*3.2. Impact of Mast Cells and Mediators on Liver Disease and Individual Cells*

Once MCs have infiltrated into the interstitial tissue of the portal tracts, MCs most likely get activated by environmental factors. Through the process of hyperplasia and degranulation (Figure 2), effector molecules are liberated, which in turn act on surrounding cells, including bile duct epithelial cells, HSCs, portal fibroblasts, and resident Kupffer cells. In addition, these mediators are further triggers that recruit circulating immune cells to the site of injury. On the other hand, it has been described that the contents liberated after degranulation of MCs can act in an endocrine fashion via the bile ductules/bile showing long-range effects [122]. In addition to the "classical" degranulation, a cell-to-cell interaction has been described and termed transgranulation. This interaction occurs with MCs and fibroblasts/vascular endothelial cells and describes the direct transfer of granules from MCs via pseudopodia to the communicating cell. This process has been observed in vitro as well as in vivo [123].

**Figure 2.** Regulation of mast cell (MC) effector functions. Upon liver damage, MCs increase in the liver tissue and degranulation is triggered by environmental factors. The released MC effectors modulate target cell functions in receptive cells, such as hepatic stellate cells, portal myofibroblasts, and cholangiocytes. There are several possible ways to interfere with MC effector functions: (1) MC stabilizers like cromolyn block the general release of MC granula, and inhibitors can inactivate liberated (2) proteases, (3) block different types of histamine receptors, or (4) can modulate downstream signaling in target cells. Abbreviations used are: APC366, *N*-(1-hydroxy-2-naphthoyl)-L-arginyl-L-prolinamide; G-Prot., G protein; HR, histamine receptor; H1R, histamine receptor H1; H2R, histamine receptor H2; PAR-2, protease-activator receptor-2; PZ-235, P2pal-18S cell-penetrating pepducin targeting the intracellular i3 loop of PAR-2; TY-51469, 2-[4-[(5-fluoro-3-methyl-1-benzothiophen-2-yl)sulfonylamino]-3-methylsulfonylphenyl]-1,3-thiazole-4-carboxylic acid.

In kinetic experiments using CCl4 intoxication, MC infiltration was associated with individual stages of fibrosis. At the time points with significant MC infiltration, HSCs were not yet activated. Therefore, the authors concluded that MCs are indicators of acute inflammation [124]. Work contributed by Takeshita [125] implied that after BDL surgery in rats, MC amounts increase in the liver but not that much in the portal fields. In contrast, there is a strong but transient increase in MC number after recanalization of the bile duct [125]. The authors concluded that MCs are not involved in the early phase of fibrosis, but rather in the apoptosis of biliary epithelial cells and removal of bile ductules after recanalization [125]. However, further data from the time-resolved analysis of BDL or DMN-treated rats imply that MCs do play a role in the fibrotic phase of liver disease [95,126]. Results from CCl4-treated rats revealed that myofibroblasts and macrophages increase in number during the first ten weeks, whereas MCs increase constantly over the period peeking in number at week 14 [127]. Nevertheless, although the function of MCs in the process of fibrosis is not completely resolved, it is common sense that most of the studies performed so far showed that MCs fulfill profibrogenic functions in diseased liver (see below).

Only one of the early works performed by Sugihara et al. implied that MCs have no consequence in the development of liver fibrosis in wild-type or MC-deprived *KitWs*/*Ws* rats and in wild-type or MC-deprived *KitW*/*Wv* mice [128]. The animals were subjected to BDL and treated with CCl4 or porcine serum (rats) and CCl4 or BDL (mice) to induce portal tract (BDL, serum) or parenchymal fibrosis (CCl4). The treatments lasted 21 days and were evaluated using a hydroxyproline measurement for collagen deposition and Alcian blue stain to detect MCs. Since the early time points (less than 21 days) were not taken into account, the reaction of the biliary tree and infiltration of MCs during the early events of fibrosis were not evaluated. All treatments caused fibrosis and increased quantities of MCs. However, although the overall MC quantities were approximately 20-fold higher in wild-type than in *KitWs*/*Ws* animals, the degree of fibrosis was equal in both groups. Therefore, the authors concluded that MCs have no role in fibrosis [128]. On the other hand, a recent paper analyzing the influence of MCs on fibrosis in a direct co-culture system or an indirect co-culture system composed of a human MC line and

primary MCs with the human HSC line LX2 showed that soluble factor(s) derived from MCs caused a decrease in ECM/collagen I abundance in the presence or absence of TGF-β, IFN-α, or IL-10 [129]. They could show by using the APC366 tryptase inhibitor and the broad range inhibitor Chymostatin, that the key MC proteases tryptase and chymase are most likely involved in collagen I degradation. In addition, the human leukocyte antigen G (HLA-G), representing a histocompatibility antigen, is increasingly expressed in MCs when co-cultured with HSCs. This is accompanied by a significant decrease in collagen production, irrespectively of whether cells were stimulated with TGF-β, IL-10, or IFN-α, or left untreated [129]. In their study, the authors proposed MC proteases as critical mediators of collagen degradation. HLA-G has been studied by the same authors before in the setting of chronic HCV infection. They could show that HLA-G in situ is produced by MCs and is up-regulated upon stimulation with IL-10 and IFN-α, leading to the recruitment of T-lymphocytes and NK cells [130].

Nevertheless, taking into account data of MC involvement in fibrosis in lung, kidney and heart, there is a great body of evidence that MCs play a pro-fibrogenic role in liver fibrosis [131–135].
