Discussion
Various studies have demonstrated that PPARγ agonists have potential in the treatment of CLD. Immune modulation by PPARγ ligands may be of therapeutic benefit in reducing biliary inflammation in PBC[38]. In addition, PPARγ inhibits the transcriptional activation of inflammatory response genes and represses cellular toll-like receptor signaling in inflammatory cells as well as in cholangiocytes[39]. Importantly, recent studies have shown that rosiglitazone improve intrahepatic cholestasis and cholestasis-associated dyslipidemia induced by ANIT[29]. These data indicated that PPARγ activation may be an effective strategy for the treatment of CLD, especially for the improvement of liver fibrosis and inflammation, thereby limiting disease progression. Despite this, troglitazone, a PPARγ ligand, was withdrawn from the market due to hepatotoxicity and no experimental or clinical data on other glitazones are available[40]. Although this evidence suggests the feasibility of PPARγ agonists as a therapeutic strategy for CLD, new agonists still need to be developed.
In this study, we demonstrated that TEC, a partial PPARγ agonist[31], can alleviate cholestasis in an experimental mouse model (Figure 1 and 2) without obvious side effects. Importantly, we confirmed that TEC alleviation of ANIT-induced liver injury was dependent on PPARγ to reduce the recruitment and activation of macrophages and enhance bile transporter expression in hepatocytes (Figure 7). Taken together, we found that TEC may be a potential therapeutic strategy for the treatment of CLD as a PPARγ agonist.
It was shown in vitro that TEC inhibited LPS-induced macrophage activation as well as LPS-induced hepatocyte dysfunction via the PPARγ/NFκb pathway (Figure 4 and 5). In accordance with our findings, multiple studies have suggested that activation of NFκb plays an important role in the development of CLD[20, 41]. Genetic or pharmacological inhibition of NFκb prevented cholestasis-induced liver damage in various experimental mice[42, 43]. However, inhibition of NFκb was found to lead to an increase in hepatocyte apoptosis after bile duct ligation (BDL)[44]. Additionally, IKK1 and IKK2 are IκB kinases which are important for NFκb activation and genetic ablation of IκB kinases could lead to inflammatory damage to portal bile ducts[45]. Therefore, appropriate targets or the identification of drugs that either exert only a moderate effect on NFκb activity or that can be specifically delivered to nonparenchymal cells are essential to avoid the increase in liver injury associated with complete NFκb blockade in hepatocytes[46].
Interestingly, PPARγ is highly expressed in macrophages where it has an important role in immune modulation, while PPARγ expression is relatively low in hepatocytes under normal physiological conditions[47, 48]. Our results showed that TEC treatment almost blocked the phosphorylation of NFκb-p65 induced by LPS in KCs (Figure S1C) and BMDMs (Figure S2D). The phosphorylation of NFκb-p65 was still 1.78-fold higher than that in control hepatocytes after TEC treatment in the presence of LPS (Figure 5A). An in vivo study also supported these findings, where TEC decreased rather than increased the apoptosis of hepatocytes in ANIT-induced model mice, as shown by the down-regulation of the positive area of TUNEL staining and reduced caspase-3 activity (Figure 1C and D). Our results suggest that TEC had a much stronger inhibitory effect on NFκb in macrophages compared with hepatocytes which could be attributed to the different levels of PPARγ expression in KCs and hepatocytes. Additionally, NFκb inhibitor (BAY 11-7085) treatment at the same dose as TEC markedly decreased the phosphorylation of NFκb-p65 induced by LPS in hepatocytes, to an even lower level than that in the control group (Figure S3A). Collectively, this dose of TEC had a stronger inhibitory effect on NFκb activation in macrophages compared with hepatocytes, and avoided liver injury induced by complete NFκb blockade in hepatocytes.
Bsep, encoded by the gene ABCB11, is a member of the adenosine triphosphate (ATP)-binding cassette (ABC) transporters. It is mainly expressed on hepatocyte canalicular membranes and is basically responsible for the secretion of bile acids, and it deficiency may result in progressive familial intrahepatic cholestasis type 2[49]. De novo or retargeted canalicular expression of Bsep has been confirmed to play an important role in bile acid canalicular export in the treatment of cholestasis[50, 51]. Previous studies have shown that troglitazone can induce intrahepatic cholestasis by increasing serum bile salt concentrations and inhibiting Bsep expression in rat liver[40, 52]. In contrast, another study showed that troglitazone, but not rosiglitazone or pioglitazone, regulated the expression of the FXR target gene Bsep[37]. In summary, this evidence could support further investigation of the relationship between Bsep and TEC. According to our findings, TEC promoted the binding of PPARγ and Bsep promoter regions and promoted their expression (Figure 5). As TEC directly increased the expression of Bsep, this may be another molecular mechanism of TEC in the treatment of CLD.
Although we found that TEC (50 mpk) significantly alleviated liver injury in ANIT and DDC-induced CLD without significant side effects, additional cholestatic models and different doses are still needed to verify the efficacy and toxicity of TEC. In conclusion, we have demonstrated that TEC exert liver protection in a PPARγ-dependent manner, which in turn inhibit macrophage activation and hepatocyte dysfunction through restrain NFκb activation as well as enhance Bsep expression, thus alleviated intrahepatic cholestasis.