Cytochrome P450 2E1 contributes to ethanol-induced fatty liver in mice

Authors


  • Potential conflict of interest: Nothing to report.

Abstract

Cytochrome P450 2E1 (CYP2E1) is suggested to play a role in alcoholic liver disease, which includes alcoholic fatty liver, alcoholic hepatitis, and alcoholic cirrhosis. In this study, we investigated whether CYP2E1 plays a role in experimental alcoholic fatty liver in an oral ethanol-feeding model. After 4 weeks of ethanol feeding, macrovesicular fat accumulation and accumulation of triglyceride in liver were observed in wild-type mice but not in CYP2E1-knockout mice. In contrast, free fatty acids (FFAs) were increased in CYP2E1-knockout mice but not in wild-type mice. CYP2E1 was induced by ethanol in wild-type mice, and oxidative stress induced by ethanol was higher in wild-type mice than in CYP2E1-knockout mice. Peroxisome proliferator-activated receptor α (PPARα), a regulator of fatty acid oxidation, was up-regulated in CYP2E1-knockout mice fed ethanol but not in wild-type mice. A PPARα target gene, acyl CoA oxidase, was decreased by ethanol in wild-type but not in CYP2E1-knockout mice. Chlormethiazole, an inhibitor of CYP2E1, lowered macrovesicular fat accumulation, inhibited oxidative stress, and up-regulated PPARα protein level in wild-type mice fed ethanol. The introduction of CYP2E1 to CYP2E1-knockout mice via an adenovirus restored macrovesicular fat accumulation. These results indicate that CYP2E1 contributes to experimental alcoholic fatty liver in this model and suggest that CYP2E1-derived oxidative stress may inhibit oxidation of fatty acids by preventing up-regulation of PPARα by ethanol, resulting in fatty liver. (HEPATOLOGY 2008.)

Alcoholic liver disease includes alcoholic fatty liver (steatosis), alcoholic hepatitis, and alcoholic cirrhosis1 and is a result of complex pathophysiological events involving various types of liver cells and injurious factors such as oxidative/nitrosative stress, lipopolysaccharide (LPS), and cytokines.2 Fatty liver is a uniform and early response of the liver to alcohol consumption, and it was previously considered to be benign. However, now it is known that fatty livers and related disorders like obesity and chronic ethanol treatment cause increased sensitivity to hepatotoxins such as LPS.3–5 There is an increased interest in and need to understand the mechanisms by which ethanol induces steatosis.

Oxidative stress has been suggested to play a central role in mechanisms of alcohol-induced damage.6 It appears that steatosis is also related to oxidative stress, as shown by the pioneering studies of Diluzio, who found that antioxidants can prevent ethanol-induced fatty liver.7 In recent studies, delivery of the Cu/Zn superoxide dismutase (Cu/Zn-SOD) gene via an adenovirus not only prevents alcohol-induced inflammation and necrosis but also reduces steatosis by 50% in rats.8 In Cu/Zn-SOD-knockout mice, liver injury including steatosis after chronic ethanol feeding is more severe.9 Many pathways have been suggested to play a key role in how oxidative stress is induced by ethanol, including induction of cytochrome P450 2E1 (CYP2E1).10 In this study, we used CYP2E1-knockout mice, the CYP2E1 inhibitor chlormethiazole (CMZ), and adenovirus-mediated expression of CYP2E1 to evaluate whether CYP2E1 plays a role in alcoholic fatty liver.

Abbreviations

ACC, acetyl CoA carboxylase; AMPK, AMP-activated protein kinase; AOX, acyl-CoA oxidase; CMZ, chlormethiazole; CYP2E1, cytochrome P450 2E1; FFAs, free fatty acids; L-FABP, L-fatty acid binding protein; FAS, fatty acyl synthase; HNE, 4-hydroxy-2-nonenal; LPS, lipopolysaccharide; 3-NT, 3-nitrotyrosine; PPARα, peroxisome proliferator-activated receptor α; SOD, superoxide dismutase; SREBP, sterol regulatory element-binding protein.

Materials and Methods

Animals and Ethanol Treatment.

SV/129-background CYP2E1-knockout and wild-type mice were kindly provided by Dr. Frank J. Gonzalez (Laboratory of Metabolism, National Cancer Institute, Bethesda, MD),11 and the female offspring of these mating pairs were used in this study. All mice were housed in temperature-controlled animal facilities with 12-hour light/12-hour dark cycles and were permitted consumption of tap water and Purina standard chow ad libitum until being fed the liquid diets. The mice received humane care, and experiments were carried out according to the criteria outlined in the Guide for the Care and Use of Laboratory Animals and with approval of the Mount Sinai Animal Care and Use Committee.

Female mice were initially fed the control liquid dextrose diet (Bio-Serv, Frenchtown, NJ) for 3 days to acclimate them to the liquid diet. Afterward, knockout and wild-type mice were fed either the liquid ethanol diet (Bio-Serv, Frenchtown, NJ) or the control liquid dextrose diet, as described by Lieber and DeCarli.12 For experiments involving 2 weeks of feeding (CMZ and adenovirus), the mice were directly subjected to the diet containing ethanol as 35% of total calories (equivalent to 6.2% [vol/vol]). In initial experiments, we observed that the mice could not tolerate ethanol at 35% of total calories for more than 2 weeks; therefore, for experiments involving 3 or 4 weeks of feeding, the content of ethanol was gradually increased every 3-4 days from 10% (1.77% [vol/vol]) of total calories to 20% (3.54% [vol/vol]), 25% (4.42% [vol/vol]), 30% (5.31% [vol/vol]), and finally 35% of total calories. The wild-type and knockout control mice were pair-fed the control dextrose diet on an isoenergetic basis. For LPS experiments, the mice were fed the dextrose or ethanol liquid diets for 4 weeks as above, LPS (Sigma, serum type 055:B5; single dose, 4 mg/kg body weight, ip) was injected, and 8 hours later, the mice were sacrificed. For CMZ experiments, during the 2 weeks of feeding with 35% ethanol or dextrose liquid diets, CMZ (Sigma; 50 mg/kg body weight, ip) was injected every other day into the mice. For adenovirus-CYP2E1 experiments, the mice were fed the 35% ethanol or dextrose liquid diets for 2 weeks after injection with the adenovirus. The ethanol-fed mice had access to their rations ad libitum, and the conditions of knockout and wild-type mice were comparable. The amount of food consumed by CYP2E1-knockout mice and wild-type mice was approximately the same, and CMZ or adenovirus injection had no effect on food consumption (data not shown). Whole liver was removed and liver weight measured; then the liver was rapidly excised into fragments and washed with cold saline, and one aliquot of tissue was placed in 10% formalin solution for paraffin blocking. Another aliquot was placed into 50 mM sodium phosphate containing 18% sucrose at 4°C overnight and then was frozen and cut into 10-μm frozen sections for oil red O staining. The remaining aliquots were stored at −80°C for further assays. Liver homogenates were prepared in ice-cold 0.15 M KCl. All samples were stored at −80°C in aliquots.

Liver Histology and Immunohistochemistry.

Liver sections were stained with H&E for pathological evaluation. Steatosis was quantified as the percentage of cells containing fatty droplets. Necroinflammation was quantified as the number of clusters of 5 or more inflammatory cells per square millimeter. Five 200× fields per liver were examined (one 200× field area = 0.95 mm2). The pathologists were unaware of the treatment groups when evaluating the slides.

Immunohistochemical staining for CYP2E1, 3-nitrotyrosine (3-NT), and 4-hydroxy-2-nonenal (HNE) adducts was performed by using anti-CYP2E1 antibody (a gift from Dr. Jerome Lasker, Hackensack Biomedical Research Institute, Hackensack, NJ), anti-3-NT adducts IgG (Upstate, Lake Placid, NY), and anti-HNE Michael adducts IgG (Calbiochem, La Jolla, CA), followed by a rabbit ABC staining system (Santa Cruz Biotechnology, Santa Cruz, CA).

Serum ALT, Ethanol, and Tumor Necrosis Factor α Determination.

Serum alanine aminotransferase (ALT) and ethanol were assayed using an Infinity kit (Thermo Electron, Melbourne, Australia) and an ethanol assay kit (Biovision, Mountain View, CA), respectively. Serum tumor necrosis factor α (TNFα) was assayed using a TNFα ELISA kit (Biosource, Camarillo, CA).

Liver Triglyceride, Free Fatty Acids, and TBAR Determination.

Liver triglyceride (TG) was determined using an Infinity kit (Thermo Electron, Melbourne, Australia). Liver free fatty acids (FFAs) were determined using an Infinity kit (Wako, Richmond, VA). Liver TBAR determination was measured as described in Lu et al.13

Cytochrome P450 2E1 Activity.

CYP2E1 activity was measured by the rate of oxidation of p-nitrophenol to p-nitrocatechol by isolated hepatic microsomes.14

Western Blotting.

Western blotting was performed to evaluate the levels of CYP2E1, peroxisome proliferator-activated receptor α (PPARα), sterol regulatory element-binding protein 1 (SREBP-1), AMP-activated protein kinase (AMPK), acetyl CoA carboxylase (ACC), phosphorylated acetyl CoA carboxylase (p-ACC), and L-fatty acid binding protein (L-FABP) using antibodies against CYP2E1, PPARα. (Rockland, Gilbertsville, PA), acyl-CoA oxidase (AOX; a gift from Professor Paul Van Veldhoven, K.U. Leuven, Belgium),15 SREBP-1 (mouse anti-SREBP-1 monoclonal antibody, clone 2A4; Santa Cruz Biotechnology, Santa Cruz, CA), AMPK, p-AMPK, ACC, p-ACC (Cell Signaling Technology, Beverly, MA), and L-FABP (Santa Cruz Biotechnology, Santa Cruz, CA), followed by horseradish peroxidase–conjugated secondary antibody. Blots were quantified using the UN-SCAN-IT automated digitizing system version 5.1 (Silk Scientific), and the results were expressed as the ratio of protein to β-actin.

RNA Isolation and RT-PCR.

Total RNA was prepared from fresh liver using TRIzol Reagent (Invitrogen, Carlsbad, CA). Reverse-transcription polymerase chain reaction (RT-PCR) to assay messenger RNA (mRNA) levels of PPARα, AOX, carnitine palmitoyltransferase (CPT1a), L-FABP, and fatty acid synthase (FAS) was performed as described,16–18 except for using β-actin as the internal control.

Recombinant Adenovirus Production.

Ad5 adenoviral vector with compensating deletions in the early region 1 (E1) was purchased from Microbix Biosystems Inc. (Ontario, Canada). An adenovirus containing CYP2E1 and β-galactosidase was developed using an Ad5 adenoviral vector as described previously.19 Purified recombinant adenovirus (3 × 109 plaque-forming units) possessing CYP2E1 (Ad-2E1) and β-galactosidase (Ad-LacZ) were diluted in 0.2 mL of normal saline and injected into the tail vein of CYP2E1-knockout mice.

Statistics.

Results are expressed as means ± SDs. Statistical evaluation was carried out by one-way ANOVA followed by the Student-Newman-Keuls post hoc test.

Results

Ethanol-Induced Liver Injury and Steatosis in Wild-Type and CYP2E1-Knockout Mice.

Mice body weight did not change during the first 2 weeks of ethanol feeding in wild-type or knockout mice (Fig. 1A). In the third week, ethanol-fed wild-type mice had lost weight compared with the dextrose-fed mice. The ethanol-fed knockout mice lost a slight but not significant amount of body weight, compared with the dextrose-fed knockout mice (Fig. 1A). After 4 weeks of ethanol feeding, the body weight of both the knockout and the wild-type mice further decreased (Fig. 1A). After 1 or 2 weeks of ethanol feeding, serum ALT levels in both wild-type and knockout mice had not changed (Fig. 1B). After 3 weeks of ethanol feeding, serum ALT levels had increased 2.2-fold in wild-type mice and 1.6-fold in knockout mice; there was no statistical difference between the knockout and wild-type mice (Fig. 1B). After 4 weeks of ethanol feeding, serum ALT levels further increased up to 3-fold in wild-type mice and 2.2-fold in knockout mice; again, there was no statistical difference between the knockout and wild-type mice (Fig. 1B). After 4 weeks of ethanol feeding, serum ethanol levels did not differ significantly between the knockout and wild-type mice (data not shown). There was no significant increase in serum TNFα levels by ethanol feeding in the wild-type and knockout mice.

Figure 1.

Mouse body weight (A), serum ALT (B), and H&E staining (C; arrows showing lipid droplets) after varying weeks of oral ethanol feeding. (D) Steatosis and necroinflammation scores after 4 weeks of oral ethanol feeding. *P < 0.05 and **P < 0.01, compared with WT DEX group; #P < 0.05 and ##P < 0.01, compared with KO Dex group; &&P < 0.01, compared with WT ethanol (ETOH) group.

Pathology examination showed no morphological change in the wild-type and CYP2E1-knockout mice after 1 week of ethanol feeding (Fig. 1C). After 2 weeks of ethanol feeding, small lipid droplets were observed in the wild-type mice but not in the knockout mice. After 3 weeks of ethanol feeding, much bigger and more lipid droplets were observed, mainly around the central vein of the wild-type mice; however, no lipid droplets were seen in the knockout mice (Fig. 1C). After 4 weeks of ethanol feeding, extensive lipid droplets were observed in the wild-type mice, but only a small number of tiny lipid droplets were observed in the knockout mice (Fig. 1C,D). A few small and variable necroinflammatory foci could be found both in wild-type and knockout mice without significant difference between the two genotypes (Fig. 1D). Oil red O staining showed that prominent lipid droplets were observed in frozen liver sections from wild-type mice, whereas only a few lipid droplets were present in the CYP2E1-knockout liver sections (Fig. 2A). Similarly, in CYP2E1-knockout mice, the liver triglyceride content did not increase after 4 weeks of ethanol feeding, whereas in the wild-type mice, triglycerides increased more than 2-fold (Fig. 2C). The liver–to–body weight ratio increased 40% by ethanol feeding in the wild-type mice, but a small, nonsignificant increase was observed in the knockout mice (Fig. 2B). Interestingly, in the wild-type mice liver FFA content did not increase after 4 weeks of ethanol feeding, whereas in the knockout mice, FFAs increased 2-fold (Fig. 2D).

Figure 2.

Effects of oral ethanol feeding on hepatic steatosis in mice. (A) Oil red O staining after 4 weeks of oral ethanol feeding (arrows showing positive staining). (B) Liver weight. *P < 0.05, compared with WT dextrose group. (C) Hepatic triglyceride content after 4 weeks of oral ethanol feeding. **P < 0.01, compared with WT dextrose group; ##P < 0.01, compared with KO ethanol group. (D) Hepatic FFA content. **P < 0.01, compared with KO dextrose group; #P < 0.05, compared with KO ethanol group.

Levels of CYP2E1 and Oxidative Stress in Wild-Type and CYP2E1-Knockout Mice.

As expected, CYP2E1 protein was absent in CYP2E1-knockout mice fed either dextrose or ethanol (Fig. 3A). In the wild-type mice, CYP2E1 protein was induced by ethanol feeding (Fig. 3A). CYP2E1 activity was increased about 4-fold after ethanol feeding in wild-type mice, but in CYP2E1-knockout mice fed dextrose or ethanol, rates of PNP oxidation were low (Fig. 3B). Induction of CYP2E1 can induce oxidative stress.6 After feeding with ethanol, TBAR levels, a marker of oxidative stress, increased about 9-fold in wild-type mice but only 3-fold in CYP2E1-knockout mice (Fig. 3C).

Figure 3.

Effects of 4 weeks of oral ethanol feeding on hepatic CYP2E1 induction and MDA levels in mice. (A) CYP2E1 protein (western blotting analysis). (B) Hepatic CYP2E1 activity. **P < 0.01, compared with WT dextrose group; ##P < 0.01, compared with KO ethanol group; &&P < 0.01, compared with KO dextrose group. (C) Hepatic oxidative stress as reflected by TBAR levels. Effects of 4 weeks of oral ethanol feeding on hepatic TBAR levels. *P < 0.05, compared with KO dextrose group; ##P < 0.01, compared with WT dextrose group; $$P < 0.01, compared with KO ethanol group.

PPARα Is Up-regulated after Ethanol Feeding in Knockout But Not in Wild-Type Mice.

PPARα, SREBP-1, or AMPK may be involved in the development of alcoholic fatty liver17, 20, 21; their expression was examined by western blotting. As shown in Fig. 4A,B, protein levels of PPARα, SREBP-1 and AMPK were similar in the dextrose-fed wild-type and CYP2E1-knockout mice. We only detected the mature, active 68-kDa form of SREBP-1, not the 125-kDa precursor form (Fig. 4A). PPARα, SREBP-1, AMPK, and p-AMPK levels were not altered by ethanol feeding in the wild-type mice. Similarly, SREBP-1, AMPK, and p-AMPK did not significantly change in the ethanol-fed knockout mice. However, PPARα was up-regulated 2-fold by ethanol feeding in CYP2E1-knockout mice. PPARα-regulated acyl-CoA oxidase (AOX) was decreased more than 2-fold after ethanol feeding in the wild-type mice but was not decreased in the knockout mice (Fig. 4A,B). There were no differences in levels of L-FABP (Fig. 4A) or ACC or p-ACC (data not shown) between the dextrose- and ethanol-fed wild-type mice and the dextrose- and ethanol-fed CYP2E1-knockout mice. RT-PCR analysis showed that after ethanol feeding, PPARα and AOX mRNA were up-regulated in the CYP2E1-knockout mice but down-regulated in the wild-type mice (Fig. 4C,D). Although in some cases there appeared to be good agreement in changes between protein and mRNA levels, in other cases, there was poor correlation of effects on the protein versus mRNA levels. For example, ethanol lowered AOX protein and mRNA in wild-type mice and elevated PPARα protein and mRNA in CYP2E1-knockout mice. However, ethanol increased AOX mRNA but not protein in the knockouts. Reasons for this are not clear, but it is suggested that mRNA level does not always reflect the corresponding protein level. RT-PCR analysis did not detect any significant change in L-FABP, CPT1a, or FAS among the 4 groups (Fig. 4A,C).

Figure 4.

Effects of 4 weeks of oral ethanol feeding on fatty acid metabolism regulating molecules. (A) PPARα, AOX, L-FABP, SREBP-1, p-AMPK, and AMPK protein (western blotting analysis). (B) PPARα, AOX, SREBP-1 (68-kDa mature form), L-FABP/β-actin ratios. **P < 0.01, compared with KO DEX group; ##P < 0.01, compared with WT DEX group. (C) PPARα, AOX, CPT1a, L-FABP, and FAS mRNA (RT-PCR analysis). (D) PPARα, AOX, L-FABP, FAS, and CPT1a mRNA/β-actin ratios. *P < 0.05, compared with KO DEX group; #P < 0.05, compared with WT DEX group.

Wild-Type But Not CYP2E1-Knockout Mice Are Vulnerable to LPS Liver Damage After Chronic Ethanol Feeding.

Vulnerability to LPS-induced damage is increased by fatty liver induction3, 4 and chronic ethanol treatment.5 We examined whether this ethanol-promoted vulnerability to LPS-induced damage was potentiated by CYP2E1. After chronic ethanol or dextrose feeding for 4 weeks, 4 mg/kg body weight of LPS was injected, and 8 hours later, the mice were sacrificed. Ethanol alone caused 3-fold and 2.2-fold increases in serum ALT levels in the wild-type and CYP2E1-knockout mice, respectively (Fig. 5A). LPS treatment caused a further increase in serum ALT level in the ethanol-fed wild-type mice but not in the knockout mice (Fig. 5A). LPS treatment had no effect in dextrose-fed mice. Pathological examination showed that more necroinflammatory foci were found in ethanol-fed wild-type mice than in ethanol-fed knockout mice (3.2 ± 1.7 versus 0.8 ± 0.5, P < 0.01) after being treated with LPS (Fig. 5B). As in the absence of LPS (Fig. 2A), considerable fat accumulation occurred in the ethanol-plus-LPS-treated wild-type mice but not in the ethanol-plus-LPS-treated CYP2E1-knockout mice (Fig. 5B). Triglyceride content was elevated comparably by ethanol or by ethanol plus LPS treatment in wild-type mice but not in knockout mice (Fig. 5C). Thus, LPS promotes injury but does not increase steatosis beyond that caused by ethanol feeding alone in the wild-type mice. LPS does not promote injury or induce steatosis in ethanol-fed CYP2E1-knockout mice.

Figure 5.

Effects of 4 weeks of oral ethanol feeding on LPS liver injury. (A) Serum ALT levels. (B) H&E staining (circles showing inflammatory cell invasion and arrows showing lipid droplets). (C) Liver triglyceride levels. **P < 0.01, compared with DEX Sal group; ##P < 0.01, compared with DEX LPS group; &&P < 0.01, compared with WT ethanol (ETOH) LPS group; $P < 0.05 and $$P < 0.01, compared with WT ETOH Sal group.

CMZ, an Inhibitor of CYP2E1, Lowers Ethanol Fatty Liver and Oxidative Stress in Wild-Type Mice.

The CYP2E1 inhibitor CMZ was used to further assess the role of CYP2E1 in ethanol-induced fatty liver in wild-type mice. Because of the high price of commercial CMZ, the feeding time of ethanol was shortened to 2 weeks. Feeding of ethanol for 2 weeks did not change ALT levels (Fig. 1B) but induced steatosis in wild-type mice (Fig. 1C). After a 2-week feeding with ethanol, moderate steatosis was observed in the wild-type mice (Fig. 6A), and lipid droplets were mainly in the area where elevated CYP2E1 was present (Fig. 6B). Liver triglyceride content increased about 2-fold (Fig. 6C). CMZ treatment lowered ethanol-induced steatosis (Fig. 6A) and decreased liver triglyceride content (Fig. 6C). Ethanol-induced CYP2E1 was decreased by CMZ (Fig. 6B,D). Interestingly, PPARα protein levels were increased by the ethanol-plus-CMZ treatment (Fig. 6D), which is consistent with the results showing an increase in PPARα level with ethanol-fed CYP2E1-knockout mice (Fig. 4A). As before, ethanol feeding did not promote triglyceride accumulation in the knockout mice, and CMZ had little effect on triglyceride levels in the knockouts (Fig. 6C).

Figure 6.

Effects of CMZ on hepatic steatosis and levels of CYP2E1 and PPARα in wild-type and CYP2E1-knockout mice. (A) H&E and oil red O staining (arrowheads show oil red–positive staining, and arrows show lipid droplets). (B) Immunohistochemical staining for CYP2E1 (arrows show lipid droplets). (C) Liver triglyceride content. *P < 0.05, compared with dextrose group; #P < 0.05 and ##P < 0.01, compared with ethanol group; $P < 0.05, compared with the dextrose or CMZ dextrose group. (D) CYP2E1 and PPARα (western blotting analysis); bottom numbers show CYP2E1/β-actin or PPARα/β-actin ratios.

Ethanol-induced oxidative stress is partially mediated by CYP2E1,6 and it would be expected that CMZ could inhibit oxidative stress by inhibiting CYP2E1.22 Indeed, feeding ethanol to wild-type mice caused an increase in TBAR levels (Fig. 7A), which was inhibited by CMZ treatment (Fig. 7A). To further substantiate the effect of CMZ on ethanol-induced oxidative stress, immunohistochemical analysis of 3-NT and HNE protein adduct formation was evaluated. Ethanol feeding caused an increase in formation of 3-NT and HNE adducts, markers of oxidative stress, and CMZ treatment inhibited this increase in 3-NT and HNE adducts (Fig. 7B)

Figure 7.

Effects of CMZ on hepatic oxidative/nitrosative stress as reflected by (A) MDA (TBAR) levels and (B) 3-NT and HNE adduct formation. *P < 0.05, compared with control group; ##P < 0.01, compared with ethanol group.

Adenovirus-Mediated Expression of CYP2E1 Produces Fatty Liver in CYP2E1-Knockout Mice.

To prove that the absence of fatty liver in ethanol-fed CYP2E1-knockout mice was a result of the lack of CYP2E1, an adenovirus containing CYP2E1 (Ad-2E1) and β-galactosidase (Ad-LacZ) was used. This Ad-2E1 has been used in studies characterizing in vitro and in vivo toxicological properties of CYP2E1.19, 23 The adenovirus was injected into CYP2E1-knockout mice, and 5 days later expression of CYP2E1 was detected in the livers of CYP2E1-knockout mice treated with Ad-2E1 group (Fig. 8A). No expression of CYP2E1 was detected in the Ad-LacZ-treated knockout mice, although β-gal staining was positive (Fig. 8A). In pilot experiments, after 3 weeks of feeding with ethanol, all the adenovirus-treated knockout mice died. Therefore, like the CMZ experiments, the feeding time with ethanol was shortened to 2 weeks. After 2 weeks of ethanol feeding, serum ALT level was elevated in the Ad-2E1-injected knockout mice but not in the Ad-LacZ-injected knockout mice (Fig. 8B). Thus, partial restoration of CYP2E1 caused some injury to the ethanol-fed CYP2E1-knockout mice. Similarly, moderate steatosis was observed in the Ad-2E1-injected knockout mice but not in the Ad-LacZ-injected knockout mice (Fig. 8C). Tiny lipid droplets were observed in the Ad-2E1 mice but not the Ad-LacZ mice (Fig. 8C), and these lipid droplets were mainly in the hepatic areas where CYP2E1 was expressed (Fig. 8D).

Figure 8.

Effect of adenovirus-mediated CYP2E1 expression in CYP2E1-knockout mice on ethanol steatosis and ALT levels. (A) CYP2E1 and β-galactosidase expression after adenovirus injection into CYP2E1-knockout mice. (B) Serum ALT levels. **P < 0.01, compared with LacZ group. (C) H&E staining after 2 weeks of feeding ethanol to CYP2E1-knockout mice treated with either Ad-LacZ or Ad-CYP2E1 (arrows show lipid droplets). (D) Immunohistochemical staining for CYP2E1 (arrows show lipid droplets).

Discussion

CYP2E1 is elevated in many pathophysiological conditions related to energy metabolism such as diabetes.24–26 CYP2E1 has been reported to be elevated in chronically obese, overfed rats,27 by prolonged starvation,28, 29 and in rats during ketosis induced by a high-fat diet.30 CYP2E1 expression in HepG2 cells is down-regulated by insulin and up-regulated by the thyroid hormone triiodothyronine.31 Hepatic CYP2E1 was increased in humans with nonalcoholic steatohepatitis.32 In rats, there is a strong correlation between the degree of steatosis and ethanol induction of CYP2E1,33 and an inhibitor of CYP2E1, CMZ, not only blunts ethanol-induced inflammation, necrosis, and fibrosis but also lowers liver fat.34 These experiments support the possibility that CYP2E1 plays a role in alcoholic fatty liver. In this study, we used a chronic oral ethanol feeding model to evaluate whether CYP2E1 contributes to experimental alcoholic fatty liver. In this oral feeding model, we observed obvious steatosis in wild-type mice but not in CYP2E1-knockout mice after 2, 3, or 4 weeks of ethanol feeding. Administration of CYP2E1 to the knockout mice restored steatosis. Another approach, use of CMZ to inhibit CYP2E1, blocked ethanol-induced steatosis in wild-type mice. These results show that CYP2E1 contributes to experimental alcoholic fatty liver.

On the other hand, in the intragastric infusion model, Kono et al.35 reported that after the feeding of ethanol for 4 weeks, there was no difference in ethanol-induced steatosis between CYP2E1-knockout mice and wild-type mice. Similarly, in this model, Wan et al.36 reported that ethanol infusion for 21 days promoted fat accumulation in CYP2E1-knockout mice but not in wild-type mice. These two studies used the intragastric infusion model, so it is not clear whether the mode of ethanol delivery, oral versus intragastric, explains these discrepant results on the role of CYP2E1 in ethanol-induced fatty liver. Using the intragastric infusion model, Yin et al.37 reported that TNFα plays an important role in alcoholic fatty liver. Further studies are needed to evaluate the role of TNFα in ethanol-induced steatosis in the oral feeding model. We speculate that because oxidative stress and lipid peroxidation are involved in ethanol-induced steatosis,7, 8 the absence of CYP2E1 lowers oxidative stress and lipid peroxidation in the oral liquid diet model used in our study (for example, Fig. 3C), which lowers ethanol-induced steatosis. Other sources of oxidative stress, for example, endotoxemia, TNFα production, and CYP4A, may occur in the intragastric model to promote ethanol-induced steatosis in the absence of CYP2E1 in that model. This will require further study but is consistent with the liver injury that occurs in the intragastric model in CYP2E1-knockout mice.35

Recent studies have focused on the effects of ethanol on transcription factors regulating fatty acid metabolism.20, 21, 38 Several enzymes involved in fatty acid oxidation are predominantly controlled by PPARα.17 We observed that PPARα was up-regulated after chronic oral feeding with ethanol in the CYP2E1-knockout mice but not in wild-type mice. FFAs are endogenous ligands for nuclear hormone receptors such as PPAR.39 FFAs were elevated 2-fold after 4 weeks of feeding ethanol to the CYP2E1-knockout but not the wild-type mice. However, despite this accumulation of FFAs, hepatic triglyceride levels were not elevated in the livers of the CYP2E1-knockout mice fed ethanol, and the accumulated FFAs were not being incorporated into triglycerides. L-FABP and FAS did not change after ethanol treatment either in the knockout or the wild-type mice, so they do not seem to be responsible for the FFA elevation in the knockout mice. Another possible mechanism in addition to low levels of oxidative stress by which CYP2E1-knockout mice might have failed to develop alcoholic fatty liver would be because PPARα-dependent oxidation of fatty acids was maintained in the knockout mice but decreased in the wild-type mice. PPARα-regulated AOX is the first enzyme of peroxisomal β-oxidation of fatty acids.15 In wild-type mice, AOX protein was decreased by ethanol feeding in association with failure to up-regulate PPARα protein. In the knockouts, AOX protein was not lowered by ethanol in association with up-regulation of PPARα. It appears that in the knockouts, up-regulated PPARα maintained AOX levels and prevented the decrease produced by ethanol in the wild-type mice. Future studies to directly assay fatty acid oxidation are necessary to evaluate this issue, especially because another target of PPARα, CPT1a, was the same in wild-type and knockout mice.

The induction of CYP2E1 has been suggested to play a role in experimental alcoholic liver injury because of the oxidative stress it generates,40, 41 although some studies suggest little or no role for CYP2E1 in early alcoholic liver injury.35 It has been reported that there is a strong association between increased liver steatosis and systemic oxidative alterations in metabolic syndrome patients,42 and lipid peroxidation is significantly increased among patients with nonalcoholic fatty liver disease.43, 44 Antioxidants have been shown to prevent alcohol-induced fatty liver.7 The lipid peroxidation marker malondialdehyde (MDA), assayed via TBAR levels, was increased after 4 weeks of ethanol feeding in the CYP2E1-knockout mice, which may be a result of elevated FFAs. However, TBAR levels were much higher in the wild-type mice than that in the knockout mice, consistent with the concept that CYP2E1 plays an important role in ethanol-induced oxidative stress. Furthermore, the elevated TBAR levels and other indices of oxidative and nitrosative stress (3-NT and HNE adducts) by ethanol feeding were inhibited by CMZ in wild-type mice in association with a decrease in CYP2E1 by the CMZ treatment. Interestingly, PPARα was up-regulated when oxidative stress was inhibited by CMZ treatment. Differences in ethanol-induced oxidative stress in wild-type versus CYP2E1-knockout mice may be important for why fatty liver occurs in the former but not in the latter. It is possible that oxidative stress inhibits fatty acid oxidation through inhibiting PPARα, resulting in fatty liver. This remains to be further evaluated.

With respect to liver injury beyond steatosis, only modest liver injury is observed in the liquid oral models, and small elevations in ALT were observed in the wild-type and knockout mice. These results show that chronic ethanol produces some toxicity in CYP2E1-knockout mice. Elevated FFAs might be one of the reasons for this CYP2E1-independent toxicity. FFAs might evoke hepatocyte damage by ROS generated from oxidation by microsomal enzymes such as CYP4A other than CYP2E1.45 Indeed, CYP4A is up-regulated in CYP2E1-knockout mice fed with a choline-methionine-deficient diet.45 Oxidized FAs themselves become sources for lipid peroxidation reactions that are directly cytotoxic.46 It was recently suggested that accumulated triglycerides may be a protective mechanism to prevent progressive liver injury from FFA lipotoxicity by “buffering the accumulated FFA.”39 The reason for FFA elevation in the knockout mice is not clear. CYP2E1 is a FA-oxidizing enzyme, but it is likely to make only a small contribution to overall fatty acid oxidation. Similarly, ethanol is a substrate of CYP2E1, but serum ethanol levels were the same in knockout and wild-type mice, most likely because only 5%-10% of ethanol is subjected to CYP2E1 oxidation.47

LPS toxicity is increased in fatty liver,4 and chronic ethanol exposure also enhances LPS liver injury.5 Is chronic ethanol exposure–enhanced LPS liver injury a result of alcoholic fatty liver, and does CYP2E1 play a role in this potentiated liver injury? Ethanol-fed wild-type and knockout mice were challenged with LPS; ALT levels were higher in wild-type mice than in CYP2E1-knockout mice. This result is consistent with our previous observations that CYP2E1-knockout mice showed less LPS liver injury after pyrazole treatment than did wild-type mice treated with pyrazole to induce CYP2E1.22 We have suggested that the induction of CYP2E1 primes or sensitizes mice or rats to LPS-induced injury.13, 22 LPS treatment did not change hepatic fat accumulation in either wild-type or knockout mice. The LPS-induced increase in ALT levels in ethanol-fed wild-type mice might be potentiated by the presence of fatty liver in these mice, that is, a higher extent of injury in the wild-type mice that exhibit fatty liver and a lesser extent of injury in the CYP2E1-knockout mice that do not display fatty liver. This is consistent with the “two-hit” theory, with the first hit being hepatic fat accumulation and the second hit LPS. Thus, either CYP2E1 directly synergizes with LPS to promote liver injury, or CYP2E1 plays a role in fat accumulation in the liver, and the latter increases LPS toxicity.

Acknowledgements

We thank Dr. Frank J. Gonzalez (Laboratory of Metabolism, National Cancer Institute, Bethesda, MD) for CYP2E1-knockout and wild-type mice; Professor Paul Van Veldhoven (K.U. Leuven, Belgium) for AOX antibody; and Drs. Stephen Ward and Swan Thung (Department of Pathology, Mount Sinai School of Medicine) for help with necroinflammation evaluation.

Ancillary