SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Adiponectin is an adipocyte-derived hormone with a wide range of beneficial effects on obesity-related medical complications. Numerous epidemiological investigations in diverse ethnic groups have identified a lower adiponectin level as an independent risk factor for nonalcoholic fatty liver diseases and liver dysfunctions. Animal studies have demonstrated that replenishment of adiponectin protects against various forms of hepatic injuries, suggesting it to be a potential drug candidate for the treatment of liver diseases. This study was designed to investigate the cellular and molecular mechanisms underlying the hepatoprotective effects of adiponectin. Our results demonstrated that in adiponectin knockout (ADN-KO) mice, there was a preexisting condition of hepatic steatosis and mitochondrial dysfunction that might contribute to the increased vulnerabilities of these mice to secondary liver injuries induced by obesity and other conditions. Adenovirus-mediated replenishment of adiponectin depleted lipid accumulation, restored the oxidative activities of mitochondrial respiratory chain (MRC) complexes, and prevented the accumulation of lipid peroxidation products in ADN-KO mice but had no obvious effects on mitochondrial biogenesis. The gene and protein levels of uncoupling protein 2 (UCP2), a mitochondrial membrane transporter, were decreased in ADN-KO mice and could be significantly up-regulated by adiponectin treatment. Moreover, the effects of adiponectin on mitochondrial activities and on protection against endotoxin-induced liver injuries were significantly attenuated in UCP2 knockout mice. Conclusion: These results suggest that the hepatoprotective properties of adiponectin are mediated at least in part by an enhancement of the activities of MRC complexes through a mechanism involving UCP2. (HEPATOLOGY 2008.)

Obesity and its related metabolic syndrome are now reaching an epidemic level worldwide. Nonalcoholic fatty liver disease (NAFLD) is being increasingly recognized as a hepatic manifestation of the metabolic syndrome, which is closely associated with the development of insulin resistance, type 2 diabetes, and cardiovascular diseases.1, 2 NAFLD and nonalcoholic steatohepatitis (NASH) predispose people to the progressive development of hepatic fibrosis, cirrhosis, and end-stage liver cancers. A growing body of evidence from animal models suggests a “two-hit” hypothesis for the development of NAFLD.3–5 In this theory, the first hit is the occurrence of a fatty liver (steatosis), and it is followed by a second event leading to the development of NASH. The secondary liver injuries can be induced by endotoxin exposure, alcohol consumption, drug compounds, and virus infections, for example. To date, there have been very few effective drugs available for the treatment of NAFLD and NASH.

Adiponectin is an adipokine abundantly produced from adipocytes.6 This adipokine has recently attracted great attention because of its antidiabetic, anti-inflammatory, and anti-atherogenic activity. A previous study from our group provided the first evidence demonstrating that adiponectin possesses potent protective effects against alcoholic fatty liver disease, NAFLD, and steatohepatitis.7 In both ethanol-fed and ob/ob obese mice, chronic treatment with recombinant adiponectin markedly attenuated hepatomegaly and steatosis and significantly decreased hepatic inflammation and serum alanine aminotransferase (ALT) levels. Consistent with our data, studies from a number of independent groups have demonstrated the hepatoprotective effects of adiponectin in different animal models with various forms of liver injuries, including those induced by carbon tetrachloride, lipopolysaccharide (LPS)/D-galactosamine, pharmacological compounds, bile duct ligations, and a methionine-deficient diet.8–13 These animal-based findings have been further corroborated by clinical observations showing an inverse association between serum levels of adiponectin and liver dysfunctions.7, 14–18 Plasma adiponectin levels are significantly lower in patients with NAFLD in comparison with sex-matched and age-matched healthy controls. Moreover, NASH patients with lower levels of adiponectin show higher grades of inflammation, and this suggests that adiponectin deficiency is an important risk factor for the development of fatty livers, steatohepatitis, and other forms of liver injuries. Therefore, adiponectin and its agonists represent emerging therapeutic agents for the treatment and/or prevention of liver dysfunctions.

Although adiponectin deficiency has been shown to predispose mice to various liver injuries,9, 10 the underlying mechanisms remain largely unknown. Here we have examined the liver functions of adiponectin knockout (ADN-KO) mice and their wild-type littermates under both normal and obese conditions. Our results show a preexisting condition of steatosis accompanied by mitochondrial defects in the liver tissues of ADN-KO mice. Adiponectin treatment improves liver functions by increasing mitochondrial respiratory chain (MRC) activities and reducing hepatic lipid accumulation in these mice. Furthermore, we also provide evidence suggesting that uncoupling protein 2 (UCP2), a mitochondrial membrane carrier protein, is critically important in mediating the hepatoprotective effects of adiponectin.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Animal Studies.

ADN-KO mice with a C57BL/6J background and leptin receptor−/−/adiponectin−/− double knockout (DKO) mice were generated as we described recently.19 Uncoupling protein 2 knockout (UCP2-KO) mice with a C57BL/6J background were purchased from the Jackson Laboratory (Bar Harbor, ME). Animals were provided with standard chow (LabDiet 5053, LabDiet, Purina Mills, Richmond, IN) unless otherwise specified. To establish a dietary obese mouse model, 4-week old male C57BL/6J and ADN-KO mice were fed a 19.33 kJ/g high-fat diet (D12451, Research Diet, New Brunswick, NJ) that was 49.85% fat, 20% protein, and 30.15% carbohydrate for 8 weeks. All animals were kept under 12-hour light-dark cycles at 22 to 24°C. For adiponectin treatment, the recombinant adenovirus expressing adiponectin or luciferase was tail-vein–injected into mice 2 weeks prior to tissue collection.19, 20 Note that the amount of the injected adenovirus (108 pfu) caused no toxicity in mice as judged by their body weight gains, food and water intake, liver functions, and other behavioral variables. The increased expression levels of adiponectin were confirmed by both enzyme-linked immunosorbent assay and Western blotting analysis. For the induction of acute liver injury, 200 μg of LPS was intraperitoneally injected into adiponectin or UCP2-KO mice, and the liver tissue was collected at different time points as indicated. All the animal experimental procedures were approved by the Committee on the Use of Live Animals for Teaching and Research of the University of Hong Kong and were carried out in accordance with the Guide for the Care and Use of Laboratory Animals.

Isolation of Mitochondria from Liver Tissues and Measurement of MRC Activities.

Mice were sacrificed under deep anesthesia, and the liver tissues were immediately collected for hepatic mitochondria isolation with the procedures that we described previously.21 The resulting mitochondrial pellet was resuspended in a 50 mM trishydroxymethylaminomethane solution (pH 8.0) to a protein concentration of 5 to 8 mg/mL and kept at −80°C for subsequent measurements of MRC activities. Briefly, complex I activity was determined by the measurement of the colorimetric changes during the oxidation of NADH at the wavelength of 340 nm. The specific activities were calculated by the subtraction of values obtained in the presence of the complex I–specific inhibitor rotenone. The dual complex activities of complexes II and III were measured by the monitoring of the colorimetric changes during the reduction of cytochrome C at the wavelength of 550 nm in the presence or absence of the specific inhibitor antimycin A. The activity of complex IV was analyzed by the measurement of the oxidation of reduced cytochrome C with a commercially available cytochrome C oxidase assay kit (Sigma, St Louis, MO). The activity of complex V was evaluated with a coupled enzymatic (pyruvate kinase and lactate dehydrogenase) assay for measuring the hydrolytic activities of ATPase. The specific activity was calculated by subtraction of the readings in the presence of oligomycin.

Electron Microscopy Analysis of Mitochondrial Ultrastructures and Histological Staining for the Evaluation of Hepatic Steatosis and Inflammation.

Transmission electron microscopy (TEM) was used to evaluate the ultrastructures of liver mitochondria. Briefly, ∼1 mm3 of liver tissues from the same anatomical locations was fixed in 2.5% glutaraldehyde in 100 mM sodium cocodylate, treated with 1% osmium tetroxide in a 0.1 M NaCo buffer, washed, dehydrated, and embedded in Epon 812 resin. Random thin sections (100 nm) were cut and stained with aqueous uranyl acetate and Reynold's lead acetate. The images were examined with a Philips EM208S transmission electron microscope (Philips, Eindhoven, The Netherlands). To evaluate hepatic steatosis, liver tissues were embedded in Tissue-Teck O.C.T. compound (Sakura Finetek USA, Inc., Torrance, CA), and 8-μm sections were prepared with a Leica 2800E Frigocut microtome cryostat. The air-dried frozen sections were used for Red Oil O and hematoxylin-eosin staining.22

Thiobarbituric Acid Reactive Substance (TBARS) and Lucigenin Assays.

The concentrations of the lipid peroxidation product malondialdehyde (MDA) in serum, liver lysates, or mitochondrial suspensions were determined with a commercial TBARS assay kit (Cayman Chemical, Ann Arbor, MI). The results were calculated against the total protein contents. The activity of nicotinamide adenine dinucleotide phosphate (NADPH) oxidase was assessed with the lucigenin assay as described.23

Quantitative Reverse-Transcription Polymerase Chain Reaction (RT-PCR).

Total RNA was extracted from the liver tissues with the TRIzol reagent (Invitrogen Corp., Carlsbad, CA) and transcribed into complementary DNA with the ImProm-II reverse transcription system (Promega, Madison, WI). Mitochondrial and nuclear genomic DNA was obtained by proteinase K digestion and phenol-chloroform extraction. Fifty nanograms of complementary DNA or 4 ng of genomic DNA was used for quantification of the gene expression or mitochondrial DNA (mtDNA) copy number, respectively. Quantitative RT-PCR was performed with GreenER qPCR Supermix with specific primers (Invitrogen). The reactions were carried out on a 7000 Sequence Detection System (Applied Biosystems, Foster City, CA). For gene expression, quantification was achieved with Ct values normalized against glyceraldehyde 3-phosphate dehydrogenase as an internal control. For mtDNA copy number determination, a T cell receptor was used as an internal control. The primer sequences are listed in Supplementary Table 1.

Western Blotting Analysis.

Fifty micrograms of mitochondrial proteins was separated by 12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis, transferred onto a polyvinylidene difluoride membrane, and probed with sheep antiserum against UCP2 (Santa Cruz, CA) at a dilution of 1:1000. After incubation with a horseradish peroxidase–conjugated rabbit anti-sheep immunoglobulin G antibody, the proteins were visualized with enhanced chemiluminescence reagents (GE Healthcare, Uppsala, Sweden). The specificity of this antibody was validated by immunoprecipitation and mass spectrometry analysis, which confirmed the purified protein as UCP2 (data not shown).

Terminal Deoxynucleotidyl Transferase-Mediated Deoxyuridine Triphosphate Nick-End Labeling (TUNEL) Assay.

The liver tissues were obtained 6 hours after LPS injection. Apoptosis was assessed by TUNEL assay with an in situ cell death detection kit (Roche, Indianapolis, IN). The liver sections were viewed at 488-nm excitation/512-nm emission under a fluorescence microscope (Leica Microsystems, Bensheim, Germany). TUNEL-positive cells were manually counted in 20 random fields (×400) per sample to determine the average number of positive-staining apoptotic cells. DNase I treatment of additional sections was performed, and they served as positive controls.

Statistical Analysis.

All the results were derived from at least three independent experiments. Values are expressed as the mean ± standard error. The statistical calculations were performed with the Statistical Package for the Social Sciences version 11.5 software package (SPSS, Inc., Chicago, IL). Differences between groups were determined by the Student t test. Comparisons with P < 0.05 were considered statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Mice Without Adiponectin Exhibit Increased Vulnerabilities to Obesity-Induced Liver Injuries and a Preexisting Condition of Hepatic Steatosis.

Although many epidemiological studies have observed a close association between low serum levels of adiponectin and increased risks of NAFLD and NASH,7, 14–18 there is no laboratory evidence showing that adiponectin deficiency can exacerbate obesity-induced NAFLD and NASH in animal models. To this end, we established two types of obese models with an adiponectin-null background (mice whose obesity was induced by a high-fat diet and mice who were genetically obese because of the lack of functional leptin receptor−/−), and we evaluated the development of fatty liver diseases in these animals. Our results demonstrated that in both dietary obese mice and genetically obese mice, adiponectin deficiency resulted in significantly increased hepatomegaly (Fig. 1A), exacerbated hepatic steatosis (Fig. 1B), and a more severe phenotype of liver injury, as reflected by the elevated levels of tumor necrosis factor alpha (TNFα) and the lipid peroxidation product MDA (Fig. 1C). In addition, ALT, a well-established marker of liver injury, was also significantly elevated in both obese models with adiponectin deficiency (data not shown). These results, along with several previous studies,7–11, 13 suggest that adiponectin deficiency is a common risk factor that increases the vulnerability of the liver to various acute and chronic injuries.

thumbnail image

Figure 1. Adiponectin deficiency exacerbates liver injury in both high fat (HF)–fed obese mice and genetically obese mice. The two types of obese models were established as described in the Materials and Methods section. The liver tissues were collected from 12-week-old C57 (C57+HF) and ADN-KO (ADN-KO+HF) mice that had been fed an HF diet for 8 weeks or from 12-week-old leptin receptor−/− (DB) and DKO mice, and they were evaluated for (A) hepatomegaly, (B) hepatic steatosis, and (C) liver inflammatory injuries as described in the text. Data are presented as fold changes against the corresponding control groups. *P < 0.05 and **P < 0.01 versus the C57+HF group; #P < 0.05 versus the DB group (n = 6-8).

Download figure to PowerPoint

We also analyzed the liver tissues of ADN-KO mice and their wild-type littermates under basal conditions. To our surprise, we found that there was a preexisting condition of hepatic steatosis in the liver tissues of 3-week-old ADN-KO mice even with the standard chow feeding (Fig. 2A). The increased lipid accumulation was also detected in the liver tissues of 13-week-old ADN-KO mice. On the other hand, TNFα levels were not altered in 3-week-old ADN-KO mice but were significantly increased in 13-week-old ADN-KO mice in comparison with those of wild-type mice (Fig. 2B). These data suggest that the lack of adiponectin might cause a “first hit” condition of hepatic steatosis, which predisposes the mice to the secondary liver injuries and inflammation induced by obesity, chemicals, and endotoxins, for example.

thumbnail image

Figure 2. Hepatic steatosis preexists in the liver tissues of ADN-KO mice. Liver sections derived from 3- or 13-week-old C57 and ADN-KO mice were subjected to (A) Red Oil O staining for evaluating the fatty liver status and (B) quantitative RT-PCR analysis for measuring TNFα mRNA levels. Data are shown as fold changes against the age-matched C57 control mice. *P < 0.05 versus C57 mice (n = 6).

Download figure to PowerPoint

Adiponectin Deficiency Is Associated with Abnormal Mitochondrial Morphologies and Decreased MRC Activities.

Mitochondrial dysfunction plays a central role in various forms of hepatic steatosis and liver injuries.24–28 Using TEM, we then compared the mitochondrial ultrastructures in the liver tissues of ADN-KO mice with those of their wild-type littermates. As shown in Fig. 3, this analysis demonstrated a profound morphological change of hepatic mitochondria in both 3- and 13-week-old ADN-KO mice. First, the electron densities of the mitochondrial matrix were kept well in wild-type C57 mice but were lost distinctly in ADN-KO mice. Second, the average sizes of mitochondria in ADN-KO mice were about 1.5- to 2-fold larger than those of the wild-type controls. Moreover, mitochondrial swelling and megamitochondria were present in ADN-KO mice. Some of the mitochondria in ADN-KO mice showed ruptures at the outer membrane, which were possibly due to a mitochondrial permeability transition. To further evaluate whether the mitochondrial functions were altered in ADN-KO mice, we purified the mitochondria from the liver tissues and measured the activities of each individual MRC complex. Our results demonstrated that the activities of complex I, complex II+III, complex IV, and complex V were significantly decreased in both 3- and 13-week-old ADN-KO mice in comparison with those of the age-matched and strain-matched C57 controls (Fig. 4). Moreover, in both dietary obese mice and genetically obese mice, targeted disruption of the adiponectin gene also led to a significant reduction in the MRC activities in comparison with those of littermate controls with normal adiponectin levels (Supplementary Fig. 1). Taken together, these data suggest that the mitochondrial abnormalities might contribute to the increased vulnerabilities of ADN-KO mice to various liver injuries.

thumbnail image

Figure 3. Representative electron photomicrographs demonstrating the altered mitochondrial ultrastructures in the liver tissues of 3- and 13-week-old ADN-KO mice. Note that there are profound mitochondrial morphological changes, including mitochondrial swelling, mega mitochondria, and mitochondrial outer membrane ruptures, in the livers of ADN-KO mice. Scale bars represent 1 μm.

Download figure to PowerPoint

thumbnail image

Figure 4. MRC activities are significantly reduced in the liver tissues of ADN-KO mice. The activities of hepatic MRC complexes I, II+III, IV, and V of both 3- and 13-week-old ADN-KO mice were measured as described in the Materials and Methods section and calculated against the reaction time and the total amount of proteins. *P < 0.05 compared to C57 mice (n = 4-7).

Download figure to PowerPoint

Adiponectin Treatment Enhances MRC Activities but Has No Obvious Effects on Mitochondrial Biogenesis.

We next investigated whether replenishment of adiponectin can restore the MRC activities and reduce the lipid accumulation in the liver tissues of ADN-KO mice. To this end, the recombinant adenoviruses encoding mouse adiponectin or luciferase (as a control) were administered to ADN-KO mice through tail-vein injection. Consistent with our previous findings,20 serum levels of adiponectin started to rise within 2 days after adenoviral injection, reached peak levels at day 8, and remained within a physiological concentration (approximately 5 μg/mL) at 2 weeks after treatment (Supplementary Fig. 2A). A nonheating and nonreducing sodium dodecyl sulfate–polyacrylamide gel electrophoresis analysis showed that circulating adiponectin in ADN-KO mice infected with the adenovirus formed all three quaternary structures, including trimer, hexamer, and high-molecular-weight oligomeric complexes (Supplementary Fig. 2B); this pattern was similar to that of endogenous adiponectin in wild-type mice.29 Furthermore, over 60% of the adiponectin expressed in the liver and circulated in serum was the high-molecular-weight oligomer, which has been proposed to be the major active form possessing hepatoprotective functions.6 This evidence suggests that the adenovirus-mediated expression system could deliver adiponectin with proper biological activities. As shown in Fig. 5A,B, adenovirus-mediated adiponectin replenishment in 13-week-old ADN-KO mice significantly decreased the lipid accumulation and enhanced the MRC activities in liver tissues. Similar results have also been observed in 3-week-old ADN-KO mice (Supplementary Fig. 3). Moreover, the elevated mitochondrial MDA contents in the liver tissues of 13-week-old ADN-KO mice were significantly reduced by adiponectin treatment (Fig. 5C). On the other hand, the activities of NADPH oxidase, a major source of cytoplasmic reactive oxygen species (ROS) production, showed no obvious changes in ADN-KO mice before and after adiponectin treatment. These results suggest that the hepatoprotective properties of adiponectin might be at least partly attributable to its stimulatory effects on mitochondrial activities and the oxidative phosphorylations of the MRC complexes.

thumbnail image

Figure 5. Adiponectin treatment reduces lipid accumulation and improves MRC activities in the liver tissues of ADN-KO mice. Thirteen-week-old ADN-KO mice were treated with 108 pfu of a recombinant adenovirus encoding luciferase (ADN-KO+Luci) or an adenovirus encoding adiponectin (ADN-KO+ADN). The liver tissues were collected after 2 weeks and subjected to (A) Red Oil O staining and (B) mitochondria isolation for MRC activity measurements. Note that the MRC complex activities of ADN-KO+Luci mice were not different from those of the ADN-KO mice in Fig. 5. (C) The MDA contents and NADPH oxidase activity were measured with the liver lysates derived from C57 and ADN-KO mice treated without or with the recombinant adenoviruses according to the procedures described in the Materials and Methods section. *P < 0.05 and **P < 0.01 versus ADN-KO+Luci; #P < 0.05 versus C57 mice (n = 5-10).

Download figure to PowerPoint

The components of oxidative respiratory chain complexes are encoded by the coordination of the nuclear and mitochondrial genomes. To investigate whether the decreased MRC activities in ADN-KO mice are due to impaired mitochondrial biogenesis, we evaluated the mtDNA copy number and the messenger RNA (mRNA) levels of several mitochondrial genome-encoded genes. This analysis demonstrated that in both 3- and 13-week-old ADN-KO mice, the mRNA levels of several MRC complexes genes, such as NADH dehydrogenase subunit 1, cytochrome B, cytochrome C oxidase subunit 1, and ATP synthase subunit 6, were significantly decreased by ∼30% in comparison with those of the wild-type mice. The mtDNA copy number was decreased by ∼35% in 13-week-old ADN-KO mice but not in 3-week-old ADN-KO mice. However, adiponectin treatment had no obvious effects on mtDNA or mRNA expressions of mitochondria-encoded genes, mitochondria transcription factor A, peroxisome proliferator-activated receptor-γ coactivator-1α, DNA polymerase γ, and mitochondrial RNA polymerase (data not shown), and this suggests that adiponectin-mediated improvement of mitochondrial functions might not be related to the de novo mitochondrial biogenesis.

Expression of UCP2 Is Decreased in ADN-KO Mice and Can Be Induced by Adiponectin Treatment.

UCP2 is a mitochondrial inner membrane carrier protein that has been suggested to regulate the proton leak, uncoupling, and mitochondrial respirations.30 A growing body of evidence suggests that UCP2 possesses hepatoprotective and anti-inflammatory activities by enhancing the mitochondrial functions, reducing mitochondrial ROS levels, and inhibiting the production of proinflammatory cytokines.31–33 Using a genome-wide microarray analysis, we found that the mRNA levels of UCP2 were significantly decreased in high-fat diet–fed ADN-KO mice and DKO mice in comparison with age-matched and strain-matched control animals (data not shown). We then evaluated the expressions of UCP2 in ADN-KO mice by comparing them with those of wild-type controls. Interestingly, our results showed that in mitochondria purified from liver tissues of both 3- and 13-week-old ADN-KO mice, the protein levels of UCP2 were significantly decreased in comparison with those of wild-type mice (Fig. 6A). Moreover, a lower level of UCP2 expression was also detected in 16- and 18-day embryos of ADN-KO mice (data not shown), and this indicated that the reduced UCP2 might contribute to the preexisting conditions of hepatic steatosis and mitochondrial dysfunction in ADN-KO mice. On the other hand, replenishment of adiponectin dramatically increased both the gene and protein levels of UCP2 (Fig. 6A,B).

thumbnail image

Figure 6. The reduced expression of UCP2 in the liver tissue of ADN-KO mice can be restored by adenovirus-mediated replenishment of recombinant adiponectin. (A) Mitochondrial proteins (50 μg) were analyzed by Western blotting using antibodies against UCP2 and single-stranded DNA binding protein 1 (SSBP1) as the loading control. (B) Quantitative RT-PCR was performed for measuring the mRNA levels of UCP2 in liver tissues of C57 and ADN-KO mice treated without or with a recombinant adenovirus encoding luciferase (ADN-KO+Luci) or adiponectin (ADN-KO+ADN). Data are presented as fold changes against the C57 control groups. *P < 0.05 versus C57 mice; #P < 0.05 versus ADN-KO+Luci (n = 4-6).

Download figure to PowerPoint

The Stimulatory Effects of Adiponectin on MRC Activities Are Abolished in UCP2-KO Mice.

To investigate the role of UCP2 in mediating the hepatic actions of adiponectin, we measured the MRC activities in UCP2-KO mice treated with or without adiponectin (Fig. 7). In comparison with the wild-type controls, the activities of the MRC complexes II+III, IV, and V were not changed in UCP2-deficient mice, whereas the activity of MRC complex I was decreased by ∼40%. In contrast to its effects in C57 and ADN-KO mice, adiponectin treatment failed to raise the activities of the MRC complexes I, II+III, IV, and V in UCP2-KO mice, and this suggested an obligatory role of this mitochondrial carrier protein in mediating the stimulatory activities of adiponectin on mitochondrial functions.

thumbnail image

Figure 7. UCP2 deficiency attenuates the stimulatory effects of adiponectin on the MRC activities. The activities of mitochondrial MRC complexes (I, II+III, IV, and V) were measured and compared among the liver samples derived from 5-week-old C57, ADN-KO, and UCP2-KO mice that had been treated with or without a recombinant adenovirus encoding adiponectin for 2 weeks through tail-vein injections. Data were calculated as fold changes against the values of C57 mice. #P < 0.05 versus C57 mice; *P < 0.05 and **P < 0.01 versus ADN-KO mice (n = 4-6).

Download figure to PowerPoint

We also evaluated the role of UCP2 in mediating the hepatoprotective properties of adiponectin in LPS-induced acute liver injury. As shown in Fig. 8, the liver sections of both ADN-KO and UCP2-KO mice exposed to LPS showed extensive areas of ballooned, hypereosinophilic hepatocytes, necrosis, histological features of apoptosis (that is, condensed nuclear chromatin with reduced cytoplasmic volume), and a large number of TUNEL-positive hepatocytes. Adenovirus-mediated adiponectin treatment prevented LPS-induced massive apoptosis of hepatocytes and decreased LPS-induced elevation of TNFα and ALT levels in ADN-KO mice but not in UCP2-KO mice. These data suggest that up-regulation of UCP2 by adiponectin might play a critical role in mediating its protective effects against liver injuries.

thumbnail image

Figure 8. The protective effects of adiponectin against LPS-induced liver injury are attenuated in UCP2-deficient mice. Recombinant adenoviruses encoding luciferase (Luci) or adiponectin (ADN) were tail-vein–injected into ADN-KO and UCP2-KO mice. One week later, the mice were treated with LPS as described in the Materials and Methods section. (A) The thin sections of the liver tissues collected 6 hours after LPS treatment were subjected to histological evaluation and TUNEL staining. The specificity of the TUNEL assay was tested with samples for which the terminal transferase was not added on the slides (data not shown). (B) Serum levels of ALT were determined with commercial reagents from Sigma-Aldrich. (C) TNFα mRNA levels in liver tissues were quantified by real-time RT-PCR. *P < 0.05 versus all other groups (n = 4).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Although the hepatoprotective properties of adiponectin have been suggested by many clinical and animal studies,7, 12, 14–18 the detailed cellular and molecular mechanisms remain elusive. Here we have found that the liver tissues of ADN-KO mice show aberrant mitochondrial ultrastructures and decreased MRC activities in comparison with those of the C57 control mice (Figs. 3 and 4). Notably, the mitochondrial dysfunction can be detected in the liver tissues of 3-week-old ADN-KO mice (Fig. 3), and this suggests that it is an early event that predisposes ADN-KO mice to NAFLD, NASH, and other forms of liver injuries (Fig. 1). Using adenovirus-mediated overexpression approaches, we have found that adiponectin treatment almost completely depletes the lipid accumulation in ADN-KO mice and restores the activities of complexes I, II+III, IV, and V in both 3- and 13-week-old ADN-KO mice (Fig. 5 and Supplementary Fig. 3). On the other hand, it has little effect on mtDNA copy numbers and the expression levels of genes involved in mitochondrial biogenesis, and this suggests that the stimulatory effects of adiponectin on mitochondrial functions might not involve de novo mitochondrial biogenesis.

A growing body of evidence suggests that mitochondrial dysfunctions might represent a central mechanism linking obesity with its related metabolic complications.34 Fatty liver disease is an important component of the metabolic syndrome associated with obesity and is causally involved in the development of insulin resistance and type 2 diabetes.35 In patients with NASH, the hepatic mitochondria exhibit ultrastructural lesions and decreased activity of the respiratory chain complexes.27, 36 Under this condition, the decreased activity of the respiratory chain results in an accumulation of ROS that oxidize fat deposits to form lipid peroxidation products, which in turn cause steatohepatitis, necrosis, inflammation, and fibrosis. The increased mitochondrial ROS formation in steatohepatitis could directly damage mtDNA and the respiratory chain polypeptides and induce nuclear factor kappa B activation and hepatic TNFα production.37 Oxidative phosphorylation reactions mediated by the MRC complexes are directly involved in regulating the intracellular ROS levels and preventing the hepatic accumulation of lipids and lipid peroxidation products. In addition, mitochondria are the major sites for fat oxidation. Mitochondrial dysfunction can increase hepatic lipid accumulation by promoting lipogenesis in the liver.28 In this study, we found increased lipid accumulation in ADN-KO mice even when the animals were fed standard chow (Fig. 2). This preexisting hepatic steatotic condition might be the direct result of dysregulated mitochondrial functions in these mice. This conclusion is supported by our observation that adiponectin-mediated reduction of hepatic lipid accumulation is associated with the restoration of the impaired MRC activities and the reduction of the elevated mitochondrial MDA levels in ADN-KO mice (Fig. 5). Therefore, stimulation of mitochondrial functions might represent an important mechanism by which adiponectin exerts its multiple beneficial effects on various obesity-related pathologies.38–40

Impaired mitochondrial oxidative phosphorylation is the major culprit for the increased ROS production and liver injury associated with steatohepatitis. It is known that UCP2 possesses antioxidant activities through inhibition of ROS production from mitochondria.30, 41 UCP2 can also inhibit the production of proinflammatory cytokines in both macrophage and Kupffer cells, the major sources of liver inflammation.31 Here we have provided evidence supporting a key role of UCP2 in mediating the beneficial effects of adiponectin on mitochondrial functions. The protein and mRNA levels of UCP2 are decreased in the liver tissues of ADN-KO mice and can be significantly up-regulated by adiponectin treatment (Fig. 6). Furthermore, the effects of adiponectin on MRC activities and endotoxin-induced liver injuries are dramatically attenuated in UCP2-deficient mice, and this suggests that increased UCP2 expression might be needed for adiponectin to elicit its hepatoprotective functions. Although this is the first report on the regulation of UCP2 expression by adiponectin in the liver, a number of studies have shown that low adiponectin levels are associated with reduced UCP2 expression in adipose tissue. For example, abnormal down-regulation of adiponectin and UCP2 expression has been reported in adipose tissue from first-degree relatives of patients with type 2 diabetes.42, 43 Lipodystrophy in antiretroviral-treated human immunodeficiency virus patients is associated with systemic insulin resistance, mitochondrial impairment, UCP2 down-regulation, and reduced adiponectin levels in adipose tissue.44 Increased insulin sensitivity in 11β-hydroxysteroid dehydrogenase type-1–deficient mice and Crebbp heterozygous mice is associated with increased adiponectin and UCP2 mRNA levels.45 In agouti yellow (Ay/a) obese mice, adiponectin treatment increases the expression of UCP1 in brown adipose tissue, UCP2 in white adipose tissue, and UCP3 in skeletal muscle.46 Chevillotte and colleagues47 showed that UCP2 controls adiponectin gene expression by regulating ROS production. Taken together, these results suggest the existence of a reciprocal relationship between uncoupling proteins and adiponectin in various tissues. However, the detailed signaling mechanisms underlying adiponectin-stimulated UCP2 expression remain poorly understood and warrant further investigation.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Supplementary material for this article can be found on the H EPATOLOGY Web site ( http://interscience.wiley.com/jpages/0270-9139/suppmat/index.html ).

FilenameFormatSizeDescription
hep22444-SupplementaryFigure1.tif1317KSupporting Information file hep22444-SupplementaryFigure1.tif
hep22444-SupplementaryFigure2.tif1475KSupporting Information file hep22444-SupplementaryFigure2.tif
hep22444-SupplementaryFigure3.tif4232KSupporting Information file hep22444-SupplementaryFigure3.tif
hep22444-SupplementaryData.doc45KSupporting Information file hep22444-SupplementaryData.doc

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.