SEARCH

SEARCH BY CITATION

Abstract

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

The importance of chemokines in alcoholic liver injury has been implicated. The role of the chemokine, monocyte chemoattractant protein-1 (MCP-1), elevated in patients with alcoholic liver disease is not yet understood. Here, we evaluated the pathophysiological significance of MCP-1 and its receptor, chemokine (C-C motif) receptor 2 (CCR2), in alcoholic liver injury. The Leiber-DeCarli diet containing alcohol or isocaloric control diets were fed to wild-type (WT) and MCP-1-deficient knockout (KO) mice for 6 weeks. In vivo and in vitro assays were performed to study the role of MCP-1 in alcoholic liver injury. MCP-1 was increased in Kupffer cells (KCs) as well as hepatocytes of alcohol-fed mice. Alcohol feeding increased serum alanine aminotransferase in WT and CCR2KO, but not MCP-1KO, mice. Alcohol-induced liver steatosis and triglyceride were attenuated in alcohol-fed MCP-1KO, but high in CCR2KO mice, compared to WT, whereas serum endotoxin was high in alcohol-fed WT and MCP-1KO mice. Expression of liver proinflammatory cytokines tumor necrosis factor alpha, interleukin (IL)-1β, IL-6, KC/IL-8, intercellular adhesion molecule 1, and cluster of differentiation 68 was induced in alcohol-fed WT, but inhibited in MCP-1KO, mice independent of nuclear factor kappa light-chain enhancer of activated B cell activation in KCs. Oxidative stress, but not cytochrome P450 2E1, was prevented in chronic alcohol-fed MCP-1KO mice, compared to WT. Increased expression of peroxisome proliferator-activated receptor (PPAR)α and PPARγ was accompanied by nuclear translocation, DNA binding, and induction of fatty acid metabolism genes acyl coenzyme A oxidase and carnitine palmitoyltransferase 1A in livers of alcohol-fed MCP-1KO mice, compared to WT controls. In vitro assays uncovered an inhibitory effect of recombinant MCP-1 on PPARα messenger RNA and peroxisome proliferator response element binding in hepatocytes independent of CCR2. Conclusion: Deficiency of MCP-1 protects mice against alcoholic liver injury, independent of CCR2, by inhibition of proinflammatory cytokines and induction of genes related to fatty acid oxidation, linking chemokines to hepatic lipid metabolism. (HEPATOLOGY 2011)

Alcoholic liver disease (ALD) is a major health concern, and approximately 90% of heavy drinkers develop fatty liver disease or steatosis. Fatty liver is occasionally accompanied by, or progresses to, inflammation in human ALD. The essential role of innate immune cell activation and circulating endotoxin/lipopolysaccharide (LPS) in ALD has been proposed.1, 2 Circulating endotoxin activates liver macrophages and leads to the induction of cytokines, chemokines, and reactive oxygen species.3 Though proinflammatory cytokine production in the alcoholic liver is extensively investigated, the importance of chemokines is still unknown. Elevation of chemokines, such as interleukin (IL)-8, monocyte chemoattractant protein-1 (MCP-1) and macrophage inflammatory protein 1 (MIP-1), in alcoholic hepatitis and cirrhotic patients and the correlation with the recruitment of polymorphonuclear leukocytes is reported.4, 5 However, the pathophysiological mechanisms affected by chemokines in ALD are yet to be determined. CC-chemokines induce the recruitment and activation of mononuclear cells, such as monocytes/macrophages, T cells and natural killer T cells,6, 7 and these cells play an important role in the development and propagation of alcoholic liver injury.8

MCP-1 or chemokine (C-C motif) ligand 2 (CCL2), an important CC-chemokine, recruits and activates monocytes/macrophages to the site of tissue injury and regulates adhesion molecules and proinflammatory cytokines tumor necrosis factor alpha (TNFα), IL-1β, and IL-6.9, 10 The pivotal role of MCP-1 in alcoholic liver injury was first recognized by studies showing higher amounts of MCP-1, as compared to other CC-chemokines, such as MIP-1α and MIP-1β, in the liver and mononuclear cells of patients with alcoholic hepatitis.4, 5 Subsequently, the pathogenic role of MCP-1 expressed by liver macrophages and endothelial cells was demonstrated in rodent models of alcoholic hepatitis.11 Besides macrophage activation, MCP-1 appears to play a significant role in hepatic steatosis or early liver injury. Recently, transgenic mice overexpressing MCP-1 in adipose tissue exhibited insulin resistance and increased hepatic triglyceride content.12 These studies were based on the observations that mice fed a high-fat diet led to MCP-1 induction in adipose tissue, but not liver.12In vitro studies also demonstrated that MCP-1 can induce lipid accumulation in hepatocyte cultures.13 In general, MCP-1 seems to play an important role in hepatic inflammatory responses and steatosis during tissue injury.

Previous studies from our laboratory and others have shown the pathophysiological importance of proinflammatory cytokines in ALD.1, 2, 14 However, the pathophysiological role of chemokines, such as MCP-1, in alcoholic liver injury is still uncertain. Based on the preferential elevation of MCP-1 among other CC-chemokines in alcoholic hepatitis patients4, 5 and its importance in the modulation of proinflammatory cytokines,9, 10 we hypothesized that MCP-1 contributes to chronic alcoholic liver injury and steatosis via the modulation of inflammatory cytokines. Using MCP-1-deficient mice, we sought to investigate whether MCP-1 and its receptor, chemokine (C-C motif) receptor 2 (CCR2), play a causative role in alcoholic liver injury.

Materials and Methods

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

Additional descriptions of methods are available in the Supporting Information.

Animal Studies.

All animals received proper care in agreement with animal protocols approved by the Institutional Animal Use and Care Committee of the University of Massachusetts Medical School (Worcester, MA). Six- to eight-week-old female wild-type (WT) (C57BL/6) and MCP-1-deficient and CCR2-deficient mice (all generated on a C57BL/6 background; The Jackson Laboratory, Bar Harbor, ME) received Lieber-DeCarli diet (Bio-Serv, Frenchtown, NJ) with 5% (v/v) ethanol (36% ethanol-derived calories) for 6 weeks; pair-fed control mice received an equal amount of calories as their alcohol-fed counterparts with the alcohol-derived calories substituted with dextrin maltose. All strains of mice consumed comparable daily calories. In some cases, mice from both the alcohol-fed and pair-fed groups were administered an intraperitoneal (IP) injection of either 0.2 mL of 0.9% saline (phosphate-buffered, pH 7.4) alone as a vehicle control or 0.2 mL 0.9% saline containing 0.5 mg/kg of nonpurified lipopolysaccharide (LPS; from Escherichia coli 0111:B4; Sigma-Aldrich, St. Louis, MO) and sacrificed 2 hours later. There were 9-12 mice in each experimental group. Serum was separated from whole blood and frozen at −80°C. Liver tissue was rapidly excised and a portion was snap-frozen in liquid nitrogen and stored at −80°C. Additional portions of the livers were stored in RNA stabilization reagent (RNAlater; Qiagen, Valencia, CA) for RNA extraction or fixed in 10% neutral-buffered formalin for histopathological analysis.

Statistical Analysis.

Statistical significance was determined by analysis of variance (in vivo) and t-test (in vitro) using the GraphPad Prism 5.01 software (GraphPad Software, Inc., La Jolla, CA). Data are shown as mean ± standard error of the mean (SEM) and were considered statistically significant at P < 0.05.

Results

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

Chronic Alcohol Consumption Induces MCP-1 in Kupffer Cells and Hepatocytes.

MCP-1 is increased during ALD; however, its cellular source in the liver is not yet identified. Here, C57Bl/6 mice were fed the Leiber-Decarli alcohol diet or its isocaloric control (pair-fed) diet to determine the expression of MCP-1 in the liver. Chronic alcohol feeding for 6 weeks induced MCP-1 messenger RNA (mRNA) (Fig. 1A) and protein (Fig. 1B) in whole livers, compared to pair-fed controls. Next, to identify the cell types expressing MCP-1, we isolated hepatocytes and Kupffer cells (KCs) and estimated MCP-1 mRNA. Figure 1C shows that isolated hepatocytes as well as KCs express high amounts of MCP-1 mRNA in chronic alcohol-fed mice, compared to isocaloric pair-fed controls, with similar expression levels of baseline MCP-1 in hepatocytes relative to KCs (Supporting Fig. 1). Expression analysis of the CC-chemokine gene family revealed a significant increase in CCL4/MIP-1β and KC/IL-8/chemokine (C-X-C motif) ligand 1, with a maximal elevation in MCP-1 in livers of chronic alcohol-fed mice, compared to pair-fed controls (Fig. 1D).

thumbnail image

Figure 1. MCP-1 is induced in hepatocytes and KCs during chronic alcohol exposure. C57Bl/6 mice were fed 5% alcohol-containing Leiber-DeCarli and isocaloric pair-fed diet for 6 weeks. (A) Total RNA from liver tissue was subjected to real-time quantitative polymerase chain reaction (qPCR) for determination of MCP-1 mRNA (*P < 0.04, compared to pair-fed control; n = 12). (B) Total liver MCP-1 protein levels were analyzed in tissue extracts by enzyme-linked immunosorbent assay (ELISA) of alcohol-fed and pair-fed mice (**P < 0.045, compared to pair-fed control; n = 12). (C) Hepatocytes and KCs isolated from alcohol-fed and pair-fed mice were subjected to real-time qPCR for determination of MCP-1 mRNA (*P < 0.05, compared to corresponding pair-fed control; n = 9). (D) Total RNA from liver tissue was subjected to real-time qPCR for determination of chemokine mRNA (*P < 0.05, compaired to pair-fed control; n = 9). Values depicted in the graph are mean fold change ± SEM.

Download figure to PowerPoint

MCP-1 Deficiency Protects Against Alcoholic Liver Injury.

To investigate the role of MCP-1 in ALD, WT and MCP-1 knockout (MCP-1KO) mice were fed the Leiber-DeCarli diet with 5% ethanol or isocaloric control diet for 6 weeks to induce ALD. Prolonged alcohol feeding resulted in liver injury, as assessed by significantly increased serum alanine aminotransferase (ALT) levels (Fig. 2A) and higher liver/body-weight ratio (Supporting Fig. 2A) in alcohol-fed WT mice, compared to pair-fed controls and MCP-1KO mice. Despite no liver damage, serum alcohol levels in MCP-1KO were comparable to alcohol-fed WT mice (Supporting Fig. 2B). Histological analysis showed micro- and macrosteatosis in chronic alcohol-fed WT mice, whereas fat deposition was not detectable in pair-fed controls and MCP-1KO mice (Fig. 2B). In agreement with the histological data, liver triglyceride levels were significantly higher in alcohol-fed WT mice, compared to pair-fed controls and MCP-1KO mice (Fig. 2C). Furthermore, chronic alcohol-fed WT and MCP-1KO mice showed significantly increased serum endotoxin, compared to pair-fed controls (Fig. 2D). Collectively, these data show that chronic alcohol feeding induces liver damage in WT mice, and regardless of high blood alcohol levels and elevated endotoxin, MCP-1KO mice are protected from alcoholic liver injury.

thumbnail image

Figure 2. MCP-1 deficiency protects against alcohol-induced liver injury. MCP-1deficient (MCP-1−/−) or WT mice were fed an isocaloric pair-fed diet or 5% alcohol-containing Leiber-DeCarli diet for 6 weeks and (A) serum ALT was analyzed (*P < 0.001; n = 12, compared to pair-fed controls), (B) liver sections were fixed in formalin and stained with hematoxylin and eosin (H&E) (magnification, 100×), (C) liver triglycerides (mg/g liver tissue) were measured (*P < 0.001, compared to pair-fed control), and (D) serum endotoxin was analyzed (*P < 0.046; n = 9, compared to pair-fed control). Values shown in the graph are mean ± SEM (9-12 mice per group).

Download figure to PowerPoint

Table 1. Real-Time Polymerase Chain Reaction Primers
Target GenesForward Primer (5′ → 3′)Reverse Primer (5′ → 3′)
  1. Abbreviations: TNFα, tumor necrosis factor alpha; IL, interleukin; MCP, monocyte chemoattractant protein; KC, Kupffer cell; MIP, macrophage inflammatory protein; CCL, chemokine (C-C motif) ligand; CCR, chemokine (C-C motif) receptor; ICAM, intercellular adhesion molecule; VCAM, vascular cell adhesion molecule; CD, cluster of differentiation; TLR4, Toll-like receptor 4; PPAR, peroxisome proliferator-activated receptor.

TNFαcac cac cat caa gga ctc aaagg caa cct gac cac tct cc
IL-1βct ttg aag ttg acg gac cctga gtg ata ctg cct gcc tg
IL-6aca acc acg gcc ttc cct act tcac gat ttc cca gag aac atg tg
MCP-1cag gtc cct gtc atg ctt ctcag gtc cct gtc atg ctt ct
KC/IL-8gga agt gtg atg act cag gtt tgcgat ggt tcc ttc cgg tgg ttt ctt c
CCL3/MIP-1αtct cag cgc cat atg gag ctttc cgg ctg tag gag aag ca
CCL4/MIP-1βccg agc aac acc atg aag ccca ttg gtg ctg aga acc ct
CCL5/RANTESgct gct ttg cct acc tct cctcg agt gac aaa cac gac tgc
CCL8/MCP-2cca gat aag gct cca gtc acc tggc act gga tat tgt tga ttc tct c
CCL7/MCP-3gca gag aag caa ggc cag cac aagc agg cac aga agc gtg gc
CCR1tag tga act tgg acc cca ggtca agg ttc aag gtc cca ac
CCR2gtg tac ata gca aca agc ctc aaa gccc cca cat agg gat cat ga
ICAMSA Biosciences catalog no. PPM03196A
VCAMSA Biosciences catalog no. PPM03208B
F4/80tgc atc tag caa tgg aca gcgcc ttc tgg atc cat ttg aa
CD68ccc aca ggc agc aca gtg gactcc aca gca gaa gct ttg gcc c
TLR4gcc ttt cag gga att aag ctc caga tca acc gat gga cgt gta a
PPARαaac atc gag tgt cga ata tgt ggagc cga ata gtt cgc cga aag
PPARγgga aga cca ctc gca ttc ctttcg cac ttt ggt att ctt gga g

Chronic Alcohol Induced Up-regulation of Proinflammatory Cytokines Is Prevented by MCP-1 Deficiency Independent of Nuclear Factor Kappa Light-Chain Enhancer of Activated B Cell Activation.

MCP-1 plays an important role in the induction of proinflammatory cytokines at the site of tissue injury.10 Here, we investigated the effect of MCP-1 deficiency on alcohol-induced expression of cytokines in the liver. We elucidated the expression of circulating endotoxin (baseline)-mediated induction of proinflammatory cytokines TNFα, IL-1β, and IL-6, as well as CC-chemokine mRNA levels in liver of alcohol-fed WT and MCP-1KO mice. Here, we show that TNFα, IL-1β, and IL-6 mRNA was increased significantly in alcohol-fed WT mice, compared to pair-fed WT controls, whereas alcohol-fed MCP-1KO mice were unable to induce proinflammatory cytokine mRNA in the liver (Fig. 3A). MCP-1 deficiency also prevented chronic alcohol-induced liver tissue TNFα, as compared to WT mice (Fig. 3B). Interestingly, among CC-chemokine genes, KC/IL-8 mRNA was significantly decreased, but CCL4/MIP-1β and CCL5/RANTES mRNA was high in alcohol-fed MCP-1KO mice, compared to pair-fed controls (Fig. 3C). Furthermore, investigation of MCP-1-mediated adhesion molecules and macrophage markers demonstrated a significant induction of intercellular adhesion molecule 1 (ICAM-1) and cluster of differentiation (CD)68, but unchanged vascular cell adhesion molecule 1 (VCAM-1) and F4/80 in livers of alcohol-fed WT, but not MCP-1KO, mice (Fig. 3D). Because nuclear factor kappa light-chain enhancer of activated B cells (NF-κB) is important in chronic alcohol-mediated proinflammatory cytokine production and macrophage activation,15 we next determined whether the inhibition of inflammatory cytokines was regulated by the lack of NF-κB activation in MCP-1-deficient mice. Interestingly, our results show that NF-κB binding activity in whole livers was significantly increased in alcohol-fed MCP-1-deficient mice (Fig. 3E), compared to alcohol-fed WT and pair-fed MCP-1KO mice. Furthermore, increased NF-κB activation was observed in isolated KCs of alcohol-fed MCP-1KO and WT mice, compared to pair-fed controls (Fig. 3F). Immunohistochemical analysis revealed NF-κB p65 staining in nonparenchymal cells of alcohol-fed WT and MCP-1KO mice (Supporting Fig. 3). On the other hand, isolated hepatocytes showed decreased NF-κB activation in alcohol-fed WT mice, compared to pair-fed controls, and this inhibition was prevented in alcohol-fed MCP-1KO mice (Fig. 3F), likely contributing to NF-κB-mediated hepatocyte survival in alcohol-fed MCP-1KO mice. These results indicate that liver proinflammatory cytokine mRNA, ICAM-1, and CD68 are significantly decreased in chronic alcohol-fed MCP-1KO mice, compared to their WT counterparts, in an NFκB-independent manner.

thumbnail image

Figure 3. Chronic alcohol-induced inflammatory cytokine production is prevented in MCP-1−/− mice. Chronic alcohol-fed WT and MCP-1−/− mice were subjected to analysis of (A) TNFα mRNA (*P < 0.05, compared to pair-fed control; n = 9), IL-1β mRNA (*P < 0.05, compared to pair-fed control n = 9), and IL-6 mRNA (*P < 0.05, compared to pair-fed control; n = 9). (B) Tissue TNFα estimated by ELISA (*P < 0.05, compared to pair-fed control; n = 9) and (C) Total RNA from liver tissue was subjected to real-time qPCR for determination of chemokine mRNA (*P < 0.05; **P < 0.001; #P < 0.01, compared to pair-fed controls; n = 9). (D) Total RNA from liver tissue was subjected to real-time qPCR for determination of adhesion molecules ICAM-1 and VCAM-1 and the macrophage markers, F4/80 and CD68 mRNA (*P < 0.04, n = 7; #P < 0.03, n = 9). (E) NFκB DNA-binding activity in whole livers (*P < 0.05; **P < 0.01; #P < 0.001, compared to corresponding pair-fed control (n = 8) (F) KCs (left panel) (*P < 0.05; n = 6, compared to pair-fed control) and primary hepatocytes (right panel) (*P < 0.03, compared to pair-fed control; n = 6) were analyzed by EMSA showing representative gel (upper panels). To confirm specificity, a 20-fold excess of unlabeled oligonucleotide was included as a cold competitor (Comp). Density units of the upper and lower bands together are shown in the bar graph (lower panel). Values shown in graphs are mean ± SEM. ns, nonsignificant.

Download figure to PowerPoint

Oxidative Stress and Sensitization to LPS-Induced Proinflammatory Cytokine Production Is Inhibited by MCP-1 Deficiency.

The classical feature of alcoholic liver injury is alcohol-mediated oxidative stress and increased sensitization to LPS, resulting in enhanced proinflammatory cytokine expression in the liver.1, 16 To further test the effect of sensitization to LPS in chronic alcohol-fed MCP-1-deficient mice, an in vivo LPS challenge (0.5 mg/kg body weight; IP) was administered at the end of the chronic alcohol feeding. Our results show that in vivo LPS challenge increased proinflammatory cytokine, TNFα, IL-1β, and IL-6 mRNA in liver of WT mice, as compared to pair-fed controls, and this induction was prevented in chronic alcohol-fed MCP-1KO mice (Fig. 4A). Interestingly, no changes were observed in Toll-like receptor 4 (TLR4) expression, a receptor for LPS (Supporting Fig. 4). We next determined whether MCP-1 deficiency would affect alcohol-induced oxidative stress and alcohol-metabolizing enzyme cytochrome P450 2E1 (CYP2E1) in the liver. Chronic alcohol-induced oxidative stress, as illustrated by increased thiobarbituric acid-reactive substances (TBARS) in WT mice, was significantly blunted in alcohol-fed MCP-1KO mice (Fig. 4B). However, CYP2E1 levels estimated in liver microsomal preparations from alcohol-fed WT and MCP-1KO mice remained similar (Fig. 4C). These results suggest that MCP-1 contributes to chronic alcohol-induced oxidative stress in a CYP2E1-independent fashion and sensitizes the liver to LPS, resulting in enhanced proinflammatory cytokine production.

thumbnail image

Figure 4. MCP-1 deficiency prevents oxidative stress and inhibits sensitization to LPS in chronic alcohol-fed mice. Chronic alcohol-fed WT and MCP-1−/− mice were injected with LPS, and livers were subjected to analysis of mRNA by real-time PCR after 2 hours for (A) TNFα mRNA (*P < 0.01, compared to pair-fed control; n = 9), IL-6 mRNA (*P < 0.05; n = 9), and IL-1β mRNA (*P < 0.02; n = 9). (B) Livers from WT and MCP-1−/− mice were subjected to analysis of TBARS, a marker of oxidative stress (**P < 0.01, compared to pair-fed). (C) CYP2E1 was detected in liver microsomal fractions by western blotting. Anticalnexin antibody was used as an internal loading control. Values shown in the graphs are mean ± SEM.

Download figure to PowerPoint

Chronic Alcohol Feeding Induces Liver Injury in CCR2-Deficient Mice.

MCP-1 is known to mediate inflammatory cell activation in the liver via its receptor, CCR2.17 The importance of CCR2, predominantly expressed in monocyte/macrophage cells, is shown in liver diseases, such as fibrosis.18 To investigate the role of CCR2 in alcoholic liver injury, we fed CCR2-deficient mice (CCR2KO) with the Leiber-DeCarli diet containing 5% ethanol for 6 weeks. Similarly to WT mice, alcohol feeding increases serum ALT in CCR2KO mice, indicating liver damage in the absence of CCR2 (Fig. 5A). Furthermore, histological examination showed that micro- and macrosteatosis were observed in alcohol-fed WT and CCR2KO mice, compared to pair-fed controls (Fig. 5B). Quantitation of liver triglycerides exhibited significantly high levels in alcohol-fed WT and CCR2KO mice, compared to pair-fed mice (Fig. 5C), supporting histological findings. Thus, it is evident that chronic alcohol feeding induces liver injury irrespective of the absence of CCR2, and suggests that MCP-1-mediated protection from alcoholic liver injury is independent of CCR2.

thumbnail image

Figure 5. CCR2 deficiency does not prevent against alcoholic liver injury. CCR2-deficient (CCR2−/−) or WT mice were fed an isocaloric pair-fed diet or 5% alcohol-containing Leiber-DeCarli diet for 6 weeks. (A) Serum ALT was analyzed (*P < 0.002; n = 12, compared to pair-fed controls), (B) Liver sections were fixed in formalin and stained with H&E (magnification, 100×), and (C) liver triglycerides (mg/g liver tissue) were measured (*P < 0.02, compared to pair-fed control; n = 12). Values shown in the graph are mean ± SEM.

Download figure to PowerPoint

PPARα and PPARγ mRNA Expression, DNA Binding, and Target Genes Are Increased in Chronic-Alcohol Fed MCP-1KO Mice.

Having observed inhibitory effects on inflammatory responses in the liver, we next wanted to determine whether the decrease in hepatic steatosis in alcohol-fed MCP-1KO mice (in Fig. 2D,E) was related to the regulation of fatty acid metabolism genes. We analyzed peroxisome proliferator-activated receptor alpha (PPARα) and peroxisome proliferator-activated receptor gamma (PPARγ), important transcription factors in metabolism as well as inflammatory responses.19 Though chronic alcohol feeding decreased PPARα mRNA in WT, alcohol-fed MCP-1KO mice showed comparable levels to pair-fed controls, indicating the prevention of PPARα down-regulation by alcohol (Fig. 6A). On the other hand, PPARγ mRNA was not affected in alcohol-fed WT mice, but was significantly increased in alcohol-fed MCP-1KO mice, as compared to controls (Fig. 6A). PPARα and PPARγ mRNA were not altered between genotypes, but showed a significant up-regulation in alcohol-fed MCP-1KO, compared to WT counterparts (Fig. 6A). Upon activation, PPARs translocate to the nucleus and bind to promoter elements of the target gene involved in fatty acid metabolism.20, 21 Nuclear PPARα and PPARγ levels were increased in alcohol-fed MCP-1KO mice, compared to pair-fed controls (Fig. 6B). Using electrophoretic mobility shift analysis (EMSA), we next analyzed the DNA-binding activity of PPARs in livers of alcohol-fed WT and MCP-1KO mice. Our results show that peroxisome proliferator response element (PPRE)-binding activity was significantly reduced in alcohol-fed WT mice, compared to pair-fed controls, whereas down-regulation of PPRE-binding activity was prevented in alcohol-fed livers of MCP-1KO mice (Fig. 6C). Similar to PPRE activation in whole livers, PPRE binding activity in isolated hepatocytes was significantly reduced in alcohol-fed WT mice, whereas this down-regulation was prevented in alcohol-fed MCP-1KO mice, compared to pair-fed controls (Fig. 6D). It is worthy to note that PPRE binding was significantly higher in alcohol-exposed hepatocytes (Fig. 6D) and whole livers (Fig. 6C) of MCP-1KO, compared to alcohol-fed, WT mice. Supershift analysis in whole livers of alcohol-fed MCP-1KO and WT mice revealed the presence of PPARα and retinoid X receptor alpha in the PPAR-binding complex (Fig. 6E). Next, to further evaluate whether increased PPRE-binding activity in MCP-1KO mice would result in target-gene induction related to fatty acid metabolism,20 we estimated mRNA levels of acyl coenzyme A (CoA) oxidase (ACOX), carnitine palmitolyltransferase (CPT-1), long-chain acyl CoA dehydrogenase (LCAD), and medium-chain acyl CoA dehydrogenase (MCAD). Our results show that ACOX (Fig. 7A) and CPT-1 (Fig. 7B) mRNA levels are significantly decreased in alcohol-fed WT mice, and this down-regulation was prevented in MCP-1KO mice. Furthermore, LCAD (Fig. 7C) and MCAD (Fig. 7D) mRNA did not show significant changes in alcohol-fed MCP-1KO and WT mice. These results indicate that MCP-1 regulates PPAR mRNA expression, nuclear translocation, DNA binding, and downstream target-gene expression related to fatty acid metabolism in alcoholic liver injury.

thumbnail image

Figure 6. Deficiency of MCP-1 restores chronic alcohol-induced PPAR expression and DNA-binding activity. Livers from chronic alcohol-fed WT and MCP-1−/− mice were subjected to analysis of (A) PPARα mRNA (*P < 0.05; **P < 0.03, compared to pair-fed control; n = 12) and PPARγ mRNA (*P < 0.05; **P < 0.04, compared to pair-fed control; n = 12) by real-time PCR. (B) Nuclear PPARα (*P < 0.01, compared to pair-fed control; n = 6) and PPARγ (*P < 0.05, compared to pair-fed control; n = 6) levels by western blotting are shown in representative gels (upper panel) and graphs of density units (lower panel). (C) PPAR-binding activity to PPRE in whole livers (*P < 0.05, compared to pair-fed control; #P < 0.05, compared to WT pair-fed mice) and (D) in isolated hepaocytes (*P < 0.01, compared to pair-fed control, n = 6; #P < 0.04, compared to WT EtOH-fed mice) by EMSA shown as a representative in gel (upper panel). To confirm specificity, a 20-fold excess of unlabeled oligonucleotide was included as cold competitor (Comp). Density unit of the band is shown in the bar graph (lower panel). (E) Supershift analysis was carried out using anti-PPARα, anti-PPARγ, anti-RXRα, and negative control (nonspecific) antibody (Ab) added 30 minutes before the PPRE oligo in EMSA. Supershifted band (SS) is indicated by bold arrow and indicates the presence of that protein in the binding complex. ns, nonsignificant.

Download figure to PowerPoint

thumbnail image

Figure 7. Down-regulation of PPARα target genes is prevented in chronic alcohol-fed MCP-1-deficient mice. Chronic alcohol-fed WT and MCP-1−/− mice were subjected to analysis of PPARα target gene expression by real-time PCR for (A) ACOX mRNA (*P < 0.02, compared to pair-fed control; n = 9), (B) CPT-1 mRNA (*P < 0.01, compared to pair-fed control; n = 9), (C) LCAD mRNA, and (D) MCAD mRNA. Values shown in the graphs are mean ± SEM. ns, nonsignificant.

Download figure to PowerPoint

MCP-1 Negatively Interferes With PPARα mRNA and PPRE Binding in Cultured Hepatocytes.

Because the lack of MCP-1 correlates with PPRE binding and expression of fatty acid oxidation genes, we wanted to next evaluate whether MCP-1 would directly affect PPARα expression and DNA-binding activity in hepatocytes. To this end, we performed in vitro experiments using recombinant MCP-1 and determined its effect on PPAR agonist WY-14,643-induced PPARα mRNA and PPRE-binding activity in human hepatocyte Huh7 cells. In accord with previous studies showing a lack of CCR2 expression in hepatocytes18 and Huh7 cells,13 our results show an absence of CCR2 expression in hepatocytes and Huh7 cells, compared to a high expression in monocyte/macrophages (Supporting Fig. 5A). Recombinant MCP-1 significantly decreases baseline and WY-induced PPARα mRNA, compared to WY treatment alone, in Huh7 hepatocytes (Fig. 8A). Further, MCP-1 treatment of Huh7 cells significantly reduced baseline and WY-induced PPRE activation (Fig. 8B; shown in triplicates in Supporting Fig. 5B). Collectively, our results suggest that MCP-1 can directly inhibit PPARα induction and activation impeding fatty acid oxidation in hepatocytes.

thumbnail image

Figure 8. MCP-1 interferes with PPARα expression and DNA-binding activity. Huh7 cells treated with 100 and 200 ng/mL of MCP-1 in the presence or absence of PPARα agonist WY14,643 (100 μM) (WY) were subjected to (A) PPARα mRNA (#P < 0.03, compared to untreated cells; *P < 0.05, compared to WY14,643, n = 9; values on bar graphs are mean fold change ± SEM). (B) PPRE DNA-binding activity by EMSA is shown in a representative gel (upper panel), and average density of the bands ± SEM from a total of nine experiments is shown. *P < 0.05 MCP-1 (100 ng) versus untreated cells; #P < 0.04 MCP-1 (200 ng) versus untreated; **P < 0.002 MCP-1 (100 ng) versus WY14,643; ***P < 0.001 MCP-1 (200 ng) versus WY14,643.

Download figure to PowerPoint

Discussion

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

The significance of MCP-1 as a master regulator of monocyte/macrophage function has been proposed in various chronic inflammatory diseases.22 MCP-1 was previously identified to direct the trafficking of immune cells to the site of tissue injury.6, 7 However, recent studies have suggested a role for MCP-1 in metabolic diseases, such as diabetes and obesity-related insulin resistance and hepatic steatosis.12, 23 Here, we report on novel data that MCP-1 contributes to alcohol-induced fatty liver, likely via the down-regulation of PPARα and its target fatty acid metabolism genes, independent of its receptor, CCR2. These results, for the first time, indicate a link between inflammatory chemokines and lipid metabolism in alcoholic liver injury. We show that chronic alcohol consumption increases MCP-1 in KCs and hepatocytes in the liver. Deficiency of MCP-1 protects against chronic alcohol-induced liver injury by reducing the expression of proinflammatory cytokines and the macrophage activation markers, ICAM-1 and CD68, and increasing PPARα expression and DNA binding, leading to the induction of fatty acid metabolism genes.

Chronic alcohol-induced liver injury is characterized by steatosis, inflammatory cell activation, and hepatocyte damage.1, 2, 24 Inflammatory cell mediators produced in the liver during chronic alcohol exposure contribute to liver injury. For instance, TNFα induces hepatocyte apoptosis in the liver,25 whereas IL-6 can generate hepatoprotective or damaging effects, based on the target cells.26 Human studies and animal models of alcoholic hepatitis show that MCP-1 is up-regulated in KCs.11 Our novel observations show that chronic alcohol feeding induces MCP-1 in KCs and hepatocytes, suggesting a functional role in inflammatory responses and steatosis. Previous studies in genetically obese and high-fat-diet–fed mice showed that MCP-1 significantly increased in adipose tissue and plasma, but not in the liver, contributing to insulin resistance and hepatic steatosis.12 Recent studies by Obstfeld et al.23 demonstrate that leptin-deficient ob/ob mice exhibit increased MCP-1 in hepatocytes only. Contrary to obesity and diabetes, chronic alcohol induces MCP-1 in KCs as well as hepatocytes, ascribing a pathogenic role for MCP-1 in the alcoholic liver. Activating signals, including LPS/TLR4 and proinflammatory cytokines, are potent inducers of MCP-1.27, 28 Interestingly, recent studies show that homocysteine, increased in alcoholic liver injury,29 also induces MCP-1 expression in hepatocytes.30 The investigation of the cell-specific regulation of MCP-1 during alcoholic liver injury will provide further insights into its functional significance.

The contribution of MCP-1 in various models of liver injury has been under investigation. Though in some cases of liver injury, such as hepatic granuloma formation and obesity-induced fatty liver, the lack of MCP-1 is protective,12, 23, 31 in other instances, such as concanavalin A–induced liver injury and lethal endotoxemia, the absence of MCP-1 worsens disease.32, 33 Here, we show that MCP-1 deficiency is protective against chronic alcohol-induced liver injury, as indicated by decreased serum ALT and reduced steatosis. Patients with severe alcoholic hepatitis and cirrhosis displayed the highest elevation of MCP-1 in liver and plasma, compared to other CC-chemokines.4, 5 Previous studies indicated that CC-chemokines, including MCP-1, played a major role in late-stage alcoholic hepatitis directing the migration of inflammatory cells and leading to fibrosis and cirrhosis.8 Studies from Seki et al.18 indicated the significance of the MCP-1/CCR2 axis in liver fibrosis. Our studies provide novel direct evidence for the importance of MCP-1 in the pathogenesis of early alcoholic liver injury.

Chronic alcohol feeding induces gut permeability and increases serum endotoxin levels, which, in turn, upregulate proinflammatory cytokine production in the liver.2, 3 Our results show that similar to alcohol-fed wild-type, MCP-1KO animals also demonstrate an elevation in serum endotoxin, suggesting that chronic alcohol does not affect mechanisms related to gut permeability in MCP-1-deficient mice. MCP-1 regulates the production of proinflammatory cytokines and adhesion molecules in monocytes/macrophages.9, 10 Despite increased endotoxin, we observed a significant reduction in mRNA expression of proinflammatory cytokines TNFα, IL-1β, IL-6, and KC/IL-8 in the liver of alcohol-fed MCP-1KO mice, compared to WT controls. In addition, we also observed a significant decrease in adhesion moelcule, ICAM-1, and the macrophage activation marker, CD68, in alcohol-fed MCP-1KO mice. Furthermore, our data indicate that the down-regulation of proinflammatory cytokines, adhesion molecule, and macrophage activation marker is independent of NF-κB activation in KCs in alcohol-fed MCP-1KO mice. Noteworthy is the lack of reduction in NF-κB DNA-binding activity in isolated hepatocytes from alcohol-fed MCP-1KO, compared to the inhibition of NF-κB activation in hepatocytes of alcohol-fed WT mice, which indicates a role for NF-κB in hepatocyte survival. Future studies will delineate the mechanism of reduction in proinflammatory responses in alcohol-fed MCP-1-deficient mice.

Oxidative stress and sensitization to LPS are hallmarks of molecular mechanisms of alcoholic liver injury.1, 2, 16 Interestingly, our results show that MCP-1 deficiency prevents the induction of chronic alcohol-induced oxidative stress, compared to WT mice. A similar correlation between MCP-1 and induction of oxidative stress in a toxic model of acute liver injury, using carbon tetrachloride, was previously observed.34 How MCP-1 modulates oxidative stress pathways or reduces antioxidants will be investigated in the future. Similar up-regulation of microsomal CYP2E135 in alcohol-fed WT and MCP-1KO mice indicate the induction of oxidative stress independent of alcohol-metabolizing CYP2E1. Besides cellular mechanisms, such as TLR expression, oxidative stress contributes to LPS sensitization in ALD and enhancement of pro inflammatory cytokine gene expression.36, 37 An in vivo LPS challenge given at the end of chronic alcohol feeding led to an augmentation of proinflammatory cytokines TNFα, IL-1β, and IL-6 in WT mice, and this was prevented in MCP-1KO mice. Our results suggest that MCP-1 deficiency inhibits oxidative stress and also impedes sensitization to LPS, independent of TLR4 expression, during alcoholic liver injury. Studies to unravel the mechanisms of LPS sensitization affected by MCP-1 during chronic alcohol exposure will be examined in the future.

Among the various mechanistic studies for alcoholic steatosis, alterations in transcription factors, such as PPARα and PPARγ, controlling lipid metabolism have been recognized.24 Previous studies showed that alcohol feeding in rats decreased PPARα activation and downstream target genes important in fatty acid oxidation.19 Here, we show that alcohol-fed MCP-1KO mice exhibit increased PPARα and PPARγ mRNA expression, enhanced nuclear PPARα and PPARγ, PPRE activation in whole liver and isolated hepatocytes, presence of PPARα/RXRα in the DNA-binding complex, and induction of target genes, such as CPT-1 and ACOX—both enzymes critical in fatty acid oxidation. Previous studies have shown that a secretory product of adipose tissue, likely MCP-1, can induce lipid accumulation in hepatoyctes.13 Our in vitro findings demonstrate that recombinant MCP-1 down-regulates PPARα mRNA expression and DNA-binding activity in hepatocytes, likely contributing to increased triglyceride accumulation in ALD. These results suggest a direct effect of MCP-1 on PPARα and its target genes and thus steatosis.

MCP-1 mediates its action via receptor CCR238 or independent of CCR2.39 Our results show that CCR2KO mice induce alcoholic liver injury similar to alcohol-fed WT mice, indicating CCR2-independent effects of MCP-1. Furthermore, because hepatocytes do not express CCR2,18 as reported here (Supporting Fig. 5A), we predict that MCP-1 mediates its effects in the liver independent of CCR2. Another lipid-modulating transcription factor with anti-inflammatory properties, PPARγ, was up-regulated in alcohol-fed MCP-1KO livers. It is likely that PPARγ inhibits proinflammatory cytokine production in chronic alcohol-exposed MCP-1KO mice. Our results here show that MCP-1 expression is directly up-regulated in hepatocytes during chronic alcohol exposure and likely regulates fatty acid oxidation, resulting in steatosis. LPS can modulate fatty acid oxidation genes via TLR4/IRAK-1 and PPARα reduction.40 We predict that alcohol-mediated increase in circulating endotoxin (i.e., LPS) induces MCP-1 in hepatocytes and macrophages to regulate fatty acid oxidation pathways in an autocrine or paracrine fashion in the liver. Future studies, using MCP-1-targeting strategies, will provide mechanistic insights into the pathophysiological mechanisms affected by MCP-1 in alcoholic liver injury.

Overall, our studies show, for the first time, that MCP-1 in the liver regulates macrophage activation, proinflammatory responses, and hepatic steatosis in alcoholic liver disease. These studies provide a link between inflammatory cell activation and pathways of fatty acid metabolism during alcoholic liver injury likely involved in the amplification and progression of disease. Therefore, it appears plausible that pharmacological approaches to block MCP-1 in the alcoholic liver may be beneficial to early alcoholic fatty liver injury and also abrogate inflammatory pathways contributing to propagation in ALD.

Acknowledgements

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

The authors thank Karen Kodys for labeling the oligonucleotides for the EMSA analysis.

References

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

Supporting Information

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

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

FilenameFormatSizeDescription
HEP_24599_sm_SuppFigs.doc1126KSupporting Information Figures

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.