Objectives: Toll-like receptors and downstream signal transduction pathways play pivotal roles in induction of inflammation, which is crucial for liver injury and regeneration.
Materials and Methods: Using a mouse model of partial hepatic ischemia-reperfusion injury followed by a 28-day time course for liver repair and regeneration, we assessed gene expression levels for Toll-like receptors, myeloid differentiation primary response 88, TIR-domain-containing adapter-inducing interferon-β, nuclear factor κB, interferon regulatory factors, tumor necrosis factor-α, and interleukins 1β and 6 at days 1, 4, 7, 14, and 28 after reperfusion in liver and blood cells by quantitative polymerase chain reaction.
Results: Mouse liver was gradually injured until 24 hours after reperfusion, and necrotic areas remained for 7 days. Concurrent with liver necrosis, overexpression of hepatocyte growth factor in blood cells (days 1-14), transient overexpression of cyclin D1 at day 7 in hepatic cells, and overexpression of transforming growth factor-β1 at days 7 and 14 in blood cells were used to characterize the priming, proliferative, and termination phases of liver regeneration. Liver regeneration was associated with significant up-regulation of Toll-like receptor 4, p65, interferon regulatory factors 1, 3, 9, tumor necrosis factor-α, and interleukin 1β at 24 hours. Liver regeneration was also associated with persistent overexpression of MyD88 (days 1-28) and with delayed TIR-domain-containing adapter-inducing interferon-β (days 4-28) in hepatic cells. In peripheral blood cells, Toll-like receptor 2 and MyD88 were up-regulated at 24 hours and Toll-like receptor 4 (days 1-14) and interferon regulatory factor 1 (days 1-7) showed persistent overexpression concomitant with interferon regulatory factor 5 (days 7-14); interleukin 1β (days 1-28) and interleukin 6 (day 4-28) also showed persistent expression.
Conclusions: We depict for the first time a prospective view of cooperative transcriptional activation of Toll-like receptors/adaptors/interferon regulatory factors/cytokines in both liver and blood cells during different phases of liver repair after ischemia-reperfusion injury.
Key words : Liver transplantation, mRNA expression, Toll-like receptor signaling
Host innate immune responses participate not only in host defense but also in tissue injury and repair as a consequence of prompt induction and activation of inflammation.1 It is well established that, during liver ischemia-reperfusion injury (IRI), Toll-like receptors (TLRs), cell surface receptors of parenchymal and nonparenchymal hepatic cells, are stimulated by danger-associated molecular patterns released from oxidant-induced hepatocellular damage.2 Downstream signal transduction of TLRs through myeloid differentiation primary response 88 (MyD88) and TIR-domain-containing adapter-inducing interferon-β (TRIF) adaptor molecules leads to activation of nuclear factor-κB (NF-κB) and several different interferon regulatory factors (IRFs) to induce the expression of proinflammatory cytokines and chemokines.3 This intense inflammation leads to recruitment of activated neutrophils and T cells and subsequently liver necrosis.4
Although IRI can damage hepatic cells, it can also elicit tissue repair, a process that is involved in the removal of necrotic tissue and hepatocyte proliferation.1 In the context of hepatic tissue repair after partial hepatectomy (PH), activation of TLRs in downstream signaling, in response to gut-derived microbial components and danger-associated molecular patterns, also results in induction of cytokine and chemokine production, including tumor necrosis factor α (TNF-α) and interleukin 6 (IL-6), which in turn triggers hepatocyte proliferation and liver regeneration.5,6 Despite the integral role of TLRs, adaptor molecules, IRFs, and cytokine axis during liver injury, there are some controversial data on the involvement of these pathways in different models of liver regeneration. Studies on models of PH knockout mice for TLR2, TLR3, TLR4, TLR9, and MyD88 genes have revealed that TLR3 plays an inhibitory role in the priming of liver regeneration.7 The TLR2, TLR4, and TLR9 genes are not essential for liver regeneration, whereas MyD88 is needed for cytokine production (especially TNF-α and IL-6) but is not needed for liver regeneration after PH.8-10 However, another study on CCl4-induced liver injury revealed that liver regeneration is suppressed in TLR4-null mice through decreased IL-6 levels in both liver tissue and blood.11 Based on our knowledge, we have found no data on the involvement of the IRF family of transcription factors in liver tissue and peripheral blood cells during liver regeneration after IRI. Microarray analysis in partially hepatectomized mice has shown that IRF1 and IRF9 may be induced in liver during the primary phase of liver regeneration.12
Although PH is the most well-studied model of liver regeneration to date, additional models of liver injury (such as IRI) represent better means to assess the reparative and regenerative mechanisms in stressed and injured hepatocytes. Our study was conducted as a result of the controversial role of TLRs and their downstream adaptor molecules and the lack of data regarding the involvement of IRFs in liver repair after acute liver injury. We measured mRNA expression levels of TLR2, TLR3, TLR4, and TLR9 as responders to endogenous ligands and levels of MyD88, TRIF, IRF1, IRF3, IRF4, IRF5, IRF7, IRF8, and IRF9 in both hepatic tissue and peripheral blood samples from mice to determine cooperative transcriptional activation of these genes in liver tissue and peripheral blood cells during different phases of liver repair.
Materials and Methods
Experiments were performed using male Balb/c mice (8-10 weeks old, 26-30 g) purchased from Shiraz University of Medical Sciences (Shiraz, Iran). Animals were housed in a room with 12:12-h light-dark cycle at 22 ± 1°C stable temperature and were allowed free access to water and food. Mice were randomly divided into 6 groups: 1 sham-operation group and 5 ischemia-reperfusion groups with 60 minutes of ischemia followed by 1, 4, 7, 14, or 28 days of reperfusion (4-6 mice/group). As described previously, a model of partial (70%) IRI in liver was used.13 Briefly, anesthesia was induced by intraperitoneal injection of 100 mg/kg ketamine and 10 mg/kg xylazine. After midline laparotomy, portal triads from left and median lobes were occluded using an atraumatic microvascular clamp. Reperfusion was initiated after 60 minutes of ischemia by removing the clamp. The abdomen was then closed using 5-0 silk sutures. Sham-operation mice underwent the same protocol without portal triad occlusion. Whole blood samples were collected from left ventricles. Median and left lobes of liver were frozen at -80°C for further analysis. All animal protocols were approved by the Ethics Committee of Shiraz University.
Real-time polymerase chain reaction
Total RNA was extracted from fresh whole blood and frozen liver samples using the RNX-Plus kit (CinnaGen, Tehran, Iran) according to the manufacturer's instruction. cDNA was synthesized using the PrimeScript reverse transcriptase reagent kit (Takara, Tokyo, Japan). Quantitative real-time polymerase chain reaction (PCR) was performed with use of the Rotor-gene 6000 real-time PCR system (Corbett Life Science, Sydney, Australia) and the SYBR Green Premix Ex Taq II kit (Takara). Primer sequences were designed using the intron inclusion method by AlleleID version 7.5 software (PREMIER Biosoft, Palo Alto, CA, USA) (Table 1) and also tested for mRNA specificity and no genomic DNA amplification. We included 25 ng of each cDNA sample in the PCR reactions (final volume of 10 µL). For IRF5 and EF-1 genes, the PCR program was set as follows: 95°C for 5 seconds, followed by 40 cycles of 95°C for 20 seconds, 62°C for 30 seconds, and 72°C for 30 seconds. For other studied genes, the program was set as follows: 95°C for 5 seconds, followed by 40 cycles of 95°C for 20 seconds and 60°C for 45 seconds. Expression levels of each gene were first normalized with EF-1 (as internal control) and then compared with the sham group using 2-ΔΔCt method.
Liver damage assessment
We evaluated hepatocellular injury and repair by assessing the plasma levels of alanine aminotransferase (ALT) using an autoanalyzer (Mindray BS-380, Trezzano Sul Naviglio, Italy) and by observing histopathologic changes. For this purpose, formalin-fixed left liver lobes were embedded in paraffin. Next, nine 5-μm sections from each mouse liver were stained with hematoxylin and eosin and scored by an expert pathologist from 0 to 4 for their sinusoidal congestion, hepatocyte vacuolation, polymorphonuclear cell infiltration, and parenchymal necrosis, as described by Suzuki and associates.14 The observer was blinded to treated groups.
Data are shown as means ± SEM. Homogeneity of variances (achieved by square root of data) and comparisons between several time points with sham group were performed using one-way analysis of variance followed by Dunnett post hoc test. Correlated gene expression levels were determined by Pearson correlation analysis. All statistical analyses were performed with SPSS software (SPSS: An IBM Company, version 16.0, IBM Corporation, Armonk, NY, USA), and P < .05 was considered statistically significant.
Liver damage and repair assessment
To confirm our partial (70%) hepatic IRI and reparative model, plasma levels of ALT and pathologic changes in liver tissue were assessed. As shown in Figure 1A, based on Suzuki score, polymorphonuclear cell infiltration and hepatocyte necrosis were evident 24 hours, 4 days, and 7 days after reperfusion. Necrotic area healed 14 and 28 days after reperfusion (Figure 1B). Plasma ALT levels also increased 3 and 24 hours after reperfusion and subsequently returned to levels shown in the sham group 4 days after reperfusion (Figure 1C). These results successfully reproduced our previous results in which we analyzed liver injury after 60 minutes of ischemia followed by 3 hours and 7 days of reperfusion.15
To characterize the regenerative response, we measured hepatic expression of cyclin D1, a marker for progression of the G1 cell cycle phase.16 As shown in Figure 1D, the expression levels of cyclin D1 increased transiently at day 7 after reperfusion and immediately down-regulated 14 days and 28 days after reperfusion. We also measured the liver and blood mRNA expression levels of hepatocyte growth factor (HGF) and transforming growth factor β1 (TGF-β1) as important stimulators and inhibitors of hepatocyte proliferation, respectively.17 Hepatocyte growth factor levels did not significantly differ in liver over the 28-day postreperfusion time course compared with that shown in the sham group (Figure 1E). However, levels measured in blood in IRI animals increased 24 hours after reperfusion and remained up-regulated until day 7 and day 14 after reperfusion (Figure 1F). The hepatic expression levels of TGF-β1 were also transiently elevated at 3 hours after reperfusion but returned to sham levels 1 day, 4 days, and 7 days later. Levels were then down-regulated 14 and 28 days after reperfusion (Figure 1E). Blood levels of TGF-β1 were enhanced at days 7 and 14 after reperfusion and returned to baseline levels at day 28 (Figure 1F). Together, these results suggest that the liver was gradually injured until 24 hours after reperfusion and the necrotic area remained until 7 days after reperfusion.
Concurrent with liver necrosis, hepatic regeneration was also primed at 24 hours (as shown by elevated mRNA levels of HGF in peripheral blood cells), which continued until day 7 (as shown by hepatocyte proliferative indications, such as cyclin D1) and then terminated at 14 to 28 days after reperfusion (as shown by overexpression of TGF-β1 and histologic findings).
mRNA expression levels of Toll-like receptors and adaptor molecules during
To determine transcriptionally activated TLRs in liver repair and regeneration after hepatic ischemia, we evaluated the mRNA levels of TLR2, TLR3, TLR4, and TLR9 as responder TLRs to endogenous ligands in both liver tissue and blood samples 1 to 28 days after reperfusion. Our results showed that TLR2 and TLR9 were down-regulated in liver samples at days 14 and 28 after reperfusion during the regenerative phase compared with that shown in the sham group (Figure 2A). We found that TLR3 was also down-regulated 4, 14, and 28 days after reperfusion (Figure 2A); however, expression levels of TLR3 and TLR9 in blood did not differ between groups. In blood samples, mRNA levels of TLR2 were up-regulated transiently at 24 hours and 14 days after reperfusion (Figure 2B), with TLR4 also showing persistent up-regulation 1, 4, 7, and 14 days after reperfusion (Figure 2B). In addition, expression levels of TLR4 in liver samples were up-regulated 24 hours after reperfusion, returning to baseline levels 4 to 28 days after reperfusion compared with that shown in the sham group (Figure 2A). These results revealed transcriptional gene activation of TLR2 in peripheral blood cells and, more notably, of TLR4 in both liver tissue and peripheral blood cells during liver regeneration after liver IRI.
To determine which adaptor molecules are transcriptionally overexpressed downstream of the TLR genes, we evaluated expression levels of MyD88 and TRIF in liver and blood samples. We found that hepatic mRNA of MyD88 was continuously overexpressed from 24 hours to 28 days after reperfusion compared with that shown in the sham group (Figure 2A). However, although its mRNA levels in blood were transiently elevated 24 hours after reperfusion, MyD88 levels returned to baseline 4 to 28 days after reperfusion (Figure 2B). Hepatic mRNA levels of TRIF were also up-regulated with a delay 4, 7, 14, and 28 days after reperfusion (Figure 2A); however, levels in blood were transiently up-regulated at 24 hours after reperfusion (Figure 2B). We found a positive correlation between transcriptionally activated MyD88 and TLR2 (r = 0.72, P < .001), MyD88 and TLR4 (r = 0.77, P < .001), TRIF and TLR4 (r = 0.41, P = .041), and TRIF and MyD88 (r = 0.66, P < .001) in peripheral blood cells. However, there were no significant correlations between TLR4 and adaptor genes in liver tissue, but MyD88 and TRIF were positively correlated with each other (r = 0.63, P = .001).
mRNA expression levels of interferon regulatory factors during liver
To clarify which IRFs are overexpressed downstream of TLRs during liver regeneration, we evaluated the expression levels of IRF1, IRF3, IRF4, IRF5, IRF7, IRF8, and IRF9 in liver and blood samples. Our results revealed that hepatic mRNA levels of IRF1, IRF3, and IRF9 were overexpressed transiently 24 hours after reperfusion, whereas IRF5 was down-regulated at this time point (Figure 3A). The mRNA levels of most studied IRFs were gradually decreased in liver tissue; of note, we observed a significant down-regulation of IRF4 at day 4, IRF3 and IRF5 at days 14 and 28, IRF8 at day 14, and IRF9 at day 28 after reperfusion (Figure 3A). However, we also observed that hepatic mRNA levels of IRF1 were transiently elevated at day 7 (Figure 3A). In blood samples, IRF1 was overexpressed at 1, 4, and 7 days after reperfusion and returned to sham levels at day 14 (Figure 3B). Interestingly, in an opposite manner, expression levels of IRF5 were not different from those shown in the sham group until 4 days after reperfusion; we found it to be overexpressed at days 7 to 14 and then down-regulated at day 28 (Figure 3B). In blood samples, expression levels of IRF3 and IRF4 were not different over the 1- to 28-day times points after reperfusion versus levels shown in the sham group. However, the mRNA levels of IRF7 (at days 4, 7, 14, and 28), IRF8 (at days 4 and 7), and IRF9 (at day 28) were down-regulated (Figure 3B).
We also evaluated the mRNA levels of the p65 subunit of NF-κB, another important transcription factor activated downstream of TLRs. Our results revealed a transient overexpression of this factor at 24 hours after liver reperfusion (Figure 3A), although levels in peripheral blood cells were not changed during the measured liver regeneration time points (Figure 3B). Overall, these data suggest a cooperative role for NF-κB, IRF1, IRF3, and IRF9 in hepatic cells and also a sequential role for IRF1 and IRF5 in peripheral blood cells during different phases of liver regeneration.
Our correlative analyses also revealed positive correlations between IRF1 (r = 0.42, P = .039), IRF3 (r = 0.60, P = .001), IRF9 (r = 0.7, P < .001), and RelA genes and TLR4 transcriptional activation in liver tissue. However, MyD88 was not correlated with these transcription factors. In the context of peripheral blood cells, IRF1 (r = 0.51, P = .011) and IRF5 (r = 0.46, P = .028) were correlated with TLR4 but not with TLR2. MyD88 was also correlated with IRF1 (r = 0.48, P = .014). We also found positive correlations between TGF-β1 and both TLR4 (r = 0.43, P = .039) and IRF5 (r = 0.50, P = .013) in blood samples.
mRNA expression levels of proinflammatory cytokines during liver
We also assessed mRNA expression levels of TNF-α, IL-6, and IL-1β (IRFs downstream proinflammatory cytokines) during the regenerative phases after hepatic IRI. Our results showed a transient overexpression of TNF-α at 24 hours and a down-regulation at 4 and 14 days after reperfusion in liver (Figure 4A). However, its mRNA levels did not change during the 4- to 28-day time course in blood samples (Figure 4B). Interleukin 6 was also transcriptionally overexpressed at the 4-, 7-, 14-, and 28-day time points after reperfusion in peripheral blood cells (Figure 4B), whereas, in liver tissue, the mRNA levels were not different from those in the sham group (Figure 4A). Hepatic mRNA levels of IL-1β showed overexpression at days 1 and 7 versus sham group levels (Figure 4A), with levels of IL-1β in blood constantly elevated at 1, 4, 7, 14, and 28 days after reperfusion (Figure 4B).
Data regarding activation of nonparenchymal hepatic cells, particularly Kupffer cells, through the TLRs/MyD88 signaling pathway during the priming phase of liver regeneration remain controversial. Previous studies on TLR2, TLR4, TLR9, and MyD88 null mice have indicated that signal transduction downstream of MyD88 (but not TLR2, TLR4, and TLR9) is essential for cytokine induction in liver tissue during the early phase of liver regeneration. However, MyD88 deficiency did not impair hepatocyte proliferation and outcomes of liver regeneration.8-10 Our results are consistent with Su and associates, who demonstrated that liver regeneration after CCl4 liver injury was impaired in TLR4-null mice.11 Correlated overexpression of IRF1, IRF3, IRF9, p65, TNF-α, and IL-1β with TLR4 but not MyD88 in liver tissue may suggest a TLR4/IRF1, IRF3, IRF9, NF-κB/TNF-α, and IL-1β signaling axis at the priming phase of liver regeneration in our hepatic IRI model. It has been recently reported that Kupffer cell maintenance is in a tolerant state, with certain products of commensal bacteria crucial for triggering liver regeneration after PH.18 This tolerant state is associated with decreased mRNA expression of TLR4 to baseline levels.19 This state also leads to impaired recruitment of MyD88 to TLR4, impaired activation of NF-κB, and decreased cytokine transcriptional activation.20 Therefore, our observations on the return of mRNA levels of TLR4 and its downstream IRFs, p65, TNF-α, and IL-1β to baseline levels in the liver may have occurred as a result of TLR4 stimulation tolerance. Nevertheless, the sustained transcriptional activation of MyD88 and TRIF in liver tissue during liver regeneration remains to be clarified.
Our results also showed a transient overexpression of IRF1 and IL-1β in hepatic cells concurrent with cyclin D1 overexpression at day 7. Previous studies have reported a negative regulatory role for IL-1β in hepatocyte proliferation during liver regeneration.21 Sun and associates have shown that the inhibitory effects of TLR3 on hepatocyte proliferation after PH is diminished in IRF1 knockout mice.22 On the other hand, it has been shown that IRF1 mRNA is up-regulated in cultured rat hepatocytes in response to IL-1β stimulation.23 Overall, previous findings that support our results may suggest the presence of an IL-1β/IRF1 axis to terminate liver regeneration after progression of hepatocyte proliferation through cyclin D1 induction. In addition to the liver priming events, it is well documented that depletion of some peripheral blood cells, particularly platelets and neutrophils, impairs hepatocyte proliferation and successful liver regeneration.24-26 Previous reports have indicated that activation of the TLR4/MyD88 signaling axis is essential for neutrophil activation, survival, and degranulation, as well as platelet activation, degranulation, and aggregation during the early phases of liver regeneration.27-30 Activation of the TLR4/MyD88 signaling pathway on platelet cells in response to lipopolysaccharide stimulation leads to IL-1β production.31 Platelets respond to IL-1β itself through an autocrine stimulatory loop to thrombocytosis, which is essential for hepatocyte proliferation during the early phase of liver regeneration.31,32 The IRF1 can also promote megakaryocytopoiesis and platelet production.33 Together, these studies that are consistent with our results suggest that TLR4-, MyD88-, TRIF-, IRF1-, and IL-1β-related pathways are associated with a functional role in peripheral blood cells at priming phase of liver regeneration. Interleukin 1β has also been shown to induce IL-6 production in peripheral blood monocytes,34 and IL-6 has been shown to promote megakaryocytopoiesis and platelet production at a later phase.18 We found that overexpression of IL-6 in peripheral blood cells was only correlated with IL-1β (r = 0.54, P = .015) but not with TLR4 and IRF1 gene overexpression. Overall, these results suggest that signal transduction during the proliferative phase of liver regeneration in peripheral blood cells is not associated with MyD88 transcriptional activation but is associated with TLR4, IRF1, IL-1β, and IL-6 genes.
Persistent overexpression of IL-6 and HGF in peripheral blood cells without their transcriptional activation in hepatic tissue may indicate that these cells are a major source of IL-6 and HGF during liver regeneration after IRI. Moles and associates showed that TLR2 but not TLR4 is required for recruitment of neutrophils to the CCl4 injured liver at the early phase of liver regeneration.35 In addition, MyD88 has been shown to be essential for early recruitment of neutrophils in response to systemic inflammatory stimulus.36,37 Consistent with our results, these findings suggest a role for TLR2/MyD88 genes in transmigration of peripheral blood cells into the liver at the early phase of liver regeneration. The mito-inhibitory properties of TGF-β1 on hepatocytes through suppression of HGF and its important role in hepatic tissue remodeling and assembly toward the end of liver regeneration indicate that this cytokine is a liver regeneration terminator.17 Interestingly, we also found overexpression of TGF-β1 at the termination phase of liver regeneration to be strongly correlated with TLR4 (r = 0.42, P = .047) and IRF5 (r = 0.5, P = .013) overexpression in peripheral blood cells, suggesting a cooperative action of these factors in the liver regeneration termination phase. Moreover, our observation of hepatic overexpression of TGF-β1 during liver injury (3 h after reperfusion) along with our recent study showing overexpression of the TLR4/IRF5/proinflammatory cytokine axis 3 hours after reperfusion leading to hepatocellular necrosis in a mouse IRI model suggests a dual role for TLR4/IRF5/TGF-β1 signaling axis in both liver injury and regeneration after hepatic IRI.15
In summary, our present investigation, in which we assessed mRNA levels of genes involved in a key pathway of innate immune response, revealed that concomitant transcriptional activation of TLR2 and TLR4/MyD88 and TRIF/IRF1, IRF3, IRF5, IRF9, and p65/TNF-α, IL-1β, and IL-6 genes could have cooperative roles in both hepatic and peripheral blood cells during different phases of liver regeneration after hepatic IRI. Although this study was a prospective view, further investigations on TLRs/adaptors/IRFs/cytokine signaling pathways during liver regeneration after IRI and additional studies using knockout mice are needed to unveil the related signaling pathways to these genes during liver regeneration.
DOI : 10.6002/ect.2017.0120
From the 1Department of Biology, College of Sciences, Shiraz
University, Shiraz, Iran; and the 2Transplant Research Center, Shiraz
University of Medical Sciences, Shiraz, Iran
Acknowledgements: The authors have no conflicts of interest to report. This study was supported by Shiraz University (93GCU4M1740).
Corresponding author: Iraj Saadat, Department of Biology, College of Sciences, Shiraz University, Shiraz 71467-13565, Iran
Phone: +98 71 36137435
Figure 1. Liver Injury and Regeneration Assessment
Figure 2. Gene Expression of Toll-Like Receptors and Their Downstream Adaptor Molecules
Figure 3. Gene Expression of Interferon Regulatory Factors and p65 Subunit of Nuclear Factor κB
Figure 4. Gene Expression of Proinflammatory Cytokines
Table 1. Primer Sequences and Amplicons