You are here

  • Home
  • Archive
  • Volume 61, Issue 6
  • Bone marrow transplantation improves hepatic fibrosis in Abcb4−/− mice via Th1 response and matrix metalloproteinase activity

Article Text

Article menu

Download PDFPDF

Hepatology
Original article
Bone marrow transplantation improves hepatic fibrosis in Abcb4−/− mice via Th1 response and matrix metalloproteinase activity
  1. Martin Roderfeld1,
  2. Timo Rath1,
  3. Sravanthi Pasupuleti1,
  4. Marc Zimmermann1,
  5. Caterina Neumann1,
  6. Yuri Churin1,
  7. Christian Dierkes2,
  8. Robert Voswinckel3,
  9. Peter J Barth2,
  10. Daniel Zahner4,
  11. Jürgen Graf5,6,
  12. Elke Roeb1
  1. 1Department of Gastroenterology, Justus-Liebig-University Giessen, Giessen, Germany
  2. 2Gerhard-Domagk-Institute of Pathology, University Hospital Münster, Münster, Germany
  3. 3Department of Pneumology, Justus-Liebig-University Giessen, Giessen, Germany
  4. 4Division of Animal Welfare and Ethology, Institute of Veterinary Physiology, Justus Liebig University Giessen, Giessen, Germany
  5. 5Aero Medical Center, Deutsche Lufthansa AG, Frankfurt am Main, Germany
  6. 6Faculty of Medicine, Philipps-University Marburg, Marburg, Germany
  1. Correspondence to Professor Dr Elke Roeb, Department of Gastroenterology, Justus-Liebig-University Giessen, Klinikstrasse 33, 35385 Gießen, Germany; elke.roeb{at}innere.med.uni-giessen.de

Abstract

Objective Reports on the effects of bone marrow-derived cells on hepatic fibrosis are contradictory. Impaired fibrosis but increased inflammation has recently been demonstrated 10 weeks after bone marrow transplantation (BM-Tx) in Abcb4−/− mice. It is hypothesised that BM-Tx might have long-term therapeutic potential by altering the immunological and matrix remodelling processes leading to hepatic regeneration.

Methods After lethal irradiation of recipient mice, BM cells from GFP+ donor mice (allogeneic Tx) or Abcb4−/− mice (syngeneic Tx) were transplanted via tail vein injection. Readouts were performed 2, 10 and 20 weeks after Tx. Liver integrity was assessed serologically and histologically. Surrogate markers for fibrogenesis, T helper (Th) response, inflammation, graft-versus-host disease and fibrolysis were analysed by quantitative real-time PCR, zymography and immunohistology.

Results 20 weeks after syngeneic and allogeneic BM-Tx, hepatic grading and staging were significantly improved. In contrast, 2 weeks after BM-Tx inflammatory grading, expression of inflammatory cell markers and associated chemokines and their receptors were increased and subsequently declined. In parallel, CD8+/GFP+ donor-derived T cells infiltrated the liver 2 weeks after BM-Tx. The Th1 cyokine interferon γ was increased 2 and 10 weeks after BM-Tx whereas the Th2 associated interleukin 13 was not altered. The gene expression of matrix metalloproteinases MMP-2, MMP-7, MMP-9 and MMP-13 was transiently upregulated and MMP-9 protein remained elevated 20 weeks after BM-Tx with enhanced gelatinase activity located within the fibrotic areas. Neutrophils were identified as major sources of MMP-9.

Conclusion These results show that BM-Tx causes an antifibrotic Th1 response combined with transient inflammatory effects and subsequently upregulated MMP activity. Antifibrotic Th polarisation and prolonged proteolytic activity, especially of MMP-9, might be responsible for long-term amelioration of hepatic fibrosis.

  • Liver fibrosis
  • T-cell polarisation
  • matrix metalloproteinase
  • cholangitis
  • stem cells
  • hepatic fibrosis
  • inflammatory bowel disease
  • liver cirrhosis
  • matrix
  • matrix metalloproteinase
  • IBD basic research
  • acute hepatitis
  • acute liver failure
  • cell cycle control
  • cell death
  • helicobacter pylori

https://doi.org/10.1136/gutjnl-2011-300608

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Significance of this study

    What is already known about this subject?

    • The therapeutic benefit of stem cell transplantation for the treatment of chronic liver disease is controversial since both a reduction in hepatic fibrosis and an increase in BM-derived fibrogenic cell pools have been shown by different experimental approaches.

    • Expression of matrix metalloproteinases MMP-9 and MMP-13 by BM-derived cells has been demonstrated during fibrolysis, and MMP-9 expression by GFP+ BM-derived donor cells and curative effects after BM-Tx in carbon tetrachloride-induced liver fibrosis have also been reported.

    What are the new findings?

    • Transiently worsened hepatic grading after bone marrow transplantation (BM-Tx) is comparable to graft-versus-host disease but both hepatic grading and staging improve as a long-term effect of BM-Tx.

    • BM-Tx causes a Th1 switch which is likely to be the key event mediating transient inflammation and antifibrotic effects and is thus crucial for the long-term amelioration of hepatic fibrosis.

    • Gene expression of MMP-2, MMP-7, MMP-9 and MMP-13 was transiently upregulated and MMP-9 protein remained elevated 20 weeks after BM-Tx with enhanced gelatinase activity located within the fibrotic areas.

    • Neutrophils were identified as major sources of MMP-9.

    How might it impact on clinical practice in the foreseeable future?

    • BM-Tx might have a therapeutic impact on clinical practice for the treatment of chronic liver disease.

    • Induction of Th1 switch as the underlying mechanism of the fibrolytic effects represents a potential tool for the treatment of liver fibrosis.

    Introduction

    The therapeutic benefit of bone marrow (BM)-derived progenitor cells for liver regeneration remains controversial.1 2 Several studies in rodents showed hepatic regeneration after transplantation of BM-derived progenitor cells.3–6 In humans, autologous stem cell therapy has been successfully used in a small number of clinical studies.7 8 In contrast, following bone marrow transplantation (BM-Tx), an increase in fibrogenic cell populations such as CD45+ fibrocytes,9 α-SMA+ myofibroblasts10 11 and hepatic stellate cells12 has been reported. We recently demonstrated that BM-derived CD34+ fibrocytes and hepatic stellate cells contribute to matrix production during fibrogenesis in BALB/c-Abcb4−/− mice.13 However, 10 weeks after allogeneic BM-Tx (ie, transplantation of GFP+→Abcb4−/− BM), fibrotic staging improved whereas hepatic grading (ie, inflammation) worsened. From those experiments it could not be concluded whether allogeneic or syngeneic BM-Tx (Abcb4−/−→Abcb4−/−) were of any long-lasting beneficial effect.13 In addition, the underlying mechanisms could not be elucidated.

    We hypothesised that BM-Tx might be able to improve long-term hepatic fibrosis in Abcb4−/− mice by as yet unknown immunomodulatory mechanisms. The aim of the present study was to determine whether BM-Tx is capable of improving long-term hepatic fibrogenesis in Abcb4−/− mice. Furthermore, we aimed to analyse the immunomodulatory mechanisms underlying altered hepatic staging and grading following BM-Tx. The post-Tx observation period was extended from 2 to 20 weeks to enable the immediate and long-term effects of both inflammation and liver fibrosis after BM-Tx to be studied.

    Materials and methods

    Animal model

    The study was performed with permission of the State of Hesse, Regierungspräsidium Giessen, according to section 8 of the German Law for the Protection of Animals and conforms to the NIH Guide for the Care and Use of Laboratory Animals. BALB/c-Abcb4−/− mice were bred and housed as described previously.14 Characterisation of BALB/c-Abcb4−/− genotype and hepatic fibrosis phenotype, sample collection and routine analysis have been described elsewhere.13

    Bone marrow transplantation (BM-Tx)

    We performed two different types of BM-Tx in Abcb4−/− mice to determine the short-term and long-term effects on fibrogenesis and inflammation. BM was transplanted from BALB/c-GFP+ reporter protein mice (allogeneic BM-Tx) or from BALB/c-Abcb4−/− mice (syngeneic control) into 6-week-old lethally irradiated (11 Gy, 60Co) BALB/c-Abcb4−/− mice. For BM-Tx, 5×106 BM donor cells were transplanted via tail vein injection. A second control group received an intravenous injection of sterile phosphate-buffered saline (PBS) without prior irradiation (sham Tx). BALB/c-GFP transgenic mice were raised from C57BL/6-TgN(ACTbEGFP)1Osb (Jackson Laboratories, Bar Harbor, USA) and crossed back on BALB/c for 10 generations. The mice were provided by Dr M Heil (Max-Planck-Institute, Bad Nauheim, Germany). Owing to the GFP marker, all tissues of these GFP transgenic mice except erythrocytes and hair follicle cells appear green under blue excitation light.15 Controls and BM-Tx mice were housed for 2, 10 and 20 weeks after BM-Tx (n=6–8 for each subgroup). The amount of GFP+ BM engraftment was evaluated by flow cytometry of peripheral blood leucocytes using FACS Calibur (Becton Dickinson, Heidelberg, Germany) as described previously.16

    Histopathology and hydroxyproline assay

    Histomorphometric analysis of inflammatory grading and hepatic fibrosis staging were independently assessed by a liver pathologist and a trained scientist (CD and MR), both of whom were blinded to the liver specimens, as described previously.13 Staging and grading using semiquantitative scores have been published elsewhere.17 18 The entire content of collagen was determined by hydroxyproline (HYP) quantification.19

    Immunohistological and immunocytological analysis

    Liver tissue samples were treated as described previously.13 Fluorescence immunostaining was performed using goat anti-matrix metalloproteinase (MMP)-9 antibodies (1:100, R&D Systems, Wiesbaden, Germany), goat anti-GFP antibodies (1:100, Rockland, Gilbertsville, USA), rat anti-CD34 (1:25, BD, Heidelberg, Germany), rat anti-CD45 antibodies (1:50, BD), rat-anti F4/80 antibodies (1:100, BMA Biochemicals, Augst, Switzerland) and mouse antimyeloperoxidase antibodies (1:50, Abcam, Cambridge, USA). Donkey- or goat-derived secondary antibodies (Fab fragment) labelled with Alexa 488 or Alexa 568 fluorescent dyes were used as secondary antibodies (1:1000 in PBS containing 5% horse (for donkey-derived antibodies) or goat serum). Prior to the use of primary mouse antibodies, cryosections were pretreated with a MOM Kit (Vector Laboratories, Burlingame, US). Co-staining of nuclei was performed with 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI; Sigma-Aldrich, Munich, Germany). The specificity of all immunofluorescence stainings was proved using equally concentrated unspecific isotype IgG instead of primary antibodies as negative controls. Fluorescence immunostaining was analysed using a fluorescence microscope (Leica DMRB, Wetzlar, Germany; camera: Nikon Coolpix 5400, Düsseldorf, Germany).

    RNA extraction, complementary DNA synthesis and quantitative real-time PCR

    mRNA isolation, cDNA synthesis, quantitative real-time PCR and quality control of all steps were performed as described before.13 Primers were purchased from Eurofins (Ebersberg, Germany) and Qiagen (Hilden, Germany). qRT-PCR data were analysed using the ΔΔCt model.20

    Interferon γ ELISA

    Interferon γ (IFNγ) ELISA was performed according to the manufacturer's instructions (eBioscience, Frankfurt, Germany).

    Zymography

    MMP-9 gelatinase activity in liver lysates was analysed by gelatin zymography in liver probes 2 weeks after BM-Tx, as described previously.21 Similar amounts of liver lysates from all mice of each group (2 weeks after sham-Tx, syngeneic and allogeneic BM-Tx) were pooled and analysed in quadruplicate. Densitometric analysis of gelatinolytic bands was perfomed with Bio Doc Analyze-Software (BiometraGmbH, Göttingen, Germany). For relative bar chart visualisation, all data were normalised to sham control.

    Statistical analysis

    Statistical analysis was performed with SPSS V.17.0 software (SPSS Inc). For non-normally distributed parameters, non-parametric tests (Mann–Whitney U test and Spearman rank test) were applied. The results are presented as mean±SD. A two-sided p<0.05 was considered significant.

    Results

    BM engraftment and hepatic infiltration by donor cells

    The infiltration of GFP+ BM cells into recipient mouse liver was demonstrated by in situ hybridisation for Y-chromosome after sex-mismatched BM-Tx (figure 1A) and applying immunofluorescence histological staining of GFP+ cells as shown previously.13

    Figure 1
    Figure 1

    Hepatic infiltration of donor cells and reduced fibrosis and inflammation 20 weeks after bone marrow transplantation (BM-Tx). (A) Hepatic infiltration of donor cells was demonstrated by in situ hybridisation for the Y chromosome after sex-mismatched BM-Tx. Green arrows indicate Y chromosome positive stained cells. Original magnification 1000×, bar=10 μm. (B,C) Photomicrographs of Masson Trichrome-stained liver sections showing reduced fibrosis after BM-Tx in (B) a 26-week-old Abcb4−/− mouse sham control and (C) a 26-week-old Abcb4−/− mouse 20 weeks after allogeneic BM-Tx. BM-Tx attenuated collagen deposition (green staining). Original magnification 100×, bar=150 μm. (D–F) H&E-stained liver sections of (D) sham-Tx control (Abcb4−/− mouse, 26-week-old sham control), (E) syngeneic BM-Tx (Abcb4−/− mouse, 26-week-old, 20 weeks after syngeneic BM-Tx) and (F) allogeneic BM-Tx (Abcb4−/− mouse, 26-week-old, 20 weeks after allogeneic BM-Tx). In BM-Tx mice the borders of the fibrotic septae are well-defined and localised within the portal tracts (dashed white line) whereas sham-treated mice have less-defined fibrotic areas exceeding the portal tracts. Enhanced piecemeal necrosis was observed in 26-week-old Abcb4−/− mice (sham control, D) while piecemeal necrosis and the grade of inflammation was ameliorated in Abcb4−/− mice of the same age 20 weeks after BM-Tx (E and F). Original magnification 400×, bar =50 μm. Stainings were performed for all animals independently. Figures are representative images for all animals of each particular group.

    Reduced hepatic fibrosis and inflammatory grading 20 weeks after BM-Tx

    Abcb4−/− mice are characterised by spontaneous and progressive biliary fibrosis. Changes in serum levels of aspartate transaminase (AST), alanine transaminase (ALT) and alkaline phosphatase are shown in online supplement 1. Progressive liver fibrosis in sham-treated Abcb4−/− mice was accompanied by hepatic fibrillar collagen accumulation (figure 1B). Compared with sham-treated mice, allogeneic BM-Tx led to a significant decrease in hepatic fibrosis (as assessed by HYP quantification and semiquantitative fibrosis staging) after 20 weeks (figure 2A,B). At this time point, reduced hepatic fibrosis was accompanied by reduced inflammatory cell infiltrates of the liver (figures 1D–F and 2C). Differences in hepatic grading 20 weeks after Tx were characterised by a considerable reduction of both piecemeal necrosis and intraparenchymal inflammation. Inflammatory grading was not altered between syngeneic and allogeneic BM-Tx groups (figure 2C).

    Figure 2
    Figure 2

    Reduced hepatic collagen levels and ameliorated pathological grading 20 weeks after bone marrow transplantation (BM-Tx). (A) Hydroxyproline (HYP) analysis revealed reduced hepatic collagen content after both syngeneic and allogeneic BM-Tx. (B) Semiquantitative fibrosis staging showed reduced fibrosis 20 weeks after allogeneic BM-Tx. (C) Inflammatory grading revealed approximately threefold enhanced inflammation in both groups 2 weeks after BM-Tx. In contrast, hepatic inflammation was ameliorated 20 weeks after BM-Tx. 6–8 mice per group were analysed independently; white column: sham control; grey column: syngeneic BM-Tx; dark column: allogeneic BM-Tx.

    Acute hepatic infiltration with inflammatory cells 2 weeks after BM-Tx

    In order to characterise the infiltrates of inflammatory cells, transcriptional levels of specific surrogate markers of cellular inflammation were analysed (figure 3). Gene expression of myeloperoxidase (surrogate for neutrophils) and CD45 (pan leucocyte surrogate marker) were significantly enhanced 2 weeks after syngeneic and allogeneic BM-Tx. In contrast, a marker for macrophages F4/80 was increased after allogeneic BM-Tx only (normalised to sham Tx; syngeneic (and allogeneic): myeloperoxidase ×4.8, p=0.001 (×11.9, p<0.001); CD45 ×4.5, p=0.002 (×7.0, p=0.001); F4/80 ×1.7, p>0.05 (×4.4, p=0.001); gene expression after allogeneic and syngeneic BM-Tx is shown in figure 3A,B). Transcriptional levels of all three surrogate markers of inflammation (ie, myeloperoxidase, CD45 and F4/80) declined from week 2 to week 20.

    Figure 3
    Figure 3

    Hepatic infiltration with inflammatory cells and increased cytokine expression 2 weeks after bone marrow transplantation (BM-Tx). (A, B) Hepatic gene regulation of specific inflammatory cell marker genes was transiently upregulated 2 weeks after (A) allogeneic and (B) syngeneic BM-Tx and declined thereafter. Neutrophil granulocyte marker myeloperoxidase, macrophage marker F4/80 and pan-leucocyte marker CD45 gene expression peaked 2 weeks after allogeneic BM-Tx. (C, D) Transcriptional regulation of cytokines and their receptors, which are associated with infiltration of neutrophil granulocytes (CXCL1, CXCL2, CXCR2) and macrophages (CCL3, CCR1, CCR2), were analysed after (C) allogeneic and (D) syngeneic BM-Tx. Except for CXCL1, regulation of chemokines and their receptors was increased 2 weeks after BM-Tx and declined thereafter to normal values in comparison with sham-treated animals. Tumour necrosis factor α (TNFα) was regulated similarly. Note the generally enhanced regulation pattern after allogeneic BM-Tx in comparison with the syngeneic BM-Tx group. All data were normalised to r18S. The mean±SEM fold increase to sham Tx is shown (6–8 mice per group were analysed independently).

    Induction of neutrophil-, granulocyte- and macrophage-associated chemokines, cytokines and their receptors

    Gene expression of tumour necrosis factor α (TNFα), CXCL2, CCL3, CXCR2, CCR1 and CCR2 was regulated comparably to neutrophil and macrophage markers. Similarly to myeloperoxidase, CD45 and F4/80, transcription of TNFα, CXCL2, CCL3, CXCR2, CCR1 and CCR2 peaked 2 weeks after BM-Tx and relapsed thereafter (figure 3C,D, allogeneic and syngeneic BM-Tx). Among all the chemokines analysed, only CCL3 remained elevated 10 weeks after syngeneic (and allogeneic) BM-Tx: CCL3 ×2.5, p=0.042 (×3.2, p=0.033). The gene expression of the neutrophil granulocyte-associated chemokine CXCL1 was not regulated over the whole post-transplantation period.

    Infiltration of donor T cells and Th1 response after BM-Tx

    To clarify whether graft-versus-host disease was present in BM-Tx-treated mice, immunostaining for CD3+ and CD8+ lymphocytes 2 weeks after allogeneic BM-Tx was performed. The co-stainings of T cell receptor CD3 and GFP revealed an infiltration of CD3+ T cells from donor BM into the fibrotic liver (figure 4A, yellow arrow; approximately 1–2 GFP+/CD3+ cells per portal tract in 3 μm slices; red cells depict host T-cells). Furthermore, GFP+/CD8a+ cells were frequently visualised within portal tracts of allogeneic transplanted mice, indicating a donor-derived cytotoxic T cell response against host liver parenchyma (figure 4B). In order to determine whether T cell polarisation influenced inflammation and fibrosis, we measured surrogate markers for Th1 and Th2 responses after allogeneic BM-Tx (figure 4C). Expression of the Th1 cytokine IFNγ was significantly enhanced 2 and 10 weeks after BM-Tx (×7.6, p=0.001 and ×2.6, p=0.002) whereas Th2-associated interleukin 13 (IL-13) was not altered. IL-10 (another Th2 cytokine) was increased 2 weeks after BM-Tx (×4.8, p=0.008) but declined thereafter. Serum IFNγ levels were upregulated 2 weeks after syngeneic and allogeneic BM-Tx in comparison with 8-week-old control animals in whom IFNγ was not detectable at all (figure 4D, 0.65 pg/ml, p=0.011, 2.48 pg/ml, p=0.010).

    Figure 4
    Figure 4

    Infiltration of donor T cells and Th1 switch after bone marrow transplantation (BM-Tx). Co-immunostaining showing the infiltration of donor-derived GFP+ T lymphocytes 2 weeks after allogeneic Tx. GFP+/CD3+ and (A) GFP+/CD8a+ (B) cells infiltrate the portal tracts 2 weeks after allogeneic BM-Tx (yellow arrows: co-stained cells). Immunostainings were performed for four animals independently. Representative micrographs of fibrotic portal tracts are shown (scale bar 25 μm; original magnification ×400). (C) Transient upregulation of the Th1 cytokine interferon γ (IFNγ) indicates a Th1 response after BM-Tx while the Th2 cytokine interleukin 13 (IL-13) is not altered following BM-Tx. Note that Th2-associated IL-10 is transiently induced 2 weeks after BM-Tx. All data were normalised to r18S. The mean±SEM fold increase to sham Tx is shown (6–8 mice per group were analysed independently). (D) Mean±SEM IFNγ serum concentrations were enhanced 2 weeks after BM-Tx in comparison with control animals.

    Transiently increased gene expression of MMPs after BM-Tx

    MMP-2, MMP-9 and MMP-13 were transiently induced 2 weeks after allogeneic and syngeneic BM-Tx and decreased thereafter (figure 5A,B). MMP-7 was increased up to 10 weeks after allogeneic BM-Tx and declined subsequently. Syngeneic BM-Tx led to an increased gene expression of MMP-7 and MMP-9 after 2 weeks and to an increased gene expression of MMP-2 and MMP-7 after 10 weeks compared with sham-treated controls. Similar to allogeneic BM-Tx mice, increased MMP expression in syngeneic mice decreased to normal levels 10 and 20 weeks after Tx, respectively. In the group of allogeneic transplanted mice, MMP-9 gene expression correlated well with IFNγ transcription (r=0.83, p<0.001) but not with IL-13 using the two-sided Spearman test (r=0.32, p=0.16).

    Figure 5
    Figure 5

    Transiently increased matrix metalloproteinase (MMP) expression and prolonged enhanced MMP-9 expression following bone marrow transplantation (BM-Tx). (A, B) Gene expression of MMP-2, MMP-7, MMP-9 and MMP-13 were induced 2 weeks after allogeneic BM-Tx. Note the enhanced regulation pattern after allogeneic BM-Tx compared with syngeneic BM-Tx. All data were normalised to r18S. The mean±SEM fold increase to the sham Tx group is shown (6–8 mice per group). (C) MMP-9 protein expression was enhanced between 2 and 20 weeks after BM-Tx. A representative western blot and loading control is shown. Equal amounts of lysate pools normalised for protein concentration were loaded. Semiquantitative analysis was conducted densitometrically (mean±SEM, 6–8 mice per group; a total of four western blots were performed).

    Prolonged and enhanced MMP-9 activity within portal tracts after BM-Tx

    Owing to the importance of MMP-9 in the context of hepatic fibrosis and fibrolysis6 22 and its strong secretion from neutrophil granulocytes, we further assessed MMP-9 activity by Western blot analysis and gelatin zymography (figure 5B and online supplement 2). MMP-9 activity complied with transcriptional regulation and thus confirmed enhanced protein expression levels even 20 weeks after BM-Tx (figure 5B; syngeneic (and allogeneic): ×1.8, p=0.021 (×3.6, p=0.021)). Gelatin zymography demonstrated enhanced MMP-9 protein expression after BM-Tx whereas MMP-2 was barely regulated.

    Furthermore, gelatin in situ zymography assays showed enhanced gelatinolytic activity within the liver tissue 20 weeks after BM-Tx (see online supplement 3). Of note, enhanced gelatinolytic activity was most prominently located in fibrotic and type I collagen-rich periportal tracts, suggesting a substantial role for MMP-9-mediated gelatinolytic activity in hepatic collagen degradation in mice undergoing BM transplantation (see online supplement 3).

    MMP-9 expression by CD45+ leucocytes and neutrophil granulocytes in portal tracts 2 weeks after BM-Tx

    To further characterise MMP-9 expressing cells, co-immunostaining of MMP-9 and CD45, myeloperoxidase, F4/80 and GFP was performed. Representative co-stained liver slices 2 weeks after allogeneic BM-Tx are shown in figure 6A–D. MMP-9 protein expression was almost always located within portal fields. Predominantly, CD45+ leucocytes and myeloperoxidase+ neutrophil granulocytes represented the major source for MMP-9 surrounding portal tracts (figure 6A–D, yellow arrows depict co-staining, red arrows depict MMP-9 staining without cell marker). Apart from MMP-9 expressing epithelial cells (figure 6D, red arrowhead), approximately 90% of MMP-9+ cells located in the portal tracts were also positive for GFP, indicating the donor origin of MMP-9 expressing cells (figure 6D).

    Figure 6
    Figure 6

    CD45+ leucocytes and myeloperoxidase+ neutrophil granulocytes were major sources of matrix metalloproteinase (MMP)-9. Co-immunostaining showed that almost all MMP-9 expressing cells (red) were also positively stained for the pan-leucocyte marker CD45 (green, A) and the neutrophil granulocyte marker myeloperoxidase (green, B). Approximately 10% of F4/80+ macrophages (green, C) exhibited co-staining for MMP-9 (red). Yellow arrows indicate co-stained cells, red arrows indicate MMP-9 expressing cells without co-staining of cellular markers (scale bar 25 μm; original magnification ×400). Most of MMP-9+ cells were also positively stained for GFP (green) and thus recruited from donor BM (D). Note that the surface of the epithelial cells in inflamed and fibrotic portal tracts was positively stained for MMP-9 (red arrowheads, D). Immunostainings were performed for four animals independently.

    Discussion

    The therapeutic benefit of stem cell transplantation to cure chronic liver disease is still controversial since both a reduction in hepatic fibrosis1 5 6 and an increase in BM-derived fibrogenic cell pools9 11 12 23 has been shown by different experimental approaches. We recently demonstrated participation of BM-derived CD34+ fibrocytes in liver fibrosis of BALB/c-Abcb4−/− mice, together with beneficial effects of BM-Tx on hepatic fibrogenesis within the same experiment.13 The underlying mechanisms, however, remained unclear. In the present study we evaluated whether immunological or proteolytic mechanisms were responsible for impaired fibrosis after BM-Tx. Our results provide evidence that BM-Tx is capable of improving hepatic fibrosis and inflammation in the long term (ie, after 20 weeks). In contrast, after 2 weeks we observed increased hepatic inflammation with donor-derived T cells, CD45+ leucocytes (especially myeloperoxidase+neutrophil granulocytes) and F4/80+ macrophages. As the underlying immunological mechanism, our results provide the first evidence of a Th1 response upon BM-Tx leading to increased expression of antifibrotic IFNγ. Furthermore, we have shown that prolonged enhanced MMP-9 activity from cellular infiltrate (mainly neutrophils) occurs within the regions of hepatic fibrosis. This enhanced and focused MMP activity may therefore be responsible for the long-term amelioration of hepatic fibrosis.

    Most fibrotic diseases have a persistent inflammatory stimulus and lymphocyte-monocyte interaction which sustains the production of profibrotic factors.24 On the other hand, chronic inflammation does not always induce fibrogenesis and the magnitude of fibrosis is tightly regulated by the polarisation of the developing Th response.24 Hepatic fibrogenesis of different aetiologies is associated with the production of the Th2-related cytokines IL-4, IL-5 and IL-13.25–27 Furthermore, it has been shown in a number of studies that the administration or induction of Th1-related cytokines such as IL-12 and IFNγ attenuates liver fibrosis.28–31 The transiently enhanced inflammation after BM-Tx described in this study reflects the enhanced expression of IFNγ, a major Th1 cytokine, while IL-13, representing the Th2 response, was not altered. In agreement with others, our observation suggests that BM-Tx induces a switch towards the antifibrotic Th1 response which is known to restrain the profibrotic Th2 activity.24 Furthermore, MMP-9 expression showed a strong positive correlation with IFNγ but not with IL-13, thereby strengthening the hypothesis that the BM-Tx-induced Th1 response was responsible for the enhanced proteolytic activity and thus for the long-term amelioration of hepatic fibrosis. Serum IFNγ levels were elevated 2 weeks after BM-Tx and declined to normal levels thereafter. Thus, the prolonged hepatic IFNγ expression is not a systemic effect but rather seems to reflect a local phenomenon in the liver.

    It was recently shown that IL-4 and IL-13 can accelerate fibrocyte differentiation towards a fibrogenic phenotype while IFNγ and IL-12 inhibit these activities.32 As fibrocytes are mainly involved in hepatic fibrogenesis in Abcb4−/− mice,13 it might be speculated that any Th1 response further exhibits the potential to diminish the profibrotic action of fibrocytes in Abcb4−/− mice.

    Furthermore, it was shown that IFNγ acts pro-apoptotic on hepatic stellate cells.33 Hepatic stellate cells, especially myofibroblasts, are considered as major fibrogenic cell population in chronic liver disease and may also be involved in the initial phase of fibrosis in Abcb−/− mice.13 34 Taken together, enhanced IFNγ expression could diminish maturation of fibrocytes and enhance apoptosis of hepatic stellate cells. This reduction in fibrogenic cells might explain the observed long-term reduction of hepatic fibrosis.

    Our observation of enhanced IL-10 expression 2 weeks after BM-Tx does not generally support the concept of a global hepatic Th1 response with BM-Tx. However, the role of IL-10 in hepatic fibrosis remains controversial as both profibrotic and antifibrotic actions have been linked to IL-10.35–42 Furthermore, the combined antifibrotic action of IL-10 and Th1 cytokines is a well-known mechanism35 43 44 which might be strengthened by the results of our study.

    It has recently been demonstrated that hepatic infiltration by neutrophils and macrophages following hepatocyte transplantation is driven by cytokines and chemokines.45 In our model we observed a transiently increased expression of TNFα, CCL3 and CXCL2 (figure 3), which are mainly secreted from neutrophils and macrophages.46 47 CCL3, a chemotactic factor for monocytes,48 binds to its receptors CCR1 and CCR5 while CXCL2, a potent attractor of neutrophils towards sites of inflammation,49 binds to CXCR2.50 In the present study, gene expression of the chemokine receptors CCR1, CXCR2 and CCR2 was regulated comparably to the respective cytokines. These transcriptional data are strongly supported by our immunohistochemical studies in which we clearly identified infiltration of neutrophils and macrophages into the liver following BM-Tx.

    The activation of neutrophils and macrophages as an initial response to TNFα secretion has already been described after hepatocyte transplantation.45 The release of TNFα by macrophages might have been stimulated by neutrophils51 and may be capable of inducing the expression of chemokines by parenchymal cells.52 Thus, TNFα is of considerable relevance in liver inflammation following cell transplantation.45

    More important in the context of hepatic fibrosis, TNFα inhibits both proliferation and apotosis in activated hepatic stellate cells, probably by keeping cells in a G1 arrest.53 Further, it has been shown recently that dendritic cells from fibrotic livers exhibit a distinct phenotype indicative of enhanced maturity. These fibrotic dendritic cells have a marked capacity to induce hepatic stellate cells, NK cells and T cells to mediate inflammation, proliferation and production of potent immune responses. Importantly, these proinflammatory and immunogenic effects of fibrotic dendritic cells were contingent on their production of TNFα.54

    On the other hand, TNFα can activate the proapoptotic caspase cascades via TRADD, FADD and cleavage of procaspase 8. In this context, recent evidence suggests that, in the liver of streptozotocin-induced diabetic rats, hepatic TNFα is increased and leads to increased apoptotic cell death mediated through caspase 8, NF-κB and JNK pathways.55

    Given the pleiotropic function of TNFα with broad inflammatory properties, a final assessment of the predominant cellular effect on liver homeostasis and pathology cannot yet be made. Proapoptotic and antiapoptotic effects, as well as no effects on apoptosis, have been described, showing that the effect of TNFα depends on the cell type investigated and also on the model in which hepatic damage and subsequent regeneration are studied.56–62 In our study, TNFα showed highest expression levels 2 weeks after BM-Tx in both allogeneic and syngeneic transplanted mice and declined subsequently. Although the pathophysiological relevance of the observed transient increase in TNFα in the current study remains unclear, it might be speculated that transiently increased TNFα in the first line results from acute hepatic infiltration with various inflammatory cells after BM-Tx rather than exhibiting sustained effects on hepatic matrix metabolism subsequent to BM-Tx.

    The influx of inflammatory cells around intrahepatic bile ducts after irradiation and subsequent allogeneic BM-Tx (but not after irradiation alone) is a well-characterised phenomenon.63 Allogeneic BM-Tx represents a therapeutic option for the treatment of hepatocellular carcinoma. The observed (and desired) graft-versus-tumour effect went along with graft-versus-host disease (GvHD).63 Subsequent to the initial damage of endothelia and epithelia by conditioning treatments, proinflammatory factors were secreted which facilitated the infiltration of donor T cells.64 Adjacent activation and proliferation of donor T cells resulted from the recognition of host minor histocompatibility (MHC) antigens presented by MHC class I and class II molecules.65 The clinical surrogate of GvHD was exemplified by cytotoxic impairment of primary target organs such as skin, gut, liver and lung.65 Acute hepatic GvHD is a frequent complication after BM-Tx and eventually resembles acute hepatitis, while the chronic features are characteristic of primary biliary cirrhosis.66–68 Two weeks after allogeneic BM-Tx we were able to identify GFP+/CD3+ donor T cells, known to be key players in GvHD. Furthermore, hepatic infiltration of donor-derived CD8+ cytotoxic T cells was seen.

    CCL3/MIP-1a is a member of the CC chemokine subfamily that has been shown to be expressed in GvHD and to affect T cell recruitment and proliferation in GvHD.69 In the final phase of GvHD, reactive effector T cells are recruited by chemokines including CCL3 into target organs such as the liver, resulting in host tissue injury.69 Recently, the relevance of CCL3 in GvHD in mice was confirmed by blockade of CCL3 with its specific binding protein evasin-1.70 In the current study, CCL3 transcription was upregulated even 10 weeks after syngeneic and allogeneic BM-Tx. This strengthens the idea that transient inflammation is caused by GvHD and that inflammation-dependent enhanced MMP expression is thereby a direct cause of BM-Tx.

    Our results therefore clearly identify an influx of donor T cells into the liver of the recipient. Thus, transient inflammation after BM-Tx in Abcb4−/− mice might be partially based on GvHD, although direct proof is not currently available.

    MMPs and TIMPs play a central role in matrix homeostasis and remodelling processes during hepatic fibrogenesis and fibrolysis.71 72 Acute hepatic inflammation subsequent to BM-Tx was accompanied by transiently enhanced expression of fibrolytic MMPs (figure 4). MMP-9 was expressed by CD45+ leucocytes, myeloperoxidase+ neutrophil granulocytes and CD34+ fibrocytes whereas only about 10% of all F4/80+ macrophages located in fibrotic portal tracts were positive for MMP-9 (figure 5). Apart from these cells, stellate cells and portal fibroblasts have been described as a source of MMP-9 secretion before,73 and in the current study we cannot rule out the possibility that both cell types represent a further source of MMP-9.

    Using gelatin in situ zymography assays, we were able to assign enhanced gelatinolytic activity most prominently to fibrotic type I collagen-rich periportal tracts. Our data indicate that MMP-9-mediated gelatinolytic activity facilitates hepatic matrix degradation and fibrolysis in mice undergoing BM-Tx.

    Higashiyama and coworkers observed the expression of MMP-9 and MMP-13 by BM-derived cells during fibrolysis22 and Sakaida et al demonstrated MMP-9 expression by GFP+ BM-derived donor cells after BM-Tx in carbon tetrachloride-induced liver fibrosis.6 These data are in accordance with the results of our study as the major amount of MMP-9 in Abcb4−/− mouse liver was expressed by GFP+/CD45+/myeloperoxidase+ cells 2 weeks after allogeneic BM-Tx. Apart from MMP-9 expression by donor-derived cells, Sakaida and coworkers demonstrated impaired hepatic fibrosis 4 weeks after BM-Tx of particular haematopoietic subsets (Liv8− cells). As Liv8 antibodies were derived by the use of fetal liver as antigen, it is difficult to discuss the character of those particular subsets of Liv8− cells.74 Nevertheless, it can be concluded that the studies available including our current report suggest that enhanced MMP expression from BM-derived cells may contribute to fibrolysis and thus to improved liver fibrosis following BM-Tx. Our study has shown for the first time that these effects are long-lasting and can improve liver staging and grading.

    Most of the inflammatory effects observed in the present study were enhanced after allogeneic BM-Tx and were less pronounced after syngeneic BM-Tx. Recently it was demonstrated that genetic drift in mouse inbred strains had a significant impact on the alloreactive immune response caused by altered MHC antigens.75 This might explain the immune reactions after syngeneic BM-Tx and the enhanced effects after allogeneic BM-Tx, although both GFP+ donor mice and Abcb4−/− mice were bred on the same genetic background (BALB/c).

    In conclusion, the results of the present study demonstrate that transiently worsened hepatic grading after BM-Tx is comparable to GvHD, whereas both hepatic grading and staging improve as a long-term effect of BM-Tx. Furthermore, our data provide evidence that transient hepatic influx of inflammatory cells subsequent to BM-Tx represents a main source of proteolytic enzymes and leads to prolonged proteolytic activity, especially of MMP-9. This enhanced proteolytic activity constitutes the basis for improvement of liver fibrosis in Abcb4−/− mice. BM-Tx causes a Th1 switch which is likely to be the key event mediating transient inflammation and antifibrotic effects and is thus crucial for the long-term amelioration of hepatic fibrosis.

    Acknowledgments

    The authors thank Dr M Heil (Max-Planck-Institute, Bad Nauheim, Germany) for BALB/c-GFP+ mice, Annette Tschuschner and Michaela Weiss for excellent technical assistance and Dr Walbott and Dr Lugert of Strahlenzentrum Justus-Liebig University Gießen for irradiation experiments.

    References

    1. ↵
      1. Higashiyama R,
      2. Moro T,
      3. Nakao S,
      4. et al
      . Negligible contribution of bone marrow-derived cells to collagen production during hepatic fibrogenesis in mice. Gastroenterology 2009;137:1459–66.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Kallis YN,
      2. Forbes SJ
      . The bone marrow and liver fibrosis: friend or foe? Gastroenterology 2009;137:1218–21.
      OpenUrlCrossRefPubMed
    3. ↵
      1. Fang B,
      2. Shi M,
      3. Liao L,
      4. et al
      . Systemic infusion of FLK1(+) mesenchymal stem cells ameliorate carbon tetrachloride-induced liver fibrosis in mice. Transplantation 2004;78:83–8.
      OpenUrlPubMed
    4. ↵
      1. Harb R,
      2. Xie G,
      3. Lutzko C,
      4. et al
      . Bone marrow progenitor cells repair rat hepatic sinusoidal endothelial cells after liver injury. Gastroenterology 2009;137:704–12.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Nakamura T,
      2. Torimura T,
      3. Sakamoto M,
      4. et al
      . Significance and therapeutic potential of endothelial progenitor cell transplantation in a cirrhotic liver rat model. Gastroenterology 2007;133:91–107.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. Sakaida I,
      2. Terai S,
      3. Yamamoto N,
      4. et al
      . Transplantation of bone marrow cells reduces CCl4-induced liver fibrosis in mice. Hepatology 2004;40:1304–11.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. am Esch JS,
      2. Knoefel WT,
      3. Klein M,
      4. et al
      . Portal application of autologous CD133+ bone marrow cells to the liver: a novel concept to support hepatic regeneration. Stem Cells 2005;23:463–70.
      OpenUrlCrossRefPubMed
    8. ↵
      1. Terai S,
      2. Ishikawa T,
      3. Omori K,
      4. et al
      . Improved liver function in patients with liver cirrhosis after autologous bone marrow cell infusion therapy. Stem Cells 2006;24:2292–8.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Kisseleva T,
      2. Uchinami H,
      3. Feirt N,
      4. et al
      . Bone marrow-derived fibrocytes participate in pathogenesis of liver fibrosis. J Hepatol 2006;45:429–38.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Forbes SJ,
      2. Russo FP,
      3. Rey V,
      4. et al
      . A significant proportion of myofibroblasts are of bone marrow origin in human liver fibrosis. Gastroenterology 2004;126:955–63.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Russo FP,
      2. Alison MR,
      3. Bigger BW,
      4. et al
      . The bone marrow functionally contributes to liver fibrosis. Gastroenterology 2006;130:1807–21.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Miyata E,
      2. Masuya M,
      3. Yoshida S,
      4. et al
      . Hematopoietic origin of hepatic stellate cells in the adult liver. Blood 2008;111:2427–35.
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Roderfeld M,
      2. Rath T,
      3. Voswinckel R,
      4. et al
      . Bone marrow transplantation demonstrates medullar origin of CD34+ fibrocytes and ameliorates hepatic fibrosis in Abcb4−/− mice. Hepatology 2010;51:267–76.
      OpenUrlCrossRefPubMed
    14. ↵
      1. Henkel C,
      2. Roderfeld M,
      3. Weiskirchen R,
      4. et al
      . Changes of the hepatic proteome in murine models for toxically induced fibrogenesis and sclerosing cholangitis. Proteomics 2006;6:6538–48.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Okabe M,
      2. Ikawa M,
      3. Kominami K,
      4. et al
      . 'Green mice' as a source of ubiquitous green cells. FEBS Lett 1997;407:313–19.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Voswinckel R,
      2. Ziegelhoeffer T,
      3. Heil M,
      4. et al
      . Circulating vascular progenitor cells do not contribute to compensatory lung growth. Circ Res 2003;93:372–9.
      OpenUrlAbstract/FREE Full Text
    17. ↵
      1. Batts KP,
      2. Ludwig J
      . Chronic hepatitis. An update on terminology and reporting. Am J Surg Pathol 1995;19:1409–17.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Ishak K,
      2. Baptista A,
      3. Bianchi L,
      4. et al
      . Histological grading and staging of chronic hepatitis. J Hepatol 1995;22:696–9.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Jamall IS,
      2. Finelli VN,
      3. Que Hee SS
      . A simple method to determine nanogram levels of 4-hydroxyproline in biological tissues. Anal Biochem 1981;112:70–5.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Pfaffl MW
      . A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 2001;29:2002–7.
      OpenUrlCrossRef
    21. ↵
      1. Roeb E,
      2. Behrmann I,
      3. Grotzinger J,
      4. et al
      . An MMP-9 mutant without gelatinolytic activity as a novel TIMP-1-antagonist. FASEB J 2000;14:1671–3.
      OpenUrlFREE Full Text
    22. ↵
      1. Higashiyama R,
      2. Inagaki Y,
      3. Hong YY,
      4. et al
      . Bone marrow-derived cells express matrix metalloproteinases and contribute to regression of liver fibrosis in mice. Hepatology 2007;45:213–22.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. Baba S,
      2. Fujii H,
      3. Hirose T,
      4. et al
      . Commitment of bone marrow cells to hepatic stellate cells in mouse. J Hepatol 2004;40:255–60.
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Wynn TA
      . Fibrotic disease and the T(H)1/T(H)2 paradigm. Nat Rev Immunol 2004;4:583–94.
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Borg BB,
      2. Seetharam A,
      3. Subramanian V,
      4. et al
      . Immune response to extracellular matrix collagen in chronic hepatitis C-induced liver fibrosis. Liver Transpl 2011;17:814–23.
      OpenUrlCrossRefPubMed
    26. ↵
      1. Coutinho HM,
      2. Acosta LP,
      3. Wu HW,
      4. et al
      . Th2 cytokines are associated with persistent hepatic fibrosis in human Schistosoma japonicum infection. J Infect Dis 2007;195:288–95.
      OpenUrlAbstract/FREE Full Text
    27. ↵
      1. Reiman RM,
      2. Thompson RW,
      3. Feng CG,
      4. et al
      . Interleukin-5 (IL-5) augments the progression of liver fibrosis by regulating IL-13 activity. Infect Immun 2006;74:1471–9.
      OpenUrlAbstract/FREE Full Text
    28. ↵
      1. Gao B,
      2. Radaeva S,
      3. Jeong WI
      . Activation of natural killer cells inhibits liver fibrosis: a novel strategy to treat liver fibrosis. Expert Rev Gastroenterol Hepatol 2007;1:173–80.
      OpenUrlCrossRefPubMed
    29. ↵
      1. Ha MH,
      2. Wei L,
      3. Rao HY,
      4. et al
      . Effect of interferon-gamma on hepatic stellate cells stimulated by acetaldehyde. Hepatogastroenterology 2008;55:1059–65.
      OpenUrlPubMed
    30. ↵
      1. Wynn TA,
      2. Cheever AW,
      3. Jankovic D,
      4. et al
      . An IL-12-based vaccination method for preventing fibrosis induced by schistosome infection. Nature 1995;376:594–6.
      OpenUrlCrossRefPubMed
    31. ↵
      1. Yao H,
      2. Pan J,
      3. Qian Y,
      4. et al
      . Enhanced effect of soluble transforming growth factor-beta receptor II and IFN-gamma fusion protein in reversing hepatic fibrosis. Eur J Med Res 2010;15:152–61.
      OpenUrlCrossRefPubMed
    32. ↵
      1. Shao DD,
      2. Suresh R,
      3. Vakil V,
      4. et al
      . Pivotal Advance: Th-1 cytokines inhibit, and Th-2 cytokines promote fibrocyte differentiation. J Leukoc Biol 2008;83:1323–33.
      OpenUrlAbstract/FREE Full Text
    33. ↵
      1. Saile B,
      2. Eisenbach C,
      3. Dudas J,
      4. et al
      . Interferon-gamma acts proapoptotic on hepatic stellate cells (HSC) and abrogates the antiapoptotic effect of interferon-alpha by an HSP70-dependant pathway. Eur J Cell Biol 2004;83:469–76.
      OpenUrlCrossRefPubMedWeb of Science
    34. ↵
      1. Popov Y,
      2. Patsenker E,
      3. Fickert P,
      4. et al
      . Mdr2 (Abcb4)-/- mice spontaneously develop severe biliary fibrosis via massive dysregulation of pro- and antifibrogenic genes. J Hepatol 2005;43:1045–54.
      OpenUrlCrossRefPubMedWeb of Science
    35. ↵
      1. Booth M,
      2. Mwatha JK,
      3. Joseph S,
      4. et al
      . Periportal fibrosis in human Schistosoma mansoni infection is associated with low IL-10, low IFN-gamma, high TNF-alpha, or low RANTES, depending on age and gender. J Immunol 2004;172:1295–303.
      OpenUrlAbstract/FREE Full Text
    36. ↵
      1. Herbert DR,
      2. Orekov T,
      3. Perkins C,
      4. et al
      . IL-10 and TGF-beta redundantly protect against severe liver injury and mortality during acute schistosomiasis. J Immunol 2008;181:7214–20.
      OpenUrlAbstract/FREE Full Text
    37. ↵
      1. Louis H,
      2. Van Laethem JL,
      3. Wu W,
      4. et al
      . Interleukin-10 controls neutrophilic infiltration, hepatocyte proliferation, and liver fibrosis induced by carbon tetrachloride in mice. Hepatology 1998;28:1607–15.
      OpenUrlCrossRefPubMedWeb of Science
    38. ↵
      1. Louis H,
      2. Le MA,
      3. Quertinmont E,
      4. et al
      . Repeated concanavalin A challenge in mice induces an interleukin 10-producing phenotype and liver fibrosis. Hepatology 2000;31:381–90.
      OpenUrlCrossRefPubMedWeb of Science
    39. ↵
      1. Louis H,
      2. Le MO,
      3. Goldman M,
      4. et al
      . Modulation of liver injury by interleukin-10. Acta Gastroenterol Belg 2003;66:7–14.
      OpenUrlPubMed
    40. ↵
      1. Tang-Feldman YJ,
      2. Lochhead GR,
      3. Lochhead SR,
      4. et al
      . Interleukin-10 repletion suppresses pro-inflammatory cytokines and decreases liver pathology without altering viral replication in murine cytomegalovirus (MCMV)-infected IL-10 knockout mice. Inflamm Res 2011;60:233–43.
      OpenUrlCrossRefPubMed
    41. ↵
      1. Thompson K,
      2. Maltby J,
      3. Fallowfield J,
      4. et al
      . Interleukin-10 expression and function in experimental murine liver inflammation and fibrosis. Hepatology 1998;28:1597–606.
      OpenUrlCrossRefPubMedWeb of Science
    42. ↵
      1. Yin S,
      2. Wang H,
      3. Park O,
      4. et al
      . Enhanced liver regeneration in IL-10-deficient mice after partial hepatectomy via stimulating inflammatory response and activating hepatocyte STAT3. Am J Pathol 2011;178:1614–21.
      OpenUrlCrossRefPubMed
    43. ↵
      1. Hoffmann KF,
      2. Cheever AW,
      3. Wynn TA
      . IL-10 and the dangers of immune polarization: excessive type 1 and type 2 cytokine responses induce distinct forms of lethal immunopathology in murine schistosomiasis. J Immunol 2000;164:6406–16.
      OpenUrlAbstract/FREE Full Text
    44. ↵
      1. Vaillant B,
      2. Chiaramonte MG,
      3. Cheever AW,
      4. et al
      . Regulation of hepatic fibrosis and extracellular matrix genes by the th response: new insight into the role of tissue inhibitors of matrix metalloproteinases. J Immunol 2001;167:7017–26.
      OpenUrlAbstract/FREE Full Text
    45. ↵
      1. Krohn N,
      2. Kapoor S,
      3. Enami Y,
      4. et al
      . Hepatocyte transplantation-induced liver inflammation is driven by cytokines-chemokines associated with neutrophils and Kupffer cells. Gastroenterology 2009;136:1806–17.
      OpenUrlCrossRefPubMed
    46. ↵
      1. Li X,
      2. Klintman D,
      3. Liu Q,
      4. et al
      . Critical role of CXC chemokines in endotoxemic liver injury in mice. J Leukoc Biol 2004;75:443–52.
      OpenUrlAbstract/FREE Full Text
    47. ↵
      1. Moreno C,
      2. Gustot T,
      3. Nicaise C,
      4. et al
      . CCR5 deficiency exacerbates T-cell-mediated hepatitis in mice. Hepatology 2005;42:854–62.
      OpenUrlCrossRefPubMedWeb of Science
    48. ↵
      1. Gaga M,
      2. Ong YE,
      3. Benyahia F,
      4. et al
      . Skin reactivity and local cell recruitment in human atopic and nonatopic subjects by CCL2/MCP-1 and CCL3/MIP-1alpha. Allergy 2008;63:703–11.
      OpenUrlCrossRefPubMed
    49. ↵
      1. Rittner HL,
      2. Mousa SA,
      3. Labuz D,
      4. et al
      . Selective local PMN recruitment by CXCL1 or CXCL2/3 injection does not cause inflammatory pain. J Leukoc Biol 2006;79:1022–32.
      OpenUrlAbstract/FREE Full Text
    50. ↵
      1. Allen SJ,
      2. Crown SE,
      3. Handel TM
      . Chemokine: receptor structure, interactions, and antagonism. Annu Rev Immunol 2007;25:787–820.
      OpenUrlCrossRefPubMedWeb of Science
    51. ↵
      1. Wilsson A,
      2. Lind S,
      3. Ohman L,
      4. et al
      . Apoptotic neutrophils containing Staphylococcus epidermidis stimulate macrophages to release the proinflammatory cytokines tumor necrosis factor-alpha and interleukin-6. FEMS Immunol Med Microbiol 2008;53:126–35.
      OpenUrlAbstract/FREE Full Text
    52. ↵
      1. Liu LY,
      2. Bates ME,
      3. Jarjour NN,
      4. et al
      . Generation of Th1 and Th2 chemokines by human eosinophils: evidence for a critical role of TNF-alpha. J Immunol 2007;179:4840–8.
      OpenUrlAbstract/FREE Full Text
    53. ↵
      1. Saile B,
      2. Matthes N,
      3. Knittel T,
      4. et al
      . Transforming growth factor beta and tumor necrosis factor alpha inhibit both apoptosis and proliferation of activated rat hepatic stellate cells. Hepatology 1999;30:196–202.
      OpenUrlCrossRefPubMedWeb of Science
    54. ↵
      1. Connolly MK,
      2. Bedrosian AS,
      3. Mallen-St CJ,
      4. et al
      . In liver fibrosis, dendritic cells govern hepatic inflammation in mice via TNF-alpha. J Clin Invest 2009;119:3213–25.
      OpenUrlPubMedWeb of Science
    55. ↵
      1. Ingaramo PI,
      2. Ronco MT,
      3. Frances DE,
      4. et al
      . Tumor necrosis factor alpha pathways develops liver apoptosis in type 1 diabetes mellitus. Mol Immunol 2011;48:1397–407.
      OpenUrlCrossRefPubMed
    56. ↵
      1. Bachem MG,
      2. Riess U,
      3. Gressner AM
      . Liver fat storing cell proliferation is stimulated by epidermal growth factor/transforming growth factor alpha and inhibited by transforming growth factor beta. Biochem Biophys Res Commun 1989;162:708–14.
      OpenUrlCrossRefPubMed
    57. ↵
      1. Badley AD,
      2. Dockrell D,
      3. Simpson M,
      4. et al
      . Macrophage-dependent apoptosis of CD4+ T lymphocytes from HIV-infected individuals is mediated by FasL and tumor necrosis factor. J Exp Med 1997;185:55–64.
      OpenUrlAbstract/FREE Full Text
    58. ↵
      1. Davis BH
      . Transforming growth factor beta responsiveness is modulated by the extracellular collagen matrix during hepatic ito cell culture. J Cell Physiol 1988;136:547–53.
      OpenUrlCrossRefPubMedWeb of Science
    59. ↵
      1. Davis BH,
      2. Rapp UR,
      3. Davidson NO
      . Retinoic acid and transforming growth factor beta differentially inhibit platelet-derived-growth-factor-induced Ito-cell activation. Biochem J 1991;278:43–7.
      OpenUrlAbstract/FREE Full Text
    60. ↵
      1. Knittel T,
      2. Muller L,
      3. Saile B,
      4. et al
      . Effect of tumour necrosis factor-alpha on proliferation, activation and protein synthesis of rat hepatic stellate cells. J Hepatol 1997;27:1067–80.
      OpenUrlCrossRefPubMedWeb of Science
    61. ↵
      1. Meyer DH,
      2. Bachem MG,
      3. Gressner AM
      . Bidirectional effects of Kupffer cells on hepatocyte proliferation in vitro. FEBS Lett 1991;283:150–4.
      OpenUrlCrossRefPubMed
    62. ↵
      1. Pinzani M,
      2. Gesualdo L,
      3. Sabbah GM,
      4. et al
      . Effects of platelet-derived growth factor and other polypeptide mitogens on DNA synthesis and growth of cultured rat liver fat-storing cells. J Clin Invest 1989;84:1786–93.
      OpenUrlCrossRefPubMedWeb of Science
    63. ↵
      1. Yoshida Y,
      2. Hirano T,
      3. Son G,
      4. et al
      . Allogeneic bone marrow transplantation for hepatocellular carcinoma: hepatocyte growth factor suppresses graft-vs.-host disease. Am J Physiol Gastrointest Liver Physiol 2007;293:1114–23.
      OpenUrlCrossRef
    64. ↵
      1. Ichiki Y,
      2. Bowlus CL,
      3. Shimoda S,
      4. et al
      . T cell immunity and graft-versus-host disease (GVHD). Autoimmun Rev 2006;5:1–9.
      OpenUrlCrossRefPubMedWeb of Science
    65. ↵
      1. Lee SJ
      . New approaches for preventing and treating chronic graft-versus-host disease. Blood 2005;105:4200–6.
      OpenUrlAbstract/FREE Full Text
    66. ↵
      1. Chiba T,
      2. Yokosuka O,
      3. Goto S,
      4. et al
      . Clinicopathological features in patients with hepatic graft-versus-host disease. Hepatogastroenterology 2005;52:1849–53.
      OpenUrlPubMed
    67. ↵
      1. Hidaka M,
      2. Iwasaki S,
      3. Matsui T,
      4. et al
      . Efficacy of bezafibrate for chronic GVHD of the liver after allogeneic hematopoietic stem cell transplantation. Bone Marrow Transplant 2010;45:912–18.
      OpenUrlCrossRefPubMed
    68. ↵
      1. Wakae T,
      2. Takatsuka H,
      3. Seto Y,
      4. et al
      . Similarity between hepatic graft-versus-host disease and primary biliary cirrhosis. Hematology 2002;7:305–10.
      OpenUrlCrossRefPubMed
    69. ↵
      1. Serody JS,
      2. Burkett SE,
      3. Panoskaltsis-Mortari A,
      4. et al
      . T-lymphocyte production of macrophage inflammatory protein-1alpha is critical to the recruitment of CD8(+) T cells to the liver, lung, and spleen during graft-versus-host disease. Blood 2000;96:2973–80.
      OpenUrlAbstract/FREE Full Text
    70. ↵
      1. Castor MG,
      2. Rezende B,
      3. Resende CB,
      4. et al
      . The CCL3/macrophage inflammatory protein-1alpha-binding protein evasin-1 protects from graft-versus-host disease but does not modify graft-versus-leukemia in mice. J Immunol 2010;184:2646–54.
      OpenUrlAbstract/FREE Full Text
    71. ↵
      1. Hemmann S,
      2. Graf J,
      3. Roderfeld M,
      4. et al
      . Expression of MMPs and TIMPs in liver fibrosis—a systematic review with special emphasis on anti-fibrotic strategies. J Hepatol 2007;46:955–75.
      OpenUrlCrossRefPubMedWeb of Science
    72. ↵
      1. Roderfeld M,
      2. Hemmann S,
      3. Roeb E
      . Mechanisms of fibrinolysis in chronic liver injury (with special emphasis on MMPs and TIMPs). Z Gastroenterol 2007;45:25–33.
      OpenUrlCrossRefPubMed
    73. ↵
      1. Knittel T,
      2. Mehde M,
      3. Kobold D,
      4. et al
      . Expression patterns of matrix metalloproteinases and their inhibitors in parenchymal and non-parenchymal cells of rat liver: regulation by TNF-alpha and TGF-beta1. J Hepatol 1999;30:48–60.
      OpenUrlCrossRefPubMedWeb of Science
    74. ↵
      1. Yamamoto N,
      2. Terai S,
      3. Ohata S,
      4. et al
      . A subpopulation of bone marrow cells depleted by a novel antibody, anti-Liv8, is useful for cell therapy to repair damaged liver. Biochem Biophys Res Commun 2004;313:1110–18.
      OpenUrlCrossRefPubMedWeb of Science
    75. ↵
      1. Fanning SL,
      2. Appel MY,
      3. Berger SA,
      4. et al
      . The immunological impact of genetic drift in the B10.BR congenic inbred mouse strain. J Immunol 2009;183:4261–72.
      OpenUrlAbstract/FREE Full Text

    Footnotes

    • MR and TR contributed equally to this paper.

    • Funding This work was supported by grants from the Deutsche Forschungsgemeinschaft (RO 957/8-1) and by a Research Grant from the University Medical Center Giessen and Marburg (UKGM 10/2010 GI). TR received starting grants from the Justus-Liebig-University Giessen.

    • Competing interests None.

    • Provenance and peer review Not commissioned; externally peer reviewed.