SMJ 2003 49(1): 3-6
Dr Jonathan A Fallowfield, MRC Clinical Research Fellow
and
Professor John P Iredale, MRC Senior Clinical Fellow and Professor of Hepatology, Liver Research Group, Division of Infection, Inflammation and Repair, University of Southampton
Correspondence to:
Professor John P Iredale, Liver Research Group, Level D (Mailpoint 811), South Block, Southampton General Hospital, Tremona Road, Southampton SO16 6YD
e-mail: jpi@soton.ac.uk
Funding: Dr Fallowfield and Professor Iredale gratefully acknowledge the support of the MRC. Professor Iredale also acknowledges the support of the CLDF and Wessex Medical Trust (Hope)
Competing interests: None declared
Abstract
Hepatic fibrosis, or scarring of the liver, is the final stage of a wound healing response that invariably follows chronic liver injury of any aetiology. Progressive fibrosis ultimately leads to a gross disruption of the normal liver architecture (cirrhosis) and subsequent impairment of liver function. Cirrhosis is associated with significant and life-threatening complications and is a major cause of morbidity and mortality worldwide. However, what was previously held as inconceivable – that cirrhosis does not inevitably lead to liver transplantation or death – is now a realistic hope. Mounting clinical and experimental evidence has demonstrated that even advanced fibrosis and cirrhosis are reversible.
The ‘activated’ hepatic stellate cell (HSC) is the pivotal cell type involved in the development of liver fibrosis, simultaneously secreting excessive scar proteins (collagens) and potent inhibitors of collagen degradation – tissue inhibitors of matrix metalloproteinases (TIMPs). Resolution of fibrosis is characterized by matrix degradation and is associated with HSC apoptosis and the increased activity of matrix metalloproteinase enzymes (MMPs). A key event is the initiation of degradation of collagen scar, presumed to be mediated by the interstitial collagenases, starting a cascade of events leading to restoration of the normal liver architecture. In this review we discuss the exciting recent advances in our understanding of the mechanisms of reversibility of liver fibrosis, citing both clinical reports of resolution and studies using experimental models of fibrosis. These important insights reveal potential and credible targets for effective antifibrotic therapies.
Introduction
Fibrosis, or scarring, is a highly conserved evolutionary response to limit tissue damage and serves as a generic response to chronic liver injury, regardless of aetiology. However, progressive scarring in response to a persisting liver insult eventually leads to ‘cirrhosis’ with disorganization of the normal liver architecture, characterized by fibrotic bands, parenchymal nodules and vascular distortion. Subsequent liver cell dysfunction and portal hypertension result in the grave clinical complications which have become all-too-familiar, such as variceal haemorrhage, encephalopathy, ascites and the hepatorenal syndrome. Were these problems not sufficient, there is the added risk of hepatocellular carcinoma, an aggressive neoplasm causing death for most within six months unless detected early. Cirrhosis, therefore, carries a significant morbidity and mortality and represents a major global health burden. Worldwide, chronic viral infection is the dominant cause, but in the developed countries of the West alcohol is the major aetiology. In the UK large rises in death rates from cirrhosis have been observed with over 4000 dying from the disease in 1999, two thirds of them before their 65th birthday (1). At present, the only curative treatment for end-stage cirrhosis is liver transplantation, but a shortage of available donors and the often poor state of health and nutrition of the potential recipient limit its clinical applicability.
For years, liver fibrosis has been considered irreversible. However, there is accumulating clinical and experimental evidence to suggest that this axiom be rejected. Data from the histological assessment of biopsy tissue from patients with chronic liver disease of various aetiologies who have been successfully treated, and from animal models of fibrosis, indicate that liver fibrosis is a dynamic, bi-directional process, wherein recovery with remodelling of scar tissue is possible. Furthermore, by understanding the cell and molecular mechanisms which mediate the reversibility of liver fibrosis we may establish the attributes required of an effective antifibrotic therapy.
Liver fibrogenesis
Liver fibrosis is characterized by quantitative and qualitative alterations to the normal hepatic extracellular matrix (ECM). The hepatic stellate cell (HSC) is the pivotal cell type in the development of liver fibrosis and appears to be the major source of the fibrillar collagens (type I, III and IV) and other matrix proteins that accumulate in chronic liver disease (2). In response to injury, these normally quiescent peri-sinusoidal vitamin-A storing cells proliferate and become ‘activated’ to a highly fibrogenic ‘myofibroblastic’ phenotype. In the activated state, HSCs orchestrate an array of changes including ECM remodelling, vascular contraction, and the release of cytokines. Progression of fibrosis is not only associated with synthesis of a HSC-derived fibrotic neomatrix rich in type-I collagen, but also inhibition of its degradation by secretion of the potent tissue inhibitors of metalloproteinases (TIMPs) (3). Furthermore, the increased expression of TIMPs precedes expression of procollagen-I mRNA, suggesting that fibrillar collagen is laid down into an extracellular milieu in which matrix degradation is already inhibited. In addition, TIMP-1 has been shown to inhibit apoptosis of some cell types, and may therefore be anti-apoptotic for activated HSCs (4). Continued HSC activation is perpetuated by autocrine and paracrine factors and positive feedback loops involving ECM components and cytokines (5).
Examining the evidence for reversibility
The concept of reversibility of liver fibrosis and cirrhosis is not a novel one. Reversal of cirrhosis was reported over 30 years ago in patients with haemochromatosis after long-term intensive venesection therapy (6). Regression of scarring has also been reported in patients with autoimmune chronic active hepatitis (7, 8) and primary biliary cirrhosis (9) after successful immunosuppressive therapy. Further support has also been accrued from lamivudine treatment in chronic hepatitis B patients (10) and following treatment of chronic delta hepatitis with longterm interferon-a (11). In addition, a recent study demonstrated regression of liver fibrosis after surgical biliary decompression in patients with chronic pancreatitis and stenosis of the common bile duct (12). A justified criticism of many of the earlier reports of reversibility of fibrosis and cirrhosis was the relatively small numbers of patients analyzed. However, the results from large-scale clinical trials in the treatment of chronic hepatitis C have more recently provided compelling and robust data. Poynard et al (13) analysed the results of 4 previous major clinical trials involving 3010 patients with chronic hepatitis C randomised to various treatment regimens with either interferon or pegylated interferon, with or without the addition of ribavirin. Pre and post treatment liver biopsies were analyzed. Major beneficial effects of antiviral therapy on liver fibrosis were observed, particularly with combination therapy. Moreover, reversal of cirrhosis was observed in 75 (49%) of 153 patients with cirrhosis at baseline.
The diverse aetiologies in which these observations of reversibility have been made suggest the liver’s capacity to remodel scar tissue is a generic phenomenon which, if harnessed and manipulated, may offer novel therapeutic approaches.
Mechanisms of reversibility of liver fibrosis
Complete restitution of the normal hepatic architecture would involve breakdown and remodelling of the fibrotic scar, of which collagen-1 is the major constituent. There is still conjecture as to the precise enzyme(s) mediating scar digestion and the mechanisms by which spontaneous resolution of liver fibrosis occurs, but the loss of activated stellate cells via apoptosis and increased collagenolytic activity within the liver appear to be central and common mechanisms. A key advance in this respect was the creation of a reproducible experimental model for the spontaneous resolution of liver fibrosis (14). In rats treated for 8 weeks with twice-weekly intraperitoneal CCl4, significant liver fibrosis develops, with formation of fibrotic septae and early nodule formation. If CCl4 treatment is continued for a further 4 weeks, the fibrosis progresses to cirrhosis. If, however, the liver injury is discontinued after 4 weeks, the liver fibrosis resolves entirely and normal liver histology is restored. In this model, there was a dramatic increase in HSC apoptosis observed, leading to a 50% reduction in the number of activated HSCs in the liver within 72 hours of withdrawal of injury. Furthermore, there was a fivefold increase in ‘collagenase’ activity in the liver, which coincided temporally with the observed degradation of fibrotic matrix.
Apoptosis (or programmed cell death) is the mechanism by which cells are eliminated from tissues without eliciting an inflammatory response. Apoptosis of HSCs has the potential to remove the source of both the fibrotic neomatrix and the metalloproteinase inhibitors (TIMPs), thus facilitating net matrix degradation. Animal models of reversibility of fibrosis have shown that HSC’s rapidly undergo apoptosis after 4 weeks of CCl4 liver injury (14), when the injurious stimulus is withdrawn, or in the bile duct ligation model of liver injury after decompressive surgery (12). Furthermore, Wright et al used gliotoxin to stimulate apoptosis of rat HSC after CCl4-induced liver fibrosis and found that the extent of fibrosis could be experimentally attenuated in vivo (15). However, the loss of activated HSCs may not in itself be sufficient to allow remodelling of the existing fibrotic scar. For this to occur, matrix degradation must be upregulated. Interestingly, apoptosis in HSCs has been shown experimentally to induce pro-MMP-2 activation and this may represent a mechanism via which matrix remodelling is controlled (16).
The matrix metalloproteinases are a family of calcium-dependent endopeptidases that specifically degrade collagens and non-collagenous substrates. Matrix degradation occurs predominantly as a consequence of the action of these enzymes which are secreted as proenzymes and activated primarily via cell surface-associated cleavage mechanisms (17). The initiation of degradation of matrix appears to represent a pivotal initial step in the process of resolution, starting a cascade of events that result in loss of fibrillar matrix and apoptosis of hepatic stellate cells. A key event appears to be the action of the interstitial collagenases (predominantly MMP-1 in humans and MMP-13 in rodents) which cleave the collagen-1 molecule between Gly775 and Ile776 in the a-1 chain and a corresponding Gly/Leu in the a-2 chain. This causes the molecule to unwind and become susceptible to degradation by gelatinases and other more promiscuous MMPs (17). Using mice bearing a mutated collagen-1 gene (r/r mice), which confers complete resistance to collagenase degradation, Issa et al (18) showed that failure to degrade collagen-1 critically impairs HSC apoptosis with failure of resolution of fibrosis and blunting of the hepatocyte regenerative response. It has also been shown in a rat model of thioacetamide-induced liver injury that adenoviral-mediated hepatic over-expression of MMP-1 is sufficient to result in resolution of fibrosis and, with it, the loss of activated HSCs (19). This further suggests that collagen-1 degradation may be critical to regression of liver fibrosis and HSC apoptosis, presumably mediated by changes in cell-matrix interactions.
The nature and origin of the key MMPs involved during regression are unclear, and recent evidence shows that two further MMPs, both of which are expressed in fibrotic liver (gelatinase A (MMP-2) (20) and MT1-MMP (MMP-14)) (21), may also have collagenolytic properties. Although there is no true rodent homolog of MMP-1, MMP-13 is considered to be the major interstitial collagenase in the rat and mouse. Previous studies have demonstrated that MMP-13 is expressed in both progressive and recovery models of liver fibrosis (3, 14, 22). In the CCl4 rat model (14), messenger RNA (mRNA) expression for MMP-13 was relatively constant in both progressive fibrosis and during recovery. In contrast, expression of TIMP-1 and TIMP-2 (both of which inhibit MMP-13) was upregulated in progressive fibrosis but reduced during recovery. Overall, net liver collagenase activity decreased with progressive injury, associated with TIMP expression, and increased during recovery associated with a decrease in TIMP expression. These data suggest that MMP-13 may be an important potential collagenase and that the latent collagenolytic activity within the liver is unharnessed when, during recovery, TIMP-1 and -2 expression decreases. Other groups have documented expression of MMP-13 during recovery from fibrosis and have suggested that HSCs are the source (23). However, the mRNA level for MMP-13 is relatively constant despite a significant decrease in HSC numbers mediated by apoptosis, and the site of origin of MMP-13 thus remains contentious. Considerable work is required to accurately dissect the relative roles of the different MMPs in degradation of the hepatic scar.
There is a close temporal correlation between HSC apoptosis and matrix degradation, suggesting that the two events may be intrinsically linked. This hypothesis is supported by data demonstrating that cell-matrix interactions, mediated by integrins, regulate the phenotype, survival and secretory activity of a variety of cell types, including HSCs (24, 25, 26). Indeed, tissue culture studies have shown that type 1 collagen matrix perpetuates the activated phenotype of HSCs (27). Additionally, HSC proliferation may be enhanced by pericellular collagen degradation, initiated by collagenase and mediated via avb3 engagement. Furthermore, contact with type 1 collagen can take hepatocytes out of the cell cycle, which they then re-enter after contact with partially degraded collagen-1, an event apparently also mediated by the avb3 integrin (17). Degradation of collagen may therefore drive the regenerative hepatocyte response which also characterizes recovery from fibrosis.
It is likely that the same mechanisms observed in rodent models are applicable in human liver disease. As discussed earlier, there is compelling histological evidence for a reduction in fibrotic matrix after treatment of various human chronic liver diseases. Although HSC apoptosis was not specifically studied in these series, there is histological evidence of a diminution in HSC numbers. Direct evidence for HSC loss during recovery from injury was provided in a study of acute paracetamol injury, where there was a clear reduction in numbers of a-smooth muscle actin positive (activated) HSCs on follow-up biopsy. There are clearly practical and ethical barriers to following the cellular mechanisms mediating recovery from fibrosis in humans by way of serial biopsy, particularly in patients who appear to be improving clinically. But, by taking the human data together with that derived from animal models, which permit frequent sampling and control over the chronology and extent of resolution, an increasingly complete picture of the critical features of spontaneous recovery from fibrosis is being constructed.
Figure 1: A summary of the processes mediating recovery from liver fibrosis and the mechanisms regulating hepatic stellate cell apoptosis. (Reproduced from Iredale et al., Seminars in Liver Disease 2001; 21, 3:427-436, by permission of Seminars in Liver Disease, copyright 2001 Thieme Medical Publishers)
Therapeutic targets
Currently, the most effective treatment for hepatic fibrosis is removal of the causative agent. For many patients, however, the liver disease is insidious in onset and they often present late when the potential for disease-specific treatment is limited. A shortage of liver transplant donors has led to a search for generically-applicable antifibrotic therapies. No treatment is yet approved for human usage and problems with drug efficacy, tissue selectivity and safety represent significant challenges. Table 1 summarises potential areas for therapeutic intervention.
Since the HSC plays a central role in the development of liver fibrosis, this cell is a major target. The delivery of biologically active molecules to HSCs has become a feasible option. Albumin-based carriers have been developed that preferentially distribute to HSCs in fibrotic rat livers (27). This would potentially allow for drug-targetting or selective gene delivery. Adenoviral vectors efficiently target normal liver cells and via effective transduction of hepatocytes can lead to high local concentrations of expressed protein. However, there is limited data on the use of such carriers in cirrhosis. Therefore, the demonstration by Garcia-Banuelos et al (28) that cirrhotic rats can be safely transduced with clinical-grade adenoviral vectors, with evidence of cirrhosis reversion, is potentially important. In addition, anti-inflammatory and superoxide agents may effectively supplement the treatments aimed at abrogating fibrosis described above (29). Finally, more information regarding the embryonic origin of the HSC will raise the possibility of using cell-lineage specific promoters to drive transgene expression selectively in HSCs in vivo.
Future prospects
Antifibrotic therapies are an emerging reality. The framework upon which these strategies are designed is the result of the burgeoning evidence base for the reversibility of liver fibrosis. However, key questions remain unanswered. For example, does fibrosis reach a stage when it is truly irreversible? This might involve complex cross-linking of collagen fibrils (30) or a critical mass which renders the hepatic scar less susceptible to collagenase activity. Animal models provide a method by which quantitative and qualitative differences in liver structure can be compared between recoverable and irreversible fibrosis. Furthermore, evidence of long-term benefits of the reversal of fibrosis on clinical outcome, such as a reduction in portal hypertension or the rate of development of hepatocellular carcinoma, will be increasingly sought.
In summary, there are compelling data demonstrating proof-of-concept that even advanced fibrosis and cirrhosis are reversible states. The molecular mechanisms mediating recovery are becoming increasingly clear, giving real hope of impacting therapeutically on this serious disease.
References
Annual Report of the Chief Medical Officer 2001, Department of Health.
Friedman SL. Molecular regulation of hepatic fibrosis, an integrated cellular response to tissue injury. Journal of Biological Chemistry 2000; 275(4):2247-2250.
Iredale JP, Benyon RC, Arthur MJP, Ferris WF, Alcolado R, Winwood PJ et al. Tissue inhibitor of metalloproteinase-1 messenger RNA expression is enhanced relative to interstitial collagenase messenger RNA in experimental liver injury and fibrosis. Hepatology 1996; 24:176-184.
Arthur MJP. Fibrogenesis II. Metalloproteinases and their inhibitors in liver fibrosis. Am J Physiol Gastrointest Liver Physiol 2000; G245-G249.
Pinzani M, Marra F. Cytokine receptors and signalling in hepatic stellate cells. Semin Liver Dis 2001: 21(3), 397-416.
Powell LW, Kerr JF. Reversal of “cirrhosis” in idiopathic haemochromatosis following long-term intensive venesection therapy. Australas Ann Med 1970; 19:54-57.
Dufour JF, DeLellis R, Kaplan MM. Reversibility of hepatic fibrosis in autoimmune hepatitis. Ann Intern Med 1997; 127:981-985.
Wanless IR. Use of corticosteroid therapy in autoimmune hepatitis resulting in the resolution of cirrhosis. J Clin Gastroenterol 2001; 32:371-372.
Kaplan MM, DeLellis R, Wolfe HJ. Sustained biochemical and histological remission of primary biliary cirrhosis in response to medical treatment. Ann Intern Med 1997; 126:682-688.
Kweon YO, Goodman ZD, Dienstag JL, Schiff ER, Brown NA, Burkhard E et al. Decreasing fibrogenesis: an immunohistochemical study of paired liver biopsies following lamivudine therapy for chronic hepatitis B. Br J Hepatol 2001: 35:749-755.
Lau DT, Kleiner DE, Park Y, Di Bisceglie AM, Hoofnagle JH. Resolution of chronic delta hepatitis after 12 years of interferon alfa therapy. Gastroenterology 1999; 117:1229-1233.
Hammel P, Couvelaed A, O’Toole D, Ratouis A, Sauvanet A, Flejou JF 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; 344:418-423.
Poynard T, McHutchinson J, Manns M, Trepo C, Lindsay K, Goodman Z et al. Impact of pegylated interferon alfa-2b and ribavirin on liver fibrosis in patients with chronic hepatitis C. Gastroenterology 2002; 122:1303-1313.
Iredale JP, Benyon RC, Pickering J, McCullen M, Northrop S, Pawley S et al. Mechanisms of spontaneous resolution of rat liver fibrosis. Hepatic stellate cell apoptosis and reduced hepatic expression of metalloproteinase inhibitors. J Clin Invest 1998; 102(3):538-549.
Wright MC, Issa R, Smart DE, Trim N, Murray GI, Primrose JN et al. Gliotoxin stimulates the apoptosis of human and rat hepatic stellate cells and enhances the resolution of liver fibrosis in rats. Gastroenterology 2001;121:685-698.
Preux AM, D’Ortho MP, Bralet MP, Laperche Y, Mavier P. Apoptosis of human hepatic myofibroblasts promotes activation of matrix metalloproteinase-2. Hepatology 2002; 36(3):615-622.
Benyon RC, Arthur MJP. Extracellular matrix degradation and the role of hepatic stellate cells. Semin Liver Dis 2001; 21(3):373-384.
Issa R, Zhou X, Trim N, Millward-Sadler H, Krane S, Benyon RC et al. Mutation in collagen-1 that confers resistance to the action of collagenase results in failure of recovery from CCl4-induced liver fibrosis, persistence of activated hepatic stellate cells, and diminished hepatocyte regeneration. FASEB J 2003; 17(1):47-49.
Iimuro Y, Nishio T, Fujii H, Oe S, Hirose T, Morimoto T et al. Over expression of matrix metalloproteinase-1 attenuates liver fibrosis caused by thioacetamide in the rat. Hepatology 1999; 30:299A.
Kerkvliet EH, Docherty AJ, Beertsen W, Everts V. Collagen breakdown in soft connective tissue explants is associated with the level of active gelatinase-A (MMP-2) but not with collagenase. Matrix boil 1999; 18:373-380.
Ohuchi E, Imai K, Fujii Y, Sato H, Seiki M, Okada Y. Membrane type 1 matrix metalloproteinase digests interstitial collagens and other extracellular matrix macromolecules. J Biol Chem 1997; 272:2446-2451.
Yata Y, Takahara T, Furui K, Zhang LP, Watanabe A. Expression of matrix metalloproteinase-13 and tissue inhibitor of metalloproteinase-1 in acute liver injury. J Hepatol 1999; 30(3):419-424.
Watanabe T, Niioka M, Hozawa S, Kameyama K, Hayashi T, Arai M et al. Gene expression of interstitial collagenase in both progressive and recovery phase of rat liver fibrosis induced by carbon tetrachloride. J Hepatol 2000; 33(2):224-235.
Iwamoto H, Sakai H, Tada S, Nakamuta M, Natawa H. Induction of apoptosis in rat hepatic stellate cells by disruption of integrin-mediated cell adhesion. J Lab Clin Med 1999; 134(1):83-89.
Iwamoto H, Sakai H, Kotoh K, Nakamuta M, Natawa H. Soluble Arg-Gly-Asp peptides reduce collagen accumulation in cultured rat hepatic stellate cells. J Hepatol 1998; 29(5):752-759.
Montgomery AM, Reisfeld RA, Cheresh DA. Integrin alpha v beta 3 rescues melanoma cells from apoptosis in three-dimensional dermal collagen. Proc Natl Acad Sci USA 1994.; 91(19):8856-8860.
Friedman SL, Roll FJ, Boyles J, Arenson DM, Bissell DM. Maintenance of differentiated phenotype of cultured rat hepatic lipocytes by basement membrane matrix. J Biol Chem 1989; 264:10756-10762.
Beljaars L, Meijer DK, Poelstra K. Targetting hepatic stellate cells for cell-specific treatment of liver fibrosis. Front Biosci 2002; 7:214-222.
Garcia-Banuelos J, Siller-Lopez F, Miranda A, Aguilar LK, Aguilar-Cordova E, Armendariz-Borunda J. Cirrhotic rat livers with extensive fibrosis can be safely transduced with clinical-grade adenoviral vectors. Evidence of cirrhosis reversion. Gene Ther 2002; 9(2):127-134.
Zhong Z, Froh M, Wheeler MD, Smutney O, Lehmann TG, Thurman RG. Viral gene delivery of superoxide dismutase attenuates experimental cholestasis-induced liver fibrosis in the rat. Gene Ther 2002; 9(3): 183-191.