Interleukin 6 alleviates hepatic steatosis and ischemia/reperfusion injury in mice with fatty liver disease

Authors

  • Feng Hong,

    1. Section on Liver Biology, Laboratory of Physiologic Studies, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, Bethesda, MD
    Search for more papers by this author
  • Svetlana Radaeva,

    1. Section on Liver Biology, Laboratory of Physiologic Studies, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, Bethesda, MD
    Search for more papers by this author
  • Hong-na Pan,

    1. Section on Liver Biology, Laboratory of Physiologic Studies, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, Bethesda, MD
    Search for more papers by this author
  • Zhigang Tian,

    1. Institute of Immunology, School of Life Science, University of Science and Technology of China, Hefei, Anhui, China
    Search for more papers by this author
  • Richard Veech,

    1. Laboratory of Metabolism and Molecular Biology, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, Bethesda, MD
    Search for more papers by this author
  • Bin Gao

    Corresponding author
    1. Section on Liver Biology, Laboratory of Physiologic Studies, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, Bethesda, MD
    • Section on Liver Biology, NIAAA/NIH, Park Bldg. Rm. 120, 12420 Parklawn Drive, MSC 8115, Bethesda, MD 20892
    Search for more papers by this author

Abstract

Fatty liver, formerly associated predominantly with excessive alcohol intake, is now also recognized as a complication of obesity and an important precursor state to more severe forms of liver pathology including ischemia/reperfusion injury. No standard protocol for treating fatty liver exists at this time. We therefore examined the effects of 10 days of interleukin 6 (IL-6) injection in 3 murine models of fatty liver: leptin deficient ob/ob mice, ethanol-fed mice, and mice fed a high-fat diet. In all 3 models, IL-6 injection decreased steatosis and normalized serum aminotransferase. The beneficial effects of IL-6 treatment in vivo resulted in part from an increase in mitochondrial β oxidation of fatty acid and an increase in hepatic export of triglyceride and cholesterol. However, administration of IL-6 to isolated cultured steatotic hepatocytes failed to decrease lipid contents, suggesting that the beneficial effects of IL-6 in vivo do not result from its effects on hepatocytes alone. IL-6 treatment increased hepatic peroxisome proliferator-activated receptor (PPAR) α and decreased liver and serum tumor necrosis factor (TNF) α. Finally, 10 days of treatment with IL-6 prevented the susceptibility of fatty livers to warm ischemia/reperfusion injury. In conclusion, long-term IL-6 administration ameliorates fatty livers and protects against warm ischemia/reperfusion fatty liver injury, suggesting the therapeutic potential of IL-6 in treating human fatty liver disease. Supplementary material for this article can be found on the Hepatology website (http://interscience.wiley.com/jpages/0270-9139/suppmat/index.html). (HEPATOLOGY 2004;40:933–941.)

The prevalence of fatty liver disease, primarily caused by obesity and alcohol consumption, is approximately 20% in the general population of developed countries and is expected to increase further as a result of the alarming overall increase of obesity in the general population.1 In the past, fatty liver disease was considered a benign condition; however, mounting evidence indicates that fatty livers increase the risk of developing progressive liver injury, such as nonalcoholic steatohepatitis, fibrosis, and cirrhosis.1–4 Moreover, hepatic steatosis is also a frequent histopathological feature of chronic hepatitis C and accelerates the progression of liver damage in these patients.5 Additionally, fatty livers tolerate poorly cold and warm ischemia/reperfusion injury and are associated with high mortality after major surgery and transplantation.6, 7

Interleukin 6 (IL-6) is elevated in the plasma and peripheral blood monocytes of patients with fatty diseases, including alcoholic liver disease and nonalcoholic steatohepatitis, and elevation of IL-6 correlates with the progression and severity of liver disease,8–11 suggesting that IL-6 may be involved in the pathogenesis of fatty liver disease. However, increasing evidence indicates that IL-6 is an important hepatoprotective cytokine for promoting liver regeneration12–14 and protecting against liver injury caused by various insults in lean animals.15–22 Although the protective effect of IL-6 in lean livers has been well documented, the effects of IL-6 on fatty liver disease remain unclear. Selzner and Clavien23 reported that treatment with IL-6 normalized proliferating-cell nuclear antigen expression in steatotic hepatocytes but failed to increase DNA synthesis and mitosis in steatotic hepatocytes in Zucker obese rats after partial hepatectomy, indicating that the mitogenic role of IL-6 is partially attenuated in Zucker rats. However, Torbenson et al.24 reported that activation of signal transducer and activator of transcription factor 3 (STAT3), a downstream IL-6 signal, was significantly higher in steatotic livers of ob/ob mice after partial hepatectomy, suggesting that IL-6 signaling is highly induced in ob/ob murine fatty livers after partial hepatectomy. Previously, we showed that in vitro IL-6 treatment prevents mortality associated with fatty liver transplants in rats by protecting against endothelial cell necrapoptosis and consequent improvement in microcirculation.25, 26 In this study, we demonstrate that long-term—but not short-term—IL-6 treatment significantly alleviates steatosis and ischemia/reperfusion injury in obesity- and ethanol-associated fatty livers. Furthermore, we show that tumor necrosis factor (TNF)-α down-regulation and peroxisome proliferator-activated receptor (PPAR)-α up-regulation may be mechanisms partially contributing to the amelioration of fatty livers by IL-6.

Abbreviations:

IL-6, interleukin 6; STAT, signal transducer and activator of transcription factor; TNF, tumor necrosis factor; PPAR, peroxisome proliferator-activated receptor; SREBP, sterol regulatory element-binding protein; HE, hematoxylin-eosin; alanine aminotransferase; RT-PCR, reverse-transcriptase polymerase chain reaction; bp, base pair; mRNA, messenger RNA; AST, aspartate aminotransferase; MCD, methionine and choline-deficient; TUNEL, terminal deoxynucleotidyl transferase-mediated nick end labeling.

Materials and Methods

Materials.

Recombinant human IL-6 was produced by recombinant DNA technology as described previously.25 Antibodies against STATs were obtained from Cell Signaling (Beverly, MA). Mcl-1, Bcl-2, and sterol regulatory element-binding protein (SREBP)-1 antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA) and Bcl-xL antibody was obtained from BD PharMingen (San Diego, CA). PPAR-α and PPAR-γ antibodies were purchased from Research Diagnostics Inc. (Flanders, NJ).

Murine Models of Fatty Livers.

Three murine models of fatty livers were used: leptin-deficient ob/ob mice, ethanol-fed mice, and high fat-fed mice. Male ob/ob mice, 8 to 10 weeks old, obtained from Jackson Laboratory (Bar Harbor, ME) were obese with severe steatosis of the liver. For the high fat-induced fatty liver model, 8- to 10-week old male C57BL/6 mice were fed a high-fat diet (Catalogue no. 88137; Harlan Teklad, Madison, WI) for 10 weeks and developed severe steatosis of the liver. For the ethanol-induced fatty liver model, 7- to 8-week old male C57BL/6NCR mice (National Cancer Institute [NCI], National Institutes of Health [NIH], Bethesda, MD) were fed an ethanol-containing liquid Lieber-DeCarli diet or control diet whereby ethanol was substituted isocalorically with dextrin maltose (BioServ Inc., Frenchtown, NJ) for 8 weeks. After 8 weeks of feeding, the mice developed significant steatosis of the liver.

Hepatic Triglyceride Secretion.

The nonionic detergent Triton WR-1339 (Sigma, St. Louis, MO) was used to measure hepatic triglyceride secretion as described previously.27 Lipoproteins in the plasma are trapped by Triton WR-1339, which allowed determination of the secretion rate of hepatic triglyceride lipoproteins. Mice were injected with either IL-6 or saline and 12 mg Triton WR-1339 in 0.9% saline. One and 2 hours later, blood was collected, and levels of triglyceride were measured.

Hematoxylin-Eosin (HE) and Oil Red O Staining of Liver Sections.

Following fixation of the livers with 10% formalin/phosphate-buffered saline, livers were sliced and stained with HE for histological examination. Liver steatosis was graded semiquantitatively based on the percentage of hepatocytes according to the following criteria: grade 0, no hepatocytes involved; grade 1, 1% to 25% of hepatocytes involved; grade 2, 26% to 50% of hepatocytes involved; grade 3, 51% to 75% of hepatocytes involved; and grade 4, 76% to 100% of hepatocytes involved. Hepatic lipid content was also determined by staining with Oil Red O (Sigma).

Determination of Liver Injury.

Liver injury caused by ischemia/reperfusion was quantified by measuring plasma enzyme activities of alanine aminotransferase (ALT) using a kit from DREW Scientific (Cumbria, UK), HE staining, and terminal deoxynucleotidyl transferase-mediated nick end labeling (TUNEL) assay. Hepatocellular necrosis was determined in HE-stained sections using a semiquantitative scale by a point counting method as previously described.16, 28 An ordinal scale was used for determination of hepatocellular injury as follows: grade 0, minimal or no evidence of injury; grade 1, mild injury consisting of cytoplasmic vacuolation and focal nuclear pycknosis; grade 2, moderate to severe injury with extensive nuclear pycknosis, cytoplasmic hypereosinophilia, and loss of intercellular borders; and grade 3, severe necrosis with disintegration of hepatic cords, hemorrhage, and neutrophil infiltration. Thirty random sections were examined per slide to determine the percentage of necrotic cells. Apoptotic cells in sections were determined by TUNEL staining using ApopTag (Oncor, Gaithersburg, MD) as described previously.20

Biological Analysis of Hepatic Lipids.

To measure hepatic lipids, livers were homogenized at 4°C in lysis buffer containing 50 mmol/L Tris (pH 8.0), 150 mmol/L NaCl, 1% Triton X-100, and 0.5% Na DOC. Lipids from the total liver homogenate were extracted using the chloroform/methanol method (2:1), evaporated, and dissolved in 2-propanol. Amounts of triglyceride, total cholesterol, and phospholipids were assayed enzymatically using kits obtained from Wako Pure Chemicals Co. (Richmond, VA).

Primary Mouse Hepatocyte Isolation and Culture.

Hepatocytes were isolated and cultured as described previously.29

Fatty Acid β-Oxidation Activity.

Fresh livers were homogenized in 4 volumes of 0.25 mol/L sucrose containing 1 mmol/L ethylenediaminetetraacetic acid in a homogenizer using a tight-fitting Teflon pestle. Fatty acid β-oxidation activity was measured as described previously30 and expressed as nmol/min/liver.

Reverse-Transcriptase Polymerase Chain Reaction (RT-PCR).

RT-PCR was performed as described previously.29 The primer sequences for TNF-α were forward primer, 5′CCA CAT CTC CCT CCA GAA AA-3′ and reverse primer, 5′AGG GTC TGG GCC ATA GAA CT-3′. The expected PCR product was 258 base pair (bp). PCR using RNA without reverse transcription did not yield amplicons, indicating a lack of genomic DNA contamination. The PCR bands were scanned using Storm PhosphoImager (Molecular Dynamics, Sunnyvale, CA).

Gel Mobility Shift Assay.

Gel mobility shift assay was performed as described previously.29 The double-stranded oligo containing the PPAR response element 5′-GAACTAGGTCAAAGGTCATCCCCT-3′ was used as the probe.

Murine Model of Partial Ischemia/Reperfusion Liver Injury.

Adult male mice were anesthetized by isoflurane inhalation, followed by a midline laparotomy. All structures in the portal triad to the left and median liver lobes were occluded with a vascular clamp to induce partial ischemia (70%). After 50 minutes, the vascular clamp was removed; animals were sacrificed 6 hours later, and the serum was collected to measure ALT levels.

Murine Model of Total Liver Ischemia/Reperfusion.

To examine the effects of IL-6 treatment on the survival after ischemia/reperfusion, the murine model of total hepatic ischemia was used as described previously.31 Briefly, after the midline laparotomy, a partial (30%) hepatectomy was performed with resection of the caudate, right lateral, and quadrate lobes, and papillary process. Total hepatic ischemia was then achieved by placing the vascular microclamps across the pedicles of the median and left lateral lobes at the level of the hilum. After 60 minutes of ischemia, the clip was removed, and the abdominal cavity was closed. Mice were then observed daily until postsurgical day 7 to determine survival.

Western Blot Analysis.

Western blot analysis was performed as described previously.29

Statistical Analysis.

For comparing values obtained in 3 or more groups, 1-way ANOVA was used, followed by post hoc Tukey test. Statistical significance was taken at the P less than .05 level.

Results

Long-Term IL-6 Treatment Ameliorates Fatty Livers in Ob/Ob Mice.

Subcutaneous injection of IL-6 for 10 days did not significantly reduce body weight (Fig. 1A) or food intake (data not shown). In contrast, liver weight and liver-to-body weight ratios were reduced in the IL-6–treated group compared to the saline treatment group (P < .05; Fig. 1B), whereas the epididymal and subcutaneous fat pad-to-body weight ratios remained unchanged (Fig. 1C). Initially, the livers of ob/ob mice were pale in appearance, but they became a homogenous red after long-term IL-6 treatment, indicating that steatosis was alleviated in IL-6-treated ob/ob mice (Fig. 1D).

Figure 1.

IL-6 treatment reduces liver weight without affecting body weight. Eight-week-old ob/ob mice were treated daily with subcutaneous injection of 1 μg/g of IL-6 (n = 10) or saline (n = 10) for 10 days. (A) Mice were weighed daily. (B) On day 11, mice were sacrificed, liver weight was measured, and liver/body weight ratios were calculated. (C) The ratios of epididymal (Ep) or subcutaneous (Sub, fat pad-to-body weight were also calculated. (D) A representative picture of saline and IL-6 treated ob/ob mouse livers is shown. Note the pale appearance of steatotic livers of saline-treated ob/ob mice and the red appearance of IL-6-treated ob/ob mouse livers. *P < .05 in comparison with ob/ob group with saline.

Histological analysis of the livers with HE staining showed that steatosis was significantly improved 5 days after treatment with IL-6 (1 μg/g) and almost completely recovered 10 days and 20 days after treatment. As shown in Fig. 2A, lipid droplets accumulated in nearly 100% of hepatocytes in both periportal and pericentral areas in untreated ob/ob mice. In comparison, lipid droplet accumulations in hepatocytes were significantly decreased after 5 days of IL-6 injections, particularly in the periportal areas. After longer treatment with IL-6 (10-20 days), lipid droplets were diminished in both periportal and perivenular areas (Figs. 2A and B). Treatment with higher doses of IL-6 (2-5 μg/g) for 5 days also significantly alleviated steatosis in the livers of ob/ob mice (Fig. 2A), but these doses caused significant reductions in food intake and body weight gain (data not shown). As added confirmation, Oil Red O staining showed that large lipid droplets observed in the livers of ob/ob mice were markedly diminished in the 10-day IL-6 treatment group (Fig. 2C), and quantitation of lipid contents revealed that IL-6 treatment reduced the levels of hepatic triglyceride (Fig. 2D). Finally, treatment of ob/ob mice with IL-6 for 10 days reduced the serum levels of ALT (Fig. 2E).

Figure 2.

IL-6 treatment ameliorates fatty livers, improves liver histology, and normalizes serum aminotransferase activity in ob/ob mice. (A) Eight-week-old ob/ob mice were administered IL-6 (1 μg/g, subcutaneous) daily for various numbers of days (upperpanel) or with various doses (lowerpanel) of IL-6 daily for 5 days. Mice were then sacrificed, (A) their livers were stained with HE, and (B) hepatic steatosis was scored. (C-E) Eight-week-old ob/ob mice were administered IL-6 (1 μg/g, subcutaneous) daily for 10 days. The mice were then killed, (C) their livers were stained with Oil Red O, (D) hepatic triglyceride levels were measured, and (E) serum ALT levels were measured. Values shown in panels B, D, and E represent means ± SEM from 5 to 10 mice in each group. **P < .01. ***P < .001 in comparison with ob/ob group treated with saline.

Long-Term IL-6 Treatment Reverses Fatty Livers in Mice Fed a High Fat Diet and Ethanol-Fed Mice.

Next, we examined whether IL-6 treatment also alleviated steatosis in mice fed a high-fat diet and ethanol-fed mice. Mice fed a high-fat diet for 10 weeks developed severe steatosis in the liver, which was significantly attenuated after long-term IL-6 treatment, as demonstrated by Oil Red O staining. As shown in Supplementary Fig. 1A, Oil Red O staining showed that large lipid droplets in the livers of mice fed a high-fat diet were diminished after 10 days of IL-6 treatment. Quantitation of the lipid contents revealed that IL-6 treatment significantly reduced hepatic levels of triglyceride in mice fed a high-fat diet (Supplementary Fig. 1B). Similarly, mice fed a liquid diet containing 5% ethanol for 8 weeks developed significant steatosis, which was significantly improved after 5 days of IL-6 treatment, as demonstrated by Oil Red O staining (Supplementary Fig. 1C) and quantitation of hepatic levels of triglyceride (Supplementary Fig. 1D).

IL-6 Does Not Alleviate Steatosis in Cultured Steatotic Hepatocytes.

To understand the molecular mechanism underlying the effects of IL-6 treatment on fatty liver in vivo, we examined the effects of IL-6 on steatosis in cultured steatotic hepatocytes. Oil Red O staining revealed that lipid droplets in steatotic hepatocytes were not affected in the absence or presence of IL-6 (500 ng/mL) after 5 days in culture (Supplementary Fig. 2A). Quantitation of lipid contents revealed that triglyceride levels in hepatocytes were not reduced after IL-6 treatment (Supplementary Fig. 2B), and the levels of triglyceride in the supernatant were similar in both IL-6–treated and -untreated cells (Supplementary Fig. 2C). We also tested the effects of various concentrations of IL-6 (50-2,000 ng/mL) on hepatic steatosis in cultured hepatocytes, and no significant effects were observed (data not shown).

IL-6 Treatment Stimulates Hepatic Triglyceride Secretion and Hepatic Fatty Acid β Oxidation.

The effects of IL-6 treatment on serum lipids were examined. As shown in Fig. 3, IL-6 administration caused rapid elevations in serum triglyceride and cholesterol levels in both ob/ob mice and lean C57BL/6 mice with peak effect at 12 hours (Figs. 3A and B). Serum triglyceride and cholesterol levels were also elevated after long-term IL-6 treatment (Fig. 3C). We next determined whether the IL-6–induced elevation in serum lipid contents was due to stimulation of hepatic lipoprotein secretion. Hepatic secretion of lipoproteins was determined using the Triton 1339 technique. As shown in Fig. 3D, triglyceride secretion increased 2.1-fold in IL-6–treated animals compared to saline-treated mice (P < .01). Finally, long-term IL-6 treatment slightly—but significantly—stimulated palmitic acid β oxidation in the liver mitochondria (Fig. 3E; P < .05). The same treatment did not affect peroxisomal oxidation of lignoceric acid (Fig. 3E).

Figure 3.

IL-6 treatment stimulates hepatic triglyceride secretion and hepatic fatty acid β oxidation. (A and B) C57BL/6 mice or ob/ob mice were administered a single dose of IL-6 (1 μg/g, subcutaneous). After various time periods, mice were sacrificed, and serum levels of (A) triglyceride and (B) cholesterol were measured. (C) Ob/ob mice were administered IL-6 (1 μg/g, subcutaneous) daily for 10 days. Mice were then sacrificed, and hepatic lipid contents were measured. (D) Ob/ob mice were treated with saline or IL-6 (1 μg/g) and Triton WR-1339 (12 mg). One and 2 hours later, blood was collected and levels of triglyceride were determined. The difference (Δ triglyceride) in serum triglyceride levels between 1 and 2 hours is equivalent to the rate of secretion of triglyceride-containing lipoproteins. (E) Ob/ob mice were administered IL-6 (1 μg/g, subcutaneous) daily for 10 days. Mice were then sacrificed, and livers were collected for determination of lignoceric acid oxidation or palmitic acid oxidation. Values shown represent means ± SEM from 3 mice (A and B), 10 mice (C), 4 mice (D), and 10 mice (E) in each group. *P < .05. **P < .01 in comparison with saline-treated group.

IL-6 Down-regulates Hepatic and Serum Levels of TNF-α.

TNF-α has been implicated in the development of fatty livers,32, 33 and IL-6 treatment has been shown to down-regulate TNF-α in several models of liver injury.16, 34 Therefore, we sought to examine whether IL-6 amelioration of fatty livers was mediated by down-regulation of TNF-α. As shown in Fig. 4A, TNF-α messenger RNA (mRNA) expression was significantly greater in the livers of ob/ob mice compared to wild type C57BL/6 mice. Long-term IL-6 treatment, however, decreased TNF-α mRNA expression in the livers of ob/ob mice (P < .001; Fig. 4A). Moreover, treatment with IL-6 for 10 days down-regulated serum TNF-α levels in ob/ob mice and mice fed a high-fat diet (Fig. 4B).

Figure 4.

IL-6 down-regulates hepatic and serum levels of TNF-α. (A) Ob/ob mice were administered IL-6 (1 μg/g, subcutaneous) or saline daily for 10 days. Total liver RNA from IL-6–treated ob/ob mice, saline-treated mice, or C57BL/6 mice were subjected to RT-PCR analysis of TNF-α and β-actin. (A) Relative density of TNF-α to β-actin was quantitated. (B) Ob/ob mice or mice fed a high-fat diet were administered IL-6 (1 μg/g, subcutaneous) or saline daily for 10 days. Serum TNF-α levels were measured. Values shown are means ± SEM from 4 mice (A) and 10 mice (B) in each group. *P < .05. ***P < .001 in comparison with saline-treated ob/ob group.

Long-Term IL-6 Treatment Up-regulates PPAR-α Protein Expression and DNA Binding Activity of PPAR.

To further understand the molecular mechanisms underlying IL-6 amelioration of fatty liver, expression of PPAR, which increases fatty-acid oxidation,35–37 and protein expression of SREBP, which increases fatty acid and cholesterol synthesis, were examined. As shown in Fig. 5A, expression of PPAR-α protein was elevated in the livers of IL-6–treated ob/ob mice, whereas expression of PPAR-γ and SREBP-1 was not affected by IL-6 treatment. Furthermore, results of the gel mobility shift assay showed that DNA binding of the PPAR/PXR heterodimer in hepatic nuclear extracts from IL-6–treated ob/ob mice was significantly higher than that from nontreated ob/ob mice (Fig. 5B). Incubation with PPAR-α antibody significantly reduced DNA binding of PPAR/PXR in hepatic nuclear extracts from IL-6–treated ob/ob mice, suggesting that IL-6 treatment enhances the DNA binding of PPAR-α/PXR. Conversely, treatment of cultured hepatocytes with various concentrations of IL-6 (100-500 ng/mL) did not affect the expression and DNA binding of PPAR-α protein (data not shown). Taken together, our findings indicate that in vivo treatment with IL-6 for 10 days increases the expression and DNA binding of PPAR-α in the liver.

Figure 5.

IL-6 treatment up-regulates PPAR-α protein expression and DNA binding activity of PPAR/PXR. Ob/ob mice were administered IL-6 (1 μg/g, subcutaneous) or saline daily for 10 days. On day 11, mice were sacrificed and protein extracts were prepared for (A) Western blot analysis of PPAR-α, PPAR-γ, or SREBP-1 or (B) gel mobility shift analysis of DNA binding of PPAR. (B) Anti–PPAR-α antibody (Ab) was included in the left-most lane for supershift analysis. Relative density of (A) PPAR-α to β-actin and (B) fold induction of PPAR/PXR binding in panel B were quantified and shown (left). The PPAR binding in the first lane (sample 1) was defined as 1. Values shown represent means ± SEM from 6 mice in each group. **P < .01 in comparison with saline-treated ob/ob group.

Long-Term, but Not Short-Term IL-6 Treatment Protects Steatotic Livers From Warm Ischemia/Reperfusion Injury.

Camargo et al.16 previously reported that IL-6 protected against warm ischemia/reperfusion injury in lean animals. However, the effects of IL-6 on warm ischemia/reperfusion injury in fatty livers remain unclear. As shown in Fig. 6A, ischemia for 50 minutes and reperfusion for 6 hours caused only a slight elevation in serum aspartate aminotransferase (AST) levels in lean animals, whereas the same treatment caused a dramatic elevation in serum AST levels in ob/ob mice. Pretreatment with IL-6 for 10 days but not 1 day protected fatty livers from warm ischemia/reperfusion injury in ob/ob mice (7,736 ± 1851 U/L in IL-6–treated ob/ob mice vs. 24,493 ± 3033 U/L in ob/ob mice; P < .001). Additionally, treatment with IL-6 for 10 days also protected fatty livers from ischemia/reperfusion injury in mice fed high-fat diets (Fig. 6B).

Figure 6.

Long-term, but not short-term IL-6 treatment protects steatotic livers from warm ischemia/reperfusion injury. (A) Ob/ob mice were treated with IL-6 (1 μg/g, subcutaneous) or saline for 1 day or daily for 10 days. (B) Mice fed a high-fat diet were administered IL-6 (1 μg/g, subcutaneous) or saline daily for 10 days. (A and B) Mice were subjected to a 50-minute ischemia/6-hour reperfusion treatment as described in Materials and Methods. Serum AST levels were measured. (C and D) After a 50-minute ischemia/6-hour reperfusion treatment, mouse livers from panel A were subjected to (C) HE staining for determination of necrosis (only grade 3 injury was counted as necrosis) or (D) to TUNEL staining for determination of apoptosis (A-D) Values shown represent means ± SEM from 7 mice in each group. *P < .05; **P < .01; ***P < .001; in comparison with the corresponding saline-treated group (white bar). (E) Effect of IL-6 treatment on survival after 30% partial hepatectomy plus 60 minutes of ischemia. Survival for 7 days was 100%, 10%, and 90% for lean male C57BL/6 mice, male ob/ob mice, and IL-6–treated male ob/ob mice, respectively.

Next, apoptotic and necrotic cell death were examined in IL-6–treated and -untreated ob/ob mouse livers. As shown in Fig. 6C, less than 10% of necrotic cells and less than 5% TUNEL-positive hepatocytes were detected in the livers of lean control mice after 50 minutes ischemia/6 hours reperfusion, while the same treatment caused necrosis and apoptosis of 80% and 5%, respectively, of hepatocytes in the livers of ob/ob mice. Both necrosis and apoptosis were reduced in the livers of ob/ob mice treated with IL-6 for 10 days but not in those treated for 1 day.

The effects of IL-6 treatment on ob/ob mouse survival after ischemia/reperfusion were also studied by using a model of total hepatic ischemia as described previously.31 As shown in Fig. 6E, all wild type C57BL/6 mice survived 60 minutes of total ischemia, but 9 of 10 ob/ob mice died within 2 days after 60 minutes total ischemia. Conversely, 9 of 10 IL-6–treated ob/ob mice survived 60 minutes of total ischemia.

IL-6 Induces Weaker Hepatic STAT3 Activation in Ob/Ob Mice Compared to Lean Mice.

IL-6 activation of antiapoptotic signals and genes have been well documented in lean animals15–22; these signals and genes appear to play important roles in the protective effects of IL-6 in various forms of liver injury. In the present study, we examined IL-6 activation of antiapoptotic STAT3 signals in fatty livers. As shown in Supplementary Fig. 3A, IL-6 injection induced STAT3 phosphorylation in the livers of ob/ob mice and lean C57BL/B6 mice, with peak effect at 1 hour and returning to basal levels at 8 hours after injection. IL-6 activation of hepatic STAT3 in ob/ob mice was lower than in lean C57BL/6 mice (Supplementary Fig. 3A). Western blot analysis showed that levels of Bcl-2 and Bcl-xL were elevated in the livers of IL-6–treated ob/ob mice compared to saline-treated ob/ob mice (Supplementary Fig. 3B).

Discussion

In the present study, we demonstrate that in addition to its well-documented antiapoptotic functions in the liver, IL-6 also ameliorates fatty liver disease. This amelioration is not due to reduced food intake and body weight, since subcutaneous injection of IL-6 (1 μg/g) for 10 days reduced liver weight and liver-to-body weight ratios but did not affect food intake, body fat pads, and body weight (Fig. 1). In contrast, chronic intracerebroventricular injection of IL-6 was shown to decrease body weight and body fat pads but did not affect liver weight.38 Taken together, these findings suggest that IL-6 amelioration of fatty livers is mediated by a peripheral effect, likely through stimulation of hepatic triglyceride secretion and hepatic fatty acid β oxidation (Fig. 3). In Fig. 3, we clearly show that long-term and short-term IL-6 treatments elevate serum levels of triglyceride and cholesterol in both lean and ob/ob mice; this is likely due to IL-6 stimulation of hepatic triglyceride secretion (Fig. 3D). At present, the molecular mechanisms for IL-6 stimulation of triglyceride secretion remain unclear. Similarly, stimulation in rats of hepatic triglyceride secretion by IL-6 was found to be independent of endogenous catecholamines.27 Treatment with IL-6 does not stimulate triglyceride secretion in cultured steatotic hepatocytes (Supplementary Fig. 2), suggesting that the stimulatory effect of IL-6 on hepatic triglyceride in vivo is mediated by an indirect mechanism.

Additional experiments suggest that at least 2 mechanisms may be involved in IL-6 amelioration of fatty liver disease. First, chronic IL-6 treatment down-regulates hepatic TNF-α expression and serum TNF-α levels. Increasing evidence suggests that TNF-α plays a critical role in the development of steatosis in obesity- and ethanol-induced fatty liver.32, 33 For example, blocking TNF-α signaling with a probiotic (modifying the intestinal flora) or anti–TNF-α antibodies alleviates fatty livers in ob/ob mice,32 and deletion of the TNF-α gene abolishes development of ethanol-induced fatty livers.33 In the present study, we show that treatment with IL-6 for 10 days down-regulates expression of hepatic TNF-α mRNA and serum TNF-α levels in ob/ob mice (Fig. 4). The inhibitory effect of IL-6 on TNF-α and soluble TNF receptor p55 production also has been reported in various models of liver injury.16, 34, 39 Taken together, IL-6 treatment in vivo alleviates hepatic steatosis, at least in part, by inhibition of TNF-α activity. The second potential mechanism by which IL-6 ameliorates fatty liver may be through up-regulation of PPAR-α. The inhibitory effect of PPAR-α in the development and progression of fatty liver is well documented. Mice deficient in PPAR-α develop severe hepatic steatosis after fasting or feeding on a high-fat diet or a methionine and choline-deficient (MCD) diet,35–37 but activation of PPAR-α by the agonist Wy14,643 ameliorated alcoholic fatty liver30 and MCD-induced steatohepatitis.37, 40 The critical role of PPAR-α in ameliorating steatosis is mediated through regulation of a wide variety of genes involved in peroxisomal, mitochondrial, and microsomal fatty acid oxidation systems in the liver.41

In this study, we demonstrate that treatment of ob/ob mice with IL-6 for 10 days enhances PPAR-α protein expression and DNA binding activity of PPAR-α/PXR in the liver (Fig. 5), and increases palmitic acid oxidation in the liver, suggesting that up-regulation of PPAR-α may be an important mechanism in the IL-6 amelioration of hepatic steatosis.

Another important finding is that IL-6 treatment protects steatotic livers from ischemia/reperfusion injury. High mortality (14%) is associated with fatty livers after major surgery compared to 2% in normal livers6, 7 and there is currently no known treatment to reduce such high mortality. In rodents, fatty livers are more susceptible to warm or cold ischemia/reperfusion injury6, 42, 43; it was also shown in the present study that warm ischemia/reperfusion induced much higher levels of ALT and mortality in ob/ob mice compared to lean control mice (Fig. 6). Moreover, we demonstrate that necrosis is the major form of cell death in the livers of ob/ob mice after ischemia/reperfusion (Fig. 6); this is consistent with previous studies.42–45 Long-term IL-6 treatment markedly prevents ischemia/reperfusion-induced liver injury in ob/ob mice and mice fed high-fat diets (Fig. 6). This hepatoprotective effect likely results from a decrease of hepatic steatosis and TNF-α and activation of the antiapoptotic STAT3 signal after IL-6 treatment. First, steatosis has been recognized as an important risk factor for hepatic surgery and ischemia/reperfusion injury.6, 43, 45 Thus, a decrease of hepatic lipid content after long-term IL-6 treatment could be an important mechanism in IL-6 protection against hepatic ischemia/reperfusion injury in ob/ob mice. Second, TNF-α has been shown to be an essential mediator not only for the development of hepatic ischemia/reperfusion injury but also for the hepatoprotective effects of ischemic preconditioning.46–49 Down-regulation of TNF-α has been implicated in the hepatoprotective effect of IL-6 in hepatic ischemia/reperfusion injury in lean animals.16 Therefore, it is plausible that inhibition of TNF-α also contributes partly to IL-6 amelioration of hepatic ischemia/reperfusion in ob/ob mice in addition to IL-6 alleviation of hepatic steatosis. Finally, the hepatoprotective role of STAT3 has been well documented in a variety of models of liver injury by means of activation of antiapoptotic proteins (such as Bcl-2 and Bcl-xL),15–22 and lean transgenic mice overexpressing Bcl-2 are resistant to hepatic ischemia and reperfusion.50 In the present study, we showed that IL-6 treatment also activated STAT3, Bcl-2, and Bcl-xL in the steatotic livers of ob/ob mice (Supplementary Fig. 3), suggesting that activation of these antiapoptotic signals could be an important mechanism in the hepatoprotective effect of IL-6 in ischemia/reperfusion injury of steatotic livers.

In summary, our results show that treatment with IL-6 has beneficial effects on fatty livers, including the alleviation of steatosis, improvement of liver history, normalization of serum aminotransferase activity, and prevention of ischemia/reperfusion injury. In addition to the previously identified therapeutic potential of IL-6 in preventing fatty liver transplant failure25, 26 and ischemia/reperfusion injury of lean livers,15, 16 these findings suggest another clinical application of IL-6 in protecting fatty livers from warm ischemia/reperfusion injury and ameliorating fatty liver disease.

Ancillary