3.1 TLR5 is up-regulated in the liver after PHx
We used 2/3 PHx as an established model of liver regeneration to determine the effect of TLR5 [27]. Significant increases of bacterial flagellin in the serum and liver were observed at 6 hours post PHx (Fig. 1a). We next investigated the expression pattern of TLR5 in the liver of mice after PHx. Global transcriptome analysis of the mouse liver at various time points following PHx revealed a significant up‐regulation of TLR5 along with other TLRs (GSE95135) [29] (Fig. 1b). Consistently, real-time PCR analysis showed that the expression of TLR5 in the liver was significantly increased following PHx, with two peaks occurring at 1 hour and 12 hours after PHx (Fig. 1c). By flow cytometry, we found that TLR5 was expressed on hepatic neutrophils, KCs, and recruited macrophages (Fig. 1d-e), which is line with the previous results of our own and others [8,23]. The mean fluorescence intensity of TLR5 on these cells was unchanged, but the numbers of neutrophils and recruited macrophages were significantly increased following PHx, which might account for the up-regulation of hepatic TLR5 expression in the early phase upon PHx (Fig. 1f).
3.2 TLR5 deficiency decreases hepatocyte proliferation after PHx
To assess the effect of TLR5 on liver regeneration, we performed 2/3 PHx in Tlr5−/− mice and their WT littermates. The hepatocyte proliferation, as assessed by the incorporation of BrdU, was remarkably decreased in Tlr5−/− mice at 36 and 48 hours after PHx (Fig. 2a-b). Consistent with this result, PCNA staining showed a reduced number of proliferating hepatocytes in Tlr5−/− mice at 36 hours after PHx (Fig. 2c-d). The serum ALT and AST levels were rapidly elevated in mice after PHx, but no significant difference was observed between WT and Tlr5−/− mice(Fig. 2e), indicating a similar degree of liver injury in the two genotypes. Together, these data suggest that deficiency of TLR5 significantly decreases hepatocyte proliferation in the first 48 hours after PHx.
3.3 TLR5 deficiency suppresses hepatocyte priming in PHx-induced liver regeneration
As flagellin concentration and TLR5 expression in the liver increased at early time points after PHx, we examined the effects of TLR5 on the priming of liver regeneration following PHx. Quantitative PCR was used to examine gene expressions involved in the initial stage of liver regeneration. The results showed that PHx increased the hepatic mRNA levels of c-Myc, c-Jun, and c-Fos in both Tlr5−/− and WT mice at 30 to 60 min following PHx, but the increase was significantly blunted in Tlr5−/− mice (Fig. 3a), suggesting that loss of TLR5 suppresses PHx-induced immediate early gene expression.
We further investigated the effect of TLR5 signaling on proinflammatory cytokines expression upon PHx. As shown in Fig. 3b, serum levels of TNF-α, IL-6, TGF-α, and HGF in Tlr5−/− mice were similar to those in WT mice before PHx. PHx rapidly increased the serum levels of these cytokines in both WT and Tlr5−/− mice; however, these effects were greater in WT mice at 6 and 12 hours after PHx (Fig. 3b). Consistently, TNF-α, IL-6, TGF-α, and HGF mRNA levels in the liver were inhibited in Tlr5−/− mice at 1 and 3 hours after PHx (Fig. 3c). These results indicate that TLR5 signaling contributes to cytokines production induced by PHx.
Activation of NF-κB and STAT3 are well known as one of the major priming events during liver regeneration [3,14]. NF-κB signaling is activated in mouse liver by PHx as indicated by the phosphorylation and degradation of IκBα, and the p65 phosphorylation. Compared with WT mice, PHx-induced NF-κB activation in the liver was inhibited in Tlr5−/− mice at 30 and 60 min after PHx (Fig. 3d). Both WT and Tlr5−/− mice displayed increased hepatic STAT3 phosphorylation levels at 30 and 60 min following PHx, while Tlr5−/− mice showed lower levels of phosphorylated STAT3 than WT mice (Fig. 3d). Our results indicate that TLR5 contributes to the regulation of hepatocyte priming in liver regeneration after PHx.
3.4 The TLR5 agonist CBLB502 enhances hepatocyte proliferation in mice following PHx
To further assess the role of the TLR5 pathway in liver regeneration, we investigated whether administration of CBLB502, a TLR5 agonist derived from Salmonella flagellin, affects liver regeneration in mice after PHx. The liver/body weight ratio of CBLB502-pretreated mice was significantly higher than that of control mice during the first 72 hours post-PHx, suggesting that activation of the TLR5 pathway is involved in the early recovery of liver mass after PHx (Fig. 4a). Accordingly, PHx-induced liver damage was reduced by CBLB502 administration at 24 and 72 hours after PHx as revealed by decreased serum transaminases (Fig. 4b). Hepatocyte proliferation was markedly enhanced in CBLB502 treated mice at 36, 48, and 72 hours after hepatectomy by BrdU staining and 36 and 48 hours by PCNA staining (Fig. 4c-f). Taken together, activation of TLR5 signaling by CBLB502 enhances hepatocyte proliferation in mice following PHx.
3.5 Enhanced inflammatory response in CBLB502-pretreated mice after PHx
In line with previous reports [23], the administration of CBLB502 in mice rapidly increased serum TNF-α, IL-6, and G-CSF (Fig. 5a). Interestingly, TGF-α and HGF, which are critical for liver regeneration [30,31], were also significantly up-regulated in the serum at 3 and 6 hours after CBLB502 injection (Fig. 5b). We next performed PHx 1 hour after CBLB502 administration. Much higher levels of serum TNF-α, IL-6, G-CSF, TGF-α, and HGF were observed in CBLB502-pretreated mice than in control mice right before PHx (Fig. 5c-d). After PHx, the serum concentrations of these growth factors were significantly increased in control mice, but the levels were still much lower than those in CBLB502-pretreated mice at 6 hours after PHx (Fig. 5c-d). Moreover, the administration of CBLB502 significantly increased hepatic mRNA levels of c-Fos, c-Myc, c-Jun, TNF-α, and IL-6 at 1 hour post-PHx (Fig. 5e), which is in agreement with the result that TLR5 deficiency suppresses PHx-induced immediate early gene expression. We next examined whether CBLB502 affects the hepatic recruitment of immune cells after PHx. As shown in Fig. 5f, the number of hepatic mononuclear cells (MNCs), neutrophils, and recruited macrophages was significantly higher in CBLB502 treated mice than that in control mice before PHx, which was in line with previous reports that CBLB502 treatment induces recruitment of various types of immune cells into the liver [22,23]. PHx also induced a significant increase in the total number of hepatic MNCs, as well as the number of recruited macrophages and neutrophils, and mice pretreated with CBLB502 showed further increased number of these cells compared to mice pretreated with PBS. However, the number of KCs was not affected (Fig. 5f). Taken together, these data indicate that CBLB502 pretreatment increases the hepatic inflammatory response in mice after PHx.
3.6 TLR5 signaling contributes to hepatic lipid accumulation induced by PHx
The liver transiently accumulates lipids during liver regeneration, which is associated with hepatocyte proliferation [32,33,34]. Histological analysis and Oil Red O staining showed an obvious decrease of hepatic lipid accumulation in Tlr5−/− mice at 24 hours post-PHx (Fig. 6a). The levels of triglyceride, free fatty acids, and cholesterol in liver homogenates were comparable between WT and Tlr5−/− mice under normal conditions (Fig. 6b). Significant increases of triglyceride and free fatty acids levels in liver homogenates were observed in both WT and Tlr5−/− mice at 24 hours after PHx, but the levels of triglyceride and free fatty acids in Tlr5−/− mice were remarkably lower than those in WT mice (Fig. 6b). Deficiency of TLR5 did not affect the levels of hepatic cholesterol and serum triglyceride, free fatty acids, and cholesterol no matter before or after PHx (Fig. 6b-c). Consistent with these results, CBLB502 treatment increased hepatic triglyceride and free fatty acids accumulation at 24 and 36 hours following PHx, but had no significant effect on hepatic cholesterol content (Fig. 6d). Histological examination and Oil Red O staining of liver sections from CBLB502-treated and WT mice at different time points after PHx further confirmed these results (Fig. 6e). Collectively, these data indicate that the TLR5 signaling pathway is involved in the regulation of the hepatic lipid transient accumulation induced by PHx.