Preventive Effects of Indole-3-carbinol against Alcohol-Induced Liver Injury in Mice via Antioxidant, Anti-inflammatory, and Anti-apoptotic mechanisms: Role of Gut-Liver-Adipose Tissue Axis

Abstract

Indole-3-carbinol (I3C), found in Brassica family vegetables, exhibits antioxidant, anti-inflammatory, and anti-cancerous properties. Here, we aimed to evaluate the preventive effects of I3C against ethanol (EtOH)-induced liver injury and study the protective mechanism(s) by using the well-established chronic-plus-binge alcohol exposure model. The preventive effects of I3C were evaluated by conducting various histological, biochemical, and real-time PCR analyses in mouse liver, adipose tissue, and colon, since functional alterations of adipose tissue and intestine can also participate in promoting EtOH-induced liver damage. Daily treatment with I3C alleviated EtOH-induced liver injury and hepatocyte apoptosis, but not steatosis, by attenuating elevated oxidative stress, as evidenced by the decreased levels of hepatic lipid peroxidation, hydrogen peroxide, CYP2E1, NADPH-oxidase, and protein acetylation with maintenance of mitochondrial complex I, II, and III protein levels and activities. I3C also restored the hepatic antioxidant capacity by preventing EtOH-induced suppression of glutathione contents and mitochondrial aldehyde dehydrogenase-2 activity. I3C preventive effects were also achieved by attenuating the increased levels of hepatic proinflammatory cytokines, including IL1β, and neutrophil infiltration. I3C also attenuated EtOH-induced gut leakiness with decreased serum endotoxin levels through preventing EtOH-induced oxidative stress, apoptosis of enterocytes, and alteration of tight junction protein claudin-1. Furthermore, I3C alleviated adipose tissue inflammation and decreased free fatty acid release. Collectively, I3C prevented EtOH-induced liver injury via attenuating the damaging effect of ethanol on the gut-liver-adipose tissue axis. Therefore, I3C may also have a high potential for translational research in treating or preventing other types of hepatic injury associated with oxidative stress and inflammation.

Graphical Abstract

1. Introduction

Alcoholic beverages are widely consumed all over the world. While moderate alcohol [ethanol (EtOH)] drinking is beneficial for mood enhancing and preventing cardiovascular and neurological diseases [1], chronic excessive alcohol consumption can cause various sociomedical and public health problems worldwide [2, 3]. The onset and progression of alcoholic liver disease (ALD) is multifactorial. In the liver, chronic alcohol drinking promotes hepatic steatosis (fat accumulation) and this can progress to more severe forms of liver disease, including liver inflammation [i.e., alcoholic steatohepatitis (ASH)], with a relatively high mortality rate, fibrosis, cirrhosis, hepatocarcinoma and hepatic failure [2, 4]. These facts indicate the critical value of preventing the development of ALD at the early stages better than struggling with its treatment at later time points due to relatively short life expectancy of the patients after clinical diagnosis of ASH [2, 3]. Hepatocyte ballooning and apoptotic hepatocytes are the common features in the liver of patients with ASH [5]. Excessive hepatocyte apoptosis stimulates inflammation and stimulates infiltration of phagocytic neutrophils and/or the production of proinflammatory cytokines and reactive oxygen species (ROS) by innate immune cells [6]. Additionally, oxidative stress-induced hepatocyte apoptosis represents one of the consequences of acute alcohol injury [7]. Thus, it seems like that there is a vicious cycle of hepatocyte apoptosis and immune cell activation that promotes the progression of liver injury in ALD.

Direct injurious effects by ethanol on the liver via increasing hepatic oxidative stress, inflammation, and cell apoptosis might not be the only route via which ethanol produces its deleterious consequences. Recently emerging reports show that adipose tissue dysfunction can influence hepatic metabolism [8]. Ethanol is also known to stimulate adipose tissue inflammation, apoptosis of adipocytes, and release of free fatty acids (FAAs) [9], all of which can participate in EtOH-induced liver injury. Additionally, It is now widely-accepted that endotoxin, released from the gut into portal circulation due to compromised intestinal integrity, plays an important role in mediating inflammatory reactions and advancing liver injury, including cell apoptosis, in response to ethanol [10–14]. Thus, both adipose tissue and gut are likely to play an important role in the development and/or the progression of ALD.

Recently, the beneficial effects of natural compounds with anti-oxidant capacities on preventing ALD in experimental models have been reported, and the bioactive, protective compounds predominantly belong to polyphenols, including flavonoid compounds [15–19]. Indole-3-carbinol (I3C) is a naturally occurring compound produced from glucobrassicin in Brassica (cruciferous) vegetables such as cabbage, broccoli, cauliflower and brussels sprouts. As a nutritional supplement, I3C has received much attention lately due to its promising preventive and treatment properties against various types of cancer [20].

I3C has been suggested as a potential anti-obesity and anti-inflammatory agent [21]. For instance, I3C improved hyperglycemia and hyperinsulinemia in mice fed a n-6 fatty acid containing high-fat diet (HFD) and decreased the expression of many proin ammatory cytokines. The protective action of I3C against HFD-induced hepatic steatosis is mediated, at least in part, through the up-regulation of the sirtuin1 (SIRT1)-AMP-activated protein kinase (AMPK) signaling pathway in the livers of HFD-fed mice [22]. However, the preventive benefits of I3C against the development of ALD have been poorly characterized.

An ongoing challenge for the management of ALD, including ASH with a relatively high rate of mortality [2, 3, 23], is the understanding of the early liver injury mechanisms and the identification of a dietary supplement that can prevent this process at early stages. This is particularly important since early intervention strategies can prevent serious and irreversible damage and deaths. The real dilemma is that many alcoholics cannot control their drinking habits due to addiction and/or physical dependency and still suffer from various medical conditions including ALD with inflammation. Consequently, identification of a safe and preventive agent against ALD is still needed. The aims of this study were: (1) To examine the hypothesis that I3C can prevent or minimize the development of ALD via anti-inflammatory and antioxidant mechanisms, (2) to evaluate the hypothesis that I3C preventive effect may extend beyond its direct action on the liver by evaluating other organs such as adipose tissue and intestine, which are known to contribute to the liver damage, and (3) to establish the pathological role of the gut-liver-adipose tissue axis in promoting ALD. To achieve these aims, we used a well-established model, namely chronic-plus-single-binge ethanol exposure model [24] in the absence or presence of the I3C treatment. Our findings show that I3C may represent a novel, protective strategy against alcoholic liver injury by attenuating oxidative stress, inflammatory response, and apoptosis and that this preventive action is mediated, at least partially, through the gut-liver-adipose tissue axis.

2. Materials and methods

2.1. Materials

All chemicals used in this study were purchased from Sigma Chemical (St. Louis, MO, USA), unless indicated otherwise. Specific antibodies against CYP2E1 (catalog # ab28146), inducible nitric oxide synthase (iNOS) (catalog # ab3523), 3-nitrotyrosine (3-NT) (catalog # ab7084), nicotinamide adenine dinucleotide phosphate oxidase-4 (NADPH-oxidase) (catalog # ab133303), total OXPHOS (catalog # ab110413), IL1β (catalog # 9722), osteopontin (OPN) (catalog # AB8448, Ly6g (catalog # ab25377), and claudin-1 (catalog # ab15098) were from Abcam Inc. (Cambridge, MA, USA). Specific antibodies against poly(ADP-ribose) polymerase-1 (PARP-1) (catalog # sc7150), mitochondrial aldehyde dehydrogenase-2 (ALDH2) (catalog # sc48838), ATP synthase subunit beta (ATP5B) (catalog # sc16690), and secondary antibodies conjugated with horse radish peroxidase were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Antibodies against acetylated lysine (catalog # 9441S) and β-actin antibody (catalog # 4967) were from Cell Signaling Technology, Danvers, MA, USA, while anti-myeloperoxidase antibody (MPO) (catalog # SKU: PP023AA) was purchased from Biocare Medical, Concord, CA, USA.

2.2. Animals and experimental design

Age-matched (8–10 weeks of age) male C57BL/6J mice were obtained from Jackson Laboratories. After one week of acclimation, the mice were randomly divided into four groups (n=10/group): (1) Control, (2) control+I3C, (3) EtOH, and (4) EtOH+I3C. The mice were fed the Lieber-DeCarli control (dextrose) liquid diet ad libitum for 5 d to allow mice to acclimate to the liquid diet and tube feeding. Mice were then subjected to the chronic-plus-single binge ethanol exposure model [24]. Lieber-DeCarli ‘82 Shake and Pour control liquid diet (Bio-Serv, product no. F1259SP) and Lieber-DeCarli ‘82 Shake and Pour ethanol liquid diet (Bio-Serv, product no. F1258SP) were used for diet preparation. Both the control+I3C and EtOH+I3C groups were subjected to I3C (40 mg/kg body weight/day) via daily oral gavage during the acclimation period and throughout the liquid diet feeding period. Afterward, alcohol-fed mice were allowed free access to the Lieber-DeCarli alcohol liquid diet containing 5% (vol/vol) ethanol for 10 days, and control groups were pair-fed with the isocaloric dextrose-control liquid diet, as described [24]. To ensure the adequate I3C delivery, mice were treated with freshly-prepared I3C once a day by oral gavage at a dose of 40 mg/kg body weight/day just before the daily administration of the ethanol or dextrose-control liquid diet. This I3C oral dosage was well tolerated and exhibited preventive effects on hyperglycemia-induced oxidative stress in mice treated for 35 days [25]. In the early morning on day 11, both alcohol-fed and pair-fed control mice were exposed to a single oral (gavage) dose of alcohol (5 g/kg BW) or isocaloric dextrose, respectively, and euthanized 9 h later for tissue collection. Both groups were subjected to a final dose of I3C (40 mg/kg body weight) 1.5 h prior to the last oral dose of binge ethanol (5 g/kg) or dextrose. Liver, epididymal adipose tissues, colon, and serum were collected from each mouse. Parts of the tissues (liver, epididymal adipose tissue, and colon) were fixed in 10% formalin for histological analysis, while the rest of the tissues and the serum were quickly frozen and stored at −80°C until analysis. Animal experiments were performed in accordance with the National Institutes of Health guidelines and approved by the Institutional Animal Care and Use Committee.

2.3. Immunoblots analysis for detecting target proteins

Total liver homogenates were prepared in ice-cold extraction buffer (50 mM Tris–Cl, pH 7.5, 1 mM EDTA, and 1% CHAPS), while mitochondrial fractions were prepared by differential centrifugation, as described previously [26–28]. The extraction buffers were subjected to nitrogen gas to remove the dissolved oxygen. The bicinchoninic acid protein assay reagent (Pierce, Rockford, IL, USA) was used to measure protein concentration for each sample (total protein lysate or mitochondrial protein). Equal amounts of lysates were subjected to 10% SDS-PAGE gels and transferred to nitrocellulose membranes (Hercules, CA, USA). Membranes were blocked with 5% (w/v) nonfat milk proteins. Following the washing steps, membrane was exposed to a primary antibody specific for the target protein in 3.5% (w/v) bovine serum albumin (BSA) overnight at 4°C. Membranes were then washed three separate times to remove the primary antibody and subjected to incubation with the secondary antibody IgG conjugated with horse radish peroxidase (Santa Cruz Biotechnology, Dallas, TX, USA) for 1 h at room temperature. Image was detected by SuperSignal West Pico Kit (Thermo Fisher Scientific, Waltham, MA). β-Actin was used as a loading control for whole liver extracts, while ATP5B was used for mitochondrial proteins. Intestinal homogenates were prepared in RIPA buffer (Thermo Fisher Scientific, Waltham, MA), followed by sonication. Lysates were then subjected to centrifugation 10,000 x g for 20 minutes and the supernatant was used for analysis, and β-actin was used as a loading control.

2.4. Biochemical assays

Liver tissues (50 mg wet weight) were homogenized in 5% Triton X-100 solution and heated in 80–100°C water bath for 2–5 min to solubilize the triglyceride (TG). The samples were then centrifuged at 10,000 × g for 10 min, and the resulting supernatant was used to determine the TG level by following the manufacturer’s protocol (catalog # 10010303), (Cayman Chemical, Ann Arbor, Michigan, USA). Serum concentrations of alanine aminotransferase (ALT) (catalog # 700260) (Cayman Chemical, Ann Arbor, Michigan, USA). Serum concentrations of free fatty acid (FFA) (catalog # K612-100) were determined enzymatically using commercial kits (BioVision Inc., Milpitas, CA). Commercial kits (catalog # MAKO55-1KT, Sigma Aldrich, MO, USA) were used to measure the levels of serum aspartate aminotransferase (AST) and serum ethanol (catalog # ab65343, Abcam, Cambridge, MA, USA), respectively. Serum adiponectin and leptin concentrations were measured using the ELISA kits (catalog # KMP0041 and catalog # KMP0041, respectively) (Thermo Fisher Scientific, Waltham, MA, USA). Serum endotoxin levels were measured using the commercial kit from Lonza (catalog # 50-647U) (Walkersville, MD). Commercial kits were used to measure caspase-1 (catalog # K111-100), hydrogen peroxide (H₂O₂) (catalog # K265) (BioVision Inc., Milpitas, CA), caspase-3 (catalog # ab39401) (Abcam, Cambridge, MA, USA), and glutathione (GSH) levels (catalog # 703002) (Cayman chemical, Ann arbor, MI, USA) in liver homogenate by following the manufacturer’s protocols. Commercially available kits were used to determine the activities of mitochondrial complexes I (catalog # 700930), II and III (catalog # 700950) (Cayman Chemical Inc., Ann Arbor, MI, USA), and mitochondrial ALDH2 (catalog # ab115348) per manufacturer’s protocol. Hepatic and intestinal levels of malondialdehyde+4-hydroxyalkenals (MDA+HAE) (indicators for lipid peroxidation) (catalog # FR22) in the lysates from different tissues were determined by using a commercial kit (Oxford Biomedical Research, Oxford, MI, USA) by following manufacturer’s instructions.

2.5. Histopathology, Immunohistochemistry, immunofluorescence, and terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay

Liver tissue sections from the large lobe were stained with hematoxylin and eosin (H&E). For lipid staining, individual, frozen liver samples embedded in optimal cutting temperature compound were cut (10 μm) and stained with Oil Red O (Sigma-Aldrich, St. Louis, MO, USA).

For immunohistochemistry analysis, formalin-fixed liver, adipose tissue, and colon were processed for immunohistochemistry where 5-μm thick paraffin sections were used. Briefly, deparaffinized tissue (liver, adipose, and colon) sections were exposed to 3% hydrogen peroxide followed by antigen retrieval. Each section was then blocked with 2% (w/v) non-fat skim milk solution. Specific primary antibody for each target protein, such as CYP2E1, 4-HNE, Ly6G, MPO, and tight junction protein claudin-1, was used to incubate overnight at 4°C. Following appropriate washing, the attached primary antibody was linked to the dextran polymer per manufacturer’s protocol (Envision kit, Dako, Carpinteria, CA, USA). The sections were then subjected to immersion into a solution of 3,3′-diaminobenzidine (DAB) for final reaction. For confocal imaging (immunofluorescence), the sections were incubated with a goat anti-rabbit IgG secondary antibody conjugated with Alexa Fluor® 594 (catalog # A11037) or Alexa Fluor® 488 (catalog # A32731) (Thermo Scientific, MA, USA). Nuclei were counter-stained with the DAPI (4′,6-Diamidino-2-Phenylindole, Dihydrochloride) (catalog # D1306) (Thermo Scientific, MA, USA) and visualized under a Zeiss LSM710 confocal microscope (Jena, Germany).

For detection of DNA strand breaks to evaluate the number of apoptotic cells in liver, adipose, and colon tissues, we used the ApopTag Peroxidase in situ apoptosis detection kit (catalog # S7100) (Millipore, Billerica, MA). TUNEL-positive cells were counted in multiple high-power microscope fields for all groups.

2.6. Real-time PCR

Total RNA was isolated from 50 mg of frozen liver or 100 mg of frozen epididymal adipose tissues using Trizol® from Life Technologies (Grand Island, NY) by following the manufacturer’s recommendations. The concentration of RNA samples was measured by Nanodrop® ND-1000 (Thermo Scientific, Wilmington, DE). Real-time quantitative PCR amplification reactions were carried out in 7900HT Sequence Detection System from Applied Biosystems (Foster City, CA) and Eco Real-Time PCR system from Illumina (San Diego, CA) in a 20 μl volume. The reaction was conducted using Power SYBR® Green RNA-to-CT^™ 1-Step Kit from Life Technologies (Grand Island, NY) per manufacturer’s recommendations. Two hundred nM each of forward (F) and reverse (R) primers, and 40 ng of template RNA were used. All reactions were carried out in four biological replicates. The PCR reactions were conducted by following the manufacturer’s recommendations. To distinguish specific amplicons from non-specific amplifications, a dissociation curve was generated and examined. The Ct-values were calculated with SDS 2.3, RQ Manager 1.2 (Applied Biosystems), and Eco® software V4.0 (Illumina) with an automatic adjustment of base line and determination of Ct. The resulting Ct-values were imported to Microsoft Excel worksheet for further analysis. The primers used for the analysis were designed using Primer-BLAST software (www.ncbi.nlm.nih.gov/tools/primer-blast/). The primer sequences were designed to include the intron region of a target gene to avoid amplification of trace amounts of genomic DNA in the samples. The sequences were as follows: TNFα-F, CCCTCACACTCAGATCATCTTCT, TNFα-R, GCTACGACGTGGGCTACAG; IL6-F, TAGTCCTTCCTACCCCAATTTCC, and IL6-R, TTGGTCCTTAGCCACTCCTTC; MCP1-F, AGGTCCCTGTCATGCTTCTG, MCP1-R, TCTGGACCCATTCCTTCTTG; CD11c-F, CTGGATAGCCTTTCTTCTGCTG, CD11c-R, GCACACTGTGTCCGAACTCA; Cyclophilin-F, CAGACGCCACTGTCGCTTT, Cyclophilin-R, TGTCTTTGGAACTTTGTCTGCAA.

2.7. Statistical analysis

Data represent results from at least two separate measurements, unless otherwise stated. Each point represents the mean ± SEM (n=10/group). The significance of differences between groups was determined by One-way ANOVA followed by Tukey’s analysis. Statistical analysis was conducted using Graphpad Prism software (GraphPad Software Inc., version 7) and values with p < 0.05 were considered significant.

3. Results

3.1. I3C treatment decreased EtOH-induced hepatic injury but not hepatic steatosis

H&E and oil red O staining revealed greater steatosis (fat deposition) in the livers of mice after chronic-plus-single binge ethanol exposure compared with the pair-fed control group (Fig. 1A and B). However, I3C treatment did not decrease the elevated accumulation of lipid droplets in ethanol-exposed mice compared with those without I3C. These results were also confirmed by the hepatic TG levels (Fig. 1C). For assessment of liver damage, the serum transaminase ALT and AST activities were measured. Alcohol exposure significantly increased the serum ALT and AST levels compared with dextrose-controls, and both elevations were attenuated by I3C treatment (Fig. 1D and E). To determine whether this preventive effect of I3C is associated with the different rates of alcohol absorption in the digestive organs, the serum alcohol concentrations in the four mouse groups were measured. After chronic-plus-single-binge ethanol exposure, the serum alcohol concentration was markedly elevated in EtOH-fed group, but I3C did not change the increased serum ethanol level after alcohol exposure (Fig. 1F). Compared with the alcohol-exposed control group, I3C administration significantly reduced the serum FFA concentration in the alcohol-exposed mice (Fig. 1G). EtOH-fed mice had a significantly lower serum adiponectin concentration compared with their control, while I3C treatment did not display any effect (Fig. 1H). In contrast, EtOH exposure increased the serum leptin level compared with pair-fed control, and I3C treatment significantly attenuated the increased leptin level (Fig. 1I). Ethanol exposure significantly elevated the serum endotoxin contents compared to the dextrose-control and I3C markedly suppressed the ethanol-mediated elevation of serum endotoxin (Fig. 1J).

Fig. 1. Histological and biochemical analyses for the beneficial effects of I3C on ethanol-induced hepatic injury.

Representative photomicrographs (200×) following (A) H&E, and (B) Oil Red O staining are shown for the indicated mouse groups. The levels of (C) hepatic triglyceride (TG), (D) serum ALT, (E) serum AST, (F) serum ethanol, (G) serum FFA, (H) serum adiponectin, (I) serum leptin and (J) serum endotoxin in different groups are presented. Significance was determined by one-way ANOVA followed by Tukey’s analysis (P<0.05). Labeled characters without a common letter represent significant differences from the other group(s).

3.2. I3C attenuated EtOH-induced hepatocyte apoptosis

To evaluate any difference in apoptosis of the hepatocytes in all four groups, the TUNEL assay was performed. TUNEL-positive apoptotic hepatocytes were remarkably elevated in the EtOH-fed group compared with dextrose-fed control (Fig. 2A and B). I3C treatment significantly prevented the increased hepatocyte apoptosis caused by ethanol exposure (Fig. 2A and B). The preventive effects of I3C on EtOH-induced hepatocyte apoptosis were then confirmed by evaluating both capase-3 activity and cleaved (activated) PARP-1 levels (Fig. 2C–E). Indeed, I3C treatment significantly prevented EtOH-mediated increases in capsase-3 activity (Fig. 2C) and cleaved PARP-1 levels (Fig. 2D and E). Thus, I3C treatment significantly prevented EtOH-mediated effects on hepatocyte apoptosis.

Fig. 2. Preventive effects of I3C on ethanol-induced hepatocyte apoptosis.

(A) TUNEL-positive hepatocytes were marked by black arrows and (B) quantified in high-power fields (×200) for the indicated mouse groups. (C) Caspase 3-activity, (D) representative results of immunoblot and (E) densitometric analysis for cleaved PARP-1 normalized to β-actin are shown. Significance was determined by one-way ANOVA followed by Tukey’s analysis (P<0.05). Labeled characters without a common letter represent significant differences from the other group(s).

3.3. I3C mitigated the increased oxidative stress levels and decreased anti-oxidant capacity in EtOH-fed mice

Since oxidative stress is likely to result from the imbalance between oxidative radical production and their removal by the antioxidant defense system, we first evaluated whether I3C has an antioxidant effect on the chronic-plus-single-binge ethanol exposure model. The immunohistochemical levels of hepatic lipid peroxidation product 4-HNE, a well-established oxidative stress marker, were markedly increased in EtOH-fed group compared with dextrose-control (Fig. 3A). This data was further confirmed by the significantly elevated levels of MDA+HAE in EtOH-fed group than control (Fig. 3B). In addition, hepatic H₂O₂ levels were also significantly increased in EtOH-fed group compared with control (Fig. 3C). I3C treatment completely prevented the ethanol-induced increases in the amounts of lipid peroxidation and H₂O₂ (Fig. 3A–C), suggesting the antioxidant effects of I3C. Following the confirmation of the increased oxidative stress after ethanol exposure and the antioxidant effects of I3C, we evaluated two well-established cellular defense systems that protect the cells against oxidative radicals and that are known to be altered in response to ethanol, namely GSH and mitochondrial ALDH2 [29]. We observed that I3C treatment significantly prevented the decreased GSH levels in response to ethanol (Fig. 3D). However, there was no significant change in the levels of mitochondrial ALDH2 among the four groups (Fig. 3E). However, significantly decreased mitochondrial ALDH2 activity was observed in EtOH-fed group compared with dextrose-control, and that I3C administration completely restored the suppressed ALDH2 activity following ethanol exposure (Fig. 3F).

Fig. 3. Preventive effects of I3C against the increased amounts of lipid peroxidation and H₂O₂ with decreased GSH levels and ALDH2 activity following ethanol exposure.

(A) Representative photomicrographs (200×) of immunohistochemistry for 4-HNE-protein adducts (marker for lipid peroxidation), (B) the amounts of MDA+HAE, (C) H₂O₂, (D) reduced GSH are presented for the indicated mouse groups. (E) Representative results of immunoblot and densitometric analyses for mitochondrial ALDH2 normalized to ATP5B, and (F) ALDH2 activity are shown. Significance was determined by one-way ANOVA followed by Tukey’s analysis (P<0.05). Labeled characters without a common letter represent significant differences from the other group(s).

We then evaluated the potential protective effects of I3C treatment on EtOH-stimulated induction of proteins known to be sources of oxidative/nitrative radicals such as CYP2E1, NADPH-oxidase, and iNOS. Immunohistochemical evaluation showed remarkably elevated levels of CYP2E1 in EtOH-fed mice compared with dextrose-control, and that this increase was significantly attenuated in I3C+EtOH group (Fig. 4A). The increased levels of hepatic CYP2E1 in ethanol-exposed mice were further confirmed by immunoblot analysis, where I3C addition significantly decreased the elevated levels of CYP2E1 following ethanol treatment (Fig. 4B and C). In addition, NADPH-oxidase levels were slightly but significantly elevated in response to ethanol compared with dextrose-control, and that the I3C treatment completely reversed the increased levels comparable to those of control (Fig. 4B and D). Although there was no significant difference in the levels of iNOS among the four groups, its level was remarkably decreased in I3C+EtOH group compared with the other groups (Fig. 4B and E).

Fig. 4. Preventive effects of I3C on increased oxidative stress-producing proteins following ethanol exposure.

(A) Representative photomicrographs (200×) of immunohistochemistry for CYP2E1 and (B) representative results of immunoblot and (C,D,E) densitometric analysis for CYP2E1, NADPH-oxidase, and iNOS, respectively, normalized to β-actin, are shown for the indicated mouse groups. Significance was determined by one-way ANOVA followed by Tukey’s analysis (P<0.05). Labeled characters without a common letter represent significant differences from the other group(s).

Alteration of mitochondria function provides another, well-established source of oxidative stress. Thus, we determined the effects of I3C treatment on EtOH-induced alteration of mitochondrial complex protein levels and activities (Fig. 5). We analyzed the levels of mitochondrial complex proteins using the total OXPHOS rodent WB antibody cocktail, which contains 5 mouse monoclonal antibodies against CI subunit NDUFB8, CII-SDHB, CIII-Core protein 2 (UQCR2), CIV subunit 1 (MTCO1) and CV alpha subunit (ATP-5A). Immunoblot and statistical analyses showed that the levels of CI subunit NDUFB8, CII-SDHB, and CIII-UQCR2 were significantly decreased in EtOH-exposed group compared with dextrose-control, and that I3C prevented these suppressions in all three complexes (Fig. 5A, B, C, and D). However, we did not observe any significant changes in the levels of CIV-MTCO and CV-ATP-5A among the four groups (Fig. 5A, E, and F). We further confirmed the preventive effects of I3C on EtOH-induced alterations of mitochondrial complexes by measuring the activities of complexes I (Fig 5G), II (Fig. 5H) and III (Fig. 5I). In agreement with the immunoblot data, treatment with I3C slightly but significantly prevented EtOH-induced declines of these complex protein activities and maintained the activities to the levels comparable to dextrose-control levels (Fig. 5G–I).

Fig. 5. Preventive effects of I3C against altered levels of mitochondrial complex proteins and activities induced by ethanol.

(A, B) Representative results of immunoblot and densitometric analysis for CI subunit NDUFB8, (A, C) CII-SDHB, (A, D) CIII-core protein 2 (UQCR2), (A, E) CIV subunit 1 (MTCO1), and (A, F) CV alpha subunit (ATP5A) using the specific total OXPHOS antibody are presented for the indicated mouse groups. ATP5B protein was used for normalization of the mitochondrial proteins. The relative activities of (G) complex I, (H) complex II, and (I) complex III are shown for the different mouse groups. Significance was determined by one-way ANOVA followed by Tukey’s analysis (P<0.05). Labeled characters without a common letter represent significant differences from the other group(s).

3.4. I3C mitigated the increased levels of protein acetylation in EtOH-fed mice

Ethanol metabolism results in increased production of acetate, which can lead to acetyl-CoA formation, and this change can contribute to the formation of protein acetylation [30]. In addition, protein acetylation is considered one of the major proteomic footprints of ALD [31]. In line with this, we monitored markedly increased levels on hepatic protein acetylation in EtOH-treated mice compared with dextrose-control and the elevated protein acetylation was significantly attenuated by I3C treatment (Fig. 6A and B).

Fig. 6. Preventive effects of I3C on ethanol-induced protein acetylation.

(A) Representative results of immunoblot and (B) densitometric analysis for acetyl-lysine proteins normalized to β-actin are presented for the indicated mouse groups. Significance was determined by one-way ANOVA followed by Tukey’s analysis (P<0.05). Labeled characters without a common letter represent significant differences from the other group(s).

Collectively, these results (Fig. 1–6) showed that ethanol increased oxidative stress and hepatocyte apoptosis and that I3C treatment decreased the increased oxidative stress and hepatic cell death by preventing EtOH-mediated suppression of antioxidant defense capacity and increased production of ROS.

3.5. Treatment with I3C protected against liver inflammation and hepatic neutrophil infiltration in EtOH-fed mice

Chronic plus binge ethanol feeding causes liver inflammation [24]. IL1β has gained great interest since it has been suggested as a major player in promoting ASH since the use of IL1β inhibitor (Anakinra) decreased the ASH rate with increased liver recovery [32, 33]. In this study, the level of cleaved (active) IL1β was markedly increased in EtOH-fed mice compared with dextrose-control and the elevation of cleaved IL1β in EtOH-exposed group was significantly prevented by I3C treatment (Fig. 7A and B, respectively). Similar patterns of change were monitored with the cleaved (activated) OPN, which promotes neutrophil infiltration and proinflammatory action [34, 35] (Fig. 7A and C). We also observed the activation of inflammatory caspase-1, known as IL1β converting enzyme as it cleaves the precursor IL1β [36], in EtOH-fed mice compared with dextrose-control and the caspase-1 activation was significantly prevented by the I3C treatment (Fig. 7D). We did not monitor significant changes in the levels of TNF-α (data not shown). Neutrophil infiltration, important for the progression of ALD [35], was characterized by evaluating the markers of neutrophil such as Ly6G and MPO. Immunohistochemical staining for Ly6G (Fig. 7E) and MPO (Fig. 7F) confirmed the infiltration of a significantly higher number of neutrophils into the liver of EtOH-treated mice compared with dextrose-control, and this neutrophil infiltration was largely attenuated by I3C treatment. These results suggest that I3C exhibits an anti-inflammatory effect to protect against ethanol-induced liver injury.

Fig. 7. Protective effects of I3C against ethanol-induced hepatic inflammation.

(A and B) Representative results of immunoblot and densitometric analysis for cleaved IL1β, (A and C) cleaved osteopontin (OPN) and (D) Caspase-1 activities are presented for the indicated mouse groups. β-Actin was used as a loading control for protein normalization. (E and F) Representative photomicrographs (200×) of immunohistochemistry for Ly6g and MPO, respectively, for the different groups. Significance was determined by one-way ANOVA followed by Tukey’s analysis (P<0.05). Labeled characters without a common letter represent significant differences from the other group(s).

3.6. I3C treatment prevented ethanol-induced endotoxemia, oxidative stress, enterocyte apoptosis, and alteration of a tight junction protein in the colon

Histopathological analysis revealed markedly increased levels of epithelial loss of the lamina propria of large intestine (colon) villi in EtOH-fed mice compared with dextrose-control while the ethanol-mediated loss was largely prevented by I3C treatment (Fig. 8A). Excessive amounts of alcohol are known to affect the intestinal integrity and promote the release of endotoxin, which is an etiological factor in the development and/or progression of inflammatory ALD [12, 37, 38]. Thus, we studied whether I3C treatment prevents ethanol-induced endotoxemia by determining the levels of serum endotoxins. Interestingly, treatment with I3C completely prevented the significantly increased serum endotoxin levels in EtOH-fed mice, when compared with dextrose-control (Fig. 8B). Increased serum endotoxin via gut leakiness in response to ethanol might result from many factors, such as increased oxidative stress, apoptosis of enterocytes, and alteration of intestinal tight junctions, all of which can affect intestinal integrity, and thus we evaluated the preventive effects of I3C on these parameters. To confirm the role of I3C in preventing EtOH-induced oxidative stress in the colon, we evaluated its effect on EtOH-induced lipid peroxidation. Immunohistochemical analysis revealed that the levels of 4-HNE protein adducts were markedly increased in EtOH-exposed mice compared with dextrose-control (Fig. 8C). This data was further confirmed by the significantly higher levels of MDA+HAE in EtOH-fed group than dextrose-control (Fig. 8D) and I3C treatment significantly prevented this increase (Fig. 8C and D), suggesting the antioxidant effects of I3C on the intestine. To evaluate the rates of apoptosis of colon enterocytes in the four different groups, the TUNEL assay was performed. TUNEL-positive apoptotic enterocytes were significantly higher in the EtOH-fed group compared with dextrose-control and the elevated enterocyte apoptosis was significantly attenuated by I3C treatment (Fig. 8E and F). Finally, we observed significantly decreased levels of claudin-1, used as an example of gut tight junction proteins, in ethanol-exposed mice compared with dextrose-control, as shown by immunofluorescence microscopy, where green fluorescence represents claudin-1 and blue staining indicates nuclei (Fig. 8G) and immunoblot analysis (Fig. 8H). Treatment with I3C significantly prevented ethanol-mediated decreased levels of claudin-1 (Fig. 8G and H). Thus, I3C seems to prevent ethanol-induced increase in the serum endotoxin levels by decreasing oxidative stress, apoptosis of enterocytes, and alteration of tight junction protein levels in the colon.

Fig. 8. Histological and biochemical analyses for the preventive effects of I3C on ethanol-induced colon injury.

(A) Representative photomicrographs (200×) of H&E staining of colon are presented for the indicated mouse groups. (B) Serum endotoxin, (C) Representative photomicrographs (200×) of immunohistochemistry for 4-HNE-protein adducts, (D) MDA+HAE levels (as markers for lipid peroxidation), (E) TUNEL-positive colon enterocytes and (F) quantification in high-power fields (×200) are shown for the indicated mouse groups. Colon levels of (G) claudin-1 were determined by immunofluorescent microscopy after labelling the protein with anti-claudin-1 antibody followed by green fluorescence-labeled secondary antibody. Nuclei were counter-stained with DAPI (blue). (H) Representative immunoblot and densitometric analyses of claudin-1 in the different groups are shown. Significance was determined by one-way ANOVA followed by Tukey’s analysis (P<0.05). Labeled characters without a common letter represent significant differences from the other group(s).

3.7. I3C attenuated the adipose tissue inflammation and apoptosis in EtOH-fed mice

We next determined whether I3C treatment prevents chronic-binge ethanol-induced adipose tissue inflammation and adipocyte apoptosis in mice. Ethanol feeding significantly increased the mRNA expressions of IL6, MCP1 and CD11c, but not TNF-α, compared to control group (Fig. 9A). The adipose tissue derived cytokines and chemokines are likely to contribute to chronic inflammation in the livers of EtOH-fed mice [39]. Expression of CD11c by adipose macrophages has been reported to be associated with adipose tissue inflammation in models of metabolic syndrome [40]. Indeed, I3C treatment significantly decreased EtOH-mediated increased mRNA transcripts of IL6, MCP1 and CD11c (Fig. 9A). It has been suggested that adipocyte death accompanied by macrophage infiltration into adipose tissues are mechanistically associated with the pathogenesis of adipose tissue inflammation [41]. Indeed, our immunohistochemical analysis showed significantly higher levels of TUNEL-positive apoptotic adipocytes in the EtOH-fed mice compared to dextrose-control group and the elevated adipocyte deaths were significantly prevented by I3C treatment (Fig. 9B and C).

Fig. 9. Beneficial effects of I3C against ethanol-induced epididymal adipose tissue inflammation and adipocyte apoptosis.

(A) The mRNA levels of the indicated genes involved in inflammation and macrophage infiltration and relative to cyclophilin transcript are shown for the different mouse groups. (B) TUNEL positive adipocytes identified by black arrows and (C) quantification in high-power fields are shown. Data are presented as mean ± SEM. Significance was determined by one-way ANOVA followed by Tukey’s analysis (P<0.05). Labeled characters without a common letter represent significant differences from the other group(s).

4. Discussion

To investigate the beneficial effects of I3C on attenuating alcoholic fatty liver and liver injury, we checked the hepatic fat accumulation and serum levels of ALT and AST in the experimental groups. Unexpectedly, although I3C treatment failed to reduce hepatic TG levels in the chronic-plus-binge ethanol feeding model, it significantly decreased the serum concentrations of ALT and AST. The dissociation between hepatic steatosis and injury has been reported previously. For instance, deletion of E-selectin prevented hepatic injury, but not hepatic steatosis in chronic-binge ethanol model [42]. In contrast, hepatic peroxisome proliferator–activated receptor gamma (PPARγ) deficiency reduced hepatic steatosis, but not serum levels of ALT and AST following HFD-plus-binge ethanol (a murine steatohepatitis model) [43]. Thus, steatosis and inflammatory liver injury may be dissociated. In the current model, I3C is not necessarily working through the conventional method by alleviating steatosis first followed by preventing inflammatory liver injury as described in the “two-hit” hypothesis where steatosis primes the liver to secondary insults, including oxidative stress and inflammation, leading to the progression of the hepatic disease [44]. To our surprise, I3C seems to display its protective effects in this model mainly through preventing the secondary insults such as oxidative stress and inflammation, leading to the prevention and/or the slow progression of ALD. It has been reported that I3C prevented HFD-induced hepatic steatosis and obesity [21, 22], which is in contrast to the lack of preventive effect of I3C on ethanol-induced hepatic steatosis observed in this study. These different results suggest that I3C-preventive effects on hepatic steatosis may differ between NAFLD and AFLD. This view is not uncommon and has been reported previously where creatine has been reported to prevent HFD- and choline methionine-deficient diet-induced hepatic steatosis [45, 46], but not in ethanol-induced hepatic steatosis [47]. This might be attributed to the possible different mechanisms by which ethanol or HFD stimulates hepatic steatosis or distinct interactions between I3C and ethanol or HFD. In addition, the preventive effects of I3C do not seem to be related to its effects of ethanol per se since the serum levels of ethanol were similar in both ethanol and ethanol + I3C groups. These results suggest that I3C could have prevented the ethanol-damaging effects independent from serum ethanol levels. In fact, restoration of ALDH2 activity with presumably decreased levels of acetaldehyde and suppression of the elevated CYP2E1 and lipid peroxides could have contributed to the I3C-mediated prevention of gut leakiness, since acetaldehyde, CYP2E1 and lipid peroxides were shown to stimulate epithelial barrier dysfunction and gut leakiness as will be discussed later.

It has been shown that I3C pre-treatment significantly attenuated clonidine-induced oxidative stress and neuronal damage by increasing the levels of antioxidant GSH and decreasing the oxidative stress markers such as apoptosis, lipid peroxidation, and inflammation, resulting in neuroprotection [48]. We thus decided to evaluate the preventive effects of I3C on ethanol-induced apoptosis, oxidative stress, and inflammation in the liver as well as in the gut and adipose tissue since dysfunction or injury of the latter tissues is also known to contribute to and/or aggravate the ethanol-induced liver damage.

Hepatocyte apoptosis is an important pathologic feature of ALD [12, 49, 50]. Once apoptosis of parenchymal cells is activated, caspases initiate cell death by cleaving and activating effector caspases which drive the process of apoptosis [51]. Although more than a dozen caspases have been identified up to date, caspase-3 stands out because it has many cellular targets and its activation produces the morphologic features of apoptosis [52, 53]. Caspase mediated apoptotic cell death is accomplished through the cleavage of several key proteins required for cellular functioning and survival [54]. PARP-1 is one of several known cellular substrates of caspases. Cleavage of PARP-1 by caspases is considered a hallmark of apoptosis [55]. This theory is in accordance with our results since we monitored significantly increased hepatocyte apoptosis in mice subjected to chronic-binge ethanol exposure, at least partially by activation of caspase-3, resulting in PARP-1 cleavage, all of which were significantly prevented by I3C. Together, these results indicate that I3C prevented ethanol-induced hepatotoxic effects partly by suppressing hepatocyte apoptosis.

Increasing evidence shows that oxidative stress is a feature of alcohol-mediated hepatoxicity and contributes to the progression of ALD by stimulating hepatocellular injury directly [56]. Increased oxidative stress can result from the imbalance between their production and removal. They can be produced from the induction and/or activation of various sources such as CYP2E1, NADPH oxidase, iNOS, mitochondrial electron transport chain (ETC, especially complexes I and III), and leptin, all which have been reported to play a role in the development of ALD while it can be reduced or scavenged by many cellular antioxidant defense systems, such as GSH and mitochondrial ALDH2 [57–59]. ROS promote the formation of lipid peroxidation products, such as MDA and 4-HNE, which exhibit both proinflammatory and profibrotic properties, and thus modify essential cellular proteins, resulting in loss of protein function and cellular homeostasis [60]. In addition, it has been reported that antioxidants not only prevent lipid peroxidation but also reduce hepatic focal necrosis and inflammation [61]. Mitochondrial ALDH2, a key enzyme in the ethanol metabolism, has been found to possess the antioxidant property to prevent aldehyde-induced toxicities and decrease levels of lipid peroxidation [29, 62, 63]. It has been suggested that the activity of ALDH2 can be modified via several post-translational modifications including oxidation, nitration, phosphorylation, methylation, adduct formation, etc, and that its functional activity might be suppressed without the change in its protein levels [64–69]. For instance, our laboratory previously showed in detail via systematic investigation with purification of oxidatively-modified proteins followed by mass spectrometry, and then evaluated the expression and activity levels of these modified proteins. Using this systematic method, we clearly demonstrated that both acute ethanol and chronic binge ethanol administration to rodents decreased the activity of mitochondrial ALDH2 although its protein levels were not altered. We further demonstrated that the inhibition was due to oxidative modifications of the active site cysteine residue but not affecting its amounts [67]. This fact is in agreement with the current results where we also demonstrated that the chronic-plus-single-binge ethanol exposure similarly decreased the mitochondrial ALDH2 activities without affecting its protein levels, most probably due to oxidative post-translational modification(s), and all these results were significantly prevented by I3C administration. Recent studies have also indicated that ethanol exposure stimulates global protein hyperacetylation [30]. This reversible, post-translational modification on the ε-amino group of lysine residues of various proteins has been shown to modulate multiple, diverse cellular processes ranging from transcriptional activation to microtubule stability. Due to the reversibility of lysine acetylation of many proteins have led some researchers to suggest that it might rival the protein phosphorylation in its ability to regulate cellular processes [70]. Thus, alcohol-induced protein hyperacetylation likely results in major physiological consequences that play at least a partial role to the progression of hepatotoxicity [30]. Although the exact role of hyperacetylation in response to ethanol is not well defined, it can be considered as the very least one of the markers of ethanol exposure and/or toxicity. In the present study, alcohol significantly increased the levels of hepatic lipid peroxidation and protein acetylation. The development of ethanol-induced oxidative stress was associated with significantly increased levels of CYP2E1 and NADPH-oxidase, but not iNOS. The involvement of mitochondrial dysfunction following exposure to either acute or chronic ethanol has been well-characterized and suggested to be one of main mechanisms by which ethanol induces its injurious effects with elevated oxidative stress-mediated events and inflammation. It has been shown that many anti-oxidant substances exhibited their protective effects against ethanol-induced tissue injury by preventing mitochondrial dysfunction [67, 71]. Mitochondrial impairment has been attributed largely to increased mitochondrial oxidative stress that might result from alteration on the mitochondrial ETC via decreased levels of mitochondrial complexes, alteration of their activities, mitochondrial DNA damage, and inhibition of mitochondrial antioxidant enzymes ALDH2, glutathione peroxidase, etc [67, 71]. This notion is in accordance with our current results where significantly altered amounts and activities of the mitochondrial ETC I, II, and III, in addition to the inactivated mitochondrial ALDH2 and decreased GSH levels were observed in ethanol-exposed mice compared to other groups. Intriguingly, I3C treatment significantly decreased the hepatic MDA+HAE levels compared with those in ethanol-only treatment. I3C also prevented the suppressed mitochondrial complexes activities, ALDH2 activity, and decreased GSH levels, while blocking the elevated amounts of CYP2E1 and NADPH oxidase when compared with ethanol-only exposed group. Furthermore, I3C prevented the significantly increased levels of serum leptin, known to be involved in oxidative stress development [72, 73]. Thus, I3C may scavenge free radicals and/or decrease the levels of lipid peroxidation and H₂O₂ through the prevention of the development of mitochondrial dysfunction, the inhibition of the proteins directly involved in the production of ROS, and the maintenance of the antioxidant defense.

Apoptotic hepatocytes have been shown to co-localize with infiltrating neutrophils, and thus it is likely that apoptosis and inflammation are correlated [49, 50]. Neutrophils have emerged as critical stimulators of hepatic injury by generating oxidative stress and cytotoxic mediators [74, 75], and have been suggested a key player in mediating chronic-binge ethanol-induced hepatic damage [76]. The levels of activated neutrophils are also increased by inflammatory reactions [77–79]. ALD is characterized by upregulation of pro-inflammatory cytokines, including IL6 and IL1β [76]. It has been reported that activation of cleaved IL1β in ALD was caspase 1–dependent and that IL1β signaling was required for the pathogenesis of ASH and fibrosis [33]. In addition, inflammatory cytokines, including IL1β, could be released from: (1) damaged hepatocytes by increased oxidative stress, (2) ethanol-induced adipose tissue inflammation, and (3) ethanol-induced gut permeability change and released endotoxins including lipopolysaccharide, which then activates hepatic Kupffer cells to produce proinflammatory cytokines [32, 33]. Interestingly, the significantly higher neutrophil infiltration and inflammatory liver injury strongly correlated with increased levels of hepatic and circulating levels of cleaved (activated) OPN in ASH [80]. Additionally, activated OPN has been proposed to play a causal role in mediating ASH [35, 76], although this role has been challenged in another study [81]. We observed that mice treated with I3C were protected against hepatic neutrophil infiltration in the chronic-binge alcohol exposure model, as demonstrated by MPO and Ly6G staining, and this effect was associated with decreased levels of activated OPN along with decreased hepatic caspase 1 activity and cleaved IL1β levels. Thus, inhibition of hepatic inflammation may represent another mechanism by which I3C prevented ethanol-induced liver injury.

Hepatic inflammation might be induced in response to ethanol via direct hepatotoxicity or proinflammatory cytokines released from the damaged liver, gut and adipose tissue. The importance of the gut-liver axis in alcohol-mediated liver pathology has been gaining much interest [12, 82–86]. Chronic and binge alcohol intake promotes intestinal permeability change and allows the release of bacterial endotoxin [e.g., lipopolysaccharide (LPS)], which was transported to the liver via portal vein [87, 88]. LPS is likely to activate the hepatic Kupffer cells to release additional amounts of ROS, proinflammatory cytokines and chemokines. These cytokines and chemokines contribute to the intrahepatic recruitment and activation of granulocytes, such as neutrophils, that are found abundantly in acute, severe ASH [89]. These positive feed-forward circles might ultimately lead to hepatocellular apoptosis and/or necrosis. Hence, suppression of the growth of intestinal Gram negative bacteria and preservation of the gut permeability with thereby blocked transfer of endotoxin from the intestine to general blood circulation may be considered as a therapeutic approach to prevent ALD. Intestinal oxidative stress, apoptosis and dysbiosis have been reported to play a critical role in alcohol-induced intestinal barrier dysfunction [90–92]. Recent studies demonstrated that disruption of tight junction proteins appears to be a common mechanism involved in the pathogenesis of endotoxemia in ALD [93]. In addition, Dr. Nagy’s group reported that ethanol disorganized intestinal integrity and causes gut dysbiosis during combined chronic-binge ethanol exposure as evidenced by disruption and decreased levels of tight junction proteins such as claudin-1 in mice and Caco-2 colonic monolayer cells [94]. The same group also reported that intestinal expression of tight junction proteins such as claudin-1 was altered in ileum and colon of C57BL/6J mice treated with 3 different EtOH feeding protocols: chronic feeding (for 25 days), short-term (for 2 days), and acute single gavage (5 g/kg) [95]. Furthermore, Dr. Rao’s group reported that ethanol feeding significantly altered distribution of tight junction proteins, resulting in their depletion in colonic epithelium and that the disruption and depletion were accompanied by decreased amounts of reduced protein thiols and elevated oxidized protein thiols [96]. It has also been shown that acetaldehyde, the metabolite of ethanol, induced alteration in the tight junction proteins in both Caco-2 cell monolayers and human colonic mucosa [97, 98]. In accordance with the aforementioned reports [94–98], ethanol-induced gut leakiness was prevented by I3C, as evidenced by the decreased serum LPS levels with restored claudin-1 levels along with significantly blunted ethanol-induced intestinal oxidative stress with decreased lipid peroxidation and 4-HNE immunohistochemistry. In addition, I3C treatment decreased apoptosis of gut enterocytes caused by the chronic-plus-binge ethanol exposure. The prevention of ethanol-induced oxidative stress and apoptosis of gut enterocytes might, at least partially, contribute to the restoration of the altered levels of tight junction protein claudin-1. These results suggest that I3C preventive effects on ethanol-induced liver injury, were mediated, at least partially, through its protection against ethanol-induced gut leakiness and tight junction alteration.

It has been reported that ethanol can increase apoptosis of adipocytes which may cause adipose tissue inflammation and release of inflammatory cytokines, which might lead to hepatocyte apoptosis and liver damage [99, 100]. Free fatty acids released from adipose tissues may also contribute to increased levels of lipotoxicity in the liver and may play a role in apoptosis of hepatocytes [101, 102]. Ethanol feeding was not associated with adiposity, or change in adipocyte size [103, 104]. This data suggested that there seems to be a strong proinflammatory and proapoptotic milieu in adipose tissues of rodents exposed to ethanol feeding compared with the dextrose-control. Herein, we found that chronic-binge ethanol feeding model increased levels of serum FFAs, apoptosis of adipocytes, and inflammatory cytokines such as IL6 and MCP1, all of which were significantly prevented by I3C treatment. Thus, I3C preventive effects might also be mediated via the prevention of ethanol-induced damaging effects on adipose tissues. The exact underlying mechanism(s) by which I3C exerts its preventive antioxidant and anti-inflammatory effects warrant further investigation.

In conclusion, increased ROS levels in the chronic-binge exposure model were mediated by the combined effects of (1) elevated amounts of the microsomal ethanol-oxidizing enzyme CYP2E1, NADPH-oxidase, and altered mitochondrial complex proteins, and (2) decreased cell antioxidant defense such as GSH and mitochondrial ALDH2 activity. Increased ROS levels may damage and stimulate hepatocyte apoptosis which might then active immune cells as one of the major sources of cytokines including IL1β. Chronic-binge alcohol exposure increased gut leakiness with elevated serum endotoxin that reaches the liver via portal circulation. This might also sensitize hepatic Kupffer cells, leading to the increased production of proinflammatory cytokines. In addition, alcohol feeding induced adipocyte apoptosis and inflammation with elevated release of FFAs, leading to the production of cytokines which can reach the liver via the hepatic artery. Thus, oxidative stress and inflammation might mediate ethanol damaging effects on the liver from multiple sources such as liver, gut and adipose tissue. All these events were significantly prevented by I3C treatment. Collectively, our results show for the first time that I3C significantly prevented ethanol-mediated hepatic injury via its antioxidant, anti-inflammatory, and antiapoptotic effects on the gut-liver-adipose tissue axis (Fig. 10).

Fig. 10. Schematic diagram for the protective effects of I3C against inflammatory liver injury in the chronic-binge ethanol exposure model.

Ethanol increased the ROS levels by the combined effects of elevated amounts of CYP2E1 and NADPH-oxidase with altered mitochondrial complex proteins and by decreased the cellular antioxidant defense such as GSH and mitochondrial ALDH2 activity. Increased ROS may directly damage and/or stimulate apoptosis/necrosis of hepatocytes which might then activate immune cells to produce proinflammatory cytokines, including IL1β, and chemokines. Chronic-binge alcohol exposure also increased gut leakiness with elevated serum endotoxin which reaches the liver via portal vein, contributing to activation of hepatic Kupffer cells to produce large amounts of proinflammatory cytokines. In addition, chronic-binge alcohol feeding stimulated adipocyte apoptosis and inflammation to produce cytokines/adipokines and release FFAs which reach the liver via the hepatic artery. Thus, oxidative stress and inflammation from multiple sources such as liver, gut and adipose tissue are likely to exert ethanol-mediated inflammatory liver injury in a cooperative manner. On the other hand, I3C significantly prevented ethanol-mediated hepatic injury via its antioxidant and anti-inflammatory effects through the gut-liver-adipose tissue axis.