Clinics in Liver Disease, Vol 12 PDF

Hepatic Fibrosis: Pathogenesis, Diagnosis, and Emerging Therapies
Preface
Scott L. Friedman, MD
Guest Editor
The field of hepatic fibrosis continues to inch closer to effective therapies because of
sustained investment in understanding the basic, translational, and clinical compo-
nents of this topic. Gratifying progress has been made on many fronts, and it is
my intention to capture these advances in this issue of Clinics in Liver Disease. I
am extremely privileged to have secured the participation of the world’s leading
experts in this area, and the result is a comprehensive and truly up-to-date compi-
lation that spans the full spectrum of science and practice in hepatic fibrosis.
First, Drs. Kim and Lim have outlined the global challenges presented by the world-
wide epidemics of viral, metabolic, and neoplastic liver disease, which make painfully
clear how urgent the need is for progress. Yet, not all individuals with chronic liver dis-
ease are at equal risk of fibrosis progression, and Dr. Lammert and coauthors review
the exciting inroads in defining the genetic basis for risk of fibrosis. At the same time,
studies elucidating the molecular basis of fibrosis have more clearly defined therapeu-
tic targets. In this context, Drs. Wells, Nieto, Yee, George, and Bataller and their coau-
thors review current concepts regarding the cellular sources of extracellular matrix,
stimuli to fibrogenesis and cellular contractility, and the role of cytokines and the re-
nin-angiotensin system. Among the most intriguing areas of recent progress has
been the emerging role of hepatic stellate cells—long recognized as a key source of
extracellular matrix in liver injury—in interacting with, and possibly trans-differentiating
into, hepatic progenitors. To address this topic, Dr. Roskams offers unique expertise
by highlighting intriguing new concepts in liver regeneration and repair. Another hot
area is the interactions between immune and fibrogenic cells in liver, an area thought-
fully summarized by Dr. Adams and his colleagues.
In addition to these basic advances, progress has continued in defining novel
approaches to the diagnosis and therapy of hepatic fibrosis. Drs. Guha and Rosenberg
offer a concise, yet comprehensive, summary of non-invasive diagnostics that are pro-
gressively likely to replace liver biopsy for the staging of fibrosis. Some of these tech-
niques are based on features of hepatic stiffness and contractility, the underlying basis
of which is beautifully summarized by Drs. Pinzani and Vizzuti. If anti-fibrotic therapies
are to succeed, they will do so based on a clearer understanding of mechanisms
Clin Liver Dis 12 (2008) xiii–xiv

doi:10.1016/j.cld.2008.07.009 liver.theclinics.com
1089-3261/08/$ – see front matter ª 2008 Elsevier Inc. All rights reserved.
xiv Preface
underlying fibrosis reversibility, and Dr. Mann and coauthors present a thoughtful and
detailed summary of this rapidly advancing body of knowledge. Finally, Dr. Rockey
reviews the many points of potential therapeutic attack in an increasingly promising
panoply of emerging treatment options.
While this volume is a unique compendium of state-of-the-art reviews on the topic
of hepatic fibrosis, it is my sincere hope that progress will be so rapid as to render the
content here out of date in only a few years. At that point, I am confident that any
contemporary review of the field will include recommendations based on successful
clinical trials that have begun to transform the outlook for millions of patients world-
wide with chronic liver disease.
Scott L. Friedman, MD
Division of Liver Diseases
Box 1123, Mount Sinai School of Medicine
1425 Madison Ave., Room 1170C
New York, NY 10029
E-mail address:
Scott.Friedman@mssm.edu (S.L. Friedman)
The Global I mpact
of Hepatic Fibrosis
a nd End - St age Liver
Dis eas e
Young-Suk Lim, MD, W. Ray Kim, MD*
KEYWORDS
Cirrhosis End-stage liver disease Epidemiology
Global impact Hepatic fibrosis
Cirrhosis is the end result of many types of liver disease and it consists of fibrosis and
nodular regeneration. Both fibrosis and cirrhosis are the consequences of a sustained
wound-healing response to chronic liver injury.1 Cirrhosis eventually leads to dimin-
ished hepatic function and altered blood flow. Hepatic fibrosis and cirrhosis are, in
principle, defined morphologically and the pattern and extent of the morphologic
changes depend on the cause and stage of fibrosis. Accordingly, there is a wide
spectrum in the degree of fibrosis and in the severity of clinical symptoms. Clinical pre-
sentation may vary widely, ranging from absent to nonspecific symptoms to life-
threatening conditions. In most cases, no clear dividing line can be drawn between
cirrhosis and the preceding liver disease because the transition is gradual and
unapparent.
Chronic liver disease and cirrhosis occur throughout the world, regardless of race,
age or gender. However, there are marked geographic variations in incidence and
prevalence, largely depending on the prevalence of causative factors. Although virtu-
ally any chronic liver disease may progress to cirrhosis, the most common causes of
liver fibrosis and cirrhosis globally are thought to be hepatitis B virus (HBV), hepatitis C
virus (HCV), and alcohol. Other causes include: immune-mediated liver injury, hepato-
toxic drugs, cholestatic diseases, genetic abnormalities, and nonalcoholic steatohe-
patitis. Viral hepatitis, especially HBV, is the leading cause of cirrhosis in developing
countries, whereas alcohol, HCV, and, more recently, nonalcoholic steatohepatitis
are the most significant causes of cirrhosis in the developed world.2
This work was supported by NIH Grant DK-34238.

Division of Gastroenterology and Hepatology (PL 6), Mayo Clinic College of Medicine, 200 First
Street SW, Rochester, MN 55905, USA
* Corresponding author.
E-mail address: kim.ray@mayo.edu (W.R. Kim).
Clin Liver Dis 12 (2008) 733–746

734 Lim & Kim
SOURCES AND LIMITATIONS OF GLOBAL EPIDEMIOLOGIC DATA OF CHRONIC LIVER DISEASE
The estimated prevalence of cirrhosis, as identified from autopsy studies ranges from
4.5% to 9.5% of the general population, which would project to hundreds of millions of
patients affected with cirrhosis worldwide.3,4 However, the precise incidence or prev-
alence of cirrhosis is difficult to ascertain because cirrhosis is often clinically silent.2
Up to 40% of patients with cirrhosis are asymptomatic and may remain so for more
than a decade.5,6 While a liver biopsy is required to establish the diagnosis of cirrhosis,
the absence of accurate, validated, noninvasive diagnostic tools makes population
screening difficult.2
The public health impact of chronic liver disease can be gauged from the reported
death rate. However, data based on death certificates have a number of limitations.
First, mortality statistics based on death certificates can only be obtained in settings
when there is necessary infrastructure for data collection. Second, even if these re-
sources are in place, variability in the documentation of cirrhosis will influence its re-
porting.2 The numbers of deaths by cirrhosis or chronic liver disease are likely to be
greatly underestimated, especially if only the primary cause of death is considered.
In fact, deaths from cirrhosis may be classified as a number of other categories. For
example, in a decedent who died of cirrhosis, the primary cause of death may be listed
under the underlying etiology such as, hepatitis B or C or alcoholic liver disease, or un-
der complications of cirrhosis such as, primary liver cancer or upper gastrointestinal
bleeding.7 In a recent report that analyzed chronic liver disease, mortality rates using
databases from a health maintenance organization in northern California, a compre-
hensive approach using a broad spectrum of death codes, including viral hepatitis
and hepatocellular carcinoma, led to detection of almost twice as many confirmed
deaths related to chronic liver disease as did the conventional method that included
only limited codes such as chronic hepatitis and cirrhosis.8 This suggests that
methods used conventionally for national and statewide statistics in the United States
substantially underestimate the true rates of mortality associated with cirrhosis.
Obviously, the death rate may not always be a valid surrogate for prevalence as
there may not necessarily be a 1:1 correlation between prevalence of liver disease
and mortality. For example, if treatment significantly improves over time, the preva-
lence of a condition in the community will rise, but the death rate from the given con-
dition may appear to fall.2 In the case of chronic liver disease and cirrhosis,
widespread application of antiviral agents and liver transplantation, as well as im-
provement in the treatment of complications of cirrhosis, may lead to a decrease in
cirrhosis mortality, especially in developed countries.
All of these factors make it difficult to estimate the true prevalence and burden of
cirrhosis, especially in the absence of hepatic decompensation. Patients with com-
pensated cirrhosis, however, often progress to decompensation including develop-
ment of ascites, variceal hemorrhage, or encephalopathy. In such patients, the
probability of eventual death from liver failure is high. The 5-year mortality has been
reported to be 50% and approximately 70% of these deaths are directly attributable
to liver disease.9
GLOBAL BURDEN OF CHRONIC LIVER DISEASE

Cirrhosis
In 2001, 771,000 people are estimated to have died from cirrhosis worldwide, ranking
14th and 10th as leading cause of death in the world and in developed countries, re-
spectively.10 Worldwide, deaths from cirrhosis have been projected to increase to
make it the 12th cause of death in 2020.11 In the 2001 data, cirrhosis was the sixth
Global Impact of Hepatic Fibrosis 735
most common cause of death among adults in developed countries, accounting for
4.4% of all deaths. (Fig. 1) In low to middle income countries, cirrhosis was ranked
as the 9th cause of death, claiming 320,000 lives.10
According to a World Health Organization (WHO) database that incorporates mor-
tality data from 41 countries worldwide, age-adjusted mortality rates from cirrhosis
were the highest in some countries in Central and South America (eg, Mexico and
Chile, 55/100,000 in men and 14/100,000 in women) and in southern Europe (eg,
France, Italy, Portugal, Austria, Hungary and Romania, 30–35/100,000 in men and
10–15/100,000 in women) in the early 1980s.12 Since the mid- to late-1970s, mortality
from cirrhosis showed decreasing trends in most countries of Europe and America, co-
inciding with a reduction in alcohol consumption (annual percent change, APC, -5.0%
to -1.5%).12,13 The steadily declining mortality from cirrhosis was particularly obvious in
southern European countries.12,14 In these countries, rates in the early 2000s de-
creased by more than 50% compared with earlier decades. Meanwhile, mortality rates
in central and eastern European countries rose to reach a peak in the mid-1990s (eg,
rates over 58/100,000 men and 22/100,000 women in Hungary in 1990) and then
decreased thereafter. Mortality rates in Northern European countries reportedly show
a gradual yet continued increase.
A substantial increase in cirrhosis mortality over the last two decades has also been
reported for the United Kingdom (UK) (APC around 17% in men and 13% in women in
England and Wales; 19% in men and 17% in women in Scotland). However, cirrhosis
mortality in the UK remained lower than the level reported in Central and Eastern
European countries. Mortality rates from cirrhosis were lower in women from all coun-
tries, but the time trend was similar in both genders.12 Similar to data from neighboring
European countries, the increase in cirrhosis mortality in the UK has largely been at-
tributed to a recent rise in alcohol consumption.12–15 A correlation between the total
alcohol consumption and mortality from cirrhosis has been demonstrated, regardless
of the type of alcoholic beverages consumed (Fig. 2).12,16
The data discussed so far are limited because they did not include data from many
developing countries with high prevalence of Hepatitis B virus (HBV) infection, such as
Southeast Asia, western Pacific countries, and sub-Saharan African countries. This
omission obviously leads to underestimation of the worldwide burden of chronic liver
diseases. HBV infection is a global health problem, affecting approximately 2 billion
people (ie, one third of the world’s population) based on serologic evidence of past
or present HBV infection, including 360 million people with chronic HBV infection.17,18
Of subjects who have chronic HBV infection, 15% to 40% develop serious sequelae
during their lifetime, such as, cirrhosis, hepatic decompensation, and hepatocellular
carcinoma (HCC).19 Globally, most HBV-related deaths result from the chronic se-
quelae of infection. For the year 2000, it was estimated that, of the total 620,000
deaths from HBV-related causes, 580,000 (94%) were from cirrhosis and HCC.20
Although HBV is globally present, its endemic areas include: Southeast Asia, China,
Pacific Islands, Sub-Saharan Africa, Alaska, Peru, and northwestern Brazil.19,21 Peo-
ple living in these areas occupy about 45% of the global population. In developed
countries, HBV infection tends to be more prevalent in certain population groups,
such as, immigrants from endemic areas.22
In endemic areas with HBV infection, almost all infections occur during either the
perinatal period or early in childhood; this pattern accounts for the high rates of
chronic HBV infection in these populations.19 Safe and effective vaccines have
been available to prevent HBV infection since 1981.23,24 In 1992, the WHO recommen-
ded that all countries include HBV vaccine in their routine infant immunization pro-
grams.20 However, in 2000, only 116 of 215 countries had such a policy,
736
Lim & Kim
High Income Countries Low Income Countries
% of total deaths % of total deaths
0 2 4 6 8 10 12 0 2 4 6 8 10 12 14 16
Ischemic heart disease HIV/AIDS
Self-inflected injuries Ischemic heart disease
Road traffic accidents Tuberculosis
Trachea, bronchus, and lung cancers Road traffic accidents
Cerebrovascular disease Cerebrovascular disease
Cirrhosis of the liver 4.4% Self inflicted injuries
Breast cancer Violence
Colon and rectal cancers Lower respiratory infections
Diabetes mellitus Cirrhosis of the liver 2.2%
Stomach cancer Chronic obstructive pulmonary disease
Fig.1. The 10 Leading Causes of Death in Adults Ages 15 to 59, by Broad Income Group, 2001. (Adapted from Mathers C, Lopez A, Murray C. The burden
of disease and mortality by condition: data, methods, and results for 2001. In: Lopez A, Mathers C, Ezzati M, et al, editors. Global burden of disease and
risk factors. Washington (DC): Oxford University Press and the World Bank; 2006. p. 45–93; with permission of Oxford University Press, Copyright ª
2006.)
Fig. 2. Correlation between cirrhosis mortality (^) and per capital consumption of ethanol
( ): (A) United Kingdom and (B) United States. (From Kerr WC, Fillmore KM, Marvy P. Bev-
erage-specific alcohol consumption and cirrhosis mortality in a group of English-speaking
beer-drinking countries. Addiction 2000;95(3):339–46. Reprinted with permission of Wiley-
Blackwell Publishing, Copyright ª 2007.)
representing 31% of the global birth cohort.25 Although the counties that had adopted
the universal childhood HBV vaccination policies increased to 79% by 2003, vaccina-
tion coverage rate remained low in many countries that had introduced the vaccine.26
Thus, despite the availability of an effective vaccine, a significant number of the
world’s children remain at risk for HBV infection. In addition, because HBV-associated
cirrhosis usually does not manifest until the fifth decade in life or later, chronic liver dis-
ease secondary to HBV is expected to remain a major public health problem world-
wide until the cohort of vaccinated children reach adulthood.27
Hepatocellular Carcinoma (HCC)

The majority of patients with HCC worldwide have underlying liver disease, most com-
monly, cirrhosis.28 With the exception of some areas of the world where the role of cer-
tain oncogenic agents (ie, aflatoxin) may be important, it is uncommon to find HCC in
the absence of advanced hepatic fibrosis or cirrhosis.28 Therefore, HCC is a serious
complication of chronic liver diseases.
Estimates from the 2000 WHO mortality databank indicate that primary liver cancer
was the fifth most common malignancy in men and the ninth in women, accounting for
approximately 5.6% of all human cancers.29 The mortality burden of liver cancer is
738 Lim & Kim
projected to increase through 2020.11 The number of new cases is estimated to be

564,000 per year, including 398,000 in men and 166,000 in women.30 With few excep-
tions, such as areas in the Philippines where intrahepatic cholangiocarcinoma caused
by liver flukes is common, HCC accounts for 70% to 85% of primary liver cancers.31
Liver cancer is rapidly fatal in the majority of patients, which makes its incidence-to-
mortality ratio very close to one.31
As the development of HCC is generally the end result of long-standing, typically
asymptomatic chronic liver disease,32 the incidence of HCC is linked closely to the
prevalence of major causes of cirrhosis, such as, chronic viral hepatitis and excessive
alcohol consumption. By and large, the global geographic variability in the incidence
of HCC is largely explained by the distribution and the natural history of HBV and HCV.
The attributable risk estimates for these infections account for well over 80% of liver
cancer cases worldwide.30 Infection with HBV is the principal risk factor in most
high-risk countries, whereas HCV and alcohol consumption are more important risk
factors in low-risk countries. Finally, during the last two decades, increases of primary
liver cancer incidence rates have been reported from some low-risk regions, such as
Australia, Central Europe, the United Kingdom, Japan, and North America.30
BURDEN OF CHRONIC LIVER DISEASE IN THE UNITED STATES

Mortality from Cirrhosis
In 2002, chronic liver disease was the 12th leading cause of death in the United States,
accounting for 27,045 (1.1%) of the 2,447,862 deaths (mortality rate, 9.4/100,000).33
For the age group of 25–64 years, it was the seventh leading cause of death for all
races combined.33 These data may underestimate the true cirrhosis mortality in the
United States, because they included only death associated with the ICD-10 code
for chronic liver disease or cirrhosis (K70 and K73-K74 codes) as the underlying cause
of death. Thus, 5,706 (0.2%) deaths related to viral hepatitis (B15-B19 codes) and
14,046 (0.6%) deaths from malignant neoplasms of liver, the great majority of which
are HCC that arises in cirrhotic liver, were not included. If these additional diagnoses
were incorporated in the number of deaths attributable to chronic liver disease, the es-
timate would increase to 46,797 (1.9% of all deaths) for 2002, which would place
chronic liver diseases at the ninth leading cause of death in the United States.
Alcohol and HCV are thought to be the most important etiologies of chronic liver dis-
ease in the United States. Vong and colleagues34 analyzed patterns of deaths associ-
ated with chronic liver disease from 1990 through 1998, using an expanded definition
that included deaths from chronic liver disease, viral hepatitis, or other chronic liver
disease-related sequelae. Of 30,933 deaths considered to be from chronic liver dis-
ease in 1998, about 40% of deaths were alcohol-related; about 15% were HCV-re-
lated; and about 4% were HBV-related. In a large proportion (44%), no underlying
cause of liver disease was recorded (Fig. 3).34 Age-adjusted rates were higher among
males (47.6/100,000) than females (32.2/100,000).
According to Vong and colleagues, overall chronic liver disease mortality declined
5% from 1990 through 1994 (12.1 to 11.6/100,000) but remained unchanged from
1995 through 1998. These trends were similar for all causes except HCV, for which
rates increased 220% from 1993 to 1998 (0.57 to 1.67/100,000).34 Projections showed
that although the prevalence of HCV infection may be decreasing currently because of
the decline in incidence in the 1990s, the number of persons infected for 20 years has
been projected to increase substantially before reaching a peak in 2015.35 In a recent
analysis of the United States data about HCV mortality,36 the most dramatic increases
in HCV-related mortality rates between 1995 and 2004 occurred among persons
Fig. 3. Age-adjusted death rates of chronic liver disease by year and cause in the United
States, 1990–1998. (From Vong S, Bell BP. Chronic liver disease mortality in the United States,
1990–1998. Hepatology 2004;39(2):476–83. Reprinted with permission of Wiley-Liss, Inc.,
a subsidiary of John Wiley & Sons, Inc., Copyright ª 2004.)
45–54 years of age with the rate increasing 376% from 1.76 to 8.01 per 100,000, fol-
lowed by people 55–64 years of age, who had a 188% increase from 2.22 to 6.05 per
100,000 (Fig. 4). Although the last two years of observation suggested a small decline
in overall mortality, the increase continued among persons aged 55–64 years.
Fig. 4. Annual hepatitis C mortality rates and 95% confidence intervals for selected age groups,
United States, 1995–2004. (From Wise M, Bialek S, Finelli L, et al. Changing trends in hepatitis
C-related mortality in the United States, 1995–2004. Hepatology 2008;47(4):1128–35. Reprinted
with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc., Copyright ª 2008.)
740 Lim & Kim
Currently, the vast majority of mortality from HCV-related disease is occurring in peo-
ple under the age of 60 years, especially in men.36
Morbidity and Economic Burden of Chronic Liver Diseases

People who have chronic liver diseases, in particular, cirrhosis, have a poor prognosis.
Recent evidence suggests that cirrhosis reduces the estimated life expectancy by 38
years among 30 year-olds and by 20 years among 50 year-olds.37 In addition to the
decrease in longevity, chronic liver disease typically results in a significant reduction
in the quality of life of the patient. One study reported that patients with Child’s class
B and C cirrhosis showed reductions in all aspects of quality of life similar to patients
who survive a stroke.38 Such data confirm the substantial socioeconomic impact of
liver disease on patients and their families.39
In addition, cirrhosis and complications from portal hypertension are recognized as
significant causes of health care resource use. A report by the American Gastroentero-
logical Association (AGA) estimated the direct cost of chronic liver disease and cirrhosis
to be $1.4 billion in the United States, excluding patients with chronic hepatitis C.7,40 The
majority (90%) of these expenses were from hospital or inpatient-based medical care. In
addition, indirect costs were estimated to be $222 million in 1998, with the total costs
attributable to chronic liver disease and cirrhosis at $1.6 billion. These estimates are
likely substantially larger today because of the increased prevalence of chronic liver dis-
ease and health care cost inflation. Also, the study methodology employed in the AGA
report is thought to have led to underestimation of the true burden.7,40
Liver transplantation is by far the most expensive procedure applied in patients with
chronic liver disease. It has become a widely accepted therapy for improving survival
and quality of life for patients with either end-stage liver disease or early HCC, show-
ing a substantial gain in longevity and quality of life compared with conservative treat-
ment. The number of liver transplantations performed each year in the United States
has been steadily increasing (Fig. 5).7,41 The average cost of a liver transplantation is
well over $100,000 per year, which gives an overall expenditure of about $500 million
per year.7,41
Hepatocellular Carcinoma
In the United States, the age-adjusted incidence rates of HCC increased 2-fold be-
tween 1985 and 1998 (Fig. 6).42,43 The increase in HCC became recognizable in the
mid-1980s, while the greatest proportional increase occurred during the latter part
of the 1990s.44 The increase in HCC incidence has been attributed to the aging cohort
of individuals with chronic HCV infection, the incidence of which peaked in the 1980s.
Although the incidence of end-stage liver disease from HCV may level off in the next
few years, that of HCC may continue to rise,44 as the pool of individuals with long-
standing HCV increases over time.30
Simultaneous with the steady rise in mortality from HCC, hospital service use re-
lated to HCC increased substantially in the United States. In the authors’ study that
examined two nationally representative databases, the estimated total charges for
HCC hospitalizations nationwide increased from $241 million in 1988 to $509 million
in 2000 after inflation adjustment.45 This increase was attributable not only to the num-
ber of hospitalizations but also to the intensity of resource use per hospitalization. For
example, the average daily charges increased $2,140 (95% CI: 1966–2315) to $4,007
(95% CI: 3563–4453) after inflation adjustment, indicating more health care services
provided per hospitalization day.
Fig. 5. Number of deceased donor liver transplantations by age, 1997–2006. The number
receiving a deceased donor liver transplant has gradually increased, more steeply in
2004–2006. (From Freeman RB Jr, Steffick DE, Guidinger MK, et al. Liver and intestine trans-
plantation in the United States, 1997–2006. Am J Transplant 2008;8(4 Pt 2):958–76. Reprinted
with permission of Wiley-Blackwell Publishing, Copyright ª 2007.)
Fig. 6. Incidence of HCC in the United States. (Data from El-Serag HB. Hepatocellular carci-
noma: recent trends in the United States. Gastroenterology 2004;127(5 Suppl 1):S27–34.)
742 Lim & Kim
GROWING BURDEN OF NONALCOHOLIC FATTY LIVER DISEASE
Nonalcoholic fatty liver disease (NAFLD) is a clinicopathological condition that com-

prises a wide spectrum of liver damage, ranging from steatosis alone to steatohepa-
titis and to advanced fibrosis and cirrhosis, which occurs in patients who do not
consume significant amounts of alcohol.46 Simple hepatic steatosis is believed to
have a generally benign course, whereas nonalcoholic steatohepatitis (NASH) can
progress to cirrhosis, leading to liver failure and HCC.
The significance of NAFLD as a cause of serious liver disease is increasing because
of the rising prevalence of individuals at risk in the general population47–49 as well as its
potential to progress to end-stage liver disease and HCC.50,51 Because of their causal
association with the metabolic syndrome, NAFLD and NASH commonly occur in pa-
tients with obesity, diabetes, and hyperlipidemia. The likelihood of having NAFLD is
directly proportional to body weight.46 In the past 25 years, the prevalence of obesity
in North America has more than doubled, and it continues to rise without any evidence
of slowing down.52 In parallel, NAFLD is being increasingly recognized as a major
cause of liver-related morbidity and mortality.53 It currently is regarded the most com-
mon form of chronic liver disease in the United States and in many parts of the world.54
In unselected populations, the estimated prevalence of NAFLD ranges from 3% to
36.9%.54 In the United States, data from the Third National Health and Nutrition Exam-
ination Survey (NHANES III) showed that as many as 23% of Americans have unex-
plained elevations in liver enzymes. The majority of these persons are presumed to
have NAFLD.53,55 A recent study in the United States found that 34% of the population
had hepatic steatosis.53
In other parts of the world, the prevalence estimates of NAFLD in the general pop-
ulation ranges from 9% to 37%.54 Data from Japan suggest a greater-than-twofold in-
crease in the prevalence of NAFLD in the past 12 years.56,57 In a recent review, using
the ratio of NASH to NAFLD of 1:3 in liver biopsy studies, the prevalence of NAFLD was
estimated to be 17% in the general population and that of NASH was estimated to be
5.7%.54
In selected patient groups, the prevalence of NAFLD is much higher. Although there
is substantial ethnic variation, it is estimated that 90% of people with obesity (body
mass index [BMI] > 30 kg/m2) have some form of fatty liver, ranging from simple stea-
tosis to more severe forms of NASH, including cirrhosis.58 In patients who have abnor-
mal liver enzymes in the absence of serologic or biochemical markers of liver disease,
the prevalence of NASH ranges from 34% to 40%.58,59
NAFLD is increasingly recognized as a risk factor for HCC.50 The association be-
tween NASH and HCC has been examined in studies that correlated the risk of
HCC with diabetes and obesity, established risk factors for NASH.60,61 The incidence
of HCC was more than two-fold higher among patients with diabetes; independent of
age, gender, and race, diabetes was associated with a hazard rate of 2.16 (95% CI:
1.86 –2.52; P<.0001).31,43 Obesity is also recognized as a significant risk for the devel-
opment of liver cancer.62,63 Clearly, HCC is a part of the spectrum of NAFLD. In most
cases, HCC in NASH is thought to occur through the fibrosis-cirrhosis-HCC path-
way.64 In addition, diabetes is also associated with increased levels of insulin and in-
sulin-like growth factors, which are potential cancer promoting factors.65
SUMMARY
Hepatic fibrosis is an integral part in the progression of chronic liver disease, ultimately
leading to cirrhosis and hepatocellular carcinoma. Globally, alcohol consumption,
HBV and HCV have been the main causes of cirrhosis. More recently, the increasing
prevalence of obesity and the metabolic syndrome has resulted in increasing inci-
dence of cirrhosis secondary to NAFLD, especially in developed countries.
Chronic liver disease and cirrhosis are important causes of morbidity and mortality
in the world. Moreover, the burden of chronic liver disease is projected to increase,
due in part to the increasing prevalence of end-stage liver disease and HCC second-
ary to NAFLD and HCV.
The economic burden of chronic liver diseases, cirrhosis, and HCC is high. Preven-
tion of hepatic fibrosis and thus, cirrhosis may be the key to reducing health care costs
and overall disease burden. Globally, immunization against HBV and preventive mea-
sures against HCV infection, as well as identification and treatment of individuals with
chronic hepatitis should all contribute to this objective. In contrast, the burden of liver
disease associated with excessive alcohol consumption, obesity and metabolic syn-
drome both in the United States and globally is less amenable to preventive measures.
In the foreseeable future, it is likely that hepatic fibrosis will remain an important public
health concern. The limitations in preventive measures and lack of universally effective
and affordable treatment for these liver diseases highlight the importance of alterna-
tive therapeutic strategies, such as novel antifibrotic agents.
REFERENCES
1. Rockey DC, Friedman SL. Hepatic fibrosis and cirrhosis. In: Boyer TD, Wright TL,
Manns MP, editors. 5th edition. Zakim and Boyer’s hepatology, vol.1. New York:
Elsevier; 2006. p. 87–109.
2. Guha IH, Iredale JP. Clinical and diagnostic aspects of cirrhosis. In: Rodes J,
Benhamou JP, Blei AT, et al, editors. Textbook of hepatology, from basic science
to clinical practice. 3rd edition. Oxford: Blackwell Publishing Ltd.; 2007. p.
604–19.
3. Graudal N, Leth P, Marbjerg L, et al. Characteristics of cirrhosis undiagnosed dur-
ing life: a comparative analysis of 73 undiagnosed cases and 149 diagnosed
cases of cirrhosis, detected in 4929 consecutive autopsies. J Intern Med 1991;
230(2):165–71.
4. Melato M, Sasso F, Zanconati F. Liver cirrhosis and liver cancer. A study of their
relationship in 2563 autopsies. Zentralbl Pathol 1993;139(1):25–30.
5. Falagas ME, Vardakas KZ, Vergidis PI. Under-diagnosis of common chronic dis-
eases: prevalence and impact on human health. Int J Clin Pract 2007;61(9):
1569–79.
6. Friedman SL. Liver fibrosis—from bench to bedside. J Hepatol 2003;38(Suppl 1):
S38–53.
7. Kim WR, Brown RS Jr, Terrault NA, et al. Burden of liver disease in the United
States: summary of a workshop. Hepatology 2002;36(1):227–42.
8. Manos MM, Leyden WA, Murphy RC, et al. Limitations of conventionally derived
chronic liver disease mortality rates: Results of a comprehensive assessment.
Hepatology 2008;47(4):1150–7.
9. Fattovich G, Giustina G, Degos F, et al. Morbidity and mortality in compensated
cirrhosis type C: a retrospective follow-up study of 384 patients. Gastroenterology
1997;112(2):463–72.
10. Mathers C, Lopez A, Murray C. The burden of disease and mortality by condi-
tion: data, methods, and results for 2001. In: Lopez A, Mathers C, Ezzati M,
et al, editors. Global burden of disease and risk factors. Washington (DC):
Oxford University Press and the World Bank; 2006. p. 45–93.
744 Lim & Kim
11. Murray CJ, Lopez AD. Alternative projections of mortality and disability by
cause 1990–2020: Global Burden of Disease Study. Lancet 1997;349(9064):
1498–504.
12. Bosetti C, Levi F, Lucchini F, et al. Worldwide mortality from cirrhosis: an update to
2002. J Hepatol 2007;46(5):827–39.
13. Singh GK, Hoyert DL. Social epidemiology of chronic liver disease and cirrhosis
mortality in the United States, 1935–1997: trends and differentials by ethnicity, so-
cioeconomic status, and alcohol consumption. Hum Biol 2000;72(5):801–20.
14. Ramstedt M. Per capita alcohol consumption and liver cirrhosis mortality in 14 Eu-
ropean countries. Addiction 2001;96(Suppl 1):S19–33.
15. Kerr WC, Fillmore KM, Marvy P. Beverage-specific alcohol consumption and cir-
rhosis mortality in a group of English-speaking beer-drinking countries. Addiction
2000;95(3):339–46.
16. Williams JG, Roberts SE, Ali MF, et al. Gastroenterology services in the UK. The
burden of disease, and the organisation and delivery of services for gastrointes-
tinal and liver disorders: a review of the evidence. Gut 2007;56(Suppl 1):1–113.
17. Kane M. Global programme for control of hepatitis B infection. Vaccine 1995;
13(Suppl 1):S47–9.
18. Lee WM. Hepatitis B virus infection. N Engl J Med 1997;337(24):1733–45.
19. Lok AS, McMahon BJ, Chronic hepatitis B. Hepatology 2007;45(2):507–39.
20. Goldstein ST, Zhou F, Hadler SC, et al. A mathematical model to estimate global
hepatitis B disease burden and vaccination impact. Int J Epidemiol 2005;34(6):
1329–39.
21. Seeff LB, Hoofnagle JH. Epidemiology of hepatocellular carcinoma in areas of
low hepatitis B and hepatitis C endemicity. Oncogene 2006;25(27):3771–7.
22. McQuillan GM, Coleman PJ, Kruszon-Moran D, et al. Prevalence of hepatitis B vi-
rus infection in the United States: the National Health and Nutrition Examination
Surveys, 1976 through 1994. Am J Public Health 1999;89(1):14–8.
23. Wong VC, Ip HM, Reesink HW, et al. Prevention of the HBsAg carrier state in new-
born infants of mothers who are chronic carriers of HBsAg and HBeAg by admin-
istration of hepatitis-B vaccine and hepatitis-B immunoglobulin. Double-blind
randomised placebo-controlled study. Lancet 1984;1(8383):921–6.
24. CDC. Recommendations for protection against viral hepatitis. MMWR Morb Mor-
tal Wkly Rep 1985;34(22):313–24, 29–35.
25. Alter MJ. Epidemiology of hepatitis B in Europe and worldwide. J Hepatol 2003;
39(Suppl 1):S64–9.
26. CDC. Global progress toward universal childhood hepatitis B vaccination, 2003.
MMWR Morb Mortal Wkly Rep 2003;52(36):868–70.
27. Yim HJ, Lok AS. Natural history of chronic hepatitis B virus infection: what we
knew in 1981 and what we know in 2005. Hepatology 2006;43(2 Suppl. 1):
S173–81.
28. Fattovich G, Stroffolini T, Zagni I, et al. Hepatocellular carcinoma in cirrhosis: in-
cidence and risk factors. Gastroenterology 2004;127(5 Suppl. 1):S35–50.
29. Parkin DM, Bray F, Ferlay J, et al. Estimating the world cancer burden: Globocan
2000. Int J Cancer 2001;94(2):153–6.
30. Bosch FX, Ribes J, Diaz M, et al. Primary liver cancer: worldwide incidence and
trends. Gastroenterology 2004;127(5 Suppl. 1):S5–16.
31. El-Serag HB. Epidemiology of hepatocellular carcinoma. Clin Liver Dis 2001;5(1):
87–107.
32. Seeff LB. Introduction: the burden of hepatocellular carcinoma. Gastroenterology
2004;127(5 Suppl. 1):S1–4.
33. Kochanek KD, Smith BL. Deaths: preliminary data for 2002. Natl Vital Stat Rep
2004;52(13):1–47.
34. Vong S, Bell BP. Chronic liver disease mortality in the United States, 1990–1998.
Hepatology 2004;39(2):476–83.
35. Armstrong GL, Alter MJ, McQuillan GM, et al. The past incidence of hepatitis C
virus infection: implications for the future burden of chronic liver disease in the
United States. Hepatology 2000;31(3):777–82.
36. Wise M, Bialek S, Finelli L, et al. Changing trends in hepatitis C-related mortality in
the United States, 1995–2004. Hepatology 2008;47(4):1128–35.
37. Orlewska E. The cost-effectiveness of alternative therapeutic strategies for the
management of chronic hepatitis B in Poland. Value Health 2002;5(5):405–21.
38. Younossi ZM, Boparai N, McCormick M, et al. Assessment of utilities and health-
related quality of life in patients with chronic liver disease. Am J Gastroenterol
2001;96(2):579–83.
39. Burroughs A, McNamara D. Liver disease in Europe. Aliment Pharmacol Ther
2003;18(Suppl. 3):54–9.
40. AGA. The burden of gastrointestinal diseases. Bethesda: AGA; 2001. p. 41–60.
41. Freeman RB Jr, Steffick DE, Guidinger MK, et al. Liver and intestine transplanta-
tion in the United States, 1997–2006. Am J Transplant 2008;8(4 Pt 2):958–76.
42. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the
United States. N Engl J Med 1999;340(10):745–50.
43. El-Serag HB, Richardson PA, Everhart JE. The role of diabetes in hepatocellular
carcinoma: a case-control study among United States Veterans. Am J Gastroen-
terol 2001;96(8):2462–7.
44. El-Serag HB. Hepatocellular carcinoma: recent trends in the United States. Gas-
troenterology 2004;127(5 Suppl. 1):S27–34.
45. Kim WR, Gores GJ, Benson JT, et al. Mortality and hospital utilization for hepato-
cellular carcinoma in the United States. Gastroenterology 2005;129(2):486–93.
46. Sanyal AJ. AGA technical review on nonalcoholic fatty liver disease. Gastroenter-
ology 2002;123(5):1705–25.
47. Flegal KM, Carroll MD, Ogden CL, et al. Prevalence and trends in obesity among
US adults, 1999–2000. JAMA 2002;288(14):1723–7.
48. Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome among US
adults: findings from the third National Health and Nutrition Examination Survey.
JAMA 2002;287(3):356–9.
49. Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and
obesity-related health risk factors, 2001. JAMA 2003;289(1):76–9.
50. Bugianesi E. Non-alcoholic steatohepatitis and cancer. Clin Liver Dis 2007;11(1):
191–207.
51. Teli MR, James OF, Burt AD, et al. The natural history of nonalcoholic fatty liver:
a follow-up study. Hepatology 1995;22(6):1714–9.
52. Klein S. The national obesity crisis: a call for action. Gastroenterology 2004;
126(1):6.
53. Browning JD, Szczepaniak LS, Dobbins R, et al. Prevalence of hepatic steatosis
in an urban population in the United States: impact of ethnicity. Hepatology 2004;
40(6):1387–95.
54. Ong JP, Younossi ZM. Epidemiology and natural history of NAFLD and NASH.
Clin Liver Dis 2007;11(1):1–16.
55. Ruhl CE, Everhart JE. Determinants of the association of overweight with elevated
serum alanine aminotransferase activity in the United States. Gastroenterology
2003;124(1):71–9.
746 Lim & Kim
56. Kojima S, Watanabe N, Numata M, et al. Increase in the prevalence of fatty liver in
Japan over the past 12 years: analysis of clinical background. J Gastroenterol
2003;38(10):954–61.
57. Kuntz E, Kuntz H-D. Liver cirrhosis. Hepatology, Principles and Practice. Ger-
many: Springer; 2006. p. 716–49.
58. Neuschwander-Tetri BA, Caldwell SH. Nonalcoholic steatohepatitis: summary of
an AASLD Single Topic Conference. Hepatology 2003;37(5):1202–19.
59. Skelly MM, James PD, Ryder SD. Findings on liver biopsy to investigate abnormal
liver function tests in the absence of diagnostic serology. J Hepatol 2001;35(2):
195–9.
60. Bugianesi E, Leone N, Vanni E, et al. Expanding the natural history of nonalco-
holic steatohepatitis: from cryptogenic cirrhosis to hepatocellular carcinoma.
Gastroenterology 2002;123(1):134–40.
61. Marrero JA, Fontana RJ, Su GL, et al. NAFLD may be a common underlying liver
disease in patients with hepatocellular carcinoma in the United States. Hepatol-
ogy 2002;36(6):1349–54.
62. Calle EE, Rodriguez C, Walker-Thurmond K, et al. Overweight, obesity, and mor-
tality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med
2003;348(17):1625–38.
63. Ratziu V, Bonyhay L, Di Martino V, et al. Survival, liver failure, and hepatocellular
carcinoma in obesity-related cryptogenic cirrhosis. Hepatology 2002;35(6):
1485–93.
64. Caldwell SH, Crespo DM, Kang HS, et al. Obesity and hepatocellular carcinoma.
Gastroenterology 2004;127(5 Suppl. 1):S97–103.
65. Alexia C, Fallot G, Lasfer M, et al. An evaluation of the role of insulin-like growth
factors (IGF) and of type-I IGF receptor signalling in hepatocarcinogenesis and in
the resistance of hepatocarcinoma cells drug-induced apoptosis. Biochem Phar-
macol 2004;68(6):1003–15.
Genetic Determina nts
in Hepatic Fibrosis :
From Exp eriment al
Models to Fibro genic
Gene Signatures
in Huma ns
Susanne Weber, PhDa,*, Olav A. Gressner, MDb, Rabea Hall, MSa,
Frank Grˇnhage, MDa, Frank Lammert, MDa
KEYWORDS
Fibrogenesis Cirrhosis Quantitative trait locus
Inbred mouse strains
Hepatic fibrosis, or scarring of the liver, is a nonspecific reaction in response to chronic

liver injury. The damage to hepatocytes causes activation of inflammatory cells and an
abnormal deposition of extracellular matrix (ECM) in the space of Disse between he-
patocytes and sinusoidal endothelia.1–4 The ECM proteins comprise mainly collagen
but also glycoproteins, proteoglycans, and glycosaminoglycans, all of which are pro-
duced by activated hepatic stellate cells (HSCs or myofibroblasts) or portal fibroblasts.
HSC are vitamin A–storing cells that usually remain in a quiescent stage and are acti-
vated by the infiltration of inflammatory cells releasing profibrogenic mediators such
as transforming growth factor b1 (TGF-b1) and angiotensinogen. Activation of HSCs
leads to a myofibroblast-like phenotype characterized by the enhanced production
and secretion of ECM proteins.2,3,5
Hepatic fibrosis is commonly caused by exogenous factors such as viral hepatitis or
alcohol abuse, but recent studies also indicate a genetic predisposition.4 Although
some patients who have chronic liver diseases show only minor morphologic and
functional alterations of the liver and are characterized by slow progression of disease
with mild clinical symptoms, others develop pronounced hepatic fibrosis rapidly,
a
Department of Medicine II, Saarland University Hospital, Saarland University, Kirrberger St.,
66424 Homburg/Saar, Germany
b
Institute for Clinical Chemistry and Pathobiochemistry, University Hospital Aachen, Aachen
University (RWTH), Aachen, Germany
E-mail address: susanne.weber@uks.eu (S. Weber).
Clin Liver Dis 12 (2008) 747–757

748 Weber et al
culminating in cirrhosis, liver failure, or hepatocellular carcinoma, respectively. Hence,

a distinction between ‘‘rapid fibrosers’’ and ‘‘slow fibrosers’’ has been proposed.6
These well-known differences in progression of hepatic fibrosis persist when con-
trolling for age (at infection), gender, and exogenous factors in multivariate analysis,
indicating that genetic factors might play important roles in the modulation of hepatic
fibrosis and contribute to the variability in fibrosis progression.7,8 Furthermore, epide-
miologic studies indicate that the prevalence of cryptogenic cirrhosis among Hispanic
and African American patients is 3.1-fold higher and 3.9-fold lower, respectively, than
among European Americans (Fig. 1), despite similar prevalence rates of diabetes
mellitus.9 These ethnic differences support the hypothesis that fibrosis induced by
nonalcoholic fatty liver disease, a common cause of cryptogenic cirrhosis, is at least
partly genetically determined.
In contrast to these epidemiologic studies, twin studies are a better source for
dissecting the complex genetics because the interaction of genotypes and
phenotypes with age, gender, and lifestyle factors can be analyzed. Comparisons of
concordance rates between monozygotic and dizygotic pairs of twins can provide in-
formation on whether the familial pattern is caused by hereditary or environmental fac-
tors.10,11 However, few studies of liver diseases in twins have been reported. The large
twin study of the NAS-NRC Twin Registry (N 5 15924)12 showed that the concordance
rates for the development of alcohol-induced liver cirrhosis were significantly higher in
monozygotic (14.6%) compared with dizygotic twins (5.4%) (Fig. 2), but this was also
the case for alcohol abuse, and no further histologic analyses were provided. Maxi-
mum-likelihood modeling indicated that approximately 50% of the overall variance
was caused by additive genetic effects, but only a minor portion of the genetic vari-
ance of the individual complications was independent of the genetic predisposition
for alcoholism.12 Hence, further family-based studies are needed.
With respect to genetic determinants of human diseases, monogenic and polygenic
diseases can be distinguished. Monogenic diseases are caused by a single gene
defect following Mendelian rules of inheritance. These disorders are relatively rare,
Fig. 1. Ethnic differences in the prevalence of cryptogenic cirrhosis. Prevalence rates of

diabetes mellitus (DM, white bars) and cryptogenic cirrhosis (CC, black bars) in 41 patients
(African Americans, European Americans, and Hispanics). (Data from Browning JD, Kumar
KS, Saboorian MH, et al. Ethnic differences in the prevalence of cryptogenic cirrhosis. Am
J Gastroenterol 2004;99:292–8.)
Genetic Determinants in Hepatic Fibrosis 749
Fig. 2. Genetic predisposition to organ-specific endpoints of alcoholism. Medical records of

15,924 twin-pairs in the National Academy of Sciences-National Research Council (NAS-NRC)
twin registry aged 67 to 77 years were collected. Concordance rates for alcoholism, alcoholic
psychosis, and cirrhosis for monozygotic twins (black bars) and dizygotic twins (white bars).
(Data from Reed T, Page WF, Viken RJ, et al. Genetic predisposition to organ-specific end-
points of alcoholism. Alcohol Clin Exp Res 1996;20:1528–33.)
mostly severe, characterized by early onset, and usually not determined by environ-
mental factors. In contrast, polygenic diseases such as coronary heart disease, can-
cer, diabetes, and osteoporosis result from the interaction of multiple gene variants,
and environmental factors play a major role in their development.13 Accordingly, the
current paradigm predicts that a genetic predisposition for rapid fibrosis is mediated
by a combination of different variants in many genes.5 Hence, the goal is to identify
genes that enhance the susceptibility to hepatic fibrosis and represent molecular tar-
gets for novel antifibrotic therapies. Furthermore, predictive gene signatures might
help define the subgroup of patients who are at-risk for rapid fibrosis progression
and would benefit most from antifibrotic therapies.14,15
EXPERIMENTAL GENETICS
Mouse models are of great interest for the study of human disorders, including hepatic
fibrosis, because mice have similar physiology, anatomy, and metabolism to hu-
mans.16 These similarities are also reflected in the mouse genome. Almost every
gene in the human genome has a counterpart in the mouse. Researchers have allied
this to powerful genetic tools and developed a diverse set of inbred mouse strains with
variants that mirror those seen in human genetic diseases.17,18 The generation of
inbred mouse strains is a great achievement for the research of polygenic diseases.
Inbred mice are the result of at least 20 brother-sister matings and are therefore
homozygous for all alleles (ie, genetically identical).
Several hundred inbred strains currently exist, each with a different background.
The most common mouse strains that differ in their susceptibility toward fibrosis are
A/J, AKR, BALB/c, C3H/He, C57BL/6, DBA/2, and FVB/N.7,19,20 Hepatic fibrosis in
mice can be induced either surgically or chemically using different protocols.21 A sur-
gical method inducing fibrosis is bile duct ligation (BDL). Here, the common bile duct is
ligated and the mice display hepatic fibrosis approximately 2 to 4 weeks after surgery,
which leads to stimulation of portal myofibroblasts, ductular reaction,22 and pro-
nounced periportal and portoportal septal fibrosis.23,24 An alternative way to induce
hepatic fibrosis chemically is the administration of carbon tetrachloride (CCl4),7,19 thi-
oacetamide ethanol,25 or dimethynitrosamine.26 Low-dose CCl4 injection leads to
750 Weber et al
formation of CCl3 radicals, which damage hepatocytes27 and cause necrosis, inflam-
mation, and centrilobular fibrosis, which develops from activated HSC between the
surrounding parenchyma and results in portocentral septal fibrosis. Thioacetamide
in the drinking water induces fibrosis but also tumors (ie, cholangiocarcinoma),28
which must be considered when planning the experiments. Injection of
dimethynitrosamine also leads to tumor formation besides fibrosis,26 and therefore
CCl4 has become the most common method for fibrosis studies in mice.
GENOME-WIDE QUANTITATIVE TRAIT LOCI ANALYSIS
Analysis of polygenic diseases and identification of disease-related gene loci have

become possible using quantitative trait loci (QTL) analysis.29–31 This method com-
bines molecular techniques and classic genetic linkage analysis and has already led
to the identification of many QTL for polygenic diseases.13 Experimental crossing of
inbred mouse strains can be regarded as the central procedure in identifying the ge-
netic loci that determine complex diseases.
QTL are regions of DNA associated with a particular trait (eg, cancer or fibrosis). Al-
though not necessarily genes themselves, QTL are stretches of DNA that include at
least one gene that underlies the trait in question.32
Actual QTL analysis in mice comprises the following steps: first, two inbred mouse
strains that are preferably clearly distinguishable with respect to their susceptibility to-
ward is hepatic fibrosis, are mated to obtain many F2 progeny. Whereas the F1 mice
are heterozygous and uniform, a recombination of the parental chromosomes may oc-
cur during meiosis of the F1 genomes and appear genotypically and phenotypically
visible in the F2 progeny. All F2 mice are phenotyped according to quantitative trait
characteristics, such as concentration of hepatic collagen (according to photometric
measurement of the collagen-specific amino acid hydroxyproline), histologic fibrosis
stages (after Picrosirius red staining), or mRNA expression profiles. The F2 progeny
are genotyped for genetic markers that differ between the parental strains and that
densely cover the whole genome (eg, microsatellites or single nucleotide polymor-
phisms). Although unlinked genetic loci segregate independently from each other in
the F2 mice, markers that lie near disease-associated genes are inherited together
with the specific phenotype.33
Linkage analysis of fibrosis-specific phenotypes and marker genotypes is per-
formed using genetic mapping software such as MapManager QTX34 or R/qtl,35 which
allow user-friendly identification (eg, by regression analysis or interval mapping)36 and
statistical validation of linkage. The logarithm of the odds ratio (LOD) score, the tradi-
tional statistics for genetic linkage, is a measure of the strength of evidence for the
presence of a quantitative trait locus. The LOD score is the odds ratio comparing
the hypothesis of a quantitative trait locus at a particular location versus that of
none. LOD scores of 3 or higher are generally considered to indicate significant re-
sults, although theoretic thresholds have been calculated for specific cross-designs37
and alternatively, empiric thresholds, which correct for multiple testing, can be calcu-
lated using permutation tests.38
TRANSLATIONAL STUDIES FROM MICE TO HUMANS
Hillebrandt and colleagues7 performed a systematic inbred strain survey to identify

unknown susceptibility loci for hepatic fibrosis. For QTL analysis, F1 hybrids of sus-
ceptible BALB/c and resistant A/J inbred strains were intercrossed to obtain 358 F2
progeny. Linkage analysis of phenotypes and genotypes identified QTL on chromo-
somes 2 and 15 (Table 1) that significantly affect the histologic stage of fibrosis and
Table 1
Fibrogenic gene loci in polygenic experimental models
Mouse QTL Chromosome Candidate Gene(s)

Hfib1 15 A1bg, Mtss1
Hfib2 2 C5 (Hc)
Hfib3 1 —
5 Cxcl9
hepatic collagen contents.7,20 Using studies with congenic and transgenic mice, the
locus on chromosome 2 could be refined to the gene encoding complement factor
C5. Small molecule inhibitors of the C5a receptor displayed antifibrotic effects in
vivo, and common haplotype-tagging polymorphisms of the human gene C5 were as-
sociated with advanced fibrosis in a small cohort of patients who had chronic hepatitis
C virus infection. Thus, the mouse quantitative trait gene analysis led to the identifica-
tion of an unknown gene possibly underlying human susceptibility to hepatic
fibrosis.20
Another fibrosis susceptibility locus identified through QTL analysis is located on
chromosome 15. This locus, designated hepatic fibrogenic gene 1 (Hfib1), significantly
affected the stage of fibrosis and hepatic collagen contents in the F2 progeny of
BALB/c and A/J mice.7 In an intercross between the fibrosis-susceptible strain
BALB/cJ and another resistant strain FVB/NJ, a quantitative trait locus was detected
on mouse chromosome 1. According to standard nomenclature, this major quantita-
tive trait locus with significant effects on the progression of hepatic fibrosis (LOD 3.9)
was named Hfib3.39
Recently, functional variants of the chemokines CXCL9 and CCL5 (RANTES) were
shown to be linked to enhanced fibrosis progression in experimental crosses of
mice and knock-out mice and to be associated with the stage of fibrosis in indepen-
dent cohorts of patients, providing evidence for chemokines as genetic risk factors for
hepatic fibrosis.40,41
ASSOCIATION (IN SILICO) MAPPING IN MICE
An alternative way to identify genes or genomic regions linked to a specific phenotype

is association, or in silico mapping (Fig. 3).20,42,43 In contrast to QTL (linkage) analysis
in experimental crosses, the phenotypes of a set of inbred strains are correlated to
dense maps of single nucleotide polymorphism (SNP) genotypes. After entry of the
phenotypic information of the different inbred mouse strains, large mouse SNP data-
bases are scanned using different algorithms to predict chromosomal regions that de-
termine phenotypic traits.44 The trait under study and its genetic component, the
number of strains, the number of genetic markers, and the algorithms used are impor-
tant factors to consider.43
To illustrate the application of an integrated approach, Cervino and colleagues43
used hepatic fibrosis as an example. The study used the phenotypes determined by
Hillebrandt and colleagues7,20 and performed an in silico analysis using a 140 k
SNP map and a cladistic analysis of haplotypes (ie, unique combinations of SNP al-
leles located closely together on the same chromosome and tend to be inherited to-
gether).45 The two most likely susceptibility loci identified on chromosome 2 contained
nine candidate genes, as filtered with a median threefold increase for liver expression.
This list included the C5 gene identified by experimental QTL mapping, illustrating the
power of the integrated in silico approach. In addition, chromosome 15 showed
752 Weber et al
Fig. 3. An integrated strategy to narrowing down the genome to a list of candidate genes in
experimental models.43 The integrated workflow brings in information from linkage (QTL)
analysis, identical by descent (IBD) blocks, in silico analysis, and gene expression to produce
a short list of candidate genes.
a strong association pattern at around 68 Mb, harboring two creedal candidate genes
(a-1-B glycoprotein [A1bg] and metastasis suppressor 1 [Mtss1]) (see Table 1).43
HUMAN ASSOCIATION (CASE-CONTROL) STUDIES
In contrast to mouse models, which can be studied under defined environmental

conditions, identifying disease-related genes in humans is much more complicated.
This complexity is because of the large variety of environmental factors and the lack
of standardized noninvasive phenotyping of hepatic fibrosis in patients who have
chronic liver diseases. Genetic studies of hepatic fibrosis in humans are based on
associations between phenotypes and gene variants.46 In these case-control studies,
the genotype frequencies in patients who have severe fibrosis are compared with
patients who have normal liver tissue or mild fibrosis.
The most common variants investigated are SNPs. An SNP is a DNA sequence
variation, occurring in at least 1% of the human population. SNPs are generally
considered to be a form of point mutation that was evolutionarily successful enough
to recur in a significant proportion of humans. Once an SNP is observed at
a significantly higher frequency in patients compared with controls, it is considered
to be associated with the disease.
For efficient association studies, the choice of SNP is essential. Completion of the
International HapMap Project45,47 has made a lot of information readily available about
the occurrence and frequency of polymorphisms throughout the human genome. The
International HapMap Project determined the common patterns of DNA sequence
variation and has published a HapMap of many millions of sequence variants, includ-
ing their frequencies, linkage disequilibrium, and haplotypes in specific populations
with ancestry from Africa, Asia, and Europe.
Recent scholarly reviews48–50 summarized and evaluated all currently available
association studies (Table 2), which investigated the role of polymorphisms in several
Table 2
Replicated fibrosis risk genes in humans
Gene Mechanism Liver Disease

AGT Profibrogenic HBV, HCV, NAFLD
APOE Viral entry HCV, PBC
CCL2 (RANTES) Chemotaxis HCV
CCR5 Chemotaxis HCV
CD14 Endotoxin receptor ALD, NAFLD
CTLA4 Immune response ALD, HCV
GSTM1 Oxidative stress ALD
IL1B Proinflammatory ALD, HCV, PBC
IL1RN Proinflammatory ALD, HCV, PBC
IL10 Anti-inflammatory ALD, HCV
KRT8 Cytoskeleton HCV
MPO Immune response, oxidative stress HCV, HHC
MTP Lipid metabolism HCV, NAFLD
SERPINA1 (A1AT) Autophagic response ALD, CCI, HCV
SOD2 Oxidative stress HCV, NAFLD
TGFB1 Profibrogenic ALD, HCV, NAFLD
TNF Proinflammatory ALD, HCV, HHC, NAFLD, PBC
Abbreviations: ALD, alcoholic liver disease; CCI, cryptogenic cirrhosis; HBV, chronic hepatitis B virus
infection; HCV, chronic hepatitis C virus infection; HHC, hereditary hemochromatosis; NAFLD,
nonalcoholic liver disease; PBC, primary biliary cirrhosis.
Data from Refs.48–50
candidate genes on the progression of hepatic fibrosis or development of cirrhosis in

patients who had different types of chronic liver diseases. Particularly, functional SNPs
of genes coding for proinflammatory modulators or profibrogenic cytokines were in-
vestigated. However, many publications on the role of certain genetic variants in
Table 3
Fibrogenic gene signatures
Odds Ratio for Individual Gene

Gene Signature Variant (Range) Reference
AQP2 2.2 (1.4–3.3) Huang et al.15
AZIN1 3.2 (1.8–6.1)
TLR4 2.1 (1.2–2.8)
TRPM5 1.9 (1.4–3.1)
rs2290351 1.9 (1.2–2.8)
rs4290029 2.4 (1.5–3.6)
rs17740066 2.8 (1.5–5.2)
APOE 1.6 (0.8–3.3) Richardson et al.14
CCR5 2.1 (0.9–4.7)
CTLA4 2.3 (1.2–4.5)
HFE 1.8 (0.9–3.6)
MTP 2.7 (1.3–5.4)
SOD2 2.5 (1.2–4.5)
754 Weber et al
modulating the progression of hepatic fibrosis have been limited by small sample
sizes, low minor allele frequencies, and low statistical power.49 This finding is reflected
by a recent meta-analysis51 on hemochromatosis genotypes and risk for disease end
points in 6969 patients who had liver diseases and 41,017 controls. This meta-analysis
did not detect any association between heterozygous or compound mutations of the
hemochromatosis gene HFE and hepatitis C, hepatocellular carcinoma, or nonalco-
holic steatohepatitis, although several much smaller previous studies reported these
associations.
Only recently have genetic studies of hepatic fibrosis taken into account the poly-
genic nature of the disorder, which is not affected by a single major gene but combi-
nations of several genes, all contributing to the fibrosis phenotypes (Table 3). A study
from Australia14 detected associations between polymorphisms in six genes (APOE,
CCR5, CTLA4, HFE, MTP, SOD2) and progressive hepatic fibrosis in chronic hepatitis
C virus infection, with individual odds ratios ranging from 2.1 to 4.5.
Fig. 4. Study design for association study between genetic markers and fibrosis phenotypes
in humans and building of the cirrhosis risk score (CRS) signature. (From Huang H, Shiffman
ML, Friedman S, et al. A 7 gene signature identifies the risk of developing cirrhosis in
patients with chronic hepatitis C. Hepatology 2007;46:297–306; with permission.)
Using logistic regression analysis, a predictive equation could be developed and

tested using a second cohort of patients who have advanced fibrosis. The odds ratio
between rapidly progressing fibrosis and possession of at least 3, 4, or 5 at-risk alleles
increased from 9.1 and 15.5 to 24.1, respectively,14 consistent with an additive mode
of inheritance. Huang and colleagues15 used a ‘‘machine learning’’ approach to
develop and validate a gene signature consisting of markers most predictive for cir-
rhosis risk in Caucasian patients (Fig. 4). The area under the receiver operating char-
acteristic curves of the Cirrhosis Risk Score (see Table 3) were 0.73 to 0.75, and
a better predictor than clinical factors in differentiating high risk versus low risk for
cirrhosis,15 but prospective studies are needed to further validate these findings.
Genome-wide association studies in large, well-characterized cohorts with chronic
liver diseases are anticipated to allow the determination and validation of additional
gene signatures that predict rapid fibrosis progression and help to select patients
for personalized antifibrotic therapies.15,52,53
REFERENCES
1. Gressner AM. The cell biology of liver fibrogenesis—an imbalance of

proliferation, growth arrest and apoptosis of myofibroblasts. Cell Tissue Res
1998;292:447–52.
2. Friedman SL. Mechanisms of disease: mechanisms of hepatic fibrosis and
therapeutic implications. Nat Clin Pract Gastroenterol Hepatol 2004;1:
98–105.
3. Gressner AM, Weiskirchen R. Modern pathogenetic concepts of liver fibrosis sug-
gest stellate cells and TGF-beta as major players and therapeutic targets. J Cell
Mol Med 2006;10:76–99.
4. Schupan D, Afdhal NH. Liver cirrhosis. Lancet 2008;371:838–51.
5. Bataller R, Brenner DA. Liver fibrosis. J Clin Invest 2005;115:209–18.
6. Poynard T, Bedossa P, Opolon P. Natural history of liver fibrosis progression in
patients with chronic hepatitis C. Lancet 1997;349:825–32.
7. Hillebrandt S, Goos C, Matern S, et al. Genome-wide analysis of hepatic fibrosis
in inbred mice identifies the susceptibility locus Hfib1 on chromosome 15. Gastro-
enterology 2002;123:2041–51.
8. Lammert F. Genetic determinants of complex liver diseases: mouse models and
quantitative trait locus analysis. In: Rodes J, Benhamou JP, Blei A, et al, editors.
Textbook of hepatology: from basic science to clinical practice. Oxford (UK):
Blackwell Publishing; 2007. p. 371–83.
9. Browning JD, Kumar KS, Saboorian MH, et al. Ethnic differences in the preva-
lence of cryptogenic cirrhosis. Am J Gastroenterol 2004;99:292–8.
10. Neale MC, Cardon LR. Methodology for genetic studies of twins and families
Dordrecht: Kluwer Academic Publishers; 1992.
11. Perola M, Sammalisto S, Hiekkalinna T, et al. Combined genome scans for body
stature in 6,602 European twins: evidence for common Caucasian loci. PLoS
Genet 2007;3:e97.
12. Reed T, Page WF, Viken RJ, et al. Genetic predisposition to organ-specific end-
points of alcoholism. Alcohol Clin Exp Res 1996;20:1528–33.
13. Korstanje R, Paigen B. From QTL to gene: the harvest begins. Nat Genet 2002;
31:235–6.
14. Richardson MM, Powell EE, Barrie HD, et al. A combination of genetic polymor-
phisms increases the risk of progressive disease in chronic hepatitis C. J Med
Genet 2005;42:e45.
756 Weber et al
15. Huang H, Shiffman ML, Friedman S, et al. A 7 gene signature identifies the risk of
developing cirrhosis in patients with chronic hepatitis C. Hepatology 2007;46:
297–306.
16. Paigen K. A miracle enough: the power of mice. Nat Med 1995;1:215–20.
17. Mouse Genome Sequencing Consortium. Initial sequencing and comparative
analysis of the mouse genome. Nature 2002;420:520–62.
18. Bogue M. Mouse phenome project: understanding human biology through
mouse genetics and genomics. J Appl Physiol 2003;95:1335–7.
19. Shi Z, Wakil AE, Rockey DC. Strain-specific differences in mouse hepatic wound
healing are mediated by divergent T helper cytokine responses. Proc Natl Acad
Sci USA 1997;94:10663–8.
20. Hillebrandt S, Wasmuth HE, Weiskirchen R, et al. Complement factor 5 is a quan-
titative trait gene that modifies liver fibrogenesis in mice and humans. Nat Genet
2005;37:835–43.
21. Iredale JP. Models of liver fibrosis: exploring the dynamic nature of inflammation
and repair in a solid organ. J Clin Invest 2007;117:539–48.
22. Roskams TA, Theise ND, Balabaud C, et al. Nomenclature of the finer branches of
the biliary tree: canals, ductules, and ductular reactions in human livers. Hepatol-
ogy 2004;39:1739–45.
23. Accatino L, Contreras A, Fernandez S, et al. The effect of complete biliary
obstruction on bile flow and bile acid excretion: postcholestatic choleresis in
the rat. J Lab Clin Med 1979;93:706–17.
24. Chang ML, Yeh CT, Chang PY, et al. Comparison of murine cirrhosis models
induced by hepatotoxin administration and common bile duct ligation. World
J Gastroenterol 2005;11:4167–72.
25. Kornek M, Raskopf E, Guetgemann I, et al. Combination of systemic thioaceta-
mide (TAA) injections and ethanol feeding accelerates hepatic fibrosis in C3H/
He mice and is associated with intrahepatic up regulation of MMP-2, VEGF
and ICAM-1. J Hepatol 2006;45:370–6.
26. Weiler-Normann C, Herkel J, Lohse AW. Mouse models of liver fibrosis. Z Gastro-
enterol 2007;45:43–50.
27. Muriel P, Escobar Y. Kupffer cells are responsible for liver cirrhosis induced by
carbon tetrachloride. J Appl Toxicol 2003;23:103–8.
28. Yeh CN, Maitra A, Lee KF, et al. Thioacetamide-induced intestinal-type cholangio-
carcinoma in rat: an animal model recapitulating the multi-stage progression of
human cholangiocarcinoma. Carcinogenesis 2004;25:631–6.
29. Darvasi A. Experimental strategies for the genetic dissection of complex traits in
animal models. Nat Genet 1998;18:19–24.
30. Flint J, Valdar W, Shifman S, et al. Strategies for mapping and cloning quantitative
trait genes in rodents. Nat Rev Genet 2005;6:271–86.
31. Peters LL, Robledo RF, Bult CJ, et al. The mouse as a model for human biology:
a resource guide for complex trait analysis. Nat Rev Genet 2007;8:58–69.
32. The Complex Trait Consortium. The nature and identification of quantitative trait
loci: a community’s view. Nat Rev Genet 2003;4:911–6.
33. Hillebrandt S, Matern S, Lammert F. Mouse models for genetic dissection of poly-
genic gastrointestinal diseases. Eur J Clin Invest 2003;33:155–60.
34. Manly KF, Cudmore RH Jr, Meer JM. Map Manager QTX, cross-platform software
for genetic mapping. Mamm Genome 2001;12:930–2.
35. Broman KW. Mapping expression in randomized rodent genomes. Nat Genet
2005;37:209–10.
36. Lander ES, Botstein D. Mapping Mendelian factors underlying quantitative traits
using RFLP linkage maps. Genetics 1989;121:185–99.
37. Lander E, Kruglyak L. Genetic dissection of complex traits: guidelines for inter-
preting and reporting linkage results. Nat Genet 1995;11:241–7.
38. Churchill GA, Doerge RW. Empirical threshold values for quantitative trait map-
ping. Genetics 1994;138:963–71.
39. Hillebrandt S, Sauerbruch T, Lammert F. Identification of multiple interacting
quantitative trait loci (QTL) for hepatic fibrosis in mice. Gastroenterology 2006;
130:A427.
40. Wasmuth HE, Wiederholt T, Dahl E, et al. The chemokine CXCL9 (MIG) is
a genetic modifier of liver fibrogenesis in chronic hepatitis C. Hepatology 2006;
44:568A.
41. Berres ML, Rüland A, Moreno Zaldivar M, et al. The CC chemokine RANTES
(CCL5) mediates liver fibrosis in mice and humans. Z Gastroenterol 2008;46.
42. Pletcher MT, McClurg P, Batalov S, et al. Use of a dense single nucleotide poly-
morphism map for in silico mapping in the mouse. PLoS Biol 2004;2:e393.
43. Cervino AC, Darvasi A, Fallahi M, et al. An integrated in silico gene mapping
strategy in inbred mice. Genetics 2007;175:321–33.
44. Grupe A, Germer S, Usuka J, et al. In silico mapping of complex disease-related
traits in mice. Science 2001;292:1915–8.
45. Wasmuth HE, Matern S, Lammert F. From genotypes to haplotypes in hepatobili-
ary diseases: one plus one equals (sometimes) more than two. Hepatology 2004;
39:604–7.
46. Day CP. Genetic studies to identify hepatic fibrosis genes and SNPs in human
populations. Methods Mol Med 2005;117:315–31.
47. The International HapMap Consortium. The International HapMap Project. Nature
2003;426:789–96.
48. Wasmuth HE, Lammert F, Matern S. Genetische Risikofaktoren der Fibrogenese
bei chronischen Lebererkrankungen. Med Klin 2003;98:754–62.
49. Österreicher CH, Stickel F, Brenner DA. Genomics of liver fibrosis and cirrhosis.
Semin Liver Dis 2007;27:28–43.
50. Gressner OA, Weiskirchen R, Gressner AM. Biomarkers of hepatic fibrosis, fibro-
genesis and genetic pre-disposition pending between fiction and reality. J Cell
Mol Med 2007;11:1031–51.
51. Ellervik C, Birgens H, Tybjaerg-Hansen A, et al. Hemochromatosis genotypes and
risk of 31 disease endpoints: meta-analyses including 66,000 cases and 226,000
controls. Hepatology 2007;46:1071–80.
52. Hellerbrand C, Wasmuth HE. Genomewide genetic association studies in hepa-
tology: the end of searching for the needle in the haystack? Hepatology 2007;
46:1661–3.
53. Kruglyak L. The road to genome-wide association studies. Nat Rev Genet 2008;9:
314–8.
Cellular Sources of
Ex tracellular Matrix
in Hepatic Fibrosis
Rebecca G.Wells, MDa,b,*
KEYWORDS
Collagen Hepatic stellate cells Portal fibroblasts
Sinusoidal endothelial cells Liver fibrosis
Extracellular matrix Epithelial to mesenchymal transition
The extracellular matrix (ECM) is a complex and dynamic component of the liver that
has multiple functions. ECM proteins have architectural and mechanical roles, provid-
ing tensile strength and resilience, modulating diffusion and vascular flow, and regu-
lating cell movement. The mechanics of the ECM are increasingly recognized as
key determinants of normal and pathologic cell behavior.1 ECM proteins additionally
regulate signaling, serving as ligands, storage depots, and receptors.
The interaction between the cells and matrix of the liver is bidirectional. Most cells
produce matrix and respond phenotypically to that matrix. Defining the cellular sour-
ces of ECM in the normal and diseased liver is thus critical to understanding the
pathophysiology of liver fibrosis.
ECM quantity in the fibrotic liver can be up to eightfold higher than in the normal
liver. Perhaps even more dramatic are changes in the quality of the matrix expressed.
There are significant increases in the ECM proteins that produce the fibrotic scar, in-
cluding the fibrillar collagens, which provide rigidity, and fibronectin splice variants
and proteoglycans.2 Basement membrane proteins are similarly increased, reflecting
the capillarization of the sinusoids. Less well understood are changes in matrix protein
modifications, including glycosylation, intermolecular cross-linking, and glycosamino-
glycan side chain sulfation.3
Understanding these changes in ECM proteins and the cells that synthesize them is
critical to understanding fibrosis and, ultimately, identifying new therapies. Although
increasing evidence indicates significant limitations in the final common pathway
This work was supported by grant #DK58123 from the National Institutes of Health.
a
Department of Medicine (Gastroenterology), University of Pennsylvania School of Medicine,
600 CRB/6140, 415 Curie Boulevard, Philadelphia, PA 19104-6140, USA
b
Department of Pathology and Laboratory Medicine, University of Pennsylvania School of
Medicine, 600 CRB/6140, 415 Curie Boulevard, Philadelphia, PA 19104-6140, USA
* Department of Medicine (Gastroenterology), University of Pennsylvania School of Medicine,
600 CRB/6140, 415 Curie Boulevard, Philadelphia, PA 19104-6140.
E-mail address: rgwells@mail.med.upenn.edu
Clin Liver Dis 12 (2008) 759–768

760 Wells
model of fibrosis, knowledge of cell- and disease-specific matrix synthesis is rudimen-

tary. This article summarizes the current literature on the cellular source of ECM pro-
teins in fibrosis, highlighting recent advances that broaden the concept of the
fibrogenic cell and outlining areas of ongoing and future research.
STELLATE CELLS AND OTHER MYOFIBROBLAST PRECURSORS
Myofibroblasts are the workhorses of fibrosis. Wound healing in general and patho-
logic tissue fibrosis in particular result primarily from myofibroblast-mediated ECM
synthesis and deposition, and the state of the myofibroblast population in a diseased
tissue reflects the direction of fibrosis. Myofibroblasts are contractile and secretory
cells that express de novo the microfilament protein a-smooth muscle actin
(a-SMA), which is used in practice as a myofibroblast marker.4
Three conditions are required for myofibroblastic differentiation: high levels of the
growth factor transforming growth factor b (TGF-b), the presence of the fibronectin
splice variant EDA, and increased local mechanical tension.5 These factors act on var-
ious potential myofibroblast precursors in the body, including fibroblasts and smooth
muscle cells. An appreciation for the heterogeneity of the myofibroblast precursor
population in the fibrotic liver is one of the most dramatic results of research in the
past decade.
Hepatic Stellate Cells

Hepatic stellate cells (HSCs), the stellate-shaped, vitamin A–storing cells located in the
space of Disse, are without question the most studied fibrogenic population in the
liver. HSCs have important functions in liver development, metabolism, the immune
response, and angiogenesis, but are best known for their key role in fibrosis.6 Several
seminal papers established that HSC-derived myofibroblasts produce most of the
abnormal ECM, including fibrillar collagens and basement membrane proteins, in
the fibrotic liver, particularly in nonbiliary disease.7–9 Although many studies over
the past decade failed to use appropriately HSC-specific markers, the overwhelming
consensus of the extensive HSC literature, summarized in several excellent recent
reviews, is that HSCs are highly fibrogenic participants in liver fibrosis.10–12
HSCs are easily isolated from normal liver and undergo spontaneous myofibroblas-
tic differentiation when cultured under standard conditions. This model system has
resulted in the detailed characterization of the ECM deposited by HSC myofibroblasts,
although HSC populations are recognized to be heterogeneous. HSCs produce most
major and minor matrix proteins of the fibrotic liver, including fibrillar and nonfibrillar
collagens, components of the organized sinusoidal basement membrane (collagen
IV, laminin, and perlecan), cellular fibronectin, tenascin-C, the elastic fibril component
fibrillin, and many of the small proteoglycans.2,10–12
Portal Fibroblasts and Other Resident Liver Fibroblast Populations

A notable controversy in the hepatology literature over the past decade has been
whether non–HSC-derived myofibroblasts contribute to fibrosis. Using marker analy-
sis, several authors have now convincingly and elegantly shown in rodent model
systems that there are fibrogenic, a-SMA–positive myofibroblast populations
distinct from HSCs.13,14
These cells, many of which are derived from portal fibroblasts, seem to be particu-
larly abundant in biliary fibrosis.14–17 In one important recent study, the livers of rats
with biliary fibrosis of two different origins (bile duct ligation and arterial ischemia)
were immunostained for a-SMA as a marker for myofibroblasts, and desmin as
Cellular Sources of Extracellular Matrix 761
a marker for HSCs. Most myofibroblasts in fibrotic regions were found to be desmin-
negative and derived from portal mesenchymal cells rather than HSCs.14
The matrix expression profile of portal fibroblasts and myofibroblasts has not been
investigated, with the exception of one study that showed up-regulation of the a1
subunits of collagens I, III, and IV when primary portal fibroblasts underwent myofibro-
blastic differentiation in vitro.18 Portal myofibroblasts, however, are likely to secrete
a broad variety of ECM proteins. Many studies examining HSC matrix expression
in vivo used a-SMA as a marker of fibrogenic cells, and thus the conclusions are
applicable to portal myofibroblasts as well as HSCs.
Resident non–portal fibroblast populations in the liver that may activate to myofibro-
blasts and contribute to fibrogenesis include the fibroblasts of Glasson’s capsule,
smooth muscle cells of the vasculature, and so-called ‘‘second layer cells’’ around
central veins.19 In some forms of fibrosis, a significant population of a-SMA–negative
activated fibroblasts are also present.20,21 The matrix-secreting profile of these cells
has not been studied, and their relative contributions to liver diseases of different
origins is not known.
Bone Marrow–Derived Myofibroblast Precursors

Although fibrogenic cells in the liver have been assumed to arise from resident popu-
lations, recent evidence suggests that circulating cells derived from the bone marrow
contribute, possibly significantly, to fibrogenesis. One group showed the presence of
green fluorescent protein (GFP)–positive HSCs in the livers of mice that had received
bone marrow transplants with GFP-expressing cells. These HSCs could undergo
myofibroblastic differentiation in culture, and seemed to deposit collagen.22
Human liver specimens from patients who underwent gender-mismatched bone
marrow or liver transplantations also seemed to have a significant population of mar-
row-derived myofibroblasts.23 An elegant study using mice that had undergone bone
marrow transplantations followed by CCl4- or thioacetamide-induced liver injury
showed that close to 70% of HSCs and myofibroblasts in the fibrotic liver were derived
from the bone marrow, and that these cells produced type I collagen.24
Other groups, however, have been unable to identify HSCs derived from the
bone marrow and instead have observed a population of fibrocytes (bone marrow–
derived circulating mesenchymal cell precursors) that seem to undergo myofibroblastic
differentiation and deposit collagen after lodging in the liver.25 Therefore, the role of
circulating cells in the fibrotic response to damage is yet to be determined definitively.
THE FIBROGENIC HEPATOCYTE CONTROVERSY
Hepatocytes are the major cell population in the liver and a significant potential source
of ECM proteins. The role of hepatocytes in fibrogenesis has been the source of in-
tense debate in the literature. Although the controversy was initially resolved in favor
of myofibroblasts as the primary matrix-producing cells of the liver, recent reports
suggest that hepatocytes can undergo an epithelial-to-mesenchymal transition
(EMT) and may contribute to fibrogenesis in the form of fibroblasts and myofibroblasts.
Throughout the 1980s and early 1990s, various groups arrived at different conclu-
sions about the role of hepatocytes in the deposition of fibrillar collagens (collagen
I and III) and basement membrane proteins (collagen IV, laminin, and perlecan) in fibro-
sis.8,26 Definitive identification of the cells responsible for depositing specific ECM
proteins proved difficult using tissue sections because secreted matrix proteins can
adhere to adjacent cells, whereas cells in culture have a propensity to adopt new
and potentially nonphysiologic phenotypes.
762 Wells
Several studies using different techniques to overcome these limitations were partic-
ularly influential in showing ultimately that mature hepatocytes have little if any role in
the synthesis of fibrillar collagens and basement membrane proteins. In situ hybridiza-
tion, alone or in combination with immunostaining to mark specific cell populations,
yielded no evidence that hepatocytes, as opposed to nonparenchymal cells, expressed
collagen I, III, or IV in CCl4-induced fibrosis in the rat or in human livers affected by vari-
able degrees of fibrosis from various causes (eg, alcoholic, biliary, and viral).9,26,27 Sim-
ilarly, mRNA expression analyses of freshly isolated, noncultured hepatocytes,
sinusoidal endothelial cells, and HSCs showed minimal synthesis of laminin or collagens
I, III, or IV by hepatocytes from either normal, bile duct–ligated, or CCl4-treated livers.8
New data have again raised the specter of the fibrogenic hepatocyte. This work is
based on studies from the renal fibrosis field showing that tubular epithelial cells
undergo EMT, and that these cells comprise a large percentage of fibrogenic cells
in the diseased kidney. Initial studies on fibrotic kidneys showed that renal tubular
epithelial cells expressed fibroblast-specific protein-1 (FSP1, also known as S100
A4), believed to be both a mediator and a marker of EMT.28 Tubular epithelial cells
also expressed HSP47, a collagen-specific chaperone indicative of ongoing collagen
synthesis, and the myofibroblast marker a-SMA.28,29 Lineage tracing studies, in which
transgenic reporter mice were used to identify all epithelial cell descendents, provided
definitive evidence in support of the contribution of EMT to kidney fibrosis. In one
model, 36% of FSP1-positive fibroblasts in the fibrotic kidney were derived from
epithelial cells.30
Neonatal and adult hepatocytes can undergo EMT in culture, losing epithelial char-
acteristics and becoming a-SMA–expressing myofibroblasts.31–35 Hepatocyte EMT in
vivo is less well established. Human explant livers from patients who have various
biliary and nonbiliary diseases exhibited no colocalization between hepatocyte
markers and the EMT markers FSP1, HSP47, and a-SMA.36
Lineage tracing studies using transgenic mice in which GFP was expressed under
the alpha-fetoprotein promoter failed to show evidence of a-SMA expression in la-
beled cells 4 weeks after bile duct ligation (R. Diaz, A. Chu, R. G. Wells, unpublished
results, 2008). A recently published lineage tracing study using transgenic mice ex-
pressing b-galactosidase under the control of the albumin promoter showed, how-
ever, that hepatocytes undergo EMT in CCl4-induced fibrosis.37 The authors of this
work suggested that a significant population of hepatocyte-derived, FSP1–positive,
a-SMA–negative fibroblasts is present in the fibrotic liver. Further work is required
to determine the magnitude and timing of the contribution of these cells to ECM syn-
thesis in fibrosis.
Hepatocytes are the major source of plasma fibronectin, one of the two major prod-
ucts (along with cellular fibronectin) of the fibronectin gene.38 In the normal liver, plasma
fibronectin is the most abundant matrix protein in the spaces of Disse, coating hepato-
cyte membranes and collagen fibrils.39,40 In the setting of fibrosis, fibronectin increases
dramatically and is one of the first matrix proteins to do so; however, almost all of this
increase is in cellular fibronectin.40 Although cellular fibronectin is thought to play an
important role in myofibroblastic differentiation in vitro, the function of plasma fibronec-
tin in either the normal liver or the context of the altered milieu of the injured liver is not
well understood.
BILIARY EPITHELIAL CELLS
Defining the role of biliary epithelial cells (BECs) in fibrosis and, more specifically, their
direct contributions to fibrogenesis is an area of intense and evolving research.
Although the major contribution of BEC to matrix synthesis (in the normal and dis-
eased liver) was once believed to be limited to the production of basement membrane,
new research suggests that BECs may play a direct role in synthesis of the fibrotic
scar, particularly in conditions involving a ductular reaction.
BECs, in contrast to hepatocytes, rest on a fully organized basement membrane. In
vivo and in vitro data indicate that this basement membrane is synthesized by portal
mesenchymal cells on one side and BECs on the other.26,41 In culture, human BECs
isolated from normal cystic ducts showed intense immunoreactivity for two major
components of the basement membrane: collagen IV and laminin.42
BECs in normal and fibrotic rat and human liver tissue (from patients who had var-
ious liver diseases, including from viral, alcoholic, and biliary causes) expressed
mRNA for the B1 chain of laminin, the a1 chain of collagen IV, and the basement mem-
brane proteoglycan perlecan.26,43 BECs, along with hepatocytes, also synthesize col-
lagen XVIII, a newly appreciated collagenous component of the basement membrane
that is a precursor of the angiogenesis inhibitor endostatin.44 BEC synthesis of base-
ment membrane components was observed in normal and fibrotic livers, with slight
increases seen in fibrosis.26,41
Studies have consistently shown that BECs do not directly synthesize significant
amounts of the fibrillar collagens or fibronectin.26,41 In the past several years, however,
data on the role of EMT in renal fibrosis (see earlier section on ‘‘The Fibrogenic Hepa-
tocyte Controversy’’) have raised the possibility that EMT also contributes to liver
fibrosis. It has been known for at least 20 years that, in the setting of biliary injury,
newly formed BECs (especially cells within the ductular reaction) express the interme-
diate filament vimentin and it has been postulated that this reflects cellular reorganiza-
tion of BECs.45 Recent data suggest that this reorganization reflects EMT, with BECs
in the damaged liver undergoing a process analogous to that of renal tubular epithelial
cells in the damaged kidney, as discussed below. Preliminary experiments with BECs
in vitro and in vivo are suggestive of EMT, although definitive lineage tracing studies
have not been reported.
EMT is driven largely by TGF-b, and multiple studies have now shown that
TGF-b–treated BECs in culture undergo EMT and exhibit changes, including the
acquisition of a myofibroblast-like morphology and de novo expression of a-SMA
and collagen I.46,47 As is the case for hepatocytes, in vivo studies are less definitive.
In the mouse, CK19-positive BECs 12 weeks after bile duct ligation showed synthe-
sis of collagen I and morphologic changes and basement membrane disruption
consistent with EMT.47 Similarly, in human liver tissue taken from patients with var-
ious different diseases, BECs were shown to coexpress epithelial cytokeratin
markers and FSP-1, HSP47, and vimentin, with associated cytoplasmic redistribu-
tion of E-cadherin and nuclear localization of one or both of the TGF-b downstream
signaling molecules Smad2 and Smad3.36,46,48 The colocalization was particularly
marked in small ducts and cells of the ductular reaction, and in diseases such as
primary biliary cirrhosis and biliary atresia in which the ductular reaction is
prominent.36,46,48
Few cells in any of these in vivo analyses coexpressed epithelial markers and
a-SMA, and evidence for direct fibrogenesis is largely circumstantial (eg, expression
of the collagen chaperone HSP47).49 Whether expression of all mesenchymal
markers is required to show the presence of fully fibrogenic cells is unclear.
Thus, the data so far are consistent with mesenchymal changes but not yet full
EMT. Lineage tracing studies are required before concluding definitively that biliary
EMT occurs and to determine the relative contribution of EMT to early and late stages
of fibrosis.
764 Wells
SINUSOIDAL ENDOTHELIAL CELLS
Sinusoidal endothelial cells (SECs) line the sinusoids and are in constant contact with
sinusoidal blood flow and closely associated with hepatocytes, HSC, and Kupffer
cells. Changes in SECs can be detected significantly before fibrosis is visible using
light microscopy,36,50,51 leading some authors to suggest that SECs might drive or
even initiate fibrosis.43,51,52 The study of SECs in fibrosis has been hindered, however,
by controversies regarding their isolation, culture, and identification,53 and correlating
in vivo and in vitro findings has been particularly difficult. Additionally, SECs have
a high endocytic capacity and are located adjacent to highly fibrogenic HSCs, making
the interpretation of immunostains prone to error.
Whether SECs synthesize fibrillar collagens and other components of the fibrotic
scar is controversial. Early studies showed that isolated SECs expressed mRNA for
collagens I and III, and that SECs isolated from fibrotic liver had higher type I collagen
expression than cells from normal liver.8 More recent studies have questioned the
contribution of SECs to the production of the fibrillar collagens,26 and the literature
as a whole suggests that in most cases, HSCs and other myofibroblasts, not SECs,
produce the bulk of the fibrotic ECM.
SECs are likely, however, to play two important roles in fibrosis, particularly in the
early stages before HSCs undergo myofibroblastic differentiation. First, SECs have
an important role in capillarization of the sinusoids. This process, characterized by
the loss of typical SEC fenestrations and the formation of an organized basement
membrane in the space of Disse, has been recognized as one of the hallmarks of liver
fibrosis since it was first described in 1963.54
The literature is unclear on the relative contributions of HSCs and SECs to this pro-
cess, but various techniques have shown that SECs secrete significant amounts of the
components required for an organized basement membrane, including collagen IV,
laminin, entactin, and perlecan.8,26,43,55–60 SECs seem to be the major perisinusoidal
source of perlecan, a large and complex heparan sulfate proteoglycan that binds other
matrix components, including fibronectin.43,61 The formation of an organized base-
ment membrane with different organization and type IV collagen composition results
in hepatocyte dedifferentiation and dysfunction and therefore may facilitate hepatic
injury and the perpetuation of fibrosis.50,62
The second important role attributed to SECs is the production of fibronectin extra
domain A (EDA), a splice variant of cellular fibronectin expressed primarily during
development and in response to injury. In vitro studies suggest that fibronectin
EDA, acting in concert with the potent profibrogenic mediator TGF-b, is necessary
for the general process of fibroblast-to-myofibroblast differentiation.5,63 Fibronectin
EDA is produced early in rodent models of fibrosis, preceding collagen deposition,
and persists at high levels in advanced fibrosis.40,64 Fibronectin EDA induces the
myofibroblastic differentiation of HSC in culture, and studies have elegantly shown
that TGF-b acts on SECs to rapidly up-regulate production of fibronectin EDA, thus
linking TGF-b, SEC, and HSC activation.64,65 Treatment of SECs using albumin
modified with malondialdehyde-acetaldehyde (an ethanol byproduct) resulted in in-
creased fibronectin EDA expression, suggesting multiple potential inducers of
SECs.66 Although HSCs themselves, and potentially hepatocytes also, produce fi-
bronectin EDA, SECs are the first responders and therefore may play a crucial
role in the early stages of fibrosis.67–69 Why fibronectin EDA, unlike plasma fibronec-
tin or other forms of cellular fibronectin, is able to induce myofibroblast activation is
unknown.
SUMMARY
The cellular basis of liver fibrosis appeared a decade ago to be a closed question.
HSCs had been shown to be the source of fibrillar collagens and basement membrane
proteins, and the fibrosis field moved on to study HSC regulation. Today, HSCs are still
considered to be important myofibroblast precursors in the diseased liver; however,
recent literature has forced a reevaluation of the view that HSCs are the dominant
fibrogenic cells in all forms of fibrosis. Other mesenchymal cell populations, most
notably portal fibroblasts and circulating cells from the bone marrow, are emerging
as significant matrix-producing cells. Hepatocytes and BECs, previously relegated
to minor roles in fibrosis, are hypothesized to undergo EMT and participate in fibrogen-
esis in the form of myofibroblasts. The challenges of the next decade are to define the
role of different fibrogenic cells in different forms of fibrosis, characterize more
systematically the ECM produced by distinct fibroblast and myofibroblast popula-
tions, define the source and function of newly appreciated and minor matrix proteins,
and identify the cell-based, soluble, and mechanical factors underlying fibrogenic cell
differentiation.
REFERENCES
1. Wells RG. The role of matrix stiffness in regulating cell behavior. Hepatology 2008;
47(4):1394–400.
2. Wells RG. Function and metabolism of collagen and other extracellular matrix pro-
teins. In: Rhodes J, editor. The textbook of hepatology: from basic science to clin-
ical practice. 3rd edition. Oxford (UK): Blackwell Publishing; 2007.
3. Gressner OA, Weiskirchen R, Gressner AM. Evolving concepts of liver fibro-
genesis provide new diagnostic and therapeutic options. Comp Hepatol 2007;
6:7.
4. Gabbiani G. The myofibroblast in wound healing and fibrocontractive diseases.
J Pathol 2003;200(4):500–3.
5. Hinz B, Phan SH, Thannickal VJ, et al. The myofibroblast: one function, multiple
origins. Am J Pathol 2007;170(6):1807–16.
6. Friedman SL. Hepatic stellate cells: protean, multifunctional, and enigmatic cells
of the liver. Physiol Rev 2008;88(1):125–72.
7. Friedman SL, Roll FJ, Boyles J, et al. Hepatic lipocytes: the principal collagen-
producing cells of normal rat liver. Proc Natl Acad Sci U S A 1985;82(24):8681–5.
8. Maher JJ, McGuire RF. Extracellular matrix gene expression increases preferen-
tially in rat lipocytes and sinusoidal endothelial cells during hepatic fibrosis in
vivo. J Clin Invest 1990;86(5):1641–8.
9. Nakatsukasa H, Nagy P, Evarts RP, et al. 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 1990;85(6):1833–43.
10. Friedman SL. Molecular regulation of hepatic fibrosis, an integrated cellular
response to tissue injury. J Biol Chem 2000;275(4):2247–50.
11. Bataller R, Brenner DA. Liver fibrosis. J Clin Invest 2005;115(2):209–18.
12. Wallace K, Burt AD, Wright MC. Liver fibrosis. Biochem J 2008;411(1):1–18.
13. Knittel T, Kobold D, Saile B, et al. Rat liver myofibroblasts and hepatic stellate
cells: different cell populations of the fibroblast lineage with fibrogenic potential.
766 Wells
14. Beaussier M, Wendum D, Schiffer E, et al. Prominent contribution of portal mes-

enchymal cells to liver fibrosis in ischemic and obstructive cholestatic injuries.
Lab Invest 2007;87(3):292–303.
15. Tuchweber B, Desmouliere A, Bochaton-Piallat ML, et al. Proliferation and pheno-
typic modulation of portal fibroblasts in the early stages of cholestatic fibrosis in
the rat. Lab Invest 1996;74(1):265–78.
16. Kinnman N, Housset C. Peribiliary myofibroblasts in biliary type liver fibrosis.
Front Biosci 2002;7:d496–503.
17. Tang L, Tanaka Y, Marumo F, et al. Phenotypic change in portal fibroblasts in
biliary fibrosis. Liver 1994;14(2):76–82.
18. Li Z, Dranoff JA, Chan EP, et al. Transforming growth factor-beta and substrate
stiffness regulate portal fibroblast activation in culture. Hepatology 2007;46(4):
1246–56.
19. Guyot C, Lepreux S, Combe C, et al. Hepatic fibrosis and cirrhosis: the
(myo)fibroblastic cell subpopulations involved. Int J Biochem Cell Biol 2006;
38(2):135–51.
20. Kalluri R, Zeisberg M. Fibroblasts in cancer. Nat Rev Cancer 2006;6(5):392–401.
21. Magness ST, Bataller R, Yang L, et al. A dual reporter gene transgenic mouse
demonstrates heterogeneity in hepatic fibrogenic cell populations. Hepatology
2004;40(5):1151–9.
22. Baba S, Fujii H, Hirose T, et al. Commitment of bone marrow cells to hepatic
stellate cells in mouse. J Hepatol 2004;40(2):255–60.
23. Forbes SJ, Russo FP, Rey V, et al. A significant proportion of myofibroblasts are of
bone marrow origin in human liver fibrosis. Gastroenterology 2004;126(4):
955–63.
24. Russo FP, Alison MR, Bigger BW, et al. The bone marrow functionally contributes
to liver fibrosis. Gastroenterology 2006;130(6):1807–21.
25. Kisseleva T, Uchinami H, Feirt N, et al. Bone marrow-derived fibrocytes partici-
pate in pathogenesis of liver fibrosis. J Hepatol 2006;45(3):429–38.
26. Herbst H, Frey A, Heinrichs O, et al. Heterogeneity of liver cells expressing
procollagen types I and IV in vivo. Histochem Cell Biol 1997;107(5):399–409.
27. Milani S, Herbst H, Schuppan D, et al. In situ hybridization for procollagen types I,
III and IV mRNA in normal and fibrotic rat liver: evidence for predominant expres-
sion in nonparenchymal liver cells. Hepatology 1989;10(1):84–92.
28. Strutz F, Okada H, Lo CW, et al. Identification and characterization of a fibroblast
marker: FSP1. J Cell Biol 1995;130(2):393–405.
29. Liu Y. Epithelial to mesenchymal transition in renal fibrogenesis: pathologic signif-
icance, molecular mechanism, and therapeutic intervention. J Am Soc Nephrol
2004;15(1):1–12.
30. Iwano M, Plieth D, Danoff TM, et al. Evidence that fibroblasts derive from epithe-
lium during tissue fibrosis. J Clin Invest 2002;110(3):341–50.
31. Del Castillo G, Murillo MM, Alvarez-Barrientos A, et al. Autocrine production of
TGF-beta confers resistance to apoptosis after an epithelial-mesenchymal transi-
tion process in hepatocytes: role of EGF receptor ligands. Exp Cell Res 2006;
312(15):2860–71.
32. Ju W, Ogawa A, Heyer J, et al. Deletion of Smad2 in mouse liver reveals novel
functions in hepatocyte growth and differentiation. Mol Cell Biol 2006;26(2):
654–67.
33. Kaimori A, Potter J, Kaimori JY, et al. Transforming growth factor-beta1 induces
an epithelial-to-mesenchymal transition state in mouse hepatocytes in vitro.
J Biol Chem 2007;282(30):22089–101.
34. Kojima T, Takano K, Yamamoto T, et al. Transforming growth factor-beta induces

epithelial to mesenchymal transition by down-regulation of claudin-1 expression
and the fence function in adult rat hepatocytes. Liver Int 2008;28(4):534–45.
35. Valdes F, Alvarez AM, Locascio A, et al. The epithelial mesenchymal transition
confers resistance to the apoptotic effects of transforming growth factor Beta in
fetal rat hepatocytes. Mol Cancer Res 2002;1(1):68–78.
36. Diaz R, Kim JW, Hui JJ, et al. Evidence for the epithelial to mesenchymal transi-
tion in biliary atresia fibrosis. Hum Pathol 2008;39(1):102–15.
37. Zeisberg M, Yang C, Martino M, et al. Fibroblasts derive from hepatocytes in liver
fibrosis via epithelial to mesenchymal transition. J Biol Chem 2007;282(32):
23337–47.
38. Tamkun JW, Hynes RO. Plasma fibronectin is synthesized and secreted by hepa-
tocytes. J Biol Chem 1983;258(7):4641–7.
39. Hahn E, Wick G, Pencev D, et al. Distribution of basement membrane proteins in
normal and fibrotic human liver: collagen type IV, laminin, and fibronectin. Gut
1980;21(1):63–71.
40. Odenthal M, Neubauer K, Meyer zum Buschenfelde KH, et al. Localization and
mRNA steady-state level of cellular fibronectin in rat liver undergoing a CCl4-
induced acute damage or fibrosis. Biochim Biophys Acta 1993;1181(3):266–72.
41. Milani S, Herbst H, Schuppan D, et al. Cellular localization of laminin gene tran-
scripts in normal and fibrotic human liver. Am J Pathol 1989;134(6):1175–82.
42. Schier C, Schier F, Voss B, et al. Characterization of human extrahepatic biliary
duct epithelial cells in culture. Exp Mol Pathol 1988;48(3):301–10.
43. Rescan PY, Loreal O, Hassell JR, et al. Distribution and origin of the basement
membrane component perlecan in rat liver and primary hepatocyte culture. Am
J Pathol 1993;142(1):199–208.
44. Jia JD, Bauer M, Sedlaczek N, et al. Modulation of collagen XVIII/endostatin ex-
pression in lobular and biliary rat liver fibrogenesis. J Hepatol 2001;35(3):386–91.
45. Milani S, Herbst H, Schuppan D, et al. Vimentin expression of newly formed rat
bile duct epithelial cells in secondary biliary fibrosis. Virchows Arch A Pathol
Anat Histopathol 1989;415(3):237–42.
46. Rygiel KA, Robertson H, Marshall HL, et al. Epithelial-mesenchymal transition
contributes to portal tract fibrogenesis during human chronic liver disease. Lab
Invest 2008;88(2):112–23.
47. Xia JL, Dai C, Michalopoulos GK, et al. Hepatocyte growth factor attenuates liver
fibrosis induced by bile duct ligation. Am J Pathol 2006;168(5):1500–12.
48. Robertson H, Kirby JA, Yip WW, et al. Biliary epithelial-mesenchymal transition in
posttransplantation recurrence of primary biliary cirrhosis. Hepatology 2007;
45(4):977–81.
49. Sicklick JK, Choi SS, Bustamante M, et al. Evidence for epithelial-mesenchymal
transitions in adult liver cells. Am J Physiol Gastrointest Liver Physiol 2006;
291(4):G575–83.
50. Horn T, Junge J, Christoffersen P. Early alcoholic liver injury: changes of the Disse
space in acinar zone 3. Liver 1985;5(6):301–10.
51. Bardadin KA, Desmet VJ. Ultrastructural observations on sinusoidal endothelial
cells in chronic active hepatitis. Histopathology 1985;9(2):171–81.
52. DeLeve LD. Hepatic microvasculature in liver injury. Semin Liver Dis 2007;27(4):
390–400.
53. Elvevold K, Smedsrod B, Martinez I. The liver sinusoidal endothelial cell: a cell
type of controversial and confusing identity. Am J Physiol Gastrointest Liver Phys-
iol 2008;294(2):G391–400.
768 Wells
54. Schaffner F, Popper H. Capillarization of hepatic sinusoids in man. Gastroenterol-

ogy 1963;44:239–42.
55. Roskams T, Moshage H, De Vos R, et al. Heparan sulfate proteoglycan expres-
sion in normal human liver. Hepatology 1995;21(4):950–8.
56. Roskams T, Rosenbaum J, de Vos R, et al. Heparan sulfate proteoglycan expres-
sion in chronic cholestatic human liver diseases. Hepatology 1996;24(3):524–32.
57. Clement B, Rescan PY, Baffet G, et al. Hepatocytes may produce laminin in
fibrotic liver and in primary culture. Hepatology 1988;8(4):794–803.
58. Geerts A, Greenwel P, Cunningham M, et al. Identification of connective tissue
gene transcripts in freshly isolated parenchymal, endothelial, Kupffer and fat-stor-
ing cells by northern hybridization analysis. J Hepatol 1993;19(1):148–58.
59. Nakayama Y, Takahara T, Miyabayashi CR, et al. Ultrastructural localization of
type IV collagen and laminin in the Disse space of rat liver with carbon tetrachlo-
ride induced fibrosis. Liver 1991;11(5):260–71.
60. Tsutsumi M, Takada A, Takase S, et al. Connective tissue components in cultured
parenchymal and nonparenchymal cells of rat liver. Immunohistochemical stud-
ies. Lab Invest 1988;58(1):88–92.
61. Hopf M, Gohring W, Mann K, et al. Mapping of binding sites for nidogens, fibulin-
2, fibronectin and heparin to different IG modules of perlecan. J Mol Biol 2001;
311(3):529–41.
62. Zeisberg M, Kramer K, Sindhi N, et al. De-differentiation of primary human hepa-
tocytes depends on the composition of specialized liver basement membrane.
Mol Cell Biochem 2006;283(1–2):181–9.
63. Serini G, Bochaton-Piallat ML, Ropraz P, et al. The fibronectin domain ED-A is
crucial for myofibroblastic phenotype induction by transforming growth factor-
beta1. J Cell Biol 1998;142(3):873–81.
64. George J, Wang SS, Sevcsik AM, et al. Transforming growth factor-beta initiates
wound repair in rat liver through induction of the EIIIA-fibronectin splice isoform.
Am J Pathol 2000;156(1):115–24.
65. Jarnagin WR, Rockey DC, Koteliansky VE, et al. Expression of variant fibronectins
in wound healing: cellular source and biological activity of the EIIIA segment in rat
hepatic fibrogenesis. J Cell Biol 1994;127(6 Pt 2):2037–48.
66. Thiele GM, Duryee MJ, Freeman TL, et al. Rat sinusoidal liver endothelial cells
(SECs) produce pro-fibrotic factors in response to adducts formed from the
metabolites of ethanol. Biochem Pharmacol 2005;70(11):1593–600.
67. Odenthal M, Neubauer K, Baralle FE, et al. Rat hepatocytes in primary culture
synthesize and secrete cellular fibronectin. Exp Cell Res 1992;203(2):289–96.
68. Ramadori G, Knittel T, Odenthal M, et al. Synthesis of cellular fibronectin by rat
liver fat-storing (Ito) cells: regulation by cytokines. Gastroenterology 1992;
103(4):1313–21.
69. Xu G, Niki T, Virtanen I, et al. Gene expression and synthesis of fibronectin
isoforms in rat hepatic stellate cells. Comparison with liver parenchymal cells
and skin fibroblasts. J Pathol 1997;183(1):90–8.
Oxidative a nd
Nitrosative Stress a nd
Fibro genic Respons e
R. Urtasun, PhD, L. Conde de la Rosa, PhD, N. Nieto, PhD*
KEYWORDS
CYP2E1, Cytochrome P450 2E1 ECM, Extracellular matrix
ERK 1/2, Extracellular signal-regulated kinase 1/2
GSH, Glutathione HSC, Hepatic stellate cells
H2O2, Hydrogen peroxide IFNg, Interferon g IL, Interleukin
MMPs, Matrix metalloproteinases
MAPK, Mitogen-activated protein kinase NO$, Nitric oxide
NOS2, Nitric oxide synthase 2 ONOO , Peroxynitrite
PI3K, Phosphatidylinositol 3-kinase
PDGF, Platelet-derived growth factor
RNS, Reactive nitrogen species
ROS, Reactive oxygen species
O2˙ , Superoxide anion
TIMPs, Tissue inhibitor of metalloproteinases
TGFb, Transforming growth factor-beta
Hepatic fibrosis represents a significant global health problem for which no adequate ther-
apy exists.1,2 Alcoholic fibrosis is the liver’s wound-healing response to injury and it can
lead to cirrhosis characterized by scar accumulation and nodule formation. Thus far, there
is no proven therapy for hepatic fibrosis, which can be further complicated by hepatocel-
lular carcinoma. Hence, prevention of liver fibrosis could help to ameliorate the complica-
tions of cirrhosis and improve the quality of life of many patients worldwide.1,2
Uncontrolled production of collagen I is the main feature of liver fibrosis.1–13 Follow-
ing a fibrogenic stimulus such as alcohol, hepatic stellate cells (HSC) transform into an
activated collagen-producing cell. In alcoholic liver disease, numerous changes in
gene expression are associated with HSC activation, including the induction of several
intracellular signaling cascades, which help maintain the activated phenotype and
control the fibrogenic and proliferative state of the cell.1–16 Detailed analyses for
understanding the molecular basis of the collagen I gene regulation have revealed
Urtasun R and Conde de la Rosa L contributed equally.

Mount Sinai School of Medicine, Box 1123, Department of Medicine/Division of Liver Diseases,
1425 Madison Avenue, Room 11-76, New York, NY 10029, USA
E-mail address: natalia.nieto@mssm.edu (N. Nieto).
Clin Liver Dis 12 (2008) 769–790

770 Urtasun et al
a complex process involving reactive oxygen species (ROS) as key mediators.1,2 Less
is known, however, about the contribution of reactive nitrogen species (RNS). In
addition, a series of cytokines, growth factors, and chemokines, which activate
extracellular matrix (ECM)-producing cells through paracrine and autocrine loops,
contribute to the fibrogenic response.17
RELEVANCE OF OXIDATIVE STRESS IN LIVER DISEASE
Following alcohol consumption, cholestasis, and iron overload, ROS and lipid peroxida-
tion products are generated in large amounts, leading to Kupffer cell activation,18–23
a key event in the liver inflammatory and profibrogenic response, and to the secretion
of a myriad of growth factors, cytokines, and prostaglandins.24,25 Among other stimuli,
hepatocyte-derived lipid peroxidation induced by ethanol plays an important role in the
pathogenesis of liver fibrosis by up-regulating collagen I synthesis.26 Acetaldehyde is
the first metabolite of ethanol and is a profibrogenic agent that stimulates intracellular
accumulation of H2O2.27 In addition, H2O2 has been also implicated in the onset of
scleroderma, an event consistent with earlier clinical evidence for ROS participation
in disease pathology.28,29 Furthermore, ROS-sensitive cytokines contribute to HSC ac-
tivation during inflammatory through paracrine signals released from immune cells.30
Alcohol metabolism by cytochrome P450 2E1 (CYP2E1) generates ROS.31 Binge
drinking triggers steatosis, a fibrogenic response, and apoptosis in rats fed a choline-
deficient diet through increased oxidant stress, elevated phosphorylation of p38, and
down-regulation of ECM proteolytic enzymes.32 To better understand how HSC
become activated in the presence of oxidative stress and to evaluate whether
CYP2E1-derived ROS could play a role in HSC activation, our laboratory co-incubated
primary HSC with HepG2 cells overexpressing or not CYP2E1.8,9 There were
enhanced proliferation rates and induction of a-smooth muscle actin, intracellular
and secreted collagen I protein, and intra- and extracellular H2O2 and lipid peroxida-
tion products in HSC co-incubated with HepG2 cells overexpressing CYP2E1
compared with HSC incubated with control HepG2 cells.8,9 These effects were
prevented by antioxidants and by CYP2E1 inhibitors, suggesting a role for CYP2E1-
derived diffusible mediators on these effects.8,9
A causal relationship also exists between oxidative stress and chronic alcohol-
induced liver injury, with sustained activation of Kupffer cells and HSC. A recent study4
demonstrated that Kupffer cells induced a more activated phenotype, greater prolifer-
ation rates, and increased intra- and extracellular collagen I protein, H2O2, and IL-6 in
HSC co-cultured with Kupffer cells when compared with HSC cultured alone. All these
features were prevented by catalase, indicating a role for H2O2.4 In addition, MMP-13,
which degrades extracellular collagen I, decreased in the co-culture of HSC with
Kupffer cells, while there was up-regulation of tissue inhibitor of metalloproteinase 1
(TIMP1), a MMP-13 inhibitor. A novel dual mechanism mediated by H2O2 and IL-6
was proposed for how Kupffer cells could modulate the fibrogenic response in HSC.4
Supporting the role of oxidative stress in fibrosis, a novel study showed a correlation
between cellular activation and oxidative stress in the GRX cell line and how vitamin E
and N-acetylcysteine prevented the activation.33 Parola and colleagues34 also have
demonstrated that activated HSC are more vulnerable than quiescent HSC to
aldehyde end products.
SYNERGISM BETWEEN REACTIVE OXYGEN SPECIES AND GROWTH FACTORS
The role of pro-fibrogenic cytokines and growth factors is central for the development
of liver fibrosis because they allow cross-talk between ECM-producing cells.3–6 ROS
Oxidative-Nitrosative Stress and Fibrosis 771
play a decisive role in the initial phase of fibrosis by integrating different profibrotic
stimuli independently of TGFb. By contrast, progression to the subsequent stage
depends on TGFb production and the canonical Smad pathway.35 Liver fibrosis
involves elevated expression of TGFb in rodents36,37 and in humans.38 In normal liver
and in CCl4-induced fibrosis, TGFb1 is mainly produced by Kupffer cells and HSC
although a small amount is generated by endothelial cells.36,39 TGFb is secreted as
a latent complex that is trapped among matrix fibbers, from which it is released and
activated during tissue remodeling.40,41 Activated TGFb signals through transmem-
brane serine/threonine kinases and intracellular Smad proteins, which translocate
into the nucleus, activate gene transcription by binding the ‘‘CAGA’’ consensus
sequence, and interact with promoter-specific transcription factors and general co-
activators42 to initiate and perpetuate the fibrogenic response.43 TGFb1 is a redox
sensitive gene;44 indeed, ROS (eg, H2O2) up-regulate TGFb1 expression in rat HSC,
which is blocked by catalase.45 In addition, TGFb per se increases O2˙ production
in fibroblasts.46 Other studies have described how TGFb increases ROS production
by activating the membrane-bound enzyme NADPH oxidase47,48 and by impairing
complex IV in the mitochondrial respiratory chain.49 The canonical signal transduction
pathway of TGFb plays a central role in fibrosis42,43,50,51 even though phosphatidylino-
sitol 3-kinase (PI3K)/Akt is a well-documented example of an alternative pathway that
is induced by TGFb in different cell lines and profibrogenic conditions.35,52
PDGF is the most potent mitogen for transdifferentiated HSC during liver fibrosis,
and the expression of PDGF receptors is important for mitogenesis and chronic
inflammation of the liver,53–55 and particularly chemo-attraction of mononuclear
cells.56 The expression of PDGF is also influenced by intracellular redox changes.57
TGFb1 regulates the PDGFRb in human HSC.58 Moreover, the PDGF-dependent
activation of HSC is followed by phosphorylation of PI3K.59,60 PI3K activation is essen-
tial for both mitogenesis and chemotaxis induced by PDGF during liver injury in vivo.61
The 70-kDa ribosomal S6 kinase is activated in a PI3K-dependent manner and plays
an important role in HSC proliferation, collagen expression, and cell cycle control,
representing a potential therapeutic target for liver fibrosis.62 FoxO1 inhibits PDGF-
induced HSC proliferation via G1 cell cycle arrest suggesting that FoxO1 is a crucial
downstream target of the PI3K/Akt pathway in regulating HSC proliferation.60 PI3K is
induced by oxidative stress in rat HSC following exposure to CCl4.63 HSC isolated
from CCl4-treated rats or from acute liver damage in vivo show increased ERK activ-
ity, which modulates HSC proliferation and chemotaxis and regulates nuclear
signaling.64
The proliferative effect of PDGF requires the activation of PDGFRb.65,66 Further-
more, the activation of cultured rat HSC by Kupffer cell-conditioned medium directly
enhances matrix synthesis and stimulates HSC proliferation via induction of PDGF
receptors.65 The myristoylated alanine-rich protein kinase C substrate is a downstream
effector in PDGF-induced motility of activated human HSC.67 Different compounds
that activate the AMPK pathway inhibit PDGF-stimulated proliferation and migration
of human HSC and reduce the secretion of monocyte chemoattractant protein-1,
suggesting that AMPK negatively modulates the activated phenotype of HSC.68 Nitric
oxide (NO$) donors also exert a direct antifibrogenic action by inhibiting PDGF-
induced proliferation, motility, and contractility in HSC, in addition to lowering ECM
proteins.69 A regulatory mechanism of reactive aldehydes on PDGFRb signaling and
biologic actions is relevant to liver fibrosis.70
TNFa secreted by macrophages, Kupffer cells, and HSC, as well as IFNg secreted
by T cells, are well-known antifibrogenic signals.71,72 TNFa reduces ECM deposition
by inhibiting the synthesis of structural components, including elastin, osteocalcin,
772 Urtasun et al
and collagen I.73–76 TNFa counteracts TGFb-stimulation of collagen I gene in different

cell types.77,78 Both TNFa and IFNg blunt the TGFb-mediated up-regulation of the
COL1A2 promoter by interfering with the formation of the TGFb1-responsive element
complex.79–81 TGFb antagonism by TNFa involves c-Jun N-terminal kinase-1 phos-
phorylation of c-Jun, which leads to off-DNA interference of Smad3 binding to the
cognate DNA site and/or interaction with the p300/CBP co-activators.82 Moreover,
a previous report has shown that p38 MAPK is a key mediator of the antifibrogenic
effect of TNFa in regulating the expression of col1A1 in HSC in response to
cytokines.83
In addition, interferon regulatory factor-binding site IF3 is a novel target of the path-
ways elicited by IFNg to blunt COL1A2 promoter transcription.84 IFNg promotes occu-
pancy of the COL1A2 transcription start site by the RFX5/CIITA complex, which
interacts with CBF/NFY and/or YB-1.79 These antifibrogenic cytokines down-regulate
COL1A2 transcription and antagonize TGFb through multiple pathways, which
converge on the same promoter elements.85,86 TNFa–mediated down-regulation of
the mouse col1a1 gene is associated with the activation and binding of C/EBPd-
and C/EBPb-containing complexes to the -370 to -344 region of the mouse col1a1
promoter (Table 1).87
EFFECTS OF REACTIVE OXYGEN SPECIES ON THE COL1A1 AND COL1A2 PROMOTERS
Collagen I, the most abundant collagen type found in liver fibrosis, is a heterotrimeric
protein composed of two a1 chains and one a2 chain forming a triple helical structure.
The human a-chains genes are located as single copies on different chromosomes.
The a1(I) chain gene is on chromosome 17q21-22 whereas the a2(I) chain gene is
located on 7q21-22.88,89 In normal tissue, both genes are co-ordinately expressed,90
whereas a homotrimer of three a1(I) chains occasionally occurs in tumors91 and in
cultured cells.92 Structural and metabolic deficiencies of the collagen I chains lead
Table 1
Summary of different factors that modulate collagen I expression
Collagen I
ROS(O˙2 and H2O2)11,29,47,64,91,133,157–160 [
IL-4100 [
IL-6101,102 [
Acetaldehyde29,106,128–131 [
Arachidonic acid11 [
Malondialdehyde134 [
PDGF61,62,67 [
Nitric oxide71 Y
TGFb47,112,113 [
TNFa78,85,89 Y
INFg81,83,86 Y
Fli-1120–122 Y
Sp1/Sp3118,119 [
Adiponectin123,124 Y
Leptin125–127 [
to several heritable and acquired disorders of connective tissue,93 in general, and

impair liver function, in particular.94
The IL-4-induced transcriptional activator STAT6 binds to various sequences within
the COL1A1 and COL1A2 promoters.95 An AP-2 site adjacent to the reverse-oriented
STAT6 consensus motif TTCN3/4 GCT is located within 205 bp from the transcription
start site and seems to support the moderate IL-4-induced COL1A1 gene activation.
Furthermore, IL-6 up-regulates the expression of type I collagen in vivo96 and in
cultured HSC.97–99 HSC respond to IL-6 with a transient increase in col1a1 mRNA
expression.97,100 A TGFb1-responsive element is found in the -370 to -344 bp region
of the mouse col1a1 gene.45 TGFb1 induces the activation and binding to the TGFb1-
responsive element of a protein complex that contains C/EBPb, and H2O2 acts as
a second messenger for the TGFb1-mediated col1a1 gene up-regulation.45 In fact,
acetaldehyde induces col1a1 up-regulation via H2O2.101 However, IFNg and TNFa
down-regulate transcription of the COL1A1 promoter.95 The TNFa-mediated down-
regulation of the col1a1 gene is associated with activation and binding of C/EBPd-
and C/EBPb-containing complexes to the -370 to -344 bp region of the mouse
col1a1 promoter. Indeed, over-expression of C/EBPd or p20C/EBPb down-regulates
the expression of a reporter construct driven by the -412 to 1110 bp sequence of the
col1a1 promoter, validating the relevance of these two transcription factors in the
TNFa-mediated col1a1 down-regulation.87
The functional properties of the proximal promoter of the COL1A2 gene have been
studied in transfection experiments,102 which have described the minimal upstream
sequence directing high and cell type-specific expression of the CAT-reporter gene.
The proximal promoter of the COL1A2 spans from 380 to 154 bp relative to the tran-
scription start site and contains several overlapping DNA elements that are bound by
ubiquitous transcription factors. Constitutive transcription of the proximal promoter of
the COL1A2 is under the control of four clusters of cis-acting elements, which are
involved in mediating the transcriptional response to cytokines implicated in tissue
remodeling and fibrosis.103 Recent studies have underscored the importance of the
proximal promoter in proper COL1A2 expression, describing, in addition, that a far-
upstream enhancer and a downstream silencer are also part of the regulatory network
of the COL1A2 gene.104,105 A model has been proposed whereby COL1A2 expression
is the result of combinatorial interactions amongst trans-acting factors bound within
the promoter, enhancer, and repressor sequences.85 The downstream repressor
resides within the first intron of COL1A2 and contains a DNase I hypersensitive site
located around a cluster of three cis-acting elements, which contain binding sites
for ‘GATA’ (FIi1 an FIi2) and IRF (FIi3) nuclear proteins.105
Many nuclear factors that have been implicated in regulating the COL1A2 proximal
promoter in fibroblasts, including Sp1/Sp3, NFkB, C/EBPd and b, AP1, Fli-1/Ets-1,
CBF/NFY, YB-1 and the Smads 3/4 and RFX5/CIITA complexes.76,79,81,106 Specifi-
cally, either single or multiple ‘‘CAGA’’ boxes are present in the TGFb -responsive
element on the COL1A2 gene; hence, the Smad3/4 complex represents a common
mediator of the TGFb signaling for ECM accumulation.107 The ubiquitous transcription
factor Sp1, the Smad3/4 complex, and the co-activators p300/CBP mediate the
response of the COL1A2 promoter to TGFb stimulation.85,108,109 A Smad complex
may represent the alleged Sp1 co-factor involved in COL1A2 trans-activation.110
Transient over-expression of Smad3 and Smad4 can transactivate the COL1A2
promoter.111 Sp1 and Smad3/Smad4 cooperate synergistically in transactivating the
COL1A2 promoter after binding to the TGFb1-responsive element. Furthermore, there
is additional evidence for a critical role of Sp1 in constitutive COL1A2 expression, and
in integrating the transcriptional responses of the gene to antagonistic cytokines.112
774 Urtasun et al
Sp1/Sp3 proteins bound to the COL1A2 promoter interact with CBF/NFY to

strengthen promoter activity and patterned transgene expression.113,114 Fli-1, a mem-
ber of Ets transcriptional factors, is a negative regulator of the COL1A2 gene expres-
sion in dermal fibroblasts.115 Competition between Fli-1 and Ets-1 for binding to the
same promoter sequence is associated both with modulating constitutive COL1A2
activity and inhibiting TGFb signaling.116 In addition, a recent study has shown that
TGFb -dependent acetylation and inhibition of Fli-1 may represent the principal mech-
anisms responsible for the TGFb -induced dissociation of Fli-1 from the COL1A2
promoter.117
Several studies indicate that adiponectin has antifibrotic properties because hepatic
fibrosis produced by chronic CCl4 was enhanced in adiponectin-knockout mice as
compared with wild-type mice,118,119 while leptin has an opposite effect of enhancing
fibrogenesis.120–122
Acetaldehyde up-regulates type I collagen in HSC123–125 and the COL1A1 and
COL1A2 induction occurs through a TGFb-dependent mechanism.126 In addition,
acetaldehyde stimulates intracellular accumulation of H2O2 and COL1A2 promoter
activity in HSC,29 strongly suggesting that the early stage of ethanol-induced liver
fibrosis induces a H2O2-dependent loop, which triggers and amplifies autocrine
TGFb production, via activation of the Sp1-Smad3/4 complex, conceivably through
the PI3K pathway.29,127 Another report has implicated H2O2 in the pathology of sclero-
derma, demonstrating that PDGF treatment of primary human fibroblasts triggers an
intracellular loop involving Ha-Ras, ERK1/2, and ROS, ultimately leading to COL1A2
up-regulation.128 The TGFbRI-dependent program and the up-regulation of collagen
I does not involve Smad2/3 activation but is mediated by ALK1/Smad1 and ERK1/2
pathways in scleroderma fibroblasts.128 The involvement of H2O2 in the induction of
the COL1A1 promoter under TGFb treatment has been defined.45,76,101 Primary
Kupffer cells in co-culture with HSC induce a profibrogenic response mediated by
H2O2 that leads to up-regulation of COL1A1 and COL1A2 trans-activation and simul-
taneously prevents collagen I protein degradation via an IL-6-dependent mechanism.4
Work from our group has described a role for H2O2 in the up-regulation of COL1A2
expression by ethanol and arachidonic acid whereby COX-2 appears to mediate the
arachidonic acid-mediated induction of COL1A2 expression (see Table 1).11
STABILITY AND DEGRADATION OF COLLAGEN I: ROLE OF METALLOPROTEINASES
Liver fibrosis is characterized by activation of HSC and subsequent ECM synthesis,

along with insufficient degradation. Matrix remodeling occurs mainly due to the action
of MMPs, a multidomain family of zinc-dependent endopeptidases. These enzymes
are secreted into the extracellular space as zymogens that require activation by
a variety of stimuli. The active enzymes can, in turn, be inhibited by the family of tissue
inhibitors of metalloproteinases (TIMPs). ECM remodeling is, therefore, highly regu-
lated under physiologic conditions. Alteration of the balance between MMPs plays
a role in scarring.129,130
The three most relevant MMPs are gelatinase A (MMP-2), gelatinase B (MMP-9),
and stromelysin (MMP-3). In liver fibrosis, the expression of the MMPs involved in
fibrillar collagen degradation (eg, MMP-1 in humans and rodents and MMP-13 also
in rodents) is limited, whereas the expression of MMP-2 is markedly increased.131
TGFb1 modulates MMP-13 expression in HSC by complex mechanisms involving
p38 MAPK, PI3K/AKT and p70-ribosomal S6 kinase.132 MMP-2 can degrade several
components of the subendothelial matrix, including collagen IV, laminin, and fibronec-
tin, and it may be important in the remodeling of matrix during tissue repair
processes.133 A recent study has shown that oxidative stress induces MMP-2
expression, proliferation, and invasiveness of HSC; these effects could be prevented
by specific MMP inhibitors and antioxidants.62 Increased expression of pro-gelatinase
and formation of active enzyme occurs in human liver disease and in animal models of
liver fibrosis.134,135 Sustained over-expression of MMPs like gelatinase A, with the
consequent degradation of basement-membrane collagen IV, represents a basic
mechanism in the remodeling of the space of Disse with capillarization of the sinu-
soids.136 In progressive liver fibrosis, the overall MMP activity decreases,137 due to
increased expression of TIMPs and other anti-proteases expressed by HSC and
hepatocytes.62,138 In liver fibrosis, hepatic TIMP-1 expression is markedly up-regu-
lated both in humans and in murine fibrosis models.130,139,140 Both TIMP-1 and
TIMP-2 are released by fully activated HSC.141 The increased expression of TIMPs
is important in advanced liver fibrosis both in rodents and in humans.142,143 In fact,
increased TIMP-1 and TIMP-2 mRNA levels have been demonstrated by in situ hybrid-
ization in CCl4-induced rat fibrosis,144 as well as in primary biliary cirrhosis and biliary
atresia.140
Plasmin is a broad-spectrum protease capable of directly degrading matrix compo-
nents, including fibronectin, laminin, and proteoglycans145 and also participates in
matrix degradation indirectly by activating MMP-13, MMP-1, MMP-3, interstitial colla-
genase, and stromelysin.146,147 PAI-1, a physiologic inhibitor of plasminogen activa-
tor, inhibits protease-dependent fibrinolytic activity and subsequent ECM
degradation. PAI-1 expression is up-regulated in a variety of fibrotic diseases as
well as in experimental animal models such as CCl4-induced liver fibrosis,148 and
PAI-1-deficient mice develop less severe fibrosis in lung,149 suggesting a role for
PAI-1 in the progression of fibrosis. Furthermore, GSH inhibits TGFb -induced colla-
gen I accumulation by blocking TGFb-induced PAI-1 expression, and thus stimulating
collagen degradation.150 Many other studies have also described TGFb as an inducer
of ROS production and ROS mediate PAI-1 induction by different stimuli.151–154
SOURCES OF REACTIVE NITROGEN SPECIES IN THE LIVER
Nearly all cell types in the liver, including hepatocytes, Kupffer cells, HSC, and endo-
thelial cells, have the capacity to generate NO$.155 The reactivity of NO$ per se has
been greatly overestimated in vitro because no drain is provided to remove NO$.156
NO$ remains in solution for several minutes in micromolar concentrations before it
reacts with O2 to form much stronger oxidants like nitrogen dioxide and others.157
Most biological actions of NO$ appear to be mediated by interactions with paramag-
netic centers in effector proteins, such as heme- or iron-sulfur centers, but NO$ is also
known to react rapidly, via radical termination reactions, with other targets that carry
unpaired electrons.158 These reactions include interactions with ROS such as O2˙ , or
with radical intermediates in proteins or lipids.156 Furthermore, NO$ reacts with O2 to
form higher oxides of nitrogen, in a relatively slow reaction (Fig. 1).159 The eventual
biologic fate of NO$ is oxidation to nitrite and nitrate, end products of NO$ metabolism
that are rapidly distributed throughout the body and excreted in urine.158
There are a number of RNS derived from NO$.160 Of these, peroxynitrite (ONOO ) is
the best characterized and appears to have the highest biological activity.160 ONOO
is formed by the bi-radical reaction of NO$ and O2˙ . The reaction is extremely fast and
will occur at a near diffusion-limited rate.161 NO$ is the only biological molecule
produced in concentrations large enough to compete with superoxide dismutase for
O2˙ .156 ONOO reacts relatively slowly with most biological molecules, which
defines it as a selective oxidant. Effects of ONOO may also be beneficial or
776 Urtasun et al
NO. + O2 NO2
N2O3
NO -
2
NOS NO·
NO. + O2.- ONOO-
L-Arginine L-Citrulline
ONOO- + H+ ONOOH NO3-
Protein nitration
Fig. 1. Generation of NO$ by NOS in liver cells. NO$ is relatively unstable in the presence of
O2 and will rapidly and spontaneoulsy auto-oxidize to yield a variety of nitrogen oxides.
NO$ also reacts with O˙2 to generate ONOO . Although ONOO is relatively stable, it has
a pKa of 6.8, which implies that substantial amounts of ONOO will be protonated at phys-
iologic pH to yield peroxynitrous acid. This conjugate acid rapidly decomposes to yield
NO3 . Nitration of tyrosine residues by ONOO forms the stable product, 3-nitrotyrosine
(3-NT, the footprint for ONOO ) by addition of a nitro group to the 3-position adjacent
to the hydroxyl group of tyrosine.
detrimental depending on the concentration and local environment, the level of cellular
activation, and the endogenous GSH pool, which acts as a natural ONOO scaven-
ger.162 On the other hand, direct in vivo and in vitro evidence that, at low concentra-
tions, ONOO is actively involved in triggering cellular survival signals has been
reported in studies demonstrating protection against myocardial ischemia-reperfusion
injury and neuronal apoptosis.163,164
REACTIVE NITROGEN SPECIES AND HEPATOTOXICITY
NO$ is a short-life gaseous free radical known to exert many actions in the liver as well
as in other tissues and organs.165 This review summarizes only the major notions, with
special reference to interactions with ROS at the molecular level leading to collagen I
regulation, and effects at the cellular level (ie, HSC activation). In normal liver, low
fluxes of NO$ are produced by constitutive endothelial nitric oxide synthase (eNOS,
mainly in endothelial cells) and are considered sufficient to maintain perfusion of liver
sinusoids by acting on vascular tone (ie, vasodilatation) and on vascular permeabil-
ity.166 NO$ regulates leukocyte adhesion to sinusoidal endothelium and inhibits plate-
let adhesion and aggregation.167 In pathologic conditions, including endotoxemia and
chronic inflammation, nitric oxide synthase 2 (NOS2) is up-regulated in almost all liver
cells, including HSC,168 by several mediators and, consequently, NO$ generation
increases.166 Under these conditions, NO$ acts either as cytoprotective or as cyto-
toxic depending on the cellular microenvironment.169
The molecular regulation of NOS2 expression is complex and occurs at multiple
levels. NOS2 expression requires the transcription factor NFkB and is down-regulated
by steroids, TGFb, the heat shock response, p53, and NO$ itself.169 NO$ also presents
a protective effect both in vivo and in vitro by blocking TNFa-induced apoptosis and
hepatotoxicity, in part by a thiol-dependent inhibition of caspase-3-like protease
activity.170 These studies demonstrate the cytoprotective effects of NO$ in the liver
and suggest that hepatic NOS2 expression may function as an adaptive response
to minimize inflammatory injury.170 Thus, numerous mechanisms have evolved to reg-
ulate NOS2 expression during hepatocellular injury.171 The activation of NO$ synthesis
can be considered as an early adaptive response, which may become a mediator of
tissue damage in excess. Whether or not NO$ or secondary oxidants generated
from NO$ act as mediators of tissue injury or protect against toxicity will likely depend
on the precise targets of these RNS, the levels of O2˙ , and the extent to which tissue
injury is mediated by ROS.155
SYNERGISM BETWEEN REACTIVE NITROGEN SPECIES AND REACTIVE OXYGEN SPECIES
ROS may synergize with or antagonize RNS during liver injury and inflammation. ROS
and RNS are important in the process of energy generation, lipid peroxidation, protein
and DNA oxidation, nitration, nitrosation, or nitrosylation and catecholamine
response.172 ROS and RNS also strongly interact with reactive sulfur species, ie, de-
rivatives of reduced thiols (RSH) including the thiolate anion (RS ), thiyl radical (RS$)˙,
sulfenic acid (R-SOH), sulfinic acid (R-SOO), and sulfonic acid (R-SOOH) derivatives.
Among the many types of oxidative modifications induced by ONOO and other
RNS are the characteristic addition or substitution products in which NO$ is essentially
incorporated into the target molecule (ie, nitrosation and nitration reactions).158 For in-
stance, reactions with thiol residues to form S-nitrosothiols have been proposed as
a mechanism of either enzyme regulation or NO$ transport, and may provide a unique
signaling mechanism induced by nitrosative stress. S-Nitrosothiols in proteins
(eg, albumin) or in low-molecular-weight thiols, such as GSH, have been detected in
the circulation, bile, as well as in respiratory tract lining fluids.158 There is a mechanism
for pro-MMPs activation caused by S-glutathiolation whereby the GSH adduct of pro-
MMP may be produced through disulfide S-oxide formation involving generation of
S-nitrosoglutathione (GSNO2) by ONOO .173
The amino acid tyrosine appears to be a particularly susceptible target for nitration,
and the formation of free or protein-associated 3-nitrotyrosine has received much
recent interest as a potential biomarker for the generation of RNS in vivo.158 Further-
more, there is considerable evidence in the protein chemistry literature that nitration of
essential tyrosine residues can inactivate many enzymes or prevent phosphorylation
of tyrosine kinase substrates,174 and these findings have supported the hypothesis
that tyrosine nitration might result not only in the formation of inactive ‘‘footprints’’
of RNS but might also be functionally related to the pathobiology of inflammatory
diseases.175 For example, ONOO promoted nitration and/or phosphorylation of
regulatory sites at tyrosine kinase receptors coupled to the well-known anti-apoptotic
pathways involving PI3K/Akt or MAPK.176,177 Moreover, one of the most interesting
protective effects of NO$ is represented by the NO-dependent blocking of hepatocyte
apoptosis induced either by removal of growth factors or by exposure to TNFa or anti-
Fas antibody. This anti-apoptotic effect has been ascribed to S-nitrosylation of cas-
pase-3 and -8, with the subsequent inhibition of their activity.178
REACTIVE NITROGEN SPECIES AND HEPATIC STELLATE CELL ACTIVATION
Paracrine signaling is also important for nitrosative stress as it is for oxidative stress.
Kupffer cells also produce NO$, which can counterbalance the stimulatory effects of
ROS by reducing HSC proliferation, contractility, and collagen I production.179
Neutrophils are an important source of ROS, which have a direct stimulatory effect
on HSC collagen I synthesis. Activated neutrophils increased HSC collagen synthesis
3-fold over control levels. O2˙ was identified as the principal mediator of the neutro-
phils’ effect. Activated neutrophils also produce NO$, which dampened the effect of
O2˙ on collagen I expression but did not abrogate it completely.180
Activated HSC have contractile features181 that may contribute to increased intra-
hepatic portal hypertension via constriction of the sinusoid or by contraction of fibrous
ECM rich in collagen I with concomitant disruption of lobular architecture.182
778 Urtasun et al
Endothelin and NO$ play a major role in the modulation of HSC contractility, and are
therefore important in the pathogenesis of intrahepatic portal hypertension.182
Therefore, NO$ and NO$ donors are capable of preventing or reducing proliferative
responses of activated HSC. NO$ donors can efficiently inhibit PDGF-dependent pro-
liferation and chemotaxis in activated human HSC by activating an ibuprofen-sensi-
tive, prostaglandin E2 and cAMP-dependent pathway which interferes negatively
with PDGF signaling.69 Similar results (ie, inhibition of stimulated proliferation of
HSC by NO$ donors) have been found with angiotensin II as proliferative stimulus.183
There is recent work showing lipopolysaccharide-induced synthesis of IL-6, TNFa,
and NO$ via NOS2 in HSC.184 This group presented evidence that activation of p38 by
lipopolysaccharide initiates signaling via NFkB and ROS (eg, H2O2) leading to the
induction of NOS2 and expression of IL-6 and TNFa, major players in hepatic hemo-
dynamic regulation, inflammation, and immune responses.
REACTIVE NITROGEN SPECIES AND COLLAGEN I
Oxidative stress may represent a direct or indirect relevant profibrogenic stimulus for
HSC, as suggested by in vivo experimental studies in which administration of antiox-
idants prevents oxidative stress, lipid peroxidation, and liver fibrosis.185 Furthermore,
oxidative stress and lipid peroxidation are concomitant or precede HSC activation and
collagen I deposition.4 Exposure of cultured human or rat HSC to pro-oxidants or to
medium containing products released from hepatocytes undergoing oxidative stress
(ie, to mimic a possible paracrine effect by damaged parenchymal cells) is followed by
increased pro-collagen I gene expression.10
In contrast to ROS, which have been typically considered pro-fibrogenic agents,4,11
NO$ may be anti-fibrogenic.69,186 In the wound-healing response that restores tissue
integrity, NO$ is synthesized in the early phase by inflammatory cells, mainly macro-
phages.187,188 However, many cells participate in NO$ synthesis during the prolifera-
tive phase after wounding. NO$ released via NOS2 regulates collagen formation, cell
proliferation, and wound contraction in distinct ways in animal models of wound heal-
ing. Although NOS2 gene deletion delays, and arginine and NO$ administration,
improve healing, the exact mechanisms of action of NO$ on wound-healing
parameters are still unknown.187,188
ONOO can down-regulate type I collagen in dermal and cardiac fibroblasts,189
smooth muscle cells,190 and other cell types.191 In addition, ONOO can act as
a potent antifibrotic effector in animal models of experimental fibrosis,192 and in the
long-term inhibition of NOS2 in rats.186 However, in early traumatic wound-healing
conditions, ONOO favors collagen synthesis and the formation of granulation tis-
sue.193 There are different mechanisms to explain the inhibition of collagen by
ONOO , ie, ONOO may act through direct inhibition of collagen synthesis by proline
hydroxylation,191 stimulation of MMPs,194,195 reduced production of TGFb,189,195
initiation of fibroblast apoptosis,196 and/or neutralization of profibrogenic ROS.156,164
Recent data from our group indicate that ONOO and its secondary products may
have potential beneficial effects in the early fibrogenic response of HSC, which are
exposed to reactive species (ie, O2˙ and NO$) per se and also able to generate
them (Fig. 2). The authors found a time- and dose-dependent down-regulation of in-
tra- and extracellular collagen I protein along with an up-regulation of MMP-1 and
TNFa in HSC treated with either pure ONOO or a ONOO donor, which were blocked
by ONOO scavengers. The addition of ONOO increased nitration of MMP-1 and
MMP-13 leading to increased activity as the cleaved active isoforms 22/25 kDa for
MMP-1 and 44/48 kDa for MMP-13 were undetected in the absence of ONOO .
Fig. 2. In liver injury, ROS and RNS are generated in all liver cells. O2˙ can react with NO$ to
generate ONOO and other metabolites that may impact the HSC fibrogenic response in the
early stages of cellular activation. ONOO and its metabolites lower collagen I protein by
increasing TNFa and inducing nitration of MMP1 with the subsequent cleavage of collagen
I. These effects can be reverted by ONOO chelating agents.
However, the protective role of ONOO occurs only in the early fibrogenic response,
and it is lost at more advanced stages of the disease when other factors may hit in
a synergistic way.197
The temporal expression profiles of profibrogenic genes in HSC and their coordina-
tion concerning cell proliferation in alcoholic liver disease still needs further clarifica-
tion. Stress-derived mediators may activate seemingly contradictory signaling
pathways and the ultimate outcome may be dependent on the balance between these
stress-activated pathways because they could determine whether the cell proliferates
or undergoes a fibrogenic response.
REFERENCES
of the liver. Physiol Rev 2008;88:125–72.
2. Friedman SL. Mechanisms of disease: mechanisms of hepatic fibrosis and ther-
apeutic implications. Nat Clin Pract Gastroenterol Hepatol 2004;1:98–105.
3. Nieto N. Ethanol and fish oil induce NFkappaB trans-activation of the collagen
alpha2(I) promoter through lipid peroxidation-driven activation of the PKC-
PI3K-Akt pathway. Hepatology 2007;45:1433–45.
4. Nieto N. Oxidative-stress and IL-6 mediate the fibrogenic effects of rodent
Kupffer cells on stellate cells. Hepatology 2006;44:1487–501.
5. Nieto N, Cederbaum AI. S-adenosylmethionine blocks collagen I production by
preventing transforming growth factor-beta induction of the COL1A2 promoter.
J Biol Chem 2005;280:30963–74.
6. Nieto N, Cederbaum AI. Increased Sp1-dependent trans-activation of the
LAMgamma 1 promoter in hepatic stellate cells co-cultured with HepG2 cells
overexpressing cytochrome P450 2E1. J Biol Chem 2003;278:15360–72.
780 Urtasun et al
7. Nieto N, Dominguez-Rosales JA, Fontana L, et al. Rat hepatic stellate cells con-
tribute to the acute-phase response with increased expression of alpha1(I) and
alpha1(IV) collagens, tissue inhibitor of metalloproteinase-1, and matrix-metallo-
proteinase-2 messenger RNAs. Hepatology 2001;33:597–607.
8. Nieto N, Friedman SL, Cederbaum AI. Stimulation and proliferation of primary rat
hepatic stellate cells by cytochrome P450 2E1-derived reactive oxygen species.
Hepatology 2002;35:62–73.
9. Nieto N, Friedman SL, Cederbaum AI. Cytochrome P450 2E1-derived reactive
oxygen species mediate paracrine stimulation of collagen I protein synthesis
by hepatic stellate cells. J Biol Chem 2002;277:9853–64.
10. Nieto N, Friedman SL, Greenwel P, et al. CYP2E1-mediated oxidative stress
induces collagen type I expression in rat hepatic stellate cells. Hepatology
1999;30:987–96.
11. Nieto N, Greenwel P, Friedman SL, et al. Ethanol and arachidonic acid increase
alpha 2(I) collagen expression in rat hepatic stellate cells overexpressing cyto-
chrome P450 2E1. Role of H2O2 and cyclooxygenase-2. J Biol Chem 2000;275:
20136–45.
12. Anania FA, Womack L, Potter JJ, et al. Acetaldehyde enhances murine alpha2(I)
collagen promoter activity by Ca21-independent protein kinase C activation in
cultured rat hepatic stellate cells. Alcohol Clin Exp Res 1999;23:279–84.
13. Saxena NK, Saliba G, Floyd JJ, et al. Leptin induces increased alpha2(I) colla-
gen gene expression in cultured rat hepatic stellate cells. J Cell Biochem 2003;
89:311–20.
14. Wu J, Zern MA. Hepatic stellate cells: a target for the treatment of liver fibrosis.
15. Rippe RA, Brenner DA. From quiescence to activation: gene regulation in
hepatic stellate cells. Gastroenterology 2004;127:1260–2.
16. Safadi R, Friedman SL. Hepatic fibrosis–role of hepatic stellate cell activation.
MedGenMed 2002;4:27.
17. Gressner AM. Cytokines and cellular crosstalk involved in the activation of fat-
storing cells. J Hepatol 1995;22:28–36.
18. Balasubramaniyan V, Shukla R, Murugaiyan G, et al. Mouse recombinant leptin
protects human hepatoma HepG2 against apoptosis, TNF-alpha response and
oxidative stress induced by the hepatotoxin-ethanol. Biochim Biophys Acta
2007;1770:1136–44.
19. Wu J, Danielsson A, Zern MA. Toxicity of hepatotoxins: new insights into mech-
anisms and therapy. Expert Opin Investig Drugs 1999;8:585–607.
20. Wu D, Zhai Q, Shi X. Alcohol-induced oxidative stress and cell responses.
J Gastroenterol Hepatol 2006;21(Suppl 3):S26–9.
21. Wang AL, Wang JP, Wang H, et al. A dual effect of N-acetylcysteine on acute
ethanol-induced liver damage in mice. Hepatol Res 2006;34:199–206.
22. Videla LA, Fernandez V, Tapia G, et al. Oxidative stress-mediated hepatotoxicity
of iron and copper: role of Kupffer cells. Biometals 2003;16:103–11.
23. Perez MJ, Velasco E, Monte MJ, et al. Maternal ethanol consumption during
pregnancy enhances bile acid-induced oxidative stress and apoptosis in fetal
rat liver. Toxicology 2006;225:183–94.
24. Kunkel SL, Chensue SW, Phan SH. Prostaglandins as endogenous mediators of
interleukin 1 production. J Immunol 1986;136:186–92.
25. Funaki N, Arii S, Monden K, et al. Chemical mediators released from hepatic
macrophages in primary culture–basic characteristics of human hepatic macro-
phages and changes in liver cirrhosis. J Surg Res 1993;54:222–9.
26. Kamimura S, Gaal K, Britton RS, et al. Increased 4-hydroxynonenal levels in

experimental alcoholic liver disease: association of lipid peroxidation with liver
fibrogenesis. Hepatology 1992;16:448–53.
27. Rojkind M, Dominguez-Rosales JA, Nieto N, et al. Role of hydrogen peroxide
and oxidative stress in healing responses. Cell Mol Life Sci 2002;59:
1872–91.
28. Murrell DF. A radical proposal for the pathogenesis of scleroderma. J Am Acad
Dermatol 1993;28:78–85.
29. Svegliati-Baroni G, Inagaki Y, Rincon-Sanchez AR, et al. Early response of
alpha2(I) collagen to acetaldehyde in human hepatic stellate cells is TGF-beta
independent. Hepatology 2005;42:343–52.
30. Friedman SL. Cytokines and fibrogenesis. Semin Liver Dis 1999;19:129–40.
31. Ekstrom G, Ingelman-Sundberg M. Rat liver microsomal NADPH-supported
oxidase activity and lipid peroxidation dependent on ethanol-inducible
cytochrome P-450 (P-450IIE1). Biochem Pharmacol 1989;38:1313–9.
32. Nieto N, Rojkind M. Repeated whiskey binges promote liver injury in rats fed
a choline-deficient diet. J Hepatol 2007;46:330–9.
33. Guimaraes EL, Franceschi MF, Grivicich I, et al. Relationship between oxidative
stress levels and activation state on a hepatic stellate cell line. Liver Int 2006;26:
477–85.
34. Parola M, Robino G. Oxidative stress-related molecules and liver fibrosis.
J Hepatol 2001;35:297–306.
35. Svegliati-Baroni G, Ridolfi F, Di Sario A, et al. Intracellular signaling pathways
involved in acetaldehyde-induced collagen and fibronectin gene expression in
human hepatic stellate cells. Hepatology 2001;33:1130–40.
36. De Bleser PJ, Niki T, Rogiers V, et al. Transforming growth factor-beta gene
expression in normal and fibrotic rat liver. J Hepatol 1997;26:886–93.
37. Poli G, Parola M. Oxidative damage and fibrogenesis. Free Radic Biol Med
1997;22:287–305.
38. Annoni G, Weiner FR, Zern MA. Increased transforming growth factor-beta
1 gene expression in human liver disease. J Hepatol 1992;14:259–64.
39. Gressner AM, Bachem MG. Molecular mechanisms of liver fibrogenesis–
a homage to the role of activated fat-storing cells. Digestion 1995;56:335–46.
40. Ross JJ, Tranquillo RT. ECM gene expression correlates with in vitro tissue
growth and development in fibrin gel remodeled by neonatal smooth muscle
cells. Matrix Biol 2003;22:477–90.
41. Shi Y, O’Brien JE Jr, Fard A, et al. Transforming growth factor-beta 1 expression
and myofibroblast formation during arterial repair. Arterioscler Thromb Vasc Biol
1996;16:1298–305.
42. Massague J, Seoane J, Wotton D. Smad transcription factors. Genes Dev 2005;
19:2783–810.
43. Leask A, Abraham DJ. TGF-beta signaling and the fibrotic response. FASEB
J 2004;18:816–27.
44. Leonarduzzi G, Scavazza A, Biasi F, et al. The lipid peroxidation end product
4-hydroxy-2,3-nonenal up-regulates transforming growth factor beta1 expres-
sion in the macrophage lineage: a link between oxidative injury and fibrosclero-
sis. FASEB J 1997;11:851–7.
45. Garcia-Trevijano ER, Iraburu MJ, Fontana L, et al. Transforming growth factor
beta1 induces the expression of alpha1(I) procollagen mRNA by a hydrogen
peroxide-C/EBPbeta-dependent mechanism in rat hepatic stellate cells. Hepa-
tology 1999;29:960–70.
782 Urtasun et al
46. Liu RM, Liu Y, Forman HJ, et al. Glutathione regulates transforming growth
factor-beta-stimulated collagen production in fibroblasts. Am J Physiol Lung
Cell Mol Physiol 2004;286:L121–8.
47. Thannickal VJ, Day RM, Klinz SG, et al. Ras-dependent and -independent
regulation of reactive oxygen species by mitogenic growth factors and TGF-
beta1. FASEB J 2000;14:1741–8.
48. Thannickal VJ, Fanburg BL. Activation of an H2O2-generating NADH oxidase in
human lung fibroblasts by transforming growth factor beta 1. J Biol Chem 1995;
270:30334–8.
49. Yoon YS, Lee JH, Hwang SC, et al. TGF beta1 induces prolonged mitochondrial
ROS generation through decreased complex IV activity with senescent arrest in
Mv1Lu cells. Oncogene 2005;24:1895–903.
50. Verrecchia F, Mauviel A. TGF-beta and TNF-alpha: antagonistic cytokines
controlling type I collagen gene expression. Cell Signal 2004;16:873–80.
51. Abraham DJ, Varga J. Scleroderma: from cell and molecular mechanisms to
disease models. Trends Immunol 2005;26:587–95.
52. Runyan CE, Schnaper HW, Poncelet AC. The phosphatidylinositol 3-kinase/Akt
pathway enhances Smad3-stimulated mesangial cell collagen I expression in
response to transforming growth factor-beta1. J Biol Chem 2004;279:2632–9.
53. Vlahos CJ, Kriauciunas TD, Gleason PE, et al. Platelet-derived growth factor
induces proliferation of hyperplastic human prostatic stromal cells. J Cell
Biochem 1993;52:404–13.
54. Grappone C, Pinzani M, Parola M, et al. Expression of platelet-derived growth
factor in newly formed cholangiocytes during experimental biliary fibrosis in
rats. J Hepatol 1999;31:100–9.
55. Marra F, Choudhury GG, Pinzani M, et al. Regulation of platelet-derived growth
factor secretion and gene expression in human liver fat-storing cells. Gastroen-
terology 1994;107:1110–7.
56. Kinnman N, Hultcrantz R, Barbu V, et al. PDGF-mediated chemoattraction of
hepatic stellate cells by bile duct segments in cholestatic liver injury. Lab Invest
2000;80:697–707.
57. Ruef J, Rao GN, Li F, et al. Induction of rat aortic smooth muscle cell growth by
the lipid peroxidation product 4-hydroxy-2-nonenal. Circulation 1998;97:1071–8.
58. Pinzani M, Gentilini A, Caligiuri A, et al. Transforming growth factor-beta 1
regulates platelet-derived growth factor receptor beta subunit in human liver
fat-storing cells. Hepatology 1995;21:232–9.
59. Lechuga CG, Hernandez-Nazara ZH, Hernandez E, et al. PI3K is involved in
PDGF-beta receptor upregulation post-PDGF-BB treatment in mouse HSC. Am
J Physiol Gastrointest Liver Physiol 2006;291:G1051–61.
60. Adachi M, Osawa Y, Uchinami H, et al. The forkhead transcription factor FoxO1
regulates proliferation and transdifferentiation of hepatic stellate cells. Gastroen-
terology 2007;132:1434–46.
61. Marra F, Gentilini A, Pinzani M, et al. Phosphatidylinositol 3-kinase is required for
platelet-derived growth factor’s actions on hepatic stellate cells. Gastroenterol-
ogy 1997;112:1297–306.
62. Galli A, Svegliati-Baroni G, Ceni E, et al. Oxidative stress stimulates proliferation
and invasiveness of hepatic stellate cells via a MMP2-mediated mechanism.
Hepatology 2005;41:1074–84.
63. Pinzani M, Marra F, Carloni V. Signal transduction in hepatic stellate cells. Liver
1998;18:2–13.
64. Marra F, Arrighi MC, Fazi M, et al. Extracellular signal-regulated kinase activation
differentially regulates platelet-derived growth factor’s actions in hepatic stellate
cells, and is induced by in vivo liver injury in the rat. Hepatology 1999;30:951–8.
65. Friedman SL, Arthur MJ. Activation of cultured rat hepatic lipocytes by Kupffer
cell conditioned medium. Direct enhancement of matrix synthesis and stimula-
tion of cell proliferation via induction of platelet-derived growth factor receptors.
J Clin Invest 1989;84:1780–5.
66. Wong L, Yamasaki G, Johnson RJ, et al. Induction of beta-platelet-derived
growth factor receptor in rat hepatic lipocytes during cellular activation in vivo
and in culture. J Clin Invest 1994;94:1563–9.
67. Rombouts K, Lottini B, Caligiuri A, et al. MARCKS is a downstream effector in
platelet-derived growth factor-induced cell motility in activated human hepatic
stellate cells. Exp Cell Res 2008;314(7):1444–54.
68. Caligiuri A, Bertolani C, Guerra CT, et al. Adenosine monophosphate-activated
protein kinase modulates the activated phenotype of hepatic stellate cells.
Hepatology 2008;47:668–76.
69. Failli P, De FR, Caligiuri A, et al. Nitrovasodilators inhibit platelet-derived growth
factor-induced proliferation and migration of activated human hepatic stellate
cells. Gastroenterology 2000;119:479–92.
70. Robino G, Parola M, Marra F, et al. Interaction between 4-hydroxy-2,3-alkenals
and the platelet-derived growth factor-beta receptor. Reduced tyrosine phos-
phorylation and downstream signaling in hepatic stellate cells. J Biol Chem
2000;275:40561–7.
71. Riches DW, Chan ED, Winston BW. TNF-alpha-induced regulation and signalling
in macrophages. Immunobiology 1996;195:477–90.
72. Bach EA, Aguet M, Schreiber RD. The IFN gamma receptor: a paradigm for
cytokine receptor signaling. Annu Rev Immunol 1997;15:563–91.
73. Sivakumar P, Gupta S, Sarkar S, et al. Upregulation of lysyl oxidase and MMPs
during cardiac remodeling in human dilated cardiomyopathy. Mol Cell Biochem
2008;307:159–67.
74. Zhang XH, Sun HM, Yuan JQ. Extracellular matrix production of lens epithelial
cells. J Cataract Refract Surg 2001;27:1303–9.
75. Li YY, Feng YQ, Kadokami T, et al. Myocardial extracellular matrix remodeling in
transgenic mice overexpressing tumor necrosis factor alpha can be modulated
by anti-tumor necrosis factor alpha therapy. Proc Natl Acad Sci U S A 2000;97:
12746–51.
76. Greenwel P, Tanaka S, Penkov D, et al. Tumor necrosis factor alpha inhibits type I
collagen synthesis through repressive CCAAT/enhancer-binding proteins. Mol
Cell Biol 2000;20:912–8.
77. Czaja MJ, Weiner FR, Flanders KC, et al. In vitro and in vivo association of trans-
forming growth factor-beta 1 with hepatic fibrosis. J Cell Biol 1989;108:2477–82.
78. Son G, Iimuro Y, Seki E, et al. Selective inactivation of NF-kappaB in the liver
using NF-kappaB decoy suppresses CCl4-induced liver injury and fibrosis.
Am J Physiol Gastrointest Liver Physiol 2007;293:G631–9.
79. Xu Y, Wang L, Buttice G, et al. Major histocompatibility class II transactivator
(CIITA) mediates repression of collagen (COL1A2) transcription by interferon
gamma (IFN-gamma). J Biol Chem 2004;279:41319–32.
80. Ghosh AK, Bhattacharyya S, Mori Y, et al. Inhibition of collagen gene expression
by interferon-gamma: novel role of the CCAAT/enhancer binding protein beta
(C/EBPbeta). J Cell Physiol 2006;207:251–60.
784 Urtasun et al
81. Higashi K, Inagaki Y, Fujimori K, et al. Interferon-gamma interferes with trans-

forming growth factor-beta signaling through direct interaction of YB-1 with
Smad3. J Biol Chem 2003;278:43470–9.
82. Feinberg MW, Cao Z, Wara AK, et al. Kruppel-like factor 4 is a mediator of proin-
flammatory signaling in macrophages. J Biol Chem 2005;280:38247–58.
83. Varela-Rey M, Montiel-Duarte C, Oses-Prieto JA, et al. p38 MAPK mediates the
regulation of alpha1(I) procollagen mRNA levels by TNF-alpha and TGF-beta in
a cell line of rat hepatic stellate cells(1). FEBS Lett 2002;528:133–8.
84. Tanaka S, Ramirez F. The first intron of the human alpha2(I) collagen gene
(COL1A2) contains a novel interferon-gamma responsive element. Matrix Biol
2007;26:185–9.
85. Ramirez F, Tanaka S, Bou-Gharios G. Transcriptional regulation of the human
alpha2(I) collagen gene (COL1A2), an informative model system to study fibrotic
diseases. Matrix Biol 2006;25:365–72.
86. Kahari VM, Chen YQ, Su MW, et al. Tumor necrosis factor-alpha and interferon-
gamma suppress the activation of human type I collagen gene expression by
transforming growth factor-beta 1. Evidence for two distinct mechanisms of in-
hibition at the transcriptional and posttranscriptional levels. J Clin Invest 1990;
86:1489–95.
87. Iraburu MJ, Dominguez-Rosales JA, Fontana L, et al. Tumor necrosis factor
alpha down-regulates expression of the alpha1(I) collagen gene in rat hepatic
stellate cells through a p20C/EBPbeta- and C/EBPdelta-dependent mechanism.
Hepatology 2000;31:1086–93.
88. Dalgleish R, Trapnell BC, Crystal RG, et al. Copy number of a human type I
alpha 2 collagen gene. J Biol Chem 1982;257:13816–22.
89. Huerre C, Junien C, Weil D, et al. Human type I procollagen genes are located
on different chromosomes. Proc Natl Acad Sci U S A 1982;79:6627–30.
90. Hata R. Transfection of normal human skin fibroblasts with human alpha 1(I) and
alpha 2(I) collagen gene constructs and evidence for their coordinate expres-
sion. Cell Biol Int 1995;19:735–41.
91. Moro L, Smith BD. Identification of collagen alpha1(I) trimer and normal type I
collagen in a polyoma virus-induced mouse tumor. Arch Biochem Biophys
1977;182:33–41.
92. Gan YB. [Expression of type I and type III procollagen genes in human scirrhous
carcinoma of breast by in situ hybridization]. Zhonghua Zhong Liu Za Zhi 1992;
14:348–50 [in Chinese].
93. Colige A, Li SW, Sieron AL, et al. cDNA cloning and expression of bovine procol-
lagen I N-proteinase: a new member of the superfamily of zinc-metalloprotei-
nases with binding sites for cells and other matrix components. Proc Natl
Acad Sci U S A 1997;94:2374–9.
94. Myllyharju J, Kivirikko KI. Collagens and collagen-related diseases. Ann Med
2001;33:7–21.
95. Buttner C, Skupin A, Rieber EP. Transcriptional activation of the type I collagen
genes COL1A1 and COL1A2 in fibroblasts by interleukin-4: analysis of the func-
tional collagen promoter sequences. J Cell Physiol 2004;198:248–58.
96. Greenwel P, Iraburu MJ, Reyes-Romero M, et al. Induction of an acute phase
response in rats stimulates the expression of alpha 1(I) procollagen messenger
ribonucleic acid in their livers. Possible role of interleukin-6. Lab Invest 1995;72:
83–91.
97. Greenwel P, Rubin J, Schwartz M, et al. Liver fat-storing cell clones obtained
from a CCl4-cirrhotic rat are heterogeneous with regard to proliferation,
expression of extracellular matrix components, interleukin-6, and connexin 43.

Lab Invest 1993;69:210–6.
98. Solis-Herruzo JA, Brenner DA, Chojkier M. Tumor necrosis factor alpha inhibits
collagen gene transcription and collagen synthesis in cultured human fibro-
blasts. J Biol Chem 1988;263:5841–5.
99. Armendariz-Borunda J, Katayama K, Seyer JM. Transcriptional mechanisms of
type I collagen gene expression are differentially regulated by interleukin-1
beta, tumor necrosis factor alpha, and transforming growth factor beta in Ito
cells. J Biol Chem 1992;267:14316–21.
100. Greenwel P, Schwartz M, Rosas M, et al. Characterization of fat-storing cell lines
derived from normal and CCl4-cirrhotic livers. Differences in the production of
interleukin-6. Lab Invest 1991;65:644–53.
101. Greenwel P, Dominguez-Rosales JA, Mavi G, et al. Hydrogen peroxide: a link
between acetaldehyde-elicited alpha1(I) collagen gene up-regulation and
oxidative stress in mouse hepatic stellate cells. Hepatology 2000;31:109–16.
102. Boast S, Su MW, Ramirez F, et al. Functional analysis of cis-acting DNA
sequences controlling transcription of the human type I collagen genes. J Biol
Chem 1990;265:13351–6.
103. Florin L, Knebel J, Zigrino P, et al. Delayed wound healing and epidermal hyper-
proliferation in mice lacking JunB in the skin. J Invest Dermatol 2006;126:
902–11.
104. Antoniv TT, De Val S, Wells D, et al. Characterization of an evolutionarily
conserved far-upstream enhancer in the human alpha 2(I) collagen (COL1A2)
gene. J Biol Chem 2001;276:21754–64.
105. Antoniv TT, Tanaka S, Sudan B, et al. Identification of a repressor in the first
intron of the human alpha2(I) collagen gene (COL1A2). J Biol Chem 2005;
280:35417–23.
106. Inagaki Y, Truter S, Ramirez F. Transforming growth factor-beta stimulates alpha
2(I) collagen gene expression through a cis-acting element that contains an
Sp1-binding site. J Biol Chem 1994;269:14828–34.
107. Dennler S, Itoh S, Vivien D, et al. Direct binding of Smad3 and Smad4 to critical
TGF beta-inducible elements in the promoter of human plasminogen activator
inhibitor-type 1 gene. EMBO J 1998;17:3091–100.
108. Poncelet AC, Schnaper HW. Sp1 and Smad proteins cooperate to mediate
transforming growth factor-beta 1-induced alpha 2(I) collagen expression in
human glomerular mesangial cells. J Biol Chem 2001;276:6983–92.
109. Verrecchia F, Rossert J, Mauviel A. Blocking sp1 transcription factor broadly
inhibits extracellular matrix gene expression in vitro and in vivo: implications
for the treatment of tissue fibrosis. J Invest Dermatol 2001;116:755–63.
110. Greenwel P, Inagaki Y, Hu W, et al. Sp1 is required for the early response of
alpha2(I) collagen to transforming growth factor-beta1. J Biol Chem 1997;272:
19738–45.
111. Chen SJ, Yuan W, Mori Y, et al. Stimulation of type I collagen transcription in hu-
man skin fibroblasts by TGF-beta: involvement of Smad 3. J Invest Dermatol
1999;112:49–57.
112. Zhang W, Ou J, Inagaki Y, et al. Synergistic cooperation between Sp1 and
Smad3/Smad4 mediates transforming growth factor beta1 stimulation of alpha
2(I)-collagen (COL1A2) transcription. J Biol Chem 2000;275:39237–45.
113. Ihn H, Ohnishi K, Tamaki T, et al. Transcriptional regulation of the human
alpha2(I) collagen gene. Combined action of upstream stimulatory and
inhibitory cis-acting elements. J Biol Chem 1996;271:26717–23.
786 Urtasun et al
114. Tanaka S, Antoniv TT, Liu K, et al. Cooperativity between far upstream enhancer
and proximal promoter elements of the human {alpha}2(I) collagen (COL1A2)
gene instructs tissue specificity in transgenic mice. J Biol Chem 2004;279:
56024–31.
115. Czuwara-Ladykowska J, Shirasaki F, Jackers P, et al. Fli-1 inhibits collagen type I
production in dermal fibroblasts via an Sp1-dependent pathway. J Biol Chem
2001;276:20839–48.
116. Czuwara-Ladykowska J, Sementchenko VI, Watson DK, et al. Ets1 is an effector
of the transforming growth factor beta (TGF-beta) signaling pathway and an
antagonist of the profibrotic effects of TGF-beta. J Biol Chem 2002;277:
20399–408.
117. Asano Y, Czuwara J, Trojanowska M. Transforming growth factor-beta regulates
DNA binding activity of transcription factor Fli1 by p300/CREB-binding protein-
associated factor-dependent acetylation. J Biol Chem 2007;282:34672–83.
118. Kamada Y, Tamura S, Kiso S, et al. Enhanced carbon tetrachloride-induced liver
fibrosis in mice lacking adiponectin. Gastroenterology 2003;125:1796–807.
119. Wang Y, Xu A, Knight C, et al. Hydroxylation and glycosylation of the four
conserved lysine residues in the collagenous domain of adiponectin. Potential
role in the modulation of its insulin-sensitizing activity. J Biol Chem 2002;277:
19521–9.
120. Honda H, Ikejima K, Hirose M, et al. Leptin is required for fibrogenic responses
induced by thioacetamide in the murine liver. Hepatology 2002;36:12–21.
121. Potter JJ, Rennie-Tankesley L, Mezey E. Influence of leptin in the development of
hepatic fibrosis produced in mice by Schistosoma mansoni infection and by
chronic carbon tetrachloride administration. J Hepatol 2003;38:281–8.
122. Saxena NK, Ikeda K, Rockey DC, et al. Leptin in hepatic fibrosis: evidence for
increased collagen production in stellate cells and lean littermates of ob/ob
mice. Hepatology 2002;35:762–71.
123. Anania FA, Potter JJ, Rennie-Tankersley L, et al. Activation by acetaldehyde of
the promoter of the mouse alpha2(I) collagen gene when transfected into rat
activated stellate cells. Arch Biochem Biophys 1996;331:187–93.
124. Casini A, Cunningham M, Rojkind M, et al. Acetaldehyde increases procollagen
type I and fibronectin gene transcription in cultured rat fat-storing cells through
a protein synthesis-dependent mechanism. Hepatology 1991;13:758–65.
125. Moshage H, Casini A, Lieber CS. Acetaldehyde selectively stimulates collagen
production in cultured rat liver fat-storing cells but not in hepatocytes. Hepatol-
ogy 1990;12:511–8.
126. Chen A. Acetaldehyde stimulates the activation of latent transforming growth
factor-beta1 and induces expression of the type II receptor of the cytokine in
rat cultured hepatic stellate cells. Biochem J 2002;368:683–93.
127. Svegliati-Baroni G, Saccomanno S, van Goor H, et al. Involvement of reactive
oxygen species and nitric oxide radicals in activation and proliferation of rat he-
patic stellate cells. Liver 2001;21:1–12.
128. Pannu J, Nakerakanti S, Smith E, et al. Transforming growth factor-beta receptor
type I-dependent fibrogenic gene program is mediated via activation of Smad1
and ERK1/2 pathways. J Biol Chem 2007;282:10405–13.
129. Arthur MJ. Fibrosis and altered matrix degradation. Digestion 1998;59:376–80.
130. Arthur MJ, Mann DA, Iredale JP. Tissue inhibitors of metalloproteinases,
hepatic stellate cells and liver fibrosis. J Gastroenterol Hepatol 1998;
13(Suppl):S33–8.
131. Milani S, Herbst H, Schuppan D, et al. Differential expression of matrix-metallo-

proteinase-1 and -2 genes in normal and fibrotic human liver. Am J Pathol 1994;
144:528–37.
132. Lechuga CG, Hernandez-Nazara ZH, Dominguez Rosales JA, et al. TGF-{beta}1
modulates matrix metalloproteinase-13 expression in hepatic stellate cells by
complex mechanisms involving p38MAPK, PI3K, AKT and p70S6K. Am J Phys-
iol Gastrointest Liver Physiol 2004;287(5):G974–87.
133. Wang JC. Importance of plasma matrix metalloproteinases (MMP) and tissue
inhibitors of metalloproteinase (TIMP) in development of fibrosis in agnogenic
myeloid metaplasia. Leuk Lymphoma 2005;46:1261–8.
134. Takahara T, Furui K, Funaki J, et al. Increased expression of matrix metallopro-
teinase-II in experimental liver fibrosis in rats. Hepatology 1995;21:787–95.
135. Takahara T, Furui K, Yata Y, et al. Dual expression of matrix metalloproteinase-2
and membrane-type 1-matrix metalloproteinase in fibrotic human livers. Hepa-
tology 1997;26:1521–9.
136. Burggraf D, Martens HK, Dichgans M, et al. Matrix metalloproteinase (MMP)
induction and inhibition at different doses of recombinant tissue plasminogen
activator following experimental stroke. Thromb Haemost 2007;98:963–9.
137. Okazaki I, Maruyama K. Collagenase activity in experimental hepatic fibrosis.
Nature 1974;252:49–50.
138. Giannelli G, Bergamini C, Marinosci F, et al. Antifibrogenic effect of IFN-alpha2b
on hepatic stellate cell activation by human hepatocytes. J Interferon Cytokine
Res 2006;26:301–8.
139. Iredale JP. Tissue inhibitors of metalloproteinases in liver fibrosis. Int J Biochem
Cell Biol 1997;29:43–54.
140. Benyon RC, Iredale JP, Goddard S, et al. Expression of tissue inhibitor of metal-
loproteinases 1 and 2 is increased in fibrotic human liver. Gastroenterology
1996;110:821–31.
141. Roeb E, Graeve L, Hoffmann R, et al. Regulation of tissue inhibitor of metallopro-
teinases-1 gene expression by cytokines and dexamethasone in rat hepatocyte
primary cultures. Hepatology 1993;18:1437–42.
142. Nie QH, Zhang YF, Xie YM, et al. Correlation between TIMP-1 expression and
liver fibrosis in two rat liver fibrosis models. World J Gastroenterol 2006;12:
3044–9.
143. Hsieh CS, Chuang JH, Huang CC, et al. Evaluation of matrix metalloproteinases
and their endogenous tissue inhibitors in biliary atresia-associated liver fibrosis.
J Pediatr Surg 2005;40:1568–73.
144. Herbst H, Wege T, Milani S, et al. Tissue inhibitor of metalloproteinase-1 and -2
RNA expression in rat and human liver fibrosis. Am J Pathol 1997;150:1647–59.
145. Werb Z, Banda MJ, Jones PA. Degradation of connective tissue matrices by
macrophages. I. Proteolysis of elastin, glycoproteins, and collagen by protein-
ases isolated from macrophages. J Exp Med 1980;152:1340–57.
146. He CS, Wilhelm SM, Pentland AP, et al. Tissue cooperation in a proteolytic cas-
cade activating human interstitial collagenase. Proc Natl Acad Sci U S A 1989;
86:2632–6.
147. Murphy G, Stanton H, Cowell S, et al. Mechanisms for pro matrix metalloprotei-
nase activation. APMIS 1999;107:38–44.
148. Zhang Z, Shen HM, Zhang QF, et al. Critical role of GSH in silica-induced oxida-
tive stress, cytotoxicity, and genotoxicity in alveolar macrophages. Am J Physiol
1999;277:L743–8.
788 Urtasun et al
149. Eitzman DT, Krauss JC, Shen T, et al. Ginsburg. Lack of plasminogen activator
inhibitor-1 effect in a transgenic mouse model of metastatic melanoma. Blood
1996;87:4718–22.
150. Vayalil PK, Olman M, Murphy-Ullrich JE, et al. Glutathione restores collagen
degradation in TGF-beta-treated fibroblasts by blocking plasminogen activator
inhibitor-1 expression and activating plasminogen. Am J Physiol Lung Cell Mol
Physiol 2005;289:L937–45.
151. Ferroni P, Guagnano MT, Manigrasso MR, et al. Increased plasminogen activa-
tor inhibitor-1 levels in android obesity: correlation with oxidative stress.
J Thromb Haemost 2005;3:1086–7.
152. Jiang Z, Seo JY, Ha H, et al. Reactive oxygen species mediate TGF-beta1-
induced plasminogen activator inhibitor-1 upregulation in mesangial cells.
Biochem Biophys Res Commun 2003;309:961–6.
153. Lee EA, Seo JY, Jiang Z, et al. Reactive oxygen species mediate high glucose-
induced plasminogen activator inhibitor-1 up-regulation in mesangial cells and
in diabetic kidney. Kidney Int 2005;67:1762–71.
154. Swiatkowska M, Szemraj J, Al-Nedawi KN, et al. Reactive oxygen species upre-
gulate expression of PAI-1 in endothelial cells. Cell Mol Biol Lett 2002;7:
1065–71.
155. Feder L, Todaro JA, Laskin DL. Characterization of interleukin-1 and interleukin-6
production by hepatic endothelial cells and macrophages. J Leukoc Biol 1993;
53:126–32.
156. Beckman JS, Koppenol WH. Nitric oxide, superoxide, and peroxynitrite: the
good, the bad, and ugly. Am J Physiol 1996;271:C1424–37.
157. Ohhashi T, Mizuno R, Ikomi F, et al. Current topics of physiology and pharmacol-
ogy in the lymphatic system. Pharmacol Ther 2005;105:165–88.
158. van der Vliet A, Eiserich JP, Shigenaga MK, et al. Reactive nitrogen species and
tyrosine nitration in the respiratory tract: epiphenomena or a pathobiologic
mechanism of disease? Am J Respir Crit Care Med 1999;160:1–9.
159. Beckman JS, Chen J, Ischiropoulos H, et al. Oxidative chemistry of peroxynitrite.
Methods Enzymol 1994;233:229–40.
160. Eiserich JP, Patel RP, O’Donnell VB. Pathophysiology of nitric oxide and related
species: free radical reactions and modification of biomolecules. Mol Aspects
Med 1998;19:221–357.
161. Huie RE, Padmaja S. The reaction of no with superoxide. Free Radic Res
Commun 1993;18:195–9.
162. Hecker M, Schott C, Bucher B, et al. Increase in serum NG-hydroxy-L-arginine
in rats treated with bacterial lipopolysaccharide. Eur J Pharmacol 1995;275:
R1–3.
163. Laude K, Thuillez C, Richard V. Peroxynitrite triggers a delayed resistance of
coronary endothelial cells against ischemia-reperfusion injury. Am J Physiol
Heart Circ Physiol 2002;283:H1418–23.
164. Bolanos JP, Garcia-Nogales P, Almeida A. Provoking neuroprotection by perox-
ynitrite. Curr Pharm Des 2004;10:867–77.
165. De Keulenaer GW, Alexander RW, Ushio-Fukai M, et al. Tumour necrosis factor
alpha activates a p22phox-based NADPH oxidase in vascular smooth muscle.
Biochem J 1998;329(Pt 3):653–7.
166. Clemens MG. Nitric oxide in liver injury. Hepatology 1999;30:1–5.
167. Pietrangelo A. Iron, oxidative stress and liver fibrogenesis. J Hepatol 1998;
28(Suppl 1):8–13.
168. Rockey DC, Chung JJ. Inducible nitric oxide synthase in rat hepatic lipocytes and
the effect of nitric oxide on lipocyte contractility. J Clin Invest 1995;95:1199–206.
169. Wink DA, Mitchell JB. Chemical biology of nitric oxide: insights into regulatory,
cytotoxic, and cytoprotective mechanisms of nitric oxide. Free Radic Biol Med
1998;25:434–56.
170. Xu Y, Bialik S, Jones BE, et al. NF-kappaB inactivation converts a hepatocyte
cell line TNF-alpha response from proliferation to apoptosis. Am J Physiol
1998;275:C1058–66.
171. Hur GM, Ryu YS, Yun HY, et al. Hepatic ischemia/reperfusion in rats induces
iNOS gene transcription by activation of NF-kappaB. Biochem Biophys Res
Commun 1999;261:917–22.
172. Smith KJ, Kapoor R, Felts PA. Demyelination: the role of reactive oxygen and
nitrogen species. Brain Pathol 1999;9:69–92.
173. Uemura M, Tsujii T, Kikuchi E, et al. Increased plasma levels of substance P and
disturbed water excretion in patients with liver cirrhosis. Scand J Gastroenterol
1998;33:860–6.
174. Carlucci F, Marinello E, Rosi F, et al. Nitric oxide generation is associated with an
unbalance of protein tyrosine phosphatases during liver transplantation. Biomed
Pharmacother 2007;61:216–21.
175. Hanazawa T, Kharitonov SA, Barnes PJ. Increased nitrotyrosine in exhaled
breath condensate of patients with asthma. Am J Respir Crit Care Med 2000;
162:1273–6.
176. Srisook K, Kim C, Cha YN. Cytotoxic and cytoprotective actions of O2- and NO
(ONOO-) are determined both by cellular GSH level and HO activity in macro-
phages. Methods Enzymol 2005;396:414–24.
177. Pesse B, Levrand S, Feihl F, et al. Peroxynitrite activates ERK via Raf-1 and
MEK, independently from EGF receptor and p21Ras in H9C2 cardiomyocytes.
J Mol Cell Cardiol 2005;38:765–75.
178. Rippe RA, Umezawa A, Kimball JP, et al. Binding of upstream stimulatory factor
to an E-box in the 3’-flanking region stimulates alpha1(I) collagen gene
transcription. J Biol Chem 1997;272:1753–60.
179. Friedman SL. Liver fibrosis – from bench to bedside. J Hepatol 2003;
38(Suppl 1):S38–53.
180. Tan AS, Berridge MV. Superoxide produced by activated neutrophils efficiently
reduces the tetrazolium salt, WST-1 to produce a soluble formazan: a simple
colorimetric assay for measuring respiratory burst activation and for screening
anti-inflammatory agents. J Immunol Methods 2000;238:59–68.
181. Rockey D. The cellular pathogenesis of portal hypertension: stellate cell
contractility, endothelin, and nitric oxide. Hepatology 1997;25:2–5.
182. Colombato L. The role of transjugular intrahepatic portosystemic shunt (TIPS) in
the management of portal hypertension. J Clin Gastroenterol 2007;41:S344–51.
183. Bataller R, Gines P, Nicolas JM, et al. Angiotensin II induces contraction and
proliferation of human hepatic stellate cells. Gastroenterology 2000;118:
1149–56.
184. Thirunavukkarasu C, Watkins S, Harvey SA, et al. Superoxide-induced apopto-
sis of activated rat hepatic stellate cells. J Hepatol 2004;41:567–75.
185. Parola M, Pinzani M, Casini A, et al. Stimulation of lipid peroxidation or 4-hydrox-
ynonenal treatment increases procollagen alpha 1 (I) gene expression in human
liver fat-storing cells. Biochem Biophys Res Commun 1993;194:1044–50.
186. Ferrini MG, Vernet D, Magee TR, et al. Antifibrotic role of inducible nitric oxide
synthase. Nitric Oxide 2002;6:283–94.
790 Urtasun et al
187. Schwentker A, Vodovotz Y, Weller R, et al. Nitric oxide and wound repair: role of
cytokines? Nitric Oxide 2002;7:1–10.
188. Witte MH, Borgs P, Way DL, et al. Alcohol, hepatic sinusoidal microcirculation,
and chronic liver disease. Alcohol 1992;9:473–80.
189. Chu AJ, Prasad JK. Up-regulation by human recombinant transforming growth
factor beta-1 of collagen production in cultured dermal fibroblasts is mediated
by the inhibition of nitric oxide signaling. J Am Coll Surg 1999;188:271–80.
190. Kolpakov V, Gordon D, Kulik TJ. Nitric oxide-generating compounds inhibit total
protein and collagen synthesis in cultured vascular smooth muscle cells. Circ
Res 1995;76:305–9.
191. Cao M, Westerhausen-Larson A, Niyibizi C, et al. Nitric oxide inhibits the synthe-
sis of type-II collagen without altering Col2A1 mRNA abundance: prolyl hydrox-
ylase as a possible target. Biochem J 1997;324(Pt 1):305–10.
192. Genovese T, Cuzzocrea S, Di Paola R, et al. Inhibition or knock out of inducible
nitric oxide synthase result in resistance to bleomycin-induced lung injury.
Respir Res 2005;6(58):1–17.
193. Thornton FJ, Schaffer MR, Witte MB, et al. Enhanced collagen accumulation
following direct transfection of the inducible nitric oxide synthase gene in cuta-
neous wounds. Biochem Biophys Res Commun 1998;246:654–9.
194. Trachtman H, Futterweit S, Garg P, et al. Nitric oxide stimulates the activity of
a 72-kDa neutral matrix metalloproteinase in cultured rat mesangial cells.
195. Rai RM, Lee FY, Rosen A, et al. Impaired liver regeneration in inducible nitric
oxide synthasedeficient mice. Proc Natl Acad Sci U S A 1998;95:13829–34.
196. Valente EG, Vernet D, Ferrini MG, et al. L-arginine and phosphodiesterase (PDE)
inhibitors counteract fibrosis in the Peyronie’s fibrotic plaque and related
fibroblast cultures. Nitric Oxide 2003;9:229–44.
197. Urtasun R, Cubero FJ, Vera M, et al. Reactive nitrogen species switch on early ex-
tracellular matrix remodeling via induction of MMP1 and TNFa. Gastroenterology,
in press.
Stellate Cell
Contrac tion : Role,
Re g ulation, a nd
Potential Therap eutic
Target
Russell K. Soon, Jr., BSa, Hal F.Yee, Jr., MD, PhDa,b,*
KEYWORDS
Hepatic stellate cell Contraction Fibrosis Pericyte
Rho-associated kinase Sinusoid
Contractile force generation by hepatic stellate cells is recognized to play a key role in
the liver’s response to injury. This cellular behavior is consequently believed to contrib-
ute to normal healing and the development of hepatic fibrosis. Improved understanding
of stellate cell contraction and its regulation would therefore be predicted to facilitate
development of clinical strategies for the treatment of liver disease. Despite more
than 15 years of study, however, effective therapies based on targeting the generation
of contractile force by stellate cells have remained elusive. This article examines the
current state of knowledge regarding stellate cell contraction, its role, its regulation,
and its potential as a therapeutic target, by addressing three fundamental questions:
What do we believe we know?

What do we really know?
What would we like to know?
WHAT DO WE BELIEVE WE KNOW ABOUT HEPATIC STELLATE CELL CONTRACTION?
Several observations about the human liver have led to the concept that stellate cells
are contractile and that generation of contractile force by these cells mediates hepatic
This work was supported in part by NIH, R01 DK61532, the Technical Training Foundation, and
the William and Mary Ann Rice Memorial Distinguished Professorship.
a
Department of Medicine and Liver Center, University of California, San Francisco, 1001
Potrero Avenue, SFGH Building 40, Room 4102, San Francisco, CA 94110, USA
b
Division of Gastroenterology and Hepatology, San Francisco General Hospital and Trauma
Center, 1001 Potrero Avenue, SFGH Building 40, Room 4102, San Francisco, CA 94110, USA
* Corresponding author. Division of Gastroenterology and Hepatology, San Francisco General
Hospital and Trauma Center, 1001 Potrero Avenue, SFGH Building 40, Room 4102 San Francisco,
CA 94110.
E-mail address: hyee@medsfgh.ucsf.edu (H.F. Yee).
Clin Liver Dis 12 (2008) 791–803

792 Soon & Yee
pathophysiology. Stellate cells express a-smooth muscle actin, a marker of non-

muscle cell contractility, in patients who have various forms of chronic liver injury.1–6
Moreover, stellate cells of normal human livers do not express a-smooth muscle actin,
suggesting that contractility may be induced by liver injury. Human stellate cells
express various receptors for well-known contractile agonists. Studies have demon-
strated stellate cell expression of receptors for endothelin 1, arginine-vasopressin,
and angiotensin II, all of which induce generation of contractile force by contractile
cell types.7–10 The presence of these receptors suggests that stellate cells are capable
of transducing chemical signals into changes in mechanical force. These observations
indicate that stellate cells contain the cellular machinery necessary for generation of
contractile force.
The anatomic location of stellate cells in the normal human liver also suggests an
important role for contraction by these cells in the regulation of sinusoidal blood
flow. Immunohistochemical studies of normal human liver show that stellate cells
reside in the perisinusoidal space and extend elongate protrusions that run along
and encircle one or more sinusoids.3,4,11 This anatomy is similar to that of tissue peri-
cytes, such as the mesangial cells of the kidney, which modulate vascular tone by
contracting around their capillaries.12–14 In the same way, hepatic stellate cells have
been theorized to regulate sinusoidal resistance, and consequently blood flow, by
contracting around sinusoids.
Studies of the injured human liver have demonstrated that stellate cells appear
prominently in fibrotic bands of collagen (scar tissue) remote from their normal loca-
tion.1–5,11,15 This finding raised the possibility that stellate cells may function similar
to cutaneous myofibroblasts that participate in wound healing of the skin by contract-
ing scar tissue as healing and regeneration progresses.16–18 It has thus been proposed
that contractile force generation by stellate cells may permit remodeling of extracellu-
lar matrix during the liver’s injury response.19
In summary, conventional wisdom holds that hepatic stellate cells are contractile,
and their capacity to generate contractile force mediates the liver’s injury response
through modulation of sinusoidal blood flow and scar contracture (Fig. 1). This current
understanding of the role of stellate cell contraction is, however, largely based on
circumstantial evidence and common logic. In the following section we dissect the
scientific methods that have been used to study stellate cell contraction, and discuss
the model (Fig. 2) for the regulation of stellate cell contraction that is best supported by
the evidence.
WHAT DO WE REALLY KNOW ABOUT HEPATIC STELLATE CELL CONTRACTION?
Current understanding of hepatic stellate cell contraction has been advanced largely
through use of in vitro experimental methods. Several methods have been used
to study stellate cell contraction, each with its own inherent limitations in spatial and
temporal resolution, accuracy and precision, and fidelity to what occurs in vivo. To
properly appreciate what we really know about stellate cell contraction it is worthwhile
to appraise the strengths and limitations of the main assays used to study this impor-
tant stellate cell behavior.
An early technique for the study of stellate cell contraction examined the decrease in
two-dimensional cell surface area of cultured cells as a surrogate marker for generation
of contractile force.20–23 In this assay stellate cells in culture were grown to subconflu-
ence on a glass coverslip and visualized with transmission light videomicroscopy.
Reductions in stellate cell surface area could be measured in response to exposure
to various agonists and inhibitors. Although this assay permitted quantitative
Stellate Cell Contraction 793
Sinusoidal Scar
Resistance Contracture
Fig. 1. Role of hepatic stellate cell contraction. Stellate cell contractile force generation is
believed to mediate the liver’s response to injury through constricting sinusoids and con-
tracting scar.
measurements with subcellular spatial resolution and temporal resolution in seconds,

changes in stellate cell surface area were not a direct gauge of contractile force. In-
deed, shrinkage of stellate cell surface area could result from several noncontractile
events, including changes in cellular adhesion or volume, three-dimensional shape
changes, or even focusing artifacts, making it difficult to discern how accurate a mea-
sure of the contractile force generated by stellate cells this method provided.
Hepatic stellate cell contraction has also been evaluated using a model in which
stellate cells in culture are grown in a monolayer on silicone rubber-coated cover-
slips.24–27 In this assay the silicone rubber substrate on which stellate cells were grown
was visualized with transmission light microscopy and wrinkling of the substrate
was examined. This method permitted semiquantitative determination of substrate
ET-1
GPCR
Ca2+ RhoA
Hepatic Stellate Cell
ROK
MLC PO4
MLCK MLCP
MLC PO4
NO NO
Contractile Force Generation
Fig. 2. Regulation of hepatic stellate cell contraction. Soluble factors associated with liver
injury, such as endothelin 1 and nitric oxide, are transduced primarily through Rho signaling
pathways that promote the myosin II–powered generation of contractile force by stellate
cells. Dashed arrows indicate subordinate Ca21 signaling pathway. ET-1, endothelin 1; GPCR,
G protein coupled receptor; MLC, myosin light chain; MLCK, myosin light chain kinase;
MLCP, myosin light chain phosphatase; NO, nitric oxide; RhoA, Rho family GTPase; ROK,
Rho-associated kinase; PO4, phosphate.
794 Soon & Yee
wrinkling, as a surrogate measure of the tension developed by the population of stel-

late cells across the silicone rubber, within seconds of exposure to various agonists
and inhibitors. Increases in substrate wrinkling likely reflect gross changes in stellate
cell contractile force generation, but the tightness of that correlation was unknown.
Although determination of substrate wrinkling was subjective and imprecise, this
assay did permit examination of the formation and loss of wrinkles, putative correlates
of contractile force generation and relaxation, respectively.
Another method for examining stellate cell contraction used a model in which stellate
cells in culture were grown on top of or within gel lattices composed of type I
collagen.27–35 In this assay shrinkage of the collagen gels by the population of stellate
cells was used as a surrogate measure of cellular contraction. After release of the gel
from a supporting culture dish and exposure to different agonists and inhibitors
changes in gel diameter could be measured with a ruler after at least an overnight
incubation. This technique did not permit measurement of relaxation or acute changes
in contractile force development. In addition, reductions in gel area were not revers-
ible. It is uncertain how closely changes in lattice area correlate with alterations in
contractile force generation, and this model may not differentiate between active con-
traction and passive changes in stellate cell tension (eg, passive tension across the
lattice generated by cellular spreading or increases in stellate cell number).
High resolution intravital videomicroscopy of sinusoids within the isolated rat liver
has been used as a model for the study of stellate cell contraction.36–39 In this assay
changes in sinusoidal diameter were visualized adjacent to stellate cell bodies demar-
cated by retinoid autofluorescence as a surrogate measure of contractile force gener-
ation by stellate cells encircling these vessels. These sinusoidal changes could be
observed in real time within minutes of exposure to different agonists and inhibitors.
By finely measuring and controlling input and output pressures across the major
hepatic vessels and estimating sinusoidal blood flow, estimates of increases and de-
creases in sinusoidal resistance could be derived. This technique, however, did not
permit direct measurement of contractile force generation by stellate cells, nor did
it rule out the possibility that other cell types (eg, colocalized endothelial cells or
upstream or downstream vascular smooth muscle) or noncontractile events (eg,
cellular swelling) might contribute to the observed changes in sinusoidal diameter.
The contraction of hepatic stellate cells in culture has been directly and quantita-
tively measured.40–42 In this assay stellate cells were grown within a three-dimensional
type I collagen gel lattice, which was placed in an organ bath and attached to a sensi-
tive force transducer. This model permitted real-time measurement in actual force
units of the contractile tension generated by stellate cells within the gel in response
to various agonists and inhibitors. The method exhibited temporal resolution in the
range of seconds, and it permitted precise quantification of both contraction and
relaxation within the same sample. Although this technique allowed direct measure-
ment of the contractile forces exerted by stellate cells populating a collagen gel,
determination of the contractile force generated by a single stellate cell could only
be estimated.
When evaluating the methods used to study hepatic stellate cell contraction, a crit-
ical consideration is how closely the assay is likely to reflect what actually occurs
within the human liver—in other words, the fidelity of the experimental model. Despite
the diverse experimental models and substantial efforts to better understand stellate
cell contraction important factors limit what we really know about stellate cell contrac-
tion in vivo. Except for intravital microscopy, all of the methods used to study stellate
cell contraction used cells in culture, raising the concern that isolated stellate cells do
not function in an identical fashion to stellate cells in the liver. This concern is
particularly true in studies that used secondary cultures of stellate cells.7,33,41,43 It is

thus generally recognized that the results of experiments performed using stellate
cells in primary culture may have greater clinical relevance. Another point is that
experiments performed on stellate cells in monolayer do not replicate the normal
three-dimensional environment in which stellate cells reside in vivo. Studies of stellate
cells grown within collagen gels more likely replicate the authentic milieu within the
liver. Intravital microscopy of hepatic sinusoids in intact liver, both in vivo and ex
vivo,36–39 are most likely to be relevant to human pathophysiology, however. Despite
the shared and unique limitations of each of the different methods used to study stel-
late cell contraction, together these assays complement each other and have contrib-
uted to a robust understanding of stellate cell contraction.
Several chemicals have been demonstrated to stimulate stellate cell contraction,
including endothelin 1, arginine vasopressin, angiotensin II, thrombin, eicosanoids,
and a1-adrenergic agonists.9,10,20,24,35,40–42 The best-studied and most prominent
agonist for stellate cell contraction is endothelin 1. Circulating levels of this peptide
are elevated in patients who have liver disease7,44,45 and increased in animal models
of liver injury.46,47 Endothelin 1 can induce markers of stellate cell contraction in every
one of the assays discussed earlier.20,25,29,36,40 In particular, the magnitude and speed
of the contractile force generated by stellate cells in response to endothelin 1 has been
predicted to be sufficient to regulate sinusoidal resistance to blood flow.40 Even more
significant, perfusion of isolated rodent livers with endothelin 1 caused a reduction in
sinusoidal diameter colocalized with stellate cells that was paralleled by an increase in
portal pressure.36,48–51 Administration of endothelin 1 receptor antagonists decreased
portal pressure in portal hypertensive rats.52 These experimental findings indicate that
endothelin 1 is a potent agonist of stellate cell contraction and suggest an important
contribution of this mediator to the regulation of hepatic blood flow.
Several agents, including nitric oxide, carbon monoxide, and prostaglandins, may
counteract the effects of contraction-inducing stimuli by causing stellate cell relaxa-
tion.24,25,38,53–55 Nitric oxide production is reduced in the injured liver.56–58 In vitro
studies have suggested that activation of nitric oxide signaling (through nitric oxide
donors or cytokine stimulation of nitric oxide production) causes relaxation in stellate
cells and attenuates agonist-induced contraction,10,25,53,56,59,60 a process that might
occur through cGMP-dependent activation of myosin light chain phosphatase, similar
to what has been demonstrated in smooth muscle cells.61–63 Finally, nitric oxide
donors can attenuate elevations in portal pressure in the perfused rodent liver induced
by endothelin 1 or other contraction-inducing stimuli.36,48,64 These observations have
led to a proposed model in which sinusoidal tone is finely modulated by the net bal-
ance of agents that induce stellate cell relaxation, such as nitric oxide, and agonists
of stellate cell contraction, such as endothelin 1.65–67
It has long been known that the motor protein complex, myosin II, powers contrac-
tile force generation in smooth muscle and fibroblasts through its action on the actin
cytoskeleton.68,69 Numerous studies observed that hepatic stellate cells in culture ex-
press both myosin II31,41,42,70–73 and a fully formed actin cytoskeleton.31,41–43,70–74
Myosin II activation, as assessed by myosin regulatory light chain phosphorylation,
correlates with various surrogate measures of stellate cell contraction31,43,71 and
with the actual contractile force generated by stellate cells.41 Moreover, antagonism
of myosin phosphorylation inhibited contractile force generation by stellate cells.42
Finally, the myosin regulatory light chain expressed by stellate cells is phosphorylated
at serine 19,73 the consensus activation site for myosin II. Taken together these results
indicate that stellate cell contraction is powered by myosin II, which is activated by
phosphorylation of its myosin regulatory light chain.
796 Soon & Yee
Evidence suggests that Ca21 signaling pathways regulate stellate cell contraction
by activating myosin light chain kinase, which selectively phosphorylates the myosin
regulatory light chain,20,75–77 similar to what has been demonstrated in smooth mus-
cle. This notion was supported by several experimental observations. First, ligands,
including endothelin 1, thrombin, and angiotensin II, that induced transient increases
in cytosolic Ca21 concentration also stimulated stellate cell contraction.7,10,20,25,40,41
Second, plasma membrane Ca21 channel expression, Ca21 influx through these
channels, and cytosolic Ca21 concentration each correlated with reductions in stellate
cell surface area.23,60,77 Third, inhibitors of Ca21-dependent myosin light chain kinase
attenuated the shrinkage of collagen gels populated with stellate cells.35,43 Although
these findings suggested an important role for Ca21 signaling in the control of stellate
cell contraction, they did not provide any direct evidence to support this model.
In contrast to previously held views, current data indicate that Ca21 signaling path-
ways play a subordinate role in the regulation of contractile force generation by stellate
cells. The contribution of Ca21 signaling pathways to the regulation of stellate cell con-
traction was directly tested by modulating cytosolic Ca21 and directly measuring the
contractile force generated by this cell type.42 Increases in cytosolic Ca21 induced by
depolarizing the plasma membrane did not provoke contractile force generation.
Superphysiologic elevations in cytosolic Ca21 triggered by a calcium ionophore
induced minimal increases in contractile force. Eliminating increases in cytosolic
Ca21 with a calcium chelator had no effect on endothelin 1–induced contractile force
generation. This study provided surprising evidence indicating that Ca21 signaling is
neither necessary for contractile force generation by stellate cells nor sufficient to
provoke stellate cell contraction. This fresh perspective was supported by the recent
observation that stellate cell contraction was stimulated by the inhibition of myosin
phosphatase despite the absence of any changes in cytosolic Ca21 concentration.35
Over the past decade substantial data have emerged demonstrating that contractile
force generation by certain non-muscle cell types, including fibroblasts and endothe-
lial cells, is predominantly regulated by transduction pathways that signal through the
ras-like GTPase, RhoA, rather than Ca21.69,78–80 Mounting evidence indicates that
Rho signaling pathways also control stellate cell contraction. Stellate cells express
RhoA and Rho-associated kinase.31,70,71 Specific inhibition of RhoA caused derange-
ment of the stellate cell actin cytoskeleton.70,74 Highly selective antagonism of the
RhoA effector protein, Rho-associated kinase, impeded shrinkage of collagen gels
populated with stellate cells,31,43,71 inhibited myosin regulatory light chain phosphor-
ylation,31,41–43,71,73 and blocked contractile force generation by stellate cells.41,42
Attenuation of myosin phosphatase also reduced stellate cell contraction as assessed
by the shrinkage of stellate cell–populated collagen gels.35 In combination with studies
of the contribution of Ca21 signaling, these studies support a model in which contrac-
tile force generation by stellate cells is mediated primarily by Rho signal transduction
pathways.
With regard to the putative roles that stellate cell contraction may play in the path-
ophysiology of the liver, the strongest data pertain to their contribution to the modu-
lation of sinusoidal blood flow. The concept that stellate cells modulate resistance
to hepatic blood flow by contracting around sinusoids is supported by several obser-
vations. First, stellate cells in situ exhibit a pericyte-like morphology with protrusions
encircling the sinusoids.81–83 Second, the number and spacing of stellate cells and
their characteristic protrusions overlay the entire sinusoidal network.84 Third, ex vivo
perfusion of the liver with endothelin 1–induced reductions in sinusoidal caliber colo-
calized with stellate cells.36,37,85 Fourth, direct measurement of contractile force gen-
eration by stellate cells within collagen gels suggests that the magnitude and rate of
stellate cell contraction and relaxation are capable of modulating blood flow by way of
sinusoidal constriction.40 Taken together these findings obtained from several com-
plementary methods indicate that stellate cells contribute to the regulation of sinusoi-
dal blood flow.
WHAT WOULD WE LIKE TO KNOW ABOUT HEPATIC STELLATE CELL CONTRACTION?
What we would really like to know is how understanding the emerging pathobiology of
stellate cell contraction can be used to develop new strategies for the prevention and
treatment of hepatic fibrosis in humans. Despite 15 years of intensive investigation and
a great deal of new information about the role and regulation of stellate cell contrac-
tion, no effective stellate cell contraction–targeted therapies for hepatic fibrosis
have been validated. In fact, no fibrosis-directed treatments of any sort have yet
been developed for the treatment of chronic liver disease.86–89 As discussed in other
articles in this issue, the only proven therapies for fibrosis so far are directed at the pre-
vention or removal of a specific cause of chronic hepatic injury, such as treatment of
hepatitis C or biliary obstruction.
Two logical approaches for developing new treatments for hepatic fibrosis are (1) to
destroy stellate cells or disable their function, and (2) to modulate specific molecular
targets within key signal transduction pathways used by stellate cells. There are, how-
ever, serious real and theoretic challenges to these general therapeutic approaches.
Stellate cells mediate the response of the liver to acute and chronic injury. They are
thus believed to contribute to the normal wound-healing process and to the develop-
ment of hepatic fibrosis and subsequent cirrhosis. If this is accurate, then destruction
or disabling of stellate cells could impair healthy and essential responses of the liver to
injury in addition to the anticipated prevention or attenuation of fibrosis. One solution
to the paradox that the presence of intact stellate cells may be necessary for both nor-
mal wound healing and fibrogenesis might be to selectively target influential signaling
pathways used by stellate cells. The problem with this therapeutic strategy is that im-
portant signaling pathways are generally shared by different cell types. For example,
use of a mitogen-activated protein kinase antagonist to inhibit stellate cell proliferation
would also be predicted to influence the proliferation of hepatocytes and numerous
other cell types. It would therefore be especially advantageous to identify and develop
treatment strategies precisely and specifically targeted to stellate cell behaviors that
mediate hepatic fibrosis.
The contractile force exerted by stellate cells contributes to the regulation of sinu-
soidal blood flow and the development of fibrosis. As discussed, emerging evidence
indicates that generation of contractile force by stellate cells may be differentially reg-
ulated by transduction pathways that signal through Rho and Rho-associated kinase
rather than Ca21 and myosin light chain kinase as it is in vascular smooth muscle. This
differential regulation of stellate cell contraction offers the possibility that novel thera-
peutic strategies could be developed that selectively target the generation of contrac-
tile force by stellate cells. In fact, commercially available highly selective small
molecule inhibitors of Rho-associated kinase attenuate the increases in intrahepatic
vascular resistance and portal hypertension35,71,90,91 and lessen the development of
hepatic fibrosis91–94 in diverse rodent models of liver injury. These studies provide
proof of principle that stellate cell contraction can be selectively targeted to treat
hepatic fibrosis, at least in rodent models of chronic liver injury.
A significant concern with targeting Rho signal transduction pathways as a strategy
for treating hepatic fibrosis is that this pathway is ubiquitous in playing vital roles in
diverse cell types throughout the body. This concern could be circumvented by
798 Soon & Yee
delivering inhibitors directly to the liver or stellate cells. Possible methods for directed
delivery include portal or hepatic venous injection, coupling drugs to carriers (eg, anti-
bodies, peptides, lectins, or lipids) with affinity for the liver or stellate cells, and the use
of particular viruses to selectively deliver therapeutic genes or ribonucleic acids71,95–99
to stellate cells or the liver. By integrating new technologies for liver-directed delivery
with pharmaceutical or genetic agents that selectively target stellate cell contraction it
may be possible to develop effective strategies for the prevention and treatment of
hepatic fibrosis.
SUMMARY
The contractile force generated by stellate cells within the liver may contribute to the
development of hepatic fibrosis by modulating sinusoidal blood flow and participating
in extracellular matrix remodeling. For more than 15 years, the role and regulation of
stellate cell contraction have been areas of substantial research. Diverse but comple-
mentary experimental methods have been used to elucidate the pathophysiology of
stellate cell contraction. Although each technique for studying the contraction of stel-
late cells has its own limitations, taken together the published studies have provided
a robust model for the regulation of stellate cell contractile force generation. In this
model, soluble factors associated with liver injury, including endothelin 1 and nitric
oxide, are transduced primarily through Rho signaling pathways that promote the
myosin II–powered generation of contractile force by stellate cells. Moreover, compel-
ling data support a role for stellate cells in the control of hepatic blood flow by con-
tracting around sinusoids. Our enhanced understanding of the role and differential
regulation of stellate cell contraction may facilitate the discovery of new and targeted
strategies for the prevention and treatment of hepatic fibrosis.
REFERENCES
1. Schmitt-Graff A, Kruger S, Bochard F, et al. Modulation of alpha smooth muscle

actin and desmin expression in perisinusoidal cells of normal and diseased
human livers. Am J Pathol 1991;138(5):1233–42.
2. Nouchi T, Tanaka Y, Tsukada T, et al. Appearance of alpha-smooth-muscle-actin-
positive cells in hepatic fibrosis. Liver 1991;11(2):100–15.
3. Yamaoka K, Nouchi T, Marumo F, et al. Alpha-smooth-muscle actin expression in
normal and fibrotic human livers. Dig Dis Sci 1993;38(8):1473–9.
4. Enzan H, Himeno H, Iwamura S, et al. Immunohistochemical identification of Ito
cells and their myofibroblastic transformation in adult human liver. Virchows
Arch 1994;424(3):249–56.
5. Enzan H, Himeno H, Iwamura S, et al. Sequential changes in human Ito cells and
their relation to postnecrotic liver fibrosis in massive and submassive hepatic
necrosis. Virchows Arch 1995;426(1):95–101.
6. Baroni GS, Pastorelli A, Manzin A, et al. Hepatic stellate cell activation and liver
fibrosis are associated with necroinflammatory injury and Th1-like response in
chronic hepatitis C. Liver 1999;19(3):212–9.
7. Pinzani M, Milani S, De Franco R, et al. Endothelin 1 is overexpressed in human
cirrhotic liver and exerts multiple effects on activated hepatic stellate cells.
8. Yokomori H, Oda M, Yasogawa Y, et al. Enhanced expression of endothelin B
receptor at protein and gene levels in human cirrhotic liver. Am J Pathol 2001;
159(4):1353–62.
9. Bataller R, Nicolas JM, Gines P, et al. Arginine vasopressin induces contraction

and stimulates growth of cultured human hepatic stellate cells. Gastroenterology
1997;113(2):615–24.
10. Bataller R, Gines P, Nicolas JM, et al. Angiotensin II induces contraction and pro-
liferation of human hepatic stellate cells. Gastroenterology 2000;118(6):1149–56.
11. Hautekeete ML, Geerts A. The hepatic stellate (Ito) cell: its role in human liver
disease. Virchows Arch 1997;430(3):195–207.
12. Sims DE. The pericyte—a review. Tissue Cell 1986;18(2):153–74.
13. Sims DE. Recent advances in pericyte biology—implications for health and
disease. Can J Cardiol 1991;7(10):431–43.
14. Pallone TL, Silldorff EP. Pericyte regulation of renal medullary blood flow. Exp
Nephrol 2001;9(3):165–70.
15. Novo E, Cannito S, Zamara E, et al. Proangiogenic cytokines as hypoxia-dependent
factors stimulating migration of human hepatic stellate cells. Am J Pathol 2007;
170(6):1942–53.
16. Martin P. Wound healing—aiming for perfect skin regeneration. Science 1997;
276(5309):75–81.
17. Nedelec B, Ghahary A, Scott PG, et al. Control of wound contraction. Basic and
clinical features. Hand Clin 2000;16(2):289–302.
18. Darby IA, Hewitson TD. Fibroblast differentiation in wound healing and fibrosis.
Int Rev Cytol 2007;257:143–79.
20. Pinzani M, Failli P, Ruocco C, et al. Fat-storing cells as liver-specific pericytes.
Spatial dynamics of agonist-stimulated intracellular calcium transients. J Clin
Invest 1992;90(2):642–6.
21. Sakamoto M, Ueno T, Kin M, et al. Ito cell contraction in response to endothelin-1
and substance P. Hepatology 1993;18(4):978–83.
22. Tao J, Mallat A, Gallois C, et al. Biological effects of C-type natriuretic peptide in
human myofibroblastic hepatic stellate cells. J Biol Chem 1999;274(34):23761–9.
23. Bataller R, Gasull X, Gines P, et al. In vitro and in vivo activation of rat hepatic
stellate cells results in de novo expression of L-type voltage-operated calcium
channels. Hepatology 2001;33(4):956–62.
24. Kawada N, Klein H, Decker K. Eicosanoid-mediated contractility of hepatic
stellate cells. Biochem J 1992;285(Pt 2):367–71.
25. Kawada N, Tran-Thi TA, Klein H, et al. The contraction of hepatic stellate (Ito) cells
stimulated with vasoactive substances. Possible involvement of endothelin 1 and
nitric oxide in the regulation of the sinusoidal tonus. Eur J Biochem 1993;213(2):
815–23.
26. Kawada N, Inoue M. Effect of adrenomedullin on hepatic pericytes (stellate cells)
of the rat. FEBS Lett 1994;356(1):109–13.
27. Kawada N, Kuroki T, Kobayashi K, et al. Action of endothelins on hepatic stellate
cells. J Gastroenterol 1995;30(6):731–8.
28. Housset C, Rockey DC, Bissell DM. Endothelin receptors in rat liver: lipocytes as
a contractile target for endothelin 1. Proc Natl Acad Sci U S A 1993;90(20):9266–70.
29. Rockey DC, Housset CN, Friedman SL. Activation-dependent contractility of rat
hepatic lipocytes in culture and in vivo. J Clin Invest 1993;92(4):1795–804.
30. Mashiba S, Mochida S, Ishikawa K, et al. Inhibition of hepatic stellate cell contrac-
tion during activation in vitro by vascular endothelial growth factor in association
with upregulation of FLT tyrosine kinase receptor family, FLT-1. Biochem Biophys
Res Commun 1999;258(3):674–8.
800 Soon & Yee
31. Yanase M, Ikeda H, Matsui A, et al. Lysophosphatidic acid enhances collagen gel
contraction by hepatic stellate cells: association with rho-kinase. Biochem
Biophys Res Commun 2000;277(1):72–8.
32. Reynaert H, Vaeyens F, Qin H, et al. Somatostatin suppresses endothelin-1-
induced rat hepatic stellate cell contraction via somatostatin receptor subtype 1.
33. Kharbanda KK, Rogers DD 2nd, Wyatt TA, et al. Transforming growth factor-beta
induces contraction of activated hepatic stellate cells. J Hepatol 2004;41(1):
60–6.
34. Perri RE, Langer DA, Chatterjee S, et al. Defects in cGMP-PKG pathway contrib-
ute to impaired NO-dependent responses in hepatic stellate cells upon activa-
tion. Am J Physiol Gastrointest Liver Physiol 2006;290(3):G535–42.
35. Laleman W, Van Landeghem L, Severi T, et al. Both Ca21-dependent and -inde-
pendent pathways are involved in rat hepatic stellate cell contraction and intrahe-
patic hyperresponsiveness to methoxamine. Am J Physiol Gastrointest Liver
Physiol 2007;292(2):G556–64.
36. Zhang JX, Pegoli W Jr, Clemens MG. Endothelin-1 induces direct constriction of
hepatic sinusoids. Am J Physiol 1994;266(4 Pt 1):G624–32.
37. Bauer M, Paquette NC, Zhang JX, et al. Chronic ethanol consumption increases
hepatic sinusoidal contractile response to endothelin-1 in the rat. Hepatology
1995;22(5):1565–76.
38. Suematsu M, Goda N, Sano T, et al. Carbon monoxide: an endogenous modulator
of sinusoidal tone in the perfused rat liver. J Clin Invest 1995;96(5):2431–7.
39. Vanheule E, Geerts AM, Reynaert H, et al. Influence of somatostatin and octreo-
tide on liver microcirculation in an experimental mouse model of cirrhosis studied
by intravital fluorescence microscopy. Liver Int 2008;28(1):107–16.
40. Thimgan MS, Yee HF Jr. Quantitation of rat hepatic stellate cell contraction: stellate
cells’ contribution to sinusoidal resistance. Am J Physiol 1999;277(1 Pt 1):G137–43.
41. Saab S, Tam SP, Tran BN, et al. Myosin mediates contractile force generation by
hepatic stellate cells in response to endothelin-1. J Biomed Sci 2002;9(6 Pt 2):
607–12.
42. Melton AC, Datta A, Yee HF Jr. [Ca21]i-independent contractile force generation
by rat hepatic stellate cells in response to endothelin-1. Am J Physiol Gastrointest
Liver Physiol 2006;290(1):G7–13.
43. Yanase M, Ikeda H, Ogata I, et al. Functional diversity between Rho-kinase- and
MLCK-mediated cytoskeletal actions in a myofibroblast-like hepatic stellate cell
line. Biochem Biophys Res Commun 2003;305(2):223–8.
44. Moller S, Emmeluth C, Henriksen JH. Elevated circulating plasma endothelin-1
concentrations in cirrhosis. J Hepatol 1993;19(2):285–90.
45. Alam I, Bass NM, Bacchetti P, et al. Hepatic tissue endothelin-1 levels in chronic
liver disease correlate with disease severity and ascites. Am J Gastroenterol
2000;95(1):199–203.
46. Gandhi CR, Sproat LA, Subbotin VM. Increased hepatic endothelin-1 levels and
endothelin receptor density in cirrhotic rats. Life Sci 1996;58(1):55–62.
47. Rockey DC, Fouassier L, Chung JJ, et al. Cellular localization of endothelin-1 and
increased production in liver injury in the rat: potential for autocrine and paracrine
effects on stellate cells. Hepatology 1998;27(2):472–80.
48. Tran-Thi TA, Kawada N, Decker K. Regulation of endothelin-1 action on the per-
fused rat liver. FEBS Lett 1993;318(3):353–7.
49. Okumura S, Takei Y, Kawano S, et al. Vasoactive effect of endothelin-1 on rat liver
in vivo. Hepatology 1994;19(1):155–61.
50. Rockey DC, Weisiger RA. Endothelin induced contractility of stellate cells from
normal and cirrhotic rat liver: implications for regulation of portal pressure and
resistance. Hepatology 1996;24(1):233–40.
51. Kaneda K, Ekataksin W, Sogawa M, et al. Endothelin-1-induced vasoconstriction
causes a significant increase in portal pressure of rat liver: localized constrictive
effect on the distal segment of preterminal portal venules as revealed by light and
electron microscopy and serial reconstruction. Hepatology 1998;27(3):735–47.
52. De Gottardi A, Shaw S, Sagesser H, et al. Type A, but not type B, endothelin
receptor antagonists significantly decrease portal pressure in portal hypertensive
rats. J Hepatol 2000;33(5):733–7.
53. Rockey DC, Chung JJ. Inducible nitric oxide synthase in rat hepatic lipocytes and
the effect of nitric oxide on lipocyte contractility. J Clin Invest 1995;95(3):
1199–206.
54. Goda N, Suzuki K, Naito M, et al. Distribution of heme oxygenase isoforms in rat
liver. Topographic basis for carbon monoxide-mediated microvascular relaxation.
J Clin Invest 1998;101(3):604–12.
55. Wang XE, Watanabe S, Oide H, et al. Hepatic stellate cell contraction is
inhibited by lipo-prostaglandin E1 in vitro. J Gastroenterol Hepatol 1998;
13(Suppl):S14–8.
56. Rockey DC, Chung JJ. Reduced nitric oxide production by endothelial cells in
cirrhotic rat liver: endothelial dysfunction in portal hypertension. Gastroenterology
1998;114(2):344–51.
57. Sarela AI, Mihaimeed FM, Batten JJ, et al. Hepatic and splanchnic nitric oxide
activity in patients with cirrhosis. Gut 1999;44(5):749–53.
58. Yu Q, Shao R, Qian HS, et al. Gene transfer of the neuronal NO synthase isoform
to cirrhotic rat liver ameliorates portal hypertension. J Clin Invest 2000;105(6):
741–8.
59. Sakamoto M, Ueno T, Sugawara H, et al. Relaxing effect of interleukin-1 on rat
cultured Ito cells. Hepatology 1997;25(6):1412–7.
60. Gasull X, Bataller R, Gines P, et al. Human myofibroblastic hepatic stellate cells
express Ca(21)-activated K(1) channels that modulate the effects of endothe-
lin-1 and nitric oxide. J Hepatol 2001;35(6):739–48.
61. Surks HK, Mochizuki N, Kasai Y, et al. Regulation of myosin phosphatase by
a specific interaction with cGMP-dependent protein kinase Ialpha. Science
1999;286(5444):1583–7.
62. Etter EF, Eto M, Wardle RL, et al. Activation of myosin light chain phosphatase in
intact arterial smooth muscle during nitric oxide-induced relaxation. J Biol Chem
2001;276(37):34681–5.
63. Nakamura K, Koga Y, Sakai H, et al. cGMP-dependent relaxation of smooth muscle
is coupled with the change in the phosphorylation of myosin phosphatase. Circ Res
2007;101(7):712–22.
64. Laleman W, Van Landeghem L, Van der Elst I, et al. Nitroflurbiprofen, a nitric oxide-
releasing cyclooxygenase inhibitor, improves cirrhotic portal hypertension in rats.
65. Racine-Samson L, Rockey DC, Bissell DM. The role of alpha1beta1 integrin in
wound contraction. A quantitative analysis of liver myofibroblasts in vivo and in
primary culture. J Biol Chem 1997;272(49):30911–7.
66. Rockey DC. New concepts in the pathogenesis of portal hypertension: hepatic
wounding and stellate cell contractility. Clin Liver Dis 1997;1(1):13–29.
67. Rockey DC. Cellular pathophysiology of portal hypertension and prospects for
management with gene therapy. Clin Liver Dis 2001;5(3):851–65.
802 Soon & Yee
68. Somlyo AP, Somlyo AV. Signal transduction and regulation in smooth muscle.
Nature 1994;372(6503):231–6.
69. Somlyo AP, Somlyo AV. Ca21 sensitivity of smooth muscle and nonmuscle myosin II:
modulated by G proteins, kinases, and myosin phosphatase. Physiol Rev 2003;
83(4):1325–58.
70. Yee HF Jr. Rho directs activation-associated changes in rat hepatic stellate cell
morphology via regulation of the actin cytoskeleton. Hepatology 1998;28(3):
843–50.
71. Kawada N, Seki S, Kuroki T, et al. ROCK inhibitor Y-27632 attenuates stellate cell
contraction and portal pressure increase induced by endothelin-1. Biochem
Biophys Res Commun 1999;266(2):296–300.
72. Tangkijvanich P, Tam SP, Yee HF Jr. Wound-induced migration of rat hepatic stel-
late cells is modulated by endothelin-1 through rho-kinase-mediated alterations in
the acto-myosin cytoskeleton. Hepatology 2001;33(1):74–80.
73. Melton AC, Yee HF. Hepatic stellate cell protrusions couple platelet-derived
growth factor-BB to chemotaxis. Hepatology 2007;45(6):1446–53.
74. Kato M, Iwamoto H, Higashi N, et al. Role of Rho small GTP binding protein in the
regulation of actin cytoskeleton in hepatic stellate cells. J Hepatol 1999;31(1):
91–9.
75. Yee HF Jr. Ca21 and rho signaling pathways: two paths to hepatic stellate cell
contraction. Hepatology 2001;33(4):1007–8.
76. Reynaert H, Thompson MG, Thomas T, et al. Hepatic stellate cells: role in micro-
circulation and pathophysiology of portal hypertension. Gut 2002;50(4):571–81.
77. Bataller R, Nicolas JM, Ginees P, et al. Contraction of human hepatic stellate cells
activated in culture: a role for voltage-operated calcium channels. J Hepatol
1998;29(3):398–408.
78. Kolodney MS, Thimgan MS, Honda HM, et al. Ca21-independent myosin II phos-
phorylation and contraction in chicken embryo fibroblasts. J Physiol 1999;
515(Pt 1):87–92.
79. Parizi M, Howard EW, Tomasek JJ. Regulation of LPA-promoted myofibroblast
contraction: role of Rho, myosin light chain kinase, and myosin light chain phos-
phatase. Exp Cell Res 2000;254(2):210–20.
80. Yee HF Jr, Melton AC, Tran BN. RhoA/rho-associated kinase mediates fibroblast
contractile force generation. Biochem Biophys Res Commun 2001;280(5):1340–5.
81. Blomhoff R, Wake K. Perisinusoidal stellate cells of the liver: important roles in
retinol metabolism and fibrosis. FASEB J 1991;5(3):271–7.
82. Ekataksin W, Kaneda K. Liver microvascular architecture: an insight into the path-
ophysiology of portal hypertension. Semin Liver Dis 1999;19(4):359–82.
83. Oikawa H, Masuda T, Kawaguchi J, et al. Three-dimensional examination of hepatic
stellate cells in rat liver and response to endothelin-1 using confocal laser scanning
microscopy. J Gastroenterol Hepatol 2002;17(8):861–72.
84. Wake K. Liver perivascular cells revealed by gold and silver impregnation
methods and electron microscopy. In: Motta P, editor. Biopathology of the liver:
an ultrastructural approach. Dordrecht (Netherlands): Kluwer; 2001. p. 23–6.
85. Zhang JX, Bauer M, Clemens MG. Vessel- and target cell-specific actions of
endothelin-1 and endothelin-3 in rat liver. Am J Physiol 1995;269(2 Pt 1):
G269–77.
86. Tangkijvanich P, Yee HF Jr. Cirrhosis—can we reverse hepatic fibrosis? Eur J Surg
Suppl 2002;168(Suppl 587):100–12.
87. Murphy F, Arthur M, Iredale J. Developing strategies for liver fibrosis treatment.
Expert Opin Investig Drugs 2002;11(11):1575–85.
88. Rockey DC. Hepatic fibrosis, stellate cells, and portal hypertension. Clin Liver Dis
2006;10(3):459–79.
89. Friedman SL, Bansal MB. Reversal of hepatic fibrosis – fact or fantasy? Hepatology
2006;43(2 Suppl 1):S82–8.
90. Zhou Q, Hennenberg M, Trebicka J, et al. Intrahepatic upregulation of RhoA and
Rho-kinase signalling contributes to increased hepatic vascular resistance in rats
with secondary biliary cirrhosis. Gut 2006;55(9):1296–305.
91. Kitamura K, Tada S, Nakamoto N, et al. Rho/Rho kinase is a key enzyme system
involved in the angiotensin II signaling pathway of liver fibrosis and steatosis.
J Gastroenterol Hepatol 2007;22(11):2022–33.
92. Tada S, Iwamoto H, Nakamuta M, et al. A selective ROCK inhibitor, Y27632, pre-
vents dimethylnitrosamine-induced hepatic fibrosis in rats. J Hepatol 2001;34(4):
529–36.
93. Murata T, Arii S, Nakamura T, et al. Inhibitory effect of Y-27632, a ROCK inhibitor,
on progression of rat liver fibrosis in association with inactivation of hepatic stel-
late cells. J Hepatol 2001;35(4):474–81.
94. Murata T, Arii S, Mori A, et al. Therapeutic significance of Y-27632, a Rho-kinase
inhibitor, on the established liver fibrosis. J Surg Res 2003;114(1):64–71.
95. Beljaars L, Molema G, Schuppan D, et al. Successful targeting to rat hepatic stel-
late cells using albumin modified with cyclic peptides that recognize the collagen
type VI receptor. J Biol Chem 2000;275(17):12743–51.
96. Gonzalo T, Beljaars L, van de Bovenkamp M, et al. Local inhibition of liver fibrosis
by specific delivery of a platelet-derived growth factor kinase inhibitor to hepatic
stellate cells. J Pharmacol Exp Ther 2007;321(3):856–65.
97. Kinoshita K, Iimuro Y, Fujimoto J, et al. Targeted and regulable expression of
transgenes in hepatic stellate cells and myofibroblasts in culture and in vivo using
an adenoviral Cre/loxP system to antagonise hepatic fibrosis. Gut 2007;56(3):
396–404.
98. Schoemaker MH, Rots MG, Beljaars L, et al. PDGF-receptor beta-targeted ade-
novirus redirects gene transfer from hepatocytes to activated stellate cells. Mol
Pharm 2008;5:399–406.
99. Sato Y, Murase K, Kato J, et al. Resolution of liver cirrhosis using vitamin A-coupled
liposomes to deliver siRNA against a collagen-specific chaperone. Nat Biotechnol
2008;26:431–42.
Dis eas e -Sp e c ific
Mecha nisms of
Fibrosis : Hepatitis
C Virus a nd
Nonalcoholic
Steatohepatitis
David van der Poorten, BSc (med) MBBSa,b, Jacob George, MBBS, PhDa,b,*
KEYWORDS
Fibrosis Hepatitis C Nonalcoholic steatohepatitis
Nonalcoholic fatty liver disease Pathogenesis
Insulin resistance
Hepatic fibrosis and cirrhosis are the endpoints of most types of chronic liver disease
and the result of repeated injury with the deposition of high-density extracellular matrix
proteins.1 Although historically believed to be an irreversible process because of
replacement of liver tissue with collagenous scar,2 fibrosis is now considered as
a wound-healing response to chronic injury, and thus potentially reversible.3 The
mechanisms of liver injury are multifactorial and disease specific. The stimuli may
include the generation of reactive oxygen species (ROS) with the subsequent develop-
ment of oxidative stress, hypoxia, inflammation, immune stimulation, apoptosis, stea-
tosis, and alterations within the extracellular matrix.4 Hepatic stellate cells (HSCs) are
the key effector cells of fibrogenesis and their activation is crucial to the fibrogenic
response.5 In the normal liver HSCs are quiescent, but with ongoing liver injury they
become activated and transform into myofibroblast-like cells capable of proliferation,
chemotaxis, matrix degradation, and fibrogenesis.5 In addition, HSCs are an important
source of matrix metalloproteinases (MMPs) and their physiologic inhibitors, tissue
inhibitors of metalloproteinases (TIMPs). Irrespective of cause, any repeated injury
to the liver over time results in fibrosis because of an imbalance between fibrolytic
This work was supported in part by the Westmead Millennium Institute initiating grant and the
Robert W. Storr Bequest to the University of Sydney.
a
Storr Liver Unit, Westmead Millennium Institute, Darcy Road, Westmead NSW 2145, Australia
b
Department of Medicine, University of Sydney at Westmead Hospital, Darcy Road, Westmead
NSW 2145, Australia
E-mail address: j.george@usyd.edu.au (J. George).
Clin Liver Dis 12 (2008) 805–824

1089-3261/08/$ – see front matter. Crown Copyright ª 2008 Published by Elsevier Inc. All rights reserved.
806 van der Poorten & George
and fibrogenic processes. This article focuses on the mechanisms of fibrosis that are
specific to hepatitis C and nonalcoholic steatohepatitis (NASH) and the overlap that is
present at times between these two conditions (Fig. 1). An excellent review of the gen-
eral mechanisms of hepatic fibrosis can be found elsewhere.4
Hepatitis C and NASH are two of the most common causes of hepatic fibrosis and
cirrhosis on a global scale. The World Health Organization estimates that up to 3%
(180 million people) of the world’s population are infected with hepatitis C, 130 million
of whom are at risk for developing cirrhosis.6 Although hepatic steatosis and NASH are
conditions predominantly affecting affluent populations, their impact cannot be under-
stated. In the United States alone, up to 30% of the population (90 million people) have
nonalcoholic fatty liver disease (NAFLD) and 3% to 4% have NASH.7,8 Similar rates are
now being reported in Europe and Asia.9–11 Far from being a benign entity as once
considered, it is now appreciated that NASH can cause progressive fibrosis12 and
is responsible for most cases of cryptogenic cirrhosis.13 The long-term prognosis is
no better than that of hepatitis C–related cirrhosis and most patients who have
NASH cirrhosis succumb to liver-related death.14,15
HEPATITIS C
Hepatitis C virus (HCV) is a positive sense, single-stranded RNA virus whose genome
encodes a large polyprotein precursor of more than 3000 amino acids, which is
cleaved to generate 10 key proteins: HCV core, envelope glycoproteins E1 and E2,
p7 protein, and the nonstructural proteins 2 (NS2), NS3, NS4A, NS4B, NS5A, and
NS5B. These proteins are important for replication and affect various cellular
Fig.1. Mechanisms of fibrosis in in chronic hepatitis C (CHC) and nonalcoholic steatohepatitis

(NASH).
Disease-Specific Mechanisms of Fibrosis 807
functions.16 Core protein has several biologic actions within hepatocytes and leuko-
cytes, including the control of apoptosis, cell growth, and the generation of oxidative
stress.17,18 These actions are driven by the generation of free radicals, stimulation of
mitogen-activated protein kinases, and NFkB activation.19,20 NS3 has been the best
characterized nonstructural protein and seems essential for viral replication but can
also regulate host cell functions, such as growth and differentiation.21
Immunopathogenic Mechanisms of Liver Damage and Fibrosis

Significant evidence exists that immune activation and modulation is a key mechanism
in the pathogenic process that leads to tissue injury, scarring, and ultimately cirrhosis
in chronic hepatitis C (CHC).22 Studies in mice and chimpanzees initially suggested
that HCV was not directly cytopathic,23,24 a contention supported by human studies
wherein the percentage of HCV-infected hepatocytes and the intrahepatic levels of
HCV-RNA had no correlation with liver damage.25,26 In addition it was shown that
HCV replication and high levels of viremia could occur in the absence of any inflamma-
tory response.27 Reports of patients who had HCV taking immunosuppressive therapy
have documented an amelioration of hepatic inflammation and damage despite
increased viral replication, then subsequent flares of disease and severe hepatic nec-
roinflammation when immunosuppressive therapy was stopped.22,28,29 These data
strongly suggest that the immune response is central in mediating hepatic inflamma-
tion and subsequent hepatic fibrosis in CHC as detailed later in this article.
Toll-like receptors
Initiation of the immune response by HCV is complex and involves the direct
recognition of viral components as specific foreign antigens and general activation
of the innate immune system.22 Initial detection of the virus by the innate immune sys-
tem occurs by way of toll-like receptors (TLRs), which recognize pathogen-associated
molecular patterns in viral RNA.30 The up-regulation of TLRs ultimately leads to
enhanced expression of Major Histocompatability Complex (MHC) class I molecules,
which are central to the cell-mediated immune response.22 Further stimulation of TLRs
on B lymphocytes and macrophages leads to induction of interleukin (IL)–1, IL-6, IL-8,
and tumor necrosis factor (TNF)a.30 IL-8 is of particular importance because it inhibits
type I antiviral interferon production while at the same time acting as a proinflammatory
and chemotactic cytokine.31,32 HCV further reduces the production and the effects of
antiviral interferons by inhibiting TLR3 and retinoic acid–inducible gene-I pathways,
which cause disruption of intracellular interferon signaling.22 The cumulative effect
of inadequate antiviral cellular immunity combined with proinflammatory cytokine se-
cretion leads to hepatocyte damage and subsequent fibrogenesis.22
Cytotoxic lymphocytes
These are the most common cells seen in the livers of individuals who have CHC and
comprise cytotoxic CD81 T cells, T cells, natural killer (NK) T cells, and NK cells.33
These lymphocytes are responsible for killing infected or damaged cells and thus
are vital to the immune response to viral infection.34 In CHC their activation may be
by recognition of viral peptide/MHC I complex, by lack of MHC, or by direct binding
with viral protein. HCV envelope proteins E1 and E2 have been shown to bind directly
to CD81 receptors, which are present on all these cells.35 In gd T cells, this leads to the
release of inflammatory cytokines, such as TNFa and INFg without T-cell receptor
activation.22 gd T cells are important because they are known to be present in
increased proportions in HCV-infected livers with high necroinflammatory activity.36
A good cytotoxic T cell response in general is associated with control of HCV
replication.37 In the context of inefficient viral clearance, however, they may still cause
tissue damage through perforins/granzymes, Fas/FasL, and TNF pathways, which in
addition to killing infected hepatocytes may cause significant bystander damage.34
NK cells play an important role as the first line of defense against viral infections
through rapid recognition and lysis of infected cells, coupled with secretion of
proinflammatory cytokines.22 Despite the potential for tissue damage and
inflammation induced by these functions, it seems that NK cells in fact play a protective
role in CHC. In patients who have CHC with low or defective NK cell function, there is
both viral persistence38 and accelerated progression to liver fibrosis,39 whereas
increased NK cytolytic activity has been associated with HSC killing and reduced
fibrosis.40,41 That alcohol inhibits NK activity may in part explain the faster progression
to cirrhosis of alcoholics who have CHC.37
Direct Cytopathic Effects of Hepatitis C Virus Leading to Fibrosis

Despite the key role of the immune system and previous evidence suggesting that
HCV may not be directly cytopathic, a great deal of work has now demonstrated
that HCV can directly mediate apoptosis, stellate cell activation, and liver damage.
Apoptosis
Apoptosis, or programmed cell death, is a highly regulated process by which cells are
eliminated with minimal inflammation or leakage of intracellular components.42 Inter-
nal and external pathways lead to the activation of caspase enzymes, which along with
DNA-degrading enzymes are responsible for degrading intracellular components,
ultimately leading to the formation of small apoptotic bodies. Engulfment of these
apoptotic bodies by phagocytes, including stellate cells, is intensely profibrogenic.34
Levels of caspases are consistently up-regulated in patients who have HCV compared
with controls, and to correlate with both inflammatory and fibrotic liver injury.43,44
Apoptosis has been shown to be profibrogenic in in vitro models by way of activation
of stellate cells and up-regulation of collagen and transforming growth factor (TGF)–b
genes.45 A key mechanism behind HCV-induced apoptosis occurs by way of
external Fas/FasL (otherwise known as CD95/CD95L ‘‘death receptor’’) mediated
pathways.42 Activation of these pathways leads to the recruitment of caspase
enzymes and subsequent apoptosis.46 Levels of Fas and FasL are higher in patients
who have HCV compared with controls and their expression correlates with the
degree of hepatic inflammation.47,48 Moreover, HCV envelope proteins 1 and 2 have
recently been shown to directly enhance FasL expression in hepatocytes.49 HCV
core and NS5A proteins have also been implicated in the regulation of apoptosis.
Although results have been conflicting, it seems these HCV proteins may affect intrin-
sic pathways to increase apoptosis and cause liver injury, and in other circumstances
to inhibit apoptosis and increase their own survival.34
Direct stellate cell activation

Good experimental evidence now exists to suggest that HCV can directly stimulate
fibrogenesis by way of interactions between infected hepatocytes, viral proteins,
and HSCs (Fig. 2). This evidence was first demonstrated in experiments involving
isolated human and rat HSCs by Bataller and colleagues.50 In the first experiment,
activated human HSCs were incubated with recombinant core and NS3 proteins.
The investigators were able to demonstrate increases in intracellular calcium concen-
tration and ROS production, both of which are involved in the proinflammatory func-
tions of HSCs.50 ROS in particular has been shown to have a crucial role in the
biologic actions of HSCs and in experimental liver fibrosis.51 Infection of HSCs with
Fig. 2. Direct and indirect profibrogenic interactions between the hepatitis C virus (HCV) and
hepatic stellate cells (HSCs). (A) HCV core protein and nonstructural proteins 3-5 (NS3-NS5)
induce HSC production of cytokines and extracellular matrix growth factors, such as TGF-b1
and procollagen-a1. (B) Hepatocytes infected with NS3-NS5 produce TGF-b1, which in turn
stimulates HSC production of ECM-promoting factors, TIMP-1 and MMP-2. There is also
a concurrent down-regulation of the fibrinolytic matrix metalloproteinases MMP-3 and
MMP-13. (C) HCV envelope glycoprotein 2 (E2) directly binds to transmembrane CD81 on
HSCs to cause an up-regulation in the destructive MMP-2.
nonstructural HCV proteins (NS3-NS5), led to a marked increase in proinflammatory

chemokines and adhesion molecules that was mediated by increases in intracellular
calcium and ROS.50 The stimulation of chemokines and adhesion molecules, such
as IL-8, MCP-1, and ICAM-1, is important because these molecules participate in
the recruitment and activation of lymphocytes, the main inflammatory cell type in
chronic hepatitis C.52 In the final experiment, early cultured rat HSCs were infected
with either core or NS3-NS5 adenovirus.50 Both core and to a lesser extent NS3-
NS5 induced accelerated HSC activation and stimulated the secretion of TGF-b1
and procollagen a1. It is worth noting that simply activating HSCs is important,
because activated HSCs alone exert inflammatory actions that contribute to the path-
ogenesis of chronic liver disease.53
Other work has defined some of the mechanisms by which HCV-infected
hepatocytes can indirectly stimulate fibrogenesis. Using the human hepatoma cell
line Huh-7 5-15, which stably expresses HCV viral genes NS3-NS5, and HSC/myofi-
broblasts as effector cells, Schulze-Krebs and colleagues investigated the expression
of key profibrogenic genes and factors.54 In HCV-infected hepatocytes they
demonstrated a threefold increase in TGF-b1 mRNA levels and a sixfold increase in
TGF-b1 protein levels when compared with noninfected controls. When HCV replicon
cells were incubated with activated HSCs, there was a definite shift to a more
fibrogenic state with increases in connective tissue growth factor (CTGF), procollagen
a1, and TIMP-1. CTGF is an important marker of hepatic fibrogenesis and HSC
activation status and seems to be a strong promoter of extracellular matrix (ECM)
accumulation.55,56 Also seen was an increase in the activity of MMP-2, which is
considered a destructive and profibrogenic MMP because of its involvement in the
degradation of the basal lamina,57 and a substantial down-regulation of MMP-3 and
MMP-13. Both MMP-3 and MMP-13 are involved in the degradation of fibril-forming
collagens, whereas MMP-3 is also an activator of other MMPs.58 Differential modula-
tion of MMPs in conjunction with an increase in the inhibitor of most MMPs, TIMP-1,
strongly reflects a shift to net profibrogenic state. TGF-b1 is known to be an important
mediator of liver fibrogenesis by potently inducing the deposition of ECM compo-
nents59 and by directly activating HSCs.60 The demonstration by other groups that
HCV core protein also up-regulates TGF-b161 emphasizes the likely importance of
this as a mechanism of HCV-induced fibrosis in the absence of inflammation.
An additional HCV protein, E2, is of importance in HCV-related fibrogenesis by way
of up-regulation of the destructive metalloproteinase MMP-2.62 MMP-2 plays an inte-
gral part in the progression of HCV-related fibrosis by degrading and remodeling the
normal liver ECM, such as collagen (I and IV), fibronectin, and laminin.63 Degradation
of the normal ECM allows penetration of inflammatory cells and may be a key event
mediating ongoing tissue injury. The major site of E2 binding is the transmembrane
molecule CD81, which acts as a facilitator for downstream intracellular signaling.64
This receptor is expressed in high concentrations on the cell surface of HSCs.65,66
In experiments performed by Mazzocca and colleagues,62 E2 glycoprotein added to
activated stellate cells strongly bound to the CD81 molecule. This interaction led to
significant up-regulation of the synthesis and activity of MMP2.
Steatosis
Hepatic steatosis is a common histologic feature of CHC and may contribute to fibro-
genesis in this disease. The prevalence of steatosis in patients who have CHC ranges
from 40% to 86%,67–70 as compared with rates of 20% to 30% seen in people who
have other liver diseases.71,72 Although steatosis may be associated with obesity,
diabetes, alcohol abuse, or other causes, good evidence now suggests that HCV
can be directly involved in its induction.73,74 In culture, core protein localizes to lipid
droplets in hepatocytes and can stimulate the production of de novo lipid75,76 and
its redistribution within the cell.77 In transgenic mice, core protein inhibits microsomal
triglyceride transfer protein, which leads to the accumulation of intrahepatic trig-
lyceride.78 Cross-sectional studies have demonstrated that the presence of hepatic
steatosis in CHC is associated with genotype 3 infection and higher body mass index
(BMI), suggesting that there is ‘‘viral steatosis’’ (especially in patients who have
genotype 3) and ‘‘metabolic steatosis’’ (predominantly in patients who have
genotype 1).73,79
The results of human studies have been conflicting with regard to the influence of
steatosis on fibrosis progression. A major drawback of most published reports has
been their failure to consider covariates that may coexist in the setting of hepatic stea-
tosis, such as insulin resistance (IR), which in and of itself may be the fibrogenic factor,
rather than steatosis per se. In one review, 10 of 14 studies found an association
between steatosis and increased fibrosis, but the associations were weak and the
studies heterogeneous.73 Some studies have only found an association between
genotype 3 viral steatosis and fibrosis, whereas others have suggested that only
genotype 1 metabolic steatosis affects fibrosis progression.80
Possibly the most compelling evidence comes from a recent meta-analysis of more
than 3000 patients who had CHC from Europe, Australia, and the United States.81 On
multivariate analysis steatosis was a strong independent predictor of fibrosis stage
and was also associated with necroinflammation, BMI, genotype 3, and increasing
age. Unfortunately, IR was not evaluated in this cohort. There was no association
between steatosis and fibrosis in obese patients, suggesting that in these patients
the association between steatosis and fibrosis was indirect, possibly linked to
necroinflammation, IR, or enhanced hepatic susceptibility to other insults, such as
apoptosis.80,81 A recent study has elegantly illustrated this point by showing that in
CHC with steatosis, apoptosis was associated with HSC activation and increased
fibrosis, whereas in the absence of steatosis, apoptosis had no influence on HSCs
or fibrosis.82
The profibrogenic potential of steatosis is strengthened by several studies in which
a link has been demonstrated between necroinflammatory activity and steatosis
grade.79,83,84 Several mechanisms may account for this link. First, it has been shown
in cell lines that HCV core protein, which induces steatosis, also generates oxidative
stress by way of generation of ROS, the effects of which are magnified and propa-
gated in the steatotic liver.73,85 TNFa levels, which are increased in patients who
have CHC, correlate with necroinflammatory activity86 and may further increase
ROS. In addition, a small subgroup of patients who have CHC has true steatohepati-
tis,87 the pathogenesis of which is discussed later. The specific role of IR to fibrogen-
esis in CHC is discussed next.
Insulin resistance
It is now well recognized that there is a strong association between CHC and type 2
diabetes.88–90 Moreover, diabetes has been associated with increased fibrosis
progression in patients who have CHC.91,92 IR is the precursor to type 2 diabetes
and precedes the onset of diabetes by 10 to 20 years.93 Previous work from our
unit has helped to clarify the role and importance of IR in hepatitis C disease progres-
sion. In 260 patients who had CHC we demonstrated that the extent and the rate of
fibrosis was associated with IR as measured by the homeostasis model.94 In a subset
of patients who had mild CHC (fibrosis 0–1) insulin, c-peptide, and Homeostasis
model assessment of insulin resistance (HOMA-IR) levels were all significantly higher
(P < .01) than healthy matched controls. Increasing HOMA-IR was associated with
higher BMI and inflammatory grade, whereas the HOMA-IR scores in patients who
had genotype 3 were significantly lower. These findings suggest that HCV can induce
IR independent of disease severity and that IR contributes to fibrosis progression in
hepatitis C. Moreover, despite there being an association between steatosis and fibro-
sis, the profibrogenic effects of IR were independent of steatosis grade. There was
also a suggestion that the relationship between hepatic steatosis, IR, and fibrosis
may be genotype specific.95 Although genotype 3 subjects had more extensive stea-
tosis, they had lower IR compared with other genotypes for each stage of fibrosis. IR
remained an important predictor of progressive fibrosis even when genotype 3 pa-
tients were considered in isolation,94 and in a subsequent, larger cohort.96 At the mo-
lecular level, IR and the resulting hyperinsulinemia have been shown to directly
stimulate fibrogenesis. Hyperinsulinemia thus promotes HSC proliferation and secre-
tion of ECM components and up-regulates the key profibrogenic cytokine CTGF.97,98
TNFa also seems to be intimately involved in the pathophysiology linking steatosis,
IR, inflammation, and fibrosis in CHC. TNFa levels are increased in patients who have
CHC and associated with increased hepatic inflammation.86 Likewise, TNFa can
induce IR by interference with insulin receptor substrates 1 and 2, which are essential
for insulin signaling.99,100 Finally, certain TNF polymorphisms have been linked to
fibrosis progression.101 The importance of TNF to IR in CHC has been underscored
by studies in HCV transgenic mice in which anti-TNF antibody was able to ameliorate
IR.102 In a recent human study, however, we were unable to demonstrate any associ-
ation between cytokines such as TNFa in serum and the extent of IR as determined by
the homeostasis model.103 Despite these data a role for hepatic TNFa in mediating
hepatic IR in CHC is not excluded.
NONALCOHOLIC STEATOHEPATITIS
NAFLD is defined by the presence of macrovesicular fat in greater than 5% of

hepatocytes, in the absence of significant alcohol use or other secondary causes of
steatosis.104 NASH represents the potentially progressive form of NAFLD, a spectrum
that ranges from fat alone, to fat with inflammation, hepatocyte injury, and fibrosis.105
Patients who have steatosis alone are believed to have a benign clinical course with
only 3% developing cirrhosis after long-term follow-up.106,107 In contrast, up to 50%
of patients who have NASH have progressive fibrosis108–110 and 15% to 25% develop
cirrhosis.105,111 Although the mechanisms behind the development of steatosis and
the progression to NASH and cirrhosis remain incompletely understood, the last few
years have seen significant progress, the details of which are discussed here.
Insulin Resistance
NAFLD is strongly linked to overweight and obesity, and most patients have features
of the metabolic syndrome.112,113 In case series, up to 80% of patients who have
NASH are obese, 50% to 70% are hypertensive, and up to 70% have dyslipide-
mias.114,115 It has thus become accepted that NAFLD is the hepatic manifestation
of the metabolic syndrome. The pathophysiologic process that ties all these
conditions together is IR, which is now considered to be an intrinsic defect in
NAFLD.116 According to the ‘‘two hit’’ theory of NASH,117 IR is the initiating event
that causes an increase in hepatic triglyceride synthesis and steatosis.118 In turn, IR
per se is proinflammatory and in conjunction with other pathophysiologic processes
operating in a fatty liver may provide the second or additional hits.119 Several clinical
studies have supported this contention by demonstrating a strong independent asso-
ciation between IR and fibrosis severity in well-characterized NASH cohorts.120,121
The pathophysiologic mechanisms behind insulin’s toxicity may include the genera-
tion of oxidative stress by way of ROS122 and an ability to worsen steatosis by way
of up-regulation of sterol regulatory element-binding protein.123 Moreover, IR and
hyperinsulinemia can stimulate fibrogenesis by way of up-regulation of CTGF and
direct interactions with HSCs, effects that are particularly marked in the presence of
hyperglycemia.97,98 These finding may explain why patients who have type 2 diabetes
and NASH have rapidly progressive disease and a poor prognosis.105,124
Oxidative Stress
It is widely accepted that oxidative stress is one of the vital second hits involved in the
progression of hepatic steatosis to NASH.105 Numerous studies in animal models of
NAFLD have demonstrated evidence of increases in ROS production and lipid perox-
idation.125–128 In human studies, immunohistochemical staining for the byproducts of
oxidation were increased in NAFLD compared with controls, and were significantly
increased in advanced NASH when compared with simple steatosis.129 An increase
in the hepatic and serum level of lipid peroxidation products has also been reported
in patients who have NAFLD.12,130 More recently, high levels of lipid peroxidation
antibodies have been demonstrated in patients who had advanced NASH fibrosis.131
There was a significant incremental increase in antibody titers from controls to those
who had steatosis, and from steatosis to NASH, highlighting the importance of oxida-
tive stress in progressive disease. The profibrogenic properties of oxidative stress are
mediated by direct DNA and mitochondrial damage, induction of hepatocyte apopto-
sis, and amplification of the inflammatory response.4 Mitochondrial damage is of par-
ticular importance as the changes induced by oxidative injury impair mitochondrial
function and lead to the promotion of lipid peroxidation and further generation of
ROS.105 ROS also stimulates the production of profibrogenic mediators from Kupffer
cells, such as TNFa, and directly stimulates proliferation and activation of HSCs.4 The
mechanisms of oxidative stress production in NAFLD, which include hyperinsulinemia,
free fatty acid (FFA) peroxidation, and steatosis, are discussed elsewhere.
Apoptosis
Hepatocyte apoptosis is a key mediator of progressive liver injury and fibrosis in
NASH, as it is for hepatitis C. Markers of apoptosis, such as activated caspase-3
and Fas receptor, are significantly up-regulated in patients who have NASH compared
with steatosis and controls.132 In addition, there seems to be a strong correlation
between apoptotic activity in NASH and inflammatory and fibrosis grades.132,133 A
recent study has even suggested that markers of caspase activation in the serum
can be used to noninvasively differentiate simple steatosis from NASH.134 Important
mediators of apoptosis in NASH include hyperinsulinemia, oxidative stress, and
excess FFAs.
Steatosis and Fat Compartmentalization

Steatosis results from the excessive storage of lipids within the liver. In NAFLD these
are predominantly triglycerides (TGs), although FFAs, cholesterol, and phospholipids
may also accumulate. Recent evidence from experimental studies suggests that com-
partmentalization of lipids and the quality rather than quantity of lipids accumulating
may play a vital role in the development of progressive disease.135 Studies in mice
have shown that hepatic accumulation of saturated FFA causes a decrease in hepa-
tocyte viability and increased apoptosis, whereas the accumulation of TGs alone was
protective.136,137 FFAs seem to be toxic by several mechanisms. First, FFAs have
been shown to up-regulate proapoptotic death receptors, such as Fas138 and TRAIL
receptor 5 (DR5), and both of these receptors have been shown to be increased in
patients who have NASH.132,139 Second, mitochondrial dysfunction, which is a central
abnormality behind the progression from steatosis to NASH,140 can be induced by sat-
urated FFAs by a dose- and saturation-dependent mechanism.137 Finally, overaccu-
mulation of saturated FFAs may result in endoplasmic reticulum stress and
apoptosis.141 Hepatic free cholesterol may also have a toxic role in the progression
of NASH. Murine studies wherein hepatic content of either TG or cholesterol was in-
creased showed that TNF-related apoptosis and liver damage occurred only in those
who had increased liver cholesterol.142 When the mice were treated with an HMG-CoA
reductase inhibitor, hepatic free cholesterol was reduced and liver damage amelio-
rated. This reduction in free hepatic cholesterol may in part explain the beneficial
effects HMG-CoA reductase inhibitors have elicited in human NASH trials.143,144
Although the exact importance of lipid partitioning in human NASH remains unre-
solved, one recent lipidomic study reported that both increased free cholesterol and
an increase in n-6/n-3 polyunsaturated fatty acid ratio was associated with NASH
compared with NAFLD.145
Adipokines
Adiponectin
Adiponectin is the most highly abundant adipokine in human serum and has insulin
sensitizing and anti-inflammatory properties.146 Two receptors, AdipoR1 and Adi-
poR2, located predominantly in muscle cells and the liver, respectively, mediate the
actions of adiponectin.147 PPARg is involved in transcriptional up-regulation of the adi-
ponectin gene and the PPARg agonist thiazolidinediones increase serum adiponectin
levels.148 Adiponectin stimulates PPARa, which has several beneficial effects, includ-
ing increased fatty acid oxidation, a reduction in hepatic triglycerides, and inhibition of
cytokines, such as Il-6 and COX-2.119 Further anti-inflammatory effects of adiponectin
are mediated by inhibition of macrophages and direct blockade of TNF release.149 It is
not surprising therefore that hypoadiponectinemia has been associated with increases
in BMI, fasting insulin levels, serum triglycerides, and body fat. Work from our group
has demonstrated that adiponectin levels are significantly reduced in patients who
have NAFLD and NASH compared with controls and are predictive of more advanced
grades of steatosis and necroinflammation, independent of IR.120 Subsequent studies
have underscored this relationship, and in addition have shown a significant correla-
tion between low adiponectin levels and NASH-associated fibrosis.150,151 Of particular
note is the demonstration that the mRNA expression of the main hepatic adiponectin
receptor, AdipoR2, correlates inversely with liver fibrosis in NASH.152
The importance of hypoadiponectinemia has been demonstrated in adiponectin
knockout mice, in which enhanced hepatic fibrosis was seen in response to hepato-
toxic insults accompanied by significantly increased levels of TGF-b1 and CTGF.153
In contrast, adiponectin treated ob/ob mice show improvement in steatosis, liver
enzyme levels, and hepatic inflammation.154 HSCs express both adiponectin recep-
tors155 and when incubated with adiponectin, HSCs undergo several anti-fibrogenic
changes, including apoptosis,155 reversal of activation, and resistance to the profibro-
genic effects of platelet derived growth factor and TGF-b1.153 Taken together, all
these findings suggest an important role for adiponectin in modulating hepatic fibrosis
in NASH.
Leptin
Leptin (from the Greek word leptos, meaning thin), discovered as the ob gene product
in 1994, was initially considered to be solely an anorexigenic hormone with the poten-
tial to decrease food intake and increase energy expenditure.156 It soon became clear,
however, that despite its deficiency causing severe obesity in mice, most obese
humans had elevated leptin levels in association with leptin resistance.157 Leptin
receptors are located throughout the body, including the liver, and mediate a wide
variety of effects, including energy homeostasis, anti-steatogenesis, immune modula-
tion, and hepatic fibrogenesis.156 The profibrogenic properties of leptin are highlighted
by experiments in mice in which administration of leptin augments hepatotoxin-
induced fibrosis,158 whereas leptin-deficient mice remain resistant to fibrosis.159,160
The mechanisms underlying these profibrogenic properties relate to induction of
TGF-b1 and CTGF in Kupffer cells161 and a direct interaction with HSCs that causes
proliferation and up-regulation of collagen genes.162,163 Although in one study leptin
levels were higher in patients who had NASH compared with controls,164 this has
not been a consistent finding.165,166 Moreover, an association between leptin levels
and fibrosis stage in NASH has never been demonstrated.164–166 It is certainly possible
that the profibrogenic actions of leptin depend more on intrahepatic production and
subsequent autocrine and paracrine actions.167 At this stage a definite role for leptin
in NASH-related fibrogenesis is uncertain, and further studies examining the intrahe-

patic expression of leptin are needed.
Visceral Fat
Intriguing recent evidence suggests that visceral fat may play an important
proinflammatory and profibrogenic role in the pathogenesis of NASH. Visceral obesity
is the cornerstone of the metabolic syndrome, of which NAFLD has been considered
the hepatic manifestation. The portal/fatty acid flux theory suggests that visceral fat,
by its unique location and enhanced lipolytic activity, releases large amounts of toxic
FFAs, which are delivered in high concentrations directly to the liver.168,169 This
process leads to the accumulation and storage of hepatic fat and the development
of hepatic IR.168,169 The effects of steatosis and excess FFAs induce oxidative stress,
apoptosis, and liver damage. Obesity has recently been shown to be an inflammatory
condition with macrophage accumulation in adipose tissue, a process that is exagger-
ated in visceral fat.170 These macrophages are responsible for the production of proin-
flammatory cytokines and the modulation of adipocyte-derived cytokines. Harmful
factors, such as IL-6 and TNF-a, are expressed in greater amount in visceral fat
than subcutaneous fat.168 In this context, recent work from our laboratory has
examined the relationship between increasing visceral fat and liver histology in
a well-defined cohort of obese patients who had NAFLD. We demonstrated that the
extent of visceral fat significantly correlated with increasing hepatic necroinflammation
and fibrosis, and that this effect was independent of IR and hepatic steatosis.171 The
levels of IL-6 correlated with the extent of visceral fat and with liver inflammation, flag-
ging an important role for this cytokine. Certainly further studies are needed, but it
seems that visceral fat, by various mechanisms, may be an important player in
NASH fibrogenesis.
SUMMARY
This article delineates the disease-specific mechanisms responsible for hepatic fibro-
sis operating in people who have CHC infection and in those who have NASH. For
those who have CHC, the virus seems to mediate fibrogenesis through cytopathic ef-
fects on hepatocytes, direct interactions with hepatic stellate cells, and activation of
the immune system. There has also been recognition of the importance of hepatic
steatosis and IR to fibrogenesis in CHC. Meanwhile, NASH and its precursor lesion,
hepatic steatosis, seem to be the diseases of our times, as rates of obesity skyrocket
worldwide. IR and steatosis are intrinsic deficits in this disease that create the milieu in
which all subsequent damage occurs. We now appreciate that not all fats are created
equal, and that FFAs along with free cholesterol may be the most profibrogenic. Most
important, however, visceral fat looms as the genesis of most if not all metabolic
derangement and may be the underlying orchestrator of the alterations in adipokines,
IR, fatty acids, steatosis, and inflammation that are central to NASH. These insights
serve not only to better our understanding of the natural history and progression of
CHC and NASH fibrosis but also to permit rational attempts to try to interfere with
the disease processes itself and fibrosis in general.
REFERENCES

2. Schaffner F, Klion FM. Chronic hepatitis. Annu Rev Med 1968;19:25–38.
3. Albanis E, Friedman SL. Hepatic fibrosis. Pathogenesis and principles of
therapy. Clin Liver Dis 2001;5:315–34, v–vi.
4. Guo JS, Friedman SL. Hepatic fibrogenesis. Semin Liver Dis 2007;27:413–26.
5. Friedman SL. Mechanisms of disease: mechanisms of hepatic fibrosis and ther-
apeutic implications. Nat Clin Pract Gastroenterol Hepatol 2004;1:98–105.
6. World Health Organisation. Initiative for vaccine research (IVR); hepatitis C
page. Available at: http://www.who.int/vaccine_research/diseases/viral_cancers.
Accessed January 15, 2008.
7. Riley P, O’Donohue J, Crook M. A growing burden: the pathogenesis, investiga-
tion and management of non-alcoholic fatty liver disease. J Clin Pathol 2007;
60(12):1384–91.
8. Browning JD, Szczepaniak LS, Dobbins R, et al. Prevalence of hepatic steatosis
in an urban population in the United States: impact of ethnicity. Hepatology
2004;40:1387–95.
9. Bedogni G, Miglioli L, Masutti F, et al. Prevalence of and risk factors for nonalco-
holic fatty liver disease: the Dionysos nutrition and liver study. Hepatology 2005;
42:44–52.
10. Amarapurkar DN, Hashimoto E, Lesmana LA, et al. How common is non-
alcoholic fatty liver disease in the Asia-Pacific region and are there local differ-
ences? J Gastroenterol Hepatol 2007;22:788–93.
11. Chitturi S, Farrell GC, George J. Non-alcoholic steatohepatitis in the Asia-Pacific
region: future shock? J Gastroenterol Hepatol 2004;19:368–74.
12. Loguercio C, De Simone T, D’Auria MV, et al. Non-alcoholic fatty liver disease:
a multicentre clinical study by the Italian Association for the Study of the Liver.
Dig Liver Dis 2004;36:398–405.
13. Ayata G, Gordon FD, Lewis WD, et al. Cryptogenic cirrhosis: clinicopathologic
findings at and after liver transplantation. Hum Pathol 2002;33:1098–104.
14. Hui JM, Kench JG, Chitturi S, et al. Long-term outcomes of cirrhosis in nonalco-
holic steatohepatitis compared with hepatitis C. Hepatology 2003;38:420–7.
15. Sanyal AJ, Banas C, Sargeant C, et al. Similarities and differences in outcomes
of cirrhosis due to nonalcoholic steatohepatitis and hepatitis C. Hepatology
2006;43:682–9.
16. Rosenberg S. Recent advances in the molecular biology of hepatitis C virus.
J Mol Biol 2001;313:451–64.
17. Lai MM. Hepatitis C virus proteins: direct link to hepatic oxidative stress, steato-
sis, carcinogenesis and more. Gastroenterology 2002;122:568–71.
18. McLauchlan J. Properties of the hepatitis C virus core protein: a structural pro-
tein that modulates cellular processes. J Viral Hepat 2000;7:2–14.
19. Erhardt A, Hassan M, Heintges T, et al. Hepatitis C virus core protein induces
cell proliferation and activates ERK, JNK, and p38 MAP kinases together with
the MAP kinase phosphatase MKP-1 in a HepG2 Tet-Off cell line. Virology
2002;292:272–84.
20. Yoshida H, Kato N, Shiratori Y, et al. Hepatitis C virus core protein activates
nuclear factor kappa B-dependent signaling through tumor necrosis factor
receptor-associated factor. J Biol Chem 2001;276:16399–405.
21. Bartenschlager R. The NS3/4A proteinase of the hepatitis C virus: unravelling
structure and function of an unusual enzyme and a prime target for antiviral ther-
apy. J Viral Hepat 1999;6:165–81.
22. Spengler U, Nattermann J. Immunopathogenesis in hepatitis C virus cirrhosis.
Clin Sci 2007;112:141–55.
23. Maeda N, Watanabe M, Okamoto S, et al. Hepatitis C virus infection in human
liver tissue engrafted in mice with an infectious molecular clone. Liver Int
2004;24:259–67.
24. Bradley DW. Studies of non-A, non-B hepatitis and characterization of the hep-
atitis C virus in chimpanzees. Curr Top Microbiol Immunol 2000;242:1–23.
25. Rodriguez-Inigo E, Bartolome J, de Lucas S, et al. Histological damage in
chronic hepatitis C is not related to the extent of infection in the liver. Am J Pathol
1999;154:1877–81.
26. McGuinness PH, Bishop GA, Painter DM, et al. Intrahepatic hepatitis C RNA levels
do not correlate with degree of liver injury in patients with chronic hepatitis C.
Hepatology 1996;23:676–87.
27. Brillanti S, Foli M, Gaiani S, et al. Persistent hepatitis C viraemia without liver dis-
ease. Lancet 1993;341:464–5.
28. Gruber A, Lundberg LG, Bjorkholm M. Reactivation of chronic hepatitis C after
withdrawal of immunosuppressive therapy. J Intern Med 1993;234:223–5.
29. Melisko ME, Fox R, Venook A. Reactivation of hepatitis C virus after chemother-
apy for colon cancer. Clin Oncol (R Coll Radiol) 2004;16:204–5.
30. Schwabe RF, Seki E, Brenner DA. Toll-like receptor signaling in the liver. Gastro-
enterology 2006;130:1886–900.
31. Khabar KS, Al-Zoghaibi F, Al-Ahdal MN, et al. The alpha chemokine, interleukin
8, inhibits the antiviral action of interferon alpha. J Exp Med 1997;186:1077–85.
32. Polyak SJ, Khabar KS, Rezeiq M, et al. Elevated levels of interleukin-8 in serum
are associated with hepatitis C virus infection and resistance to interferon ther-
apy. J Virol 2001;75:6209–11.
33. Waterhouse NJ, Clarke CJ, Sedelies KA, et al. Cytotoxic lymphocytes; instiga-
tors of dramatic target cell death. Biochem Pharmacol 2004;68:1033–40.
34. Mengshol JA, Golden-Mason L, Rosen HR. Mechanisms of disease: HCV-
induced liver injury. Nat Clin Pract Gastroenterol Hepatol 2007;4:622–34.
35. Levy S, Shoham T. The tetraspanin web modulates immune-signalling com-
plexes. Nat Rev Immunol 2005;5:136–48.
36. Agrati C, D’Offizi G, Narciso P, et al. Vdelta1 T lymphocytes expressing a Th1
phenotype are the major gammadelta T cell subset infiltrating the liver of HCV-
infected persons. Mol Med 2001;7:11–9.
37. Teixeira R, Marcos LA, Friedman SL. Immunopathogenesis of hepatitis C
infection and hepatic fibrosis: New insights into antifibrotic therapy in chronic
hepatitis C. Hepatol Res 2007;37:579–95.
38. Deignan T, Curry MP, Doherty DG, et al. Decrease in hepatic CD56(1) T cells
and V alpha 24(1) natural killer T cells in chronic hepatitis C viral infection.
J Hepatol 2002;37:101–8.
39. Berenguer M, Ferrell L, Watson J, et al. HCV-related fibrosis progression
following liver transplantation: increase in recent years. J Hepatol 2000;32:
673–84.
40. Morishima C, Paschal DM, Wang CC, et al. Decreased NK cell frequency in
chronic hepatitis C does not affect ex vivo cytolytic killing. Hepatology 2006;
43:573–80.
41. Radaeva S, Sun R, Jaruga B, et al. Natural killer cells ameliorate liver fibrosis by
killing activated stellate cells in NKG2D-dependent and tumor necrosis factor-
related apoptosis-inducing ligand-dependent manners. Gastroenterology
2006;130:435–52.
42. Bantel H, Schulze-Osthoff K. Apoptosis in hepatitis C virus infection. Cell Death
Differ 2003;10(Suppl 1):S48–58.
43. Bantel H, Lugering A, Heidemann J, et al. Detection of apoptotic caspase acti-
vation in sera from patients with chronic HCV infection is associated with fibrotic
liver injury. Hepatology 2004;40:1078–87.
44. Bantel H, Lugering A, Poremba C, et al. Caspase activation correlates with the
degree of inflammatory liver injury in chronic hepatitis C virus infection. Hepatol-
ogy 2001;34:758–67.
45. Canbay A, Taimr P, Torok N, et al. Apoptotic body engulfment by a human stel-
late cell line is profibrogenic. Lab Invest 2003;83:655–63.
46. Zhang N, Hartig H, Dzhagalov I, et al. The role of apoptosis in the development
and function of T lymphocytes. Cell Res 2005;15:749–69.
47. Okazaki M, Hino K, Fujii K, et al. Hepatic Fas antigen expression before and after
interferon therapy in patients with chronic hepatitis C. Dig Dis Sci 1996;41:
2453–8.
48. Hiramatsu N, Hayashi N, Katayama K, et al. Immunohistochemical detection of
Fas antigen in liver tissue of patients with chronic hepatitis C. Hepatology 1994;
19:1354–9.
49. Iken K, Huang L, Bekele H, et al. Apoptosis of activated CD41 and CD81
T cells is enhanced by co-culture with hepatocytes expressing hepatitis C
virus (HCV) structural proteins through FasL induction. Virology 2006;346:
363–72.
50. Bataller R, Paik YH, Lindquist JN, et al. Hepatitis C virus core and nonstructural
proteins induce fibrogenic effects in hepatic stellate cells. Gastroenterology
2004;126:529–40.
51. Bataller R, Sancho B, Ginès P, et al. Activated human hepatic stellate cells ex-
press the renin-angiotensin system and synthesize angiotensin II. Gastroenterol-
ogy 2003;125:117–25.
52. Lalor PF, Shields P, Grant A, et al. Recruitment of lymphocytes to the human liver.
Immunol Cell Biol 2002;80(1):52–64.
53. Marra F. Hepatic stellate cells and the regulation of liver inflammation. J Hepatol
1999;31:1120–30.
54. Schulze-Krebs A, Preimel D, Popov Y, et al. Hepatitis C virus-replicating hepato-
cytes induce fibrogenic activation of hepatic stellate cells. Gastroenterology
2005;129:246–58.
55. Paradis V, Dargere D, Bonvoust F, et al. Effects and regulation of connective tis-
sue growth factor on hepatic stellate cells. Lab Invest 2002;82:767–74.
56. Rachfal AW, Brigstock DR. Connective tissue growth factor (CTGF/CCN2) in
hepatic fibrosis. Hepatol Res 2003;26:1–9.
response to tissue injury. J Biol Chem 2000;275:2247–50.
58. Benyon RC, Arthur MJ. Extracellular matrix degradation and the role of hepatic
stellate cells. Semin Liver Dis 2001;21:373–84.
59. Parsons CJ, Takashima M, Rippe RA. Molecular mechanisms of hepatic fibro-
genesis. J Gastroenterol Hepatol 2007;22:S79–84.
60. Gressner AM, Weiskirchen R, Breitkopf K, et al. Roles of TGF-beta in hepatic
fibrosis. Front Biosci 2002;7:d793–807.
61. Taniguchi H, Kato N, Otsuka M, et al. Hepatitis C virus core protein upregulates
transforming growth factor-beta 1 transcription. J Med Virol 2004;72:52–9.
62. Mazzocca A, Sciammetta SC, Carloni V, et al. Binding of hepatitis C virus
envelope protein E2 to CD81 up-regulates matrix metalloproteinase-2 in human
hepatic stellate cells. J Biol Chem 2005;280:11329–39.
63. Werb Z. ECM and cell surface proteolysis: regulating cellular ecology. Cell 1997;
91:439–42.
64. Maecker HT, Todd SC, Levy S. The tetraspanin superfamily: molecular facilita-
tors. FASEB J 1997;11:428–42.
65. Petracca R, Falugi F, Galli G, et al. Structure-function analysis of hepatitis C virus

envelope-CD81 binding. J Virol 2000;74:4824–30.
66. Pileri P, Uematsu Y, Campagnoli S, et al. Binding of hepatitis C virus to CD81.
Science 1998;282:938–41.
67. Mihm S, Fayyazi A, Hartmann H, et al. Analysis of histopathological
manifestations of chronic hepatitis C virus infection with respect to virus geno-
type. Hepatology 1997;25:735–9.
68. Hui JM, Kench J, Farrell GC, et al. Genotype-specific mechanisms for hepatic
steatosis in chronic hepatitis C infection. J Gastroenterol Hepatol 2002;17:873–81.
69. Rubbia-Brandt L, Fabris P, Paganin S, et al. Steatosis affects chronic hepatitis C
progression in a genotype specific way. Gut 2004;53:406–12.
70. Poynard T, Ratziu V, McHutchison J, et al. Effect of treatment with peginterferon
or interferon alfa-2b and ribavirin on steatosis in patients infected with hepatitis
C. Hepatology 2003;38:75–85.
71. Clark JM, Brancati FL, Diehl AM. Nonalcoholic fatty liver disease. Gastroenterol-
ogy 2002;122:1649–57.
72. Sanyal AJ. AGA technical review on nonalcoholic fatty liver disease. Gastroen-
terology 2002;123:1705–25.
73. Asselah T, Rubbia-Brandt L, Marcellin P, et al. Steatosis in chronic hepatitis C:
why does it really matter? Gut 2006;55:123–30.
74. Kumar D, Farrell GC, Fung C, et al. Hepatitis C virus genotype 3 is cytopathic to
hepatocytes: Reversal of hepatic steatosis after sustained therapeutic response.
Hepatology 2002;36:1266–72.
75. Shi ST, Polyak SJ, Tu H, et al. Hepatitis C virus NS5A colocalizes with the core
protein on lipid droplets and interacts with apolipoproteins. Virology 2002;292:
198–210.
76. Barba G, Harper F, Harada T, et al. Hepatitis C virus core protein shows a cyto-
plasmic localization and associates to cellular lipid storage droplets. Proc Natl
Acad Sci U S A 1997;94:1200–5.
77. Douglas MW, Goldring CE, Li K, et al. The effect of HCV genotype on steatosis –
a model for inducing core protein in hepatoma cells. In Proceedings of the 14th
International symposium on hepatitis C virus and related viruses; 2007. p. 77.
78. Perlemuter G, Sabile A, Letteron P, et al. Hepatitis C virus core protein inhibits
microsomal triglyceride transfer protein activity and very low density lipoprotein
secretion: a model of viral-related steatosis. FASEB J 2002;16:185–94.
79. Asselah T, Boyer N, Guimont MC, et al. Liver fibrosis is not associated with stea-
tosis but with necroinflammation in French patients with chronic hepatitis C. Gut
2003;52:1638–43.
80. Bjornsson E, Angulo P. Hepatitis C and steatosis. Arch Med Res 2007;38:621–7.
81. Leandro G, Mangia A, Hui J, et al. Relationship between steatosis, inflammation,
and fibrosis in chronic hepatitis C: a meta-analysis of individual patient data.
Gastroenterology 2006;130:1636–42.
82. Walsh MJ, Vanags DM, Clouston AD, et al. Steatosis and liver cell apoptosis in
chronic hepatitis C: a mechanism for increased liver injury. Hepatology 2004;39:
1230–8.
83. Hourigan LF, Macdonald GA, Purdie D, et al. Fibrosis in chronic hepatitis C
correlates significantly with body mass index and steatosis. Hepatology 1999;
29:1215–9.
84. Adinolfi LE, Gambardella M, Andreana A, et al. Steatosis accelerates the
progression of liver damage of chronic hepatitis C patients and correlates with
specific HCV genotype and visceral obesity. Hepatology 2001;33:1358–64.
85. Okuda M, Li K, Beard MR, et al. Mitochondrial injury, oxidative stress, and
antioxidant gene expression are induced by hepatitis C virus core protein.
86. Neuman MG, Benhamou JP, Malkiewicz IM, et al. Kinetics of serum cytokines
reflect changes in the severity of chronic hepatitis C presenting minimal fibrosis.
J Viral Hepat 2002;9:134–40.
87. Clouston AD, Jonsson JR, Powell EE. Steatosis as a cofactor in other liver dis-
eases: hepatitis C virus, alcohol, hemochromatosis, and others. Clin Liver Dis
2007;11:173–89, x.
88. Mehta SH, Brancati FL, Sulkowski MS, et al. Prevalence of type 2 diabetes
mellitus among persons with hepatitis C virus infection in the United States.
Ann Intern Med 2000;133:592–9.
89. Mehta SH, Brancati FL, Strathdee SA, et al. Hepatitis C virus infection and inci-
dent type 2 diabetes. Hepatology 2003;38:50–6.
90. Allison ME, Wreghitt T, Palmer CR, et al. Evidence for a link between hepatitis C
virus infection and diabetes mellitus in a cirrhotic population. J Hepatol 1994;21:
1135–9.
91. Ortiz V, Berenguer M, Rayon JM, et al. Contribution of obesity to hepatitis
C-related fibrosis progression. Am J Gastroenterol 2002;97:2408–14.
92. Monto A, Alonzo J, Watson JJ, et al. Steatosis in chronic hepatitis C: relative con-
tributions of obesity, diabetes mellitus, and alcohol. Hepatology 2002;36:
729–36.
93. Martin BC, Warram JH, Krolewski AS, et al. Role of glucose and insulin
resistance in development of type 2 diabetes mellitus: results of a 25-year fol-
low-up study. Lancet 1992;340:925–9.
94. Hui JM, Sud A, Farrell GC, et al. Insulin resistance is associated with chronic
hepatitis C virus infection and fibrosis progression [corrected]. Gastroenterology
2003;125:1695–704.
95. McCaughan GW, George J. Fibrosis progression in chronic hepatitis C virus
infection. Gut 2004;53:318–21.
96. Bugianesi E, Marchesini G, Gentilcore E, et al. Fibrosis in genotype 3 chronic
hepatitis C and nonalcoholic fatty liver disease: role of insulin resistance and
hepatic steatosis. Hepatology 2006;44:1648–55.
97. Paradis V, Perlemuter G, Bonvoust F, et al. High glucose and hyperinsulinemia
stimulate connective tissue growth factor expression: a potential mechanism in-
volved in progression to fibrosis in nonalcoholic steatohepatitis. Hepatology
2001;34:738–44.
98. Svegliati-Baroni G, Ridolfi F, Di Sario A, et al. Insulin and insulin-like growth fac-
tor-1 stimulate proliferation and type I collagen accumulation by human hepatic
stellate cells: differential effects on signal transduction pathways. Hepatology
1999;29:1743–51.
99. Ozes ON, Akca H, Mayo LD, et al. A phosphatidylinositol 3-kinase/Akt/mTOR
pathway mediates and PTEN antagonizes tumor necrosis factor inhibition of
insulin signaling through insulin receptor substrate-1. Proc Natl Acad Sci U S A
2001;98:4640–5.
100. Aytug S, Reich D, Sapiro LE, et al. Impaired IRS-1/PI3-kinase signaling in
patients with HCV: a mechanism for increased prevalence of type 2 diabetes.
Hepatology 2003;38:1384–92.
101. Kusumoto K, Uto H, Hayashi K, et al. Interleukin-10 or tumor necrosis factor-
[alpha] polymorphisms and the natural course of hepatitis C virus infection in
a hyperendemic area of Japan. Cytokine 2006;34:24–31.
102. Shintani Y, Fujie H, Miyoshi H, et al. Hepatitis C virus infection and diabetes:
direct involvement of the virus in the development of insulin resistance.
103. Cua IH, Hui JM, Bandara P, et al. Insulin resistance and liver injury in hepatitis C
is not associated with virus-specific changes in adipocytokines. Hepatology
2007;46:66–73.
104. Brunt EM. Nonalcoholic steatohepatitis: definition and pathology. Semin Liver
Dis 2001;21:3–16.
105. Edmison J, McCullough AJ. Pathogenesis of non-alcoholic steatohepatitis:
human data. Clin Liver Dis 2007;11:75–104, ix.
106. Teli MR, James OF, Burt AD, et al. The natural history of nonalcoholic fatty liver:
a follow-up study. Hepatology 1995;22:1714–9.
107. Dam-Larsen S, Franzmann M, Andersen IB, et al. Long term prognosis of fatty
liver: risk of chronic liver disease and death. Gut 2004;53:750–5.
108. Hui AY, Wong VW, Chan HL, et al. Histological progression of non-alcoholic fatty
liver disease in Chinese patients. Aliment Pharmacol Ther 2005;21:407–13.
109. Adams LA, Lymp JF, St Sauver J, et al. The natural history of nonalcoholic fatty liver
disease: a population-based cohort study. Gastroenterology 2005;129:113–21.
110. Harrison SA, Torgerson S, Hayashi PH. The natural history of nonalcoholic fatty liver
disease: a clinical histopathological study. Am J Gastroenterol 2003;98:2042–7.
111. Matteoni CA, Younossi ZM, Gramlich T, et al. Nonalcoholic fatty liver disease:
a spectrum of clinical and pathological severity. Gastroenterology 1999;116:
1413–9.
112. Chitturi S, Abeygunasekera S, Farrell GC, et al. NASH and insulin resistance:
Insulin hypersecretion and specific association with the insulin resistance syn-
drome. Hepatology 2002;35:373–9.
113. Marchesini G, Brizi M, Morselli-Labate AM, et al. Association of nonalcoholic
fatty liver disease with insulin resistance. Am J Med 1999;107:450–5.
114. Bacon BR, Farahvash MJ, Janney CG, et al. Nonalcoholic steatohepatitis: an
expanded clinical entity. Gastroenterology 1994;107:1103–9.
115. Fan J-G, Zhu J, Li X-J, et al. Prevalence of and risk factors for fatty liver in a gen-
eral population of Shanghai, China. J Hepatol 2005;43:508–14.
116. Bugianesi E, McCullough AJ, Marchesini G. Insulin resistance: a metabolic
pathway to chronic liver disease. Hepatology 2005;42:987–1000.
117. Day CP, James OF. Steatohepatitis: a tale of two ‘‘hits’’? Gastroenterology 1998;
114:842–5.
118. Bugianesi E, Zannoni C, Vanni E, et al. Non-alcoholic fatty liver and insulin
resistance: a cause-effect relationship? Dig Liver Dis 2004;36:165–73.
119. van der Poorten D, George J. Current and novel therapies for the treatment of
nonalcoholic steatohepatitis. Hepatology International 2007;1(3):342–54.
120. Hui JM, Hodge A, Farrell GC, et al. Beyond insulin resistance in NASH: TNF-
alpha or adiponectin? Hepatology 2004;40:46–54.
121. Bugianesi E, Manzini P, D’Antico S, et al. Relative contribution of iron burden,
HFE mutations, and insulin resistance to fibrosis in nonalcoholic fatty liver. Hep-
atology 2004;39:179–87.
122. Goldstein BJ, Mahadev K, Wu X. Redox paradox: insulin action is facilitated by
insulin-stimulated reactive oxygen species with multiple potential signaling tar-
gets. Diabetes 2005;54:311–21.
123. Li XL, Man K, Ng KT, et al. Insulin in UW solution exacerbates hepatic ischemia/
reperfusion injury by energy depletion through the IRS-2/SREBP-1c pathway.
Liver Transpl 2004;10:1173–82.
124. Younossi ZM, Gramlich T, Matteoni CA, et al. Nonalcoholic fatty liver disease in
patients with type 2 diabetes. Clin Gastroenterol Hepatol 2004;2:262–5.
125. Hensley K, Kotake Y, Sang H, et al. Dietary choline restriction causes complex I
dysfunction and increased H(2)O(2) generation in liver mitochondria. Carcino-
genesis 2000;21:983–9.
126. Laurent A, Nicco C, Tran Van Nhieu J, et al. Pivotal role of superoxide anion and
beneficial effect of antioxidant molecules in murine steatohepatitis. Hepatology
2004;39:1277–85.
127. Letteron P, Fromenty B, Terris B, et al. Acute and chronic hepatic steatosis lead
to in vivo lipid peroxidation in mice. J Hepatol 1996;24:200–8.
128. Leclercq IA, Farrell GC, Field J, et al. CYP2E1 and CYP4A as microsomal cata-
lysts of lipid peroxides in murine nonalcoholic steatohepatitis. J Clin Invest 2000;
105:1067–75.
129. Seki S, Kitada T, Yamada T, et al. In situ detection of lipid peroxidation and
oxidative DNA damage in non-alcoholic fatty liver diseases. J Hepatol 2002;
37:56–62.
130. Garcia-Monzon C, Martin-Perez E, Iacono OL, et al. Characterization of patho-
genic and prognostic factors of nonalcoholic steatohepatitis associated with
obesity. J Hepatol 2000;33:716–24.
131. Albano E, Mottaran E, Occhino G, et al. Review article: role of oxidative
stress in the progression of non-alcoholic steatosis. Aliment Pharmacol Ther
2005;22(Suppl 2):71–3.
132. Feldstein AE, Canbay A, Angulo P, et al. Hepatocyte apoptosis and fas
expression are prominent features of human nonalcoholic steatohepatitis. Gas-
troenterology 2003;125:437–43.
133. Ramalho RM, Cortez-Pinto H, Castro RE, et al. Apoptosis and Bcl-2 expression
in the livers of patients with steatohepatitis. Eur J Gastroenterol Hepatol 2006;
18:21–9.
134. Wieckowska A, Zein NN, Yerian LM, et al. In vivo assessment of liver cell
apoptosis as a novel biomarker of disease severity in nonalcoholic fatty liver dis-
ease. Hepatology 2006;44:27–33.
135. Feldstein A: Pathophysiology of fatty liver: implications for treatment. In Pro-
ceedings of the AASLD postgraduate course: Pathophysiologic basis for the
therapy of liver disease, p. 55–60.
136. Listenberger LL, Han X, Lewis SE, et al. Triglyceride accumulation protects
against fatty acid-induced lipotoxicity. Proc Natl Acad Sci U S A 2003;100:
3077–82.
137. Li Z, McIntyre T, Feldstein A. Free fatty acids induce hepatocyte mitochondrial
dysfunction in a lysosomal-dependent fashion: a link between steatosis and liver
injury. Gastroenterology 2006;130:A762.
138. Tilg H, Hotamisligil GS. Nonalcoholic fatty liver disease: cytokine-adipokine
interplay and regulation of insulin resistance. Gastroenterology 2006;131:934–45.
139. Malhi H, Barreyro FJ, Isomoto H, et al. Free fatty acids sensitise hepatocytes to
TRAIL mediated cytotoxicity. Gut 2007;56:1124–31.
140. Caldwell SH, Chang CY, Nakamoto RK, et al. Mitochondria in nonalcoholic fatty
liver disease. Clin Liver Dis 2004;8:595–617, x.
141. Wei Y, Wang D, Topczewski F, et al. Saturated fatty acids induce endoplasmic
reticulum stress and apoptosis independently of ceramide in liver cells.
Am J Physiol Endocrinol Metab 2006;291:E275–81.
142. Mari M, Caballero F, Colell A, et al. Mitochondrial free cholesterol loading sensi-
tizes to TNF- and Fas-mediated steatohepatitis. Cell Metab 2006;4:185–98.
143. Antonopoulos S, Mikros S, Mylonopoulou M, et al. Rosuvastatin as a novel

treatment of non-alcoholic fatty liver disease in hyperlipidemic patients. Athero-
sclerosis 2006;184:233–4.
144. Gomez-Dominguez E, Gisbert JP, Moreno-Monteagudo JA, et al. A pilot study of
atorvastatin treatment in dyslipemid, non-alcoholic fatty liver patients [see com-
ment]. Aliment Pharmacol Ther 2006;23:1643–7.
145. Puri P, Baillie RA, Wiest MM, et al. A lipidomic analysis of nonalcoholic fatty liver
disease. Hepatology 2007;46:1081–90.
146. Pajvani UB, Du X, Combs TP, et al. Structure-function studies of the adipocyte-
secreted hormone Acrp30/adiponectin. Implications fpr metabolic regulation
and bioactivity. J Biol Chem 2003;278:9073–85.
147. Yamauchi T, Kamon J, Ito Y, et al. Cloning of adiponectin receptors that mediate
antidiabetic metabolic effects. Nature 2003;423:762–9.
148. Iwaki M, Matsuda M, Maeda N, et al. Induction of adiponectin, a fat-derived
antidiabetic and antiatherogenic factor, by nuclear receptors. Diabetes 2003;
52:1655–63.
149. Ouchi N, Kihara S, Arita Y, et al. Adiponectin, an adipocyte-derived plasma
protein, inhibits endothelial NF-kappaB signaling through a cAMP-dependent
pathway. Circulation 2000;102:1296–301.
150. Musso G, Gambino R, Biroli G, et al. Hypoadiponectinemia predicts the severity
of hepatic fibrosis and pancreatic Beta-cell dysfunction in nondiabetic
nonobese patients with nonalcoholic steatohepatitis. Am J Gastroenterol 2005;
100:2438–46.
151. Targher G, Bertolini L, Rodella S, et al. Associations between plasma
adiponectin concentrations and liver histology in patients with nonalcoholic fatty
liver disease. Clin Endocrinol (Oxf) 2006;64:679–83.
152. Kaser S, Moschen A, Cayon A, et al. Adiponectin and its receptors in non-
alcoholic steatohepatitis. Gut 2005;54:117–21.
fibrosis in mice lacking adiponectin. Gastroenterology 2003;125:1796–807.
154. Xu A, Wang Y, Keshaw H, et al. The fat-derived hormone adiponectin alleviates
alcoholic and nonalcoholic fatty liver diseases in mice. J Clin Invest 2003;112(1):
91–100.
155. Ding X, Saxena NK, Lin S, et al. The roles of leptin and adiponectin: a novel
paradigm in adipocytokine regulation of liver fibrosis and stellate cell biology.
Am J Pathol 2005;166:1655–69.
156. Tsochatzis E, Papatheodoridis GV, Archimandritis AJ. The evolving role of leptin
and adiponectin in chronic liver diseases. Am J Gastroenterol 2006;101:
2629–40.
157. Mantzoros CS. The role of leptin in human obesity and disease: a review of cur-
rent evidence. Ann Intern Med 1999;130:671–80.
158. Ikejima K, Honda H, Yoshikawa M, et al. Leptin augments inflammatory and pro-
fibrogenic responses in the murine liver induced by hepatotoxic chemicals.
Hepatology 2001;34:288–97.
159. Honda H, Ikejima K, Hirose M, et al. Leptin is required for fibrogenic responses
induced by thioacetamide in the murine liver. Hepatology 2002;36:12–21.
160. Leclercq IA, Farrell GC, Schriemer R, et al. Leptin is essential for the hepatic
fibrogenic response to chronic liver injury. J Hepatol 2002;37:206–13.
161. Wang J, Leclercq I, Brymora J, et al. Kupffer cell-derived connective tissue
growth factor (CTGF) and transforming growth factor beta (TGFB) mediate the
profibrogenic effects of leptin in vitro. Hepatology 2007;46(S1):716A.
162. Saxena NK, Saliba G, Floyd JJ, et al. Leptin induces increased alpha2(I) colla-
gen gene expression in cultured rat hepatic stellate cells. J Cell Biochem 2003;
89:311–20.
163. Lang T, Ikejima K, Yoshikawa M, et al. Leptin facilitates proliferation of hepatic
stellate cells through up-regulation of platelet-derived growth factor receptor.
164. Chitturi S, Farrell G, Frost L, et al. Serum leptin in NASH correlates with hepatic
steatosis but not fibrosis: a manifestation of lipotoxicity? Hepatology 2002;36:
403–9.
165. Chalasani N, Crabb DW, Cummings OW, et al. Does leptin play a role in the
pathogenesis of human nonalcoholic steatohepatitis? Am J Gastroenterol
2003;98:2771–6.
166. Angulo P, Alba LM, Petrovic LM, et al. Leptin, insulin resistance, and liver fibrosis
in human nonalcoholic fatty liver disease. J Hepatol 2004;41:943–9.
167. Marra F, Aleffi S, Bertolani C, et al. Review article: the pathogenesis of fibrosis in
non-alcoholic steatohepatitis. Aliment Pharmacol Ther 2005;22(Suppl 2):44–7.
168. Lafontan M, Berlan M. Do regional differences in adipocyte biology provide new
pathophysiological insights? Trends Pharmacol Sci 2003;24:276–83.
169. Bergman RN, Kim SP, Catalano KJ, et al. Why visceral fat is bad: mechanisms of
the metabolic syndrome. Obesity (Silver Spring) 2006;14(Suppl 1):16S–9S.
170. Bruun JM, Lihn AS, Pedersen SB, et al. Monocyte chemoattractant protein-1
release is higher in visceral than subcutaneous human adipose tissue (AT):
implication of macrophages resident in the AT. J Clin Endocrinol Metab 2005;
90:2282–9.
171. van der Poorten D, Milner KL, Hui J, et al. Visceral fat: a key mediator of steato-
hepatitis in metabolic liver disease. Hepatology 2008;48(2):449–57.
Cy tokines a nd
Renin - Angiotensin
System Signaling
in Hepatic Fibrosis
Montserrat Moreno, PhD, Ramon Bataller, MD*
KEYWORDS
Liver fibrosis Cytokines Angiotensin II
Hepatic stellate cells Renin-angiotensin system
Hepatic fibrosis is the wound healing response of the liver to repeated injury.1 Fibrosis
is the result of a complex interplay among resident hepatic cells, infiltrating inflamma-
tory cells, and several locally acting peptides called cytokines. Cytokines are a family
of proteins that function as mediators of cell communication.2 They include chemo-
kines, interleukins, interferons, growth factors, angiogenic factors, vasoactive sub-
stances, soluble receptors, and soluble proteases. Unregulated cytokine synthesis
and release coordinate the hepatic response to injury and participate in the initiation,
progression, and maintenance of fibrosis. Understanding the complexity of the cyto-
kine-driven mechanisms of fibrosis is important for identifying potential molecular tar-
gets for future pharmacologic interventions in prevention and treatment. Key
mediators include transforming growth factor b1 (TGF-b1), platelet derived growth
factor (PDGF), adipokines, and several inflammatory cytokines and chemokines.
The cellular source of cytokines in liver diseases probably depends on the type of
disease. In chronic viral diseases, infected hepatocytes and infiltrating lymphocytes
release reactive oxygen species (ROS), inflammatory chemokines, and fibrogenic me-
diators.3 In alcoholic liver disease, damaged hepatocytes, Kupffer cells, and infiltrating
neutrophils secrete large amounts of ROS and cytokines, such as tumor necrosis fac-
tor-a (TNF-a) and interleukin (IL)-8, favoring hepatocellular death and myofibroblast
accumulation.4 Recent data indicate that adipokines also play an important role in liver
fibrogenesis.5,6 They are locally produced by liver resident cells (eg, activated hepatic
This work was supported by a grant from the Institut d’Investigacions Biomèdiques August Pi i
Sunyer (IDIBAPS) and the Mintisterio de Ciencia y Tecnologıa (SAF 2005 06245).
Liver Unit, Institut Clınic de Malalties Digestives i Metabòliques, Hospital Clınic, Institut d’Inves-
tigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Villarroel 170, 08036-Barcelona, Catalonia,
Spain
E-mail address: bataller@clinic.ub.es (R. Bataller).
Clin Liver Dis 12 (2008) 825–852

826 Moreno & Bataller
stellate cells [HSCs]) and amplify inflammatory and fibrogenic signals in fibrogenic
myofibroblats.
The past few years have seen an explosion of knowledge about signal transduction
pathways in liver fibrogenesis, involving virtually all events in tissue repair, such as
myofibroblast accumulation, hepatocyte regeneration, and scar tissue formation.6
Most of these pathways have been identified in cultured HSCs, the main target cell
for fibrogenic cytokines. Most importantly, drugs interfering with intracellular path-
ways involved in increased collagen production are considered potential therapies
for liver fibrosis.
Accumulating evidence indicates that the renin–angiotensin system (RAS) is a major
mediator in liver fibrogenesis.7 Key components of the RAS are locally expressed in
chronically injured livers and activated HSCs de novo generate angiotensin II, the
main effector peptide of this system.8 Angiotensin II induces an array of fibrogenic ac-
tions in activated HSCs, including cell proliferation, migration, secretion of proinflam-
matory cytokines, and collagen synthesis.9,10 These actions are largely mediated by
ROS generated by a nonphagocytic form of nicotinamide adenine dinucleotide phos-
phate (NADPH) oxidase.1 Pharmacologic or genetic ablation of the RAS attenuates ex-
perimental liver fibrosis.11–14 Currently, RAS inhibitors are being tested as antifibrotic
drugs in patients who have chronic liver diseases.
This article provides a succinct and current overview of cytokines implicated in liver
fibrogenesis and the signaling pathways involved, and describes the role of the RAS
and angiotensin II.
CYTOKINES INVOLVED IN LIVER FIBROGENESIS
Hepatic fibrosis is regulated by host of factors, including interactions with the extracel-
lular matrix, surface of inflammatory cells, hormones, and an extremely complex and
redundant network of profibrogenic cytokines.1,15 The nature of mechanisms through
which cytokines regulate fibrosis is dual: indirect, through attraction of inflammatory
cells, and direct, through binding to specific receptors on myofibroblasts and stimu-
lating proliferation, collagen production, and secretion of autocrine factors.
Main cytokines involved in liver fibrogenesis are depicted in Table 1. They include
classical proinflammatory cytokines and chemokines and growth factors. Moreover,
vasoactive substances traditionally considered hormones regulating arterial pressure
homeostasis are currently viewed as true cytokines that participate in the wound heal-
ing response to injury.16
Adipokines are another family of cytokines that are locally produced in the liver and
regulate liver fibrosis.5 All of these mediators do not work alone but rather in a complex
network of intracellular signaling and interaction with cells and extracellular matrix
components.
Inflammatory Cytokines
During liver inflammation and fibrosis, secretion of cytokines is dysregulated, promot-
ing an inflammatory state. Potential sources of inflammatory cytokines in the hepatic
wound healing response are Kupffer cells, hepatocytes, HSCs, natural killer cells, and
lymphocytes, including CD41 T helper (Th).1 Th cells can differentiate into Th1 and Th2
subsets, a classification that is based on the pattern of cytokines produced. In general,
Th1 cells produce cytokines that promote cell-mediated immunity (interferon [IFN]-g,
TNF-a, and IL-2) and protect against fibrosis, whereas Th2 cells promote humoral im-
munity (IL-4, IL-5, IL-6, and IL-13) and induce fibrosis, as evidenced by a study using
Cytokines and Renin-Angiotensin System 827
two mice strains with different polarity of Th cells and different susceptibility to liver
fibrosis.17
Classic cytokines may be divided into chemokines (monocyte chemotactin protein 1
[MCP-1], RANTES, IL-8), interferons (IFN-a, IFN-g) and interleukins (IL-1, IL-6, IL-10).
Chemokines are divided into four groups depending on the spacing of their first cys-
teine residues: CC (eg, MCP-1, RANTES), CXC (eg, IL-8, GRO-a), C (lymphotactins),
and CX3C (fractalkine).18 TNF-a participates in the activation process of HSCs and
is a critical factor for the proinflammatory role of HSCs.19,20 Another powerful cytokine,
IL-1, also exerts profibrogenic actions by stimulating metalloproteinase secretion.21
Administration of IL-1 receptor antagonist reduces matrix deposition in a rat model
of liver fibrosis.22 In contrast, IL-10 exerts net antifibrogenic effects in the liver in
vivo23 and in vitro.24 IFN-a has been shown to exert a direct antifibrotic effect in vitro
over HSCs25 and in vivo in different animal models of hepatic fibrosis, and is used as
antiviral therapy in patients who have chronic hepatitis C.26–28 IFN-g inhibits HSC pro-
liferation and procollagen mRNA expression in vitro and reduces liver fibrosis in
rodents.29,30
Among CC chemokines, MCP-1 is a profibrogenic chemokine overexpressed in the
injured liver.31 It induces chemotaxis of HSCs32 and participates in experimentally in-
duced fibrosis in rats.33 RANTES is produced by HSCs, is up-regulated in livers of pa-
tients who have HCV, and induces HSC proliferation in vitro.34,35 CXC chemokines are
also involved in liver fibrosis, as evidenced by several studies. For instance, serum IL-8
is increased in alcoholic patients36 and those who have nonalcoholic steatohepatitis
(NASH), and primary biliary cirrhosis,37 and GRO-a expression is up-regulated in livers
of patients who have alcoholic hepatitis.17,38
Growth Factors
Growth factors play a key role in liver fibrogenesis by promoting activation and accu-
mulation of HSCs and stimulating collagen synthesis. PDGF and TGF-b are the most
important mediators because of their effects on HSC proliferation and extracellular
matrix protein production, respectively. PDGF is a dimeric protein composed of two
polypeptide chains (mainly A and B) that can combine to form PDGF-A and PDGF-
B.39,40 They signal through the tyrosine kinase receptors PDGF receptor a and b
(PDGFRa and PDGFRb, respectively). PDGFb is the most potent mitogen factor for
HSCs by acting through PDGFRb.41 All isoforms of PDGF and PDGFR are up-regu-
lated in injured livers and correlate with the degree of inflammation and fibrosis.42–44
Moreover, inhibition of PDGF-B attenuates experimental liver fibrogenesis.45,46
TGF-b1 is a key mediator of liver fibrogenesis. In the injured liver, TGF-b1 is up-reg-
ulated47 and favors the transition of resident HSCs into myofibroblast-like cells, stim-
ulating synthesis of extracellular matrix proteins and inhibiting their degradation.
Strategies aimed at disrupting TGF-b1 synthesis or signaling pathways markedly
decreased fibrosis in experimental models.48,49 A newly discovered regulator of
TGF-b activity is bone morphogenic protein and activin membrane-bound inhibitor
(BAMBI), which is a TGF-b pseudoreceptor that inhibits TGF-b signaling by preventing
the formation of receptor complexes.50 Down-regulation of BAMBI seems to be
a mechanism of fibrogenesis induced by lipopolysaccharide through toll-like receptor
(TLR)-4.51
Established angiogenic growth factors such as vascular endothelial growth factor
(VEGF) and fibroblast growth factor (FGF) play a central role in not only angiogenesis
but also chronic wound-healing conditions. VEGF and its receptors (VEGFR-1 and
VEGFR-2) are up-regulated in chronic liver injury and promote fibrogenic effects in
HSCs by stimulating cell proliferation, collagen production, and migration.52
828
Moreno & Bataller
Table 1
Main cytokines regulating liver fibrogenesis
Expressed by Hepatic Effect in Hepatic

Stellate Cells Stellate Cells Effect on Animal Models in Rodents
Inflammatory cytokines
MCP-1 Yes Profibrogenic Blockade of MCP-1 suppresses hepatic fibrosis
RANTES Yes Profibrogenic ?
IL-1 ? Profibrogenic Antagonism of IL-1 reduces liver fibrosis
IL-4 ? Controversial ?
IL-8 Yes ? ?
IL-10 Yes Antifibrogenic IL-10 treatment reduces hepatic fibrosis
IL-13 ? Controversial ?
TNF-a Yes Controversial Treatment with an anti–TNF-a agent reduces liver fibrosis
IFN-g ? Antifibrogenic IFN-g treatment attenuates liver fibrosis
IFN-a ? ? IFN-a attenuates liver fibrosis
Growth factors
TGF-b1 Yes Profibrogenic Inhibition of TGF-b1 signaling decreases liver fibrosis
PDGF-BB Yes Profibrogenic Inhibition of PDGF-B prevents liver fibrosis
CTGF Yes Profibrogenic Down-regulation of CTGF by siRNA reduces liver fibrosis
VEGF Yes Profibrogenic Antibodies against VEGF receptors inhibit liver fibrosis
FGF-1 and -2 ? Profibrogenic Double knockout exhibit reduced liver fibrosis
HGF Yes Antifibrogenic HGF gene therapy attenuates progression of liver fibrosis
IGF-1 Yes Antifibrogenic IGF-1 treatment attenuates liver fibrosis
Vasoactive substances
Angiotensin II Yes Profibrogenic RAS inhibition attenuates liver fibrosis
Endothelin-1 Yes Profibrogenic? ETA receptor antagonism reduces hepatic fibrosis
Vasopressin ? Profibrogenic ?
Thrombin Yes Profibrogenic Thrombin antagonism reduces liver fibrosis
Norepinephrine Yes Profibrogenic a-adrenoreceptor blocker treatment attenuates liver fibrosis
Nitric oxide Yes Antifibrogenic Mice lacking iNOS exhibit exacerbated liver fibrosis
Relaxin ? Antifibrogenic Relaxin administration reduces hepatic fibrosis
Prostaglandin E2 Yes Antifibrogenic Hepatoprotection in the liver
Adrenomedullin Yes Antifibrogenic ?
Atrial natriuretic peptide Yes Antifibrogenic ?
Adipokines
Leptin Yes Profibrogenic Leptin deficient rats do not develop liver fibrosis
Resistin ? Profibrogenic? ?
Adiponectin Yes Antifibrogenic Adiponectin-deficient mice show enhanced liver fibrosis
Others
Tetrahydrocannabinol Yes Profibrogenic CB1 receptor antagonism reduces liver fibrosis
Cytokines and Renin-Angiotensin System

Abbreviations: CTGF, connective tissue growth factor; ETA, endothelin A; HGF, hepatocyte growth factor, IGF-1, insulin-like growth factor 1; iNOS, inducible nitric
oxide synthase; MCP-1, monocyte chemotactin protein 1; siRNA, small interfering RNA strands; VEGF, vascular endothelial growth factor.
829
Moreover, VEGFR-1 and VEGFR-2 signaling is required for liver fibrosis develop-
ment.53 FGF-1 and -2 exert profibrogenic effects in vivo as double knockout mice ex-
hibit reduced liver fibrosis.54 Other growth factors, including hepatocyte growth factor
(HGF) and insulin-like growth factor 1 (IGF-1), attenuate liver fibrosis in rodents.55–57
Vasoactive Substances
Several vasoactive substances are locally produced in the injured liver and have an
autocrine or paracrine effect on HSCs. Vasodilator substances (eg, nitric oxide, pros-
taglandin E2, atrial natriuretic peptide, adrenomedullin, and relaxin) exert antifibrotic
effects, whereas vasoconstrictors (eg, endothelin-1, norepinephrine, angiotensin II,
and thrombin) have opposite effects.1
Livers with advanced fibrosis have a predominance of vasoconstrictors compared
with vasodilators, favoring collagen deposition. Among vasodilatory substances, nitric
oxide has received special attention. It is produced by all nonparenchymal cells and
inhibits liver fibrosis in vitro and in vivo. Advanced fibrosis is associated with endothe-
lial dysfunction and decreased nitric oxide production, which may contribute to dis-
ease progression.58
Prostaglandin E2 is a vasodilatory molecule synthesized by virtually all liver cells that
inhibits HSC proliferation and TGF-b1–mediated collagen synthesis and attenuates fi-
brosis in vivo.59,60 Drugs delivering either NO or prostaglandin E2 have been proposed
to treat patients who have liver fibrosis. Among vasoconstrictors, angiotensin II is the
most widely studied and is discussed extensively later.
Thrombin is a multifunctional serine protease that binds to specific cell surface re-
ceptors called protease-activated receptors (PAR). Thrombin is produced by activated
HSCs to regulate cell migration, growth, and fibrogenic actions. Both thrombin and
PAR-1, its main receptor, are overexpressed in fibrotic livers. Moreover, antagonism
of thrombin attenuates liver fibrosis in animal models.61–63
Endothelin is another important vasoconstrictor implicated in liver fibrosis. Three
isoforms of endothelin1–3 act through two receptors (ETA and ETB). Endothelin and
its receptors are up-regulated in the fibrotic liver and their expression correlates
with the severity of the disease.64 In the early phase of activation, HSCs have most
ETA receptors, which stimulate an increase in intracellular-free calcium in HSCs cou-
pled with cell contraction and proliferation. This process is linked to stimulation of
fibrogenesis. In later stages, ETB receptors become more abundant and their stimula-
tion promotes an antiproliferative effect. The use of ETA/ETB receptor blockers have
yielded conflicting results,16,65 possibly because of the different relative activities to-
ward each of the two receptors.
Norepinephrine is a catecholamine with a dual role as a neurotransmitter and a hor-
mone. Evidence indicates that norepinephrine stimulates liver fibrogenesis. Activated
HSCs are capable of secreting mature norepinephrine, which induces proinflamma-
tory and fibrogenic effects. Moreover, a1 adrenoreceptors are up-regulated in livers
with advanced fibrosis and its blockade attenuates the development of liver fibrosis
in rats with chronic liver injury.66,67
Adipokines
Adipokines are biologically active peptides mainly secreted by adipose tissue. Main
adipokines include leptin, resistin, visfatin, and adiponectin. Circulating adipokines se-
creted by excessive fat accumulation may regulate hepatic fibrosis in diseases such
as NASH.5,68 Moreover, several adipokines are locally synthesized in the liver and
may regulate fibrogenesis in an autocrine/paracrine manner. Leptin is secreted by ac-
tivated HSCs and stimulates cell proliferation, secretion of chemokines, and collagen
synthesis. Moreover, leptin is required for fibrosis development.69,70 In contrast, adi-

ponectin markedly inhibits liver fibrogenesis in vitro and in vivo.71 Resistin is up-regu-
lated in alcoholic liver disease and exerts proinflammatory effects on HSCs,
suggesting a role in liver fibrogenesis.68
Endogenous Cannabinoids
In addition to their central effects, cannabinoids display a wide variety of peripheral
functions, including regulation of wound healing response to injury. The endogenous
cannabinoid system has been implicated in liver fibrosis. Both CB1 and CB2 receptors
and endocannabinoids are up-regulated in chronic liver diseases.72 Pharmacologic or
genetic inactivation of CB1 reduces fibrosis in different models of chronic liver injury.73
In contrast, activation of CB2 receptors attenuates liver injury, inflammation, and ox-
idative stress, and CB2 knockout mice exposed to carbon tetrachloride (CCl4) show
enhanced liver fibrosis.74 Globally, cannabinoids may worsen liver injury because daily
cannabis use exacerbates liver fibrosis in patients who have chronic hepatitis C.75
CYTOKINE SIGNALING PATHWAYS INVOLVED IN LIVER FIBROGENESIS
All molecules implicated in liver fibrosis activate different intracellular pathways

(Fig. 1). A complex cross-talk exists between them that determines the global effect
on liver fibrosis. Data on intracellular pathways regulating liver fibrogenesis are mainly
derived from studies using cultured HSCs, whereas understanding of their role in vivo
is progressing through experimental fibrogenesis studies using knockout mice.
PI3K/Akt Pathway
The focal adhesion kinase (FAK) phosphoinositol-3-phosphate kinase (PI3K)/Akt–sig-
naling pathway mediates various profibrogenic actions in HSCs, including prolifera-
tion, chemotaxis, and transcription of profibrogenic genes.76,77 This pathway may
be activated by growth factors that trigger tyrosine kinase activity (PDGF, VEGF) or ac-
tivation of cytokine receptors (MCP-1), but also by other signals, including integrins,
stimulators of G-protein-coupled receptors (angiotensin II, thrombin), and adipokines
(leptin).10,78,79 As an example, when PDGF binds to its receptor (a tyrosine kinase re-
ceptor), the receptor dimerizes and autophosphorylates.80 Then, PI3K associates with
the activated receptor and becomes activated by phosphorylation. PI3K activation re-
sults in the phosphorylation of inositol lipids.
Phosphatase and tensin homolog (PTEN) functions as an antagonist of PI3K,
thereby impairing the generation of phosphoinositol-3,4,5-triphosphate (PIP3) from
phosphoinositol-4,5-biphosphate (PIP2).81 The phosphoinositols bind to Akt and in-
duce its translocation to the plasma membrane where it becomes phosphorylated
by the phosphoinositide-dependent kinase, and thus activated. Activated Akt induces
mammalian target of rapamycin (mTOR) activity, and signals through mTOR increase
the phosphorylation of p70S6 kinase, which phosphorylates a ribosomal subunit and
4E-BP1 leading to up-regulation of protein synthesis and stimulation of cell growth
signals. Rapamycin, which inhibits mTOR activity, attenuates liver fibrosis, possibly
through decreasing growth of HSCs.82 Moreover, inhibition of PI3-K by wortmannin
blocks mitogenesis and chemotaxis in response to PDGF, supporting the involvement
of this pathway in HSC accumulation in vivo.77
Mitogen-Activated Protein Kinase Pathway

Members of the mitogen-activated protein kinases (MAPK) family, including extracel-
lular-regulated kinase (ERK), c-Jun N-terminal kinase (JNK), and p38 MAPK, are
Fig. 1. Main cytokine signaling pathways in the hepatic stellate cell. After binding of cyto-
kines such as TGF-b1, PDGF, or TNF-a to their cell surface receptors, activation of several
intracellular signaling pathways occurs. The PDGF stimulation can induce activation of mito-
gen-activated protein kinase (MAPK) signaling and the phosphatidylinositol 3-kinase/Akt/
p70S6 kinase (PI3K/Akt/p70S6K) signaling pathway. The matrix-associated focal adhesion ki-
nase also stimulates the PI3K/Akt/p70S6K signaling pathway. TGF-b1 stimulates transcription
of profibrogenic genes through activating the Smad signaling pathway. Fatty acids and
other agonists activate peroxisome proliferator–activated receptors to regulate gene ex-
pression. The Wnt/b-catenin pathway is also involved in transcriptional regulation. Bacterial
products such as lipopolysaccharide bind to TLR4 and stimulate IL-1 associated kinase to in-
duce fibrogenic signals. TNF-a binds to the protein TNFR-associated death domain to acti-
vate c-Jun N-terminal kinase and nuclear factor-kB. Finally, agonists such as prostaglandin
E2 induce cAMP production and protein kinase A activation to inhibit MAPK signaling.
activated by several growth factors and vasoactive peptides and subsequently trans-
located to the nucleus where they phosphorylate several transcription factors, result-
ing in cellular responses.83 In HSCs, ERKs regulate cell proliferation, secretion of
chemokines, cell migration, and collagen synthesis. This pathway is basically acti-
vated by peptides that induce proliferation (PDGF, thrombin, angiotensin II, VEGF,
and leptin) and by chemokines.10,80,84 On activation, tyrosine kinase and G-protein–
coupled receptors recruit the signaling molecule Ras, causing the sequential activa-
tion of Raf, MEK, and ERK1 and -2.
Activated ERK induces activation of transcription factors implicated in cell prolifer-
ation, such as activating protein type-1 (AP-1). ERK activation is induced in vivo in rats
with chronic liver injury and after chronic exposure to angiotensin II.85,86 In HSCs, JNK
or stress-activated protein kinase (SAPK) is activated in response to cellular stress,
bacterial products, FasL, oxidative stress, vasoactive substances (angiotensin II), adi-
pokines (leptin), chemokines (RANTES), and growth factors (TGF-b1 and PDGF).78
JNK activity is regulated upstream by MAPK kinase 4 (MKK4) and MAPK kinase 7
(MKK7); it is a profibrogenic pathway in HSCs through modulating cell growth and
secretion of inflammatory cytokines.10,87–89
MAPK kinase 6 (MKK6) and MKK3 act directly upstream p38 MAPK and lead to the
phosphorylation of p38, which subsequently regulates gene expression through acti-
vating transcription factors. The p38 pathway seems to have an antiproliferative role in
HSCs, because blocking p38 activity increases cellular proliferation89 Moreover, p38
activity has profibrogenic effects, because collagen expression induced by TGF-b or
other molecules is partially mediated by p38 MAPK signaling in HSCs.90,91
Smad Pathway
Smad pathway plays a major role in liver fibrosis through signaling TGF-b1 in activated
HSC16397841. TGF-b1–dependent Smad signaling also mediates other fibrogenic
factors, such as hypoxia.92 TGF-b1 binds to its type II receptor that becomes activated
and dimerized with type I receptor. Smad2 and -3 bind the resulting complex to
become phosphorylated, and are then released to the cytosol and associate with
Smad4. The heterotrimer translocates into the nucleus and activates profibrogenic
transcription factors (eg, Sp1), which bind to the promoter region of target genes.93
Smad6 and -7 are endogenous inhibitors of Smad signaling through preventing the
binding of Smad2 and -3 to the TGF-b receptor.
Inhibitory Smad proteins mediate the effect of IFN-g on TGF-b signaling in the
liver.94 Smad signaling participates in TGF-b1–dependent mesenchymal-to-epithelial
transition in hepatocytes, a novel mechanism implicated in liver fibrogenesis.95 In vivo,
Smad signaling mediates liver fibrogenesis induced by chronic cholestasis,96 and in-
hibition of Smad signaling suppresses collagen gene expression and hepatic fibrosis
in mice.97
Nuclear Factor-kB Signaling

Nuclear factor-kB (NF-kB) is a major downstream effector of proinflammatory cyto-
kines such as TNF-a.98 Other peptides, such as angiotensin II and leptin, also activate
NF-kB signaling.91 NF-kB is a transcription factor composed of homo- or hetero-
dimers of the Rel protein family (p65, p50, p52, c-Rel, and RelB). NF-kB activity in
the cytoplasm is regulated by its inhibitor, IkBa. After IkBa degradation, the active
form of NF-kB translocates into the nucleus where it regulates transcriptional activity
of target genes. After HSC activation, NF-kB becomes persistently activated and
many NF-kB–responsive genes (eg, IL-6, intercellular adhesion molecule-1, ICAM-1)
are constitutively expressed.99,100 NF-kB plays a pivotal role in the inflammatory ef-
fects of TNF-a and other mediators on HSCs. Its activity is not required for proliferation
or activation, but it protects activated HSCs against TNF-a–induced apoptosis.100
Nuclear Receptors
Nuclear receptors can directly bind to DNA and regulate the expression of adjacent
genes, and therefore are classified as transcription factors. Several nuclear receptors
have been described in HSCs. The pregnane X receptor (PXR) is a nuclear receptor
that seems to exert an antifibrotic role. Pregnane X receptor activators inhibit the pro-
liferation, transdifferentiation, and expression of TGF-b1 in HSCs.101,102 In addition,
treatment with a PXR activator markedly reduces the degree of liver fibrosis in animal
models.103
Peroxisome proliferator–activated receptors (PPARs) regulate HSC’s biologic ac-
tions and are potential targets for antifibrotic therapy.104 Three isoforms are encoded
by three different genes: PPARa, PPARb, and PPARg. Fatty acids and eicosanoids
bind to PPAR, which dimerizes with the retinoid receptor and migrates into the nu-
cleus, where they bind to peroxisome proliferator response elements in the promoter
region of target genes and recruit cofactors that modulate transcriptional activity.
PPARs regulate mainly metabolic functions in the liver but also inflammation and fibro-
genesis. After HSC activation, expression of PPARg diminishes as PPARb expression
increases. PPARg activation inhibits the proinflammatory and profibrogenic actions in
HSCs and attenuate liver fibrosis in vivo,105,106 whereas PPARb seems to exert oppo-
site effects.107 The mechanisms involve attenuation of TGF-b signaling in HSC.108
Most importantly, PPARg ligands (eg, thiazolidinediones) are currently being tested
to treat liver fibrosis in the context of NASH.
Wnt/b-Catenin Pathway
The Wnt/b-catenin pathway is crucial in normal development, including embryogene-
sis. This pathway also signals cytokines and promotes inflammation.109 It was recently
implicated in hepatic fibrogenesis 17,464,972. Wnt is an extracellular-secreted glyco-
protein that binds to the cell surface receptor Frizzled (Fz) and induces specific down-
stream events.110 In a normal state, the monomeric form of b-catenin in the cytoplasm
is targeted for degradation by ubiquitinitation, keeping free levels of b-catenin low and
preventing it from translocating to the nucleus to induce target gene transcription.
When any Wnt proteins bind to their seven-transmembrane receptor, a complex
cascade of reactions occurs until b-catenin becomes hypophosphorylated and re-
leased and translocated into the nucleus, where it binds to T-cell factor/lymphoid-en-
hancing factor. Once formed, this complex transcriptionally regulates several target
genes. In the liver, evidence indicates that Wnt signaling has a profibrogenic role. In
cultured activated HSCs, mRNA for Wnt genes and coreceptors increase and protect
cells from apoptosis.111 Moreover, Wnt activity is enhanced in liver cirrhosis. These
observations suggest that Wnt signaling promotes hepatic fibrosis through enhancing
HSC activation and survival.111,112
CD14/TLR-4 Pathway
Toll-like receptors (TLRs) are pattern recognition receptors that recognize pathogen-
associated molecular patterns and signal through adaptor molecules, including mye-
loid differentiation factor 88 (MyD88) and TRIF-related adaptor molecule (TRAM), to
activate transcription factors, NF-kB, and interferon regulatory factors (IRFs), leading
to innate immunity. Although Kupffer cells are considered the primary cells to respond
to pathogen-associated molecular patterns in the liver, recent studies show TLR sig-
naling in hepatic nonimmune cell populations, including hepatocytes and hepatic stel-
late cells.113
Recent studies suggest a role for intracellular pathways driven by TLRs in liver in-
flammation.114 In particular, TLR4 is implicated in liver fibrogenesis and lipopolysac-
charide signaling. Mice lacking TLR4 have a reduced liver fibrosis compared with
wild-type mice. The mechanism showing a role for TLR4 in liver fibrogenesis was re-
cently uncovered. TLR4 activation in HSCs reduces BAMBI expression, which is
a TGF-b pseudoreceptor, and therefore TLR4 activation enhances TGF-b signaling
in HSCs.51 The intracellular domain of TLR is similar to that of IL-1 receptor, and
thus they share intracellular pathways. Stimulated Toll/IL-1 receptors activate
MyD88, and then the receptor recruits IL-1–associated kinase that becomes acti-
vated.115 This process leads to phosphorylation of the TNF receptor–associated factor
(TRAF), which then activates proinflammatory transcription factors (AP-1 and NF-kB).
Death Pathways
Several pathways implicated in cell death mediate cytokine signaling in activated HSCs
(see article by Gieling and Mann elsewhere in this issue). TNF receptors (TNFR) belong
to a superfamily that includes several transmembrane molecules that bind cytokines
and other molecules.116 Receptors with a dead domain include TNFR1, Fas, and p75
receptor for nerve growth factor. TNFR2 and CD40 lack the death domain.
TNFR1 plays a major role in mediating biologic actions of TNF-a.117 In HSCs, TNF-a
activates pathways that regulate gene transcription and inflammation and other path-
ways leading to cell death.118 Binding of TNF-a induces homotrimerization of TNFR1,
which binds to the death domain containing protein TNFR-associated death domain
(TRADD). It associates with receptor-interacting protein and TNFR-associated fac-
tor-2 (TRAF2) to activate NF-kB and JNK, respectively.
TNF-a is a critical factor for the proinflammatory role of HSCs. Quiescent HSCs ex-
press TNFR1, and TNF binds to the cell surface. However, the receptor in quiescent
cells seems to be only partially functional, because activity of NF-kB in response to
TNF-a is only seen in activated HSCs. TNF-a activates JNK both in quiescent and ac-
tivated HSC. TNF-a also activates ERK1/2 and p38 MAPK, which regulates collagen
synthesis in HSC.119 Other receptor, CD40, interacts with its ligand to amplify the in-
flammatory behavior of HSC through TRAF2- and IKK2-dependent pathways.87 Cell
death is mediated by the interaction of TRADD with Fas-associated protein with
dead domain (FADD), which stimulates caspases leading to apoptotic cell death.
Fas (CD95) is also expressed in quiescent HSCs and drives proliferation and resis-
tance to apoptosis.120 Another ligand, TNF-related apoptosis-inducing ligand (TRAIL)
binds to TRAIL receptor 2 in activated HSCs to induce apoptosis.121
JAK/STAT Pathway
Janus kinases (JAKs) can bind to both tyrosine receptors and G-protein–coupled re-
ceptors. They phosphorylate tyrosine residues on the receptor and create sites for in-
teraction with proteins that contain phosphotyrosine-binding SH2 domain. Signal
transducer and activator of transcription (STAT) proteins possessing SH2 domain
are recruited to the receptors and are phosphorylated at tyrosine residues by JAKs.
Activated STAT dimers accumulate in the cell nucleus and activate transcription of
their target genes. In HSCs, this pathway is stimulated by various cytokines and me-
diators, including INF-g and leptin.122 Activation of STAT1 plays an important role in
liver injury, inflammation, and inhibition of liver regeneration.123 Mice lacking STAT1
exhibit accelerated liver fibrosis from inhibition of HSC proliferation, suppression of
PDGF expression, and inhibition of TGF-b/Smad3 signaling.124
AMP-Activated Protein Kinase Pathway

AMP-activated protein kinase (AMPK) is a fuel-sensing enzyme that can cellular me-
tabolism in response to different stimuli. Once activated, AMPK activates catabolic
pathways, leading to ATP generation, and inactivates ATP-consuming processes
not essential for short-term survival. AMP-activated protein inhibits cell proliferation,
migration, chemokine secretion, and collagen production in HSCs.125 This pathway
mediates the antifibrogenic effect of adiponectin in HSCs.126
THE RENIN^ANGIOTENSIN SYSTEM AND LIVER FIBROSIS

General Concepts
The RAS has been traditionally considered an endocrine system that regulates arterial
pressure homeostasis.127 According to this concept, the precursor angiotensinogen is
synthesized by the hepatocytes and released into the bloodstream, where it is trans-
formed to angiotensin I by renin. Angiotensin I is then cleaved to angiotensin II by an-
giotensin-converting enzyme (ACE), an ectoenzyme abundant in endothelial cells.
Angiotensin II binds to angiotensin type 1 (AT1) receptors to induce vasoconstriction
and, in glomerulosa cells, to release aldosterone, which causes sodium and water
reabsorption in the kidneys (Fig. 2).
Besides this endocrine action, the RAS components are expressed in damaged tis-
sues that de novo generate mature angiotensin II.128 Key enzymes for local synthesis
of angiotensin II include ACE type 1 (ACE-1) and chymase. Angiotensin II accumulates
at the sites of parenchymal injury and binds to AT1 receptors in myofibroblasts to pro-
mote recruitment of inflammatory cells, secretion of extracellular matrix proteins, and
inhibition of collagen degradation.129 Moreover, angiotensin II regulates the local
microcirculation through inducing contraction of vascular cell types.130 Angiotensin
II also binds to angiotensin type 2 (AT2) receptors that are typically found in many or-
gans during embryogenesis and are re-expressed in chronically inflamed tissues.131
These receptors oppose the actions of AT1 receptors by inducing vasodilatation
and tissue growth inhibition.132 ACE type 2 is overexpressed in tissues with fibrosis
and converts angiotensin II into angiotensin,1–7 a smaller peptide with vasodilatory ac-
tions that counteracts the actions of angiotensin II.133
The RAS is currently viewed as part of a system of interconnected cytokines that
become activated after tissue injury to promote tissue repair.134 This new understand-
ing of the RAS has important clinical implications. It explains why blockade of the RAS
with ACE inhibitors, the newer AT1 receptor antagonists, or both together significantly
Angiotensinogen
renin
Angiotensin-I Chymase
ACE ACE-1
inhibitors
AT1-R Angiotensin-II Angiotensin 1-7

blockers ACE-2
AT1-R AT2-R
Sodium retention Development

Vasoconstriction Apoptosis
Tissue growth Vasodilation
Oxidative stress Growth inhibition
Fibrogenesis
Inflammation Aldosterone
MC-R antagonists
MC-R
Sodium retention
Fibrogenesis
Fig. 2. The renin–angiotensin pathway and related pathogenic actions. ACE, angiotensin-
concerting enzyme; AT1-R, angiotensin type 1 receptor; AT2-R, angiotensin type 2 receptor;
MC-R, mineralocorticoid receptor.
slows the progression of many fibrotic diseases. This antifibrotic effect has been
shown in coronary heart disease, heart failure, diabetic nephropathy, and stroke.135
The beneficial effect of RAS inhibitors in reducing morbidity and mortality in these pa-
tients is not necessarily associated with a reduction in arterial pressure, indicating that
angiotensin II induces tissue injury through mechanisms other than arterial
hypertension.
Role of the Renin-Angiotensin System in Liver Fibrosis: Intracellular Mechanisms

The role of RAS in liver fibrosis was first suggested when HSCs were found to bear
functional AT1 receptors.9,136 After phenotypic activation, HSCs highly express AT1
receptors, the activation of which mediates cell contraction, migration, and prolifera-
tion. Moreover, angiotensin II stimulates collagen synthesis and TGF-b1 expression.
Angiotensin II also induces proinflammatory effects in HSCs, stimulating the expres-
sion of cell adhesion molecules and the secretion of inflammatory chemokines.9 All
these effects are blocked by AT1 receptor antagonists and are blunted in mouse
HSCs lacking AT1a receptors.137,138
An important finding is that an intrahepatic RAS is expressed in the fibrotic liver.139
Although angiotensinogen is the only component of the RAS expressed in the normal
rat liver, ACE and AT1 are markedly expressed in fibrotic rat livers. In humans, ACE
and chymase are up-regulated in livers with significant fibrosis, whereas AT1 receptor
expression is shifted to fibrotic areas.38,140,141 The cellular source of the RAS in the in-
jured liver is uncertain. In other tissues (eg, heart), myofibroblasts accumulated at the
areas of tissue remodeling express all components of the RAS and generate angioten-
sin II, which participates in the tissue repair process. In the human liver, quiescent
HSCs neither express the RAS components nor secrete angiotensin II.8 In contrast,
after cell activation in culture and in vivo, myofibroblastic HSCs express key compo-
nents of the RAS and generate mature angiotensin II.
The molecular mechanisms mediating the inflammatory and fibrogenic effects of
angiotensin II in HSC have been partially uncovered (Fig. 3).10,142,143 Angiotensin II in-
creases intracellular calcium concentration and induces ROS formation, stimulating
key intracellular pathways, such as PI3k/AKT, Rho kinase, and MAPKs. It increases
AP-1 DNA binding, whereas NF-kB activity is stimulated in rat but not human HSCs.
Angiotensin II also stimulates the Smad signaling pathway through up-regulation of
TGF-b1.142 As a consequence, angiotensin II stimulates the expression of numerous
genes involved in hepatic tissue remodeling, such as extracellular matrix components,
inflammatory cytokines, and collagenolysis inhibitors.
The stimulation of intracellular signaling pathways and the biologic actions stimu-
lated by angiotensin II are redox-sensitive. In HSCs, a nonphagocytic form of NADPH
oxidase mediates angiotensin II–induced ROS formation. NADPH oxidases present in
vascular cell types are constitutively active, producing relatively low levels of ROS un-
der basal conditions and generating higher levels of oxidants in response to cytokines
such as angiotensin II, stimulating redox-sensitive intracellular pathways.144 Disrup-
tion of active NADPH oxidase protects mice from developing severe fibrosis after
bile duct ligation, indicating that NADPH oxidase plays an important role in liver
fibrosis.10
The most convincing evidence supporting a role for the RAS in experimental liver
fibrosis is the finding that blockade of the generation of angiotensin II or its binding
to AT1 receptors markedly attenuates experimental liver fibrosis. Remarkably, at least
27 studies using different experimental models of liver fibrosis have yielded similar re-
sults (Table 2).11–14,141,143,145–161
Source of Ang II Locally generated Ang II Systemic Ang II
Increased Ang II
AT1 receptors
Receptors in HSC
NADPH oxidase PKC
MAP kinases Calcium channels

Signaling PI3K/AKT IP3 receptors
pathways
NFκB
Intracellular calcium
AP-1
Procollagen α1(I)
Gene PAI-1
transcription TIMP-1
MCP-1/RANTES
Cell functions INCREASED COLLAGEN SYNTHESIS CELL CONTRACTION

DECREASED COLLAGEN DEGRADATION
INFLAMMATION
GROWTH/MIGRATION
Clinical
consequences LIVER FIBROSIS PORTAL HYPERTENSION
Fig. 3. Mechanisms of the pathogenic effect of the renin–angiotensin system in the liver. In-
creased angiotensin II binds to AT1 receptors located in activated HSCs. AT1 receptors acti-
vate a nonphagocytic NADPH oxidase to generate ROS that stimulate redox-sensitive
intracellular pathways. Increased gene transcription leads to mitogenic, fibrogenic, and
inflammatory properties, promoting fibrogenesis. Angiotensin II increases intracellular
calcium and induces cell contraction, increasing intrahepatic vascular resistance and partici-
pating in the pathogenesis of portal hypertension. AP-1, activating protein type-1; MAPk,
mitogen-activated protein kinase; MCP-1, monocyte chemotactic protein type 1; NF-kB, nu-
clear factor kB; PI3k, phosphoinositol 3 kinase; PAI-1, plasminogen activator inhibitor type 1;
PKC, protein kinase C; TIMP-1, tissue inhibitor of metalloproteinases type 1.
RAS inhibition is associated with a decrease in TGF-b1 expression in the injured

liver, a key effector mediating the fibrogenic effect of angiotensin II in other tissues.162
Moreover, it causes a decrease in connective tissue growth factor and AT1 receptor
expression.149 RAS inhibition prevents the accumulation of myofibroblasts positive
for smooth muscle a–actin.13,14,149,154,156,159,160 A role for AT1 receptors in liver fibro-
sis is also supported by the finding that mice lacking AT1a receptors are protected to
develop liver fibrosis after prolonged bile duct ligation.137,138 In contrast, mice lacking
AT2 receptors show increased susceptibility to liver fibrosis.150
The Renin-Angiotensin System and Hepatic Fibrosis: Clinical Implications

Although the experimental evidence supporting a role for the RAS in liver fibrogenesis
is overwhelming, clinical evidence is limited. Components required to synthesize an-
giotensin II, such as ACE-1 and chymase, are expressed in the livers of patients who
have alcoholic liver disease and chronic hepatitis C.38,163 A polymorphism in the an-
giotensinogen gene that confers increased angiotensin II synthesis influences fibrosis
progression in patients who have chronic hepatitis B, hepatitis C, and NASH.157,164,165
Few reports have investigated the potential antifibrotic effect of RAS inhibition in pa-
tients who have chronic liver diseases. Preliminary uncontrolled studies suggest that
losartan, an AT1 receptor blocker, may attenuate fibrosis progression in patients who
have chronic hepatitis C or NASH.147,166–168 Moreover, a retrospective study in pa-
tients who underwent transplantation and experienced hepatitis C recurrence showed
that those receiving RAS inhibitors as antihypertensive therapy showed less fibrosis
progression than patients receiving other types of drugs.169 This study is potentially
relevant, because fibrosis progression is aggressive in patients who underwent trans-
plantation who experienced hepatitis C recurrence and is the main cause of graft
loss.170
The rationale supporting the use of RAS inhibitors is that they markedly attenuate
experimentally induced liver fibrosis and are safe and effective in preventing renal or
cardiac fibrosis progression in patients who have type II diabetes and arterial hyper-
tension.7,171 Moreover, RAS inhibitors are reasonably inexpensive and can be safely
administered for prolonged periods. However, clinical evidence supporting their use
in patients who have liver diseases is only based on pilot studies that included
a low number of patients and retrospective data. Randomized clinical trials are needed
before this approach can be recommended.
The target population for clinical trials is patients who have chronic liver diseases for
which the causative agent cannot be removed (eg, chronic hepatitis C not responding
to antiviral therapy, chronic cholestatic conditions). Alcohol-induced liver fibrosis and
NASH, conditions associated with marked oxidative stress, are also potential targets
for RAS blockade. The expected benefits in these patients include decreased fibrosis
progression and decreased inflammation. As a result, RAS inhibitors may slow pro-
gression to advanced fibrosis and therefore prevent development of portal hyperten-
sion and related complications. The duration of antifibrotic therapy to show changes in
liver fibrosis should be considered, depending on the rate of fibrosis progression of the
underlying disease. Obviously, patients who have undergone liver transplantation rep-
resent the ideal population to evaluate the efficacy of antifibrotic drugs, including RAS
inhibitors.
Finally, experts have suggested that the RAS may participate in the development
and progression of hepatocellular carcinoma through promoting fibrosis and angio-
genesis, respectively.160 Although experimental studies indicate that RAS inhibition
prevents liver carcinogenesis,160 no clinical data support this hypothesis. Whether
this approach is useful in patients who have advanced cirrhosis is unknown and
should be evaluated in clinical trials.
Side effects are a potential limitation for the use of ACE inhibitors in patients who
have chronic liver diseases. The antifibrotic profile of ACE inhibitors and AT1 antago-
nists is similar, but AT1 antagonists are usually better tolerated. Hepatotoxicity can oc-
cur, although its incidence in patients who have chronic hepatitis is unknown.172 RAS
inhibitors can cause arterial hypotension or renal impairment in patients who have ad-
vanced cirrhosis and subsequent activation of the systemic RAS. In this clinical set-
ting, the efficacy of RAS inhibitors is probably very limited.
840
Moreno & Bataller
Table 2
Studies assessing the role of the renin-angiotensin system in liver fibrosis in rats: proposed mechanisms
Reference RAS Inhibitor Experimental Model Proposed Mechanism

Ramos et al.153 Captopril Pig serum Decreased mast cell accumulation
Ohishi et al.151 Lisinopril CCl4 Decreased stimulation of HSCs
Decreased TGF-b expression
Jonsson et al.13 Captopril BDL Decreased TGF-b expression
Regulation MMPs/TIMPs
Yoshiji et al.159 Candesartan Pig-serum Reduced aSMA–positive cells
Perindopril Decreased TGF-b expression
Paizis et al.152 Irbesartan BDL Decreased TGF-b expression
AT1 down-regulation
Wei et al.156 Losartan CCl4 Reduced aSMA–positive cells
Ramalho et al.14 Losartan BDL Reduced aSMA–positive cells
Croquet et al.145 Losartan CCl4 Not assessed
Yoshiji et al.173 PerindoprilCandesartan Pig serum Decreased TIMP-1 expression
Li et al.174 Perindopril CCl4 Decreased MMP2,9 expression
Decreased TGF-b, NF-kB
Ueki et al.175 Candesartan BDL Decreased CTGF expression
Yoshiji et al.176 Perindopril CCl4 Decreased aSMA expression
Li et al.174 Perindopril CCl4 Decreased MCP-1 expression
Candesartan Decreased TGF-b expression
Xu et al.158 Perindopril CCl4 Decreased Smad3 expression
Valsartan
Kurikawa et al.149 Olmesartan BDL Reduced aSMA–positive cells
Decreased CTGF expression
Effect in HSCs
Tuncer et al.154 Candesartan CCl4 Reduced aSMA-positive cells
Captopril
Ibanez et al.148 Losartan CDD Reduced TGF-b expression
El Demerdash et al.146 Lisinopril CCl4 Reduced oxidative stress
Losartan
Kitamura et al.143 TCV-116 CCD Decreased activation Rho kinase
Turkay et al.155 Enalapril BDL Decreased TGF-b/MMP2 expression
Hirose et al.11 Olmesartan CDD Decreased TGF-b expression
Rseduced aSMA–positive cells
Jin et al.12 Telmisartan CDD Decreased TIMP-1/MMP13 expression
Nabeshima et al.150 Losartan CCl4 Decreased TGF-b and TNF-a expression
Abbreviations: aSMA, a-smooth muscle actin; BDL, bile duct ligation; CCl4, carbon tetrachloride; CDD, choline deficient diet; MMP, matrix metallopeptidase;
TIMP-1, tissue inhibitor of metalloproteinases type 1.
Cytokines and Renin-Angiotensin System

841
CONCLUSIONS
Hepatic wound healing response to injury involves several cytokines that regulate my-
ofibroblast proliferation, angiogenesis, and collagen synthesis. Dysregulated cytokine
synthesis contributes to the initiation and progression of fibrosis. Multiple signaling
pathways are activated by cytokines liver fibrogenesis. Most of these pathways
have been identified in cultured HSCs, the main target cell for fibrogenic cytokines.
Understanding the complexity of the cytokine-driven mechanisms of fibrosis and
related signaling pathways is important for identifying potential molecular targets for
future pharmacologic interventions in prevention and treatment. A growing body of
evidence indicates that the RAS plays a key role in liver fibrogenesis. Angiotensin II
exerts prooxidant, fibrogenic, and proinflammatory actions in the liver. Although the
molecular mechanisms underlying the fibrogenic effect of angiotensin II in the liver
are unknown, NADPH oxidase–derived ROS seem to play an important role. Although
preliminary clinical data suggest that RAS inhibition can be useful as an antifibrotic
therapy in patients who have chronic liver diseases, randomized clinical trials are
needed before this approach can be routinely recommended.
SUMMARY
Hepatic fibrosis is the result of a complex interplay between resident hepatic cells, in-
filtrating inflammatory cells, and several locally acting peptides called cytokines. Dys-
regulated cytokine release underlies the hepatic response to injury and participates in
the initiation, progression, and maintenance of fibrosis. Key mediators include
TGF-b1, vasoactive substances, adipokines, and several inflammatory cytokines
and chemokines.
Multiple signal transduction pathways are involved in cytokine signaling. Most path-
ways have been identified in cultured hepatic stellate cells, the main target cell for
fibrogenic cytokines. Drugs interfering intracellular pathways involved in increased
collagen production are potential therapies for liver fibrosis. Accumulating evidence
indicates that angiotensin II, the main effector of the RAS, is a true cytokine that plays
a major role in liver fibrosis. An intrahepatic RAS is expressed in chronically damaged
livers, and angiotensin II induces an array of fibrogenic actions in hepatic stellate cells,
including increased collagen synthesis and secretion of inflammatory mediators.
These effects are mediated by NADPH oxidase–generated ROS and are prevented
by angiotensin type 1 receptor blockers.
Inhibition of the RAS markedly attenuates experimentally induced liver fibrosis in ro-
dents. Preliminary studies in patients who have chronic hepatitis C and NASH suggest
that RAS blocking agents may have beneficial effects on fibrosis progression. This ar-
ticle outlines the main cytokines involved in liver fibrosis, including angiotensin II, and
describes the signaling pathways involved.
REFERENCES

2. Saile B, Ramadori G. Inflammation, damage repair and liver fibrosis—role of
cytokines and different cell types. Z Gastroenterol 2007;45(1):77–86.
3. Schuppan D, Krebs A, Bauer M, et al. Hepatitis C and liver fibrosis. Cell Death
Differ 2003;10(Suppl 1):S59–67.
4. Reuben A. Alcohol and the liver. Curr Opin Gastroenterol 2008;24(3):328–38.
5. Marra F, Aleffi S, Bertolani C, et al. Adipokines and liver fibrosis. Eur Rev Med
Pharmacol Sci 2005;9(5):279–84.
6. Pinzani M, Marra F. Cytokine receptors and signaling in hepatic stellate cells.

Semin Liver Dis 2001;21(3):397–416.
7. Bataller R, Sancho-Bru P, Gines P, et al. Liver fibrogenesis: a new role for the
renin-angiotensin system. Antioxid Redox Signal 2005;7(9–10):1346–55.
8. Bataller R, Sancho-Bru P, Gines P, et al. Activated human hepatic stellate cells
express the renin-angiotensin system and synthesize angiotensin II. Gastroen-
terology 2003;125(1):117–25.
9. Bataller R, Gines P, Nicolas JM, et al. Angiotensin II induces contraction and
proliferation of human hepatic stellate cells. Gastroenterology 2000;118(6):
1149–56.
10. Bataller R, Schwabe RF, Choi YH, et al. NADPH oxidase signal transduces an-
giotensin II in hepatic stellate cells and is critical in hepatic fibrosis. J Clin Invest
2003;112(9):1383–94.
11. Hirose A, Ono M, Saibara T, et al. Angiotensin II type 1 receptor blocker inhibits
fibrosis in rat nonalcoholic steatohepatitis. Hepatology 2007;45(6):1375–81.
12. Jin H, Yamamoto N, Uchida K, et al. Telmisartan prevents hepatic fibrosis and
enzyme-altered lesions in liver cirrhosis rat induced by a choline-deficient
L-amino acid-defined diet. Biochem Biophys Res Commun 2007;364(4):801–7.
13. Jonsson JR, Clouston AD, Ando Y, et al. Angiotensin-converting enzyme inhibi-
tion attenuates the progression of rat hepatic fibrosis. Gastroenterology 2001;
121(1):148–55.
14. Ramalho LN, Ramalho FS, Zucoloto S, et al. Effect of losartan, an angiotensin II
antagonist, on secondary biliary cirrhosis. Hepatogastroenterology 2002;49(48):
1499–502.
15. Friedman SL. Mechanisms of hepatic fibrogenesis. Gastroenterology 2008;
134(6):1655–69.
16. Rockey DC, Chung JJ. Endothelin antagonism in experimental hepatic fibrosis.
Implications for endothelin in the pathogenesis of wound healing. J Clin Invest
1996;98(6):1381–8.
17. Shi Z, Wakil AE, Rockey DC. Strain-specific differences in mouse hepatic wound
healing are mediated by divergent T helper cytokine responses. Proc Natl Acad
Sci U S A 1997;94(20):10663–8.
18. Galliera E, Corsi MM, Bonecchi R, et al. Chemokines as pharmacological tar-
gets. Mini Rev Med Chem 2008;8(7):638–46.
19. Hernandez-Munoz I, de la TP, Sanchez-Alcazar JA, et al. Tumor necrosis factor
alpha inhibits collagen alpha 1(I) gene expression in rat hepatic stellate cells
through a G protein. Gastroenterology 1997;113(2):625–40.
20. Eng FJ, Friedman SL. Transcriptional regulation in hepatic stellate cells. Semin
Liver Dis 2001;21(3):385–95.
21. Han YP, Zhou L, Wang J, et al. Essential role of matrix metalloproteinases in in-
terleukin-1-induced myofibroblastic activation of hepatic stellate cell in collagen.
J Biol Chem 2004;279(6):4820–8.
22. Mancini R, Benedetti A, Jezequel AM. An interleukin-1 receptor antagonist
decreases fibrosis induced by dimethylnitrosamine in rat liver. Virchows Arch
1994;424(1):25–31.
23. Zhang LJ, Zheng WD, Chen YX, et al. Antifibrotic effects of interleukin-10 on
experimental hepatic fibrosis. Hepatogastroenterology 2007;54(79):2092–8.
24. Zhang LJ, Chen YX, Chen ZX, et al. Effect of interleukin-10 and platelet-derived
growth factor on expressions of matrix metalloproteinases-2 and tissue inhibitor
of metalloproteinases-1 in rat fibrotic liver and cultured hepatic stellate cells.
World J Gastroenterol 2004;10(17):2574–9.
25. Giannelli G, Bergamini C, Marinosci F, et al. Antifibrogenic effect of IFN-alpha2b

on hepatic stellate cell activation by human hepatocytes. J Interferon Cytokine
Res 2006;26(5):301–8.
26. Chang XM, Chang Y, Jia A. Effects of interferon-alpha on expression of hepatic
stellate cell and transforming growth factor-beta1 and alpha-smooth muscle ac-
tin in rats with hepatic fibrosis. World J Gastroenterol 2005;11(17):2634–6.
27. Inagaki Y, Nemoto T, Kushida M, et al. Interferon alfa down-regulates collagen
gene transcription and suppresses experimental hepatic fibrosis in mice. Hep-
atology 2003;38(4):890–9.
28. Wu CS, Piao XX, Piao DM, et al. Treatment of pig serum-induced rat liver fibrosis
with Boschniakia rossica, oxymatrine and interferon-alpha. World J Gastroenterol
2005;11(1):122–6.
29. Shen H, Zhang M, Minuk GY, et al. Different effects of rat interferon alpha, beta
and gamma on rat hepatic stellate cell proliferation and activation. BMC Cell Biol
2002;3:9.
30. Yoshiji H, Kuriyama S, Noguchi R, et al. Combination of interferon-beta and an-
giotensin-converting enzyme inhibitor, perindopril, attenuates the murine liver
fibrosis development. Liver Int 2005;25(1):153–61.
31. Marra F, DeFranco R, Grappone C, et al. Increased expression of monocyte che-
motactic protein-1 during active hepatic fibrogenesis: correlation with monocyte
infiltration. Am J Pathol 1998;152(2):423–30.
32. Marra F, Romanelli RG, Giannini C, et al. Monocyte chemotactic protein-1 as a che-
moattractant for human hepatic stellate cells. Hepatology 1999;29(1):140–8.
33. Tsuruta S, Nakamuta M, Enjoji M, et al. Anti-monocyte chemoattractant protein-1
gene therapy prevents dimethylnitrosamine-induced hepatic fibrosis in rats. Int J
Mol Med 2004;14(5):837–42.
34. Nischalke HD, Nattermann J, Fischer HP, et al. Semiquantitative analysis of intra-
hepatic CC-chemokine mRNas in chronic hepatitis C. Mediators Inflamm 2004;
13(5–6):357–9.
35. Schwabe RF, Bataller R, Brenner DA. Human hepatic stellate cells express
CCR5 and RANTES to induce proliferation and migration. Am J Physiol Gastro-
intest Liver Physiol 2003;285(5):G949–58.
36. Swiatkowska-Stodulska R, Bakowska A, Drobinska-Jurowiecka A. Interleukin-8
in the blood serum of patients with alcoholic liver disease. Med Sci Monit
2006;12(5):CR215–20.
37. Yamashiki M, Kosaka Y, Nishimura A, et al. Analysis of serum cytokine levels in pri-
mary biliary cirrhosis patients and healthy adults. J Clin Lab Anal 1998;12(2):77–82.
38. Colmenero J, Bataller R, Sancho-Bru P, et al. Hepatic expression of candidate
genes in patients with alcoholic hepatitis: correlation with disease severity. Gas-
troenterology 2007;132(2):687–97.
39. Li X, Ponten A, Aase K, et al. PDGF-C is a new protease-activated ligand for the
PDGF alpha-receptor. Nat Cell Biol 2000;2(5):302–9.
40. LaRochelle WJ, Jeffers M, McDonald WF, et al. PDGF-D, a new protease-acti-
vated growth factor. Nat Cell Biol 2001;3(5):517–21.
41. Pinzani M, Gesualdo L, Sabbah GM, et al. Effects of platelet-derived growth fac-
tor and other polypeptide mitogens on DNA synthesis and growth of cultured rat
liver fat-storing cells. J Clin Invest 1989;84(6):1786–93.
42. Borkham-Kamphorst E, van Roeyen CR, Ostendorf T, et al. Pro-fibrogenic poten-
tial of PDGF-D in liver fibrosis. J Hepatol 2007;46(6):1064–74.
43. Pinzani M, Milani S, Herbst H, et al. Expression of platelet-derived growth factor

and its receptors in normal human liver and during active hepatic fibrogenesis.
Am J Pathol 1996;148(3):785–800.
44. Wong L, Yamasaki G, Johnson RJ, et al. Induction of beta-platelet-derived
growth factor receptor in rat hepatic lipocytes during cellular activation in vivo
and in culture. J Clin Invest 1994;94(4):1563–9.
45. Borkham-Kamphorst E, Stoll D, Gressner AM, et al. Antisense strategy against
PDGF B-chain proves effective in preventing experimental liver fibrogenesis. Bi-
ochem Biophys Res Commun 2004;321(2):413–23.
46. Borkham-Kamphorst E, Herrmann J, Stoll D, et al. Dominant-negative soluble
PDGF-beta receptor inhibits hepatic stellate cell activation and attenuates liver
fibrosis. Lab Invest 2004;84(6):766–77.
47. De Bleser PJ, Niki T, Rogiers V, et al. Transforming growth factor-beta gene ex-
pression in normal and fibrotic rat liver. J Hepatol 1997;26(4):886–93.
48. Shek FW, Benyon RC. How can transforming growth factor beta be targeted use-
fully to combat liver fibrosis? Eur J Gastroenterol Hepatol 2004;16(2):123–6.
49. Chen HN, Fan S, Weng CF. Down-regulation of TGFbeta1 and leptin ameliorates
thioacetamide-induced liver injury in lipopolysaccharide-primed rats. J Endo-
toxin Res 2007;13(3):176–88.
50. Onichtchouk D, Chen YG, Dosch R, et al. Silencing of TGF-beta signalling by the
pseudoreceptor BAMBI. Nature 1999;401(6752):480–5.
51. Seki E, De Minicis S, Osterreicher CH, et al. TLR4 enhances TGF-beta signaling
and hepatic fibrosis. Nat Med 2007;13(11):1324–32.
52. Novo E, Cannito S, Zamara E, et al. Proangiogenic cytokines as hypoxia-depen-
dent factors stimulating migration of human hepatic stellate cells. Am J Pathol
2007;170(6):1942–53.
53. Yoshiji H, Kuriyama S, Yoshii J, et al. Vascular endothelial growth factor and re-
ceptor interaction is a prerequisite for murine hepatic fibrogenesis. Gut 2003;
52(9):1347–54.
54. Yu C, Wang F, Jin C, et al. Role of fibroblast growth factor type 1 and 2 in carbon
tetrachloride-induced hepatic injury and fibrogenesis. Am J Pathol 2003;163(4):
1653–62.
55. Canturk NZ, Canturk Z, Ozden M, et al. Protective effect of IGF-1 on experimen-
tal liver cirrhosis-induced common bile duct ligation. Hepatogastroenterology
2003;50(54):2061–6.
56. Sanz S, Pucilowska JB, Liu S, et al. Expression of insulin-like growth factor I by
activated hepatic stellate cells reduces fibrogenesis and enhances regeneration
after liver injury. Gut 2005;54(1):134–41.
57. Xia JL, Dai C, Michalopoulos GK, et al. Hepatocyte growth factor attenuates liver
fibrosis induced by bile duct ligation. Am J Pathol 2006;168(5):1500–12.
58. Rockey DC, Chung JJ. Reduced nitric oxide production by endothelial cells in
cirrhotic rat liver: endothelial dysfunction in portal hypertension. Gastroenterol-
ogy 1998;114(2):344–51.
59. Hui AY, Dannenberg AJ, Sung JJ, et al. Prostaglandin E2 inhibits transforming
growth factor beta 1-mediated induction of collagen alpha 1(I) in hepatic stellate
cells. J Hepatol 2004;41(2):251–8.
60. Koide S, Kobayashi Y, Oki Y, et al. Prostaglandin E2 inhibits platelet-derived
growth factor-stimulated cell proliferation through a prostaglandin E receptor
EP2 subtype in rat hepatic stellate cells. Dig Dis Sci 2004;49(9):1394–400.
61. Fiorucci S, Antonelli E, Distrutti E, et al. PAR1 antagonism protects against ex-
perimental liver fibrosis. Role of proteinase receptors in stellate cell activation.
Hepatology 2004;39(2):365–75.
62. Duplantier JG, Dubuisson L, Senant N, et al. A role for thrombin in liver fibrosis.
Gut 2004;53(11):1682–7.
63. Rullier A, Gillibert-Duplantier J, Costet P, et al. Protease-activated receptor 1
knockout reduces experimentally induced liver fibrosis. Am J Physiol Gastroint-
est Liver Physiol 2008;294(1):G226–35.
64. Pinzani M, Milani S, De Franco R, et al. Endothelin 1 is overexpressed in human
cirrhotic liver and exerts multiple effects on activated hepatic stellate cells. Gas-
troenterology 1996;110(2):534–48.
65. Poo JL, Jimenez W, Maria MR, et al. Chronic blockade of endothelin receptors in
cirrhotic rats: hepatic and hemodynamic effects. Gastroenterology 1999;116(1):
161–7.
66. Oben JA, Yang S, Lin H, et al. Norepinephrine and neuropeptide Y promote pro-
liferation and collagen gene expression of hepatic myofibroblastic stellate cells.
Biochem Biophys Res Commun 2003;302(4):685–90.
67. Sancho-Bru P, Bataller R, Colmenero J, et al. Norepinephrine induces calcium
spikes and proinflammatory actions in human hepatic stellate cells. Am J Physiol
Gastrointest Liver Physiol 2006;291(5):G877–84.
68. Bertolani C, Sancho-Bru P, Failli P, et al. Resistin as an intrahepatic cytokine:
overexpression during chronic injury and induction of proinflammatory actions
in hepatic stellate cells. Am J Pathol 2006;169(6):2042–53.
69. Marra F. Leptin and liver fibrosis: a matter of fat. Gastroenterology 2002;122(5):
1529–32.
70. Ikejima K, Takei Y, Honda H, et al. Leptin receptor-mediated signaling regulates
hepatic fibrogenesis and remodeling of extracellular matrix in the rat. Gastroen-
terology 2002;122(5):1399–410.
fibrosis in mice lacking adiponectin. Gastroenterology 2003;125(6):1796–807.
72. Siegmund SV, Schwabe RF. Endocannabinoids and liver disease. II. Endocan-
nabinoids in the pathogenesis and treatment of liver fibrosis. Am J Physiol Gas-
trointest Liver Physiol 2008;294(2):G357–62.
73. Teixeira-Clerc F, Julien B, Grenard P, et al. CB1 cannabinoid receptor antagonism:
a new strategy for the treatment of liver fibrosis. Nat Med 2006;12(6):671–6.
74. Julien B, Grenard P, Teixeira-Clerc F, et al. Antifibrogenic role of the cannabinoid
receptor CB2 in the liver. Gastroenterology 2005;128(3):742–55.
75. Hezode C, Roudot-Thoraval F, Nguyen S, et al. Daily cannabis smoking as a risk
factor for progression of fibrosis in chronic hepatitis C. Hepatology 2005;42(1):
63–71.
76. Gabele E, Reif S, Tsukada S, et al. The role of p70S6K in hepatic stellate cell col-
lagen gene expression and cell proliferation. J Biol Chem 2005;280(14):
13374–82.
77. Reif S, Lang A, Lindquist JN, et al. The role of focal adhesion kinase-phospha-
tidylinositol 3-kinase-akt signaling in hepatic stellate cell proliferation and type I
collagen expression. J Biol Chem 2003;278(10):8083–90.
78. Aleffi S, Petrai I, Bertolani C, et al. Upregulation of proinflammatory and proan-
giogenic cytokines by leptin in human hepatic stellate cells. Hepatology 2005;
42(6):1339–48.
79. Lechuga CG, Hernandez-Nazara ZH, Hernandez E, et al. PI3K is involved in

PDGF-beta receptor upregulation post-PDGF-BB treatment in mouse HSC. Am
J Physiol Gastrointest Liver Physiol 2006;291(6):G1051–61.
80. Pinzani M. PDGF and signal transduction in hepatic stellate cells. Front Biosci
2002;7:d1720–6.
81. Zhang X, Jin B, Huang C. The PI3K/Akt pathway and its downstream transcrip-
tional factors as targets for chemoprevention. Curr Cancer Drug Targets 2007;
7(4):305–16.
82. Zhu J, Wu J, Frizell E, et al. Rapamycin inhibits hepatic stellate cell proliferation
in vitro and limits fibrogenesis in an in vivo model of liver fibrosis. Gastroenterol-
ogy 1999;117(5):1198–204.
83. Goldsmith ZG, Dhanasekaran DN. G protein regulation of MAPK networks.
Oncogene 2007;26(22):3122–42.
84. Bonacchi A, Romagnani P, Romanelli RG, et al. Signal transduction by the che-
mokine receptor CXCR3: activation of Ras/ERK, Src, and phosphatidylinositol 3-
kinase/Akt controls cell migration and proliferation in human vascular pericytes.
J Biol Chem 2001;276(13):9945–54.
85. Bataller R, Gabele E, Schoonhoven R, et al. Prolonged infusion of angiotensin II
into normal rats induces stellate cell activation and proinflammatory events in
liver. Am J Physiol Gastrointest Liver Physiol 2003;285(3):G642–51.
86. Marra F, Arrighi MC, Fazi M, et al. Extracellular signal-regulated kinase activation
differentially regulates platelet-derived growth factor’s actions in hepatic stellate
cells, and is induced by in vivo liver injury in the rat. Hepatology 1999;30(4):
951–8.
87. Schwabe RF, Schnabl B, Kweon YO, et al. CD40 activates NF-kappa B and c-
Jun N-terminal kinase and enhances chemokine secretion on activated human
hepatic stellate cells. J Immunol 2001;166(11):6812–9.
88. Schwabe RF, Uchinami H, Qian T, et al. Differential requirement for c-Jun NH2-
terminal kinase in TNFalpha- and Fas-mediated apoptosis in hepatocytes.
FASEB J 2004;18(6):720–2.
89. Schnabl B, Bradham CA, Bennett BL, et al. TAK1/JNK and p38 have opposite
effects on rat hepatic stellate cells. Hepatology 2001;34(5):953–63.
90. Tsukada S, Westwick JK, Ikejima K, et al. SMAD and p38 MAPK signaling path-
ways independently regulate alpha1(I) collagen gene expression in unstimu-
lated and transforming growth factor-beta-stimulated hepatic stellate cells.
J Biol Chem 2005;280(11):10055–64.
91. Tsukada S, Parsons CJ, Rippe RA. Mechanisms of liver fibrosis. Clin Chim Acta
2006;364(1–2):33–60.
92. Shi YF, Fong CC, Zhang Q, et al. Hypoxia induces the activation of human he-
patic stellate cells LX-2 through TGF-beta signaling pathway. FEBS Lett 2007;
581(2):203–10.
93. Cohen MM Jr. TGF beta/Smad signaling system and its pathologic correlates.
Am J Med Genet A 2003;116A(1):1–10.
94. Weng H, Mertens PR, Gressner AM, et al. IFN-gamma abrogates profibrogenic
TGF-beta signaling in liver by targeting expression of inhibitory and receptor
Smads. J Hepatol 2007;46(2):295–303.
95. Kaimori A, Potter J, Kaimori JY, et al. Transforming growth factor-beta1 induces
an epithelial-to-mesenchymal transition state in mouse hepatocytes in vitro.
J Biol Chem 2007;282(30):22089–101.
96. Seyhan H, Hamzavi J, Wiercinska E, et al. Liver fibrogenesis due to cholestasis

is associated with increased Smad7 expression and Smad3 signaling. J Cell Mol
Med 2006;10(4):922–32.
97. Inagaki Y, Kushida M, Higashi K, et al. Cell type-specific intervention of trans-
forming growth factor beta/Smad signaling suppresses collagen gene expres-
sion and hepatic fibrosis in mice. Gastroenterology 2005;129(1):259–68.
98. Schwabe RF, Brenner DA. Mechanisms of liver injury. I. TNF-alpha-induced liver
injury: role of IKK, JNK, and ROS pathways. Am J Physiol Gastrointest Liver
Physiol 2006;290(4):G583–9.
99. Hellerbrand C, Jobin C, Licato LL, et al. Cytokines induce NF-kappaB in activated
but not in quiescent rat hepatic stellate cells. Am J Phys 1998;275(2 Pt 1):
G269–78.
100. Lang A, Schoonhoven R, Tuvia S, et al. Nuclear factor kappaB in proliferation,
activation, and apoptosis in rat hepatic stellate cells. J Hepatol 2000;33(1):
49–58.
101. Haughton EL, Tucker SJ, Marek CJ, et al. Pregnane X receptor activators inhibit
human hepatic stellate cell transdifferentiation in vitro. Gastroenterology 2006;
131(1):194–209.
102. Wright MC. The impact of pregnane X receptor activation on liver fibrosis. Bio-
chem Soc Trans 2006;34(Pt 6):1119–23.
103. Marek CJ, Tucker SJ, Konstantinou DK, et al. Pregnenolone-16alpha-carbonitrile
inhibits rodent liver fibrogenesis via PXR (pregnane X receptor)-dependent and
PXR-independent mechanisms. Biochem J 2005;387(Pt 3):601–8.
104. Hellemans K, Michalik L, Dittie A, et al. Peroxisome proliferator-activated recep-
tor-beta signaling contributes to enhanced proliferation of hepatic stellate cells.
105. Marra F, Efsen E, Romanelli RG, et al. Ligands of peroxisome proliferator-acti-
vated receptor gamma modulate profibrogenic and proinflammatory actions in
hepatic stellate cells. Gastroenterology 2000;119(2):466–78.
106. Galli A, Crabb DW, Ceni E, et al. Antidiabetic thiazolidinediones inhibit collagen
synthesis and hepatic stellate cell activation in vivo and in vitro. Gastroenterol-
ogy 2002;122(7):1924–40.
107. Hellemans K, Rombouts K, Quartier E, et al. PPARbeta regulates vitamin A me-
tabolism-related gene expression in hepatic stellate cells undergoing activation.
J Lipid Res 2003;44(2):280–95.
108. Zhao C, Chen W, Yang L, et al. PPARgamma agonists prevent TGFbeta1/
Smad3-signaling in human hepatic stellate cells. Biochem Biophys Res Com-
mun 2006;350(2):385–91.
109. Gustafson B, Smith U. Cytokines promote Wnt signaling and inflammation and
impair the normal differentiation and lipid accumulation in 3T3-L1 preadipo-
cytes. J Biol Chem 2006;281(14):9507–16.
110. Bowerman B. Cell signaling. Wnt moves beyond the canon. Science 2008;
320(5874):327–8.
111. Myung SJ, Yoon JH, Gwak GY, et al. Wnt signaling enhances the activation and
survival of human hepatic stellate cells. FEBS Lett 2007;581(16):2954–8.
112. Cheng JH, She H, Han YP, et al. Wnt antagonism inhibits hepatic stellate cell ac-
tivation and liver fibrosis. Am J Physiol Gastrointest Liver Physiol 2008;294(1):
G39–49.
113. Schwabe RF, Seki E, Brenner DA. Toll-like receptor signaling in the liver. Gastro-
enterology 2006;130(6):1886–900.
114. Seki E, Brenner DA. Toll-like receptors and adaptor molecules in liver disease:
Update. Hepatology 2008;48(1):322–35.
115. Paik YH, Schwabe RF, Bataller R, et al. Toll-like receptor 4 mediates inflamma-
tory signaling by bacterial lipopolysaccharide in human hepatic stellate cells.
Hepatology 2003;37(5):1043–55.
116. Ware CF. The TNF superfamily-2008. Cytokine Growth Factor Rev 2008;19(3–4):
183–6.
117. Rothe J, Lesslauer W, Lotscher H, et al. Mice lacking the tumour necrosis factor
receptor 1 are resistant to TNF-mediated toxicity but highly susceptible to infec-
tion by Listeria monocytogenes. Nature 1993;364(6440):798–802.
118. Gallois C, Habib A, Tao J, et al. Role of NF-kappaB in the antiproliferative effect
of endothelin-1 and tumor necrosis factor-alpha in human hepatic stellate cells.
Involvement of cyclooxygenase-2. J Biol Chem 1998;273(36):23183–90.
119. Varela-Rey M, Montiel-Duarte C, Oses-Prieto JA, et al. p38 MAPK mediates the
regulation of alpha1(I) procollagen mRNA levels by TNF-alpha and TGF-beta in
a cell line of rat hepatic stellate cells(1). FEBS Lett 2002;528(1–3):133–8.
120. Reinehr R, Sommerfeld A, Haussinger D. CD95 ligand is a proliferative and anti-
apoptotic signal in quiescent hepatic stellate cells. Gastroenterology 2008;
134(5):1494–506.
121. Taimr P, Higuchi H, Kocova E, et al. Activated stellate cells express the TRAIL
receptor-2/death receptor-5 and undergo TRAIL-mediated apoptosis. Hepatol-
ogy 2003;37(1):87–95.
122. Cao Q, Mak KM, Ren C, et al. Leptin stimulates tissue inhibitor of metalloprotei-
nase-1 in human hepatic stellate cells: respective roles of the JAK/STAT and
JAK-mediated H2O2-dependant MAPK pathways. J Biol Chem 2004;279(6):
4292–304.
123. Gao B. Cytokines, STATs and liver disease. Cell Mol Immunol 2005;2(2):92–100.
124. Jeong WI, Park O, Radaeva S, et al. STAT1 inhibits liver fibrosis in mice by inhib-
iting stellate cell proliferation and stimulating NK cell cytotoxicity. Hepatology
2006;44(6):1441–51.
protein kinase modulates the activated phenotype of hepatic stellate cells. Hep-
atology 2008;47(2):668–76.
126. Adachi M, Brenner DA. High molecular weight adiponectin inhibits proliferation
of hepatic stellate cells via activation of adenosine monophosphate-activated
protein kinase. Hepatology 2008;47(2):677–85.
127. Lavoie JL, Sigmund CD. Minireview: overview of the renin-angiotensin system–
an endocrine and paracrine system. Endocrinology 2003;144(6):2179–83.
128. Varagic J, Frohlich ED. Local cardiac renin-angiotensin system: hypertension
and cardiac failure. J Mol Cell Cardiol 2002;34(11):1435–42.
129. Weber MA. Comparison of type 1 angiotensin II receptor blockers and angioten-
sin converting enzyme inhibitors in the treatment of hypertension. J Hypertens
Suppl 1997;15(6):S31–6.
130. Miyazawa K, Fukuyama J, Misawa K, et al. Tranilast antagonizes angiotensin II
and inhibits its biological effects in vascular smooth muscle cells. Atherosclero-
sis 1996;121(2):167–73.
131. Kaschina E, Unger T. Angiotensin AT1/AT2 receptors: regulation, signalling and
function. Blood Press 2003;12(2):70–88.
132. Kurisu S, Ozono R, Oshima T, et al. Cardiac angiotensin II type 2 receptor activates
the kinin/NO system and inhibits fibrosis. Hypertension 2003;41(1):99–107.
133. Keidar S, Kaplan M, Gamliel-Lazarovich A. ACE2 of the heart: from angiotensin I

to angiotensin (1–7). Cardiovasc Res 2007;73(3):463–9.
134. Ma L, Fogo AB. Role of angiotensin II in glomerular injury. Semin Nephrol 2001;
21(6):544–53.
135. Brenner BM, Zagrobelny J. Clinical renoprotection trials involving angiotensin II-
receptor antagonists and angiotensin-converting-enzyme inhibitors. Kidney Int
Suppl 2003;83:S77–85.
136. Pinzani M, Failli P, Ruocco C, et al. Fat-storing cells as liver-specific pericytes.
Spatial dynamics of agonist-stimulated intracellular calcium transients. J Clin
Invest 1992;90(2):642–6.
137. Kanno K, Tazuma S, Nishioka T, et al. Angiotensin II participates in hepatic in-
flammation and fibrosis through MCP-1 expression. Dig Dis Sci 2005;50(5):
942–8.
138. Yang L, Bataller R, Dulyx J, et al. Attenuated hepatic inflammation and fibrosis in
angiotensin type 1a receptor deficient mice. J Hepatol 2005;43(2):317–23.
139. Paizis G, Cooper ME, Schembri JM, et al. Up-regulation of components of the
renin-angiotensin system in the bile duct-ligated rat liver. Gastroenterology
2002;123(5):1667–76.
140. Ikura Y, Ohsawa M, Shirai N, et al. Expression of angiotensin II type 1 receptor in
human cirrhotic livers: its relation to fibrosis and portal hypertension. Hepatol
Res 2005;32(2):107–16.
141. Komeda K, Jin D, Takai S, et al. Significance of chymase-dependent angiotensin
II formation in the progression of human liver fibrosis. Hepatol Res 2008;38(5):
501–10.
142. Kamada Y, Tamura S, Kiso S, et al. Angiotensin II stimulates the nuclear trans-
location of Smad2 and induces PAI-1 mRNA in rat hepatic stellate cells. Hepatol
Res 2003;25(3):296–305.
143. Kitamura K, Tada S, Nakamoto N, et al. Rho/Rho kinase is a key enzyme system
involved in the angiotensin II signaling pathway of liver fibrosis and steatosis.
J Gastroenterol Hepatol 2007;22(11):2022–33.
144. De Minicis S, Brenner DA. NOX in liver fibrosis. Arch Biochem Biophys 2007;
462(2):266–72.
145. Croquet V, Moal F, Veal N, et al. Hemodynamic and antifibrotic effects of losartan
in rats with liver fibrosis and/or portal hypertension. J Hepatol 2002;37(6):
773–80.
146. El Demerdash E, Salam OM, El Batran SA, et al. Inhibition of the renin-angioten-
sin system attenuates the development of liver fibrosis and oxidative stress in
rats. Clin Exp Pharmacol Physiol 2008;35(2):159–67.
147. Fabris C, Toniutto P, Bitetto D, et al. Sex-related influence of angiotensin-con-
verting enzyme polymorphisms on fibrosis progression due to recurrent hepati-
tis C after liver transplantation. J Gastroenterol 2007;42(7):543–9.
148. Ibanez P, Solis N, Pizarro M, et al. Effect of losartan on early liver fibrosis devel-
opment in a rat model of nonalcoholic steatohepatitis. J Gastroenterol Hepatol
2007;22(6):846–51.
149. Kurikawa N, Suga M, Kuroda S, et al. An angiotensin II type 1 receptor antag-
onist, olmesartan medoxomil, improves experimental liver fibrosis by suppres-
sion of proliferation and collagen synthesis in activated hepatic stellate cells.
Br J Pharmacol 2003;139(6):1085–94.
150. Nabeshima Y, Tazuma S, Kanno K, et al. Anti-fibrogenic function of angiotensin II
type 2 receptor in CCl4-induced liver fibrosis. Biochem Biophys Res Commun
2006;346(3):658–64.
151. Ohishi T, Saito H, Tsusaka K, et al. Anti-fibrogenic effect of an angiotensin con-

verting enzyme inhibitor on chronic carbon tetrachloride-induced hepatic fibro-
sis in rats. Hepatol Res 2001;21(2):147–58.
152. Paizis G, Gilbert RE, Cooper ME, et al. Effect of angiotensin II type 1 receptor
blockade on experimental hepatic fibrogenesis. J Hepatol 2001;35(3):376–85.
153. Ramos SG, Montenegro AP, Goissis G, et al. Captopril reduces collagen and
mast cell and eosinophil accumulation in pig serum-induced rat liver fibrosis.
Pathol Int 1994;44(9):655–61.
154. Tuncer I, Ozbek H, Ugras S, et al. Anti-fibrogenic effects of captopril and can-
desartan cilexetil on the hepatic fibrosis development in rat. The effect of AT1-R
blocker on the hepatic fibrosis. Exp Toxicol Pathol 2003;55(2–3):159–66.
155. Turkay C, Yonem O, Arici S, et al. Effect of angiotensin-converting enzyme inhi-
bition on experimental hepatic fibrogenesis. Dig Dis Sci 2008;53(3):789–93.
156. Wei HS, Lu HM, Li DG, et al. The regulatory role of AT 1 receptor on activated
HSCs in hepatic fibrogenesis:effects of RAS inhibitors on hepatic fibrosis in-
duced by CCl(4). World J Gastroenterol 2000;6(6):824–8.
157. Xiao F, Wei H, Song S, et al. Polymorphisms in the promoter region of the angio-
tensinogen gene are associated with liver cirrhosis in patients with chronic hep-
atitis B. J Gastroenterol Hepatol 2006;21(9):1488–91.
158. Xu W, Song S, Huang Y, et al. Effects of perindopril and valsartan on expression
of transforming growth factor-beta-Smads in experimental hepatic fibrosis in
rats. J Gastroenterol Hepatol 2006;21(8):1250–6.
159. Yoshiji H, Kuriyama S, Yoshii J, et al. Angiotensin-II type 1 receptor interaction is a ma-
jor regulator for liver fibrosis development in rats. Hepatology 2001;34(4 Pt 1):
745–50.
160. Yoshiji H, Yoshii J, Ikenaka Y, et al. Inhibition of renin-angiotensin system atten-
uates liver enzyme-altered preneoplastic lesions and fibrosis development in
rats. J Hepatol 2002;37(1):22–30.
161. Yoshiji H, Kuriyama S, Noguchi R, et al. Angiotensin-II and vascular endothelial
growth factor interaction plays an important role in rat liver fibrosis development.
Hepatol Res 2006;36(2):124–9.
162. Gaedeke J, Peters H, Noble NA, et al. TGF-beta and renal fibrosis. Contrib
Nephrol 2001;135:153–60.
163. Shimizu S, Satomura K, Aramaki T, et al. Hepatic chymase level in chronic hep-
atitis: co-localization of chymase with fibrosis. Hepatol Res 2003;27(1):62–6.
164. Dixon JB, Bhathal PS, Jonsson JR, et al. Pro-fibrotic polymorphisms predictive
of advanced liver fibrosis in the severely obese. J Hepatol 2003;39(6):967–71.
165. Powell EE, Edwards-Smith CJ, Hay JL, et al. Host genetic factors influence dis-
ease progression in chronic hepatitis C. Hepatology 2000;31(4):828–33.
166. Terui Y, Saito T, Watanabe H, et al. Effect of angiotensin receptor antagonist on
liver fibrosis in early stages of chronic hepatitis C. Hepatology 2002;36(4 Pt 1):
1022.
167. Yokohama S, Yoneda M, Haneda M, et al. Therapeutic efficacy of an angiotensin
II receptor antagonist in patients with nonalcoholic steatohepatitis. Hepatology
2004;40(5):1222–5.
168. Yoshiji H, Noguchi R, Kojima H, et al. Interferon augments the anti-fibrotic activ-
ity of an angiotensin-converting enzyme inhibitor in patients with refractory
chronic hepatitis C. World J Gastroenterol 2006;12(42):6786–91.
169. Rimola A, Londono MC, Guevara G, et al. Beneficial effect of angiotensin-block-
ing agents on graft fibrosis in hepatitis C recurrence after liver transplantation.
Transplantation 2004;78(5):686–91.
170. Zekry A, Whiting P, Crawford DH, et al. Liver transplantation for HCV-associated
liver cirrhosis: predictors of outcomes in a population with significant genotype 3
and 4 distribution. Liver Transpl 2003;9(4):339–47.
171. Schiffrin EL. Vascular and cardiac benefits of angiotensin receptor blockers. Am
J Med 2002;113(5):409–18.
172. Bosch X. Losartan-induced hepatotoxicity. JAMA 1997;278(19):1572.
173. Yoshiji H, Kuriyama S, Yoshii J, et al. Angiotensin-II induces the tissue inhibitor of
metalloproteinases-1 through the protein kinase-C signaling pathway in rat liver
fibrosis development. Hepatol Res 2003;27(1):51–6.
174. Li X, Meng Y, Yang XS, et al. ACEI attenuates the progression of CCl4-induced
rat hepatic fibrogenesis by inhibiting TGF-beta1, PDGF-BB, NF-kappaB and
MMP-2,9. World J Gastroenterol 2005;11(31):4807–11.
175. Ueki M, Koda M, Yamamoto S, et al. Preventive and therapeutic effects of angio-
tensin II type 1 receptor blocker on hepatic fibrosis induced by bile duct ligation
in rats. J Gastroenterol 2006;41(10):996–1004.
176. Yoshiji H, Kuriyama S, Noguchi R, et al. Amelioration of liver fibrogenesis by dual
inhibition of PDGF and TGF-beta with a combination of imatinib mesylate and
ACE inhibitor in rats. Int J Mol Med 2006;17(5):899–904.
Relationships Among
Stellate Cell Ac tivation,
Progenitor Cells, a nd
Hepatic Re generation
Tania Roskams, MD, PhD
KEYWORDS
Hepatic stellate cells Liver progenitor cells
Hepatic progenitor cells Liver stem cell niche
Liver progenitor cell niche Hepatic regeneration
DIFFERENT TYPES OF STEM CELLS AND THEIR NICHES
A stem cell is an undifferentiated cell capable throughout life of renewing itself and
generating one or more types of differentiated cells.1–3 The ultimate stem cells are
those of the early embryo, which are totipotential. Farther down the developmental
line, in the fetus and adult, F multipotential (pluripotential) stem cells are present.
The cells closer to final differentiation are called progenitor, committed, or transit cells.
In continuously renewing lining epithelia, such as the epidermis, stem cells are easy
to find because they abut the basement membrane. In other tissues, such as the liver,
the search is frustrating because, by definition, stem cells look bland and
undifferentiated. However, they can be visualized with immunohistochemistry, which
shows distinctive proteins (antigens) on their surface or in their cytoplasm (Fig. 1). The
liver contains different cell types with stem cell properties: hepatocytes, progenitor
cells in the most peripheral branches of the biliary tree, and hepatic stellate cells
(HSCs).4–8
HEPATOCYTES HAVE STEM CELL PROPERTIES
In normal circumstances, the liver hardly proliferates and was therefore classified as
a stable organ. Hepatocytes in normal adult liver have a life span of more than 1 year.
After partial hepatectomy, however, proliferation of the main epithelial compartments
(hepatocytes and cholangiocytes), followed by proliferation of the mesenchymal cells
(HSCs and endothelial cells), quickly restores the liver. In rodents, the liver can restore
its original volume after two-thirds hepatectomy in approximately 10 days.9,10
Head Liver Research Unit, Department of Morphology and Molecular Pathology, University of
Leuven, Minderbroederstraat 12, B-3000 Leuven, Belgium
E-mail address: tania.roskams@uz.kuleuven.ac.be
Clin Liver Dis 12 (2008) 853–860

854 Roskams
Fig. 1. Cytokeratin 7 immunohistochemical stain illustrating liver progenitor cells (short

arrows) and intermediate hepatocytes (long arrows) in a liver after severe hepatocyte
damage.
Serial transplantation experiments have shown that hepatocytes have almost an

infinite capacity to proliferate.11,12 At least 69 doublings can occur, confirming the
clonogenic potential of hepatocytes, one of the crucial properties of a stem cell.
HSCs play a crucial role in the hepatocyte regeneration process because they pro-
duce not only mitogens for hepatocytes (eg, hepatocyte growth factor, transforming
growth factor a (TNF-a), insulin growth factor) but also inhibitory growth factors,
such as TNF-b to stop the regeneration process.10,13,14 This feature and their close
anatomic relationship with hepatocytes suggests that they are part of the local
‘‘stem cell niche’’ for hepatocytes.
HEPATIC PROGENITOR CELLS: LOCALIZATION AND ACTIVATION
When the mature epithelial cell compartments of the liver, hepatocytes, or

cholangiocytes are damaged or inhibited in their replication, a reserve cell compart-
ment is activated.6 This compartment, called the progenitor cell compartment in
humans and the oval cell compartment in rodents, resides in the smallest and most
peripheral branches of the biliary tree, the ductules, and the canals of Hering.7
The canal of Hering is a channel partly lined by hepatocytes and partly by
cholangiocytes. It represents the anatomic and physiologic link between the intralob-
ular canalicular system and the biliary tree. Cells of morphology and immunophenotype
intermediate between hepatocytes and cholangiocytes (intermediate cells) are not rec-
ognized in normal tissue. A corollary is that the true interface of hepatocytes and the
biliary tree does not reside, as was assumed, at the limiting plate, but rather along
an array of sites that project star-like from the portal tracts, along the canals of Hering.
Wilson and Leduc15 were the first to describe this activation of a reserve cell com-
partment in mice after severe dietary injury. Subsequently, several models of so-called
‘‘oval cell reaction’’ in rodents have been described. These models mainly used poten-
tial carcinogenic agents to inhibit the proliferation of mature hepatocytes after a regen-
erative stimulus (eg, acetaminofluorene intoxication of rats, partial hepatectomy, or
administration of a necrogenic dose of carbon tetrachloride)16 or ethionine intoxication
in mice. Furthermore, models of fatty liver disease such as Ob/Ob-mice or PARP1( / )
mice are characterized by inhibition of the replication of mature hepatocytes, caused
by oxidative stress, and show a striking oval cell response.17,18
Activation of oval cells (called ductular reaction in the human liver) comprises expan-
sion of a transit amplifying cell compartment of small biliary cells, which can differen-
tiate into at least biliary epithelial cells and hepatocytes. The progenitor cells are
labeled by biliary-type cytokeratins CK7 and CK19, oval cell markers OV6 and OV1,
Relationships Among Liver Cell Types 855
neuroendocrine markers chromogranin-A, neural cell adhesion molecule, and the

parathyroid hormone-related peptide connexin 43. A subpopulation of ductular/pro-
genitor cells express markers of hematopoietic cells (CD34, C-kit, flt-3, CD133), which
raised the question whether progenitor cells would be directly derived from bone mar-
row stem cells. Most studies, however, are concordant with a progenitor cell niche in
the ductules/canals of Hering, at the interface between the parenchyma and the portal
tract mesenchyme. This location is also where, during embryonic development, bipo-
tential hepatoblasts form the primitive ductal plate, which has the same phenotype as
progenitor cells in adult life: ductal plate cells express biliary markers, (immature) hep-
atocytic markers such as alpha-fetoprotein, and hematopoietic markers such as CD34.
Progenitor cell activation or ductular reaction is seen in most chronic human liver
diseases. Reactive ductules form strands of small cholangiocytes with an oval nucleus
and a small rim of cytoplasm.19 The degree of progenitor cell activation increases with
the severity of disease.20 In chronic hepatitis, progenitor cell activation correlates with
the degree of inflammation.21 In various chronic liver diseases, such as chronic hep-
atitis C, hemochromatosis, and nonalcoholic steatohepatitis, the degree of progenitor
cell activation also correlates with degree of fibrosis and stage of disease.17,20
In moderate and severe degrees of inflammation, intermediate hepatocytes occur,
having a phenotype intermediate between progenitor and ductular cells and mature
hepatocytes (see Fig. 1). The number of these intermediate hepatocytes gradually
increases with higher degrees of inflammation, higher degrees of necrosis in necrotiz-
ing hepatitis, and more advanced stages of nonalcoholic steatohepatitis.17,20–22 This
correlation strongly suggests a higher degree of differentiation of progenitor cells
into hepatocytes when more hepatocyte damage is present.
In cirrhotic livers, and especially in nonalcoholic steatohepatitis, whole cirrhotic
nodules can be composed of intermediate hepatocytes, which ultrastructurally look
strikingly normal, without Mallory body formation and fatty change. This finding sug-
gests that these intermediate hepatocyte nodules originate from progenitor cells.17
In a three-dimensional reconstruction study, Falkowski and colleagues23 showed
that sequestered hepatocyte buds in cirrhosis are always in continuity with reactive
ductules, strongly suggesting a progenitor cell origin.
A trigger for progenitor cell activation is certainly e mature cell compartments’ lack
of ability to proliferate. In parallel to what is known from rodent models, inhibition of
replication of mature hepatocytes also occurs in human liver diseases. In the cirrhotic
stage of a wide variety of chronic human liver diseases, hepatocytes were recently
shown to be senescent from telomere shortening. However, mesenchymal cells,
such as endothelial cells and HSCs, do not show this replicative senescence.24,25
Hepatocyte replicative senescence is probably partly the result of ongoing prolifera-
tion over 20 to 30 years of chronic liver disease. Chronic inflammation, growth factors,
DNA-damaging agents such as reactive oxygen species, and nitrogen species also
play a role. Similar to rodent models, replicative senescence of hepatocytes in man
triggers progenitor cell activation.6,26
Several papers have illustrated the role of bone marrow stem cells in producing hepa-
tocytes, both in animal and human models. Their precise role is not clear, because cell fu-
sion of bone marrow stem cells with damaged hepatocytes occurred in several models.27
THE SOCIETY OF LIVER CELLS AND THE SURROUNDING STROMA: THE LIVER ‘‘STEM CELL NICHE’’
The liver consists of epithelial cells types (hepatocytes, cholangiocytes, and progen-
itor cells), mesenchymal cell types (Kupffer cells, endothelial cells, and HSCs), and
stroma. Stroma is present in the portal tracts and Disse’s spaces (Fig. 2).
856 Roskams
Circulating growth factors Kupffer Cells: TNFalpha, IL6

EGF, HGF
VEGF+
VEGF+
BD
V
PT
Matrix
A HSC/MF bound HGF
produce HGF +,
TGFbeta -,
-
Neurotrophins,
TIMP
TWEAK, …
Fig. 2. Society of liver cells. When hepatocytes are injured, Kupffer cells secrete interleukin-6
and transforming growth factor (TGF) a, which activate hepatic stellate cells (HSCs). HSCs/
myofibroblasts produce growth factors for hepatocytes and progenitor cells (eg, hepatocyte
growth factor). They also produce matrix components and TGF-b, which inhibit the growth
of hepatocytes. Growth factors such as hepatocyte and epidermal growth factors also come
from the circulation. The matrix harbors growth factors for hepatocytes and progenitor
cells, bound to proteoglycans, for example, which can be quickly released when needed.
Release of hepatocyte growth factor by the matrix is then inhibited by tissue inhibitor of
metalloproteinase. Hepatocytes and progenitor cells produce growth factors for endothelial
cells (flat green cells) such as vascular endothelial growth factor to maintain their
vascularization.
The importance of this society of cells and the stroma is illustrated by the fact that
hepatocytes, which have a huge growth potential in vivo, are very hard to keep alive
and differentiated when isolated and put in culture. When hepatocytes are injured,
Kupffer cells secrete interleukin-6 and TNF-a, which activate HSCs. HSCs produce
growth factors for hepatocytes and progenitor cells (eg, hepatocyte growth factor).
They also produce matrix components and TNF-b, which inhibits the growth of
hepatocytes.
Growth factors such as hepatocyte and epidermal growth factors also derive from
the circulation. The matrix harbors growth factors for hepatocytes and progenitor
cells, bound to proteoglycans (for example), which can be quickly released when
needed. Hepatocyte growth factor release by the matrix is then inhibited by tissue
inhibitor of metalloproteinase (TIMP).28 Hepatocytes and progenitor cells produce
growth factors such as vascular endothelial growth factor for endothelial cells to main-
tain their vascularization.
A specific growth factor for oval cells, TGF-like weak inhibitor of apoptosis (TWEAK),
was recently described, which does not act on hepatocytes.29 This finding is an
important step in understanding the regulation of oval cell reaction versus hepatocyte
proliferation. This growth factor is produced by sinusoidal lining cells. Sinusoidal lining
cells like HSCs proliferate in close anatomic relationship with progenitor cells; in addi-
tion, stellate cells produce growth factors for which progenitor cells have receptors,
suggesting a true interaction between these cell compartments.6 In general, defining
the so-called ‘‘stem cell niche’’ for adult human liver progenitor cells will be an impor-
tant goal for the near future.
Different subtypes of HSC/myofibroblasts (MFs) are known: portal myofibroblasts,
interface HSC/MFs, and lobular stellate cells.30,31 These subtypes are activated to
a different extent in different types of diseases, and therefore the target cells in
antifibrotic therapy might also be heterogeneous. Around activated progenitor cells,
mainly glial fibrillary acidic protein (GFAP)/a-smooth muscle actin (SMA)–reactive

HSC/MFs, are seen in human diseases. In biliary diseases, a-SMA-expressing portal
MFs are the main subtype coproliferating with reactive ductules and inducing
expansion of portal tracts and formation of portal–portal connections.
At the edge of the ductular fronts, interface HSC/MFs are also present. Activation of
lobular HSCs is only seen in a very late stage in biliary diseases. In chronic viral and
autoimmune hepatitis, lobular HSCs expressing neurotrophins and their receptors,
neural cell adhesion cell molecule, GFAP, and a-SMA are activated in lobular hepatitis
areas and bridging necrosis. Activation of lobular HSCs and interface HSC/MFs is
seen at the edges of portal tracts where interface hepatitis—the hallmark of chronic
hepatitis—is present. The most prominent activation of lobular HSCs is seen in chronic
alcoholic and nonalcoholic steatohepatitis. In these diseases, the activation of HSCs
and deposition of matrix components in the perisinusoidal spaces of Disse leads to
perisinusoidal and even pericellular fibrosis (so-called ‘‘chicken-wire fibrosis’’). It is
very possible that these different subtypes transition into one another, because
intermediate forms between the subtypes have been described (Fig. 3).30,32
Progenitor cells seem capable of surviving when hepatocytes are lost from toxic
damage or viral infections. ATP-binding cassette transporters play a secretory and
protective role, and a strong up-regulation of apical multidrug resistance gene-1
(MDR1) and basolateral multidrug resistance-associated protein (MRP1) and MRP3
in human hepatocytes and progenitor cell–related bile ductules has been
observed.33,34 The authors hypothesized that this change in expression offers
protection against the accumulation of toxic bile constituents and may render these
cells resistant to oxidative stress. In addition, breast cancer resistance protein
(BCRP) is expressed by human and rat hepatic progenitor cells.35 BCRP is able to
pump out the Hoechst dye, thereby defining the ‘‘side population;’’ cells having this
pump, are ‘‘side sorted’’ from the main population in fluorescent-activated cell sorting,
because they pump out the dye. The side population contains stem cells in many
organs, and increasing evidence shows that cancer stem cells are also part of the
side population.
Fig. 3. a-Smooth muscle stain (a-SMA) showing coproliferation of progenitor cells (arrows)
with a-SMA positive hepatic stellate cells.
858 Roskams
HEPATIC STELLATE CELLS: NICHE CELLS OR STEM CELLS?
HSCs produce growth factors for epithelial cells (hepatocytes and progenitor
cells).10,13 Their close anatomic relationship with hepatocytes in the spaces of Disse
and around progenitor cells36–38 place them in a perfect location to be ‘‘niche cells;’’
likewise, for hepatocytes, they could serve as a kind of liver stem cell.
Recent work even suggests that HSCs themselves might be progenitors for liver
epithelial cells. This theory is based on data showing that quiescent HSCs from adults
express markers of all three embryonic germ layers, such as GFAP (an ectodermal
marker),39,40 desmin (a mesodermal marker),41 and various stem cell markers, includ-
ing nestin, CD105, p75 neurotrophin receptor, c-kit, and CD133.40,42–44 Cell culture
data support the plasticity of HSCs, showing differentiation of quiescent rat HSCs
into not only MFs but also cells with hepatocyte characteristics, those with biliary fea-
tures, and endothelial-like cells, depending on the culture conditions.45 In culture sys-
tems, however, the apparent pluripotency of HSCs could be the result of selective
outgrowth of epithelial progenitor cells that are present in the original isolates. How-
ever, recent evidence using fate-mapping approaches demonstrates that HSCs can
be epithelial progenitor cells in adult mouse livers. This study also suggests that
they can be a type of epithelial progenitor, transitioning through a mesenchymal stage
and differentiating into hepatocytes during liver regeneration.8 The field of (stem) cell
plasticity becomes intriguingly complex, and further study will hopefully shed more
light.
SUMMARY
HSCs play an important role in liver fibrogenesis. They are also key players in liver
regeneration as part of the stem cell niche of hepatocytes and hepatic progenitor cells.
They produce growth stimulating and inhibiting factors for these epithelial cell com-
partments. In addition, recent studies suggest a role for HSCs themselves for being
progenitors of epithelial cells through a transitional mesenchymal phase.
REFERENCES
1. Watt FM, Hogan BL. Out of Eden: stem cells and their niches. Science 2000;287:
1427–30.
2. Alison M, Sarraf C. Hepatic stem cells. J Hepatol 1998;29:676–82.
3. Hall PA. What are stem cells and how are they controlled? J Pathol 1989;158:
275–7.
4. Kuwahara R, Kofman AV, Landis CS, et al. The hepatic stem cell niche: identifica-
tion by label-retaining cell assay. Hepatology 2008;47:1810–2.
5. Theise ND, Kuwahara R. The tissue biology of ductular reactions in human
chronic liver disease. Gastroenterology 2007;133:350–2.
6. Roskams TA, Libbrecht L, Desmet VJ. Progenitor cells in diseased human liver.
7. Roskams TA, Theise ND, Balabaud C, et al. Nomenclature of the finer branches of
the biliary tree: canals, ductules, and ductular reactions in human livers. Hepatol-
ogy 2004;39:1739–45.
8. Yang L, Jung Y, Omenetti A, et al. Fate-mapping evidence that hepatic stellate
cells are epithelial progenitors in adult mouse livers. Stem Cells 2008;26:2104–43.
9. Michalopoulos GK, DeFrances MC. Liver regeneration. Science 1997;276:60–6.
10. Fausto N. Liver regeneration and repair: hepatocytes, progenitor cells, and stem
cells. Hepatology 2004;39:1477–87.
11. Overturf K, Al-Dhalimy M, Finegold M, et al. The repopulation potential of

hepatocyte populations differing in size and prior mitotic expansion. Am J Pathol
1999;155:2135–43.
12. Overturf K, Al-Dhalimy M, Manning K, et al. Ex vivo hepatic gene therapy of
a mouse model of Hereditary Tyrosinemia Type I. Hum Gene Ther 1998;9:
295–304.
13. Taub R. Liver regeneration: from myth to mechanism. Nat Rev Mol Cell Biol 2004;
5:836–47.
14. Fausto N, Campbell JS, Riehle KJ. Liver regeneration. Hepatology 2006;43:
S45–53.
15. Wilson JW, Leduc EH. Role of cholangioles in restoration of the liver of the mouse
after dietary injury. J Pathol Bacteriol 1958;76:441–9.
16. Alison MR, Golding M, Sarraf CE, et al. Liver damage in the rat induces
hepatocyte stem cells from biliary epithelial cells. Gastroenterology 1996;110:
1182–90.
17. Roskams T, Yang SQ, Koteish A, et al. Oxidative stress and oval cell accumulation
in mice and humans with alcoholic and nonalcoholic fatty liver disease. Am
J Pathol 2003;163:1301–11.
18. Yang S, Koteish A, Lin H, et al. Oval cells compensate for damage and replicative
senescence of mature hepatocytes in mice with fatty liver disease. Hepatology
2004;39:403–11.
19. Roskams T, Desmet V. Ductular reaction and its diagnostic significance. Semin
Diagn Pathol 1998;15:259–69.
20. Lowes KN, Brennan BA, Yeoh GC, et al. Oval cell numbers in human chronic
liver diseases are directly related to disease severity. Am J Pathol 1999;154:
537–41.
21. Libbrecht L, Desmet V, Van Damme B, et al. Deep intralobular extension of human
hepatic ‘progenitor cells’ correlates with parenchymal inflammation in chronic
viral hepatitis: can ‘progenitor cells’ migrate? J Pathol 2000;192:373–8.
22. Katoonizadeh A, Nevens F, Verslype C, et al. Liver regeneration in acute severe
liver impairment: a clinicopathological correlation study. Liver Int 2006;26:
1225–33.
23. Falkowski O, An HJ, Ianus IA, et al. Regeneration of hepatocyte ‘buds’ in cirrhosis
from intrabiliary stem cells. J Hepatol 2003;39:357–64.
24. Wiemann SU, Satyanarayana A, Tsahuridu M, et al. Hepatocyte telomere
shortening and senescence are general markers of human liver cirrhosis. Faseb
J 2002;16:935–42.
25. Marshall A, Rushbrook S, Davies SE, et al. Relation between hepatocyte G1
arrest, impaired hepatic regeneration, and fibrosis in chronic hepatitis C virus
infection. Gastroenterology 2005;128:33–42.
26. Roskams T. Progenitor cell involvement in cirrhotic human liver diseases: from
controversy to consensus. J Hepatol 2003;39:431–4.
27. Thorgeirsson SS, Grisham JW. Hematopoietic cells as hepatocyte stem cells:
a critical review of the evidence. Hepatology 2006;43:2–8.
28. Perugorria MJ, Latasa MU, Nicou A, et al. The epidermal growth factor receptor
ligand amphiregulin participates in the development of mouse liver fibrosis.
Hepatology 2008; in press.
29. Jakubowski A, Ambrose C, Parr M, et al. TWEAK induces liver progenitor cell
proliferation. J Clin Invest 2005;115:2330–40.
30. Cassiman D, Libbrecht L, Desmet V, et al. Hepatic stellate cell/myofibroblast sub-
populations in fibrotic human and rat livers. J Hepatol 2002;36:200–9.
860 Roskams
31. Gressner OA, Weiskirchen R, Gressner AM. Evolving concepts of liver fibrogen-
esis provide new diagnostic and therapeutic options. Comp Hepatol 2007;6:7.
32. Cassiman D, Roskams T. Beauty is in the eye of the beholder: emerging concepts
and pitfalls in hepatic stellate cell research. J Hepatol 2002;37:527–35.
33. Ros JE, Libbrecht L, Geuken M, et al. High expression of MDR1, MRP1, and
MRP3 in the hepatic progenitor cell compartment and hepatocytes in severe
human liver disease. J Pathol 2003;200:553–60.
34. Ros JE, Roskams TA, Geuken M, et al. ATP binding cassette transporter gene
expression in rat liver progenitor cells. Gut 2003;52:1060–7.
35. Vander Borght S, Libbrecht L, Katoonizadeh A, et al. Breast cancer resistance
protein (BCRP/ABCG2) is expressed by progenitor cells/reactive ductules and
hepatocytes and its expression pattern is influenced by disease etiology and
species type: possible functional consequences. J Histochem Cytochem 2006;
54:1051–9.
36. Yin L, Lynch D, Ilic Z, et al. Proliferation and differentiation of ductular progenitor
cells and littoral cells during the regeneration of the rat liver to CCl4/2-AAF injury.
Histol Histopathol 2002;17:65–81.
37. Yin L, Lynch D, Sell S. Participation of different cell types in the restitutive
response of the rat liver to periportal injury induced by allyl alcohol. J Hepatol
1999;31:497–507.
38. Roskams T. Different types of liver progenitor cells and their niches. J Hepatol
2006;45:1–4.
39. Neubauer K, Knittel T, Aurisch S, et al. Glial fibrillary acidic protein–a cell type
specific marker for Ito cells in vivo and in vitro. J Hepatol 1996;24:719–30.
40. Geerts A. History, heterogeneity, developmental biology, and functions of quies-
cent hepatic stellate cells. Semin Liver Dis 2001;21:311–35.
41. Yokoi Y, Namihisa T, Kuroda H, et al. Immunocytochemical detection of desmin in
fat-storing cells (Ito cells). Hepatology 1984;4:709–14.
42. Fujio K, Evarts RP, Hu Z, et al. Expression of stem cell factor and its receptor,
c-kit, during liver regeneration from putative stem cells in adult rat. Lab Invest
1994;70:511–6.
43. Niki T, Pekny M, Hellemans K, et al. Class VI intermediate filament protein Nestin
is induced during activation of rat hepatic stellate cells. Hepatology 1999;29:
520–7.
44. Cassiman D, Denef C, Desmet VJ, et al. Human and rat hepatic stellate cells
express neurotrophins and neurotrophin receptors. Hepatology 2001;33:148–58.
45. Kordes C, Sawitza I, Muller-Marbach A, et al. CD1331 hepatic stellate cells are
progenitor cells. Biochem Biophys Res Commun 2007;352:410–7.
I mmune I nterac tions
in Hepatic Fibrosis
Andrew P. Holt, MRCP, PhDa,b, Mike Salmon, PhD, MRCPathc,
Christopher D. Buckley, FRCP, PhDc, David H. Adams, MD, FRCP, FMedScia,*
KEYWORDS
Liver Inflammation Stroma Lymphocytes Fibrosis
Hepatic stellate cell
Far from being a static process, chronic inflammation is a dynamic aggregate of lym-
phocytes, macrophages, and stromal cells held together by autocrine and paracrine
interactions. Consequently, the role of leukocytes in wound healing must be
considered within the context of these organ-specific stromal microenvironments.
This relationship is particularly the case in the liver, where interactions between
stromal cells and infiltrating leukocytes are critical in determining the outcome of liver
injury, as demonstrated by the observations that animals deficient in macrophages,
T cells, or B cells all show reduced fibrotic responses. The ability of leukocyte-derived
cytokines to modulate the behavior of liver fibroblasts is widely accepted, but
fibroblasts also contribute to the processes of leukocyte recruitment, positioning,
and survival at sites of chronic inflammation.1 Neutrophil recruitment in the immediate/
early response to liver injury is preceded by rapid up-regulation of adhesion molecules
on stellate cells within the subendothelial space, whereas in animals recovering from
liver fibrosis, resolution of liver inflammation is prefaced by apoptosis of activated
liver myofibroblasts.2,3 Thus, the induction, maintenance, and resolution of liver
inflammation and fibrosis is determined by interactions between infiltrating leukocytes
and activated fibroblasts within the liver stroma.4,5 This article discusses the immuno-
logic mechanisms that drive liver fibrosis and, in particular, the complex interactions
that exist between cellular components of the stroma (macrophages and fibroblasts
particularly) and the innate and adaptive immune systems.
Dr. Holt is supported by grants from the Medical Research Council, British Liver Trust, and Core.
a
Liver Research Group, MRC Centre for Immune Regulation, University of Birmingham,
Birmingham, UK
b
Liver Unit, Queen Elizabeth Hospital Birmingham, Edgbaston, Birmingham, B152TH, UK
c
Department of Rheumatology, MRC Centre for Immune Regulation, University of
Birmingham, Birmingham, UK
* Corresponding author. Liver Research Group, 5th floor IBR, University of Birmingham, Queen
Elizabeth Hospital, Edgbaston, Birmingham B15 2TH, UK.
E-mail address: d.h.adams@bham.ac.uk (D.H. Adams).
Clin Liver Dis 12 (2008) 861–882

862 Holt et al
STROMAL MICROENVIRONMENTS IN CHRONIC INFLAMMATORY DISEASE
A characteristic feature of chronic liver disease is the presence of a persistent

inflammatory infiltrate associated with regions of fibrosis and matrix remodeling.
Although macrophages play an important role in the initiation and resolution of fibrosis,
fibroblasts are the most numerous cells within the liver stroma and are primarily
responsible for the synthesis of the extracellular matrix (ECM).6 Fibroblasts provide
an architectural framework for tissues, and regulate important homeostatic and devel-
opmental functions.7 Until recently, resident stromal cells were not thought to make
significant contributions to immune responses within the liver, but it has become clear
that fibroblasts and macrophages within the liver stroma play critical roles in regulating
the inflammatory and fibrotic reactions at sites of tissue injury.1,8,9 Disruption of these
processes results in abnormal wound healing, scarring, and perpetuation of inflamma-
tion.10 This situation is particularly the case in human liver disease, where reiterative
cycles of tissue injury perpetuate stromal cell activation and provoke a switch from
acute reversible inflammation to chronic inflammatory disease.
Monocytes demonstrate considerable plasticity and can differentiate into macro-
phages, dendritic cells (DC), and stromal cell lineages in response to microenviron-
mental activation signals.11,12 In humans, CD161 monocytes expressing high levels
of the fractalkine receptor CX3CR1 are recruited to tissues where they undergo further
site-specific differentiation into resident populations of cells such as Kupffer cells and
interstitial DC,13 whereas CD16low cells are recruited during inflammation (Fig. 1).
Most ‘‘inflammatory’’ monocytes differentiate into macrophages in tissue and mediate
Fig. 1. Based on work in other tissues, the figure shows the two main monocyte subsets in
blood and how they can give rise to functionally diverse subsets of Kupffer cells, macrophages,
and myeloid DC after recruitment into the liver. Two distinct subpopulations of monocytes can
be identified in the human circulation that show differences in expression of chemokine re-
ceptors and adhesion molecules involved in recruitment through sinusoidal endothelium.
CCR, CC chemokine receptor; GM-CFU, granulocyte-macrophage colony-forming units; IL, in-
terleukin. (Data from Gordon S, Taylor PR. Monocyte and macrophage heterogeneity. Nat Rev
Immunol 2005;5:953–64.)
Immune Interactions in Hepatic Fibrosis 863
the clearance of pathogens and the resolution of inflammation, but some CC chemo-
kine receptor (CCR)71 cells can emigrate from tissues by way of lymphatics to the
draining lymph nodes, where they acquire DC function. In the absence of inflamma-
tion, CD161CX3CR1high monocytes enter the liver to replenish Kupffer cells and DC
populations.12 In addition, a population of adherent CD141 monocytes can be in-
duced to differentiate along fibroblast lineages when cultured with T cells. The result-
ing cells, known as fibrocytes, express both hematopoietic and mesenchymal markers
(collagen-11, CD11a/b, CD451, CD341). When recruited to inflamed tissues and ex-
posed to cytokines such as transforming growth factor beta (TGFb)-1,11,14 they can
differentiate into a smooth muscle actin (SMA)1 collagen-secreting contractile cells
with characteristics of myofibroblasts. Fibrocytes are implicated in wound healing
and inflammation in various diseases including chronic liver disease15,16 and exemplify
the cellular plasticity and divergence exhibited within the stroma.
Transcriptional profiling of human fibroblasts from different anatomic sites has dem-
onstrated patterns of gene expression as divergent as those seen among the different
leukocyte lineages17,18 and this biologic diversity is reflected in differential patterns of
growth factor and matrix expression.19 The liver contains several phenotypically dis-
tinct fibroblasts20,21 that may be derived from distinct progenitor cells.22 The intrinsic
variation in phenotype and function of macrophages and fibroblasts within tissues
may be an important factor that predisposes tissues to scarring, and could explain
why relapses in chronic inflammatory disease are often tissue and site specific.
MACROPHAGES REGULATE THE INDUCTION AND RESOLUTION OF LIVER FIBROSIS
Many of the cells that play an important role in the immediate innate immune response,
including neutrophils and mast cells, make little contribution to tissue fibrosis, and nat-
ural killer (NK) cells actually suppress fibrosis by killing activated myofibroblasts, pos-
sibly by tumor necrosis factor–related apoptosis-inducing ligand (TRAIL)–dependent
mechanisms,23–25 which may explain why many acute inflammatory reactions in the
liver resolve without scarring.26,27 On the other hand, macrophages are critical for
fibrosis, as demonstrated by depletion studies in animals that implicate hepatic mac-
rophages in the progression and resolution of fibrosis. Activation of fibroblasts by
macrophage-derived TGFb-1 or insulin-like growth factor is an early feature of
fibrogenesis, which promotes a switch in fibroblast gene expression to initiate matrix
remodeling,28 and macrophage-derived cytokines including interleukin (IL)-6 and
TGFb maintain the proliferation and differentiation of liver fibroblasts.29,30
The divergent roles played by macrophages within the liver are a consequence of
distinct pathways of activation and differentiation determined by local cytokines and
interactions with other liver cells.31 Pathogen-associated molecular patterns are
conserved sequences in microbes that stimulate Toll-like receptors and other pattern
recognition receptors on macrophages and thereby activate innate immune
responses. Classic activation, resulting in so-called ‘‘M1 macrophages,’’ is driven
by interferons (IFN) and characterized by a signature of proinflammatory genes
associated with T helper 1 (Th1) lymphocyte responses and cell-mediated immunity.
An alternative activation profile results in M2 macrophages associated with a T helper
2 (Th2)-type polarized response characterized by signal transducer and activator of
transcription (STAT)-6 signaling, IL-4 and IL-13 secretion, and up-regulation of
endocytic lectin receptors, including the mannose receptor. Alternative activation is
characteristic of parasitic infections, allergy, humoral immunity, and fibrosis. The
deactivation of macrophages is also important to promote resolution and can be trig-
gered by exposure to apoptotic cells, IL-10, and other regulatory mediators and
864 Holt et al
glucocorticoids. Although a clear distinction is evident in vitro between proinflamma-

tory matrilytic M1 and profibrotic M2 macrophages,32 the situation in vivo is more
complex, with classic and alternatively differentiated macrophages present in chron-
ically inflamed liver tissue.33
The importance of macrophages in liver fibrosis has been demonstrated by animal
studies where macrophages were selectively depleted at different stages of fibrosis.
Macrophage depletion at the induction of fibrosis, either by inhibiting CC chemokine
ligand (CCL)2-dependent recruitment or selective depletion of tissue macrophage
populations with gadolinium chloride or diphtheria toxin, reduced scarring and re-
sulted in fewer activated (aSMA1) hepatic stellate cells (HSC).4,34,35 By contrast, de-
pleting macrophages during the immediate/early phase of tissue recovery delayed
matrix degradation and prolonged fibrosis. These divergent roles suggest that either
different populations of macrophages are involved at each stage or, more likely,
macrophage function switches during fibrogenesis in response to cytokines in the
microenvironment.
Whether macrophages revert to a quiescent phenotype following tissue recovery or
undergo unidirectional activation followed by apoptotic death is unknown. The
inhibitor (I)-kappaB kinase complex, which controls the activation of nuclear factor
(NF)-kappaB transcription factors, plays a pivotal role in switching off macrophage
activation and the resolution of inflammation,36,37 and in vitro macrophage
deactivation can be induced by various signals, including IL-10 and TGFb, or ligation
of the inhibitory receptors CD172 and CD200, which leads to anti-inflammatory cyto-
kine production and a reduction in major histocompatibility complex (MHC) class II
expression.12
Cross-talk between macrophages and liver fibroblasts leads to changes in function
and in the patterns of cytokines secreted by both cell types.38,39 Survival of activated
HSC in the fibrotic liver depends on expression of tissue inhibitors of metalloprotei-
nases (TIMPs) (which prevent degradation of matrix proteins) and survival signals
from accumulated fibrillar type 1 and 3 collagens. The failure to degrade collagen-1
critically impairs HSC apoptosis, thereby perpetuating fibrosis.40 Recovery from liver
fibrosis is associated with the secretion of macrophage-derived proteases that
degrade cross-linked fibrillar collagens, allowing other, more promiscuous, matrix
metalloproteinases (MMP) to accelerate matrilysis and thereby remove the survival
signals on which activated HSC depend.4,41 In addition, macrophages may actively
promote apoptosis of HSC by expressing TRAIL and other apoptotic stimuli that
accelerate clearance of myofibroblasts at sites of scar formation.42
Thus, the resident populations of macrophages within the liver direct different stro-
mal responses to injury at different times, but whether this reflects functionally distinct
populations of cells derived from distinct blood monocyte precursors or is the result of
a local intrahepatic switch in function is unknown.
STROMAL CELLS MODULATE THE DIFFERENTIATION AND FUNCTION

OF ANTIGEN-PRESENTING CELLS
The noninflamed liver contains a subset of myeloid DC that preferentially secrete IL-10
and induce IL-10 secretion in responding T cells.43,44 These DC may play an important
role in suppressing damaging immune responses in noninflammatory states. In the
normal human liver, most DC are CD161 and express high levels of CX3CL1.
CD161 circulating monocytes can differentiate into DC during the process of transen-
dothelial migration45 and this process may be enhanced by interactions with fibro-
blasts.46 The interaction between stroma and leukocytes in human tissues has
received little attention, although accumulating evidence suggests that HSC and ECM
provide differentiation signals to monocytes during recruitment into the liver. Human
spleen–derived myofibroblasts express IL-15, which drives CD341 blood cells to
differentiate into activated NK cells and myeloid DC after 3 to 4 weeks in culture, dem-
onstrating how stromal cells in tissues can shape the innate immune system.47 Such
interactions will also determine the nature of the DC generated. Murine splenic DC can
be generated from hematopoietic precursors by culture on splenic stromal cells46 and
these DC have regulatory properties similar to those seen in human liver.44,48 Matura-
tion depended on fibronectin-mediated adhesion and CXC chemokine ligand
(CXCL)12 secreted by the fibroblasts. The DC generated on the stromal cells mediated
suppression of T-cell proliferation in response to antigen presentation by mature DC.
These two studies show how different fibroblast-derived cytokines, IL-15 in the former
study and CXCL12 in the latter, may drive different differentiation pathways of interact-
ing DC and thereby shape the subsequent immune response.
Similar processes have been described in the skin, where interactions between
blood monocytes and human dermal fibroblasts mediated by b2 integrins, intracellular
adhesion molecule (ICAM)-1 and thymocyte-1 (Thy)-1 (CD90) induce the maturation of
DC with potent T cell–activating capabilities.49 The importance of the local
microenvironment is emphasized by the finding that culturing human monocyte-
derived DC in liver-conditioned media induced a regulatory phenotype similar to
that seen in DC isolated directly from the liver, whereas culturing in skin-conditioned
media had no such effect.44 IL-10 is the critical cytokine mediating this effect because
defective hepatic DC function could be restored by inhibiting IL-10, but the precise
role played by HSC in modulating these responses has not been determined.50
Thus, the outcome of interactions between DC precursors and fibroblasts is likely to
be tissue-specific responses modulated by local inflammatory signals that regulate
DC differentiation and function.
Recent studies have demonstrated that HSC can themselves act as professional
antigen-presenting cells (APC) by efficiently presenting antigens to MHC-I– and
MHC-II–restricted T cells and lipid antigens to CD1-restricted natural killer T (NKT)
cells. HSC share many phenotypic similarities with follicular DC, which are specialized
APC located in secondary lymphoid tissue.51 Unlike other resident liver APC, notably
sinusoidal endothelium and hepatocytes, the outcome of presentation by HSC ap-
pears to be full effector responses rather than tolerance. Thus, HSC not only modulate
the differentiation of other APC but can also directly stimulate immune responses. As
immune responses develop and HSC differentiate into activated myofibroblasts, their
role may shift from direct antigen presentation to indirect effects on monocyte
differentiation, resulting in complex effects on intrahepatic immune responses.52
Gene array studies have shown that as HSC mature, they switch gene transcription
from a fibrotic to a proinflammatory phenotype, providing evidence that senescent
HSC continue to modulate local inflammatory responses long after they have stopped
remodeling the ECM in response to liver injury.53
HEPATIC STELLATE CELL ACTIVATION IN THE LIVER PROMOTES CHRONIC INFLAMMATION
Liver myofibroblasts contribute to leukocyte recruitment to the injured liver by secret-

ing cytokines, chemokines, and other chemoattractants, such as platelet-activating
factor and complement proteins.54–56 In most circumstances, inflammation resolves
once the initiating stimulus has been removed, but in some conditions, inflammation
persists, resulting in chronic hepatitis and fibrogenesis.57 The switch from acute
resolving to chronic persistent hepatitis is associated with the development
866 Holt et al
of a specialized microenvironmental niche that maintains leukocytes, particularly lym-

phocytes in close proximity to scar-associated fibroblasts at sites of tissue injury.5
Chronic inflammation in other tissues including the kidney, lung, and eye is also asso-
ciated with the presence of closely associated lymphocytes and activated fibroblasts,
indicating that these links are a generic feature of fibrotic diseases.1,58–60
To date, the role of lymphocyte–fibroblast interactions in driving chronic
inflammation has been most closely studied in rheumatoid arthritis, but it is likely
that general mechanisms defined in the joint will apply to chronic inflammation in
the liver. Fibroblasts in rheumatoid joints shape the inflammatory infiltrate in several
ways.61 Fibroblast-derived chemokines can be presented on the endothelium of ves-
sels in the joint, thereby contributing to inflammatory cell recruitment from blood. In
addition, the local secretion of CXCL12 activates integrin-mediated adhesion of lym-
phocytes to fibroblast vascular cell adhesion molecule (VCAM)-1, thereby retaining
lymphocytes in the joint.62,63 Other cytokines secreted by fibroblasts, particularly
type 1 IFN, provide survival signals for activated T cells and prevent them from dying
by apoptosis.64 Thus, fibroblasts contribute to the persistence of chronic joint
inflammation by promoting lymphocyte recruitment from blood, by retaining infiltrating
lymphocytes, and by providing them with survival signals that prevent death by apo-
ptosis, a critical component of immune resolution.1
The interactions between fibroblasts and lymphocytes are bidirectional, and T cell–
derived cytokines are potent activators of fibroblasts in many chronic inflammatory dis-
eases.65–67 The outcome of fibroblast activation will be determined by the activation of
disease-specific and tissue-specific signaling pathways; in humans, polymorphisms in
regulatory genes can result in major differences in response to particular stimuli.68
Thus, lipopolysaccharide-mediated NF-kappaB activation and CXCL8 secretion by
liver myofibroblasts vary among patients who have liver disease.69,70 Failure to regulate
activation of NF-kappaB leads to sustained inflammation and a tendency toward a Th1-
polarized immune response,36,37 which may, in part, be driven by uncontrolled fibro-
blast activation. Thus, mice that have a selective deletion of RelB (a transcriptional
regulator that stabilizes the NF-kappaB inhibitor I-kappaB) in fibroblasts develop an
overwhelming systemic inflammatory response to endotoxin, which is characterized
by persistent chemokine induction and leukocyte recruitment. Transfecting the wild-
type gene back into these animals attenuates inflammation by restoring RelB-mediated
stromal regulation.71 More recently, it has been shown that mice lacking NF-kappaB-1
(p50) treated with repeated injections of carbon tetrachloride (CCl4) develop a more se-
vere inflammatory and fibrotic reaction than wild-type controls. This reaction is asso-
ciated with increased stellate cell expression of tumor necrosis factor (TNF)a in
areas of inflammation, suggesting that the p50 subunit of NF-kappaB may protect
the injured liver by limiting TNFa-dependent inflammatory cell recruitment.72
Thus, iterative cycles of fibroblast activation at sites of chronic injury result in the
development of a stromal phenotype that perpetuates inflammation and leukocyte
recruitment, resulting in dysregulation of the normal processes of tissue repair.10
HEPATIC STELLATE CELL ACTIVATION IS ASSOCIATED WITH LEUKOCYTE RECRUITMENT
Quiescent HSC are specialized pericytes found immediately adjacent to the sinusoids
in the space of Disse. They maintain endothelial integrity by secreting angiogenic pep-
tides in response to bidirectional interactions with the overlying endothelium.73 The
expression of markers associated with stem cells such as CD133, Thy-1 (CD90),
nanog, and Oct-4, and their ability to transdifferentiate into hepatocytes and endothe-
lial-like cells suggest that HSC should also be regarded as progenitor cells.73,74
Quiescent HSC undergo a phenotypic switch in response to liver injury, becoming

spindle-shaped and secreting proinflammatory cytokines and up-regulating programs
of genes that result in excess deposition of fibrillar collagens.6,75 Although evidence is
increasing that human liver fibrosis is the product of both intra- and extrahepatic pop-
ulations of fibroblasts, activated HSC are still regarded as the principal collagen-pro-
ducing cells in the fibrotic liver.76
In liver fibrosis, HSC-like cells are seen immediately adjacent to fibrovascular mem-
branes at the margins of fibrotic septa associated with the inflammatory infiltrate.77–79
Thus, although paracrine interactions with quiescent HSC maintain the overlying
endothelium, HSC activation results in fibrogenesis associated with an inflammatory
infiltrate. HSC activation is precipitated in various ways; reactive oxygen intermediates
released from injured cells activate quiescent HSC while activation signals are
generated within the stroma as latent TGFb-1 is activated by way of plasmin released
by injured endothelial cells.6,80 Perpetuation of HSC activation is associated with
increased sensitivity to local cytokine signals (particularly platelet-derived growth
factor [PDGF]) and the effects of remodeling the existing ECM.81 These enhanced
cytokine responses typify activated HSC and are principally mediated by up-
regulation of receptor tyrosine kinases.80,82
MIGRATION AND POSITIONING OF T CELLS ARE DETERMINED BY PATTERNS OF CYTOKINE

AND CHEMOKINE SECRETION WITHIN THE STROMA
The liver responds to injury by recruiting leukocytes from the circulation, expanding
populations of scar-forming fibroblasts, and positioning the inflammatory infiltrate
within the modified neomatrix. An effective hepatic immune response requires that
leukocytes be recruited to the liver and then appropriately positioned at sites of
injury.57 Leukocyte adhesion to endothelial cells is a prerequisite for recruitment into
tissue and is regulated by a sequence of interactions in which the leukocyte is cap-
tured from the flowing blood by carbohydrate-dependent mechanisms and activated
by chemokines presented on the endothelial glycocalyx, resulting in integrin activation
and arrest on the vessel wall, allowing the cell to then migrate through endothelium
into tissue by as yet poorly understood mechanisms. To trigger leukocyte recruitment,
chemokines must be presented at the endothelium where they can be delivered by
transcytosis from underlying cells and then presented bound to proteoglycans in
the endothelial glycocalyx.83 Thus, the chemokines present on sinusoidal endothelium
during liver inflammation may be derived from paracrine secretion by Kupffer cells,
other infiltrating leukocytes, hepatocytes, and activated HSC. The position of HSC im-
mediately beneath and in close association with endothelium places them in an ideal
site to ‘‘post’’ chemokines onto the overlying endothelium.
As a result of the complex microvascular anatomy of the liver, leukocytes can enter
the liver by way of endothelial cells at three anatomic sites, the portal tract, the sinu-
soids, or the terminal hepatic veins.84 Although some branches of the hepatic artery
and portal veins terminate in postcapillary venules, most blood is diverted to the liver
parenchyma via the sinusoids before returning to the systemic circulation by way of
hepatic veins, and the sinusoidal bed provides the major route of entry into the paren-
chyma. Recent studies show that recirculating lymphocytes enter the liver by way of
sinusoids and, if not retained, then migrate by way of the space of Disse to portal
tracts before moving on to draining lymph nodes. DC, which enter by way of the sinu-
soids as monocyte precursors, take the same route by way of portal tracts to draining
lymph nodes after activation.85 The lumen of the sinusoids is narrow and blood flow is
regulated in part by stellate cells acting as pericytes, which facilitates the attachment
868 Holt et al
of flowing leukocytes.86–88 Leukocytes interact with sinusoids in a modified version of

the adhesion cascade in which the capture phase is attenuated and does not involve
selectins that capture leukocytes in other tissues.89,90 A different adhesion molecule,
vascular adhesion protein 1 (VAP-1), which is constitutively expressed on hepatic en-
dothelium, mediates capture and transmigration of leukocytes on hepatic endothe-
lium.91,92 Once captured and firmly adherent, a lymphocyte must breach the
endothelial barrier to enter tissue and, under physiologic shear stress, VAP-1 and
ICAM-1 promote lymphocyte transendothelial migration through sinusoidal
endothelium.92 Selection at the endothelial level is only one aspect of the processes
that govern the development of inflammatory infiltrate cells; retention of leukocytes
within tissue is equally important, although much less studied. Based on studies in
other organs, particularly the thymus and bone marrow, it is likely that stromal cells
will be integral to this process.10,93
HSC accumulate at sites of tissue injury in response to PDGF and CCL2,94 and once
activated, secrete proinflammatory cytokines and chemokines that up-regulate adhe-
sion molecules such as ICAM-1, VCAM-1, and CXC3CR1, which facilitate interactions
with local leukocytes.3,5,49,95 HSC can modulate leukocyte recruitment in at least three
ways: (1) by paracrine interactions, which prime overlying sinusoidal endothelial cells
to recruit leukocytes; (2) by secreting matrix, which provides a substrate through
which infiltrating leukocytes can migrate or be retained; and (3) by direct interactions
with leukocytes to modulate their migratory behavior or retain them at sites of injury.
Leukocytes entering the subendothelial space interact with stromal cells and ECM
by way of cell surface integrins, which not only provide anchorage but also mediate the
‘‘outside-in’’ signals that direct cellular activation and differentiation.10 Injury to the
liver stimulates vigorous infiltration of leukocytes as a result of cytokines and chemo-
kines displayed on the endothelium and within the subendothelial stroma.96 Once in
the subendothelial space, lymphocytes undergo directed migration within the stroma
by responding to signals displayed within the ECM. The distribution and spatial
organization of the hepatic infiltrate varies according to the inflammatory stimulus
and reflects the expression of chemokines and adhesion molecules within the tissue.57
Thus, in inflamed liver, adhesion molecules and chemokines are increased, and in the
sinusoids, both endothelial cells and activated HSC secrete the potent lymphocyte-
recruiting chemokines CXCL9 and CXCL10 in response to TNFa and IFNg.97
Simply recruiting inflammatory cells is not enough to maintain an effective immune
response because chronic inflammation requires that effector cells be retained in tis-
sues. Adhesion molecules and chemokines, including CXCL16 and fractalkine
(CX3CR1), are expressed by target epithelial cells and stromal cells at sites of injury
where they localize infiltrating T cells.98–101 Thus, stromal expression of chemokines
determines the position and shape of the resulting inflammatory infiltrate and serves
to localize leukocytes to areas of fibrosis.
ACTIVATED HEPATIC STELLATE CELLS MODULATE RECRUITMENT AND POSITIONING

OF LEUKOCYTES WITHIN THE LIVER
As a consequence of being situated between the endothelium and parenchyma, HSC

can participate in the recruitment of leukocytes into the liver. The first evidence that
activated liver myofibroblasts contributed to liver inflammation came from the obser-
vation that CCL2 in HSC-conditioned media promoted vigorous monocyte chemo-
taxis.54 CCL2 also recruits memory T cells, basophils, and DC, suggesting it may be
an important recruitment signal involved in chronic inflammation.102,103 Studies in
rat models of liver fibrosis revealed that nearly all of the CCL2 generated came from
activated HSC.104 Moreover, activated HSC isolated from fibrotic human livers con-
tinue to secrete high concentrations of CCL2 in vitro even in the absence of proinflam-
matory cytokines. Activated myofibroblasts secrete a wide variety of other
chemokines, cytokines, and matricellular proteins that regulate recruitment and sub-
sequent differentiation of neutrophils, macrophages, and lymphocytes (Fig. 2). The
authors’ own unpublished data show that human hepatic myofibroblasts secrete
not only CXCL8 and CCL2 but also CCL5, CCL11, and CCL3, and in response to
INFg and TNFa, large amounts of CXCL9 and CXCL10.105 These factors in cell super-
natants from liver fibroblasts exert a powerful chemotactic effect on human lympho-
cytes in real-time chemotaxis studies (Fig. 3), increasing the rate and magnitude of
lymphocyte migration and demonstrating the potential of liver fibroblasts to direct lym-
phocyte migration within the stroma.
Fig. 2. Activated HSC control many aspects of tissue inflammation. Following activation, HSC
undergo a phenotypic switch, adopting a myofibroblast morphology and up-regulating
proinflammatory cytokines that regulate stromal immune responses and secreting fibrillar
collagens and neomatricellular proteins that generate tissue scarring. G-CSF, granulocyte
colony stimulating factor; GM-CSF, granulocyte monocyte colony stimulating factor; HGF,
hepatocyte growth factor; PAF, platelet activating factor.
870 Holt et al
Fig. 3. Real-time lymphocyte chemotaxis to liver fibroblast-conditioned media. Directed real-

time migration of fluorescently labeled human lymphocytes toward liver fibroblast-condi-
tioned media.147 Conditioned supernatant taken from human liver fibroblasts isolated
from diseased livers contains soluble factors that promote rapid lymphocyte chemotaxis
even in the absence of proinflammatory cytokine stimulation. Data represents mean stan-
dard error 3 replicate experiments.
The importance of lymphocytes in fibrosis is illustrated by the fact that animals with
targeted disruptions to chemokines that promote lymphocyte recruitment show re-
duced fibrotic responses to experimentally induced liver injury, whereas inhibition of
neutrophil chemotaxis has no effect on fibrosis.27,35 Cytokine therapy with IL-10
antagonizes proinflammatory chemokines and cytokines secreted by activated liver
fibroblasts and reduces fibrosis in humans who have chronic viral hepatitis and ani-
mals that have experimentally induced hepatitis.106,107 Fibroblasts secrete other cyto-
kines that can promote lymphocyte migration in addition to chemokines. IL-6,
hepatocyte growth factor, and TGFb, all of which are secreted by aSMA1 liver myofi-
broblasts,108 have lymphocyte chemotactic activity and could contribute to the
recruitment and positioning of lymphocytes within the stroma.109–112
Chronic inflammation is associated with the accumulation of T cells, B cells, and DC in
ectopic or tertiary lymphoid structures that resemble lymph nodes.113 Their develop-
ment is a consequence of aberrant and sustained expression of cytokines that drive
lymph node development, including lymphotoxin and homeostatic chemokines.
CCL21 is an important chemokine in lymph node development and function because
it is critical for the recruitment of naive T cells and DC to lymph nodes.114 CCL21 is se-
creted by liver fibroblasts, including those found associated with tertiary lymphoid struc-
tures in chronic inflammatory diseases including primary biliary cirrhosis (PBC), primary
sclerosing cholangitis (PSC), and chronic hepatitis C.99,115,116 These findings suggest
that activated fibroblasts may be responsible for the development and persistence of
lymphoid follicles, providing another mechanism by which they can modulate the re-
cruitment of lymphocytes and determine the nature of chronic hepatic inflammation.
Thus, activated HSC and liver fibroblasts at sites of tissue injury contribute to leuko-
cyte recruitment by secreting chemokines and nonchemokine chemotactic factors
that drive G-protein coupled receptor associated and G-protein coupled receptor
independent and mechanisms, respectively. These mechanisms operate throughout

the inflammatory response, playing a role in early leukocyte recruitment and then in
the remodeling of the stromal microenvironment to promote the recruitment and sur-
vival of T cells, B cells, and DC as part of the chronic inflammation that accompanies
fibrogenesis. Because fibroblasts are not all the same, site-specific differences among
stromal cells may play an important role in determining the distinct patterns of inflam-
mation seen in particular diseases or tissues.17,19
DIRECT INTERACTIONS WITH ACTIVATED LIVER FIBROBLASTS RETAIN LYMPHOCYTES

IN TISSUE AND PREVENT RESOLUTION OF INFLAMMATION
The close temporal and spatial association between infiltrating lymphocytes and
fibrosis has been recognized for many years, but the cellular interactions are poorly
understood. In other tissues, T cells and B cells rely on cellular elements within the
stroma for structural anchorage and directed migration. Thus, within secondary lymph
nodes, fibroblastic reticular cells provide the adhesive scaffold that facilitates lympho-
cyte migration within the lymph node and that maintains the zonal separation of T cells
and B cells.117 These stromal cells bring together T cells and DC by secreting
chemokines that attract both cell types within the T zone in the medulla.118,119 Simi-
larly, stromal cells play a critical role in recruiting and positioning thymocytes within
the thymus during lymphocyte development. It is likely that liver fibroblasts play
a similar role in directing lymphocyte recruitment to the inflamed liver. In human liver
disease, infiltrating lymphocytes are closely associated with myofibroblasts at the
margins of fibrotic septa and with aSMA1 myofibroblasts within portal tracts. In
mice that have experimentally induced liver fibrosis, lymphocytes are only found in
the proximity of activated HSC/activated liver myofibroblasts suggesting a direct inter-
action exists between liver myofibroblasts and infiltrating lymphocytes.5 Similarly, NK
cells are found attached to aSMA1 HSC following fibrosis induction along the interface
zone of fibrotic septa,24 suggesting that paracrine interactions between fibroblasts
and lymphocytes at sites of tissue damage facilitate the matricellular changes associ-
ated with fibrosis and chronic inflammation.
Studies in animals indicate that the release of CCL2 and expression of ICAM-1 by
stellate cells immediately precede leukocyte infiltration in the early phase of liver injury,
indicating that HSC activation promotes the recruitment of leukocytes into the
parenchyma.120 Part of this response is mediated by the secretion of cytokines and
chemokines that promote recruitment through the overlying endothelium, but recent
evidence suggests that fibroblasts also modulate lymphocyte migration by direct
contacts with the infiltrating cells. The authors have shown that if lymphocytes are
placed in contact with myofibroblasts isolated from diseased human liver, they
show increased motility and migrate rapidly over and through the fibroblast monolayer
in a process termed ‘‘pseudoemperopolesis.’’ Treatment of the fibroblasts with
proinflammatory cytokines increases the kinetics of stromal transmigration, which is
dependent on increased expression of ICAM-1 and VCAM-1.95 Murine studies have
demonstrated that HSC support lymphocyte adhesion in an ICAM-1–dependent
manner, and human HSC also express high levels of CD90, which supports adhesive
interactions with DC.49,95,121 In response to tissue injury, liver myofibroblasts provide
an adhesive template that facilitates stromal migration and approximates lymphocytes
and fibrotic HSC-like cells in regions of matrix remodeling, thus perpetuating the in-
flammatory reaction by promoting leukocyte recruitment at sites of tissue damage.
The development of a stable chronic inflammatory infiltrate depends on the recruit-
ment of leukocytes from blood and their subsequent survival at the site of
872 Holt et al
inflammation outweighing the mechanisms of resolution, which include local death of

inflammatory cells by apoptosis and emigration out of the inflammatory site by lym-
phatics.1 Stromal cells modulate all of these processes, including the survival of the
inflammatory infiltrate, as interactions with synovial fibroblasts induce expression of
anti-apoptotic genes in T cells that prevent apoptosis. Fibroblasts can also keep neu-
trophils alive, particularly by secreting cytokines, including granulocyte-macrophage
colony-stimulating factor (GM-CSF) and type 1 IFN. These survival signals are ampli-
fied by adhesive interactions between neutrophil a9b1 integrin and VCAM-1 on stro-
mal cells.63,122–124 The cycles of tissue injury and repair associated with chronic
liver disease lead to local secretion of cytokines, including IL-6, IL-12, and TNFa, by
activated HSC, all of which can promote antigen-independent bystander T-cell activa-
tion. Thus, the cytokine milieu in the liver of patients who have chronic hepatitis is
shaped by activated HSC and favors not only lymphocyte survival but also bystander
activation of T cells, promoting effector T-cell function even in the absence of the orig-
inal stimulating antigen.125,126 This process amplifies antigen-specific responses
within the tissue, but also increases the resistance of responder cells to apoptotic
stimuli.127 As a result, infiltrating T cells become increasingly resistant to apoptosis,
whereas bystander activation increases their effector function, thereby exacerbating
tissue damage and fibrosis.128
STROMAL INTERACTIONS WITH INFILTRATING LYMPHOCYTES SHAPE FIBROGENESIS

IN CHRONIC HEPATITIS
Although interactions among macrophages and fibroblasts are critical for fibrosis
induction and resolution,129 the role of the adaptive immune system is less well
defined (Fig. 4). Nevertheless, good evidence suggests that T cells and B cells are
implicated in sustaining fibrotic responses in chronic liver disease.130 In animal models
and diseased human liver, lymphocytes are seen in close association with fibroblasts
Fig. 4. Liver fibrosis is the product of interactions between stromal cells and the innate and
adaptive immune systems. aHSC, activated hepatic stellate cells.
in inflamed portal tracts and fibrous septa, suggesting the involvement of the adaptive
immune response in fibrogenesis.5,96 Functional support for this hypothesis comes
from animal studies in which depletion of T cells and B cells protects animals from liver
fibrosis.106,131 Moreover, mice deficient in B cells and T cells (recombination activating
gene 2 [RAG2] / ) are resistant to fibrosis in acute and chronic models of liver in-
jury.131 The specific role of particular lymphocyte subsets during fibrogenesis is less
clear.
The contribution of intrahepatic T cells to fibrogenesis is accepted, and adoptive
transfer studies in various murine models confirm the involvement of CD41 and
CD81 T cells. CD41 T cells were found to secrete a fibrotic factor in a Th2-polarized
mouse model of liver fibrosis, which was subsequently shown to be IL-13.132,133 The
polarity of the inflammatory response is critical for fibrogenesis, and different pro-
grams of gene expression are induced when chronic inflammatory responses are
dominated by Th1 or Th2 cytokines. Although Th1 cytokines characteristically gen-
erate an intense cellular response, they cause little fibrosis,134 whereas Th2 cyto-
kines increase transcription of several genes involved in fibrogenesis, including
procollagen I and III, MMP2, MMP9, and TIMPs,30 all of which are expressed in
CCl4 liver injury, suggesting that mechanisms of fibrosis may share a common cyto-
kine profile. The importance of Th2-polarized T-cell subsets is confirmed by studies
demonstrating that IL-13Ra2–blocking antibodies reduce liver fibrosis132 and inhibi-
tion of Th2 responses using targeted mutations of IL-4, STAT-6, and IL-4Ra atten-
uate fibrosis in a mouse model of scleroderma.32 Other studies have
demonstrated that transfer of CD81 (but not CD41) T cells from mice that have
CCl4-induced liver fibrosis provokes significant liver injury in severe combined immu-
nodeficient (SCID)-mice recipients.106 Conversely, fibrosis is attenuated in trans-
genic mice that overexpress IL-10 in hepatocytes and in animals depleted of
T lymphocytes. Although CD41 T cells and CD81 make important contributions
to fibrogenesis, their precise role is unclear. CD4 T cells can modulate the local cy-
tokine response to favor fibrogenesis by way of direct effects on stromal macro-
phages and fibroblasts, and CD8 T cells may promote fibrosis by amplifying
tissue injury as part of a bystander response driven by the local cytokine milieu.
The role of B lymphocytes in hepatic fibrosis is poorly understood, although their
ability to activate T cells at low antigen loads and to secrete cytokines suggests
they could be involved in shaping and maintaining the inflammatory response. Regu-
latory B cells can direct the polarity of the local inflammatory response away from
a Th1-like phenotype in autoimmunity and may contribute to local production of
TGFb-1.135,136 A recent study demonstrated an antibody-independent role for B cells
in liver fibrosis induced by two different agents, CCl4 or a-naphthylisothiocyanate. This
study was unable to show any protective effect in CD4, CD8, or gd T cell–deplete mice
but found that mice lacking T cells and B cells were resistant to experimental fibrosis
despite mounting a similar inflammatory response to wild-type animals.131 A role for
antibody was excluded because mice that were unable to secrete immunoglobulin still
showed reduced fibrosis when B cells were depleted.131 The fact that T cell–deficient
mice showed normal fibrotic responses in this model argues against a role for B cells in
antigen presentation. Therefore, it is most likely that B cells promote fibrosis by secret-
ing cytokines or by contact-dependent interactions with other cells that favor a profi-
brotic microenvironment. Differentiation of naive B cells into Th1-like or Th2-like
patterns of cytokine secretion is determined by cytokine microenvironments similar
to those that regulate corresponding T-cell differentiation, and B cells secrete the pro-
fibrotic cytokines IL-4, IL-6, and IL-13 at levels similar to CD41 T cells.137 Although
B cell receptor (BCR) ligation is required for optimal cytokine secretion, repetitive
874 Holt et al
BCR-independent stimulation is sufficient to stimulate naive B cells to secrete cyto-

kines in a polarized fashion.138,139 Thus, the repeated cycles of tissue injury seen in
human liver disease could maintain B-cell activation and allow the process to become
self-perpetuating. Interactions among NKT cells, macrophages, and B cells within this
fibrotic environment will also reinforce local cellular responses and secretion of
TGFb.140 Costimulatory signals provided by B cells to T cells, NKT cells, and Kupffer
cells may also be important.
CD40 is a strong candidate for a central role in these paracrine interactions. The
ligand for CD40, CD154, is strongly expressed on lymphocytes and macrophages
at sites of liver injury in several diseases31,141 and could activate B-cell CD40, resulting
in cell survival, IL-6 secretion, and paracrine activation of HSC proliferation and colla-
gen synthesis. Dysregulated bystander activation of B cells is usually prevented by ac-
tivation of Fas-dependent apoptosis, but survival signals such as B cell activating
factor of TNF family (BAFF)/B lymphocyte stimulator that are increased in the fibrotic
microenvironment may inhibit Fas-mediated apoptosis, allowing inappropriate sur-
vival of effector B cells and amplification of fibrosis.1 However, other animal studies
have failed to show a major role for B cells in liver fibrosis. B-cell depletion in a Th2-
dominated model of parasitic liver fibrosis did not provide any protection against fibro-
sis, which might be because the strongly polarized Th2 response in this model was
unaffected by B cells. The example illustrates that different cellular mechanisms
may be involved, depending on the nature of liver injury and the pattern of
inflammation.142
Recent interest has focused the role of NKT cells in fibrogenesis. Invariant NKT cells
are a subset of lymphocytes that recognize endogenous lipid ligands presented by
CD1d. They regulate host responses to cell injury, tissue damage, and viral infection,
and are potent secretors of cytokines. NKT cells are found in normal liver tissues but
their numbers increase with necroinflammatory activity, and in the context of chronic
viral hepatitis they have been shown to secrete type 2 profibrotic cytokines including
IL-4 and IL-13, which could drive progression of fibrosis. Part of this may be driven by
increased expression of CD1d on APC in chronically inflamed liver tissue.143 These
studies are supported by animal experiments in which administration of the lipid
b-glucosylceramide results in reduced intrahepatic NKT cells associated with amelio-
ration of fibrosis.144 However, further complexity is likely because in other models
IFNg secreting NKT cells have an antifibrotic role.145 A recent study in the double-neg-
ative TGFb-RII mouse model of primary biliary cirrhosis shows a marked increase of
NKT cells within the liver associated with tissue injury. When the same mice were
crossed onto a CD1 / background, in which no CD1d-restricted NKT cells were lo-
cated, liver inflammation and injury were markedly decreased, suggesting NKT cells
are involved in liver injury in this model. Furthermore, they showed age-dependent dif-
ferences in IFNg secretion by hepatic CD1d-restricted NKT cells, suggesting that their
precise role may change with ageing.146,147 Thus, it is the combinations of leukocytes
recruited and the outcome of complex cell–cell crosstalk and cytokine expression
within the stroma that determine the nature and progression of fibrosis, regardless
of the specific lymphocytes in the inflammatory infiltrate.
SUMMARY
Advances in our understanding of liver fibrosis have highlighted the importance of in-
teractions between liver fibroblasts and the innate and adaptive arms of the immune
system in determining the outcome following tissue injury. Although many acute
injuries to the liver result in self-limiting nonfibrotic inflammatory reactions, chronic
hepatitis is associated with persistent inflammation and scarring because the devel-
opment of a stromal microenvironment comprising modified neomatrix, activated fi-
broblasts, and macrophages recruits, retains, and maintains lymphocytes at the site
of tissue injury, resulting in chronic, persistent inflammation. The lymphocytes, in
turn, secrete cytokines and interact directly with cellular components of the stroma
to maintain the local profibrotic environment. Strategies that aim to restore liver func-
tion, either by preventing fibrosis progression or by replenishing hepatocytes from
stem cell precursors, must take these complex, bidirectional interactions into account.
Future antifibrotic interventions in liver disease may use the complex interactions be-
tween fibroblasts and T cells to ameliorate the production of scar tissue and prevent
inflammation from persisting. In this context, the use of specific protein tyrosine-
kinases inhibitors such as imatinib (Glivec) to treat fibrotic liver diseases may offer par-
ticular promise as a means of preventing proliferation of fibroblasts in response to
chronic hepatitis and also of inhibiting stromal patterns of cytokine secretion.
REFERENCES
1. Buckley CD, Pilling D, Lord JM, et al. Fibroblasts regulate the switch from acute
resolving to chronic persistent inflammation. Trends Immunol 2001;22:199–204.
2. Iredale JP, Benyon RC, Pickering J, et al. Mechanisms of spontaneous resolution
of rat liver fibrosis. Hepatic stellate cell apoptosis and reduced hepatic expres-
sion of metalloproteinase inhibitors. J Clin Invest 1998;102:538–49.
3. Knittel T, Dinter C, Kobold D, et al. Expression and regulation of cell adhesion
molecules by hepatic stellate cells (HSC) of rat liver: involvement of HSC in
recruitment of inflammatory cells during hepatic tissue repair. Am J Pathol
1999;154:153–67.
4. Duffield JS, Forbes SJ, Constandinou CM, et al. Selective depletion of
macrophages reveals distinct, opposing roles during liver injury and repair.
J Clin Invest 2005;115:56–65.
5. Muhanna N, Horani A, Doron S, et al. Lymphocyte-hepatic stellate cell proximity
suggests a direct interaction. Clin Exp Immunol 2007;148:338–47.
response to tissue injury. J Biol Chem 2000;275:2247–50.
7. Filer A, Pitzalis C, Buckley CD. Targeting the stromal microenvironment in
chronic inflammation. Curr Opin Pharmacol 2006;6:393–400.
of the liver. Physiol Rev 2008;88:125–72.
9. Iredale JP. Hepatic stellate cell behavior during resolution of liver injury. Semin
Liver Dis 2001;21:427–36.
10. Parsonage G, Filer AD, Haworth O, et al. A stromal address code defined by
fibroblasts. Trends Immunol 2005;26:150–6.
11. Abe R, Donnelly SC, Peng T, et al. Peripheral blood fibrocytes: differentiation
pathway and migration to wound sites. J Immunol 2001;166:7556–62.
12. Gordon S, Taylor PR. Monocyte and macrophage heterogeneity. Nat Rev Immu-
nol 2005;5:953–64.
13. Geissmann F, Jung S, Littman DR. Blood monocytes consist of two principal
subsets with distinct migratory properties. Immunity 2003;19:71–82.
14. Phillips RJ, Burdick MD, Hong K, et al. Circulating fibrocytes traffic to the lungs
in response to CXCL12 and mediate fibrosis. J Clin Invest 2004;114:438–46.
15. Quan TE, Cowper SE, Bucala R. The role of circulating fibrocytes in fibrosis. Curr
Rheumatol Rep 2006;8:145–50.
876 Holt et al
16. Kisseleva T, Uchinami H, Feirt N, et al. Bone marrow-derived fibrocytes

participate in pathogenesis of liver fibrosis. J Hepatol 2006;45:429–38.
17. Spanakis E, Brouty-Boye D. Discrimination of fibroblast subtypes by multivariate
analysis of gene expression. Int J Cancer 1997;71:402–9.
18. Chang HY, Chi JT, Dudoit S, et al. Diversity, topographic differentiation, and
positional memory in human fibroblasts. Proc Natl Acad Sci U S A 2002;99:
12877–82.
19. Brouty-Boye D, Doucet C, Clay D, et al. Phenotypic diversity in human fibro-
blasts from myelometaplasic and non-myelometaplasic hematopoietic tissues.
Int J Cancer 1998;76:767–73.
20. Magness ST, Bataller R, Yang L, et al. A dual reporter gene transgenic mouse
demonstrates heterogeneity in hepatic fibrogenic cell populations. Hepatology
2004;40:1151–9.
21. Cassiman D, Roskams T. Beauty is in the eye of the beholder: emerging con-
cepts and pitfalls in hepatic stellate cell research. J Hepatol 2002;37:527–35.
22. Marinova-Mutafchieva L, Taylor P, Funa K, et al. Mesenchymal cells expressing
bone morphogenetic protein receptors are present in the rheumatoid arthritis
joint. Arthritis Rheum 2000;43:2046–55.
23. Dunn C, Brunetto M, Reynolds G, et al. Cytokines induced during chronic
hepatitis B virus infection promote a pathway for NK cell-mediated liver damage.
J Exp Med 2007;204:667–80.
24. Melhem A, Muhanna N, Bishara A, et al. Anti-fibrotic activity of NK cells in exper-
imental liver injury through killing of activated HSC. J Hepatol 2006;45:60–71.
2006;130:435–52.
26. Saito JM, Bostick MK, Campe CB, et al. Infiltrating neutrophils in bile duct-
ligated livers do not promote hepatic fibrosis. Hepatol Res 2003;25:180–91.
27. Xu J, Lee G, Wang H, et al. Limited role for CXC chemokines in the pathogenesis
of alpha-naphthylisothiocyanate-induced liver injury. Am J Physiol Gastrointest
Liver Physiol 2004;287:G734–41.
28. Friedman SL. Cytokines and fibrogenesis. Semin Liver Dis 1999;19:129–40.
29. Pauleau AL, Rutschman R, Lang R, et al. Enhancer-mediated control of
macrophage-specific arginase I expression. J Immunol 2004;172:7565–73.
30. Sandler NG, Mentink-Kane MM, Cheever AW, et al. Global gene expression pro-
files during acute pathogen-induced pulmonary inflammation reveal divergent
roles for Th1 and Th2 responses in tissue repair. J Immunol 2003;171:3655–67.
31. Alabraba EB, Lai V, Boon L, et al. Coculture of human liver macrophages and
cholangiocytes leads to CD40-dependent apoptosis and cytokine secretion.
Hepatology 2008;47:552–62.
32. McGaha TL, Bona CA. Role of profibrogenic cytokines secreted by T cells in
fibrotic processes in scleroderma. Autoimmun Rev 2002;1:174–81.
33. Henderson NC, Iredale JP. Liver fibrosis: cellular mechanisms of progression
and resolution. Clin Sci (Lond) 2007;112:265–80.
34. Ide M, Kuwamura M, Kotani T, et al. Effects of gadolinium chloride (GdCl(3)) on
the appearance of macrophage populations and fibrogenesis in thioacetamide-
induced rat hepatic lesions. J Comp Pathol 2005;133:92–102.
35. Imamura M, Ogawa T, Sasaguri Y, et al. Suppression of macrophage infiltration
inhibits activation of hepatic stellate cells and liver fibrogenesis in rats. Gastro-
enterology 2005;128:138–46.
36. Barnes PJ, Karin M. Nuclear factor-kappaB: a pivotal transcription factor in

chronic inflammatory diseases. N Engl J Med 1997;336:1066–71.
37. Lawrence T, Bebien M, Liu GY, et al. IKKalpha limits macrophage NF-kappaB
activation and contributes to the resolution of inflammation. Nature 2005;434:
1138–43.
38. Bonecchi R, Facchetti F, Dusi S, et al. Induction of functional IL-8 receptors by
IL-4 and IL-13 in human monocytes. J Immunol 2000;164:3862–9.
39. Song E, Ouyang N, Horbelt M, et al. Influence of alternatively and classically
activated macrophages on fibrogenic activities of human fibroblasts. Cell Immu-
nol 2000;204:19–28.
40. Issa R, Zhou X, Trim N, et al. Mutation in collagen-1 that confers resistance to the
action of collagenase results in failure of recovery from CCl4-induced liver fibro-
sis, persistence of activated hepatic stellate cells, and diminished hepatocyte
regeneration. FASEB J 2003;17:47–9.
41. Fallowfield JA, Mizuno M, Kendall TJ, et al. Scar-associated macrophages are
a major source of hepatic matrix metalloproteinase-13 and facilitate the resolu-
tion of murine hepatic fibrosis. J Immunol 2007;178:5288–95.
42. Friedman SL. Mac the knife? Macrophages- the double-edged sword of hepatic
fibrosis. J Clin Invest 2005;115:29–32.
43. Cremer I, Dieu-Nosjean MC, Marechal S, et al. Long-lived immature dendritic
cells mediated by TRANCE-RANK interaction. Blood 2002;100:3646–55.
44. Goddard S, Youster J, Morgan E, et al. Interleukin-10 secretion differentiates
dendritic cells from human liver and skin. Am J Pathol 2004;164:511–9.
45. Randolph GJ, Sanchez-Schmitz G, Liebman RM, et al. The CD16(1) (Fcgam-
maRIII(1)) subset of human monocytes preferentially becomes migratory
dendritic cells in a model tissue setting. J Exp Med 2002;196:517–27.
46. O’Neill HC, Ni K, Wilson H. Long-term stroma-dependent cultures are a consis-
tent source of immunostimulatory dendritic cells. Immunol Cell Biol 1999;77:
434–41.
47. Briard D, Azzarone B, Brouty-Boye D. Importance of stromal determinants in the
generation of dendritic and natural killer cells in the human spleen. Clin Exp
Immunol 2005;140:265–73.
48. Zhang M, Tang H, Guo Z, et al. Splenic stroma drives mature dendritic cells to
differentiate into regulatory dendritic cells. Nat Immunol 2004;5:1124–33.
49. Saalbach A, Klein C, Sleeman J, et al. Dermal fibroblasts induce maturation of
dendritic cells. J Immunol 2007;178:4966–74.
50. Lai WK, Curbishley SM, Goddard S, et al. Hepatitis C is associated with
perturbation of intrahepatic myeloid and plasmacytoid dendritic cell function.
J Hepatol 2007;47:338–47.
51. Munoz-Fernandez R, Blanco FJ, Frecha C, et al. Follicular dendritic cells are
related to bone marrow stromal cell progenitors and to myofibroblasts. J Immu-
nol 2006;177:280–9.
52. Winau F, Hegasy G, Weiskirchen R, et al. Ito cells are liver-resident
antigen-presenting cells for activating T cell responses. Immunity 2007;26:
117–29.
53. Schnabl B, Purbeck CA, Choi YH, et al. Replicative senescence of activated
human hepatic stellate cells is accompanied by a pronounced inflammatory
but less fibrogenic phenotype. Hepatology 2003;37:653–64.
54. Marra F, Valente AJ, Pinzani M, et al. Cultured human liver fat-storing cells
produce monocyte chemotactic protein-1. Regulation by proinflammatory cyto-
kines. J Clin Invest 1993;92:1674–80.
878 Holt et al
55. Pinzani M, Carloni V, Marra F, et al. Biosynthesis of platelet-activating factor and

its 1O-acyl analogue by liver fat-storing cells. Gastroenterology 1994;106:
1301–11.
56. Marra F. Hepatic stellate cells and the regulation of liver inflammation. J Hepatol
1999;31:1120–30.
57. Eksteen B, Afford SC, Wigmore SJ, et al. Immune-mediated liver injury. Semin
Liver Dis 2007;27:351–66.
58. Sakai N, Wada T, Yokoyama H, et al. Secondary lymphoid tissue chemokine
(SLC/CCL21)/CCR7 signaling regulates fibrocytes in renal fibrosis. Proc Natl
Acad Sci U S A 2006;103:14098–103.
59. Smith TJ, Koumas L, Gagnon A, et al. Orbital fibroblast heterogeneity may
determine the clinical presentation of thyroid-associated ophthalmopathy.
J Clin Endocrinol Metab 2002;87:385–92.
60. Schmidt M, Sun G, Stacey MA, et al. Identification of circulating fibrocytes as
precursors of bronchial myofibroblasts in asthma. J Immunol 2003;171:380–9.
61. Parsonage G, Falciani F, Burman A, et al. Global gene expression profiles in
fibroblasts from synovial, skin and lymphoid tissue reveals distinct cytokine
and chemokine expression patterns. Thromb Haemost 2003;90:688–97.
62. Buckley CD, Amft N, Bradfield PF, et al. Persistent induction of the chemokine
receptor CXCR4 by TGF-beta 1 on synovial T cells contributes to their
accumulation within the rheumatoid synovium. J Immunol 2000;165:3423–9.
63. Pilling D, Akbar AN, Girdlestone J, et al. Interferon-beta mediates stromal cell
rescue of T cells from apoptosis. Eur J Immunol 1999;29:1041–50.
64. Scheel-Toellner D, Pilling D, Akbar AN, et al. Inhibition of T cell apoptosis by IFN-
beta rapidly reverses nuclear translocation of protein kinase C-delta. Eur
J Immunol 1999;29:2603–12.
65. Azzarone B, Failly-Crepin C, Daya-Grosjean L, et al. Abnormal behavior of
cultured fibroblasts from nodule and nonaffected aponeurosis of Dupuytren’s
disease. J Cell Physiol 1983;117:353–61.
66. Azzarone B, Mareel M, Billard C, et al. Abnormal properties of skin fibroblasts
from patients with breast cancer. Int J Cancer 1984;33:759–64.
67. Pap T, Muller-Ladner U, Gay RE, et al. Fibroblast biology. Role of synovial
fibroblasts in the pathogenesis of rheumatoid arthritis. Arthritis Res 2000;2:
361–7.
68. Day CP. From fat to inflammation. Gastroenterology 2006;130:207–10.
69. Doucet C, Jasmin C, Azzarone B. Unusual interleukin-4 and -13 signaling in
human normal and tumor lung fibroblasts. Oncogene 2000;19:5898–905.
70. Muhlbauer M, Weiss TS, Thasler WE, et al. LPS-mediated NFkappaB activation
varies between activated human hepatic stellate cells from different donors.
71. Xia Y, Pauza ME, Feng L, et al. RelB regulation of chemokine expression mod-
ulates local inflammation. Am J Pathol 1997;151:375–87.
72. Oakley F, Mann J, Nailard S, et al. Nuclear factor-kappaB1 (p50) limits the
inflammatory and fibrogenic responses to chronic injury. Am J Pathol 2005;
166:695–708.
73. Armulik A, Abramsson A, Betsholtz C. Endothelial/pericyte interactions. Circ Res
2005;97:512–23.
74. Nissen LJ, Cao R, Hedlund EM, et al. Angiogenic factors FGF2 and PDGF-BB
synergistically promote murine tumor neovascularization and metastasis.
J Clin Invest 2007;117:2766–77.
76. Pinzani M, Rombouts K. Liver fibrosis: from the bench to clinical targets. Dig
Liver Dis 2004;36:231–42.
77. Cassiman D, Libbrecht L, Desmet V, et al. Hepatic stellate cell/myofibroblast
subpopulations in fibrotic human and rat livers. J Hepatol 2002;36:200–9.
78. Rappaport AM, MacPhee PJ, Fisher MM, et al. The scarring of the liver acini
(cirrhosis). Tridimensional and microcirculatory considerations. Virchows Arch
A Pathol Anat Histopathol 1983;402:107–37.
79. Wanless IR, Nakashima E, Sherman M. Regression of human cirrhosis. Morpho-
logic features and the genesis of incomplete septal cirrhosis. Arch Pathol Lab
Med 2000;124:1599–607.
81. Pinzani M, Marra F, Carloni V. Signal transduction in hepatic stellate cells. Liver
1998;18:2–13.
82. Gonzalo T, Beljaars L, van de BM, et al. Local inhibition of liver fibrosis by
specific delivery of a platelet-derived growth factor kinase inhibitor to hepatic
stellate cells. J Pharmacol Exp Ther 2007;321:856–65.
83. Middleton J, Patterson AM, Gardner L, et al. Leukocyte extravasation: chemo-
kine transport and presentation by the endothelium. Blood 2002;100:3853–60.
84. Takasaki S, Hano H. Three-dimensional observations of the human hepatic
artery (arterial system in the liver). J Hepatol 2001;34:455–66.
85. Kudo S, Matsuno K, Ezaki T, et al. A novel migration pathway for rat dendritic
cells from the blood: hepatic sinusoids-lymph translocation. J Exp Med 1997;
185:777–84.
86. Crispe IN. Hepatic T cells and liver tolerance. Nat Rev Immunol 2003;3:51–62.
87. Adams DH, Eksteen B. Aberrant homing of mucosal T cells and extra-intestinal
manifestations of inflammatory bowel disease. Nat Rev Immunol 2006;6:244–51.
88. Elvevold KH, Nedredal GI, Revhaug A, et al. Scavenger properties of cultivated
pig liver endothelial cells. Comp Hepatol 2004;3:4.
89. Wong J, Johnston B, Lee SS, et al. A minimal role for selectins in the recruitment
of leukocytes into the inflamed liver microvasculature. J Clin Invest 1997;99:
2782–90.
90. Adams DH, Hubscher SG, Fisher NC, et al. Expression of E-selectin and
E-selectin ligands in human liver inflammation. Hepatology 1996;24:533–8.
91. McNab G, Reeves JL, Salmi M, et al. Vascular adhesion protein 1 mediates
binding of T cells to human hepatic endothelium. Gastroenterology 1996;110:
522–8.
92. Lalor PF, Edwards S, McNab G, et al. Vascular adhesion protein-1 mediates
adhesion and transmigration of lymphocytes on human hepatic endothelial cells.
J Immunol 2002;169:983–92.
93. Suniara RK, Jenkinson EJ, Owen JJ. An essential role for thymic mesenchyme in
early T cell development. J Exp Med 2000;191:1051–6.
94. Marra F, Romanelli RG, Giannini C, et al. Monocyte chemotactic protein-1 as
a chemoattractant for human hepatic stellate cells. Hepatology 1999;29:140–8.
95. Hellerbrand C, Wang SC, Tsukamoto H, et al. Expression of intracellular adhe-
sion molecule 1 by activated hepatic stellate cells. Hepatology 1996;24:670–6.
96. Lalor PF, Shields P, Grant A, et al. Recruitment of lymphocytes to the human liver.
Immunol Cell Biol 2002;80:52–64.
97. Curbishley SM, Eksteen B, Gladue RP, et al. CXCR3 activation promotes lym-
phocyte transendothelial migration across human hepatic endothelium under
fluid flow. Am J Pathol 2005;167:887–99.
880 Holt et al
98. Heydtmann M, Lalor PF, Eksteen JA, et al. CXC chemokine ligand 16 promotes
integrin-mediated adhesion of liver-infiltrating lymphocytes to cholangiocytes
and hepatocytes within the inflamed human liver. J Immunol 2005;174:1055–62.
99. Heydtmann M, Hardie D, Shields PL, et al. Detailed analysis of intrahepatic CD8
T cells in the normal and hepatitis C-infected liver reveals differences in specific
populations of memory cells with distinct homing phenotypes. J Immunol 2006;
177:729–38.
100. Efsen E, Grappone C, Defranco RM, et al. Up-regulated expression of fractal-
kine and its receptor CX3CR1 during liver injury in humans. J Hepatol. 2002;
37:39–47.
101. Isse K, Harada K, Zen Y, et al. Fractalkine and CX3CR1 are involved in the
recruitment of intraepithelial lymphocytes of intrahepatic bile ducts. Hepatology
2005;41:506–16.
102. Roth SJ, Carr MW, Springer TA. C-C chemokines, but not the C-X-C chemokines
interleukin-8 and interferon-gamma inducible protein-10, stimulate transendo-
thelial chemotaxis of T lymphocytes. Eur J Immunol 1995;25:3482–8.
103. Taub DD, Proost P, Murphy WJ, et al. Monocyte chemotactic protein-1 -2 and -3
are chemotactic for human T lymphocytes. J Clin Invest 1995;95:1370–6.
104. Czaja MJ, Geerts A, Xu J, et al. Monocyte chemoattractant protein-1 expression
occurs in toxic rat liver injury and human liver disease. J Leukoc Biol 1994;55:
120–6.
105. Holt A, et al, submitted.
106. Safadi R, Ohta M, Alvarez CE, et al. Immune stimulation of hepatic fibrogenesis
by CD8 cells and attenuation by transgenic interleukin-10 from hepatocytes.
107. Nelson DR, Lauwers GY, Lau JY, et al. Interleukin 10 treatment reduces fibrosis
in patients with chronic hepatitis C: a pilot trial of interferon nonresponders. Gas-
108. Tiggelman AM, Boers W, Linthorst C, et al. Interleukin-6 production by human
liver (myo)fibroblasts in culture. Evidence for a regulatory role of LPS, IL-1
beta and TNF alpha. J Hepatol 1995;23:295–306.
109. Adams DH, Hathaway M, Shaw J, et al. Transforming growth factor-beta induces
human T lymphocyte migration in vitro. J Immunol 1991;147:609–12.
110. Hathaway M, Burnett D, Elias E, et al. Secretion of soluble chemotactic factors,
including interleukin-6: a mechanism for the recruitment of CD8-positive T
lymphocytes to human liver allografts during rejection. Hepatology 1993;18:
511–8.
111. Delaney B, Koh WS, Yang KH, et al. Hepatocyte growth factor enhances B-cell
activity. Life Sci 1993;53:89–93.
112. Adams DH, Harvath L, Bottaro DP, et al. Hepatocyte growth factor and macro-
phage inflammatory protein 1 beta: structurally distinct cytokines that induce
rapid cytoskeletal changes and subset-preferential migration in T cells. Proc
Natl Acad Sci U S A 1994;91:7144–8.
113. Drayton DL, Liao S, Mounzer RH, et al. Lymphoid organ development: from
ontogeny to neogenesis. Nat Immunol 2006;7:344–53.
114. Hjelmstrom P, Fjell J, Nakagawa T, et al. Lymphoid tissue homing chemokines
are expressed in chronic inflammation. Am J Pathol 2000;156(4):1133–8.
115. Bonacchi A, Petrai I, Defranco RM, et al. The chemokine CCL21 modulates lym-
phocyte recruitment and fibrosis in chronic hepatitis C. Gastroenterology 2003;
125:1060–76.
116. Grant AJ, Goddard S, Ahmed-Choudhury J, et al. Hepatic expression of

secondary lymphoid chemokine (CCL21) promotes the development of por-
tal-associated lymphoid tissue in chronic inflammatory liver disease. Am
J Pathol 2002;160:1445–55.
117. von Andrian UH, Mempel TR. Homing and cellular traffic in lymph nodes. Nat
Rev Immunol 2003;3:867–78.
118. Bajenoff M, Egen JG, Koo LY, et al. Stromal cell networks regulate lymphocyte
entry, migration, and territoriality in lymph nodes. Immunity 2006;25:989–1001.
119. Mempel TR, Henrickson SE, von Andrian UH. T-cell priming by dendritic cells in
lymph nodes occurs in three distinct phases. Nature 2004;427:154–9.
120. Neubauer K, Eichhorst ST, Wilfling T, et al. Sinusoidal intercellular adhesion
molecule-1 up-regulation precedes the accumulation of leukocyte function
antigen-1-positive cells and tissue necrosis in a model of carbontetrachloride-
induced acute rat liver injury. Lab Invest 1998;78:185–94.
121. Kordes C, Sawitza I, Muller-Marbach A, et al. CD1331 hepatic stellate cells are
progenitor cells. Biochem Biophys Res Commun 2007;352:410–7.
122. McGettrick HM, Lord JM, Wang KQ, et al. Chemokine- and adhesion-dependent
survival of neutrophils after transmigration through cytokine-stimulated
endothelium. J Leukoc Biol 2006;79:779–88.
123. Ross EA, Douglas MR, Wong SH, et al. Interaction between integrin
alpha9beta1 and vascular cell adhesion molecule-1 (VCAM-1) inhibits neutro-
phil apoptosis. Blood 2006;107:1178–83.
124. Filer A, Parsonage G, Smith E, et al. Differential survival of leukocyte subsets
mediated by synovial, bone marrow, and skin fibroblasts: site-specific versus
activation-dependent survival of T cells and neutrophils. Arthritis Rheum 2006;
54:2096–108.
125. Tough DF, Borrow P, Sprent J. Induction of bystander T-cell proliferation by
viruses and type-I Interferon in-vivo. Science 1996;272:1947–50.
126. Abrignani S. Bystander activation by cytokines of intrahepatic T cells in chronic
viral hepatitis. Semin Liver Dis 1997;17:319–22.
127. Mitchell T, Kappler J, Marrack P. Bystander virus infection prolongs activated
T cell survival. J Immunol 1999;162:4527–35.
128. Inaba M, Kurasawa K, Mamura M, et al. Primed T cells are more resistant to Fas-
mediated activation-induced cell death than naive T cells. J Immunol 1999;163:
1315–20.
129. Iredale JP. Models of liver fibrosis: exploring the dynamic nature of inflammation
and repair in a solid organ. J Clin Invest 2007;117:539–48.
130. Holt AP, Stamataki Z, Adams DH. Attenuated liver fibrosis in the absence of
B cells. Hepatology 2006;43:868–71.
131. Novobrantseva TI, Majeau GR, Amatucci A, et al. Attenuated liver fibrosis in the
absence of B cells. J Clin Invest 2005;115:3072–82.
132. Chiaramonte MG, Donaldson DD, Cheever AW, et al. An IL-13 inhibitor blocks
the development of hepatic fibrosis during a T-helper type 2-dominated inflam-
matory response. J Clin Invest 1999;104:777–85.
133. Chiaramonte MG, Schopf LR, Neben TY, et al. IL-13 is a key regulatory cytokine
for Th2 cell-mediated pulmonary granuloma formation and IgE responses
induced by Schistosoma mansoni eggs. J Immunol 1999;162:920–30.
134. Hoffmann KF, McCarty TC, Segal DH, et al. Disease fingerprinting with cDNA
microarrays reveals distinct gene expression profiles in lethal type 1 and type
2 cytokine-mediated inflammatory reactions. FASEB J 2001;15:2545–7.
882 Holt et al
135. Tian J, Zekzer D, Hanssen L, et al. Lipopolysaccharide-activated B cells down-

regulate Th1 immunity and prevent autoimmune diabetes in nonobese diabetic
mice. J Immunol 2001;167:1081–9.
136. Parekh VV, Prasad DV, Banerjee PP, et al. B cells activated by lipopolysaccha-
ride, but not by anti-Ig and anti-CD40 antibody, induce anergy in CD81 T cells:
role of TGF-beta 1. J Immunol 2003;170:5897–911.
137. Mosmann T. Complexity or coherence? Cytokine secretion by B cells. Nat Immu-
nol 2000;1:465–6.
138. Harris DP, Haynes L, Sayles PC, et al. Reciprocal regulation of polarized cyto-
kine production by effector B and T cells. Nat Immunol 2000;1:475–82.
139. Harris DP, Goodrich S, Mohrs K, et al. Cutting edge: the development of IL-4-
producing B cells (B effector 2 cells) is controlled by IL-4, IL-4 receptor alpha,
and Th2 cells. J Immunol 2005;175:7103–7.
140. Fichtner-Feigl S, Strober W, Kawakami K, et al. IL-13 signaling through the
IL-13alpha(2) receptor is involved in induction of TGF-beta(1) production and fi-
brosis. Nat Med 2006;12:99–106.
141. Afford SC, Randhawa S, Eliopoulos AG, et al. CD40 activation induces apopto-
sis in cultured human hepatocytes via induction of cell surface fas ligand ex-
pression and amplifies fas-mediated hepatocyte death during allograft
rejection. J Exp Med 1999;189:441–6.
142. Ferru I, Roye O, Delacre M, et al. Infection of B-cell-deficient mice by the para-
site Schistosoma mansoni: demonstration of the participation of B cells in gran-
uloma modulation. Scand J Immunol 1998;48:233–40.
143. de LC, Galli G, Aldrighetti L, et al. Production of profibrotic cytokines by invariant
NKT cells characterizes cirrhosis progression in chronic viral hepatitis. J Immu-
nol 2004;173:1417–25.
144. Safadi R, Zigmond E, Pappo O, et al. Amelioration of hepatic fibrosis via beta-
glucosylceramide-mediated immune modulation is associated with altered CD8
and NKT lymphocyte distribution. Int Immunol 2007;19:1021–9.
145. Kim JH, Kim HY, Kim S, et al. Natural killer T (NKT) cells attenuate bleomycin-
induced pulmonary fibrosis by producing interferon-gamma. Am J Pathol
2005;167:1231–41.
146. Chuang YH, Lian ZX, Yang GX, et al. Natural killer T cells exacerbate liver injury
in a transforming growth factor beta receptor II dominant-negative mouse model
of primary biliary cirrhosis. Hepatology 2007;47:571–80.
147. Johnson LA, Clasper S, Holt AP, et al. An inflammation-induced mechanism for
leukocyte transmigration across lymphatic vessel endothelium. J Exp Med 2006;
203:2763–77.
Noninvasive
Ass essment of Liver
Fibrosis : Serum
Marker s, I maging,
and Other Modalities
IN. Guha, MD, PhDa,*, W.M. Rosenberg, MD, PhD, FRCPb
KEYWORDS
Liver fibrosis Diagnosis Serum markers Non-invasive
Imaging
INDICATIONS FOR ASSESSING LIVER FIBROSIS
It is important to have a safe and effective diagnostic tool for liver fibrosis for several
reasons. Patients who have liver fibrosis represent the group at greatest risk for devel-
oping liver-associated morbidity and mortality and are therefore in greatest need of
therapy for the underlying etiology (eg, antiviral treatment for chronic hepatitis C
[CHC], alcohol abstinence, and weight loss or possibly glitazones for nonalcoholic
fatty liver disease [NAFLD]). Secondly, prognostic information for patient and clinician
use depends on assessment of the disease severity and understanding factors asso-
ciated with progression. Thirdly, when the underlying insult can be removed, it may
soon be possible to offer patients specific antifibrotic therapy to halt or reverse liver
damage. For this promise to become reality once drugs are available for the treatment
of fibrosis, the potential target population will need to be identified, which will accel-
erate the need for accurate biomarkers to measure the efficacy of antifibrotic treat-
ment effects.
Finally, liver fibrosis is generally asymptomatic until the final stages of disease. Thus
the estimation of the prevalence of significant disease (ie, fibrosis or cirrhosis) requires
the use of diagnostic tests. Current reports of the prevalence of chronic liver disease
(CLD) probably underestimate the burden of disease because of underascertainment
on death certification, the social stigma of certain causes of liver disease, and reliance
on an invasive method of diagnosis. To plan better preventative and therapeutic inter-
ventions, the incidence and prevalence of liver disease of different severities (from
a
Liver Group, University of Southampton, Mail Point 805, Level C, Southampton, S016 6YD, UK
b
Institute of Hepatology, University College London, 69-75 Chenies Mews, London WC1E 6HX, UK
E-mail address: guhaneil@hotmail.com (I. Guha).
Clin Liver Dis 12 (2008) 883–900

884 Guha & Rosenberg
mild fibrosis to cirrhosis) will need to be established in the population. Additionally,

causal factors associated with fibrosis will need to be determined. These epidemio-
logic studies require simple, robust, noninvasive measures.
NATURAL HISTORY OF FIBROSIS
The liver biopsy has been useful in providing information on the natural history of the
progression of fibrosis. Yano and colleagues1 performed a longitudinal study on 70
patients who had CHC. The patients underwent between 2 to 10 liver biopsies over
1 to 26 years (mean, 9 years). Fibrosis at the initial biopsy (index biopsy) was divided
into three stages: A (none/mild), B (moderate), and C (severe). These three stages of
fibrosis had differing rates of progression and provided evidence that the evolution
of fibrosis to cirrhosis may not occur in a linear fashion. This concept gained further
support from a comprehensive meta-analysis of fibrosis progression in CHC by Thein
and colleagues.2 Using Markov maximum likelihood estimation, these investigators
found different stage-specific transition probabilities; the highest rate was between
F2 and F3.
Advantages of Biopsy
The biopsy provides information in three major areas. Firstly, it can elucidate cause; for
example marked accumulation of iron in the hepatic parenchyma suggests hemach-
romatosis, whereas biliary duct damage associated with granulomas raises the suspi-
cion of primary biliary cirrhosis. Secondly, it provides information on underlying
pathologic processes (eg, steatosis, necroinflammation and fibrosis) and provides
the opportunity to stage the severity of disease (eg, presence of cirrhosis). Thirdly, it
can also act as a prognostic indicator. For example, in the context of hepatitis C,
necroinflammation in the biopsy suggests future fibrosis.3 It is unrealistic to believe
that biopsy will ever be superfluous in the clinical management of liver disease in all
patients, because it provides a wealth of information.
DISADVANTAGES OF LIVER BIOPSY
Liver biopsy does have its limitations, and some have questioned whether it truly
represents a gold standard reference test. In large studies of patients undergoing
biopsy, pain was reported in 20% and severe complications in 0.57%.4 Sampling error
exists, which is not surprising considering the average biopsy specimen represents 1/
50,000 of the organ. Studies have shown discordance rates of up to 30% when right
and left lobes are sampled laparoscopically, even in homogenously distributed
disease.5 Biopsy interpretation is open to both intra- and interobserver error.6,7
The quantification of fibrosis using scoring systems presents several problems.
Many scoring systems were developed for assessing treatment response in the
context of CHC (eg, Metavir, Scheuer, Ishak, and Knodell) and have subsequently
been adapted for other diseases and for diagnostic studies. The scores are ordinal,
categorical variables, and only semiquantitative descriptors.
To quantify fibrosis more accurately, automated morphometry was investigated.8,9
Liver biopsies are scored for the quantity of fibrosis using digital imaging and comput-
erized image analysis packages. The technology is sensitive to the degree of contrast
obtained with staining material used, which may contribute to reading variation
between the same core of liver tissue. Moreover, although the quantity of fibrosis
detected with morphometry correlates to the stages of fibrosis, the nature of this
relationship is not linear. For example, the progression from no fibrosis (Ishak stage 0)
to bridging fibrosis (Ishak stage 4) to cirrhosis (Ishak stage 6) is associated with fibrosis
Assessment of Liver Fibrosis 885
measurements of 1.9%, 13.7%, and 27.8%, respectively.10 This variation suggests

that not only the quantity of matrix but also the distribution of fibrosis, architectural dis-
turbance, and vascular involvement are important for staging of fibrosis. Thus, in the
earlier example, the progression from Ishak stage 0 to 4 may only be associated with
an approximately 12% increase of fibrosis, but the connection of fibrotic tissue be-
tween the portal tract and central vein (with the associated angiogenesis and subse-
quent shunting of blood) determines function and prognosis.
As technology continues to develop, incorporation of topography and quantification
of fibrosis may increase the value of this automated technology. Sampling bias inher-
ent in the reluctance to perform a biopsy will remain an issue, and the approach will
carry all of the hazards associated with the biopsy. Thus, the systems used to score
biopsies have inherent constraints because they are based on ordinal categorical vari-
ables and may not truly reflect the dynamic progression and regression of fibrosis;
thus, the use of numeric scores can provide a misleading picture of fibrosis progres-
sion. Furthermore, because of the invasive nature of biopsy, ethical considerations
exist regarding the frequency of biopsy and intervals between biopsies.
NATURE OF NONINVASIVE TESTS
Noninvasive tests can be classified in several ways, including based on the modality of
the test (eg, serum blood tests versus imaging), the constituents of the test (eg, direct
markers versus indirect markers of fibrosis); or the derivation of the test (eg, a candi-
date versus a discovery approach). With the evolution of noninvasive tests, some of
theses boundaries are starting to merge, particularly with the use of combination or
serial noninvasive tests.
PRINCIPLES OF DERIVATION OF BIOMARKERS
Increasing knowledge of the pathophysiology of liver fibrosis has led directly to the
wide spectrum of serum biomarkers that currently exists. Biomarkers of the structural
elements of fibrogenesis and the key inflammatory mediators involved in the genesis
or degradation of scar tissue are often referred to as direct components. Indirect
markers can reflect the accompanying perturbations in hepatic function, such as
portal hypertension, that correlate with the evolution of fibrosis. This division may
be overly simplistic and some biomarkers may reflect both structural and functional
changes. For example, hyaluronic acid is intuitively considered a direct structural
component, but may also reflect the decreased extraction occurring because of the
capillarization of sinusoids, and thus may also be a surrogate functional component.
TESTS TO ASSESS DIAGNOSTIC DISCRIMINATION
Several diagnostic tests are used in the analysis of noninvasive markers. Sensitivity,
specificity, and positive and negative predictive values provide a measure of perfor-
mance at individual thresholds. The positive and negative likelihood ratios are argu-
ably better summary measures of individual thresholds, combining two values from
the classical 2 2 diagnostic table. A value of 10 is generally accepted as an excellent
positive likelihood ratio (1LR) and less than 0.1 as an excellent negative likelihood ratio
( LR).
An important issue is how to compare different noninvasive markers. The area under
the receiving operator curve (AUROC) is widely used to characterize diagnostic tests;
the closer the score is to 1, the better the test. Area under the curve (AUC) statistics
between different studies can be compared for statistical significance or to help
meta-analysis summary receiving operator curves (SROCs) that have been created.
Several caveats exist to the AUROC. The proximity of the AUC to 1 or 0.5 helps pro-
vide a summary of performance, but being certain of the significance of small changes
in AUC is difficult (eg, an increase in 0.80–0.83). Moreover, in reality, diagnostic tests
are used at certain optimal thresholds rather than in entirety.
The diagnostic odds ratio is the 1LR divided by the –LR and may thus allow similar
optimal thresholds from different studies to be either compared directly or pooled for
meta-analysis.11 The authors recently proposed a practical method for comparing
different tests using a clinical utility tool (Fig. 1)12,13 Thresholds are chosen using prede-
termined sensitivity and specificity rates and subjects are classified as correct,
incorrect, or indeterminate. The advantages of the clinical utility model is that it is trans-
parent, reflects how a test would be used in clinical practice, and allows accuracy to be
tailored to necessity of certainty. The major disadvantage is that a spectrum exists of
what individual clinicians regard as acceptable rates of incorrect or indeterminate clas-
sification. As the statistical methodology becomes increasingly sophisticated and com-
plex in this area, care must be taken not to forget that the gold standard test may have
a significant, and possibly unquantifiable, deviation from the truth.
VALIDATION
The common approach in diagnostic studies is to test several potential biomarkers in

a derivation cohort. The most promising markers, usually isolated by univariate
analysis, are then combined and the final algorithm created (usually by multivariate
analysis). The panel should then be tested in a separate population, the validation
cohort. If recruitment of the validation cohort is similar to that of the derivation cohort,
problems may arise with bias. To reduce this validation in a separate center, external
validation is often undertaken.
Spectrum bias is another common issue to all noninvasive tests. Requiring a liver
biopsy as the comparative test will naturally concentrate studies in secondary or ter-
tiary care settings. This limitation will result in similarities in prevalence of the severity
of disease and, more pertinently, a narrower range of alternative diagnoses than if the
test were performed in a community setting. Although this may not detract from using
the test in a similar clinical setting, it will limit its general applicability.
CANDIDATE BLOOD BIOMARKERS
Initial studies of noninvasive markers largely consisted of single components, but the
field has evolved into combining these single components into panel markers.14
Fig. 1. Clinical utility model.

Table 1 shows some examples of the single components and how many of the panels
have combined several components from the different biomarker groups to enhance
diagnostic accuracy.
The diagnostic performance of the panel markers depends on the stage of fibrosis
that is distinguished and disease origin. Table 2 below shows some examples of the
common panel markers. The panel markers have been most extensively studied in
hepatitis C. Although the diagnostic performance of the panel markers is broadly
similar, there seems to be a trend of superior performance in alcohol, followed by
CHC and NAFLD, and then chronic hepatitis B (CHB). The performance at extreme
thresholds in certain conditions is sufficient to allow a proportion of patients to be
classified accurately and potentially avoid liver biopsy (30% to 40% in CHC).13
In diagnosing cirrhosis, all panel tests have an improved performance, but the
added diagnostic benefit must be considered in the context of other diagnostic
Table 1
Examples of biomarker components and constituents of panel markers
Examples of Panel
Examples of Individual Biomarkers using at Least
Biomarker Groups Components One Component
Direct
Collagen and HA ELF
extracellular matrix MMP 1 Fibrospect
components MMP 8 HA score
PIII NP Hepascore
Laminin Leroy score
Hepatic stellate cell TIMP 1 ELF
and fibrogenic TGF b Fibrospect
cell mediators Angiotensin II
YKL 40
Indirect
Portal hypertension Platelet count Fibrometer
Spleen size Fibroindex
FIB 4
Pohl index
Testa index
Wai score
Synthetic parameters Albumin PGAA index
Platelet count Fibrometer
Liver enzymes AST APRI
and bilirubin ALT BAAT score
AST/ALT ratio Fibrometer
GGT Fibroindex
Bilirubin Fibrotest
Forns
Hepascore
HA score
NAFLD simple index
Pohl index
Wai score
Miscellaneous Cholesterol NAFLD simple index
Insulin resistance BAAT score
Forns
Table 2
Example of panel markers for the detection of significant liver fibrosis in different etiologies
AUC in AUC in
Training Validation
End Point Prevalence Cohort Cohort
Disease of Fibrosis of Significant (No. in (No. in Lower Upper
Study Measure Fibrosis Study) Study) Threshold Threshold
HCV
APRI53 F2/3/4 64 0.8 0.88 91 (Se) 41 (Se)
(192) (78) 47 (Sp) 95 (Sp)
1.7 (1LR) 8.2 (1LR)
0.2 (–LR) 0.6 (–LR)
ELF54 3/4 Scheuer 27 n/a 0.77 95 (Se) 38 (Se)
(261) 29 (Sp) 95 (Sp)
Fibrotest55 F2/3/4 40 0.84 0.87 97 (Se) 29 (Se)
(205) (134) 24 (Sp) 95 (Sp)
1.3 (1LR) 5.7 (1LR)
0.1 (–LR) 0.7 (–LR)
Fibrometer56 F2/3/4 55 0.88 0.91 n/a 81 (Se)
(383) (<120) 84 (Sp)
Forns57 F2/3/4 25 0.86 0.81 94 (Se) 44 (Se)
(351) (125) 45 (Sp) 96 (Sp)
1.7 (1LR) 11.6 (1LR)
0.1 (–LR) 0.6 (–LR)
Hepascore58 F2/3/4 44 0.85 0.82 n/a 67 (Se)
(117) (104) 92 (Sp)
Alcohol
APRI59 Septal 37 n/a 0.70 94 (Se) 9 (Se)
fibrosis (507) 26 (Sp) 97 (Sp)
Ishak 1.3 (1LR) 3.1 (1LR)
0.24 (–LR) 0.94 (–LR)
ELF54 3/4 scheuer 27 n/a 0.94 100 (Se) 93 (Se)
(64) 17 (Sp) 100 (Sp)
Fibrotest60 F2/3/4 64 n/a 0.84 84 (Se) 55 (Se)
(221) 66 (Sp) 93 (Sp)
2.5 (1LR) 7.4 (1LR)
0.25 (–LR) 0.5 (–LR)
Fibrometer56 F2/3/4 80 n/a 0.96 n/a 92 (Se)
(95) 93 (Sp)
13 (1LR)
0.09 (–LR)
NAFLD
ELF12 2/3/4 Kleiner 40 n/a 0.82 95 (Se) 45 (Se)
(192) 22 (Sp) 95 (Sp)
1.21 (1LR) 8.71 (1LR)
0.24 (–LR) 0.57 (–LR)
Fibrotest61 2/3/4 Kleiner 27 n/a 0.81 77 (Se) 15 (Se)
(267) 77 (Sp) 98 (Sp)
NAFLD fibrosis 3/4 Kleiner 26 0.88 0.82 82 (Se) 51 (Se)
score45 (480) (253) 77 (Sp) 98 (Sp)
3.57 (1LR) 26 (1LR)
0.23 (–LR) 0.49 (–LR)
(continued on next page)
Table 2
(continued)
AUC in AUC in
Training Validation
End Point Prevalence Cohort Cohort
Disease of Fibrosis of Significant (No. in (No. in Lower Upper
Study Measure Fibrosis Study) Study) Threshold Threshold
HBV
APRI62 2/3/4 68 n/a 0.72 71 (Se) 27 (Se)
(110) 87 (Sp) 96 (Sp)
5.45 (1LR) 6.3 (1LR)
0.36 (–LR) 0.76 (–LR)
Fibrotest63 2/3/4 29 n/a 0.78 89 (Se) 18 (Se)
(209) 52 (Sp) 99 (Sp)
1.85 (1LR) 26.7 (1LR)
Hui and colleagues64 3/4/5/6 27 0.80 0.76 93 (Se) 41 (Se)
Ishak (150) (85) 49 (Sp) 90 (Sp)
Abbreviations: ALT, alanine aminotransferase; APRI, aspartate aminotransferase-to-platelet ratio

index; AST, aspartate transaminase; ELF, Enhanced Liver Fibrosis Panel; GGT, Gamma glutamyl
transferase; HA, hyaluronic acid; HBV, hepatitis B virus; LR, likelihood ratio; MMP, Matrix metallo-
proteinase; NAFLD, nonalcoholic fatty liver disease; PPIII NP, aminoterminal type three procollagen
peptide; Se, sensitivity; Sp, specificity; TGF B, transforming growth factor beta; TIMP, tissue inhib-
itor of metalloproteinase
modalities, such as clinical signs and standard radiology. This fact is exemplified by an
algorithm devised by Kaul and colleagues15 that includes parameters such as platelet
count, spider nevi, and aspartate transaminase (AST); the AUC for cirrhosis was 0.93.
DISCOVERY
In contrast to the candidate marker approach, discovery-based technologies such as

genomics, proteomics, metabonomics, and lipidomics use a hypothesis generating
approach to diagnostics. These techniques may not only reveal novel biomarkers
but may also show associations that may not have been believed influential. These
technologies will not necessarily show causal relationships, and therefore further stud-
ies are often required to illustrate mechanistic processes. The large data sets
produced, because of the extremely large number of variables generated, necessitate
the use of specialized statistical methods.
Using genomics investigators have found several single nucleotide polymorphisms
(SNPs) that predict the progression to advanced liver fibrosis in CHC;16 for example an
SNP in the dead box polypeptide 5 gene was significantly associated with severe
fibrosis, and an SNP in carnitine palmitoyltransferase 1A was associated with
a decreased risk for severe fibrosis. A proteomic index for detecting severe fibrosis
has been shown to be superior to Fibrotest (AUC of 0.88 versus 0.81); the index
consisted of 8 peaks out of 1000 tested with identification awaited.17 Poon and
colleagues18 found that a proteomic algorithm in CHB, using a fingerprint composed
of 30 proteomic features, had an AUC of 0.91 for detecting significant fibrosis in 46
patients. Glycomics has been used to assess liver fibrosis through modifying DNA
sequencer/fragment analyzers to measure serum protein N-glycans.19 Using this
technology, an AUC of 0.87 was achieved in separating fibrosis from compensated
cirrhosis.
The technologies show much promise; however, their acceptance has been limited
by the practicality and cost because of the need for specialized equipment, sample
processing, and labor-intensive analysis, and because of a high false-positive discov-
ery rate. These constraints currently limit the true potential of these approaches and
enhance their appeal, primarily in providing insights into disease mechanisms rather
than as first-line diagnostics.
IMAGING
The role of conventional ultrasound, CT, and MRI has largely been to detect cirrhosis,
particularly portal hypertension. Although the radiologic features of splenomegaly,
reversal of hepatic blood flow, change in caudate to right lobe ratio, and hepatic
vein narrowing aid the sensitivity of detecting severe disease, they are less useful in
earlier disease. However, modification of conventional imaging and development of
novel techniques are challenging this concept.
CONTRAST ULTRASONOGRAPHY
Microbubble ultrasound contrast agents have been used to assess disease severity in
a number of conditions. The microbubbles are injected peripherally and the hepatic
vein transit time (HVTT) calculated, which represents the difference in arrival times
between the hepatic vein and artery. The underlying principle is that HVTT should
decrease with progressive fibrosis, reflecting increasing intrahepatic shunting and
capillarization of the sinusoids. Lim and colleagues20 showed separation of HVTT
times in three groups with hepatitis C virus (HCV), based on using necroinflammation
or fibrosis to evaluate mild hepatitis, chronic hepatitis, and cirrhosis. The separation of
cirrhosis from mild disease was excellent and larger validation studies are awaited to
confirm the findings of this promising pilot study.
Contrast ultrasonography can be used to discern lesions within hepatic paren-
chyma, because metastases and hepatocellular carcinoma show a reduced uptake
of agents such as Levovist (Shering, Berlin, Germany) compared with the surrounding
normal tissue.21,22 Interest has been shown in determining whether this modality can
be used to differentiate disease severity. A recent study (N 5 66) found that delayed
parenchymal enhancement of Levovist was helpful in distinguishing NAFLD from non-
alcoholic steatohepatitis, but contrast ultrasound did not correlate with the histologic
features of fibrosis.23,24
ULTRASOUND ELASTOGRAPHY (TRANSIENT ELASTOGRAPHY)
Transient elastography is a technique whereby a shear wave, at a low frequency of

50 Hz, is created by a vibrating probe and transducer applied to the skin overlying
the liver. The velocity of the propagated wave is correlated with the stiffness or elas-
ticity of the underlying liver; simplistically, the propagated wave travels faster with
increasing fibrosis. A pulse–echo ultrasound allows measurement of the wave velocity
and the results are presented as kilopascals (kPa). The validity of measurement is
assessed by the interquartile range and ratio of successful measurements to unsuc-
cessful measurements (should be >60%).
Transient elastography technique has several advantages. Stiffness is measured
within a cylinder, measuring 1 cm in width and 4 cm in length, producing an estimated
sampling area that is 100 times greater than biopsy. It is a simple technique to learn,
with the reproducibility from a large study, 800 examinations in 200 patients who had
heterogeneous liver disease, suggesting an intraclass correlation coefficient (ICC) of
0.98 by two operators.23 The ICC in this study was lower in less-severe fibrosis,
increased body mass index (BMI), and increased steatosis. Finally, the test is inexpen-
sive; the equipment has a capital cost, but the running cost thereafter is low compared
with other noninvasive tests.
Table 3 highlights the performance of transient elastography in detecting significant
fibrosis in several causes. The threshold for detecting significant fibrosis varies from
4 to 9 kPa in the selected studies, and variation occurs between and within causes.
Good diagnostic performance occurs above these critical thresholds. Similar to blood
tests, transient elastography shows better performance in detecting cirrhosis.
Several potential issues with transient elastography are starting to be addressed
with emerging research. The effect of other variables on liver stiffness measurements
(LSM), other than fibrosis, may be an important consideration. Three studies have
shown that acute hepatitis elevates LSM independently of fibrosis.25–27 As exemplified
in the study by Arena and colleagues,25 the peak LSM corresponded to values of
cirrhosis in noncirrhotic patients at the peak of necroinflammation, before returning
to the normal range in parallel with the aminotransferases. This finding may be more
relevant in causes such as CHB, in which flares occur more commonly, but nonethe-
less supports the argument that LSM must be interpreted carefully.28 Hepatic
Table 3
Performance of transient elastography in detecting significant fibrosis in different etiologies of liver
disease
Prevalence of
Significant AUC Threshold Above
Study Disease Fibrosis (No. in Study) (kPa) Threshold
Fraquelli and colleagues.23 Mixed 50 0.86 7.6 81 (Se)
(200) 76 (Sp)
3.37 (1LR)
0.25 (–LR)
Gomez-Dominguez Mixed 82 0.74 4.0 94 (Se)
and colleagues.65 (94) 33 (Sp)
1.22 (1LR)
0.18 (–LR)
Chang and colleagues.66 Mixed 44 0.86 9.0 83 (Se)
(120) 85 (Sp)
5.58 (1LR)
0.20 (–LR)
Castera and colleagues.46 HCV 74 0.83 7.1 67 (Se)
(183) 89 (Sp)
6.09 (1LR)
0.37 (–LR)
Ziol and colleagues.67 HCV 65 0.79 8.8 56 (Se)
(251) 91 (Sp)
6.63 (1LR)
0.48 (–LR)
Yoneda and colleagues.68 NAFLD 49 0.87 6.6 83 (Se)
(67) 81 (Sp)
4.4 (1LR)
0.21 (–LR)
Abbreviations: HCV, hepatitis C virus; LR, likelihood ratio; NAFLD, nonalcoholic fatty liver disease;
Se, sensitivity; Sp, specificity.
steatosis may also have an independent effect on LSM; however, large studies, spe-
cifically including patients who have massive steatosis, are required to address this
issue.
Some technical issues may also arise when obtaining scans in very thin (eg, need to
insert the probe in the intercostal space) or obese individuals (eg, attenuation of signal
by surface adiposity). A BMI of greater than 28 was found to be an independent factor
for failure to obtain LSM in a multivariate analysis.29 Modifying the transducer design
may address these issues.
CT
CT use for measuring liver fibrosis has been limited by the inability to detect fibrosis
(rather than consequences of cirrhosis) and concerns about ionizing radiation expo-
sure if this modality were adopted for longitudinal assessment. The advent of multi-
slice and helical CT scanning have reduced acquisition times. Furthermore, the new
Fibro-CT tool was recently developed, which was shown to have an AUC of 0.83
and 0.93 for detecting moderate and severe fibrosis, respectively, in patients who
have CHC.30 This technique does not require contrast and uses optical digital analysis
of conventional images. The effect of steatosis, necroinflammation, and reproducibility
are potential caveats to this technique and further studies are awaited.
MRI
MRI techniques can be divided into several categories: conventional MRI, diffusion-
weighted MRI, contrast enhanced MRI, MRI spectroscopy, and MRI elastography.
The obvious advantage of all techniques is the lack of ionizing radiation. Diffusion-
weighted MRI is based on the concept of measuring free water protons; the water
content has been postulated to decrease with increasing extracellular matrix. Prelim-
inary studies have suggested that the technique is useful in distinguishing cirrho-
sis.31,32 The effects of steatosis, necroinflammation, and iron on this modality must
be further elucidated. MRI studies have largely used the contrast agents, gadolinium
or superparamagnetic iron oxide (SPIO). In one of the larger contrast-enhanced MRI
studies, Aguirre and colleagues33 used a combination of gadolinium and SPIO and
found accuracy rates greater than 90%.
The ability to extract additional information, such as metabolism (MR spectroscopy)
and elasticity (MR elastography), further advances the potential of this modality be-
cause it allows in vivo measurement of structure and function.
MR Spectroscopy
MR spectroscopy studies of the human liver have been based on the ubiquitous
protons hydrogen (1H) and phosphorus (31P). The magnetic properties of these
protons are influenced by their surrounding chemical environment, and thus individual
metabolites will have signature spectroscopic signals. The challenges of the technique
include obtaining adequate resolution of these signals and maximizing the signal-to-
noise ratio. Quantification of metabolites detected in vivo is difficult, because it
requires the analysis of an internal or external reference standard and lengthens acqui-
sition times, and the data is often presented as metabolite ratios. The major publica-
tions in this area have used 31P MR spectroscopy.
The dominant metabolic signals reported include phosphomonesters, phospho-
diesters, inorganic phosphate, and the nucleoside triphosphates. Several studies
have found that the phosphomonoesters-to-phosphodiesters ratio increased in
severe fibrosis,34,35 and this may reflect changing cell turnover and remodeling that
occurs in fibrogenesis.
MR elastography
MR elastography is based on similar principles to those of ultrasound elastography.
A shear wave is created by a driver (pneumatic or electromechanical) attached to
the abdominal wall. A specialized MR sequence is then used to measure the propa-
gated waves and analysis is performed to quantify these sequences into elastograms.
As the entire liver is sequenced, the area of sampling is greatly increased and the
heterogeneous distribution of fibrosis is more commonly appreciated. The number
of published studies is relatively small (<5 at the time of this writing), but the data
look encouraging. A recent study comparing MR elastography with aspartate amino-
transferase-to-platelet ratio index (APRI) in a cohort with viral and alcoholic liver
disease found that the ROC curves were significantly greater for distinguishing mod-
erate and severe fibrosis using elastography compared with the biochemical test.36
Issues common to all MR techniques include the increased acquisition time of scan-
ning (examinations can take up to 60 minutes), the capital investment required for the
equipment, the expertise in analysis, reproducibility, and standardized thresholds of
measurement. Although these techniques may not be the most pragmatic first-line
noninvasive assessment tools for liver fibrosis, they hold much promise, particularly
if larger validation studies confirm their diagnostic accuracy.
SYSTEMATIC REVIEWS OF NONINVASIVE MARKERS
The first systematic review in the field of noninvasive markers was by Gebo and
colleagues14 in 2002. This review concentrated on blood tests in CHC, and most stud-
ies were of single markers. The conclusions were that noninvasive markers still lacked
sufficient diagnostic power to obviate the need for liver biopsy, but panel markers
showed promise. This conclusion formed the foundation for the next major systematic
review in hepatitis C by Parkes and colleagues in 2006.13 This review was specifically
for panel markers in hepatitis C. Ten separate panel markers were identified and
results analyzed using likelihood ratios, diagnostic odds ratios, and SROC statistics.
Additionally, the QUADAS tool was used to measure quality of the included studies.
Meta-analysis was limited in this review because of heterogeneity of the studies,
but the summary diagnostic odds ratio was 9.96 for detecting significant fibrosis,
and using the clinical utility model the panels were found to ‘‘rule-in’’ or ‘‘rule-out’’
fibrosis in 35% of the population.
The group led by Myers37–39 subsequently performed two further systematic
reviews in CHC, examining (1) APRI (2) Fibrotest, and (3) Fibroscan (transient elastog-
raphy); and a further systematic review in (4) HIV/HCV coinfection. These reviews
included meta-analysis, including summary sensitivities and specificities and meta-
regression analysis for heterogeneity. In detecting significant fibrosis, the following
values were obtained (1) summary diagnostic odds ratio (SDOR) of 5.7; SROC of
0.76 (2) SDOR of 7.6, SROC of 0.81 (3) SDOR of 10.2, SROC of 0.82, and (4) SDOR
of 7.9, SROC of 0.82, respectively. Studies for heterogeneity were significant, and in
one review meta-regression identified significant fibrosis as a significant factor.
A meta-analysis of Fibrotest by Poynard and colleagues40 showed a mean AUC of
0.8 across different disease etiologies.
In NAFLD, a systematic review by Guha and colleagues41 suggested that the evo-
lution of noninvasive markers lagged behind CHC with fewer panel markers. Variables
found to be significant in the detection of fibrosis included the presence of diabetes,
age, homeostasis model assessment of insulin resistance, ratio of AST to alanine
transaminase, decreased platelet count, presence of hyaluronic acid, and increased

BMI.
A recent meta-analysis using only transient elastography42 found seven studies that
distinguished moderate fibrosis, with pooled estimates for sensitivity and specificity of
70%, and 84%, respectively. Analysis for heterogeneity was also significant.
Common themes emerging from the systematic reviews are the heterogeneity that
exists in the conduct, reporting, and analysis of diagnostic studies. Factors such as
prevalence of disease have a major impact on diagnostic tests, yet vary widely
between diagnostic studies. Regarding reporting and publication, guidelines such
as standards for reporting diagnostic accuracy (STARD) exist but are not uniformly
implemented, in direct comparison with the Consolidated Standards of Reporting Tri-
als (CONSORT) guidelines used in randomized controlled trials. This variation has
made it difficult to extract meaningful information, such as diagnostic statistics to cre-
ate 2 2 tables in many previously published diagnostic studies. The reviews also
highlight the performance variation of noninvasive tests at the extreme thresholds
compared with mid-thresholds.
EVOLUTION OF NONINVASIVE MARKERS

Combining Noninvasive Tests
Several studies have now explored the serial use of panel markers for detection.
Sebastini and colleagues43 looked at APRI, Forns index, and Fibrotest in a group of
patients who had chronic hepatitis (N 5 290). The initial use of APRI followed by Fibrot-
est was the optimal combination for detecting significant fibrosis. An accuracy of 94%
was reported, avoiding approximately 50% of liver biopsies. Leroy and colleagues44
recently looked at six panel markers: APRI, MP3, Forns, Fibrotest, Hepascore, and
Fibrometer. The panels had very similar performances for detecting significant fibrosis
and cirrhosis, but combinations of MP3, APRI, and Fibrotest (any two out of three)
improved accuracy and avoided approximately one third of biopsies for detecting
significant fibrosis.
In NAFLD, a simple panel of markers identified by Angulo and colleagues45 was
recently combined with the original European Liver Fibrosis panel for detecting early,
moderate, and severe fibrosis. This combined panel had an AUC of 0.93 for detecting
significant fibrosis and could potentially avoid biopsy in more than 80% of patients.12
The combination of different modalities of noninvasive markers is a particularly
attractive notion, because it may measure different aspects of structure and function
and hence reduce ‘‘blind spots.’’ Castera and colleagues46 prospectively examined
the performance of Fibroscan (transient elastography), Fibrotest, and APRI. The com-
bination of Fibrotest and Fibroscan resulted in an AUC of 0.88 for detecting significant
fibrosis and the possibility of avoiding approximately 80% of liver biopsies. Further
studies combining different blood tests and imaging are awaited.
ALTERNATIVE OUTCOME MEASURES

Clinical End Points
Severity of fibrosis detected with liver biopsy has been shown to predict progression
to cirrhosis. In a long-term cohort of approximately 3000 patients who had CHC,
Neal47 and Irving recently showed that liver fibrosis was an independent predictor
of liver-related mortality. Therefore, because histology is a surrogate of clinical
outcome, there is a logic to measuring the performance of noninvasive markers
against histologic fibrosis directly. An issue with planning and performing these direct
diagnostic studies is that the complications of liver disease take time to develop, and
therefore studies must be conducted over time. Nonetheless, efforts are starting to
emerge. Ngo and colleagues48 examined the Fibrotest at baseline in predicting clinical
outcomes at 5 years, showing that the AUC of Fibrotest was 0.96 for predicting death
(AUC of biopsy 0.87) and 0.96 for the complications of disease (AUC of biopsy 0.91).
Hepatic Venous Pressure Gradient

Hepatic venous pressure gradient (HVPG) is an important indicator of portal hyperten-
sion and has been shown to predict outcome measures, such as acute variceal hem-
orrhage49 and recurrence of hepatitis C after orthotopic liver transplantation.50 Carrion
and colleagues51 published data showing a close correlation between liver stiffness on
transient elastography and HVPG. An excellent study by Vizzutti and colleagues52
added a note of caution to this relationship; in their study, HVPG measurements
greater than 10 mm Hg did not correlate well with stiffness. This result is not surprising
because many other components determine portal hypertension aside from obvious
structural disturbance.
FUTURE PERSPECTIVES
Liver fibrosis is a result of chronic injury to the liver, and the progression to cirrhosis is
associated with changes in several key functions of the liver; Fig. 2 represents a sim-
plified summary. Biomarkers that assess liver fibrosis could be derived from several
broad areas, as represented by the boxes in red. Host factors that interact with the
insult will determine if and how quickly fibrosis will progress (eg, age, sex, genetic
Host factors:
INSULT
age, gender,
AST, ALT
genomics,
obesity
STEATOSIS INFLAMMATION
Structure:
Serum markers
CT
U/S
MRI
CIRRHOSIS FIBROSIS
Reduced synthesis Reduced extraction Haemodynamic Metabolic changes

e.g. clotting and elimination changes e.g. protein,
factors, albumin e.g. toxins, drugs, e.g. ascites, HRS, carbohydrate and
ammonia varices lipid metabolism
PT time, Aminopyrine Microbubble Proteomics,

albumin breath test, U/S Lipidomics
sorbitol Transient Metabonomics
clearence elastography
Fig. 2. Simplified summary of the causes and consequences of liver fibrosis and potential
areas of biomarker development.
polymorphisms). Other potential sources of biomarkers are measures of pathologic

processes preceding fibrosis (eg, ALT as a measure of necroinflammation). Directly
measuring structural alterations associated with fibrosis (eg, collagen turnover prod-
ucts measured through serum markers, imaging techniques such as ultrasound and
CT) is analogous to histologic assessment. Finally, biomarkers measuring the
functional consequence of fibrosis (eg, reduced clotting, albumin) will be of clinical
relevance for planning intervention such as liver transplantation. Combining
biomarkers that measure different components may advance diagnostic accuracy.
Noninvasive tests have concentrated on cross-sectional diagnosis, but this
technique may be ultimately flawed by the ‘‘glass ceiling’’ effect of an imperfect
gold standard. However, the ultimate goal of any noninvasive goal should be the ability
to detect change early and better inform prognosis to optimize management deci-
sions. Therefore, longitudinal measurement of noninvasive tests, and especially
changes within individual patients, may be more informative. How this is best
measured is a matter for debate, because serial histology is unlikely to be a useful
comparator. Ultimately, the use of well-designed studies assessing the changes in
noninvasive tests against clinical end points is likely to provide the strongest evidence
for prognostic efficacy.
SUMMARY
The diagnosis of liver fibrosis using noninvasive tools is important for epidemiologic,
prognostic, therapeutic, pragmatic, and economical reasons. Noninvasive tests for
liver fibrosis are broadly divided into blood tests, imaging, and novel technologies.
Noninvasive tests have concentrated on defining fibrosis at a fixed point, and the abil-
ity to define severe disease is excellent. The serial and combination use of different
noninvasive tests has continued to improve performance and increased the ability
to classify more patients and maintain accuracy. The use of clinical end points and
longitudinal measurement are promising areas of development and will probably
show the true potential of noninvasive tests.
REFERENCES
1. Yano M, Kumada H, Kage M, et al. The long-term pathological evolution of

chronic hepatitis C. Hepatology 1996;23(6):1334–40.
2. Thein HH, Yi Q, Dore GJ, et al. Estimation of stage-specific fibrosis progression
rates in chronic hepatitis C virus infection: a meta-analysis and meta-regression.
Hepatology 2008;48(2):418–31.
3. Ryder SD, Irving WL, Jones DA, et al. Progression of hepatic fibrosis in patients
with hepatitis C: a prospective repeat liver biopsy study. Gut 2004;53(3):451–5.
4. Cadranel JF, Rufat P, Degos F. Practices of liver biopsy in France: results of a pro-
spective nationwide survey. For the Group of Epidemiology of the French Associ-
ation for the study of the liver (AFEF). Hepatology 2000;32(3):477–81.
5. Regev A, Berho M, Jeffers LJ, et al. Sampling error and intraobserver variation in
liver biopsy in patients with chronic HCV infection. Am J Gastroenterol 2002;
97(10):2614–8.
6. Intraobserver and interobserver variations in liver biopsy interpretation in patients
with chronic hepatitis C. The French METAVIR Cooperative Study Group. Hepatol-
ogy 1994;20(1 Pt 1):15–20.
7. Bedossa P, Dargere D, Paradis V. Sampling variability of liver fibrosis in chronic
hepatitis C. Hepatology 2003;38(6):1449–57.
8. O’Brien MJ, Keating NM, Elderiny S, et al. An assessment of digital image

analysis to measure fibrosis in liver biopsy specimens of patients with chronic
hepatitis C. Am J Clin Pathol 2000;114(5):712–8.
9. Wright M, Thursz M, Pullen R, et al. Quantitative versus morphological assess-
ment of liver fibrosis: semi-quantitative scores are more robust than digital image
fibrosis area estimation. Liver Int 2003;23(1):28–34.
10. Standish RA, Cholongitas E, Dhillon A, et al. An appraisal of the histopathological
assessment of liver fibrosis. Gut 2006;55(4):569–78.
11. Deeks J. Evaluations of diagnostic and screening tests. In: Egger M, Smith GD,
AItman DG, editors. Systematic reviews in health care. UK: BMJ; 2006.
12. Guha IN, Parkes J, Roderick P, et al. Noninvasive markers of fibrosis in nonalco-
holic fatty liver disease: validating the European Liver Fibrosis Panel and explor-
ing simple markers. Hepatology 2008;47(2):455–60.
13. Parkes J, Guha IN, Roderick P, et al. Performance of serum marker panels for liver
fibrosis in chronic hepatitis C. J Hepatol 2006;44(3):462–74.
14. Gebo KA, Herlong HF, Torbenson MS, et al. Role of liver biopsy in management of
chronic hepatitis C: a systematic review. Hepatology 2002;36(5 Suppl 1):
S161–72.
15. Kaul V, Friedenberg FK, Braitman LE, et al. Development and validation of
a model to diagnose cirrhosis in patients with hepatitis C. Am J Gastroenterol
2002;97(10):2623–8.
16. Huang H, Shiffman ML, Cheung RC, et al. Identification of two gene variants
associated with risk of advanced fibrosis in patients with chronic hepatitis C.
17. Morra R, Munteanu M, Bedossa P, et al. Diagnostic value of serum protein profil-
ing by SELDI-TOF ProteinChip compared with a biochemical marker, FibroTest,
for the diagnosis of advanced fibrosis in patients with chronic hepatitis C. Aliment
Pharmacol Ther 2007;26(6):847–58.
18. Poon TC, Hui AY, Chan HL, et al. Prediction of liver fibrosis and cirrhosis in
chronic hepatitis B infection by serum proteomic fingerprinting: a pilot study.
Clin Chem 2005;51(2):328–35.
19. Callewaert N, Van Vlierberghe H, Van Hecke A, et al. Noninvasive diagnosis of
liver cirrhosis using DNA sequencer-based total serum protein glycomics. Nat
Med 2004;10(4):429–34.
20. Lim AK, Taylor-Robinson SD, Patel N, et al. Hepatic vein transit times using
a microbubble agent can predict disease severity non-invasively in patients
with hepatitis C. Gut 2005;54(1):128–33.
21. Blomley M, Albrecht T, Cosgrove D, et al. Stimulated acoustic emission in liver
parenchyma with Levovist. Lancet 1998;351(9102):568.
22. Blomley MJ, Albrecht T, Cosgrove DO, et al. Improved imaging of liver metasta-
ses with stimulated acoustic emission in the late phase of enhancement with
the US contrast agent SH U 508A: early experience. Radiology 1999;210(2):
409–16.
23. Fraquelli M, Rigamonti C, Casazza G, et al. Reproducibility of transient elastog-
raphy in the evaluation of liver fibrosis in patients with chronic liver disease.
Gut 2007;56(7):968–73.
24. Iijima H, Moriyasu F, Tsuchiya K, et al. Decrease in accumulation of ultrasound
contrast microbubbles in non-alcoholic steatohepatitis. Hepatol Res 2007;37(9):
722–30.
25. Arena U, Vizzutti F, Corti G, et al. Acute viral hepatitis increases liver stiffness
values measured by transient elastography30. Hepatology 2008;47(2):380–4.
26. Bolkenius U, Hahn D, Gressner AM, et al. Glucocorticoids decrease the bioavail-
ability of TGF-beta which leads to a reduced TGF-beta signaling in hepatic
stellate cells. Biochem Biophys Res Commun 2004;325(4):1264–70.
27. Coco B, Oliveri F, Maina AM, et al. Transient elastography: a new surrogate
marker of liver fibrosis influenced by major changes of transaminases. J Viral
Hepat 2007;14(5):360–9.
28. Cobbold JF, Taylor-Robinson SD. Transient elastography in acute hepatitis: all
that’s stiff is not fibrosis. Hepatology 2008;47(2):370–2.
29. Foucher J, Castera L, Bernard PH, et al. Prevalence and factors associated with
failure of liver stiffness measurement using FibroScan in a prospective study of
2114 examinations5. Eur J Gastroenterol Hepatol 2006;18(4):411–2.
30. Romero-Gomez M, Gomez-Gonzalez E, Madrazo A, et al. Optical analysis of
computed tomography images of the liver predicts fibrosis stage and distribution
in chronic hepatitis C. Hepatology 2008;47(3):810–6.
31. Boulanger Y, Amara M, Lepanto L, et al. Diffusion-weighted MR imaging of the
liver of hepatitis C patients. NMR Biomed 2003;16(3):132–6.
32. Girometti R, Furlan A, Bazzocchi M, et al. Diffusion-weighted MRI in evaluating
liver fibrosis: a feasibility study in cirrhotic patients. Radiol Med (Torino) 2007;
112(3):394–408.
33. Aguirre DA, Behling CA, Alpert E, et al. Liver fibrosis: noninvasive diagnosis with
double contrast material-enhanced MR imaging. Radiology 2006;239(2):425–37.
34. Lim AK, Patel N, Hamilton G, et al. The relationship of in vivo 31P MR spectros-
copy to histology in chronic hepatitis C. Hepatology 2003;37(4):788–94.
35. Menon DK, Sargentoni J, Taylor-Robinson SD, et al. Effect of functional grade and
etiology on in vivo hepatic phosphorus-31 magnetic resonance spectroscopy in
cirrhosis: biochemical basis of spectral appearances. Hepatology 1995;21(2):
417–27.
36. Huwart L, Sempoux C, Salameh N, et al. Liver fibrosis: noninvasive assessment
with MR elastography versus aspartate aminotransferase-to-platelet ratio index.
Radiology 2007;245(2):458–66.
37. Shaheen AA, Myers RP. Diagnostic accuracy of the aspartate aminotransferase-
to-platelet ratio index for the prediction of hepatitis C-related fibrosis: a systematic
review. Hepatology 2007;46(3):912–21.
38. Shaheen AA, Myers RP. Systematic review and meta-analysis of the diagnostic
accuracy of fibrosis marker panels in patients with HIV/hepatitis C coinfection.
HIV Clin Trials 2008;9(1):43–51.
39. Shaheen AA, Wan AF, Myers RP. FibroTest and FibroScan for the prediction
of hepatitis C-related fibrosis: a systematic review of diagnostic test accuracy.
Am J Gastroenterol 2007;102(11):2589–600.
40. Poynard T, Imbert-Bismut F, Munteanu M, et al. Overview of the diagnostic value
of biochemical markers of liver fibrosis (FibroTest, HCV FibroSure) and necrosis
(ActiTest) in patients with chronic hepatitis C. Comp Hepatol 2004;3(1):8.
41. Guha IN, Parkes J, Roderick PR, et al. Non-invasive markers associated with liver
fibrosis in non-alcoholic fatty liver disease. Gut 2006;55(11):1650–60.
42. Talwalkar JA, Kurtz DM, Schoenleber SJ, et al. Ultrasound-based transient elas-
tography for the detection of hepatic fibrosis: systematic review and meta-analy-
sis. Clin Gastroenterol Hepatol 2007;5(10):1214–20.
43. Sebastiani G, Vario A, Guido M, et al. Stepwise combination algorithms of non-
invasive markers to diagnose significant fibrosis in chronic hepatitis C. J Hepatol
2006;44(4):686–93.
44. Leroy V, Hilleret MN, Sturm N, et al. Prospective comparison of six non-invasive
scores for the diagnosis of liver fibrosis in chronic hepatitis C. J Hepatol 2007;
46(5):775–82.
45. Angulo P, Hui JM, Marchesini G, et al. The NAFLD fibrosis score: a noninvasive
system that identifies liver fibrosis in patients with NAFLD. Hepatology 2007;
45(4):846–54.
46. Castera L, Vergniol J, Foucher J, et al. Prospective comparison of transient elas-
tography, Fibrotest, APRI, and liver biopsy for the assessment of fibrosis in
chronic hepatitis C. Gastroenterology 2005;128(2):343–50.
47. Neal K, Irving WL. Excess mortality rates in a cohort of patients infected with the
hepatitis C virus: a prospective study. Gut 2007;56(8):1098–104.
48. Ngo Y, Munteanu M, Messous D, et al. A prospective analysis of the prognostic
value of biomarkers (FibroTest) in patients with chronic hepatitis C. Clin Chem
2006;52(10):1887–96.
49. Ripoll C, Groszmann R, Garcia-Tsao G, et al. Hepatic venous pressure gradient
predicts clinical decompensation in patients with compensated cirrhosis. Gastro-
enterology 2007;133(2):481–8.
50. Blasco A, Forns X, Carrion JA, et al. Hepatic venous pressure gradient identifies
patients at risk of severe hepatitis C recurrence after liver transplantation. Hepa-
tology 2006;43(3):492–9.
51. Carrion JA, Navasa M, Bosch J, et al. Transient elastography for diagnosis of
advanced fibrosis and portal hypertension in patients with hepatitis C recurrence
after liver transplantation. Liver Transpl 2006;12(12):1791–8.
52. Vizzutti F, Arena U, Romanelli RG, et al. Liver stiffness measurement predicts
severe portal hypertension in patients with HCV-related cirrhosis. Hepatology
2007;45(5):1290–7.
53. Wai CT, Greenson JK, Fontana RJ, et al. A simple noninvasive index can predict
both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepa-
tology 2003;38(2):518–26.
54. Rosenberg WM, Voelker M, Thiel R, et al. Serum markers detect the presence of
liver fibrosis: a cohort study. Gastroenterology 2004;127(6):1704–13.
55. Imbert-Bismut F, Ratziu V, Pieroni L, et al. Biochemical markers of liver fibrosis in
patients with hepatitis C virus infection: a prospective study. Lancet 2001;
357(9262):1069–75.
56. Cales P, Oberti F, Michalak S, et al. A novel panel of blood markers to assess the
degree of liver fibrosis. Hepatology 2005;42(6):1373–81.
57. Forns X, Ampurdanes S, Llovet JM, et al. Identification of chronic hepatitis C
patients without hepatic fibrosis by a simple predictive model. Hepatology
2002;36(4 Pt 1):986–92.
58. Adams LA, Bulsara M, Rossi E, et al. Hepascore: an accurate validated predictor
of liver fibrosis in chronic hepatitis C infection. Clin Chem 2005;51(10):1867–73.
59. Lieber CS, Weiss DG, Morgan TR, et al. Aspartate aminotransferase to platelet
ratio index in patients with alcoholic liver fibrosis. Am J Gastroenterol 2006;
101(7):1500–8.
60. Naveau S, Raynard B, Ratziu V, et al. Biomarkers for the prediction of liver fibrosis
in patients with chronic alcoholic liver disease. Clin Gastroenterol Hepatol 2005;
3(2):167–74.
61. Ratziu V, Massard J, Charlotte F, et al. Diagnostic value of biochemical markers
(FibroTest-FibroSURE) for the prediction of liver fibrosis in patients with non-
alcoholic fatty liver disease. BMC Gastroenterol 2006;6:66.
62. Sebastiani G, Vario A, Guido M, et al. Sequential algorithms combining non-

invasive markers and biopsy for the assessment of liver fibrosis in chronic hepa-
titis B. World J Gastroenterol 2007;13(4):525–31.
63. Myers RP, Tainturier MH, Ratziu V, et al. Prediction of liver histological lesions with
biochemical markers in patients with chronic hepatitis B. J Hepatol 2003;39(2):
222–30.
64. Hui AY, Chan HL, Wong VW, et al. Identification of chronic hepatitis B patients
without significant liver fibrosis by a simple noninvasive predictive model. Am
J Gastroenterol 2005;100(3):616–23.
65. Gomez-Dominguez E, Mendoza J, Rubio S, et al. Transient elastography: a valid
alternative to biopsy in patients with chronic liver disease. Aliment Pharmacol
Ther 2006;24(3):513–8.
66. Chang PE, Lui HF, Chau YP, et al. Prospective evaluation of transient elastogra-
phy for the diagnosis of hepatic fibrosis in Asians: comparison with liver biopsy
and aspartate transaminase platelet ratio index. Aliment Pharmacol Ther 2008;
28(1):51–61.
67. Ziol M, Handra-Luca A, Kettaneh A, et al. Noninvasive assessment of liver fibrosis
by measurement of stiffness in patients with chronic hepatitis C. Hepatology
2005;41(1):48–54.
68. Yoneda M, Yoneda M, Fujita K, et al. Transient elastography in patients with non-
alcoholic fatty liver disease (NAFLD). Gut 2007;56(9):1330–1.
Fibrosis a nd Cirrhosis
Rever sibilit y :
Clinic al Features
and I mplic ations
Massimo Pinzani, MD, PhD*, Francesco Vizzutti, MD, PhD
KEYWORDS
Fibrosis Cirrhosis Chronic liver disease
Antifibrotic therapy
FIBROSIS AND CIRRHOSIS: THE NEED FOR A CLEAR DISTINCTION
In the past decade, practicing hepatologists have finally directed their attention to the
most relevant outcome of most chronic liver diseases (CLD), ie, the progressive sub-
stitution of the functioning hepatic parenchyma with fibrotic tissue. Undoubtedly, the
significant advancements in the knowledge of cellular and molecular mechanisms of
hepatic fibrogenesis have greatly contributed to this change and, currently, major ef-
forts are directed at translating these acquisitions in diagnostic and therapeutic appli-
cations. Nevertheless, fibrosis and cirrhosis are often used as synonyms and this
causes obvious confusion. In particular, this problem emerges when dealing with
the question: are fibrosis and cirrhosis reversible?1
Tissue fibrosis is the consequence of a chronic wound-healing reaction occurring in
response to chronic damage and chronic inflammation in a biological context, charac-
terized by a limited repertoire of responses. In other words, the deposition of fibrillar
extracellular matrix (ECM) is the simplest, fastest and, only solution, and it is, eventu-
ally, an ‘‘intention-to-treat’’ process aimed at preserving tissue continuity. In addition,
the detectable amount of fibrosis is the net result of the continuous deposition of new
fibrillar ECM associated with a continuous, but obviously not efficient, attempt of deg-
radation and remodelling. Although tissue fibrosis is an essential element in the cir-
rhotic transformation of the liver, it is per se devoid of significant functional (and
clinically relevant) effects.
Cirrhosis is a diffuse process characterized by fibrosis and the conversion of normal
liver architecture into structurally abnormal nodules.2 Key morphologic features of cir-
rhosis include: diffuse fibrosis, regenerative nodules, altered lobular architecture, and
Dipartimento di Medicina Interna, Center for Research, High Education and Transfer, Università
degli Studi di Firenze, Florence, Italy
* Corresponding author. Dipartimento di Medicina Interna, Università degli Studi di Firenze,
Viale G.B. Morgagni, 85, 50134 Firenze, Italy.
E-mail address: m.pinzani@dmi.unifi.it (M. Pinzani).
Clin Liver Dis 12 (2008) 901–913

902 Pinzani & Vizzutti
establishment of intrahepatic vascular shunts between afferent (portal vein and

hepatic artery) and efferent (hepatic vein) vessels of the liver.3 The vascular shunts
are determined by the topography of the vascularized fibrotic septa and they repre-
sent an essential feature of cirrhosis.4 Other relevant characteristics are comprised
of: capillarization of sinusoids and perisinusoidal fibrosis; vascular thrombosis and
obliterative lesions in portal tracts and hepatic veins; under-perfusion of lobular paren-
chyma; and consequent tissue hypoxia.5,6 Together, these changes are responsible
for the development of portal hypertension and relative complications. Portal hyper-
tension is indeed the principal mechanism leading to the death of cirrhotic patients.
FIBROGENESIS AND ANGIOGENESIS: TOWARD THE POINT-OF-NO-RETURN
Vascular structures in cirrhotic septa originate either from pre-existing sinusoids,

which persist in areas of postnecrotic collapse of the connective tissue framework,
or from pathologic angiogenesis, which, irrespective of the etiology, have been exten-
sively described in CLD.7–9 From a mechanistic point of view, angiogenesis in CLD can
be explained along two main pathways. First, the process of chronic wound healing
typical of fibrogenic CLD is characterized by an over-expression of several growth
factors, cytokines and metalloproteinases (MMPs) with an inherent pro-angiogenic
action. In particular, platelet-derived growth factor (PDGF), transforming growth
factor-B1 (TGF- B1), fibroblast growth factor (FGF), and vascular endothelial growth
factor (VEGF) have been shown to exert potent profibrogenic and pro-angiogenic
roles.10–13 Second, neo-angiogenesis is stimulated in hepatic tissue by the progres-
sive increase of tissue hypoxia. This mechanism is strictly linked to the anatomic mod-
ifications following the establishment of periportal fibrosis with an increased
contribution of the hepatic artery to the formation of sinusoidal blood. Accordingly,
sinusoidal blood flow becomes increasingly arterialized with hepatocytes adjusting
to an abnormally high oxygen concentration. Subsequently, the progressive capillari-
zation of sinusoids leads to an impairment of oxygen diffusion from the sinusoids to
hepatocytes with the consequent up-regulation of pro-angiogenic pathways.14–17
Practically, tissue hypoxia promotes angiogenesis and fibrogenesis; fibrosis and hyp-
oxia aggravates each other in the presence of persistent parenchymal injury.
Evidence obtained by morphologic studies suggests that angiogenesis occurring in
hepatic tissue undergoing chronic wound healing is characterized by branching of
neo-vessels from the existing vasculature. The large majority of these neo-vessels
originate from the fine portal vein branches and tend to establish a connection
between the portal system and the hepatic veins.12,15 The role of bone marrow-
derived endothelial precursors (vasculogenesis) in hepatic angiogenesis has been
suggested by studies employing animal models of hepatic fibrogenesis16,18 and the
role needs to be substantiated in human CLD.
A possible pro-angiogenic role of hepatic stellate cells (HSC) during the fibrogenic
process has been suggested by several reports.19–24 In addition to the demonstration
that pro-angiogenic growth factors such as VEGF and angiopoietin 1 induce HSC
proliferation and motility, key conclusions of these studies are that HSC express
and release these factors in response to hypoxia and that neo-angiogenesis occurs
in parallel to fibrogenesis in the development of fibrous septa. In this context, the
promotion of a pro-angiogenic phenotype in activated HSC is also stimulated by non-
hypoxic conditions and particularly by the exposure to the profibrogenic adipokine
leptin.25 An elegant and convincing demonstration of the interplay among inflamma-
tory response, angiogenesis, and fibrogenesis has been recently provided by an
Fibrosis and Cirrhosis Reversibility 903
experimental study in which all these features have been significantly reduced by the
treatment with the multitargeted receptor tyrosine kinase inhibitor Sunitinib.26
The evidence so far accumulated suggests that the association of fibrogenesis and
angiogenesis should be regarded as crucial in the modern evaluation of disease pro-
gression and in the search for therapeutic targets. In addition, depending on the dif-
ferent pattern of fibrogenic evolution distinctive of different CLD (ie, post-necrotic,
biliary, centrolobular, pericellular/perisinusoidal), the extent of neo-angiogenesis
may have profound consequences on the rate of disease progression to cirrhosis
and it represents a key determinant affecting reversibility of fibrosis. Although there
is no definitive information on the features of neo-angiogenesis in relation to the differ-
ent patterns of fibrosis development, possibly the earliest is the alteration of sinusoidal
blood flow and the presence of tissue hypoxia and the fastest is the progression
toward the profound angio-architectural changes typical of cirrhotic liver. Accordingly,
it is quite possible that intralobular perisinusoidal fibrosis or biliary fibrosis are more
reversible than septal fibrosis, possibly due to differences in neo-angiogenesis.
THE REGRESSION OF CIRRHOSIS
‘‘Reversal’’ and ‘‘regression’’ are two terms characterizing the largely semantic ground
on which the current debate is based. In practical terms, ‘‘reversal’’ should be used
with caution because the term implies a return to normal. On the other hand, ‘‘regres-
sion,’’ implying a reduction in ECM content of any degree, without necessarily return-
ing the histology to normal, allows a more flexible interpretative approach. Other
frequently used terms that lead to further confusion include: ‘‘minimal,’’ ‘‘partial,’’
‘‘extensive,’’ and ‘‘total’’ reversal. Unfortunately, none of these terms is adaptable to
a reliable clinical evaluation.
Overall, the issue of regression or reversibility of cirrhosis originates from evidence
obtained in animal models upon the discontinuation of the cause of liver damage or
following treatment with a putative antifibrotic agent. This issue is reviewed in detail
in this issue of Clinics in Liver Disease by Gieling and colleagues, but it has been clear
for many years that experimental cirrhosis is characterized by reversible and
irreversible changes. In their enlightening report dated 1965, Quinn and Higginson27
described that, in the majority of the currently employed experimental models of cir-
rhosis, while fibrosis, inflammation, and bile duct proliferation decrease markedly on
withdrawal of the damaging stimulus, regenerative nodules become autonomous
with continued progressive growth. In another seminal paper published in 1979,
Perez-Tamayo28 provided evidence for the reversibility of fibrosis and cirrhosis in
both animal models and in human CLD; the author highlighted the notion that all ex-
perimental models of cirrhosis are reversible providing the causative agent is discon-
tinued and sufficient time is allowed. In addition, this author remarked that cirrhosis
that has been developing for weeks rather than years is characterized by different fea-
tures and, particularly, ‘‘increased reticulum fibers are more easily reabsorbed than
thick collagen bundles.’’ However, these two pioneering reports did not mention the
possible role of tissue hypoxia and neo-angiogenesis, which could indeed represent
key mechanisms linking cirrhosis to its reversibility or irreversibility.
Although regression has been shown in animal models of cirrhosis, this possibility is
not yet fully substantiated in humans. Evidence of either fibrotic or cirrhotic regression
has now been reported in CLD of different etiologies, including viral hepatitis,29–35
autoimmune hepatitis,36 alcoholic and nonalcoholic steatohepatitis.37–39 However,
when performing an accurate analysis of the results of these studies, the only prudent
conclusion is that, in most cases, there was a variable degree of fibrosis regression in
cirrhosis but not a reversal of cirrhosis.40,41 There is no convincing evidence that the
abnormalities of the intrahepatic vasculature regress in human cirrhotic liver. Actually,
the available evidence suggest that the so-called ‘‘veno-portal adhesions’’ persist
even in cases of extensive fibrosis regression, and evident ‘‘arterialized’’ sinusoids
appear in the context of intrahepatic arterio-venous shunts.42
The most obvious problems when discussing the issue of fibrosis regression in
cirrhosis or even cirrhosis reversal are the lack of a clear and common language
and, ultimately, problems with: 1) the precise distinction of advanced fibrosis (‘‘pre-
cirrhosis’’) from true cirrhosis; and 2) the possibility of staging cirrhosis. The problems
are fundamentally based on the use of semi-quantitative scoring systems for staging
fibrosis and, in particular, in the fact that cirrhosis is always represented by the highest
score and is indeed considered as an end-stage of CLD.41–43 Indeed, cirrhosis
appears in a very broad spectrum of variants (early, fully developed, ‘‘active’’, and
‘‘inactive’’) and more than one study has documented the transition from micronodular
to macronodular cirrhosis following the discontinuation of the causative agent.44,45
Practically, as clearly stated by Desmet and Roskams41, there is a fundamental differ-
ence between a diagnosis of cirrhosis and a score of cirrhosis. For example, mostly
due to potential sampling error, a low score does not exclude a cirrhosis of the macro-
nodular or incomplete septal type. From a clinical point of view, patients with cirrhosis
can experience widely variable clinical courses; the cirrhotic stage, defined as ‘‘com-
pensated cirrhosis’’, includes anything from the initial histopathological demonstration
of ‘‘early cirrhosis’’ to the development of complications of portal hypertension. This
oversight is mainly caused by the fact that, until recently, it hardly mattered whether
a patient had early or late cirrhosis because the only viable option was liver transplan-
tation, and that situation is clearly reflected by clinical staging systems such as the
MELD score.46
THE REGRESSION OF FIBROSIS
Although it is doubtful that an accurately-defined cirrhosis is able to reverse to normal,

there is sound evidence concerning the capacity of the healing liver to reabsorb scar
tissue and, accordingly, the issue of fibrosis regression has become central in hepa-
tology, both in terms of diagnostic and therapeutic approaches.
Even when asking the question ‘‘does fibrosis regress?’’ it is important to make
some practical considerations. Indeed, fibrotic deposition related to recent disease
and characterized by the presence of thin reticulin fibers, often in the presence of a dif-
fuse inflammatory infiltrate, is likely to be fully reversible, whereas long-standing fibro-
sis, branded by extensive collagen cross-linking by tissue transglutaminase, presence
of elastin, dense acellular/paucicellular ECM, and decreased expression and/or activ-
ity of specific metalloproteinases, is not.47–49
Progress in understanding HSC biology obtained in the past two decades provides
important support to the understanding of the mechanisms regulating fibrosis regres-
sion.50 In chronic liver injury, activated HSC are, together with other ECM-producing
cells, the major source of fibrillar ECM as well as of the tissue inhibitors of metallopro-
teinases (TIMPs), which inhibit collagen degradation.1,10,51 Studies performed in ani-
mal models of fibrogenesis have shown that recovery from acute and chronic injury
is characterized by apoptosis of activated HSC, which removes ECM-producing cells
that are also expressing TIMPs, thereby relieving the inhibition of matrix degrada-
tion.52 Both survival and apoptosis of HSC are regulated by growth factors expressed
during fibrotic liver injury and by components of the fibrotic ECM.53,54
However, the possibility of a straightforward induction of HSC apoptosis should be

considered with caution when translating these findings to human CLD. Work by Novo
and colleagues55 showed that activated HSC isolated from human liver and activated
in culture are resistant to most pro-apoptotic stimuli due to a rapid and persistent
overexpression of the anti-apoptotic protein Bcl-2. In addition, the expression of
Bcl-2 is markedly evident in myofibroblast-like cells present in areas of fibrosis in liver
tissue obtained from patients with hepatitis C (HCV)-related cirrhosis. It is therefore
plausible that long-term fibrogenesis is characterized, in addition to the biochemical
evolution of scar tissue49 and the lack of an appropriate degradation machinery,56
by the immovability of a critical mass of profibrogenic cells.
Other considerations concerning the possibility of fibrosis regression in the long-
term clinical course of CLD emerge from observations suggesting the key involvement
of macrophages in both fibrogenesis and fibrosis regression.57 It is still unclear
whether these opposite actions of macrophages are dependent on the existence of
different cellular subpopulations or if they are just secondary to the biological context
(ie, persistence or disappearance of tissue damage). It is speculated reasonably that,
in the setting of a successful eradication of the cause of damage (eg, by antiviral
therapy), the presence of macrophages in the inflammatory infiltrate is a guarantee
for fibrosis regression. ECM degradation by macrophages may indeed be faster and
more extensive than that ensured by HSC, whose degradation capability is progres-
sively impaired during activation.54 This issue becomes interesting when considering
the results of studies suggesting that bone marrow transplantation may represent an
attractive antifibrotic therapy.58,59 The molecular mechanism underlying such BMC-
related matrix degradation possibly includes the action of metalloproteinases,
although the precise role of diverse cell types needs further clarification.60,61
In the past decade, several biochemical markers for the prediction of the fibrotic
stage of CLD have been proposed. Some markers reflect alterations in hepatic
function but do not directly reflect ECM metabolism, ie, ‘‘indirect markers.’’ Others
are directly linked to the modifications in ECM turnover occurring during fibrogenesis,
ie, ‘‘direct markers.’’62,63 Fig. 1 illustrates the relationships existing among some of
the proposed panels of markers, which are generally structured in mathematic algo-
rithms, the stage of fibrosis (in this case according to the METAVIR scoring system
for chronic HCV hepatitis) and pathophysiological hallmarks, from parenchymal
damage to complications of portal hypertension. Clearly, algorithms including only
‘‘extreme’’ variables (ie, ALT increase and any of the parameters related to hepatocel-
lular failure or to portal hypertension) will be able to distinguish normality or very limited
fibrosis from advanced fibrosis and cirrhosis but not fibrosis in the intermediate
stages. The inclusion of parameters more closely related to inflammation and fibro-
genesis, particularly proteins and enzymes involved in ECM turnover, may improve
the performance of the algorithm; however, at the current stage of validation, all the
proposed biochemical systems remain characterized by a sufficient to excellent diag-
nostic accuracy for the detection (or exclusion) of advanced fibrosis and cirrhosis, yet
none is able to allow a step-wise follow-up of the fibrogenic evolution of CLD accord-
ing to the existing histopathologic staging systems.
Similar considerations apply to the use of transient elastography for the measure-
ment of liver stiffness (LSM), a physical measure supposed to be largely dependent
on the fibrotic transformation of liver tissue (reviewed in62). At least in part, the reasons
for this drawback lay in the absence of a true gold standard and, currently, the defini-
tion of 90% diagnostic accuracy remains a goal for the future. In addition, none of the
currently available tests has a well-defined prognostic value such as the prediction of
decompensation or death. Regardless, it is increasingly evident that rational, prudent
Fig. 1. Biochemical markers for the evaluation of fibrosis progression and their relationship
with the stage of disease and pathophysiologic hallmarks, from parenchymal damage to
complications of portal hypertension. Stages of fibrosis F0 to F4 are, according to the META-
VIR scoring system, for chronic HCV hepatitis. Abbreviations: ALT, alanine aminotransferase;
APO-A1, apolipoprotein A1; APRI, AST-platelet ratio index; AST, aspartate aminotransferase;
ELF, European liver fibrosis; gGT, g-glutamyltransferase; HA, hyaluronan; INR, international
normalized ratio; a2-MC, a2-macroglobulin; MMP, matrix metalloproteinases; PIIINP, N-ter-
minal propeptide of type III procollagen; PI, prothrombin index; TIMP, tissue inhibitors of
metalloproteinases. (Data from Pinzani M, Vizzutti M, Arena U, et al. Technology insight:
non-invasive assessment of liver fibrosis by biochemical scores and elastography. Nat Clin
Pract Gastroenterol Hepatol 2008;5:95–106; with permission.)
use of the proposed methodologies will reduce the need of liver biopsy in a significant
percentage of patients and this use represents a diagnostic advantage.
Although there is an extensive debate on the clinical usefulness of noninvasive
methods for the evaluation of fibrosis progression, it is completely unclear whether
methods for analyzing fibrosis regression should be the same as those used to assess
progression. Indeed, when dealing with this issue, the problem becomes even more
complex because there has been no validation of the use of liver biopsy to assess
regression rather than progression; additionally, no definitive information is available
regarding the speed and the modalities of fibrosis regression in human CLD. Beyond
consideration of fibrosis progression, the evaluation of fibrosis regression cannot ig-
nore the fact that liver biopsy, when analyzed by standard methodologies, provides
a predominantly static view of disease evolution.
Along these lines, efforts should be directed at translating the knowledge of the
cellular and molecular mechanisms of fibrogenesis into diagnostic systems able to
monitor fibrogenesis and fibrolysis, rather than just fibrosis. For example, the detec-
tion of fibrogenic/angiogenic growth factors receptor expression and the activation
of the relative intracellular signaling pathways could represent a marker of disease
progression in terms of fibrogenetic potential. An example of this possibility is offered
by the excellent correlation existing between platelet-derived growth factor (PDGF)
receptor expression and histopathological activity in human CLD.64 Similar
considerations could be made, in terms of fibrosis regression, for factors involved in

ECM degradation. Practically, this approach would not be very feasible nor easy to
standardize on liver biopsy specimens because of sampling error. However, future
technical advancements in integrated imaging systems, ie, SPECT/PET scan, may
offer sound opportunities in this direction,65 mostly because, in addition of being non-
invasive, they would allow an evaluation of the whole liver rather than of a very limited
portion.
REGRESSION OF FIBROSIS: WHICH END-POINTS FOR ANTIFIBROTIC THERAPY?
To answer the question about regression of fibrosis and end-points for antifibrotic
therapy, it is necessary to make a distinction between regression of fibrosis in a non-
cirrhotic liver and regression of fibrosis in cirrhosis.
Because the degree of fibrosis in a noncirrhotic liver is devoid of clinically corre-
sponding manifestations, the main end-point should be down-staging or, at least, sta-
bilizing fibrosis, ie, inducing a lack of progression. As sensibly stated by other
authors,1 adequate therapy might ensure that the patient will die of other causes
‘‘with’’ liver fibrosis rather than dying ‘‘of’’ cirrhosis. This is basically what the antiviral
treatment of chronic viral hepatitis has tried to realistically achieve in the past few
years.29–35 Likely, this objective will be even more feasible when antifibrotic drugs
could be combined with antiviral treatment. Unfortunately, although an impressive
number of compounds have been shown to be effective in reducing fibrosis in cell
and animal models,63 no drug is presently available for the treatment of patients
and, even more problematic, no large clinical trials are currently in progress. This
largely reflects the lag between the increasing knowledge on the mechanisms of fibro-
sis in any organ and system, and the lagging interest of the pharmaceutical industry,
which in part is constrained by the absence of robust markers of fibrosis as an alter-
native to histopathology.
At present, regardless of the type of therapeutic approach, the evaluation of fibrosis
down-staging or stabilization can only rely on histopathologic analysis, with all the well
known limitations, and its integration with the available noninvasive methods when
employed in a longitudinal assessment.
The increasing clinical awareness that cirrhosis represents a new dimension in the
clinical course of CLD and not just the extreme stage of fibrosis, together with the
more and more realistic possibility of reducing fibrosis even in a cirrhotic liver, have
led to the assumption that liver transplantation is no longer the only possible option
to increase patient survival. Importantly, according to epidemiologic data concerning
two of the most common CLD, ie, HCV and NASH, and the relative estimates of dis-
ease progression, the number of patients with definite cirrhosis will increase exponen-
tially in the next 10–15 years, with this increase representing the most frequent clinical
entity in hepatology practice.66,67
The possibility of monitoring fibrosis regression in cirrhosis faces the already men-
tioned lack of a system able to classify cirrhosis in different stages. A first distinction
should be made between ‘‘compensated’’ (ie, complication-free) and ‘‘decompen-
sated’’ (ie, with clinically evident complications of portal hypertension) cirrhosis. In
this context, a classification of compensated cirrhosis represents the major clinical
need when analyzing the effect of a causative and/or antifibrotic therapy aimed at pro-
longing complication-free survival.
Different approaches have been proposed in order to reach this goal. First, as
suggested by Goodman collegue,68 morphometric image analysis may help to
quantify the extension of fibrosis in cirrhotic liver, thus overcoming the biases of the
semi-quantitative scores and providing a staging of cirrhosis beyond an end-stage

score. However, morphometric analysis of hepatic collagen content may be also lim-
ited by potential sampling variability and is not reliable when dealing with fragmented
specimens.
Second, the measurement of hepatic venous pressure gradient (HVPG), a validated,
safe and highly reproducible technique, has been proposed for monitoring the
progression of the disease from the precirrhotic to advanced stages of cirrhosis.69
In absence of significant fibrotic evolution, HVPG, as an expression of intrahepatic
resistance, does not exceed 5 mm Hg, whereas a gradient of more than 5 mm Hg
is always associated with significant fibrosis. Cross-sectional studies have shown
that portal pressure (estimated by the HVPG) must reach certain thresholds for the
development of complications of portal hypertension: 10 mmHg for the development
of varices and ascites (‘‘clinically significant’’ portal hypertension) and 12 mmHg for
variceal bleeding (‘‘clinically severe’’ portal hypertension).70 Therefore, in broad terms,
at least cirrhotic patients with a HVPG within the range 5–10 mm Hg should be com-
plication-free, ie, ‘‘compensated.’’ Although it is difficult to ascertain whether HVPG
holds prognostic information independent of liver function, a correlation between
the features of cirrhosis, particularly the thickness of fibrous septa and the size of cir-
rhotic nodules, and HVPG has been reported by Nagula and coworkers.71 Therefore,
because most of the lethal complications of cirrhosis are related to portal hyperten-
sion, a primary end-point in this setting would be to achieve enough fibrosis regression
to obtain a reduction in portal pressure. Along these lines, several recent studies have
suggested that changes in HVPG could represent a realistic end-point for the thera-
peutic evaluation of antiviral therapy for chronic hepatitis C.72–74 In addition, it is pos-
sible that a reduction of fibrosis and portal pressure will lead to a general improvement
in liver function.75
Third, the majority of the noninvasive methodologies proposed for the evaluation of
fibrosis progression in CLD, including serum biomarkers and transient elastography
(reviewed in62), seem to have an adequate diagnostic accuracy for the prediction of
advanced fibrosis and cirrhosis and they could be further implemented for staging cir-
rhosis. Along these lines, a three-variable algorithm consisting of hyaluronic acid,
TIMP-1, and platelet count was recently shown to correlate with hystopathological
scores better than with the hepatic collagen content measured by morphometric anal-
ysis.76 In addition, in patients with compensated HCV cirrhosis, with a HVPG in the
range 5–12 mm Hg, LSM by transient elastography shows an excellent correlation
with HVPG values and may be a good predictor of significant and severe portal hyper-
tension.77 Therefore, it is conceivable that the use of noninvasive methodologies could
represent a feasible way to complement traditional/morphometric hystopathological
analysis and the measurement of HVPG in the attempt to stage compensated cirrhosis
and assess the response to treatment.
To summarize some of these concepts, Fig. 2 illustrates the relationship between
histopathology, the degree of portal pressure expressed as HVPG, different proposed
cut-off values of LSM, and the biological and clinical features of chronic HCV hepatitis
and cirrhosis. It appears clear that most of the clinical manifestations of the disease
are present once the METAVIR stage F4 is reached. Within this stage, a definition of
disease progression is broadly obtainable by measuring HVPG. Available evidence
suggest that LSM may be able, once thoroughly validated, to provide complementary
information concerning fibrosis progression in patients with compensated cirrhosis,77
whereas its validity in a more advanced stage remains very speculative.78 Although the
MELD score is routinely used for ranking the patients according to their death risk for
organ allocation, the inclusion of HVPG does not significantly improve the
Fig. 2. Relationship between histopathology, the degree of portal pressure expressed as

HVPG, different proposed cut-off values of LSM, and the biological and clinical features
of chronic HCV hepatitis and cirrhosis. Abbreviations: LSM, liver stiffness measurement;
HVPG, hepatic vein pressure gradient.
discriminative ability of MELD.79 This indirectly indicates that, in decompensated cir-

rhosis, there is no precise association between the degree of portal hypertension and
clinical parameters reflecting liver failure. Therefore, there is an additional need for de-
veloping staging systems able to integrate invasive and noninvasive measures of por-
tal hypertension with parameters reflecting the failure of complex metabolic pathways.
REFERENCES
1. Friedman SL, Bansal MB. Reversal of hepatic fibrosis-fact or fantasy? Hepatology

2006;43:S82–8.
2. Anthony PP, Ishak KG, Nayak NC, et al. The morphology of cirrhosis. Working
group sponsored by the World Health Organization. J Clin Pathol 1978;31:
395–414.
3. Millward-Sadler GH, Hanh EG, Wright R. Cirrhosis: an appraisal. In: Wright R, Mill-
ward-Sadler GH, Alberti KGMM, editors. Liver and biliary disease. 2nd edition.
London: Baillière Tindall WB Saunders; 1985. p. 821–60.
4. Rappaport AM, McPhee PJ, Fisher MM, et al. The scarring of the liver acini
(cirrhosis). Tridimensional and microcirculatory considerations. Virchows Arch A
Pathol Anat Histopathol 1983;402:107–37.
5. Reichen J, Egger B, Ohara N, et al. Determinants of hepatic function in liver
cirrhosis in the rat. Multivariate analysis. J Clin Invest 1988;82:2069–76.
6. Wanless IR, Wong F, Blendis LM, et al. Hepatic and portal vein thrombosis in
cirrhosis: possible role in development of parenchymal extinction and portal
hypertension. Hepatology 1995;21:1238–47.
7. Garcia-Monzon C, Sanchez-Madrid F, Garcia-Buey L, et al. Vascular adhesion

molecule expression in viral chronic hepatitis: evidence of neoangiogenesis in
portal tracts. Gastroenterology 1995;108:231–41.
8. Medina J, Arroyo AG, Sanchez-Madrid F, et al. Angiogenesis in chronic inflamma-
tory liver disease. Hepatology 2004;39:1185–95.
9. Medina J, Sanz-Cameno P, Garcia-Buey L, et al. Evidence of angiogenesis in pri-
mary biliary cirrhosis: an immunohistochemical descriptive study. J Hepatol 2005;
42:124–31.
10. Pinzani M, Rombouts K. Liver fibrosis: from the bench to clinical targets. Dig Liver
Dis 2004;36:231–42.
Semin Liv Dis 2001;21:397–416.
12. Shackel NA, McGuinnes P, Abbott CA, et al. Insights into the pathobiology of hep-
atitis C virus-associated cirrhosis: analysis of intrahepatic gene expression. Am
J Pathol 2002;160:642–54.
13. Shackel NA, McGuinnes P, Abbott CA, et al. Identification of novel molecules and
pathogenic pathways in primary biliary cirrhosis: cDNA array analysis of intrahe-
patic differential gene expression. Gut 2001;49:567–76.
14. Hoofring A, Boitnott J, Torbenson M. Three-dimensional reconstruction of hepatic
bridging fibrosis in chronic hepatitis C viral infection. J Hepatol 2003;39:738–41.
15. DeLeve LD. Hepatic microvasculature in liver injury. Semin Liver Dis 2007;27:
390–400.
16. Rosmorduc O, Wendum D, Corpechot C, et al. Hepatocellular hypoxia-induced
vascular endothelial growth factor expression and angiogenesis in experimental
biliary fibrosis. Am J Pathol 1999;155:1065–73.
17. Onori P, Morini S, Franchitto A, et al. Hepatic microvascular features in
experimental cirrhosis: a structural and morphometrical study in CCl4-treated
rats. J Hepatol 2000;33:555–63.
18. Jin YL, Enzan H, Kuroda N, et al. Vascularization in tissue remodeling after rat
hepatic necrosis induced by dimethylnitrosamine. Med Mol Morphol 2006;39:
33–43.
19. Corpechot C, Barbu V, Wendum D, et al. Hypoxia–induced VEGF and collagen I
expressions are associated with angiogenesis and fibrogenesis in experimental
cirrhosis. Hepatology 2002;35:1010–21.
20. Nakamura T, Torimura T, Sakamoto M, et al. Significance and therapeutic
potential of endothelial progenitor cell transplantation in a cirrhotic liver rat model.
21. Ankoma-Sey V, Matli M, Chang KB, et al. Coordinated induction of VEGF recep-
tors in mesenchymal cell types during rat hepatic wound healing. Oncogene
1998;17:115–21.
22. Wang YQ, Luk JM, Ikeda K, et al. Regulatory role of vHL/HIF-1a in hypoxia-
induced VEGF production in hepatic stellate cells. Biochem Biophys Res
Commun 2004;317:358–62.
23. Lee JS, Semela D, Iredale J, et al. Sinusoidal remodeling and angiogenesis:
a new function for the liver specific pericyte? Hepatology 2007;45:817–25.
2007;170:1942–53.
25. Aleffi S, Petrai I, Bertolani C, et al. Upregulation of proinflammatory and proangio-
genic cytookines by leptin in human hepatic stellate cells. Hepatology 2005;42:
1339–48.
26. Tugues S, Fernandez-Varo G, Muñoz-Luque J, et al. Antiangiogenic treatment

with Sunitinib ameliorated inflammatory infiltrate, fibrosis, and portal pressure in
cirrhotic rats. Hepatology 2007;46:1919–26.
27. Quinn PS, Higginson J. Reversible and irreversible changes in experimental cir-
rhosis. Am J Pathol 1965;47:353–69.
28. Perez-Tamayo R. Cirrhosis of the liver: a reversible disease? Pathol Annu 1979;
14(Pt2):183–213.
29. Shiratori Y, Imazeki F, Moriyama M, et al. Histological improvement of fibrosis in
patients with hepatitis C who have sustained response to interferon therapy.
Ann Intern Med 2000;132:517–24.
30. Kweon YO, Goodman ZD, Dienstag JL, et al. Decreasing fibrogenesis: an immu-
noistochemical study of paired liver biopsies following lamivudine therapy for
chronic hepatitis B. J Hepatol 2001;35:749–55.
31. Dienstag JL, Goldin RD, Heathcote EJ, et al. Histological outcome during long-
term lamivudine therapy. Gastroenterology 2003;124:105–17.
32. Poynard T, McHutchison J, Manns M, et al. Impact of pegylated interferon alfa-2b
and ribavirin on liver fibrosis in patients with chronic hepatitis C. Gastroenterology
2002;122:1303–13.
33. Farci P, Roskams T, Chessa L, et al. Long-term benefit of interferon alpha therapy
of chronic hepatitis D: regression of advanced hepatic fibrosis. Gastroenterology
2004;126:1740–9.
34. Serpaggi J, Carnot F, Nalpas B, et al. Direct and indirect evidence for the revers-
ibility of cirrhosis. Hum Pathol 2006;37:1519–26.
35. Hui CK, Leung N, Shek TW, et al. Sustained disease remission after spontaneous
HBeAg seroconversion is associated with reduction in fibrosis progression in
chronic hepatitis B Chinese patients. Hepatology 2007;46:690–8.
36. Dufour JF, DeLellis R, Kaplan MM. Reversibility of hepatic fibrosis in autoimmune
hepatitis. Ann Intern Med 1997;127:981–5.
37. Wakim-Fleming J, Mullen KD. Long-term management of alcoholic liver disease.
Clin Liver Dis 2005;9:135–49.
38. Dixon JB, Bhathal PS, Hughes NR, et al. Nonalcoholic fatty liver disease: improvement
in liver histological analysis with weight loss. Hepatology 2004;39:1647–54.
39. Kral JG, Thung SN, Biron S, et al. Effects of surgical treatment of the metabolic
syndrome on liver fibrosis, and cirrhosis. Surgery 2004;135:48–58.
40. Desmet VJ, Roskams T. Reversal of cirrhosis: evidence based medicine? Gastro-
enterology 2003;125:629–30.
41. Desmet VJ, Roskams T. Cirrhosis reversal: a duel between dogma and myth.
J Hepatol 2004;40:860–7.
42. Wanless IR, Nakashima E, Sherman M. Regression of human cirrhosis: morpho-
logic features and the genesis of incomplete septal cirrhosis. Arch Pathol Lab
Med 2000;124:1599–607.
43. Desmet VJ. Milestones in liver disease. Scoring chronic hepatitis. J Hepatol 2003;
38:382–6.
44. Popper H. What are the major types of hepatic cirrhosis? In: Ingelfinger F,
Relman A, Finland M, editors. Controversies in internal medicine. Philadelphia:
WB Saunders; 1966. p. 233–43.
45. Fauerholdt L, Schlichting P, Christensen E, et al. The Copenhagen study group
for liver disease. Conversion of micronodular into macronodular cirrhosis.
Hepatology 1983;3:928–31.
46. Wiesner R, Edwards E, Freeman R, et al. Model for end-stage liver disease
(MELD) and allocation of donor livers. Gastroenterology 2003;124:91–6.
47. Ricard-Blum S, Bresson-Hadni S, Vuitton DA, et al. Hydroxypyridinium collagen

cross-links in human liver fibrosis: study of alveolar echinococcosis. Hepatology
1992;15:599–602.
48. Hayasaka A, Ilda S, Suzuki N, et al. Pyridinoline collagen cross-links in patients
with chronic viral hepatitis and cirrhosis. J Hepatol 1996;24:692–8.
49. Issa R, Zhou X, Constandinou CM, et al. Spontaneous recovery from
micronodular cirrhosis: evidence for incomplete resolution associated with matrix
cross-linking. Gastroenterology 2004;126:1795–808.
51. Iredale JP. Hepatic stellate cell behavior during resolution of liver injury. Semin
Liver Dis 2001;21:427–36.
sion of metalloproteinase inhibitors. J Clin Invest 1998;102:538–49.
53. Issa R, Williams E, Trim N, et al. Apoptosis of hepatic stellate cells: involvement in
resolution of biliary fibrosis and regulation by soluble growth factors. Gut 2001;
48:548–57.
action of collagenase results in failure of recovery from CCl4-induced liver
fibrosis, persistence of activated hepatic stellate cells, and diminished hepato-
cyte regeneration. FASEB J 2003;17:47–9.
55. Novo E, Marra F, Zamara E, et al. Overexpression of Bcl-2 by activated human
hepatic stellate cells: resistance to apoptosis as a mechanism of progressive
hepatic fibrogenesis in humans. Gut 2006;55:1174–82.
56. Milani S, Herbst H, Schuppan D, et al. Differential expression of matrix-
metalloproteinase-1 and -2 genes in normal and fibrotic human liver. Am J Pathol
1994;144:528–37.
57. Duffield JS, Forbes SJ, Constandinou CM, et al. Selective depletion of macro-
phages reveals distinct, opposing roles during liver injury and repair. J Clin Invest
2005;115:56–65.
58. Sakaida I, Terai S, Yamamoto N, et al. Transplantation of bone marrow cells
reduces CCl4-induced liver fibrosis in mice. Hepatology 2004;40:1304–11.
59. Terai S, Ishikawa T, Omori K, et al. Improved liver function in patients with liver
cirrhosis after autologous bone marrow cell infusion therapy. Stem Cells 2006;
24:2292–8.
60. Higashiyama R, Inagaki Y, Hong YY, et al. Bone marrow-derived cells express
matrix metalloproteinases and contribute to regression of liver fibrosis in mice.
Hepatology 2007;45:213–22.
61. Russo FP, Kallis YN, Alison MR, et al. Bone marrow cells in the liver: diverse cells,
diverse effects. Hepatology 2007;46:604–5.
62. Pinzani M, Vizzutti M, Arena U, et al. Technology insight: noninvasive assessment
of liver fibrosis by biochemical scores and elastography. Nat Clin Pract Gastroen-
terol Hepatol 2008;5:95–106.
63. Pinzani M, Rombouts K, Colagrande S. Fibrosis in chronic liver diseases: diagno-
sis and management. J Hepatol 2005;42:S22–36.
Am J Pathol 1996;148:785–800.
65. Wagner S, Breyholz HJ, Faust A, et al. Molecular imaging of matrix metalloprotei-
nases in vivo using small molecule inhibitors for SPECT and PET. Curr Med Chem
2006;13:2819–38.
66. Seef LB, Lie BA. Is cirrhosis an inevitable consequence of chronic hepatitis C
infection? Clin Gastroenterol Hepatol 2005;3:840–2.
67. Angulo P. Nonalcoholic fatty liver disease. N Engl J Med 2002;346:1221–31.
68. Goodman ZD, Becker RL, Pockros PJ, et al. Progression of fibrosis in advanced
chronic hepatitis C: evaluation by morphometric image analysis. Hepatology
2007;45:886–94.
69. Bosch J, Garcia-Pagan JC, Berzigotti A, et al. Measurement of portal pressure
and its role in the management of chronic liver disease. Semin Liver Dis 2006;
26:348–62.
70. Garcia-Tsao G, Groszmann RJ, Fisher RL, et al. Portal pressure, presence of gas-
troesophageal varices and variceal bleeding. Hepatology 1985;5:419–24.
71. Nagula S, Jain D, Groszmann RJ, et al. Histological-hemodynamic correlation in
cirrhosis-a histological classification of the severity of cirrhosis. J Hepatol 2006;
44:111–7.
72. Rincon D, Ripoll C, Lo Iacono O, et al. Antiviral therapy decreases hepatic venous
pressure gradient in patients with chronic hepatitis C and advanced fibrosis. Am
73. Roberts S, Gordon A, McLean C, et al. Effect of sustained viral response on
hepatic venous pressure gradient in hepatitis C-related cirrhosis. Clin Gastroen-
terol Hepatol 2007;5:932–7.
74. Carrion JA, Navasa M, Garcıa-Retortillo M, et al. Efficacy of antiviral therapy on
hepatitis C recurrence after liver transplantation: a randomized controlled study.
75. Veldt BJ, Heathcote EJ, Wedemeyer H, et al. Sustained virologic response and
clinical outcomes in patients with chronic hepatitis C and advanced fibrosis.
Ann Intern Med 2007;147:677–84.
76. Fontana RJ, Goodman ZD, Dienstag J, et al. Relationship of serum markers with
liver fibrosis stage and collagen content in patients with advanced chronic hep-
atitis C. Hepatology 2008;47:789–98.
77. Vizzutti F, Arena U, Romanelli RG, et al. Liver stiffness measurement predicts
severe portal hypertension in patients with HCV-related cirrhosis. Hepatology
2007;45:1290–7.
78. Foucher J, Chanteloup E, Vergnion J, et al. Diagnosis of cirrhosis by transient
elastography (FibroScan): a prospective study. Gut 2006;55:403–38.
79. Ripoll C, Banares R, Rincon R, et al. Influence of hepatic venous gradient on the
prediction of survival of patients with cirrhosis in the MELD era. Hepatology 2005;
42:793–801.
Fibrosis a nd Cirrhosis
Rever sibilit y ^ Mole cular
Mecha nisms
Roben G. Gieling, PhDa, Alastair D. Burt, MDb, Derek A. Mann, PhDa,*
KEYWORDS
Hepatic myofibroblasts Apoptosis NF-kB Fibrosis
Extracellular matrix Neovascularisation
Early investigators considered the structural changes in the cirrhotic liver to represent
collapse of the normal ‘‘reticulin’’ framework due to loss of parenchymal substance,
with the nodules arising as a passive consequence. Popper and others later began
to demonstrate that the evolution of cirrhosis involves several parallel phenomena in-
cluding fibrogenesis – a very active process – and nodular regeneration of surviving
mature hepatocytes (now recognized as proliferation of progenitor cell populations
and, in some situations, bone marrow derived precursors). However, until the end of
the 1970s, a central dogma of hepatology was that chronic liver disease is a relent-
lessly progressive condition which at the more severe end of the spectrum and
certainly when there was established cirrhosis is irreversible. This dogma was first
challenged by Perez-Tamayo1 who described reversal of advanced fibrosis (pre-
cirrhosis) in three patients. This observation served to establish a new paradigm
that the fibrotic process is highly dynamic with potential for regression as well as pro-
gression. There is now a general consensus that liver fibrosis is potentially reversible.2
However, there remains skepticism that cirrhosis can be truly reversed. The aim of this
review is to briefly survey the evidence for fibrosis/cirrhosis reversion and consider the
key cellular and molecular mechanisms that dictate fibrosis progression versus
regression. The unique features of the cirrhotic liver that may hinder reversion will
be discussed. Possibilities for therapeutic regression of fibrosis will be investigated
together with the hurdles that need to be overcome before such therapies are brought
to the clinic.
This work was supported by the UK Medical Research Council, British Liver Trust, and the
Wellcome Trust.
a
Liver Research Group, Institute of Cellular Medicine, Level 4, Cookson Building, Newcastle
University, Framlington Place, Newcastle upon Tyne, NE24HH, UK
b
University of Newcastle upon Tyne, Peacock Hall, Royal Victoria Infirmary, Newcastle upon
Tyne, NE14LP, UK
E-mail address: derek.mann@ncl.ac.uk (D.A. Mann).
Clin Liver Dis 12 (2008) 915–937

916 Gieling et al
THE DISTINCTION BETWEEN FIBROSIS AND CIRRHOSIS
It is important to carefully distinguish between the terms ‘‘fibrosis’’ and ‘‘cirrhosis’’ to

ensure they are not used interchangeably, especially when considering reversion of
liver disease. Liver fibrosis is essentially the net deposition of excess extracellular
matrix (ECM) arising from an imbalance between the hepatic fibrogenic and fibrolytic
activities. Fibrosis is also characterized by qualititative changes in the hepatic ECM,
with a shift toward the expression and maintenance of fibril-forming collagens (princi-
pally collagen types I and III).3 Crucially, fibrosis is accompanied by markedly reduced
activity of the collagenases that normally promote degradation of fibrotic matrix. If this
state continues, then the normal hepatic matrix associated with the perisinusoidal
and/or periportal spaces is progressively remodeled to a fibrotic matrix. The pivotal
cellular event responsible for tipping the balance toward fibrogenesis is the appear-
ance of hepatic myofibroblasts (HM), which can be detected by stains for smooth
muscle a-actin (a-SMA) and (in some species) desmin. In normal liver, very few HM
are detected; however, in response to liver injury, large numbers of HM are found in
areas of necroinflammation. The principal source of HM is via the transdifferentiation
of perisinusoidal hepatic stellate cells;4 however, other cellular origins of HM are
increasingly recognized including periportal fibroblasts, so-called ‘‘interface myofibro-
blasts,’’ fibrocytes, transdifferentiated epithelia, and possibly bone marrow-derived
stem cells.5–10 Collectively, HM make a major contribution to fibrogenesis by secreting
vast quantities of collagen I/III and they also prevent the normal turnover of this matrix
by expressing high levels of the tissue inhibitor of metalloproteinase-1 (TIMP-1), which
is a potent and broad-specificity inhibitor of matrix metalloproteinases (MMPs) inclu-
sive of collagenases. The accumulation of matrix proteins is not restricted to interstitial
collagens; proteoglycans and noncollagenous glycoproteins including laminin, undu-
lin, fibronectin, and nidogen are all increased in amount as are other collagen mole-
cules such as type VI.11
The degree of fibrosis can be assessed in liver biopsy specimens; this is now fre-
quently used in clinical practice. Several semiquantitative scoring systems have
been developed including the Metavir scheme, which describes fibrosis according
to stages F0 (normal) to F4 (cirrhosis); other systems such as the Ishak system attempt
to be more discriminatory with a greater number of stages. The pitfalls of using such
scoring are well documented (see).12 There are very real sampling problems when
assessing needle biopsies. Bedossa and colleagues13 have shown that specimens
of >2.5 cm in length are needed to provide anything that resembles a representative
and reproducible assessment. Several studies have also shown that there can be sig-
nificant inter- and intra-observer variations when applying such approaches;14 the
kappa values are rarely better than 0.6 even when experienced hepatopathologists
are involved. Furthermore, a numeric system is used for descriptors and there is no
evidence that these represent a linear progression. Indeed, all systems in routine
use confound the assessment of the amount of fibrosis with the presence or absence
of cirrhosis. An alternative approach using biopsy material for assessing the amount of
fibrosis alone (as opposed to also assessing architectural changes) is morphometric
analysis of sections stained for fibril-forming collagen (Sirius red); a-SMA to assess
HM load can be used to provide further information. In experimental models of rodent
fibrosis, the ‘‘gold standard’’ measurement of fibrosis is quantification of hepatic
hydroxyproline content, which provides a biochemical readout of the level of hepatic
collagen. The current lack of a sensitive and robust assay for regular repeated
measurements of dynamic changes in hepatic collagen content is an impediment to
determining the extent to which fibrosis is progressing or regressing.
The evolution of fibrosis to cirrhosis involves a great deal more than the spread and
thickening of fibrotic bands across the liver. Additionally, there are gross distortions of
normal tissue architecture with alteration in the relationship between vascular struc-
tures and establishment of shunts between the afferent and efferent hepatic vessels.15
Portal-systemic shunting is likely to be a major contributor to progressive and irrevers-
ible hepatic failure in cirrhosis. Micronodular cirrhosis appears as thin fibrotic septae
surrounding small (<3 mm diameter) nodules of regenerating hepatocytes. In macro-
nodular cirrhosis, the fibrotic bands are thicker and envelop nodules of varying size
(up to 5 cm diameter), often incorporating terminal hepatic venules and portal tracts.
A pathological feature that is unique to cirrhosis is the development of fibrous vascu-
larized septa linking portal tracts and central veins. These structures, which are in part
formed by angiogenesis, function to shunt blood between the hepatic artery/portal
veins and the centrilobular veins, effectively bypassing the lobular parenchyma. In
advanced cirrhosis, the shunts can measure 1–2 mm in diameter and carry a large
volume of the hepatic blood supply.
Another important vascular change contributes to the development of cirrhosis:
venous occlusion. In many forms of chronic liver disease, there is obliteration of portal
and hepatic veins; the former is seen particularly in conditions in which there is intense
portal inflammation such as primary sclerosing cholangitis and the latter is frequently
observed in alcoholic liver disease.16 The occlusion occurs in part by intraluminal
thrombosis, but changes in the vein wall contribute and veno-occlusive lesions with
intimal thickening and inflammation have been observed. In alcoholic liver disease
(and NAFLD), there is also obliteration of vessels by progressive perivascular fibrosis
(so-called ‘‘phlebosclerosis’’).16,17 Vascular occlusion leads to hypoxia and, particu-
larly where there is obliteration of portal vessels, this may lead to loss of areas of liver
tissue through ischemia. Thus large areas may be seen in the end-stage cirrhotic liver
where there is complete loss of liver cell plates and accompanying sinusoids: so-
called ‘‘parenchymal extinction’’. It is possible that the degree to which this occurs
may influence when cirrhosis passes a point of ‘‘no-return’’. Some have suggested
that this should be assessed in liver biopsy specimens, and some have argued that
Stage 4 in the Metavir system should be subdivided according to the size of septa
and amount of parenchymal extinction; this has yet to be fully validated.18
Cirrhosis is, therefore, very distinct in pathology terms from fibrosis and will require
more than a redress of the imbalance in fibrogenesis versus fibrinolysis to achieve
reversion. Furthermore, portal hypertension, ascites formation, and varices are all
clinical signs commonly observed in cirrhotic patients, which must be considered
when assessing the clinical success of cirrhosis reversion.
THE EVIDENCE FOR FIBROSIS REGRESSION
The terms ‘‘regression’’ and ‘‘reversion’’ are often used loosely to define a partial or
complete return to normal liver structure and function and these terms may be unhelp-
ful unless defined. For the purpose of this review, regression will describe changes as-
sociated with fibrotic ECM remodeling and the dissolution of fibrotic septae. Reversion
will be used when referring to an apparent return to a near-normal liver architecture.
Clinical Evidence
Clinical evidence for spontaneous reversion of cirrhosis is rare in humans. Bortolotti
and colleagues19,20 reported two cases of hepatitis B virus-associated cirrhosis that
developed during childhood and which underwent spontaneous regression to the
extent that the patients were considered ‘‘cured.’’ Other examples of apparent
918 Gieling et al
reversion of cirrhosis were in the context of treatment of the underlying cause of liver
disease. Massarrat and colleagues,21 describe a 65-year-old patient, diagnosed with
hepatitis B-induced cirrhosis who received antiviral therapy (furosemide and spirono-
lactone), twice-a-week for 8 years. However, the question remains as to whether the
anatomical architecture returned to normal (as angiography could have revealed).
A retrospective study of 113 patients with biopsy-proven cirrhosis showed reversion
of cirrhosis in fourteen patients (12.4%) based on histological scoring and biochemical
parameters after prolonged treatment with immunosuppressive therapy.22 Among 36
patients with genetic hemochromatosis, venesection therapy decreased the level of
fibrosis in patients diagnosed with F3 fibrosis and cirrhosis.23 A major issue with these
studies is the extent to which return to normal pathology has been fully established
and whether sampling error associated with the biopsy gave the false impression
that cirrhosis has been reversed.
In contrast to cirrhosis, there is now a substantial body of evidence for regression of
fibrosis in humans, and indeed, this evidence can be drawn from studies with almost
every type of liver disease.2 However, large scale studies leading to routine analysis of
progression versus regression of fibrosis remain a challenge due to the continued
diagnostic use of the liver biopsy.
Experimental Evidence
Experimental evidence for reversion of cirrhosis is rare and complicated by the fact
that the animal models currently available do not accurately recapitulate the complex
architectural changes observed in human liver disease. In addition, these models tend
to develop fibrosis over a relatively short period of time; in rats, 4 weeks for fibrosis and
12 weeks for ‘‘cirrhosis.’’ This does not compare with human liver disease which
develops over many years during which time long-term and possibly irreversible
remodeling of matrix and vascular structures may occur that might not have time to
develop in rats or mice.
The best-characterized work was by Issa and colleagues24 who used a 12-week
carbon tetrachloride (CCl4) injury model to establish the fibrotic appearances of mac-
ronodular and micronodular cirrhosis. Liver pathology revealed mixed micronodular
and macronodular cirrhosis with the former able to undergo extensive regression
upon cessation of injury, but the latter persisting for 1 year post-injury. The authors
proposed that fibrotic matrix associated with macronodular cirrhosis was chronolog-
ically deposited before the ‘‘younger’’ matrix associated with micronodular fibrosis.
The more mature thick fibrotic matrix was relatively hypocellular and highly cross-
linked, in part due to the action of HM-derived tissue transglutaminase. This important
study suggests that there is limited potential for spontaneous regression of mature
fibrotic matrix in the cirrhotic liver; for this to be potentiated, there will be a requirement
for therapeutic fibrolysis.
There is substantial experimental evidence for almost complete regression of fibro-
sis, which includes studies describing spontaneous and stimulated remodeling of
fibrotic matrix in the rodent liver established by toxin- and bile duct ligation (BDL)-
induced injuries. These models of fibrosis regression have provided key insights
into two molecular processes thought to regulate remodeling of fibrotic ECM: HM
apoptosis and fibrolysis.
HEPATIC MYOFIBROBLAST APOPTOSIS: GUILT BY ASSOCIATION?
This review appears a decade after the seminal original study by Iredale and col-
leagues3 that first proposed HM apoptosis as a cellular mechanism for enabling
spontaneous regression of established CCl4-induced fibrosis in rats. HM apoptosis

was also observed during recovery from BDL-induced liver disease suggesting it to
be a conserved mechanism operating during resolution of fibrosis.25 HM apoptosis
provides an attractive and plausible process for enabling fibrosis regression because
clearance of HM removes the main hepatic supply of newly synthesized collagens and
TIMP-1.26 An alternative mechanism would be reverse differentiation of HM back to
their original nonmyofibroblastic phenotype. Cell culture studies have shown that
HSC-derived HM can at least partially revert to their quiescent nonfibrogenic state, ei-
ther through manipulation of their interactions with their ECM or by re-expression of
regulatory genes associated with the quiescent phenotype such as PPARg.27,28 How-
ever, there is currently no evidence to support reversion of HM differentiation in vivo
during fibrosis regression. To the contrary, there is evidence against this mechanism.
Kim and colleagues,29 observed decreased numbers of aSMA and desmin positive
cells during HGF-stimulated recovery from dimethylnitrosamine (DEN) induced fibro-
sis in rats. Because desmin is a marker for both activated rat HM and quiescent rat
HSC, it can be concluded that apoptosis rather than phenotype reversion was respon-
sible for the reduction in numbers of HM.
Following the study by Iredale and colleagues, there has been an explosion of
research describing stimulators and inhibitors of HM apoptosis, leading to the sugges-
tion that experimental stimulation of apoptosis may be of therapeutic value.26 Unfor-
tunately, definitive evidence is still lacking that HM apoptosis is of mechanistic
importance for fibrosis regression, especially in humans. At present, the only con-
firmed conclusion is that HM apoptosis is associated with regression of experimental
fibrosis, as there remains the possibility that HM apoptosis is simply a bystander effect
of ECM remodeling. Of relevance to this alternative scenario and, as described in more
detail below, both collagen I and TIMP-1 protect HM from apoptosis, and their expres-
sion is diminished during recovery from liver injury.30,31 In favor of HM apoptosis as the
mechanistic driver of apoptosis, compounds such as gliotoxin and sulfasalazine,
which promote HM apoptosis in vitro, are also able to stimulate HM apoptosis in
vivo, and both significantly accelerate regression of established fibrosis in rats.32–36
However, neither compound is specific for HM, with some apoptotic loss of Kupffer
cells also observed in vivo.37–39 As such, alternative mechanisms of action for gliotoxin
and sulfasalazine can not be ruled out. More definitive evidence would be provided
where stimulation or prevention of apoptosis is specifically targeted to HM. With the
development of in vivo targeting technologies such as the antibody C1-3, PDGF-
albumin, and mannose-6-phosphate-albumin, all of which selectively target HM, it
should now be possible to more definitively test the HM apoptosis hypothesis.40–42
An additional caveat is the limited published evidence of HM apoptosis in the recov-
ering human liver. Moreover, human HM may be less susceptible to apoptosis than
rodent HM.43 To strengthen the evidence for a prominent role for HM apoptosis and
build a case for therapeutic targeting of this process, convincing detection of apopto-
sis of human HM and its association with fibrosis regression is required.
REGULATORS OF HEPATIC MYOFIBOBROBLAST APOPTOSIS
Regardless of its contribution to the control of fibrogenesis, HM apoptosis is also likely

to be an important regulator of other processes involved in the reversion of liver
pathology. The contractile properties of HM are implicated in one of the major compli-
cations of cirrhosis, portal hypertension.44 Apoptotic clearance of HM may therefore
relieve one of the potential contributors to increased vascular resistance which would
be of enormous clinical benefit. Because HM are also implicated as regulators of
920 Gieling et al
angiogenesis,45 clearance of HM may therefore attenuate remodeling of the hepatic

vasculature, which is a hallmark of progression to cirrhosis. Given the potential for
HM apoptosis to influence multiple features of liver physiology, it is therefore reason-
able to consider the key regulators of HM apoptosis. Detailed reviews of the ever-
increasing list of naturally occurring as well as pharmacological inhibitors and
stimulators of HM apoptosis have been recently published.26,46 For the purpose of
this review, discussion is limited to those regulators that have been functionally ana-
lyzed using in vivo models providing some validation of mechanistic relevance to the
process of fibrosis regression.
Transcriptional Regulators of Hepatic Myofibroblast Apoptosis

Several transcription factors have been implicated in controlling the reversal of fibrosis
including: NF-kB, p53, JunD, FXR, and CEBP-b. The best characterized of these
factors is NF-kB, which is expressed in HM in a constitutive active state (both in vitro
and in vivo) due to transcriptional repression of its naturally occurring inhibitor
IkB-a.47,48 The authors’ group recently employed a cell-permeable mutated IkB-a
that forms stable transcriptional inactive complexes with NF-kB to show that the tran-
scription factor is critical for survival of mature rodent and human HM.49 In the
absence of transcriptionally active NF-kB, the HM loses expression of at least two
key anti-apoptotic proteins, Gadd45b and Bcl2. The precise mechanism that drives
apoptosis when NF-kB is suppressed is unclear but requires Jun-N-terminal kinase
(JNK) activation, which is under negative control by the NF-kB regulated protein
Gadd45b.37 JNKs are three closely related kinases (JNK1-3) originally identified as
regulators of Jun transcription factors. However JNKs have broader substrate speci-
ficity and can regulate the activity of several proapoptotic factors including p53, p73
and the BH3-only subgroup of Bcl2-related proteins, Bim and Bmf.50–52 JNK activa-
tion in HM stimulates translocation of the pro-apoptotic Bcl2 family protein Bax into
the mitochondria, cytochrome c release, and caspase 3-dependant apoptosis.53 Pre-
liminary data from our laboratory reveals p53 as an additional downstream mediator of
JNK-dependent HM apoptosis.49 p53 stimulates expression of pro-apoptotic Bcl2-
family proteins Bax and PUMA.54 NF-kB therefore most likely operates to prevent
activation of the intrinsic stress-activated mitochondrial apoptosis pathway by main-
taining expression of anti-apoptotic Bcl2 and suppressing JNK activation of p53
required for expression of pro-apoptotic p53 target genes Bax and PUMA (Fig. 1).
Several laboratories have utilized pharmacological and molecular inhibitors of
NF-kB as either stimulators of fibrosis regression or as preventive treatments admin-
istered during the course of liver injury. A single high-dose administration of NF-kB
inhibitors gliotoxin and sulfasalazine will stimulate accelerated recovery of CCl4-
induced fibrosis, and this effect is associated with significantly elevated rates of HM
apoptosis.32,37 Of note, treatment of fibrotic rats with sulfasalazine resulted in reduced
hepatic expression of TIMP-1 and elevated MMP activity.37 As sulfasalazine does not
directly influence TIMP-1 gene transcription in HM, this observation supports the
hypothesis that stimulated apoptotic clearance of HM is a mechanism for promoting
fibrolytic activity. NF-kB is a target for thalidomide, which can induce HM apoptosis
and ameliorate liver fibrosis induced by CCl4, BDL, thioacetamide (TAA) and
DEN.55–58 It is more likely that the primary effect of thalidomide is to suppress hepatic
inflammation; effects of the drug in models of fibrosis regression, however, are lack-
ing. Similarly, although double-stranded oligonucleotide NF-kB decoys accelerate
TNF-a-induced HM apoptosis and inhibit CCl4-induced fibrosis, the major in vivo
effect of the NF-kB decoy is likely to be via suppression of inflammation.59 More
definitive in vivo evidence for NF-kB as an HM pro-survival factor and suppressor of
Fig. 1. A model for the anti-apoptotic actions of NF-kB in the hepatic myofibroblasts. Path-
ways stimulating cell survival are denoted by unbroken arrows; those promoting apoptosis
are indicated by broken arrows. Constitutive active nuclear NF-kB interacts with its binding
site in the upstream regulatory regions of anti-apoptotic genes such as Bcl2 and Gadd45b,
which suppress different component of the intrinsic mitochondrial apoptosis pathway.
Gadd45b operates as a repressor of JNK-mediated phosphorylation and activation of
apoptosis-promoting proteins such as p53. Bcl2 functions by inhibiting the actions of mito-
chondrial regulators Bax and Puma.
fibrosis regression will hopefully emerge from studies that target genetic or pharmaco-
logical knockout of NF-kB to HM.
JunD is the predominant Jun-family protein expressed in HM and it regulates tran-
scription of two key pro-fibrogenic genes, TIMP-1 and IL-6.60 JunD expression is
induced with HM activation both in vitro and in vivo and is detected in HM of diseased
human liver.61 Chronic CCl4 injury of JunD knockout mice was associated with
reduced hepatic expression of TIMP-1 and attenuation of fibrosis development and
recovery, the latter being remarkably rapid in JunD deficient mice.61 The discovery
that the ERK1/2 pathway is involved in regulation of JunD-stimulated TIMP-1 expres-
sion in an apparent HM-selective manner raises the potential for inhibitors of this path-
way to be tested as stimulators of fibrosis regression.61
The farnesoid X receptor (FXR) is a member of the nuclear receptor superfamily of
ligand-activated transcription factors and is selectively activated upon binding bile
acids, with chenodeoxycholic acid being its most active natural ligand.62 FXR is
expressed by rodent and human HM and, in response to bile acid binding, represses
the expression of TGF-b1, a1(I) collagen and TIMP-1.63 In the case of TIMP-1, FXR
negatively regulates gene transcription via induction of the short heterodimer partner
922 Gieling et al
(SHP) orphan nuclear receptor, which directly interacts with JunD and prevents its
binding to an AP-1 site in the TIMP-1 promoter.63,64 FXR ligands are potent stimulators
of apoptosis for the T6 HSC line and this effect was prevented by incubation of the
cells with recombinant TIMP-1.64 This latter observation provides further support for
a JunD/TIMP-1 pathway to play a key role in preventing fibrosis regression via stimu-
lation of HM survival and indicates activation of FXR as a means to suppress this path-
way. Administration of FXR ligand 6-ECDCA to rats during liver injury (chronic CCl4)
protected against fibrosis development. Furthermore, 6-ECDCA treatment of rats
with established fibrosis decreased hepatic TIMP-1 content, increased MMP activity,
increased numbers of apoptotic non-parenchymal liver cells, and accelerated the rate
of regression of fibrosis following cessation of injury.64 FXR ligands therefore hold
great promise not only as antifibrotics but also as stimulators of fibrosis regression.
CCAAT/enhancer-binding protein beta (CEBP-b) is a basic ZIP transcription factor
that is expressed in adipose, hepatic, and immune tissues and a variety of other
tissues. Buck and Chojkier65 recently revealed a role for C/EBP-b and its regulatory
molecule Ribosomal S-6 Kinase (RSK) as regulators of HM apoptosis and fibrosis
regression. RSK phosphorylates murine C/EBP-b at Thr217, mice carrying a mutation
(Thr217Ala) that prevents RSK phosphorylation of C/EBP-b are protected against
CCl4-induced inflammation and fibrosis. A cell-permeable RSK-inhibitory peptide car-
rying the Ala217 mutant phosphoacceptor motif of C/EBP-b stimulated regression of
established severe fibrosis in the context of ongoing liver injury. The RSK-inhibitory
peptide blocked phosphorylation of endogenous C/EBP-b at Thr217 in HSC-derived
HM, elevated caspase 8 activation and induced caspase-3 dependent apoptosis.
Precisely how phosphorylated C/EBP-b stimulated caspase 8 and apoptosis was
not determined; however, in an earlier study the authors described an interaction
between C/EBP-b-PhosphoThr217 and inactive caspase 8 suggesting that the phos-
phorylation status of C/EBP-b may help release caspase 8 from a complex that
renders the enzyme in an inactive state.66 The authors also suggest that granzyme
B may be an alternative mediator of the effects of phosphorylated C/EBPb-Thr217
because the enzyme is regulated by C/EBPb and is implicated as a regulator of HM
survival.
Receptor-Ligand–Mediated Apoptosis of Hepatic Myofibroblasts

A variety of receptor-ligand interactions have been implicated as regulators of fibrotic
matrix turnover via stimulation of HM apoptosis:
Cannabinoids can function as either pro- or anti-fibrogenic molecules depending on
whether they signal via CB1 or CB2 receptors respectively.67,68 CB1 expression is
highly elevated in cirrhotic liver with localization to nonparenchymal cells and is
induced with transdifferentiation of HSC to HM.68 CB1 knockout mice are resistant
to fibrosis induced by CCl4, TAA and bile duct ligation.68 Similar protection was also
provided by the CB1 antagonist SR141716A and this effect (as with CB1 knockout
mice) was associated with reduced numbers of HM. CB1 deficient HM were highly
susceptible to apoptosis; however, this effect was not observed for wild type HM
treated with SR141716, which instead were inhibited for proliferation induced by
platelet derived growth factor (PDGF). CB2 is also strongly induced in HM of fibrotic
liver, but, in this case, gene knockout enhances fibrosis induced by CCl4.67 Stimulation
of CB2 inhibited HM proliferation and induced HM apoptosis via induction of intracel-
lular oxidative stress. Munoz-Luque and colleagues69 recently demonstrated that
long-term in vivo activation of hepatic CB2 receptors stimulated regression of fibrosis
and improved mean arterial pressure in rats with decompensated cirrhosis. Although
the role of CB1 and CB2 as regulators of HM function and fibrogenesis are not in
doubt, it should also be noted that cannabinoids may additionally stimulate HM apo-
ptosis via alternative mechanisms. In this respect, the endocannabinoid 2-arachidonyl
glycerol (1-AG) is highly elevated in the chronic injured liver and appears to be a selec-
tive stimulator of HM (but not hepatocyte) apoptosis via a mechanism requiring mem-
brane cholesterol and mitochondrial reactive oxygen species-mediated caspase
activation.70 In summary, although the cannabinoid system requires further mechanis-
tic dissection, there is already a good case for progressing specific antagonists or
agonists of this system to clinical studies.
Nerve growth factor (NGF) is a member of the mammalian neurotrophin family,
which exert their biological effects via interaction with the high affinity TrkA, TrkB
and TrkC neurotrophin receptors and the low affinity p75NTR that belongs to the tumor
necrosis family (TNF) receptor superfamily. p75NTR carries cytoplasmic death and
Chopper domains that trigger apoptosis; this receptor was first identified as a selective
marker for rodent HM and human HM in cirrhotic liver by Trim and colleagues71 who
also reported the ability of NGF to induce HM apoptosis. NGF is expressed by hepa-
tocytes of the injured liver and will suppress NF-kB activity in HM providing a paracrine
mechanism for stimulating HM apoptosis.72 These observations were recently compli-
cated by the discovery that HSC lacking p75NTR fail to transdifferentiate into HM which
raises the possibility of a profibrogenic as well as antifibrogenic role.73 In addition,
human HM appear to resist NGF-induced apoptosis;43 the therapeutic opportunities
for manipulation of NGF and p75NTR may therefore be limited.
Hepatocyte growth factor (HGF) was originally identified as a potent mitogen for
hepatocytes, but HGF is now recognized as a stimulator of diverse cellular responses
via the c-Met receptor. Several in vivo studies reveal an antifibrogenic role for HGF.
Recombinant HGF treatment of rats with established fibrosis stimulates regression
of fibrotic matrix by 60% compared with saline treated animals.29 In an earlier study,
Ueki and colleagues74 reported similar findings in DEN-induced fibrosis treated with
repeated transfection of the human HGF gene into skeletal muscle; the authors dem-
onstrated regression of established fibrosis in the context of ongoing injury. Similarly,
Ozawa and colleagues achieved regression of severe dimethylnitrosamine (DMN)-
induced fibrosis by adenoviral delivery of HGF.75 In vitro studies have confirmed the
ability of HGF to suppress PDGF-stimulated HM proliferation and promote HM apo-
ptosis, with the latter possibly occurring via a JNK-dependent mechanism.29 HGF
administration in vivo was also associated with elevated rates of HM apoptosis. How-
ever, mechanisms other than HM apoptosis have been proposed as responsible for
the antifibrogenic activities of HGF, including inhibition of TGF-b expression and stim-
ulation of the hepatic recruitment of bone marrow-derived MMP-expressing cells. Of
relevance to a direct effect of HGF on HM, portal fibroblast and HSC-derived HM
express c-Met receptor.29 However, the roles played by the receptor in HM prolifera-
tion/apoptosis and fibrosis regression remain to be determined. Of perhaps more
concern regarding potential for translation to the clinic, HGF and c-Met signaling
may help promote hepatocellular carcinoma, which is already at high risk for develop-
ment in the cirrhotic liver.
The adipocyte-derived adipocytokines, adiponectin and leptin are emerging as
important regulators of hepatic fibrogenesis with apparently opposing roles. Leptin
promotes fibrosis; in contrast, adiponectin is able to suppress HM proliferation,
TGF-b1 expression and stimulate HM apoptosis.76,77 The in vivo importance of adipo-
nectin as a natural occurring antifibrotic was demonstrated by enhanced CCl4-
induced fibrosis in adiponectin knockout mice.77 Furthermore, adenoviral-mediated
over-expression of hepatic adiponectin prevented development of fibrosis and also
stimulated regression of established fibrosis.77 Interestingly, HSC can express
924 Gieling et al
adiponectin in their quiescent state but lose expression upon activation to their HM
phenotype. HSC also express the adiponectin receptors, AdipoR1 and AdipoR2, in
both their quiescent and activated phenotypic states.76 Hence, loss of autocrine
adiponectin signaling during transdifferentiation of HSC to HM is a mechanism for pro-
moting survival of fibrogenic HM and progression of fibrosis. A downstream effector of
adiponectin is adenosine monophosphate-activated kinase (AMPK) which, upon acti-
vation, can mediate antifibrogenic effects in human HM.78 As AMPK activation inhibits
NF-kB activation and RSK phosphorylation, it is tempting to speculate that adiponec-
tin may stimulate HM apoptosis and fibrosis regression by, at least in part, inhibiting
the prosurvival properties of NF-kB and RSK.78
Evidence for immune regulation of liver fibrosis is gathering pace and, in particular,
there is excitement regarding regulatory functions for natural killer (NK) cells, which are
highly enriched in the liver where their primary function is to eliminate virally infected
hepatocytes. NK cells have the ability to directly interact with HM and induce apopto-
sis; by contrast, NK cells do not kill quiescent hepatic stellate cells.79,80 Enhancement
of NK cells’ killing activities may therefore selectively deplete HM and promote regres-
sion of fibrosis. In support of this concept, Horani and colleagues81 report that the
immune modulatory drug glatiramer acetate (Copaxone) can stimulate regression of
established fibrosis; this therapeutic effect is associated with increased numbers of
hepatic NK cells and decreased numbers of aSMA positive HM. Furthermore IFN-g,
which has well documented antifibrotic properties including the ability to stimulate
regression of fibrosis in patients with hepatitis B and hepatitis C viral infections,82,83
stimulates NK cell killing of HSC-derived HM both in vitro and in vivo.79,84 Expression
of NKG2D receptor and the tumor necrosis factor-related apoptosis-inducing ligand
(TRAIL) on NK cells is a requirement for stimulation of HM apoptosis.79 Modulation
of the expression of these molecules or their interacting partner molecules on the sur-
face of HM provides mechanisms that regulate HM apoptosis and fibrogenesis. For
example, the NKG2D ligand retinoic acid early inducible gene 1 (RAE-1) is up-
regulated with HSC transdifferention and sensitizes early activated HM to NK cell-
mediated killing.80 However, RAE-1 expression is lost as HM mature—most likely as
a result of depletion of retinol in these cells—and as a result, mature fibrogenic HM
become resistant to NK cell killing. NK2GD, TRAIL and IFN-g expression on NK cells
are diminished in ethanol-fed mice which are attenuated for NK-mediated killing of HM
and develop more severe CCl4-induced fibrosis than mice in which ethanol was
substituted for dextrin maltose.85 Hence, as fibrogenesis progresses, the natural anti-
fibrogenic properties of NK cells are diminished as a consequence of HM maturation
but can be further attenuated by environmental factors such as ethanol. Experimental
stimulation of NK cell numbers and activity via manipulation of IFN-g signaling,
NK2GD/RAE-1 interactions, and TRAIL/TRAIL-R interactions may offer opportunities
for enhancing fibrosis regression.
Extracellular Matrix and its Regulators as Attenuators of Hepatic

Myofibroblast Apoptosis
The hepatic ECM and its various regulatory factors (eg, MMPs and TIMPs) are directly
involved in the fibrogenic process via their influence on ECM remodeling, and they
may also indirectly influence fibrogenesis by contributing to HM survival mechanisms.
A strong positive association between matrix turnover and HM apoptosis was demon-
strated with studies using mice carrying a mutation in the a1 (I) collagen gene (r/r
collagen) that generates type I collagen which is resistant to cleavage by interstitial
collagenase.30 Regression of liver fibrosis was not observed in these mice over an
extended 28-day period of recovery and this absence of matrix remodelling was
associated with reduced numbers of apoptotic HM. This remarkable observation sug-
gests that intact type I collagen, which is a key component of the fibrotic matrix, acts
as a survival signal for fibrogenic HM. Interactions of pericellular ECM with cells are
largely mediated via the a/b integrin family members, many of which are expressed
on HM. The concept that a/b integrins can regulate the balance between HM prolifer-
ation and apoptosis has been determined with in vitro studies on the a3b2 integrin.86
Disruption of a3b2 integrin with echistatin, neutralizing antibodies, or siRNA knock-
down inhibited HM proliferation and stimulated apoptosis. The latter effect was asso-
ciated with an increased ratio of BAX to Bcl2 and increased caspase-3 activation
suggesting that a3b2 prevents activation of the intrinsic mitochondrial apoptosis path-
way. This idea is further supported by an earlier study in which HM apoptosis was
stimulated by a peptide antogonist of integrins; this effect was accompanied by in-
creased expression of p53 and a reduced Bcl2/Bax ratio.87 Disruption of a3b2 signal-
ing increased the collagenolytic activity of HM via reduction of TIMP-1 expression and
increased synthesis of MMP-9.86 This shift in balance from fibrogenic to fibrolytic phe-
notype of HM would, in effect, serve as a positive feedback mechanism, promoting
further loss of ECM-HM interactions to stimulate apoptosis. Of relevance to the spe-
cific role of a3b2, this integrin binds preferentially to partially degraded and unwound
collagen I, which is a requirement for cell proliferation. HM plated onto nondegradable
r/r collagen proliferated more slowly than HM plated onto wild type collagen.88
In addition to its profibrogenic role as a stimulator of net deposition of fibrotic ECM
(described in more detail elsewhere),89,90 TIMP-1 may also limit fibrosis regression
by promoting HM survival.31,91,92 By contrast, MMPs may contribute to fibrosis
regression indirectly by stimulating apoptosis at least in part via disruption of prosur-
vival interactions between HM and their surrounding matrix.30,31 Transgenic mice
over-expressing hepatic TIMP-1 are resistant to spontaneous regression of CCl4-
induced fibrosis, have reduced hepatic MMP activity relative to controls, and dis-
played reduced numbers of apoptotic non-parenchymal cells.91 In vitro studies
confirmed that TIMP-1 suppresses HM apoptosis via inhibition of MMP activities.31
Based on these discoveries, Roderfeld and colleagues93 recently demonstrated the
potential for an innovative and promising therapeutic strategy in which the highly el-
evated levels of TIMP-1 in the diseased liver are effectively sequestered from endog-
enous MMPs by adenovirus-mediated expression of proteolytically inactive mutant
MMP-9 proteins. Fibrogenesis was inhibited, transdifferentiation of HSC to HM
was suppressed, and HM apoptosis was stimulated. There is, therefore, sufficient
endogenous MMP expression in the diseased liver to stimulate fibrosis regression
providing that inhibitory TIMP-1 can be neutralized. Furthermore, the therapeutic
application of proteolytically inactive MMPs may be preferable to the alternative
proposal of over-expression of proteolytically active MMPs, which would have
potential deleterious effects, such as stimulation of the development and spread
of cancers.94,95
REGULATORS OF FIBROTIC EXTRACELLULAR MATRIX DEPOSITION AND DEGRADATION
The concept that fibrogenesis is driven by increased synthesis of fiber-forming colla-

gens coupled with decreased hepatic activity of collagenolytic enzymes is well estab-
lished and described in detail elsewhere.89 It is also widely acknowledged that
elevated hepatic TIMP-1 expression derived from HM plays a major role in tipping
the balance of ECM turnover toward net deposition via blockade of the activity of
key MMPs.90 In this review, the focus is mainly on recent reports describing the func-
tions of ECM regulators in fibrosis regression.
926 Gieling et al
Profibrogenic Growth Factors

Nonparenchymal liver cells, including HM, produce a variety of growth factors that
stimulate ECM deposition via a combination of autocrine- and paracrine-signaling
pathways. Type b1 transforming growth factor (TGF-b1) and platelet derived growth
factor-BB (PDGF) are the most potent and well studied of the fibrosis promoting
growth factors; recent studies have implicated these two factors as targets for thera-
peutic regression of fibrosis. Expression of receptors for TGF-b and PDGF are induced
with acquisition of the HM phenotype, as is production of both growth factors.4,96,97
Treatment of HM with TGF-b1 stimulates collagen I and TIMP-1 expression, while
transgenic conditional over-expression of hepatic TGF-b1 stimulates collagen synthe-
sis and deposition and, in the absence of liver injury, is sufficient to induce progressive
liver fibrosis.98 These effects were associated with elevated apoptosis and reduced
proliferation of hepatocytes, which act as triggers for the fibrogenic response. In
contrast, fibrogenesis was not only halted by switching off TGF-b1 production, but,
in addition, established fibrosis was reversed with regeneration of normal liver struc-
ture, highlighting the central role played by TGF-b1 in fibrogenesis. BDL-injured rats
were protected from fibrosis by treatment with an adenovirus expressing TGF-b1 an-
tisense mRNA.99 These experimental studies on TGF-b highlight potential for a number
of therapeutic strategies (see review).100 The TGF-b signaling pathway can be
inhibited by agents that block TGF-b function, including: the administration of TGF-b
antibodies, anti-sense oligonucleotides and soluble truncated TGF-b receptor II
(TbRII). Alternatively, over-expression of Smad7 (inhibitory Smad) blocked
TGF-b signal transduction by inhibiting Smad2/3 phosphorylation.100 Currently, phase
I/II trials to study the effect of targeting the TGF-b signaling pathway in patients are
limited. Significantly, CAT-192, a recombinant human antibody that neutralizes
TGF-b1, was studied in cutaneous systemic sclerosis; significant morbidity and mor-
tality were observed with no evidence of a therapeutic effect for CAT-192.101 Detailed
safety and efficiency studies are, therefore, required to improve the therapeutic use of
anti-TGF-b1 therapy.
Induction of PDGF-B and the PDGF receptor are early events in fibrogenesis and
drive HM proliferation as well as stimulating synthesis of ECM molecules. Transgenic
over-expression of PDGF-B induced transdifferentiation of HSCs and liver fibrosis.102
In a similar way PDGF-C, a new PDGF family member, promoted HM activation and
fibrosis when over-expressed in vivo.103 Specific delivery of a PDGF kinase inhibitor,
using the HSC-selective carrier mannose-6-phosphate modified human serum albu-
min (M6PHSA), significantly reduced BDL-induced fibrosis.104 The PDGFs, their com-
mon receptor, and downstream signaling molecules are, therefore, pivotal regulators
of fibrogenesis and may be excellent targets for stimulating fibrosis regression.
Modifying the Balance Between Matrix Mettaloproteinases and Their Main

Inhibitor TIMP-1
During liver injury and fibrogenesis, there is a substantive increase in hepatic expres-
sion of TIMP-1 while expression levels of interstitial collagenases remain constant,
although their collagenolytic activity is diminished due to the repressive properties
of TIMP-1.105 Following cessation of injury and onset of recovery, TIMP-1 levels
rapidly diminish back to pre-injury levels, enabling release of collagenase activities.3
These highly dynamic shifts in the balance of MMP activity and TIMP-1 expression
are widely thought to be a major molecular determinant of the balance between
hepatic fibrogenesis versus fibrinolysis. Hence, strategies that tip the balance toward
elevated MMP activity have potential to promote fibrosis regression. Bueno and
colleagues106 over-expressed human urokinase-type plasminogen (uPA) to promote

hepatic ECM remodeling and fibrosis regression in rats with pre-established CCl4-
induced disease. uPA plays an important role in ECM remodeling by converting plas-
min into plasminogen, which is a powerful and broad-specificity activator of MMPs.89
The precise identity of the MMPs responsible for matrix remodeling and the cellular
sources of collagenolytic activity in the regressing phase of fibrosis are unclear.
Duffield and colleagues107 identified so-called ‘‘scar-associated macrophages’’
(SAMs) as being critical for fibrosis regression in mice; more recently, the same labo-
ratory have shown a functional association between SAM-expressed MMP-13 and
remodeling of fibrotic ECM.108 However, although MMP-13 knockout mice have a
delay in their rate of fibrosis regression, matrix remodeling still takes place quite
effectively in these mice, suggesting a predominant role for a different MMP or, alter-
natively, redundancy for collagenolytic activity between members of the MMP fam-
ily.108 Another potential cellular source of MMPs are bone marrow-derived cells
(BMC) which are recruited to the injured liver and may functionally contribute to matrix
remodeling via the expression of several MMPs with collagenolytic activities including
MMP-9 and MMP-14.109,110 This has led to the proposal for BMC-based therapies for
fibrosis; however, the suggestion that BMC may also be a source of fibrogenic HM
highlights the need for further basic research.10,111 Viral vectors have been exploited
by several groups to demonstrate the antifibrotic potential for MMP gene therapy
approaches. Adenoviral expression of the human interstitial collagenase MMP-1 pro-
moted regression of established fibrosis in rats112 and was also associated with
reduction in numbers of HM. This, again, highlights the potential for MMP-mediated
disruption of matrix–HM interactions to promote HM apoptosis and accelerate fibrosis
regression. Gene delivery of the human neutrophil collagenase, MMP-8, also stimu-
lated regression of established fibrosis.113 Mice treated with MMP-8 also displayed
increased levels of hepatic MMP2 and MMP3 and elevated free TIMP-1, suggesting
the treatment triggers complimentary endogenous fibrolytic mechanisms. Alterna-
tively, increased MMP-8 may stimulate a positive feedback loop as MMPs can acti-
vate subsequent MMPs as shown by in vitro reconstitution experiments.114
Although promising, these MMP gene therapy strategies will require controlled delivery
and considerable safety trials given the potential for MMPs to disrupt the structure and
function of normal hepatic matrix and to promote cancer development. The alternative
approach of therapy with proteolytically-inactive MMP genes as sinks for TIMP-1
described above may be a safer strategy.93 Neutralizing antibodies against human
TIMP-1 may also offer an alternative approach, as there is experimental evidence of
a modest therapeutic effect of anti-TIMP-1 on established fibrosis in rodents.115 Other
approaches may be developed to exploit the activity of natural occurring modulators of
TIMP-1 expression including leptin, opioids and FXR/SHP.64,116,117
Remodeling Vascular Changes in Fibrotic Liver

As noted above, neovascularization contributes to shunting in the fibrotic/cirrhotic
liver in human disease and experimental models of injury.118 Angiogenesis is an inte-
gral part of wound-healing processes, in general. Hypoxia is the main driver, which in
chronic liver disease may be a consequence of impaired microvascular blood flow
through the architectural changes of fibrosis and nodular regenerative activity. This
leads to stimulation of hypoxia-inducible factors, which, in turn, up-regulate a range
of molecules that are pro-fibrogenic.
The molecular control of angiogenesis has been extensively studied (see review).119
The process involves several overlapping phenomena, including: sprouting and
928 Gieling et al
budding of small vessels; endothelial cell migration and proliferation; lumen formation;
and vessel stabilization. VEGF plays a central role, orchestrating a number of these
events. VEGF-C and D are involved in the proliferation and maturation of new lym-
phatic channels during repair; lymphatic angiogenesis is thought to occur in chronic
liver disease.120 Several other cytokines and growth factors stimulate endothelial
cell proliferation including acidic and basic fibroblast growth factors, as well as hepa-
tocyte growth factor.121 Although the molecular control of angiogenesis in liver is likely
to be very similar to that documented for other settings, there is evidence that there
may be liver-specific pro-angiogenic factors;122 this may reflect the unique (or cer-
tainly highly restricted) microcirculatory properties such as the fenestrated nature of
the sinusoidal endothelium and the diverse properties of the surrounding pericytes,
the hepatic stellate cells.123
Resolution of fibrosis in other tissues is accompanied by vascular remodeling. There
are reductions in the levels of pro-angiogenic signals and this is accompanied by
regression of small vessels that involves endothelial cell apoptosis. Angiopoietin 2
appears to play a crucial role, but there are other inhibitors of angiogenesis which con-
tribute to remodeling including thrombospondin 1.119 These events are likely to also
occur during regression of liver fibrosis, although they are less well documented
than that in other model systems. New vessels, including lymphatics, persist in the
cirrhotic liver, contributing to vascular shunting and suggesting that the process
may therefore be incomplete in some situations even when there is cessation of the
original injurious stimulus.
In addition to small vessel changes in active chronic liver disease, there are also
changes in the portal vessels and hepatic veins. These become occluded through
perivascular fibrosis and by intraluminal thrombosis; the latter is particularly seen in
conditions where inflammatory disease affects adjacent structures within portal tracts,
eg, primary sclerosing cholangitis. The changes in larger vessels are associated with
hypoxia-involved larger microcirculatory areas than those associated with for example
peri-sinusoidal fibrosis and may lead to parenchymal extinction. The degree to which
this occurs may be crucial as to whether or not advanced disease reaches a ‘‘point of
no return’’. Strategies to prevent the alterations to portal and hepatic veins may, thus,
turn out to be important for limiting the extent of injury in progressive chronic liver
injury.
With respect to the discussions regarding the possibility that HM apoptosis may be
a bystander effect of ECM remodelling and that stimulation of fibrosis regression by
gliotoxin may operate via alternative mechanisms from induction of HM apoptosis,
Douglass and colleagues recently used C1-3 to target gliotoxin to HM in vivo and
demonstrated selective induction of HM apoptosis that provoked regression of fibro-
sis under conditions of sustained liver injury. This argues against HM apoptosis as
a simple bystander effect of ECM remodelling and provides strong experimental sup-
port that selective induction of HM apoptosis by agents such as gliotoxin and sulfasa-
lazine will effectively stimulate regression of established fibrosis even where liver injury
persists.124
Krizhanovsky and colleagues125 have recently reported that hepatic stellate cells
can undergo cell cycle arrest (senescence) and that this provides a mechanism
for preventing proliferation of myofibroblasts, activating their expression of colla-
gen-degrading MMPs and targets the myofibroblast for clearance by NK cells.
This important study raises the possibility of therapeutic targeting of the activities
of regulators of senescence (eg, p53 and Rb) to provoke remodelling of fibrotic
tissue.
SUMMARY
The increasing awareness that fibrosis is a reversible process, at least in terms of

regression of scar tissue, is paving the way for a new era in hepatology in which ther-
apeutic reversion for fibrosis of mild to moderate severity may soon be offered.
Improved understanding of the cellular and molecular mechanisms that fine-tune
the balance between accumulation of fibril-forming collagens and their degradation
will be key to ushering in this new era. At present, the major mechanisms under study
are the signaling pathways that regulate the survival and proliferation of fibrogenic HM
and the matrix remodeling processes that regulate the equilibrium between fibrogen-
esis and fibrolysis. The current state of knowledge already provides a wealth of poten-
tial molecular targets and related therapeutics to take beyond pre-clinical tests
(Fig. 2). What, then, are the hurdles that remain to be overcome before this knowledge
can be translated into clinical practice? First and foremost, improved and preferably
minimal invasive diagnostic systems for real-time monitoring of progression and
Fig. 2. Potential molecular therapeutic strategies to reverse severe liver fibrosis and cirrhosis
have to deal with three different components (hepatic myofibroblast [HM], excessive ECM
and neovascularization), all major determinants for the deteriorated liver function in
patients with cirrhosis. (A) HM numbers can be effectively reduced by either suppressing
transdifferentiation to HM (first arrow) or supporting apoptosis of HM (second arrow).
(B) Molecular therapies to target the excessive fibrotic ECM will be aimed at tipping the
imbalance between the hepatic fibrogenic and fibrolytic activities toward fibrolysis.
(C) An important component of molecular therapy for cirrhosis must tackle the develop-
ment of intrahepatic vascular shunting which causes a large percentage of intrahepatic
blood to bypass the liver parenchyma. The molecular strategies shown in boxes are proven
to be effective by in vitro and in vivo experimental studies.
930 Gieling et al
regression of fibrosis are required. As discussed in detail elsewhere,46 the ongoing

developments with serum markers of fibrosis, metabolomics, and deep molecular
imaging of tissues will soon provide the analytical tools required for real-time monitor-
ing of the efficacy of cellular and molecular therapies aimed at reversion of fibrosis. For
safe and effective reversion of fibrosis, it is highly likely that cellular and molecular ther-
apeutics will have to be targeted to specific structures (eg, fibrotic septae) or specific
cell types (eg, HM or macrophages). This will be particularly important with apoptosis-
promoting drugs that may be highly beneficial if targeted to HM but be potentially
harmful if applied in a non–cell-selective manner. In a similar way, administration of
therapies that promote proteolysis of fibrotic matrix should be directed selectively
at the liver to prevent disruption of crucial matrix structures in other organs, but the
therapies may need more specialized targeting to fibrotic collagen.
When considering the ultimate challenge of achieving reversion of severe fibrosis/
cirrhosis, the gross architectural and physiological transformations that become
established functional features of the liver must be taken into account. To simply
promote regression of fibrotic ECM alone is unlikely to be sufficient for restoration
of normal liver function and may even do considerable harm. For example, what would
be the consequences of rapid proteolytic clearance of fibrotic ECM on the vascular
shunts that can carry substantive volumes of blood and which may rely on the physical
support of the matrix for structural and functional integrity? Additionally, how would
matrix remodeling influence the new network of smaller vessels that have formed in
conjunction with fibrotic septae and which support residual hepatic functions?
Research should be directed to the often under-appreciated problem of how to
remodel and repair the disrupted vasculature of cirrhotic liver. If there is eventually
development of technologies to encourage proteolytic breakdown of the thick mature
highly cross-linked collagen typical of cirrhosis, then attention should be paid to how
the liver would literally ‘‘fill in the gaps’’ to prevent tissue collapse. A solution may be to
design controlled fibrolytic strategies that allow gradual regression of matrix over
months or years and to perhaps couple this with stem cell-based therapeutics that
stimulate regeneration of normal hepatic tissue.
In summary, the body of knowledge has come a long way from the old dogma that
fibrosis was a process that could at best be halted; now there are useful handles on
the molecular processes that govern fibrosis reversion versus progression. However,
to turn the dream of fibrosis reversion into a clinical reality, yet more discoveries and
advances are needed.
REFERENCES
1. Perez-Tamayo R. Cirrhosis of the liver: a reversible disease? Pathol Annu 1979;

14(Pt 2):183–213.
2. Friedman SL, Bansal MB. Reversal of hepatic fibrosis—fact or fantasy? Hepatol-
ogy 2006;43(2 Suppl 1):S82–8.
sion of metalloproteinase inhibitors. J Clin Invest 1998;102(3):538–49.
6. Parola M, Marra F, Pinzani M. Myofibroblast - like cells and liver fibrogenesis:
emerging concepts in a rapidly moving scenario. Mol Aspects Med 2008;29:
58–66.
7. Cassiman D, Libbrecht L, Desmet V, et al. Hepatic stellate cell/myofibroblast

subpopulations in fibrotic human and rat livers. J Hepatol 2002;36(2):200–9.
fibrosis via epithelial to mesenchymal transition. J Biol Chem 2007;282(32):
23337–47.
9. Robertson H, Kirby JA, Yip WW, et al. Biliary epithelial-mesenchymal transition in
posttransplantation recurrence of primary biliary cirrhosis. Hepatology 2007;
45(4):977–81.
10. Forbes SJ, Russo FP, Rey V, et al. A significant proportion of myofibroblasts are
of bone marrow origin in human liver fibrosis. Gastroenterology 2004;126(4):
955–63.
11. Schuppan D, Ruehl M, Somasundaram R, et al. Matrix as a modulator of hepatic
fibrogenesis. Semin Liver Dis 2001;21(3):351–72.
12. Scheuer PJ, Standish RA, Dhillon AP. Scoring of chronic hepatitis. Clin Liver Dis
2002;6(2):335–47.
hepatitis C. Hepatology 2003;38(6):1449–57.
14. Westin J, Lagging LM, Wejstal R, et al. Interobserver study of liver histopathol-
ogy using the Ishak score in patients with chronic hepatitis C virus infection.
Liver 1999;19(3):183–7.
15. Desmet VJ, Roskams T. Cirrhosis reversal: a duel between dogma and myth.
J Hepatol 2004;40(5):860–7.
16. Burt AD, MacSween RN. Hepatic vein lesions in alcoholic liver disease: retro-
spective biopsy and necropsy study. J Clin Pathol 1986;39(1):63–7.
17. Goodman ZD, Ishak KG. Occlusive venous lesions in alcoholic liver disease.
A study of 200 cases. Gastroenterology 1982;83(4):786–96.
18. Yip WW, Burt AD. Alcoholic liver disease. Semin Diagn Pathol 2006;23(3–4):
149–60.
19. Bortolotti F, Guido M, Cadrobbi P, et al. Spontaneous regression of hepatitis B
virus-associated cirrhosis developed in childhood. Dig Liver Dis 2005;37(12):
964–7.
20. Desmet VJ. Comments on cirrhosis reversal. Dig Liver Dis 2005;37(12):909–16.
21. Massarrat S, Fallahazad V, Kamalian N. Clinical, biochemical and imaging-
verified regression of hepatitis B-induced cirrhosis. Liver Int 2004;24(2):
105–9.
22. Serpaggi J, Carnot F, Nalpas B, et al. Direct and indirect evidence for the revers-
ibility of cirrhosis. Hum Pathol 2006;37(12):1519–26.
23. Falize L, Guillygomarc’h A, Perrin M, et al. Reversibility of hepatic fibrosis in
treated genetic hemochromatosis: a study of 36 cases. Hepatology 2006;
44(2):472–7.
24. Issa R, Zhou X, Constandinou CM, et al. Spontaneous recovery from micronod-
ular cirrhosis: evidence for incomplete resolution associated with matrix cross-
linking. Gastroenterology 2004;126(7):1795–808.
25. Issa R, Williams E, Trim N, et al. Apoptosis of hepatic stellate cells: Involvement
in resolution of biliary fibrosis and regulation by soluble growth factors. Gut
2001;48(4):548–57.
26. Elsharkawy AM, Oakley F, Mann DA. The role and regulation of hepatic stellate
cell apoptosis in reversal of liver fibrosis. Apoptosis 2005;10(5):927–39.
27. Sohara N, Znoyko I, Levy MT, et al. Reversal of activation of human myofibro-
blast-like cells by culture on a basement membrane-like substrate. J Hepatol
2002;37(2):214–21.
932 Gieling et al
28. Hazra S, Miyahara T, Rippe RA, et al. PPAR gamma and hepatic stellate cells.
Comp Hepatol 2004;3(Suppl 1):S7.
29. Kim WH, Matsumoto K, Bessho K, et al. Growth inhibition and apoptosis in liver
myofibroblasts promoted by hepatocyte growth factor leads to resolution from
liver cirrhosis. Am J Pathol 2005;166(4):1017–28.
action of collagenase results in failure of recovery from CCl4-induced liver fibro-
sis, persistence of activated hepatic stellate cells, and diminished hepatocyte
regeneration. FASEB J 2003;17(1):47–9.
31. Murphy FR, Issa R, Zhou X, et al. Inhibition of apoptosis of activated hepatic
stellate cells by tissue inhibitor of metalloproteinase-1 is mediated via effects
on matrix metalloproteinase inhibition: implications for reversibility of liver
fibrosis. J Biol Chem 2002;277(13):11069–76.
32. Wright MC, Issa R, Smart DE, et al. Gliotoxin stimulates the apoptosis of human
and rat hepatic stellate cells and enhances the resolution of liver fibrosis in rats.
33. Dekel R, Zvibel I, Brill S, et al. Gliotoxin ameliorates development of fibrosis and
cirrhosis in a thioacetamide rat model. Dig Dis Sci 2003;48(8):1642–7.
34. Kweon YO, Paik YH, Schnabl B, et al. Gliotoxin-mediated apoptosis of activated
human hepatic stellate cells. J Hepatol 2003;39(1):38–46.
35. Orr JG, Leel V, Cameron GA, et al. Mechanism of action of the antifibrogenic
compound gliotoxin in rat liver cells. Hepatology 2004;40(1):232–42.
36. Habens F, Srinivasan N, Oakley F, et al. Novel sulfasalazine analogues with en-
hanced NF-kB inhibitory and apoptosis promoting activity. Apoptosis 2005;
10(3):481–91.
37. Oakley F, Meso M, Iredale JP, et al. Inhibition of inhibitor of kappaB kinases stim-
ulates hepatic stellate cell apoptosis and accelerated recovery from rat liver
fibrosis. Gastroenterology 2005;128(1):108–20.
38. Hagens WI, Olinga P, Meijer DK, et al. Gliotoxin non-selectively induces apopto-
sis in fibrotic and normal livers. Liver Int 2006;26(2):232–9.
39. Anselmi K, Stolz DB, Nalesnik M, et al. Gliotoxin causes apoptosis and necrosis
of rat Kupffer cells in vitro and in vivo in the absence of oxidative stress: exac-
erbation by caspase and serine protease inhibition. J Hepatol 2007;47(1):
103–13.
40. Elrick LJ, Leel V, Blaylock MG, et al. Generation of a monoclonal human single
chain antibody fragment to hepatic stellate cells–a potential mechanism for tar-
geting liver anti-fibrotic therapeutics. J Hepatol 2005;42(6):888–96.
41. Hagens WI, Beljaars L, Mann DA, et al. Cellular targeting of the apoptosis-induc-
ing compound Gliotoxin to fibrotic rat livers. J Pharmacol Exp Ther 2008;324(3):
902–10.
42. Hagens WI, Mattos A, Greupink R, et al. Targeting 15d-prostaglandin J2 to
hepatic stellate cells: two options evaluated. Pharm Res 2007;24(3):566–74.
43. Novo E, Marra F, Zamara E, et al. Overexpression of Bcl-2 by activated human
hepatic stellate cells: resistance to apoptosis as a mechanism of progressive
hepatic fibrogenesis in humans. Gut 2006;55(8):1174–82.
44. Rockey DC. Hepatic fibrosis, stellate cells, and portal hypertension. Clin Liver
Dis 2006;10(3):459–79.
2007;170(6):1942–53.
46. Muddu AK, Guha IN, Elsharkawy AM, et al. Resolving fibrosis in the diseased
liver: translating the scientific promise to the clinic. Int J Biochem Cell Biol
2007;39(4):695–714.
47. Oakley F, Mann J, Ruddell RG, et al. Basal expression of IkappaBalpha is con-
trolled by the mammalian transcriptional repressor RBP-J (CBF1) and its activa-
tor Notch1. J Biol Chem 2003;278(27):24359–70.
48. Mann J, Oakley F, Akiboye F, et al. Regulation of myofibroblast transdifferentia-
tion by DNA methylation and MeCP2: implications for wound healing and fibro-
genesis. Cell Death Differ 2007;14(2):275–85.
49. Watson MR, Wallace K, Gieling RG, et al. NF-kB is a critical regulator of the
survival of rodent and human hepatic myofibroblasts. J Hepatol 2008;48(4):
589–97.
50. Lei K, Davis RJ. JNK phosphorylation of Bim-related members of the Bcl2 family
induces Bax-dependent apoptosis. Proc Natl Acad Sci U S A 2003;100(5):2432–7.
51. Jones EV, Dickman MJ, Whitmarsh AJ. Regulation of p73-mediated apoptosis by
c-Jun N-terminal kinase. Biochem J 2007;405(3):617–23.
52. Oleinik NV, Krupenko NI, Krupenko SA. Cooperation between JNK1 and JNK2 in
activation of p53 apoptotic pathway. Oncogene 2007;26(51):7222–30.
53. Park EJ, Zhao YZ, Kim YC, et al. Bakuchiol-induced caspase-3-dependent
apoptosis occurs through c-Jun NH2-terminal kinase-mediated mitochondrial
translocation of Bax in rat liver myofibroblasts. Eur J Pharmacol 2007;
559(2–3):115–23.
54. Schuler M, Green DR. Mechanisms of p53-dependent apoptosis. Biochem Soc
Trans 2001;29(Pt 6):684–8.
55. Muriel P, Fernandez-Martinez E, Perez-Alvarez V, et al. Thalidomide ameliorates
carbon tetrachloride induced cirrhosis in the rat. Eur J Gastroenterol Hepatol
2003;15(9):951–7.
56. Yeh TS, Ho YP, Huang SF, et al. Thalidomide salvages lethal hepatic necroinflam-
mation and accelerates recovery from cirrhosis in rats. J Hepatol 2004;41(4):
606–12.
57. Fernandez-Martinez E, Morales-Rios MS, Perez-Alvarez V, et al. Effects of thalid-
omide and 3-phthalimido-3-(3,4-dimethoxyphenyl)-propanamide on bile duct
obstruction-induced cirrhosis in the rat. Drug Dev Res 2001;54(4):209–18.
58. Chong LW, Hsu YC, Chiu YT, et al. Anti-fibrotic effects of thalidomide on hepatic
stellate cells and dimethylnitrosamine-intoxicated rats. J Biomed Sci 2006;13(3):
403–18.
59. Son G, Iimuro Y, Seki E, et al. Selective inactivation of NF-kappaB in the liver
using NF-kappaB decoy suppresses CCl4-induced liver injury and fibrosis.
Am J Physiol Gastrointest Liver Physiol 2007;293(3):G631–9.
60. Smart DE, Vincent KJ, Arthur MJ, et al. JunD regulates transcription of the tissue
inhibitor of metalloproteinases-1 and interleukin-6 genes in activated hepatic
stellate cells. J Biol Chem 2001;276(26):24414–21.
61. Smart DE, Green K, Oakley F, et al. JunD is a profibrogenic transcription factor
regulated by Jun N-terminal kinase-independent phosphorylation. Hepatology
2006;44(6):1432–40.
62. Parks DJ, Blanchard SG, Bledsoe RK, et al. Bile acids: natural ligands for an
orphan nuclear receptor. Science 1999;284(5418):1365–8.
63. Fiorucci S, Antonelli E, Rizzo G, et al. The nuclear receptor SHP mediates inhi-
bition of hepatic stellate cells by FXR and protects against liver fibrosis. Gastro-
enterology 2004;127(5):1497–512.
934 Gieling et al
64. Fiorucci S, Rizzo G, Antonelli E, et al. A farnesoid x receptor-small heterodimer

partner regulatory cascade modulates tissue metalloproteinase inhibitor-1 and
matrix metalloprotease expression in hepatic stellate cells and promotes resolu-
tion of liver fibrosis. J Pharmacol Exp Ther 2005;314(2):584–95.
65. Buck M, Chojkier M. C/EBPbeta phosphorylation rescues macrophage dysfunc-
tion and apoptosis induced by anthrax lethal toxin. Am J Physiol, Cell Physiol
2007;293(6):C1788–96.
66. Buck M, Poli V, Hunter T, et al. C/EBPbeta phosphorylation by RSK creates
a functional XEXD caspase inhibitory box critical for cell survival. Mol Cell
2001;8(4):807–16.
receptor CB2 in the liver. Gastroenterology 2005;128(3):742–55.
68. Teixeira-Clerc F, Julien B, Grenard P, et al. CB1 cannabinoid receptor antago-
nism: a new strategy for the treatment of liver fibrosis. Nat Med 2006;12(6):
671–6.
69. Munoz-Luque J, Ros J, Fernandez-Varo G, et al. Regression of fibrosis after
chronic stimulation of cannabinoid Cb2 receptor in cirrhotic rats. J Pharmacol
Exp Ther 2008;324(2):475–83.
70. Siegmund SV, Qian T, de Minicis S, et al. The endocannabinoid 2-arachidonoyl
glycerol induces death of hepatic stellate cells via mitochondrial reactive oxygen
species. Faseb J 2007;21(11):2798–806.
71. Trim N, Morgan S, Evans M, et al. Hepatic stellate cells express the low affinity
nerve growth factor receptor p75 and undergo apoptosis in response to nerve
growth factor stimulation. Am J Pathol 2000;156(4):1235–43.
72. Oakley F, Trim N, Constandinou CM, et al. Hepatocytes express nerve growth
factor during liver injury: evidence for paracrine regulation of hepatic stellate
cell apoptosis. Am J Pathol 2003;163(5):1849–58.
73. Passino MA, Adams RA, Sikorski SL, et al. Regulation of hepatic stellate cell
differentiation by the neurotrophin receptor p75NTR. Science 2007;315(5820):
1853–6.
74. Ueki T, Kaneda Y, Tsutsui H, et al. Hepatocyte growth factor gene therapy of liver
cirrhosis in rats. Nat Med 1999;5(2):226–30.
75. Ozawa S, Uchiyama K, Nakamori M, et al. Combination gene therapy of HGF
and truncated type II TGF-beta receptor for rat liver cirrhosis after partial hepa-
tectomy. Surgery 2006;139(4):563–73.
76. Ding X, Saxena NK, Lin S, et al. The roles of leptin and adiponectin: a novel
paradigm in adipocytokine regulation of liver fibrosis and stellate cell biology.
Am J Pathol 2005;166(6):1655–69.
77. Kamada Y, Tamura S, Kiso S, et al. Enhanced carbon tetrachloride-induced
liver fibrosis in mice lacking adiponectin. Gastroenterology 2003;125(6):
1796–807.
protein kinase modulates the activated phenotype of hepatic stellate cells.
Hepatology 2008;47(2):668–76.
2006;130(2):435–52.
80. Radaeva S, Wang L, Radaev S, et al. Retinoic acid signaling sensitizes hepatic
stellate cells to NK cell killing via upregulation of NK cell activating ligand RAE1.
Am J Physiol Gastrointest Liver Physiol 2007;293(4):G809–16.
81. Horani A, Muhanna N, Pappo O, et al. Beneficial effect of glatiramer acetate

(Copaxone) on immune modulation of experimental hepatic fibrosis. Am J Phys-
iol Gastrointest Liver Physiol 2007;292(2):G628–38.
82. Weng HL, Wang BE, Jia JD, et al. Effect of interferon-gamma on hepatic fibrosis
in chronic hepatitis B virus infection: a randomized controlled study. Clin Gastro-
enterol Hepatol 2005;3(8):819–28.
83. Muir AJ, Sylvestre PB, Rockey DC. Interferon gamma-1b for the treatment of
fibrosis in chronic hepatitis C infection. J Viral Hepat 2006;13(5):322–8.
84. Jeong WI, Park O, Radaeva S, et al. STAT1 inhibits liver fibrosis in mice by inhib-
iting stellate cell proliferation and stimulating NK cell cytotoxicity. Hepatology
2006;44(6):1441–51.
85. Jeong WI, Park O, Gao B. Abrogation of the antifibrotic effects of natural killer
cells/interferon-gamma contributes to alcohol acceleration of liver fibrosis.
86. Zhou X, Murphy FR, Gehdu N, et al. Engagement of alphavbeta3 integrin regu-
lates proliferation and apoptosis of hepatic stellate cells. J Biol Chem 2004;
279(23):23996–4006.
87. Iwamoto H, Sakai H, Tada S, et al. Induction of apoptosis in rat hepatic stellate
cells by disruption of integrin-mediated cell adhesion. J Lab Clin Med 1999;
134(1):83–9.
88. Zhou X, Jamil A, Nash A, et al. Impaired proteolysis of collagen I inhibits prolif-
eration of hepatic stellate cells: implications for regulation of liver fibrosis. J Biol
Chem 2006;281(52):39757–65.
89. Benyon RC, Arthur MJ. Extracellular matrix degradation and the role of hepatic
stellate cells. Semin Liver Dis 2001;21(3):373–84.
90. Roderfeld M, Hemmann S, Roeb E. Mechanisms of fibrinolysis in chronic liver
injury (with special emphasis on MMPs and TIMPs). Z Gastroenterol 2007;
45(1):25–33.
91. Yoshiji H, Kuriyama S, Yoshii J, et al. Tissue inhibitor of metalloproteinases-1
attenuates spontaneous liver fibrosis resolution in the transgenic mouse. Hepa-
tology 2002;36(4 Pt 1):850–60.
92. Murphy F, Waung J, Collins J, et al. N-Cadherin cleavage during activated
hepatic stellate cell apoptosis is inhibited by tissue inhibitor of metalloprotei-
nase-1. Comp Hepatol 2004;3(Suppl 1):S8.
93. Roderfeld M, Weiskirchen R, Wagner S, et al. Inhibition of hepatic fibrogenesis
by matrix metalloproteinase-9 mutants in mice. FASEB J 2006;20(3):444–54.
94. Sakamoto Y, Mafune K, Mori M, et al. Overexpression of MMP-9 correlates with
growth of small hepatocellular carcinoma. Int J Oncol 2000;17(2):237–43.
95. Määttä M, Soini Y, Liakka A, et al. Differential expression of matrix metalloprotei-
nase (MMP)-2, MMP-9, and membrane type 1-MMP in hepatocellular and pan-
creatic adenocarcinoma: implications for tumor progression and clinical
prognosis. Clin Cancer Res 2000;6:2726–34.
96. Gressner AM, Weiskirchen R. Modern pathogenetic concepts of liver fibrosis
suggest stellate cells and TGF-beta as major players and therapeutic targets.
J Cell Mol Med 2006;10(1):76–99.
Am J Pathol 1996;148(3):785–800.
98. Ueberham E, Low R, Ueberham U, et al. Conditional tetracycline-regulated
expression of TGF-beta1 in liver of transgenic mice leads to reversible interme-
diary fibrosis. Hepatology 2003;37(5):1067–78.
936 Gieling et al
99. Arias M, Sauer-Lehnen S, Treptau J, et al. Adenoviral expression of a transform-

ing growth factor-beta1 antisense mRNA is effective in preventing liver fibrosis in
bile-duct ligated rats. BMC Gastroenterol 2003;3:29.
100. Liu X, Hu H, Yin JQ. Therapeutic strategies against TGF-beta signaling pathway
in hepatic fibrosis. Liver Int 2006;26(1):8–22.
101. Denton CP, Merkel PA, Furst DE, et al. Recombinant human anti-transforming
growth factor beta1 antibody therapy in systemic sclerosis: a multicenter,
randomized, placebo-controlled phase I/II trial of CAT-192. Arthritis Rheum
2007;56(1):323–33.
102. Czochra P, Klopcic B, Meyer E, et al. Liver fibrosis induced by hepatic overex-
pression of PDGF-B in transgenic mice. J Hepatol 2006;45(3):419–28.
103. Campbell JS, Hughes SD, Gilbertson DG, et al. Platelet-derived growth factor C
induces liver fibrosis, steatosis, and hepatocellular carcinoma. Proc Natl Acad
Sci U S A 2005;102(9):3389–94.
104. Gonzalo T, Beljaars L, van de Bovenkamp M, et al. Local inhibition of liver fibro-
sis by specific delivery of a platelet-derived growth factor kinase inhibitor to
hepatic stellate cells. J Pharmacol Exp Ther 2007;321(3):856–65.
105. Iredale JP, Benyon RC, Arthur MJ, et al. Tissue inhibitor of metalloproteinase-1
messenger RNA expression is enhanced relative to interstitial collagenase mes-
senger RNA in experimental liver injury and fibrosis. Hepatology 1996;24(1):
176–84.
106. Bueno M, Salgado S, Beas-Zarate C, et al. Urokinase-type plasminogen activa-
tor gene therapy in liver cirrhosis is mediated by collagens gene expression
down-regulation and up-regulation of MMPs, HGF and VEGF. J Gene Med
2006;8(11):1291–9.
107. Duffield JS, Forbes SJ, Constandinou CM, et al. Selective depletion of macro-
phages reveals distinct, opposing roles during liver injury and repair. J Clin
Invest 2005;115(1):56–65.
108. Fallowfield JA, Mizuno M, Kendall TJ, et al. Scar-associated macrophages are
a major source of hepatic matrix metalloproteinase-13 and facilitate the resolu-
tion of murine hepatic fibrosis. J Immunol 2007;178(8):5288–95.
109. Sakaida I, Terai S, Yamamoto N, et al. Transplantation of bone marrow cells
reduces CCl4-induced liver fibrosis in mice. Hepatology 2004;40(6):1304–11.
110. Higashiyama R, Inagaki Y, Hong YY, et al. Bone marrow-derived cells express
matrix metalloproteinases and contribute to regression of liver fibrosis in mice.
Hepatology 2007;45(1):213–22.
to liver fibrosis. Gastroenterology 2006;130(6):1807–21.
112. Iimuro Y, Nishio T, Morimoto T, et al. Delivery of matrix metalloproteinase-1 atten-
uates established liver fibrosis in the rat. Gastroenterology 2003;124(2):445–58.
113. Siller-Lopez F, Sandoval A, Salgado S, et al. Treatment with human metallopro-
teinase-8 gene delivery ameliorates experimental rat liver cirrhosis. Gastroenter-
ology 2004;126(4):1122–33.
114. Han YP, Yan C, Zhou L, et al. A matrix metalloproteinase-9 activation cascade by
hepatic stellate cells in trans-differentiation in the three-dimensional extracellular
matrix. J Biol Chem 2007;282(17):12928–39.
115. Parsons CJ, Bradford BU, Pan CQ, et al. Antifibrotic effects of a tissue inhibitor
of metalloproteinase-1 antibody on established liver fibrosis in rats. Hepatology
2004;40(5):1106–15.
116. Cao Q, Mak KM, Lieber CS. DLPC and SAMe combined prevent leptin-stimu-
lated TIMP-1 production in LX-2 human hepatic stellate cells by inhibiting HO-
mediated signal transduction. Liver Int 2006;26(2):221–31.
117. Ebrahimkhani MR, Kiani S, Oakley F, et al. Naltrexone, an opioid receptor antag-
onist, attenuates liver fibrosis in bile duct ligated rats. Gut 2006;55(11):1606–16.
118. Corpechot C, Barbu V, Wendum D, et al. Hypoxia-induced VEGF and collagen I
cirrhosis. Hepatology 2002;35(5):1010–21.
119. Medina J, Arroyo AG, Sanchez-Madrid F, et al. Angiogenesis in chronic inflam-
matory liver disease. Hepatology 2004;39(5):1185–95.
120. Yamauchi Y, Ikeda R, Michitaka K, et al. Morphometric analysis of lymphatic ves-
sels in primary biliary cirrhosis. Hepatol Res 2002;24(2):107.
121. Carmeliet P. Angiogenesis in health and disease. Nat Med 2003;9(6):653–60.
122. Camenisch G, Pisabarro MT, Sherman D, et al. ANGPTL3 stimulates endothelial
cell adhesion and migration via integrin alpha vbeta 3 and induces blood vessel
formation in vivo. J Biol Chem 2002;277(19):17281–90.
123. Iredale JP. Regulating hepatic inflammation: pathogen-associated molecular
patterns take their toll. Hepatology 2003;37(5):979–82.
124. Douglass A, Wallace K, Parr R, et al. Antibody-targeted myofibroblast apoptosis
reduces fibrosis during sustained liver injury. J Hepatol 2008;49(1):88–98.
125. Krizhanovsky V, Yon M, Dickins RA, et al. Senescence of activated stellate cells
limits liver fibrosis. Cell 2008;134(4):657–67.
Current a nd Future
Anti - Fibrotic Therapies
for Chronic Liver
Dis eas e
Don C. Rockey, MD
KEYWORDS
Fibrosis Cirrhosis Stellate cell Extracellular matrix
Myofibroblast Liver biopsy
Complication Portal hypertension
Advances in the understanding of the cellular and molecular basis of hepatic fibrogen-
esis over the past 2 decades have allowed the emergence of a field dedicated to anti-
fibrotic therapy. The liver responds to injury with wound healing and, subsequently,
fibrosis. This response occurs after essentially all kinds of injury (eg, from virus or
alcohol) and ultimately leads to cirrhosis in some patients. The observation that any
of several types of liver diseases and their injury result in cirrhosis suggests a common
pathogenesis.
Experts now recognize that a population or populations of effector cells play a crit-
ical role in the fibrogenic process. A classic effector cell, the hepatic stellate cell, is one
of the most important fibrogenic cells in the liver. This cell undergoes a transformation
during injury, termed activation. The activation process is complex, but one of its most
prominent features is the synthesis of large amounts of extracellular matrix (ECM), re-
sulting in deposition of scar or fibrous tissue. Thus, the hepatic stellate cell or other
fibrogenic cell types have been a therapeutic target.
The fibrogenic process is dynamic, and even advanced fibrosis is reversible. The
best antifibrotic therapy is elimination of the underlying disease process. For example,
elimination of hepatitis B and hepatitis C virus (HBC and HCV, respectively) can lead to
reversal of fibrosis. In situations in which treating the underlying process is not possi-
ble, specific antifibrotic therapy would be highly desirable. Many specific antifibrotic
treatments have been tested, but none have succeeded. Nonetheless, because of
the importance of fibrosis, the field of antifibrotic compounds is rapidly growing.
This article emphasizes mechanisms underlying fibrogenesis as they relate to putative
antifibrotic therapy and reviews current and potential future antifibrotic therapies.
Division of Digestive and Liver Diseases, Department of Internal Medicine, The University
of Texas, Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390, USA
E-mail address: don.rockey@utsouthwestern.edu
Clin Liver Dis 12 (2008) 939–962

940 Rockey
FIBROGENESIS: PATHOPHYSIOLOGY
The Fibrogenic Process
A fundamental concept is that although the wounding process is complicated, it is
characterized by common features, including increased production of extracellular
matrix as a result of a coordinated response that includes the action of various events
on effector cells, which then lead to ECM synthesis. In the liver and in most organs,
inflammation often drives the response. Excellent examples include HBV and HCV in-
fection, autoimmune hepatitis, and alcoholic hepatitis. The chronicity of inflammation,
the type of inflammation (ie, Th2 vs. Th1), and the interplay of inflammation with envi-
ronmental/metabolic/genetic factors are often important in many types of liver
disease.
The effectors of the fibrogenic response in the liver are diverse and include different
cell types, including activated stellate cells, periportal and pericentral fibroblasts, my-
ofibroblasts (which may be derived from all 3 of the above cell types), bone marrow
derived cells, fibroblasts derived from epithelial cells, and even bile duct epithelial
and endothelial cells.1–8 Considerable attention has focused on hepatic stellate cells,
which transform from a quiescent (normal) to an activated (injured liver) state (Fig. 1).
(See the article by Wells elsewhere in this issue.) Although straightforward in concept,
the activation process is remarkably complex and consists of many important cellular
changes. Characteristic features of this transition include loss of vitamin A, acquisition
of stress fibers, and development of prominent rough endoplasmic reticulum.
Because the stellate cell has been identified as a key effector of the fibrogenic re-
sponse, one of the most prominent features of activation is a striking increase in se-
cretion of ECM proteins, including types I, III, and IV collagens; fibronectin; laminin;
and proteoglycans. Some ECM molecules are increased by greater than 50-fold, con-
sistent with the conclusion that stellate cells are the cellular source of the enhanced
ECM production at the whole organ level.9 A further critical feature of activation is
Fig. 1. Stellate cell activation. The current consensus is that the key pathogenic feature un-
derlying liver fibrosis and cirrhosis is activation of hepatic stellate cells. This process is com-
plex, both in terms of the events that induce activation and the effects of activation.
Multiple and varied stimuli participate in the induction and maintenance of activation, in-
cluding cytokines, peptides, and the ECM. Key phenotypic features of activation include
production of ECM, loss of retinoids, proliferation, up-regulation of smooth muscle pro-
teins, secretion of peptides and cytokines (which have autocrine effects), and up-regulation
of various cytokine and peptide receptors.179 Other effector cells (fibroblasts, fibrocytes,
bone marrow derived-cells) probably undergo similar activation and also contribute to
the fibrogenic response. (From Rockey DC. Antifibrotic therapy in chronic liver disease.
Clin Gastroenterol Hepatol 2005;3:95; with permission.)
Current and Future Anti-Fibrotic Therapies 941
de novo expression of smooth muscle–specific proteins, such as smooth muscle

a-actin.10 This feature identifies activated stellate cells as liver-specific myofibroblasts.
The field of stellate cell biology has exploded over the past 20 years, and a review of
this area can be found elsewhere in this issue in the article by Wells. The science has
led to multiple therapeutic approaches based on an understanding of this cell’s biol-
ogy. For example, many pathways lead specifically to stellate cell fibrogenesis, and
these have been or can be targeted. Thus, the complexity of the wounding response
allows for multiple different therapeutic interventions, including those based on stel-
late cell biology, but also based on other mechanisms active in the wounding milieu
(Table 1). Therefore, more than one antifibrotic agent may be prescribed or an antifi-
brotic agent may be taken along with another agent that has a different mechanism of
action (eg, anti-inflammatory, antioxidant).
Pathophysiology of the Fibrogenic Process and Considerations for Therapy

When considering antifibrotic therapy, it is important to recognize that fibrosis is a dy-
namic process. Although the fibrotic lesion was once believed to be static, abundant
evidence indicates otherwise. A prominent feature of liver fibrosis is ECM turnover, in-
cluding not only synthesis but also degradation.11 During fibrosis progression, expres-
sion of matrix metalloproteinases (MMPs) and their tissue inhibitors (TIMPs) increases.
Furthermore, and imbalanced seems to be present between MMPs that degrade
‘‘good’’ or normal matrix and those that degrade ‘‘bad’’ or abnormal matrix. Early in
the process, MMPs seem to degrade normal matrix proteins, and this itself may per-
petuate the fibrogenic phenotype of effector cells.11–16 In advanced fibrosis, overex-
pression of MMP8 was shown to lead to partial reversal of fibrosis, providing proof
of a therapeutic role for overexpressing MMPs.17,18
An enormous body of literature in animal models13,18–21 and a surprisingly robust
amount of data in human liver disease emphasize that fibrosis is reversible. The
data are primarily from treatment studies in which the disease was removed or elim-
inated (Table 2). For example, eradication or inhibition of HBV22–24 or HCV25,26 leads
to reversion of fibrosis, even in some patients who have histologic cirrhosis. Addition-
ally, fibrosis (and cirrhosis) is reversible in patients who have autoimmune hepatitis
who respond to medical treatment (prednisone or equivalent).27 Fibrosis may improve
Table 1
Diseases in which fibrosis can be reduced by treating the underlying disorder
Disease Therapy
Hepatitis B Lamivudine, others
Hepatitis C Interferon alphaa
Autoimmune hepatitis Corticosteroids
Bile duct obstruction Surgical decompression
Hemochromatosis Iron depletion
Alcoholic hepatitisb Corticosteroids
Primary biliary cirrhosisb Ursodeoxycholic acid, MTX
Non-alcholic steatohepatitisc PPAR gamma ligands
Abbreviations: MTX, methotrexate; PPAR, peroxisomal proliferator activated receptor

a
or PEG-interferon alpha, with or without ribavirin
b
The effect is minimal if present
c
Evidence is preliminary at this point
Data from Rockey DC. Antifibrotic therapy in chronic liver disease. Clin Gastroenterol Hepatol
2005;3:95–107.
942 Rockey
Table 2
Potential therapeutic approaches for patients with hepatic fibrosis
Eliminate or treat the underlying disease process
Block inflammation (if present and if driving the fibrotic response)
Eliminate effector cells
Block or inhibit effector cell function (i.e. fibrogenesis, proliferation, cell contraction,
motility)
Other
in patients who have alcoholic liver disease whose disease responds to anti-inflamma-
tory therapy such as corticosteroids.28,29 Fibrosis reverts in patients who have hemo-
chromatosis during iron depletion30–32 and after relief of bile duct obstruction.33
Finally, in patients who have nonalcoholic steatohepatitis (NASH) treated with the per-
oxisomal proliferator–active receptor g (PPARg) agonist, rosiglitazone reduced both
steatosis and fibrosis.34 Based on our current understanding of the pathophysiology
of fibrosis, there are several potential approaches to treatment (see Table 1).
APPROACH TO THERAPY FOR FIBROSIS

Monitoring of Hepatic Fibrosis
Currently, one of the major challenges in the field of fibrosis therapy is to monitor fibro-
sis. Emerging evidence suggests that the presence of fibrosis has important prognos-
tic implications. For example, in patients who had HCV infection after liver
transplantation, adverse clinical events seemed to be increased in those who had
the greatest degree of fibrosis.35 Also, progression of nonalcoholic fatty liver disease
and even liver-related mortality also seemed to be related to initial fibrosis stage.36
Additionally, in patients who have many different types of liver disease, histologic
grade and stage may be helpful in identifying those who should undergo therapy
and those who should not, particularly for patients who have HCV infection.37,38 None-
theless, use of fibrosis data in treatment algorithms for patients who have HCV re-
mains controversial. Patients who have advanced fibrosis (eg, Batts and Ludwig,
stages 3 and/or 4) may be less likely to respond to antiviral therapy than those who
have less advanced fibrosis, and it is well appreciated that patients who have ad-
vanced fibrosis are more likely to experience greater side effects and often require
more aggressive supportive measures (eg, growth factors) to maintain adequate blood
cell counts during treatment. Furthermore, patients who have advanced fibrosis may
have poorer response rates, and should probably be managed differently (eg, treated
with gradual increments in drug doses or for longer periods) than patients who have
absent or mild fibrosis. Experts have also proposed that patients who have absent
or minimal fibrosis should simply be observed and perhaps undergo periodic liver bi-
opsy for follow-up staging.
Liver biopsy and histologic analysis of the liver has long been considered the gold
standard for determining the extent of fibrosis and assessing fibrosis progression.
Qualitative assessment of fibrosis has been made simply through the widespread
use of connective tissue stains, such as reticulin, Masson’s trichrome, and picrosirius
red (which each readily identify ECM within tissue). Quantitative measure of collagen
content can be performed using colorimetric assay of sirius red in liver tissue or image
analytic quantitation of collagen containing tissue.39 Additionally, scoring systems can
quantitate fibrosis40–42 and help standardize the interpretation of biopsies among dif-
ferent centers. These systems are most useful for standardization and comparison of
fibrosis in large studies. In single patients, simple inspection of biopsies over time is
often the most helpful.
Although histologic analysis of the liver has been traditionally considered the gold
standard tool to assess fibrosis, it is not perfect. First, liver biopsy is associated
with significant potential morbidity, including a finite risk for death.43 Additionally, it
is subject to interobserver variability, and sampling error may be important, as evi-
denced by studies examining samples from different regions of the liver.44,45 There-
fore, noninvasive tools to measure fibrosis would be ideal.46 Noninvasive methods
used to assess fibrosis include routine clinical parameters, such as physical examina-
tion findings, laboratory tests,47,48 radiographic tests,49 combinations of laboratory
tests,48,50 and specific serum markers.46,51,52 Serum marker panels, including those
that use mathematical algorithms,48,50 were recently emphasized.
Most recently, transient elastography, an ultrasound-based technology, has gained
considerable attention.53 This examination involves acquisition of pulse–echo ultra-
sound signals to measure liver stiffness54 following the simple placement of an ultra-
sound transducer probe between two ribs, over the right lobe of the liver. The probe
transmits a low amplitude (vibration and frequency) signal to the liver, which induces
an elastic shear wave that propagates through the liver. This pulse–echo ultrasound
measurement provides a measure of liver stiffness (reported in kilopascals).
In addition to its noninvasive nature, this technique has the advantage of allowing
stiffness to be measured across a large area of the liver (1–2 cm), which is at least
100 times greater than with a liver biopsy. Normal liver stiffness is reported to range
from 4 to 6 kPa, whereas cirrhosis is generally present at levels above 12 to 14 kPa.
The higher the level, the more likely the patient has cirrhosis.55–57
Overview of Treatment
Preclinical studies have reported a scientific rationale and experimental evidence sup-
porting the use of many potential therapies for fibrosis. These therapies have been tar-
geted to any of several different biologic targets (eg, inhibition of collagen synthesis,
interruption of matrix deposition, stimulation of matrix degradation, modulation of stel-
late cell activation, induction of stellate cell death). In general, these therapies have
been highly effective in animal models. Several of these preclinical approaches
have been transitioned to clinical trials in humans, which are highlighted below and
in Tables 1 and 3. Therapies have been divided into those that specifically target fibro-
sis (see Table 3) and those that target a more general component of the liver disease
Table 3
Potential antifibrotic therapies with ‘‘specific’’ effects
Agent Disease Efficacy Safety Comments

Colchicine Misc 1/ 1111 Inhibits collagen synthesis
Interferon gamma HCV 1/ 11 Stellate cell–specific effects
ARBs Misc 11 Stellate cell-specific effects
PPAR ligands NASH 11 11 Stellate cell–specific effects
Pirfenidone Misc 1/ 111 Mechanism through cytokines?
The scale for efficacy and safety is to 1111, with – being the lowest rating and 1111 the
highest rating. The ratings are empiric based on the aggregate available literature. See text for
references and discussion of mechanisms.
Abbreviations: ARBs, angiotensin receptor blockers; HCV, hepatitis C virus; Misc, miscellaneous;
NASH, nonalcoholic steatohepatitis; PPAR, peroxisomal proliferator–activated receptor.
944 Rockey
process (ie, oxidative stress) (see Table 3). The following sections highlight the major
antifibrotic agents that have been examined in clinical studies in humans.
Specific Antifibrotic Therapies (Studied in Human Subjects)

Angiotensin II antagonists
The angiotensin II system represents an extremely attractive antifibrotic target. Abun-
dant experimental evidence indicates overproduction of angiotensin II in the injured
liver, and a role of angiotensin II in stimulation of stellate cell activation and fibrogen-
esis.58,59 Several studies have also shown specific antifibrotic effects of angiotensin II
inhibition in various animal models.60–62 Angiotensin II may also play a role in the path-
ogenesis of portal hypertension,63 and thus its inhibition could potentially abrogate not
only fibrosis but also portal hypertension.
Several human studies have examined the effects of angiotensin receptor blockers
in humans,64–68 most in the setting of advanced liver disease and most often in an at-
tempt to reduce portal pressure. A 6-week trial of losartan in 25 patients did not sig-
nificantly reduce hepatic venous pressure gradient (HVPG) compared with propanolol
in patients who had cirrhosis treated after a variceal bleeding episode.66 In a random-
ized trial of 36 patients who had cirrhosis and portal hypertension, irbesartan reduced
the hepatic venous pressure gradient by 12.2% 6.6% after 7 days but also induced
arterial hypotension.67
In a trial of 39 patients who had cirrhosis randomized to losartan (n 5 19) or propran-
olol (n 5 20), HVPG was measured at baseline and on day 14 of therapy.64 A 20% or
greater reduction in HVPG occurred in 15 of 19 (79%) patients treated with losartan,
compared with 9 of 20 (45%) treated with propranolol (P<.05). Although the hepatic
venous pressure gradient reduction (ie, percentage from baseline) with losartan
(27% 22%) was higher than with propranolol (15% 32%), the difference was
not significant.
In another study, 47 compensated Child A and Child B (8) cirrhotic patients were ran-
domly assigned to receive candesartan (8 mg/d; n 5 24) and no treatment (n 5 23) for
1 year. In patients treated with candesartan, the HVPG decreased significantly ( 8.4%
2.4 mm Hg) and 25% of patients experienced a reduction greater than 20%, com-
pared with an increase of 15.6% 2.9 mm Hg in the untreated group. Plasma hyalur-
onic acid levels were also significantly reduced in patients treated with candesartan in
whom HVPG diminished, whereas they rose in untreated patients in whom HVPG
increased.65
The data in humans are therefore mixed, and further studies have been performed in
small numbers of patients. Given the particularly supportive preclinical data, evidence
suggests that the angiotensin receptor blockers (or perhaps angiotensin-conver-
ting enzyme inhibitors) have some antifibrotic effect in humans. Larger and longer
studies seem to be warranted, several of which have recently been closed or are cur-
rently underway (http://clinicaltrials.gov/;ClinicalTrials.gov Identifier: NCT00298714,
NCT00265642).
Interferon-g
The interferons (IFNs) consist of a family of three major isoforms, including a, b, and g.
Many different IFN-a subtypes exist, whereas there appear to be only single IFN-b
and -g species. IFN-a and -b bind to the same receptor and therefore share many
common properties. IFN-a has much more potent antiviral effects than IFN-g.
IFN-g has been shown to specifically inhibit ECM synthesis in fibroblasts.69 Preclin-
ical work with IFN-g in hepatic stellate cells showed that this cytokine inhibited multiple
aspects of stellate cell activation.70,71 These data led to an initial pilot study showing
that IFN-g 1b was safe and well tolerated in humans who had HCV infection and ad-
vanced fibrosis.72 In addition, it led to reduction in fibrosis in selected patients.72
A subsequent double-blind, placebo-controlled, multicenter study examined IFN-g
1b in 488 patients who had an Ishak fibrosis score of 4 to 6 assigned to one of three treat-
ment groups: IFN-g 1b 100 mg (group 1, n 5 169), IFN-g 1b 200 mg (group 2, n 5 157), or
placebo (group 3, n 5 162) three times weekly for 48 weeks.73 Most patients (83.6%)
had cirrhosis at baseline (Ishak score 5 5 or 6). Among the 420 patients in whom pre-
and posttreatment liver biopsies were evaluable, no improvement in Ishak score was
seen among the three groups. Analysis of IFN-g–inducible biomarkers showed that
IFN-inducible T cell a chemoattractant (I-TAC), an IFN-g–inducible CXCR3 chemokine,
was an independent predictor of stable or improving Ishak score. IFN-g was well toler-
ated, suggesting that it could be effective in certain patient subgroups.
In a randomized, open-labeled, multicenter trial of IFN-g in patients who had HBV
infection and biopsy-proven hepatic fibrosis,74 99 patients who were not receiving
anti-HBV antiviral medications were divided into those receiving diammone glycyrrhi-
zinate and potassium magnesium aspartate alone (n 5 33) and those receiving these
medications plus 50 mg IFN-g intramuscularly on a daily basis for 3 months, and on
alternate days the subsequent 6 months (n 5 66). Most patients had follow-up biop-
sies at 9 months. Hepatic fibrosis scores were significantly reduced in 63% of IFN-g
treated patients compared with 24.1% in the control group. Using a semiquantitative
scoring system combining the systems developed by Chevallier and colleagues75 and
Knodell and colleagues,76 mean total fibrosis scores decreased from 13.8 5.8 to
10.1 5.1 in the IFN-g group (P 5 .0001), whereas they were unchanged in control
subjects (13.2 6.8 vs. 12.6 4.8; P 5 .937). Using the Scheuer histologic grading
system, 12 of 54 patients improved one stage or more in the IFN-g group compared
with 1 of 29 in the control group. Of 35 patients who had compensated cirrhosis (26
receiving IFN-g and 9 in the control group), 5 in the IFN-g group were found to have
histologic reversal of cirrhosis, whereas no patient in the control group showed an im-
provement in fibrosis at the 9-month follow-up biopsy.
In aggregate, the biologic rationale for using IFN-g is strong. Data suggest that there
are likely to be subgroups of patients for whom IFN-g may be effective, although
whether it would be cost-effective is unclear.
Peroxisomal proliferator–activated receptor g ligands

The PPAR family of nuclear hormone receptors consists of three subgroups: a, g, and
d (b). These receptors heterodimerize with the retinoid X receptor and bind to specific
regions on target-gene DNA. PPARg ligands have received great attention because
hepatic stellate cell activation during liver injury is associated with reduced PPARg ex-
pression,77 and activation of this receptor by exogenous PPARg ligands77 or reex-
pression of PPARg itself in stellate cells reversed the activated phenotype.78
Additionally, expression of PPARg during liver injury led to substantial improvements
in fibrosis.79
In a pilot study, 30 subjects who had histologic evidence of NASH received the
PPARg ligand, rosiglitazone, for 48 weeks,34 and 26 underwent posttreatment biop-
sies. Overall, significant improvement was seen in hepatocellular ballooning and
zone 3 perisinusoidal fibrosis. For the 25 patients completing 48 weeks of treatment,
insulin sensitivity and mean serum alanine aminotransferase (ALT) levels (104 initially,
42 U/L at end of treatment) improved significantly. Weight gain occurred in 67% of pa-
tients during treatment. Liver tests returned to baseline after treatment termination,
consistent with data from a subsequent study that showed a return of histologic
NASH after cessation of a PPARg ligand used for therapy.80
946 Rockey
In another small study, pioglitazone was examined in subjects who had impaired
glucose tolerance or type 2 diabetes and liver biopsy–confirmed NASH.81 Subjects
were randomized to 6 months of a hypocaloric diet (a reduction of 500 kcal/d in rela-
tion to the calculated daily intake required to maintain body weight) plus pioglitazone
(45 mg/d) or a hypocaloric diet plus placebo.81 Compared with placebo, diet plus pio-
glitazone improved glycemic control and glucose tolerance and led to normalization of
aminotransferase levels. Pioglitazone also decreased hepatic fat content, histologic
evidence of steatosis (P 5 .003), ballooning necrosis (P 5 .02), and inflammation
(P 5 .008), but did not reduce fibrosis significantly compared with placebo (P 5 .08).
The findings of these small studies suggest that treatment of underlying NASH may
be associated with an improvement in fibrosis, and warrant larger studies, particularly
given the preclinical data suggesting a direct affect of PPARg ligands on stellate cell
activation. One study examining features of stellate cell biology is currently underway
(see http://clinicaltrials.gov/;ClinicalTrials.gov Identifier: NCT00244751), with prelimi-
nary results expected in 2008.
Pirfenidone
Pirfenidone (5-methyl-1-phenyl-2-(1H)-pyridone) is a small orally bioavailable mole-
cule that seems to inhibit collagen synthesis,82,83 although its molecular mechanism
of action is not clearly understood. Pirfenidone has been shown to have antifibrotic ef-
fects in various fibrogenic animal models, including the lung, kidney, and liver.84–87
Pirfenidone has been evaluated in a small number of patients who have fibrosing pa-
renchymal organ diseases. In a double-blind, randomized, placebo-controlled trial of
107 patients who had idiopathic pulmonary fibrosis, pirfenidone led to an improve-
ment in vital capacity and a reduction in the number of episodes of acute exacerbation
of idiopathic pulmonary fibrosis compared with placebo (P 5 .0031).88 Significant
adverse events were associated with pirfenidone. In 15 patients who had treatment-
naı̈ve HCV related fibrogenesis, the compound was administered for 12 months
(1200 mg/d)89 and pre- and posttreatment biopsies were compared. Fibrosis was re-
duced in 5 of 15 patients (30%) by the end of 12 months of treatment.
Colchicine
Colchicine is a plant alkaloid that inhibits polymerization of microtubules, a process
believed to be required for collagen secretion. Thus, this compound is believed to
be antifibrotic through preventing collagen secretion and deposition. Colchicine effec-
tively inhibits collagen synthesis and fibrosis in experimental animal models.90–92
Given the rationale for use of colchicine as an antifibrotic and its presumed favorable
safety profile, colchicine has been studied in several clinical trials,93–96 including in pri-
mary biliary cirrhosis, alcoholic cirrhosis, and various other liver diseases.94 In one pri-
mary biliary cirrhosis trial, improvements were noted in several biochemical markers,
but colchicine failed to reduce fibrosis.93 In a study comparing colchicine and metho-
trexate for primary biliary cirrhosis in 42 subjects, no change was seen in fibrosis after
24 months of treatment with colchicine (or methotrexate), and interleukin (IL)-1b syn-
thesis was elevated in peripheral blood mononuclear cells in those who had stable or
improving histologic stage.97
In a double-blind, randomized, controlled trial of colchicine versus placebo in 100
patients who had different liver diseases (mostly alcohol or posthepatitic), colchicine
led to improved fibrosis and a dramatic improvement in survival when patients were
followed up for up to 14 years.94 However, this study has been questioned because
many patients were lost to follow-up, and substantial unexplained excess mortality
occurred in the control group from causes unrelated to liver disease. A meta-analysis
involving 1138 subjects found that colchicine had no effect on fibrosis or mortality.96
In a multicenter study comparing colchicine (0.6 mg orally twice daily) to placebo in

549 patients who had alcoholic liver disease, active treatment had no effect on survival
(histologic data were not obtained).95 A more recent small trial of colchicine random-
ized 38 subjects who had miscellaneous liver diseases to receive either colchicine
1 mg/d (n 5 21) or no agent (n 5 17) for at least 12 months.98 Liver biopsy was per-
formed before beginning colchicine and after 12 months. Mean albumin serum levels
increased within 12 months posttreatment only in the colchicine group (P<.05). Al-
though Knodell fibrosis scores remained unchanged at 12 months, seven patients
were noted to have a reduction in mean serum PIIINP (N-terminal propeptide of
type III procollagen) levels during the 24-month posttreatment follow-up period.
In summary, the data indicate that colchicine is generally safe (although one report
suggested that it may alter the response to IFN-a–based anti-HCV therapy)99 and may
improve markers of liver disease and even mortality from liver disease. Overall, how-
ever, colchicine does not seem to reduce hepatic fibrosis and therefore cannot be rec-
ommended as a primary antifibrotic treatment. In small randomized studies, colchicine
did not seem to be effective in treating pulmonary fibrosis.100,101
Herbal medicines
Several herbal medicines have been shown to have antifibrotic properties in experi-
mental animal models.21,102–105 Many of these medications have arisen from China.106
Although the mechanisms of many of these agents is unknown, these compounds are
being used extensively in a wide array of patients who have liver diseases.107 Medica-
tions containing herbs of the Salvia genus have been popular as antifibrotics; salvia-
nolic acid B, a major water-soluble polyphenolic acid, seems to be the major active
ingredient.106,107 This compound seems to have specific effects on stellate cells.
The active ingredients of other agents, including curcumin, glycyrrhizin, celastrol, tet-
randrine, berberine, and oxymatrine, seem to have a wide variety of biologic effects,
accounting for their purported activity in human disease.
Although some studies have suggested effectiveness of specific herbal medi-
cines,106,107 rigorous evidence is sorely lacking. Because it is well appreciated that
these herbal medicines may have significant toxicity, including hepatotoxicity,108
they should be used with extreme caution.
Compounds with a Potential Antifibrotic Effect Occurring from Upstream Effects

Several compounds seem capable of affecting fibrogenesis; not through a direct ef-
fect on stellate cells or on matrix synthesis per se, but rather from having an effect
on other important biologic events, such as on lipid peroxidation, the immune system,
or others. These compounds are highlighted briefly in this section and in Table 4.
Silymarin
The major active component of the milk thistle Silybum marianum, silymarin extract
(the major component of which is silybinin), reduces lipid peroxidation and inhibits fi-
brogenesis in small animals109,110 and baboons.111 Although fibrosis was not studied
as a primary outcome, the compound has been found to be safe. It has been reported
to have variable effects.112,113 One study showed a putative benefit on mortality in
patients who had alcohol-induced liver disease.112 Those who had early stages of
cirrhosis also seemed to benefit. However, in another study focused solely on alco-
holics, no survival benefit could be identified.113 Given the apparent safety of silymarin
and its common use as a complementary and alternative medicine, studies have been
initiated in patients who have NASH or whose HCV infection failed to respond
to conventional antiviral treatment (http://clinicaltrials.gov/;ClinicalTrials.gov
948 Rockey
Table 4
Potential antifibrotic therapies with ‘‘general’’effects
Agent Disease Efficacy Safety Comments

Silymarin ETOH 1 1 Anti-fibrotic effects may
be secondary to
anti-oxidative effects
Interleukin-10 HCV 11 1 Increased viral load
Malotilate ETOH 111
Polyenylphosphatidylcholine ETOH 1111
Propylthiouracil ETOH 11
S-adenosylmethionine ETOH 1 111
Anti-TNF-a compounds ETOH 11 1 Clearly anti-inflammatory
Ursodeoxycholic acid Multiple 1 1111 Most studied in PBC
Vitamin E HCV/NASH 1111
The scale for efficacy and safety to 1111, with being the lowest rating and 1111 the
highest. The ratings are empiric based on the aggregate available literature. See text for references
and discussion of mechanisms.
Abbreviations: ETOH, alcohol; HCV, hepatitis C virus; NASH, nonalcoholic steatohepatitis; PBC,
primary biliary cirrhosis; TNF, tumor necrosis factor.
Data from Rockey DC. Antifibrotic therapy in chronic liver disease. Clin Gastroenterol Hepatol
2005;3:95–107.
Identifier: NCT00680407 and NCT00680342). Although fibrosis is not a primary out-

come measure, histologic data are planned, and thus information about the effect
of silymarin on liver fibrosis is anticipated.
Polyenylphosphatidylcholine
Polyenylphosphatidylcholine has gained considerable interest in the treatment of pa-
tients who have liver injury. The therapeutic compound is derived from purified soy-
bean extract, consisting of 95% to 96% polyunsaturated phosphatidylcholines.
Polyenylphosphatidylcholine has both antioxidant and antifibrotic properties. It is at-
tractive in alcoholic liver injury because this disease is often associated with oxidative
stress. Oxidative stress then leads to lipid peroxidation, cellular injury, inflammation,
and subsequently, fibrogenesis. Experts have thus proposed that because phospha-
tidylcholine is a prominent component of cell membranes, its supplementation should
protect cell membranes and might lead to reduced cellular injury and fibrogenesis.
Experimental data support this concept.114
Several large studies have been undertaken in an attempt to determine whether poly-
enylphosphatidylcholine is beneficial. A multicenter, prospective, randomized, double-
blind, placebo-controlled Department of Veterans Affairs cooperative clinical trial
randomized 789 alcoholics (average alcohol intake of 16 drinks per day)115 to either poly-
enylphosphatidylcholine or placebo for 2 years. Although most subjects substantially
reduced their ethanol consumption during the trial (which was believed to improve fibrosis
in the control group), polyenylphosphatidylcholine failed to cause a comparative
improvement in fibrosis. A subsequent study examining the effect of polyenylphosphati-
dylcholine is currently underway (http://clinicaltrials.gov/;ClinicalTrials.gov Identifier:
NCT00211848).
Ursodeoxycholic acid
Ursodeoxycholic acid, a nontoxic bile acid, binds to hepatocyte membranes and is
presumably cytoprotective, thereby reducing inflammation and therefore downstream
fibrogenesis.116 Neither experimental data nor human studies indicate a primary anti-
fibrotic effect of ursodeoxycholic acid in the liver. However, an extensive body of lit-
erature on various of liver diseases has generally examined ursodeoxycholic.117–125
The aggregate data suggest that ursodeoxycholic acid may impede progression of fi-
brosis in primary biliary cirrhosis by way of effects on biliary ductal inflammation, par-
ticularly if initiated early in the disease course. Ursodeoxycholic acid is safe, and
although it is expensive, this author believes that in the absence of definitively effective
agents, the available data justify its use as an antifibrotic, primarily in patients who
have primary biliary cirrhosis.
Interleukin-10
IL-10 is a potent immunomodulatory cytokine that can down-regulate production of
proinflammatory T-cell cytokines, such as tumor necrosis factor (TNF)-a, IL-1, IFN-g,
and IL-2. Endogenous IL-10 seems to attenuate the intrahepatic inflammatory response
and reduce fibrosis in several models of liver injury.126 A direct antifibrotic effect for IL-
10 has not been established. Notwithstanding, in an effort shift the intrahepatic immu-
nologic balance away from Th1 cytokine predominance (subcutaneous IL-10 given
daily or three times weekly), IL-10 was given to 30 subjects who had HCV infection
and advanced fibrosis whose infection failed to respond to antiviral therapy for
12 months.127 This therapy decreased hepatic inflammatory activity and fibrosis, but
led to increased HCV viral levels. Therefore, although these results suggest that IL-10
might be an attractive antifibrotic agent, the adverse effects on HCV viral levels are
problematic.
Miscellaneous antioxidants and anti-inflammatory compounds

Oxidative stress has been implicated in a wide variety of biologic processes in liver in-
jury (including stellate cell activation and stimulation of ECM production).128 Thus,
antioxidants have received considerable attention as putative antifibrotics.129–133
Compounds such as the vitamin E precursor d-alpha-tocopherol (1200 IU/d for
8 weeks),131 vitamin E,132,133 malotilate,134 propylthiouracil,135 penicillamine,136–138
and S-adenosylmethionine139,140 have all been tested in humans, although evidence
supporting their effectiveness remains lacking. However, for many of these com-
pounds, fibrosis was not typically measured as a specific outcome and therefore it is
not appropriate to consider these agents as primary antifibrotics but rather as com-
pounds that could have secondary effects on fibrogenesis because of other properties.
Methotrexate is also considered to be an anti-inflammatory compound. However, it
has also been believed to be profibrogenic,141,142 although the risk for fibrosis pro-
gression when used in patients who have skin or rheumatologic disease may be
less than commonly believed.141,143 Nonetheless, it has been studied in a substantial
number of human trials as a therapeutic agent in patients who had primary biliary cir-
rhosis.97,143–146 Although improvements in disease and fibrosis have been reported,
including reversion of fibrosis,147 most data on methotrexate are either negative144,146
or show that its effects are marginal, either alone146 or in combination with colchi-
cine.148 If methotrexate is used to treat patients (who have primary biliary cirrhosis),
an experienced hepatologist must manage its use.
Experimental evidence suggests that inhibition of TNF-a signaling during liver injury
may ameliorate fibrosis.149,150 TNF-a is up-regulated in alcoholic liver disease, and thus
an anti–TNF-a compound would be attractive because it should reduce inflammation
and thus fibrosis. Several studies have examined the effect of anti–TNF-a compounds
in patients who have alcohol-induced liver disease.151–154 Available evidence suggests
an improvement in inflammation and acute injury (which presumably precede fibrosis in
950 Rockey
this disease).154 A randomized, double-blind, placebo-controlled trial of etanercept in

patients who had moderate to severe alcoholic hepatitis showed no improvement in
mortality at 1 month, and patients treated with etanercept had a greater mortality after
6 months; however, this study did not evaluate liver histology.155
WHY HAVE SO MANY POTENTIAL THERAPIES BEEN EFFECTIVE IN ANIMAL MODELS,

YET SO INEFFECTIVE IN HUMANS THUS FAR?
This key question is a major conundrum for the field of antifibrotics. A critical consid-
eration is that experimental models and conditions are dramatically different from real-
life situations. First, in most animal experiments, antifibrotic agents have been tested
for their ability to prevent development of fibrosis, which almost never happens in the
clinical arena (patients present with advanced fibrosis or cirrhosis, and little or no op-
portunity presents to treat patients during fibrosis progression). Second, patients in
most human trials performed have had advanced fibrosis, if not cirrhosis, and because
the duration of treatment has been relatively short, even if a compound actively in-
hibited fibrosis, it seems unlikely that a demonstrable benefit will be apparent within
a short timeframe (1 or 2 years).
It is also possible that the pharmacokinetics of therapeutic agents are different be-
tween animals and humans. For example, drug levels may be pushed to very high
levels in animals, but these levels are not realistically attainable in humans. It is also
possible that compounds that truly have antifibrotic features in animals are simply
not antifibrotic in humans; this may be because of differences in basic cell or molecular
aspects of the fibrogenic platforms.
Finally, the duration of injury differs markedly between rodent models and human
disease, which could lead to significant differences in the cross-linking of ECM, and
thus its potential for degradation. Although human diseases that lead to fibrosis re-
quire decades, this process is condensed into weeks or months in rodents, providing
less time for the ECM to mature. Therefore, less chemical cross linking occurs and in-
stead the scar remains highly cellular and resorbable.
FUTURE SPECIFIC TARGETS
A comprehensive discussion of the many different putative pathways that could lead
to novel antifibrotic therapeutics is beyond the scope of this article. However, several
systems/areas are particularly attractive (Box 1). The most central of fibrogenic path-
ways involve the cytokine transforming growth factor b (TGF-b). Several approaches
to inhibit the action of TGF-b can interrupt its signaling pathway.156–158 The concept
is clear, although theoretic concerns include the potential (pro-proliferative) effect of
inhibiting TGF-b signaling in vivo.
Recent data implicate the cannabinoid system in fibrogenesis. In the injured liver,
the endogenous endocannabinoid receptors CB1 and CB2 are up-regulated and
thus facilitate endocannabinoid signaling.159 Additionally, in patients who have
chronic HCV infection, daily cannabis use is an independent predictor of fibrosis pro-
gression.160 Although up-regulation of endogenous hepatic cannabinoid CB2 recep-
tors is associated with progression of experimental liver fibrosis,161 CB1 receptors
were induced in human cirrhotic samples and liver fibrogenic cells.162 Furthermore,
in animals undergoing liver injury, a CB1 receptor antagonist inhibited fibrosis, pre-
sumably through inhibiting expression of TGF-b1 and either inhibiting growth hepatic
myofibroblasts or stimulating apoptosis.162
Emerging data suggest that angiogenesis is important in the fibrogenic response to
injury,163,164 and therefore antiangiogenic compounds are attractive therapeutic targets.
Box 1
951
Experimental antifibrotic therapies according to mechanism of action
Inhibit stellate cell activation and fibrogenesis

Cannabinoid receptor antagonists
Rapamycin
Mycophenolate
Gliotoxin
Pentoxifylline
HOE 77
Endothelin antagonists
Prazosin
Angiotensin II–converting enzyme inhibitors
Sho-saiko-to (TJ-9)
Retinoic acid
Arginine-glycine-aspartate peptides
Inhibit stellate cell fibrogenesis

Halofuginone
Anti–TGF-b compounds
Follistatin
Other/unknown or generalized effect

Prostaglandins
Adiponectin
COX-2 inhibitors
Estradiol
Curcumin
Anticoagulation
Puerarin
Salvia miltiorrhiza
IL-1 antagonist
Octreotide
Glycyrrhizin
Hepatocyte growth factor
Stellate cell activation-associated protein
PAR antagonists
PPARs
C-5 depletion
5-lipoxygenase inhibitors
Angiostatin
Anti–TNF-a compounds
Hepatocyte growth factor (inhibition)
Connective tissue growth factor (inhibition)
Relaxin
952 Rockey
Likewise, as biology uncovering stellate cell signaling pathways continues to emerge,

therapy targeted at these pathways will become attractive.165 However, the signaling
pathways are extremely complicated and may vary among models of injury.166
An important therapeutic concept is directed or targeted therapy. Because many
compounds have adverse effects on collateral cells or organs outside the fibrogenic re-
sponse, specifically targeting fibrogenic cells, particularly hepatic stellate cells, would
be most desirable.102,167–172 The ability to specifically stimulate stellate cell apoptosis
and enhance the resolution of fibrosis is especially attractive.20 Additionally, the ability
to potentially specifically target small interfering RNAs to the liver also makes this ap-
proach appealing.173–175 MicroRNAs may also be important in fibrogenesis;176 addi-
tional investigation in liver injury models is expected to lead to potential therapies for
fibrosis. Several other specific targets are of considerable interest (see Box 1).
Farnesoid X receptor (FXR) is a member of the nuclear hormone receptor superfam-
ily or transcription factors that is bile acid-activated. It is not only hepatoprotective in
various experimental models of liver injury,177,178 but it may also ameliorate fibrosis.
FXR activators may be particularly useful in patients with cholestatic injury.
SUMMARY
Elucidation of the mechanisms responsible for fibrogenesis, with particular emphasis

on stellate cell biology, has generated great hope that novel therapies will evolve, and
in fact the field of antifibrotic compounds is growing rapidly. A central event in
fibrogenesis is the activation of effector cells (with hepatic stellate cells the most
prominent). The activation process is characterized by several important features,
including enhanced matrix synthesis and transition to a myofibroblast-like (and con-
tractile) phenotype. Factors controlling activation are multifactorial and complex,
and thus multiple potential therapeutic interventions are possible. A further critical
concept is that even advanced fibrosis is dynamic and may be reversible. Currently,
the most effective therapy for hepatic fibrogenesis is to attenuate or clear the under-
lying disease. The most effective specific antifibrotic therapies will most likely be
directed at fibrogenic effector cells, either in a targeted fashion or using generalized
approaches that account for biologic differences between fibrogenic cells and their
nonfibrogenic neighbors (Box 1). Additionally, approaches that address matrix remod-
eling (ie, through enhancing matrix degradation or inhibiting factors that prevent matrix
breakdown) will be pursued. Thus, although no specific, effective, safe, and inexpen-
sive antifibrotic therapies are yet available, multiple potential targets have been
identified and effective therapies are expected to emerge.
ACKNOWLEDGMENTS
This work was supported by the NIH (grants R01 DK 50574 and R01 DK 60338).
REFERENCES
1. Caligiuri A, Glaser S, Rodgers RE, et al. Endothelin-1 inhibits secretin-stimulated

ductal secretion by interacting with ETA receptors on large cholangiocytes. Am
J Physiol 1998;275:G835–46.
2. Jarnagin WR, Rockey DC, Koteliansky VE, et al. Expression of variant fibronec-
tins in wound healing: cellular source and biological activity of the EIIIA segment
in rat hepatic fibrogenesis. J Cell Biol 1994;127:2037–48.
3. Kinnman N, Francoz C, Barbu V, et al. The myofibroblastic conversion of peribili-
ary fibrogenic cells distinct from hepatic stellate cells is stimulated by platelet-
derived growth factor during liver fibrogenesis. Lab Invest 2003;83:163–73.
4. Kisseleva T, Uchinami H, Feirt N, et al. Bone marrow-derived fibrocytes partici-

pate in pathogenesis of liver fibrosis. J Hepatol 2006;45:429–38.
to liver fibrosis. Gastroenterology 2006;130:1807–21.
6. Wang X, Kanel GC, DeLeve LD. Support of sinusoidal endothelial cell glutathione
prevents hepatic veno-occlusive disease in the rat. Hepatology 2000;31:428–34.
7. Wells RG, Kruglov E, Dranoff JA. Autocrine release of TGF-beta by portal fibro-
blasts regulates cell growth. FEBS Lett 2004;559:107–10.
fibrosis via epithelial to mesenchymal transition. J Biol Chem 2007;282:
23337–47.
9. Maher JJ, McGuire RF. Extracellular matrix gene expression increases preferen-
tially in rat lipocytes and sinusoidal endothelial cells during hepatic fibrosis in
vivo. J Clin Invest 1990;86:1641–8.
10. Rockey DC, Boyles JK, Gabbiani G, et al. Rat hepatic lipocytes express smooth
muscle actin upon activation in vivo and in culture. J Submicrosc Cytol Pathol
1992;24:193–203.
11. Arthur MJ. Fibrogenesis II. Metalloproteinases and their inhibitors in liver fibro-
sis. Am J Physiol Gastrointest Liver Physiol 2000;279:G245–9.
12. Benyon RC, Iredale JP, Goddard S, et al. Expression of tissue inhibitor of metal-
loproteinases 1 and 2 is increased in fibrotic human liver. Gastroenterology
1996;110:821–31.
13. Emonard H, Grimaud JA. Active and latent collagenase activity during reversal
of hepatic fibrosis in murine schistosomiasis. Hepatology 1989;10:77–83.
14. Friedman SL, Rockey DC, Bissell DM. Hepatic fibrosis 2006: report of the Third
AASLD Single Topic Conference. Hepatology 2007;45:242–9.
15. Preaux AM, Mallat A, Van Nhieu JT, et al. Matrix metalloproteinase-2 activation in
human hepatic fibrosis regulation by cell-matrix interactions. Hepatology 1999;
30:944–50.
16. Takahara T, Furui K, Funaki J, et al. Increased expression of matrix metallopro-
teinase-II in experimental liver fibrosis in rats. Hepatology 1995;21:787–95.
17. Garcia-Banuelos J, Siller-Lopez F, Miranda A, et al. Cirrhotic rat livers with exten-
sive fibrosis can be safely transduced with clinical-grade adenoviral vectors. Ev-
idence of cirrhosis reversion. Gene Ther 2002;9:127–34.
18. Siller-Lopez F, Sandoval A, Salgado S, et al. Treatment with human metallopro-
teinase-8 gene delivery ameliorates experimental rat liver cirrhosis. Gastroenter-
ology 2004;126:1122–33; discussion 949.
19. Issa R, Williams E, Trim N, et al. Apoptosis of hepatic stellate cells: involvement
in resolution of biliary fibrosis and regulation by soluble growth factors. Gut
2001;48:548–57.
20. Wright MC, Issa R, Smart DE, et al. Gliotoxin stimulates the apoptosis of human
and rat hepatic stellate cells and enhances the resolution of liver fibrosis in rats.
21. Zhang S, Ji G, Liu J. Reversal of chemical-induced liver fibrosis in Wistar rats by
puerarin. J Nutr Biochem 2006;17:485–91.
22. Hadziyannis SJ, Tassopoulos NC, Heathcote EJ, et al. Adefovir dipivoxil for the
treatment of hepatitis B e antigen-negative chronic hepatitis B. N Engl J Med
2003;348:800–7.
23. Lai CL, Chien RN, Leung NW, et al. A one-year trial of lamivudine for chronic
hepatitis B. Asia Hepatitis Lamivudine Study Group [see comments]. N Engl J
Med 1998;339:61–8.
954 Rockey
24. Mallet VO, Dhalluin-Venier V, Verkarre V, et al. Reversibility of cirrhosis in HIV/

HBV coinfection. Antivir Ther 2007;12:279–83.
25. Abergel A, Darcha C, Chevallier M, et al. Histological response in patients
treated by interferon plus ribavirin for hepatitis C virus-related severe fibrosis.
Eur J Gastroenterol Hepatol 2004;16:1219–27.
26. Poynard T, McHutchison J, Manns M, et al. Impact of pegylated interferon alfa-
2b and ribavirin on liver fibrosis in patients with chronic hepatitis C. Gastroenter-
ology 2002;122:1303–13.
27. Dufour JF, DeLellis R, Kaplan MM. Reversibility of hepatic fibrosis in autoimmune
hepatitis. Ann Intern Med 1997;127:981–5.
28. Ramond MJ, Poynard T, Rueff B, et al. A randomized trial of prednisolone in pa-
tients with severe alcoholic hepatitis. N Engl J Med 1992;326:507–12.
29. Spahr L, Rubbia-Brandt L, Pugin J, et al. Rapid changes in alcoholic hepatitis
histology under steroids: correlation with soluble intercellular adhesion mole-
cule-1 in hepatic venous blood. J Hepatol 2001;35:582–9.
30. Blumberg RS, Chopra S, Ibrahim R, et al. Primary hepatocellular carcinoma in
idiopathic hemochromatosis after reversal of cirrhosis. Gastroenterology 1988;
95:1399–402.
31. Powell LW, Kerr JF. Reversal of ‘‘cirrhosis’’ in idiopathic haemochromatosis fol-
lowing long-term intensive venesection therapy. Australas Ann Med 1970;19:
54–7.
32. Wu SF, Peng CT, Wu KH, et al. Liver fibrosis and iron levels during long-term de-
feriprone treatment of thalassemia major patients. Hemoglobin 2006;30:215–8.
33. Hammel P, Couvelard A, O’Toole D, et al. Regression of liver fibrosis after biliary
drainage in patients with chronic pancreatitis and stenosis of the common bile
duct. N Engl J Med 2001;344:418–23.
34. Neuschwander-Tetri BA, Brunt EM, Wehmeier KR, et al. Improved nonalcoholic
steatohepatitis after 48 weeks of treatment with the PPAR-gamma ligand rosigli-
tazone. Hepatology 2003;38:1008–17.
35. Blasco A, Forns X, Carrion JA, et al. Hepatic venous pressure gradient identifies
patients at risk of severe hepatitis C recurrence after liver transplantation. Hep-
atology 2006;43:492–9.
36. Ekstedt M, Franzen LE, Mathiesen UL, et al. Long-term follow-up of patients with
NAFLD and elevated liver enzymes. Hepatology 2006;44:865–73.
37. Dienstag JL, McHutchison JG. American Gastroenterological Association tech-
nical review on the management of hepatitis C. Gastroenterology 2006;130:
231–64; quiz 214–7.
38. Strader DB, Wright T, Thomas DL, et al. Diagnosis, management, and treatment
of hepatitis C. Hepatology 2004;39:1147–71.
39. Rockey DC, Chung JJ. Endothelin antagonism in experimental hepatic fibrosis.
Implications for endothelin in the pathogenesis of wound healing. J Clin Invest
1996;98:1381–8.
hepatitis C. Hepatology 2003;38:1449–57.
41. O’Brien MJ, Keating NM, Elderiny S, et al. An assessment of digital image anal-
ysis to measure fibrosis in liver biopsy specimens of patients with chronic hep-
atitis C. Am J Clin Pathol 2000;114:712–8.
42. Pilette C, Rousselet MC, Bedossa P, et al. Histopathological evaluation of liver
fibrosis: quantitative image analysis vs semi-quantitative scores. Comparison
with serum markers. J Hepatol 1998;28:439–46.
43. Thampanitchawong P, Piratvisuth T. Liver biopsy: complications and risk factors.

World J Gastroenterol 1999;5:301–4.
44. Ratziu V, Charlotte F, Heurtier A, et al. Sampling variability of liver biopsy in non-
alcoholic fatty liver disease. Gastroenterology 2005;128:1898–906.
45. Regev A, Berho M, Jeffers LJ, et al. Sampling error and intraobserver variation in
liver biopsy in patients with chronic HCV infection. Am J Gastroenterol 2002;97:
2614–8.
46. Rockey DC, Bissell DM. Noninvasive measures of liver fibrosis. Hepatology
2006;43:S113–20.
47. Poynard T, Bedossa P, Opolon P. Natural history of liver fibrosis progression in
patients with chronic hepatitis C. The OBSVIRC, METAVIR, CLINIVIR, and DOS-
VIRC groups. Lancet 1997;349:825–32.
48. Wai CT, Greenson JK, Fontana RJ, et al. A simple noninvasive index can predict
both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepa-
tology 2003;38:518–26.
49. Colli A, Fraquelli M, Andreoletti M, et al. Severe liver fibrosis or cirrhosis: accu-
racy of US for detection–analysis of 300 cases. Radiology 2003;227:89–94.
50. Imbert-Bismut F, Ratziu V, Pieroni L, et al. Biochemical markers of liver fibrosis in
patients with hepatitis C virus infection: a prospective study. Lancet 2001;357:
1069–75.
51. Callewaert N, Van Vlierberghe H, Van Hecke A, et al. Noninvasive diagnosis of
liver cirrhosis using DNA sequencer-based total serum protein glycomics. Nat
Med 2004;10:429–34.
52. McHutchison JG, Blatt LM, de Medina M, et al. Measurement of serum hyalur-
onic acid in patients with chronic hepatitis C and its relationship to liver histology.
Consensus Interferon Study Group. J Gastroenterol Hepatol 2000;15:945–51.
53. Rockey DC. Noninvasive assessment of liver fibrosis and portal hypertension
with transient elastography. Gastroenterology 2008;134:8–14.
54. Sandrin L, Fourquet B, Hasquenoph JM, et al. Transient elastography: a new
noninvasive method for assessment of hepatic fibrosis. Ultrasound Med Biol
2003;29:1705–13.
55. Castera L, Foucher J, Bertet J, et al. FibroScan and FibroTest to assess liver
fibrosis in HCV with normal aminotransferases. Hepatology 2006;43:373–4;
author reply 375–6.
56. Fraquelli M, Rigamonti C, Casazza G, et al. Reproducibility of transient elastog-
raphy in the evaluation of liver fibrosis in patients with chronic liver disease. Gut
2007;56:968–73.
57. Ziol M, Handra-Luca A, Kettaneh A, et al. Noninvasive assessment of liver fibro-
sis by measurement of stiffness in patients with chronic hepatitis C. Hepatology
2005;41:48–54.
58. Bataller R, Gabele E, Parsons CJ, et al. Systemic infusion of angiotensin II exac-
erbates liver fibrosis in bile duct-ligated rats. Hepatology 2005;41:1046–55.
59. Bataller R, Sancho-Bru P, Gines P, et al. Activated human hepatic stellate cells
express the renin-angiotensin system and synthesize angiotensin II. Gastroen-
terology 2003;125:117–25.
60. Hirose A, Ono M, Saibara T, et al. Angiotensin II type 1 receptor blocker inhibits
fibrosis in rat nonalcoholic steatohepatitis. Hepatology 2007;45:1375–81.
61. Jin H, Yamamoto N, Uchida K, et al. Telmisartan prevents hepatic fibrosis and
enzyme-altered lesions in liver cirrhosis rat induced by a choline-deficient
L-amino acid-defined diet. Biochem Biophys Res Commun 2007;364:801–7.
956 Rockey
62. Nabeshima Y, Tazuma S, Kanno K, et al. Anti-fibrogenic function of angiotensin II

type 2 receptor in CCl4-induced liver fibrosis. Biochem Biophys Res Commun
2006;346:658–64.
63. Rockey DC. Vascular mediators in the injured liver. Hepatology 2003;37:4–12.
64. De BK, Bandyopadhyay K, Das TK, et al. Portal pressure response to losartan
compared with propranolol in patients with cirrhosis. Am J Gastroenterol 2003;
98:1371–6.
65. Debernardi-Venon W, Martini S, Biasi F, et al. AT1 receptor antagonist Candesar-
tan in selected cirrhotic patients: effect on portal pressure and liver fibrosis
markers. J Hepatol 2007;46:1026–33.
66. Gonzalez-Abraldes J, Albillos A, Banares R, et al. Randomized comparison of
long-term losartan versus propranolol in lowering portal pressure in cirrhosis.
67. Schepke M, Werner E, Biecker E, et al. Hemodynamic effects of the angiotensin
II receptor antagonist irbesartan in patients with cirrhosis and portal hyperten-
sion. Gastroenterology 2001;121:389–95.
68. Tripathi D, Therapondos G, Lui HF, et al. Chronic administration of losartan, an
angiotensin II receptor antagonist, is not effective in reducing portal pressure
in patients with preascitic cirrhosis. Am J Gastroenterol 2004;99:390–4.
69. Czaja MJ, Weiner FR, Eghbali M, et al. Differential effects of gamma-interferon
on collagen and fibronectin gene expression. J Biol Chem 1987;262:13348–51.
70. Rockey DC, Chung JJ. Interferon gamma inhibits lipocyte activation and extra-
cellular matrix mRNA expression during experimental liver injury: implications for
treatment of hepatic fibrosis. J Investig Med 1994;42:660–70.
71. Rockey DC, Maher JJ, Jarnagin WR, et al. Inhibition of rat hepatic lipocyte acti-
vation in culture by interferon-gamma. Hepatology 1992;16:776–84.
72. Muir AJ, Sylvestre PB, Rockey DC. Interferon gamma-1b for the treatment of
fibrosis in chronic hepatitis C infection. J Viral Hepat 2006;13:322–8.
73. Pockros PJ, Jeffers L, Afdhal N, et al. Final results of a double-blind, placebo-
controlled trial of the antifibrotic efficacy of interferon-gamma1b in chronic hep-
atitis C patients with advanced fibrosis or cirrhosis. Hepatology 2007;45:569–78.
74. Weng HL, Wang BE, Jia JD, et al. Effect of interferon-gamma on hepatic fibrosis
in chronic hepatitis B virus infection: a randomized controlled study. Clin Gastro-
enterol Hepatol 2005;3:819–28.
75. Chevallier M, Guerret S, Chossegros P, et al. A histological semiquantitative
scoring system for evaluation of hepatic C fibrosis in needle liver biopsy spec-
imens: comparison with morphometric C studies. Hepatology 1994;20:349–55.
76. Knodell RG, Ishak KG, Black WC, et al. Formulation and application of a numer-
ical scoring system for assessing histological activity in asymptomatic chronic
active hepatitis. Hepatology 1981;1:431–5.
77. Miyahara T, Schrum L, Rippe R, et al. Peroxisome proliferator-activated recep-
tors and hepatic stellate cell activation. J Biol Chem 2000;275:35715–22.
78. Hazra S, Xiong S, Wang J, et al. Peroxisome proliferator-activated receptor
gamma induces a phenotypic switch from activated to quiescent hepatic stellate
cells. J Biol Chem 2004;279:11392–401.
79. Yang L, Chan CC, Kwon OS, et al. Regulation of peroxisome proliferator-acti-
vated receptor-gamma in liver fibrosis. Am J Physiol Gastrointest Liver Physiol
2006;291:G902–11.
80. Lutchman G, Modi A, Kleiner DE, et al. The effects of discontinuing pioglitazone
in patients with nonalcoholic steatohepatitis. Hepatology 2007;46:424–9.
81. Belfort R, Harrison SA, Brown K, et al. A placebo-controlled trial of pioglitazone

in subjects with nonalcoholic steatohepatitis. N Engl J Med 2006;355:2297–307.
82. Di Sario A, Bendia E, Macarri G, et al. The anti-fibrotic effect of pirfenidone in rat
liver fibrosis is mediated by downregulation of procollagen alpha1(I), TIMP-1
and MMP-2. Dig Liver Dis 2004;36:744–51.
83. Di Sario A, Bendia E, Svegliati Baroni G, et al. Effect of pirfenidone on rat hepatic
stellate cell proliferation and collagen production. J Hepatol 2002;37:584–91.
84. Garcia L, Hernandez I, Sandoval A, et al. Pirfenidone effectively reverses exper-
imental liver fibrosis. J Hepatol 2002;37:797–805.
85. Liu H, Drew P, Gaugler AC, et al. Pirfenidone inhibits lung allograft fibrosis
through L-arginine-arginase pathway. Am J Transplant 2005;5:1256–63.
86. Shimizu T, Kuroda T, Hata S, et al. Pirfenidone improves renal function and fibro-
sis in the post-obstructed kidney. Kidney Int 1998;54:99–109.
87. Tada S, Nakamuta M, Enjoji M, et al. Pirfenidone inhibits dimethylnitrosamine-in-
duced hepatic fibrosis in rats. Clin Exp Pharmacol Physiol 2001;28:522–7.
88. Azuma A, Nukiwa T, Tsuboi E, et al. Double-blind, placebo-controlled trial of pir-
fenidone in patients with idiopathic pulmonary fibrosis. Am J Respir Crit Care
Med 2005;171:1040–7.
89. Armendariz-Borunda J, Islas-Carbajal MC, Meza-Garcia E, et al. A pilot study in
patients with established advanced liver fibrosis using pirfenidone. Gut 2006;55:
1663–5.
90. Poo JL, Feldmann G, Moreau A, et al. Early colchicine administration reduces
hepatic fibrosis and portal hypertension in rats with bile duct ligation. J Hepatol
1993;19:90–4.
91. Rodriguez L, Cerbon-Ambriz J, Munoz ML. Effects of colchicine and colchiceine
in a biochemical model of liver injury and fibrosis. Arch Med Res 1998;29:
109–16.
92. Rojkind M, Kershenobich D. Effect of colchicine on collagen, albumin and trans-
ferrin synthesis by cirrhotic rat liver slices. Biochim Biophys Acta 1975;378:
415–23.
93. Kaplan MM, Alling DW, Zimmerman HJ, et al. A prospective trial of colchicine for
primary biliary cirrhosis. N Engl J Med 1986;315:1448–54.
94. Kershenobich D, Vargas F, Garcia-Tsao G, et al. Colchicine in the treatment of
cirrhosis of the liver. N Engl J Med 1988;318:1709–13.
95. Morgan TR, Nemchausky B, Schiff ER, et al. Colchicine does not prolong life in
patients with advanced alcoholic cirrhosis: results of a prospective, randomized,
placebo-controlled trial. Gastroenterology 2002;641A.
96. Rambaldi A, Gluud C. Colchicine for alcoholic and non-alcoholic liver fibrosis or
cirrhosis. Liver 2001;21:129–36.
97. Miller LC, Sharma A, McKusick AF, et al. Synthesis of interleukin-1 beta in pri-
mary biliary cirrhosis: relationship to treatment with methotrexate or colchicine
and disease progression. Hepatology 1995;22:518–24.
98. Nikolaidis N, Kountouras J, Giouleme O, et al. Colchicine treatment of liver fibro-
sis. Hepatogastroenterology 2006;53:281–5.
99. Angelico M, Cepparulo M, Barlattani A, et al. Unfavourable effects of colchicine
in combination with interferon-alpha in the treatment of chronic hepatitis C. Ali-
ment Pharmacol Ther 2000;14:1459–67.
100. Douglas WW, Ryu JH, Swensen SJ, et al. Colchicine versus prednisone in the
treatment of idiopathic pulmonary fibrosis. A randomized prospective study.
Members of the Lung Study Group. Am J Respir Crit Care Med 1998;158:220–5.
958 Rockey
101. Selman M, Carrillo G, Salas J, et al. Colchicine, D-penicillamine, and prednisone

in the treatment of idiopathic pulmonary fibrosis: a controlled clinical trial. Chest
1998;114:507–12.
102. Mandal AK, Das S, Basu MK, et al. Hepatoprotective activity of liposomal flavo-
noid against arsenite-induced liver fibrosis. J Pharmacol Exp Ther 2007;320:
994–1001.
103. Sakaida I, Tsuchiya M, Kawaguchi K, et al. Herbal medicine Inchin-ko-to
(TJ-135) prevents liver fibrosis and enzyme-altered lesions in rat liver cirrhosis
induced by a choline-deficient L-amino acid-defined diet. J Hepatol 2003;38:
762–9.
104. Shimizu I, Ma YR, Mizobuchi Y, et al. Effects of Sho-saiko-to, a Japanese herbal
medicine, on hepatic fibrosis in rats. [see comments]. Hepatology 1999;29:
149–60.
105. Zhang XL, Liu L, Jiang HQ. Salvia miltiorrhiza monomer IH764-3 induces
hepatic stellate cell apoptosis via caspase-3 activation. World J Gastroenterol
2002;8:515–9.
106. Luk JM, Wang X, Liu P, et al. Traditional Chinese herbal medicines for treatment
of liver fibrosis and cancer: from laboratory discovery to clinical evaluation. Liver
Int 2007;27:879–90.
107. Wang BE. Treatment of chronic liver diseases with traditional Chinese medicine.
J Gastroenterol Hepatol 2000;15(15):E67–70.
108. Stedman C. Herbal hepatotoxicity. Semin Liver Dis 2002;22:195–206.
109. Boigk G, Stroedter L, Herbst H, et al. Silymarin retards collagen accumulation in
early and advanced biliary fibrosis secondary to complete bile duct obliteration
in rats. Hepatology 1997;26:643–9.
110. Jia JD, Bauer M, Cho JJ, et al. Antifibrotic effect of silymarin in rat secondary
biliary fibrosis is mediated by downregulation of procollagen alpha1(I) and
TIMP-1. J Hepatol 2001;35:392–8.
111. Lieber CS, Leo MA, Cao Q, et al. Silymarin retards the progression of alcohol-
induced hepatic fibrosis in baboons. J Clin Gastroenterol 2003;37:336–9.
112. Ferenci P, Dragosics B, Dittrich H, et al. Randomized controlled trial of silymarin
treatment in patients with cirrhosis of the liver. J Hepatol 1989;9:105–13.
113. Pares A, Planas R, Torres M, et al. Effects of silymarin in alcoholic patients with
cirrhosis of the liver: results of a controlled, double-blind, randomized and mul-
ticenter trial [see comments]. J Hepatol 1998;28:615–21.
114. Aleynik SI, Leo MA, Ma X, et al. Polyenylphosphatidylcholine prevents carbon
tetrachloride-induced lipid peroxidation while it attenuates liver fibrosis. J Hep-
atol 1997;27:554–61.
115. Lieber CS, Weiss DG, Groszmann R, et al. II. Veterans Affairs Cooperative Study
of Polyenylphosphatidylcholine in alcoholic liver disease. Alcohol Clin Exp Res
2003;27:1765–72.
116. Nava-Ocampo AA, Suster S, Muriel P. Effect of colchiceine and ursodeoxycholic
acid on hepatocyte and erythrocyte membranes and liver histology in experimen-
tally induced carbon tetrachloride cirrhosis in rats. Eur J Clin Invest 1997;27:77–84.
117. Combes B, Carithers RL Jr, Maddrey WC, et al. A randomized, double-blind,
placebo-controlled trial of ursodeoxycholic acid in primary biliary cirrhosis. Hep-
atology 1995;22:759–66.
118. Degott C, Zafrani ES, Callard P, et al. Histopathological study of primary biliary
cirrhosis and the effect of ursodeoxycholic acid treatment on histology progres-
sion. Hepatology 1999;29:1007–12.
119. Goulis J, Leandro G, Burroughs AK. Randomised controlled trials of ursodeox-

ycholic-acid therapy for primary biliary cirrhosis: a meta-analysis. Lancet 1999;
354:1053–60.
120. Jacquemin E, Hermans D, Myara A, et al. Ursodeoxycholic acid therapy in pe-
diatric patients with progressive familial intrahepatic cholestasis. Hepatology
1997;25:519–23.
121. Lindblad A, Glaumann H, Strandvik B. A two-year prospective study of the effect
of ursodeoxycholic acid on urinary bile acid excretion and liver morphology in
cystic fibrosis-associated liver disease. Hepatology 1998;27:166–74.
122. Lindor KD, Kowdley KV, Heathcote EJ, et al. Ursodeoxycholic acid for treatment
of nonalcoholic steatohepatitis: results of a randomized trial. Hepatology 2004;
39:770–8.
123. Poupon RE, Bonnand AM, Chretien Y, et al. Ten-year survival in ursodeoxycholic
acid-treated patients with primary biliary cirrhosis. The UDCA-PBC Study
Group. Hepatology 1999;29:1668–71.
124. Poupon RE, Lindor KD, Pares A, et al. Combined analysis of the effect of treat-
ment with ursodeoxycholic acid on histologic progression in primary biliary cir-
rhosis. J Hepatol 2003;39:12–6.
125. Stiehl A. Ursodeoxycholic acid in the treatment of primary sclerosing cholangi-
tis. Ann Med 1994;26:345–9.
126. Thompson K, Maltby J, Fallowfield J, et al. Interleukin-10 expression and func-
tion in experimental murine liver inflammation and fibrosis [see comments].
Hepatology 1998;28:1597–606.
127. Nelson DR, Tu Z, Soldevila-Pico C, et al. Long-term interleukin 10 therapy in
chronic hepatitis C patients has a proviral and anti-inflammatory effect. Hepatol-
ogy 2003;38:859–68.
128. Albano E. Oxidative mechanisms in the pathogenesis of alcoholic liver disease.
Mol Aspects Med 2008;29:9–16.
129. Brown KE, Poulos JE, Li L, et al. Effect of vitamin E supplementation on hepatic
fibrogenesis in chronic dietary iron overload. Am J Physiol 1997;272:G116–23.
130. Hasegawa T, Yoneda M, Nakamura K, et al. Plasma transforming growth factor-
beta1 level and efficacy of alpha-tocopherol in patients with non-alcoholic stea-
tohepatitis: a pilot study. Aliment Pharmacol Ther 2001;15:1667–72.
131. Houglum K, Venkataramani A, Lyche K, et al. A pilot study of the effects of d-al-
pha-tocopherol on hepatic stellate cell activation in chronic hepatitis C. Gastro-
enterology 1997;113:1069–73.
132. Mezey E, Potter J, Rennie-Tankersley L, et al. A randomized placebo controlled
trial of vitamin E in alcoholic hepatitis. Hepatology 2003;38:264A.
133. Stewart S, Prince M, Bassendine M, et al. A trial of antioxidant therapy alone or
with corticosteroids in acute alcoholic hepatitis. J Hepatol 2002;36(S):16.
134. Anonymous. The results of a randomized double blind controlled trial evaluating
malotilate in primary biliary cirrhosis. A European multicentre study group.
J Hepatol 1993;17:227–35.
135. Rambaldi A, Gluud C. Meta-analysis of propylthiouracil for alcoholic liver dis-
ease—a Cochrane Hepato-Biliary Group Review. Liver 2001;21:398–404.
136. Bodenheimer HC Jr, Schaffner F, Sternlieb I, et al. A prospective clinical trial
of D-penicillamine in the treatment of primary biliary cirrhosis. Hepatology
1985;5:1139–42.
137. Dickson ER, Fleming TR, Wiesner RH, et al. Trial of penicillamine in advanced
primary biliary cirrhosis. N Engl J Med 1985;312:1011–5.
960 Rockey
138. Schaff Z, Lapis K, Szende B, et al. The effect of D-penicillamine on CCl4-

induced experimental liver cirrhosis. Exp Pathol 1991;43:111–20.
139. Lu SC, Tsukamoto H, Mato JM. Role of abnormal methionine metabolism in
alcoholic liver injury. Alcohol 2002;27:155–62.
140. Mato JM, Camara J, Fernandez de Paz J, et al. S-adenosylmethionine in alco-
holic liver cirrhosis: a randomized, placebo-controlled, double-blind, multicenter
clinical trial. J Hepatol 1999;30:1081–9.
141. Aithal GP, Haugk B, Das S, et al. Monitoring methotrexate-induced hepatic
fibrosis in patients with psoriasis: are serial liver biopsies justified? Aliment
Pharmacol Ther 2004;19:391–9.
142. Nohlgard C, Rubio CA, Kock Y, et al. Liver fibrosis quantified by image
analysis in methotrexate-treated patients with psoriasis. J Am Acad Dermatol
1993;28:40–5.
143. Te HS, Schiano TD, Kuan SF, et al. Hepatic effects of long-term methotrexate use
in the treatment of inflammatory bowel disease. Am J Gastroenterol 2000;95:
3150–6.
144. Bach N, Bodian C, Bodenheimer H, et al. Methotrexate therapy for primary bil-
iary cirrhosis. Am J Gastroenterol 2003;98:187–93.
145. Bach N, Thung SN, Schaffner F. The histologic effects of low-dose methotrexate
therapy for primary biliary cirrhosis. Arch Pathol Lab Med 1998;122:342–5.
146. Hendrickse MT, Rigney E, Giaffer MH, et al. Low-dose methotrexate is ineffec-
tive in primary biliary cirrhosis: long-term results of a placebo-controlled trial.
[see comments]. Gastroenterology 1999;117:400–7.
147. Kaplan MM, DeLellis RA, Wolfe HJ. Sustained biochemical and histologic remis-
sion of primary biliary cirrhosis in response to medical treatment. Ann Intern
Med 1997;126:682–8.
148. Kaplan MM, Cheng S, Price LL, et al. A randomized controlled trial of colchicine
plus ursodiol versus methotrexate plus ursodiol in primary biliary cirrhosis: ten-
year results. Hepatology 2004;39:915–23.
149. Bahcecioglu IH, Koca SS, Poyrazoglu OK, et al. Hepatoprotective effect of
Infliximab, an anti-TNF-alpha agent, on carbon tetrachloride-induced hepatic
fibrosis. Inflammation 2008;31:215–21.
150. Koca SS, Bahcecioglu IH, Poyrazoglu OK, et al. The treatment with antibody of
TNF-alpha reduces the inflammation, necrosis and fibrosis in the non-alcoholic
steatohepatitis induced by methionine- and choline-deficient diet. Inflammation
2008;31:91–8.
151. Akriviadis E, Botla R, Briggs W, et al. Pentoxifylline improves short-term survival
in severe acute alcoholic hepatitis: a double-blind, placebo-controlled trial. Gas-
152. Menon KV, Stadheim L, Kamath PS, et al. A pilot study of the safety and tolera-
bility of etanercept in patients with alcoholic hepatitis. Am J Gastroenterol 2004;
99:255–425.
153. Spahr L, Rubbia-Brandt L, Frossard JL, et al. Combination of steroids with inflix-
imab or placebo in severe alcoholic hepatitis: a randomized controlled pilot
study. J Hepatol 2002;37:448–55.
154. Tilg H, Jalan R, Kaser A, et al. Anti-tumor necrosis factor-alpha monoclonal an-
tibody therapy in severe alcoholic hepatitis. J Hepatol 2003;38:419–25.
155. Boetticher NC, Piene CJ, Kwo P, et al. Gastroenterology 2008;134:A765.
156. George J, Roulot D, Koteliansky VE, et al. In vivo inhibition of rat stellate cell ac-
tivation by soluble TGF beta type II receptor: a potential new therapy for hepatic
fibrosis. Proc Nat Acad Sci U S A 1999;96:12719–24.
157. Isaka Y, Brees DK, Ikegaya K, et al. Gene therapy by skeletal muscle expression
of decorin prevents fibrotic disease in rat kidney. Nat Med 1996;2:418–23.
158. Yata Y, Gotwals P, Koteliansky V, et al. Dose-dependent inhibition of hepatic fi-
brosis in mice by a TGF-beta soluble receptor: implications for antifibrotic ther-
apy. Hepatology 2002;35:1022–30.
159. Siegmund SV, Schwabe RF. Endocannabinoids and liver disease. II. Endocan-
nabinoids in the pathogenesis and treatment of liver fibrosis. Am J Physiol Gas-
trointest Liver Physiol 2008;294:G357–62.
160. Hezode C, Roudot-Thoraval F, Nguyen S, et al. Daily cannabis smoking as a risk
factor for progression of fibrosis in chronic hepatitis C. Hepatology 2005;42:
63–71.
receptor CB2 in the liver. Gastroenterology 2005;128:742–55.
162. Teixeira-Clerc F, Julien B, Grenard P, et al. CB1 cannabinoid receptor antag-
onism: a new strategy for the treatment of liver fibrosis. Nat Med 2006;12:
671–6.
163. Corpechot C, Barbu V, Wendum D, et al. Hypoxia-induced VEGF and collagen I
cirrhosis. Hepatology 2002;35:1010–21.
164. Tugues S, Fernandez-Varo G, Munoz-Luque J, et al. Antiangiogenic treatment
with sunitinib ameliorates inflammatory infiltrate, fibrosis, and portal pressure
in cirrhotic rats. Hepatology 2007;46:1919–26.
165. Buck M, Chojkier M. A ribosomal S-6 Kinase-mediated signal to C/EBP-beta is
critical for the development of liver fibrosis. PLoS ONE 2007;2:e1372.
166. Khimji AK, Shao R, Rockey DC. Divergent transforming growth factor-beta sig-
naling in hepatic stellate cells after liver injury: functional effects on ECE-1 reg-
ulation. Am J Pathol 2008;173:716–27.
167. Beljaars L, Molema G, Schuppan D, et al. Successful targeting to rat hepatic
stellate cells using albumin modified with cyclic peptides that recognize the col-
lagen type VI receptor. J Biol Chem 2000;275:12743–51.
168. Douglass A, Wallace K, Parr R, et al. Antibody-targeted myofibroblast apoptosis
reduces fibrosis during sustained liver injury. J Hepatol 2008;49:88–98.
169. Du SL, Pan H, Lu WY, et al. Cyclic Arg-Gly-Asp peptide-labeled liposomes for
targeting drug therapy of hepatic fibrosis in rats. J Pharmacol Exp Ther 2007;
322:560–8.
170. Gonzalo T, Beljaars L, van de Bovenkamp M, et al. Local inhibition of liver fibro-
sis by specific delivery of a platelet-derived growth factor kinase inhibitor to he-
patic stellate cells. J Pharmacol Exp Ther 2007;321:856–65.
171. Hagens WI, Beljaars L, Mann DA, et al. Cellular targeting of the apoptosis-induc-
ing compound gliotoxin to fibrotic rat livers. J Pharmacol Exp Ther 2008;324:
902–10.
172. Luk JM, Zhang QS, Lee NP, et al. Hepatic stellate cell-targeted delivery of M6P-
HSA-glycyrrhetinic acid attenuates hepatic fibrogenesis in a bile duct ligation rat
model. Liver Int 2007;27:548–57.
173. Akinc A, Zumbuehl A, Goldberg M, et al. A combinatorial library of lipid-like
materials for delivery of RNAi therapeutics. Nat Biotechnol 2008;26:561–9.
174. Sato Y, Murase K, Kato J, et al. Resolution of liver cirrhosis using vitamin A-cou-
pled liposomes to deliver siRNA against a collagen-specific chaperone. Nat
Biotechnol 2008;26:431–42.
175. Wolfrum C, Shi S, Jayaprakash KN, et al. Mechanisms and optimization of in vivo
delivery of lipophilic siRNAs. Nat Biotechnol 2007;25:1149–57.
962 Rockey
176. van Rooij E, Sutherland LB, Qi X, et al. Control of stress-dependent cardiac

growth and gene expression by a microRNA. Science 2007;316:575–9.
177. Fiorucci S, Antonelli E, Rizzo G, et al. The nuclear receptor SHP mediates
inhibition of hepatic stellate cells by FXR and protects against liver fibrosis.
178. Liu Y, Binz J, Numerick MJ, et al. Hepatoprotection by the farnesoid X receptor
agonist GW4064 in rat models of intra- and extrahepatic cholestasis. J Clin
Invest 2003;112:1678–87.
179. Rockey DC. Antifibrotic therapy in chronic liver disease. Clin Gastroenterol
Hepatol 2005;3:95–107.

Clinics in Liver Disease, Vol 12 PDF

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Clinics in Liver Disease, Vol 12 PDF

Încărcat de

Drepturi de autor:

Formate disponibile

Hepatic Fibrosis: Pathogenesis, Diagnosis, and Emerging Therapies

Clin Liver Dis 12 (2008) xiii–xiv

This work was supported by NIH Grant DK-34238.

Clin Liver Dis 12 (2008) 733–746

SOURCES AND LIMITATIONS OF GLOBAL EPIDEMIOLOGIC DATA OF CHRONIC LIVER DISEASE

GLOBAL BURDEN OF CHRONIC LIVER DISEASE

Ischemic heart disease HIV/AIDS

Self-inflected injuries Ischemic heart disease

Road traffic accidents Tuberculosis

Trachea, bronchus, and lung cancers Road traffic accidents

Cerebrovascular disease Cerebrovascular disease

Cirrhosis of the liver 4.4% Self inflicted injuries

Breast cancer Violence

Colon and rectal cancers Lower respiratory infections

Diabetes mellitus Cirrhosis of the liver 2.2%

Stomach cancer Chronic obstructive pulmonary disease

Hepatocellular Carcinoma (HCC)

projected to increase through 2020.11 The number of new cases is estimated to be

BURDEN OF CHRONIC LIVER DISEASE IN THE UNITED STATES

Morbidity and Economic Burden of Chronic Liver Diseases

GROWING BURDEN OF NONALCOHOLIC FATTY LIVER DISEASE

Nonalcoholic fatty liver disease (NAFLD) is a clinicopathological condition that com-

Hepatic fibrosis, or scarring of the liver, is a nonspecific reaction in response to chronic

Clin Liver Dis 12 (2008) 747–757

culminating in cirrhosis, liver failure, or hepatocellular carcinoma, respectively. Hence,

Fig. 1. Ethnic differences in the prevalence of cryptogenic cirrhosis. Prevalence rates of

Fig. 2. Genetic predisposition to organ-specific endpoints of alcoholism. Medical records of

GENOME-WIDE QUANTITATIVE TRAIT LOCI ANALYSIS

Analysis of polygenic diseases and identification of disease-related gene loci have

TRANSLATIONAL STUDIES FROM MICE TO HUMANS

Hillebrandt and colleagues7 performed a systematic inbred strain survey to identify

Mouse QTL Chromosome Candidate Gene(s)

ASSOCIATION (IN SILICO) MAPPING IN MICE

An alternative way to identify genes or genomic regions linked to a specific phenotype

HUMAN ASSOCIATION (CASE-CONTROL) STUDIES

In contrast to mouse models, which can be studied under defined environmental

Gene Mechanism Liver Disease

candidate genes on the progression of hepatic fibrosis or development of cirrhosis in

Odds Ratio for Individual Gene

Using logistic regression analysis, a predictive equation could be developed and

1. Gressner AM. The cell biology of liver fibrogenesis—an imbalance of

Clin Liver Dis 12 (2008) 759–768

model of fibrosis, knowledge of cell- and disease-specific matrix synthesis is rudimen-

STELLATE CELLS AND OTHER MYOFIBROBLAST PRECURSORS

Hepatic Stellate Cells

Portal Fibroblasts and Other Resident Liver Fibroblast Populations

Bone Marrow–Derived Myofibroblast Precursors

THE FIBROGENIC HEPATOCYTE CONTROVERSY

BILIARY EPITHELIAL CELLS

SINUSOIDAL ENDOTHELIAL CELLS

14. Beaussier M, Wendum D, Schiffer E, et al. Prominent contribution of portal mes-

34. Kojima T, Takano K, Yamamoto T, et al. Transforming growth factor-beta induces

54. Schaffner F, Popper H. Capillarization of hepatic sinusoids in man. Gastroenterol-

Urtasun R and Conde de la Rosa L contributed equally.

Clin Liver Dis 12 (2008) 769–790

RELEVANCE OF OXIDATIVE STRESS IN LIVER DISEASE

SYNERGISM BETWEEN REACTIVE OXYGEN SPECIES AND GROWTH FACTORS

and collagen I.73–76 TNFa counteracts TGFb-stimulation of collagen I gene in different

EFFECTS OF REACTIVE OXYGEN SPECIES ON THE COL1A1 AND COL1A2 PROMOTERS

to several heritable and acquired disorders of connective tissue,93 in general, and

Sp1/Sp3 proteins bound to the COL1A2 promoter interact with CBF/NFY to

STABILITY AND DEGRADATION OF COLLAGEN I: ROLE OF METALLOPROTEINASES

Liver fibrosis is characterized by activation of HSC and subsequent ECM synthesis,