Learning Center
Plans & pricing Sign in
Sign Out

Reversibility of liver fibrosis


                                     Reversibility of liver fibrosis.                            CONCISE
                                                                                     , 2009; 8 (4): 283-291       REVIEW
                                                                                          October-December, Vol. 8 No.4, 2009: 283-291

                                          Reversibility of liver fibrosis
                                                Prakash Ramachandran,*John P. Iredale*

                         * Centre for Inflammation Research to MRC, University of Edinburgh Centre for Inflammation Research


            Liver cirrhosis is a major cause of morbidity and mortality worldwide and has very limited therapeutic op-
            tions. Regardless of the aetiology, hepatic fibrosis is a characteristic feature of chronic liver disease. Our
            knowledge regarding the pathogenesis of this scarring has grown exponentially in the past 25 years. It has
            now clear that this is a highly dynamic process and the long-held dogma that it is irreversible and relent-
            lessly progressive is now being challenged. In this review, we will summarise the key pathogenic mechanis-
            ms at play and will focus on the evidence demonstrating that liver fibrosis is reversible in humans and
            animal models. In particular, we will examine the role of hepatic stellate cells, MMPs, TIMPs and macro-
            phages in this process. Finally, we will discuss some of the studies aimed to therapeutically target the re-
            solution of fibrosis and their potential for translation into a badly-needed treatment modality in the
            clinical setting.

Key words. TIMP. MMP. Macrophage. Stellate cell. Apoptosis.

                      INTRODUCTION                                           of complications and removal of the aetiological
                                                                             agent, barring liver transplantation for a select few.
   Liver fibrosis is the wound healing response of                           With an increasing incidence of cirrhosis and an al-
the liver to a variety of insults. In clinical practice                      ready severe shortage of donor organs, it is clear no-
these injurious stimuli commonly take the form of                            vel therapies are vital.
chronic alcohol abuse, chronic viral infection with                             The progression from liver fibrosis to cirrhosis
Hepatitis C or B, Non Alcoholic Fatty Liver Disease                          was long thought to be inexorable. However, recent
(NAFLD) in association with obesity and the meta-                            research into the mechanisms of fibrogenesis has
bolic syndrome, parasitic infection and a series of                          highlighted the dynamic nature of the process and is
other autoimmune and metabolic conditions. If these                          beginning to challenge this dogma. In this concise
insults remain chronically unchecked, ultimately                             review, we will focus on the evidence that hepatic fi-
the scarring can progress to liver cirrhosis, with                           brosis is potentially reversible and highlight possi-
the pathological hallmarks of advanced fibrosis, no-                         ble therapeutic strategies that may prove invaluable
dular hepatocyte regeneration, and significant ar-                           in the future.
chitectural disruption. Such patients become
susceptible to the complications of chronic liver di-                               HEPATIC MYOFIBROBLASTS
sease, including portal hypertension, hepatocellular                           ARE THE KEY SCAR PRODUCING CELLS
carcinoma and liver failure. Unfortunately current                                    IN THE FIBROTIC LIVER
therapeutic options remain limited to management
                                                                                Following injury to the liver, there is cellular da-
                                                                             mage, activation of the resident inflammatory cells
Correspondence and reprint request: Professor John P Iredale.                as well as recruitment of additional inflammatory
Professor of Medicine                                                        cells. In the acute setting the extraordinary homeos-
Tissue Injury and Regeneration Group
Centre for Inflammation Research
                                                                             tatic and regenerative capacity of the liver takes
to MRD/ University of Edinburgh                                              over, and normal architecture is restored with only
47 Little France Crescent, Edinburgh, United Kingdom, EH16 4TJ               transient extracellular matrix (ECM) deposition ob-
Tel.: 0131-242-6686                                                          served.1,2
Fax: 0131-242-6578
                                                                                With repeated and chronic injury, inflammation is
                                 Manuscript received: November 12, 2009.     persistent or iterative and the normal wound hea-
                                 Manuscript accepted: November 12, 2009.     ling response becomes aberrant with excessive depo-
                                  Ramachandran P. et al.           , 2009; 8 (4): 283-291

sition of ECM, primarily fibrillar collagens I and         some of these enzymes may in fact serve to further
III.3,4 The principal cellular source of this ECM has      enhance HSC activation, it is clear that there is sig-
been identified as the activated hepatic stellate cell     nificant matrix degrading potential already present
(HSC). When quiescent, HSCs reside within the spa-         within the fibrotic liver. So what is it that keeps
ce of Disse and principally store retinoids. Follo-        these enzymes in check during progressive fibrosis?
wing injury, HSCs undergo transdifferentiation into           Tissue Inhibitors of Metalloproteinases (TIMPs)
an activated myofibroblast-like phenotype, associa-        are potent inhibitors of MMPs. During activation
ted with loss of the retinoid droplets and expression      HSCs upregulate TIMP-1 and TIMP-2, even prior to
of the filament alpha smooth muscle actin (α-              expression of collagen I. This HSC derived TIMP-1,
SMA).5-8 They then adopt a number of pro-fibrotic          inhibits secreted MMP-2 and MMP-9 activity more
features including proliferation, contractility, ECM       than 20-fold.22,23 Subsequent studies demonstrated
synthesis, secretion of pro-inflammatory mediators         elevated levels of TIMP-1 and TIMP-2 in progressive
and inhibition of matrix degradation (see below).9         fibrosis in animal models as well as explanted hu-
Thus, they are critical in the progression of liver fi-    man liver.21,22,24 This has even lead to the proposed
brosis.                                                    use of TIMP-1 as a serum marker of fibrosis.25 An
   In addition to HSCs, other sources of hepatic           elegant study Yoshiji et al confirmed a functional
myofibroblasts have also been identified. Portal           role for TIMP-1 in fibrosis progression. By utilising
myofibroblasts have been defined as a distinct popu-       a transgenic mouse model with liver specific TIMP-1
lation, which may have enhanced significance in            overexpression, they demonstrated that no increase
chronic cholestatic disease or periportal fibrosis.10-12   in basal fibrosis was seen but following injury the
Bone-marrow derived myofibroblasts also contribute         TIMP-1 overexpressing mouse had markedly increa-
significantly to liver fibrosis. This interesting fin-     sed levels of scarring.26 This combination of data
ding was demonstrated by studying male patients            has lead to the hypothesis that the overall MMP-
who had received a liver transplant from a female          TIMP balance is critical for fibrosis progression.
donor, with recurrent fibrosis in the graft. The au-
thors were able to demonstrate that a significant                          IS HUMAN LIVER
proportion of the myofibroblasts in these livers were                   FIBROSIS REVERSIBLE?
Y-chromosome positive, confirming their origin
from circulating cells.13 This has subsequently been          A careful review of the literature reveals that ane-
confirmed in murine models including tangible evi-         cdotal evidence for the reversibility of liver fibrosis
dence that these cells are functional in vivo.14 Epi-      has existed for over 30 years. Subsequently, the de-
thelial-mesenchymal transition (EMT), where                velopment of effective treatments for chronic hepati-
hepatic myofibroblasts arise from differentiation of       tis B and hepatitis C lead to clinical trial data with
hepatocytes or cholangiocytes, has also been propo-        liver biopsies before and after therapy. These studies
sed in liver fibrosis,15-17 although the relative func-    convincingly demonstrated, for the first time in lar-
tional contribution of these cells remains to be           ge cohorts of patients, that resolution of fibrosis
elucidated.                                                was a real phenomenon.27-29 Similar findings have
                                                           also been demonstrated in chronic liver diseases of
   FIBROSIS PROGRESSION RESULTS                            other aetiologies including alcohol,30 autoimmune
FROM A FAILURE IN MATRIX DEGRADATION                       hepatitis,31,32 chronic biliary obstruction,33 heredita-
                                                           ry haemochromatosis34 and NAFLD.35 The diversity
   Matrix metalloproteinases (MMPs) are a group of         in the range of conditions represented in this list
enzymes capable of degrading different components          would suggest that reversibility is potentially gene-
of the ECM. Studies in whole liver of both animal          ric rather than disease specific and raises hope that
models and human cirrhosis have demonstrated an            a therapeutic intervention could be effective.
increase in MMPs active against a range of ECM
proteins.2,18-20 Furthermore, tissue culture studies of       MYOFIBROBLAST APOPTOSIS AND
isolated HSCs have shown that when activated they            MATRIX DEGRADATION ARE CENTRAL
can express a plethora of these MMPs. Early during           FEATURES OF FIBROSIS RESOLUTION
activation, transient upregulation of MMP-13 (co-
llagenase-3) and MMP-3 (stromelysin) is seen, follo-          Rodent studies of bile duct ligation (BDL) follo-
wed by increases in MMP-2 (gelatinase A), MMP-9            wed by bilio-jejunal anastamosis or chronic carbon
(gelatinase B) and MMP-14 (MT-MMP).21 Whilst               tetrachloride (CCl4) administration followed by cess-
                             Reversibility of liver fibrosis.         , 2009; 8 (4): 283-291

ation of dosing are both excellent models to study              solution. Cells of the monocyte/macrophage lineage
the dynamics of fibrogenesis and spontaneous reso-              are central in this. Macrophages are capable of
lution. In each of these situations, an established             adopting different activation states in response to di-
hepatic fibrosis can resolve to near normal liver ar-           fferent environmental cues, including pro-and anti-
chitecture within 4-6 weeks.36,37 Two key histologi-            inflammatory and wound healing phenotypes.46
cal changes are observed during this resolution                 Zamara and colleagues have demonstrated that mice
phase: degradation of fibrotic matrix (see below) and           lacking the monocyte chemoattractant protein MCP-
a dramatic loss of the activated HSC/myofibroblast              1, have reduced levels of hepatic inflammation follo-
population. Closer analysis indicates that the HSCs             wing acute injury. 47 Furthermore, inhibition of
are lost by the process of programmed cell death, or            MCP-1 with a dominant-negative mutant followed
apoptosis.36,37 This finding lead to a series of in vitro       by chronic hepatic injury with dimethylnitrosamine
studies examining the signals regulating HSC apop-              in a rat model, clearly shows a reduction in liver
tosis.38 These signals include pro-apoptotic soluble            macrophage infiltration and protection from fibroge-
mediators such as Nerve Growth Factor (NGF)                     nesis. 48 The seminal work in this field was publis-
which can be released from infiltrating inflamma-               hed by Duffield, et al.49 Using a transgenic mouse
tory cells or regenerating hepatocytes.39,40 Further-           model enabling specific depletion of macrophages,
more, direct cellular contact between macrophages               they were able to mirror the above findings,
and activated HSCs has been demonstrated to pro-                showing a reduction in fibrosis and myofibroblast
mote HSC death.41 Intriguingly, interactions bet-               number by depletion during progressive injury with
ween the ECM itself and HSCs can modulate                       CCl4. Intriguingly, depletion during spontaneous re-
apoptosis. Binding between collagen-1 and HSCs                  solution abrogated the improvement in fibrosis, sug-
promote activation whereas loss of integrin-media-              gesting that macrophages also have a critical
ted contact can lead to the HSCs entering the apop-             involvement in matrix remodelling. This data
totic pathway.42 Similar findings have been                     showed, for the first time, the existence of functio-
demonstrated in vivo by Issa and colleagues, where              nally distinct subpopulations of macrophages in the
a transgenic mouse expressing a non-degradable                  liver during injury and recovery.
form of collagen-1 showed a failure of spontaneous                 The mechanism of these macrophage driven
fibrosis resolution and a persistence of activated              effects has also been the subject of scrutiny (Figure
HSCs in the scar following chronic CCL 4.43 Taken               1). Further analysis has revealed that following in-
together, this evidence highlights the critical impor-          jury, there is a contribution to the hepatic macro-
tance of the myobroblast-matrix interaction in fibro-           phage pool from both liver resident Kupffer cells as
sis resolution-ECM degradation will result in                   well as circulating bone-marrow derived cells.49 In
reduced survival/increased apoptotic signals to the             addition, review of the histology shows that the ma-
HSCs, which in turn will undergo cell death leading             crophages are located closely associated with the he-
to a reduction in ECM production.                               patic scar, again confirming their central role.49,50
   Clearly, matrix degradation is required in addi-             In vitro studies have demonstrated that macropha-
tion to HSC apoptosis for adequate fibrosis resolu-             ges are capable of promoting HSC activation via
tion. Once again studies in animal models have                  TGF-β production51 and increasing PDGF receptor
given us critical clues into this process. Studying a           expression,52 potentially explaining their profibrotic
series of timepoints during fibrosis resolution indi-           role. Conversely, they can also promote HSC apop-
cates a significant reduction in the levels of TIMP-1           tosis via expression of molecules such as TRAIL41 or
and TIMP-236,44 with an associated increase in ove-             MMP-9. 42 Moreover, Fallowfield and colleagues
rall collagenase activity due to a switch in the                showed that scar-associtated macrophages in the li-
MMP-TIMP balance, resulting in net ECM degrada-                 ver are the predominant source of MMP-13, the ma-
tion.44 Furthermore, TIMP-1 in itself is a pro-survi-           jor rodent interstitial collagenase, and that
val signal for HSCs,45 highlighting the importance              MMP-13-deficient mice fail to resolve the scar.50
of this reduction in concentration during resolution.           These findings raise the possibility of manipulating
                                                                macrophages as a therapeutic modality.
              MACROPHAGES IN                                       The relative contribution of other cells of the in-
            FIBROSIS RESOLUTION                                 nate and adaptive immune system in liver fibrosis is
                                                                less well established, although a complex interplay
   Inflammatory cells, both resident and recruited,             clearly exists.1 Inhibition of neutrophil recruitment,
also have a key role to play in both fibrosis and re-           using either anti-neutrophil serum53 or CXCR2
                                       Ramachandran P. et al.                  , 2009; 8 (4): 283-291

                                   PROGRESIVE FIBROSIS

  Quiescent HSC                     Activated HSC

                  HSC activation                   HSC proliferation                                 Figure 1. The central role of Scar
                                                                                                 Associated Macrophages (SAMs) in
                              TG-β                PDGF                                           liver fibrosis progression and resolu-
                                                                                                 tion. During progressive liber fibro-
                                                                                                 sis, SAMs can produce factors
                                                                                                 enhancing hepatic stellate cell (HSC)
                                                                                                 activation (e.g. TGF-β) and prolife-
                                                                                                 ration (e.g. Platelet Derived Growth
                                                                                                 Factor (PDGF), leading to a popula-
                         Scar Associated Macrophage                                              tion of matrix-producing activated
                                                                                                 myofibroblasts. During the resolu-
                                                                                                 tion phase, SAMs enhance loss of
                    TRAIL                                MMP-13               ECM                HSCs by apoptosis (e.g. via expres-
                                                                              Degradation        sion of death ligands such as TRAIL)
                                                                                                 and have a direct effect on extrace-
                  HSC apoptosis                                                                  llular matrix (ECM) degradation by
                                                                                                 expression of matrix metalloprotei-
                                                    Loss of HSC-ECM
                                                                                                 nases such as MMP-13. ECM degra-
                                                                                                 dtion further enhaces HSC apoptosis
                                       RESOLUTION                                                by loss of survival signals mediated
                                                                                                 by HSC-ESC interactions.

knockout mice54 seems to have minimal effect on fi-                    ced fibrosis, but also encompasses abnormal hepato-
brosis. In a B Cell deficient mouse, however, there                    cyte regeneration, architectural disturbance and
is reduced liver fibrosis in response to chronic CCL4                  vascular changes. Once again experimental models
administration, while no difference was seen in T                      have given us insight into this key question. Issa et
cell deficient mice by this group.55 A proposed me-                    al induced cirrhosis in rats by 12 weeks of CCL4 ad-
chanism for this finding is that there may be enhan-                   ministration, followed by a protracted recovery pe-
ced clearance of apoptotic hepatocytes by                              riod of up to 1 year.44 After 12 weeks of injury,
macrophages.56 On the other hand, B cell deficient                     micronodules were clearly visible surrounded by
mice show increased hepatic fibrosis in response to                    dense mature fibrotic septa. Following recovery, sig-
chronic schistosoma infection57 whilst adoptive                        nificant remodelling was observed, primarily in the
transfer of CD8+ve T Cells from CCl4 treated mice                      freshly formed fibrosis immediately adjacent to
into immunodeficient SCID mice causes activation                       the nodules. The more mature scars, however, failed
of HSCs.58 Such findings demonstrate the difficulty                    to resolve completely even after a prolonged period.
in unpicking the key mechanisms involved and it is                     By comparing the reversible with the potentially
certain that disease- or model-specific factors will be                irreversible scars, some of the features leading to
at play. There also remains a lack of mechanistic                      “irreversibility” could be defined. Principally, the
data on the relative contribution of these cell types                  irreversible septae contained greater quantities of
during the resolution from fibrosis.                                   the ECM component elastin. Furthermore, tissue
                                                                       transglutaminase mediated matrix cross-linking
              WHAT RENDERS                                             could be demonstrated within the irreversible
          FIBROSIS IRREVERSIBLE?                                       scars.44 This would suggest the possibility that the
                                                                       more mature scar contains cross-linked matrix ren-
   Having described some of the evidence confirming                    dering it resistant to proteolytic degradation, simi-
the reversibility of liver fibrosis in human and ani-                  lar to fibrosis in other organ systems.59 The other
mal models and the key mechanistic events in this                      distinguishing feature of the irreversible scars was
process, what remains uncertain is whether advan-                      that they were relatively paucicellular, containing
ced cirrhosis can return to normal liver architecture                  fewer active myofibroblasts and macrophages than
or whether there is a “point of no return”. It is im-                  their reversible counterparts.44 This again could be
portant to note that cirrhosis is not simply advan-                    critical, the implication being that the absence of ce-
                             Reversibility of liver fibrosis.         , 2009; 8 (4): 283-291

lls within the scar leads to reduced delivery of ma-            the management of type 2 diabetes and the in vitro
trix degrading enzymes to the area.                             findings tally nicely with the recently published cli-
   Similar evidence in human liver disease is still             nical trials in NAFLD, demonstrating histological
lacking. However, a study by Wanless et al looking              improvement in the liver following treatment.65-67
at explanted cirrhotic human livers, did indicate               Angiotensin II also has an activating effect on
partial regression even from advanced micronodular              HSCs, probably mediated via TGF-β expression, and
cirrhosis. 60 Strikingly, the histological patterns             blockade of the renin-angiotensin system seems to
were remarkably similar to that described in the rat            ameliorate experimental liver fibrosis68-70 as well as
model,44 suggesting common mechanisms might be                  potentially having benefits in human disease.71 This
prevalent.                                                      correlates well with the reported effects on cardiac
                                                                and renal fibrosis.72,73
         TARGETED THERAPIES                                        Given critical nature of HSC apoptosis in fibrosis
   IN FIBROGENESIS AND RESOLUTION                               resolution, several authors have attempted to stimu-
                                                                late this as an anti-fibrotic therapy. As discussed
   As described there is extensive data on the rever-           above activated HSCs respond to a variety of apop-
sibility of fibrosis and at least in part of cirrhosis.         totic signals and express several death receptors in-
This raises hopes that an effective anti-fibrotic the-          cluding p75, FAS and TNF-α, in addition to the
rapy can soon be translated into the clinic. As our             critical interaction with ECM.2 Attempts to specifi-
understanding of the pathogenic mechanisms has                  cally target this have been made, but are limited by
evolved, numerous researchers have attempted to                 potential off target effects due to the pleotropic natu-
target different fibrogenic pathways. A review of the           re of these molecules. An observation that activated
literature would reveal hundreds of such anti-fibro-            HSCs persistently express high levels of NF-κB, a
tic interventions in animal models. These can                   pro-inflammatory transcription factor, and that this
broadly be divided into 3 main categories:                      confers protection from HSC apoptosis has lead to
                                                                additional strategies to target this process.74 Sul-
1. Targeting the injurious stimulus or the inflam-              phasalazine, a widely used anti-inflammatory drug,
   matory process ultimately leading to myofibro-               has been shown to induce HSC apoptosis in vitro
   blast activation and ECM deposition.                         and in vivo via an NF-κB dependent mechanism and
2. Reducing myofibroblast activation and promo-                 accelerates fibrosis resolution in a rat model.75 Glio-
   ting myofibroblast loss by apoptosis.                        toxin, a fungal metabolite, also induces HSC apopto-
3. Modulating the MMP-TIMP balance to favour                    sis, at least in part due to NF-κB inhibition, and
   ECM degradation.                                             thus has anti-fibrotic effects.76 This also had off tar-
                                                                get effects causing depletion of hepatic macrophages.
   Given the importance of the resolution phase be-             To circumvent this problem authors have used an
cause patients invariably present to medical atten-             antibody raised against myofibroblasts to target the
tion with established disease, this review will now             gliotoxin to HSCs and gain a much more specific
focus on the latter 2 aspects.                                  effect.77 This certainly represents a significant ad-
   Various molecules have been shown to reduce the              vance, and similar methods may be applied more
pro-fibrotic nature of HSCs. One particular example             widely in the future.
is relaxin. This hormone is normally produced by                   Having described the importance of the MMP-
the corpus luteum during the 3rd trimester of preg-             TIMP balance in the progression and resolution of
nancy and acts to soften the symphysis pubis. Stu-              fibrosis, not surprisingly various studies have at-
dies using relaxin on activated HSCs have shown                 tempted to target this. Initially attempts were made
that it modulates TIMP and collagen expression in               to increase hepatic MMP levels. Siller-Lopez and co-
vitro and that it is anti-fibrotic in vivo,61 suggesting        lleagues administered adenovirus causing hepatic
that recombinant relaxin or synthetic analogues                 overexpression of MMP-8 (neutrophil collagenase)
may have a therapeutic role. The peroxisome prolife-            and showed that this significantly reduced fibrosis
rator-activated receptors (PPAR) also have an im-               following BDL in rats.78 TIMP inhibition as a thera-
portant role in regulating HSC behaviour. During                peutic avenue has also received some attention. A
HSC activation a marked reduction of PPARγ expre-               TIMP-1 neutralising antibody resulted in reduced
ssion is seen and synthetic PPARγ agonists reduce               progression in a rat CCL4 model.79 Roeb and collea-
HSC proliferation, chemotaxis and collagen-1 expre-             gues used a particularly novel method of TIMP inhi-
ssion.62-64 These drugs are already widely used in              bition. They developed a series of MMP-9 mutants,
                                  Ramachandran P. et al.            , 2009; 8 (4): 283-291

which lacked MMP gelatinolytic activity, but retai-        renal injury, using macrophages either stimulated
ned their TIMP binding properties. These mutants           with cytokines85 or virally transduced to express
were able to inhibit TIMP-1 effects in vitro and in        anti-inflammatory cytokines86,87 prior to adoptive
vivo and reduced fibrosis in a murine model.80,81          transfer. Studies such as these are still lacking in li-
This strategy would once again have the advantage          ver fibrosis models.
of minimising off-target effects, by specifically ha-
ving action in areas where TIMP-1 would be upre-                                  CONCLUSION
gulated during fibrogenesis.
   Clearly a full discussion of the myriad of poten-          Our understanding of the pathogenesis of hepatic
tial anti-fibrotic therapies that have been tried in ex-   fibrosis has progressed exponentially in the past 30
perimental models is beyond the scope of this              years. In particular, we now have cogent and con-
article. However, what is clear from the selected          sistent evidence that fibrosis (and to a lesser degree
examples, is that our greater understanding of the         advanced cirrhosis) is potentially reversible in hu-
pathogenesis has enabled specific and targetted the-       mans and animal models, and have an insight into
rapies to be developed. The challenge remains in the       the critical molecular and cellular pathways invol-
translation of these to the clinical setting, in parti-    ved. We have also gained knowledge of some of the
cular the avoidance of off-target effects which could      features preventing resolution of mature scars. Uti-
potentially cause adverse events in humans.                lising these findings, several novel therapies have
                                                           been trialled in reproducible animal models of liver
    THE POTENTIAL OF CELL THERAPY                          disease. What remains elusive is effective translation
                                                           into the human setting. In the future, cellular the-
   As discussed above, a cardinal feature of the irre-     rapies or treatments targeting multiple points in
versible scar is its hypocellularity. Many groups are      the complex fibrogenic pathway may help to bridge
now exploring the use of cellular therapy for liver fi-    this gap.
brosis. This could potentially have the benefit of see-
ding the irreversible scar, enabling modification of                           ABBREVIATIONS
the local microenvironment and possible degrada-
tion. A variety of cell types have been used for this      •   ECM: Extracellular matrix.
purpose, with a focus on a stem cell fractions.            •   HSC: Hepatic stellate cell.
Sakaida and colleagues, for example, used adoptive         •   MMP: Matrix metalloproteinase.
transfer of GFP-expressing bone marrow cell frac-          •   TIMP: Tissue Inhibitor of metalloproteinases.
tion into mice during progressive CCL4 induced in-         •   TRAIL: TNF-related apoptosis inducing ligand.
jury. They were able to demonstrate a reduction in
fibrosis, and also that the injected cells localised                                 FUNDING
around the scars and co-expressed MMP-9.82 The
implication of this is that they may become pro-reso-         Prakash Ramachandran is funded by The Wellco-
lution scar-associated macrophages. It has also been       me Trust.
proposed that injected stem cell fractions could diffe-       John P Iredale is consultan for GE Healhcare and
rentiate and contribute to the regenerating hepato-        is funded by The Wellcome Trust, Medical Research
cyte pool, although this remains controversial.83          Council United Kingdom and the Childrens Liver
Clinical trials of cell therapy have been conducted-       Disease Foundation.
whilst these are generally small, non-randomised
and do not show unequivocal benefit, they do offer                                   REFERENCES
some hope for the future.84
   A criticism of the cell therapy strategy would be       1. Henderson NC, Iredale JP. Liver fibrosis: cellular mechanis-
                                                              ms of progression and resolution. Clin Sci (Lond) 2007;
that following adoptive transfer, the cells will en-          112(5): 265-80.
counter the existing microenvironment in the liver,        2. Iredale J. Defining therapeutic targets for liver fibrosis:
meaning their effect could be abrogated by phenotypic         exploiting the biology of inflammation and repair. Pharma-
reprogramming according to the local conditions.              col Res 2008; 58(2): 129-36.
                                                           3. Bataller R, Brenner DA. Liver fibrosis. J Clin Invest 2005;
A way to circumvent this problem would be to modi-            115(2): 209-18.
fy the cells ex vivo into an anti-inflammatory or pro-     4. Schuppan D. Structure of the extracellular matrix in nor-
resolution phenotype, prior to delivery. Similar              mal and fibrotic liver: collagens and glycoproteins. Semin
experiments have been effective in animal models of           Liver Dis 1990; 10(1): 1-10.
                                     Reversibility of liver fibrosis.             , 2009; 8 (4): 283-291

5. Friedman SL, Rockey DC, McGuire RF, Maher JJ, Boyles JK,                   talloproteinases-1. Implications for regulation of matrix
    Yamasaki G. Isolated hepatic lipocytes and kupffer cells                  degradation in liver. J Clin Invest 1992; 90(1): 282-7.
    from normal human liver-morphological and functional-cha-             24. Benyon RC, Iredale JP, Goddard S, Winwood PJ, Arthur MJ.
    racteristics in primary culture. Hepatology 1992; 15(2):                  Expression of tissue inhibitor of metalloproteinases 1 and
    234-43.                                                                   2 is increased in fibrotic human liver. Gastroenterology
6. Friedman SL, Roll FJ, Boyles J, Bissell DM. Hepatic lipo-                  1996; 110(3): 821-31.
    cytes: the principal collagen-producing cells of normal rat           25. Patel K, Gordon SC, Jacobson I, Hezode C, Oh E, Smith KM,
    liver. Proc Natl Acad Sci U S A 1985; 82(24): 8681-5.                     et al. Evaluation of a panel of non-invasive serum markers
7. Maher JJ, Bissell DM, Friedman SL, Roll FJ. Collagen measu-                to differentiate mild from moderate-to-advanced liver fi-
    red in primary cultures of normal rat hepatocytes derives                 brosis in chronic hepatitis C patients. J Hepatol 2004;
    from lipocytes within the monolayer. J Clin Invest 1988;                  41(6): 935-42.
    82(2): 450-9.                                                         26. Yoshiji H, Kuriyama S, Miyamoto Y, Thorgeirsson UP,
8. Maher JJ, McGuire RF. Extracellular-matrix gene-expres-                    Gomez DE, Kawata M, et al. Tissue inhibitor of metallo-
    sion increases preferentially in rat lipocytes and sinusoidal             proteinases-1 promotes liver fibrosis development in a
    endothelial-cells during hepatic-fibrosis invivo. J Clin In-              transgenic mouse model. Hepatology 2000; 32(6):
    vest 1990; 86(5): 1641-8.                                                 1248-54.
9. Friedman SL. Hepatic stellate cells: protean, multifunctio-            27. Dienstag JL, Goldin RD, Heathcote EJ, Hann HW, Woess-
    nal, and enigmatic cells of the liver. Physiol Rev 2008;                  ner M, Stephenson SL, et al. Histological outcome during
    88(1): 125-72.                                                            long-term lamivudine therapy. Gastroenterology 2003;
10. Kinnman N, Housset C. Peribiliary myofibroblasts in biliary               124(1): 105-17.
    type liver fibrosis. Front Biosci 2002; 7: 496-503.                   28. Iredale JP. Hepatic stellate cell behavior during resolution
11. Knittel T, Kobold D, Saile B, Grundmann A, Neubauer K, Pisca-             of liver injury. Semin Liver Dis 2001; 21(3): 427-36.
    glia F, et al. Rat liver myofibroblasts and hepatic stellate cells:   29. Poynard T, McHutchison J, Manns M, Trepo C, Lindsay K,
    different cell populations of the fibroblast lineage with fibro-          Goodman Z, et al. Impact of pegylated interferon alfa-2b
    genic potential. Gastroenterology 1999; 117(5): 1205-21.                  and ribavirin on liver fibrosis in patients with chronic he-
12. Tuchweber B, Desmouliere A, Bochaton-Piallat ML, Rubbia-                  patitis C. Gastroenterology 2002; 122(5): 1303-13.
    Brandt L, Gabbiani G. Proliferation and phenotypic modula-            30. Pares A, Caballeria J, Bruguera M, Torres M, Rodes J. His-
    tion of portal fibroblasts in the early stages of cholestatic             tological course of alcoholic hepatitis. Influence of absti-
    fibrosis in the rat. Lab Invest 1996; 74(1): 265-78.                      nence, sex and extent of hepatic damage. J Hepatol
13. Forbes SJ, Russo FP, Rey V, Burra P, Rugge M, Wright NA,                  1986; 2(1): 33-42.
    et al. A significant proportion of myofibroblasts are of              31. Dufour JF, DeLellis R, Kaplan MM. Reversibility of hepatic
    bone marrow origin in human liver fibrosis. Gastroentero-                 fibrosis in autoimmune hepatitis. Ann Intern Med 1997;
    logy 2004; 126(4): 955-63.                                                127(11): 981-5.
14. Russo FP, Alison MR, Bigger BW, Amofah E, Florou A, Amin              32. Serpaggi J, Carnot F, Nalpas B, Canioni D, Guechot J, Le-
    F, et al. The bone marrow functionally contributes to liver               bray P, et al. Direct and indirect evidence for the reversi-
    fibrosis. Gastroenterology 2006; 130(6): 1807-21.                         bility of cirrhosis. Hum Pathol 2006; 37(12): 1519-26.
15. Ikegami T, Zhang Y, Matsuzaki Y. Liver fibrosis: possible             33. Hammel P, Couvelard A, O’Toole D, Ratouis A, Sauvanet A,
    involvement of EMT. Cells Tissues Organs 2007; 185(1-3):                  Flejou JF, et al. Regression of liver fibrosis after biliary
    213-21.                                                                   drainage in patients with chronic pancreatitis and steno-
16. Kaimori A, Potter J, Kaimori JY, Wang C, Mezey E, Koteish                 sis of the common bile duct. N Engl J Med 2001; 344(6):
    A. Transforming growth factor-beta1 induces an epithe-                    418-23.
    lial-to-mesenchymal transition state in mouse hepatocytes             34. Falize L, Guillygomarc’h A, Perrin M, Laine F, Guyader D,
    in vitro. J Biol Chem 2007; 282(30): 22089-101.                           Brissot P, et al. Reversibility of hepatic fibrosis in treated
17. Rygiel KA, Robertson H, Marshall HL, Pekalski M, Zhao L,                  genetic hemochromatosis: a study of 36 cases. Hepatolo-
    Booth TA, et al. Epithelial-mesenchymal transition contri-                gy 2006; 44(2): 472-7.
    butes to portal tract fibrogenesis during human chronic li-           35. Dixon JB, Bhathal PS, O’Brien PE. Weight loss and non-alco-
    ver disease. Lab Invest 2008; 88(2): 112-23.                              holic fatty liver disease: falls in gamma-glutamyl transfera-
18. Arthur MJ, Iredale JP, Mann DA. Tissue inhibitors of meta-                se concentrations are associated with histologic
    lloproteinases: role in liver fibrosis and alcoholic liver di-            improvement. Obes Surg 2006; 16(10): 1278-86.
    sease. Alcohol Clin Exp Res1999; 23(5): 940-3.                        36. Iredale JP, Benyon RC, Pickering J, McCullen M, Northrop
19. Arthur MJ, Mann DA, Iredale JP. Tissue inhibitors of meta-                M, Pawley S, et al. Mechanisms of spontaneous resolution
    lloproteinases, hepatic stellate cells and liver fibrosis. J              of rat liver fibrosis. Hepatic stellate cell apoptosis and re-
    Gastroenterol Hepatol 1998; 13(Suppl.): S33-8.                            duced hepatic expression of metalloproteinase inhibitors.
20. Benyon RC, Arthur MJP. Extracellular matrix degradation                   J Clin Invest 1998; 102(3): 538-49.
    and the role of hepatic stellate cells. Semin Liver Dis 2001;         37. Issa R, Williams E, Trim N, Kendall T, Arthur MJ, Reichen J,
    21(3): 373-84.                                                            et al. Apoptosis of hepatic stellate cells: involvement in re-
21. Iredale JP. Models of liver fibrosis: exploring the dynamic               solution of biliary fibrosis and regulation by soluble growth
    nature of inflammation and repair in a solid organ. J Clin                factors. Gut 2001; 48(4): 548-57.
    Invest 2007; 117(3): 539-48.                                          38. Elsharkawy AM, Oakley F, Mann DA. The role and regula-
22. Iredale JP, Benyon RC, Arthur MJ, Ferris WF, Alcolado R,                  tion of hepatic stellate cell apoptosis in reversal of liver fi-
    Winwood PJ, et al. Tissue inhibitor of metalloproteinase-1                brosis. Apoptosis 2005; 10(5): 927-39.
    messenger RNA expression is enhanced relative to inters-              39. Oakley F, Trim N, Constandinou CM, Ye W, Gray AM, Fran-
    titial collagenase messenger RNA in experimental liver in-                tz G, et al. Hepatocytes express nerve growth factor du-
    jury and fibrosis. Hepatology1996; 24(1): 176-84.                         ring liver injury: evidence for paracrine regulation of
23. Iredale JP, Murphy G, Hembry RM, Friedman SL, Arthur MJ.                  hepatic stellate cell apoptosis. Am J Pathol 2003; 163(5):
    Human hepatic lipocytes synthesize tissue inhibitor of me-                1849-58.
                                        Ramachandran P. et al.                  , 2009; 8 (4): 283-291

40. Trim N, Morgan S, Evans M, Issa R, Fine D, Afford S, et al.       56. Bhogal RK, Bona CA. B cells: no longer bystanders in liver
    Hepatic stellate cells express the low affinity nerve                 fibrosis. J Clin Invest 2005; 115(11): 2962-5.
    growth factor receptor p75 and undergo apoptosis in res-          57. Ferru I, Roye O, Delacre M, Auriault C, Wolowczuk I. Infec-
    ponse to nerve growth factor stimulation. Am J Pathol                 tion of B-cell-deficient mice by the parasite Schistosoma
    2000; 156(4): 1235-43.                                                mansoni: demonstration of the participation of B cells in
41. Fischer R, Cariers A, Reinehr R, Haussinger D. Caspase 9-             granuloma modulation. Scand J Immunol 1998;48(3): 233-40.
    dependent killing of hepatic stellate cells by activated Ku-      58. Safadi R, Ohta M, Alvarez CE, Fiel MI, Bansal M, Mehal WZ,
    pffer cells. Gastroenterology 2002; 123(3): 845-61.                   et al. Immune stimulation of hepatic fibrogenesis by CD8
42. Zhou X, Murphy FR, Gehdu N, Zhang J, Iredale JP, Benyon               cells and attenuation by transgenic interleukin-10 from
    RC. Engagement of alphavbeta3 integrin regulates prolife-             hepatocytes. Gastroenterology 2004; 127(3): 870-82.
    ration and apoptosis of hepatic stellate cells. J Biol Chem       59. Griffin M, Casadio R, Bergamini CM. Transglutaminases:
    2004; 279(23): 23996-4006.                                            nature’s biological glues. Biochem J 2002; 368: 377-96.
43. Issa R, Zhou X, Trim N, Millward-Sadler H, Krane S, Benyon        60. Wanless IR, Nakashima E, Sherman M. Regression of human
    C, et al. Mutation in collagen-1 that confers resistance to           cirrhosis. Morphologic features and the genesis of incom-
    the action of collagenase results in failure of recovery              plete septal cirrhosis. Arch Pathol Lab Med 2000; 124(11):
    from CCl4-induced liver fibrosis, persistence of activated            1599-607.
    hepatic stellate cells, and diminished hepatocyte regene-         61. Williams EJ, Benyon RC, Trim N, Hadwin R, Grove BH, Ar-
    ration. FASEB J 2003; 17(1): 47-9.                                    thur MJ, et al. Relaxin inhibits effective collagen deposi-
44. Issa R, Zhou X, Constandinou CM, Fallowfield J, Millward-             tion by cultured hepatic stellate cells and decreases rat
    Sadler H, Gaca MD, et al. Spontaneous recovery from mi-               liver fibrosis in vivo. Gut 2001;49(4): 577-83.
    cronodular cirrhosis: evidence for incomplete resolution          62. Yavrom S, Chen L, Xiong S, Wang J, Rippe RA, Tsukamoto
    associated with matrix cross-linking. Gastroenterology                H. Peroxisome proliferator-activated receptor gamma su-
    2004; 126(7): 1795-808.                                               ppresses proximal alpha1(I) collagen promoter via inhibi-
45. Murphy F, Waung J, Collins J, Arthur MJ, Nagase H, Mann               tion of p300-facilitated NF-I binding to DNA in hepatic
    D, et al. N-Cadherin cleavage during activated hepatic                stellate cells. J Biol Chem 2005; 280(49): 40650-9.
    stellate cell apoptosis is inhibited by tissue inhibitor of me-   63. Zhao C, Chen W, Yang L, Chen L, Stimpson SA, Diehl AM.
    talloproteinase-1. Comp Hepatol 2004; 3 (Suppl. 1): S8.               PPARgamma agonists prevent TGFbeta1/Smad3-signaling in
46. Mosser DM, Edwards JP. Exploring the full spectrum of ma-             human hepatic stellate cells. Biochem Biophys Res Commun
    crophage activation. Nature Reviews Immunology 2008;                  2006; 350(2): 385-91.
    8(12): 958-69.                                                    64. Marra F, Efsen E, Romanelli RG, Caligiuri A, Pastacaldi S,
47. Zamara E, Galastri S, Aleffi S, Petrai I, Aragno M, Mastro-           Batignani G, et al. Ligands of peroxisome proliferator-ac-
    cola R, et al. Prevention of severe toxic liver injury and            tivated receptor gamma modulate profibrogenic and pro-
    oxidative stress in MCP-1-deficient mice. J Hepatol 2007;             inflammatory       actions    in    hepatic    stellate    cells.
    46(2): 230-8.                                                         Gastroenterology 2000; 119(2): 466-78.
48. Imamura M, Ogawa T, Sasaguri Y, Chayama K, Ueno H. Su-            65. Belfort R, Harrison SA, Brown K, Darland C, Finch J, Har-
    ppression of macrophage infiltration inhibits activation of           dies J, et al. A placebo-controlled trial of pioglitazone in
    hepatic stellate cells and liver fibrogenesis in rats. Gas-           subjects with nonalcoholic steatohepatitis. N Engl J Med
    troenterology 2005; 128(1): 138-46.                                   2006; 355(22): 2297-307.
49. Duffield JS, Forbes SJ, Constandinou CM, Clay S, Partolina        66. Ratziu V, Giral P, Jacqueminet S, Charlotte F, Hartemann-
    M, Vuthoori S, et al. Selective depletion of macrophages              Heurtier A, Serfaty L, et al. Rosiglitazone for nonalcoholic
    reveals distinct, opposing roles during liver injury and re-          steatohepatitis: one-year results of the randomized pla-
    pair. J Clin Invest 2005; 115(1): 56-65.                              cebo-controlled Fatty Liver Improvement with Rosiglita-
50. Fallowfield JA, Mizuno M, Kendall TJ, Constandinou CM,                zone Therapy (FLIRT) Trial. Gastroenterology 2008;
    Benyon RC, Duffield JS, et al. Scar-associated macropha-              135(1): 100-10.
    ges are a major source of hepatic matrix metalloproteina-         67. Aithal GP, Thomas JA, Kaye PV, Lawson A, Ryder SD, Spen-
    se-13 and facilitate the resolution of murine hepatic                 dlove I, et al. Randomized, placebo-controlled trial of pio-
    fibrosis. J Immunol 2007 Apr 15;178(8):5288-95.                       glitazone in nondiabetic subjects with nonalcoholic
51. Matsuoka M, Tsukamoto H. Stimulation of hepatic lipocyte              steatohepatitis. Gastroenterology 2008; 135(4): 1176-84.
    collagen production by Kupffer cell-derived transforming          68. Warner FJ, Lubel JS, McCaughan GW, Angus PW. Liver fibro-
    growth factor beta: implication for a pathogenetic role in al-        sis: a balance of ACEs? Clin Sci (Lond) 2007; 113(3): 109-18.
    coholic liver fibrogenesis. Hepatology 1990; 11(4): 599-605.      69. Paizis G, Gilbert RE, Cooper ME, Murthi P, Schembri JM,
52. Friedman SL, Arthur MJ. Activation of cultured rat hepatic            Wu LL, et al. Effect of angiotensin II type 1 receptor
    lipocytes by Kupffer cell conditioned medium. Direct en-              blockade on experimental hepatic fibrogenesis. J Hepatol
    hancement of matrix synthesis and stimulation of cell pro-            2001; 35(3): 376-85.
    liferation via induction of platelet-derived growth factor        70. Yoshiji H, Kuriyama S, Yoshii J, Ikenaka Y, Noguchi R,
    receptors. J Clin Invest 1989; 84(6): 1780-5.                         Nakatani T, et al. Angiotensin-II type 1 receptor interac-
53. Saito JM, Bostick MK, Campe CB, Xu J, Maher JJ. Infiltra-             tion is a major regulator for liver fibrosis development in
    ting neutrophils in bile duct-ligated livers do not promote           rats. Hepatology 2001; 34(4 Pt 1): 745-50.
    hepatic fibrosis. Hepatol Res 2003; 25(2): 180-91.                71. Yokohama S, Yoneda M, Haneda M, Okamoto S, Okada M,
54. Xu J, Lee G, Wang H, Vierling JM, Maher JJ. Limited role              Aso K, et al. Therapeutic efficacy of an angiotensin II re-
    for CXC chemokines in the pathogenesis of alpha-na-                   ceptor antagonist in patients with nonalcoholic steatohe-
    phthylisothiocyanate-induced liver injury. Am J Physiol               patitis. Hepatology 2004; 40(5): 1222-5.
    Gastrointest Liver Physiol 2004; 287(3): G734-41.                 72. Peng H, Carretero OA, Vuljaj N, Liao TD, Motivala A, Pe-
55. Novobrantseva TI, Majeau GR, Amatucci A, Kogan S, Bren-               terson EL, et al. Angiotensin-converting enzyme inhibi-
    ner I, Casola S, et al. Attenuated liver fibrosis in the ab-          tors: a new mechanism of action. Circulation 2005;
    sence of B cells. J Clin Invest 2005; 15(11): 072-82.                 112(16): 2436-45.
                                   Reversibility of liver fibrosis.           , 2009; 8 (4): 283-291

73. Wolf G. Renal injury due to renin-angiotensin-aldosterone              by matrix metalloproteinase-9 mutants in mice. FASEB J
    system activation of the transforming growth factor-beta               2006; 20(3): 444-54.
    pathway. Kidney Int 2006; 70(11): 19140-9.                         81. Roeb E, Behrmann I, Grotzinger J, Breuer B, Matern S. An
74. Elsharkawy AM, Oakley F, Mann DA. The role and regula-                 MMP-9 mutant without gelatinolytic activity as a novel
    tion of hepatic stellate cell apoptosis in reversal of liver fi-       TIMP-1-antagonist. FASEB J 2000; 14(12): 1671-3.
    brosis. Apoptosis 2005; 10(5): 927-39.                             82. Sakaida I, Terai S, Yamamoto N, Aoyama K, Ishikawa T,
75. Oakley F, Meso M, Iredale JP, Green K, Marek CJ, Zhou X,               Nishina H, et al. Transplantation of bone marrow cells re-
    et al. Inhibition of inhibitor of kappaB kinases stimulates            duces CCl4-induced liver fibrosis in mice. Hepatology
    hepatic stellate cell apoptosis and accelerated recovery               2004; 40(6): 1304-11.
    from rat liver fibrosis. Gastroenterology 2005; 128(1):            83. Kanazawa Y, Verma IM. Little evidence of bone marrow-
    108-20.                                                                derived hepatocytes in the replacement of injured li-
76. Wright MC, Issa R, Smart DE, Trim N, Murray GI, Primrose               ver. Proc Natl Acad Sci U S A 2003; 100(Suppl. 1):
    JN, et al. Gliotoxin stimulates the apoptosis of human and rat         11850-3.
    hepatic stellate cells and enhances the resolution of liver fi-    84. Houlihan DD, Newsome PN. Critical review of clinical trials
    brosis in rats. Gastroenterology 2001; 121(3): 685-98.                 of bone marrow stem cells in liver disease. Gastroentero-
77. Douglass A, Wallace K, Parr R, Park J, Durward E, Broad-               logy 2008; 135(2): 438-50.
    bent I, et al. Antibody-targeted myofibroblast apoptosis           85. Wang Y, Wang YP, Zheng G, Lee VW, Ouyang L, Chang
    reduces fibrosis during sustained liver injury. J Hepatol              DH, et al. Ex vivo programmed macrophages ameliorate
    2008; 49(1): 88-98.                                                    experimental chronic inflammatory renal disease. Kidney
78. Siller-Lopez F, Sandoval A, Salgado S, Salazar A, Bueno M,             Int 2007; 72(3): 290-9.
    Garcia J, et al. Treatment with human metalloproteinase-8          86. Kluth DC, Ainslie CV, Pearce WP, Finlay S, Clarke D, Ane-
    gene delivery ameliorates experimental rat liver cirrhosis.            gon I, et al. Macrophages transfected with adenovirus to
    Gastroenterology 2004; 126(4): 1122-33.                                express IL-4 reduce inflammation in experimental glomeru-
79. Parsons CJ, Bradford BU, Pan CQ, Cheung E, Schauer M,                  lonephritis. J Immunol 2001; 166(7): 4728-36.
    Knorr A, et al. Antifibrotic effects of a tissue inhibitor of      87. Wilson HM, Stewart KN, Brown PA, Anegon I, Chettibi
    metalloproteinase-1 antibody on established liver fibrosis             S, Rees AJ, et al. Bone-marrow-derived macrophages
    in rats. Hepatology 2004; 40(5): 1106-15.                              genetically modified to produce IL-10 reduce injury in
80. Roderfeld M, Weiskirchen R, Wagner S, Berres ML, Henkel                experimental glomerulonephritis. Mol Ther 2002; 6(6):
    C, Grotzinger J, et al. Inhibition of hepatic fibrogenesis             710-7.

To top