The authors have no conflicts of interest to declare.
This project was supported by the Pennsylvania Tobacco Settlement Funds.
Address reprint requests to Tadahiro Uemura, M.D., Ph.D., Division of Transplantation, Department of Surgery, College of Medicine, Pennsylvania State University, 500 University Drive, H062, Hershey, PA17033. Telephone: 717-531-5921; FAX: 717-531-5851; E-mail: email@example.com or Samuel Shao-Min Zhang, M.D., Ph.D., Neural and Behavioral Sciences, College of Medicine, Pennsylvania State University, 500 University Drive, H109, Hershey, PA 17033. Telephone: 717-531-8480; FAX: 717-531-5184; E-mail: firstname.lastname@example.org
phosphorylated signal transducer and activator of transcription
signal transducer and activator of transcription
wild type; CFF, TTR-Cre/STAT3(F/F); Z/EG, lacZ/EGFP; CV, Central Vein; PT, Portal Triad; FloxP, Flanked LoxP; C20, anti-STAT3 antibody; AKT, Protein Kinase B
Warm ischemia/reperfusion (I/R) is a common clinical problem during liver transplantation. I/R injury in liver transplantation is strongly associated with a nonfunctional liver, which is a life-threatening condition and requires emergent liver retransplantation. I/R injury also happens during liver resection. To minimize blood loss during liver resection, the hepatic vascular inflow is temporarily clamped (the Pringle maneuver). The Pringle maneuver is a well-established and routine procedure during liver surgery. However, the Pringle maneuver causes liver I/R injury, which can lead to rapid hepatocellular injury and tissue necrosis. Warm ischemia also occurs during trauma and shock. The liver is the most commonly injured abdominal organ in blunt and penetrating abdominal trauma. The management of trauma and shock frequently involves exposing the liver to various periods of warm ischemia followed by reperfusion. I/R injury is significantly associated with morbidity and mortality under such conditions.
The prevention or minimization of liver I/R injury was the clinical priority of this study because there is still no safe and promising strategy for protecting the liver from I/R injury. Signal transducer and activator of transcription 3 (STAT3) is a major signaling molecule for a variety of genes in response to cell stimuli, and it plays a key role in many cellular processes such as cell growth and apoptosis. Recent studies have suggested that STAT3 activation is involved in the protection of hepatocytes against I/R injury. Several studies have examined extracellular signaling (interleukin-6 and interleukin-22) in the STAT3 cascade during liver I/R injury.[4, 5] However, no study has shown a direct effect of endogenous STAT3 on liver I/R injury. In this study, we created hepatocyte-specific STAT3-deficient mice, and we investigated the mechanisms of STAT3 in liver I/R injury with partial liver ischemia models.
MATERIALS AND METHODS
Generation of Hepatocyte Tissue–Specific STAT3-Deficient Mice
Classic knockout (KO) of the STAT3 gene in a mouse results in early embryonic lethality. Using the Cre-LoxP system, we created a conditional STAT3 allele in which exons 18 to 20 were flanked by 2 LoxP sequences.[6-8] The removal of exons 19 and 20, which encoded the Src homology 2 domain of STAT3, destroyed the function of the protein by blocking its dimer formation. A hepatocyte tissue–specific STAT3 deficiency was generated via the crossing of STAT3 Flanked LoxP/Flanked LoxP (F/F) mice with transferrin (TTR)-Cre mice (strain 4-18). The TTR promoter drove hepatocyte-specific expression of Cre recombinase. We examined the efficiency of Cre activation by crossing TTR-4-18Cre(+/−) mice with a green fluorescent protein (GFP) report FloxP mouse line (Z/EG), and this allowed us to confirm that TTR-driven Cre recombinase could specifically delete FloxP in hepatocytes.
STAT3(F/F) mice and TTR-Cre(+/−) mice with a C57BL/6 background were bred at the Penn State College of Medicine Animal Core in accordance with the guidelines of the Institutional Animal Care and Use Committee of Penn State College of Medicine. Mice were housed in pairs in plastic cages in a pathogen-free environment with access to food and water on a 12-hour light-dark schedule.
Genotyping and Grouping
The genotyping of STAT3(F/F) mice and TTR-Cre(+/−) mice was confirmed 21 days after birth with a polymerase chain reaction analysis (T100 thermal cycler, Bio-Rad Laboratories, Hercules, CA). Male homozygous and wild-type (WT) littermates were used for this study. Mice were divided into 2 groups: a STAT3-deficient group and a control group. At least 4 animals were used at each point in these experimental groups.
70% Partial Liver Ischemia
Mice were anesthetized intraperitoneally with a mixture of 100 mg/kg ketamine and 10 mg/kg xylazine. A midline incision was made in the abdomen. The liver was exposed with retractors placed in the flanks. The blood supply to the medial largest lobe and lateral lobe of the liver (70% of the liver) was temporarily interrupted (ischemia) with the application of a microclamp (FD562, Aesculap, South San Francisco, CA) to the vascular pedicle (Fig. 1). The blood supply was resumed by the removal of the microclamp after 90 minutes of ischemia.
Sampling and Gross Observation
Sera and livers were collected at each point after reperfusion. Portions of these livers were fixed with 4% paraformaldehyde in phosphate-buffered saline for at least 6 hours at 4°C for immunohistochemistry studies. Fresh liver tissues were also used for full cell lysates. For monitoring liver injury, serum was collected from each animal for enzymatic analysis.
Serum Markers of Reperfusion Injury
The serum level of alanine aminotransferase (ALT) was used as a marker of I/R injury. ALT was measured with a serum multiple-biochemical analyzer in clinical laboratories.
Pathological Evaluation of Liver Injury
Histological changes were scored from 0 to 4 according to the degree of cytoplasmic vacuolation, congestion, and necrosis with a score of I/R injury (the Suzuki score) 6 hours after reperfusion. Images were taken from all stained slides from STAT3-deficient and control groups. At least 3 images were recorded from each field. Polymorphonuclear neutrophils (PMNs) were counted blindly in 40 high-power fields, and the results were expressed as the total number of PMNs in 5 high-power fields.
Cell Culture and 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium Bromide (MTT) Cell Proliferation Assay
HepG2 cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (FBS). The culture medium was discarded every 2 to 3 days. Cultures were incubated at 37°C in a humidified atmosphere with 5% carbon dioxide. Passaged cells were seeded into 3 rows of a 96-well plate (1 × 104 cells/well) for 24 hours before treatment (100 μL/well). After 24 hours in the culture, cells in the first row without medium replacement served as controls. The medium of the second row was replaced with a culture medium without 10% FBS. The medium of the third row was replaced with 100 μL of a culture medium with 100 μM (E)-2-cyano-3-(3,4-dihydrophenyl)-N-(phenylmethyl)-2-propenamide (AG490; catalog number 0414, Tocris Bioscience), which is a STAT3 inhibitor. The MTT assay followed the manufacturer's protocol. The cells were cultured for another 12 or 24 hours. Then, an MTT solution (with a final concentration of 0.5 mg/mL in phosphate-buffered saline) was added to each well, and it was incubated for another 4 hours [an MTT solution. The medium was removed, and 100 μL of dimethyl sulfoxide was added to each well to dissolve purple crystals of formazan. Ten minutes later, the absorbance was measured with a spectrophotometer at a wavelength of 450 nm with a microplate reader (model 680XR, Bio-Rad Laboratories). The MTT reagent had to be kept at 4°C in the dark. The detergent reagent could be stored at either 4°C or the ambient temperature. When the detergent reagent was kept at 4°C, the bottle had to be warmed for 5 minutes at 37°C, and the reagent had to be gently mixed by inversion before use (to prevent the creation of bubbles).
Cells or fresh liver tissues were used for full cell lysates. Whole cell extracts were prepared, and western blot assays were performed as described previously. Briefly, livers or cells from each experimental group and each control group were suspended in approximately 80 μL of a T-PER tissue protein extraction buffer (Pierce, Rockford, IL) with a protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN). The tissue samples were homogenized and centrifuged at 10,000g and 4°C for 5 minutes. The supernatant was collected, and the protein concentrations were measured. The whole cell extract (15-25 μg) was separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to an Immun-Blot polyvinylidene difluoride membrane (Bio-Rad Laboratories). After they were blocked with 3% nonfat milk in a washing buffer, the membranes were incubated with primary antibodies. The primary antibodies included caspase-3 (Cell Signaling, Danvers, MA), cleaved caspase-3 (Cell Signaling), STAT3 (Santa Cruz Biotechnology, Santa Cruz, CA), and phosphorylated signal transducer and activator of transcription 3 (pSTAT3; Cell Signaling). After the washes, they were incubated in anti-rabbit or anti-mouse immunoglobulin G coupled to horseradish peroxidase. The immunoreactive bands were visualized with SuperSignal chemiluminescent substrate (Pierce, Rockford, IL) via film development or with the ChemiDoc XRS system (Bio-Rad Laboratories). The signal intensity was analyzed with Quantity One (Bio-Rad Laboratories).
Densitometry analysis was used for all gel images from immunoblotting assays. Images from western blotting were analyzed semiquantitatively with ImageJ software. After background correction, the spots representing STAT3 (C20; catalog number sc-482, Santa Cruz Biotechnology), pSTAT3 (Tyr705; catalog number sc-7993, Santa Cruz Biotechnology), and beta-actin (catalog number A5316, Sigma-Aldrich) were measured for their integrated density. At least 3 replicates on average were obtained for statistical analysis.
Data Analysis and Interpretation
An analysis of variance and the Student t test were applied for the statistical analysis. P < 0.05 was considered to indicate a significant change in this study.
Inhibition of STAT3 Signaling Reduces Hepatocyte Survival In Vitro
We first tested whether AG490, a pharmacological STAT3 inhibitor, could sufficiently inhibit STAT3 tyrosine phosphorylation in the HepG2 cell line (Fig. 2). AG490 significantly inhibited STAT3 tyrosine phosphorylation with (P = 0.04) or without 10% FBS stimulation (P = 0.001) at 24 hours (Fig. 2A,B). However, there were no changes in the total STAT3 protein levels at either tested time point (Fig. 2A,C). Serum withdrawal can mimic ischemic injury in many cultured cells.[14-17] Using the HepG2 cell line, we examined the effects of STAT3 signaling on cell survival in vitro with MTT assays. Figure 3A,B shows the results of MTT assays after 12 and 24 hours of incubation, respectively. As shown in Fig. 3A, in comparison with 10% FBS, serum withdrawal significantly reduced cell survival (P = 0.001). The addition of AG490 to 10% FBS also significantly reduced cell survival (P = 0.01; Fig. 3A). The combination of AG490 and serum withdrawal remarkably and synergistically reduced cell survival (Fig. 3A). These results indicate that STAT3 plays an important role in cell survival and that its effect is independent of the serum condition, and they suggest that STAT3 provides hepatocyte protection against ischemic injury in vitro. Similar findings were observed after 24 hours of incubation (Fig. 3B).
Immediate Increase in STAT3 Tyrosine Phosphorylation by Warm I/R In Vivo
STAT3 was first identified as an immediate-response gene in the liver. To understand the time course of STAT3 activation, we examined the expression of STAT3 and its activated form tyrosine pSTAT3 in WT livers with a partial liver I/R model (Fig. 1). The expression was analyzed in nonischemic quadrate lobes as well as ischemic lobes. As shown in Fig. 4A,B, the levels of pSTAT3 (Tyr705) in ischemic lobes were increased 6 (P < 0.01) and 18 hours (P < 0.01) after reperfusion in comparison with 0.5 hours after reperfusion. The expression showed a trend of returning to the normal level within 48 hours, but the level was still higher (P < 0.01; left panel in Fig. 4B). Interestingly, the expression of pSTAT3 (Tyr705) temporarily decreased 1 hour after reperfusion (P < 0.01; left panel in Fig. 4B). The total STAT3 (C20) protein level in ischemic lobes reached a peak 18 hours after reperfusion (P < 0.01; right panel in Fig. 4B). As shown in Fig. 4A,C, there were different patterns of pSTAT3 (Tyr705) and STAT3 (C20) responses in nonischemic lobes versus ischemic lobes. pSTAT3 (Tyr705) levels in nonischemic lobes increased from 1 hour (P < 0.05), and this continued to 48 hours (P < 0.01), but the increased levels (left panel in Fig. 4C) were mild in comparison with those for ischemic lobes. A significant increase in the total STAT3 (C20) protein level was observed with a peak at 48 hours (P < 0.01), and this was different from the case for ischemic lobes.
Creation of Hepatocyte-Specific STAT3-Deficient Mice
Using the FloxP-Cre system (Fig. 5A,B), we generated hepatocyte-specific STAT3-deficient mice. The exons of STAT3 at 18 to 20 that encoded Src homology 2 domains were deleted (Fig. 5A) after the specific expression of Cre in hepatocytes (Fig. 5B). As shown in Fig. 5C, we examined and compared the levels of pSTAT3 and total STAT3 in STAT3-deficient and WT livers under rest (control) and I/R conditions. Even under rest (control) conditions, pSTAT3 (P < 0.01) and total STAT3 (P < 0.001) proteins were observed in WT mice versus hepatocyte-specific STAT3-deficient mice (Fig. 5C,D). The faint baseline levels of STAT3 and pSTAT3 in the western blots might reflect the negligible levels in nonparenchymal liver tissues other than hepatocytes.
Absence of STAT3 Tyrosine Phosphorylation in Livers With a Hepatocyte-Specific STAT3 Deficiency in Response to Warm I/R
After I/R injury, almost no expression of pSTAT3 was confirmed in STAT3-deficient livers (Fig. 5C,D). Under I/R conditions, the protein levels of both total STAT3 (P < 0.001) and pSTAT3 (P < 0.001) were significantly increased in WT livers versus STAT3-deficient livers (Fig. 5C,D). An approximately 4-fold increase in the total STAT3 protein level was observed in the STAT3-deficient livers under I/R conditions versus resting (control) conditions (Fig. 5D). This indicates that total STAT3 was expressed by nonparenchymal tissues of the liver in response to warm I/R injury.
Depletion of STAT3 Signaling in Hepatocytes Aggravates Liver Damage in Response to Warm I/R
We investigated the role of STAT3 in liver I/R injury in vivo with hepatocyte-specific STAT3-deficient mice. In WT mice, the ALT levels were 841 ± 433 and 484 ± 95 IU/L 3 and 6 hours after reperfusion, respectively. In the hepatocyte-specific STAT3-deficient group, the ALT levels were significantly elevated at 2310 ± 853 and 1086 ± 231 IU/L 3 (P < 0.01) and 6 hours after reperfusion (P < 0.001), respectively (Fig. 6A). Thus, the ALT level in the STAT3-deficient group significantly increased in comparison with the level for the WT littermates. Histopathological examinations of livers showed that the hepatocyte-specific STAT3-deficient mice had moderate to severe sinusoidal congestion (P = 0.02), cytoplasmic vacuolization (P < 0.01), and hepatocyte necrosis (P = 0.001) in comparison with their WT littermates (Fig. 6B,C). These results indicate that livers with the STAT3 deficiency in hepatocytes were more susceptible to warm I/R injury than livers in the WT group. Furthermore, we analyzed a number of inflammatory PMNs between WT and STAT3-deficient groups. The PMN counts were 59 ± 26 and 106 ± 54 per field for the WT group and 51 ± 47 and 118 ± 54 per field for the STAT3-deficient groups 3 and 6 hours after reperfusion, respectively (Fig. 6D). There was no difference in the PMN counts between these 2 groups; this suggests that inflammatory responses are similar for WT and hepatocyte-specific STAT3-deficient mice and that endogenous STAT3 signaling in hepatocytes is required for liver protection in response to warm I/R.
This study has demonstrated that STAT3 plays an important role in hepatocyte protection in response to warm I/R injury. STAT proteins were originally identified and isolated with cell culture systems that responded to cytokines such as interferon.[3, 18] It is well established that the STAT proteins have a range of essential roles in the generation and function of the immune system. Tissue-specific STAT3 disruption causes several phenotypes related to human diseases, such as Crohn's disease–like pathogenesis, heart failure, severe inflammation, and lethal hyperoxic lung injury.[6-8, 19] Thus, it is believed that STAT3 evolved to mediate signaling from cytokines and growth factors. In particular, STAT3, the mediator of interleukin-6 signaling, has been shown to be critical in cell progression from the G1 phase to the S phase, and it is significantly activated in the regenerating mouse liver after partial hepatectomy.
Recently, it has been reported that STAT3 is also involved in liver I/R injury. Ke et al. used small interfering RNA to knock down STAT3 to investigate the heme oxygenase 1–STAT3 axis in response to I/R injury. They showed that the STAT3 small interfering RNA group suffered significant hepatocyte damage during I/R injury in comparison with a control group. Yu et al. reported that the canonical notch pathway protects hepatocytes from I/R injury by repressing reactive oxide species production through Janus kinase 2 (JAK2)/STAT3 signaling. Kuboki et al. showed that STAT3 activation is augmented during I/R injury in chemokine (C-X-C motif) receptor 2–KO mice, and they suggested that STAT3 is associated with an early stage of liver recovery or regeneration after I/R injury. STAT3 has also been reported as a protective mediator for preconditioning against I/R injury.[4, 25] Thus, STAT3 appears to play an important role in liver protection and recovery from I/R injury. On the other hand, Clarke et al. recently reported that STAT3 was not a protective factor during acute liver injury induced by I/R with a direct inhibitor of STAT3 phosphorylation (6-Nitrobenzo[b]thiophene-1,1-dioxide). Freitas et al. also reported that a blockade of JAK2 signaling inhibited STAT1 and STAT3 and ameliorated liver I/R injury. Thus, the function of STAT3 in liver I/R injury is still controversial. Most studies have indirectly investigated the function of STAT in the signaling pathway. Therefore, we tried to study the direct effects of STAT3 on liver I/R injury with STAT3-KO mice. Because classic KO of the STAT3 gene in mice results in early embryonic lethality, we created conditional hepatocyte-specific STAT3-deficient mice. In this study, liver I/R injury induced STAT3 activation in hepatocytes, and hepatocyte-specific STAT3-KO mice showed more severe damage from liver I/R injury than WT mice. These results indicate that STAT3 activation in hepatocytes induced by I/R injury works for hepatocyte protection. The uniqueness of our model is that STAT3 was knocked out only in hepatocytes, and STAT3 expression was maintained in nonparenchymal liver cells. Therefore, STAT3 activation should be maintained in inflammatory cells in the liver suffering I/R. In this study, we analyzed the PMN counts, and there was no difference between WT mice and STAT3-KO mice. These results suggest that a loss of STAT3 in hepatocytes leads to a loss of hepatoprotection by inflammatory cells.
We also examined STAT3 expression in the nonischemic lobes of WT mice, and the pattern was different from the pattern for ischemic lobes. It is well known that a disruption of the portal flow to one side of the liver by ligation or embolization causes atrophy on that side, but it causes hypertrophy and hepatocyte proliferation on the contralateral side.[28, 29] These results suggest that STAT3 in nonischemic lobes may be related to the signal for liver regeneration or proliferation.
However, we also recognize the limitations of this study. I/R injury is caused by a multifactorial process, and numerous signaling cascades are involved. STAT3 functions in the JAK/STAT pathway or the heme oxygenase 1–STAT3 axis, and STAT3 activates phosphoinositide 3-kinase/Protein Kinase B signaling and the expression of a number of cytokine genes. In this study, we did not study upstream or downstream of STAT3, and further study is needed.
In summary, our present study demonstrates the role of hepatocyte-specific STAT3 in the protection of the liver from I/R injury with a conditional model of hepatocyte-specific STAT3 KO. This study supports the idea that STAT3-targeting therapy could be a therapeutic approach to combating liver I/R injury.