1E,F). These data demonstrate that NOX plays a crucial role in the pathogenesis of liver fibrosis in two complementary models of liver injury. To investigate the expression of NOX in different cell populations GSI-IX chemical structure of the liver, we performed RT-PCR analysis for the mRNA levels of the main components of NOX activity in different cells of the liver: fibrocytes, KCs, HEPs, endothelial cells, and HSCs. The data showed that the p47phox component is mainly expressed in KCs, among the BM-derived cells, and in HSCs,

among the non–BM-derived cells (Fig. 2A). Surprisingly, HEP expressed p47phox. Furthermore, the data showed a lower expression of this component in other cell types of the liver (Fig. 2A). Concordantly with the p47phox mRNA expression in the different hepatic cell types, we performed double immunofluorescence staining between 4-HNE (a lipid peroxidation product as a general marker of ROS) and cell-type markers F4/80, αSMA, and pan-cytokeratin for KCs, HSCs, and parenchymal cells, respectively. The lipid peroxidation product 4-HNE was expressed in KCs and HSCs in the livers of BDL mice (Fig. 2B,C). Some HEPs also express ROS in BDL mice (Fig. 2D). We further confirmed that these cell types express ROS in the livers of CCl4-treated selleck inhibitor mice (Supporting Fig. 1). To investigate the role of NOX from different hepatic cell types in liver fibrosis, BMT was employed

to generate chimeric mice. BM from p47phox-sufficient mice was transplanted into p47phox-deficient mice and vice versa, obtaining mice with NOX-sufficient BM-derived cells, including KCs and NOX-deficient endogenous liver cells (non–BM-derived), including HSCs, and vice versa. As controls, BM from p47phox-deficient mice was transplanted into p47phox-deficient mice and BM from p47phox-sufficient mice was transplanted into p47phox-sufficient mice.

This strategy produced four different groups of mice: (1) WT BM p47phox KO mice, triclocarban (2) WT BM WT mice, (3) p47phox KO BM WT mice, and (4) p47phox KO BM p47phox KO mice. We have previously shown that this strategy is useful in discriminating signaling pathways between HSCs and KCs.25 In summary, this resulted in mice with HSCs deficient for p47phox but with normal KCs (group 1), mice with intact NOX in all cells (group 2), mice with NOX-deficient KCs but WT HSCs (group 3), and mice deficient for NOX in all cells (group 4). These mice were then subjected to BDL for 3 weeks. As expected from our above results, mice with intact NOX (group 2) showed the most fibrosis and mice deficient for p47phox in all cells (KO KO, group 4) showed the least fibrosis (Fig. 3A-C). Deficiency for p47phox in either BM-derived or non–BM-derived cells resulted in attenuated liver fibrosis following 3 weeks of BDL as evaluated by collagen deposition and hydroxyproline content compared to mice with functional p47phox in all cells (WT BM WT mice, group 2) (Fig. 3A-C).