Interleukin-33 Is hepatoprotective during liver ischemia/reperfusion in mice†
Article first published online: 4 OCT 2012
Copyright © 2012 American Association for the Study of Liver Diseases
Volume 56, Issue 4, pages 1468–1478, October 2012
How to Cite
Sakai, N., Van Sweringen, H. L., Quillin, R. C., Schuster, R., Blanchard, J., Burns, J. M., Tevar, A. D., Edwards, M. J. and Lentsch, A. B. (2012), Interleukin-33 Is hepatoprotective during liver ischemia/reperfusion in mice. Hepatology, 56: 1468–1478. doi: 10.1002/hep.25768
Potential conflict of interest: Nothing to report.
- Issue published online: 4 OCT 2012
- Article first published online: 4 OCT 2012
- Accepted manuscript online: 10 JUL 2012 09:23AM EST
- Manuscript Accepted: 2 APR 2012
- Manuscript Received: 14 DEC 2011
- National Institutes of Health. Grant Number: DK56029 and AG025881
Interleukin (IL)-33 is a recently identified member of the IL-1 family that binds to the receptor, ST2L. In the current study, we sought to determine whether IL-33 is an important regulator in the hepatic response to ischemia/reperfusion (I/R). Male C57BL/6 mice were subjected to 90 minutes of partial hepatic ischemia, followed by up to 8 hours of reperfusion. Some mice received recombinant IL-33 (IL-33) intraperitoneally (IP) before surgery or anti-ST2 antibody IP at the time of reperfusion. Primary hepatocytes and Kupffer cells were isolated and treated with IL-33 to assess the effects of IL-33 on inflammatory cytokine production. Primary hepatocytes were treated with IL-33 to assess the effects of IL-33 on mediators of cell survival in hepatocytes. IL-33 protein expression increased within 4 hours after reperfusion and remained elevated for up to 8 hours. ST2L protein expression was detected in healthy liver and was up-regulated within 1 hour and peaked at 4 hours after I/R. ST2L was primarily expressed by hepatocytes, with little to no expression by Kupffer cells. IL-33 significantly reduced hepatocellular injury and liver neutrophil accumulation at 1 and 8 hours after reperfusion. In addition, IL-33 treatment increased liver activation of nuclear factor kappa light-chain enhancer of activated B cells (NF-κB), p38 mitogen-activated protein kinase (MAPK), cyclin D1, and B-cell lymphoma 2 (Bcl-2), but reduced serum levels of CXC chemokines. In vitro experiments demonstrated that IL-33 significantly reduced hepatocyte cell death as a result of increased NF-κB activation and Bcl-2 expression in hepatocytes. Conclusion: The data suggest that IL-33 is an important endogenous regulator of hepatic I/R injury. It appears that IL-33 has direct protective effects on hepatocytes, associated with the activation of NF-κB, p38 MAPK, cyclin D1, and Bcl-2 that limits liver injury and reduces the stimulus for inflammation. (HEPATOLOGY 2012)
Ischemia/reperfusion (I/R) injury of the liver is a major complication of liver resection, transplantation, and hypovolemic shock.1, 2 The injury response can be subdivided into two distinct phases. The acute phase of injury occurs during the initial few hours after reperfusion and is related to the production of reactive oxygen species from Kupffer cells, leading to mild hepatocellular injury.3, 4 The subacute phase of injury is initiated by inflammatory mediators, including interleukin (IL)-12/23, tumor necrosis factor alpha (TNF-α), and IL-1, released by activated Kupffer cells and hepatocytes.5-8 These mediators lead to the expression of CXC chemokines and adhesion molecules, which recruit activated neutrophils from the liver microcirculation into the parenchyma.9-11 The activated neutrophils directly injure hepatocytes and vascular endothelial cells (ECs) by releasing oxidants and proteases.3, 12
IL-33 is a newly identified member of the IL-1 family that binds to the receptor, ST2L.13 It is well known that IL-1 family cytokines have a significant effect on inflammatory, infectious, and immunological responses. Although IL-33 has structural similarities with other IL-1 family members, such as IL-1β and IL-18, which are known to contribute to the inflammatory response to liver I/R,14, 15 IL-33 is also known to induce T-helper 2 cytokine production.13 Some previous studies have demonstrated that IL-33 is detected in various organs and cell types, including ECs and epithelial cells,13, 16, 17 and has diverse effects in immune and inflammatory responses.18-22 In addition to various functions of IL-33, it has been proposed that IL-33 should be referred as an “alarmin,” a protein rapidly released from dead or dying cells in response to infection or tissue damage.23-25
IL-33 binds to ST2L and recruits myeloid differentiation primary response gene 88 and IL-1 receptor-associated kinases 1/4, leading to the activation of nuclear factor kappa light-chain enhancer of activated B cells (NF-κB) and mitogen-activated protein kinase (MAPK) pathways.24 Recent studies have shown that IL-33 is expressed not only in ECs, but also in hepatocytes during Concanavalin A (ConA)-induced acute hepatitis.26 Additional studies have demonstrated that IL-33/ST2 signaling is protective against ConA-induced hepatitis.27 In the present study, we sought to determine the role of IL-33 during acute liver inflammatory injury induced by I/R.
Materials and Methods
Male C57BL/6J mice (Jackson Laboratory, Bar Harbor, ME), weighing 20-26 g, were used in all experiments. This project was approved by the University of Cincinnati Animal Care and Use Committee and was in compliance with the National Institutes of Health guidelines. Animals underwent either sham surgery or I/R. Partial hepatic ischemia was induced, as described previously.11 Briefly, mice were anesthetized with sodium pentobarbital (60 mg/kg, intraperitoneally; IP). A midline laparotomy was performed, and an atraumatic clip was used to interrupt blood supply to the left lateral and median lobes of the liver. After 90 minutes of partial hepatic ischemia, the clip was removed to initiate hepatic reperfusion. Sham control mice underwent the same protocol without vascular occlusion. Some mice were injected IP with 5 or 10 μg/mouse of recombinant mouse IL-33 (BioLegend, San Diego, CA) or phosphate-buffered saline (PBS) 16 hours and 1 hour before ischemia. Some mice were injected IP with 100 μg/mouse of antimouse ST2/IL-1 R4 antibody (R&D Systems, Minneapolis, MN) or goat immunoglobulin G dissolved in PBS at the time of clip removal. Mice were sacrificed after the indicated periods of reperfusion, and blood and samples of the left lateral lobe were taken for analysis.
Blood and Tissue Analysis.
Blood was obtained by cardiac puncture for analysis of serum alanine aminotransferase (ALT) as an index of hepatocellular injury. Measurements of serum ALT were made using a diagnosis kit by bioassay (Wiener Laboratories, Rosario, Argentina). Serum levels of TNF-α, macrophage inflammatory protein 2 (MIP-2), and keratinocyte chemokine (KC) were assessed by enzyme-linked immunosorbent assay (ELISA; R&D Systems, Minneapolis, MN). Liver tissues were fixed in 10% neutral-buffered formalin, processed, and then embedded in paraffin for light microscopy. Sections were stained with hematoxylin and eosin for histological examination. To assess neutrophil accumulation in the liver, immunohistochemistry (IHC) for neutrophils was performed. Paraffin-embedded sections were deparaffinized, rehydrated, and immersed in citrate buffer for antigen retrieval. The sections were then incubated with antimouse Ly-6B.2 alloantigen antibody (AbD Serotec, Raleigh, NC). Liver content of IL-33 was assessed by ELISA (R&D Systems). Liver samples were weighed and immediately placed in 10 volumes (w/v) of a protease inhibitor cocktail containing 10 nmol/L of ethylene diamine tetraacetic acid (EDTA), 2 mmol/L of phenylmethylsulfonyl fluoride (PMSF), 0.1 mg/mL of soybean trypsin inhibitor (SBTI), 1.0 mg/mL of bovine serum albumin, and 0.002% sodium azide in isotonic PBS (pH 7.0). Tissues were disrupted with a tissue homogenizer, and lysates were incubated at 4°C for 2 hours. Samples were clarified by two rounds of centrifugation at 12,500×g for 10 minutes at 4°C.
Liver Neutrophil Accumulation.
Liver myeloperoxidase (MPO) content was assessed by methods described elsewhere.28 Briefly, liver tissue (100 mg) was homogenized in 2 mL of buffer A (3.4 mmol/L of KH2HPO4 and 16 mmol/L of Na2HPO4; pH 7.4). After being centrifuged for 20 minutes at 10,000×g, the pellet was resuspended in 10 volumes of buffer B (43.2 mmol/L of KH2HPO4, 6.5 mmol/L of Na2HPO4, 10 mmol/L of EDTA, and 0.5% hexadecyltrimethylammonium; pH 6.0) and sonicated for 10 seconds. After being heated for 2 hours at 60°C, the supernatant was reacted with 3,3′,3,5′-tetramethylbenzidine, and the optical density was read at 655 nm.
Western Blotting Analyses.
Liver samples were homogenized in lysis buffer (10 mM of HEPES [pH 7.9], 150 mM of NaCl, 1 mM of EDTA, 0.6% NP-40, 0.5 mM of PMSF, 1 μg/mL of leupeptin, 1 μg/mL of aprotonin, 10 μg/mL of SBTI, and 1 μg/mL of pepstatin). Samples were then sonicated and incubated for 30 minutes on ice. Cellular debris was removed by centrifugation at 10,000 rpm. Protein concentrations of each sample were determined. Samples containing equal amounts of protein in equal volumes of sample buffer were separated in a denaturing 10% polyacrylimide gel and transferred to a 0.1-μm pore nitrocellulose membrane. Nonspecific binding sites were blocked with Tris-buffered saline (TBS; 40 mM of Tris [pH 7.6] and 300 mM of NaCl) containing 5% nonfat dry milk for 1 hour at room temperature. Membranes were then incubated with antibodies to ST2 (R&D), phospho-p38 (Abcam, Inc., Cambridge, MA), p38 (Abcam), phospho-p44/42 (Cell Signaling Technology, Inc., Danvers, MA), p44/42 (Abcam), cyclin D1 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), or B-cell lymphoma 2 (Bcl-2; Abcam) in TBS with 0.1% Tween 20. Membranes were washed and incubated with secondary antibodies conjugated to horseradish peroxidase. Immunoreactive proteins were detected by enhanced chemiluminescence.
Electrophoretic Mobility Shift Assay.
Nuclear extracts of liver tissue were prepared by the method of Deryckere and Gannon29 and analyzed by electrophoretic mobility shift assay (EMSA). Briefly, double-stranded consensus oligonucleotides to NF-κB (Promega, Madison, WI) were end-labeled with gamma adenosine triphosphate (3,000 Ci/mmol at 10 mCi/mL; PerkinElmer, Waltham, MA). Binding reactions (total volume, 15 μL), containing equal amounts of nuclear protein extract (20 μg) and 35 fmols (∼50,000 cpm, Cherenkov counting) of oligonucleotide, were incubated at room temperature for 30 minutes. Binding reaction products were separated in a 4% polyacrylamide gel and analyzed by autoradiography.
Hepatocyte and Kupffer Cell Isolation.
Hepatocytes were isolated from mice by nonrecirculating collagenase perfusion through the portal vein. Livers were perfused in situ with 45 mL of Gibco Liver Perfusion Media (Invitrogen, Carlsbad, CA), followed by 45 mL of Gibco Liver Digestion Media (Invitrogen). The liver was excised, minced, and strained through a steel mesh sieve. The dispersed hepatocytes were collected by centrifugation at 50×g for 2 minutes at 4°C and washed twice with Williams' media (Invitrogen). Hepatocytes were isolated by Percoll separation and washed twice with Williams' media. The final pellet was resuspended with Williams' media. Hepatocytes were counted, and viability was checked by trypan blue exclusion. Kupffer cells were contained in the supernatants from the above-described wash. Cells were pelleted by centrifugation at 500×g for 9 minutes, resuspended in sterile Ca2+- and Mg2+-free Hank's balanced salt solution (pH 7.4), and subjected to fractionation by elutriation. Centrifugal elutriation was performed using a Beckman Coulter J20-XPI centrifuge with a JE 5.0 elutriator rotor at a constant speed of 3,200 rpm with stepwise increases in perfusion rates (Beckman Coulter, Danvers, MA). Kupffer cells were collected at the 44-mL/min fraction. The resulting cell isolates were washed, and viability was checked by trypan blue exclusion. To determine the cell production of TNF-α, MIP-2, and KC, hepatocytes or Kupffer cells were distributed onto 24-well, flat-bottomed plates (TPP, Trasadingen, Switzerland) at a concentration of 2.0 × 105 cells/500 μL/well and incubated overnight to allow cell adherence. Cells were treated with 1 μg/mL of lipopolysaccharide (LPS) or 50 ng/mL of TNF-α for 24 hours to induce cytokine and chemokine production. Cells were treated with recombinant IL-33 (rIL-33) concurrently. Culture media were collected and analyzed by an ELISA kit for TNF-α, MIP-2, and KC (R&D Systems). Hepatocyte cytotoxicity was determined by lactate dehydrogenase (LDH) assay, according to the manufacturer's instructions (Roche, Mannheim, Germany). Primary hepatocytes were distributed onto 96-well, flat-bottomed plates (TPP) at a concentration of 1.8 × 104 cells/200 μL/well and incubated overnight to allow cell adherence. Cells were treated with 0, 10, or 200 ng/mL of rIL-33, 50 ng/mL of TNF-α, and 300 or 600 μM of H2O2 for 24 hours. Culture media were collected and analyzed by an assay kit. To determine NF-κB activation and Bcl-2 expression in primary hepatocytes, cells were distributed onto 60-mm dishes at a concentration of 2 × 106 cells/5 mL/dish and incubated overnight to allow cell adherence. Cells were treated with 200 ng/mL of rIL-33 and harvested for western blotting after indicated periods of incubation.
All data are expressed as mean ± standard error of the mean (SEM). Data were analyzed with a one-way analysis of variance with a subsequent Student-Newman-Keuls test. Differences were considered significant when P < 0.05.
Hepatic IL-33 and ST2L Protein Expression During Hepatic I/R.
Previous studies have shown that IL-33 protein is expressed in various organs and cell types, including hepatocytes.13, 26 We examined the temporal expression of IL-33 protein during hepatic I/R injury. Liver samples taken from sham mice and mice after ischemia and 1, 4, 8, or 24 hours of reperfusion were processed and assessed by ELISA. We found that IL-33 protein was endogenously expressed in the liver. The expression level was increased within 4 hours after reperfusion and remained elevated for up to 24 hours (Fig. 1A).
We next examined the hepatic expression of ST2L, the transmembrane receptor for IL-33. ST2L protein was detected in the liver, and expression was increased after 1 or 4 hours of reperfusion (Fig. 1B). ST2L expression returned to baseline levels within 24 hours after reperfusion. To further examine specific liver cell expression of ST2L, we isolated hepatocytes and Kupffer cells from healthy liver and probed for ST2L. Protein expression of ST2L was strong in hepatocytes, but nearly undetectable in Kupffer cells.
IL-33 Is an Endogenous Anti-inflammatory Cytokine That Regulates Hepatic I/R Injury.
To determine whether the IL-33/ST2L system is functionally relevant to hepatic I/R injury, we injected anti-ST2L antibodies IP at the time of reperfusion. Blockade of ST2L resulted in increased neutrophil accumulation and liver injury (Fig. 2). These biochemical findings were confirmed by histological examination. After 8 hours of reperfusion, the control group and the anti-ST2L antibody treatment group showed marked necrosis and neutrophil accumulation. However, the area of necrosis and the degree of neutrophil accumulation were more significant in the antibody treatment group.
To investigate the manner in which the IL-33/ST2L system regulates hepatic I/R injury, mice were injected with rIL-33 16 hours and 1 hour before ischemia. Administration of rIL-33 dose dependently reduced liver injury after 1 and 8 hours of reperfusion (Fig. 3A). This was accompanied by decreased accumulation of neutrophils after 8 hours of reperfusion (Fig. 3B). These biochemical findings were confirmed by histological examination. After 1 or 8 hours of reperfusion, both necrotic area and accumulation of neutrophils were greatly reduced by treatment with IL-33 (Fig. 3A,B).
To determine whether the protective effects of IL-33 were the result of effects on inflammatory mediators, we assessed the expression of TNF-α and the CXC chemokines, MIP-2 and KC, which are all known as critical mediators of hepatic I/R. Treatment with IL-33 had no effect on serum levels of TNF-α (Fig. 3C). However, IL-33 dose dependently reduced serum levels of MIP-2 and KC levels after 1 and 8 hours reperfusion (Fig. 4).
IL-33 Increases the Activation of NF-κB and p38 After Hepatic I/R.
Because MAPK and NF-κB pathways have been shown to be central to injury response after hepatic I/R,30, 31 we evaluated the effects of endogenous and exogenous IL-33 on signaling pathways. Treatment with anti-ST2L antibody significantly reduced liver NF-κB activation (Fig. 4A) and p38 MAPK expression (Fig. 4B), whereas anti-ST2 antibody had no effect on p44/42 MAPK (Fig. 4C). Treatment with IL-33 significantly increased liver NF-κB activation after both 1 and 8 hours of reperfusion (Fig. 4D). In addition, though IL-33 did not change the activation of p38 MAPK after 1 hour of reperfusion, it increased p38 MAPK activation after 8 hours of reperfusion (Fig. 4E). In contrast to NF-κB and p38 MAPK, IL-33 had no effect on p44/42 MAPK at any time point (Fig. 4F).
IL-33 Increases Bcl-2 and Cyclin D1 Expression After Hepatic I/R.
p38 MAPK is known to induce cyclin D1 expression,32 one of the critical mediators in cell proliferation, and we evaluated the effect of IL-33 on cyclin D1 expression. In addition, several antiapoptotic genes, such as Bcl-2, are known to have a hepatoprotective effect in I/R injury.33-35 Because Bcl-2 expression could be induced by NF-κB,33 we also evaluated the effect of IL-33 on Bcl-2 expression. Treatment with anti-ST2L antibody significantly reduced liver cyclin D1 and Bcl-2 expression after 8 hours of reperfusion (Fig. 5A,B). Cyclin D1 expression was significantly higher in the IL-33 treatment group, compared to the control group, after 1 and 8 hours of reperfusion (Fig. 5C). There were no significant differences in Bcl-2 expression levels after 1 hour of reperfusion. However, Bcl-2 expression level was significantly higher in the IL-33 treatment group, compared to the control group, after 8 hours of reperfusion (Fig. 5D).
IL-33 Directly Protects Hepatocytes In Vitro.
We next examined the effects of IL-33 on primary hepatocytes and Kupffer cells. To assess whether IL-33 directly inhibited proinflammatory cytokine expression, we stimulated cells with LPS or TNF-α. Stimulation of Kupffer cells with 1 μg/mL of LPS resulted in robust production of TNF-α, MIP-2, and KC (Fig. 6). However, cotreatment with IL-33 had no effect on the expression of any of these mediators (Fig. 6). In hepatocytes, treatment with LPS resulted in a mild increase in TNF-α release and marked increases in MIP-2 and KC release (Fig. 7A). Similar to Kupffer cells, treatment with IL-33 had no effect on the expression of these mediators. We also assessed the effects of IL-33 on TNF-α-mediated chemokine release. IL-33 had no effect on TNF-α-induced MIP-2 or KC release from hepatocytes (Fig. 7B).
Finally, we determined whether IL-33 might have direct protective effects on hepatocytes. Primary hepatocytes were isolated from healthy mice and incubated with H2O2 and TNF-α to induce oxidative cytotoxicity. Cells were cotreated with a dose range of IL-33. Incubation of hepatocytes with 300 μM of H2O2 and 50 ng/mL of TNF-α induced moderate cytotoxicity (Fig. 7A), whereas incubation with 600 μM of H2O2 and 50 ng/mL of TNF-α induced severe cytotoxicity (Fig. 7B). Cotreatment with IL-33 dose dependently reduced hepatocyte cytotoxicity. At the 200-ng/mL dose, IL-33 reduced hepatocyte cell death by approximately 75% in the moderate model (Fig. 7A) and 35% in the severe model (Fig. 7B) of cytotoxicity. To further investigate the cell-protective mechanisms of IL-33, we assessed NF-κB activation and Bcl-2 expression in primary hepatocytes. Cells were treated with 200 ng/mL of IL-33, and nuclear extracts and cell lysates were prepared for EMSA and western blotting, respectively. Treatment with IL-33 significantly increased NF-κB activation (Fig. 8A) and Bcl-2 expression (Fig. 8B).
The present study is the first to examine the role of IL-33 in hepatic I/R injury. Our data demonstrate that IL-33 and ST2L are constitutively expressed in healthy liver and that their expression levels are increased quickly by hepatic I/R. Recently, Arshad et al. reported that IL-33 is constitutively expressed in liver vascular ECs and sinusoidal ECs. Moreover, they showed that hepatocytes could be a major source of IL-33 during acute hepatitis in mice.26 Because IL-33 has been proposed to function as an alarmin, a protein rapidly released from dead or dying cells in response to infection or tissue damage, it is plausible that IL-33 is released from injured ECs and hepatocytes during I/R. Regardless of the cellular source of IL-33 in the liver, we found that IL-33/ST2L signaling serves as an endogenous regulatory mechanism. Blockade of ST2L increased liver injury, whereas exogenous administration of IL-33 reduced liver injury.
The role of IL-33 in inflammatory diseases is still not clear. There are conflicting reports of pro- and anti-inflammatory effects of IL-33 in multiple models of inflammatory injury.18-22 In the current study, we found that IL-33 reduced liver inflammation. However, it is entirely possible, and our data support the concept, that the protective effects of IL-33 during liver I/R injury may be unrelated to inflammation. We found no direct effects of IL-33 on proinflammatory cytokine or chemokine production by Kupffer cells or hepatocytes in vitro. At the same time, we found marked protection of hepatocytes from cytokine/oxidant-induced cell death. These data suggest that IL-33 may have direct hepatoprotective effects on hepatocytes that might limit the inflammatory response by reducing the number of necrotic cells needing to be cleared by phagocytes. This would explain why we observed no effect of IL-33 on cytokine and chemokine production in vitro, yet observed decreased chemokine expression in vivo.
Alternatively, a recent study showed that IL-33 down-regulates IL-17 production during ConA-induced hepatitis.27 Because we have previously shown that IL-17 regulates chemokine expression during hepatic I/R injury,36 it is possible that the reduced chemokine levels in mice treated with IL-33 occurred through reduced IL-17 expression. Another alternative explanation may be that IL-33 has effects on other liver cell types, such as ECs and/or stellate cells, that down-regulate their production of chemokines.
Previous studies from our lab and others have suggested that NF-κB plays a cell-protective role in hepatocytes during I/R injury, whereas NF-κB activation in Kupffer cells is harmful.23, 37-41 Our current data show that IL-33 administration increases hepatic NF-κB activation during I/R and significantly suppresses I/R injury. Because we showed that ST2L expression is highly expressed in hepatocytes and not Kupffer cells, it seems that IL-33 binding to ST2L on hepatocytes increases NF-κB activation in hepatocytes, which contributes to hepatoprotection against I/R injury. We also found that IL-33 increases phospho-p38 MAPK, but not p44/42 MAPK. Interestingly, Teoh et al. showed that the protective effects of ischemic preconditioning against I/R injury are associated with increased NF-κB activation and p38 MAPK expression.32 These changes then induce cyclin D1 expression, which is associated with hepatocyte proliferation.32 Consistent with these findings, our data showed that IL-33 treatment increases cyclin D1 expression during I/R. Besides cyclin D1, our data also show that IL-33 increases Bcl-2 expression during I/R. Bcl-2 is an antiapoptotic gene known to be regulated by NF-κB and p38 MAPK.33-35, 42 It has been demonstrated that adenovirus-mediated transfer of Bcl-2 reduced liver injury after I/R.35 Therefore, it appears that IL-33 administration contributes to protect against I/R injury by activation of NF-κB and increased phospho-p38 MAPK, which ultimately leads to the induction of key mediators of both cell proliferation and antiapoptotic effects, such as cyclin D1 and Bcl-2.
In conclusion, the current data demonstrate that IL-33 is an important endogenous regulator of the acute inflammatory response induced by I/R in the liver. Based on our data, we suggest that IL-33 regulates inflammation indirectly as a side effect of its direct cell-protective effects on hepatocytes. Furthermore, it appears that the protective effects of IL-33 are associated with the activation of NF-κB and p38 MAPK, cyclin D1, and Bcl-2. The precise mechanism(s) by which IL-33 activates these signaling pathways remains to be elucidated, but the IL-33/ST2L system may be a useful therapeutic target after liver surgery or transplantation.