Gastrointestinal
A Critical Role for Matrix Metalloproteinases in Liver Regeneration1

https://doi.org/10.1016/j.jss.2007.04.002Get rights and content

Background

Matrix metalloproteinases (MMPs), tumor necrosis factor-alpha (TNF-α), and interleukin-6 (IL-6) are mediators of liver regeneration. To determine whether MMPs are required for normal hepatic regeneration, we performed 67% hepatectomies on mice treated with a broad-spectrum MMP-inhibitor, and assessed the effect on liver regeneration and urinary MMP activity.

Methods

Mice were subjected to sham operations, 67% hepatectomy, or 67% hepatectomy plus treatment with the broad-spectrum MMP inhibitor Marimastat. Urine collected preoperatively and for 8 d postoperatively was tested for MMP-2 and MMP-9 activity using zymography. Serum aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, bilirubin, TNF-α, IL-6, and hepatocyte growth factor levels were measured. Liver sections were analyzed by CD31 immunohistochemistry and microvessel density. Mitotic index and proliferating cell nuclear antigen labeling index were determined.

Results

The mean regenerating liver weight on postoperative day 8 was 0.72 ± 0.01 grams for the hepatectomy Marimastat group, and 0.83 ± 0.02 grams for the hepatectomy control group (P < 0.001). Urinary MMP-9 activity was elevated during hepatic regeneration, and decreased on postoperative day 8 when the liver returned to its preoperative mass. In contrast, urine from hepatectomy Marimastat mice, in which liver regeneration was successfully inhibited, showed consistently low levels of MMP-2 and MMP-9 activity. The hepatectomy Marimastat group also exhibited elevated serum IL-6 levels on post-operative day 8, while serum TNF-α soluble receptor II levels were unchanged. Hepatocyte growth factor levels were not significantly different between the control hepatectomy and hepatectomy Marimastat groups at days 2, 4, and 8. Liver microvessel density was reduced in the hepatectomy Marimastat group at day 4. Mitotic index and proliferating cell nuclear antigen index were significantly decreased in the Marimastat hepatectomy group at post-operative day 2.

Conclusions

The broad-spectrum MMP-inhibitor Marimastat inhibits liver regeneration. Microvessel density is reduced at day 4. Furthermore, urinary MMP-9 is elevated during liver regeneration, and this effect is not observed when regeneration is inhibited by the broad-spectrum MMP-inhibitor Marimastat.

Introduction

Liver regeneration is an essential physiological response after hepatic injury, resection, or transplantation. The liver is a unique organ in that regeneration occurs by DNA replication and mitosis, whereas most other organs simply hypertrophy [1]. Stimulators of hepatic regeneration include hepatocyte (HGF) and epidermal growth factor, transforming growth factor-alpha, and the potent stimulators of angiogenesis, fibroblast growth factor and vascular endothelial growth factor [2]. In addition, interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α) are associated with liver regeneration [3, 4, 5].

IL-6 and TNF-α are both considered to be priming factors for regeneration because they prepare hepatocytes for replication [6]. IL-6 initiates the acute phase response in the liver, and activates STAT3, a transcription factor that initiates hepatocyte replication [7]. Therefore, it appears that IL-6 plays both a priming and mitogenic role in hepatic regeneration. Further demonstrating the importance of this cytokine in the regenerative process, IL-6 deficient mice demonstrate abnormal regeneration after partial hepatectomy [8]. Although it has been difficult to test the result of giving mice large doses of IL-6 due to systemic toxicity, lower doses administered over a long period of time have resulted in livers that regenerate beyond their original mass [6]. While IL-6 activates STAT3, TNF-α activates nuclear factor kappa B (NF-κB) a similar transcription factor [9, 10]. The main action of TNF-α is thought to be the release of IL-6, resulting in the cascade described above. While TNF-α deficient mice do not regenerate their livers successfully, preoperative administration of IL-6 produces normal regeneration, suggesting that TNF-α affects hepatic regeneration principally through its effects on IL-6 [11]. Lastly, the metalloproteinase, ADAM 17, actively releases TNF- α from the cell surface.

Matrix metalloproteinases (MMPs) are a family of zinc-dependent endopeptidases that work in concert with their endogenous inhibitors, the tissue inhibitors of metalloproteinases to control the rate and degree of extracellular matrix degradation [12, 13]. The effects of this activity are widespread and include tissue regeneration, embryologic development, ovulation, tumor growth, and metastasis [14, 15, 16]. Extracellular matrix (ECM) remodeling by MMPs is critical for the growth of new capillaries from pre-existing vessels (angiogenesis) [15, 17, 18].

In a recent study, we investigated whether urinary MMPs or tissue inhibitors of metalloproteinases could predict the status of liver regeneration after partial hepatectomy. Rodent livers are known to regenerate to their original mass 8 days after a 67% hepatectomy [19, 20]. We showed that during hepatic regeneration in mice, there is an elevation in MMP-9 activity in the urine. At postoperative day 8, loss of MMP-9 activity was consistent with the cessation of hepatic regeneration [14].

Because MMPs appear to be a key mediator of hepatic regeneration, we more recently hypothesized that a broad-spectrum MMP inhibitor would reduce the rate of organ regeneration. Successful inhibition of hepatic regeneration through this route would support the concept that liver regeneration is dependent on MMP activity and would provide additional insight into the complex mechanisms of hepatic regeneration. To test our hypothesis, mice underwent a partial hepatectomy and were subsequently treated with Marimastat, a broad-spectrum MMP inhibitor.

Section snippets

Partial Hepatectomy

Experiments were performed on 7 to 8 wk old male C57BL/6 mice (The Jackson Laboratory, Bar Harbor, ME). The mice were housed five animals to a cage in a barrier room. Mice were acclimated to their environment for at least 72 h prior to the initiation of each experiment and allowed food and water ad libitum. Animal protocols complied with the NIH Animal Research Advisory Committee guidelines and were approved by the Children’s Hospital Institutional Animal Care and Use Committee. Mice underwent

Mouse and Liver Weights

There was no statistical difference in mouse weights between any of the groups when comparing preoperative weights and post-hepatectomy day 8 weights (data not shown). There was no statistical difference in mean liver weights in the sham control and sham Marimastat mice at post operative day 4 and day 8 (data not shown). Mean post-hepatectomy liver weights after sacrifice at day 4 were 0.84 ± 0.07 grams for the hepatectomy control group and 0.74 ± 0.03 grams for the hepatectomy Marimastat

Discussion

It has become increasingly clear that organ regeneration, like tumor growth, is dependent upon angiogenesis. The endothelial cell proliferation and apoptosis that accompany hepatic regeneration, for example, can be modulated through the administration of angiogenic agents [2]. In this report we show that matrix metalloproteinases are also a critical requirement in the normal hepatic regenerative process. This hypothesis was founded on the fact that MMPs are active in both extracellular matrix

Acknowledgments

JEV is supported by the American Society of Transplant Surgeons-Roche Laboratories Scientist Research Award and the Robert E. Wise Research Foundation. MP is supported by the CHMC Surgical Foundation and by NIH grant DK069621-01. IPJA is supported by the Dutch Cancer Society. MAM is supported by P01 CA45548 and 1 P50 DK065298.

References (31)

  • F.G. Court et al.

    The mystery of liver regeneration

    Br J Surg

    (2002)
  • A.K. Greene et al.

    Endothelial-directed hepatic regeneration after partial hepatectomy

    Ann Surg

    (2003)
  • N.L. Bucher

    Liver regeneration: An overview

    J Gastroenterol Hepatol

    (1991)
  • T. Kaido et al.

    Interleukin-6 augments hepatocyte growth factor-induced liver regeneration: Involvement of STAT3 activation

    Hepatogastroenterology

    (2004)
  • R. Taub

    Hepatoprotection via the IL-6/Stat3 pathway

    J Clin Invest

    (2003)
  • Cited by (34)

    • Pre-Operative Fasting Provides Long Term Protection Against Chronic Renal Damage Induced by Ischaemia Reperfusion Injury in Wild Type and Aneurysm Prone Fibulin-4 Mice

      2020, European Journal of Vascular and Endovascular Surgery
      Citation Excerpt :

      Yet, fasting increased MMP expression levels in Fibulin-4+/R kidneys compared with AL fed Fibulin-4+/R mice after IRI, which were comparable to those of WT fasted animals.30 Recent studies have highlighted the beneficial role of MMPs in tissue remodelling and recovery after IRI.31,32 Kaneko et al.33 demonstrated that after renal IRI, MMP-2 expression increased in the medulla on day 7–14 and the damaged tubules recovered within 14 days.

    • Fibroblast growth factor-2 signaling modulates matrix reorganization and cell cycle turnover rate in the regenerating tail of Hemidactylus flaviviridis

      2020, Acta Histochemica
      Citation Excerpt :

      The major group of proteolytic enzymes responsible for ECM degradation during processes that require ECM turnover, including regeneration, are the matrix metalloproteinases (MMPs). MMPs participate in regenerative response across a wide range of animals including hydra foot and head regeneration (Leontovich et al., 2000; Shimizu et al., 2002), zebrafish fin regeneration (Bai et al., 2005), amphibian limb regeneration (Vinarsky et al., 2005), and mouse liver regeneration (Alwayn et al., 2008). MMP-1, 2, 3, and 9 have all been implicated in corneal wound healing and ulceration (Matsubara et al., 1991; Fini et al., 1998; Ye and Azar, 1998).

    • Matrix metalloproteinases-9 deficiency impairs liver regeneration through epidermal growth factor receptor signaling in partial hepatectomy mice

      2015, Journal of Surgical Research
      Citation Excerpt :

      MMPs have multiple important biological functions in normal and pathologic processes that include promoting growth, tissue remodeling, and regeneration [14,19,23]. Previous studies have found increased MMP-9 and MMP-2 in liver tissue and urine after PH, and a broad-spectrum MMP inhibitor reduced liver regeneration [13,17,24]. Specific MMPs regulate EGFR signaling, such as MMP-10 (stromelysin-2) induction of EGFR ligands [25], and MMP-9 (gelatinase B) regulation of EGFR ligands [18,26].

    • Role of biomaterials, therapeutic molecules and cells for hepatic tissue engineering

      2012, Biotechnology Advances
      Citation Excerpt :

      However, it is essential to understand the sequence of events leading to regeneration of hepatocytes in vivo in order to mimic these factors in vitro. Hepatocytes are known to regenerate in a series of steps (initiation, expansion, proliferation and termination) through the action of growth factors and certain cytokines in vivo (Alwayn et al., 2008). In the initiation phase, cytokines (Interleukin 6 (IL-6) and Tumor Necrosis Factor-alpha (TNF-α)) are required for hepatocytes to proliferate (Alwayn et al., 2008).

    • Synchronized expressions of hepatic stellate cells and their transactivation and liver regeneration during liver injury in an animal model of cholestasis

      2011, Journal of Pediatric Surgery
      Citation Excerpt :

      Interleukin-6 also uses glycoprotein 130 (gp 130) as a receptor to induce the nuclear translocation of STAT3 as well as to activate the Ras/mitogen activated protein pathway. Glycoprotein 130/STAT3 demonstrated a protective effect against liver fibrogenesis during chronic cholestasis and controlled hepatocyte proliferation after a partial hepatectomy [5,12]. However, the underlying mechanism of liver regeneration could not be elucidated in this study.

    View all citing articles on Scopus
    1

    Ian P.J. Alwayn and Jennifer E. Verbesey contributed equally to this work.

    View full text