This study identified GW501516-activated PPARβ/δ as a promoter of uncontrolled liver repair, which results in fibrosis, most likely via p38- and JNK-dependent stimulation of HSC proliferation. This healing function of PPARβ/δ is reminiscent of its role in skin wound healing . Fibrosis is a response to a variety of chronic damaging stimuli. It can cause an alteration in liver structure that may lead to excessive deposition of ECM, apoptosis of hepatocytes and inflammation [2, 3]. During the fibrogenic response, activated HSCs proliferate and indeed produce an excess of ECM and pro-inflammatory proteins.
To date, the role of PPARβ/δ has been unclear in this hepatic repair process, which often degenerates to liver disease. In the present study, mice were exposed to both CCl4 and a PPARβ/δ selective ligand for a long period of time (i.e., 6 weeks). We found that agonist-activated PPARβ/δ had an additive or synergistic effect with CCl4 on the production of inflammatory cytokines, pro-fibrotic ECM proteins and HSC markers, and on the accumulation of hepatic triglycerides. In line with our results, a recent study also demonstrated a profibrotic effect of the PPARβ/δ ligand GW501516 after short-term CCl4 administration in mice . In contrast to our data, this study did not identify the molecular mechanisms by which GW501516-activated PPARβ/δ induced the fibrotic process. Furthermore, this profibrotic action of activated PPARβ/δ was consistent with our previous study in a rat model of acute CCl4-induced liver injury treated with a different PPARβ/δ ligand (L165041). In this model, we found increased expression of Col1Î±1, Î±-SMA, and lysyl oxidase with CCl4/L165041 treatment . In contrast, other studies concluded that GW0742- or KD3010-activated PPARβ/δ attenuated CCl4-induced hepatotoxicity [13, 15]. Our present findings suggest that CCl4 treatment alone causes only a weak activation of PPARβ/δ. For example, we showed that CCl4 did not result in important differences in the expression of several genes when wild type and PPARβ/δ-null mice were compared, but GW501516/CCl4 co-treatment strongly induced these genes only in wild type mice. Importantly, we found that several genes that were strongly stimulated by the combined action of CCl4 and GW501516 were also expressed at higher levels in hepatic tissue of patients with confirmed alcohol-induced liver fibrosis/cirrhosis. The discrepancies among different studies may be due to differences in the ligands used, the dose applied, and duration of administration. For instance, different ligands may present different pharmacophore features resulting in different physiological outcomes. In future studies, it will be interesting to use cell type-specific deletion of PPARβ/δ in vivo to evaluate the individual contribution of stellate cells, macrophages/Kupffer cells and hepatocytes to the observed PPARβ/δ-dependent profibrotic or protective effects.
The underlying mechanism of the PPAR-dependent stimulation of HSC proliferation in vivo was unveiled in the human LX-2 HSC line. These cells express key genes involved in hepatic fibrosis . Addition of the GW501516 ligand activated PPARβ/δ in these cells and increased proliferation after 48 h, whereas no change in proliferation was observed in the PPARβ/δ KD cells. Similar to the in vivo results, the PPARβ/δ ligand also increased the expression of pro-inflammatory and profibrotic factors. These results were consistent with our previous study, which showed that L165041-induced activation of PPARβ/δ in cultured activated primary HSCs enhanced proliferation and profibrotic factor expression .
In the present study, we also investigated genes that were not direct targets of PPARβ/δ, but rather reflected the activation of PPARβ/δ-dependent signaling pathways. We found that PPARβ/δ regulated the PI3K, p38 MAPK, and SAPK/JNK pathway, which is known to be involved in cell proliferation. We also found that Erk1/2 MAPK and nuclear factor-κB (NF-κB) signaling did not contribute to PPARβ/δ-induced HSC proliferation (data not shown). In fact, it was previously shown that MAPKs p38 and JNK were positive regulators of HSC proliferation [1, 24, 25]. Those studies showed that multiple stress stimuli increased the activity of SAPK/JNK and p38 MAPKs, which in turn activated several transcription factors implicated in cell proliferation and differentiation . The present study revealed the novel finding that, during fibrosis, these factors were regulated by GW501516-activated PPARβ/δ. Thus, it was of interest to unveil how PPARβ/δ controlled this paramount signaling pathway.
Our results showed that GW501516-activated PPARβ/δ enhanced phosphorylation of p38 and SAPK/JNK MAPKs without changing their expression levels. This suggested that PPARβ/δ was involved in the transcriptional regulation of upstream kinases. In fact, PPARβ/δ-dependent phosphorylation of p38 and JNK was suppressed by inhibitors of PI3K (LY294002) and PKC (Gö6983). In addition, we observed a PPARβ/δ-dependent phosphorylation of Akt at Ser473. Consistent with this finding, ligand-activated PPARβ/δ in skin increased keratinocyte survival upon exposure to stress through PI3K signaling; this was reflected by increased Akt1 activity . Interestingly, PKCs are downstream targets of activated PI3K. It was previously demonstrated that acetaldehyde induced PKC activation, which then increased HSC proliferation and activation [28–31] and collagen production [32, 33]. Thus, we hypothesized that, in HSCs, PPARβ/δ might also upregulate PDPK1 (Pdpk1) and downregulate phosphatase and tensin homolog (Pten) expression. This would activate, via PI3K and PKC, the Ser/Thr protein kinase MLK3, a cytokine-activated MAP3K known to regulate JNK, p38, and Erk1/2 [21, 34]. We showed for the first time that GW501516 increased MLK3 protein expression and phosphorylation in a PPARβ/δ-dependent manner; furthermore, PKC inhibitors blocked MLK3 activation. Thus, GW501516 indirectly activated MLK3, a downstream target of PKC. It was previously established that MLK3 phosphorylates and activates the MAP2K isoforms MKK4/7 and MKK3/6, which then activate JNK and p38, respectively [19, 20, 34, 35]. Interestingly, because GW501516 increased both the phosphorylation and expression levels of MLK3 (Figure 7C), MLK3 may be both a direct and indirect target of activated PPARβ/δ. Recent studies demonstrated that RNAi-mediated knockdown of MLK3 inhibited serum-stimulated cell proliferation, tumor cell proliferation, and growth factor/cytokine-induced JNK, p38, and Erk1/2 activation [21, 22]. These cells also exhibited destabilized B-Raf/Raf1 complexes . Furthermore, CEP-1347, the small-molecule inhibitor of all MLK members, caused reductions in pulmonary fibrosis , pancreatitis , and neurodegeneration  by inhibiting JNK activation.
In conclusion, this report is the first to show that GW501516-activated PPARβ/δ could enhance both the p38 and JNK MAPKs signaling pathways, and thus, increase HSC proliferation in liver injuries. Furthermore, we showed that PPARβ/δ activated p38 and JNK by phosphorylating PI3K/PKC/MLK3 components (Figure 7E). We propose that activated PPARβ/δ increased HSC proliferation, which then exacerbated inflammatory and fibrotic processes during liver injuries. Taken together, these findings showed that GW501516-activated PPARβ/δ represents an important regulatory step in HSC proliferation. Finally, the role of PPARβ/δ and its activation in HSC proliferation in liver fibrosis should be considered when evaluating PPARβ/δ agonists as potential therapeutic agents for broad applications; for example, a phase II clinical trial is currently testing GW501516 as a treatment for dyslipidemia. Furthermore, it will be important, in the future, to evaluate whether natural ligands can achieve effects similar to those of GW501516.