j1阿托伐他汀和饮食对高脂血鸡非酒精性脂肪肝的影响

lc ra Biomedicine & Pharmacotherapy 64 (2010) 275–281 Department of Animal Medicine and Surgery, University of Murcia, 30100 Murcia, Spain 1. Introduction Sedentary lifestyles and poor dietary choices are contributing to a weight gain epidemic in westernized societies. Recent epide- miological studies suggest an increased risk of cardiovascular disease and type II diabetes in overweight and obese individuals. Unfortunately, incidence of the metabolic syndrome and non- alcoholic fatty liver disease (NAFLD), which can precede the development of cardiovascular disease and type II diabetes, are also increasing [1]. Non-alcoholic steatohepatitis (NASH) is part of the spectrum of NAFLD, which includes different lesion grades, from simple steatosis and steatohepatitis, to the most severe cirrhosis and hepatocellular carcinoma, which develops in the absence of excessive alcohol intake. NAFLD is themost common liver disorder in affluent societies, representing the hepatic metabolic conse- quence of relative overnutrition and reduced physical activity [2,3]. NAFLD is a complex disorder involving environmental factors and genetic predisposition. As a result of this complexity, animal models of the spectrum of NAFLD provide the necessary tools to overcome confounding variables, such as genetic heterogeneity, gender differences, and environmental factors, including diet and lifestyle [4]. Much is still unknown about the pathophysiology of steatohepatitis in humans. Studies in animalmodelsmight provide crucial insights in the pathogenesis and therapeutic options of this disease. Given the difficulty of studying all the factors involved in food intake in human populations, studies in animal models allow manipulation of dietary composition in order to research the role of diet in the pathogenesis of steatohepatitis. Chickens are predisposed to fat deposition in the liver [5]. Furthermore, the chicken has been considered as a suitable model A R T I C L E I N F O Article history: Received 29 May 2009 Accepted 7 June 2009 Available online 22 October 2009 Keywords: Non alcoholic fatty liver disease activity score (NAS) Steatohepatitis Atorvastatin Hyperlipidemic diet Chicken A B S T R A C T Non-alcoholic steatohepatitis (NASH) is part of the spectrumof non-alcoholic fatty liver disease (NAFLD), which includes from simple steatosis and steatohepatitis, to the most severe cirrhosis and carcinoma, which develops in the absence of excessive alcohol intake. NAFLD is the most common liver disorder in affluent societies. There is no proven treatment for NAFLD/NASH. One of the most frequent adverse effects of statins is an increase in hepatic aminotransferases. Studies that evaluate if the benefits of statins overcome the risks in NASH are lacking. The present study was conceived to explore the effect of both atorvastatin and diet on regression of steatohepatitis, using a chicken experimental model induced by a hyperlipidemic diet (HD). Plasma lipid levels, liver enzymes and hepatic histopathology, as well as image analysis were performed to determine changes in liver lipid deposits and inflammatory infiltration. Features of steatosis, cell-ballooning, and inflammation were scored to obtain the NAFLD activity score (NAS). A severe level of steatosis was found in animals fed on HD. Atorvastatin treated groups showed smaller size of lipid deposits and a lower level of inflammation than non-treated groups. Atorvastatin therapy induced a significant reduction of hepatocellular damage, even though in the animals which continuously received a hyperlipidemic diet. The combination of atorvastatin therapy and a standard diet produced the lowest decrease of NAS. Our results show that atorvastatin therapy not only decreased plasmatic levels of cholesterol and triglycerides, but also induced a reduction of liver steatosis, inflammation and hepatocellular damage, without increasing plasmatic amynotransferase levels. � 2009 Elsevier Masson SAS. All rights reserved. * Corresponding author. Tel.: +34 968 367070; fax: +34 968 364147. E-mail address: iayape@um.es (I. Ayala). 0753-3322/$ – see front matter � 2009 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.biopha.2009.06.003 Original article Effect of atorvastatin and diet on non-a score in hyperlipidemic chickens Antonia Martı´n-Castillo a, Maria Teresa Castells b, G Bartolome´ Garcı´a Pe´rez c, Ignacio Ayala d,* aDigestive Service, Virgen del Rosell Hospital, Murcia, Spain bDepartment of Cell Biology and Histology, University of Murcia, 30100 Murcia, Spain c Internal Medicine Service, Virgen de la Arrixaca Hospital, Murcia, Murcia, Spain d oholic fatty liver disease activity cia Ada´nez c, Maria Teresa Sa´nchez Polo c, A. Martı´n-Castillo et al. / Biomedicine & Pharmacotherapy 64 (2010) 275–281276 for studies on the comparative lipid metabolism because it is highly sensitive to dietary modifications [6,7]. Therefore, the chicken model offers technical advantages over mammalian models, and may help in the development of a more rational treatment strategy. With no proven treatment for NAFLD/NASH, the focus of several investigations has been on the treatment of components of the metabolic syndrome (obesity, hypertension, dyslipidemia, and diabetes). Lipid loweringagents canalso lower risks of themetabolic syndrome and NAFLD. It is well-known that statins combat dyslipidemia, a hallmark of the metabolic syndrome, by reducing serum triglycerides (TG) and increasing high-density lipoproteins (HDL) levels. However, one of the most frequent adverse effects of statins is an increase in hepatic aminotransferases and caution is needed when prescribing statins to patients with liver disease [8]. Furthermore, liver injury has been associated with these drugs [9]. Studies that evaluate if the benefits of statins overcome the risks in NASH are lacking. To our knowledge, experimental studies on the potential hepatoprotective effect of atorvastatin and diet in NASH have not been reported. Therefore, the present study in an animal model was conceived to explore the effect of both atorvastatin and diet on regression of steatohepatitis. Plasma lipid levels, liver enzymes and hepatic histopathology, as well as semiquantitative and quantitative assessment by image analysis were performed to determine changes in liver lipid deposits and inflammatory infiltration. Features of steatosis, cell-ballooning, and inflammation were scored to obtain the NAFLD activity score (NAS). 2. Materials and methods 2.1. Animals and treatments One hundred male 3-week-old White Leghorn chickens (Pollos Pujante, Murcia, Spain) were housed under controlled conditions. Each room had air-conditioning and thermostatic control in order to minimize variations in temperature and humidity (approxi- mately 23 8C and 60%, respectively). The chickens were randomly assigned to two kinds of diet (they received a standard growth diet during the first 3 weeks of their life). Water was given ad libitum: � standard diet (SD): a standard growing mash. The weekly amount of this was increased with the age of the animals; � hyperlipidemic diet (HD): a standard growing mash with pure cholesterol (2% of the mixture) and 20% of the mixture of saturated oil (palm oil). After a 3-month induction period, 10 chickens in each group were sacrificed to evaluate the hyperlipidemic effect. Afterwards, the chickens fed on HD were randomly divided into four groups and were kept for another 3-month period with different diets. Thus, the groups of our study were as follows: � group A (n = 16): SD for 6 months (healthy control); � group B (n = 16): HD for 6 months (hyperlipidemic control); � group C (n = 16): HD for 3 months and SD during the next 3 months (spontaneous regression group); � group D (n = 16): HD for three months and SD during the next 3 months, when they received oral atorvastatin at clinical doses (pharmacological regression group); � group E (n = 16): HD for the whole 6 months, and oral atorvastatin at clinical doses during the last 3 months (progres- sion group). Atorvastatin was orally administered at doses of 3 mg/kg/day. Animals were weekly body-weighed in order to calculate the doses. Medications were administered (force-fed) daily at 8 a.m. 2.2. Blood sampling Blood samples (1 ml) were extracted after an overnight fasting period from the axillary vein. In all cases, bloodwas collected into 10 mM trisodium citrate-containing tubes. Plasmawas separated and analyzed for the determination of total cholesterol, low- density lipoprotein (LDL), HDL, TG, aspartate aminotransferase (AST), alanine aminotransferase (ALT), gamma glutamyl-trans- ferase (g-GT), alkaline phosphatase (AP), lactate dehydrogenase (LDH), creatine kinase (CCK), C-reactive protein (CRP) and fibrinogen. Total cholesterol, LDL, HDL, triglycerides, AST, ALT, g-GT, AP, LDH, and CCK were measured using a D-2400 and P800 analyzers (Hitachi Ltd., Tokyo, Japan) and commercially available assays from Roche Diagnostics (Manheim, Germany). The method described by Kostner et al. [10] was used for precipita- tion of HDL. 2.3. Tissue collection All animals were sacrificed by intraperitoneal administration of pentobarbital, after 6 months of receiving both diets and/or treatments. Livers were removed for histological exam- inations. All experimental procedures were approved by the University ofMurcia institutional Animal Care Committee, in accordancewith the guidelines for ethical care of experimental animals of the European Union. Liver samples were fixed in 10% formaldehyde in phosphate- buffered saline (0.1 M PBS, pH 7.4) for 10 h and embedded in paraffin; afterwards, 5 m-thick paraffin sections were cut and stained with haematoxylin and eosin (H&E) and Verhoeff Van Giesson staining techniques. A histological assessment of the tissue was performed for each animal by a pathologist who was blinded to the study. 2.4. Steatosis analysis Lipid deposits were evaluated semiquantitatively in 10 animals (100 fields (�400) in each experimental group). Liver sampleswere classified assigning a score relative to the level of lipid deposits in the sample, according to the histologic classification by Brunt et al. [11] and modified by Angulo [12]: � 0 corresponds to normal, with absence of lipid deposits or a level lower than 5%; � 1 or mild, with lipid deposits lower than 33%; � 2 or moderate, with lipid deposits between 33% and 66%; � 3 or severe, with lipid deposit levels over 66%. Percentages of samples within each semiquantitative score were determined for each experimental group and statistical analysis was performed. A more detailed evaluation of lipid deposits was carried out by quantification of the percentage of steatosis area in liver parenchyma: lobular and centrilobular zones in 10 microscopic fields (square fields of 134 mm2), obtaining 100 determinations for experimental group and zone. Mean and standard error were determined for each group and zone, and a comparative statistical analysis was also carried out. These parameters were quantified by image analysis using the MIP 4.5 (Microm, Image Processing software, Consulting Image Digital, Barcelona). Briefly, the image analysis system consisted of a light microscope (Zeiss Axioskop, Madrid) connected to a video camera 151-AP (Sony, Madrid) and a control computer. After obtaining a digital image, fat depositswere chosen interactively by a graphic line, and percentages of steatosis were measured. 2.5. Inflammatory infiltration analysis Number of inflammatory foci was assessed microscopically (200�) in 10 fields for each animal. Appearance of inflammatory foci was classified as 1, or low density; 2,moderate; 3, high density. Furthermore, area and maximal diameter of inflammatory foci were evaluated in 10microscopic fields (square fields of 267mm2), obtaining 100 determinations for each experimental group, by image analysis using the MIP 4.5 (Microm, Image Processing software, Consulting Image Digital, Barcelona). Inflammatory density was calculated with the following ratio: area of inflam- matory infiltration (obtained by image analysis)/area of the entire field. Measurements were made in five square fields of 267 mm2 for each animal. Welch test, and Bonferroni or Games-Howell post-hoc tests. Statistics were performed using SPSS v14. A p-value <0.05 was considered as statistically significant. 3. Results 3.1. Effects of hyperlipidemia on circulating lipid levels and hepatic function test. Animals fed on the hyperlipidemic diet for 6 months (group B) showed an increase in all lipid parameters in the serum when compared to those of chickens fed the standard diet (group A) (Table 1). The return to the SD for 3 months reverted partly this effect (groups C and D, p < 0.001 in all cases) (Table 1). Moreover, animals fed on HD (group B) had comparatively higher levels of from rans terna A. Martı´n-Castillo et al. / Biomedicine & Pharmacotherapy 64 (2010) 275–281 277 2.6. Hepatocyte ballooning analysis Ballooning classification (0–2) was made following a histolo- gical scoring system [13]: � 0: none; � 1: few balloon cells, i.e. rare but definite ballooned hepatocytes being present, as well as cases that are diagnostically borderline; � 2: many cells or prominent ballooning. Evaluationsweremade in five square fields of 134mm2 for each animal. 2.7. NAFLD activity score Features of steatosis, cell-ballooning, and inflammation were scored as above and single grades were summed up to obtain the NAS, ranging from 0 to 8. A semiquantitative-NAS was obtained by measured of semiquantitative steatosis, number of foci per microscopic field (200�) and frequency of ballooning [13]. A NAS �5 was considered diagnostic of NASH, NAS �2 excluded NASH (simple steatosis), and NAS in between was considered indeterminate [13]. Besides this qualitative score, we also obtained a quantitative NAS, based on results of the same parameters and image analysis (steatosis percentages, the same ballooning classification (0–2), and lobular inflammatory density: 1, or infiltrate <2.5%; 2, with infiltrate between 2.5% and 5.2%; and 3, with infiltrate over 5.2%). 2.8. Statistical analysis Results are expressed as mean � standard error. Mann-Whitney and Kruskal-Wallis non parametric tests were used for assessment of statistical significance in semiquantitative analysis, while statistical significance for quantitative analysis was evaluated by ANOVA or Table 1 Values of the main lipids, enzymatic and hepatic proteins measured in the serum Experimental Groups A B Cholesterol (mg/dl) 104.4�5.5a 980.3�141.3 Triglycerides (mg/dl) 51.7�18.8a 351.8�18.0 HDL (mg/dl) 67.9�6.1a 353.4�32.5 LDL (mg/dl) 26.1�2.6a 656.5�112.6 AST (IU/l) 206.8�36.2 267.3�37.2 ALT (IU/l) 3.4�1.0 17.8�5.1 g-GT (IU/l) 15.4�2.7 9.3�2.4 AP (IU/l) 634.7�161.7 406.3�94.8 LDH (IU/l) 568.6�91.4 1115.0�133.6 CRP (REU/ml) 1.07� 0.29a 2.75� 0.26 HDL: high-density lipoprotein; LDL: low-density lipoprotein; AST: aspartate aminot alkaline phosphatase; LDH: lactate dehydrogenase; CRP: C reactive protein; IU: in a Statistical analysis was performed vs. HD-fed animals (group B). p<0.001. CRP (p < 0.001) than those fed on the SD. No diet and/or treatment significantly decreased these parameters. In our model there was no significant increase in concentrations of the analyzed enzymes (AST, ALT, g-GT, AP or CCK). 3.2. Histology Histological analysis showed that the liver samples of healthy control chickens (group A) presented neither fat accumulation, nor inflammatory infiltration, nor significant hepatocyte ballooning (Figs. 1 et 2). On the contrary, animals fed on HD (group B) developed steatosis with abundant fat deposition. It was pre- dominantly macrovesicular, with isolated single droplets that resulted in nuclear eccentricity because they occupied the entire cell cytoplasm, and involved up to 66% of the lobules, although some hepatocytes showed also microvesicular steatosis. Balloon- ing degeneration of hepatocytes resulting from accumulation of intracellular fluid was characterized by swollen cells, often closely associated with the most distended hepatocytes by the esteatosis. A moderate grade of inflammation (based on observations of the number of foci) was found in lobular and portal zones; it was more evident in intra-acinar location. The return to the SD (group C) partially ameliorated histological findings, reaching minimum criteria for the diagnosis of steatohepatitis. Steatosis was usually lower than 33% of the sample with some degree of lobular and portal mild inflammation. Microgranulomas and lipogranulomas were occasionally found. Cell-ballooning was scarce in this group. Microvesicular steatosis (clusters of hepatocytes with intracyto- plasmatic septations) was found in group D (return to the SD with atorvastatin) but to a minimal extent (<5%). Scarce or none inflammationwas observed in lobular zone, whereas it wasmild in portal location. Cell-ballooning was absent. Administration of atorvastatin to chicks fed on HD (group E) did not improve histological parameters as in animals fed on SD: steatosis (macrovesicular and microvesicular), lobular and portal inflam- animals of all different experimental groups. C D E 204.2�40.8a 197.0�74.3a 413.8�109.6 253.4�90.9a 31.6�7.2a 356.9�145.6 95.4�20.3a 88.5�19.3a 99.4�22.1 85.5�20.3a 77.6�26.4a 242.9�79.3 231.8�47.8 371.5�158.1 372.4�74.2 11.1�4.5 4.1�1.1 18.3�6.8 16.67�4.1 12.3�2.1 20.8�9.8 142.4�50.2 275.2�123.8 385.4�36.9 499.4�109.6 617.9�182.1 1069.0�268.9 2.55�109.6 2.00�0.38 2.01�0.33 ferase; ALT: alanine aminotransferase; g-GT: gamma-glutamyl transpeptidase; AP: tional units; REU: relative ELISA units. A. Martı´n-Castillo et al. / Biomedicine & Pharmacotherapy 64 (2010) 275–281278 mation (higher than in group C), presence of lipogranulomas and microgranulomas, and hepatocyte ballooning were found. No histological signs of cirrhosis were found under any diet/treatment in any group of chicks. 3.3. Steatosis analysis Semiquantitative analysis of steatosis showed a score 0 in the group A (healthy control) (Table 2). Statistically significant differences existed between groups C (spontaneous regression group; score 2) and D (pharmacological regression group; score 1), whereas a severe level of steatosis was found in groups B (hyperlipidemic control) and E (progression group), without significant differences between them. Besides, quantitative analysis of steatosis was carried out. Decreasing percentages of lobular steatosis were found from group B (with the maximumpercentage) to groups E, C, D, and A (healthy Fig. 1. Development of hepatic steatosis in the different experimental groups. (a) Grou vacuolation of a few liver cells in groups D (pharmacological regression) (b, c) and C (spon B (hyperlipidemic control) (f). Note the big extension of the steatosis region in hyperlip staining. Bars: 40 mm. control, absence of steatosis). Atorvastatin treated groups (D and E) showed smaller size of lipid deposits than non-treated groups (C and B, respectively). No significant differences were found between lobular and centrilobular zones for each experimental group. 3.4. Inflammatory infiltration and cell ballooning Inflammatory infiltration in lobular zone was present in all the experimental groups, except in group A (Table 3). Statistically significant differences (p < 0.05) were observed for the number of foci between groups A and B, and B (hyperlipidemic control, with the highest number of foci) and the rest of groups (C, D, and E). Morphometric analysis of the inflammatory foci (i.e. area and maximal diameter) showed a lower level of inflammation in atorvastatin treated groups (D and E) than in non-treated groups (B and C, respectively). Significant differences (p < 0.05) for lobular p A, healthy control; no lipid deposits were observed. Fatty changes ranges from taneous regression) (d) to severe steatosic changes in groups E (progression) (e) and ide