Focus

Abstract

Molecular MR imaging of liver fibrosis: A feasibility study using rat and mouse modelsJournal of HepatologyVol. 57Issue 3PreviewLiver biopsy, the current clinical gold standard for fibrosis assessment, is invasive and has sampling errors, and is not optimal for screening, monitoring, or clinical decision-making. Fibrosis is characterized by excessive accumulation of extracellular matrix proteins including type I collagen. We hypothesize that molecular magnetic resonance imaging (MRI) with a probe targeted to type I collagen could provide a direct and non-invasive method of fibrosis assessment. Full-Text PDF Hepatitis C virus (HCV) protein expression enhances hepatic fibrosis in HCV transgenic mice exposed to a fibrogenic agentJournal of HepatologyVol. 57Issue 3PreviewDuring chronic HCV infection, activation of fibrogenesis appears to be principally related to local inflammation. However, the direct role of hepatic HCV protein expression in fibrogenesis remains unknown. Full-Text PDF Stimulating and seeing the scar in hepatic fibrosisTwo articles in this month’s issue of the Journal highlight progress in understanding intracellular pathways of fibrogenesis due to hepatitis C virus (HCV), and a potential new approach to imaging hepatic fibrosis using magnetic resonance imaging (MRI). Together, they illustrate the steady, successful assault on the problem of hepatic fibrosis, both in uncovering its mechanisms and advancing its clinical management.The study by Chouteau and colleagues from Paris demonstrates that hepatocyte-specific expression of the HCV structural and non-structural proteins increases the degree of fibrosis and stimulates the hepatic progenitor cell response in mice with liver injury due to CCl4, a standard hepatotoxin. The original transgenic model was developed by Stanley Lemon’s group several years ago [[1]Lerat H. Honda M. Beard M.R. Loesch K. Sun J. Yang Y. et al.Steatosis and liver cancer in transgenic mice expressing the structural and nonstructural proteins of hepatitis C virus.Gastroenterology. 2002; 122: 352-365Abstract Full Text Full Text PDF PubMed Scopus (405) Google Scholar], which demonstrated progressive steatosis and then adenomas and carcinomas as the animals aged. Expression levels of the HCV proteins were within the range associated with native infection, and there was virtually no baseline inflammation, indicating the lack of an immune response, and no fibrosis. However, the transgenic livers expressed greater levels of lipid peroxidation products, implicating interaction of HCV core protein with some cellular components, possibly mitochondria or the intracellular domains of a tumor necrosis factor alpha receptor [[1]Lerat H. Honda M. Beard M.R. Loesch K. Sun J. Yang Y. et al.Steatosis and liver cancer in transgenic mice expressing the structural and nonstructural proteins of hepatitis C virus.Gastroenterology. 2002; 122: 352-365Abstract Full Text Full Text PDF PubMed Scopus (405) Google Scholar].In the Chouteau study, animals were given CCl4 for 4 weeks, an intermediate dosing regimen that is not sufficient to induce cirrhosis, and while fibrosis was present in both the control and HCV transgenic mice, mice with transgenic HCV protein expression had more fibrosis, with an increased ductular reaction associated with greater numbers of hepatic progenitor cells, as assessed by CK-19 staining. At the same time, hepatocyte replication was reduced based on Ki67 staining, and there was a strong inverse relationship between the number of Ki67-positive hepatocytes and the extent of fibrosis, indicating that impaired hepatocyte regeneration was associated with increased fibrosis. There was no difference in inflammation between control and HCV transgenic animals. Mechanistic insights were limited to evidence of increased baseline and CCl4-induced levels of reactive oxygen species in the HCV transgenic mice, and lower MMP7 mRNA expression in these same animals, with no change in several other MMP transcripts.These results are intriguing for at least two reasons. First, they reinforce evidence of the disruptive effect of HCV proteins on hepatocyte homeostasis that leads to enhanced oxidant stress, with greater susceptibility to fibrosis. Second, they reframe the idea that replicative block of hepatocytes, from a variety of causes, drives the hepatic progenitor response, which in turn is linked to increased fibrosis in animals and humans [2Chobert M.N. Couchie D. Fourcot A. Zafrani E.S. Laperche Y. Mavier P. et al.Liver precursor cells increase hepatic fibrosis induced by chronic carbon tetrachloride intoxication in rats.Lab Invest. 2012; 92: 135-150Crossref PubMed Scopus (30) Google Scholar, 3Leandro G. Mangia A. Hui J. Fabris P. Rubbia-Brandt L. Colloredo G. et al.Relationship between steatosis, inflammation, and fibrosis in chronic hepatitis C: a meta-analysis of individual patient data.Gastroenterology. 2006; 130: 1636-1642Abstract Full Text Full Text PDF PubMed Scopus (487) Google Scholar, 4Clouston A.D. Powell E.E. Walsh M.J. Richardson M.M. Demetris A.J. Jonsson J.R. Fibrosis correlates with a ductular reaction in hepatitis C: roles of impaired replication, progenitor cells and steatosis.Hepatology. 2005; 41: 809-818Crossref PubMed Scopus (298) Google Scholar].It remains unclear whether the response of hepatocytes to transgenic HCV protein expression is identical to that of native infection. While the levels of HCV proteins are reportedly comparable to infection, the response of hepatocytes suggest activation of the unfolded protein response (UPR), a cellular mechanism that manages misfolded or excessive accumulation of proteins [[5]Hetz C. The unfolded protein response: controlling cell fate decisions under ER stress and beyond.Nat Rev Mol Cell Biol. 2012; 13: 89-102PubMed Google Scholar]; thus, the mitochondrial involvement, leading to steatosis and lipid peroxidation could reflect the protective response of a cell confronted with a misfolded HCV protein that is not seen in native infection. It would be extremely interesting to evaluate these livers for evidence of UPR activation, whose features and methods of analysis are well characterized in both animal models and human liver disease [[6]Malhi H. Kaufman R.J. Endoplasmic reticulum stress in liver disease.J Hepatol. 2011; 54: 795-809Abstract Full Text Full Text PDF PubMed Scopus (814) Google Scholar]. Also, the mechanistic studies in the Chouteau paper are limited to interrogation of a few candidate pathways, whereas a more unbiased analysis using conventional arrays, deep sequencing and/or proteomics methods could yield novel insights into the impact of HCV proteins on hepatocytes and how they interact with other cells.More intriguing, and also meriting further exploration in the Chouteau study is the evidence of a brisk progenitor response and its correlation with reduced hepatocyte proliferation and increased fibrosis. Our understanding of the relationship between injury, hepatocyte regeneration and fibrosis remains murky but ripe for a breakthrough. Study after study continue to demonstrate a link between ductular reaction, progenitor cells, and fibrosis, yet the sequence of events and key underlying mediators are elusive. Epithelial–mesenchymal transition (EMT) is implicated by some, but not all studies as a central driver of this response, and one candidate mediator in this context may be the molecule Snail [[7]Rowe R.G. Lin Y. Shimizu-Hirota R. Hanada S. Neilson E.G. Greenson J.K. et al.Hepatocyte-derived Snail1 propagates liver fibrosis progression.Mol Cell Biol. 2011; 31: 2392-2403Crossref PubMed Scopus (99) Google Scholar]. More refined genetic models to track and manipulate progenitor cells either in this HCV model or others should clarify the meaning of the ductular reaction and the link between impaired regeneration that activates the progenitor cell response, and its association with increased fibrosis. Perhaps this series of events is unique to liver and accounts for its unique regenerative capacity compared to other tissues.The study by Polasek et al. in this issue of the Journal provides a scintilla of hope for improved fibrosis assessment using a standard non-invasive tool, MRI. The investigators explored the utility of a gadolinium-based contrast agent containing a 16 amino acid cyclic peptide that specifically binds collagen, to quantify this protein in two rodent models of fibrosis. The agent, EP-3533, was optimized to maintain the highest affinity to collagen while carrying the largest gadolinium payload [8Caravan P. Das B. Dumas S. Epstein F.H. Helm P.A. Jacques V. et al.Collagen-targeted MRI contrast agent for molecular imaging of fibrosis.Angew Chem Int Ed Engl. 2007; 46: 8171-8173Crossref PubMed Scopus (180) Google Scholar, 9Caravan P. Das B. Deng Q. Dumas S. Jacques V. Koerner S.K. et al.A lysine walk to high relaxivity collagen-targeted MRI contrast agents.Chem Commun (Camb). 2009; 4: 430-432Crossref PubMed Scopus (25) Google Scholar]. The agent was tested initially in a rat model of weekly DEN for 4 weeks, an unusual dosing scheme for a drug that is typically given only once intraperitoneally to induce hepatic tumors 6–12 months later. Thus, this new DEN model of fibrosis itself merits further evaluation, as it appears to reproducibly induce fibrosis and presumably leads to cancer as well. The imaging method was further evaluated in a CCl4 model in mice treated thrice weekly for 20 weeks, leading to dense cirrhosis. As this was a pilot feasibility study only single time points were assessed for each model, correlating delayed gadolinium persistence in liver with either hydroxyproline content (an amino acid unique to collagen) or Ishak fibrosis score. The correlation coefficient were promising but not outstanding, and might have been improved if the authors had correlated the images with either Sirius red staining or collagen I immunohistochemistry. Moreover, the authors point out that that the sensitivity of the method would likely be enhanced when used for clinical studies in humans, where the magnets used for MRI are more powerful. They also emphasize that this study was not powered to stage fibrosis, and indeed the numbers of animals used was small. Still, there is good reason to further develop this methodology for a number of reasons.Currently there is no established non-invasive test to quantify fibrosis content that can supplant liver biopsy, despite its well known limitations, including invasiveness, sampling variability, and the inability to obtain serial measurements over time. Therefore, a non-invasive test that could quantify collagen at progressive intervals using a standard technology would have enormous value not only in clinical practice, but even before then in clinical trials assessing antifibrotic therapies. Perhaps the rate of change in collagen content over time could provide a new, quantitative, and dynamic metric for assessing both disease progression and regression. Moreover, as the authors point out, an MRI test for collagen content might be combined during the same exam with studies to assess fat content and/or stiffness, providing a more comprehensive assessment of liver structure in complex conditions like non-alcoholic fatty liver disease.First, however, more animal data about this gadolinium-based MRI agent are needed, and then concerns about safety in humans will need to be thoroughly addressed. Animal studies should be extended to other models of fibrosis (including biliary fibrosis), and the collagen content and Sirius red staining must be correlated at several time points in a large number of animals, to confirm that the intensity of the MR signal corresponds exactly to the amount of hepatic collagen at each stage of disease. Studies in large animals might lend further support to the value of this new methodology, for example in pigs, where models of liver disease have been established [[10]Lee L. Alloosh M. Saxena R. Van Alstine W. Watkins B.A. Klaunig J.E. et al.Nutritional model of steatohepatitis and metabolic syndrome in the Ossabaw miniature swine.Hepatology. 2009; 50: 56-67Crossref PubMed Scopus (156) Google Scholar].If the technology moves on to human studies, there is lingering concern about the propensity for gadolinium-based contrast agents to induce nephrogenic systemic fibrosis, especially in patients with concurrent renal impairment [[11]Hellman R.N. Gadolinium-induced nephrogenic systemic fibrosis.Semin Nephrol. 2011; 31: 310-316Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar]. While the incidence of this syndrome appears to be waning with careful patient selection, it remains a worry, and perhaps the collagen-binding cyclic peptide could be linked to a positron emission tomography (PET)-isotope rather than gadolinium, for PET imaging. Of course, PET when combined with CT raises concerns about radiation exposure. Also, patients with very advanced fibrosis and cirrhosis have intrahepatic shunting and vascular collapse that might undermine the sensitivity or accuracy of this technology, a prospect that should first be addressed in animals by testing the technology in animals with very advanced fibrosis and portal hypertension.The final question is whether even a test that measures collagen content perfectly is sufficient for use as a non-invasive diagnostic tool. It has become clear that in patients where the underlying disease is rapidly controlled, for example, in those treated with antiviral therapy, there may be a lag of up to 2 years before fibrosis scores improve after the inflammation and necrosis subside [[12]Chang T.T. Liaw Y.F. Wu S.S. Schiff E. Han K.H. Lai C.L. et al.Long-term entecavir therapy results in the reversal of fibrosis/cirrhosis and continued histological improvement in patients with chronic hepatitis B.Hepatology. 2010; 52: 886-893Crossref PubMed Scopus (744) Google Scholar]. The ideal fibrosis test would be one that detects early reductions in fibrogenic activity or increases in proteolytic activity, which occur well before the actual fibrosis content changes, yet are accurate harbingers of an improved histology and fibrosis score at a later interval. This would be an extremely powerful advance, especially for clinical trials where early evidence of efficacy could incentivize investigators and pharmaceutical sponsors to support long-term trials. Clearly, however, any MRI test suitable for clinical usage that serially measures collagen content would represent a major step forward, and could accelerate progress in the staging and treatment of patients with fibrosis liver disease, particularly those where biopsy is either contraindicated or not available.Conflict of interestThe author declared that he does not have anything to disclose regarding funding or conflict of interest with respect to this manuscript. Stimulating and seeing the scar in hepatic fibrosisTwo articles in this month’s issue of the Journal highlight progress in understanding intracellular pathways of fibrogenesis due to hepatitis C virus (HCV), and a potential new approach to imaging hepatic fibrosis using magnetic resonance imaging (MRI). Together, they illustrate the steady, successful assault on the problem of hepatic fibrosis, both in uncovering its mechanisms and advancing its clinical management.The study by Chouteau and colleagues from Paris demonstrates that hepatocyte-specific expression of the HCV structural and non-structural proteins increases the degree of fibrosis and stimulates the hepatic progenitor cell response in mice with liver injury due to CCl4, a standard hepatotoxin. The original transgenic model was developed by Stanley Lemon’s group several years ago [[1]Lerat H. Honda M. Beard M.R. Loesch K. Sun J. Yang Y. et al.Steatosis and liver cancer in transgenic mice expressing the structural and nonstructural proteins of hepatitis C virus.Gastroenterology. 2002; 122: 352-365Abstract Full Text Full Text PDF PubMed Scopus (405) Google Scholar], which demonstrated progressive steatosis and then adenomas and carcinomas as the animals aged. Expression levels of the HCV proteins were within the range associated with native infection, and there was virtually no baseline inflammation, indicating the lack of an immune response, and no fibrosis. However, the transgenic livers expressed greater levels of lipid peroxidation products, implicating interaction of HCV core protein with some cellular components, possibly mitochondria or the intracellular domains of a tumor necrosis factor alpha receptor [[1]Lerat H. Honda M. Beard M.R. Loesch K. Sun J. Yang Y. et al.Steatosis and liver cancer in transgenic mice expressing the structural and nonstructural proteins of hepatitis C virus.Gastroenterology. 2002; 122: 352-365Abstract Full Text Full Text PDF PubMed Scopus (405) Google Scholar].In the Chouteau study, animals were given CCl4 for 4 weeks, an intermediate dosing regimen that is not sufficient to induce cirrhosis, and while fibrosis was present in both the control and HCV transgenic mice, mice with transgenic HCV protein expression had more fibrosis, with an increased ductular reaction associated with greater numbers of hepatic progenitor cells, as assessed by CK-19 staining. At the same time, hepatocyte replication was reduced based on Ki67 staining, and there was a strong inverse relationship between the number of Ki67-positive hepatocytes and the extent of fibrosis, indicating that impaired hepatocyte regeneration was associated with increased fibrosis. There was no difference in inflammation between control and HCV transgenic animals. Mechanistic insights were limited to evidence of increased baseline and CCl4-induced levels of reactive oxygen species in the HCV transgenic mice, and lower MMP7 mRNA expression in these same animals, with no change in several other MMP transcripts.These results are intriguing for at least two reasons. First, they reinforce evidence of the disruptive effect of HCV proteins on hepatocyte homeostasis that leads to enhanced oxidant stress, with greater susceptibility to fibrosis. Second, they reframe the idea that replicative block of hepatocytes, from a variety of causes, drives the hepatic progenitor response, which in turn is linked to increased fibrosis in animals and humans [2Chobert M.N. Couchie D. Fourcot A. Zafrani E.S. Laperche Y. Mavier P. et al.Liver precursor cells increase hepatic fibrosis induced by chronic carbon tetrachloride intoxication in rats.Lab Invest. 2012; 92: 135-150Crossref PubMed Scopus (30) Google Scholar, 3Leandro G. Mangia A. Hui J. Fabris P. Rubbia-Brandt L. Colloredo G. et al.Relationship between steatosis, inflammation, and fibrosis in chronic hepatitis C: a meta-analysis of individual patient data.Gastroenterology. 2006; 130: 1636-1642Abstract Full Text Full Text PDF PubMed Scopus (487) Google Scholar, 4Clouston A.D. Powell E.E. Walsh M.J. Richardson M.M. Demetris A.J. Jonsson J.R. Fibrosis correlates with a ductular reaction in hepatitis C: roles of impaired replication, progenitor cells and steatosis.Hepatology. 2005; 41: 809-818Crossref PubMed Scopus (298) Google Scholar].It remains unclear whether the response of hepatocytes to transgenic HCV protein expression is identical to that of native infection. While the levels of HCV proteins are reportedly comparable to infection, the response of hepatocytes suggest activation of the unfolded protein response (UPR), a cellular mechanism that manages misfolded or excessive accumulation of proteins [[5]Hetz C. The unfolded protein response: controlling cell fate decisions under ER stress and beyond.Nat Rev Mol Cell Biol. 2012; 13: 89-102PubMed Google Scholar]; thus, the mitochondrial involvement, leading to steatosis and lipid peroxidation could reflect the protective response of a cell confronted with a misfolded HCV protein that is not seen in native infection. It would be extremely interesting to evaluate these livers for evidence of UPR activation, whose features and methods of analysis are well characterized in both animal models and human liver disease [[6]Malhi H. Kaufman R.J. Endoplasmic reticulum stress in liver disease.J Hepatol. 2011; 54: 795-809Abstract Full Text Full Text PDF PubMed Scopus (814) Google Scholar]. Also, the mechanistic studies in the Chouteau paper are limited to interrogation of a few candidate pathways, whereas a more unbiased analysis using conventional arrays, deep sequencing and/or proteomics methods could yield novel insights into the impact of HCV proteins on hepatocytes and how they interact with other cells.More intriguing, and also meriting further exploration in the Chouteau study is the evidence of a brisk progenitor response and its correlation with reduced hepatocyte proliferation and increased fibrosis. Our understanding of the relationship between injury, hepatocyte regeneration and fibrosis remains murky but ripe for a breakthrough. Study after study continue to demonstrate a link between ductular reaction, progenitor cells, and fibrosis, yet the sequence of events and key underlying mediators are elusive. Epithelial–mesenchymal transition (EMT) is implicated by some, but not all studies as a central driver of this response, and one candidate mediator in this context may be the molecule Snail [[7]Rowe R.G. Lin Y. Shimizu-Hirota R. Hanada S. Neilson E.G. Greenson J.K. et al.Hepatocyte-derived Snail1 propagates liver fibrosis progression.Mol Cell Biol. 2011; 31: 2392-2403Crossref PubMed Scopus (99) Google Scholar]. More refined genetic models to track and manipulate progenitor cells either in this HCV model or others should clarify the meaning of the ductular reaction and the link between impaired regeneration that activates the progenitor cell response, and its association with increased fibrosis. Perhaps this series of events is unique to liver and accounts for its unique regenerative capacity compared to other tissues.The study by Polasek et al. in this issue of the Journal provides a scintilla of hope for improved fibrosis assessment using a standard non-invasive tool, MRI. The investigators explored the utility of a gadolinium-based contrast agent containing a 16 amino acid cyclic peptide that specifically binds collagen, to quantify this protein in two rodent models of fibrosis. The agent, EP-3533, was optimized to maintain the highest affinity to collagen while carrying the largest gadolinium payload [8Caravan P. Das B. Dumas S. Epstein F.H. Helm P.A. Jacques V. et al.Collagen-targeted MRI contrast agent for molecular imaging of fibrosis.Angew Chem Int Ed Engl. 2007; 46: 8171-8173Crossref PubMed Scopus (180) Google Scholar, 9Caravan P. Das B. Deng Q. Dumas S. Jacques V. Koerner S.K. et al.A lysine walk to high relaxivity collagen-targeted MRI contrast agents.Chem Commun (Camb). 2009; 4: 430-432Crossref PubMed Scopus (25) Google Scholar]. The agent was tested initially in a rat model of weekly DEN for 4 weeks, an unusual dosing scheme for a drug that is typically given only once intraperitoneally to induce hepatic tumors 6–12 months later. Thus, this new DEN model of fibrosis itself merits further evaluation, as it appears to reproducibly induce fibrosis and presumably leads to cancer as well. The imaging method was further evaluated in a CCl4 model in mice treated thrice weekly for 20 weeks, leading to dense cirrhosis. As this was a pilot feasibility study only single time points were assessed for each model, correlating delayed gadolinium persistence in liver with either hydroxyproline content (an amino acid unique to collagen) or Ishak fibrosis score. The correlation coefficient were promising but not outstanding, and might have been improved if the authors had correlated the images with either Sirius red staining or collagen I immunohistochemistry. Moreover, the authors point out that that the sensitivity of the method would likely be enhanced when used for clinical studies in humans, where the magnets used for MRI are more powerful. They also emphasize that this study was not powered to stage fibrosis, and indeed the numbers of animals used was small. Still, there is good reason to further develop this methodology for a number of reasons.Currently there is no established non-invasive test to quantify fibrosis content that can supplant liver biopsy, despite its well known limitations, including invasiveness, sampling variability, and the inability to obtain serial measurements over time. Therefore, a non-invasive test that could quantify collagen at progressive intervals using a standard technology would have enormous value not only in clinical practice, but even before then in clinical trials assessing antifibrotic therapies. Perhaps the rate of change in collagen content over time could provide a new, quantitative, and dynamic metric for assessing both disease progression and regression. Moreover, as the authors point out, an MRI test for collagen content might be combined during the same exam with studies to assess fat content and/or stiffness, providing a more comprehensive assessment of liver structure in complex conditions like non-alcoholic fatty liver disease.First, however, more animal data about this gadolinium-based MRI agent are needed, and then concerns about safety in humans will need to be thoroughly addressed. Animal studies should be extended to other models of fibrosis (including biliary fibrosis), and the collagen content and Sirius red staining must be correlated at several time points in a large number of animals, to confirm that the intensity of the MR signal corresponds exactly to the amount of hepatic collagen at each stage of disease. Studies in large animals might lend further support to the value of this new methodology, for example in pigs, where models of liver disease have been established [[10]Lee L. Alloosh M. Saxena R. Van Alstine W. Watkins B.A. Klaunig J.E. et al.Nutritional model of steatohepatitis and metabolic syndrome in the Ossabaw miniature swine.Hepatology. 2009; 50: 56-67Crossref PubMed Scopus (156) Google Scholar].If the technology moves on to human studies, there is lingering concern about the propensity for gadolinium-based contrast agents to induce nephrogenic systemic fibrosis, especially in patients with concurrent renal impairment [[11]Hellman R.N. Gadolinium-induced nephrogenic systemic fibrosis.Semin Nephrol. 2011; 31: 310-316Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar]. While the incidence of this syndrome appears to be waning with careful patient selection, it remains a worry, and perhaps the collagen-binding cyclic peptide could be linked to a positron emission tomography (PET)-isotope rather than gadolinium, for PET imaging. Of course, PET when combined with CT raises concerns about radiation exposure. Also, patients with very advanced fibrosis and cirrhosis have intrahepatic shunting and vascular collapse that might undermine the sensitivity or accuracy of this technology, a prospect that should first be addressed in animals by testing the technology in animals with very advanced fibrosis and portal hypertension.The final question is whether even a test that measures collagen content perfectly is sufficient for use as a non-invasive diagnostic tool. It has become clear that in patients where the underlying disease is rapidly controlled, for example, in those treated with antiviral therapy, there may be a lag of up to 2 years before fibrosis scores improve after the inflammation and necrosis subside [[12]Chang T.T. Liaw Y.F. Wu S.S. Schiff E. Han K.H. Lai C.L. et al.Long-term entecavir therapy results in the reversal of fibrosis/cirrhosis and continued histological improvement in patients with chronic hepatitis B.Hepatology. 2010; 52: 886-893Crossref PubMed Scopus (744) Google Scholar]. The ideal fibrosis test would be one that detects early reductions in fibrogenic activity or increases in proteolytic activity, which occur well before the actual fibrosis content changes, yet are accurate harbingers of an improved histology and fibrosis score at a later interval. This would be an extremely powerful advance, especially for clinical trials where early evidence of efficacy could incentivize investigators and pharmaceutical sponsors to support long-term trials. Clearly, however, any MRI test suitable for clinical usage that serially measures collagen content would represent a major step forward, and could accelerate progress in the staging and treatment of patients with fibrosis liver disease, particularly those where biopsy is either contraindicated or not available. Two articles in this month’s issue of the Journal highlight progress in understanding intracellular pathways of fibrogenesis due to hepatitis C virus (HCV), and a potential new approach to imaging hepatic fibrosis using magnetic resonance imaging (MRI). Together, they illustrate the steady, successful assault on the problem of hepatic fibrosis, both in uncovering its mechanisms and advancing its clinical management. The study by Chouteau and colleagues from Paris demonstrates that hepatocyte-specific expression of the HCV structural and non-structural proteins increases the degree of fibrosis and stimulates the hepatic progenitor cell response in mice with liver injury due to CCl4, a standard hepatotoxin. The original transgenic model was developed by Stanley Lemon’s group several years ago [[1]Lerat H. Honda M. Beard M.R. Loesch K. Sun J. Yang Y. et al.Steatosis and liver cancer in transgenic mice expressing the structural and nonstructural proteins of hepatitis C virus.Gastroenterology. 2002; 122: 352-365Abstract Full Text Full Text PDF PubMed Scopus (405) Google Scholar], which demonstrated progressive steatosis and then adenomas and carcinomas as the animals aged. Expression levels of the HCV proteins were within the range associated with native infection, and there was virtually no baseline inflammation, indicating the lack of an immune response, and no fibrosis. However, the transgenic livers expressed greater levels of lipid peroxidation products, implicating interaction of HCV core protein with some cellular components, possibly mitochondria or the intracellular domains of a tumor necrosis factor alpha receptor [[1]Lerat H. Honda M. Beard M.R. Loesch K. Sun J. Yang Y. et al.Steatosis and liver cancer in transgenic mice expressing the structural and nonstructural proteins of hepatitis C virus.Gastroenterology. 2002; 122: 352-365Abstract Full Text Full Text PDF PubMed Scopus (405) Google Scholar]. In the Chouteau study, animals were given CCl4 for 4 weeks, an intermediate dosing regimen that is not sufficient to induce cirrhosis, and while fibrosis was present in both the control and HCV transgenic mice, mice with transgenic HCV protein expression had more fibrosis, with an increased ductular reaction associated with greater numbers of hepatic progenitor cells, as assessed by CK-19 staining. At the same time, hepatocyte replication was reduced based on Ki67 staining, and there was a strong inverse relationship between the number of Ki67-positive hepatocytes and the extent of fibrosis, indicating that impaired hepatocyte regeneration was associated with increased fibrosis. There was no difference in inflammation between control and HCV transgenic animals. Mechanistic insights were limited to evidence of increased baseline and CCl4-induced levels of reactive oxygen species in the HCV transgenic mice, and lower MMP7 mRNA expression in these same animals, with no change in several other MMP transcripts. These results are intriguing for at least two reasons. First, they reinforce evidence of the disruptive effect of HCV proteins on hepatocyte homeostasis that leads to enhanced oxidant stress, with greater susceptibility to fibrosis. Second, they reframe the idea that replicative block of hepatocytes, from a variety of causes, drives the hepatic progenitor response, which in turn is linked to increased fibrosis in animals and humans [2Chobert M.N. Couchie D. Fourcot A. Zafrani E.S. Laperche Y. Mavier P. et al.Liver precursor cells increase hepatic fibrosis induced by chronic carbon tetrachloride intoxication in rats.Lab Invest. 2012; 92: 135-150Crossref PubMed Scopus (30) Google Scholar, 3Leandro G. Mangia A. Hui J. Fabris P. Rubbia-Brandt L. Colloredo G. et al.Relationship between steatosis, inflammation, and fibrosis in chronic hepatitis C: a meta-analysis of individual patient data.Gastroenterology. 2006; 130: 1636-1642Abstract Full Text Full Text PDF PubMed Scopus (487) Google Scholar, 4Clouston A.D. Powell E.E. Walsh M.J. Richardson M.M. Demetris A.J. Jonsson J.R. Fibrosis correlates with a ductular reaction in hepatitis C: roles of impaired replication, progenitor cells and steatosis.Hepatology. 2005; 41: 809-818Crossref PubMed Scopus (298) Google Scholar]. It remains unclear whether the response of hepatocytes to transgenic HCV protein expression is identical to that of native infection. While the levels of HCV proteins are reportedly comparable to infection, the response of hepatocytes suggest activation of the unfolded protein response (UPR), a cellular mechanism that manages misfolded or excessive accumulation of proteins [[5]Hetz C. The unfolded protein response: controlling cell fate decisions under ER stress and beyond.Nat Rev Mol Cell Biol. 2012; 13: 89-102PubMed Google Scholar]; thus, the mitochondrial involvement, leading to steatosis and lipid peroxidation could reflect the protective response of a cell confronted with a misfolded HCV protein that is not seen in native infection. It would be extremely interesting to evaluate these livers for evidence of UPR activation, whose features and methods of analysis are well characterized in both animal models and human liver disease [[6]Malhi H. Kaufman R.J. Endoplasmic reticulum stress in liver disease.J Hepatol. 2011; 54: 795-809Abstract Full Text Full Text PDF PubMed Scopus (814) Google Scholar]. Also, the mechanistic studies in the Chouteau paper are limited to interrogation of a few candidate pathways, whereas a more unbiased analysis using conventional arrays, deep sequencing and/or proteomics methods could yield novel insights into the impact of HCV proteins on hepatocytes and how they interact with other cells. More intriguing, and also meriting further exploration in the Chouteau study is the evidence of a brisk progenitor response and its correlation with reduced hepatocyte proliferation and increased fibrosis. Our understanding of the relationship between injury, hepatocyte regeneration and fibrosis remains murky but ripe for a breakthrough. Study after study continue to demonstrate a link between ductular reaction, progenitor cells, and fibrosis, yet the sequence of events and key underlying mediators are elusive. Epithelial–mesenchymal transition (EMT) is implicated by some, but not all studies as a central driver of this response, and one candidate mediator in this context may be the molecule Snail [[7]Rowe R.G. Lin Y. Shimizu-Hirota R. Hanada S. Neilson E.G. Greenson J.K. et al.Hepatocyte-derived Snail1 propagates liver fibrosis progression.Mol Cell Biol. 2011; 31: 2392-2403Crossref PubMed Scopus (99) Google Scholar]. More refined genetic models to track and manipulate progenitor cells either in this HCV model or others should clarify the meaning of the ductular reaction and the link between impaired regeneration that activates the progenitor cell response, and its association with increased fibrosis. Perhaps this series of events is unique to liver and accounts for its unique regenerative capacity compared to other tissues. The study by Polasek et al. in this issue of the Journal provides a scintilla of hope for improved fibrosis assessment using a standard non-invasive tool, MRI. The investigators explored the utility of a gadolinium-based contrast agent containing a 16 amino acid cyclic peptide that specifically binds collagen, to quantify this protein in two rodent models of fibrosis. The agent, EP-3533, was optimized to maintain the highest affinity to collagen while carrying the largest gadolinium payload [8Caravan P. Das B. Dumas S. Epstein F.H. Helm P.A. Jacques V. et al.Collagen-targeted MRI contrast agent for molecular imaging of fibrosis.Angew Chem Int Ed Engl. 2007; 46: 8171-8173Crossref PubMed Scopus (180) Google Scholar, 9Caravan P. Das B. Deng Q. Dumas S. Jacques V. Koerner S.K. et al.A lysine walk to high relaxivity collagen-targeted MRI contrast agents.Chem Commun (Camb). 2009; 4: 430-432Crossref PubMed Scopus (25) Google Scholar]. The agent was tested initially in a rat model of weekly DEN for 4 weeks, an unusual dosing scheme for a drug that is typically given only once intraperitoneally to induce hepatic tumors 6–12 months later. Thus, this new DEN model of fibrosis itself merits further evaluation, as it appears to reproducibly induce fibrosis and presumably leads to cancer as well. The imaging method was further evaluated in a CCl4 model in mice treated thrice weekly for 20 weeks, leading to dense cirrhosis. As this was a pilot feasibility study only single time points were assessed for each model, correlating delayed gadolinium persistence in liver with either hydroxyproline content (an amino acid unique to collagen) or Ishak fibrosis score. The correlation coefficient were promising but not outstanding, and might have been improved if the authors had correlated the images with either Sirius red staining or collagen I immunohistochemistry. Moreover, the authors point out that that the sensitivity of the method would likely be enhanced when used for clinical studies in humans, where the magnets used for MRI are more powerful. They also emphasize that this study was not powered to stage fibrosis, and indeed the numbers of animals used was small. Still, there is good reason to further develop this methodology for a number of reasons. Currently there is no established non-invasive test to quantify fibrosis content that can supplant liver biopsy, despite its well known limitations, including invasiveness, sampling variability, and the inability to obtain serial measurements over time. Therefore, a non-invasive test that could quantify collagen at progressive intervals using a standard technology would have enormous value not only in clinical practice, but even before then in clinical trials assessing antifibrotic therapies. Perhaps the rate of change in collagen content over time could provide a new, quantitative, and dynamic metric for assessing both disease progression and regression. Moreover, as the authors point out, an MRI test for collagen content might be combined during the same exam with studies to assess fat content and/or stiffness, providing a more comprehensive assessment of liver structure in complex conditions like non-alcoholic fatty liver disease. First, however, more animal data about this gadolinium-based MRI agent are needed, and then concerns about safety in humans will need to be thoroughly addressed. Animal studies should be extended to other models of fibrosis (including biliary fibrosis), and the collagen content and Sirius red staining must be correlated at several time points in a large number of animals, to confirm that the intensity of the MR signal corresponds exactly to the amount of hepatic collagen at each stage of disease. Studies in large animals might lend further support to the value of this new methodology, for example in pigs, where models of liver disease have been established [[10]Lee L. Alloosh M. Saxena R. Van Alstine W. Watkins B.A. Klaunig J.E. et al.Nutritional model of steatohepatitis and metabolic syndrome in the Ossabaw miniature swine.Hepatology. 2009; 50: 56-67Crossref PubMed Scopus (156) Google Scholar]. If the technology moves on to human studies, there is lingering concern about the propensity for gadolinium-based contrast agents to induce nephrogenic systemic fibrosis, especially in patients with concurrent renal impairment [[11]Hellman R.N. Gadolinium-induced nephrogenic systemic fibrosis.Semin Nephrol. 2011; 31: 310-316Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar]. While the incidence of this syndrome appears to be waning with careful patient selection, it remains a worry, and perhaps the collagen-binding cyclic peptide could be linked to a positron emission tomography (PET)-isotope rather than gadolinium, for PET imaging. Of course, PET when combined with CT raises concerns about radiation exposure. Also, patients with very advanced fibrosis and cirrhosis have intrahepatic shunting and vascular collapse that might undermine the sensitivity or accuracy of this technology, a prospect that should first be addressed in animals by testing the technology in animals with very advanced fibrosis and portal hypertension. The final question is whether even a test that measures collagen content perfectly is sufficient for use as a non-invasive diagnostic tool. It has become clear that in patients where the underlying disease is rapidly controlled, for example, in those treated with antiviral therapy, there may be a lag of up to 2 years before fibrosis scores improve after the inflammation and necrosis subside [[12]Chang T.T. Liaw Y.F. Wu S.S. Schiff E. Han K.H. Lai C.L. et al.Long-term entecavir therapy results in the reversal of fibrosis/cirrhosis and continued histological improvement in patients with chronic hepatitis B.Hepatology. 2010; 52: 886-893Crossref PubMed Scopus (744) Google Scholar]. The ideal fibrosis test would be one that detects early reductions in fibrogenic activity or increases in proteolytic activity, which occur well before the actual fibrosis content changes, yet are accurate harbingers of an improved histology and fibrosis score at a later interval. This would be an extremely powerful advance, especially for clinical trials where early evidence of efficacy could incentivize investigators and pharmaceutical sponsors to support long-term trials. Clearly, however, any MRI test suitable for clinical usage that serially measures collagen content would represent a major step forward, and could accelerate progress in the staging and treatment of patients with fibrosis liver disease, particularly those where biopsy is either contraindicated or not available. Conflict of interestThe author declared that he does not have anything to disclose regarding funding or conflict of interest with respect to this manuscript. The author declared that he does not have anything to disclose regarding funding or conflict of interest with respect to this manuscript.