NAFLD (non-alcoholic fatty liver disease) involves significant changes in liver metabolism characterized by oxidative stress, lipid accumulation and fibrogenesis. Mitochondrial dysfunction and bioenergetic defects also contribute to NAFLD. In the present study, we examined whether differences in mtDNA influence NAFLD. To determine the role of mitochondrial and nuclear genomes in NAFLD, MNX (mitochondrial–nuclear exchange) mice were fed an atherogenic diet. MNX mice have mtDNA from C57BL/6J mice on a C3H/HeN nuclear background and vice versa. Results from MNX mice were compared with wild-type C57BL/6J and C3H/HeN mice fed a control or atherogenic diet. Mice with the C57BL/6J nuclear genome developed more macrosteatosis, inflammation and fibrosis compared with mice containing the C3H/HeN nuclear genome when fed the atherogenic diet. These changes were associated with parallel alterations in inflammation and fibrosis gene expression in wild-type mice, with intermediate responses in MNX mice. Mice with the C57BL/6J nuclear genome had increased State 4 respiration, whereas MNX mice had decreased State 3 respiration and RCR (respiratory control ratio) when fed the atherogenic diet. Complex IV activity and most mitochondrial biogenesis genes were increased in mice with the C57BL/6J nuclear or mitochondrial genome, or both fed the atherogenic diet. These results reveal new interactions between mitochondrial and nuclear genomes and support the concept that mtDNA influences mitochondrial function and metabolic pathways implicated in NAFLD.

INTRODUCTION

NAFLD (non-alcoholic fatty liver disease) is the most common chronic liver disease [1] with 10–30% prevalence rates in the general population and higher rates in patients with metabolic syndrome [2]. The histopathological spectrum of NAFLD includes simple steatosis alone (i.e. fatty liver), steatosis with inflammation [NASH (non-alcoholic steatohepatitis)] and NASH that includes necro-inflammation with or without fibrosis [3,4]. NASH is a risk factor for cirrhosis and potentially hepatocellular carcinoma [5,6]. Thus a better understanding of NAFLD is needed for improved diagnosis, prevention and treatment.

Experimental and clinical research supports the hypothesis that altered mitochondrial metabolism contributes to NAFLD [7]. Although many of the molecular changes responsible for these mitochondrial changes remain poorly defined, they include alterations in mitochondrial bioenergetics, fatty acid oxidation, biogenesis and mtDNA content. Sookoian et al. [8] have shown that the hepatic mtDNA to nDNA ratio is significantly reduced in NAFLD patients compared with control subjects and correlated with peripheral insulin resistance, a key risk factor for NAFLD/NASH. In contrast, the liver mtDNA to nDNA ratio was significantly increased in mice fed on a high-fat diet for 20 weeks, indicating a possible adaptive response to chronic metabolic stress [9]. Further, mtDNA deletions and mutations have long been associated with metabolic diseases such as diabetes [10,11]. Epigenetic modification of mtDNA and nuclear-encoded mitochondrial genes may also contribute to NAFLD/NASH. For example, the mitochondrial-encoded Complex I subunit ND6 (NADH dehydrogenase 6) gene was found to be highly methylated in the liver of patients with NASH, but not in patients with simple steatosis [12]. This was associated with significantly decreased ND6 transcript and protein in the liver of NASH patients [12], which would decrease Complex I function and mitochondrial energy production. Together, these results strongly implicate mitochondria and mtDNA alterations in NAFLD pathobiology.

Although it is acknowledged that certain mtDNA haplotypes are associated with metabolic diseases [13,14], it has been difficult to identify the precise molecular mechanisms that underpin how mtDNA differences influence disease susceptibility. This is especially the case for complex diseases such as NAFLD. Therefore we used the recently developed MNX (mitochondrial–nuclear exchange) mice model [15] to test whether differences in the mtDNA genetic background influence mitochondrial function and NAFLD. MNX mice have interchanged nuclear and mitochondrial genomes from different mice strains and, therefore, are different from conplastic [16] and xeno-mitochondrial animal models [17,18]. MNX mice have 100% of the nDNA and mtDNA from their respective donor strains (C57BL/6J and C3H/HeN) through nuclear transfer techniques [15]. For example, MNX mice have the nuclear genome of a C57BL/6J (C57n) or a C3H/HeN (C3Hn) mouse, and the mitochondrial genome of a C3H/HeN (C3Hmt) or C57BL/6J (C57mt) mouse respectively. One previous study has shown that C57BL/6J mice are more, and C3H/HeN mice are less, susceptible to NAFLD [19]. Using the MNX model, Ballinger and colleagues showed that mtDNA background modulated mitochondrial function, oxidant production and response to injury in cardiac tissue [15]. In the present study, hepatic mitochondrial bioenergetics, histopathology and gene expression were examined in liver of MNX mice and WT (wild-type) C57BL/6J and WT C3H/HeN mice fed on an atherogenic diet. Our results show that the nuclear genome is a key determinant of inflammation and fibrosis; however, the mitochondrial genome influences certain aspects of lipid metabolism and mitochondrial bioenergetic function in the liver.

EXPERIMENTAL

Mice and dietary treatments

Male 6-week-old C57BL/6J (Jackson Laboratories), C3H/HeNHsd (Harlan Research Models) and MNX (MNX C57=C57n:C3Hmt and MNX C3H=C3Hn:C57mt) mice were housed in a temperature controlled environment (22±2°C) with a 12 h light–dark cycle. A detailed description of how MNX mice were generated has been provided previously [15]. Mice were ad libitum fed on a control diet (PicoLab Rodent Chow 20; 4.5% fat and 0.02% cholesterol) or on an atherogenic diet (Teklad Atherogenic Rodent Diet TD.02028; 21% fat, 1.25% cholesterol and 0.5% cholic acid) for 12 weeks. For experiments, mice were anaesthetized by ketamine (60 mg/kg of body mass) and xylazine (10 mg/kg of body mass) injection (intraperitoneal), and blood and liver were collected. Studies were approved by the Institutional Animal Care and Use Committee at the University of Alabama at Birmingham (APN09319).

Serum chemistries and liver triacylglycerol measurement

Serum and liver TAG (triacylglycerol) and ALT (alanine aminotransferase) content were measured using the TAG-GPO and ALT reagent kits (Pointe Scientific). Serum total cholesterol was measured by fluorescence detection using the Amplex Red Cholesterol assay kit (Invitrogen). Serum-free cholesterol was measured with the Free Cholesterol E kit (Wako Diagnostics). Serum adiponectin was determined with the Quantikine® Adiponectin/Acrp30 kit (R&D Systems).

Histopathology

H&E (haematoxylin and eosin)-stained sections were labelled using a numeric code and evaluated by a pathologist blinded to the experimental design and status of individual animals. Sections were examined for steatosis and inflammatory activity. The presence or absence of steatosis was noted and the degree of macrosteatosis was expressed as the percentage of lobular parenchyma occupied by macrovesicular fat. Inflammatory activity was graded using the modified Knodell histology activity index for grading chronic hepatitis in humans [20]. The absence of intra-acinar inflammation was given a score of 0, one inflammatory focus or less per 10× field was scored 1, two to four foci was scored 2, five to ten foci was scored 3 and more than ten foci was scored 4. Sirius Red-stained sections were evaluated for fibrosis as described in [21]. Sirius Red (Direct Red 80; Sigma-Aldrich) reacts with collagen and does not stain other matrix proteins. Four non-overlapping fields per slide were randomly selected under ×400 magnification by an investigator blinded to sample identity. Fibrosis area was calculated using ImageJ software (NIH). Images were converted into HSB stacks with saturation set at 155 as MinThreshold and 255 as MaxThreshold to segment the collagen-stained area. The percentage of the surface area that fell in the threshold range within the whole image was taken as the fibrosis area.

Mitochondria isolation and respiration measurements

Mitochondria were prepared from fresh liver by differential centrifugation techniques [22]. Oxygen consumption of isolated liver mitochondria was monitored using a Clark-type oxygen electrode (Hansatech Instruments). Respiratory capacity was assessed by measuring State 3 (i.e. ADP-dependent) and State 4 (i.e. ADP-independent) respiration using succinate as the oxi-dizable substrate. Succinate-driven respiration was done in the presence of rotenone (1 μM). The RCR (respiratory control ratio) was calculated as the ratio of State 3 to State 4 respiration rates. Mitochondrial complex (I, II–III, IV and V) and citrate synthase activities were determined using standard spectrophotometric assays [23].

RNA isolation and gene expression analysis

Total RNA was isolated from liver using TriReagent® (Sigma). Reverse transcription was performed using the High Capacity cDNA Reverse Transcription kit (Applied Biosystems). Real-time PCR was performed using an Applied Biosystems 7900 HT instrument with Taqman Gene Expression assays containing gene-specific primers and probes. Relative gene expression was determined using the comparative cycle threshold method. Data were normalized to GAPDH (glyceraldehyde-3-phosphate dehydrogenase) and expressed as the fold-change from C57BL/6J mice fed on the control diet for statistical analysis. Gene expression data were transformed (log2) and plotted in a heatmap generated with Genespring version 12.6 (Agilent Technologies) to aid visualization of the results (Figure 3).

Statistical analysis

Statistical differences were determined by ANOVA using SAS (SAS Institute) and SigmaStat (Systat Software). Three-way ANOVA was used to determine the significance of the main effect of diet, genotype and haplotype, and interactions (diet X genotype, diet X haplotype, genotype X haplotype and diet X genotype X haplotype). The ANOVA results are shown in Supplementary Tables S1 and S2 (http://www.biochemj.org/bj/461/bj4610223add.htm). P values for key pair-wise comparisons between groups (post-hoc tests) are shown in Supplementary Table S3 (http://www.biochemj.org/bj/461/bj4610223add.htm). Only the results for the pair-wise comparisons within each group as a function of diet (control against atherogenic) are provided in Table 1 and Figures 1(B), 1(C), 2(B), 4 and 5. Criterion for significance for all statistical analyses was set at P≤0.05.

Effect of diet, genotype and haplotype on liver histology

Figure 1
Effect of diet, genotype and haplotype on liver histology

Liver macrosteatosis and inflammatory foci were assessed in H&E-stained sections for mice groups: WT C57, WT C3H, MNX C57 and MNX C3H. (A) Representative photomicrographs of H&E-stained sections are shown from WT C57 (a; n=5), WT C3H (c; n=4), MNX C57 (e; n=5) and MNX C3H (g; n=6) mice fed on the control diet and WT C57 (b; n=16), WT C3H (d; n=15), MNX C57 (f; n=7) and MNX C3H (h; n=7) mice fed on the atherogenic diet (magnification ×400). The narrow and wide arrows indicate hepatocyte steatosis and inflammatory infiltration respectively. (B and C) Histograms representing the percentage of hepatocytes with macrosteatosis and inflammation score. Steatosis was not present in livers from mice fed on the control diet and was scored as 0 and thus no bars are given for control diet groups. Results are expressed as means±S.E.M. *P<0.05 compared with the corresponding control diet counterpart.

Figure 1
Effect of diet, genotype and haplotype on liver histology

Liver macrosteatosis and inflammatory foci were assessed in H&E-stained sections for mice groups: WT C57, WT C3H, MNX C57 and MNX C3H. (A) Representative photomicrographs of H&E-stained sections are shown from WT C57 (a; n=5), WT C3H (c; n=4), MNX C57 (e; n=5) and MNX C3H (g; n=6) mice fed on the control diet and WT C57 (b; n=16), WT C3H (d; n=15), MNX C57 (f; n=7) and MNX C3H (h; n=7) mice fed on the atherogenic diet (magnification ×400). The narrow and wide arrows indicate hepatocyte steatosis and inflammatory infiltration respectively. (B and C) Histograms representing the percentage of hepatocytes with macrosteatosis and inflammation score. Steatosis was not present in livers from mice fed on the control diet and was scored as 0 and thus no bars are given for control diet groups. Results are expressed as means±S.E.M. *P<0.05 compared with the corresponding control diet counterpart.

Effect of diet, genotype and haplotype on liver fibrosis

Figure 2
Effect of diet, genotype and haplotype on liver fibrosis

Staining of collagen fibres was assessed for WT C57, WT C3H, MNX C57 and MNX C3H mice. (A) Representative photomicrographs of Sirius Red-stained liver sections are shown for WT C57 (a; n=5), WT C3H (c; n=3), MNX C57 (e; n=4) and MNX C3H (g; n=4) mice fed on the control diet and WT C57 (b; n=7), WT C3H (d; n=5), MNX C57 (f; n=6) and MNX C3H (h; n=5) mice fed on the atherogenic diet (magnification ×400). The arrows indicate Sirius Red-stained collagen fibres. (B) Histogram representing the percentage of area stained with Sirius Red. Results are expressed as means±S.E.M. *P<0.05 and **P<0.0001 compared with the corresponding control diet counterpart. ANOVA results are provided in Supplementary Table S1 and P values for pair-wise comparisons are provided in Supplementary Table S3 (at http://www.biochemj.org/bj/461/bj4610223add.htm).

Figure 2
Effect of diet, genotype and haplotype on liver fibrosis

Staining of collagen fibres was assessed for WT C57, WT C3H, MNX C57 and MNX C3H mice. (A) Representative photomicrographs of Sirius Red-stained liver sections are shown for WT C57 (a; n=5), WT C3H (c; n=3), MNX C57 (e; n=4) and MNX C3H (g; n=4) mice fed on the control diet and WT C57 (b; n=7), WT C3H (d; n=5), MNX C57 (f; n=6) and MNX C3H (h; n=5) mice fed on the atherogenic diet (magnification ×400). The arrows indicate Sirius Red-stained collagen fibres. (B) Histogram representing the percentage of area stained with Sirius Red. Results are expressed as means±S.E.M. *P<0.05 and **P<0.0001 compared with the corresponding control diet counterpart. ANOVA results are provided in Supplementary Table S1 and P values for pair-wise comparisons are provided in Supplementary Table S3 (at http://www.biochemj.org/bj/461/bj4610223add.htm).

Table 1
Effect of diet, genotype and haplotype on body parameters, liver measurements and select serum chemistries

Mice were fed on a control diet or on an atherogenic diet for 12 weeks. Data are presented as means±S.E.M. Body mass, liver mass and liver/body mass ratio, n=5–15; liver TAG, n=3–4; serum TAG, n=4–9; total serum cholesterol, n=3–5; free serum cholesterol, n=5–11; serum adiponectin, n=5; and serum ALT, n=4–9. Statistical differences as compared with corresponding control counterparts. *P<0.05; **P< 0.0001 for differences between dietary treatments (atherogenic diet compared with control diet). Other pair-wise comparisons are included in Supplementary Table S3 (http://www.biochemj.org/bj/461/bj4610223add.htm).

Parameter WT C57 (C57n:C57mtWT C3H (C3Hn:C3HmtMNX C57 (C57n:C3HmtMNX C3H (C3Hn:C57mt
 Control Atherogenic Control Atherogenic Control Atherogenic Control Atherogenic 
Body mass (g) 28.8±1.3 24.9±0.3* 31.3±0.8 42.4±0.4** 28.5±0.8 24.5±1.04* 32.9±0.8 42.4±0.8** 
Liver mass (g) 1.4±0.1 1.8±0.01* 1.7±0.02 2.9±0.1** 1.3±0.03 2.1±0.1** 1.5±0.07 2.9±0.1** 
Liver mass/body mass 4.6±0.1 7.3±0.2** 5.6±0.2 6.9±0.2* 4.7±0.1 8.7±0.3** 4.6±0.2 7.6±0.5** 
Liver TAG (mg/mg of protein) 3.1±0.5 5.4±1.4 3.9±0.6 4.5±0.4 2.7±0.07 3.5±1.5 2.1±0.12 3.3±0.05* 
Serum TAG (mg/dl) 31.0±2.8 20.0±3.8 87.4±9.4 33.5±4.2** 21.4±1.2 13.2±2.00 93.8±11.7 63.2±12.4** 
Total serum cholesterol (mg/dl) 56.5±8.8 125.1±9.9* 148.1±7.7 196.9±29.8 55.8±12.7 113.4±12.3* 169.4±9.1 298.4±7.5** 
Free serum cholesterol (mg/dl) 6.7±1.3 16.6±1.5* 14.9±0.8 16.8±0.7 7.6±0.9 15.5±1.3* 16.8±1.07 18.2±2.4 
Serum adiponectin (μg/ml) 6.1±0.4 5.7±0.7 2.8±0.3 4.6±0.4* 8.0±0.5 5.7±0.5* 5.0±0.6 3.9±0.4 
Serum ALT (IU/l) 13.9±3.5 51.2±4.9* 22.4±5.2 35.5±5.8 27.8±4.6 88.2±17.9* 33.9±9.1 95.8±11.3** 
Parameter WT C57 (C57n:C57mtWT C3H (C3Hn:C3HmtMNX C57 (C57n:C3HmtMNX C3H (C3Hn:C57mt
 Control Atherogenic Control Atherogenic Control Atherogenic Control Atherogenic 
Body mass (g) 28.8±1.3 24.9±0.3* 31.3±0.8 42.4±0.4** 28.5±0.8 24.5±1.04* 32.9±0.8 42.4±0.8** 
Liver mass (g) 1.4±0.1 1.8±0.01* 1.7±0.02 2.9±0.1** 1.3±0.03 2.1±0.1** 1.5±0.07 2.9±0.1** 
Liver mass/body mass 4.6±0.1 7.3±0.2** 5.6±0.2 6.9±0.2* 4.7±0.1 8.7±0.3** 4.6±0.2 7.6±0.5** 
Liver TAG (mg/mg of protein) 3.1±0.5 5.4±1.4 3.9±0.6 4.5±0.4 2.7±0.07 3.5±1.5 2.1±0.12 3.3±0.05* 
Serum TAG (mg/dl) 31.0±2.8 20.0±3.8 87.4±9.4 33.5±4.2** 21.4±1.2 13.2±2.00 93.8±11.7 63.2±12.4** 
Total serum cholesterol (mg/dl) 56.5±8.8 125.1±9.9* 148.1±7.7 196.9±29.8 55.8±12.7 113.4±12.3* 169.4±9.1 298.4±7.5** 
Free serum cholesterol (mg/dl) 6.7±1.3 16.6±1.5* 14.9±0.8 16.8±0.7 7.6±0.9 15.5±1.3* 16.8±1.07 18.2±2.4 
Serum adiponectin (μg/ml) 6.1±0.4 5.7±0.7 2.8±0.3 4.6±0.4* 8.0±0.5 5.7±0.5* 5.0±0.6 3.9±0.4 
Serum ALT (IU/l) 13.9±3.5 51.2±4.9* 22.4±5.2 35.5±5.8 27.8±4.6 88.2±17.9* 33.9±9.1 95.8±11.3** 

RESULTS

The following abbreviations will be used to identify the four different strains of mice used: (i) WT C57=C57n:C57mt; (ii) WT C3H=C3Hn:C3Hmt; (iii) MNX C57=C57n:C3Hmt; and (iv) MNX C3H=C3Hn:C57mt. WT C57 and MNX C57 mice contain the nuclear genome of C57 mice, and WT C3H and MNX C3H contain the nuclear genome of C3H mice. Further, WT C57 and MNX C3H contain the mitochondrial genome of C57 mice, and WT C3H and MNX C57 contain the mitochondrial genome of C3H mice.

Body parameters and serum chemistries

Wild-type C3H and MNX C3H mice weighed more when fed on the atherogenic diet, whereas WT C57 and MNX C57 mice weighed less, compared with their control diet counterparts (Table 1). The liver mass and liver/body mass ratio were increased in all groups fed on the atherogenic diet compared with mice fed on the control diet (Table 1). Serum TAG and cholesterol (total and free) levels were significantly higher in control-fed WT C3H and MNX C3H mice than in WT C57 and MNX C57 mice (Table 1). The atherogenic diet decreased serum TAG levels in all groups (Table 1), whereas total cholesterol was increased in all groups fed on the atherogenic diet, except WT C3H mice (Table 1). The atherogenic diet increased serum adiponectin in WT C3H mice, but not in WT C57 mice (Table 1). Serum adiponectin was decreased in MNX C57 mice fed on the atherogenic diet, whereas no changes were observed in MNX C3H mice. The atherogenic diet significantly increased serum ALT in mice with the C57 genotype and haplotype (Table 1). Results for three-way ANOVA on serum chemistries are shown in Supplementary Table S1. A significant effect of diet was observed for all serum chemistries except adiponectin. A significant effect of genotype was observed for all serum chemistries except ALT. A significant effect of haplotype was observed for serum TAG and total cholesterol.

Histopathology

WT and MNX mice fed on the atherogenic diet developed steatosis (Figure 1A); however, the degree of steatosis in WT C3H mice fed on the atherogenic diet did not reach statistical significance (Figure 1B). Inflammation (Figure 1C) and pericellular fibrosis (Figure 2A) were present in the livers of WT C57 and MNX C57 mice fed on the atherogenic diet, whereas WT C3H and MNX C3H mice were resistant to inflammation and fibrosis. Collagen staining was 3–4-fold higher in WT C57 and MNX C57 mice fed on the atherogenic diet compared with the control diet counterparts (Figure 2B).

Expression of lipid metabolism, inflammation and fibrosis genes

Having observed different effects of the atherogenic diet on liver pathology, we examined mRNA expression of well-established lipid metabolism, inflammation and fibrosis mediators (Figure 3). All genes measured in the present study were connected by network analysis (Supplementary Figure S1 at http://www.biochemj.org/bj/461/bj4610223add.htm). The ANOVA and post-hoc test results for gene expression are shown in Supplementary Tables S2 and S3 respectively.

Effect of diet, genotype and haplotype on lipid metabolism, inflammation and fibrosis genes

Figure 3
Effect of diet, genotype and haplotype on lipid metabolism, inflammation and fibrosis genes

Gene expression was determined for WT C57, WT C3H, MNX C57 and MNX C3H mice fed on the control and atherogenic diets. Results were normalized based on a change in expression from WT C57 mice fed on the control diet. The sample size for each measurement was n=4 mice per group. Gene expression results were transformed (log2) and plotted in a heatmap generated using Genespring version 12.6 (Agilent Technologies). Down-regulated genes (relative to WT C57 control diet) are shown in green, and up-regulated genes (relative to WT C57 control diet) are shown in red. The ANOVA results are provided in Supplementary Table S2 and P values for the pair-wise comparisons are provided in Supplementary Table S3 (at http://www.biochemj.org/bj/461/bj4610223add.htm).

Figure 3
Effect of diet, genotype and haplotype on lipid metabolism, inflammation and fibrosis genes

Gene expression was determined for WT C57, WT C3H, MNX C57 and MNX C3H mice fed on the control and atherogenic diets. Results were normalized based on a change in expression from WT C57 mice fed on the control diet. The sample size for each measurement was n=4 mice per group. Gene expression results were transformed (log2) and plotted in a heatmap generated using Genespring version 12.6 (Agilent Technologies). Down-regulated genes (relative to WT C57 control diet) are shown in green, and up-regulated genes (relative to WT C57 control diet) are shown in red. The ANOVA results are provided in Supplementary Table S2 and P values for the pair-wise comparisons are provided in Supplementary Table S3 (at http://www.biochemj.org/bj/461/bj4610223add.htm).

Differences in the basal expression of PPARG (perioxisome-proliferator-activated receptor γ), FABP1 (fatty acid-binding protein 1), FASN (fatty acid synthase), PNPLA3 (patatin-like phospholipase domain-containing protein 3) and SREBP1 (sterol-regulatory-element-binding protein 1c) were observed between WT C57 and WT C3H mice. Interestingly, the basal expression of PPARG was markedly lower in WT C3H and MNX C3H mice compared with WT C57 and MNX C57 mice. WT and MNX mice fed on the atherogenic diet had increased expression of the lipogenesis transcription factor SREBP1 and one of its downstream gene targets DGAT1 (diacylglycerol O-acyltransferase 1). FASN and PNPLA3 were increased only in MNX mice fed on the atherogenic diet. Increased expression of the β-oxidation enzyme ACADM (medium-chain acyl-CoA dehydrogenase) and CPT1A (carnitine palmitoyltransferase 1A) were increased in mice fed on the atherogenic diet. PPARA was up-regulated by the atherogenic diet in WT C57, MNX C57, and MNX C3H mice. The atherogenic diet had no effect FABP1 and PPARG.

Higher levels of inflammatory genes were observed in WT C57 and MNX C57 mice compared with WT C3H and MNX C3H mice fed on the atherogenic diet. Importantly, significant main effects of diet and genotype were observed for all inflammation genes. It is important to note that C3H/HeN mice are TLR (Toll-like receptor) competent, unlike C3H/HeJ mice that have non-functional TLR4 signalling and do not respond to lipopolysaccharide [24,25]. TNFA (tumour necrosis factor α) and NFKB (nuclear factor κB) were increased by the atherogenic diet in all four mice strains with the highest levels measured in WT C57 and MNX C57 mice. HMOX1 (haem oxygenase 1) and NOS2 (inducible nitric oxide synthase) were significantly increased in WT C57, MNX C57 and WT C3H mice fed on the atherogenic diet compared with the controls. WT C57 and MNX C57 mice also showed increased expression of TLR2, TLR4 and MYD88 (myeloid differentiation primary response gene 88) in response to the atherogenic diet. TLR9 expression was only up-regulated in WT C57 mice fed on the atherogenic diet compared with the control diet mice. A significant effect of haplotype was found for NOS2 and TLR9.

Higher levels of fibrosis genes were observed in WT C57 and MNX C57 mice in response to the atherogenic diet compared with the WT C3H and MNX C3H mice. Significant effects of diet and genotype were observed for all three fibrosis genes [COL1A1 (collagen, type I, α1), COL4A1 (collagen, type IV, α1) and TGFB1 (transforming growth factor β)]

Mitochondrial bioenergetics

Feeding mice on the atherogenic diet caused a small decrease in State 3 respiration in WT mice. State 3 respiration was significantly decreased in MNX mice fed on the atherogenic diet compared with the control-fed mice (Figure 4A). In contrast, State 4 respiration was significantly increased in WT C57 and MNX C57 mice fed on the atherogenic diet compared with the controls (Figure 4B). The RCR was significantly decreased in MNX mice fed on the atherogenic diet compared with the controls (Figure 4C). A significant effect of diet was observed for State 3 respiration and RCR. A significant effect of genotype was observed for State 4 respiration. Additional ANOVA results are provided in Supplementary Table S1.

Effect of diet, genotype and haplotype on mitochondrial function

Figure 4
Effect of diet, genotype and haplotype on mitochondrial function

Mitochondrial respiration and RCR was measured for WT C57, WT C3H, MNX C57 and MNX C3H mice fed on the control and atherogenic diets. State 3 (A) and state 4 (B) respiration were measured and the RCR (C) was determined. Succinate was used as the oxidizable substrate. Results are expressed as means±S.E.M. (n=4–8 for the three parameters measured). *P<0.05 and **P<0.0001 compared with the corresponding control diet counterpart. The ANOVA results are provided in Supplementary Table S1 and P values for the pair-wise comparisons are provided in Supplementary Table S3 (at http://www.biochemj.org/bj/461/bj4610223add.htm).

Figure 4
Effect of diet, genotype and haplotype on mitochondrial function

Mitochondrial respiration and RCR was measured for WT C57, WT C3H, MNX C57 and MNX C3H mice fed on the control and atherogenic diets. State 3 (A) and state 4 (B) respiration were measured and the RCR (C) was determined. Succinate was used as the oxidizable substrate. Results are expressed as means±S.E.M. (n=4–8 for the three parameters measured). *P<0.05 and **P<0.0001 compared with the corresponding control diet counterpart. The ANOVA results are provided in Supplementary Table S1 and P values for the pair-wise comparisons are provided in Supplementary Table S3 (at http://www.biochemj.org/bj/461/bj4610223add.htm).

Higher and lower activities of Complex I and Complex II–III were measured in mitochondria from control-fed WT C57 mice compared with WT C3H mice (Figures 5A and 5B). Complex I activity was decreased in response to the atherogenic diet in WT C57, WT C3H and MNX C3H mice (Figure 5A). Complex II–III activity was markedly decreased in WT C3H mice, but increased in MNX mice fed on the atherogenic diet compared with the controls (Figure 5B). Complex IV activity was significantly increased by the atherogenic diet in WT C57, MNX C57 and MNX C3H mice (Figure 5C). Complex V activity was increased by the atherogenic diet only in MNX C3H mice (Figure 5D). Citrate synthase activity was increased in WT C57 and MNX C57 mice fed on the atherogenic diet (Figure 5E). A significant effect of diet was observed for all enzyme activities except Complex II–III. A significant effect of genotype was observed for Complex I. Significant diet X genotype interactions for Complexes II–III and IV, and diet X haplotype interactions for Complexes II–III and IV and citrate synthase were observed. Additional ANOVA results are provided in Supplementary Table S1.

Effect of diet, genotype and haplotype on respiratory complex activities

Figure 5
Effect of diet, genotype and haplotype on respiratory complex activities

Mitochondrial respiratory complexes activities were measured for (A) Complex I, (B) Complex II–III, (C) Complex IV, (D) Complex V and (E) citrate synthase (CS) activity in liver mitochondria of WT and MNX mice fed on control and atherogenic diets. Results are expressed as mean±S.E.M. nmol/s per mg of protein [n=3–4 (Complex I), n=4–6 (Complex II–III), n=4–12 (Complex IV), n=4–6 (Complex V) and n=4–12 (citrate synthase)]. *P<0.05 and **P<0.0001 compared with the corresponding control diet counterpart. The ANOVA results are provided in Supplementary Table S1 and P values for the pair-wise comparisons are provided in Supplementary Table S3 (at http://www.biochemj.org/bj/461/bj4610223add.htm).

Figure 5
Effect of diet, genotype and haplotype on respiratory complex activities

Mitochondrial respiratory complexes activities were measured for (A) Complex I, (B) Complex II–III, (C) Complex IV, (D) Complex V and (E) citrate synthase (CS) activity in liver mitochondria of WT and MNX mice fed on control and atherogenic diets. Results are expressed as mean±S.E.M. nmol/s per mg of protein [n=3–4 (Complex I), n=4–6 (Complex II–III), n=4–12 (Complex IV), n=4–6 (Complex V) and n=4–12 (citrate synthase)]. *P<0.05 and **P<0.0001 compared with the corresponding control diet counterpart. The ANOVA results are provided in Supplementary Table S1 and P values for the pair-wise comparisons are provided in Supplementary Table S3 (at http://www.biochemj.org/bj/461/bj4610223add.htm).

To determine whether changes in bioenergetics might be associated with alterations in mitochondrial biogenesis genes, we measured the following: PPARGC1A (PPARG co-activator 1α), PPARGC1B, NRF1 (nuclear respiratory factor 1), TFB1M (transcription factor B1, mitochondrial), SIRT1 (sirtuin 1) and HIF1A (hypoxia-inducible factor 1α). WT C57 and MNX C57 mice fed on the atherogenic diet had higher levels of all mitochondrial biogenesis genes (Figure 3). The atherogenic diet increased NRF1 in WT C3H and MNX C3H mice, albeit at lower levels. A significant effect of diet was observed for all biogenesis genes. A significant genotype effect was observed for all genes, except PPARGC1B. Additional ANOVA results are given in Supplementary Table S2.

DISCUSSION

Oxidative damage, ER (endoplasmic reticulum) stress, lipotoxicity, insulin resistance and mitochondrial dysfunction have all been implicated as contributors to NAFLD/NASH [26,27]. Although genetic and dietary animal models have been very useful in studying NAFLD, most of these models fail to mimic many features of the human disease [28]. Most experimental animal studies have also not specifically considered the importance of mitochondrial genetic background in the disease process. To help address this question we used a new experimental model, the MNX mouse, in combination with an atherogenic diet, to investigate whether differences in mitochondrial genetic background influence some of the established metabolic pathways and targets implicated in NAFLD/NASH.

MNX mice have the nuclear genome of one mouse strain and the mitochondrial genome of another mouse strain [15]. C57BL/6J and C3H/HeN mice were chosen as the parental strains for these studies as they are generally susceptible and resistant respectively to many metabolic syndrome disorders, including NAFLD [19]. C57BL/6J and C3H/HeN mice differ in mtDNA coding for the ND3 subunit of Complex I, subunit III of Complex IV and position 9818 in the tRNAArg gene [29]. Thus MNX mice are a unique model to test the potential influence and/or interaction of the nuclear and mitochondrial genetic backgrounds in disease. Our results support the concept that the nuclear genome is a key determinant for inflammation and fibrosis because changing mtDNA had little effect on these pathologies. Specifically, mice with C57 nDNA (WT C57 and MNX C57) had higher inflammation and fibrosis scores when fed on the atherogenic diet compared with mice with C3H nDNA (WT C3H and MNX C3H). The present study also shows that mtDNA background affected serum TAG and cholesterol, two measurements linked to liver lipid metabolism and accumulation. Mitochondrial haplotype also influenced some mitochondrial bioenergetic parameters including State 3 respiration and Complex IV activity.

With regards to bioenergetics, we observed differences in mitochondrial function among the four distinct mice strains under both control and stress (atherogenic diet) conditions. For example, State 3 respiration and the RCR were decreased in MNX mice fed on the atherogenic diet, whereas these parameters remained unchanged in WT mice. In contrast, State 4 respiration was significantly increased in WT C57 and MNX C57 mice fed on the atherogenic diet. Increased State 4 respiration typically suggests an uncoupling effect, i.e. leakage of protons across the inner membrane. Uncoupling, if severe, can decrease efficiency for ATP synthesis [30] and lead to necrotic cell death [31]. Thus this could partially explain increased liver injury in the livers of WT C57 and MNX C57 mice. Alternatively, uncoupling can function as an adaptive response aimed to reduce mitochondrial superoxide anion production in response to chronic fatty acid overload in vivo [32]. Although we did not measure oxidant production, oxidative stress/damage in WT C57 and MNX C57 mice fed on the atherogenic diet is anticipated because hepatic inflammation and inflammatory gene expression (e.g. TNFA and NOS2) was much higher in these mice. Studies by Ballinger and colleagues showed increased ROS (reactive oxygen species) in cardiomyocytes isolated from WT C57 compared with WT C3H mice [15]. Future studies are needed to address the influence of diet and genetics (nuclear and/or mitochondrial) on ROS production in hepatocytes.

Chronic consumption of the atherogenic diet led to a significant increase in Complex IV activity in WT C57 mice, but not in WT C3H mice. Complex IV activity was also increased in both strains of MNX mice fed on the atherogenic diet. This increase suggests an adaptive response to metabolic stress (e.g. mitochondrial biogenesis) in mice with C57 nDNA or mtDNA, or both. Further, it appears that the stress-mediated increase in Complex IV was influenced by C57 mtDNA because increased activity was observed in MNX C3H mice. As mentioned earlier, WT C57 and WT C3H mice differ in the coding sequence of subunit III of Complex IV. This mitochondrial-encoded subunit contains no redox centres and is thought to regulate complex activity and participate in proton uptake [33]. It is possible that the unique combination of C57 mtDNA and C3H nDNA altered retrograde signalling mechanisms (e.g. ROS production), which contributed to different adaptive responses in Complex IV. In contrast, different effects were observed for Complex I. Both WT C57 and WT C3H strains showed decreased Complex I activity in response to the atherogenic diet. These mice differ in the coding sequence of the mtDNA subunit ND3. Interestingly, the atherogenic diet-mediated decrease in Complex I activity did not occur in MNX C57 mice, again revealing a different response to stress in this novel mouse model. Although the molecular signals responsible for these distinct changes are presently not known, our results show that the same stress (atherogenic diet) elicits different bioenergetic responses depending on the combination of mitochondrial and nuclear genetic backgrounds.

To complement these functional studies in mitochondria, we examined expression of genes known to be involved in mitochondrial maintenance and NAFLD, including PPARGC1A, PPARGC1B, SIRT1, NRF1 and TFB1M. PGC1α and PGC1β are transcriptional co-activators that serve as key regulators of metabolism, and function as mechanistic links between the mitochondrial and nuclear regulatory pathways of mitochondrial biogenesis [34]. We observed increased gene expression for both PPARGC1A and PPARGC1B in WT C57 and MNX C57 mice in response to the atherogenic diet. Similarly, the atherogenic diet increased hepatic SIRT1 gene expression only in WT C57 and MNX C57 mice. SIRT1, an NAD+-dependent deacetylase, promotes mitochondrial biogenesis by PGC1α activation [35]. Together, these data suggest a potential defect or block in mitochondrial biogenesis in mice with both C3H nDNA and mtDNA (i.e. WT C3H). Large increases in NRF1 were also measured in WT C57 and MNX C57 mice fed on the atherogenic diet with a smaller increase detected in MNX C3H mice. NRF1 is a transcription factor that activates transcription of cytochrome c, respiratory chain complex subunits, and the transcriptional machinery of the mitochondrion including TFB1M [36]. Notably, NRF1 and TFB1M gene expression patterns positively correlated with Complex IV activities. This co-ordinated gene expression pattern supports the role of a mitochondrial biogenesis program that can be activated by metabolic stress (i.e. liver pathology) in mice with C57 nDNA (WT C57 and MNX C57) or C57 mtDNA only (MNX C3H).

In the present study it was noteworthy that the expression of the transcription factor PPARG was very low in WT C3H and MNX C3H mice compared with mice with the C57 nuclear genome. In contrast, mRNA levels of other classic mediators of lipid and TAG synthesis (SREBP1 and DGAT1) and fatty acid oxidation (PPARA, ACADM and CPAT1A) were essentially the same in all control-fed mice and induced to the same levels in response to the atherogenic diet. PPARγ plays an important role in lipid metabolism by promoting lipid accumulation in adipose and liver [37]. For example, inactivation of liver PPARG reduced hepatic steatosis in lipoatrophic AZIP mice (a model of diabetes with insulin resistance) and WT mice fed on a high-fat diet [38]. Previously, Krek and colleagues reported that during stress HIF1α activated PPARγ, which, in turn, induced cardiac hypertrophy and TAG accumulation [39]. Similarly, hypoxia and/or HIF1α has been implicated in the development of liver steatosis [9,4042] and fibrosis [43,44]. In the present study, we observed stress-mediated increase in HIF1A in WT C57 and MNX C57 mice, but not in WT C3H and MNX C3H mice. Further, HIF1A and PPARG expression positively correlated with the presence of fibrosis and fibrosis gene expression in WT C57 and MNX C57 mice fed on the atherogenic diet. Together, these data support the mechanism proposed by Krek and colleagues whereby a pathologic stress activates a HIF1α–PPARγ signalling axis resulting in tissue injury [39]. Thus the lack of activation of HIF1A and PPARG in WT C3H and MNX C3H mice may have contributed, in part, to a disease-resistant phenotype in these mice.

Along these same lines, we observed a similar pattern in the expression of inflammation genes. For example, TNFA, NOS2, TLR2, TLR4 and MYD88 were greatly increased in the liver of WT C57 and MNX C57 mice fed on the atherogenic diet, with smaller increases detected in WT C3H and MNX C3H mice. These results demonstrate a blunted inflammatory response in WT C3H and MNX C3H mice even though C3H/HeN mice are TLR competent [24,25]. We also observed a large increase in TLR9 in WT C57 fed on the atherogenic diet, but not in MNX C57 mice. The reason for this difference is not known. Notably, TLR9 signalling has been shown to contribute to the development of NASH [45,46]. Our finding regarding TLR9 may be of particular interest as mtDNA may act as a DAMP (damage-associated molecular pattern) molecule to activate TLR9 signalling [47]. Whether different mtDNA sequences or differences in mitochondria–nuclear interactions confer differential sensitivity or resistance to disease via activation of TLR signalling is an intriguing concept, and one that will require further investigation.

In summary, we show that the nuclear genome is an important determinant of liver inflammation and fibrosis. Utilization of the new MNX mice model [15], in combination with a NAFLD-inducing diet, indicates that the mitochondrial genome influences some aspects of lipid metabolism and mitochondrial respiratory function. Therefore the present study supports the hypothesis that differences in mtDNA contribute to the susceptibility and severity of NAFLD. These findings also provide a solid starting point for additional investigations focused on understanding how mitochondrial genetics influence cellular metabolism, redox signalling and liver disease susceptibility.

Abbreviations

     
  • ACADM

    medium chain acyl-CoA dehydrogenase

  •  
  • ALT

    alanine aminotransferase

  •  
  • DGAT1

    diacylglycerol O-acyltransferase 1

  •  
  • FABP1

    fatty-acid-binding protein 1

  •  
  • FASN

    fatty acid synthase

  •  
  • H&E

    haematoxylin and eosin

  •  
  • HIF1A

    hypoxia-inducible factor 1α

  •  
  • MNX

    mitochondrial–nuclear exchange

  •  
  • MYD88

    myeloid differentiation primary response gene 88

  •  
  • NAFLD

    non-alcoholic fatty liver disease

  •  
  • NASH

    non-alcoholic steatohepatitis

  •  
  • ND

    NADH dehydrogenase

  •  
  • NOS2

    inducible nitric oxide synthase

  •  
  • NRF1

    nuclear respiratory factor 1

  •  
  • PNPLA3

    patatin-like phospholipase domain-containing protein 3

  •  
  • PPAR

    perioxisome-proliferator-activated receptor

  •  
  • PPARGC1

    PPARG co-activator 1

  •  
  • ROS

    reactive oxygen species

  •  
  • SREBP1

    sterol-regulatory-element-binding transcription factor 1

  •  
  • RCR

    respiratory control ratio

  •  
  • SIRT1

    sirtuin 1

  •  
  • TAG

    triacylglycerol

  •  
  • TFB1M

    transcription factor B1, mitochondrial

  •  
  • TLR

    toll-like receptor

  •  
  • TNFA

    tumour necrosis factor α

  •  
  • WT

    wild-type

AUTHOR CONTRIBUTION

Angela Betancourt, Adrienne King and Jessica Fetterman performed the majority of the experimental work. Telisha Millender-Swain, Rachel Finley and Claudia Oliva assisted in experiments. David Ralph Crowe analysed liver histopathology (steatosis and inflammation) and Angela Betancourt analysed Sirius Red staining (fibrosis). Jessica Fetterman, Scott Ballinger and Shannon Bailey contributed to study design. Angela Betancourt and Shannon Bailey wrote the paper with helpful suggestions from all co-authors.

We thank Dr David Crossman, Department of Genetics and the Heflin Center for Genomic Science Core Laboratories, University of Alabama at Birmingham, Birmingham, AL, U.S.A., for preparation of the gene expression heatmap and network analysis.

FUNDING

This work was supported by the National Institutes of Health [grant numbers AA015172 and AA018841 (to S.M.B.) and HL094518 and HL0103859 (to S.W.B.)] and a US Army Medical Research & Material Command award [grant number W81XWH-07-1-0540d (to S.W.B.)]. A.L.K. was supported by a National Institutes of Health Research Supplement to Promote Diversity in Health-Related Research Award (linked to AA015172). J.L.F. was supported by an American Heart Association Fellowship [grant number PRE2240046].

References

References
1
Browning
J. D.
Szczepaniak
L. S.
Dobbins
R.
Nuremberg
P.
Horton
J. D.
Cohen
J. C.
Grundy
S. M.
Hobbs
H. H.
Prevalence of hepatic steatosis in an urban population in the United States: impact of ethnicity
Hepatology
2004
, vol. 
40
 (pg. 
1387
-
1395
)
[PubMed]
2
Vernon
G.
Baranova
A.
Younossi
Z. M.
Systematic review: the epidemiology and natural history of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis in adults
Aliment. Pharmacol. Ther.
2011
, vol. 
34
 (pg. 
274
-
285
)
[PubMed]
3
Ludwig
J.
Viggiano
T. R.
McGill
D. B.
Oh
B. J.
Nonalcoholic steatohepatitis: Mayo Clinic experiences with a hitherto unnamed disease
Mayo Clin. Proc.
1980
, vol. 
55
 (pg. 
434
-
438
)
[PubMed]
4
Schaffner
F.
Thaler
H.
Nonalcoholic fatty liver disease
Prog. Liver Dis.
1986
, vol. 
8
 (pg. 
283
-
298
)
[PubMed]
5
Marrero
J. A.
Fontana
R. J.
Fu
S.
Conjeevaram
H. S.
Su
G. L.
Lok
A. S.
Alcohol, tobacco and obesity are synergistic risk factors for hepatocellular carcinoma
J. Hepatol.
2005
, vol. 
42
 (pg. 
218
-
224
)
[PubMed]
6
Powell
E. E.
Jonsson
J. R.
Clouston
A. D.
Steatosis: co-factor in other liver diseases
Hepatology
2005
, vol. 
42
 (pg. 
5
-
13
)
[PubMed]
7
Begriche
K.
Massart
J.
Robin
M. A.
Bonnet
F.
Fromenty
B.
Mitochondrial adaptations and dysfunctions in nonalcoholic fatty liver disease
Hepatology
2013
, vol. 
58
 (pg. 
1497
-
1507
)
[PubMed]
8
Sookoian
S.
Rosselli
M. S.
Gemma
C.
Burgueno
A. L.
Fernandez Gianotti
T.
Castano
G. O.
Pirola
C. J.
Epigenetic regulation of insulin resistance in nonalcoholic fatty liver disease: impact of liver methylation of the peroxisome proliferator-activated receptor gamma coactivator 1α promoter
Hepatology
2010
, vol. 
52
 (pg. 
1992
-
2000
)
[PubMed]
9
Carabelli
J.
Burgueno
A. L.
Rosselli
M. S.
Gianotti
T. F.
Lago
N. R.
Pirola
C. J.
Sookoian
S.
High fat diet-induced liver steatosis promotes an increase in liver mitochondrial biogenesis in response to hypoxia
J. Cell. Mol. Med.
2011
, vol. 
15
 (pg. 
1329
-
1338
)
[PubMed]
10
Ballinger
S. W.
Shoffner
J. M.
Hedaya
E. V.
Trounce
I.
Polak
M. A.
Koontz
D. A.
Wallace
D. C.
Maternally transmitted diabetes and deafness associated with a 10.4 kb mitochondrial DNA deletion
Nat. Genet.
1992
, vol. 
1
 (pg. 
11
-
15
)
[PubMed]
11
van den Ouweland
J. M.
Lemkes
H. H.
Ruitenbeek
W.
Sandkuijl
L. A.
de Vijlder
M. F.
Struyvenberg
P. A.
van de Kamp
J. J.
Maassen
J. A.
Mutation in mitochondrial tRNALeu(UUR) gene in a large pedigree with maternally transmitted type II diabetes mellitus and deafness
Nat. Genet.
1992
, vol. 
1
 (pg. 
368
-
371
)
[PubMed]
12
Pirola
C. J.
Gianotti
T. F.
Burgueno
A. L.
Rey-Funes
M.
Loidl
C. F.
Mallardi
P.
Martino
J. S.
Castano
G. O.
Sookoian
S.
Epigenetic modification of liver mitochondrial DNA is associated with histological severity of nonalcoholic fatty liver disease
Gut
2013
, vol. 
62
 (pg. 
1356
-
1363
)
[PubMed]
13
Dunham-Snary
K. J.
Ballinger
S. W.
Mitochondrial genetics and obesity: evolutionary adaptation and contemporary disease susceptibility
Free Radic. Biol. Med.
2013
, vol. 
65
 (pg. 
1229
-
1237
)
[PubMed]
14
Wallace
D. C.
Bioenergetics in human evolution and disease: implications for the origins of biological complexity and the missing genetic variation of common diseases
Philos. Trans. R. Soc. Lond. B Biol. Sci.
2013
, vol. 
368
 pg. 
20120267
 
[PubMed]
15
Fetterman
J. L.
Zelickson
B. R.
Johnson
L. W.
Moellering
D. R.
Westbrook
D. G.
Pompilius
M.
Sammy
M. J.
Johnson
M.
Dunham-Snary
K. J.
Cao
X.
, et al. 
Mitochondrial genetic background modulates bioenergetics and susceptibility to acute cardiac volume overload
Biochem. J.
2013
, vol. 
455
 (pg. 
157
-
167
)
[PubMed]
16
Yu
X.
Gimsa
U.
Wester-Rosenlof
L.
Kanitz
E.
Otten
W.
Kunz
M.
Ibrahim
S. M.
Dissecting the effects of mtDNA variations on complex traits using mouse conplastic strains
Genome Res.
2009
, vol. 
19
 (pg. 
159
-
165
)
[PubMed]
17
McKenzie
M.
Trounce
I. A.
Cassar
C. A.
Pinkert
C. A.
Production of homoplasmic xenomitochondrial mice
Proc. Natl. Acad. Sci. U.S.A.
2004
, vol. 
101
 (pg. 
1685
-
1690
)
[PubMed]
18
Trounce
I. A.
Pinkert
C. A.
Cybrid models of mtDNA disease and transmission, from cells to mice
Curr. Top. Dev. Biol.
2007
, vol. 
77
 (pg. 
157
-
183
)
[PubMed]
19
Yamazaki
Y.
Kakizaki
S.
Takizawa
D.
Ichikawa
T.
Sato
K.
Takagi
H.
Mori
M.
Interstrain differences in susceptibility to non-alcoholic steatohepatitis
J. Gastroenterol. Hepatol.
2008
, vol. 
23
 (pg. 
276
-
282
)
[PubMed]
20
Ishak
K.
Baptista
A.
Bianchi
L.
Callea
F.
De Groote
J.
Gudat
F.
Denk
H.
Desmet
V.
Korb
G.
MacSween
R. N.
, et al. 
Histological grading and staging of chronic hepatitis
J. Hepatol.
1995
, vol. 
22
 (pg. 
696
-
699
)
[PubMed]
21
Bailey
S. M.
Mantena
S. K.
Millender-Swain
T.
Cakir
Y.
Jhala
N. C.
Chhieng
D.
Pinkerton
K. E.
Ballinger
S. W.
Ethanol and tobacco smoke increase hepatic steatosis and hypoxia in the hypercholesterolemic apoE−/− mouse: implications for a “multihit” hypothesis of fatty liver disease
Free Radic. Biol. Med.
2009
, vol. 
46
 (pg. 
928
-
938
)
[PubMed]
22
Spach
P. I.
Parce
J. W.
Cunningham
C. C.
Effect of chronic ethanol administration on energy metabolism and phospholipase A2 activity in rat liver
Biochem. J.
1979
, vol. 
178
 (pg. 
23
-
33
)
[PubMed]
23
Darley-Usmar
V. M.
Capaldi
R. A.
Takamiya
S.
Millet
F.
Wilson
M. T.
Malatesta
F.
Sarti
P.
Darley-Usmar
V. M.
Rickwood
D.
Wilson
M. T.
Mitochondria: a Practical Approach
1987
Oxford
IRL
(pg. 
113
-
152
)
24
Poltorak
A.
He
X.
Smirnova
I.
Liu
M. Y.
Van Huffel
C.
Du
X.
Birdwell
D.
Alejos
E.
Silva
M.
Galanos
C.
, et al. 
Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene
Science
1998
, vol. 
282
 (pg. 
2085
-
2088
)
[PubMed]
25
Vazquez-Torres
A.
Vallance
B. A.
Bergman
M. A.
Finlay
B. B.
Cookson
B. T.
Jones-Carson
J.
Fang
F. C.
Toll-like receptor 4 dependence of innate and adaptive immunity to Salmonella: importance of the Kupffer cell network
J. Immunol.
2004
, vol. 
172
 (pg. 
6202
-
6208
)
[PubMed]
26
Fuchs
M.
Sanyal
A. J.
Lipotoxicity in NASH
J. Hepatol.
2012
, vol. 
56
 (pg. 
291
-
293
)
[PubMed]
27
Feldstein
A. E.
Bailey
S. M.
Emerging role of redox dysregulation in alcoholic and nonalcoholic fatty liver disease
Antioxid. Redox Signal.
2011
, vol. 
15
 (pg. 
421
-
424
)
[PubMed]
28
Maher
J. J.
New insights from rodent models of fatty liver disease
Antioxid. Redox Signal.
2011
, vol. 
15
 (pg. 
535
-
550
)
[PubMed]
29
Bayona-Bafaluy
M. P.
Acin-Perez
R.
Mullikin
J. C.
Park
J. S.
Moreno-Loshuertos
R.
Hu
P.
Martos
A.
Fernandez-Silva
P.
Bai
Y.
Enriquez
J. A.
Revisiting the mouse mitochondrial DNA sequence
Nucleic Acids Res.
2003
, vol. 
31
 (pg. 
5349
-
5355
)
[PubMed]
30
Cunningham
C. C.
Coleman
W. B.
Spach
P. I.
The effects of chronic ethanol consumption on hepatic mitochondrial energy metabolism
Alcohol Alcohol.
1990
, vol. 
25
 (pg. 
127
-
136
)
[PubMed]
31
Ivester
P.
Lide
M. J.
Cunningham
C. C.
Effect of chronic ethanol consumption on the energy state and structural stability of periportal and perivenous hepatocytes
Arch. Biochem. Biophys.
1995
, vol. 
322
 (pg. 
14
-
21
)
[PubMed]
32
Eccleston
H. B.
Andringa
K. K.
Betancourt
A. M.
King
A. L.
Mantena
S. K.
Swain
T. M.
Tinsley
H. N.
Nolte
R. N.
Nagy
T. R.
Abrams
G. A.
Bailey
S. M.
Chronic exposure to a high-fat diet induces hepatic steatosis, impairs nitric oxide bioavailability, and modifies the mitochondrial proteome in mice
Antioxid. Redox Signal.
2011
, vol. 
15
 (pg. 
447
-
459
)
[PubMed]
33
Varanasi
L.
Hosler
J. P.
Subunit III-depleted cytochrome c oxidase provides insight into the process of proton uptake by proteins
Biochim. Biophys. Acta
2012
, vol. 
1817
 (pg. 
545
-
551
)
[PubMed]
34
Scarpulla
R. C.
Vega
R. B.
Kelly
D. P.
Transcriptional integration of mitochondrial biogenesis
Trends Endocrinol. Metab.
2012
, vol. 
23
 (pg. 
459
-
466
)
[PubMed]
35
Nemoto
S.
Fergusson
M. M.
Finkel
T.
SIRT1 functionally interacts with the metabolic regulator and transcriptional coactivator PGC-1α
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
16456
-
16460
)
[PubMed]
36
Scarpulla
R. C.
Transcriptional paradigms in mammalian mitochondrial biogenesis and function
Physiol. Rev.
2008
, vol. 
88
 (pg. 
611
-
638
)
[PubMed]
37
Ahmadian
M.
Suh
J. M.
Hah
N.
Liddle
C.
Atkins
A. R.
Downes
M.
Evans
R. M.
PPARγ signaling and metabolism: the good, the bad and the future
Nat. Med.
2013
, vol. 
19
 (pg. 
557
-
566
)
[PubMed]
38
Gavrilova
O.
Haluzik
M.
Matsusue
K.
Cutson
J. J.
Johnson
L.
Dietz
K. R.
Nicol
C. J.
Vinson
C.
Gonzalez
F. J.
Reitman
M. L.
Liver peroxisome proliferator-activated receptor γ contributes to hepatic steatosis, triglyceride clearance, and regulation of body fat mass
J. Biol. Chem.
2003
, vol. 
278
 (pg. 
34268
-
34276
)
[PubMed]
39
Krishnan
J.
Suter
M.
Windak
R.
Krebs
T.
Felley
A.
Montessuit
C.
Tokarska-Schlattner
M.
Aasum
E.
Bogdanova
A.
Perriard
E.
, et al. 
Activation of a HIF1α-PPARγ axis underlies the integration of glycolytic and lipid anabolic pathways in pathologic cardiac hypertrophy
Cell. Metab.
2009
, vol. 
9
 (pg. 
512
-
524
)
[PubMed]
40
Mantena
S. K.
Vaughn
D. P.
Andringa
K. K.
Eccleston
H. B.
King
A. L.
Abrams
G. A.
Doeller
J. E.
Kraus
D. W.
Darley-Usmar
V. M.
Bailey
S. M.
High fat diet induces dysregulation of hepatic oxygen gradients and mitochondrial function in vivo
Biochem. J.
2009
, vol. 
417
 (pg. 
183
-
193
)
[PubMed]
41
Nath
B.
Levin
I.
Csak
T.
Petrasek
J.
Mueller
C.
Kodys
K.
Catalano
D.
Mandrekar
P.
Szabo
G.
Hepatocyte-specific hypoxia-inducible factor-1α is a determinant of lipid accumulation and liver injury in alcohol-induced steatosis in mice
Hepatology
2011
, vol. 
53
 (pg. 
1526
-
1537
)
[PubMed]
42
Yin
J.
Gao
Z.
He
Q.
Zhou
D.
Guo
Z.
Ye
J.
Role of hypoxia in obesity-induced disorders of glucose and lipid metabolism in adipose tissue
Am. J. Physiol. Endocrinol. Metab.
2009
, vol. 
296
 (pg. 
E333
-
E342
)
[PubMed]
43
Copple
B. L.
Bai
S.
Moon
J. O.
Hypoxia-inducible factor-dependent production of profibrotic mediators by hypoxic Kupffer cells
Hepatol. Res.
2010
, vol. 
40
 (pg. 
530
-
539
)
[PubMed]
44
Copple
B. L.
Bustamante
J. J.
Welch
T. P.
Kim
N. D.
Moon
J. O.
Hypoxia-inducible factor-dependent production of profibrotic mediators by hypoxic hepatocytes
Liver Int.
2009
, vol. 
29
 (pg. 
1010
-
1021
)
[PubMed]
45
Miura
K.
Kodama
Y.
Inokuchi
S.
Schnabl
B.
Aoyama
T.
Ohnishi
H.
Olefsky
J. M.
Brenner
D. A.
Seki
E.
Toll-like receptor 9 promotes steatohepatitis by induction of interleukin-1β in mice
Gastroenterology
2010
, vol. 
139
 (pg. 
323
-
334
)
[PubMed]
46
Roh
Y. S.
Seki
E.
Toll-like receptors in alcoholic liver disease, non-alcoholic steatohepatitis and carcinogenesis
J. Gastroenterol. Hepatol.
2013
, vol. 
28
 (pg. 
38
-
42
)
[PubMed]
47
Zhang
Q.
Raoof
M.
Chen
Y.
Sumi
Y.
Sursal
T.
Junger
W.
Brohi
K.
Itagaki
K.
Hauser
C. J.
Circulating mitochondrial DAMPs cause inflammatory responses to injury
Nature
2010
, vol. 
464
 (pg. 
104
-
107
)
[PubMed]

Supplementary data