Elsevier

Human Pathology

Volume 39, Issue 1, January 2008, Pages 102-115
Human Pathology

Original contribution
Evidence for the epithelial to mesenchymal transition in biliary atresia fibrosis

https://doi.org/10.1016/j.humpath.2007.05.021Get rights and content

Summary

The epithelial to mesenchymal transition has recently been implicated as a source of fibrogenic myofibroblasts in organ fibrosis, particularly in the kidney. There is as yet minimal evidence for the epithelial to mesenchymal transition in the liver. We hypothesized that this process in biliary epithelial cells plays an important role in biliary fibrosis and might be found in patients with especially rapid forms, such as is seen in biliary atresia. We therefore obtained liver tissue from patients with biliary atresia as well as a variety of other pediatric and adult liver diseases. Tissues were immunostained with antibodies against the biliary epithelial cell marker CK19 as well as with antibodies against proteins characteristically expressed by cells undergoing the epithelial to mesenchymal transition, including fibroblast-specific protein 1, the collagen chaperone heat shock protein 47, the intermediate filament protein vimentin, and the transcription factor Snail. The degree of colocalization was quantified using a multispectral imaging system. We observed significant colocalization between CK19 and other markers of the epithelial to mesenchymal transition in biliary atresia as well as other liver diseases associated with significant bile ductular proliferation, including primary biliary cirrhosis. There was minimal colocalization seen in healthy adult and pediatric livers, or in livers not also demonstrating bile ductular proliferation. Multispectral imaging confirmed significant colocalization of the different markers in biliary atresia. In conclusion, we present significant histologic evidence suggesting that the epithelial to mesenchymal transition occurs in human liver fibrosis, particularly in diseases such as biliary atresia and primary biliary cirrhosis with prominent bile ductular proliferation.

Introduction

The epithelial to mesenchymal transition (EMT) is a process in which mature epithelial cells lose the appearance, cell-cell contacts, and unique protein expression patterns of epithelia and acquire the phenotypic characteristics of mesenchymal cells. The role of EMT in embryonic development and tumor metastasis is well established. Only recently, however, has it been shown to play a role in tissue fibrosis, serving as a source of fibrogenic myofibroblasts responsible for abnormal matrix deposition. Most evidence for EMT in fibrosis comes from the kidney. Strutz et al [1] initially reported in 1995 that renal tubular epithelial cells in a mouse model of interstitial fibrosis expressed a calcium-binding protein, fibroblast-specific protein 1 (FSP1, also known as S100 A4), that otherwise appeared to be unique to fibroblasts. There have since been multiple reports that a subset of renal tubular epithelial cells in rodent models of fibrosis and in fibrotic human kidneys express both epithelial and mesenchymal markers, including FSP1, vimentin, the collagen chaperone heat shock protein 47 (HSP47), and the myofibroblast marker α-smooth muscle actin (α-SMA), and appear to be undergoing EMT (for review, see references [2], [3], [4]). Iwano et al [5] used an elegant genetic approach to demonstrate convincingly that EMT occurs in murine kidney fibrosis, lending validity to the results obtained using EMT marker proteins.

Evidence for EMT in organs other than the kidney is still emerging. In the lung, tissue from patients with idiopathic pulmonary fibrosis demonstrated coexpression of both epithelial and mesenchymal markers, suggesting that alveolar epithelial cells are a source of myofibroblasts in this disease [6]. EMT in the liver is less well established. Several groups have now shown that neonatal and adult hepatocytes can undergo EMT in vitro, although there is no published evidence of hepatocyte EMT in vivo [7], [8], [9]. EMT of cells in the biliary tree was clearly demonstrated for the first time by Xia et al [10] who showed that biliary epithelial cells (BECs) in the mouse liver 12 weeks after bile duct ligation express HSP47 and synthesize type I collagen; in vitro, a line of human intrahepatic BECs could be induced to undergo EMT after treatment with transforming growth factor β (TGF-β). In the human liver, co-staining for epithelial and mesenchymal marker proteins suggested the occurrence of EMT in a single case of recurrent primary biliary cirrhosis (PBC) after liver transplantation [11]. Another group demonstrated that the bile ductules of a single human patient with biliary atresia fibrosis expressed α-SMA and therefore concluded that these cells were undergoing EMT [12].

The increasing evidence for EMT in tissue fibrosis suggested that it might contribute to human liver disease. We hypothesized that EMT might play an important role in biliary atresia. Defined pathologically as an inflammatory sclerosing obliteration of the extrahepatic bile ducts with bile ductular proliferation and periportal fibrosis, biliary atresia is the most common indication for liver transplant in children [13]. Although adequate bile drainage can be achieved in many patients with performance of a Kasai portoenterostomy, fibrosis nonetheless generally persists and often will progress even after resolution of cholestasis. The fibrosis seen in biliary atresia is remarkable for rapidity of progression (many infants have cirrhosis at presentation—before 2 months of age). Significant fibrosis complicated by clinical signs of portal hypertension occurs in most cases, and more than 70% require liver transplantation [14]. Because there is pronounced bile ductular proliferation in biliary atresia, particularly early, we hypothesized that BECs in biliary atresia undergo EMT, potentially explaining the rapidity of the associated fibrosis. We therefore undertook a study of human liver tissue from patients with biliary atresia and a variety of other pediatric and adult liver fibroses. We provide evidence for significant EMT in biliary atresia and other diseases with bile duct proliferation.

Section snippets

Human samples

Fixed liver tissue from adult and pediatric patients (Table 1) was obtained from the surgical pathology archives of the departments of pathology at the Children's Hospital of Philadelphia (Philadelphia, PA) and the Hospital of the University of Pennsylvania (Philadelphia, PA), with the approval of the institutional review boards at both institutions. Samples were either from diagnostic biopsies or from liver explant tissue. Normal pediatric samples were obtained from autopsies in cases where

Biliary epithelial cells in biliary atresia express FSP1

EMT in tissue fibrosis was first identified in the rodent and later human kidney by staining with marker proteins typical of the EMT proteome [16]. Although a single specific EMT marker protein has not been identified, expression of the calcium-binding protein FSP1 by epithelial cells has been considered emblematic of EMT in the kidney and may be a proximal activator of EMT [16], [17]. We therefore initially used immunohistochemical staining to examine fibrotic biliary atresia livers for

Discussion

We provide here in vivo evidence for EMT in human liver fibrosis. Our data show that BECs in biliary atresia express multiple mesenchymal markers consistent with active collagen synthesis and ongoing EMT. These markers included the EMT-associated transcription factor Snail, the collagen chaperone HSP47, the intermediate filament protein vimentin, and the transcription factor FSP1, which is not only indicative of EMT in renal fibrosis but is thought to represent activation of the EMT proteome

Acknowledgments

We are grateful to Joanne Taylor and the staff of The Children's Hospital of Philadelphia histologic laboratory for assistance obtaining pediatric specimens; to Michael Feldman for assistance with microscopy; and to David Lynch and Richard Levenson of CRI, Inc, for advice on quantification.

References (46)

  • M. Beaussier et al.

    Prominent contribution of portal mesenchymal cells to liver fibrosis in ischemic and obstructive cholestatic injuries

    Lab Invest

    (2007)
  • T. Lamireau et al.

    Expression of collagens type I and IV, osteonectin and transforming growth factor beta-1 (TGFbeta1) in biliary atresia and paucity of intrahepatic bile ducts during infancy

    J Hepatol

    (1999)
  • S.Y. Lee et al.

    Identification of transforming growth factors actively transcribed during the progress of liver fibrosis in biliary atresia

    J Pediatr Surg

    (2004)
  • G. Malizia et al.

    Growth factor and procollagen type I gene expression in human liver disease

    Gastroenterology

    (1995)
  • G.A. Ramm et al.

    Contribution of hepatic parenchymal and nonparenchymal cells to hepatic fibrogenesis in biliary atresia

    Am J Pathol

    (1998)
  • R.G. Wells et al.

    Autocrine release of TGF-beta by portal fibroblasts regulates cell growth

    FEBS Lett

    (2004)
  • F. Lopez-Casillas et al.

    Betaglycan presents ligand to the TGF beta signaling receptor

    Cell

    (1993)
  • M.K. Jansen et al.

    Intracellular localization of the matrix enzyme lysyl oxidase in polarized epithelial cells

    Matrix Biol

    (2007)
  • F. Strutz et al.

    Identification and characterization of a fibroblast marker: FSP1

    J Cell Biol

    (1995)
  • J. Zavadil et al.

    TGF-beta and epithelial-to-mesenchymal transitions

    Oncogene

    (2005)
  • Y. Liu

    Epithelial to mesenchymal transition in renal fibrogenesis: pathologic significance, molecular mechanism, and therapeutic intervention

    J Am Soc Nephrol

    (2004)
  • M. Iwano et al.

    Evidence that fibroblasts derive from epithelium during tissue fibrosis

    J Clin Invest

    (2002)
  • W. Ju et al.

    Deletion of Smad2 in mouse liver reveals novel functions in hepatocyte growth and differentiation

    Mol Cell Biol

    (2006)
  • Cited by (0)

    This work was supported by National Institutes of Health R01 grant DK58123 (to R. G. W.), by a grant from the Fred and Suzanne Biesecker Center for the Study of Pediatric Liver Diseases at the Children's Hospital of Philadelphia (Philadelphia, PA), and by the Morphology Core of the University of Pennsylvania NIDDK Center for the Study of Digestive and Liver Diseases (Philadelphia, PA; P30 DK50306). R. D. is supported by the National Institutes of Health institutional training grant T32 DK07066.

    View full text