SEARCH

SEARCH BY CITATION

Abstract

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

Kupffer cells have been implicated in the pathogenesis of various liver diseases. However, their involvement in metabolic disorders of the liver, including fatty liver disease, remains unclear. The present study sought to determine the impact of Kupffer cells on hepatic triglyceride storage and to explore the possible mechanisms involved. To that end, C57Bl/6 mice rendered obese and steatotic by chronic high-fat feeding were treated for 1 week with clodronate liposomes, which cause depletion of Kupffer cells. Loss of expression of marker genes Cd68, F4/80, and Clec4f, and loss of Cd68 immunostaining verified almost complete removal of Kupffer cells from the liver. Also, expression of complement components C1, the chemokine (C-C motif) ligand 6 (Ccl6), and cytokines interleukin-15 (IL-15) and IL-1β were markedly reduced. Importantly, Kupffer cell depletion significantly decreased liver triglyceride and glucosylceramide levels concurrent with increased expression of genes involved in fatty acid oxidation including peroxisome proliferator-activated receptor alpha (PPARα), carnitine palmitoyltransferase 1A (Cpt1α), and fatty acid transport protein 2 (Fatp2). Treatment of mice with IL-1β decreased expression of PPARα and its target genes, which was confirmed in primary hepatocytes. Consistent with these data, IL-1β suppressed human and mouse PPARα promoter activity. Suppression of PPARα promoter activity was recapitulated by overexpression of nuclear factor κB (NF-κB) subunit p50 and p65, and was abolished upon deletion of putative NF-κB binding sites. Finally, IL-1β and NF-κB interfered with the ability of PPARα to activate gene transcription. Conclusion: Our data point toward important cross-talk between Kupffer cells and hepatocytes in the regulation of hepatic triglyceride storage. The effect of Kupffer cells on liver triglycerides are at least partially mediated by IL-1β, which suppresses PPARα expression and activity. (HEPATOLOGY 2010.)

The liver contains numerous different cell types including parenchymal cells, stellate cells, sinusoidal endothelial cells, cholangiocytes, and Kupffer cells. Kupffer cells are derived from circulating monocytes that arise from bone marrow progenitors. Once localized within the liver, they differentiate to perform specialized functions, including phagocytosis, antigen processing, and presentation. Kupffer cells also generate various products, including cytokines, prostanoids, nitric oxide, and reactive oxygen intermediates, which influence the phenotypes of neighboring parenchymal, stellate, and endothelial cells.1 Kupffer cells have been implicated in the pathogenesis of various liver diseases ranging from viral hepatitis to steatohepatitis, alcoholic liver disease, intrahepatic cholestasis, activation or rejection of the liver during liver transplantation, and liver fibrosis.2 However, their role and function in metabolic disorders of the liver remains poorly explored.

Nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease in Western countries and is considered as the hepatic manifestation of the metabolic syndrome.3 NAFLD varies from the relatively benign hepatic steatosis characterized by accumulation of triglycerides in hepatocytes to nonalcoholic steatohepatitis (NASH), which might progress to end-stage liver diseases such as cirrhosis and liver cancer.4 As dictated by the two-hit model of NASH pathogenesis, once steatosis has developed, the liver is “sensitized” to secondary insults leading to NASH.5 Cytokines, mitochondrial dysfunction, oxidative stress, and lipid peroxidation are considered to be the main “second hits” in the induction of NASH. Recently, increasing emphasis has been placed on lipotoxicity as a core event in the development of NASH.6 Interestingly, in certain mouse models of NASH, inflammation as characterized by macrophage inflammation has been found to precede steatosis.7 Because steatosis is essential for development of NASH, it is important to gain more insight into the mechanisms that influence hepatic triglyceride accumulation, including the role of the various liver cell types. This study was aimed at determining the impact of Kupffer cells on hepatic triglyceride storage and exploring the possible mechanisms involved. The results reveal that Kupffer cells promote hepatic triglyceride storage via interleukin-1β (IL-1β)–dependent suppression of peroxisome proliferator-activated receptor α (PPARα) activity, leading to decreased expression of PPARα target genes and decreased fat oxidation. The inhibitory effect of IL-1β on PPARα expression was found to be mediated by nuclear factor κB (NF-κB) binding to specific binding sites in the promoter of the human and mouse PPARα gene.

Materials and Methods

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

Animal Experiments.

Eight-week-old to 12-week-old male C57Bl/6J mice were fed a low-fat diet (LFD; n = 8) or a high-fat diet (HFD; n = 20) for 20 weeks, providing 10% or 45% energy percent as fat, respectively (D12450B or D12451; Research Diets, New Brunswick, NJ). The lard component in these diets was replaced by palm oil. In the last week of the diet intervention, half of the animals fed the HFD (n = 10) received an intraperitoneal injection with clodronate liposomes twice. At the end of the diet intervention, blood was collected from the orbital plexus under isoflurane anesthesia. Immediately thereafter, animals were sacrificed by cervical dislocation. Livers were removed and immediately frozen in liquid nitrogen. The animal experiments were approved by the Local Committee for Care and Use of Laboratory Animals at Wageningen University.

For the in vivo injection of IL-1β, 8-week-old to 12-week-old male C57BL/6 mice were injected intraperitoneally with 1 μg of recombinant human IL-1β dissolved in saline, followed by sacrifice and collection of livers 16 hours thereafter. Control mice received saline only.

Plasmids.

Cloning of the mouse PPARα promoter, human PPARα promoter and serial deletion constructs have been described previously.8 The p50 and p65 expression plasmids cloned into a cytomegalovirus vector were a generous gift from G. Haegeman (Gent University, Belgium). PPRE-tk-LUC containing three copies of acyl-coenzyme A (CoA) oxidase peroxisome proliferator response element (PPRE) was a generous gift from Ronald Evans (Salk Institute, La Jolla, CA).

Immunohistochemistry/Histology.

To stain Kupffer cells, immunohistochemistry was performed using an antibody against Cd68 (Serotec, Oxford, UK). Sections were preincubated with 20% normal goat serum followed by overnight incubation at 4°C with the primary antibody diluted 1:50 in phosphate-buffered saline (PBS)/1% bovine serum albumin (BSA). After incubation with the primary antibody, a goat anti-rat immunoglobulin G conjugated to horseradish peroxidase (Serotec) was used as secondary antibody. Visualization of the complex was done using AEC Substrate Chromogen for 6 minutes. Negative controls were prepared by omitting the primary antibody. Oil red O and hematoxylin & eosin staining of sections was performed using standard protocols.

Cell Culture and Transient Transfection Assays.

Human hepatoma HepG2 cells were obtained from ATCC (Manassas, VA) and grown in Dulbecco's modified Eagle medium (Cambrex, Verviers, Belgium) containing 10% fetal bovine serum (FBS) and antibiotics. Reporter vectors containing the mouse or human PPARα promoter were transfected into HepG2 cells, together with an expression vector (pSG5) for mPPARα. Transfections were carried out via calcium-phosphate precipitation. A β-galactosidase reporter was cotransfected to normalize for differences in transfection efficiency. After transfection, cells were treated with 5 ng/mL of IL-1β, tumor necrosis factor alpha (TNFα), and IL-6 (R&D Systems Europe Ltd., Abingdon, UK) for 8 hours or with 50 μM of Wy-14643 (ChemSyn Laboratories, Lenexa, KS) for 24 hours prior to lysis. Luciferase activity was measured using the Promega luciferase assay kit on a Fluoroskan Ascent FL (Thermo labsystems, Breda, The Netherlands). β-Galactosidase activity was measured in the cell lysate by a standard assay using 2-nitrophenyl-β D-galactopyranoside as a substrate.

Isolation of Primary Hepatocytes.

Primary mouse hepatocytes were isolated from Sv129 mice as described.9 Briefly, after cannulation of the portal vein, the liver was perfused with calcium-free Hank's balanced salt solution (HBSS) which was pregassed with 95% O2/5% CO2. Next, the liver was perfused with a collagenase solution (Sigma-Aldrich) until swelling and degradation of the internal liver structure was observed. The hepatocytes were released, filtered, and washed several times using Krebs buffer. The viability was assessed by using trypan blue (Sigma-Aldrich) and was around 80%. Cells were brought into culture using Williams E Medium supplemented with 10% FBS (Lonza Bioscience, Verviers, Belgium), penicillin/streptomycin/fungizone, insulin, and dexamethasone. Cells were plated on collagen-coated (Serva Feinbiochemica, Heidelberg, Germany) wells with a density of 0.5 × 106 cells/mL. After 4 hours of incubation, the medium was removed and replaced with fresh medium. The next day, hepatocytes were used for experiments. Human hepatocytes and Hepatocyte Culture Medium Bullet kit were purchased from Lonza Bioscience (Verviers, Belgium).

Separation of Mouse Liver into Hepatocytes and Kupffer Cell–Enriched Fraction.

Liver was collected and washed using HBSS and digested in 20 mL digestion buffer (0.05% collagenase, 2% FBS, 50 μg/L deoxyribonuclease-I, and 0.06% BSA in HBSS) for 45 minutes. Solution was filtered with a micro-gauge followed by 2 minutes centrifugation at 50g at 4°C. The pellet representing the hepatocytes was washed three times with HBSS before resuspension into TRIzol. The supernatant enriched in Kupffer cells was centrifuged for 10 minutes at 800g. The pellet was resuspended into TRIzol.

RNA Isolation and Real-Time Polymerase Chain Reaction.

Total liver RNA was isolated with TRIzol reagent (Invitrogen, Breda, The Netherlands) according to manufacturer's instructions. A NanoDrop ND-1000 spectrophotometer (Isogen, Maarssen, The Netherlands) was used to determine RNA concentrations. Total RNA (1 μg) was reverse-transcribed using iScript (Bio-Rad, Veenendaal, The Netherlands). Complementary DNA was amplified on a Bio-Rad MyIQ or iCycler polymerase chain reaction (PCR) machine using Platinum Taq DNA polymerase (Invitrogen, Breda, The Netherlands). PCR primer sequences were taken from the PrimerBank10 and ordered from Eurogentec (Seraing, Belgium). Sequences of the primers used are presented in Supporting Table 1.

Microarray Analysis.

Total RNA was prepared using TRIzol reagent and pooled from all animals within each group prior to further purification using RNeasy micro columns (Qiagen, Venlo, The Netherlands). Further processing of RNA and hybridization to Affymetrix mouse genome 430 2.0 arrays were carried out as previously described.11 Scans of the Affymetrix arrays were processed using packages from the Bioconductor project.12 Expression levels of probe sets were calculated using GCRMA.

Liver Triglycerides and Ceramides.

Liver triglycerides were determined in 10% liver homogenates prepared in buffer containing 250 mM sucrose, 1 mM ethylene diamine tetraacetic acid, and 10 mM Tris-HCl at pH 7.5 using a kit from Instruchemie (Delfzijl, The Netherlands). Liver ceramide and glucosylceramide content were determined as previously described.13

Plasma Metabolites.

Plasma was obtained from blood by centrifugation for 10 minutes at 10,000g. Plasma glucose concentration was determined using a kit from Elitech (Sopachem, Wageningen, the Netherlands). Plasma triglycerides were determined using a kit from Instruchemie (Delfzijl, the Netherlands). Plasma free fatty acids were determined using a kit from WAKO Chemicals (Instruchemie).

Statistical Analysis.

Statistical significant differences were calculated using Student t test. The cut-off for statistical significance was set at a P value of 0.05 or below.

Results

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

Liver Depletion of Kupffer Cells Using Clodronate Liposomes.

To induce obesity and hepatic steatosis, C57Bl/6 mice were fed a high-fat diet (HFD) for 20 weeks. HFD accelerated weight gain and increased fat weight when compared with LFD (Fig. 1A,B). In addition, HFD markedly increased hepatic lipid storage (Fig. 1C), whereas liver weight remained unchanged (Fig. 1D). Histological examination of the livers showed minimal signs of inflammation (data not shown).

thumbnail image

Figure 1. Effect of high-fat diet on weight gain and liver triglycerides. C57Bl/6 mice were fed a LFD or HFD for 20 weeks. (A) Changes in body weight during the course of the LFD or HFD feeding. Gray squares = LFD (n = 8), Black squares = HFD (n = 20). (B) Weight of epididymal fat depot. (C) Liver triglyceride concentration. (D) Liver weights. Error bars represent standard error (SE). Differences were evaluated using Student t test. *P < 0.05.

Download figure to PowerPoint

To investigate the importance of Kupffer cells in steatosis in the context of HFD-induced obesity, liver Kupffer cell depletion was performed. To that end, mice fed HFD for 20 weeks received two intraperitoneal injections of either PBS or clodronate liposomes during the last week of HFD. Clodronate is a bisphosphonate that specifically destroys Kupffer cells in liver and leaves other hepatic cell types unharmed.14–16 Depletion of Kupffer cells was confirmed by dramatically reduced expression of Kupffer cells marker genes Cd68, F4/80 (Emr), and Kupffer cell receptor (Clec4f) (Fig. 2A), as well as by immunohistochemistry using a specific antibody against Cd68 (Fig. 2B). Microarray analysis confirmed the markedly reduced expression of numerous macrophage-related genes (Fig. 2C). Importantly, Kupffer cell depletion was associated with a significant reduction in expression of several cytokines, including IL-15 and IL-1β. As expected, high-fat feeding markedly raised hepatic expression of Cidea and Adipsin,17, 18 whereas expression of Pnpla3, which has been linked to fatty liver disease,19, 20 was markedly reduced. No effect of clodronate treatment on expression of these genes was observed. In adipose tissue, HFD increased expression of macrophage and proinflammatory markers, which was not affected by clodronate treatment (Fig. 2D). The suppressive effect of clodronate liposome on IL-1β expression was verified by quantitative PCR (Fig. 2E). Similar to Kupffer cell markers Cd68 and F4/80, IL-1β expression was higher in the Kupffer cell–enriched liver fraction compared to the hepatocyte fraction of mouse liver, confirming that Kupffer cells are the primary source of IL-1β (Fig. 2F).

thumbnail image

Figure 2. Successful depletion of Kupffer cells using clodronate. C57Bl/6 mice were fed a LFD or HFD for 20 weeks. During the last week, half of the animals fed HFD were injected twice with clodronate. (A) Expression of Kupffer cell marker genes Cd68, F4/80, and Clec4f in livers as determined by quantitative PCR (qPCR). (B) Representative immunohistochemical staining of liver using antibody against the specific macrophage marker Cd68. (C) Heat map showing changes in expression of selected immune and inflammation-related genes in liver as determined by microarray. (D) Heat map showing changes in expression of selected cytokines, chemokines, and adipokines in epididymal white adipose tissue as determined by microarray. The selected genes exceeded minimal threshold expression level. (E) Expression of IL-1β as determined by qPCR. Differences were evaluated using Student t test. *P < 0.05. (F) Expression of Cd68, F4/80, and IL-1β was assessed in hepatocyte and Kupffer cell enriched fractions of mouse liver using qPCR. Error bars represent SE.

Download figure to PowerPoint

Kupffer Cell Depletion Reduces Hepatic Steatosis by Activating PPARα.

To examine the effect of Kupffer cell depletion on hepatic steatosis, liver triglycerides were measured. Interestingly, quantitative analysis showed a significant reduction in hepatic triglyceride content in mice treated with clodronate liposomes (Fig. 3A), which was confirmed by histological staining using oil red O or hematoxylin and eosin (Fig. 3B,C). Furthermore, clodronate liposomes significantly reduced hepatic content of glucosylceramide (Fig. 3D), whereas ceramide content remained unchanged (Fig. 3E). Because clodronate liposomes were administered after 19 weeks of HFD, the data suggest that Kupffer cell depletion leads to dissipation of existing triglyceride stores. A possible explanation is via stimulation of hepatic fatty acid oxidation. To explore such an effect, we measured the expression of a number of key genes involved in fatty acid oxidation. Consistent with increased fatty acid oxidation, expression of PPARα and its targets carnitine palmitoyltransferase 1a (Cpt1a), Fatp2, and fatty acid binding protein 1 (Fabp1) was significantly increased in mice treated with clodronate liposomes (Fig. 4A). No effect of clodronate liposomes was observed on expression of genes involved in lipogenesis (Fasn, Pparg), whereas expression of genes involved in lipid export was either not changed (Apob) or slightly reduced (Mttp) upon clodronate treatment (Fig. 4B). Treatment with clodronate liposomes did not alter plasma free fatty acid levels (Fig. 4C), suggesting a lack of effect on adipose tissue lipolysis, and caused a trend toward a decrease in plasma triglyceride, although the effect was not significant (Fig. 4D). Plasma glucose levels were not affected by clodronate liposomes (Fig. 4E). Overall, these results suggest that Kupffer cell depletion reduces hepatic triglyceride storage by stimulating PPARα-mediated fatty acid oxidation in hepatocytes.

thumbnail image

Figure 3. Kupffer cell depletion reduces hepatic triglyceride content. (A) Quantitative measurement of liver triglycerides. Representative oil red O (B) and hematoxylin-eosin (C) staining of liver. Quantitative measurement of (D) hepatic glucosylceramide and (E) ceramide. Error bars represent SE. Differences between HFO-PBS and HFO-clodronate were evaluated using Student t test. *P < 0.05.

Download figure to PowerPoint

thumbnail image

Figure 4. Kupffer cell depletion increases expression of PPARα and its targets. (A) Changes in hepatic expression of PPARα and its target genes, as determined by qPCR. (B) Changes in hepatic expression of genes involved in lipogenesis and lipid export, as determined by qPCR. Gene expression levels in mice fed LFD were set at 1. (C) Plasma free fatty acid levels. (D) Plasma triglyceride levels. (E) Plasma glucose levels. Error bars represent SE. Differences between HFD-PBS and HFD-clodronate were evaluated using Student t test. *P < 0.05.

Download figure to PowerPoint

IL-1β Negatively Interferes with PPARα Action.

We next addressed the potential mechanism underlying the impact of Kupffer cells on PPARα-dependent gene regulation, focusing on factors derived from Kupffer cells. Because Kupffer cell depletion reduced expression of IL-1β, it was of interest to study the effect of IL-1β on PPARα messenger RNA (mRNA) and PPARα-dependent gene regulation in the hepatocyte. To that end, mice were treated with IL-1β via intraperitoneal injection. Compared to control treatment, IL-1β significantly decreased expression of PPARα as well as several of its target genes, including Acyl-CoA oxidase and Cpt1a, whereas expression of the positive control gene serum amyloid A was dramatically induced by IL-1β (Fig. 5A). In agreement with these data, IL-1β down-regulated expression of PPARα and its target genes Cpt1a and cytochrome P450 4A14 (Cyp4A14) in primary mouse hepatocytes (Fig. 5B). Cpt1a and Cyp4A11 but not PPARα were also down-regulated by IL-1β in primary human hepatocytes (Fig. 5C). These data suggests that IL-1β negatively interferes with PPARα action in both human and mouse primary hepatocytes, leading to reduced expression of PPARα target genes.

thumbnail image

Figure 5. IL-1β reduced PPARα and target gene expression. (A) Expression of PPARα, Acox1, Cpt1, and Saa3 was determined in livers collected 16 hours after intraperitoneal injection with IL-1β. Gray bars = saline, black bars = IL-1b. N = 3 per group. Expression of PPARα, Cyp4a, and Cpt1 was determined in primary mouse (B) or human (C) hepatocytes treated with IL-1β (5 ng/mL) for 24 hours. Error bars represent SE. *Significantly different from control-treated cells (P < 0.05).

Download figure to PowerPoint

IL-1β Inhibits PPARα Promoter Activity Via NF-κB.

To elucidate the mechanism underlying down-regulation of PPARα mRNA by IL-1β, transactivation studies were carried out using the human and mouse PPARα promoter placed in front of a luciferase reporter. Treatment with IL-1β markedly decreased human and mouse PPARα promoter activity, whereas TNFα and IL-6 had little or no effect (Fig. 6A). Because IL-1β is a potent inducer of NF-κB activity (Fig. 6B), we studied the effect of the NF-κB subunits p50 and p65 on PPARα promoter activity. Only a minor reduction in promoter activity was observed upon overexpression of p50. In contrast, p65 overexpression significantly decreased human and mouse PPARα promoter activity, which was further reduced by p50 (Fig. 6C).

thumbnail image

Figure 6. IL-1β suppresses PPARα promoter activity via NF-κB. (A) HepG2 cells were transfected with human and mouse PPARα promoter reporter constructs followed by treatment with IL-6, TNFα, or IL-1β (5 ng/mL). After 24 hours, luciferase activity was determined in the cell lysates. Luciferase activity was related to β-galactosidase activity to correct for differences in transfection efficiency. Reporter activity from untreated cells was set to 1. (B) HepG2 cells were transfected with the NF-κB reporter construct followed by treatment with IL-6, TNFα, or IL-1β (5 ng/mL). (C) HepG2 were transfected with the human and mouse PPARα promoter reporter construct together with expression vectors for NF-κB subunits p50 and p65 followed by treatment with IL-1β (5 ng/mL). (D) Sequence comparison of the consensus NF-κB binding site and the identified NF-κB binding sites within the human PPARα promoter. (E) HepG2 were transfected with human PPARα promoter deletion constructs together with expression vectors for p50 and p65. Error bars represent SE. *Significantly different from control-treated cells (P < 0.05).

Download figure to PowerPoint

Previously, it was shown that the human PPARα promoter contains two NF-κB binding sites located at −646 base pairs and −1016 base pairs from the transcription start site (Fig. 6D).21 To study whether these NF-κB sites are involved in inhibition of human PPARα promoter activity by IL-1β, deletion constructs were prepared lacking one or both NF-κB response elements. Importantly, deletion of one or both NF-κB elements strongly abrogated suppression of PPARα promoter activity by p50 and p65 (Fig. 6E), suggesting that both NF-κB elements are involved in the negative regulation of PPARα promoter activity. Together, our data show that IL-1β inhibits PPARα promoter activity by stimulating NF-κB.

In human hepatocytes, IL-1β markedly reduced expression of PPARα target genes without changing expression of PPARα itself, suggesting possible inhibition of PPARα activity by IL-1β. To further explore IL-1β-dependent inhibition of PPARα activity, we performed transactivation assays in HepG2 cells using a reporter construct containing three copies of the PPAR response element (PPRE) in front of the luciferase gene. Reporter activity, which serves as a marker of PPARα activity, was markedly induced by transfection of PPARα and treatment with Wy14643 (Fig. 7A). Consistent with inhibition of PPARα activity by IL-1β, treatment of the cells with IL-1β significantly reduced reporter activity (Fig. 7A). Similar reduction in luciferase activity results were observed upon transfection of p50 and p65 (Fig. 7B), suggesting that the inhibitory effect of IL-1β on PPARα activity is mediated by NF-κB. These data indicate that IL-1β is able to suppress PPARα activity independently of changes in PPARα expression.

thumbnail image

Figure 7. NF-κB suppresses PPARα-activity. (A) HepG2 cells were co-transfected with a PPRE-tk-LUC reporter construct together with a PPARα expression vector followed by treatment with IL-1β (5 ng/mL) and/or the PPARα agonist Wy14643. After 24 hours, luciferase activity was determined in the cell lysates. *Significantly different from cells not treated with IL-1β (P < 0.05). B) HepG2 cells were cotransfected with the PPRE-tk-LUC reporter together with expression vectors for PPARα, p50 and/or p65 followed by treatment with Wy14643. After 24 hours, luciferase activity was determined in the cell lysates. Luciferase activities were related to β-galactosidase activity to correct for differences in transfection efficiency. Reporter activity from untreated cells was set to 1. *Significantly different from cells not transfected with p50 and p65 (P < 0.05). Error bars represent SE.

Download figure to PowerPoint

Discussion

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

Because Kupffer cells mediate the hepatic response to numerous inflammatory stimuli, these cells may be suspected to play a role in the progression of steatosis to steatohepatitis.22 Additionally, it can be hypothesized that via production of specific cytokines, Kupffer cells may influence lipid metabolism and accordingly lipid storage in hepatocytes.23 Interestingly, it has been reported that Kupffer cell products and specifically IL-1β stimulate very low density lipoprotein–apolipoprotein B (VLDL-apoB) and lipid secretion in cultured hepatocytes.24 Similarly, IL-6 treatment was shown to reduce hepatic lipid storage by increasing mitochondrial fatty acid oxidation as well as hepatic export of triglyceride and cholesterol.25 In this article, we show that Kupffer cell depletion leads to reduced liver steatosis in the context of chronic high-fat feeding in mice, indicating a stimulatory effect of Kupffer cells on hepatic triglyceride storage. This effect is achieved via IL-1β–dependent inhibition of PPARα expression and activity, leading to decreased expression of PPARα target genes involved in fatty acid catabolism, including Cpt1a. Kupffer cell depletion did not significantly alter expression of genes involved in lipogenesis or lipid export.

The observed suppressive effect of clodronate liposomes on lipid storage in mice rendered steatotic by chronic high-fat feeding parallels a similar finding in mice fed a methionine-deficient and choline-deficient diet.26 In the latter study, Kupffer cell depletion reduced histological evidence of steatosis and steatohepatitis, as well as the expression of marker genes of fibrosis and inflammation. Furthermore, inhibition of Kupffer cell using GdCl3 lowered hepatic steatosis in mice fed a HFD.27 Together, these studies underscore the crucial role of Kupffer cells in the development of NAFLD.

Down-regulation of PPARα is known to lead to hepatic triglyceride accumulation.28, 29 Furthermore, it has been shown that moderate changes in Cpt1a activity are sufficient to substantially alter hepatic triglycerides via changes in fatty acid oxidation.30 We found that the inhibitory effect of IL-1β on PPARα expression is mediated by NF-κB binding to specific binding sites in the promoter of the PPARα gene. Our data demonstrate an important link between Kupffer cells and hepatocytes in the regulation of hepatic lipid metabolism and point toward Kupffer cell–derived IL-1β as a major modulator of PPARα expression and activity.

A growing body of evidence supports the role of PPARα in the development of NAFLD and as a putative target for the treatment of steatohepatitis. Disabling the PPARα gene is known to increase hepatic triglyceride accumulation, especially under conditions of fasting.31–33 By reducing steatosis and by direct down-regulation of inflammatory gene expression, PPARα protects against obesity-induced chronic inflammation in liver.29 Furthermore, pharmacological PPARα activation has been shown to lower hepatic triglyceride levels and effectively attenuate steatohepatitis.34–38 However, a clear benefit of PPARα agonists on liver histology in human patients with NAFLD has yet to be demonstrated.39

Previously, it was shown that lipopolysaccharide (LPS) inhibits hepatic PPARα expression and negatively interferes with PPARα-dependent gene activation.40 A follow-up study suggested that the effects of LPS on hepatic PPARα activity were likely mediated by IL-1β and TNFα.41 Specifically, treatment of mice with IL-1β and TNFα decreased retinoid X receptor α, PPARα, and PPAR-gamma coactivator 1α mRNA in the liver. In cultured Hep3B cells, IL-1β and TNFα reduced binding of nuclear extracts to PPAR response elements and decreased PPARα-driven reporter activity. As a consequence, expression of Cpt1a was decreased in Hep3B cells treated with IL-1β.41 Furthermore, it has been shown that IL-1β represses induction of PPARα mRNA elicited by clofibrate and dexamethasone.42 Recently, it was observed that IL-1β reduces PPARα mRNA in HepG2 cells and primary mouse hepatocytes. Our data confirm these findings and additionally indicate that suppression of PPARα expression by IL-1β is mediated by NF-κB binding to specific binding sites in the promoter of the PPARα gene.

Although inflammation is generally considered to follow steatosis in the induction of NASH, in certain mouse models inflammation was found to precede lipid accumulation.7 Interestingly, recent data indicate that inflammatory cytokines can stimulate lipid accumulation in cultured hepatocytes, thereby exacerbating hepatic steatosis in vivo.43 Our data suggest that Kupffer cells and thus Kupffer cell–derived cytokines also have a general stimulatory effect on liver steatosis, which may be primarily mediated by IL-1β.

It has been suggested that lipotoxic intermediates, which include ceramides, play a key role in the progression of steatosis to NASH. Recent studies indicate that ceramides also influence lipid storage.6 Indeed, it has been shown that inhibition of ceramide or glucosylceramide synthesis reduces hepatic steatosis.44–46 In our study, the increase in PPARα-dependent fatty acid oxidation following Kupffer cell depletion can be hypothesized to lower the availability of fatty acid substrate for the synthesis of ceramides. Although hepatic ceramide content was not affected by Kupffer cell depletion, glucosylceramide content was markedly decreased. Because impaired synthesis of glucosylceramide is associated with decreased steatosis, the suppressive effect of Kupffer cell depletion on hepatic lipid storage may be partially accounted for by the reduction in glucosylceramide.

Although NF-κB predominantly acts via activation of gene transcription, evidence is growing that NF-κB can also suppress transcription of genes.47, 48 Furthermore, studies have indicated that NF-κB may down-regulate cellular responsiveness to specific cytokines.49, 50 Our data indicate that suppression of PPARα promoter activity by NF-κB subunits p50 and p65 is dependent on two NF-κB binding sites present within the PPARα promoter. How binding of p50 and p65 leads to down-regulation of PPARα promoter activity is unclear, but the mechanism may involve interference with binding of other transcription factors including HNF4α as well as PPARα itself, which are essential to maintain normal PPARα expression levels.21 Besides down-regulating PPARα promoter activity and expression, IL-1β and NF-κB also suppressed PPARα-dependent reporter activity independent of changes in PPARα expression, suggesting interference of NF-κB with normal PPARα signaling. Previously it has been shown that PPARα is able to physically interact with NF-κB and prevent its activation,51–53 which importantly contributes to the anti-inflammatory activity of PPARα. Our present data suggest that the physical interaction between PPARα and NF-κB also negatively impacts PPARα-dependent gene regulation.

Macrophages take up clodronate liposomes via endocytosis, leading to apoptosis. Depending on the method of administration, different macrophage populations throughout the body can be targeted by clodronate treatment. Intraperitoneal injection has been shown to selectively destroy liver resident macrophages, as well as macrophages present in the peritoneal cavity, bone marrow, and spleen.54 Although formal evidence excluding a role of nonhepatic macrophages in the observed reduction in hepatic triglycerides and activation of PPARα activity is lacking, our studies using IL-1β in vivo and in vitro support the effects of Kupffer cell–derived cytokines on PPARα activity and expression, thereby providing a plausible mechanism for the effect of clodronate treatment on liver triglycerides.

In conclusion, our data point toward important cross-talk between Kupffer cells and hepatocytes in the regulation of hepatic triglyceride storage. The effect of Kupffer cells on liver triglycerides is at least partially mediated by IL-1β, which potently suppresses PPARα expression and activity via NF-κB–dependent inhibition of PPARα promoter activity.

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

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
HEP_23337_sm_SupTab.doc25KSupplementary Table

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.