Involvement of retinol metabolism in regulation of hepatic maturation and injury

Abstract

Hepatic stellate cells (HSCs) store about 80% retinol of our body as lipid droplets. The cells were involved in retinol homeostasis by controlling retinol storing. Especially in hepatic fibrogenesis, HSCs were activated and transformed to myofibroblast-like cells as well as lose their retinols. In there, the cells were interacted with various immune cells by metabolizing and secreting retinols, and controlled liver fibrosis. In injured liver, the remaining hepatocytes were proliferated to compensate injuries. Although morphology of the cell is similar, metabolically functional maturation process is needed. It is well known that hepatic stellate cells were involved to cellular differentiation. Hepatocyte-like cells (HLCs), derived from human embryonic stem cells and resembled with developing fetal hepatocytes must attained functional maturity to use of drug screening. In this study, we evaluate the effects of mouse HSCs (mHSCs) on the expression and activity of drug metabolic enzymes in HLCs. In microarray analyses, expressions of major drug metabolic enzymes in HLCs were specifically down-regulated compared with primary human hepatocyte. However, metabolic enzyme expressions were significantly increased in co-cultured HLCs with retinol-storing mHSCs. Chenodeoxycholic acids (CDCAs) were secreting from HLCs and it activates farnesoid X receptor (FXR) of mHSCs. As a result, retinol storage gene expression in co-cultured mHSCs was down-regulated, while retinol production and delivery gene expressions were significantly increased by effect of FXR activation. In contrast, retinoic acid deficient LX-2, hTERT and Raldh1-depleted HSCs did not affect expression of metabolic enzymes in HLCs, suggesting that retinoic acid might be a crucial factor for functional maturation of HLCs. In western blotting, protein levels of CYP1A2, CYP2E1, and CYP3A4 were increased in co-cultured HLCs. In consistent with protein levels, metabolic activities of CYP1A2 and CYP3A4 were significantly improved in co-cultured or retinoic acid-treated HLCs. Therefore, co-culturing with mHSCs or treatments of retinoic acids could be useful ways to adapt immature HLCs for the application of drug toxicity screening system. However, the most important thing is that CDCA secreted from HLCs trigger the activation of retinol metabolic signaling in mHSCs. In addition, accumulating evidence suggests that retinol and its metabolites are closely associated with liver fibrogenesis. Recently, we demonstrated that genetic ablation of alcohol dehydrogenase 3 (ADH3), a retinol metabolizing gene that is expressed in hepatic stellate cells (HSCs) and natural killer (NK) cells, attenuated liver fibrosis in mice. In the current study, we investigated whether pharmacological ablation of ADH3 has therapeutic effects on experimentally induced liver fibrosis in mice. Liver fibrosis was induced by intraperitoneal injections of carbon tetrachloride ($CCl_4$) or bile duct ligation (BDL) for two weeks. To inhibit ADH3-mediated retinol metabolism, $10 \mu g 4$-methylpyrazole (4-MP)/g of body weight was administered to mice treated with $CCl_4$ or subjected to BDL. HSCs and NK cells were isolated from control and treated mice livers for molecular and immunological studies. Treatment with 4-MP attenuated $CCl_{4-}$ and BDL-induced liver fibrosis in mice, without any adverse effects. HSCs from 4-MP treated mice depicted decreased levels of retinoic acids and increased retinol content than HSCs from control mice. In addition, the expression of $\alpha$-SMA, transforming growth factor-$\beta 1$ (TGF-$\beta 1$), and type I collagen $\alpha 1$ was significantly reduced in the HSCs of 4-MP treated mice compared to the HSCs from control mice. Furthermore, inhibition of retinol metabolism by 4-MP increased interferon-$\gamma$ production in NK cells, resulting in increased apoptosis of activated HSCs. Based on our data, we conclude that inhibition of retinol metabolism by 4-MP ameliorates liver fibrosis in mice through activation of NK cells and suppression of HSCs. Therefore, retinol and its metabolizing enzyme, ADH3, might be potential targets for therapeutic intervention of liver fibrosis. The HSCs were taken major roles controlling in hepatic fibrosis by regulating retinol metabolism and by controlling retinol metabolizing enzyme expression, it is possible that hepatic fibrosis is effectively controlled. In addition, the hepatic stellate cells were involved in the maturation of developing hepatocytes by controlling FXR-mediated retinoic acid signaling and it is applied to hepatocyte-like cell for using as drug screening tools.