CCAAT/enhancing binding protein β deletion in mice attenuates inflammation, endoplasmic reticulum stress, and lipid accumulation in diet-induced nonalcoholic steatohepatitis


  • Shaikh Mizanoor Rahman,

    1. Department of Pediatrics, University of Colorado at Denver and Health Sciences Center, Aurora, CO
    Search for more papers by this author
  • Jill M. Schroeder-Gloeckler,

    1. Department of Pediatrics, University of Colorado at Denver and Health Sciences Center, Aurora, CO
    Search for more papers by this author
  • Rachel C. Janssen,

    1. Department of Pediatrics, University of Colorado at Denver and Health Sciences Center, Aurora, CO
    Search for more papers by this author
  • Hua Jiang,

    1. Department of Pediatrics, University of Colorado at Denver and Health Sciences Center, Aurora, CO
    Search for more papers by this author
  • Ishtiaq Qadri,

    1. Department of Pediatrics, University of Colorado at Denver and Health Sciences Center, Aurora, CO
    Search for more papers by this author
  • Kenneth N. Maclean,

    1. Department of Pediatrics, University of Colorado at Denver and Health Sciences Center, Aurora, CO
    Search for more papers by this author
  • Jacob E. Friedman

    Corresponding author
    1. Department of Pediatrics, University of Colorado at Denver and Health Sciences Center, Aurora, CO
    2. Biochemistry and Molecular Genetics, University of Colorado at Denver and Health Sciences Center, Aurora, CO
    • Dept of Pediatrics, Biochemistry and Molecular Genetics, University of Colorado at Denver and Health Sciences Center, Mail Stop 8106, PO Box 6511, Aurora, CO 80045
    Search for more papers by this author
    • fax: 303-724-3920

  • Potential conflict of interest: Nothing to report.


Nonalcoholic steatohepatitis (NASH) is characterized by steatosis, inflammation, and oxidative stress. To investigate whether the transcription factor CCAAT/Enhancer binding protein (C/EBPβ) is involved in the development of NASH, C57BL/6J wild-type (WT) or C/EBPβ knockout (C/EBPβ−/−) mice were fed either a methionine and choline deficient (MCD) diet or standard chow. These WT mice fed a MCD diet for 4 weeks showed a 2- to 3-fold increase in liver C/EBPβ messenger RNA and protein, along with increased expression of lipogenic genes peroxisome proliferators-activated receptor γ and Fas. WT mice also showed increased levels of the endoplasmic reticulum stress pathway proteins phosphorylated eukaryotic translation initiation factor α, phosphorylated pancreatic endoplasmic reticulum kinase, and C/EBP homologous protein, along with inflammatory markers phosphorylated nuclear factor κB and phosphorylated C-jun N-terminal kinase compared to chow-fed controls. Cytochrome P450 2E1 protein and acetyl coA oxidase messenger RNA involved in hepatic lipid peroxidation were also markedly increased in WT MCD diet-fed group. In contrast, C/EBPβ−/− mice fed a MCD diet showed a 60% reduction in hepatic triglyceride accumulation and decreased liver injury as evidenced by reduced serum alanine aminotransferase and aspartate aminotransferase levels, and by H&E staining. Immunoblots and real-time qPCR data revealed a significant reduction in expression of stress related proteins and lipogenic genes in MCD diet-fed C/EBPβ−/− mice. Furthermore, circulating TNFα and expression of acute phase response proteins CRP and SAP were significantly lower in C/EBPβ−/− mice compared to WT mice. Conversely, C/EBPβ over-expression in livers of WT mice increased steatosis, nuclear factor-κB, and endoplasmic reticulum stress, similar to MCD diet-fed mice. Conclusion: Taken together, these data suggest a previously unappreciated molecular link between C/EBPβ, hepatic steatosis and inflammation and suggest that increased C/EBPβ expression may be an important factor underlying events leading to NASH. (HEPATOLOGY 2007;45:1108–1117.)

Nonalcoholic steatohepatitis (NASH) is a metabolic liver disease commonly associated with obesity, type 2 diabetes and the metabolic syndrome.1, 2 A 2-hit model has been proposed in the pathogenesis of NASH.1, 3 The first insult is hepatic steatosis or fatty liver and is associated with accumulation of excess triglycerides (TG) in liver. The second hit consists of inflammation, cell death and fibrosis.1, 3 Recently, it was shown that a large majority of obese subjects have fatty liver and it has been suggested that 30% have NASH.4 The pathophysiological mechanisms underlying progression from steatosis to NASH remains obscure, due in part to lack of an appropriate experimental animal model. The methionine-choline deficient (MCD) diet has been used most often to produce a dietary model of NASH in animals. Accumulated data reveal that this model produces steatohepatitis, inflammation, and liver fibrosis that is histologically similar to human NASH.5, 6 It has been shown that both alcoholic and MCD diet-induced steatohepatitis contribute to TG accumulation in liver.7, 8

There is intense interest in the role of transcription factors in the pathogenesis of NASH, particularly PPARα and PPARγ. PPARα activation increases fatty acid oxidation, reduces adiposity, and improves hepatic steatosis.5 In addition to PPARα, PPARγ, SREBP1c and the nuclear receptors LXR and PGC-1 family have been implicated in pathways for fatty liver development.9–12 More recently, the CCAAT/enhancer binding protein (C/EBP) family of transcription factors have emerged as important regulators of hepatic lipid metabolism. Matsusue et al.13 produced a liver-specific C/EBPα null mouse on a leptin-deficient ob/ob background and found significantly reduced TG and lipogenic genes despite obesity in the ob/ob mouse. Similar reductions in hepatic gene expression were observed in leptin receptor deficient genetically obese Leprdb/db mice treated with a C/EBPα siRNA delivered by adenovirus to the liver.14 Interestingly however, disruption of hepatic C/EBPα in normal adult mice appears to cause an exacerbation of hyperglycemia and age-dependent increase in hepatosteatosis.15, 16 The reason(s) for these differences in C/EBPα−/− mice are unclear, but highlight important differences between age and the metabolic phenotype.

The role of C/EBPβ in hepatic lipogenesis or the pathogenesis of NASH has never been studied. C/EBPβ plays a key role in initiation of adipogenesis in adipose tissue,17 and the metabolic and gene regulatory responses to diabetes in liver through its regulation of the gene encoding the key gluconeogenic enzyme PEPCK.18 The absence of C/EBPβ leads to lower blood glucose and reduced adiposity,19 indicating that deleting C/EBPβ may have antidiabetic as well as antiobesity effects. Using a gene replacement strategy where C/EBPβ was expressed from the C/EBPα gene locus, Chen et al.20 showed that C/EBPβ rescued the role of C/EBPα in liver, but not in adipose tissue, emphasizing the unique role of C/EBPα in adipogenesis, and C/EBPβ in gluconeogenesis.

Unlike C/EBPα, C/EBPβ plays a major role in regulating several aspects of inflammation in the liver,21 and in the pathway for sensing endoplasmic reticulum (ER) stress.22 Indeed, C/EBPβ was originally identified as nuclear factor interleukin-6 (NFIL6) because of its inducibility by IL-6 and its important role in activation of acute inflammatory response genes in human hepatoma cells.21, 23 Previous studies have demonstrated that activators of the NF-κB pathway, such as lipopolysaccharide, TNFα and IL-1, up-regulate C/EBPβ expression,24, 25 suggesting that C/EBPβ may possibly modulate the transcription of target genes involved in the inflammatory response. In the present study, we report that C/EBPβ is highly induced in the liver during MCD diet-induced model of NASH, along with expression of genes involved in fatty acid synthesis, oxidative stress, and inflammation. In contrast, the livers from C/EBPβ−/− mice on the MCD diet show decreased TG deposition, reduced markers of liver inflammation and ER stress, and decreased expression of genes involved in the acute phase response. Conversely, C/EBPβ overexpression in the liver of wild-type (WT) mice leads to increased PPARγ, elevated TG levels, ER stress, and NF-κB activation. These data demonstrate that C/EBPβ may play a key role in steatosis and the pathophysiology of MCD diet-mediated model of NASH, through its effect on genes involved in lipid synthesis, inflammation, and possibly ER stress in mice.


C/EBPβ, CCAAT/enhancer binding proteinβ; LAP, Liver Activating Protein; LIP, Liver Inhibitory Protein; MCD diet, methionine and choline deficient diet; NASH, nonalcoholic steatohepatitis; TG, triglyceride; PPARγ, peroxisome proliferators-activated receptor γ; ER, endoplasmic reticulum; FAS, fatty acid synthase; SAP, serum amyloid P-component; CRP, C-reactive protein; TNFα, tumor necrosis factor-α; JNK, C-jun N-terminal kinase; NF-κB, nuclear factor κB; ALT, alanine aminotransferase; AST, aspartate aminotransferase; ACO, acetyl coA oxidase; CYP2E1, cytochrome P450 2E1; H&E, hematoxylin and eosin, CHOP, C/EBP homologous protein, PERK, pancreatic ER kinase; eIF2α, eukaryotic factor α.

Materials and Methods

Animals and Diets.

Male 8 week-old C57BL/6J mice (The Jackson Laboratory, Bar Harbor, ME) were used in these experiments. Mice were housed in colony cages and maintained on a 12-hour light/dark cycle. After an acclimatization period of 7 days, mice were divided into 2 groups (n = 6-8). The control group received regular chow (Harlan Teklad, Madison, WI, No 2018) while the other group received methionine-choline deficient (MCD) diet (Dyets Inc., Bethlehem, PA, No 518810). The feeding period was continued for 4 weeks. Mice were fasted for 6 hours before killing. Blood was obtained from the thoracic aorta. Serum was separated from cellular elements by centrifugation. Livers were removed, rinsed in ice-cold saline solution, and divided for various assays as outlined below. In experiment 2, C/EBPβ−/− mice26 and WT littermates on a C57BL/6J background were placed on the MCD diet for 3 weeks. All mice were kept on a 12-hour light/dark cycle, and were fed ad libitum. Experiments and sample collection took place after 6 hours of fasting. The animal care and procedures were approved by the Animal Care and Use Committee of the University of Colorado at Denver and Health Sciences Center.

Adenovirus Purification.

Adenoviruses for full-length C/EBPβ (Ad-C/EBPβ) or green fluorescent protein (Ad-GFP) control were propagated in HEK293 cells. Cells were harvested when cytopathic effects (CPE) were visible in more than 90% of the cells. Adenovirus was released from cells through rapid freeze/thawing. Adenoviruses were purified via CsCl gradients and dialyzed into virion dialysis buffer (10 mM Tris-HCl, pH 8.0, 135 mM NaCl, 1 mM MgCl2, and 50% glycerol). Titer was measured using both OD260 and plaque assays or Adeno-X Rapid Titer Kit (Clontech).

Adenovirus Transduction in Animals.

Adenovirus (1 × 1010-1 × 1011 pfu/ml adenovirus in 150 μl PBS) expressing C/EBPβ (Ad-LAP) or GFP (Ad-GFP) (control) were injected via tail vein into mice. Animals were killed 4 days after injection. Liver was removed and stored at −80°C until use.

Measurement of Serum Metabolites.

Blood glucose levels were measured using an automatic glucose monitor (Roche Diagnostics, Basel, Switzerland). Serum levels of AST and ALT were assayed using commercial kits (Biotron Diagnostics Inc., Hemet, CA). Serum FFA and TG were determined using commercial kits (WAKO Chemicals Inc., Richmond, VA and Sigma-Aldrich, St. Louis, MO). Serum level of insulin was measured by kits from ALPCO (Windham, NH). Serum levels of TNFα were determined using a Versamax tunable microplate reader (Molecular Devices, Sunnyvale, CA) and an endogen mouse TNFα ELISA kit (Pierce Biotechnology Inc., Rockford, IL) according to the manufacturer's standard protocol.

RNA Extraction and Real-Time Quantitative PCR.

Total RNA was extracted from homogenized mouse liver using the RNeasy kit (Qiagen, Valencia, CA). cDNA was prepared by reverse transcription of 250-1000 ng of total RNA using the Superscript III enzyme and random hexamers (Invitrogen, Carlsbad, CA). cDNAs were amplified using Platinum qPCR SuperMix-UDG (Invitrogen) and TaqMan Gene Expression Assays (Applied Biosystems [ABI], Foster City, CA) or custom assays designed using Primer Express (ABI) software, on an Opticon 2 (Bio-Rad, Hercules, CA) or ABI 7700 real-time PCR system. RNA expression data were normalized to levels of 18S RNA.

The TaqMan ID number for genes analyzed are as follows: ACO, Mm00443579_m1; FAS, Mm00662319_m1; PPARα, Mm00440939_m1; PPARγ, Mm00440945_m1. The custom designed assays from ABI are as follows: C/EBPβ (forward primer 5′-AAGAGCCGCGACAAGGC-3′, reverse primer 5′-GTCAGCTCCAGCACCTTGTG-3′, probe 5′-AAGATGCGCAACC-TGGA- GACGCA-3′), CRP (forward primer 5′-TGGATTGATGGGAAACCCAA-3′, reverse primer 5′-GCATCTGGCCCCACAGTG-3′, probe 5′-TGCGGAAAAGTCTGC- ACAAGGGCT-3′) and SAP (forward primer 5′-GGACCAAGCATGGACAAGCTA-3′, reverse primer 5′-TGTCTGACAAAAGGCTTCTGAAAG-3′, probe: 5′-TGCTTTGGATGTTTGTCTTCACCAGCC-3′).

Preparation of Cytosolic and Nuclear Fractions.

In order to measure nuclear NF-κB and C/EBPβ protein levels, liver nuclear extracts were prepared from frozen liver samples. 50-70 mg of liver tissue was homogenized in 300-500 μl of hypotonic buffer (10 mM HEPES, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride (PMSF), 2 μg/ml each of aprotinin and leupeptin, and 0.5 mg/ml benzamidine). The supernatants (cytoplasmic extracts) were saved. The pellets were resuspended in 40 μl of high salt buffer (20 mM HEPES, 400 mM PMSF, 2 μg/ml aprotinin, 2 μg/ml leupeptin, and 0.5 mg/ml benzamidine) for 30 minutes on ice, with occasional vortexing every 10 minutes. After centrifugation at 16,000 ×g for 30 minutes, the supernatants (nuclear extracts) were saved. For phosphor protein assay, liver samples were homogenized and centrifuged at 16,000 ×g for 30 minutes in ice-cold 50 mM HEPES buffer (pH 7.4) containing 150 mM NaCl, 10 mM sodium pyrophosphate, 2 mM Na3VO4, 10 mM NaF, 2 mM EDTA, 2 mM PMSF, 5 μg/ml leupeptin, 1% Nonidet P-40, and 10% glycerol. Supernatants were collected, and protein concentration was measured with the Bradford protein assay reagent (Bio-Rad), using BSA as standard.

Immunoblot Analysis.

Lysates were subjected to SDS/PAGE on a 10% gradient gel. Proteins were transferred and immobilized on Immobilon-P transfer membrane. We used 5% nonfat dairy milk in Tris-buffered saline (20 mM Tris, 150 mM NaCl, pH 7.4) with 0.1% Tween-20 to block nonspecific binding sites. Primary antibodies used in this study were antibodies to C/EBPβ, CHOP, actin, and GAPDH, (Santa Cruz Biotechnology, Santa Cruz, CA) and PERK, P-PERK (thr980) P-JNK, JNK, P-eIF2α (ser51), eIF2α, P-NF-κB, and NF-κB (Cell Signaling Technology, Danvers, MA). Secondary antibodies used were goat anti-mouse and goat anti-rabbit IgG-HRP conjugate (Bio-Rad) and were diluted 1:10000 in 5% nonfat dairy milk with 0.1% Tween-20 to detect antigen-antibody complexes. Immune complexes were visualized using enhanced chemiluminescence (GE Healthcare, Piscataway, NJ) and quantified by densitometry using an imaging densitometer (Model GS-800, Bio-Rad). Results were expressed relative to actin or GAPDH used as loading controls.

Extraction of Liver Lipid and Measurement of TG.

Liver lipid was extracted using the procedure of Bligh and Dyer.27 Triacylglycerol content of each sample was measured in duplicate after evaporation of the organic solvent using an enzymatic method (Sigma-Aldrich).

Liver Histology.

Liver samples were fixed in 10% buffered formaldehyde solution and processed by the paraffin slice technique. Sections about 5 μm thick were stained with hematoxylin and eosin (H&E) staining.

Statistical Analyses.

All data are presented as means ± SE. Statistical analyses were performed by using the unpaired Student's t test. A P value of less than 0.05 was considered statistically significant.


Effects of MCD Diet in WT Mice on Body Weight and Serum Metabolites.

As previously reported5, 6 despite similar food intake (Fig. 1A), MCD diet-fed mice had lower final body weight compared to standard chow- fed mice after 4 weeks of the trial period (Fig. 1B). However, the general condition of the mice remained healthy and the behavior was otherwise similar to chow-fed control mice. Serum levels of AST and ALT which are markers of hepatic injury were significantly elevated (3.2-fold and 10-fold, respectively), in MCD diet-fed mice compared to control (Fig. 1C). Moreover, serum levels of TG and FFA were also increased in the MCD diet-fed group compared to chow-fed controls (Fig. 1D,E). Conversely, blood glucose levels were significantly lower in the MCD group (Fig. 1F), while serum insulin levels did not differ (Fig. 1G).

Figure 1.

Body weight and serum metabolites in MCD diet-fed mice. Mice were fed chow diet or MCD diet for 4 weeks and serum analyzed as outlined under Materials and Methods. Values are means ± SE. n = 4-6 mice/group. *P < 0.05 versus control (chow-fed wild-type mice).

MCD Diet-Induced Steatosis Is Associated with Increased Expression of Enzymes Involved in Fatty Acid Metabolism, Including C/EBPβ and CYP2E1.

The levels of hepatic TG were increased 4-fold in MCD diet-fed mice compared to controls (Fig. 2A). To investigate whether the MCD diet affected the expression of C/EBPβ, the level of the protein was measured by immunoblot analysis (Fig. 2B). There are 3 alternate C/EBPβ translation products from a single C/EBPβ mRNA: the 38- and 34-kDa forms of C/EBPβ, also known as “Liver-enriched Activator Proteins” (LAP), and the 20-kDa form of C/EBPβ, known as “Liver Inhibitory Protein” (LIP).28 The 38-kDa LAP isoform is the predominant C/EBPβ protein in fasted control livers, however, the truncated 20-kDa LIP isoform is also present at detectable levels. The levels of the full-length C/EBPβ LAP protein were increased several fold in MCD diet-fed mice compared to the LIP isoform. As described,6 we also observed a 2-fold increase in the pro-oxidant CYP2E1 protein in the liver of MCD diet-fed mice (Fig. 2C), consistent with hepatic lipid accumulation.

Figure 2.

MCD diet increases hepatic TG levels, C/EBPβ, CYP2E1, and genes involved in hepatic lipid metabolism. Mice were fed chow diet or MCD diet for 4 weeks and livers were obtained and processed as outlined under Materials and Methods. (A) Liver TG. (B) Immunoblot for C/EBPβ (both LAP and LIP isoforms). (C) Immunoblot and densitometric quantification for CYP2E1. CYP2E1 protein expression was presented as fold change over chow-fed WT mice after normalization to GAPDH. Representative blots are shown. Values are means ± SE. n = 4-6 mice/group. *P < 0.05 versus control. (D) Relative hepatic gene expression. mRNA levels were quantified by real-time qPCR, normalized to levels of 18S RNA, and expressed as fold change versus control chow-fed mice. Values are means ± SE. n = 4-6 mice/group. * P < 0.05 versus control.

We then examined the effects of the MCD diet on C/EBPβ mRNA and other genes involved in lipogenesis and fatty acid turnover. The MCD diet significantly increased C/EBPβ mRNA expression 2.5-fold (Fig. 2D), along with the key lipogenic transcription factor PPARγ (Fig. 2D). Fatty acid synthase (FAS), a key gene in the de novo fatty acid synthesis pathway, was also profoundly increased 3.5-fold in the MCD diet-fed mice (Fig. 2D). Acetyl-CoA oxidase (ACO) mRNA, a rate-limiting enzyme in the pathway for peroxisomal beta-oxidation, was also stimulated 3-fold in the MCD group compared to controls, while PPARα mRNA, associated with increased fatty acid oxidation, was similar between the 2 groups (Fig. 2D).

The MCD Diet Is Associated with Activation of the Endoplasmic Reticulum Stress Pathway PERK, eIF2α, and CHOP, and Inflammatory Signals JNK and NF-κB.

Recent evidence has shown that chronic ER stress in liver and fat accompanies the metabolic syndrome of obesity and insulin resistance.29 Metabolic disorders including obesity and alcohol ingestion can cause protein misfolding or overloading in the ER, triggering a stress cascade with pathological consequences including inflammation and cell death.29, 30 When mammalian cells are subjected to protein overload in the ER, an immediate response is activation of the pancreatic ER kinase (PERK) to inhibit protein biosynthesis through phosphorylation of eukaryotic translation initiation factor eIF2α.31 Our results showed 1.5-fold and 5.2-fold increase in the phosphorylated forms of PERK and eIF2α in liver of MCD diet-fed mice (Fig. 3A,B), consistent with activation of the early ER stress pathway. If the overload of unfolded or misfolded proteins in the ER is not resolved, prolonged activation leads to programmed cell death requiring increased C/EBP homologous protein (CHOP/GADD153), a transcription factor that potentiates apoptosis.31 Mice fed the MCD diet showed a significant 1.5-fold increase in CHOP protein levels (Fig. 3C), confirming that the MCD diet activates down-stream mediators of the ER stress pathway.32

Figure 3.

The MCD diet induces the ER stress pathway and related proteins. (A-E) Immunoblots and densitometric quantification. Representative blots are shown. Equivalent amount of protein isolated from liver of chow-fed and MCD diet-fed mice were subjected to SDS-PAGE gel and blotted with respective antibodies. Values are means ± SE (n = 4-6) and expressed as fold change over WT control after normalizing to PERK, eIF2α, NF-κB, and JNK for P-PERK, P-eIF2α, P-NF-κB, and P-JNK, respectively, and actin for CHOP *P < 0.05 versus control (set to a value of 1.0).

The p65 subunit of NF-κB plays an essential role in inducing target genes for the NF-κB pathway in response to cell stress. The level of the P-NF-κB subunit (p65) in the nucleus was increased by 1.8-fold in the MCD diet-fed mice compared to chow-fed mice (Fig. 3D). This is consistent with increased TNFα and the well-described role for NF-κB in promoting liver inflammation in response to a MCD diet.33 Increased phosphorylation of JNK has been shown to be critical for MCD diet-induced steatohepatitis,34 and was also activated 2.5-fold in MCD diet-fed mice (Fig. 3E).

C/EBPβ−/− Mice on a MCD Diet Have Improved Liver Histology.

Having shown that the MCD diet can induce C/EBPβ, we went on to test whether deleting C/EBPβ might alter steatosis or inflammation on the MCD diet. We compared C/EBPβ−/− mice and WT littermates fed an MCD diet for 3 weeks. As shown by H&E staining, both WT control and C/EBPβ−/− mice on chow showed minimal steatosis and no signs of inflammation (Fig. 4A,B). Conversely, WT mice on the MCD diet developed histological evidence of gross hepatic steatosis and mixed inflammatory infiltrate (Fig. 4C). C/EBPβ−/− mice on the MCD diet developed markedly reduced hepatic steatosis and inflammation suggesting C/EBPβ is an important element in the induction of hepatic steatosis and inflammation (Fig. 4D).

Figure 4.

C/EBPβ−/− mice fed MCD diet demonstrate reduced steatosis and improved liver histology. C/EBPβ−/− mice or their WT littermates were fed standard chow or the MCD diet for 3 weeks. Mice were anesthetized and livers taken and processed for H&E staining as outlined under Materials and Methods. The photomicrographs show liver sections taken for: (A) chow-fed WT control, (B) chow-fed C/EBPβ−/− mice, (C) MCD diet-fed WT control, and (D) MCD diet-fed C/EBPβ−/− mice. 100× original magnification. Representative histological features are shown.

Decreased Liver and Serum TG Accumulation and Reduced TNFα Levels in C/EBPβ−/− Mice Fed the MCD Diet.

To corroborate the histological findings, biochemical analysis of TG levels were performed in serum and liver of C/EBPβ−/− mice on the MCD diet. These mice showed a significant reduction in both serum and hepatic TG levels compared to MCD diet-fed WT mice (Fig. 5A,B). Consistent with reduced lipid accumulation, AST and ALT enzymatic activity were substantially reduced compared to MCD diet-fed WT controls (Fig. 5C,D) suggesting that C/EBPβ deletion reduced liver injury in MCD diet-fed mice. Circulating TNFα, a cytokine known for its role in insulin resistance and inflammation35 was significantly lower in C/EBPβ−/− mice on the MCD diet (Fig. 5E).

Figure 5.

Serum metabolites and hepatic TG levels in C/EBPβ−/− mice fed MCD diet. Both WT and C/EBPβ−/− mice were fed MCD diet as described in Materials and Methods and fasted for 6 hours prior to sacrifice. A. Serum TG. B. Liver TG. C. Serum AST. D. Serum ALT. E. Serum TNFα. Values are means ± SE of 2 independent experiments. n= 4-5 mice/group. *P<0.05 versus MCD diet-fed WT mice.

C/EBPβ Deletion Reduces Induction of Genes Involved in Fatty Acid Metabolism and the Acute Phase Response.

Consistent with a reduction in liver lipids and enzymatic markers of liver injury in C/EBPβ−/− mice, the mRNA levels of PPARγ, ACO, and FAS were significantly lower in MCD diet-fed C/EBPβ−/− mice compared to MCD diet-fed WT controls (Fig. 6A). C/EBPβ activates transcription of the major acute phase response genes.36 The mRNA expression for C-reactive protein (CRP) and serum amyloid P-component (SAP) were significantly reduced by 60%-65% in MCD diet-fed C/EBPβ−/− mice compared to MCD diet-fed WT mice (Fig. 6B).

Figure 6.

Reduced expression of genes involved in fatty acid metabolism and acute phase response in C/EBPβ−/− mice fed MCD diet. Both WT and C/EBPβ−/− mice were fed MCD diet as described in Materials and Methods. The mRNA levels were quantified by real-time qPCR as described above. (A) FAS, PPARγ, ACO, and (B) acute-phase response genes CRP and SAP. Messenger RNA expression data was normalized to levels of 18S RNA and expressed as percent change over MCD diet-fed WT control mice. Values are means ± SE of 2 independent experiments. N = 4-5 mice/group. *P < 0.05 versus control (set to a value of 100).

Reduced Activation of ER Stress and Inflammatory Signals JNK, NF-κB and CYP2E1 in C/EBPβ−/− Mice on the MCD Diet.

In order to shed light on whether deleting C/EBPβ might attenuate the ER stress response to the MCD diet, we measured the activation of eIF2α and PERK in the liver of C/EBPβ−/− mice on the MCD diet. The phosphorylation of both PERK and eIF2α were significantly reduced by 40%-60% in C/EBPβ−/− mice compared to MCD diet-fed WT mice (Fig. 7B,C). Furthermore, the activation of NF-κB and JNK, were substantially reduced by 75% in C/EBPβ −/− mice on the MCD diet (Fig. 7D,E). Lastly, CYP2E1, one of the main enzymes responsible for initiating oxidative stress6 was significantly reduced in C/EBPβ−/− mice on the MCD diet (Fig.7F).

Figure 7.

C/EBPβ deletion prevents MCD diet-induced stress-related proteins. (A) Immunoblot for C/EBPβ, (B-F). Immunoblots and densitometric quantification for ER stress and related proteins. Values are means ± SE of 2 independent experiments. n = 4-5 mice/group and are expressed as percent change over MCD diet-fed WT control mice after normalizing to PERK, eIF2α, NF-κB, and JNK for P-PERK, P-eIF2α, P-NF-κB, and P-JNK, respectively, and GAPDH for CYP2E1. Representative blots are shown. *P < 0.05 versus control (set to a value of 100).

C/EBPβ Over-Expression Increases Hepatic TG and ER Stress Related Proteins In Vivo.

To investigate whether increasing liver C/EBPβ alone can drive hepatic TG deposition and related responses in WT mice in vivo, we administered an adenovirus carrying the full-length C/EBPβ gene or GFP control to WT mice. Adenovirus-treated mice were examined 4 days later and the level of C/EBPβ protein was increased up to 4-fold in the liver of mice receiving C/EBPβ (Fig. 8A). No changes in C/EBPβ expression were found in brain, kidney, muscle, or adipose tissue in Ad-C/EBPβ-treated mice (not shown). Importantly, mice with increased C/EBPβ in their livers had normal body weight (not shown) but showed a 3-fold increase in liver TG content (Fig. 8B). C/EBPβ over-expression also significantly increased the expression of PPARγ protein (Fig. 8C), and induced the markers of ER stress and NF-κB phosphorylation (Fig. 8D-F).

Figure 8.

C/EBPβ over-expression in livers of WT mice increases hepatic TG levels, PPARγ, ER stress and P-NFκB. Mice received adenovirus containing full-length C/EBPβ (Ad-C/EBPβ) or GFP (Ad-GFP) as described in Materials and Methods and killed 4 days later. (A) Immunoblot for LAP and LIP isoform of C/EBPβ. (B) Liver TG. (C) Immunoblot and densitometric quantification of PPARγ. (D-F) Immunoblots and densitometric values of ER and stress related proteins. Representative blots are shown. Values are means ± SE, n = 5-7 mice/group and are expressed as fold change over Ad-GFP injected control mice after normalizing to GAPDH, PERK, eiF2α, and NF-κB for PPARγ, P-PERK, P-eiF2α, and P-NF-κB, respectively. *P < 0.05 versus control (Ad-GFP).


C/EBPβ is activated in the liver through the actions of cytokines, hormones, nutrients, and cell stress21, 37 and has been suggested to play an important role in pro-inflammatory and apoptotic pathways.37 C/EBPβ is also a well established transcription factor necessary for adipogenesis in adipose tissue, however there are no prior data concerning the role of C/EBPβ in hepatic lipogenesis or NASH. Here, we tested the hypothesis that C/EBPβ, a key molecular component of inflammation and gluconeogenesis in the liver, can play a critical role in the pathogenesis of MCD diet-induced model of NASH. We found that the NASH diet significantly induced expression of C/EBPβ in the liver, while mice lacking C/EBPβ were substantially protected from steatosis, proinflammatory cytokines, acute phase response, and ER stress markers. Conversely, C/EBPβ over-expression alone in liver of WT mice markedly increased hepatic TG, PPARγ, and NF-κB similar to levels seen in the MCD diet-fed liver.

Consistent with decreased steatosis, C/EBPβ−/− mice on the MCD diet showed reduced expression of lipogenic genes FAS and PPARγ compared to WT mice, whereas C/EBPβ overexpression in the liver of WT mice increased PPARγ and TG levels. It is well known that C/EBPβ induces adipogenesis in noncommitted fibroblasts by activating expression of PPARγ.17 Although PPARγ is only minimally expressed in hepatocytes, hepatic TG accumulation is associated with a dramatic increase in PPARγ expression.38 Increased PPARγ has been shown to be involved in high fat diet-induced liver steatosis,39 while liver specific PPARγ knockout in ob/ob mice results in decreased lipid stores in liver of these animals.40 In contrast, overexpression of PPARγ in mice induced liver steatosis.9, 41 Our data showing that both the MCD diet and liver-specific C/EBPβ over-expression increases PPARγ and TG levels suggests that C/EBPβ may play a prominent role in a lipogenic program in the liver similar to the adipogenic cascade in adipose tissue. The results do not suggest that C/EBPβ is the exclusive direct mechanism for steatosis associated with NASH however, as TG accumulation and acute phase mRNA expression were increased, albeit to a lesser degree in MCD diet-fed C/EBPβ−/− mice.

Histologically, there was evidence for massive steatosis and mixed inflammatory infiltrate in WT mice on the MCD diet, but this was minimized to a large extent in C/EBPβ−/− mice. We found that NF-κB and JNK were highly active in MCD diet-fed mice and in C/EBPβ overexpressing mice. Strikingly, this was substantially suppressed in C/EBPβ knockout mice on the MCD diet, and coincided with less liver injury. At this time, it is unclear by what exact mechanism(s) these changes arise secondary to changes in C/EBPβ expression. It is possible that C/EBPβ deficiency leads to less lipid accumulation and therefore reduces subsequent liver injury, however C/EBPβ plays a fundamental role in regulating macrophage functions, particularly expression of TNFα, IL-6 and IL-1β.42 Whether deleting C/EBPβ solely in the macrophage can prevent hepatic inflammation is currently under investigation.

Several plausible theories have been advanced to explain the ability of increased TG to activate intracellular inflammatory signals, either through the metabolism of lipids to species such as long-chain CoA derivatives or via their action as ligands to modify nuclear receptors, such as NF-κB.43 Proof that such mechanisms underlie TG-induced inflammatory signaling has yet to be provided. Recent work has also shown that the toll-like receptor4 (TLR4), a key receptor for lipopolysaccharide which leads to the activation of the NF-κB pathway can play an additional role as well, acting as a sensor for endogenous lipids.44 Partial protection from development of steatohepatitis was reported in the MCD diet-induced model of NASH in TLR4 mutant mice,45 suggesting that TLR4-induced activation may also play a role in this form of liver disease.

We found that the levels of CYP2E1 protein, along with expression of the rate-limiting peroxisomal beta-oxidation gene ACO were significantly lower in C/EBPβ−/− mice on the MCD diet compared to WT MCD diet-fed mice. CYP2E1 is a microsomal fatty acid oxidase gene that generates reactive oxygen species.46 Recent data suggest that oxidative stress may recruit inflammation via activation of NF-κB.47 These findings have led to the hypothesis that in the presence of steatosis, oxidative stress may be the second hit in the progression of nonalcoholic fatty liver disease (NAFLD). The overall effect of increased extra-mitochondrial fatty acid oxidation is an increase of oxidative stress and mitochondrial impairment.3, 6 Indeed, increased reactive oxygen species production and lipid peroxidation have been reported in the MCD model of NASH.6 These results suggest that C/EBPβ deletion reduces expression of CYP2E1 and ACO and we speculate it could have a potential role in the lower NF-κB activation observed in the C/EBPβ−/− mice on the MCD diet potentially lowering oxidative stress.

We also observed a marked increase in activation of ER stress marker proteins P-PERK, P-eIF2α and CHOP in the livers of MCD diet-fed mice, and a significantly lower pattern of ER stress activation in C/EBPβ−/− mice on the MCD diet. The ER is regulated by signaling pathways that respond to an accumulation of unfolded or misfolded proteins in the organelle to protect cells from various stressors allowing for cell survival. Emerging evidence also suggests that the ability to counter-regulate inflammatory responses plays an important role in metabolic control through regulation of ER stress.31, 48 Reducing ER stress using chemical chaperones was recently shown in obese and diabetic mice to result in normalization of hyperglycemia, restoration of systemic insulin sensitivity, and resolution of fatty liver disease.48 Thus, C/EBPβ deletion may protect the liver from ER stress on the MCD diet implying an indirect role of C/EBPβ in part by relieving ER stress.

In summary, we have shown that mice lacking C/EBPβ are partially protected from steatosis and liver injury induced by the MCD diet, due to a combination of reduced lipogenesis and decreased inflammatory activation. In addition, C/EBPβ over-expression is capable of inducing PPARγ, steatosis, NF-κB, and ER stress, similar to MCD diet-fed mice. Taken together, these data suggest a previously unappreciated molecular link between C/EBPβ, hepatic steatosis and inflammation through several potential pathways (as suggested in Fig. 9) and suggest that increased C/EBPβ expression may be involved in adverse events that contribute to the development of NASH.

Figure 9.

Model for the involvement of C/EBPβ in the MCD diet-mediated induction of fatty liver, ER stress, and inflammation. Feeding a MCD diet can induce inflammation and ER stress by overloading the ER with proteins that cause a rise in expression of proinflammatory mediators such as JNK and NF-κB. One or more of these mechanisms probably accounts for the stimulation of C/EBPβ expression which can activate expression of inflammatory cytokines and lipogenic genes such as PPARγ, thereby modulating inflammatory responses in the early process of NASH, such as CRP and SAP.