Nanoformulated SOD1 ameliorates the combined NASH and alcohol-associated liver disease partly via regulating CYP2E1 expression in adipose tissue and liver

Abstract

Enhanced free fatty acid (FFA) flux from adipose tissue (AT) to liver plays an important role in the development of nonalcoholic steatohepatitis (NASH) and alcohol-associated liver disease (AALD). We determined the effectiveness of nanoformulated superoxide dismutase 1 (Nano) in attenuating liver injury in a mouse model exhibiting a combination of NASH and AALD. Male C57BL6/J mice were fed a chow diet (CD) or a high-fat diet (HF) for 10 wk followed by pair feeding of the Lieber-DeCarli control (control) or ethanol (ET) diet for 4 wk. Nano was administered once every other day for the last 2 wk of ET feeding. Mice were divided into 1) CD + control diet (CD + Cont), 2) high-fat diet (HF) + control diet (HF + Cont), 3) HF + Cont + Nano, 4) HF + ET diet (HF + ET), and 5) HF + ET + Nano. The total fat mass, visceral AT mass (VAT), and VAT perilipin 1 content were significantly lower only in HF + ET-fed mice but not in HF + ET + Nano-treated mice compared with controls. The HF + ET-fed mice showed an upregulation of VAT CYP2E1 protein, and Nano abrogated this effect. We noted a significant rise in plasma FFAs, ALT, and monocyte chemoattractant protein-1 in HF + ET-fed mice, which was blunted in HF + ET + Nano-treated mice. HF + ET-induced increases in hepatic steatosis and inflammatory markers were attenuated upon Nano treatment. Nano reduced hepatic CYP2E1 and enhanced catalase levels in HF + ET-fed mice with a concomitant increase in SOD1 protein and activity in liver. Nano was effective in attenuating AT and liver injury in mice exhibiting a combination of NASH and AALD, partly via reduced CYP2E1-mediated ET metabolism in these organs.

NEW & NOTEWORTHY Increased free fatty acid flux from adipose tissue (AT) to liver accompanied by oxidative stress promotes nonalcoholic steatohepatitis (NASH) and alcohol-associated liver injury (AALD). Obesity increases the severity of AALD. Using a two-hit model involving a high-fat diet and chronic ethanol feeding to mice, and treating them with nanoformulated superoxide dismutase (nanoSOD), we have shown that nanoSOD improves AT lipid storage, reduces CYP2E1 in AT and liver, and attenuates the combined NASH/AALD in mice.

INTRODUCTION

Alcohol-associated liver disease (AALD) is diagnosed in 2 million Americans each year, accounting for 44% of all liver-related deaths. It is the second leading reason for liver transplantation in the United States (26), making AALD a significant public health burden. Nonalcoholic steatohepatitis, another liver pathology that is often associated with obesity, is the third most common indication for liver transplantation in the United States (5). Emerging evidence suggests that the severity of AALD and its progression to steatohepatitis are strongly enhanced by obesity (7, 11). Oxidative stress is a major risk factor in AALD and obesity-related NASH (reviewed in Refs. 21, 46). Mounting evidence suggests that obesity (visceral fat accumulation) and insulin resistance are strongly associated with the severity of AALD (2, 12, 25, 30, 31, 35, 36). Therefore, it is important to study liver injury in a two-hit model involving obesity and ethanol (ET) feeding.

Emerging evidence suggests the importance of adipose tissue (AT) in AALD. An intimate link exists between AT lipolysis and the development of AALD (13, 42, 48). ET promotes AT lipolysis thereby leading to mobilization of free fatty acids (FFAs) to liver. The FFAs are reesterified and stored in the liver leading to hepatic steatosis, an initial event in the pathogenesis of AALD (49). Thus it is reasonable to postulate that in the presence of excess body fat, chronic ET consumption causes increased mobilization of FFAs from AT to liver. It should be pointed out that superoxide generated during CYP2E1-catalyzed ethanol metabolism has an important role in ethanol-induced liver injury (1, 44). Moreover, ethanol metabolism decreases SOD1 levels in liver (29), which can cause further elevation of superoxide levels. CYP2E1 also oxidizes a variety of other substrates, including free fatty acids (FFAs) (16, 37), which likely contribute to obesity + ethanol-induced oxidative stress thereby leading to NASH and AALD (3, 8, 40). Because superoxide is a major contributor to development of liver injury, it is likely that increased FFA flux from AT to liver promotes the pathogenesis of NASH and AALD and that targeting superoxide may be effective in ameliorating these liver pathologies.

Nanoformulated antioxidant enzymes have emerged as promising tools to effectively deliver active proteins to a variety of cells and tissues (32, 45, 47). We previously reported that nanoformulated superoxide dismutase (nanoSOD) was effective in ameliorating AT inflammation and obesity-related nonalcoholic fatty liver disease (28). Recently, we showed that nanoSOD was also effective in attenuating AALD in mice subjected to chronic ethanol feeding (24). However, the effectiveness of nanoSOD on obesity + ET-induced liver injury is unknown and is the focus of the study reported here. As both obesity and ethanol consumption are prevalent in our population, it is important to determine the effectiveness of nanoSOD against NASH and AALD in combination.

Studies investigating the interaction between obesity and ET utilized a model where short- or long-term high-fat (HF) feeding is followed by a single acute ET binge (4, 43). Here, we used a model in which obesity is first induced by feeding mice a HF diet, followed by chronic ET administration for 4 wk with or without nanoSOD treatment. As HF diet feeding (before ethanol administration) increases fat mass, we envisioned that this model would more accurately represent chronic ET + HF diet-induced alterations in both AT and liver. Because ethanol feeding mobilizes FFAs from AT and because both ET and FFAs have the propensity to generate superoxide during their metabolism, we hypothesized that scavenging superoxide using nanoSOD will attenuate liver injury seen in a combination of NASH and AALD.

MATERIALS AND METHODS

NanoSOD preparation.

Nano was synthesized, purified, and characterized as we reported previously (20). Briefly, Cu/Zn SOD (SOD1) (Sigma-Aldrich) was mixed with poly l-lysine-polyethylene glycol copolymer (PLL-PEG). The Cu/Zn SOD-PLL-PEG complex was covalently stabilized using a reducible cross-linker, 3,3′-dithiobis(sulfosuccinimidylproprionate) (Thermo Fisher Scientific). The size of Nano was estimated to be ~44 nm.

Mice and diet.

Six- to eight-week-old male, wild-type C57BL/6 mice were purchased from Jackson Laboratories (Bar Harbor, ME). The mice were divided into five groups. First, we fed mice a chow diet (CD) or a HF diet (45% fat calories, Research Diets) for 10 wk. Ten weeks postdiet, mice were divided into five groups: 1) CD + Lieber-DeCarli control diet (CD + Cont), 2) high-fat diet (HF) + Lieber-DeCarli control diet (HF + Cont), 3) HF + Cont + nanoSOD (Nano), 4) HF + Lieber-DeCarli ET diet (HF + ET), and 5) HF + ET + Nano. The ET-fed group received 5% ET (vol/vol) ad libitum for 4 wk after a 7-day ramp-up period from control diet to the full-strength ET (1% vol/vol ET to a final of 5% vol/vol ET) diet. All groups fed control liquid diet were pair fed with ET-fed mice to maintain equal caloric consumption (17). Two weeks postcontrol or ET diet, mice in the HF + Cont + Nano and HF + ET + Nano groups began receiving intraperitoneal injections of nanoSOD (diluted in 10 mM HEPES) at 1,000 units/kg body wt once every 2 days for a period of 2 wk. The remaining control and ET-fed mice received intraperitoneal injections of 10 mM HEPES alone (vehicle). Both control and ET-fed mice were continued on their respective diets during the treatment period. Thus each mouse received a total of eight injections during this period. A day after the final injection, all mice were fasted for 5 h and then euthanized. All protocols were approved by the Institutional Animal Care and Use Committee at the Veterans Affairs Nebraska-Western Iowa Health Care System.

Body weight and body composition.

Body weight was recorded and body composition (lean and fat mass) was analyzed using EchoMRI body composition analyzer.

Plasma variables.

Markers of liver injury including alanine amino transferase (ALT) and aspartate amino transferase were determined using the Vitros Analyzer. Plasma triglycerides (TGs) and FFAs were measured using kits from Thermo Fisher Scientific and Wako, respectively. Plasma levels of monocyte chemoattractant protein-1 (MCP-1), leptin, and insulin were measured by Luminex xMAP Technology using a Mouse Metabolic Magnetic Bead Panel (no. MMHMAG-44K) as we described previously (24). Briefly, 10 µl of plasma samples in duplicate were mixed with sonicated beads for each marker along with matrix solution and assay buffer in each well of a 96-well plate and incubated overnight on an orbital shaker (110 rpm) at 4°C. The following day, detection antibodies were added and a magnetic plate holder was used to retain the beads during the washing steps. Quantification of beads was with analyte standards, and samples were measured by a Luminex MAGPIX system. Quantification of each analyte was calculated using the manufacturer’s software.

Thiobarbituric acid-reactive substances measurement.

Lipid peroxidation was assessed by quantifying thiobarbituric acid-reactive substances (TBARS) following the procedure of Uchiyama and Mihara using malondialdehyde as a standard (23).

SOD1 activity.

Tissue lysates were collected by sonication and cytosolic fraction was isolated for SOD1 enzyme activity using the SOD Assay Kit from Cayman Chemical Company.

RNA isolation and real-time PCR.

Total RNA was extracted from AT and liver using the TRIzol reagent (Ambion, Life Technologies). RNA was reverse-transcribed to cDNA using 5× iScript Reverse Transcription Supermix (Bio-Rad). Real-time PCR was performed to determine the mRNA levels encoding proteins involved in lipid metabolism and inflammation. A ΔΔCT method was used to calculate gene expression with values normalized to 18S ribosomal RNA.

Western blot analyses.

AT and liver were homogenized in a buffer containing 20 mM Tris·HCl, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 0.5% NP-40, 2.5 mM sodium pyrophosphate, 1 mM sodium orthovanadate, and 1 mM β-glycerophosphate and proteinase inhibitor cocktail (Roche Diagnostics). Proteins were separated by SDS-polyacrylamide gel electrophoresis and immunoblotted onto PVDF membranes, and proteins of interest were detected using antibodies. The following proteins were detected using specific antibodies from Santa Cruz Biotechnology: SOD1 and cytochrome P450 2E1 (CYP2E1). Antibodies against alcohol dehydrogenase 1/2 (ADH) and catalase (CAT) were purchased from Abcam. Antibodies against hormone-sensitive lipase (HSL) and phospho-HSL (Ser 660), perilipin 1, and β-actin were purchased from Cell Signaling Technology. GAPDH antibody was from EMD Millipore. Protein bands were visualized using the Odyssey System (LI-COR Biosciences).

Histology.

Liver tissues were fixed in 10% neutral buffered formalin. The tissues were embedded in paraffin and cut into 4-µm sections. Hematoxylin and eosin (H&E) staining was carried out using our established protocol. Sections were reviewed by Dr. Geoffrey Talmon, a board-certified anatomic pathologist to evaluate for liver injury markers (i.e., steatosis, inflammation, cell death, and fibrosis).

For immunohistochemistry, sections were deparaffinized, rehydrated, followed by antigen retrieval using the DeCloaking Chamber (Biomedical Care). Sections were treated with 3% H₂O₂ for 10 min to quench endogenous peroxidase activity. Sections were incubated with anti-MCP1 primary antibody (Thermofisher Scientific) overnight at 4°C. After being washed, sections were incubated with biotinylated secondary antibody (Vector IHC kit PK-4000) for 1 h at room temperature. After being washed, sections were incubated with Vectastain ABC reagent (avidin-biotinylated horseradish peroxidase) for 30 min followed by incubation with peroxidase substrate (DAB; Vector SK-4100) until desired color intensity developed. Sections were counterstained with hematoxylin.

For immunofluorescence, after incubation with anti-SOD1 primary antibody, sections were washed and incubated with FITC-conjugated secondary antibody for 1 h at room temperature. Autofluorescence was quenched using 0.1% Sudan Black B in 70% ethanol. The sections were mounted with ProLong gold antifade DAPI. For both immunohistochemistry and immunofluorescence analysis, images (×10 and ×20) were captured using Nikon Eclipse 80i inverted microscope.

Statistics.

All data are expressed as means ± SE and analyzed by one-way ANOVA followed by Newman-Keuls multiple comparison test. Differences at P < 0.05 are considered statistically significant.

RESULTS

Effect of ethanol and/or nanoSOD on high-fat diet-induced increase in fat mass.

To study the impact of preexisting obesity on AALD and AT lipid storage, we fed mice a HF diet for 10 wk followed by 4-wk ethanol feeding after a week of acclimatization to ethanol diet. The change in body weight after a 10-wk HF feeding trended toward an increase in HF-fed mice compared with CD-fed controls (Fig. 1A). As expected, the fat mass was significantly increased in all HF-fed groups compared with CD-fed controls (Fig. 1B). Interestingly, we noted a significant reduction in fat mass only in HF + ET-fed mice but not in HF + ET + Nano-treated mice as analyzed by EchoMRI at the end of the treatment period (Fig. 1C). Moreover, the visceral AT (VAT) mass was significantly lower in HF + ET-fed mice compared with both CD + Cont (P < 0.05) and HF + Cont-fed mice (P < 0.01). On the other hand, VAT mass was not significantly altered in HF + ET + Nano-treated mice (Fig. 1D). Together, these data show that an HF diet increases fat mass and VAT mass both of which are reduced by ethanol and that nanoSOD treatment preserves the overall AT mass upon ethanol feeding.

Fig. 1.

Effect of nanoformulated superoxide dismutase [nanoSOD (Nn)] on body weight (BW) and fat mass in high-fat (HF) and/or ethanol (ET)-fed mice. The body weight of mice was recorded at baseline (BL) and at 10 wk post-HF diet and the difference is shown (A). Fat mass was assessed at baseline and at 10 wk postdiet by an EcoMRI body composition analyzer as described in materials and methods, and the difference is shown (B). The fat mass (C) and visceral adipose tissue (VAT) mass (D) were measured at the end of the experimental period. Values are expressed as means ± SE of 6–8 samples in each group. CD, chow diet; CONT, control diet.

Effect of HF + ethanol and/or nanoSOD on markers of AT lipolysis and plasma metabolic and inflammatory variables.

We next analyzed markers of AT lipolysis by Western blot analysis. As shown in Fig. 2, A and B, the levels of phosphorylated hormone-sensitive lipase (HSL), an active form of HSL that promotes lipolysis, showed a trend toward an increase in HF + ET-fed mice compared with both CD and HF controls, and this effect was blocked in HF + ET + Nano-treated mice. Moreover, the level of perilipin 1 (PLIN1), a lipid droplet protein, was significantly lower in HF + ET-treated mice compared with CD control (Fig. 2, A and C). Interestingly, the PLIN1 level was not altered significantly in HF + ET + Nano-treated mice. We determined plasma FFAs, another marker of AT lipolysis, and noted that the total FFAs were significantly higher (P < 0.001) in HF + ET-fed mice compared with both CD + Cont- and HF + Cont-fed mice (Fig. 2D). However, nanoSOD treatment blunted the effect of HF + ET-induced increase in plasma FFAs. Together, these data indicate that reduced lipolysis and improved lipid storage is a potential mechanism by which Nano increases fat mass in these mice. It is noteworthy that plasma leptin levels declined significantly in HF + ET-fed mice compared with controls whereas this effect was blunted by nanoSOD treatment (Fig. 2E). Plasma MCP-1 levels, a marker of inflammatory response, were significantly higher in HF + ET-fed mice compared with control mice whereas nanoSOD treatment diminished this effect (Fig. 2F).

Fig. 2.

Effect of nanoformulated superoxide dismutase [nanoSOD (Nn)] on high-fat (HF) and ethanol (ET)-induced changes in adipose tissue (AT) lipolytic markers and plasma metabolic and inflammatory variables. Visceral adipose tissue (VAT) homogenates were used to analyze markers of lipolysis such as phospho-hormone-sensitive lipase (HSL) and perilipin 1 levels. Protein levels were quantified and representative bands from each group are shown. Vertical lines indicate that the bands were rearranged (A). Phospho-HSL was normalized to total HSL (B) and perilipin 1 was normalized to β-actin (C). Plasma free fatty acid (FFA), a marker of AT lipolysis, was measured (D). The plasma levels of leptin, an AT secretory factor, are shown (E). Changes in plasma monocyte chemoattractant protein 1 (MCP-1), a key marker of ethanol-induced inflammation, were measured and shown (F). Values are represented as means ± SE of 6–8 samples in each group. CD, chow diet; CONT, control diet.

Effect of nanoSOD on HF + ET-induced changes in ethanol-metabolizing enzymes in VAT.

Sebastian et al. (34) showed that ethanol induces CYP2E1 expression in AT which, in turn, promotes AT inflammation and apoptosis. Because we noted an increase in AT mass indicating improved adipocyte survival in HF + ET + Nano (Nn)-treated mice compared with HF + ET-fed mice, we quantified the protein levels of CYP2E1 along with those of other ethanol-metabolizing enzymes in VAT. There was a prominent increase in CYP2E1 protein in HF + ET-fed mice compared with controls. Interestingly, this effect was abolished by nanoSOD treatment (Fig. 3, A and B). On the other hand, the level of ADH trended toward a decrease and a significant decrease in HF + ET and HF + ET + Nn groups, respectively (Fig. 3, A and C). Catalase levels did not vary among the groups (Fig. 3, A and D). These data indicate that CYP2E1 in AT may be involved in the ET-induced decline in AT mass and its storage function and that nanoSOD can prevent AT dysfunction via downregulating CYP2E1 in AT.

Fig. 3.

Effect of nanoformulated superoxide dismutase [nanoSOD (Nn)] on high-fat (HF) and ethanol (ET)-induced changes in alcohol metabolizing enzymes in visceral adipose tissue (VAT). VAT homogenates were used to analyze key markers involved in alcohol metabolism such as CYP2E1, ADH, and catalase (CAT). Protein levels were quantified and normalized to β-actin, and representative bands from each group are shown. Vertical lines indicate that the bands were rearranged (A). Normalized protein values of CYP2E1 (B), ADH (C), and CAT (D) are shown. Values are represented as means ± SE of 6–8 samples in each group. CD, chow diet; CONT, control diet.

Effect of nanoSOD on HF + ET-induced changes in hepatic oxidative stress and plasma transaminases.

We next analyzed markers of NASH/AALD, including oxidant stress and the transaminase levels. Liver weights were not altered significantly among the five groups (Fig. 4A). Analysis of liver TBARS, a measure of oxidant stress, revealed no significant changes among the different groups (Fig. 4B). Notably, plasma ALT, a marker of liver injury was significantly higher in HF + ET-fed mice compared with CD + Cont-fed mice and this effect was blocked by nanoSOD treatment (Fig. 4C). Plasma AST showed only a trend toward an increase in HF + ET-fed mice compared with CD + Cont-fed mice (Fig. 4D).

Fig. 4.

Effect of nanoformulated superoxide dismutase [nanoSOD (Nn)] on high-fat (HF) and ethanol (ET)-induced alterations in hepatic oxidative stress, plasma transaminases, and hepatic fatty acid oxidation genes. Liver weight is shown (A). The degree of lipid peroxidation was evaluated through measurement of thiobarbituric acid-reactive substances (TBARS) and is depicted (B). The levels of transaminases including ALT and AST, markers of plasma liver injury, were assessed (C and D). Quantitative real-time PCR was performed for the mRNA levels of genes involved in fatty acid oxidation including Pparα, Acox1, and Acot1 (E–G). Values are normalized to 18S and expressed as means ± SE of 6–8 samples in each group. CD, chow diet; CONT, control.

Effect of nanoSOD on HF + ET-induced alterations in genes involved in fatty acid oxidation and hepatic steatosis.

We next analyzed genes involved in fatty acid oxidation in liver by realtime PCR. As shown in Fig. 4, E–G, the mRNA levels of Pparα were higher in all four HF-fed groups compared with CD-fed controls. The HF + ET + Nn-treated mice showed a maximum increase in Pparα levels. The mRNA levels of Acox1, a gene involved in peroxisomal fatty acid β-oxidation, was higher in HF + ET and HF + ET + Nn group compared with CD and HF controls. Interestingly, a maximal increase in Acox1 level was noted in the HF + ET + Nn group, indicating increased fatty acid oxidation in these mice. In addition, the mRNA level of Acot1, a PPARα target gene (9), was increased to a greater extent in HF + ET + Nn-treated mice compared with HF + ET-fed mice. Together, these data suggest that nanoSOD was effective in increasing PPARα-mediated fatty acid oxidation in the liver of mice exhibiting a combination of NASH/AALD.

H&E staining of liver sections showed microvesicular steatosis in all the groups and macrovesicular steatosis was more prominent only in the HF + ET and HF + ET + Nn groups (Fig. 5, A–E). The extent of lipid accumulation based on the presence of macrovesicular steatosis was lower in HF + ET + Nn-treated mice compared with HF + ET-fed mice (Fig. 5F). Inflammatory nodules were seen in four out of six mice in HF + ET-fed mice, whereas only two out of six mice showed inflammatory nodules in HF + ET + Nn-treated mice. Of note, three out of six mice in HF + ET group showed hepatocyte ballooning, whereas none of the HF + ET + Nn-treated mice showed this feature. However, no significant difference in these markers were seen among different groups. Fibrosis was not seen in any of the mice in all five groups. Together, these data show that markers of liver injury including steatosis and plasma ALT were higher in HF + ET-fed mice, which were attenuated by nanoSOD treatment.

Fig. 5.

Effect of nanoformulated superoxide dismutase [nanoSOD (Nn)] on high-fat (HF) and ethanol (ET)-induced hepatic steatosis. Liver sections (4 µm) from chow diet (CD) + control diet (CONT) (A), HF + CONT (B), HF + CONT + Nn (C), HF + ET (D), and HF + ET + Nn (E) groups were stained with hematoxylin and eosin. Pictures were taken at ×20 magnification using Nikon Eclipse 80i microscope. Histological grade for the degree of steatosis is shown (F).

Effect of nanoSOD on HF + ET-induced liver inflammation.

We noted that the mRNA levels of Ccl2, a gene encoding MCP-1, increased significantly (P < 0.01) in HF + ET-fed mice compared with both CD + Cont- and HF-Cont-fed mice. On the other hand, treatment with nanoSOD blunted this effect (Fig. 6A). We also observed that Mmp12, another inflammatory gene, increased significantly only in HF + ET-fed mice (P < 0.05) but not in HF + ET + Nn-treated mice compared with CD + Cont-fed mice (Fig. 6B). Furthermore, the mRNA level of Mt2, an antioxidant and anti-inflammatory gene, showed a trend toward an increase in HF + ET-fed mice and a significant increase in HF + ET + Nn-treated mice compared with CD + Cont- (P < 0.01) and HF + Cont-fed mice (P < 0.05) (Fig. 6C). Adgre1 mRNA, which encodes F4/80, a macrophage marker, did not change in any groups (Fig. 6D). Analysis of hepatic MCP-1 protein levels by immunohistochemistry (Fig. 7, A–F) showed that the MCP-1 staining was significantly higher in HF + ET-fed mice compared with CD controls. On the other hand, MCP-1 levels were significantly lower in HF + ET + Nn-treated mice compared with HF + ET-fed mice. Notably, this corresponds with plasma MCP-1 levels as shown in Fig. 2F. These data suggest that HF + ET induces hepatic inflammation while nanoSOD treatment exerts potent anti-inflammatory effects in liver.

Fig. 6.

Effect of nanoformulated superoxide dismutase [nanoSOD (Nn)] on the mRNA expression of inflammatory markers in liver. Quantitative real-time PCR was performed for the mRNA levels of genes altering hepatic inflammation including Ccl2, Mmp12, Mt2, and Adgre (A–D). Values are normalized to 18S and expressed as means ± SE of 6–8 samples in each group. CD, chow diet; HF, high-fat diet; CONT, control; ET, ethanol.

Fig. 7.

Effect of nanoformulated superoxide dismutase [nanoSOD (Nn)] on the protein level of monocyte chemoattractant protein 1 (MCP-1) in liver. Immunohistochemistry was performed to detect MCP-1 in liver sections from chow diet (CD) + control diet (CONT) (A), high-fat diet (HF) + CONT (B), HF + CONT + Nn (C), HF + ethanol (ET) (D), and HF + ET + Nn (E) groups. Representative images at ×20 magnification are shown. Images taken at ×10 magnification in three fields per section were used for quantification (F).

Effect of nanoSOD on HF + ET-induced changes in ethanol-metabolizing enzymes in liver.

We next determined whether nanoSOD treatment altered the enzymes involved in ethanol metabolism in liver as it did in adipose tissue. As shown in Fig. 8, A and B, the protein level of CYP2E1 increased significantly in both HF + ET and HF + ET + Nn groups (P < 0.001) compared with controls. Interestingly, CYP2E1 level was significantly lower in HF + ET + Nn-treated mice compared with untreated HF + ET-fed mice (P < 0.05). The hepatic content of ADH increased significantly in both HF + ET- and HF + ET + Nn-treated mice compared with both CD + Cont- and HF + Cont-fed mice (Fig. 8, A and C). Catalase levels rose significantly in HF + ET-fed mice compared with CD + Cont- (P < 0.001) and HF + Cont-fed mice (P < 0.05). However, catalase levels were higher in HF + ET + Nn-treated mice compared with both CD + Cont- (P < 0.001) and HF + Cont-fed mice (P < 0.001), suggesting enhanced catalase expression (Fig. 8, A and D).

Fig. 8.

Effect of nanoformulated superoxide dismutase [nanoSOD (Nn)] on the key markers of alcohol metabolism in liver. Liver homogenates were subjected to Western blot analysis for alcohol metabolizing enzymes including CYP2E1, ADH, and catalase (CAT). Representative bands from each group are shown, and vertical lines indicate that the bands were rearranged (A). Bands were quantified and values were normalized to GAPDH. Protein levels of CYP2E1 (B), ADH (C), and CAT (D) are shown. Values are expressed as means ± SE of 6–8 samples in each group. CD, chow diet; HF, high-fat diet; CONT, control; ET, ethanol.

Effect of nanoSOD on liver SOD1 protein and activity.

To determine whether Nano treatment increased the hepatic content or activity of SOD1 in liver, we analyzed liver samples for these parameters. Immunofluorescence analysis of liver samples (Fig. 9, A–E) showed that HF + Cont + Nn group showed a maximum increase in SOD1 protein in liver compared with CD and HF controls. On the other hand, HF + ET-fed mice showed a decrease in SOD1 protein compared with HF controls. Interestingly, we noted an increased SOD1 protein level in HF + ET + Nn-treated mice compared with HF + ET-fed mice (Fig. 9F). In line with the immunofluorescence data, we also noted that the SOD1 enzyme activity was significantly increased in HF + ET + Nn-treated mice compared with HF + ET-fed mice (P < 0.05) indicating HF + ET-induced suppression of SOD1 activity is restored by nanoSOD treatment (Fig. 9G).

Fig. 9.

Effect of nanoformulated superoxide dismutase [nanoSOD (Nn)] on SOD1 levels and activity in liver. Immunofluorescence analysis was performed to detect SOD1 in liver sections from chow diet (CD) + control diet (CONT) (A), high-fat (HF) + CONT (B), HF + CONT + Nn (C), HF + ethanol (ET) (D), and HF + ET + Nn (E) groups. Representative images at ×20 magnification are shown. Images taken at ×10 magnification in 3 fields per section were used for quantification (F). Cytosolic SOD1 activity was measured as described in materials and methods (G). Values are expressed as means ± SE of 6–8 samples in each group.

DISCUSSION

Using a novel model of NASH/AALD, induced by HF and ET feeding, we have demonstrated that in addition to liver AT is an important site for ethanol action. We showed that HF diet-induced AT mass significantly declined when these animals were subsequently fed ethanol, which was associated with a rise in plasma FFAs, indicating increased AT lipolysis in HF + ET-fed mice compared with controls. These changes were associated with a prominent increase in AT CYP2E1, an inducible ethanol-metabolizing enzyme. Moreover, HF + ET-fed mice exhibited higher levels of steatosis and hepatic inflammation, associated with elevated plasma ALT and MCP-1. Our findings further demonstrated that HF + ET-induced AT dysfunction and liver injury were both ameliorated after nanoSOD treatment. Together, our data suggest that ethanol reduces the lipid storage capacity of AT and mobilizes FFAs to the liver leading to liver injury and that nanoSOD would be an effective therapeutic agent in ameliorating obesity + ET-induced liver injury (Fig. 10).

Fig. 10.

Effect of nanoformulated superoxide dismutase (nanoSOD) on alcohol-associated liver disease (AALD) in obesity. Using a combined nonalcoholic steatohepatitis (NASH)/AALD model, we have demonstrated that high-fat (HF) diet-induced fat mass was significantly reduced upon subsequent feeding with an ethanol (ET) diet. This was accompanied by an increase in plasma free fatty acids (FFAs), indicating increased adispose tissue (AT) lipolysis in HF + ET-fed mice compared with controls. There was a prominent increase in AT levels of CYP2E1, an ethanol-metabolizing enzyme. Next, we have shown that liver expression of CYP2E1 and inflammatory markers was increased in HF + ET-fed mice. These changes were associated with an increase in plasma monocyte chemoattractant protein 1 (MCP-1) and ALT. Using nanoSOD, we have shown that Nano improves AT mass and reduces AT CYP2E1. We have also shown that HF + ET-induced liver inflammation and liver CYP2E1 level were attenuated and catalase level was enhanced by Nano. We further demonstrated that the HF + ET-induced increase in plasma ALT and MCP-1 was reduced upon Nano treatment. Together, our data suggest that the HF + ET-induced AT dysfunction and liver injury were diminished by Nano treatment and that Nano would be an effective therapeutic agent in ameliorating obesity + ET-induced liver injury.

Growing evidence indicates that AT is an important site of ethanol action. First, AT lipolysis, stimulated by ethanol, contributes to chronic-alcohol-induced hepatic steatosis (48, 49). Second, ethanol induces inflammation in AT (6, 14) although the effect appears to vary between male and female mice (10). Third, the level of ethanol-metabolizing enzymes, in particular, CYP2E1, rises in the AT of ethanol-fed mice, which promotes oxidative stress, apoptosis, and/or inflammation in AT (34, 38). Using a model of obesity + ET-induced organ injury, we have clearly demonstrated that ET feeding causes a loss of fat mass, indicating increased lipolysis. For example, prior HF diet feeding increased fat mass in all groups. However, subsequent ethanol feeding led to a drastic reduction in total fat mass as well as VAT mass in HF + ET-fed mice. Moreover, these mice exhibited a reduction in PLIN1 protein in VAT. Evidence suggests that degradation of perilipin protein is one mechanism for enhanced adipocyte lipolysis (15, 27). We also noted a concomitant increase in plasma FFA levels, further confirming increased AT lipolysis in these animals. Interestingly, nanoSOD treatment of HF + ET mice blocked the decline in fat mass and in PLIN1, which correlated with the reduction in plasma FFAs. Furthermore, the plasma leptin levels reduced drastically in HF + ET-fed mice whereas leptin levels remained elevated in HF + ET + Nano-treated mice. Of note, serum leptin levels reflect the amount of fat stored in AT (reviewed in Ref. 22). Together, these data suggest that nanoSOD preserves the fat storage function of AT, which, in turn, may have a role at least in part, in ameliorating AALD in obesity.

Another significant observation is that the AT CYP2E1 level is greatly increased in HF + ET-fed mice compared with controls. On the other hand, nanoSOD treatment abolished the rise in CYP2E1 in HF + ET-fed mice (Fig. 3, A and B). Superoxide generated during CYP2E1-catalyzed ethanol metabolism contributes significantly to ethanol-induced liver injury (1, 19, 44). While the role of hepatic CYP2E1 in provoking AALD is well known, evidence for the involvement of CYP2E1 in AT derangements by ethanol emerged only a few years ago. Chen et al. (6) showed that ethanol feeding for only seven days increased CYP2E1 levels in rat subcutaneous AT. This rise was associated with an increase in oxidative stress (6). Tang et al. (38) further demonstrated that ethanol induces CYP2E1 in epididymal AT with a concomitant increase in markers of oxidative stress. Moreover, using CYP2E1^−/− mice, Sebastian et al. (34) showed that CYP2E1 is causally associated with AT oxidative stress and apoptosis. Here, we have clearly provided evidence that CYP2E1 expression increased in VAT after HF + ethanol feeding. Additionally, our data showed that nanoSOD has a prominent effect in suppressing CYP2E1 expression in AT. Because AT CYP2E1 promotes oxidative stress and apoptosis, we hypothesize that the improvement in total and VAT mass in HF + ET + Nn-treated mice is due, in part, to lower levels of AT CYP2E1. Further studies are required to prove this notion.

The liver is the major organ involved in ethanol metabolism. Our study showed that not only AT but also liver CYP2E1 is greatly increased in HF + ET-fed mice and that nanoSOD treatment attenuated this rise, albeit to a lesser extent than its effect in reducing CYP2E1 in AT (Fig. 8, A and B). Because CYP2E1 expression is downregulated by nanoSOD in both AT and liver, a question arises as to whether ethanol metabolism is inhibited by nanoSOD. It is interesting to note that the level of catalase, the peroxisomal enzyme involved in ethanol oxidation, is enhanced in livers of HF + ET + Nano-treated mice compared with HF + ET-fed mice (Fig. 8, A and D), indicating that ethanol metabolism via catalase may be increased upon nanoSOD treatment, which may compensate for reduced levels of CYP2E1. In fact, we noted that plasma ethanol was undetectable in both HF + ET- and HF + ET + Nn-treated groups, indicating that ethanol metabolism was not hampered by nanoSOD treatment.

Regarding hepatic steatosis, we did not see a remarkable increase in steatosis in HF + Cont-fed mice compared with CD + Cont-fed mice (Fig. 5, A and B). This could be attributed to the fact that after the HF diet feeding, the mice were not fed ad libitum but were switched to the Lieber-DeCarli liquid control diet the amount of which was matched with the amount of ethanol diet consumed by HF + ET-fed mice the previous day. As for the ethanol effect, we noted a marked increase in hepatic steatosis, which was reduced by nanoSOD treatment. Of note, the mRNA levels of genes encoding proteins involved in fatty acid oxidation rose significantly in HF + ET-fed mice compared with controls, indicating increased fatty acid oxidation upon ethanol feeding in this combined NASH/AALD model. However, these mRNAs were increased to a greater extent in HF + ET + Nn-treated group compared with HF + ET-fed mice. Together, these data suggest increased lipid metabolism in HF + ET-fed mice receiving the Nano treatment.

We previously reported that nanoSOD attenuated ethanol-induced hepatic inflammation (24). Here, we provide evidence that hepatic inflammation induced by feeding a high-fat diet followed by ethanol feeding can also be attenuated by nanoSOD, thereby establishing the anti-inflammatory effect of nanoSOD against ethanol and/or obesity-induced hepatic inflammation. We noted that hepatic levels of Ccl2 and Mmp12 mRNAs were lower after nanoSOD treatment. In addition, we noted that Mt2 mRNA is prominently increased in HF + ET + Nn-treated mice. Mt2 has received considerable attention due to its antioxidant and anti-inflammatory properties (18). It should also be pointed out that nanoSOD exerted similar effects in invoking Mt2 gene expression in heart and aorta of HF diet-fed obese mice in our previous study (33), indicating that nanoSOD has the propensity to induce Mt2 expression in the HF-fed conditions in the presence or absence of ethanol.

As for the putative mechanism by which nanoSOD exerts hepatoprotective effects against AALD, our findings suggest that increased hydrogen peroxide (H₂O₂) generated by SOD1 may have a role in mediating the beneficial effects of nanoSOD, at least in part, against AALD. For example, we noted higher SOD1 catalytic activity in livers of HF + ET + Nn-treated mice compared with HF + ET-fed mice, indirectly suggesting higher production of H₂O₂ (Fig. 9G). Moreover, our data show an enhanced expression of catalase, which converts H₂O₂ to water and molecular oxygen, in livers of HF + ET + Nn-treated mice. Furthermore, we noted enhanced expression of Mt2, an antioxidant and anti-inflammatory gene whose expression is increased by H₂O₂ (39). Although H₂O₂ is a reactive oxygen species, it is now well recognized that it is an important signaling molecule that initiates protective responses to limit or repair oxidative damage (reviewed in Ref. 41). Because H₂O₂ is highly unstable in tissues, limitations exist for conducting precise measurements of its content in vivo. Nevertheless, the fact that H₂O₂ is the end product of SOD1 enzyme activity suggests the involvement of this redox regulatory species in partly mediating the hepatoprotective effects of nanoSOD.

Using a combined model of NASH and AALD, we have demonstrated that nanoSOD is effective in preserving AT mass upon HF + ethanol feeding and in limiting plasma FFA levels. NanoSOD is also effective in blocking CYP2E1 protein expression in VAT and liver. In addition, we provide evidence that nanoSOD is effective in ameliorating systemic and liver inflammation. NanoSOD enhances the protein levels of catalase, which catalyzes H₂O₂ clearance as well as ethanol oxidation. Together, our data suggest that nanoSOD is an effective treatment to ameliorate liver inflammation seen in NASH/AALD and that inhibition of CYP2E1 expression may play a role in part in mediating the hepatoprotective effects of nanoSOD. Our study also indicates that the two-hit model employed in this study is valuable in assessing ethanol-induced alterations in AT as ethanol feeding alone limits the expansion of AT posing challenges to properly analyze its effects in AT.

GRANTS

This project was supported by National Institute on Alcohol Abuse and Alcoholism (NIAAA) Grant R21-AA-025445. V. Saraswathi is also supported by NIAAA Grant R21-AA-027367) and National Cancer Institute Grant R21-CA-238953 and a seed grant from Nebraska Research Initiatives. A. V. Kabanov was partly supported by Carolina Partnership, a strategic partnership between the University of North Carolina at Chapel Hill Eshelman School of Pharmacy and the University Cancer Research Fund through the Lineberger Comprehensive Cancer Center. This study is the result of work conducted with the resources and the facilities at the Veterans Afffairs Nebraska-Western Iowa Health Care System (Omaha, NE).

DISCLOSURES

A. V. Kabanov is the co-inventor of the nanozyme technology at University of Nebraska Medical Center (Patent No. WO2008141155A1). The technology has been out-licensed to NeuroNano Pharma, a start-up company located in Chapel Hill, NC. A. V. Kabanov is a co-founder, shareholder, and director of this company. No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

C.A.C., T.M.D., A.V.K., and V.S. conceived and designed research; T.G., N.K., C.P.-O., T.M.D., E.N.H., A.V.K., and V.S. performed experiments; T.G., N.K., C.P.-O., E.N.H., and G.T. analyzed data; T.G., N.K., G.T., and V.S. interpreted results of experiments; T.G., N.K., C.P.-O., and V.S. prepared figures; T.G. and V.S. drafted manuscript; T.G., N.K., C.A.C., T.M.D., E.N.H., and V.S. edited and revised manuscript; T.G., N.K., C.P.-O., C.A.C., T.M.D., E.N.H., G.T., A.V.K., and V.S. approved final version of manuscript.