Dietary fat sources differentially modulate intestinal barrier and hepatic inflammation in alcohol-induced liver injury in rats

Abstract

Endotoxemia is a causal factor in the development of alcoholic liver injury. The present study aimed at determining the interactions of ethanol with different fat sources at the gut-liver axis. Male Sprague-Dawley rats were pair fed control or ethanol liquid diet for 8 wk. The liquid diets were based on a modified Lieber-DeCarli formula, with 30% total calories derived from corn oil (rich in polyunsaturated fatty acids). To test the effects of saturated fats, corn oil in the ethanol diet was replaced by either cocoa butter (CB, rich in long-chain saturated fatty acids) or medium-chain triglycerides (MCT, exclusively medium-chain saturated fatty acids). Ethanol feeding increased hepatic lipid accumulation and inflammatory cell infiltration and perturbed hepatic and serum metabolite profiles. Ethanol feeding with CB or MCT alleviated ethanol-induced liver injury and attenuated ethanol-induced metabolic perturbation. Both CB and MCT also normalized ethanol-induced hepatic macrophage activation, cytokine expression, and neutrophil infiltration. Ethanol feeding elevated serum endotoxin level, which was normalized by MCT but not CB. In accordance, ethanol-induced downregulations of intestinal occludin and zonula occludens-1 were normalized by MCT but not CB. However, CB normalized ethanol-increased hepatic endotoxin level in association with upregulation of an endotoxin detoxifying enzyme, argininosuccinate synthase 1 (ASS1). Knockdown ASS1 in H4IIEC3 cells resulted in impaired endotoxin clearance and upregulated cytokine expression. These data demonstrate that the protection of saturated fats against alcohol-induced liver injury occur via different actions at the gut-liver axis and are chain length dependent.

alcoholic liver disease (ALD) is a major health problem worldwide (19, 41). Endotoxemia is closely associated with the pathogenesis of ALD through the stimulation of proinflammatory cytokine production (6, 37). Significantly increased levels of endotoxin were found in both alcoholic patients and experimental animal models of ALD, and the levels correlated well with the development of liver injury (7). Our previous study showed that neutralization of endotoxin by endotoxin neutralizing protein attenuates acute alcohol-induced cytokine production and liver injury, suggesting the importance of endotoxins in mediating alcoholic hepatotoxicity (51). Mechanistic studies have suggested that intestinal bacterial overgrowth and intestinal permeability increase contribute to the development of endotoxemia in alcoholics (35, 37). Our previous work showed that orally administrated lipopolysaccharide was detectable in the plasma of alcohol-intoxicated mice but not in control mice (52), providing direct evidence that alcohol increases gut permeability to endotoxins. Animal studies also showed that prevention of gut leakiness resulted in suppression of alcohol exposure-induced endotoxemia and liver damage (18, 44), suggesting that gut leakiness is a causal factor in the development of alcoholic endotoxemia and liver injury.

Accumulating evidence has indicated that dietary fats are important determinants in the pathogenesis of ALD. The experimental models of ALD are commonly generated by feeding animals the Lieber-DeCarli liquid diet, in which fats in the diet are rich in unsaturated fatty acids (3, 25). Studies of alcohol exposure with dietary fats enriched with saturated fatty acids have been shown to produce much less liver damage compared with the originally used unsaturated fatty acids. These studies applied saturated fatty acids with different carbon chain lengths, including medium-chain triglycerides (MCT) (17, 22, 24, 30–32), long-chain saturated fats (cocoa butter, CB) (47, 48), or a mixture of long-chain saturated fats and monounsaturated fats (beef tallow) (16, 38). However, whether saturated fat sources with different chain lengths make a difference in protection against alcohol-induced liver injury has not been elucidated. Mechanistic studies on the protective effects of saturated fats against alcoholic steatosis have shown that fatty acid oxidation was accelerated through activation of adiponectin and the SIRT1-AMPK pathway and the restoration of hepatocyte nuclear receptor-4α (HNF-4α) (22, 38, 48). Dietary saturated fat reduced alcoholic hepatotoxicity in rats by altering fatty acid metabolism and membrane composition (38). Recent studies showed that dietary fats differentially modulate the intestinal tight junctions and hepatic inflammatory response, suggesting that dietary fats have an impact on the gut-liver axis (16). Although saturated fats have been shown to attenuate alcohol-induced endotoxemia and hepatic inflammation, the mechanisms have not been fully defined.

The fatty acid compositions of MCT and CB are clearly distinguished by chain length. All the fatty acids in MCT have less than 12 carbons, with the most abundant being C8:0 fatty acid (67%). All the fatty acids in CB have more than 16 carbons, and C16:0 (25.3%) and C18:0 (33.5%) are the major saturated fatty acids. The characterizations of MCT and CB enable us to address the question of whether saturated fats of different chain lengths may differentially modulate ethanol-induced pathogenesis. The present study used a liquid ethanol diet with corn oil (CO) as dietary fat to produce alcoholic liver injury in rats, and the effects of dietary CB and MCT on ethanol-induced generation of endotoxin-cytokine signaling at the gut-liver axis were investigated.

MATERIALS AND METHODS

Animals and treatments.

Animal experiments were carried out according to experimental procedures approved by the Institutional Animal Care and Use Committee. Three-month-old male Sprague-Dawley rats (Charles River, Wilmington, MA) were fed the modified Lieber-DeCarli liquid diets (3, 25) containing ethanol (alcohol-fed, AF; n = 8) or isocaloric maltose dextrin as control (pair-fed, PF; n = 6) for 8 wk. Liquid diets were prepared as 1 kcal/ml; protein content was constant at 16% of total calories, fat 34%; and each diet had identical mineral and vitamin content. For ethanol consumption groups, the ethanol content (%, wt/vol) was initially 5% for the first 2 wk and was increased by about 0.14% every 2 wk up to a concentration of 5.43% during the final 2 wk (equals 35% to 38% of total calories in the diet), whereas ethanol was replaced isocalorically with maltose dextrin in control group (Table 1). To determine whether dietary fat sources modulate alcohol-induced liver injury, 30% of total calories were provided by either CO (rich in polyunsaturated fatty acids), CB (rich in long-chain saturated fatty acids), or MCT (rich in medium-chain saturated fatty acids), respectively. Detailed fat compositions of each dietary fat are listed in Table 2. Because the ethanol feeding group with MCT showed the lowest food intake, the other three groups were pair fed the same amount that AF/MCT rats had in the previous day. All ingredients for the liquid diets were obtained from Dyets (Bethlehem, PA) with the exception of ethanol, which was purchased from Sigma-Aldrich (St. Louis, MO). At the end of the experiment, rats were anesthetized with isoflurane after 4-h fasting, and blood, liver, and intestinal samples were harvested for assays.

Table 1.

Nutrient compositions in the liquid diets

	PF/CO	AF/CO	AF/CB	AF/MCT
Protein	16	16	16	16
Carbohydrate	50	12	12	12
Ethanol		38	38	38
Safflower oil	4	4	4	4
Corn oil	30	30
Cocoa butter			30
MCT				30

Values are % of total calories. PF/CO, pair-fed with corn oil; AF/CO, alcohol-fed with corn oil; AF/CB, alcohol-fed with cocoa butter; AF/MCT, alcohol-fed with medium chain triglycerides (MCT).

Table 2.

Fat compositions of dietary fats contained in liquid diets

Fats	Corn Oil	Cocoa Butter	MCT
< C8:0			4.0%
C8:0			67.0%
C10:0			27.0%
C12:0			2.0%
C16:0	11.0%	25.7%
C18:0	2.0%	33.9%
C18:1	28.0%	36.9%
C18:2	58.0%	3.2%
C18:3		0.2%
C20:0	1.0%	0.1%
Saturated fats	14.0%	59.7%	100%
Unsaturated fats	86.0%	40.3%

MCT, alcohol-fed with medium chain triglycerides.

Cell culture and treatments.

H4IIEC3 rat hepatoma cells (American Type Culture Collection, Rockville, MD) were grown in Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA) supplemented with 10% (vol/vol) fetal bovine serum (Atlanta Biologicals, Lawrenceville, GA), 100 U/ml penicillin and 100 μg/ml streptomycin (Invitrogen), at 37°C in a 5% CO₂ environment. Culture medium was changed every 2 or 3 days. For silencing argininosuccinate synthase 1 (ASS1), ASS1 small interfering RNA (siRNA) transfection was conducted with rat ASS1 siRNA or negative control siRNA (Invitrogen) by using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instruction. The silencing effect was validated after 24 h of transfection. Fluorescein isothiocyanate (FITC)-labeled LPS (Sigma-Aldrich) was added to medium at 5 μg/ml 9 h post-siRNA transfection and treated for 15 h. Intracellular LPS was evaluated by both endotoxin assay and microscopy of FITC and 4′,6-diamidino-2-phenylindole (DAPI, Invitrogen), which counterstains the nucleus. Cell viability was determined by the Cell Counting Kit-8 (CCK-8) (Enzo Life Sciences, Farmingdale, NY) relative to the negative control group (negative control siRNA without LPS treatment).

Histopathological analysis of liver.

Liver tissues were fixed in 10% formalin and processed for paraffin embedding. Paraffin sections were cut at 5 μm and processed for hematoxylin and eosin (H&E) staining to assess the histological features of steatosis and inflammation.

Detection of hepatic macrophages and neutrophils.

For detection of hepatic macrophages, liver tissues were frozen in Tissue-Tek CRYO-OCT compound (Fisher Scientific, Waltham, MA). Cryostat tissue sections were cut at 7 μm and fixed with methanol for 10 min at −20°C. Tissue sections were incubated with mouse anti-CD68 or anti-CD163 antibody at 4°C overnight, followed by incubation with Alexa Fluor 594-conjugated donkey anti-mouse IgG (Invitrogen) for 30 min at room temperature. The nuclei were counterstained by DAPI (Invitrogen). The number of positively stained macrophages and area of fluorescence intensity were quantified by Image Pro Premier (Media Cybernetics, Rockville, MD). For detection of hepatic neutrophils, liver tissues were fixed with 10% formalin and cut at 5 μm. The sections were incubated with rabbit anti-myeloperoxidase (MPO) antibody (LifeSpan Biosciences, Seattle, WA) at 4°C overnight, followed by incubation with horseradish peroxidase (HRP)-conjugated anti-rabbit IgG for 30 min at room temperature. Diaminobenzidine was used as HRP substrate for visualization, and the nuclei were counterstained by methyl green (Sigma-Aldrich). The negative controls were conducted by omitting the primary antibody.

Determination of hepatic lipid accumulation.

Hepatic lipid accumulation was assessed by histochemical staining of neutral lipids and biochemical assay of lipid concentrations. For histochemical staining of neutral lipids, cryostat tissue sections were cut at 7 μm and fixed with 10% formalin for 10 min. Tissue sections were stained following the Oil Red O procedure. To quantitatively analyze hepatic lipids, liver tissues were extracted by use of chloroform/methanol (2:1 vol/vol). Protein in the homogenate was assayed by using protein assay reagent (Bio-Rad, Hercules, CA) to normalize the amount of lipid extracted. Triglyceride, cholesterol, and free fatty acid (FFA) levels were measured with assay kits from BioVision (Milpitas, CA).

Blood parameters.

Blood samples were drawn from the dorsal vena cava, and serum was obtained by centrifuging the blood at 8,000 g for 15 min at 4°C. Glucose and β-hydroxybutyrate were determined by using commercial kits purchased from Cayman Chemical (Ann Arbor, MI) following the manufacturer's instructions. Serum alanine aminotransferase (ALT) activity and aspartate aminotransferase (AST) activity were colorimetrically measured with Infinity ALT Reagent and Infinity AST Reagent (Fisher Scientific), respectively. Serum triglycerides were determined by Infinity Triglyceride Reagent (Fisher Scientific).

Metabolomic profiles and data analysis.

Serum and liver samples were prepared and analyzed with high-performance liquid chromatography-time of flight mass spectrometry (HPLC-TOFMS) as described previously (45). An Agilent HPLC 1200 system equipped with a binary solvent delivery manager and a sample manager (Agilent, Santa Clara, CA) was used with chromatographic separations performed on a 4.6 × 150 mm 5-μm Agilent ZoRbax Eclipse XDB-C18 chromatography column. The liquid chromatography elution conditions were optimized as follows: isocratic at 1% B (0–0.5 min), linear gradient from 1% to 20% B (0.5–9.0 min), 20–75% B (9.0–15.0 min), 75–100% B (15.0–18.0 min), isocratic at 100% B (18–19.5 min); linear gradient from 100% to 1% B (19.5–20.0 min) and isocratic at 1% B (20.0–25.0 min). For positive ion mode (ESI+) A = water with 0.1% formic acid and B = acetonitrile with 0.1% formic acid, whereas A = water and B = acetonitrile for negative ion mode (ESI−). Mass spectrometry is performed using an Agilent model 6220 MSD TOF mass spectrometer equipped with a dual-sprayer electrospray ionization source (Agilent). The TOF mass spectrometer was operated with the following optimized conditions: 1) ES+ mode, capillary voltage 3,500 V, nebulizer 45 psig, drying gas temperature 325°C, drying gas flow 11 l/min; and 2) ES− mode, similar conditions as ES+ mode except the capillary voltage is adjusted to 3,000 V. During metabolite profiling experiments, both plot and centroid data were acquired for each sample from 50 to 1,000 Da over a 25-min analysis time.

Principal component analysis (PCA) and orthogonal partial least squares-discriminant analysis (OPLS-DA) were carried out between groups. The differential metabolites were selected when they meet the requirements of variable importance in the projection > 1 in OPLS model and P < 0.05 from Student's t-test. The corresponding fold change shows how these selected differential metabolites varied between groups. Compounds identification was performed by comparing the accurate mass and retention time with reference standard available in our laboratory, or comparing the accurate mass with online database such as the Human Metabolome Database (http://www.hmdb.ca/).

Entodoxin assay.

The endotoxin levels in the blood, liver, and intestinal contents as well as in H4IIEC3 cells were assessed by use of a chromogenic kinetic limulus amoebocyte lysate (LAL) assay kit (QCL-1000, Lonza, Walkersville, MD). Blood samples were drawn from the dorsal vena cava. Serum was obtained by centrifuging the blood at 8,000 g for 15 min at 4°C. Liver samples and intestinal contents were prepared as described elsewhere (1). Briefly, liver tissues and intestinal contents from ileum, cecum, and colon lumen were weighed, collected in sterile, endotoxin-free tubes, diluted with limulus LAL reagent water (1:10 for liver tissues, 1:1,000 for intestinal contents samples), and homogenized. The samples were centrifuged at 14,000 g, for 10 min, and the supernatants were collected for the measurements of endotoxin. H4IIEC3 cells were lysed by Nonidet P-40 (NP-40) lysis buffer (20 mM Tris·HCl pH 7.4, 150 mM NaCl, 0.1 mM EDTA, 1% NP-40). After sonication and centrifugation (14,000 g, for 10 min), the supernatants were collected for assay. Inactivation of inhibitors was obtained by heating all samples for 15 min at 75°C. Standards and samples were incubated with LAL for 10 min at 37°C followed by 6-min incubation with colorimetric substrate. The reaction was stopped with 25% acetic acid, and the absorbance at 405 nm was read. The concentration of endotoxin was expressed in endotoxin units (EU) per milliliter for serum, EU per milligram hepatic tissues, and EU per 10⁶ cells.

Quantitative PCR.

Total RNA from liver or intestinal (ileum, cecum, colon) mucosa was isolated by using TRIzol reagent (Life Technologies, Grand Island, NY) according to the manufacturer's instructions. The isolated RNA was then reverse transcribed with the TaqMan Reverse Transcription Reagents (Life Technologies) after assessment of RNA quantity. Semiquantitative analysis of relative gene expressions were performed on the Applied Biosystems 7500 Real Time PCR Systems (Applied Biosystems, Carlsbad, CA) using SYBR green PCR Master Mix (Qiagen, Valencia, CA). Primers were designed and synthesized by Integrated DNA Technologies (Coralville, CA); sequences are listed in Table 3. All primer pairs were validated by demonstrating high amplification efficiency, consistent single-peak melt curve, and the presence of single product of the expected amplicon size on agarose gels. The relative gene expression was normalized to 18s rRNA expression, and calculated by the 2^−ΔΔCt method (26) setting the values of PF/CO as 1.

Table 3.

Primer sequences used for qPCR analysis

Gene	Genebank Accession No.	Forward Primer (5́-3́)/Reverse Primer (5́-3́)	Amplicon Size
CINC-1/Cxcl1	NM_030845	ACCCAAACCGAAGTCATAGCCAC/ACTAGTGTTGTCAGAAGCCAGCGT	181 bp
MCP-1/Ccl2	NM_031530	TGCTGTCTCAGCCAGATGCAGTTA/TACAGCTTCTTTGGGACACCTGCT	131 bp
Occludin/Ocln	NM_031329	ATGCGGAAAGAGTCGACAGTCCAA/AAGTCATCCACGGACAAGGTCAGA	148 bp
ZO-1/Tjp1	NM_001106266	AAGATGGGATTCTTAGGCCCAGCA/TCTTTGGCTGCAGGGCTATCTTCT	136 bp
LBP/Lbp	NM_017208	TGACATGTTACCGCCTGACTCCAA/AGACCACTGTTCCAAGAAGCTCCA	119 bp
CD14/Cd14	NM_021744	TGAGTTGGGCGAGAAAGGACTGAT/TCCCGCAGTGAATTGTGACTGAGA	174 bp
ASS1/Ass1	NM_013157	TCGTATACACCGGTTTCTGGCACA/ATGTACACCTGGCCCTTGAAGACA	121 bp
18 s rRNA/Rn18 s	NR_046237	ACGGACCAGAGCGAAAGCAT/TGTCAATCCTGTCCGTGTCC	152 bp

qPCR, quantitative PCR.

Western blot.

Protein lysates of the liver or H4IIEC3 cells were extracted by using NP-40 lysis buffer plus protease and phosphatase inhibitor. Protein concentrations were estimated by the Bradford method (Bio-Rad). Aliquots containing 60 μg total proteins were loaded onto a 8–12% sodium dodecyl sulfate-polyacrylamide gel, transblotted onto polyvinylidene difluoride membrane (Bio-Rad), blocked with 5% nonfat dry milk in Tris-buffered saline with 0.1% Tween 20, and then incubated with rabbit anti-ASS1 (Novus Biologicals, Littleton, CO), rabbit anti-CD14 (Santa Cruz Biotechnologies, Santa Cruz, CA), or mouse anti-β-actin (Sigma-Aldrich). The membrane was then incubated with HRP-conjugated donkey anti-rabbit or goat anti-mouse immunoglobulin G (Thermo Scientific, Rockford, IL). The bound complexes were detected with enhanced chemiluminescence (GE Healthcare, Piscataway, NJ) and quantified by densitometry analysis.

Statistical analysis.

All data are expressed as means ± SD (n = 8–10). The results were analyzed by Student's t-test and one-way analysis of variance (ANOVA) followed by Student-Newman-Keuls test. Differences between groups were considered significant at P < 0.05.

RESULTS

Effects of dietary fat sources on routine parameters and histopathological changes in the liver of alcohol-fed rats.

Routine parameters are listed in Table 4. There were no significant differences in average daily food intake and final body weight among all treatments. Alcohol feeding increased liver weight as well as liver-to-body weight ratio regardless of dietary fat sources. Chronic alcohol exposure with CO as the fat source significantly decreased blood glucose level, which was prevented by replacing CO with either CB or MCT. Regardless of dietary fat sources, chronic alcohol exposure dramatically elevated serum β-hydroxybutyrate level compared with controls, although the value in the AF/CB group was relatively lower than AF/CO and AF/MCT groups. Serum ALT and AST levels were increased in all the three groups fed ethanol. Serum triglyceride level was significantly decreased in AF/CO group compared with PF/CO group, which was normalized after replacement of CO with either CB or MCT.

Table 4.

Routine parameters and blood metabolites of rats fed liquid diets for 8 wk

Measurements	PF/CO	AF/CO	AF/CB	AF/MCT
Food intake, g/day per rat	68.72 ± 3.57	66.44 ± 4.52	67.40 ± 5.27	65.59 ± 4.71
Body weight, g	394.20 ± 30.35	393.14 ± 17.12	385.02 ± 29.37	375.57 ± 18.98
Liver weight, g	10.64 ± 0.69a	12.90 ± 0.96b	11.73 ± 0.73b	12.52 ± 0.80b
LW/BW, %	2.82 ± 0.15a	3.21 ± 0.15b	3.12 ± 0.12b	3.39 ± 0.11c
Blood glucose, mg/dl	151.22 ± 12.07a	132.50 ± 7.79b	151.74 ± 20.76ab	149.05 ± 16.57ab
β-Hydroxybutyrate, mM	0.18 ± 0.04a	1.10 ± 0.33b	0.72 ± 0.28c	1.31 ± 0.32b
Alanine aminotransferase, U/l	25.23 ± 3.41a	44.09 ± 4.94b	42.46 ± 2.82b	42.70 ± 2.76b
Aspartate aminotransferase, U/l	38.12 ± 6.38a	46.26 ± 5.59b	41.79 ± 6.90b	45.54 ± 4.76b
Serum triglycerides, mg/dl	150.96 ± 20.64a	98.53 ± 13.04b	139.86 ± 28.74c	154.36 ± 24.28c

Results are means ± SD (n = 6–8). LW, liver weight; BW, body weight. Significant differences (P < 0.05) between groups were determined by ANOVA. Means without a common letter differ at P < 0.05.

Histopathological changes of the liver were assessed by light microscopy of tissue sections with H&E staining. As shown in Fig. 1, alcohol exposure with CO caused liver injury, as indicated by lipid droplet accumulation (arrows) and inflammatory cell infiltration (arrowheads). These histopathological changes were alleviated by replacing CO with either CB or MCT. Hepatic lipid accumulation was further assessed by Oil Red O staining of neutral lipids and quantitative analysis of hepatic triglyceride, cholesterol, and FFAs. Ethanol feeding with CO induced remarkable accumulation of lipid droplets, whereas dietary CB or MCT substitution prevented this effect (Fig. 2A). The hepatic concentrations of triglyceride, cholesterol, and FFAs were all significantly higher in AF/CO group than PF/CO group, which were normalized in both AF/CB and AF/MCT groups (Fig. 2B).

Fig. 1.

Liver histopathology in rats fed ethanol with different dietary fats for 8 wk. Liver sections were stained by hematoxylin and eosin. Light microscopy showed lipid accumulation (arrows) and inflammatory cell infiltration (arrowheads) in the liver. Scale bar: 50 μm. CV, central vein; PF/CO, pair fed with corn oil; AF/CO, alcohol fed with corn oil; AF/CB, alcohol fed with cocoa butter; AF/MCT, alcohol fed with medium-chain triglycerides.

Fig. 2.

Lipid accumulation in the liver of rats chronically fed ethanol with different fat sources for 8 wk. A: neutral lipid accumulation in the liver. Neutral lipids were stained with Oil Red O. Scale bar = 50 μm. B: hepatic lipid concentrations. Triglyceride, cholesterol, and free fatty acid were measured with assay kits. Results are means ± SD (n = 6–8). Significant differences (P < 0.05, ANOVA) are indicated by different letters.

Impact of dietary fat sources on metabolite profiles in liver and serum in alcohol-fed rats.

A total of 220 metabolites in liver samples and 167 metabolites in serum samples were identified by HPLC-TOFMS, of which 69 liver tissue metabolites and 71 serum metabolites were further validated by reference standards. The metabolite profiles among groups were evaluated with unsupervised statistics, PCA. Trends of separation between ethanol feeding groups and pair feeding groups were found in both liver and serum metabolites (Fig. 3A). We then constructed the OPLS-DA models. Regardless of dietary fat sources, the three ethanol feeding groups were all clearly separated from the PF/CO group. Moreover, the clusters of the AF/CB and AF/MCT groups distributed closely, whereas they were separated clearly from the AF/CO group in liver samples. In serum samples, the clusters of the AF/MCT group distributed further to AF/CO than the AF/CB group (Fig. 3B).

Fig. 3.

Principal component analysis (PCA) and orthogonal partial least squares-discriminant analysis (OPLS-DA) plots of spectral data of liver and serum metabolites in rats. Metabolite profiles of liver tissue homogenates and serum were analyzed by high-performance liquid chromatography-time of flight mass spectrometry (HPLC-TOFMS) (n = 6–8). A: PCA plots of liver and serum metabolites. B: OPLS-DA plots of liver and serum metabolites.

Table 5 lists liver metabolites with significant changes greater than or equal to twofold in alcohol-fed groups. Alcohol feeding with CO as the fat source increased hepatic concentrations of 27 metabolites and decreased concentrations of 13 metabolites. The most profound increase was estrone glucuronide, with a 449-fold increase. Metabolites showing more than 10-fold increase were bile acids, including cholic acid (15.40-fold), glycocholic acid (10-fold), glycodeoxycholic acid (41-fold), 2-methylglutaric acid (25-fold), and 3-methylglutary carnitine (14-fold). Although dietary CB and MCT attenuated alcohol-increased hepatic estrone glucuronide, cholic acid level was further increased by both CB and MCT, and 3-methylglutary carnitine was elevated by MCT. Metabolites showing reduced concentration in the AF/CO group include four glycerophospholipids (lysoPC 16:1, lysoPC 18:2, lysoPC 20:4, and lysoPC 20:5), three fatty acids (palmitoleic acid, linoelaidic acid, and 5,8,11-eicosatrienoic acid), two monosaccharides (glucose 6-phosphate and 2,3-dihydroxyvaleric acid), an organic oxoanionic compound (phosphoenolpyruvic acid), a purine nucleoside (xanthosine), a keto acid (pyruvic acid), and a sulfonic acid (taurine). Dietary CB and MCT attenuated the reduced levels of all the glycerophospholipids and fatty acids induced by alcohol consumption.

Table 5.

Fold changes of hepatic metabolites in rats subjected to alcohol with different dietary fat sources for 8 wk

Compound	Library	M/Z	RT (min)	Class	FC AF/CO vs. PF/CO	P AF/CO vs. PF/CO	FC AF/CB vs. AF/CO	P AF/CB vs. AF/CO	FC AF/MCT vs. AF/CO	P AF/MCT vs. AF/CO
Phosphoenolpyruvic acid	Std	166.99	4.01	Aliphatic Acyclic Compounds	0.46	7.8E-03*	0.92	6.4E-01	1.01	9.4E-01
Ophthalmic acid	HMDB05765	290.13	4.54	Amino Acids, Peptides, and Analogues	5.52	3.4E-04*	1.19	3.8E-01	2.33	6.4E-03*
Hexanoylglycine	HMDB00701	174.11	19.04	Amino Acids, Peptides, and Analogues	4.48	1.9E-02*	0.57	1.0E-01	0.62	1.6E-01
N-Acetyl-l-phenylalanine	HMDB00512	208.10	19.28	Amino Acids, Peptides, and Analogues	2.89	4.8E-02*	0.51	7.1E-02	0.53	1.1E-01
Succinyladenosine	HMDB00912	382.10	17.68	Amino Acids, Peptides, and Analogues	2.32	3.2E-03*	1.20	2.3E-01	1.29	2.1E-01
γ-Aminobutyric acid	Std	102.07	3.85	Amino Acids, Peptides, and Analogues	2.15	6.4E-04*	0.81	7.8E-02	0.89	4.1E-01
Glutamic acid	Std	146.06	3.85	Amino Acids, Peptides, and Analogues	2.11	3.4E-04*	0.81	5.2E-02	0.89	3.9E-01
Phenylacetylglycine	HMDB00821	192.07	19.12	Amino Acids, Peptides, and Analogues	2.06	1.8E-02*	0.70	9.7E-02	1.20	3.5E-01
S-Nitrosoglutathione	HMDB04645	335.06	4.54	Amino Acids, Peptides, and Analogues	2.03	4.3E-03*	1.55	1.4E-02*	1.56	2.2E-02*
Ureidosuccinic acid	HMDB00828	175.03	13.29	Amino Acids, Peptides, and Analogues	1.53	5.5E-01	3.41	1.5E-02*	1.62	2.4E-01
Nicotinic acid	HMDB01488	124.04	6.32	Aromatic Heteromonocyclic Compounds	1.20	5.8E-01	0.63	1.6E-01	0.38	1.9E-02*
Indoxyl sulfate	HMDB00682	212.00	18.80	Aromatic Heteropolycyclic Compounds	2.55	8.4E-03*	1.33	1.2E-01	1.76	5.1E-03*
5-Hydroxyindoleacetic acid	HMDB00763	190.05	18.53	Aromatic Heteropolycyclic Compounds	2.39	9.1E-03*	0.64	2.8E-02*	0.64	3.3E-02*
N-Acetyl-7-O-acetylneuraminic acid	HMDB00785	352.12	8.82	Carbohydrates and Carbohydrate Conjugates	2.72	1.0E-03*	0.86	3.4E-01	0.53	6.4E-03*
Glucose 6-phosphate	Std	261.03	3.88	Carbohydrates and Carbohydrate Conjugates	0.50	5.0E-02*	0.98	9.4E-01	0.95	8.5E-01
2,3-Dihydroxyvaleric acid	HMDB00421	135.06	3.74	Carbohydrates and Carbohydrate Conjugates	0.37	8.9E-04*	1.10	7.2E-01	0.85	4.7E-01
Estrone glucuronide	HMDB04483	445.18	12.13	Lipids	449.12	2.4E-06*	0.63	1.7E-02*	0.41	1.5E-04*
Glycodeoxycholic acid	Std	448.30	21.93	Lipids	40.85	3.5E-04*	1.10	6.3E-01	1.10	7.5E-01
2-Methylglutaric acid	HMDB00422	145.05	15.92	Lipids	25.14	1.4E-05*	0.77	1.5E-01	1.39	3.9E-01
α-Linolenic acid	HMDB01388	277.21	24.24	Lipids	23.30	1.4E-02*	1.44	2.4E-01	0.65	2.8E-01
Cholic acid	Std	407.27	21.63	Lipids	15.40	2.4E-03*	2.55	3.6E-02*	2.64	3.6E-02*
3-Methylglutarylcarnitine	HMDB00552	288.14	9.22	Lipids	13.94	4.7E-04*	1.06	7.9E-01	2.63	1.1E-02*
Glycocholic acid	Std	464.30	20.39	Lipids	9.50	8.4E-04*	1.15	5.2E-01	0.68	1.7E-01
LysoPC(18:3(6Z;9Z;12Z))	HMDB10387	518.32	22.12	Lipids	8.28	4.8E-03*	1.02	9.7E-01	0.69	2.1E-01
Cholesterol sulfate	HMDB00653	467.32	20.39	Lipids	7.50	4.5E-04*	1.14	5.3E-01	0.62	7.2E-02
LPA(0:0/18:2(9Z;12Z))	HMDB07852	435.25	22.95	Lipids	5.35	1.4E-06*	0.90	3.4E-01	0.89	4.8E-01
Leukotriene A4	HMDB01337	317.21	24.01	Lipids	2.60	5.2E-02*	1.03	9.2E-01	0.93	8.3E-01
Stearoylcarnitine	HMDB00848	428.37	23.77	Lipids	2.17	3.1E-02*	1.42	9.2E-02	0.69	1.9E-01
Propionylcarnitine	HMDB00824	218.14	15.29	Lipids	2.04	4.0E-02*	0.54	2.5E-02*	0.43	9.1E-03*
LysoPC(22:4(7Z;10Z;13Z;16Z))	HMDB10401	572.37	20.85	Lipids	1.86	3.9E-02*	0.69	1.2E-01	0.18	5.9E-04*
α-Tocotrienol	HMDB06327	425.35	22.10	Lipids	1.56	1.2E-01	0.27	2.2E-06*	0.27	3.3E-06*
Linoleyl carnitine	HMDB06469	424.34	22.10	Lipids	1.42	2.4E-01	0.28	8.9E-05*	0.26	1.6E-04*
LysoPC(15:0)	HMDB10381	482.32	22.62	Lipids	1.28	5.6E-01	1.22	5.9E-01	0.10	2.6E-03*
Oleoylcarnitine	HMDB05065	426.36	22.52	Lipids	1.19	7.2E-01	2.54	4.2E-03*	1.38	2.9E-01
Adrenic acid	HMDB02226	333.28	22.13	Lipids	1.12	7.5E-01	0.16	4.9E-03*	0.63	1.3E-01
LysoPC(22:5(7Z;10Z;13Z;16Z;19Z))	HMDB10403	570.35	21.02	Lipids	0.94	8.9E-01	0.07	1.0E-02*	0.38	1.2E-01
l-Octanoylcarnitine	HMDB00791	288.22	18.91	Lipids	0.89	4.2E-01	1.11	5.8E-01	4.93	8.8E-05*
Palmitic acid	HMDB00220	255.23	24.51	Lipids	0.85	4.7E-01	0.18	3.4E-04*	1.26	2.9E-01
Oleamide	HMDB02117	282.28	22.08	Lipids	0.75	3.8E-01	0.38	2.1E-03*	1.43	1.4E-01
Palmitoleic acid	HMDB03229	255.23	22.51	Lipids	0.40	4.9E-02*	3.58	8.5E-04*	4.04	2.0E-02*
LysoPC(18:2(9Z;12Z))	HMDB10386	520.34	20.83	Lipids	0.31	1.5E-02*	2.96	1.8E-02*	2.40	4.1E-02*
LysoPC(20:4(5Z;8Z;11Z;14Z))	HMDB10395	544.34	20.98	Lipids	0.30	2.6E-02*	3.10	7.8E-03*	2.21	1.0E-01
LysoPC(20:5(5Z;8Z;11Z;14Z;17Z))	HMDB10397	542.32	21.68	Lipids	0.27	6.6E-04*	2.09	1.3E-02*	2.44	2.2E-03*
LysoPC(16:1(9Z))	HMDB10383	494.32	19.84	Lipids	0.25	2.8E-02*	3.84	1.6E-02*	3.74	1.6E-02*
5,8,11-Eicosatrienoic acid	HMDB10378	305.25	24.53	Lipids	0.21	1.1E-02*	1.92	1.8E-01	4.58	3.6E-03*
Linoelaidic acid	HMDB06270	279.23	23.50	Lipids	0.20	1.0E-02*	5.14	1.7E-03*	5.75	9.5E-04*
Adenosine monophosphate	HMDB00045	346.05	4.60	Nucleosides, Nucleotides, and Analogues	2.13	1.1E-04*	0.99	8.7E-01	1.16	2.8E-01
Deoxyguanosine	Std	266.10	4.13	Nucleosides, Nucleotides, and Analogues	2.02	4.3E-03*	0.76	8.8E-02	0.78	1.3E-01
Nicotinamide ribotide	HMDB00229	335.06	4.20	Nucleosides, Nucleotides, and Analogues	1.29	5.0E-01	2.37	2.3E-02*	2.19	6.3E-02*
Xanthosine	HMDB00299	285.08	17.01	Nucleosides, Nucleotides, and Analogues	0.50	2.9E-02*	0.98	8.7E-01	0.74	8.8E-02*
1,11-Undecanedicarboxylic acid	HMDB02327	245.18	3.64	Organic Acids and Derivatives	2.16	1.6E-02*	0.60	1.8E-02*	0.94	7.4E-01
Pyruvic acid	Std	87.01	4.65	Organic Acids and Derivatives	0.43	4.2E-03*	0.80	4.8E-01	1.27	4.2E-01
Taurine	Std	126.02	3.73	Organic Acids and Derivatives	0.14	8.5E-05*	0.86	4.2E-01	1.15	4.6E-01

Results are means ± SD (n = 6–8).

^*Significant differences (P < 0.05) between groups were determined by Student's t-test. M/Z, mass-to-charge ratio; RT, retention time; FC, fold change; P, P value.

Table 6 lists serum metabolites with changes greater or equal to twofold in alcohol-fed groups. Alcohol feeding with CO as the fat source increased serum concentrations of 22 metabolites. The most profound increase was pantetheine, an intermediate in the production of coenzyme A. Eight metabolites were increased more than fivefold, including five bile acids, one indole metabolite (indoxyl), one amino acid/derivative metabolite (N-acetylornithine), and one fatty acid/derivative metabolite (tetracosahexaenoic acid). Alcohol feeding with CO as the fat source decreased serum concentrations of three metabolites, including butyrylcarnitine, linoelaidic acid, and pyruvic acid. Dietary CB and MCT showed limited effects on alcohol-induced alterations in serum metabolites. Compared with the AF/CO group, alcohol feeding with CB showed positive effects on seven metabolites and negative effects on two metabolites, whereas alcohol feeding with MCT showed positive effects on nine metabolites and negative effects on six metabolites.

Table 6.

Fold changes of serum metabolites in rats subjected to alcohol with different dietary fat sources for 8 wk

					AF/CO vs. PF/CO		AF/CB vs. AF/CO		AF/MCT vs. AF/CO
Compound	Library	M/Z	RT (min)	Class	FC	P	FC	P	FC	P
Trimethylamine N-oxide	Std	76.07	3.81	Aliphatic Acyclic Compounds	2.54	4.1E-03*	2.96	1.7E-04*	0.84	3.8E-01
N-Acetylornithine	HMDB03357	175.11	4.27	Amino Acids, Peptides, and Analogues	6.69	7.9E-03*	1.27	3.3E-01	2.68	1.1E-04*
Hexanoylglycine	HMDB00701	174.11	18.98	Amino Acids, Peptides, and Analogues	2.37	2.0E-02*	0.96	8.4E-01	0.89	6.1E-01
Isoputreanine	HMDB06009	161.13	4.23	Amino Acids, Peptides, and Analogues	2.26	2.6E-02*	0.96	8.6E-01	0.66	4.8E-02*
Indoxyl	HMDB04094	134.06	19.49	Aromatic Heteropolycyclic Compounds	7.40	2.1E-02*	0.92	8.0E-01	1.89	1.7E-02*
Phenol	HMDB00228	93.03	18.66	Aromatic Homomonocyclic Compounds	3.08	3.6E-02*	2.36	1.2E-02*	1.51	2.8E-01
4-Methoxyphenylacetic acid	HMDB02072	165.05	19.53	Aromatic Homomonocyclic Compounds	2.22	2.6E-02*	1.95	3.1E-03*	0.53	4.2E-02*
Indoxyl sulfate	HMDB00682	212.00	18.84	Aromatic Heteropolycyclic Compounds	2.09	6.1E-03*	1.58	1.7E-03*	2.16	1.4E-04*
Inodxyl glucuronide	HMDB10319	308.08	18.66	Carbohydrates and Carbohydrate Conjugates	2.73	2.4E-04*	1.41	3.3E-02*	1.97	3.2E-04*
Deoxycholic acid	Std	391.28	23.86	Lipids	9.60	3.7E-03*	1.12	5.9E-01	2.45	7.5E-02
Tetracosahexaenoic acid	HMDB02007	357.28	21.80	Lipids	8.74	3.1E-02*	0.98	9.5E-01	2.16	3.1E-02*
Hyodeoxycholic acid	Std	391.28	21.84	Lipids	7.97	1.2E-03*	1.12	6.1E-01	2.20	2.8E-03*
Glycocholic acid	Std	464.29	20.39	Lipids	7.33	4.4E-02*	0.99	9.7E-01	0.81	6.3E-01
Cholic acid	Std	407.27	21.66	Lipids	6.34	6.5E-03*	1.17	5.2E-01	1.05	8.7E-01
3a;7b;12a-Trihydroxyoxocholanyl-Glycine	HMDB00331	466.32	20.35	Lipids	5.90	2.5E-02*	0.80	5.3E-01	0.80	5.4E-01
LysoPC(18:3(9Z;12Z;15Z))	HMDB10388	518.32	22.24	Lipids	3.54	7.5E-03*	1.01	9.8E-01	1.14	5.4E-01
LysoPC(14:0)	HMDB10379	468.31	22.30	Lipids	3.36	4.2E-02*	1.08	7.2E-01	1.30	2.3E-01
18-Oxocortisol	HMDB00332	375.18	21.53	Lipids	2.30	3.3E-03*	1.13	3.8E-01	0.37	5.1E-04*
Elaidic carnitine	HMDB06464	426.36	22.08	Lipids	1.21	3.6E-01	0.93	7.0E-01	0.49	4.0E-02*
Linoleyl carnitine	HMDB06469	424.34	21.55	Lipids	1.20	5.3E-01	0.22	2.6E-03*	0.23	3.2E-03*
Stearoylcarnitine	HMDB00848	428.37	22.84	Lipids	1.04	7.8E-01	2.35	2.1E-06*	0.64	6.4E-03*
Butyrylcarnitine	HMDB02013	232.15	17.86	Lipids	0.60	5.2E-03*	1.47	2.1E-02*	2.76	7.7E-06*
Linoelaidic acid	HMDB06270	279.23	23.52	Lipids	0.23	4.2E-02*	1.41	5.1E-01	4.63	1.1E-02
Pantetheine	HMDB03426	277.13	19.30	Organic Acids and Derivatives	101.62	3.4E-06*	0.47	6.5E-03*	1.06	8.4E-01
2-Hydroxybutanoic acid	Std	103.04	10.80	Organic Acids and Derivatives	3.49	9.3E-05*	0.84	2.7E-01	2.53	2.6E-03*
4-Hydroxybutyric acid	HMDB00710	103.04	12.47	Organic Acids and Derivatives	3.28	1.0E-04*	0.86	3.0E-01	2.51	3.0E-03*
3-Succinoylpyridine	HMDB00992	180.07	18.86	Organic Acids and Derivatives	2.20	5.0E-03*	0.98	9.1E-01	1.05	7.5E-01
Pyruvic acid	Std	87.01	4.56	Organic Acids and Derivatives	0.44	2.0E-03*	1.29	3.3E-01	0.93	7.2E-01

Results are means ± SD (n = 6–8).

^*Significant differences (P < 0.05) between groups were determined by Student's t-test.

Dietary saturated fats attenuated hepatic macrophage activation, chemokine expression, and neutrophil infiltration in alcohol-fed rats.

Immunofluorescence staining showed that CD68⁺ macrophages were significantly increased and enlarged in the livers of rats fed ethanol with CO as the fat source for 8 wk (Fig. 4, A and B). Although ethanol feeding with CO did not affect CD163⁺ macrophage cell numbers, the cells were significantly expanded. Dietary CB or MCT reduced the cells numbers and cell sizes of both CD68⁺ and CD163⁺ macrophages. In accordance, the gene expressions of cxcl1/CINC-1 (the rat analog to human IL-8) and ccl2/MCP-1 were significantly upregulated in the AF/CO group compared with the PF/CO group and normalized in the AF/CB and AF/MCT groups (Fig. 4C). As shown in Fig. 5, ethanol feeding with CO induced neutrophil infiltration in the liver. With dietary substitution of saturated fats, the numbers of neutrophils were decreased.

Fig. 4.

Hepatic inflammation in rats after ethanol feeding with different dietary fats for 8 wk. A: macrophages in the liver detected by immunofluorescence staining. Scale bar: 20 μm. Red: CD68⁺ or CD163⁺ macrophages; blue: 4′,6-diamidino-2-phenylindole (DAPI) counterstaining of nuclei. B: quantitative analysis of CD68⁺ and CD163⁺ macrophages in the liver by Image Pro Premier (n = 25–35). C: hepatic cytokine/chemokine expressions of cxcl1/CINC-1 and ccl2/MCP-1 measured by quantitative PCR (qPCR). The relative gene expression was normalized to 18s rRNA expression and calculated by the 2^−ΔΔCt method setting the value of PF/CO as 1. Results are means ± SD (n = 6). Significant differences (P < 0.05, ANOVA) are indicated by different letters.

Fig. 5.

Hepatic neutrophil infiltration in rats fed ethanol with different dietary fats for 8 wk. Neutrophils (arrows) were detected by immunohistochemistry using anti-myeloperoxidase antibody. Nuclei were counterstained by methyl green. Scale bar = 50 μm.

Dietary medium-chain triglyceride but not CB normalized serum endotoxin level and tight junction protein expression in alcohol-fed rats.

To determine whether CB or MCT prevention from alcohol-induced inflammatory response was through inhibition of endotoxemia, serum endotoxin levels were measured. As shown in Fig. 6A, alcohol feeding with CO as the fat source increased serum endotoxin levels by more than twofold. Although the alcohol-induced elevation of serum endotoxin levels was normalized by MCT, it was not affected by dietary CB.

Fig. 6.

Endotoxemia and intestinal tight junction expression in rats fed ethanol with different dietary fats for 8 wk. A: serum endotoxin levels. Endotoxin levels were measured by the limulus amoebocyte lysate method, and results were expressed in endotoxin units (EU) per milliliter serum. B: intestinal luminal endotoxin levels. Endotoxin levels within the intestinal lumen were expressed in EU per milligram wet weight. C: gene expressions of occludin and zonula occludens (ZO-1) measured by qPCR. The relative gene expression was normalized to 18s rRNA expression and calculated using the 2^−ΔΔCt method, setting the value of PF/CO as 1. Results are means ± SD (n = 6–8). Significant differences (P < 0.05, ANOVA) are indicated by different letters.

To discriminate whether the elevated serum endotoxin level in the AF/CO group was due to bacterial overgrowth or impaired intestinal barrier function, the intestinal luminal endotoxin levels were measured. Chronic ethanol feeding with CO did not affect intragastric endotoxin level within the ileum, cecum, or colon compared with PF/CO rats. Dietary CB lowered cecal and colonic endotoxin levels, whereas MCT decreased endotoxin level within the colonic lumen (Fig. 6B). To further define the mechanisms of how CB and MCT impact intestinal barrier function, the mRNA expression of tight junction proteins were measured. As shown in Fig. 6C, alcohol feeding with CO as the fat source led to dramatic reduction of occludin and zonula occludens (ZO)-1 in all the intestinal segments tested (ileum, cecum, and colon). Dietary MCT upregulated gene expressions of occludin and ZO-1 to normal levels, or even higher, in the ileum and cecum. However, dietary CB did not show protective effects on alcohol-perturbed gene expression of tight junction proteins.

Dietary CB normalized hepatic endotoxin level and endotoxin signaling in association with upregulating ASS1 in alcohol-fed mice.

To determine how CB abrogated the alcohol-induced inflammatory response without preventing the elevation of serum endotoxin, hepatic endotoxin level was measured. As shown in Fig. 7A, alcohol feeding with CO as the fat source increased hepatic endotoxin level, which were normalized by both dietary CB and MCT. To further determine the mechanism of how CB abrogated hepatic endotoxin elevation in alcohol-fed mice, hepatic expression of endotoxin binding (CD14) and detoxification (ASS1) genes/proteins were examined. The gene expressions of CD14 and ASS1 were upregulated in rats fed alcohol with either CO or CB as the fat source (Fig. 7B) and were normalized by MCT. The protein levels of CD14 and ASS1 were measured by immunoblot analysis (Fig. 7C). Although the protein level of CD14 was elevated by alcohol feeding with either CO or CB as the fat source, the protein level of ASS1 was only increased by alcohol feeding with CB as the fat source.

Fig. 7.

Hepatic LPS signaling pathway in rats fed ethanol with different dietary fats for 8 wk. A: hepatic endotoxin levels. Endotoxin levels in the liver were measured by the limulus amoebocyte lysate method, and results were expressed in EU per milligram liver tissue. B: expression of genes related to hepatic LPS signaling and detoxification. The mRNA levels of CD14 and argininosuccinate synthase 1 (ASS1) were measured by qPCR. The relative gene expression was normalized to 18s rRNA expression, and calculated using the 2^−ΔΔCt method setting the value of PF/CO as 1. C: Western blot analysis of hepatic ASS1 and CD14. Protein bands were quantified by densitometry analysis and the ratio to β-actin was calculated by setting the value of PF/CO as 1. Results are means ± SD (A: n = 6–8; B: n = 6; C: n = 4). Significant differences (P < 0.05, ANOVA) are indicated by different letters.

Knockdown of ASS1 caused endotoxin accumulation and inflammatory cytokine expression in hepatoma H4IIEC3 cells.

To elucidate the mechanistic link between ASS1 induction and endotoxin clearance, the effects of ASS1 knockdown on endotoxin clearance as well as cytokine expression were measured in rat hepatoma H4IIEC3 cells. As shown in Fig. 8A, siRNA transfection reduced the mRNA and protein levels of ASS1. Treatment with FITC-LPS for 15 h caused more endotoxin accumulation in ASS1 siRNA-transfected cells compared with negative siRNA-transfected cells, as indicated by quantification of cellular endotoxin level and fluorescent microscopy of intracellular FITC-endotoxin (Fig. 8B). Neither siRNA transfection nor endotoxin treatment affected cell viability (Fig. 8C). ASS1 siRNA transfection did not affect endotoxin-induced CD14 mRNA level but significantly upregulated CINC-1 expression (Fig. 8D).

Fig. 8.

Effect of ASS1 in detoxifying LPS in H4IIEC3 cells. A: ASS1 siRNA transfection reduced cellular ASS1 levels. H4IIEC3 were transfected with ASS1 siRNA or negative control for 24 h. Messenger RNA and protein levels of ASS1 were measured by qPCR and Western blot with 18s rRNA and β-actin as the internal control, respectively. B: role of ASS1 in LPS detoxification. H4IIEC3 cells were transfected with ASS1 siRNA or negative control (NG) for 24 h with FITC-LPS treatment during the last 15 h. Intracellular LPS was evaluated by both endotoxin assay and microscopy of FITC (green) and DAPI (blue) nuclear staining. C: Cell viability assay. Cell viability was determined by the Cell Counting Kit-8 (CCK-8) relative to the negative control group (negative control siRNA without LPS treatment). D: gene expression of CD14 and CINC-1. The mRNA levels of CD14 and CINC-1 were measured by qPCR. The relative gene expression was normalized to 18s rRNA expression and calculated by the 2^−ΔΔCt method, setting the values of negative control as 1. Results are means ± SD (A: n = 4; B: n = 3; C: n = 6; D: n = 6). Significant differences (P < 0.05) between groups were determined by Student's t-test (A and B; indicated by *) or ANOVA (C and D; indicated by different letters). NG, negative control.

DISCUSSION

The present study demonstrates that the dietary saturated fats CB and MCT alleviated alcohol consumption-induced liver damage through differential actions at the gut-liver axis as illustrated in Fig. 9. Dietary MCT attenuated alcohol-induced downregulation of tight junction genes of the intestinal epithelium and consequently prevented alcohol-induced endotoxemia and hepatic inflammation. Dietary CB also attenuated the alcohol-induced hepatic inflammatory response. However, it did not improve the detrimental effects of alcohol on tight junction or the development of alcoholic endotoxemia. Unexpectedly, dietary CB normalized hepatic endotoxin level in association with upregulation of ASS1. Cell culture study demonstrated that ASS1 is a critical determinant of hepatocyte endotoxin clearance and endotoxin-induced chemokine expression. These results suggest that MCT and CB prevent the alcohol-induced endotoxin-chemokine cascade through differential mechanisms.

Fig. 9.

Schematic diagram of the potential mechanisms that underlie the protective actions of dietary saturated fats against alcoholic liver injury in rats. EtOH, ethanol; CB, cocoa butter; MCT, medium-chain triglyceride.

Endotoxin is a trigger of Kupffer cell activation that leads to production of proinflammatory cytokines in the liver. Several hypotheses have been suggested to be involved in the pathogenesis of endotoxemia, including bacterial overgrowth in the intestine and increased gut permeability (6, 37). Although overgrowth of bacteria in the intestine has been reported in alcoholic patients (28), only a minority of ALD patients had intestinal bacterial overgrowth. On the other hand, all the patients with different stages of ALD, including alcoholic steatosis, hepatitis, and cirrhosis, showed an increase in gut permeability to macromolecules (34). Similarly, increased permeability of isolated gut sac to macromolecules has been shown in alcohol-fed animals (49). The present study demonstrated that alcohol feeding increased serum endotoxin level without affecting the intestinal endotoxin level. Thus intestinal endotoxin overproduction is not an indispensable cause of serum endotoxin elevation in our rat model. Dietary saturated fats have been shown to attenuate alcoholic endotoxemia by sealing the leaky gut (16). A liquid alcohol diet with MCT as the fat source reversed alcohol/fish oil feeding elevated serum endotoxin levels (30, 31). The present study also showed that dietary MCT normalized the serum endotoxin elevation. However, dietary CB had no effect on alcohol-induced serum endotoxin elevation. These data suggest that the beneficial effects of saturated fats on the gut are chain length dependent.

Prevention of endotoxin penetration from the gut lumen to the blood is controlled by the intestinal barrier. The intestinal barrier function is provided by the epithelial cells and the paracellular apical junction complex, including tight junctions and adherence junctions (20). Tight junctions are the most apical part of the apical junction complex and are primarily involved in the regulation of epithelial permeability (43). Tight junctions are assembled by the organization of proteins such as occludin, claudins, and ZO (20, 43). Disassembling of proteins at tight junctions disrupts the intestinal barrier to allow the diffusion of macromolecules such as endotoxin and other pathogens from the intestinal lumen into the blood. Decreased distribution of tight junction proteins, including ZO-1, occludin, and claudin-1, in the intestinal epithelium has been reported in ALD patients as well as animal models (2, 49). A recent study demonstrated that replacing CO with saturated fats (MCT-beef tallow 82:18) attenuates alcohol-induced downregulation of intestinal tight junction genes (16). The present study found that dietary MCT, but not CB, normalized gene expression of tight junction genes in ileum and cecum. Our previous report showed that reduction of intestinal HNF-4α accounts for alcohol-induced downregulation of tight junction proteins (50). We also found that polyunsaturated fatty acid is an inhibitor and MCT is an activator of HNF-4α (22). Thus activation of HNF-4α may account for the beneficial effects of MCT on intestinal tight junction gene expression.

Previous studies have shown that MCT alleviates hepatic cytokine production and inflammation through abrogating alcohol-induced serum endotoxin elevation. The present study demonstrated that dietary CB attenuated the alcohol-induced inflammatory response without affecting serum endotoxin level, suggesting a different mechanism from MCT. The liver plays a paramount role in systemic clearance and detoxification of endotoxin (13, 27, 29). Either Kupffer cells or hepatocytes recognize bacterial components such as LPS through Toll-like receptors after lipopolysaccharide binding protein and membrane-bound CD14 binding (29). The expression of CD14 is linked with LPS responsiveness both in experimental models and patients with ALD. The mRNA level of CD14 in the liver was increased in ethanol-fed rats (42). In patients with alcoholic cirrhosis, CD14 is present in elevated amounts in the circulation (33). Moreover, the degree of injury correlates with the level of endotoxemia and with the level of CD14 expression (42). In the present study, we also found increased hepatic CD14 in rats fed ethanol with CO or CB, which correlated with the serum endotoxin levels. However, the ethanol-elevated hepatic endotoxin level was normalized by dietary CB. These findings indicated that dietary CB abrogated the alcohol-induced inflammatory response via detoxifying hepatic endotoxin instead of preventing the elevation of serum endotoxin.

ASS1 is a soluble enzyme responsible for a critical biochemical reaction in the citrulline-urea and nitric oxide cycles, which catalyzes the reversible ATP-dependent ligation of citrulline and aspartate to generate argininosuccinate, and is highly conserved among species (11, 36). ASS1 is expressed in many tissues, with the highest levels in the liver and kidney (11). Hepatic ASS1 acts as a sensitive biomarker of liver responses to bacterial endotoxin. Lipid A is the endotoxic portion of LPS. ASS1 physically binds to the lipid A portion of LPS, inactivates the biological activity and clears circulating LPS (39, 40). ASS1 was reported to be upregulated in rat hepatocytes by chronic ethanol feeding (21). We detected increased mRNA levels of ASS1 in AF/CO and AF/CB rats. However, the protein level of ASS1 was only upregulated in the AF/CB group. Cell culture studies further demonstrated that knockdown of ASS1 in rat H4IIEC3 hepatoma cells impaired intracellular clearance of LPS and upregulated cytokine expression. In addition to ASS1, acyloxyacyl hydrolase (AOAH) also degrades LPS (29). However, we did not observe differences in AOAH mRNA or protein levels among all experimental groups (data not shown). These results indicate that CB may speed up endotoxin detoxification via activating ASS1, thereby abrogating endotoxin signaling in the liver. A recent study demonstrated that partial ASS1 ablation protected acute but not chronic ethanol-induced liver injury in mice (21). This discrepancy might be due to the use of different animal models and the stages of disease.

Metabolome analysis in the present study demonstrated that the most significantly perturbed metabolites in AF/CO rats were fat related, i.e., bile acids, lipids (especially lysoPC), and fatty acid metabolites. A lipidomic study reported significant changes in phospholipid composition in the plasma of ethanol-fed male Fischer 334 rats compared with controls (5). Another study showed that plasma glycerophosphocholines and glycerophosphoserines were decreased after ethanol administration in rodents (8). A metabolomics study of high-fat diet-induced obese mice demonstrated that, among the 16 lysoPCs detected, only 1 was increased and the other 15 were dramatically reduced in the liver (15). Low phosphatidylcholine-to-phosphatidylethanolamine (PC/PE) ratio is believed to lead to increased hepatocyte membrane permeability, which contributes to steatosis and liver failure (23). Restoring PC generation in the liver was reported to attenuate alcoholic steatosis (14). Apart from the role in regulating lipid homeostatis, lysoPCs are known to exert anti-inflammatory actions (12). Research in sepsis has shown that lysoPC inversely correlates with the severity of infection (4) and that lysoPC administration in mouse models of sepsis protects them against lethality (46). The present study showed that substituting CO with CB or MCT prevented ethanol plus CO-induced lysoPC reduction in the liver. This might contribute to the alleviated steatosis in AF/CB and AF/MCT rats through normalizing PC and balancing PC/PE ratio and also attenuated hepatic inflammation due to the anti-inflammation action of lysoPC. Moreover, the dramatically decreased palmitic acid and increased palmitoleic acid in the liver in AF/CB rats indicated the activation of stearoyl-CoA desaturase-1 (SCD1). SCD1 is a rate-limiting enzyme in the biosynthesis of monounsaturated fatty acids (MUFAs). MUFAs are believed to be protective against cardiovascular disease risk factors for decades (9). Consumption of dietary MUFA promotes healthy blood lipid profiles, mediates blood pressure, improves insulin sensitivity and regulates glucose levels (9). Guo et al. (10) reported that palmitoleate supplementation suppresses high-fat diet-induced liver inflammatory response in male C57BL/6J mice. Thus the protective roles of MUFA may also account for the beneficial effects of dietary CB substitution. However, the exact mechanisms of how the altered metabolites by dietary CB or MCT affect the progression of ALD require further investigation.

In conclusion, the present study demonstrates the important role of different types of dietary fat in the gut-liver pathology associated with chronic alcohol consumption. The protective effects of dietary saturated fats in preventing alcohol-induced cytokine production and inflammation in the liver occur via different mechanisms at the gut-liver axis. In contrast to MCT, which abrogates the inflammatory response at the intestine level, the protective actions of dietary CB against endotoxin signaling mainly occurred in the liver, at least partially through activating ASS1.

GRANTS

This research was supported by the National Institutes of Health (R01AA020212 and R01AA018844).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

W.Z., W.J., and Z.Z. conception and design of research; W.Z., Q.L., G.X., Xiuhua Sun, X.T., and Xinguo Sun performed experiments; W.Z. and G.X. analyzed data; W.Z. interpreted results of experiments; W.Z. and G.X. prepared figures; W.Z. drafted manuscript; W.Z., Q.L., Xiuhua Sun, and Z.Z. edited and revised manuscript; W.Z., Q.L., G.X., Xiuhua Sun, X.T., Xinguo Sun, W.J., and Z.Z. approved final version of manuscript.