge region. Zinc is essential for MMPs’ activities and
bound by three histidine residues in a conserved sequence as HExxHxxGxxH of the
catalytic domain (12). The pro-peptide contain the cysteine residue in a conserved
sequence as PRCGxPD which interact with the zinc and thus keep the enzyme inactive,
described as “cysteine switch” (87),(95). Only after the pre-domain is processed, can
MMPs gain the ECM degrading activity. The C-terminal domain of MMPs determines
the substrate specificity and is the site interacting with tissue inhibitor of
metalloproteinases (TIMPs). In addition, a trans-membrane domain exists in MT-MMPs
that anchors those enzymes to the cell surface. (Fig. 1)

Figure 1: General domain arrangement of MMPs.
Based on the substrate specificity and partly on the cellular localization, MMPs are
commonly grouped into collagenases, gelatinases, stromelysins and the membrane type
MMPs (Table 1). Collagenases are able to degrade triple-helical fibrillar collagens, which
7

are the major components of bone and cartilage. And gelatinases mainly process type IV
collagen and gelatin. Stromelysins can cleave various ECM but not the triple-helical
fibrillar collagens. In addition, other MMPs do exist but cannot be grouped into this
classification.
Table 1: Groups of MMPs
Collagenase MMP-1, MMP-8, MMP-13, MMP-18
Gelatinases MMP-2, MMP-9
Stromelysins MMP-3, MMP-10, MMP-11, MMP-27
Membrane type MMP-14, MMP-15, MMP-16,MMP-17, MMP-24, MMP-25

Due to their central roles in tissue breakdown, MMPs are of crucial importance in
embryonic development, tissue injury, cell differentiation and cancer metastasis. In
addition to the ECM modulating functions, MMPs have been suggested of versatile roles
such as activating growth factors (110),(90)and promoting cell migration (38),(54).
1.2.2 Regulations of MMP expressions
Dormant in developed tissues, many MMPs would be turned on by injury signals. Studies
have indicated that the MMP gene expression is regulated on multiple layers including
transcriptional, post-transcriptional control and epigenetic modification. There are several
cis-elements on the MMP promoter, which allow for the regulation of the MMP gene
expression by trans-activators such as AP-1 and NF B. Similarity can be found among
MMP gene promoters(106), indicating a possible common regulating mechanism. Post-
transcriptional regulation can also be exerted to MMP genes. Expressions of MMP-1 and
MMP-13 have been reported to be regulated by microRNA222 and microRNA27b,
8

respectively (2),(55). Additionally, epigenetic regulation has also been suggested to be
involved in MMP regulation. Chicoine et al reported an inverse correlation between
methylation level of MMP-9 promoter and MMP-9 expression in lymphoma cells (17).
And they also found that DNA methylation inhibitor increased MMP-9 mRNA and
protein secretion. Histone deacetylases (HDACs) have also been indicated in suppressing
MMP-9 gene (107),(58). The epigenetic control is not limited to MMP-9 only. For
instance, Olson group identified HDAC7 as a suppressor of MMP-10 as well (15).
1.2.3 MMPs in liver fibrosis
The competition between MMPs and their inhibitors TIMPs determines the final
outcome of ECM modification. MMPs/TIMPs imbalance would lead to pathological
disorders, including liver fibrosis. Upon liver injury, resident cells are activated and
various MMPs are produced. And among these cells HSCs have been of great interest in
recent years. According to Benyon and Arthur, upon liver injury HSCs get activated and
express a combination of MMPs and TIMPs, and the specific components are dependent
on the phase of the disease (8). In the early phase of liver injury, HSCs exhibit matrix-
degrading function by transiently expressing MMP-3, MMP-13 and uroplasminogen
activator. In the later stage of liver injury, HSCs continue producing MMPs which
degrade normal ECM but not the fibrillar collagens that accumulated in liver fibrosis.
Expressions of TIMPs are also increasing at the same time, which leads to global
inhibition of interstitial MMPs that process fibrillar collagens. At the resolution stage of
liver fibrosis, expressions of MMPs are changed again to remove the scar tissues. In
9

summary, both MMP expressions and activities are under strict control to reach precise
ECM turn over.
1.3 Hepatic stellate cells
1.3.1 An overview
Historically, hepatic stellate cells (HSCs) were described by Von Kupffer as liver
Sternzellen in 1876. The cells were stained by gold chloride that was used to reveal
nerves. However, it was erroneously recognized as phagocytes at that time, and had been
called Kupffer cells since. In 1928, Zimmerman identified dendritic perisinusoidal cells
and called the cells hepatic pericytes (30). Then in 1951, Toshio Ito reported fat storing
cells in human liver (45). In 1958, with silver impregnation method Suzuki found star-
shaped cells in the space of Disse (96). Later in 1971, by gold chloride staining, silver
impregnation method, vitamin A autofluorescence and electron microscopy, Kenjiro
Wake concluded that the Sternzallen described by Von Kupffer were the same cells as the
fat-storing cells reported by Ito and the interstitial cells described by Suzuki(97). Soon
after that, the first articles linked the cells to the pathogenesis of liver fibrosis
(65),(64),(48). In 1995, the cells started to be referred as hepatic stellate cells which has
been broadly used thereafter (1).
HSCs, the pericyte in liver, reside in the space of Disse between hepatocytes and
sinusoidal endothial cells. HSCs have been well known as a key cellular element in liver
fibrogenesis by modifying ECM (6),(81). In normal liver, HSCs are quiescent with small
10

cell body and slim extensions that enwrap endothelial liver sinusoids, and are the major
retinoid storage cells. The ECM in normal liver surrounding HSCs is basement
membrane like and is mainly composed of non-fibrillar collagen such as type IV collagen.
In response to liver injury, HSCs would undergo a process called “trans-differentiation”,
when star-shaped quiescent HSCs are converted to myofibroblast-like cells. During this
transition, HSCs secret large amounts of fibrillar collagen such as type I collagen, and
also produce MMPs to degrade the loose ECM present in normal liver, resulting in
excessive ECM deposition resulting in fibrosis(3, 59, 78). In addition, other
characteristics of HSC trans-differentiation include loss of the vitamin A droplets,
formation of stress fibers and gain of the ability of contraction (82).
1.3.2 Cytokines in HSC activation
An overview
The tissue repair process in wound healing is mediated in large part by interacting
molecules, primarily cytokines, which motivate and integrate the cellular activities
underscoring inflammation and healing. A lot of studies have been performed about the
function of different cytokines in HSC quiescence as well as activation. For example,
PPAR  has been known to be able to keep HSC quiescent(69), while TGF  has been
suggested to play a pivotal role in initiating, promoting and progression of HSC trans-
differentiation(31),(70),(53),(77). In addition, IL-1  has also been reported to induce
both morphological and biochemical changes in HSCs (36). Moreover, platelet derived
11

growth factor (PDGF) was shown to induce the release of retinol storage in HSCs(28), as
well as to stimulate HSC proliferation (50) while a PDGF receptor tyrosine kinase
inhibitor attenuates HSC growth (28).
Coordination of these cytokines depends on a proper sequential order to exert their
specific functions. In our lab it was found that when applied together to isolated HSCs,
IL-1  and TGF  antagonized each other in terms of MMP production and stress fiber
formation etc. However, if IL-1  was applied first followed by TGF treatment,
mimicking the in vivo temporal sequence, characteristic HSC activation would be clearly
observed. This is consistent with the fact that IL-1  is a pro-inflammatory cytokine that
functions in initiation of fibrosis while as a fibrotic cytokine TGF  works in the buildup
of fibrosis.
The function of cytokines can be both paracrine and autocrine. When it comes the source
of cytokines in liver injury have Kupffer cells gained most attention. Kupffer cells
internalize endotoxin like LPS, and consequently start transcription of pro-inflammatory
cytokines such as IL-1  and Tumor necrosis factor alpha (TNF ), which then stimulate
HSC activation. In addition, Kupffer cells have also been identified as the major producer
of TGF  1 (11). Besides the resident Kupffer cells, other macrophages are also recruited
from circulating monocytes and may become a substantial source of cytokines in liver
injury. Sinusoidal endothelia cells are among the first cell populations in the liver to
encounter gut-derived endotoxin in blood, and they are also able to express acute phase
cytokines such as IL-1 and IL-6(25, 52). In the in vitro culture, sinusoidal endothelial
12

cells produce those two cytokines in absence of stimulation in a time-dependent manner
(25). At the same time, HSCs do produce cytokines as well. Bissell et al reported that
HSCs expressed more TGF 3 than Kupffer cells, hepatocytes and sinusoidal endothelial
cells (11). And Gorbig et al identified an autocrine pathway of adrenomedullin in human
HSCs (33). Additionally, HSCs have been reported to secrete insulin like growth factor -
1 (75). In our lab we also found rat HSCs produce IL-1  in an autocrine loop. Besides
those liver resident cells, other cells such as platelets and lymphocytes are appreciable
sources of cytokines as well (92).
Interleukin-1 in HSC activation
Interleukin-1 (IL-1) is a pro-inflammatory cytokine that acts as the mediator to increase
host defense as well as inflammatory reactions. It’s produced by macrophages,
monocytes and other cell types. Despite its contribution to disease development, IL-1 can
also bring about protective and beneficial changes. IL-1 pre-treatment ameliorates
multiple diseases such as inflammatory bowel disease and contact dermatitis. IL-1 given
before a lethal challenge would prevent death, but does not afford protection if
administrated after the challenge initiation (21).
IL-1 family includes IL-1 , IL-1 and IL-1 receptor antagonist, which are encoded by
three different genes that locate in chromosome 2 of human genome. IL-1   and IL-1
share limited homology in amino acid sequence. They are both produced as proproteins
and proteolytic cleavage is required for the maturation, which are accomplished by
13

calpain and caspase respectively (13),(14),(91). Although in general IL-1  and IL-1 
exert the same biological effects, differences do exist in their expressions and the induced
responses in some cells (10). IL-1  mRNA is readily to be translated, while most IL-1 
mRNA requires a second signal for translation. Moreover, proIL-1  possesses biological
functions, while proIL-1  needs to be proteolytically cleaved to get its activity,
suggesting a more strict control over IL-1 expression. All that findings suggest a more
strict control over IL-1 than IL-1 .
In our lab, IL-1  has been suggested to coordinate the transition from liver injury to
fibrosis, mainly by its effect on hepatic stellate cells. First, ameliorated liver damage and
reduced fibrogenesis were found in IL-1 receptor deficient mice (32). Second, IL-1  was
shown to be able to induce HSC activation evident by (1) promoting Vitamin A depletion
(unpublished data); (2) inducing formation of stress fibers and (3) expression of matrix
metalloproteinases (36), of which MMP-9 was identified as an important mediator in
liver injury (32),(108). Other groups also reported the regulation of IL-1  on MMPs
(86),(84),(79).
The study of IL-1  in triggering HSC trans-differentiation and the subsequent MMP-9
production has the following physiological relevance: as a pro-inflammatory cytokine,
IL-1  functions in the acute phase of liver injury, which activates HSCs. One of the
consequences is the production of MMP-9, a gelatinase that degrades type IV collagen
which is a major component of ECM surrounding HSCs in the healthy liver. Therefore,
14

clearance of this existing ECM provides space for further HSC trans-
differentiation/migration as well as fibrillar collagen deposition.
1.3.3 The reciprocal influence between Extracellular matrix and HSC
HSCs have been recognized as the major ECM producing cells in liver fibrosis. At the
same time, ECM also participates in HSC activation. When cultured on uncoated plastic
with FBS, HSCs would undergo spontaneous activation (27), which is a problem to study
quiescent HSCs in vitro but provides a culturing model for HSC trans-differentiation.
Friedman et al. reported that Matrigel coating would inhibit this trans-differentiation (27),
and so did three dimensional Matrigel culturing condition (36). Besides, Benyon group
revealed that Matrigel imposed inhibition of HSC proliferation (29), and that collagen
proteolysis was required for HSC proliferation (111). Moreover, HSCs’ responsiveness to
TGF  was modulated by ECM as well (20). Not only the chemical property, but also the
physical character of the surrounding environment would affect HSCs in vitro. Wells
group suggested matrix stiffness affected the phenotype of HSCs (103). In particular, the
acquisition of SMA fibers is dependent on the mechanical stiffness. Since the formation
of focal adhesion and reorganization of SMA was dependent on TGF  signaling (93), a
correlation is established between the functions of cytokine and ECM in HSC activation.
Not only producing ECM, HSCs are also one of the major sources of MMPs that process
the ECM. The MMP expression is under strict control to cope with specific physiological
situations, indicated by studies with primarily cultured HSCs. Take MMP-9 for example,
15

in response to inflammatory cytokine, HSCs cultured on uncoated plastic only express
limited quantity of pro-MMP-9, while three-dimensional collagen cultivation would
greatly boost the expression, as well as the maturation to active MMP-9 (36).
Furthermore, consistent with diminished MMP expression in the later stage of liver
fibrosis, activated HSCs lose their ability to express MMPs, which is beneficial for ECM
accumulation. In a word, HSC and ECM have a reciprocal influence on each other rather
than a one-way effect.
1.3.4 Transcriptional controls of HSCs
Given the pivotal role of HSCs in liver fibrogenesis, the molecular mechanisms
underlining the maintenance of cell quiescence and the induction of cell activation have
been extensively investigated. Most of the studies were initially focused on characterizing
the interaction between transcription factors and gene promoters. In general, the
transcription activities were compared between quiescent and active HSCs for a number
of target genes, among which type I collagen and SMA have received the most
intensive evaluation.
Several transcription factors have been indicated as substantial modulators of HSCs’
destiny and liver physiology. Inhibitor of KappaB kinase was reported to stimulate HSC
apoptosis (72). PPAR-  was shown to inhibit HSC proliferation (105) or even reverse
activated HSCs to the quiescent state, indicated by morphological changes as well as by
inhibited expression of SMA, type I collagen and TGF  (39). Additionally, mice
16

deficient in Lim/homeobox protein 2 (Lhx2) showed spontaneous liver fibrosis and
activated HSCs (98), suggesting Lhx2 might be a potential inhibitor of HSC activation.
Myocyte enhancer factor 2 (MEF2) was reported to promote HSC activation by
increasing SMA expression, activating type I collagen promoter activity and stimulating
HSC proliferation (101). Diehl group reported that Hedgehog (Hh) signaling regulated
activation and viability of HSCs (85). In that study, Hh was found active in isolated
stellate cells, while Hh neutralizing antibodies and an Hh pharmacologic inhibitor would
inhibit HSC activation. In summary, the HSCs’ behavior is controlled by widely
divergent regulatory pathways. While initial studies of HSCs focused on evaluating
transcriptional events, epigenetic mechanisms have emerged as the major determinant of
gene activation and repression.
1.3.5 Epigenetic controls of HSCs
The significance of epigenetic regulation, in particular histone acetylation and DNA
methylation, has been realized in HSCs. Geerts group has identified histone deacetylase
(HDAC) inhibitor trichostatin A (TSA) as an inhibitor of HSC activation (71),(83). Both
morphological and phenotypical evidence showed that TSA treatment in freshly isolate
quiescent HSCs abrogated the in vitro trans-differentiation. In detail, TSA treatment
impaired the formation and reorganization of actin filament, leading to deficient cell
migration. Besides, TSA also reduced collagen synthesis. However, due to the non-
specific effects of TSA and to the versatile functions of histone deacetylases, it is not
appropriate to conclude that histone de-acetylation induces HSC activation. Take alcohol
17

stimulated HSC activation and liver fibrosis for example, Kim and Shukla showed
ethanol increased histone H3 lysine 9 acetylation in hepatic stellate cells (49). Therefore,
care should be taken to clarify the finely tuned histone acetylation in HSCs, and to
illustrate the effect of HDAC inhibitors in HSC activation and liver fibrosis.
Another aspect of epigenetic control in HSCs is DNA methylation. As reported by Mann
group, DNA methylation inhibitor 5-aza-2'-deoxycytidine (5-azadC) blocked HSC trans-
differentiation (62). In the same report, it was shown that as IkB  promoter entered a
repressive chromatin structure, the gene expression was diminished in activated HSCs,
and that the 5-azadC treatment converted the promoter to an active state. Furthermore,
the study also suggested methyl CpG binding protein 2 (MeCP2) exerted this epigenetic
control through DNA methylation. A recent publication from Lu group identified
increased expression of methionine adenosyltransferase in HSC activation, which
catalyzes biosynthesis of S-adenosylmethionine (SAMe), the principle methyl donor.
However, it was also found that the intracellular SAMe levels were markedly decreased
during HSC activation, resulting in global hypomethylation, which seems odds with the
result from Mann group. Therefore, a gene-specific methylation pattern in HSCs that
relates to the activation process is required to address the discrepancy.

18

1.4 Histone acetylation
1.4.1 An overview
Nature has developed a strategy to keep the approximate 2 meter DNA in a single cell by
winding the DNA onto histones, forming nucleosomes. The core of nucleosome is a
histone octamer including two H2A-H2B dimmers and a H3-H4 tetramer, which wrap
146 base pairs of DNA (57). This assembled structure of histones and DNA is called
chromatin. For gene expression, the DNA needs to be unpacked from the nucleosome to
facilitate the loading of transcription apparatus, which is achieved by histone
modification. Specifically, histone H3 and H4 have long tails protruding from the
nucleosomes, which can be covalently modified including methylation, acetylation and
phosphorylation. And the outcome of those epigenetic events has been under intense
investigation.
In general, histone acetylation is linked to transcriptional activation. Core histones which
are positively charged at their N-terminus interact with negatively charged phosphate
groups of DNA. In histone acetylation, a negatively charged acetyl group from acetyl-
CoA is transferred to the conserved lysine residues on histone proteins by histone acetyl
transferases (HATs). This reaction neutralizes the positive charges normally present on
histones and thus reduces the affinity between histones and DNA, making the DNA more
accessible to transcription factors. On the other hand, histone deacetylation causes
19

chromatin condensation that leads to gene repression, which is catalyzed by histone
deacetylases as described below.
A
c
A
c
A
c
HAT HDAC
Transcription Repression
Transcription Activation

Figure 2: Histone acetylation and gene expression. On one hand, HATs catalyze
histone acetylation, which neutralizes the positive charges on histones and thus reduces
the affinity between histones and DNA. Therefore, the chromatin enters a loose structure
favoring gene expression. On the other hand, HDACs remove the acetyl group leading to
chromatin condensation and gene repression.
1.4.2 Histone deacetylases
Classification of HDACs
Histone deacetylases (HDAC) are enzymes that remove the acetyl group from histones
through hydrolysis, and consequently send the chromatin into the state of gene silencing.
Not only in the context of histones, HDACs have also been shown to function on other
non-histone proteins including p53, -tubulin and MyoD (46),(41),(60). There are 18
HDACs found in human and they are grouped into different classes according to the
homologue to yeast Rpd3, Hda1 and Sir2.
20

Class I HDACs are composed of HDAC1, -2, -3 and -8, which are related to yeast Rpd3.
Class I HDACs are expressed in all cell types. Except HDAC3, the other three members
are exclusively localized in the nucleus. HDAC3 is present in both nucleus and cytoplasm,
and can be membrane associated (56).
Class II HDACs include HDAC4, -5, -6, -7, -9 and -10, which are related to Hda1 and are
zinc-dependent enzymes. Among them, HDAC4, -5, -7 and -9 are further grouped into
class IIa, which possess an N-terminal extended motif as well as a C-terminal tail
flanking the catalytic domain. The remaining two, HDAC6 and -10 are grouped as Class
IIb, specific in that HDAC6 contains two catalytic domains and that HDAC10 lacks the
extended N-terminal motif.
Sir2 family of HDACs is NAD
+
- dependent deacetylases. They remove the acetyl group
from the protein and add it to the ADP-ribose part of NAD
+
to form O-acetyl-ADP-ribose.
Class II HDACs
With a nuclear localization signal (NLS) around N-terminus while a nuclear exportation
signal (NES) near C-terminus (100),(66), Class IIa HDACs show a unique characteristic
in their localization that they are shuttling between cytoplasm and nucleus, providing a
potential target for regulation. Studies with HDAC4 have revealed that phosphorylation
by calcium/calmodulin-dependent kinase II (CaMKII) provides a docking site for protein
14-3-3, which exports HDAC4 out of nuclei (35, 47, 99). Moreover HDAC4 confers
CaMKII responsiveness to HDAC5 by oligomerization (5). In addition, it is reported that
21

nuclear HDAC4 can be sumoylated to increase its deacetylase activity (51). However, the
deacetylase function of HDAC4 is controversial: while in vitro studies suggested little
deacetylase activity in HDAC4, it was reported that HDAC4 does possess this function.
Meanwhile, continuous researches have revealed HDAC4 as a transcription co-repressor
through its N-terminal MEF binding motif.
The localization of class II HDACs have been suggested to correlate with cell
differentiation. Studies from Olson group discovered that HDAC5 resides in the nucleus
during the pre-differentiation stage of muscle cells and is transported to the cytoplasm
during differentiation. HDAC5 mutant which is resistant to CaMK mediated nuclear
export acts as a dominant inhibitor of skeletal myogenesis (66). Later in 2001, Miska et al
reported the relocalization of HDAC4 during myoblast differentiation (68). And both
HDAC4 and HDAC5 suppresses myogenic program by repressing myocyte enhancer
factor 2 (MEF2) dependent transcriptions. A study in the same period also suggested a
similar transportation pattern of HDAC7 in muscle differentiation (22). Surprisingly, it
was shown that MEF2 inhibition by HDAC7 is dependent on the N-terminal domain but
not involves the C-terminal deacetylase domain, indicating versatile functions of class II
HDACs. However, it is worth noting that at different stages of muscle cell differentiation,
HDAC4 differs from HDAC5 and HDAC7 in their subcellular localizations. So HDACs
might complement each other to fine tune the repression of gene expression in response
to specific signals and thus to control the sequence of gene expression in muscle
differentiation.
22

1.4.3 Histone deacetylase inhibitors
HDAC inhibitors (HDACis) function by blocking access to the active site of HDACs,
either reversibly or irreversibly. Among many known HDAC inhibitors, trichostatin A
(TSA) is one of, if not only, the most potent, which is effective at nanomolar
concentrations in vitro. TSA is an organic compound that originally serves as anti-fungal
antibiotic. It selectively inhibits class I and class II HDACs, but not class III HDACs.
Other HDACi include butyrate, trapoxin and etc.
HDACis have been broadly studied for their potential benefits in disease treatments, such
as their anti-proliferative and pro-apoptotic activities in cancer. HDACi treatments result
in elevated acetylation which in turn leads to increased expression of particular genes
while decreased expression of others. However, the relevant targeted HDACs in those
inhibitor treatments are still not clear. A recent study suggested that HDACi might
execute repression on HDAC1, which suppresses p21. Therefore, HDACi induces p21
expression that leads to apoptosis. As mentioned previously, Geerts group identified the
suppression of TSA on HSCs trans-differentiation (71),(83). Again, it is not known how
TSA represses those active genes in HSCs or what the targeted enzyme is in this case.
Nevertheless, this discovery has given us two hints that (1) histone acetylation should be
involved in HSC activation and (2) HDACi might be a potential treatment for liver
fibrosis.
23

Chapter 2: Results
2.1 MMP-9 and MMP-13 are silenced in the fibrotic liver
2.1.1 Injury induced MMPs are repressed in the fibrotic liver
Repression of MMP genes in liver fibrosis was confirmed in a mouse model. Mice were
divided into two groups as the acute injured group and the fibrotic group, with 6 mice in
each group. The acute injured group had been injected with saline twice a week for 8
weeks, while the fibrotic group had been on Thioacetamide (TAA) injection at the same
frequency for 8 weeks. Then both groups were challenged by another injection of TAA,
and 24 hours later livers were collected for the subsequent analysis.
In the acute injured group, hepatic damages were revealed by H&E staining as large scale
of necrosis and hemorrhage, while tissue breakdown was also evident in Sirus red
staining (Fig.3A). Induction of fibrosis in the TAA treated fibrotic group was confirmed
in Sirus red staining shown as red collagen positive areas (Fig.3A). The formation of
fibrosis was predominantly located in the area surrounding blood vessels, and seemed to
form a bridge between the ducts. In addition, quantitative real time PCR was also
performed to measure the mRNA level of interstitial type-I collagen. As seen in Fig.3B,
the mRNA level of type-I collagen was statistically significantly higher in the fibrotic
livers than that in the acute injured livers, which further validated the formation of
fibrosis.
24

final injection (TAA)
Acute injured Fibrotic
H&E staining Sirius red
400µm 400µm
400µm 400µm

0
2
4
6
Acute injured Fibrotic
Arbitrary Unit
Collagen I mRNA
*

Figure 3: Induction of liver fibrosis in a mouse model. Mice were injected with TAA
(fibrotic liver group) or saline (acute injured group) twice a week for 8 weeks (n=6 for
each group). One day prior to sacrifice, all mice were subjected to a final TAA challenge.
(A) Liver fibrosis was revealed by Sirius red staining, and liver injury was indicated by
H&E staining. (B) Liver fibrosis was also confirmed by increased expression of type-I
collagen as measured by quantitative real-time PCR. The expressions were normalized by
GAPDH mRNA and presented as the folds of mRNA level in fibrotic liver compared to
that in acute injury. The results are the mean values with standard deviations. *P < 0.05.

Then, gelatin zymography was performed to examine MMP-9 and MMP-2 activity in the
livers. It was previously observed in this lab that MMP-9 was absent from uninjured
normal liver, while MMP-2 was constitutively expressed at a low level. As clearly shown
A
B
25

in Fig.4A, a large amount of MMP-9 was induced in the acute injured livers (the identity
of MMP-9 in the zymography has been previously confirmed by western blotting
analysis). In contrast, no MMP-9 was detected in the fibrotic livers. However, a similar
low level of MMP-2 was present in both groups, which could be used as an internal
loading control. Besides protein activity, mRNA level of MMP-9 was measured by
quantitative real time PCR. Consistent with its protein expression pattern, MMP-9
mRNA was also greatly induced in the acute injured group, but not in the fibrotic groups
(about 90% reduction compared to the acute injured group) (Fig.4B). In summary, MMP-
9 was elicited by toxin induced acute liver injury, but was absent from livers with
existing fibrosis. An MMP activation cascade has been proposed in this lab as MMP-14
(membrane-type MMP) activates MMP-13 (type I collagenase), which in turn activates
MMP-9 (gelatinase). So the study was expanded to those other MMPs as well. By real-
time PCR, it was shown that similar to MMP-9, mRNA of both MMP-14 and MMP-13
was significantly reduced in the livers with established fibrosis (about 70% and 50%
reduction, respectively) compared to the acute injured livers. In addition, mRNA of
MMP-12 (macrophage elastase) was also decreased in the fibrotic livers at approximate
70% reduction. Through those findings that all the four injury type MMPs examined here
were suppressed in fibrotic liver, a common mechanism was suggested to regulate the
MMP gene expression and repression.
26

final injection (TAA)
MMP-9
MMP-2
Acute injured Fibrotic

0
0.5
1
1.5
2
Acute injured Fibrotic
Arbitrary Unit
MMP-9 mRNA
*

0
0.5
1
1.5
Acute injured Fibrotic
Arbitrary Unit
MMP-14 mRNA
* *

0
0.5
1
1.5
Acute injured Fibrotic
Arbitrary Unit
MMP-13 mRNA
*

0
0.5
1
1.5
Acute injured Fibrotic
Arbitrary Unit
MMP-12 mRNA
*

Figure 4: MMPs were induced in acute liver injury but repressed in liver fibrosis.
Liver injury and fibrosis were induced in mouse livers as described in Figure 5. (A)
Gelatinase activities in the liver tissues were accessed by zymography. (B) The mRNA
levels of MMP-9, MMP-12, MMP-13 and MMP-14 were measured by quantitative RT-
PCR and normalized by GAPDH.

A
B
27

2.1.2 Hepatic Stellate Cells are the major source of MMP-9 and MMP-13 in
acute liver injury
Hepatic stellate cells have been extensively investigated for their participations in liver
fibrogenesis. Residing in the space of Disse, the major site for the excessive ECM
deposition in fibrosis, HSCs play a crucial role in liver injury. Double immunostaining
was utilized here to measure HSCs’ contribution to the hepatic MMP expression in liver
injury. In Fig.5, HSCs were identified by positive desmin staining (green), a type III
intermediate filament that is a marker of HSCs. The cells exhibited cytoplasmic
extensions and condensed cell body. Consistent with previous findings in this lab, MMP-
9 and MMP-13 positive staining (red) were mostly co-localized with the desmin positive
cells, indicating HSCs as the major source of those two MMPs in the acute liver injury.
Also in agreement with previous results, both MMP-9 and MMP-13 staining revealed
decreased expression in the fibrotic liver compared with the acute injured liver.
In summary, MMP genes are induced by acute injury in healthy livers, but are repressed
in fibrotic liver. Such MMP gene suppression was on both mRNA level and protein level,
indicating a possibility of transcriptional repression. Further, HSCs are the major source
of such acute injury induced MMPs. Besides, HSCs in healthy livers are able to express
MMPs in response to toxin challenge, while the cells in fibrotic livers appear to lose such
capacity of MMP genes expression.
28

Acute injured liver Fibrotic liver
Desmin MMP-9 Merge
10µ m 10µ m 10µ m
10µ m 10µ m 10µ m

Acute injured liver Fibrotic liver
Desmin MMP-9 Merge
10µ m 10µ m 10µ m
10µ m 10µ m 10µ m

Figure 5. Hepatic stellate cells are the major source of MMPs in liver injury. Liver
injury and fibrosis were induced in mouse livers as described in Figure 5. Liver sections
were subjected to immuno-fluorescent staining for desmin (a marker for HSCs) and
MMP-9 (A) or MMP-13(B).

A
B
29

2.2 MMP-9 and MMP-13 are silenced in myofibroblastic hepatic stellate
cells
Since HSCs had been shown to be the major sources of MMPs in acute liver injury, the
next step was to study the detailed mechanism of MMP gene suppression in the isolated
cells. For that purpose, an in vitro model was developed to recapitulate the key
characteristics of MMP expression in HSCs of acute injured livers, as well as of MMP
suppression in HSCs of fibrotic livers.
Isolated HSCs undergo spontaneous activation in in vitro culture, providing a model of
quiescent and activated HSCs. Freshly isolated rats HSCs remained quiescent on plastic
culture for about 3 days, and were transformed into myofibroblastic cells after 10 day
culture on plastic with serum. As seen in Fig.6A, the quiescent HSCs could be
discriminated from the activated cells at least in the following aspects: first, the quiescent
HSCs had condensed cell bodies while the myofibroblastic HSCs exhibited expanded cell
bodies; second, shown by phalloidin staining (red) there was little stress fiber formation
in the quiescent cells but the positive staining of filamentous actin become obvious in the
activated cells; last but not least, there was abundant autofluorescent Vitamin A droplets
stored in the quiescent HSCs but not in the myofibroblastic cells.
Besides, a population of activated HSCs had double nuclei staining, suggesting that the
cells had gone through active cell cycle but without final division. This observation was
further confirmed by cell cycle analysis with flow cytometry. HSCs, at indicated day
30

after isolation, were analyzed by flow cytometry using propidium iodide staining. G1
indicated cells with single copy of DNA content, while G2/M represented cells with
double DNA content. From Figure 6B, it was clearly that almost all of freshly isolated
HSCs stayed in G1 phase, while more and more cells had DNA replicated and
accumulated in G2/M phase as cultured on plastic. There was also an apparent increase in
cell population of S phase, but not as distinct as the changes for G2/M phase.
31

Figure 6. In vitro HSC activation. (A) Freshly isolated rat HSCs were left to recover for
2-3 days in DMEM with 10% FBS, during which the cells remain the quiescent
phenotypes. In continuous culture on plastic up to 9-10 days, the cells were trans-
differentiated into myofibroblast like cells. In the upper panels, cells were stained for F-
actin (red) by phalloidin (red) and for nuclei by DAPI (green). In the lower panels, phase
contrast images were taken to show the cells’ morphology, while the inner pictures were
taken under ceiling lights to show the auto-fluorescent vitamin A droplets stored in HSCs.
(B) Freshly isolated rat HSCs were seeded on plastic and cultured in the presence of
serum. At indicated times after isolation, the cells were collected and stained by
propidium iodide, and then cell cycle progression was analyzed with flow cytometry.
Cells in G2/M phase possessed double amount of DNA content as cells in G1 phase.
Therefore, G1 indicated single nucleus HSCs while G2 represented double nuclei cells.

32

quiescent HSCs myofibroblastic HSCs

3 day HSC
G1
S
G2/M

4 day HSC
G1
S
G2/M

5 day HSC
G1
S
G2/M

G1 97.48%
S 0.60%
G2/M 1.56%

G1 91.27%
S 6.22%
G2/M 2.76%

G1 51.95%
S 4.88%
G2/M 42.68%

A
Figure 6, continued
B
33

An injury signal was required to stimulate the MMP production in HSCs, and
interleukin-1alpha (IL-1 ) was chosen for the following reasons: previously a serial of
cytokines were tested in the lab for their effects in inducing MMP-9 expression in HSCs,
and IL-1  was shown to be the most potent one; in addition, the usage of IL-1  also
matched the physiological condition as that IL-1  is a pro-inflammatory cytokine which
functions in the early stage of fibrogenesis, and IL-1  induced MMP-9 would help
degrade the ECM present in healthy liver to make place for the subsequent fibrillar
collagen deposition. The concentration of IL-1  had been titrated in HSCs. Quiescent
HSCs were embedded in type I collagen to mimic the in vivo microenvironment, and then
treated with IL-1  at indicated concentration. One day later, cells were collected for
RNA extraction followed by cDNA synthesis and quantitative real time PCR to measure
the mRNA of both MMP-9 and MMP-13. Conditioned medium from cells after 3 day IL-
1 treatment was analyzed by zymography for MMP-9 activity, as well as by western
blotting probing MMP-13. As seen in Fig.7A, while IL-1  at the concentration of 1ng/ml
stimulated highest level of MMP-9 mRNA, the maximum induction of MMP-13 mRNA
was reached by the treatment of 3ng/ml IL-1 . However, IL-1  at the concentration of
1ng/ml was already able to induce massive amount of MMP-9 and MMP-13 at protein
level (Fig.7B&7C). Therefore, a concentration of 2ng/ml was chosen to minimize the
potential toxic effect of IL-1  while to obtain an obvious effect for the following studies.
By phalloidin staining of the filamentous actin, it was proven that this concentration was
able to induce HSC morphological changes without killing the cell (Fig.7D).
34

Figure 7. Titration of IL-1 in HSCs cultured in type I collagen. Quiescent HSC were
embedded into collagen and treated with IL-1s at indicated concentrations. (A) 24 hours
after stimulation, the cells were collected for RNA extraction, cDNA synthesis and
quantitative real time PCR to measure the mRNA level of MMP-9 and MMP-13. 72
hours after stimulation, the conditioned medium was analyzed by zymography to
examine the production of MMP-9 (B), or by western blotting to check the induction of
MMP-13 (C). In (C), higher molecular weight bands represented proMMP-13 while the
lower bands indicated active MMP-13. (D) Quiescent HSC were cultured on plastic or in
type I collagen. 72 hours after IL-1  stimulation at the concentration of 2ng/ml, cells
were fixed for phalloidin staining to show filamentous actin. As negative control, cells
without IL-1  stimulation had been cultured with DMEM in the presence of 2%FBS for
72 hours. It was evident that this concentration was able to induce HSC morphological
changes without obvious toxicity.
35

Figure 7, continued

0
1
2
3
4
0 1 2 3 4 5 6 7 8 9 10 11
mRNA/GAPDH
IL-1 (ng/ml)
MMP-9 mRNA

0
1
2
3
4
0 1 2 3 4 5 6 7 8 9 10 11
mRNA/GAPDH
IL-1 (ng/ml)
MMP-13 mRNA

IL-1 (ng/ml) 0 1 3 5 7 9 11 IL-1 (ng/ml) 0 1 3 5 7 9 11

without IL-1  with IL-1 
on plastic in collagen I

Active MMP-13
proMMP-13
A
B
C
D

36

To compare their potential in producing MMPs, both quiescent and myofibroblastic
HSCs were cultured on plastic and treated with IL-1 . To access the contribution of three
dimensional ECM to MMP production, the cells were also embedded in type I collagen
gel and then treated by IL-1  as well. After three day IL-1  treatment, the conditioned
media was collected and analyzed by gelatin zymography. As shown in Fig.8A, quiescent
HSCs produced abundant proMMP-9 in response to IL-1 , and three dimensional
collagen would further boost proMMP-9 expression as well as its maturation to active
MMP-9. A similar pattern of MMP-13 expression was observed in quiescent HSCs,
implying a common regulation might apply to both MMP genes. In contrast,
myofibroblastic HSCs barely express MMP-9 or MMP-13 in the same culturing
conditions as the quiescent cells did, although they were still capable of producing
proMMP-2 and activate MMP-2.
To check if this MMP suppression in activated HSCs were on mRNA level, real-time
PCR was first performed to measure the mRNA of MMP-9 and MMP-13. In accordance
with the protein data, mRNAs of MMP-9 and MMP-13 were also induced by IL-1
treatment in the quiescent HSCs (Fig.8B). However, these MMP mRNA inductions were
attenuated in myofibroblastic HSCs compared with those in the quiescent cells.
In summary, through manipulating the culture conditions a model was established to
recapitulate the basic characteristics of the MMP expression and repression in quiescent
and myofibroblastic HSCs. And the underlining mechanism could be addressed with this
in vitro model.
37

quiescent HSCs myofibroblastic HSCs
plastic in collagen plastic in collagen
IL-1  - + - + - + - +
proMMP-9
MMP-9
proMMP-2
MMP-2
MMP-13

quiescent HSC
myofibroblastic HSC

0
2
4
6
8
10
12
Arbitrary Unit
MMP-9 mRNA
900
800
80
60
4
2
≈
≈
*
*
900
800
80
60
4
2
≈
≈
*
IL-1  - + - +
on plastic In 3D collagen
0
1000
2000
3000
4000
5000
6000
7000
Arbitrary Unit
MMP-13 mRNA
IL-1  - + - +
on plastic In 3D collagen
*
*

Figure 8. Repression of MMP genes during HSC trans-differentiation. (A) Quiescent
and myofibroblastic HSCs were seeded on plastic or embedded in 3D type-I collagen,
both followed by IL-1 stimulation for 3 days. Secreted MMP-9 and MMP-2 in the
conditioned medium were revealed by gelatinolytic zymography. MMP-13 was measured
by Western blot analysis. (B) The mRNA levels of MMP-9 and MMP-13 by HSCs in
quiescent or myofibroblastic states were measured by quantitative RT-PCR. Results were
the average of three independent experiments, and plotted as the folds of increase on
basal levels. *P < 0.05.

A
B
38

The steady state level of mRNA was determined by the rate of mRNA synthesis and
degradation. To test if there was a difference in RNA transcription, ChIP assay was
performed using antibody against RNA polymerase II phosphorylated at serine 5, which
is related to transcription elongation. The region covered by PCR following ChIP assay is
about 20 nucleotide away from the transcription start site (TSS) in MMP-9 promoter.
And the corresponding region in MMP-13 promoter covered the TSS. As shown, in
quiescent HSCs, RNA polymerase II began to be recruited to the MMP-9 (Fig.9A) and
MMP-13 (Fig.9B) promoters about six hours after IL-1 stimulation. And such
recruitment peaked around eight hours after IL-1 stimulation for MMP-9 genes while
stilled kept rising up to 10 hours after the stimulation for MMP-13 genes as measured
here. In opposite, there was no appreciable increase of RNA polymerase II recruitment to
either MMP promoter in activated HSCs as observed in quiescent HSCs. These results
clearly revealed a failed RNA polymerase II loading onto the MMP promoter in the
myofibroblastic HSCs, indicating transcriptional control should be one of, if not at all, the
reasons for the MMP gene repression. Of note is that RNA polymerase II recruitment to
the MMP-13 promoter appeared to reach a higher level than that to the MMP-9 promoter,
which might result from the different localizations of ChIP PCR products relative to the
TSS as describer earlier.
39

0
1
2
3
0 6 8 10
Arbitrary Unit
hr, IL-1
Pol II in MMP-9 proximal site (2)

0
1
2
3
4
5
6
7
8
9
0 6 8 10
Arbitrary Unit
hr, IL-1
Pol II in MMP-13 proximal site (3)

Figure 9. RNA polymerase II recruitment was disrupted in myofibroblastic HSCs.
Quiescent and myofibroblastic HSCs were stimulated by IL-1, and harvested at the time
as indicated. Binding of Ser-5-p-Pol II to MMP-9 (A) and MMP-13 (B) promoters were
measured by ChIP assay followed by quantitative PCR analysis. Data were original
calculated as output/input ratios and further converted to fold changes based on the levels
right before IL-1 stimulation. Results were averages of three experiments with standard
deviations.

2.3 Major signal pathways are as active in mHSCs as those in qHSCs
One of the possible reasons for failed RNA polymerase II recruitment is that
myofibroblastic HSCs lost the response to IL-1. To test that, quiescent and
myofibroblastic HSCs were seeded on plastic and in collagen, and then stimulated with
A
B
qHSC
mHSC
40

IL-1 . At indicated time, cells were collected and analyzed by western blotting probing
JNK phosphorylation, I B  and ERK phosphorylation, all of which were reported to
regulate MMP expression. As shown in Fig.10A, there were no obvious difference in
those protein expressions when quiescent and myofibroblastic HSCs were compared.
Both types of HSCs exhibited phosphorylation of JNK and ERK, and degradation of
IkB  as early as 5 minutes after IL-1 stimulation, indicating myofibroblastic HSCs were
responsive to IL-1 as quiescent HSCs did. Further, when plastic culture and collagen
culture of the same type of cells were compared, no distinctions were observed either,
suggesting that collagen boosted MMP expression might not be through enhancing
signaling transductions.
In addition, translocation of NF B p65 was also investigated by immunostaining,
comparing quiescent HSCs and myofibroblastic HSCs cultured on plastic upon IL-1
stimulation. In quiescent HSCs, p65 began to enter nuclei as early as 10 minutes after IL-
1 stimulation, and stayed in the nuclei for up to 90 minutes as measured here. A similar
pattern of such IL-1 induced p65 translocation was found in myofibroblastic cells as well
(Fig.10B). In a word, all the major signaling pathways, including NF B, JNK and ERK,
are as active in myofibroblastic HSCs as those in the quiescent cells.
41

p-JNK
JNK
IkB 
p-ERK
ERK
IL-1 (min)
plastic in collagen
0 5 15 30 0 5 15 30
quiescent HSCs
plastic in collagen
0 5 15 30 0 5 15 30
myofibroblastic HSCs
GAPDH

0 min
10 min
30 min
90 min
qHSC
mHSC

Figure 10. The signal pathways essential for MMP expression were intact in
myofibroblastic HSCs. In vitro derived quiescent (day 2) or myofibroblastic (day 9)
HSCs were seeded on plastic or embedded in 3D collagen on 24-well plates. (A)After
overnight recovery, the cells were challenged by IL-1. Activation of JNK and ERK
pathways was measured by phosphorylation, and IKK activation was indirectly measured
by degradation of IkB . (B) Cells were cultured on plastic and treated with IL-1 . At
indicated time point, cells were fixed and stained for NF B p65.

2.4 Dual phase kinetics of MMP expression in HSCs
To study the temporal control of MMP expression in HSCs, the kinetic of MMP
transcription in HSCs responding to IL-1  was measured by real-time PCR, and a unique
A
B
42

two-phase kinetic was found for both the MMP-9 and MMP-13 transcription. As shown
in Fig.11, upon IL-1 stimulation on quiescent HSCs, the steady states of both MMP-9 and
MMP-13 mRNA were gradually increasing in the first twelve hours but at a low speed.
And the levels reached 2 and 40 folds of the basal by the twelfth hour, respectively.
However, after another twelve hour period, mRNA of MMP-9 and MMP-13 dramatically
increased to 30 and 600 folds of the basal.
In contrast to this mRNA accumulation in quiescent HSCs, myofibroblastic HSCs only
expressed minimal level of MMPs. Even cultured in three dimensional collagen and with
IL-1  treatment, the condition to greatly provoke MMP expression, myofibroblastic
HSCs failed to produce as much MMPs as the quiescent cells did.
The slow onset of MMP genes in HSCs is of potential interest for future studies. Data
here have strengthened the finding that MMP genes are suppressed in myofibroblastic
HSCs, and further this repression last from the very beginning of cytokine treatment to up
to 72 hours (only first 24 hour data are shown here).
43

0
0.1
0.2
0.3
0.4
0.5
0 4 8 12 16 20 24
mRNA/GAPDH
IL-1 Treatment Time (hour)
MMP-9 mRNA

0
0.1
0.2
0.3
0.4
0.5
0 4 8 12 16 20 24
mRNA/GAPDH
IL-1 Treatment Time (hour)
MMP-13

Figure 11. Dual phase kinetics of MMP transcription in HSCs. The mRNA levels of
MMP-9 and MMP-13 in quiescent and myofibroblastic HSCs on plastic were measured
by quantitative RT-PCR. The results are normalized by the mRNA level of GAPDH.

2.5 Recruitment of c-Jun to MMP promoters is impaired in mHSCs
It has been demonstrated that AP-1 is a crucial transcriptional factor for MMP
expressions in response to cytokine stimulation. MMP-9 gene promoter possesses two
AP-1 sites in the distal region and one AP-1 site in the proximal region. Similarly, there
qHSC

mHSC
44

are three AP-1 sites localized in the distal, middle and proximal regions of MMP-13
promoter. The localization of those AP-1 sites on the MMP promoters, as well as the
PCR primer sets used in following ChIP assay, were depicted in Fig.12A. As shown in
Fig.10, JNK phosphorylation was as active in myofibroblastic HSCs as in quiescent
HSCs upon IL-1  stimulation. ChIP assay was carried out to measure the recruitment of
c-Jun to the gene promoters of MMP-9 and MMP-13. It took about 6 hours for c-Jun to
load on the promoters of both MMP-9 and MMP-13 genes (Fig.12B&C). And the
recruitment of c-Jun on the promoters reached the highest level at about 8 hours after IL-
1  stimulation. Such low speed of c-Jun recruitment to the MMP gene promoters is
consistent with the dual kinetics of the gene transcription. In myofibroblastic HSCs, the
recruitment of c-Jun was abrogated. To exclude the possibility that this failed c-Jun
recruitment was due to compromised c-Jun protein expression, western blotting was
performed and showed that c-Jun was equally expressed in the quiescent and
myofibroblastic HSCs (Fig.12D&E). These results demonstrate that although the signal
transduction is intact in myofibroblastic HSCs, the MMP promoters are inaccessible to
the transcription factors such as c-Jun. Such in accessibility underscores a possible
epigenetic regulation which is presumably built up during HSC trans-differentiation.
45

Figure 12. Time courses of c-Jun recruitment to rat MMP-9 and -13 promoters in
quiescent and myofibroblastic HSCs. (A) Schematics of rat MMP-9 and MMP-13 gene
promoters are illustrated. Indicated are potential cis-elements for NF- B and AP1, and
the locations of the primers used for ChIP assay. Binding of c-Jun to the 5’-promoters of
MMP-9 (B) and MMP-13 (C) genes in quiescent vs myofibroblastic HSCs was measured
by ChIP assay followed by quantitative PCR analysis. Data were original calculated as
output/input ratios and further converted to fold changes based on the levels right before
IL-1 stimulation. Results were average of three experiments with standard deviations. (D)
Expression of c-Jun during trans-differentiation was measured by Western blot.
46

Figure 12, continued

Rat MMP-9 proximal promoter
distal site 1
-627 to -477
proximal site 2
-209 to -21
TXN
-627:AP1 -590:NF B -533:AP1 -106:AP-1 -69:AP1

TXN
-1485:AP1 -1220:AP1 -48:AP1
middle site 2
-1214 to -1134
distal site 1
-1506 to -1375
proximal site 3
-140 to 1
Rat MMP-13 proximal promoter

0
1
2
3
0 6 8 10
Arbitrary Unit
hr, IL-1
c-Jun in MMP-9 distal site (1)
0
1
2
3
4
0 6 8 10
Arbitrary Unit
hr, IL-1
c-Jun in MMP-9 proximal site (2)

A
B
qHSC
mHSC
47

Figure 12, continued
0
1
2
3
4
0 6 8 10
Arbitrary Unit
hr, IL-1
c-Jun in MMP-13 distal site (1)
0
1
2
3
4
0 6 8 10
Arbitrary Unit
hr, IL-1
c-Jun in MMP-13 middle site (2)

0
1
2
3
4
5
0 6 8 10
Arbitrary Unit
hr, IL-1
c-Jun in MMP-13 proximal site (3)

D
 actin
c-Jun
Trans-differentiation
3 6 9 days

C
48

2.6 Global histone acetylation decreases in HSC trans-differentiation
To test the hypothesis that MMP genes were epigenetically repressed during HSC trans-
differentiation, western blotting was first performed to check the level of histone
acetylation comparing quiescent and myofibroblastic HSCs. In general, high level of
histone acetylation is correlated with active gene transcription. Vice versa, low level of
histone acetylation might indicate gene repression. The results showed that both histone
H3 and H4 acetylation decreased globally as HSCs became activated. In particular, the
level of histone H4 at lysine 8 and lysine 12 were down-regulated during HSC activation,
while acetylation at lysine 5 was barely detected. And the change on lysine 16 acetylation
was not as apparent as lysine 8 and 12 (Fig.13A). To further confirm these results, HSCs
were fractionated into cytoplasmic and membrane/nucleus pools, characterized by
GAPDH and lamin A/C respectively (Fig.13B). As shown, both histone H3 and H4
acetylations were down-regulated in HSC activation. Finally, ChIP assay was carried out
to measure the basal level of histone H4 acetylation around MMP-13 transcription start
site (Fig.13C). And it was revealed that the histone H4 acetylation level was significantly
lower in myofibroblastic HSCs than that in quiescent HSCs, indicating a condensed
chromatin structure in the MMP-13 promoter region in activated HSCs. In summary, the
basal level of histone acetylation is down-regulated during HSC trans-differentiation.
49

0
0.04
0.08
0.12
0.16
qHSC mHSC
Output/Input(%)
*
act H3
act H4
-actin
Day 2 5 9
Trans-differentiation
-SMA
Histone H3
act H4K5
act H4K8
act H4K12
act H4K16
Histone H4
lamin A/C
-actin
act H3
act H4
GAPDH
Day 2 5 9 2 5 9
Cytoplasmic Nuclear
Acetyl Histone H4 in MMP-13 proximal site (3)
A
B
C

Figure 13. Global reduction of histone acetylation during HSC trans-differentiation.
(A) Quiescent (day 2), intermediate (day 5), and myofibroblastic (day 9) HSCs were
examined for H3 and H4 histone acetylation by Western blot analysis. Trans-
differentiation was monitored by increased expression of -SMA. (B) Cells were
fractionated into cytoplasmic and nuclear/membrane fractions. Decreased acetylation of
histone H3 and H4 during trans-differentiation was shown. (C) In quiescent and
myofibroblastic HSCs, basal levels of acetyl histone H4 in the transcription start site of
MMP-13 gene was measured by ChIP assay and quantitative RT-PCR. Data represented
the output/input ratio and were the average of three independent experiments.

50

2.7 The level of histone acetylation on MMP promoters is lower in mHSCs
IL-1 was reported to regulate histone H4 acetylation at lysine residue 8 and 12(44). To
test if IL-1  also changes histone acetylation on MMP promoters in HSCs, and if there is
a difference between quiescent HSCs and activate HSCs, ChIP assay was performed. The
results clearly showed that in quiescent HSCs, IL-1  stimulated acetylation of Histone
H4 on both MMP-9 and MMP-13 promoters, which occurred mostly near the
transcription start site (Fig.14). In addition to histone H4, histone H3 was also acetylated
in quiescent HSCs upon IL-1 treatment. Similar to c-Jun recruitment, this up-regulation
of histone acetylation was a slow process as well. The increase of acetylation level
around TSS started around 6 hours and reached the peak at about 8 hours after the start of
IL-1 treatment. However, this IL-1 induced histone acetylation was not detected on either
MMP-9 or MMP-13 promoter TSS region in the myofibroblastic HSCs. Summarily, not
only did the basal level of histone acetylation decrease globally in myofibroblastic HSCs,
but also the level of histone acetylation on the MMP promoter region was not induced by
IL-1 in activated HSCs as that in quiescent HSCs.

51

Figure 14. Impaired histone acetylation in MMP gene promoters in myofibroblastic
HSCs. Quiescent and myofibroblastic HSCs were stimulated by IL-1, and harvested at
the time as indicated. Level of acetyl histone H4 and H3 on MMP-9 (A) and MMP-13
(B) promoters were measured by ChIP assay followed by quantitative PCR analysis. Data
were original calculated as output/input ratios and further converted to fold changes
based on the levels right before IL-1 stimulation. Results were averages of three
experiments with standard deviations.
52

Figure 14, continued

0
1
2
3
0 6 8 10
Arbitrary Unit
hr, IL-1
Acetyl Histone H4 in MMP-9 distal site (1)
0
1
2
3
4
5
0 6 8 10
Arbitrary Unit
hr, IL-1
Acetyl Histone H4 in MMP-9 proximal
site (2)
0
1
2
3
4
5
6
7
0 6 8 10
Arbitrary Unit
hr, IL-1
Acetyl Histone H3 in MMP-9 proximal
site (2)

0
1
2
3
0 6 8 10
Arbitrary Unit
hr, IL-1
Acetyl Histone H4 in MMP-13 distal site (1)
0
1
2
3
4
0 6 8 10
Arbitrary Unit
hr, IL-1
Acetyl Histone H4 in MMP-13middle
site (2)

0
1
2
3
4
5
6
0 6 8 10
Arbitrary Unit
hr, IL-1
Acetyl Histone H4 in MMP-13 proximal
site (3)
0
1
2
3
4
0 6 8 10
Arbitrary Unit
hr, IL-1
Acetyl Histone H3 in MMP-13 proximal
site (3)

A
B
qHSC
mHSC
53

2.8 HDAC4 accumulates during HSC trans-differentiation
The dynamic level of histone acetylation is maintained by the opposite functions of
histone acetylases (HATs) and histone deacetylases (HDACs). Since it was previously
reported that histone deacetylase inhibitor kept HSCs in quiescent state, HDAC
expressions were first investigated here by west blotting. During HSC trans-
differentiation, protein levels of class I HDACs including HDAC1, -2 and -3 were not
significantly changed. In contrast, HDAC4, a member of class II HDACs, was steadily
increased, while HDAC5 and -7, another two class II HDACs, were barely detected here
(Fig.15A). As a class II HDAC, HDAC4 is shuttling between cytoplasm and nucleus. By
immunofluorescent staining, expression of HDAC4 was found to increase as HSCs
became activated, and HDAC4 was distributed both in the cytoplasm and nucleus
(Fig.15B). Taken together, those results suggest that HDAC4 accumulates during HSC
activation, and thus may lead to impaired histone acetylation which in turn causes MMP
gene repression.
Quantitative real time PCR was performed to measure the mRNA level of HDAC4, as
well as other HDACs. As shown in Fig.17C, there was no significant variation in the
HDAC mRNA levels during HSC trans-differentiation, indicating the regulation of
HDACs might be through translational or post-translational control. Although not further
pursued in the current study, how HDAC4 is modified during HSC activation is of great
interest for future studies in the lab.
54

Figure 15. Accumulation of HDAC4 during HSC trans-differentiation. (A)
Expressions of class I (HDAC1, -2, -3) and II (HDAC4,) by HSCs during the trans-
differentiation were measured by Western blot analysis. Trans-differentiation of HSCs
was monitored by increased expression of -SMA. (B) Distribution of HDAC4 in nuclear
and cytoplasmic compartments during HSC activation was shown by immunofluorescent
staining (green), while F-actin was stained by phalloidin (red). (C) The mRNA levels of
HDACs were measured by quantitative RT-PCR analysis. Data were normalized to 18s
rRNA and shown as the average with standard deviation from three experiments.
55

Figure 15, continued
HDAC1
HDAC2
HDAC3
-SMA
HDAC4
Trans-differentiation
3 6 9 days
-actin
Phalloidin HDAC4 Merge
Day 3
Day 6
Day 9
A B

0
0.4
0.8
1.2
1.6
2
2 6 9
mRNA/18s rRNA ‰
days
HDAC1
0
0.2
0.4
0.6
0.8
1
2 6 9
mRNA/18s rRNA ‰
days
HDAC2
0
0.04
0.08
0.12
0.16
2 6 9
mRNA/18s rRNA ‰
days
HDAC3
0
0.005
0.01
0.015
0.02
0.025
2 6 9
mRNA/18s rRNA ‰
days
HDAC4
0
0.08
0.16
0.24
0.32
0.4
2 6 9
mRNA/18s rRNA ‰
days
HDAC5
0
0.08
0.16
0.24
0.32
0.4
2 6 9
mRNA/18s rRNA ‰
days
HDAC7
C
56

2.9 HDAC4 suppresses MMP expression
To demonstrate a causal relationship of HDAC4 elevation in MMP gene repression,
HDAC4 was ectopically expressed in quiescent HSCs and the expression of endogenous
MMP-9 expression was measured by gelatin zymography. After transfection, the cells
were treated with IL-1 and the conditioned medium was collected 3 days later for
zymogram analysis. Exogenous expression of HDAC4 was first confirmed by western
blotting using whole cell lysates (Fig.16A). Compared to the control cells transfected
with vector only, HSCs overexpressing HDAC4 showed lower level of endogenous
MMP-9 production (Fig.16B&16C). However, there was only a partial suppression,
which might be due to the low transfection efficiency in primary HSCs. Neither
proMMP-2 nor its matured form was affected by the ectopic expression of HDAC4,
indicating this MMP-9 suppression by HDAC4 overexpression was not due to a global
effect.
In addition, a promoter reporter assay was carried out to examine the effect of HDAC4
ectopic expression on MMP promoter activities. Primary HSCs, 2 days after isolation,
were co-transfected with luciferase reporter plasmids bearing a 1.3kb MMP-9 promoter
or bearing a 1.8kb MMP-13 promoter, together with an expression plasmid encoding
mouse HDAC4. After transfection, cells were stimulated with IL-1  for 18 hours. As
shown by luciferase reporter assay in Fig.16D, IL-1  significantly stimulated promoter
activities of both MMP-9 and MMP-13. HDAC4 overexpression thoroughly repressed
the promoter activities in cells without IL-1  treatment and also in cells treated by IL-1 .
57

Figure 16. Ectopic expression of HDAC4 in quiescent HSCs suppressed the MMP
transcription. (A) The ectopic expression of mouse HDAC4 in HSCs was confirmed by
Western blot analysis. (B) Effects of ectopic expression of HDAC4 on endogenous
MMP9 in quiescent HSCs were determined by transient transfection of the expression
plasmid followed by IL-1 stimulation for 24 hrs. Gelatinases in the conditioned medium
were concentrated by gelatin-conjugated Sepharose 4B and revealed by zymography. (C)
Semi quantitative measurements of endogenous MMP9 protein were measured by
densitometry scanning of the zymography of three repeats. (D) Quiescent HSCs were co-
transfected with three plasmids: (1) reporter plasmids (p1300MMP9-luc or
p1800MMP13-luc) encoding firefly luciferase driven by rat MMP9 or -13 promoters, (2)
an expression plasmid encoding mouse HDAC4 driven by CMV promoter, and (3) renilla
luciferase to monitor the transfection efficiency. After transfection, cells were stimulated
with or without IL-1 for additional 20 hrs. Dual luciferase assay was used to report MMP
promoter activities. The results are average of three repeats with standard deviation.
Statistic significance (*) was calculated by Student T-test, with p<0.05.
58

Figure 16, continued
HDAC4
-actin

pcDNA3 HDAC4
- + - + IL-1 
proMMP-9
MMP-9
MMP-2
0
2
4
6
8
10
12
pcDNA3 HDAC4
MMP-9 arbitrary Folds

Control
IL-1 

0
2
4
pcDNA3 HDAC4
Luciferase Assay Folds
Rat MMP-9 promoter activity
*
*
*
0
2
4
pcDNA3 HDAC4
Luciferase Assay Folds
*
*
*
Rat MMP-13 promoter activity
A
B
C
D
Control
IL-1 

59

Chapter 3: Discussion and Future Direction
3.1 MMP silencing in activated HSCs
In this study the following fundamental question in tissue fibrosis was addressed: how
injury induced MMPs are silenced in fibrotic tissues to favor ECM accumulation. First in
an animal model, it was confirmed that MMPs were induced in acute liver injury but
repressed in chronicle injury induced fibrotic liver. Then an in vitro model was developed
with isolated primary rat HSCs to recapitulate the characteristic of this MMP
expression/suppression. With this model, series of experiments were performed to
address the underlying mechanism of MMP suppression in myofibroblastic HSCs and
thus in liver fibrosis.
By western blotting and immunostaining, it was demonstrated that myofibroblastic HSCs
possessed intact signal transduction as quiescent cells. However, the MMP gene
promoters were not accessible to transcription factors in the activated HSCs, resulting in
the insufficient RNA polymerase II recruitment and thus in failed transcription. The
assemblage of transcription machinery also relies on the local chromatin structure, which
is under epigenetic modification. By western blotting as well as ChIP assay, it was
revealed that the level of histone acetylation decreased globally during HSC trans-
differentiation. Additionally, the IL-1 induced histone acetylation on MMP promoter was
abrogated in the myofibroblastic HSCs. Inversely associated with this histone acetylation,
expression of HDAC4 appeared to increase, while other examined HDACs were either
60

unchanged or not detected. Finally, ectopic expression of HDAC4 in quiescent HSCs
resulted in the suppression of IL-1 induced endogenous MMP-9 secretion, as well as in
reduced MMP promoter activities.
With those evidences a model was proposed to explain how MMP genes are
epigenetically repressed during HSC trans-differentiation in liver fibrosis. HSCs in
healthy liver are maintained quiescent by the loose ECM present in the space of Disse.
And these normal HSCs are capable of producing MMPs upon injury. The MMPs
participate in ECM degradation, which might also release growth factors such as TGF 
that are bound to ECM. This clearance of normal ECM, together with growth factors,
accelerates HSCs’ transition to myofibroblast like cells. During the HSC trans-
differentiation, HDAC4 is built up and reduce the level of histone acetylation. As a result,
in the final myofibroblastic HSCs, the MMP genes become epigenetic silenced that are
not accessible to transcription factors to favor ECM accumulation.
In this study, epigenetic regulation of MMP genes was proposed, while other possibilities
for the reduced MMP expression in fibrosis cannot be excluded. Some of the alternative
explanations are described as followings. First, the fibrotic liver might be injury-tolerant.
After chronicle hepatic damages, the immune system in the liver has already developed a
strategy to cope with the same injury challenge, probably through T regulatory cells. As a
result, the liver injury/inflammation was eliminated and the consequent decreased
cytokine production leads to reduced MMP gene expression. Actually this diminished
liver injury is already indicated by the absence of liver damages in the fibrotic liver,
61

which is present in the acute injured liver (Fig.3A). To test this hypothesis in the mouse
model used in current study, protein levels or even activities of cytokines such as IL-1 
need to be compared between the acute injured liver and the fibrotic liver. It can be
predicted that the fibrotic liver lacks those injury signals, usually present in the early
stage of liver injury, and thus the MMP suppression in fibrotic liver could be partially due
to the shortage of cytokines.
Located in the space of Disse, HSCs adjust the components of the microenvironment
surrounding them: HSCs on one hand synthesize ECM, while on the other hand
producing MMPs, the enzymes that degrade ECM. The precise regulation of those two
opposite reactions plays a fundamental role in tissue homeostasis. Noteworthy is that
MMPs are not only regulated by gene expression, but also by TIMPs, which are the
predominant regulator of MMP activities in tissue. In general, when damaged cells
produce proteases, the healthy cells or the surrounding cells produce inhibitors. However,
the same cell type could also produce both the proteases and the inhibitors at the same
moment. Besides MMPs as discussed here, HSCs do synthesize TIMP-1 and TIMP-2 as
well (4). In response to injury, MMPs are activated while TIMPs can also been seen
parallel to the MMP increase (43). TIMPs do not simply function to block MMPs’
enzymatic activity, but act as a modulator of MMP functions. The activation of MMPs
takes two step: the first step is the cleavage of short peptide from the pro-peptide located
in the N-terminus, and this can be performed by any activated protease such as plasmin;
the second step is to remove the remaining part of the pro-peptide, which must be
operated by another MMP that is already activated. TIMPs take part in this MMP
62

activation by binding to the C-terminal domain of different MMPs and allowing them to
get close and interact with each other (88). Therefore, TIMPs conduct the following
actions as to block the ECM degrading function of MMPs and to modulate the activation
process of MMPs (104). So the ratio of MMPs/TIMPs, as well as TIMP concentrations, is
crucial in determining the net proteatic activity. A high MMPs/TIMPs ratio or low TIMP
concentrations allow MMP activation, while high TIMP concentrations cause inhibition
of MMP activation. In summary, the study of MMPs and TIMPs should not be separated
from each other in the research of liver diseases or any other physiological condition. For
instance, in this study the repression of MMP genes was suggested to contribute to the
excessive ECM deposition in liver fibrosis. However, not all MMP genes are repressed
such as MMP-2 (Fig.4A&Fig.8A). As a result, how the activities of MMPs are regulated
during this process should be of great interest too. By western blotting a global decrease
of histone acetylation was shown, but it is unknown how histone acetylation changes on
the gene promoters of TIMPs. Besides, if and how the ratio of MMPs/TIMPs changes in
liver fibrosis might also be subjected to future studies.
3.2 MMP production in HSC activation and liver fibrosis
The study of HSC activation should fit into the big picture of liver fibrosis. As in any
other biological studies, a physiological relevance should be one of the highest guidelines.
Therefore, during the study of cytokines or growth factors, a fundamental question should
be first addressed as how much this factor contributes to the specific physiological event.
In this lab, long time studies have high lightened the contribution of IL-1  in inducing
63

MMP-9 expression in HSCs. However, this does not exclude the importance of other
cytokines in MMP gene induction, the essential roles of other MMPs rather than MMP-9,
or the presence of various cellular sources producing MMPs in liver injury and liver
fibrosis.
TNF , an inflammatory cytokine as IL-1 ,is capable of stimulating both MMP and
TIMP expressions in HSCs. And a recent study on HSCs showed that reactive nitrogen
species switched on early ECM remodeling via TNF  induction (94). It is still unknown
how much TNF  or IL-1  contributes to HSC activation in liver fibrogenesis,
respectively. So the contribution of TNF  and IL-1  needs to be discriminated, which
may be attained by the usage of cytokine receptor antagonists.
The function of a single cytokine should never be exaggerated. For example, the current
known function of IL-1  is to initiate HSC activation, instead of conducting the whole
process of HSC trans-differentiation. In fact, IL-1  treatment causes reduction in the
expression of -SMA, while the formation of stress fiber is one of the markers for
activated HSCs. Therefore, at least another cytokine is required for the cell
transformation, and TGF  has been indicated as a candidate. HSCs treated by TGF 
showed increased expression of -SMA. In addition, TGF  pathway also participates in
MMP expression. In a recent study, adenoviral delivery of TGF  type II receptor
decreased MMP-2 in HSCs and prevented liver fibrosis in rats (63). Therefore, these
cytokines needs to exert their functions in specific time frames to reach the final cell
trans-differentiation, which should be one of the concerns for a single cytokine study.
64

Besides corporation, cytokines may also regulate each other as well. Again take IL-1 and
TGF  as an example. TGF  is activated from its latent complex by proteolytic and non-
proteolytic mechanisms to perform its biological function, while MMPs, including MMP-
9 and MT-MMP, have been reported to be the enzymatic activator of latent TGF  (110).
Although not confirmed, preliminary data from this lab suggested that IL-1 might direct
latent TGF  activation by provoking MMP production in HSCs.
In addition to MMP-9, other MMPs such as MMP-2 and MMP-13 are also produced in
liver injury and liver fibrosis, while the temporal expression patterns of those MMPs are
different. For instant, as seen in Fig.3A, although MMP-9 disappeared, there was
detectable level of MMP-2 protein activity in the fibrotic liver, and MMP-2 activity did
not differ between the acute injured and fibrotic liver. Iredale group reported that MMP-2
promoted HSC proliferation (9), while Kaneda group showed a correlation between
MMP-2 and HSC migration (42). Taken together, the persisting expression of MMP-2
might foster liver fibrosis through activating HSCs. Unlike MMP-2 but similar to MMP-9,
MMP-13 is expressed in the acute injury phase and soon returns to the basal level. This
early temporary expression of MMP-13 might destroy the tissue present in healthy liver
in order to deposit newly synthesized ECM. Moreover, this early tissue degradation may
also release ECM-bound cytokines such as TGF , facilitating the subsequent fibrogenesis.
Besides, as mentioned before, the activation of MMPs require MMPs that are activated
already. Therefore, the early expression of MMP-13 would also help the maturation of
other metalloproteinases, such as MMP-9, to amplify the function of ECM modulation.
65

Summarily, the family of MMPs orchestrates the tissue modification through the entire
process of liver fibrosis. Besides level of gene expression at a certain time point, temporal
control of those MMP productions also exists. The sequence of MMP gene expression
and the length a single MMP lasts should be taken into consideration when the tissue
architecture modification is studied.
HSCs are not the only source of MMPs in liver injury. Kupffer cells, specialized
macrophages in the liver, are able to make MMP-2, MMP-9 and MMP-13, too. In this
lab an experiment was performed to compare Kupffer cells with HSCs in their potential
to produce MMP-9 upon IL-1  stimulation. Under this specific situation, HSCs were
more potent. However, as discussed above, IL-1  is only a member of numerous
cytokines present in liver injury, and so some other cytokines or even liver toxin could
directly induce MMPs in Kupffer cells. So the contribution of Kupffer cells to MMP
production should not be under-estimated here.
In this study, it was observed that the IL-1 induced MMP expression in HSCs was a slow
process. Although the early signaling events could be activated as early as 5 minutes after
IL-1 stimulation, mRNA was not accumulated until several hours later. In accordance
with that, the recruitments of transcription factors and RNA polymerase II started only
hours after the cytokine stimulation. It is unclear what happen during this process, which
is interesting for future investigation.
Another unsolved mystery is how ECM boosts the MMP induction in HSCs. Based on
the literature that matrix stiffness would affect HSC destiny (103), it could be
66

hypothesized that collagen interfere with MMP productions in HSCs through its
mechanistic property, probably via integrins or discoidin domain tyrosine kinase receptor
2 (DDR2) as reported by Friedman group (74). To test that assumption, HSC could be
embedded in 3D collagen gel with different stiffness, followed by the measurement of
MMP secretion under those conditions. IL-1 induced signal activations were compared
between HSCs cultured on plastic and in collagen, but no distinct difference was found.
However, it is unknown if there is a difference in the final recruitments of transcription
factors to the gene promoters. Previous studies in Mann lab revealed that collagen gel
culture enhanced NF B and AP-1 activities in HSCs (89). Therefore, ChIP assay
comparing plastic culture with 3D collagen culture would be helpful in the future
research, and the first step is to find an efficient way to release HSCs from the gel with
minimal influence on the intracellular signal transduction.
3.3 Epigenetic regulation in HSC trans-differentiation
More and more recent researches have been characterizing epigenetic events during HSC
trans-differentiation and liver fibrosis, including histone acetylation and histone
methylation. In this study, a global decrease of histone acetylation was suggested by
western blotting. Although it was demonstrated that HDAC4 accumulation resulted in the
MMP gene silencing, there is no direct evidence to prove that the increased HDAC4
causes the drop in histone acetylation, either globally or loci specifically. Meanwhile, this
declined histone deacetylation should not be applied to all the genes. For HSCs trans-
67

differentiation, some genes as -SMA and collagen I need to be actively expressed, and
so the local histone acetylation is predicted to be up-regulated.
Class II HDACs might act through domains other than the catalytic region, indicating that
they could perform other functions besides protein deacetylation. One of the possibilities
is that they worked as co-repressors that interacted with transcription factors to block
their activities(7). Moreover, a recent work with genome-wide mapping has revealed that
HDACs not only loaded on inactive genes but also on active genes, resulting in distinct
functions (102). Additionally, the production of transcription factors could also be
affected by this global histone deacetylation, leading to attenuated MMP gene expression.
And transcription factors are also subjected to acetylation/deacetylation. For instance,
acetylation of RelA/p65 abolishes its interaction with I B and thus enhances its binding
to DNA (16). It was also reported that siRNA knockdown of HDAC increased p52 lysine
acetylation in bone metastatic breast carcinoma cells (7). Those could be alternative
explanations for the abrogated MMP promoter activities caused by ectopic HDAC4
expression.
Class II HDACs keep shuttling between the cytoplasm and the nucleus, which was
bestowed by a nuclear localization signal at the N-terminal region and a nuclear export
sequence near the C-terminus (67). Such unique transportation also provides a target for
regulation. In fact, a few serine residues in class II HDACs can be phosphorylated by
calmodulin-dependent kinase, which serves as a signal to recruit nuclear chaperone 14-3-
3 that escorts the complex from nuclei into cytoplasm (35). Therefore, this
68

phosphorylation dependent transportation might also be involved in HSC trans-
differentiation.
The function of HDACs has been linked to epigenetic gene silence (109). It is unknown
how the HDACs are recruited to the chromatin in the myofibroblastic HSCs. It was
suggested that HDACs form a complex with other DNA binding proteins like N-CoR to
repress gene transcription (40). Besides, loading of HDACs onto DNA has also been
proposed to be through methyl DNA and its binding protein MeCP2 (80). In particular, a
recent report from Mann group proved that MeCP2 controls HSC activation and liver
fibrosis through an epigenetic pathway (61). In summary, large is unknown about how
HDACs bind DNA in HSCs, or about the interaction between DNA methylation and
histone modification. To address that, the time window of DNA methylation and histone
modification should be studied specifically in both the in vitro and in vivo system of HSC
trans-differentiation.
Increased HDAC4 expression in HSC activation has been observed. However, it is not
understood how the protein level is regulated yet. By real time PCR, no difference was
found in the mRNA level between quiescent and myofibroblastic HSCs, indicating that
the control might rely on translational or post-translational control such as protein
degradation. Actually it has already been reported that class II HDAC proteins could be
processed through ubiquitination mediated degradation by proteosome (76), but the
detailed mechanism of such process is still unknown. If HDAC4 does determine the fate
69

of HSCs, it would be meaningful to deliberate the precise mechanism underlying the
protein expression.
Although class I HDACs were excluded from the study due to their invariable protein
level during HSC activation, their potential involvements in this process should not be
neglected or denied. In general, the enzymatic function could not be predicted solely
based on the level of protein expression. Instead, the function of activators and inhibitors,
and the interaction of enzymes with their co-activators should be taken into consideration
too. For instance, HDAC3 was suggested to interact with HDAC4, while HDAC3 could
reside in both cytoplasm and nuclei. Therefore, depending on the sub-cellular distribution
and on its interaction with HDAC4, HDAC3 might also influence the cell trans-
differentiation despite its unchanged protein expression. What’s more, by in vitro assay,
it was suggested that class I HDACs were more enzymatic active than the class II
members. Thus a small change of class I HDACs could cause a big effect in the cellular
fate. In a word, to solidify the conclusion that HDACs affect the destiny of HSCs, it
would help to measure HDAC activities in the cell activation.
Histone acetylation is only one aspect of chromatin modification. There are other
epigenetic events, such as DNA methylation, histone methylation and phosphorylation,
which could be potent regulators of the HSC trans-differentiation as well. Further, the
exact effect of such modifications should be studies case by case depending on specific
cell types and target genes. As discussed above, during HSC activation, some genes need
to be turned on or up-regulated, while others need to be shut down or down-regulated. So
70

there should be differential modifications on the specific gene loci. In that regard, it
would be better to conduct gene specific studies to elucidate the role of a certain
epigenetic regulation.
3.4 Studies of HSCs and the implication in the treatment of liver diseases
Due to their substantial roles in ECM secretion and MMP production, HSCs have gained
special interest in the study of liver fibrosis. The utmost goal is to establish a therapy for
the liver disease through manipulating the process of HSC activation. Two of the
common approaches are to prevent and to reverse the process of HSC trans-
differentiation. Although these means seem reasonable and feasible, special cautions
should be taken into account as detailed below.
First, the study of HSC should not be isolated from the in vivo microenvironment. The
cause and effect during HSC activation of liver fibrosis should always be clarified first
and be kept in mind thereafter. Not like in vitro culture where HSCs get spontaneous
activation, in vivo diseases procession usually involve a plethora of cytokines and
chemokines as well as other factors that could take place at the same time in the same
location. For example, HSC activation can arise from toxin induced Kupffer cell
activation. HSCs may also be activated by cytokines secreted from sinusoidal endothelial
cells. Moreover, HSCs would be affected by the components of ECM surrounding them.
Which one of these factors takes effect first? Which effect is dominant? If and how are
these factors coordinated? Those are all hard to answer but essential questions. Even if a
single factor shows striking influence in isolated HSCs, it would be dangerous to
71

speculate or even conclude that this very factor would play remarkable role in vivo.
Physiological complication can never be overstated. Where does this factor come from?
Is it the reason or just the effect of HSC activation? How much is its contribution in the in
vivo situation? A simplified model is definitely required but the result should not be
extended without virtual limitation.
Second, it is risky to assume that liver fibrosis could be ameliorated by eliminating so
called “makers of HSCs activation”. Take -SMA for an example. Rising expression of
-SMA is a hallmark of HSC trans-differentiation, which helps myofibroblastic HSCs in
contraction or migration. Due to that observation, knockdown of this gene was supposed
to alleviate or even totally block liver fibrosis. However, the fact is on the contrary that
absence of -SMA brings about elevated fibrosis. That emphasizes how important it is to
understand the cause and effect in the complicated cellular transition. Increasing -SMA
is the outcome of HSC activation, not the causation. As a result, it is not surprising not to
get the expected reducing fibrosis when the gene expression is removed. Storage and loss
of Vitamin droplets is another example here. It is well known that quiescent HSCs store
abundant retinoid, while gradually they lose this storage during trans-differentiation.
Attempts have been made to elaborate the mechanism of this phenomenon, and to restore
the droplets in order to reverse HSC activation. Unfortunately, there has been no success
so far, probably due to the original problematic assumption. Diminishing Vitamin A
storage actually results from, instead of leading to, HSC activation. In that case, it would
not be realistic to retain HSC quiescent by holding the cells’ retinoid content.
72

Third, there is no absolute good or bad for the effect of a single factor in liver fibrosis.
Instead, depending on a specific spatial and temporal situation, a “good” one can become
“bad” and vice verse a “malicious” factor could be beneficial. For example, new matrix
synthesis is required for the replacement of injured tissue. TGF  contributes to the
process by recruiting fibroblasts and stimulating their synthesis of ECM, and
concurrently by inhibiting proteases. Nonetheless, uncontrolled TGF  activity would
cause fibrosis. Another example is the inflammatory cytokine, which have been
suggested to provoke HSC activation and might be simply reasoned as bad. However,
limited inflammation is good for optimal would healing by first recruiting lymphocytes
that are essential for phagocytosis of debris and microbes. And the clearance of debris
and/or infectious organisms promotes the resolution of inflammation and ensures the
repair response to continue into granulation, angiogenesis and re-epithelialization. In a
word, the problem with liver fibrosis is the aberrant cytokine pathway, instead of the
mere expression/suppression of cytokines. As the result, more weight should be given to
the evaluation of the pathway’s deviation from normal physiological conditions in its
endurance and intensity, not to the presence/absence of the very cytokine.
Fourth, even though a number of factors have been suggested to keep HSC quiescent, it is
unpredictable whether HSC activation could be reversed by restoring those factors. Take
Lhx2 for example. Since Lhx2 knockout mice develop spontaneous liver fibrosis, the
gene has been suggested to keep HSC quiescent (98). However, experimental data with in
vitro primary HSCs do not support this speculation. The expression of Lhx2 goes up in
73

HSC culture activation. Adenovirus delivered Lhx2 gene overexpression in HSCs does
not show much difference from the control (adenovirus delivered GFP expression),
including mRNA level of -SMA and type I collagen, cell morphology and the ability of
contraction in response to endothelin-1.
Finally, current researches still face the question how to apply the bench work to bedside
treatments. It is well documented that TGF-  is a fibrotic cytokine and that TGF-  is able
promote HSC activation. Plus, studies have shown the beneficial effect of blocking TGF-
 pathways in diseases or of mutating TGF- 1 gene in cutaneous wound healing.
However, only by considering the immunosuppressive power of TGF- , it is easy to
foresee the catastrophic side-effects of blocking TGF-  signaling in injured tissues with
inflammation, although fibrogenesis might be interrupted. Therefore, the question is how
and when to interfere with the cytokine’s function in a defined in vivo situation, and also
how to limit the number of side targeted cells. Besides cytokine intervention, gene
therapy also encounters the problem with the method of delivery. HSCs only constitute a
small population of liver cell mass. Previous studies in the lab already showed that
adenovirus, injected through the tail vein or the spleen, was mostly taken up by
hepatocytes. One way to solve this problem is to use hepatic specific promoter, which is
still being searched. Type I collagen promoter was used in a certain study to reach HSC
specific expression in the liver. Practical in lab research, it is not appropriate in vivo since
at least skin, the largest organ in the body, could also express collagen I.
74

In summary, exciting and encouraging as more and more has been discovered about HSC
trans-differentiation and liver disease, care should be taken in drawing the study
hierarchy and in conjecturing the possible clinical applications. A lot of questions need to
be answered first to reach the ultimate therapeutic goal.
3.5 Experimental obstacles of in vitro HSC studies
The success of isolating HSCs is a breakthrough in the study. However, there are still
practical obstacles that hinder the progress of the in vitro research.
As known, HSCs undergo spontaneous activation in in vitro culture. So it is tricky to pick
the time point when cells are still “normal”. Usually 2 or 3 day HSCs are used as
quiescent cells, but the cells already exhibit more or less phenotypes of activation.
Therefore, for some quick changing genes the usage of those cells might generate false-
positive or false-negative results. Also, the failure to retain HSC quiescent by preventing
DNA methylation or histone deacetylation might due to the fact that those epigenetic
modifications are already well established in the 2 or 3 day old cells.
As long time observed and shown in Fig.6, the cellular content of HSCs increases or even
doubles during in vitro culture. So the question is which gene would be a good control
when the protein or DNA profile is compared between the quiescent and activated HSCs.
House keep gene is not necessarily a good control, such as GAPDH which protein level
actually changes during HSC culture activation. Between 3 day and 10day HSCs, -actin
75

was shown to be a better protein control while 18s rRNA appeared to be a good control in
quantitative real time analysis.
Low efficiency of plasmid transfection , especially for myofibroblastic HSCs, has been
impeding the progress of gene manipulation in HSCs. Adenoviral transduction help solve
this issue but still with problems, one of which is the possible effect of adenovirus
themselves on HSC activation. Invasion of adenovirus induces obvious changes on HSC
phenotype including elongate cytoplasm, which creates a serious problem for
demonstrating the influence of potential quiescence-keeping genes. The success of those
kinds of studies at least partially depends on the competition between possible activation
induced by adenovirus and the ability of the target gene to hold back this cellular trans-
differentiation. Therefore, the negative outcome should be interpreted discreetly. In
quiescent HSCs, which are cells 2 days after isolation, the transfection efficiency could
be partially improved by performing the transfection 4 to 6 hours after subculture, instead
of waiting for 16 to 24 hours as usually suggested.
3.6 Future direction
Increased HDAC4 expression was found during HSC in vitro activation. However, it is
not known whether this represents the in vivo patho-physiological condition. To test that,
a comparison of the HDAC4 level between HSCs from healthy and fibrotic livers could
be conducted first. And the expression of other HDACs can be measured at the same time.
The result would tell if what has been found in the thesis study really happens in
fibrogenesis.
76

Other than that, the change of HDAC activities is still unknown, either in cell culture
model or in vivo situation, which could be accessed by a well-established HDAC activity
assay. According to the finding that histone acetylation decreases globally during HSC
activation, it could be predicted that myofibroblastic HSC show higher deacetylation
activity than the quiescent cells. Due to the shortcomings that the deacetylation assay
could not distinguish one HDAC from the others, the change of a specific HDAC activity
could be masked. So false-negative results could be possibly generated.
By gain of function assay, it was indicated a casual relationship between increased
HDACs and decreased MMP gene expression. ChIP assay can be performed to detect
possible increasing HDAC4 loading while decreased histone acetylation on the MMP
promoter, which would be direct evidence to strengthen current conclusion. In addition, a
loss of function assay would be appropriate to make the conclusion more concrete.
Specifically, shRNA could be utilized to knock down HDAC4 in quiescent HSCs, and
then several parameters could be compared between control cells and the HDAC4
knockdown cells as (1) if the HDAC4 knockdown cells show less activation during in
vitro culture; (2) after 10 day plastic culture, when the control cells lose their ability to
express MMPs, if the HDAC4 knockdown cells can still produce MMPs as the quiescent
cells; (3) if the expression/activity of other HDACs changes in the HDAC4 knockdown
cell, probably for compensation. The ideal outcome would be HDAC4 knockdown cells
retain the capacity of expressing MMP genes even after long term plastic culture.
However, a negative result may be due to the redundancy of HDACs.
77

Besides, it would be interesting to study HATs in the cellular trans-differentiation, which
has not been addressed here. In a broader picture, those studies can be applied to the other
hepatic cells, such as to study the epigenetic regulation in Kupffer cell activation and
hepatocyte re-proliferation.
Finally, as in cancer studies, epigenetic modification may be exploited to treat liver
diseases, either by accelerating the apoptosis of myofibroblast to reduce fibrillar ECM
accumulation, or by stimulating the proliferation of hepatocytes to replace the damaged
cells. Modulation of the epigenetic events can also induce genes such as MMPs to help
the resolve of liver fibrosis.
78

Chapter 4: Materials and Methods
Animal model. FVB mice were injected twice a week with thioacetamid (TAA) at
0.2mg/g body weight or saline for up to eight weeks to generate fibrotic liver and control
respectively (n=6). No mice died during the eight week injection. One day before
sacrifice, all mice were subjected to a final TAA challenge. Livers were collected for
zymography and immuno-histological studies. All animals received humane care in
compliance with the institution’s guidelines for the care and use of laboratory animals in
research.
Isolation of rat HSCs. Protocols for isolation of rat HSCs have been previously
described(26). The liver of male Wistar rat was sequentially digested with Pronase
(Roche Applied Science) and type IV collagenase (Sigma) by in situ perfusion.
Hepatocytes were removed by centrifuging the digested liver at 50 X g for 2 minutes. The
nonparenchymal cells present in the supernatant were laid on top of arabinogalactant
gradient (Sigma), followed by centrifugation at 21,400 rpm for 40 minutes at 25
o
C. A
fraction of HSCs was recovered from the interface between the medium and the lowest
density. Both the purity, examined by phase contrast microscopy, and viability of HSCs,
accessed by trypan blue staining, always exceed 95% (Provided by the Southern
California Research Center for ALPD and Cirrhosis).
Cell culture. Primary rat HSCs were cultured on plastic in DMEM with 10%FBS and
100 units/ml penicillin-streptomycin for two days before experiments. For three-
79

dimensional ECM culture, cells were first collected by trypsinization and then
resupended in DMEM at the density of 4X10
5
cells/ml. An equal volume of the cell
suspension was then mixed with collagen gel mixture, and seeded in the plate.
Myofibroblastic HSCs were prepared by culturing the primary cells on plastic for up to
ten days in DMEM with 10%FBS, wherein medium was changed every three days.
Cell fractionation. Cells were collected in PBS and first lysised in hypotonic lysis buffer
(10mM HEPES, pH 7.9, 1.5mM MgCl2, 10mM KCl, 1mM DTT, 0.6% IGEPAL@ CA-
630, protease inhibitors). After centrifugation, the supernatant was transferred as
cytoplasmic fraction. After two times washing with the lysis buffer, pelleted nuclei was
lysised in 1X reducing sample buffer and analyzed by western blotting.
Western blot. To collect whole cell lysate, after PBS wash, 1X reducing sample buffer
was added to cells cultured on plastic. For cells cultured in collagen, the gel/cell was
taken into 1.5-ml tube and centrifuged at 6000 rpm at 4
o
C for 5 minutes, and the pellet
was dissolved in 1X reducing sample buffer. For conditioned medium, equal volume of
conditioned medium was mixed with 2X reducing sample buffer. After 5 min boiling,
samples were resolved by SDS-PAGE, and then transferred to Immobilon-P membrane
(Millipore IPVH00010). After 3 hour blocking in 5% nonfat milk or BSA according to
manufacture’s datasheet, the membrane was incubated with 1 g/ml primary antibody
overnight at 4
o
C. After that, membrane was washed and then further incubated with
horseradish peroxidase-conjugated secondary antibody from Santa Cruz Biotechnology
(sc-2020, sc-2005, sc-2004), and developed by Super Signal West Femto
80

Chemiluminescent substrate (34096, Pierce). Anti-MMP13 (MAB 13426) and GAPDH
(MAB 374) antibodies were purchased from Millipore. Antibodies for I B  (9242), p-
JNK (9255) and JNK (9258) were purchased from Cell Signaling, while anti-ERK (sc-
93) and p-ERK (sc-7383) were from Santa Cruz Biotechnology. Antibodies for HDAC1,
2, 3, 4, 5, 7 were from Cell Signaling (9928). Antibody for -actin (A2228) was from
Sigma.
Zymography. Conditioned medium, in sample buffer without DTT, was resolved by
10% SDS-PAGE containing 0.1% (w/v) gelatin (G9382, Sigma). To enrich gelatinases
for zymography analysis, mouse liver tissues (50 mg) were homogenized into 1.0 ml cold
NT buffer (50 mM Tris-HCl pH7.5, 100 mM NaCl) containing 1% Triton X-100. After
20 min centrifugation at 12000 rpm at 4° C, the supernatant was taken to measure total
protein concentration by Biorad protein assay (500-0006) and samples with equal total
protein amount were loaded for analysis. After electrophoresis, the gel was washed in
2% Triton X-100 for 30 minutes and then incubated with developing buffer (5 mM CaCl
2
,
150 mM NaCl,

50 mM Tris, pH 7.5) for 16 h at 37 ° C and visualized

by Coomassie Blue
R-250 staining.
Plasmid construction. Rat genomic DNA was extracted from primary HSCs. Briefly,
cells were first collected into TE buffer (200mM Tris-HCl pH8.0, 100mM EDTA).
Washed with TE buffer, cells were treated with 0.4mg/ml proteinase K and 1% SDS in
TE buffer overnight. Then DNA was purified by phenol/chloroform extraction and
isopropyl alcohol precipitation. PCR was performed to construct the promoter regions of
81

MMP9 and -13 using primers listed in table 2. For the promoter of MMP9, the primers
include KpnI site at the 5’ end and BglII site at the other end, while for the promoter of
MMP13, the primers include KpnI site at the 5’ end and SmaI site at the other. After
restriction enzyme digestion (KpnI and BglII for MMP9 while KpnI and SmaI for
MMP13), the PCR product was cloned into pGL3-basic vector harboring a firefly
luciferase (Promega) gene as reporter plasmids, namely, p1300MMP9-luc, and
p1800MMP13-luc. To express HDAC4, cDNA encoding mouse HDAC4 was purchased
from Open Biosystems. After digestion with EcoRI and NotI, the insert was cloned into
pcDNA3 (Invitrogen), prepared by those two restriction enzymes. The inserts were
confirmed by DNA sequencing performed by the DNA core facility in Norris Cancer
Center at Univerisity of Southern California.
Table 2. Primers for cloning of MMP-9 and MMP-13 promoter from rat genomic
DNA
p1300MMP9-luc 5’ GGGGTACCTGTGGCTTGAAGGCGAAATGC 3’
5’ GAAGATCTGGTGAGAACCGAAGCTTCTGGG 3’
p1800MMP13-luc 5' GGGGTACCTGGCACAAGCTGTAATCCTAGCACT 3'
5’ TCCCCCGGGCTCAACAAGAAGAAGGTGGCCAGA 3’
Transfection and Luciferase assay. Transfection of primary HSCs was carried out with
Lipofectamine 2000 (Invitrogen) according to manufacturer’s manual. Briefly, HSCs
were first subcultured into 24 well plate at the density of 0.1 million cells per well. 6
hours later transfection was performed using 1 g of reporter plasmid, 2 g of
HDAC4/pcDNA3 together with 10 ng of phRL (CMV promoter driven renilla luciferase).
18 hours later, cells were treated with or without IL-1 for an additional 18 hrs.
Conditioned medium was collected for zymography analysis. Cells were collected for
82

luciferase assay, using dual luciferase assay system (Promega) according to
manufacturer’s suggestion.
Chromatin Immunoprecipitation (ChIP) assay. ChIP assay was performed as previous
described with modification (19). After stimulation with cytokine, HSCs were cross-
linked by 1% formaldehyde for 10 minutes at room temperature. Then cells were washed
with ice cold PBS twice, and scraped into ice cold PBS containing protease inhibitor
cocktails (S8820, Sigma) and collected by 10 min centrifugation at 3000 rpm at 4
o
C. Cell
pellets were resuspended in cell lysis buffer (5 mM Pipes/potassium pH 8.0, 85 mM KCl,
0.5% CA-630 plus protease inhibitor cocktails) and incubated on ice for 10 minutes with
occasional vortex. After 10 minute spin at 5000 rpm, the nuclei pellets were lysed in SDS
lysis buffer (50 mM Tris pH 8.0, 10 mM EDTA, 1% SDS plus protease inhibitors),
followed by 10 minute incubation on ice. Sonication was then carried out to get
chromatin fragments between 200 to1000 bp. After that, the mixture was 10 fold diluted
with RIPA ChIP buffer (10 mM Tris–HCl, pH 7.5, 140 mM NaCl, 1 mM EDTA, 0.5 mM
EGTA, 1% (vol/vol) Triton X-100, 0.1% (wt/vol) SDS, 0.1% (wt/vol), Na-deoxycholate,
protease inhibitor mix), and centrifugated at 14000rpm for 10min at 4
o
C. The
supernatant containing chromatin preparation was transferred to a new tube and aliquoted
for subsequent experiment. For each assay, the chromatin preparation was first incubated
with 1.2ug of primary antibody overnight on rotator at 4
o
C. And then 10ul of RIPA pre-
washed Dynabeads protein A beads (Invitrogen) was added and incubated for another one
hour. After that, the beads were sequentially washed with RIPA ChIP buffer for three
times and TE buffer (10 mM Tris–HCl, pH 8.0, 10 mM EDTA) twice. Pulled down
83

DNA was eluted and reverse cross-linked with elution buffer (20 mM Tris–HCl, pH 7.5,
5 mM EDTA, 50 mM NaCl, 1% (wt/vol) SDS, 50 mg ml–1 proteinase K.) twice at 68
o
C
for a total of two hours. DNA was then purified by phenol-chloroform extraction and
ethanol precipitation, followed by quantitative real-time PCR. Antibodies for acetyl-
Histone H3 (06-599) and H4 (06-598) were from Millipore. Antibodies for NF B p65
(sc-109) and c-Jun (sc-45) were from Santa Cruz Technology. Anti RNA polymerase II
CTD repeat YSPTSPS (phospho S5) (ab5131) was purchased from Abcam. For relative
quantitative ChIP assay, the thermal cycle was set at 94
o
C 10 minutes, followed by 50
cycles of 20 seconds at 94
o
C and 1 minute at 60
o
C. The final result was represented as
2
(Ct
Input
-Ct
Output
)
. The primers for ChIP are listed in table 3.
Table 3. Primers for ChIP assay
Primer set for MMP9 distal site
(1)
5’ AAG GAG TCA GCC TGC TGG G 3’
5’ CCC ACA CTG TAG GTT CTA TCC TCT 3’
Primer set for MMP9 proximal
site (2)
5’ TGA GTC AGC GTA AGC CTG GA 3’
5’ GGT GAA GCA GAA TTT GCG GAG GTT 3’
Primer set for MMP13 proximal
site (1)
5’ AAG TCC CAA ATG GTC TCG GTC TGA 3’
5’ AAA CGG TTC TGA CAA AGG CTG CTG 3’
Primer set for MMP13 middle
site (2)
5’ ATC CTG TCA GCT GTC TGC GAT CT 3’
5’ TGG ACA GCC AGC CTT AAG GAA ATG 3’
Primer set for MMP13 distal
site (3)
5’ CTG CCA CAA ACC ACA CGT ACG AAA 3’
5’ CTT CCC AGG GCA AGC ATT CTC TAT 3’
qRT-PCR analysis. RNA was extracted from mouse livers using RNAqueous 96
(Ambion) and from HSCs using Trizol Reagent (Invitrogen). Reverse transcription was
carried out with random primer and M-MLV (Invitrogen) according to the manual. Real-
time PCR was performed using SYBR master mix (Applied Biosystems) and gene
84

specific primers (table 2). The reaction was carried out in 384-well plate with ABI 7900
DNA detection system at USC Research Center for Liver Diseases. Each reaction was
performed in triplicate and 5 ng of cDNA was used in a 10ul system for each well. To
measure mRNA from the reverse transcription, the reaction was carried out at 94
o
C for
10 minutes, followed by 40 cycles of 10 seconds at 94
o
C and 1 minute at 60
o
C. The
final result of gene transcription was calculated as 2
(Ct
GAPDH
-Ct
Gene
)
. The primers for real-
time PCR are listed in table 4.
Table 4. Primers for quantitative real-time PCR
mouse MMP-9 5’ CGT GTC TGG AGA TTC GAC TTG A 3’
5’ TGG AAG ATG TCG TGT GAG TTC C 3’
mouse MMP-12 5’ AGG TGG TAC ACT AGC CCA TGC TTT 3’
5’ GCA ACA AGG AAG AGG TTT GTG CCT 3’
mouse MMP-14 5’ ATC TCA CAG CTC GGT GTG TGT TCA 3’
5’ AAG GTC AGA GGG TCT TGC CTT CAA 3’
mouse type I collagen 5’ GCA TGG CCA AGA AGA CAT CC 3’
5’ CCT CGG GTT TCC ACG TCT C 3’
mouse GAPDH 5’ GCA CAG TCA AGG CCG AGA AT 3’
5’ GCC TTC TCC ATG GTG GTG AA 3’
rat MMP-9 5’ CAG ACC AAG GGT ACA GCC TGT T 3’
5’ AGC GCA TGG CCG AAC TC 3’
rat MMP-12 5’ TCT ATG GAG CCC CAG TGA AA 3’
5’ GAC ACA CAG TTG ATG GTG GAC TTC 3’
rat MMP-13 5’ GCC CTA TCC CTT GAT GCC ATT 3’
5’ ACA GTT CAG GCT CAA CCT G 3’
rat MMP-14 5’ GCC CAA CAT CTG TGA TGG GAA CTT 3’
5’ TTA TTC CTC ACC CGC CAG AAC CAT 3’
rat GAPDH 5’ CCT GGA GAA ACC TGC CAA GTA T 3’
5’ CTC GGC CGC CTG CTT 3’
Immunofluorescent staining. Liver tissues were snap frozen in liquid nitrogen and
embedded in OCT compound. Sections were fixed in ice-cold acetone for 10 min. After
85

blocking with 5% donkey serum in PBS, specimens were further incubated overnight
with 4 g/ml of the primary antibody (anti-MMP-9, sc-6841, Santa Cruz technology),
anti-MMP-13 (AB8120, Millipore), and anti-desmin (sc-7559, Santa Cruz technology
and ac-8592, Abcam). Cy3 conjugated rabbit anti goat IgG (C2821, Sigma) and FITC-
conjugated goat anti-mouse IgG (F2057) were selectively used for visualization. For
immunofluorescent staining, HSCs were fixed with 4% formaldehyde in PBS for 10 min
at room temperature. After three washes with PBS, cells were permeabilized with 0.1%
Triton X-100 in PBS for 10min. Then the cells were blocked with 5% donkey serum in
PBS and further incubated with 4 g/ml of primary antibody overnight. Then
fluorescence conjugated secondary antibodies (A-11057, A-21206, Invitrogen) were
applied. To stain F-actin, TRITC-conjugated phalloidin (P1951, Sigma) at 0.1 g/ml was
applied. Nuclei were visualized by DAPI at 0.1 g/ml. Pictures were taken with a Zeiss
confocal microscope in the USC Research Center for Liver Diseases or a Nikon TU2500-
Eclipse in this lab.
H&E and Sirus Red staining of liver sections. Paraffin sections were deparaffinized
and rehydrated by sequential treatments of Xylene, 100% and 95%, 80% ethanol,
followed by deionized water. The slides were incubated with Hematoxylin followed by
rinsing with water and quick dips into acid ethanol to destain. For Eosin staining, the
slides were incubated shortly with Eosin, followed by 95%, 100% ethanol and Xylene.
For Sirus red staining, the sections were deparaffinized and then stained by fast green and
86

Sirus red sequentially. Lastly, the sections were dehydrated by ethanol and xylene
treatments before mounting.
Flow cytometry. Cells were collected by trypsinization. After one time PBS wash, the
cells were fixed overnight by cold 75% ethanol in PBS. Then the cells were washed with
cold PBS twice and stained by propidium iodide for 2 hours. And the stained cells were
analyzed by flow cytometry to show the population of cells in different cell cycle phases.
87

Bibliography
1. 1996. Hepatic stellate cell nomenclature. Hepatology 23:193.
2. Akhtar, N., Z. Rasheed, S. Ramamurthy, A. N. Anbazhagan, F. R. Voss, and T. M.
Haqqi. 2010. MicroRNA-27b regulates the expression of MMP-13 in human
osteoarthritis chondrocytes. Arthritis Rheum.
3. Arenson, D. M., S. L. Friedman, and D. M. Bissell. 1988. Formation of
extracellular matrix in normal rat liver: lipocytes as a major source of
proteoglycan. Gastroenterology 95:441-7.
4. Arthur, M. J., J. P. Iredale, and D. A. Mann. 1999. Tissue inhibitors of
metalloproteinases: role in liver fibrosis and alcoholic liver disease. Alcohol Clin
Exp Res 23:940-3.
5. Backs, J., T. Backs, S. Bezprozvannaya, T. A. McKinsey, and E. N. Olson. 2008.
Histone deacetylase 5 acquires calcium/calmodulin-dependent kinase II
responsiveness by oligomerization with histone deacetylase 4. Mol Cell Biol
28:3437-45.
6. Bataller, R., and D. A. Brenner. 2005. Liver fibrosis. J Clin Invest 115:209-18.
7. Bendinelli, P., E. Matteucci, P. Maroni, and M. A. Desiderio. 2009. NF-kappaB
activation, dependent on acetylation/deacetylation, contributes to HIF-1 activity
and migration of bone metastatic breast carcinoma cells. Mol Cancer Res 7:1328-
41.
8. Benyon, R. C., and M. J. Arthur. 2001. Extracellular matrix degradation and the
role of hepatic stellate cells. Semin Liver Dis 21:373-84.
9. Benyon, R. C., C. J. Hovell, M. Da Gaca, E. H. Jones, J. P. Iredale, and M. J.
Arthur. 1999. Progelatinase A is produced and activated by rat hepatic stellate
cells and promotes their proliferation. Hepatology 30:977-86.
88

10. Beuscher, H. U., U. P. Rausch, I. G. Otterness, and M. Rollinghoff. 1992.
Transition from interleukin 1 beta (IL-1 beta) to IL-1 alpha production during
maturation of inflammatory macrophages in vivo. J Exp Med 175:1793-7.
11. Bissell, D. M., S. S. Wang, W. R. Jarnagin, and F. J. Roll. 1995. Cell-specific
expression of transforming growth factor-beta in rat liver. Evidence for autocrine
regulation of hepatocyte proliferation. J Clin Invest 96:447-55.
12. Bode, W., F. X. Gomis-Ruth, and W. Stockler. 1993. Astacins, serralysins, snake
venom and matrix metalloproteinases exhibit identical zinc-binding environments
(HEXXHXXGXXH and Met-turn) and topologies and should be grouped into a
common family, the 'metzincins'. FEBS Lett 331:134-40.
13. Carruth, L. M., S. Demczuk, and S. B. Mizel. 1991. Involvement of a calpain-like
protease in the processing of the murine interleukin 1 alpha precursor. J Biol
Chem 266:12162-7.
14. Cerretti, D. P., C. J. Kozlosky, B. Mosley, N. Nelson, K. Van Ness, T. A.
Greenstreet, C. J. March, S. R. Kronheim, T. Druck, L. A. Cannizzaro, and et al.
1992. Molecular cloning of the interleukin-1 beta converting enzyme. Science
256:97-100.
15. Chang, S., B. D. Young, S. Li, X. Qi, J. A. Richardson, and E. N. Olson. 2006.
Histone deacetylase 7 maintains vascular integrity by repressing matrix
metalloproteinase 10. Cell 126:321-34.
16. Chen, L. F., Y. Mu, and W. C. Greene. 2002. Acetylation of RelA at discrete sites
regulates distinct nuclear functions of NF-kappaB. Embo J 21:6539-48.
17. Chicoine, E., P. O. Esteve, O. Robledo, C. Van Themsche, E. F. Potworowski,
and Y. St-Pierre. 2002. Evidence for the role of promoter methylation in the
regulation of MMP-9 gene expression. Biochem Biophys Res Commun 297:765-
72.
18. Consolo, M., A. Amoroso, D. A. Spandidos, and M. C. Mazzarino. 2009. Matrix
metalloproteinases and their inhibitors as markers of inflammation and fibrosis in
chronic liver disease (Review). Int J Mol Med 24:143-52.
89

19. Dahl, J. A., and P. Collas. 2008. A rapid micro chromatin immunoprecipitation
assay (microChIP). Nat Protoc 3:1032-45.
20. Davis, B. H. 1988. Transforming growth factor beta responsiveness is modulated
by the extracellular collagen matrix during hepatic ito cell culture. J Cell Physiol
136:547-53.
21. Dinarello, C. A. 1994. The interleukin-1 family: 10 years of discovery. Faseb J
8:1314-25.
22. Dressel, U., P. J. Bailey, S. C. Wang, M. Downes, R. M. Evans, and G. E. Muscat.
2001. A dynamic role for HDAC7 in MEF2-mediated muscle differentiation. J
Biol Chem 276:17007-13.
23. Eisen, A. Z., J. J. Jeffrey, and J. Gross. 1968. Human skin collagenase. Isolation
and mechanism of attack on the collagen molecule. Biochim Biophys Acta
151:637-45.
24. Eng, F. J., and S. L. Friedman. 2001. Transcriptional regulation in hepatic stellate
cells. Semin Liver Dis 21:385-95.
25. Feder, L. S., J. A. Todaro, and D. L. Laskin. 1993. Characterization of
interleukin-1 and interleukin-6 production by hepatic endothelial cells and
macrophages. J Leukoc Biol 53:126-32.
26. Friedman, S. L., D. C. Rockey, R. F. McGuire, J. J. Maher, J. K. Boyles, and G.
Yamasaki. 1992. Isolated hepatic lipocytes and Kupffer cells from normal human
liver: morphological and functional characteristics in primary culture. Hepatology
15:234-43.
27. Friedman, S. L., F. J. Roll, J. Boyles, D. M. Arenson, and D. M. Bissell. 1989.
Maintenance of differentiated phenotype of cultured rat hepatic lipocytes by
basement membrane matrix. J Biol Chem 264:10756-62.
28. Friedman, S. L., S. Wei, and W. S. Blaner. 1993. Retinol release by activated rat
hepatic lipocytes: regulation by Kupffer cell-conditioned medium and PDGF. Am
J Physiol 264:G947-52.
90

29. Gaca, M. D., X. Zhou, R. Issa, K. Kiriella, J. P. Iredale, and R. C. Benyon. 2003.
Basement membrane-like matrix inhibits proliferation and collagen synthesis by
activated rat hepatic stellate cells: evidence for matrix-dependent deactivation of
stellate cells. Matrix Biol 22:229-39.
30. Geerts, A. 2001. History, heterogeneity, developmental biology, and functions of
quiescent hepatic stellate cells. Semin Liver Dis 21:311-35.
31. George, J., D. Roulot, V. E. Koteliansky, and D. M. Bissell. 1999. In vivo
inhibition of rat stellate cell activation by soluble transforming growth factor beta
type II receptor: a potential new therapy for hepatic fibrosis. Proc Natl Acad Sci
U S A 96:12719-24.
32. Gieling, R. G., K. Wallace, and Y. P. Han. 2009. Interleukin-1 participates in the
progression from liver injury to fibrosis. Am J Physiol Gastrointest Liver Physiol
296:G1324-31.
33. Gorbig, M. N., P. Gines, R. Bataller, J. M. Nicolas, E. Garcia-Ramallo, P. Cejudo,
P. Sancho-Bru, W. Jimenez, V. Arroyo, and J. Rodes. 2001. Human hepatic
stellate cells secrete adrenomedullin: potential autocrine factor in the regulation of
cell contractility. J Hepatol 34:222-9.
34. Gross, J., and C. M. Lapiere. 1962. Collagenolytic activity in amphibian tissues: a
tissue culture assay. Proc Natl Acad Sci U S A 48:1014-22.
35. Grozinger, C. M., and S. L. Schreiber. 2000. Regulation of histone deacetylase 4
and 5 and transcriptional activity by 14-3-3-dependent cellular localization. Proc
Natl Acad Sci U S A 97:7835-40.
36. Han, Y. P., L. Zhou, J. Wang, S. Xiong, W. L. Garner, S. W. French, and H.
Tsukamoto. 2004. Essential role of matrix metalloproteinases in interleukin-1-
induced myofibroblastic activation of hepatic stellate cell in collagen. J Biol
Chem 279:4820-8.
37. Harper, E., K. J. Bloch, and J. Gross. 1971. The zymogen of tadpole collagenase.
Biochemistry 10:3035-41.
91

38. Hattori, N., S. Mochizuki, K. Kishi, T. Nakajima, H. Takaishi, J. D'Armiento, and
Y. Okada. 2009. MMP-13 plays a role in keratinocyte migration, angiogenesis,
and contraction in mouse skin wound healing. Am J Pathol 175:533-46.
39. Hazra, S., S. Xiong, J. Wang, R. A. Rippe, V. Krishna, K. Chatterjee, and H.
Tsukamoto. 2004. Peroxisome proliferator-activated receptor gamma induces a
phenotypic switch from activated to quiescent hepatic stellate cells. J Biol Chem
279:11392-401.
40. Heinzel, T., R. M. Lavinsky, T. M. Mullen, M. Soderstrom, C. D. Laherty, J.
Torchia, W. M. Yang, G. Brard, S. D. Ngo, J. R. Davie, E. Seto, R. N. Eisenman,
D. W. Rose, C. K. Glass, and M. G. Rosenfeld. 1997. A complex containing N-
CoR, mSin3 and histone deacetylase mediates transcriptional repression. Nature
387:43-8.
41. Hubbert, C., A. Guardiola, R. Shao, Y. Kawaguchi, A. Ito, A. Nixon, M. Yoshida,
X. F. Wang, and T. P. Yao. 2002. HDAC6 is a microtubule-associated
deacetylase. Nature 417:455-8.
42. Ikeda, K., T. Wakahara, Y. Q. Wang, H. Kadoya, N. Kawada, and K. Kaneda.
1999. In vitro migratory potential of rat quiescent hepatic stellate cells and its
augmentation by cell activation. Hepatology 29:1760-7.
43. Iredale, J. P., R. C. Benyon, M. J. Arthur, W. F. Ferris, R. Alcolado, P. J.
Winwood, N. Clark, and G. Murphy. 1996. Tissue inhibitor of metalloproteinase-
1 messenger RNA expression is enhanced relative to interstitial collagenase
messenger RNA in experimental liver injury and fibrosis. Hepatology 24:176-84.
44. Ito, K., P. J. Barnes, and I. M. Adcock. 2000. Glucocorticoid receptor recruitment
of histone deacetylase 2 inhibits interleukin-1beta-induced histone H4 acetylation
on lysines 8 and 12. Mol Cell Biol 20:6891-903.
45. Ito, T., and M. Nemoto. 1952. [Kupfer's cells and fat storing cells in the capillary
wall of human liver.]. Okajimas Folia Anat Jpn 24:243-58.
46. Juan, L. J., W. J. Shia, M. H. Chen, W. M. Yang, E. Seto, Y. S. Lin, and C. W.
Wu. 2000. Histone deacetylases specifically down-regulate p53-dependent gene
activation. J Biol Chem 275:20436-43.
92

47. Kao, H. Y., A. Verdel, C. C. Tsai, C. Simon, H. Juguilon, and S. Khochbin. 2001.
Mechanism for nucleocytoplasmic shuttling of histone deacetylase 7. J Biol Chem
276:47496-507.
48. Kent, G., S. Gay, T. Inouye, R. Bahu, O. T. Minick, and H. Popper. 1976.
Vitamin A-containing lipocytes and formation of type III collagen in liver injury.
Proc Natl Acad Sci U S A 73:3719-22.
49. Kim, J. S., and S. D. Shukla. 2005. Histone h3 modifications in rat hepatic stellate
cells by ethanol. Alcohol Alcohol 40:367-72.
50. Kinnman, N., O. Goria, D. Wendum, M. C. Gendron, C. Rey, R. Poupon, and C.
Housset. 2001. Hepatic stellate cell proliferation is an early platelet-derived
growth factor-mediated cellular event in rat cholestatic liver injury. Lab Invest
81:1709-16.
51. Kirsh, O., J. S. Seeler, A. Pichler, A. Gast, S. Muller, E. Miska, M. Mathieu, A.
Harel-Bellan, T. Kouzarides, F. Melchior, and A. Dejean. 2002. The SUMO E3
ligase RanBP2 promotes modification of the HDAC4 deacetylase. Embo J
21:2682-91.
52. Knolle, P. A., E. Loser, U. Protzer, R. Duchmann, E. Schmitt, K. H. zum
Buschenfelde, S. Rose-John, and G. Gerken. 1997. Regulation of endotoxin-
induced IL-6 production in liver sinusoidal endothelial cells and Kupffer cells by
IL-10. Clin Exp Immunol 107:555-61.
53. Krull, N. B., T. Zimmermann, and A. M. Gressner. 1993. Spatial and temporal
patterns of gene expression for the proteoglycans biglycan and decorin and for
transforming growth factor-beta 1 revealed by in situ hybridization during
experimentally induced liver fibrosis in the rat. Hepatology 18:581-9.
54. Langlois, S., D. Gingras, and R. Beliveau. 2004. Membrane type 1-matrix
metalloproteinase (MT1-MMP) cooperates with sphingosine 1-phosphate to
induce endothelial cell migration and morphogenic differentiation. Blood
103:3020-8.

93

55. Liu, X., J. Yu, L. Jiang, A. Wang, F. Shi, H. Ye, and X. Zhou. 2009. MicroRNA-
222 regulates cell invasion by targeting matrix metalloproteinase 1 (MMP1) and
manganese superoxide dismutase 2 (SOD2) in tongue squamous cell carcinoma
cell lines. Cancer Genomics Proteomics 6:131-9.
56. Longworth, M. S., and L. A. Laimins. 2006. Histone deacetylase 3 localizes to the
plasma membrane and is a substrate of Src. Oncogene 25:4495-500.
57. Luger, K., A. W. Mader, R. K. Richmond, D. F. Sargent, and T. J. Richmond.
1997. Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature
389:251-60.
58. Ma, Z., R. C. Shah, M. J. Chang, and E. N. Benveniste. 2004. Coordination of cell
signaling, chromatin remodeling, histone modifications, and regulator recruitment
in human matrix metalloproteinase 9 gene transcription. Mol Cell Biol 24:5496-
509.
59. Maher, J. J., D. M. Bissell, S. L. Friedman, and F. J. Roll. 1988. Collagen
measured in primary cultures of normal rat hepatocytes derives from lipocytes
within the monolayer. J Clin Invest 82:450-9.
60. Mal, A., M. Sturniolo, R. L. Schiltz, M. K. Ghosh, and M. L. Harter. 2001. A role
for histone deacetylase HDAC1 in modulating the transcriptional activity of
MyoD: inhibition of the myogenic program. Embo J 20:1739-53.
61. Mann, J., D. C. Chu, A. Maxwell, F. Oakley, N. L. Zhu, H. Tsukamoto, and D. A.
Mann. 2009. MeCP2 Controls an Epigenetic Pathway That Promotes
Myofibroblast Transdifferentiation and Fibrosis. Gastroenterology.
62. Mann, J., F. Oakley, F. Akiboye, A. Elsharkawy, A. W. Thorne, and D. A. Mann.
2007. Regulation of myofibroblast transdifferentiation by DNA methylation and
MeCP2: implications for wound healing and fibrogenesis. Cell Death Differ
14:275-85.

94

63. Marquez-Aguirre, A., A. Sandoval-Rodriguez, J. Gonzalez-Cuevas, M. Bueno-
Topete, J. Navarro-Partida, I. Arellano-Olivera, S. Lucano-Landeros, and J.
Armendariz-Borunda. 2009. Adenoviral delivery of dominant-negative
transforming growth factor beta type II receptor up-regulates transcriptional
repressor SKI-like oncogene, decreases matrix metalloproteinase 2 in hepatic
stellate cell and prevents liver fibrosis in rats. J Gene Med 11:207-19.
64. McGee, J. O., and R. S. Patrick. 1972. The role of perisinusoidal cells in
experimental hepatic fibrogenesis. J Pathol 106:Pvi.
65. McGee, J. O., and R. S. Patrick. 1972. The role of perisinusoidal cells in hepatic
fibrogenesis. An electron microscopic study of acute carbon tetrachloride liver
injury. Lab Invest 26:429-40.
66. McKinsey, T. A., C. L. Zhang, J. Lu, and E. N. Olson. 2000. Signal-dependent
nuclear export of a histone deacetylase regulates muscle differentiation. Nature
408:106-11.
67. McKinsey, T. A., C. L. Zhang, and E. N. Olson. 2001. Identification of a signal-
responsive nuclear export sequence in class II histone deacetylases. Mol Cell Biol
21:6312-21.
68. Miska, E. A., E. Langley, D. Wolf, C. Karlsson, J. Pines, and T. Kouzarides. 2001.
Differential localization of HDAC4 orchestrates muscle differentiation. Nucleic
Acids Res 29:3439-47.
69. Miyahara, T., L. Schrum, R. Rippe, S. Xiong, H. F. Yee, Jr., K. Motomura, F. A.
Anania, T. M. Willson, and H. Tsukamoto. 2000. Peroxisome proliferator-
activated receptors and hepatic stellate cell activation. J Biol Chem 275:35715-22.
70. Nakatsukasa, H., P. Nagy, R. P. Evarts, C. C. Hsia, E. Marsden, and S. S.
Thorgeirsson. 1990. Cellular distribution of transforming growth factor-beta 1 and
procollagen types I, III, and IV transcripts in carbon tetrachloride-induced rat
liver fibrosis. J Clin Invest 85:1833-43.

95

71. Niki, T., K. Rombouts, P. De Bleser, K. De Smet, V. Rogiers, D. Schuppan, M.
Yoshida, G. Gabbiani, and A. Geerts. 1999. A histone deacetylase inhibitor,
trichostatin A, suppresses myofibroblastic differentiation of rat hepatic stellate
cells in primary culture. Hepatology 29:858-67.
72. Oakley, F., M. Meso, J. P. Iredale, K. Green, C. J. Marek, X. Zhou, M. J. May, H.
Millward-Sadler, M. C. Wright, and D. A. Mann. 2005. Inhibition of inhibitor of
kappaB kinases stimulates hepatic stellate cell apoptosis and accelerated recovery
from rat liver fibrosis. Gastroenterology 128:108-20.
73. Oh, S. H., H. M. Hatch, and B. E. Petersen. 2002. Hepatic oval 'stem' cell in liver
regeneration. Semin Cell Dev Biol 13:405-9.
74. Olaso, E., K. Ikeda, F. J. Eng, L. Xu, L. H. Wang, H. C. Lin, and S. L. Friedman.
2001. DDR2 receptor promotes MMP-2-mediated proliferation and invasion by
hepatic stellate cells. J Clin Invest 108:1369-78.
75. Pinzani, M., H. E. Abboud, and D. C. Aron. 1990. Secretion of insulin-like
growth factor-I and binding proteins by rat liver fat-storing cells: regulatory role
of platelet-derived growth factor. Endocrinology 127:2343-9.
76. Potthoff, M. J., H. Wu, M. A. Arnold, J. M. Shelton, J. Backs, J. McAnally, J. A.
Richardson, R. Bassel-Duby, and E. N. Olson. 2007. Histone deacetylase
degradation and MEF2 activation promote the formation of slow-twitch
myofibers. J Clin Invest 117:2459-67.
77. Qi, Z., N. Atsuchi, A. Ooshima, A. Takeshita, and H. Ueno. 1999. Blockade of
type beta transforming growth factor signaling prevents liver fibrosis and
dysfunction in the rat. Proc Natl Acad Sci U S A 96:2345-9.
78. Ramadori, G., H. Rieder, T. Knittel, H. P. Dienes, and K. H. Meyer zum
Buschenfelde. 1987. Fat storing cells (FSC) of rat liver synthesize and secrete
fibronectin. Comparison with hepatocytes. J Hepatol 4:190-7.

96

79. Rawdanowicz, T. J., A. L. Hampton, H. Nagase, D. E. Woolley, and L. A.
Salamonsen. 1994. Matrix metalloproteinase production by cultured human
endometrial stromal cells: identification of interstitial collagenase, gelatinase-A,
gelatinase-B, and stromelysin-1 and their differential regulation by interleukin-1
alpha and tumor necrosis factor-alpha. J Clin Endocrinol Metab 79:530-6.
80. Razin, A. 1998. CpG methylation, chromatin structure and gene silencing-a three-
way connection. Embo J 17:4905-8.
81. Reeves, H. L., and S. L. Friedman. 2002. Activation of hepatic stellate cells--a
key issue in liver fibrosis. Front Biosci 7:d808-26.
82. Rockey, D. C., and R. A. Weisiger. 1996. Endothelin induced contractility of
stellate cells from normal and cirrhotic rat liver: implications for regulation of
portal pressure and resistance. Hepatology 24:233-40.
83. Rombouts, K., T. Knittel, L. Machesky, F. Braet, A. Wielant, K. Hellemans, P.
De Bleser, I. Gelman, G. Ramadori, and A. Geerts. 2002. Actin filament
formation, reorganization and migration are impaired in hepatic stellate cells
under influence of trichostatin A, a histone deacetylase inhibitor. J Hepatol
37:788-96.
84. Scheller, M., B. Zimmermann, J. P. Bernimoulin, and P. Scholz. 1995. Induction
of metalloproteinase activity, cartilage matrix degradation and inhibition of
endochondral mineralization in vitro by E. coli lipopolysaccharide is mediated by
interleukin 1 alpha. Cytokine 7:331-7.
85. Sicklick, J. K., Y. X. Li, S. S. Choi, Y. Qi, W. Chen, M. Bustamante, J. Huang, M.
Zdanowicz, T. Camp, M. S. Torbenson, M. Rojkind, and A. M. Diehl. 2005. Role
for hedgehog signaling in hepatic stellate cell activation and viability. Lab Invest
85:1368-80.
86. Singer, C. F., E. Marbaix, I. Kokorine, P. Lemoine, J. Donnez, Y. Eeckhout, and
P. J. Courtoy. 2000. The matrix metalloproteinase-1 (MMP-1) expression in the
human endometrium is inversely regulated by interleukin-1 alpha and sex steroids.
Ceska Gynekol 65:211-5.
97

87. Springman, E. B., E. L. Angleton, H. Birkedal-Hansen, and H. E. Van Wart. 1990.
Multiple modes of activation of latent human fibroblast collagenase: evidence for
the role of a Cys73 active-site zinc complex in latency and a "cysteine switch"
mechanism for activation. Proc Natl Acad Sci U S A 87:364-8.
88. Strongin, A. Y., I. Collier, G. Bannikov, B. L. Marmer, G. A. Grant, and G. I.
Goldberg. 1995. Mechanism of cell surface activation of 72-kDa type IV
collagenase. Isolation of the activated form of the membrane metalloprotease. J
Biol Chem 270:5331-8.
89. Takahra, T., D. E. Smart, F. Oakley, and D. A. Mann. 2004. Induction of
myofibroblast MMP-9 transcription in three-dimensional collagen I gel cultures:
regulation by NF-kappaB, AP-1 and Sp1. Int J Biochem Cell Biol 36:353-63.
90. Tatti, O., P. Vehvilainen, K. Lehti, and J. Keski-Oja. 2008. MT1-MMP releases
latent TGF-beta1 from endothelial cell extracellular matrix via proteolytic
processing of LTBP-1. Exp Cell Res 314:2501-14.
91. Thornberry, N. A., H. G. Bull, J. R. Calaycay, K. T. Chapman, A. D. Howard, M.
J. Kostura, D. K. Miller, S. M. Molineaux, J. R. Weidner, J. Aunins, and et al.
1992. A novel heterodimeric cysteine protease is required for interleukin-1 beta
processing in monocytes. Nature 356:768-74.
92. Tsukamoto, H. 1999. Cytokine regulation of hepatic stellate cells in liver fibrosis.
Alcohol Clin Exp Res 23:911-6.
93. Uemura, M., E. S. Swenson, M. D. Gaca, F. J. Giordano, M. Reiss, and R. G.
Wells. 2005. Smad2 and Smad3 play different roles in rat hepatic stellate cell
function and alpha-smooth muscle actin organization. Mol Biol Cell 16:4214-24.
94. Urtasun, R., F. J. Cubero, M. Vera, and N. Nieto. 2009. Reactive nitrogen species
switch on early extracellular matrix remodeling via induction of MMP1 and
TNFalpha. Gastroenterology 136:1410-22, e1-4.
95. Van Wart, H. E., and H. Birkedal-Hansen. 1990. The cysteine switch: a principle
of regulation of metalloproteinase activity with potential applicability to the entire
matrix metalloproteinase gene family. Proc Natl Acad Sci U S A 87:5578-82.
98

96. Wake, K. 1980. Perisinusoidal stellate cells (fat-storing cells, interstitial cells,
lipocytes), their related structure in and around the liver sinusoids, and vitamin A-
storing cells in extrahepatic organs. Int Rev Cytol 66:303-53.
97. Wake, K. 1971. "Sternzellen" in the liver: perisinusoidal cells with special
reference to storage of vitamin A. Am J Anat 132:429-62.
98. Wandzioch, E., A. Kolterud, M. Jacobsson, S. L. Friedman, and L. Carlsson. 2004.
Lhx2-/- mice develop liver fibrosis. Proc Natl Acad Sci U S A 101:16549-54.
99. Wang, A. H., M. J. Kruhlak, J. Wu, N. R. Bertos, M. Vezmar, B. I. Posner, D. P.
Bazett-Jones, and X. J. Yang. 2000. Regulation of histone deacetylase 4 by
binding of 14-3-3 proteins. Mol Cell Biol 20:6904-12.
100. Wang, A. H., and X. J. Yang. 2001. Histone deacetylase 4 possesses intrinsic
nuclear import and export signals. Mol Cell Biol 21:5992-6005.
101. Wang, X., X. Tang, X. Gong, E. Albanis, S. L. Friedman, and Z. Mao. 2004.
Regulation of hepatic stellate cell activation and growth by transcription factor
myocyte enhancer factor 2. Gastroenterology 127:1174-88.
102. Wang, Z., C. Zang, K. Cui, D. E. Schones, A. Barski, W. Peng, and K. Zhao.
2009. Genome-wide mapping of HATs and HDACs reveals distinct functions in
active and inactive genes. Cell 138:1019-31.
103. Wells, R. G. 2005. The role of matrix stiffness in hepatic stellate cell activation
and liver fibrosis. J Clin Gastroenterol 39:S158-61.
104. Will, H., S. J. Atkinson, G. S. Butler, B. Smith, and G. Murphy. 1996. The
soluble catalytic domain of membrane type 1 matrix metalloproteinase cleaves the
propeptide of progelatinase A and initiates autoproteolytic activation. Regulation
by TIMP-2 and TIMP-3. J Biol Chem 271:17119-23.
105. Xu, J., Y. Fu, and A. Chen. 2003. Activation of peroxisome proliferator-activated
receptor-gamma contributes to the inhibitory effects of curcumin on rat hepatic
stellate cell growth. Am J Physiol Gastrointest Liver Physiol 285:G20-30.
99

106. Yan, C., and D. D. Boyd. 2007. Regulation of matrix metalloproteinase gene
expression. J Cell Physiol 211:19-26.
107. Yan, C., H. Wang, Y. Toh, and D. D. Boyd. 2003. Repression of 92-kDa type IV
collagenase expression by MTA1 is mediated through direct interactions with the
promoter via a mechanism, which is both dependent on and independent of
histone deacetylation. J Biol Chem 278:2309-16.
108. Yan, C., L. Zhou, and Y. P. Han. 2008. Contribution of hepatic stellate cells and
matrix metalloproteinase 9 in acute liver failure. Liver Int 28:959-71.
109. Yang, X. J., and S. Gregoire. 2005. Class II histone deacetylases: from sequence
to function, regulation, and clinical implication. Mol Cell Biol 25:2873-84.
110. Yu, Q., and I. Stamenkovic. 2000. Cell surface-localized matrix
metalloproteinase-9 proteolytically activates TGF-beta and promotes tumor
invasion and angiogenesis. Genes Dev 14:163-76.
111. Zhou, X., A. Jamil, A. Nash, J. Chan, N. Trim, J. P. Iredale, and R. C. Benyon.
2006. Impaired proteolysis of collagen I inhibits proliferation of hepatic stellate
cells: implications for regulation of liver fibrosis. J Biol Chem 281:39757-65.

Abstract (if available)

Abstract It is well established that matrix metalloproteinases (MMPs) play important roles in tissue injury, cell differentiation, and cancer metastasis. In liver injury, hepatic stellate cells (HSCs) express high level of MMPs and undergo trans-differentiation to become myofibroblast like cells. However, largely unknown is that how MMP genes are repressed or silenced in tissue fibrosis to favor ECM accumulation. In this study, it was investigated how MMP genes were progressively repressed in fibrotic liver and myofibroblastic HSCs. First, it was shown that upon interleukin-1 (IL-1) stimulation, the major signaling pathways that are essential for MMP expression were as active in myofibroblastic HSCs as in quiescent HSCs, indicating the repression was attained at chromatin levels. Indeed, it was found that in myofibroblastic HSCs both MMP-9 and MMP-13 genes were less accessible for transcriptional factors, which was associated with impaired histone acetylation and RNA polymerase II recruitment to the promoters. Further, it was observed that during HSC trans-differentiation, the class II histone deacetylase HDAC4 accumulated, in accordance with global reduction of histone acetylation. To demonstrate a causal relationship of HDAC4 elevation in MMP gene repression, HDAC4 was ectopically expressed in quiescent HSCs, which resulted in sufficient repression of MMP promoter activities as well as endogenous MMP-9 protein expression. Thus, a mechanism was uncovered about how MMP genes are epigenetically silenced in tissue fibrogenesis partially through HDAC4 accumulation.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Determination of the source of myofibroblasts and therapeutic strategy against fibrosis

PDF

Regulation of matrix metalloproteinase-9 in cutaneous wound healing

PDF

Using novel small molecule modulators as a tool to elucidate the role of the Myocyte Enhancer Factor 2 (MEF2) family of transcription factors in leukemia

Asset Metadata

Creator Qin, Lan (author)

Core Title Histone deacetylase 4 represses matrix metalloproteinases in myofibroblastic hepatic stellate cells

Contributor Electronically uploaded by the author (provenance)

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Systems Biology

Publication Date 03/11/2010

Defense Date 03/05/2010

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag hepatic stellate cells,histone deacetylation,liver fibrosis,matrix metalloproteinases,OAI-PMH Harvest

Language English

Advisor Han, Yuan-Ping (committee chair), Tsukamoto, Hidekazu (committee member), Zandi, Ebrahim (committee member)

Creator Email jiujiu04@hotmail.com,lanqin@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2872

Unique identifier UC1309548

Identifier etd-Qin-3560 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-292510 (legacy record id),usctheses-m2872 (legacy record id)

Legacy Identifier etd-Qin-3560.pdf

Dmrecord 292510

Document Type Dissertation

Rights Qin, Lan

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu