- Email: [email protected]
Effect of specific amino acids on hepatic lipid metabolism in fructose-induced non-alcoholic fatty liver disease
Clinical Nutrition xxx (2015) 1e8
Contents lists available at ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Original article
Effect of specific amino acids on hepatic lipid metabolism in fructose-induced non-alcoholic fatty liver disease phanie Beutheu a, Gabrielle Ventura a, Gilles Sarfati b, Prasanthi Jegatheesan a, Ste Esther Nubret a, Nathalie Kapel c, Anne-Judith Waligora-Dupriet c, Ina Bergheim d, Luc Cynober a, b, Jean-Pascal De-Bandt a, b, * Nutrition Biology Laboratory, EA4466, Faculty of Pharmacy, Paris Descartes University, Sorbonne Paris Cit e, Paris, France ^pitaux Universitaires Paris Centre, APHP, Paris, France Clinical Chemistry Department, Ho Microbiology, EA4065, Faculty of Pharmacy, Paris Descartes University, Sorbonne Paris Cit e, Paris, France d Institut of Nutrition, SD Model Systems of Molecular Nutrition, Friedrich-Schiller University Jena, Jena, Germany a
b c
a r t i c l e i n f o
s u m m a r y
Article history: Received 14 October 2014 Accepted 29 January 2015
Background & aim: Fructose diets have been shown to induce insulin resistance and to alter liver metabolism and gut barrier function, ultimately leading to non-alcoholic fatty liver disease. Citrulline, Glutamine and Arginine may improve insulin sensitivity and have beneficial effects on gut trophicity. Our aim was to evaluate their effects on liver and gut functions in a rat model of fructose-induced nonalcoholic fatty liver disease. Methods: Male SpragueeDawley rats (n ¼ 58) received a 4-week fructose (60%) diet or standard chow with or without Citrulline (0.15 g/d) or an isomolar amount of Arginine or Glutamine. All diets were made isonitrogenous by addition of non-essential amino acids. At week 4, nutritional and metabolic status (plasma glucose, insulin, cholesterol, triglycerides and amino acids, net intestinal absorption) was determined; steatosis (hepatic triglycerides content, histological examination) and hepatic function (plasma aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, bilirubin) were assessed; and gut barrier integrity (myeloperoxidase activity, portal endotoxemia, tight junction protein expression and localization) and intestinal and hepatic inflammation were evaluated. We also assessed diets effects on caecal microbiota. Results: In these experimental isonitrogenous fructose diet conditions, fructose led to steatosis with dyslipidemia but without altering glucose homeostasis, liver function or gut permeability. Fructose significantly decreased Bifidobacterium and Lactobacillus and tended to increase endotoxemia. Arginine and Glutamine supplements were ineffective but Citrulline supplementation prevented hypertriglyceridemia and attenuated liver fat accumulation. Conclusion: While nitrogen supply alone can attenuate fructose-induced non-alcoholic fatty liver disease, Citrulline appears to act directly on hepatic lipid metabolism by partially preventing hypertriglyceridemia and steatosis. © 2015 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Keywords: Citrulline Glutamine Arginine Non-alcoholic fatty liver disease
Abbreviations: NAFLD, non-alcoholic fatty liver disease; Cit, citrulline; Gln, glutamine; Arg, arginine; AA, amino acids; NEAA, non-essential amino acids; TG, triglycerides; HTG, hepatic triglycerides; AST, aspartate aminotransferase; ALT, alanine aminotransferase; ALP, alkaline phosphatase; NASH, non-alcoholic steatohepatitis; IR, insulin resistance; TLR4, toll like receptor 4; HDNL, hepatic de novo lipogenesis; VLDL, very-low-density lipoprotein; ZO1, zonula occludens 1; NO, nitric oxide; BW, body weight. de Pharmacie, 4, avenue de l'Observatoire, 75270 Paris Cedex 06, France. Tel.: þ33 1 53 73 * Corresponding author. Laboratoire de Biologie de la Nutrition, EA 4466, Faculte 94 41; fax: þ33 1 53 73 99 52. E-mail addresses: [email protected] (P. Jegatheesan), [email protected] (S. Beutheu), [email protected] (G. Ventura), gilles. [email protected] (G. Sarfati), [email protected] (E. Nubret), [email protected] (N. Kapel), [email protected] (A.-J. WaligoraDupriet), [email protected] (I. Bergheim), [email protected] (L. Cynober), [email protected] (J.-P. De-Bandt). http://dx.doi.org/10.1016/j.clnu.2015.01.021 0261-5614/© 2015 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
2
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
1. Introduction Non-alcoholic fatty liver disease (NAFLD) is a chronic obesityassociated liver disease and a frequent sign of metabolic syndrome [1]. Its pathophysiology is multifactorial and not completely understood. However, experimental and epidemiological studies suggest that fructose intake is associated with NAFLD onset and development [1,2]. Fructose intake has been shown to affect the guteliver axis through changes in liver metabolism and gut barrier function. Indeed, both experimental and clinical studies have shown that increased fructose intake can result in hepatic steatosis together with insulin resistance (IR), elevated plasma triglycerides (TG), and oxidative stress in the liver [2]. In the gut, chronic fructose feeding has been demonstrated to induce increased intestinal permeability, through the loss of occludin expression, and increased endotoxin translocation [3]. This could activate Kupffer cells via Toll-like receptor 4 (TLR-4) on the cell membrane, leading to excessive TNF-a production and hepatic inflammation as demonstrated in mice [4]. Furthermore, evidence from clinical and experimental studies suggest gut microbiota may also play a role in NAFLD pathogenesis [5]. In this context, recent studies underline the possible interest of Arginine (Arg), Glutamine (Gln) and Citrulline (Cit) supplementation as a way to not only preserve intestinal trophicity but also support whole body metabolism in various pathophysiological situations [6]. Arg plays a key role in the regulation of epithelial barrier and the maintenance of junctions between cells [6]. It also modulates immune response and promotes tissue healing [7]. These effects of Arg may help decrease gut bacterial and endotoxin translocation [8]. Plasma Arg levels are decreased in IR settings [7] and Arg supplementation has been demonstrated to improve insulin sensitivity [7]. Gln, a preferential energy substrate for enterocytes and immune cells, helps preserve intestinal mucosa in inflammatory situations [9]. Besides its immunomodulatory properties, it preserves the intestinal barrier via stabilization of tight junction proteins [9,10]. In addition, Gln is a precursor for the synthesis of glutathione required for antioxidative defenses [11]. Cit is involved in peripheral Arg availability and helps regulate protein and energy metabolism [12]. Plasma Cit is mainly the result of its production by enterocytes and its use by the kidney for Arg synthesis [12]. It is closely correlated to functional enterocytic mass [12]. Cit activates muscle protein synthesis [13], improves insulin sensitivity, has antioxidative properties [12], and, in a model of massive (80%) small bowel resection, was shown to improve gut adaptation [14]. Given the effects of these three specific amino acids (AA) on insulin sensitivity and intestinal trophicity, we hypothesized that Arg, Cit or Gln supplementation may be able to limit the fructoseinduced alterations that lead to the development of NAFLD. Thus, the aim of this study was to investigate the effects of Arg, Gln or Cit supplementation on intestinal and hepatic functions in a rat model of fructose-induced NAFLD.
2. Materials and methods
Animal care and experimentation complied with both French and EC regulations governing animal care and experimentation. All procedures were conducted in accordance with the guidelines is Re gional d’Ile-de-France animal care commitsued by the Comite tee, which also approved the study protocol (registration number: CEEA34.CM.015.11). 2.2. Experimental design Rats were randomly allocated to 8 groups (n ¼ 7e8 rats per group) to receive either a standard rodent chow (C) or a fructoseenriched (F, 60%) diet supplemented or not with either Cit 0.15 g/ d (CCit or FCit) or an isomolar amount of Arg (CArg or FArg) or Gln (CGln or FGln) for 4 weeks. All diets were made isonitrogenous to the Arg-containing diet by addition of a mixture of non-essential amino acids (NEAA: alanine, glycine, proline, glutamate, aspartate and serine in isomolar amounts). The dose of Cit was chosen on the basis of our previous studies [15]. Diet compositions are given in Table 1. Animals were euthanatized at the end of the 4-week feeding period for blood and tissue sampling and body composition assessment. 2.3. Nutritional assessment During the 4-week feeding period, food intake and body weight gain were monitored daily. During the 2nd and the 4th week, 24-h urine was collected and nitrogen excretion was measured by pyrochemiluminescence (Antek 9000, Antek, Houston, TX). Net intestinal absorption of macronutrients was assessed on 24-h stools during the same periods. Nitrogen, fat and total energy content were determined by nitrogen elemental analysis (N analyser Flash EA1112, Thermo Scientific, Waltham, MA), by the van de Kamer method [16], and by bomb calorimetry (C200 bomb calorimeter, IKA, Staufen, Germany) respectively. Energy derived from carbohydrate was calculated as the difference between total energy and nitrogen and fat-derived energy. Net intestinal absorption was expressed as the percentage of total energy ingested from the 3 main energy sources (fat, nitrogen, and carbohydrates). Body composition, and in particular the localization and size of fat depots, was determined by dissection and weighing. 2.4. Blood and tissue sampling On day 28, the rats in the fasted state were anesthetized by isoflurane inhalation. After shaving, disinfection and laparotomy, portal blood samples were taken on endotoxin-free material to measure endotoxin levels. Arterial blood was collected from abdominal aorta into heparin-containing tubes and immediately centrifuged (10 min, 2500 g, þ4 C). An aliquot of plasma was immediately deproteinized with a 30% (w/v) sulfosalicylic acid solution for plasma AA analysis. Another aliquot was frozen at 80 C until analysis. Rats were then euthanized by exsanguination, liver samples were taken and either frozen in liquid
Table 1 Compositions of the diets.
2.1. Animals
Ingredients (% total weight)
Control diet (UAR A04, SAFE)
Fructose diet (U8960, SAFE)
Fifty-eight male SpragueDawley rats (Charles River, Villemoisson-sur-Orge, France) weighing 190e220 g were housed individually in a temperature-controlled room under a 12/12-h lightedark cycle for one week. They were given ad libitum access to water and standard rodent chow (UAR A04, SAFE, Augy, France).
Lipids (%) Proteins (%) Carbohydrates (%) Fibers, vitamins, minerals… (%) Energy content (kcal/100 g)
3 16 60 21 389
5 22 65 8 404
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
nitrogen or immediately fixed in buffered 4% PFA for hepatic triglyceride (HTG) determination and histological analysis, respectively. Segments of jejunum, ileum and liver were fixed for 24 h in buffered 4% PFA, dehydrated by soaking in 15% and 30% sucrose solutions, embedded in Tissue-Tek (OCT compound, Fischer Scientific, Sakura, USA) and frozen at 80 C for immunofluorescence and histological analysis. Other jejunum and ileum sections were washed with ice-cold saline solution flushed through the lumen, reverted to collect the mucosa using glass scrapers, frozen in liquid nitrogen and stored at 80 C until analysis. Samples of cecal content were also collected, frozen in liquid nitrogen and stored at 80 C for microbiota analysis. 2.5. Metabolic assessment Plasma AA were separated and quantified by ion exchange chromatography with post-column ninhydrin detection using a JLC-500/V AminoTac™ amino acid analyzer (Jeol Ltd, Tokyo, Japan). Plasma activities of the liver enzymes alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP) and levels of bilirubin, total cholesterol, TG, glucose and uric acid were determined using standard techniques on a multiparameter analyzer (Cobas C 6000; Roche, Meylan, France). Plasma insulin was measured by an ELISA method using a commercial kit (ultrasensitive Rat insulin ELISA, Mercodia, France) following the manufacturer's protocol. Insulin sensitivity was evaluated using the HOmeostasis Model Assessment of Insulin Resistance (HOMA-IR: [fasted insulin (mU/L) fasted glucose (mM)]/22.5). 2.6. Determination of hepatic triglyceride content Frozen liver samples were homogenized in ice-cold 2X phosphate-buffered saline (PBS). Tissue lipids were extracted with methanol/chloroform (1:2), dried and resuspended in 5% fat-free bovine serum albumin in sterile water. HTG levels were assessed rieux, Marcy using a commercially available kit (TG PAP 150, Biome l’Etoile, France). 2.7. Oil red O staining Frozen sections (10 mm) of liver were fixed with 10% formalin in PBS for 30 min and stained with Oil Red O (SigmaeAldrich, StQuentin-Fallavier, France) for 30 min, then washed four times with sterile water. Representative photomicrographs were captured at a 20X magnification using a camera-equipped microscope (DM 4000B, Leica, France). 2.8. TLR4 and TNFa mRNA expression Total RNA from liver samples was extracted using TRIzol reagent (Invitrogen, Saint-Aubin, France), and cDNA was synthesized with the QuantiTect Rev.Transcription kit (Qiagen, Courtaboeuf, France). Real-time PCR was performed using the QuantiTect SYBER Green PCR kit (Qiagen, Courtaboeuf, France) following the manufacturer's protocol. To control for variations in the reactions, all PCR data were normalized to b-actin expression. Primers for TLR4 were forward 50 -ATTCCTGGTGTAGCCATTGCT-30 and reverse 50 -ACCACCACAATAACTTTCCGG-30 ; the primers for TNFa were forward 50 GTCGTAGCAAACCACCAAGC-30 and reverse 50 -GGTATGAAGTGGCAAATCGG-30 . The comparative CT-method was used to determine the amount of target gene normalized to an endogenous reference gene and relative to a calibrator (2DDCt).
3
2.9. Determination of endotoxin levels and intestinal inflammatory status After heating for 5 min at 75 C, portal plasma endotoxin levels were determined using an endpoint Limulus Amebocyte Lysate assay (Charles River; range: 0.015e1.2 EU/mL). Myeloperoxidase (MPO) activity, a marker of PMN infiltration, was determined in jejunal and ileal mucosa as described by Barone et al. [17]. Briefly, intestinal mucosa was homogenized in 50 mmol/ L KH2PO4 buffer at pH 6 containing 0.5% hexadecyltrimethyl ammonium bromide (HTAB). After centrifugation (30 min, 12,500 g, 4 C), supernatants were separated and incubated with o-dianisidine dihydrochloride (0.167 mg/mL; SigmaeAldrich) in the presence of H2O2 (0.0005%) in 50 mmol/L PBS, pH 6. Activities were measured in duplicate at 460 nm at 25 C (MRX, Dynex technologie, Chantilly, France). One unit MPO activity hydrolyzes 1 mmol of H2O2/min. 2.10. TLR4 and intestinal tight junction protein expression Liver samples and jejunal and ileal mucosa were homogenized in lysis buffer (50 mM Hepes, 150 mM NaCl, 10 mM EDTA, 10 mM bglycerophosphate, 100 mM NaF, 2 mM Ortho-vanadate, 1% Triton x100, anti-protease and phosphatase inhibitor cocktail). After centrifugation for 15 min at 12,000 g and 4 C, the supernatants were collected and stored at 80 C. Protein levels were determined using the Bicinchoninic acid assay (Protein Assay kit, Interchim, Montluçon, France). Claudin-1, occludin, ZO-1 and TLR4 expression were determined according to the technic previously described using [18] primary antibodies mouse anti-claudin-1 (1:1000, Zymed Laboratories, South San Francisco, CA), rabbit anti-occludin (1:1000, Zymed Laboratories), rabbit anti-ZO-1 (1:1000, Zymed Laboratories), mouse anti-TLR4 (1:500, Clinisciences, Nanterre, France) and mouse anti-b-actin (1:5000, SigmaeAldrich). After 3 washes with TBS-1% Tween 20 buffer, the membranes were incubated with peroxidase-conjugated goat antirabbit or anti-mouse IgG (Dakocytomation, Les Ulis, France) for 1 h at room temperature and revealed using the enhanced chemiluminescence (ECL) detection system (GE Healthcare, Orsay, France). Protein bands were quantified by densitometry using a CCD camera (ImageQuant LAS 4000, GE Healthcare, Tokyo, Japan) and ImageQuant TL software (GE Healthcare). Protein expression levels were normalized to b-actin. 2.11. Intestinal tight junction protein localization Immunofluorescence analysis was carried out on frozen sections of ileum and jejunum as previously described [18]. After immunohistochemistry, microphotographs were acquired on a confocal laser scanning SP2 microscope (Leica TCS) with an Argon laser (453e514 nm) and a HeNe laser (543 nm). An oil immersion objective lens (63) was used, and images were processed with Leica Confocal software. All images were obtained using identical imaging parameters. 2.12. Cæcal microbiota analyses The main bacterial groups in cecal microbiota and total bacteria were quantified using real-time PCR after extraction of total DNA with Guanidium isothiocyanate using bead beating method [19]. The presence of inhibitors in each sample was estimated using a modified Taqman Exogenous Internal Positive Control kit (Applied Biosystems, Saint-Aubin, France). Real-time qPCR was performed using an ABI 7900HT Sequence Detection System and version 2.3 system software (Applied-
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
4
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
Biosystems). The amplification program consisted of 1 cycle at 95 C for 10 min and 40 amplification cycles (95 C for 15 s, 60 C for 1 min). Total bacteria populations and the dominant bacterial groups, i.e. Clostridium leptum, Bacteroides/Prevotella and Bifidobacterium were quantified using TaqMan® qPCR. The sub-dominant groups Clostridium coccoides, Clostridium cluster I-II, Escherichia coli, enterococci, staphylococci, Lactobacillus/Leuconostoc/Pediococcus group were quantified using SYBR-Green® qPCR. Primers and probes designed according to 16S rRNA sequences are described in Table S1. The numbers of CFU/g of feces were calculated. When a species or bacterial group was not detected, a value of 3 log10 CFU/g of feces, i.e. corresponding to about half of the detection limit, was used for statistical analysis.
was significantly higher vs C and CCit groups. Nitrogen balance was not modified by fructose intake. 3.2. Metabolic assessment After 4 weeks on fructose diet, plasma TG levels were significantly higher in F rats than in C rats (Table 3). Interestingly, plasma TG levels were 43% lower in FCit rats compared to F rats, and triglyceridemia became similar between FCit and C groups. Other AA supplementations appeared less effective. There were no significant differences among groups in terms of fasting glucose and insulin levels, HOMA-IR, total cholesterol (Table 3), uric acid and urea (Table S3). Plasma AST, ALT, and ALP activities and plasma total bilirubindmarkers of hepatic cell lysis and bile functiondwere similar between the 8 groups (Table S3). Most plasma-free AA levels (Table S4), including Gln, Arg and Cit (Table 4), were not modified by fructose diet. The exceptions were glycine and threonine, which decreased and increased, respectively, in the F group compared to the C group (Table 4).
2.13. Data analysis Results are expressed as means ± SEM. Differences between groups were analyzed using a two-tailed ManneWhitney or KruskaleWallis test followed by Dunn's comparisons, as appropriate. For daily food intake and BW, ANOVA for repeated measurement (ANOVArm) was used, and differences were analyzed using a Bonferroni test. p < 0.05 was considered significant. Statistical analysis was performed using Prism version 5.0 (GraphPad® software, San Diego, CA).
3.3. Hepatic steatosis Chronic intake of fructose resulted in a significant increase in absolute liver weight and liver-to-body weight ratios (F vs C). This increase was attenuated after oral Gln or Cit administration (Fig. 1AeB). This increase in liver weight in F rats was associated with a significant increase in HTG (p¼0.022) (Fig. 1C). Liver histological examination showed an increase in number of microvesicular lipid droplets compared to C rats (Fig. 1D). In Gln- and Citsupplemented fructose-fed rats, HTG levels were similar to controls.
3. Results 3.1. Nutritional assessment (Table 2) After 4 weeks, BW was significantly different between the 8 groups (ANOVArm p < 0.05). We observed a significant difference at D17 and D24 between FArg and CGln groups and at D17, D21 and D24 between FGln and CGln groups (p < 0.05). No significant changes in visceral fat mass, skin plus cutaneous fat mass or lean body mass were observed among all groups. Fructose feeding did not modify food intake. Lipid and protein absorptions were significantly higher in all fructose-fed rats (p < 0.001) compared to control group. However, fructose intake had no effect on intestinal carbohydrate absorption (Table S2). Ratio of total energy absorbed to energy intake was similar between fructose-fed and control groups, except in the FArg group where it
3.4. Intestinal permeability, gut and liver inflammatory status Fructose feeding did not alter claudin-1, occludin and ZO-1 expression in the jejunal and ileal mucosa compared to control diet (Table S5) and this was not affected by Gln, Arg or Cit supplementation (Table S5). Moreover, their subcellular localization was preserved and similar in control and fructose-fed rats (Fig. 2). Increased (68%) plasma endotoxin in F group was noted but this did not reach significance (Fig. 3) and Gln, Arg or Cit did not affect this.
Table 2 Nutritional status and dietary intake of fructose-fed and control rats supplemented with Arginine, Glutamine or Citrulline. C Body composition Body weight (g)* 364 Visceral fat (g)* 8.4 Skin þ cutaneous fat (g)* 63.1 Lean mass (g)* 236.7 Nutrient intake and absorption # Food intake (g/day) 26.3 Lipid absorption 0.83 (% energy intake)$ Protein absorption 0.84 (% energy intake)$ Total energy absorption 0.86 (% energy intake)$ * Nitrogen balance 226 (mg N/24H)
CArg ± ± ± ±
5a,b 0.9 1.8 8.9
362 12.2 72.4 247.8
CGln ± ± ± ±
6a,b 1.8 2.6 8.1
361 10.7 72.7 258.4
CCit ± ± ± ±
6a 1.1 2.3 7.0
364 11.7 73.2 245.2
F ± ± ± ±
19a,b 1.7 4.8 8.2
FArg
362 9.9 66.8 241.1
± ± ± ±
13a,b 0.8 2.4 3.4
392 9.2 62.1 232.1
FGln ± ± ± ±
12a,b 1.1 4.6 5.9
401 9.3 62.9 236.9
FCit ± ± ± ±
12b 1.0 1.9 3.0
394 9.9 65.4 237.2
± ± ± ±
13a,b 1.4 3.4 10.1
± 1.0 ± 0.03a
30.1 ± 0.9 0.83 ± 0.02a,b
30.3 ± 0.7 0.84 ± 0.02a,b
30.3 ± 1.1 0.83 ± 0.02a,b
31.8 ± 1.2 0.93 ± 0.01b
28.0 ± 0.7 0.92 ± 0.01b
32.6 ± 0.4 0.92 ± 0.01b
31.3 ± 0.7 0.90 ± 0.02b
± 0.01a
0.84 ± 0.02a,b,c
0.84 ± 0.02a,b,c
0.83 ± 0.02b
0.94 ± 0.01c
0.95 ± 0.01c
0.93 ± 0.01c
0.91 ± 0.02c
± 0.01a
0.86 ± 0.01a,b
0.87 ± 0.01a,b
0.87 ± 0.01a
0.92 ± 0.07a,b
0.94 ± 0.07b
0.92 ± 0.09a,b
0.89 ± 0.04a,b
± 72
318 ± 34
347 ± 27
331 ± 27
390 ± 67
298 ± 42
224 ± 45
201 ± 70
Results are expressed as means ± SEM (n¼6-7 per group). * Body weight, body composition and nitrogen balance on D28. # Mean daily intake throughout the feeding period. $ Mean absorption measured on D14. asbsc: p < 0.05.
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
5
Table 3 Effects of Arg, Gln or Cit supplementation on metabolic parameters in NAFLD and control rats. C TG (mmol/L) Glucose (mmol/L) Cholesterol (mmol/L) Insulin (mg/L) HOMA-IR
CArg
0.3 10.3 1.1 0.56 6.0
± ± ± ± ±
0.0a 0.4 0.1 0.11 1.8
0.4 11.4 1.1 0.76 8.1
CGln 0.1ab 0.6 0.1 0.10 1.6
± ± ± ± ±
0.4 10.8 1.1 0.89 10.8
CCit ± ± ± ± ±
0.0ab 0.4 0.0 0.19 2.7
0.5 11.8 1.1 0.47 15.1
F ± ± ± ± ±
0.1ab 1.1 0.0 0.05 3.2
FArg
0.7 12.0 1.3 0.46 8.9
± ± ± ± ±
0.2b 0.4 0.1 0.12 2.0
0.8 12.6 1.1 0.55 12.7
FGln ± ± ± ± ±
0.3ab 1.0 0.1 0.11 2.2
0.7 11.7 1.3 0.78 13.3
FCit ± ± ± ± ±
0.2ab 0.6 0.1 0.21 2.7
0.4 12.2 1.2 0.96 7.5
± ± ± ± ±
0.1ab 0.6 0.1 0.18 0.8
± ± ± ± ±
31 9 3 25a,b 29a,b
Results are expressed as means ± SEM (n¼6-7 per group). asb: p < 0.05. Table 4 Effects of Arg, Gln or Cit supplementation on plasma amino acids in NAFLD and control rats. C Glutamine Arginine Citrulline Glycine Threonine
765 160 62 442 279
CArg ± ± ± ± ±
29 3 2 23a 20a
721 175 74 424 307
CGln
± ± ± ± ±
53 13 4 29a,b 11a,b
714 159 64 451 307
± ± ± ± ±
CCit 27 8 2 23a,b 8a,b
724 168 64 412 305
F ± ± ± ± ±
23 15 4 29a,b 18a,b
FArg
709 151 68 248 449
± ± ± ± ±
30 7 3 20b 41b
704 157 62 230 391
± ± ± ± ±
FGln 32 9 5 19a,b 23a,b
734 147 66 257 418
FCit
± ± ± ± ±
17 4 2 25a,b 35a,b
679 163 63 230 386
Results (mmol/l) are expressed as means ± SEM (n ¼ 6e7 per group). asb: p < 0.05.
A
Control NEAA 14
B
Fructose Cit
Liver weight (% Body weight)
Liver weight (g)
*
3.5
10 8 6 4 2 0
*
3.0 2.5 2.0 1.5 1.0 0.5 0.0
NEAA Arg
Gln
Cit NEAA Arg
Control
Gln
Cit
NEAA Arg
Fructose
Control Arg
Gln
Cit NEAA Arg
Control
D
Fructose Arg
100 Triglycerides (μmol)/liver)
Fructose Arg
4.0
*
12
C
Control NEAA
Gln
Cit
Fructose
Oil red O Staining CNEAA
FNEAA
* 80
60
40
20
0 NEAA Arg
Gln
Control
Cit NEAA Arg
Gln
Cit
500 μm
500 μm
Fructose
Fig. 1. Effect of chronic fructose consumption and amino acid supplementation on liver weight and hepatic lipid accumulation. (A) Liver weight (g), (B) Liver to body weight ratio (%), (C) Hepatic triglycerides content. (D) Representative microphotographs of liver sections stained by Oil Red O Scale bar ¼ 500 mm. Data are means ± SEM (n ¼ 7e8). *: p < 0.05 vs C.
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
6
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
Jejunal and ileal MPO activities and liver TLR4 protein expression were not affected by fructose diet (Table S5). Expression levels of TLR4 and TNFa genes in liver were not modified by fructose diet and AA supplementation (Table S5). 3.5. Intestinal microbiota Total fecal bacteria decreased in fructose-fed rats compared to controls, but did not reach statistical significance (Fig. 4A, p ¼ 0.056). Interestingly, all fructose-fed rats had lower Bifidobacterium and Lactobacillus levels compared to controls (p < 0.05) (Fig. 4BeC). There were no differences in fecal Staphylococcus, Bacteroides, C. leptum, Clostridium Cluster XI, C. coccoides, E. coli and Enterococcus bacterial groups between fructose-fed and control rats (p > 0.05) (Fig. S1). Clostridium Cluster I/II remained undetectable in all groups. Gln, Arg and Cit supplementation failed to restore total fecal bacteria levels or fecal Bifidobacterium and Lactobacillus groups in fructose-fed rats. Fecal Bifidobacterium levels were even more decreased in FArg compared to F rats (p ¼ 0.038).
CNEAA
A
B
Claudin-1
25 μm
C
D
E
F
Occludin
4. Discussion
ZO 1
Fig. 2. Effect of chronic fructose consumption on tight junction protein distribution. Representative photomicrographs illustrating distribution of Claudin-1 (A, B), Occludin (C, D) and Zonula Occludens-1 (ZO 1) (E, F) in ileal mucosa. Scale bar ¼ 25 mm.
as in Theytaz et al. [23]. The ensuing reduction in hepatic fat accumulation probably explains the absence of hepatic IR and of steatosis-related liver inflammation, as suggested by unaltered TNFa and TLR4 mRNA or protein expression. Fructose induces NAFLD through its effect on liver but also on gut barrier and/or microbiota. Indeed gut microbiota dysbiosis has been associated with the metabolic syndrome [25] in human and experimental NAFLD [5]. Our study by showing a decrease in relative cæcal abundance of Lactobacillus and Bifidobacterium
Control NEAA
Fructose NEAA
0.10 0.09 0.08 Endotoxin (EU/ml)
This study evaluated whether Gln, Arg or Cit supplementation might prevent fructose-induced NAFLD. In our experimental isonitrogenous fructose feeding conditions, fructose induced a mild steatosis with hypertriglyceridemia and limited alterations in gut barrier function. Gln and Arg seemed to have no significant effect, whereas Cit was able to prevent dyslipidemia. Fructose feeding is a classical rat model of NAFLD with consistent similarities between fructose-induced alterations in rodents and the different features of NAFLD observed in humans [20]. Nevertheless, in our experiments fructose feeding surprisingly only led to limited features of fructose-induced dysmetabolism das presented belowd probably as a result of our isonitrogenous experimental conditions (i.e. C group was also supplemented with NEAA). Food intake was not altered by fructose feeding [2]. In parallel, despite a slightly higher calorie content and higher digestive balance of the fructose diet, fructose consumption did not promote additional weight gain [2]. Preserved body composition with fructose diet suggests dissipation of the excess energy absorbed for example via a thermogenic effect of fructose as shown by Jürgens et al. [21]. As expected, fructose caused a marked increase in liver weight and HTG and, on liver histology, with an accumulation of numerous microvesicular lipid droplet presumably through its action on hepatic de novo lipogenesis (HDNL) [22]. This was however associated with normal plasma AST, ALT, ALP, and bilirubin, indicating seemingly preserved liver function. From a metabolic standpoint, fructose promoted hypertriglyceridemia as observed in NAFLD patients [22] however this occurred without change in fasting plasma insulin, glucose or HOMA-IR. Since it has been shown that hepatic lipid accumulation induces hepatic IR and subsequently peripheral IR, our metabolic pattern might be considered as an early manifestation of NAFLD metabolic disturbances. A longer period may be necessary to observe the full panel of metabolic disorders. Alternatively, we cannot rule out a possible protective effect of the isonitrogenous NEAA supplement given in addition to the fructose diet. Indeed, dietary proteins or various AA [23,24] supplementation may modulate hepatic lipid accumulation and prevent hepatic steatosis. Here, the supply of 1 g/d extra nitrogen for 4 weeks, which represents for example less than half the dose given as a high-protein (50%) diet in Stepien et al. study [24], may decrease HTG through decreased HDNL as in Stepien et al. [24] or improved VLDL release
FNEAA
0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00 NEAA Arg
Gln
Control
Cit NEAA Arg
Gln
Cit
Fructose
Fig. 3. Effect of chronic fructose consumption and amino acid supplementation on portal endotoxin level. All values are means ± SEM (n ¼ 7e8).
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
7
Fig. 4. Effect of chronic fructose consumption and amino acid supplementation on intestinal microbiota. Quantification of all bacteria populations and dominant bacterial group represented as a box plot. The box plot shows median (central horizontal line), the 25th centile (lower bow border), and the 75th centile (upper box border). The lower and upper horizontal lines refer to the 10th and the 90th centile, respectively. Data are expressed in log10 CFU/g of cæcal samples.*: p < 0.05 F vs C.
confirms this association. This could possibly affect gut barrier integrity given the possible role of some bacterial populations [5]. However, here microbiota alterations were not associated with gut inflammation as suggested by low MPO levels. Loss of tight junction protein may also be involved in NAFLD [3]. However, our results showed neither structural alterations of the jejunal and ileal mucosa nor significant differences in claudin-1, occludin and ZO-1 expression and localization in F group. Taken together, despite the observed disturbance in cæcal microbiota, our data clearly demonstrate preserved intestinal barrier integrity. Last, gut permeability can be modified without significant alterations in gut mucosa [26], and NAFLD may be associated with endotoxin translocation and activation of TLR-4-dependent signaling in the liver [4]. However, we did not observe increased portal endotoxin and hepatic TLR-4 with fructose at variance with Bergheim et al. [27] who reported 3.6-fold increase in plasma endotoxin in mice receiving 30% fructose in drinking water for 8 weeks. Differences between Bergheim's study and our results may stem from differences in animal species (rats vs mice), fructose route of administration (fructose mixed in the diet vs fructose in drinking water) and duration of treatment (4 vs 8 weeks). Anyway this may contribute to the preservation of liver function and suggest that the contribution of gut dysbiosis to metabolic disturbances may be limited in an early stage NAFLD. It does not preclude the contribution of chronic alterations of microbiota in the progression of the disease. Taken as a whole, these data further suggest that NEAA supplementation of fructose diet could potentially limit NAFLD severity in this model. However this needs to be more fully investigated. Given the early stage of steatosis and the absence of alterations of gut barrier, the supply of specific AA showed only limited effects on NAFLD symptoms. First, while Arg or Gln were without effects on metabolic status and lipid disorders, Cit decreased HTG, and therefore liver weight, and prevented hypertriglyceridemia. The absence of effect of Arg contrasts with the results of Wu et al. [7] where a similar dose of Arg significantly decreased plasma TG in high-fat-induced obesity and genetic (ZDF) obesity. Indeed, Arg supply has been shown to be an effective strategy in metabolic syndrome, as it decreases fat mass and improves dyslipidemia in both experimental and clinical studies, presumably through modulation of NO-mediated pathways [7]. Our data suggest that Arg may be ineffective on early fructose-induced alterations in lipid metabolism; it may be explain by the fructose-induced decrease in peroxisome proliferator-
activated receptor (PPAR)-a expression, a key modulator of hepatic lipid metabolism but also of NO synthesis from Arg [28]. Cit effects on hepatic lipid metabolism have not been studied. Here Cit prevented fructose-induced hypertriglyceridemia. This can be compared with the Cit-induced reduction in plasma TG shown by El-Bassossy et al. [29] in a model of hypertension. Candidate mechanisms include a decrease in hepatic lipid accumulation or an increase in VLDL clearance. Prolonged Cit administration in old rats is associated with significant reductions in subcutaneous (14%) and intra-abdominal (42%) fat mass and a significant increase in lean body mass (þ8%) [13], suggesting increased lipid oxidation. Moreover, Cit activates lipolysis and beta-oxidation, leading to decreased free fatty acid release [30] in retroperitoneal adipose tissue explants from young rats. Second, as gut function was not altered by fructose diet, we were not able to demonstrate any effect of any AA supplementations at intestinal level. Moreover, while it has been suggested that specific AA such as Arg may affect gut microbiota [8], the fructose-induced decrease in Lactobacillus and Bifidobacterium was not prevented by AA. Plasma Gln, Arg and Cit levels were not affected neither by fructose diet, in keeping with the moderate metabolic disturbance, nor by AA supplementation, showing that they are actively metabolized. This also suggests that Cit effect on lipid metabolism might occur mainly in the postprandial phase. Further investigation is warranted. In conclusion, this study shows, in addition to a presumably direct effect of nitrogen supply that blunts fructose-induced hepatic alterations, that hepatic lipid accumulation and the related dyslipidemia were decreased by Cit but not by Gln and Arg. In the absence of noticeable changes in gut microbiota, intestinal barrier function, and insulin sensitivity, Cit might play a specific regulatory role in lipid metabolism. Since Arg had no effect, Cit action is not related to its conversion into Arg and thus to NO.
Author's contributions Study concept and design: JPDB, and IB; acquisition of data, analysis and interpretation of data: PJ, SB, GV, AJW, and NK; drafting of the manuscript: PJ, SB, and JPDB; critical revision of the manuscript for important intellectual content: all authors.
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
8
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
Conflict of interest Pr. CYNOBER L. and Pr. DE BANDT JP. are founders and shareholders of Citrage company which develops Citrulline-based dietary supplements for seniors. Financial support This project was supported by an Institut Benjamin Delessert grant and an ANR [French national research agency] grant (ANR-11BSV1-0015, NAFLD-citrulline). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.clnu.2015.01.021. References [1] Marchesini G, Brizi M, Bianchi G, Tomassetti S, Bugianesi E, Lenzi M, et al. Nonalcoholic fatty liver disease: a feature of the metabolic syndrome. Diabetes 2001;50:1844e50. ^ K-A. Does fructose consumption contribute to non-alcoholic fatty [2] Tappy L, Le liver disease? Clin Res Hepatol Gastroenterol 2012;36:554e60. [3] Spruss A, Bergheim I. Dietary fructose and intestinal barrier: potential risk factor in the pathogenesis of nonalcoholic fatty liver disease. J Nutr Biochem 2009;20:657e62. [4] Spruss A, Kanuri G, Wagnerberger S, Haub S, Bischoff SC, Bergheim I. Toll-like receptor 4 is involved in the development of fructose-induced hepatic steatosis in mice. Hepatology 2009;50:1094e104. [5] Abu-Shanab A, Quigley EMM. The role of the gut microbiota in nonalcoholic fatty liver disease. Nat Rev Gastroenterol Hepatol 2010;7:691e701. [6] Marc Rhoads J, Wu G. Glutamine, arginine, and leucine signaling in the intestine. Amino Acids 2009;37:111e22. [7] Wu G, Bazer FW, Davis TA, Kim SW, Li P, Marc Rhoads J, et al. Arginine metabolism and nutrition in growth, health and disease. Amino Acids 2009;37:153e68. [8] Adawi D, Kasravi FB, Molin G, Jeppsson B. Effect of Lactobacillus supplementation with and without arginine on liver damage and bacterial translocation in an acute liver injury model in the rat. Hepatology 1997;25:642e7. [9] De-Souza DA, Greene LJ. Intestinal permeability and systemic infections in critically ill patients: effect of glutamine. Crit Care Med 2005;33:1125e35. [10] Li N, Lewis P, Samuelson D, Liboni K, Neu J. Glutamine regulates Caco-2 cell tight junction proteins. Am J Physiol Gastrointest Liver Physiol 2004;287: G726e33. €ffier M, Marion R, Ducrotte P, De chelotte P. Modulating effect of glutamine [11] Coe on IL-1beta-induced cytokine production by human gut. Clin Nutr 2003;22: 407e13. [12] Cynober L, Moinard C, De Bandt J-P. The 2009 ESPEN Sir David Cuthbertson. Citrulline: a new major signaling molecule or just another player in the pharmaconutrition game? Clin Nutr 2010;29:545e51.
nier S, Cynober L, Raynaud-Simon A. Long-term effect of [13] Moinard C, Le Ple citrulline supplementation in healthy aged rats: effect on body composition. Clin Nutr Suppl 2009;4:12. [14] Osowska S, Neveux N, Nakib S, Lasserre V, Cynober L, Moinard C. Impairment of arginine metabolism in rats after massive intestinal resection: effect of parenteral nutrition supplemented with citrulline compared with arginine. Clin Sci 2008;115:159e66. [15] Osowska S, Duchemann T, Walrand S, Paillard A, Boirie Y, Cynober L, et al. Citrulline modulates muscle protein metabolism in old malnourished rats. Am J Physiol Endocrinol Metab 2006;291:E582e6. [16] Van De Kamer JH, Ten Bokkel Huinink H, Weyers HA. Rapid method for the determination of fat in feces. J Biol Chem 1949;177:347e55. [17] Barone FC, Hillegass LM, Price WJ, White RF, Lee EV, Feuerstein GZ, et al. Polymorphonuclear leukocyte infiltration into cerebral focal ischemic tissue: myeloperoxidase activity assay and histologic verification. J Neurosci Res 1991;29:336e45. ^ le-Feysot C, [18] Beutheu Youmba S, Belmonte L, Galas L, Boukhettala N, Bo chelotte P, et al. Methotrexate modulates tight junctions through NF-kB, De MEK, and JNK pathways. J Pediatr Gastroenterol Nutr 2012;54:463e70. [19] Kalach N, Kapel N, Waligora-Dupriet A-J, Castelain M-C, Cousin MO, Sauvage C, et al. Intestinal permeability and fecal eosinophil-derived neurotoxin are the best diagnosis tools for digestive non-IgE-mediated cow's milk allergy in toddlers. Clin Chem Lab Med 2013;51:351e61. [20] Basciano H, Federico L, Adeli K. Fructose, insulin resistance, and metabolic dyslipidemia. Nutr Metab (Lond) 2005;2:5. ~ eda TR, Schürmann A, Koebnick C, Dombrowski F, [21] Jürgens H, Haass W, Castan et al. Consuming fructose-sweetened beverages increases body adiposity in mice. Obes Res 2005;13:1146e56. [22] Basaranoglu M, Basaranoglu G, Sabuncu T, Senturk H. Fructose as a key player in the development of fatty liver disease. World J Gastroenterol 2013;19: 1166e72. [23] Theytaz F, Noguchi Y, Egli L, Campos V, Buehler T, Hodson L, et al. Effects of supplementation with essential amino acids on intrahepatic lipid concentrations during fructose overfeeding in humans. Am J Clin Nutr 2012;96: 1008e16. D, Azzout-Marniche D. [24] Stepien M, Gaudichon C, Fromentin G, Even P, Tome Increasing protein at the expense of carbohydrate in the diet down-regulates glucose utilization as glucose sparing effect in rats. PLoS ONE 2011;6:e14664. [25] De Bandt J-P, Waligora-Dupriet A-J, Butel M-J. Intestinal microbiota in inflammation and insulin resistance: relevance to humans. Curr Opin Clin Nutr Metab Care 2011;14:334e40. [26] Cani PD, Amar J, Iglesias MA, Poggi M, Knauf C, Bastelica D, et al. Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes 2007;56: 1761e72. €mer S, Volynets V, Kaserouni S, et al. Anti[27] Bergheim I, Weber S, Vos M, Kra biotics protect against fructose-induced hepatic lipid accumulation in mice: role of endotoxin. J Hepatol 2008;48:983e92. [28] Guelzim N, Mariotti F, Martin PGP, Lasserre F, Pineau T, Hermier D. A role for PPARa in the regulation of arginine metabolism and nitric oxide synthesis. Amino Acids 2011;41:969e79. [29] El-Bassossy HM, El-Fawal R, Fahmy A, Watson ML. Arginase inhibition alleviates hypertension in the metabolic syndrome. Br J Pharmacol 2013;169: 693e703. [30] Joffin N, Jaubert AM, Durant S, Bastin J, De Bandt J-P, Cynober L, et al. Citrulline induces fatty acid release selectively in visceral adipose tissue from old rats. Mol Nutr Food Res 2014;58:1765e75.
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
Contents lists available at ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Original article
Effect of specific amino acids on hepatic lipid metabolism in fructose-induced non-alcoholic fatty liver disease phanie Beutheu a, Gabrielle Ventura a, Gilles Sarfati b, Prasanthi Jegatheesan a, Ste Esther Nubret a, Nathalie Kapel c, Anne-Judith Waligora-Dupriet c, Ina Bergheim d, Luc Cynober a, b, Jean-Pascal De-Bandt a, b, * Nutrition Biology Laboratory, EA4466, Faculty of Pharmacy, Paris Descartes University, Sorbonne Paris Cit e, Paris, France ^pitaux Universitaires Paris Centre, APHP, Paris, France Clinical Chemistry Department, Ho Microbiology, EA4065, Faculty of Pharmacy, Paris Descartes University, Sorbonne Paris Cit e, Paris, France d Institut of Nutrition, SD Model Systems of Molecular Nutrition, Friedrich-Schiller University Jena, Jena, Germany a
b c
a r t i c l e i n f o
s u m m a r y
Article history: Received 14 October 2014 Accepted 29 January 2015
Background & aim: Fructose diets have been shown to induce insulin resistance and to alter liver metabolism and gut barrier function, ultimately leading to non-alcoholic fatty liver disease. Citrulline, Glutamine and Arginine may improve insulin sensitivity and have beneficial effects on gut trophicity. Our aim was to evaluate their effects on liver and gut functions in a rat model of fructose-induced nonalcoholic fatty liver disease. Methods: Male SpragueeDawley rats (n ¼ 58) received a 4-week fructose (60%) diet or standard chow with or without Citrulline (0.15 g/d) or an isomolar amount of Arginine or Glutamine. All diets were made isonitrogenous by addition of non-essential amino acids. At week 4, nutritional and metabolic status (plasma glucose, insulin, cholesterol, triglycerides and amino acids, net intestinal absorption) was determined; steatosis (hepatic triglycerides content, histological examination) and hepatic function (plasma aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, bilirubin) were assessed; and gut barrier integrity (myeloperoxidase activity, portal endotoxemia, tight junction protein expression and localization) and intestinal and hepatic inflammation were evaluated. We also assessed diets effects on caecal microbiota. Results: In these experimental isonitrogenous fructose diet conditions, fructose led to steatosis with dyslipidemia but without altering glucose homeostasis, liver function or gut permeability. Fructose significantly decreased Bifidobacterium and Lactobacillus and tended to increase endotoxemia. Arginine and Glutamine supplements were ineffective but Citrulline supplementation prevented hypertriglyceridemia and attenuated liver fat accumulation. Conclusion: While nitrogen supply alone can attenuate fructose-induced non-alcoholic fatty liver disease, Citrulline appears to act directly on hepatic lipid metabolism by partially preventing hypertriglyceridemia and steatosis. © 2015 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Keywords: Citrulline Glutamine Arginine Non-alcoholic fatty liver disease
Abbreviations: NAFLD, non-alcoholic fatty liver disease; Cit, citrulline; Gln, glutamine; Arg, arginine; AA, amino acids; NEAA, non-essential amino acids; TG, triglycerides; HTG, hepatic triglycerides; AST, aspartate aminotransferase; ALT, alanine aminotransferase; ALP, alkaline phosphatase; NASH, non-alcoholic steatohepatitis; IR, insulin resistance; TLR4, toll like receptor 4; HDNL, hepatic de novo lipogenesis; VLDL, very-low-density lipoprotein; ZO1, zonula occludens 1; NO, nitric oxide; BW, body weight. de Pharmacie, 4, avenue de l'Observatoire, 75270 Paris Cedex 06, France. Tel.: þ33 1 53 73 * Corresponding author. Laboratoire de Biologie de la Nutrition, EA 4466, Faculte 94 41; fax: þ33 1 53 73 99 52. E-mail addresses: [email protected] (P. Jegatheesan), [email protected] (S. Beutheu), [email protected] (G. Ventura), gilles. [email protected] (G. Sarfati), [email protected] (E. Nubret), [email protected] (N. Kapel), [email protected] (A.-J. WaligoraDupriet), [email protected] (I. Bergheim), [email protected] (L. Cynober), [email protected] (J.-P. De-Bandt). http://dx.doi.org/10.1016/j.clnu.2015.01.021 0261-5614/© 2015 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
2
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
1. Introduction Non-alcoholic fatty liver disease (NAFLD) is a chronic obesityassociated liver disease and a frequent sign of metabolic syndrome [1]. Its pathophysiology is multifactorial and not completely understood. However, experimental and epidemiological studies suggest that fructose intake is associated with NAFLD onset and development [1,2]. Fructose intake has been shown to affect the guteliver axis through changes in liver metabolism and gut barrier function. Indeed, both experimental and clinical studies have shown that increased fructose intake can result in hepatic steatosis together with insulin resistance (IR), elevated plasma triglycerides (TG), and oxidative stress in the liver [2]. In the gut, chronic fructose feeding has been demonstrated to induce increased intestinal permeability, through the loss of occludin expression, and increased endotoxin translocation [3]. This could activate Kupffer cells via Toll-like receptor 4 (TLR-4) on the cell membrane, leading to excessive TNF-a production and hepatic inflammation as demonstrated in mice [4]. Furthermore, evidence from clinical and experimental studies suggest gut microbiota may also play a role in NAFLD pathogenesis [5]. In this context, recent studies underline the possible interest of Arginine (Arg), Glutamine (Gln) and Citrulline (Cit) supplementation as a way to not only preserve intestinal trophicity but also support whole body metabolism in various pathophysiological situations [6]. Arg plays a key role in the regulation of epithelial barrier and the maintenance of junctions between cells [6]. It also modulates immune response and promotes tissue healing [7]. These effects of Arg may help decrease gut bacterial and endotoxin translocation [8]. Plasma Arg levels are decreased in IR settings [7] and Arg supplementation has been demonstrated to improve insulin sensitivity [7]. Gln, a preferential energy substrate for enterocytes and immune cells, helps preserve intestinal mucosa in inflammatory situations [9]. Besides its immunomodulatory properties, it preserves the intestinal barrier via stabilization of tight junction proteins [9,10]. In addition, Gln is a precursor for the synthesis of glutathione required for antioxidative defenses [11]. Cit is involved in peripheral Arg availability and helps regulate protein and energy metabolism [12]. Plasma Cit is mainly the result of its production by enterocytes and its use by the kidney for Arg synthesis [12]. It is closely correlated to functional enterocytic mass [12]. Cit activates muscle protein synthesis [13], improves insulin sensitivity, has antioxidative properties [12], and, in a model of massive (80%) small bowel resection, was shown to improve gut adaptation [14]. Given the effects of these three specific amino acids (AA) on insulin sensitivity and intestinal trophicity, we hypothesized that Arg, Cit or Gln supplementation may be able to limit the fructoseinduced alterations that lead to the development of NAFLD. Thus, the aim of this study was to investigate the effects of Arg, Gln or Cit supplementation on intestinal and hepatic functions in a rat model of fructose-induced NAFLD.
2. Materials and methods
Animal care and experimentation complied with both French and EC regulations governing animal care and experimentation. All procedures were conducted in accordance with the guidelines is Re gional d’Ile-de-France animal care commitsued by the Comite tee, which also approved the study protocol (registration number: CEEA34.CM.015.11). 2.2. Experimental design Rats were randomly allocated to 8 groups (n ¼ 7e8 rats per group) to receive either a standard rodent chow (C) or a fructoseenriched (F, 60%) diet supplemented or not with either Cit 0.15 g/ d (CCit or FCit) or an isomolar amount of Arg (CArg or FArg) or Gln (CGln or FGln) for 4 weeks. All diets were made isonitrogenous to the Arg-containing diet by addition of a mixture of non-essential amino acids (NEAA: alanine, glycine, proline, glutamate, aspartate and serine in isomolar amounts). The dose of Cit was chosen on the basis of our previous studies [15]. Diet compositions are given in Table 1. Animals were euthanatized at the end of the 4-week feeding period for blood and tissue sampling and body composition assessment. 2.3. Nutritional assessment During the 4-week feeding period, food intake and body weight gain were monitored daily. During the 2nd and the 4th week, 24-h urine was collected and nitrogen excretion was measured by pyrochemiluminescence (Antek 9000, Antek, Houston, TX). Net intestinal absorption of macronutrients was assessed on 24-h stools during the same periods. Nitrogen, fat and total energy content were determined by nitrogen elemental analysis (N analyser Flash EA1112, Thermo Scientific, Waltham, MA), by the van de Kamer method [16], and by bomb calorimetry (C200 bomb calorimeter, IKA, Staufen, Germany) respectively. Energy derived from carbohydrate was calculated as the difference between total energy and nitrogen and fat-derived energy. Net intestinal absorption was expressed as the percentage of total energy ingested from the 3 main energy sources (fat, nitrogen, and carbohydrates). Body composition, and in particular the localization and size of fat depots, was determined by dissection and weighing. 2.4. Blood and tissue sampling On day 28, the rats in the fasted state were anesthetized by isoflurane inhalation. After shaving, disinfection and laparotomy, portal blood samples were taken on endotoxin-free material to measure endotoxin levels. Arterial blood was collected from abdominal aorta into heparin-containing tubes and immediately centrifuged (10 min, 2500 g, þ4 C). An aliquot of plasma was immediately deproteinized with a 30% (w/v) sulfosalicylic acid solution for plasma AA analysis. Another aliquot was frozen at 80 C until analysis. Rats were then euthanized by exsanguination, liver samples were taken and either frozen in liquid
Table 1 Compositions of the diets.
2.1. Animals
Ingredients (% total weight)
Control diet (UAR A04, SAFE)
Fructose diet (U8960, SAFE)
Fifty-eight male SpragueDawley rats (Charles River, Villemoisson-sur-Orge, France) weighing 190e220 g were housed individually in a temperature-controlled room under a 12/12-h lightedark cycle for one week. They were given ad libitum access to water and standard rodent chow (UAR A04, SAFE, Augy, France).
Lipids (%) Proteins (%) Carbohydrates (%) Fibers, vitamins, minerals… (%) Energy content (kcal/100 g)
3 16 60 21 389
5 22 65 8 404
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
nitrogen or immediately fixed in buffered 4% PFA for hepatic triglyceride (HTG) determination and histological analysis, respectively. Segments of jejunum, ileum and liver were fixed for 24 h in buffered 4% PFA, dehydrated by soaking in 15% and 30% sucrose solutions, embedded in Tissue-Tek (OCT compound, Fischer Scientific, Sakura, USA) and frozen at 80 C for immunofluorescence and histological analysis. Other jejunum and ileum sections were washed with ice-cold saline solution flushed through the lumen, reverted to collect the mucosa using glass scrapers, frozen in liquid nitrogen and stored at 80 C until analysis. Samples of cecal content were also collected, frozen in liquid nitrogen and stored at 80 C for microbiota analysis. 2.5. Metabolic assessment Plasma AA were separated and quantified by ion exchange chromatography with post-column ninhydrin detection using a JLC-500/V AminoTac™ amino acid analyzer (Jeol Ltd, Tokyo, Japan). Plasma activities of the liver enzymes alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP) and levels of bilirubin, total cholesterol, TG, glucose and uric acid were determined using standard techniques on a multiparameter analyzer (Cobas C 6000; Roche, Meylan, France). Plasma insulin was measured by an ELISA method using a commercial kit (ultrasensitive Rat insulin ELISA, Mercodia, France) following the manufacturer's protocol. Insulin sensitivity was evaluated using the HOmeostasis Model Assessment of Insulin Resistance (HOMA-IR: [fasted insulin (mU/L) fasted glucose (mM)]/22.5). 2.6. Determination of hepatic triglyceride content Frozen liver samples were homogenized in ice-cold 2X phosphate-buffered saline (PBS). Tissue lipids were extracted with methanol/chloroform (1:2), dried and resuspended in 5% fat-free bovine serum albumin in sterile water. HTG levels were assessed rieux, Marcy using a commercially available kit (TG PAP 150, Biome l’Etoile, France). 2.7. Oil red O staining Frozen sections (10 mm) of liver were fixed with 10% formalin in PBS for 30 min and stained with Oil Red O (SigmaeAldrich, StQuentin-Fallavier, France) for 30 min, then washed four times with sterile water. Representative photomicrographs were captured at a 20X magnification using a camera-equipped microscope (DM 4000B, Leica, France). 2.8. TLR4 and TNFa mRNA expression Total RNA from liver samples was extracted using TRIzol reagent (Invitrogen, Saint-Aubin, France), and cDNA was synthesized with the QuantiTect Rev.Transcription kit (Qiagen, Courtaboeuf, France). Real-time PCR was performed using the QuantiTect SYBER Green PCR kit (Qiagen, Courtaboeuf, France) following the manufacturer's protocol. To control for variations in the reactions, all PCR data were normalized to b-actin expression. Primers for TLR4 were forward 50 -ATTCCTGGTGTAGCCATTGCT-30 and reverse 50 -ACCACCACAATAACTTTCCGG-30 ; the primers for TNFa were forward 50 GTCGTAGCAAACCACCAAGC-30 and reverse 50 -GGTATGAAGTGGCAAATCGG-30 . The comparative CT-method was used to determine the amount of target gene normalized to an endogenous reference gene and relative to a calibrator (2DDCt).
3
2.9. Determination of endotoxin levels and intestinal inflammatory status After heating for 5 min at 75 C, portal plasma endotoxin levels were determined using an endpoint Limulus Amebocyte Lysate assay (Charles River; range: 0.015e1.2 EU/mL). Myeloperoxidase (MPO) activity, a marker of PMN infiltration, was determined in jejunal and ileal mucosa as described by Barone et al. [17]. Briefly, intestinal mucosa was homogenized in 50 mmol/ L KH2PO4 buffer at pH 6 containing 0.5% hexadecyltrimethyl ammonium bromide (HTAB). After centrifugation (30 min, 12,500 g, 4 C), supernatants were separated and incubated with o-dianisidine dihydrochloride (0.167 mg/mL; SigmaeAldrich) in the presence of H2O2 (0.0005%) in 50 mmol/L PBS, pH 6. Activities were measured in duplicate at 460 nm at 25 C (MRX, Dynex technologie, Chantilly, France). One unit MPO activity hydrolyzes 1 mmol of H2O2/min. 2.10. TLR4 and intestinal tight junction protein expression Liver samples and jejunal and ileal mucosa were homogenized in lysis buffer (50 mM Hepes, 150 mM NaCl, 10 mM EDTA, 10 mM bglycerophosphate, 100 mM NaF, 2 mM Ortho-vanadate, 1% Triton x100, anti-protease and phosphatase inhibitor cocktail). After centrifugation for 15 min at 12,000 g and 4 C, the supernatants were collected and stored at 80 C. Protein levels were determined using the Bicinchoninic acid assay (Protein Assay kit, Interchim, Montluçon, France). Claudin-1, occludin, ZO-1 and TLR4 expression were determined according to the technic previously described using [18] primary antibodies mouse anti-claudin-1 (1:1000, Zymed Laboratories, South San Francisco, CA), rabbit anti-occludin (1:1000, Zymed Laboratories), rabbit anti-ZO-1 (1:1000, Zymed Laboratories), mouse anti-TLR4 (1:500, Clinisciences, Nanterre, France) and mouse anti-b-actin (1:5000, SigmaeAldrich). After 3 washes with TBS-1% Tween 20 buffer, the membranes were incubated with peroxidase-conjugated goat antirabbit or anti-mouse IgG (Dakocytomation, Les Ulis, France) for 1 h at room temperature and revealed using the enhanced chemiluminescence (ECL) detection system (GE Healthcare, Orsay, France). Protein bands were quantified by densitometry using a CCD camera (ImageQuant LAS 4000, GE Healthcare, Tokyo, Japan) and ImageQuant TL software (GE Healthcare). Protein expression levels were normalized to b-actin. 2.11. Intestinal tight junction protein localization Immunofluorescence analysis was carried out on frozen sections of ileum and jejunum as previously described [18]. After immunohistochemistry, microphotographs were acquired on a confocal laser scanning SP2 microscope (Leica TCS) with an Argon laser (453e514 nm) and a HeNe laser (543 nm). An oil immersion objective lens (63) was used, and images were processed with Leica Confocal software. All images were obtained using identical imaging parameters. 2.12. Cæcal microbiota analyses The main bacterial groups in cecal microbiota and total bacteria were quantified using real-time PCR after extraction of total DNA with Guanidium isothiocyanate using bead beating method [19]. The presence of inhibitors in each sample was estimated using a modified Taqman Exogenous Internal Positive Control kit (Applied Biosystems, Saint-Aubin, France). Real-time qPCR was performed using an ABI 7900HT Sequence Detection System and version 2.3 system software (Applied-
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
4
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
Biosystems). The amplification program consisted of 1 cycle at 95 C for 10 min and 40 amplification cycles (95 C for 15 s, 60 C for 1 min). Total bacteria populations and the dominant bacterial groups, i.e. Clostridium leptum, Bacteroides/Prevotella and Bifidobacterium were quantified using TaqMan® qPCR. The sub-dominant groups Clostridium coccoides, Clostridium cluster I-II, Escherichia coli, enterococci, staphylococci, Lactobacillus/Leuconostoc/Pediococcus group were quantified using SYBR-Green® qPCR. Primers and probes designed according to 16S rRNA sequences are described in Table S1. The numbers of CFU/g of feces were calculated. When a species or bacterial group was not detected, a value of 3 log10 CFU/g of feces, i.e. corresponding to about half of the detection limit, was used for statistical analysis.
was significantly higher vs C and CCit groups. Nitrogen balance was not modified by fructose intake. 3.2. Metabolic assessment After 4 weeks on fructose diet, plasma TG levels were significantly higher in F rats than in C rats (Table 3). Interestingly, plasma TG levels were 43% lower in FCit rats compared to F rats, and triglyceridemia became similar between FCit and C groups. Other AA supplementations appeared less effective. There were no significant differences among groups in terms of fasting glucose and insulin levels, HOMA-IR, total cholesterol (Table 3), uric acid and urea (Table S3). Plasma AST, ALT, and ALP activities and plasma total bilirubindmarkers of hepatic cell lysis and bile functiondwere similar between the 8 groups (Table S3). Most plasma-free AA levels (Table S4), including Gln, Arg and Cit (Table 4), were not modified by fructose diet. The exceptions were glycine and threonine, which decreased and increased, respectively, in the F group compared to the C group (Table 4).
2.13. Data analysis Results are expressed as means ± SEM. Differences between groups were analyzed using a two-tailed ManneWhitney or KruskaleWallis test followed by Dunn's comparisons, as appropriate. For daily food intake and BW, ANOVA for repeated measurement (ANOVArm) was used, and differences were analyzed using a Bonferroni test. p < 0.05 was considered significant. Statistical analysis was performed using Prism version 5.0 (GraphPad® software, San Diego, CA).
3.3. Hepatic steatosis Chronic intake of fructose resulted in a significant increase in absolute liver weight and liver-to-body weight ratios (F vs C). This increase was attenuated after oral Gln or Cit administration (Fig. 1AeB). This increase in liver weight in F rats was associated with a significant increase in HTG (p¼0.022) (Fig. 1C). Liver histological examination showed an increase in number of microvesicular lipid droplets compared to C rats (Fig. 1D). In Gln- and Citsupplemented fructose-fed rats, HTG levels were similar to controls.
3. Results 3.1. Nutritional assessment (Table 2) After 4 weeks, BW was significantly different between the 8 groups (ANOVArm p < 0.05). We observed a significant difference at D17 and D24 between FArg and CGln groups and at D17, D21 and D24 between FGln and CGln groups (p < 0.05). No significant changes in visceral fat mass, skin plus cutaneous fat mass or lean body mass were observed among all groups. Fructose feeding did not modify food intake. Lipid and protein absorptions were significantly higher in all fructose-fed rats (p < 0.001) compared to control group. However, fructose intake had no effect on intestinal carbohydrate absorption (Table S2). Ratio of total energy absorbed to energy intake was similar between fructose-fed and control groups, except in the FArg group where it
3.4. Intestinal permeability, gut and liver inflammatory status Fructose feeding did not alter claudin-1, occludin and ZO-1 expression in the jejunal and ileal mucosa compared to control diet (Table S5) and this was not affected by Gln, Arg or Cit supplementation (Table S5). Moreover, their subcellular localization was preserved and similar in control and fructose-fed rats (Fig. 2). Increased (68%) plasma endotoxin in F group was noted but this did not reach significance (Fig. 3) and Gln, Arg or Cit did not affect this.
Table 2 Nutritional status and dietary intake of fructose-fed and control rats supplemented with Arginine, Glutamine or Citrulline. C Body composition Body weight (g)* 364 Visceral fat (g)* 8.4 Skin þ cutaneous fat (g)* 63.1 Lean mass (g)* 236.7 Nutrient intake and absorption # Food intake (g/day) 26.3 Lipid absorption 0.83 (% energy intake)$ Protein absorption 0.84 (% energy intake)$ Total energy absorption 0.86 (% energy intake)$ * Nitrogen balance 226 (mg N/24H)
CArg ± ± ± ±
5a,b 0.9 1.8 8.9
362 12.2 72.4 247.8
CGln ± ± ± ±
6a,b 1.8 2.6 8.1
361 10.7 72.7 258.4
CCit ± ± ± ±
6a 1.1 2.3 7.0
364 11.7 73.2 245.2
F ± ± ± ±
19a,b 1.7 4.8 8.2
FArg
362 9.9 66.8 241.1
± ± ± ±
13a,b 0.8 2.4 3.4
392 9.2 62.1 232.1
FGln ± ± ± ±
12a,b 1.1 4.6 5.9
401 9.3 62.9 236.9
FCit ± ± ± ±
12b 1.0 1.9 3.0
394 9.9 65.4 237.2
± ± ± ±
13a,b 1.4 3.4 10.1
± 1.0 ± 0.03a
30.1 ± 0.9 0.83 ± 0.02a,b
30.3 ± 0.7 0.84 ± 0.02a,b
30.3 ± 1.1 0.83 ± 0.02a,b
31.8 ± 1.2 0.93 ± 0.01b
28.0 ± 0.7 0.92 ± 0.01b
32.6 ± 0.4 0.92 ± 0.01b
31.3 ± 0.7 0.90 ± 0.02b
± 0.01a
0.84 ± 0.02a,b,c
0.84 ± 0.02a,b,c
0.83 ± 0.02b
0.94 ± 0.01c
0.95 ± 0.01c
0.93 ± 0.01c
0.91 ± 0.02c
± 0.01a
0.86 ± 0.01a,b
0.87 ± 0.01a,b
0.87 ± 0.01a
0.92 ± 0.07a,b
0.94 ± 0.07b
0.92 ± 0.09a,b
0.89 ± 0.04a,b
± 72
318 ± 34
347 ± 27
331 ± 27
390 ± 67
298 ± 42
224 ± 45
201 ± 70
Results are expressed as means ± SEM (n¼6-7 per group). * Body weight, body composition and nitrogen balance on D28. # Mean daily intake throughout the feeding period. $ Mean absorption measured on D14. asbsc: p < 0.05.
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
5
Table 3 Effects of Arg, Gln or Cit supplementation on metabolic parameters in NAFLD and control rats. C TG (mmol/L) Glucose (mmol/L) Cholesterol (mmol/L) Insulin (mg/L) HOMA-IR
CArg
0.3 10.3 1.1 0.56 6.0
± ± ± ± ±
0.0a 0.4 0.1 0.11 1.8
0.4 11.4 1.1 0.76 8.1
CGln 0.1ab 0.6 0.1 0.10 1.6
± ± ± ± ±
0.4 10.8 1.1 0.89 10.8
CCit ± ± ± ± ±
0.0ab 0.4 0.0 0.19 2.7
0.5 11.8 1.1 0.47 15.1
F ± ± ± ± ±
0.1ab 1.1 0.0 0.05 3.2
FArg
0.7 12.0 1.3 0.46 8.9
± ± ± ± ±
0.2b 0.4 0.1 0.12 2.0
0.8 12.6 1.1 0.55 12.7
FGln ± ± ± ± ±
0.3ab 1.0 0.1 0.11 2.2
0.7 11.7 1.3 0.78 13.3
FCit ± ± ± ± ±
0.2ab 0.6 0.1 0.21 2.7
0.4 12.2 1.2 0.96 7.5
± ± ± ± ±
0.1ab 0.6 0.1 0.18 0.8
± ± ± ± ±
31 9 3 25a,b 29a,b
Results are expressed as means ± SEM (n¼6-7 per group). asb: p < 0.05. Table 4 Effects of Arg, Gln or Cit supplementation on plasma amino acids in NAFLD and control rats. C Glutamine Arginine Citrulline Glycine Threonine
765 160 62 442 279
CArg ± ± ± ± ±
29 3 2 23a 20a
721 175 74 424 307
CGln
± ± ± ± ±
53 13 4 29a,b 11a,b
714 159 64 451 307
± ± ± ± ±
CCit 27 8 2 23a,b 8a,b
724 168 64 412 305
F ± ± ± ± ±
23 15 4 29a,b 18a,b
FArg
709 151 68 248 449
± ± ± ± ±
30 7 3 20b 41b
704 157 62 230 391
± ± ± ± ±
FGln 32 9 5 19a,b 23a,b
734 147 66 257 418
FCit
± ± ± ± ±
17 4 2 25a,b 35a,b
679 163 63 230 386
Results (mmol/l) are expressed as means ± SEM (n ¼ 6e7 per group). asb: p < 0.05.
A
Control NEAA 14
B
Fructose Cit
Liver weight (% Body weight)
Liver weight (g)
*
3.5
10 8 6 4 2 0
*
3.0 2.5 2.0 1.5 1.0 0.5 0.0
NEAA Arg
Gln
Cit NEAA Arg
Control
Gln
Cit
NEAA Arg
Fructose
Control Arg
Gln
Cit NEAA Arg
Control
D
Fructose Arg
100 Triglycerides (μmol)/liver)
Fructose Arg
4.0
*
12
C
Control NEAA
Gln
Cit
Fructose
Oil red O Staining CNEAA
FNEAA
* 80
60
40
20
0 NEAA Arg
Gln
Control
Cit NEAA Arg
Gln
Cit
500 μm
500 μm
Fructose
Fig. 1. Effect of chronic fructose consumption and amino acid supplementation on liver weight and hepatic lipid accumulation. (A) Liver weight (g), (B) Liver to body weight ratio (%), (C) Hepatic triglycerides content. (D) Representative microphotographs of liver sections stained by Oil Red O Scale bar ¼ 500 mm. Data are means ± SEM (n ¼ 7e8). *: p < 0.05 vs C.
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
6
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
Jejunal and ileal MPO activities and liver TLR4 protein expression were not affected by fructose diet (Table S5). Expression levels of TLR4 and TNFa genes in liver were not modified by fructose diet and AA supplementation (Table S5). 3.5. Intestinal microbiota Total fecal bacteria decreased in fructose-fed rats compared to controls, but did not reach statistical significance (Fig. 4A, p ¼ 0.056). Interestingly, all fructose-fed rats had lower Bifidobacterium and Lactobacillus levels compared to controls (p < 0.05) (Fig. 4BeC). There were no differences in fecal Staphylococcus, Bacteroides, C. leptum, Clostridium Cluster XI, C. coccoides, E. coli and Enterococcus bacterial groups between fructose-fed and control rats (p > 0.05) (Fig. S1). Clostridium Cluster I/II remained undetectable in all groups. Gln, Arg and Cit supplementation failed to restore total fecal bacteria levels or fecal Bifidobacterium and Lactobacillus groups in fructose-fed rats. Fecal Bifidobacterium levels were even more decreased in FArg compared to F rats (p ¼ 0.038).
CNEAA
A
B
Claudin-1
25 μm
C
D
E
F
Occludin
4. Discussion
ZO 1
Fig. 2. Effect of chronic fructose consumption on tight junction protein distribution. Representative photomicrographs illustrating distribution of Claudin-1 (A, B), Occludin (C, D) and Zonula Occludens-1 (ZO 1) (E, F) in ileal mucosa. Scale bar ¼ 25 mm.
as in Theytaz et al. [23]. The ensuing reduction in hepatic fat accumulation probably explains the absence of hepatic IR and of steatosis-related liver inflammation, as suggested by unaltered TNFa and TLR4 mRNA or protein expression. Fructose induces NAFLD through its effect on liver but also on gut barrier and/or microbiota. Indeed gut microbiota dysbiosis has been associated with the metabolic syndrome [25] in human and experimental NAFLD [5]. Our study by showing a decrease in relative cæcal abundance of Lactobacillus and Bifidobacterium
Control NEAA
Fructose NEAA
0.10 0.09 0.08 Endotoxin (EU/ml)
This study evaluated whether Gln, Arg or Cit supplementation might prevent fructose-induced NAFLD. In our experimental isonitrogenous fructose feeding conditions, fructose induced a mild steatosis with hypertriglyceridemia and limited alterations in gut barrier function. Gln and Arg seemed to have no significant effect, whereas Cit was able to prevent dyslipidemia. Fructose feeding is a classical rat model of NAFLD with consistent similarities between fructose-induced alterations in rodents and the different features of NAFLD observed in humans [20]. Nevertheless, in our experiments fructose feeding surprisingly only led to limited features of fructose-induced dysmetabolism das presented belowd probably as a result of our isonitrogenous experimental conditions (i.e. C group was also supplemented with NEAA). Food intake was not altered by fructose feeding [2]. In parallel, despite a slightly higher calorie content and higher digestive balance of the fructose diet, fructose consumption did not promote additional weight gain [2]. Preserved body composition with fructose diet suggests dissipation of the excess energy absorbed for example via a thermogenic effect of fructose as shown by Jürgens et al. [21]. As expected, fructose caused a marked increase in liver weight and HTG and, on liver histology, with an accumulation of numerous microvesicular lipid droplet presumably through its action on hepatic de novo lipogenesis (HDNL) [22]. This was however associated with normal plasma AST, ALT, ALP, and bilirubin, indicating seemingly preserved liver function. From a metabolic standpoint, fructose promoted hypertriglyceridemia as observed in NAFLD patients [22] however this occurred without change in fasting plasma insulin, glucose or HOMA-IR. Since it has been shown that hepatic lipid accumulation induces hepatic IR and subsequently peripheral IR, our metabolic pattern might be considered as an early manifestation of NAFLD metabolic disturbances. A longer period may be necessary to observe the full panel of metabolic disorders. Alternatively, we cannot rule out a possible protective effect of the isonitrogenous NEAA supplement given in addition to the fructose diet. Indeed, dietary proteins or various AA [23,24] supplementation may modulate hepatic lipid accumulation and prevent hepatic steatosis. Here, the supply of 1 g/d extra nitrogen for 4 weeks, which represents for example less than half the dose given as a high-protein (50%) diet in Stepien et al. study [24], may decrease HTG through decreased HDNL as in Stepien et al. [24] or improved VLDL release
FNEAA
0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00 NEAA Arg
Gln
Control
Cit NEAA Arg
Gln
Cit
Fructose
Fig. 3. Effect of chronic fructose consumption and amino acid supplementation on portal endotoxin level. All values are means ± SEM (n ¼ 7e8).
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
7
Fig. 4. Effect of chronic fructose consumption and amino acid supplementation on intestinal microbiota. Quantification of all bacteria populations and dominant bacterial group represented as a box plot. The box plot shows median (central horizontal line), the 25th centile (lower bow border), and the 75th centile (upper box border). The lower and upper horizontal lines refer to the 10th and the 90th centile, respectively. Data are expressed in log10 CFU/g of cæcal samples.*: p < 0.05 F vs C.
confirms this association. This could possibly affect gut barrier integrity given the possible role of some bacterial populations [5]. However, here microbiota alterations were not associated with gut inflammation as suggested by low MPO levels. Loss of tight junction protein may also be involved in NAFLD [3]. However, our results showed neither structural alterations of the jejunal and ileal mucosa nor significant differences in claudin-1, occludin and ZO-1 expression and localization in F group. Taken together, despite the observed disturbance in cæcal microbiota, our data clearly demonstrate preserved intestinal barrier integrity. Last, gut permeability can be modified without significant alterations in gut mucosa [26], and NAFLD may be associated with endotoxin translocation and activation of TLR-4-dependent signaling in the liver [4]. However, we did not observe increased portal endotoxin and hepatic TLR-4 with fructose at variance with Bergheim et al. [27] who reported 3.6-fold increase in plasma endotoxin in mice receiving 30% fructose in drinking water for 8 weeks. Differences between Bergheim's study and our results may stem from differences in animal species (rats vs mice), fructose route of administration (fructose mixed in the diet vs fructose in drinking water) and duration of treatment (4 vs 8 weeks). Anyway this may contribute to the preservation of liver function and suggest that the contribution of gut dysbiosis to metabolic disturbances may be limited in an early stage NAFLD. It does not preclude the contribution of chronic alterations of microbiota in the progression of the disease. Taken as a whole, these data further suggest that NEAA supplementation of fructose diet could potentially limit NAFLD severity in this model. However this needs to be more fully investigated. Given the early stage of steatosis and the absence of alterations of gut barrier, the supply of specific AA showed only limited effects on NAFLD symptoms. First, while Arg or Gln were without effects on metabolic status and lipid disorders, Cit decreased HTG, and therefore liver weight, and prevented hypertriglyceridemia. The absence of effect of Arg contrasts with the results of Wu et al. [7] where a similar dose of Arg significantly decreased plasma TG in high-fat-induced obesity and genetic (ZDF) obesity. Indeed, Arg supply has been shown to be an effective strategy in metabolic syndrome, as it decreases fat mass and improves dyslipidemia in both experimental and clinical studies, presumably through modulation of NO-mediated pathways [7]. Our data suggest that Arg may be ineffective on early fructose-induced alterations in lipid metabolism; it may be explain by the fructose-induced decrease in peroxisome proliferator-
activated receptor (PPAR)-a expression, a key modulator of hepatic lipid metabolism but also of NO synthesis from Arg [28]. Cit effects on hepatic lipid metabolism have not been studied. Here Cit prevented fructose-induced hypertriglyceridemia. This can be compared with the Cit-induced reduction in plasma TG shown by El-Bassossy et al. [29] in a model of hypertension. Candidate mechanisms include a decrease in hepatic lipid accumulation or an increase in VLDL clearance. Prolonged Cit administration in old rats is associated with significant reductions in subcutaneous (14%) and intra-abdominal (42%) fat mass and a significant increase in lean body mass (þ8%) [13], suggesting increased lipid oxidation. Moreover, Cit activates lipolysis and beta-oxidation, leading to decreased free fatty acid release [30] in retroperitoneal adipose tissue explants from young rats. Second, as gut function was not altered by fructose diet, we were not able to demonstrate any effect of any AA supplementations at intestinal level. Moreover, while it has been suggested that specific AA such as Arg may affect gut microbiota [8], the fructose-induced decrease in Lactobacillus and Bifidobacterium was not prevented by AA. Plasma Gln, Arg and Cit levels were not affected neither by fructose diet, in keeping with the moderate metabolic disturbance, nor by AA supplementation, showing that they are actively metabolized. This also suggests that Cit effect on lipid metabolism might occur mainly in the postprandial phase. Further investigation is warranted. In conclusion, this study shows, in addition to a presumably direct effect of nitrogen supply that blunts fructose-induced hepatic alterations, that hepatic lipid accumulation and the related dyslipidemia were decreased by Cit but not by Gln and Arg. In the absence of noticeable changes in gut microbiota, intestinal barrier function, and insulin sensitivity, Cit might play a specific regulatory role in lipid metabolism. Since Arg had no effect, Cit action is not related to its conversion into Arg and thus to NO.
Author's contributions Study concept and design: JPDB, and IB; acquisition of data, analysis and interpretation of data: PJ, SB, GV, AJW, and NK; drafting of the manuscript: PJ, SB, and JPDB; critical revision of the manuscript for important intellectual content: all authors.
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021
8
P. Jegatheesan et al. / Clinical Nutrition xxx (2015) 1e8
Conflict of interest Pr. CYNOBER L. and Pr. DE BANDT JP. are founders and shareholders of Citrage company which develops Citrulline-based dietary supplements for seniors. Financial support This project was supported by an Institut Benjamin Delessert grant and an ANR [French national research agency] grant (ANR-11BSV1-0015, NAFLD-citrulline). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.clnu.2015.01.021. References [1] Marchesini G, Brizi M, Bianchi G, Tomassetti S, Bugianesi E, Lenzi M, et al. Nonalcoholic fatty liver disease: a feature of the metabolic syndrome. Diabetes 2001;50:1844e50. ^ K-A. Does fructose consumption contribute to non-alcoholic fatty [2] Tappy L, Le liver disease? Clin Res Hepatol Gastroenterol 2012;36:554e60. [3] Spruss A, Bergheim I. Dietary fructose and intestinal barrier: potential risk factor in the pathogenesis of nonalcoholic fatty liver disease. J Nutr Biochem 2009;20:657e62. [4] Spruss A, Kanuri G, Wagnerberger S, Haub S, Bischoff SC, Bergheim I. Toll-like receptor 4 is involved in the development of fructose-induced hepatic steatosis in mice. Hepatology 2009;50:1094e104. [5] Abu-Shanab A, Quigley EMM. The role of the gut microbiota in nonalcoholic fatty liver disease. Nat Rev Gastroenterol Hepatol 2010;7:691e701. [6] Marc Rhoads J, Wu G. Glutamine, arginine, and leucine signaling in the intestine. Amino Acids 2009;37:111e22. [7] Wu G, Bazer FW, Davis TA, Kim SW, Li P, Marc Rhoads J, et al. Arginine metabolism and nutrition in growth, health and disease. Amino Acids 2009;37:153e68. [8] Adawi D, Kasravi FB, Molin G, Jeppsson B. Effect of Lactobacillus supplementation with and without arginine on liver damage and bacterial translocation in an acute liver injury model in the rat. Hepatology 1997;25:642e7. [9] De-Souza DA, Greene LJ. Intestinal permeability and systemic infections in critically ill patients: effect of glutamine. Crit Care Med 2005;33:1125e35. [10] Li N, Lewis P, Samuelson D, Liboni K, Neu J. Glutamine regulates Caco-2 cell tight junction proteins. Am J Physiol Gastrointest Liver Physiol 2004;287: G726e33. €ffier M, Marion R, Ducrotte P, De chelotte P. Modulating effect of glutamine [11] Coe on IL-1beta-induced cytokine production by human gut. Clin Nutr 2003;22: 407e13. [12] Cynober L, Moinard C, De Bandt J-P. The 2009 ESPEN Sir David Cuthbertson. Citrulline: a new major signaling molecule or just another player in the pharmaconutrition game? Clin Nutr 2010;29:545e51.
nier S, Cynober L, Raynaud-Simon A. Long-term effect of [13] Moinard C, Le Ple citrulline supplementation in healthy aged rats: effect on body composition. Clin Nutr Suppl 2009;4:12. [14] Osowska S, Neveux N, Nakib S, Lasserre V, Cynober L, Moinard C. Impairment of arginine metabolism in rats after massive intestinal resection: effect of parenteral nutrition supplemented with citrulline compared with arginine. Clin Sci 2008;115:159e66. [15] Osowska S, Duchemann T, Walrand S, Paillard A, Boirie Y, Cynober L, et al. Citrulline modulates muscle protein metabolism in old malnourished rats. Am J Physiol Endocrinol Metab 2006;291:E582e6. [16] Van De Kamer JH, Ten Bokkel Huinink H, Weyers HA. Rapid method for the determination of fat in feces. J Biol Chem 1949;177:347e55. [17] Barone FC, Hillegass LM, Price WJ, White RF, Lee EV, Feuerstein GZ, et al. Polymorphonuclear leukocyte infiltration into cerebral focal ischemic tissue: myeloperoxidase activity assay and histologic verification. J Neurosci Res 1991;29:336e45. ^ le-Feysot C, [18] Beutheu Youmba S, Belmonte L, Galas L, Boukhettala N, Bo chelotte P, et al. Methotrexate modulates tight junctions through NF-kB, De MEK, and JNK pathways. J Pediatr Gastroenterol Nutr 2012;54:463e70. [19] Kalach N, Kapel N, Waligora-Dupriet A-J, Castelain M-C, Cousin MO, Sauvage C, et al. Intestinal permeability and fecal eosinophil-derived neurotoxin are the best diagnosis tools for digestive non-IgE-mediated cow's milk allergy in toddlers. Clin Chem Lab Med 2013;51:351e61. [20] Basciano H, Federico L, Adeli K. Fructose, insulin resistance, and metabolic dyslipidemia. Nutr Metab (Lond) 2005;2:5. ~ eda TR, Schürmann A, Koebnick C, Dombrowski F, [21] Jürgens H, Haass W, Castan et al. Consuming fructose-sweetened beverages increases body adiposity in mice. Obes Res 2005;13:1146e56. [22] Basaranoglu M, Basaranoglu G, Sabuncu T, Senturk H. Fructose as a key player in the development of fatty liver disease. World J Gastroenterol 2013;19: 1166e72. [23] Theytaz F, Noguchi Y, Egli L, Campos V, Buehler T, Hodson L, et al. Effects of supplementation with essential amino acids on intrahepatic lipid concentrations during fructose overfeeding in humans. Am J Clin Nutr 2012;96: 1008e16. D, Azzout-Marniche D. [24] Stepien M, Gaudichon C, Fromentin G, Even P, Tome Increasing protein at the expense of carbohydrate in the diet down-regulates glucose utilization as glucose sparing effect in rats. PLoS ONE 2011;6:e14664. [25] De Bandt J-P, Waligora-Dupriet A-J, Butel M-J. Intestinal microbiota in inflammation and insulin resistance: relevance to humans. Curr Opin Clin Nutr Metab Care 2011;14:334e40. [26] Cani PD, Amar J, Iglesias MA, Poggi M, Knauf C, Bastelica D, et al. Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes 2007;56: 1761e72. €mer S, Volynets V, Kaserouni S, et al. Anti[27] Bergheim I, Weber S, Vos M, Kra biotics protect against fructose-induced hepatic lipid accumulation in mice: role of endotoxin. J Hepatol 2008;48:983e92. [28] Guelzim N, Mariotti F, Martin PGP, Lasserre F, Pineau T, Hermier D. A role for PPARa in the regulation of arginine metabolism and nitric oxide synthesis. Amino Acids 2011;41:969e79. [29] El-Bassossy HM, El-Fawal R, Fahmy A, Watson ML. Arginase inhibition alleviates hypertension in the metabolic syndrome. Br J Pharmacol 2013;169: 693e703. [30] Joffin N, Jaubert AM, Durant S, Bastin J, De Bandt J-P, Cynober L, et al. Citrulline induces fatty acid release selectively in visceral adipose tissue from old rats. Mol Nutr Food Res 2014;58:1765e75.
Please cite this article in press as: Jegatheesan P, et al., Effect of specific amino acids on hepatic lipid metabolism in fructose-induced nonalcoholic fatty liver disease, Clinical Nutrition (2015), http://dx.doi.org/10.1016/j.clnu.2015.01.021