Advertisement
Mac the knife? Macrophages– the double-edged sword of hepatic fibrosis
Scott L. Friedman
Division of Liver Diseases, Mount Sinai School of Medicine, New York, New York, USA.
Address correspondence to: Scott L. Friedman, Box 1123, Mount Sinai School of Medicine, 1425 Madison Avenue, Room 1170C, New York, New York 10029, USA. Phone: (212) 659-9501; Fax: (212) 849-2574; E-mail: Scott.Friedman@mssm.edu.
Find articles by Friedman, S. in: JCI | PubMed | Google Scholar
Published January 3, 2005 - More info
J Clin Invest. 2005;115(1):29–32. https://doi.org/10.1172/JCI23928.
© 2005 The American Society for Clinical Investigation
Kupffer cells, the resident macrophage population of the liver, have in recent years lost their leading position in the pecking order of cell types known to contribute to hepatic injury and repair. Long recognized for their activity in liver inflammation, they had been increasingly overlooked while a stellate cell–centric view of hepatic fibrosis had replaced the earlier focus on macrophages (Figure 1). Symbolic of this demotion, a biannual meeting initially convened as the International Kupffer Cell Symposium changed its name in 1990 to the International Symposium on Cells of the Hepatic Sinusoid (1). This evolution has been understandable because recent studies have established the activated hepatic stellate cell and its myofibroblast counterpart as the major sources of ECM in both experimental and human liver disease (2). As a result, a comprehensive picture of hepatic stellate cell activation in liver injury has emerged, resulting in exciting new prospects for targeting a range of growth factors, receptors, and intracellular mediators in the treatment of hepatic fibrosis (3).
Figure 1Cellular events in hepatic fibrosis. As fibrosis develops in response to liver injury, stellate cell activation leads to accumulation of scar matrix. This in turn contributes to the loss of hepatocyte microvilli and endothelial fenestrae, which results in the deterioration of hepatic function. Kupffer cell (macrophage) activation contributes to the activation of stellate cells. Figure modified with permission from The Journal of Biological Chemistry (24).
The paper by Duffield and colleagues in this issue of the JCI (4), however, firmly reestablishes the hepatic macrophage as a central determinant of the liver’s response to injury and repair. The study is built upon 2 important concepts. First, macrophages have more than one pathway of activation and response, depending on the specific stimulus and biologic context; indeed, divergence of macrophage responses has been recognized for many years (5, 6) but has only recently been placed in a biologically coherent context that defines pro- and anti-inflammatory phenotypes. Proinflammatory actions of macrophages include antigen presentation, T cell activation, and cytokine and protease release, among others (7). However, anti-inflammatory actions have been increasingly appreciated, particularly those induced by IL-4, including induction of immune tolerance (8), innate immunity (9), and T cell differentiation (7, 10).
A second, more recently developed concept addresses the behavior of hepatic stellate cells in the resolution of liver injury and fibrosis. Whereas initial interest in stellate cell biology focused on the cells’ activation and fibrogenic properties during progressive liver injury, more recent efforts have explored both their fate as liver injury resolves and the mechanisms underlying persistence of fibrosis in sustained liver injury. This is a vital area of inquiry since clarification of mechanisms by which the liver naturally restores its architecture might be exploited in developing antifibrotic therapies for patients with chronic liver disease. During regression of experimental liver fibrosis, activated stellate cells undergo programmed cell death associated with loss of tissue inhibitor of metalloproteinase–1 (TIMP-1) expression (11). Because TIMP-1 not only enhances stellate cell survival but also antagonizes matrix degradation, the loss of TIMP-1–expressing stellate cells is thought to unleash latent matrix-degrading activity, leading to the breakdown of scar matrix and the reconstitution of normal hepatic architecture. When liver injury persists, TIMP-1 levels remain high, and progressive cross-linking of collagen may render the accumulating matrix relatively insoluble to proteases (12, 13). This concept is reinforced by studies demonstrating attenuation of fibrosis when TIMP-1 is inhibited (14). Significant gaps in this paradigm have persisted, however, including the source(s) and identity of salutary proteases in fibrosis resolution and the role of other cellular elements, including not only hepatic macrophages, but also sinusoidal endothelium.
-
Divergent phenotypes of hepatic macrophages – one cell type or two?
These results are fascinating, and they raise some important but unanswered questions regarding the underlying mechanisms. Are the same macrophages responsible for promoting matrix accumulation during progression and preventing it during recovery through phenotype switching, or are different subsets of macrophages recruited during this brief interval? While the former seems more likely and is consistent with evolving concepts (7, 15), an alternative explanation also supported by this (4) and much earlier studies (16) is the recruitment and differentiation of bone marrow–derived monocytes into hepatic macrophages. In fact, it is uncertain whether all these recruited monocytes become classic Kupffer cells or remain phenotypically distinguishable as a novel subpopulation. Regardless, if phenotypic conversion of the same cells accounts for this divergent behavior in progression versus recovery, what are the signals driving this switch? One approach to clarifying these questions would be to isolate CD11b-DTR–expressing cells by flow cytometry at different intervals during progression and recovery (without diptheria toxin administration) to analyze their transcriptional profile by microarray and to perform functional studies. This approach could be used in mice chimeric for bone marrow cells so that the phenotype of CD11b-DTR–expressing cells from liver could be distinguished from those derived from bone marrow.
How do the divergent macrophage phenotypes present during fibrosis progression versus recovery affect matrix production and degradation? In particular, are the effects of macrophage depletion mediated largely through their impact on activated stellate cells and/or myofibroblasts? During fibrosis progression, it is believed that macrophages are likely to promote activation of stellate cells through release of paracrine factors, including TGF-β1, since this phenomenon has been demonstrated in culture (17, 18) (Figure 2). In contrast, the simultaneous loss of macrophages (termed scar-associated macrophages by Duffield et al.; ref. 4) and activated stellate cells during recovery suggests that macrophages could also provoke apoptosis of hepatic stellate cells by the expression of TNF-related apoptosis-inducing ligand (known as TRAIL) and other apoptotic stimuli, as recently reported (19). Thus, the biphasic nature of macrophages could yield divergent effects on stellate cells, promoting their activation during progression and apoptosis during recovery. One potential regulator of this duality could be the cytokine TNF-α or its downstream signaling mediator NF-κB, since delicate temporal regulation of NF-κB signaling in hepatocytes can elicit profound and divergent effects on apoptosis and growth (20).
Figure 2The double-edged sword of hepatic macrophage activity in hepatic fibrosis progression versus recovery. The findings by Duffield et al. (4) in this issue of the JCI suggest that hepatic macrophages may induce divergent effects on liver fibrosis by promoting stellate cell activation in the face of continued injury and fibrosis; and stellate cell apoptosis during recovery associated with fibrosis regression as injury subsides. Evidence from other studies implicates TGF-β1 as one potential paracrine stimulator of stellate cell activation by macrophages, while TRAIL may mediate stellate cell apoptosis during fibrosis regression associated with recovery. Apoptosis associated with the loss of TIMP-1 may unmask latent matrix protease activity released by either macrophages, stellate cells, or other cell types. It is not certain if the same macrophages account for the divergent activities of this cell type or whether different macrophage subsets mediate these opposing pathways.
Copyright © 2025 American Society for Clinical Investigation
ISSN: 0021-9738 (print), 1558-8238 (online)