Cigarette smoking exacerbates nonalcoholic fatty liver disease in obese rats

Authors

  • Lorenzo Azzalini,

    1. Centro de Investigación Biomédica En Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Institut D'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), and Liver Unit, Hospital Clínic, Barcelona, Catalonia, Spain
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Elisabet Ferrer,

    1. Centro de Investigación Biomédica En Red de Enfermedades Respiratorias (CIBERes), IDIBAPS and Pneumology Unit, Hospital Clínic, Barcelona, Catalonia, Spain
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Leandra N. Ramalho,

    1. Department of Pathology, Faculty of Medicine of Ribeirão Preto, University of São Paulo, Ribeirão Preto, SP, Brazil
    Search for more papers by this author
  • Montserrat Moreno,

    1. Centro de Investigación Biomédica En Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Institut D'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), and Liver Unit, Hospital Clínic, Barcelona, Catalonia, Spain
    Search for more papers by this author
  • Marlene Domínguez,

    1. Centro de Investigación Biomédica En Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Institut D'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), and Liver Unit, Hospital Clínic, Barcelona, Catalonia, Spain
    Search for more papers by this author
  • Jordi Colmenero,

    1. Centro de Investigación Biomédica En Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Institut D'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), and Liver Unit, Hospital Clínic, Barcelona, Catalonia, Spain
    Search for more papers by this author
  • Víctor I. Peinado,

    1. Centro de Investigación Biomédica En Red de Enfermedades Respiratorias (CIBERes), IDIBAPS and Pneumology Unit, Hospital Clínic, Barcelona, Catalonia, Spain
    Search for more papers by this author
  • Joan A. Barberà,

    1. Centro de Investigación Biomédica En Red de Enfermedades Respiratorias (CIBERes), IDIBAPS and Pneumology Unit, Hospital Clínic, Barcelona, Catalonia, Spain
    Search for more papers by this author
  • Vicente Arroyo,

    1. Centro de Investigación Biomédica En Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Institut D'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), and Liver Unit, Hospital Clínic, Barcelona, Catalonia, Spain
    Search for more papers by this author
  • Pere Ginès,

    1. Centro de Investigación Biomédica En Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Institut D'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), and Liver Unit, Hospital Clínic, Barcelona, Catalonia, Spain
    Search for more papers by this author
  • Joan Caballería,

    1. Centro de Investigación Biomédica En Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Institut D'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), and Liver Unit, Hospital Clínic, Barcelona, Catalonia, Spain
    Search for more papers by this author
  • Ramón Bataller

    Corresponding author
    1. Centro de Investigación Biomédica En Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Institut D'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), and Liver Unit, Hospital Clínic, Barcelona, Catalonia, Spain
    • Liver Unit, Hospital Clínic, Villarroel, 170, 08036 Barcelona, Spain
    Search for more papers by this author
    • fax: +34 93 451 5522


  • Potential conflict of interest: Nothing to report.

Abstract

The prevalence of cigarette smoking (CS) is increased among obese subjects, who are susceptible to develop nonalcoholic fatty liver disease (NAFLD). We investigated the hepatic effects of CS in control and obese rats. Control and obese Zucker rats were divided into smokers and nonsmokers (n = 12 per group). Smoker rats were exposed to 2 cigarettes/day, 5 days/week for 4 weeks. The effects of CS were assessed by biochemical analysis, hepatic histological examination, immunohistochemistry, and gene expression analysis. Phosphorylation of AKT and extracellular signal-regulated kinase (ERK) and quantification of carbonylated proteins were assessed by western blotting. As expected, obese rats showed hypercholesterolemia, insulin resistance, and histological features of NAFLD. Smoking did not modify the lipidic or glucidic serum profiles. Smoking increased alanine aminotransferase serum levels and the degree of liver injury in obese rats, whereas it only induced minor changes in control rats. Importantly, CS increased the histological severity of NAFLD in obese rats. We also explored the potential mechanisms involved in the deleterious effects of CS. Smoking increased the degree of oxidative stress and hepatocellular apoptosis in obese rats, but not in controls. Similarly, smoking increased the hepatic expression of tissue inhibitor of metalloproteinase-1 and procollagen-alpha2(I) in obese rats, but not in controls. Finally, smoking regulated ERK and AKT phosphorylation. The deleterious effects of CS were not observed after a short exposure (5 days). Conclusion: CS causes oxidative stress and worsens the severity of NAFLD in obese rats. Further studies should assess whether this finding also occurs in patients with obesity and NAFLD. (HEPATOLOGY 2010.)

Cigarette smoking (CS) is considered a worldwide major cause of preventable morbidity and mortality.1 The main clinical consequences of prolonged exposure to CS are chronic respiratory diseases, increased incidence of a variety of cancers, and increased risk of atherothrombotic clinical events such as myocardial infarction.2 Hepatologists have traditionally paid scant attention to the deleterious effects of CS. This reflects the fact that smoking per se does not appear to cause liver injury and therefore is not considered a causative agent for chronic liver diseases.3 However, there is increasing evidence that CS may negatively impact the incidence, severity, and clinical course of many types of chronic liver diseases.4, 5

Chronic liver diseases are commonly characterized by continuous inflammation and oxidative stress in the hepatic parenchyma, which are two well-characterized systemic consequences of continuous exposure to CS. It is then plausible that prolonged exposure to cigarette smoke negatively impacts key pathogenic events implicated in chronic liver injury. In fact, epidemiologic studies suggest that CS could accelerate the progression of a variety of liver diseases such as hepatitis C6, 7 and primary biliary cirrhosis8, 9 and could represent a risk factor for hepatocellular carcinoma.10, 11 It is unknown whether CS also influences the severity of nonalcoholic fatty liver disease (NAFLD), the main cause of chronic liver injury in Western countries. Because the prevalence of CS is increased in obese people, who are at a risk of developing NAFLD, it is likely that CS may affect the clinical course of this entity.12

Nonalcoholic steatohepatitis (NASH) is a severe form of NAFLD characterized by inflammatory infiltrate and hepatocellular damage, with or without fibrosis. In a minority of patients this condition progresses to cirrhosis and endstage liver disease.13, 14 Both environmental and genetic factors seem to influence disease severity in patients with NAFLD.15, 16 Among environmental factors, alcohol intake has been shown to exacerbate the clinical course of patients with NASH.17 We hypothesize that heavy exposure to CS could also exacerbate NAFLD. To test this hypothesis at the experimental level we used a well-characterized model of obesity-associated NAFLD in rats. Zucker rats naturally lack leptin receptor, thereby developing obesity due to increased food intake. These rats show features of insulin resistance and NAFLD, mimicking the typical findings in patients with obesity and fatty liver. To test the effects of CS we employed an experimental approach consisting of inhaling CS for 4 weeks. We investigated the effects of CS in both control and obese rats. Besides a detailed histological examination, we studied key events in the pathogenesis of NAFLD including insulin resistance, hepatic inflammation and fibrosis, as well as oxidative stress and apoptosis.

Abbreviations

ALT, alanine aminotransferase; AST, aspartate aminotransferase; CS, cigarette smoking; CYP2E1, cytochrome p450 2E1; DNP, 2,4-dinitrophenylhydrazone; EDTA, ethylenediaminetetraacetic acid; ERK, extracellular signal-regulated kinase; ELISA, enzyme-linked immunosorbent assay; HDL, high-density lipoprotein; HNE, hydroxynonenal; ICAM, intercellular adhesion molecule; HOMA-IR, homeostasis model assessment of insulin resistance; IL, interleukin; NAFLD, nonalcoholic fatty liver disease; NAS, NAFLD activity score; NASH, nonalcoholic steatohepatitis; NF-κB, nuclear factor-kappa B; PCR, polymerase chain reaction; TGF, transforming growth factor; TIMP, tissue inhibitor of metalloproteinases; TNF, tumor necrosis factor; VEGF, vascular endothelial growth factor.

Materials and Methods

Animals.

Obese (fa/fa) and control (fa/+) male Zucker rats were obtained from Charles River Laboratories (Wilmington, MA). Rats were divided into four groups: obese smokers, obese nonsmokers, control smokers, and control nonsmokers. Twelve rats were studied per group. CS was administered as previously described.18 Briefly, smoker rats were exposed with a nose-only system to first-hand smoke of two 2R4F cigarettes (Kentucky University Research, Lexington, KY) daily, 5 days a week, for 4 weeks. To further investigate if CS could also exert acute effects on the liver, we also conducted additional experiments using a short exposure (two 2R4F cigarettes daily for 5 consecutive days). In both cases a smoking device (Protowerx Design, Langley, BC, Canada) was used to puff the smoke of the cigarettes into the inhalation chambers where the rats were held still. Four rats were treated simultaneously. Each puff contained 20 mL of CS and each cigarette was smoked in 16 puffs, with intervals of 25 seconds. Nonsmoker rats underwent the same procedure but the cigarettes were not lighted. Animals were anesthetized with isoflurane (Abbott Laboratories, Madrid, Spain) and then sacrificed on the last day of the experiment, 2 hours after the last exposure to CS. Blood samples were obtained by transdiafragmatic cardiac puncture, collected in tubes with ethylenediaminetetraacetic acid (EDTA), then centrifuged at 3000 rpm for 10 minutes. Livers were collected and stored at −80°C. All animals received human care according to the criteria outlined in the “Guide for the Care and Use of Laboratory Animals”, prepared following the guidelines by the National Academy of Sciences and published by the National Institutes of Health (NIH publication 86-23 revised 1985) and the animal protocol was approved by the Ethics Committee of Animal Experimentation of the Universitat de Barcelona.

Plasma Biochemical Measurements.

Alanine aminotransferase (ALT), aspartate aminotransferase (AST), glucose, total cholesterol, high-density lipoprotein (HDL), and triglycerides serum levels were measured using standard enzymatic procedures by the Laboratory Medicine of the Hospital Clínic. Insulin levels were assessed by enzyme-linked immunosorbent assay (ELISA) (LINCO Research, St. Charles, MO).

RNA Isolation, Reverse Transcription, and Real-Time Quantitative Polymerase Chain Reaction (PCR).

Total RNA was isolated from frozen liver samples using the RNeasy procedure as described by the manufacturer (Qiagen, Hilden, Germany). For complementary DNA (cDNA) synthesis, 0.5 μg of total RNA were retrotranscribed using MultiScribe (Applied Biosystems, Foster City, CA). cDNA templates were amplified by quantitative PCR using the TaqMan technology on an ABI Prism 7900 sequence detection system (Applied Biosystems). The Assay-On-Demand probes and primers for the quantification of 18S rRNA, procollagen α2(I), transforming growth factor beta-1 (TGF-β1), tissue inhibitor of metalloproteinases-1 (TIMP-1), intercellular adhesion molecule-1 (ICAM-1), interleukin-6 (IL-6), tumor necrosis factor alpha (TNF-α), vascular endothelial growth factor A (VEGF-A), cytochrome p450 2E1 (CYP2E1), and FAS ligand were provided by Applied Biosystems. All experiments were performed in duplicate and several negative controls were included. Results were normalized to 18S rRNA expression.

Histological Studies.

Livers were fixed in 10% phosphate-buffered formalin for 24 hours at room temperature and then embedded in paraffin. Liver inflammation was assessed in 5-μm sections, which were stained with hematoxylin and eosin. Samples were blindly evaluated by an expert pathologist (L.N.R.). A necroinflammatory score was calculated including the following parameters: portal inflammation (0, absent; 1, mild; 2, moderate; and 3, severe mononuclear cell infiltration), lobular inflammation (0, absent; 1, mild; 2, moderate; and 3, severe polymorphonuclear cell infiltration), hepatocellular injury (0, absent; 1, rare microvesicular steatosis and occasional ballooning; 2, evident microvesicular steatosis and obvious ballooning; and 3, diffuse microvesicular steatosis, marked ballooning and frequent hepatocellular apoptosis or necrosis), portal necrosis (0, absent; 1, mild; 2, moderate; and 3, severe interface injury), and central necrosis (0, absent; 1, mild; 2, moderate; and 3, severe central necrosis and hemorrhagic areas). Samples were also assessed according to the scoring system for NASH proposed by Kleiner et al.,19 which considers the sum of steatosis (0-3), lobular inflammation (0-3), and hepatocellular ballooning (0-2) scores as the NAFLD Activity Score (NAS). Additional preparations were utilized to evaluate the amount of fibrosis by staining liver sections with 1% picro-Sirius red (Sirius Red F3B, Gurr BDH Chemicals, Poole, UK). Apoptosis was assessed using an in situ apoptosis detection kit (DeadEnd TUNEL, Promega, Madison, WI). The sections were also incubated with monoclonal anti-hydroxynonenal (HNE) antibody (1:500, A.G. Scientific, San Diego, CA); anti-nuclear factor-kappa B (NF-κB)-p65 antibody (1:100; Santa Cruz Biotechnology, Santa Cruz, CA); and anti-von Willebrand factor antibody (1:500; Dako, Carpinteria, CA) for 1 hour at room temperature. As negative controls, all specimens were incubated with an isotype-matched control antibody under identical conditions. The number of NF-κB-p65, von Willebrand factor and TUNEL-positive cells was evaluated in 30 randomly chosen high-power fields and mean values were obtained. For Sirius red and HNE, positive stained area was quantified using a morphometric analysis system using an optic microscope (Nikon Eclipse E600; Nikon, Tokyo, Japan) at 40× magnification. Images were imported using an image analysis software (AnalySIS, Soft Imaging System, Lakewood, CO) and automatically merged. The total positive area was calculated as the sum of the area of all positive pixels.

Western Blot.

The levels of total and phosphorylated AKT and ERK expression, total and cleaved caspase 3, and phosphorylated pSMAD2 were assessed by western blot. The levels of oxydated proteins were assessed through detection of carbonyl groups.20, 21 Selective antibodies were used to detect total and phosphorylated AKT and ERK, total and cleaved caspase 3, and phosphorylated pSMAD2 (1:1000; Cell Signaling Technology, Beverly, MA), and carbonyl groups through derivatization to 2,4-dinitrophenylhydrazone (DNP) (anti-DNP moiety antibody, 1:150, Oxyblot kit, Chemicon International, Temecula, CA). Resulting blots were scanned with an ImageReader LAS-3000 imaging densitometer (Fujifilm, Tokyo, Japan), and optical densities of specific proteins were quantified with the ImageGauge software (Fujifilm). Ponceau red staining of crude homogenates on the membranes was used to determine equal loading/transfer across lanes.

Statistical Analysis.

Statistical analysis was performed using SPPS 16 (Chicago, IL). Results are expressed as the mean ± standard deviation. The Kruskal-Wallis test and unpaired Mann-Whitney U test were performed as appropriate to determine statistical significance. Differences were considered significant if P < 0.05.

Results

Effects of Prolonged CS on Body Weight, Insulin Sensitivity, and Lipid Profile.

Smoking was in general well tolerated by both control and obese rats and mortality was similar in all groups that were exposed to CS (10%). CS did not influence animals' body weight throughout the study. The mean weight of control and obese rats at baseline was 300.4 ± 10.3 g versus 371.8 ± 13.2 g, respectively (P < 0.001). The percentage of weight gain after CS exposure was similar in both control (18% versus 20%, nonsmokers versus smokers, respectively; P = NS) and obese (19% versus 20%, nonsmokers versus smokers, respectively, P = NS) rats. Because CS may affect metabolic parameters, we assessed its effects on insulin sensitivity and lipid profile. As expected, obese rats showed parameters indicative of insulin resistance, as assessed by homeostasis model assessment of insulin resistance (HOMA-IR), as well as hypercholesterolemia and hypertriglyceridemia (Table 1). CS did not modify insulin sensitivity nor lipid profile (including serum total cholesterol, HDL cholesterol, and serum triglycerides levels) in either control and obese rats.

Table 1. Effect of Prolonged Cigarette Smoking on Insulin Sensitivity and Lipid Profile in Control (fa/+) and Obese (fa/fa) Rats
 fa/+ Nonsmokersfa/+ Smokersfa/fa Nonsmokersfa/fa Smokers
  • Twelve rats were included per group. Results are expressed as mean ± standard deviation.

  • *

    P < 0.001 vs. control (fa/+) nonsmoker rats;

  • **

    P < 0.05 vs. control (fa/+) smoker rats. Abbreviations: HOMA-IR: homeostasis model assessment of insulin resistance. HDL: high-density lipoprotein.

HOMA-IR (mmol · mIU · L−2)11.9 ± 6.898.99 ± 5.75123.75 ± 39.67*127.53 ± 120.27**
Total cholesterol (mg/dL)86 ± 684 ± 11143 ± 6*147 ± 7**
HDL-cholesterol (mg/dL)32 ± 235 ± 263 ± 3*65 ± 4**
Triglycerides (mg/dL)66 ± 2082 ± 42316 ± 88*285 ± 32**

Prolonged CS Exacerbates Liver Injury in Obese Rats.

Obese rats exposed to CS showed a significant increase in ALT serum levels, whereas this effect was not observed in control rats (Fig. 1A). We also assessed the effect of CS on liver histology, as detailed in Materials and Methods. CS was associated with an increase in the necroinflammatory score both in control and obese rats (Fig. 1B,C; Table 2). We next studied whether CS influences the severity of NAFLD in obese rats. For this purpose we used the NAFLD activity score proposed by Kleiner et al.,19 which takes into account the degree of steatosis, hepatocellular ballooning, and lobular inflammation. As expected, control rats showed preserved liver histology, whereas obese rats showed signs of NAFLD. CS induced mild changes in control rats. In contrast, CS significantly increased the extent of hepatocellular ballooning and lobular inflammation, thereby increasing the NAFLD activity score in obese rats (Fig. 1D). These results suggest that CS exacerbates NAFLD in obese rats.

Figure 1.

(A) Effects of CS on serum aminotransferases in control and obese rats. Smoking increased ALT serum levels in obese, but not control rats, whereas it increased AST in both groups (not shown). *P < 0.01 versus obese nonsmoker rats. (B) Representative microphotographs of liver specimens from all groups of rats. Hematoxylin and eosin, 100× magnification. (C,D) Quantification of the degree of necroinflammatory injury using the necroinflammatory score and Kleiner et al.'s NAFLD activity score, as described in Materials and Methods. Arrows show inflammatory infiltrate, steatosis, and hepatocellular ballooning. CS induced increased liver damage in obese rats and induced mild changes in control rats. *P < 0.001 versus control nonsmoker rats; **P < 0.05 versus obese nonsmoker rats. NS: nonsmoker. S: smoker. PT: portal tract.

Table 2. Effect of Prolonged Cigarette Smoking on Histologic Necroinflammatory Score in Control (fa/+) and Obese (fa/fa) Rats
 fa/+ Nonsmokersfa/+ Smokersfa/fa Nonsmokersfa/fa Smokers
  • Twelve rats were studied per group. Results are expressed as mean ± standard deviation.

  • **

    P < 0.001 vs. control (fa/+) nonsmoker rats.

  • *

    P < 0.01 vs. control (fa/+) nonsmoker rats.

  • §

    P < 0.01 vs. obese (fa/fa) nonsmoker rats.

  • §§

    P < 0.001 vs. obese (fa/fa) nonsmoker rats.

  • #

    P < 0.05 vs. obese (fa/fa) nonsmoker rats. The criteria used to evaluate the necroinflammatory score are described in Materials and Methods.

Portal inflammation0 ± 00.83 ± 0.39**1.00 ± 01.00 ± 0
Lobular inflammation0 ± 00.58 ± 0.51*0.58 ± 0.511.27 ± 0.47§
Hepatocellular injury0 ± 01.00 ± 0**1.25 ± 0.452.00 ± 0§§
Portal necrosis0 ± 00 ± 00 ± 00.36 ± 0.50#
Lobular necrosis0 ± 00.58 ± 0.51*0.83 ± 0.390.91 ± 0.30

Effects of Prolonged CS on Oxidative Stress.

In order to discover the molecular mechanisms involved in the deleterious effects of CS in the liver, we first investigated the degree of oxidative stress, which is a pathogenic event induced by CS in other organs. For this purpose we used two different approaches: measurement of lipid peroxidation and protein carbonylation. CS was associated with a higher degree of oxidative stress in the liver of obese rats, as indicated by increased content of HNE protein adducts. This effect was not observed in control rats (Fig. 2A,B). The pro-oxidant effect of CS on obese rats was confirmed by detection of carbonylated proteins, a marker of oxidative stress (Fig. 2C). The pro-oxidant effect of CS was not associated with an up-regulation of CYP2E1, a key oxidant enzyme in the liver that metabolizes several components of CS (Table 3).

Figure 2.

(A) Representative microphotographs of liver specimens from all groups of rats; 200× magnification. Fixed liver sections were stained for HNE as described in Materials and Methods. Arrows show areas with positive HNE staining. (B) Quantification of HNE expression in liver specimens. CS markedly increased lipid peroxidation in obese rats, but not in control rats. *P < 0.001 versus obese nonsmoker rats. (C) Expression of carbonylated proteins, as assessed by western blot (see Materials and Methods). CS increased carbonylation of cellular proteins both in control and in obese rats. NS: nonsmoker. S: smoker. CV: central vein.

Table 3. Effect of Prolonged Cigarette Smoking on Hepatic Gene Expression in Control (fa/+) and Obese (fa/fa) Rats
 fa/+ Nonsmokersfa/+ Smokersfa/fa Nonsmokersfa/fa Smokers
  • Twelve rats were studied per group. Results are expressed as 2-ΔΔCt (mean ± standard deviation). 18s rRNA was used as housekeeping gene.

  • *

    P < 0.05 vs. control nonsmoker rats;

  • §

    P < 0.05 vs. obese nonsmoker rats;

  • **

    P < 0.01 vs. control nonsmoker rats;

  • #

    P < 0.01 vs. control smoker rats. IL-6 expression resulted below the sensitivity threshold of the technique in most animals in the four groups. TGF-β1: transforming growth factor β1. TIMP-1: tissue inhibitor of metalloproteinases 1. ICAM-1: intercellular adhesion molecule 1. TNF-α: tumor necrosis factor α. CYP2E1: cytochrome p450 2E1. VEGF: vascular endothelial growth factor. IL-6: interleukin 6.

Procollagen α2(I)1.09 ± 0.451.55 ± 0.35*1.36 ± 0.322.77 ± 1.03§
TGF-β11.01 ± 0.170.62 ± 0.21*0.88 ± 0.140.98 ± 0.50
TIMP-11.02 ± 0.210.99 ± 0.231.32 ± 0.332.14 ± 0.73§
ICAM-11.00 ± 0.290.70 ± 0.171.92 ± 0.56**1.88 ± 0.89#
TNF-α1.06 ± 0.420.57 ± 0.11**0.80 ± 0.190.50 ± 0.26
CYP2E11.04 ± 0.291.01 ± 0.310.50 ± 0.23**0.43 ± 0.20#
Fas ligand1.02 ± 0.200.58 ± 0.13**0.43 ± 0.090.30 ± 0.11
VEGF-A1.11 ± 0.540.99 ± 0.520.87 ± 0.381.18 ± 0.68

Mechanisms of the Deleterious Effects of CS in the Liver.

To investigate the potential mechanisms of the effects of CS in the liver, we next studied whether it induces hepatocellular apoptosis. We found that CS increased hepatocellular apoptosis in obese but not in control rats (TUNEL, Fig. 3A,B). We also studied if CS modifies the pattern of NF-κB expression in the liver. We found increased p65 expression in nonparenchymal cells in the livers of obese and control rats exposed to CS (Fig. 3C,D). Collectively, these data suggest that CS may sensitize hepatocytes to cell death. Next, we assessed whether CS regulates intracellular pathways involved in hepatocellular death and cell defense against apoptosis (i.e., ERK and AKT, respectively). CS reduced the activation of AKT (Fig. 4A) and stimulated the phosphorylation of ERK (Fig. 4B), suggesting that it regulates several intracellular signaling pathways regulating hepatocyte resistance to cell death. In contrast, hepatic expression of Fas ligand, a proapoptotic factor, was not up-regulated by CS (Table 3). We next explored caspase 3 cleavage and SMAD2 phosphorylation, which mediates the proapoptotic effects of TGF-β1.22 Surprisingly, we found that smoking decreases the activation of caspase 3, as assessed by the ratio between cleaved caspase 3 and caspase 3, both in control and obese rats (Fig. 4C). This finding suggests that caspase 3 does not mediate the apoptotic effects of smoking in the liver. To investigate whether smoking affects SMAD2 phosphorylation we performed both western blot and immunohistochemistry studies. The western blot analysis of whole liver extracts showed that smoking does not stimulate SMAD2 phosphorylation (Fig. 4D). However, the detailed analysis of immunohistochemistry studies showed that smoking slightly increased the expression of phospho-SMAD2 in hepatocytes in both control and obese rats, especially in those located in pericentral areas (Supporting Fig. 3). Overall, these results suggest that smoking induces apoptosis through a caspase-independent mechanism.

Figure 3.

(A) Representative microphotographs of liver specimens from all groups of rats; 200× magnification. Fixed liver sections were stained for TUNEL, as described in Materials and Methods. Arrows show cells with positive TUNEL staining. CS increased the amount of TUNEL-positive cells in obese rats, but not in control rats. (B) Quantification of TUNEL-positive cells per high-power field. CS produced an important increase in hepatocellular cell death in obese rats, but not in control rats (*P < 0.001 versus obese nonsmoker rats). (C) Representative microphotographs of liver specimens from all groups of rats; 200× magnification. Fixed liver sections were stained for the p65 subunit of NF-κB as described in Materials and Methods. Arrows show cells with positive p65 staining. CS increased the amount of p65-positive cells in obese rats, but not in control rats. (D) Quantification of p65-NF-κB-positive cells per high-power field. CS greatly increased the expression of p65 in obese rats. *P < 0.05 versus control nonsmoker rats; **P < 0.05 versus obese nonsmoker rats. NS: nonsmoker. S: smoker. CV: central vein, PT: portal tract.

Figure 4.

(A,B) Effect of CS on AKT and ERK phosphorylation as assessed by western blot (see Materials and Methods). Results are expressed as the ratio of phosphorylated protein over total protein expression. CS induced a decrease in pAKT/AKT ratio and an increase in pERK/ERK ratio. *P < 0.05 versus control nonsmoker rats; **P < 0.01 versus obese nonsmoker rats. Results are the mean with standard deviation of five independent experiments. (C,D) Effect of CS on caspase 3 and pSMAD2 as assessed by western blot (see Materials and Methods). Results are expressed as the ratio of cleaved-caspase/caspase and pSMAD2/GADPH, respectively. CS decreased caspase-3-driven apoptosis, whereas it caused little change in pSMAD2-induced apoptosis. NS: nonsmoker. S: smoker.

We next investigated whether CS is associated with angiogenesis in the liver. We did not find neovascularization in any of the studied groups, as indicated by negative staining for von Willebrand factor (data not shown). Similarly, CS did not increase VEGF-A hepatic gene expression (Table 3).

Effect of Prolonged CS on the Expression of Inflammatory and Fibrogenic Genes in the Liver.

We finally evaluated whether CS regulates the hepatic expression of proinflammatory and profibrogenic genes. CS did not increase the expression of genes involved in hepatic inflammation including ICAM-1 and TNF-α (Table 3). In contrast, CS increased the expression of key genes involved in hepatic fibrogenesis in obese rats, including procollagen α2(I) and TIMP-1, whereas TGF-β1 was unaffected. This effect was not associated with the development of liver fibrosis, as assessed by quantification of Sirius red staining (not shown). These results show that CS induces profibrogenic gene expression in the liver, yet does not induce liver fibrosis per se.

Effects of Short Exposure to CS in Control and Obese Rats.

We finally explore whether CS cause acute effects in the liver. For this purpose, control and obese rats were exposed to CS daily for 5 consecutive days. Smoking was in general well tolerated by both control and obese rats and no mortality was observed. Cigarette smoking did not modify ALT serum levels in obese rats, whereas it only induced minor changes in control rats (Supporting Fig. 1B). We next evaluated blindly all liver specimens stained with H&E to evaluate necroinflammatory changes. We found that smoking did not cause significant injury in control or obese rats (Supporting Fig. 1A; Table 4). As expected, Sirius red staining showed no sign of fibrosis in any of the studied groups (Supporting Fig. 2A). Similarly, acute cigarette exposure did not modify the hepatic expression of fibrogenic genes (Supporting Fig. 2B,C). Overall, these results indicate that smoking does not cause acute effects on the liver.

Table 4. Effect of Short-Term Cigarette Smoking on Histologic Necroinflammatory Score in Control (fa/+) and Obese (fa/fa) Rats
 fa/+ Nonsmokersfa/+ Smokersfa/fa Nonsmokersfa/fa Smokers
  1. Twelve rats were studied per group. Results are expressed as mean ± standard deviation. No significant differences were found among smokers and nonsmokers in either control and obese rats. The criteria used to evaluate the necroinflammatory score are described in Materials and Methods.

Portal inflammation1.00 ± 01.00 ± 01.25 ± 0.501.00 ± 0
Lobular inflammation0.50 ± 0.580.67 ± 0.501.50 ± 0.581.33 ± 0.50
Hepatocellular injury1.25 ± 0.501.56 ± 0.532.00 ± 02.00 ± 0
Portal necrosis0 ± 00 ± 00.75 ± 0.500.67 ± 0.50
Lobular necrosis0.50 ± 0.580.78 ± 0.440.75 ± 0.500.78 ± 0.44

Discussion

The current study investigated the hepatic effects of CS in control and obese rats. We provide evidence that smoking induces hepatocellular apoptosis and oxidative stress in obese rats as compared to control animals, thus exacerbating NAFLD. These results are in keeping with previous data showing that CS exerts proinflammatory and oxidant effects in other organs, such as the kidney and the pancreas.23–27 Here, we demonstrate that CS exacerbates liver damage in a genetic model of NAFLD. Our results suggest that CS could exert deleterious effects in the liver, acting as a cofactor favoring the progression of chronic liver diseases. Further studies in experimental and human liver injury models should confirm these findings.

We used a genetic model of obesity-induced NAFLD. We preferred this model over diet-induced obesity because CS can directly affect food intake and interfere with weight gain. To expose rats to CS, we used a well-accepted model for tobacco smoking exposure in rodents,18, 28 which we had used to study the systemic effects of CS exposure in guinea pigs.29 We chose a 4-week period to expose rats to CS because we wanted to investigate a prolonged, rather than an acute, exposure. However, the length of CS exposure is obviously shorter than occurs in humans. This may explain the fact that we did not observe any change in insulin sensitivity nor in body weight in smoker rats, whereas these abnormalities are commonly found in smoker patients.12, 30 Therefore, we cannot rule out that longer exposures would have resulted in more pronounced effects on liver histology (i.e., fibrosis) or changes in the metabolic profile. Because heavy CS is associated with tissue hypoxia, we investigated whether hypoxia may have contributed to liver damage. Our data suggest that hypoxia was not involved, because the hepatic expression of parameters indicative of tissue hypoxia (i.e., VEGF-A gene expression and von Willebrand factor immunohistochemistry) were unaffected by smoking.

The main result of our study is that CS exacerbates NAFLD in obese rats. In particular, smoking induced hepatocellular ballooning and lobular inflammation in obese rats. To our knowledge, this is the first experimental study investigating the hepatic effects of CS in the setting of obesity. Although we observed some effects of CS in control rats, obese rats were more susceptible to the deleterious effects of CS. Of note, smoking was not associated with the development of significant fibrosis. This negative result can be due to two different reasons. First, rodents lacking leptin and/or leptin receptor do not develop significant fibrosis, revealing that leptin is required to develop collagen deposition in the liver. Second, we exposed rats to CS for 4 weeks and cannot rule out that longer exposures would have resulted in hepatic fibrosis. In fact, the finding that CS increases collagen synthesis at the mRNA level strongly suggests that it may exert profibrogenic effects in obese rats. It would be interesting to explore the fibrogenic effects of CS in other animal models of NAFLD that develop liver fibrosis.

Different mechanisms could have contributed to the deleterious effects of smoking in obese rats. First, we found that CS caused oxidative stress in obese rats. This finding could be relevant because oxidative stress is a well-characterized mechanism of injury in chronic liver diseases, including NAFLD.31, 32 In fact, antioxidant supplements have been proposed to treat patients with NAFLD.33, 34 We also found increased hepatocellular apoptosis in obese rats exposed to CS. This mechanism of cell death has been recently implicated in the pathogenesis of NASH in humans.35 Surprisingly, the results of our study do not support a proinflammatory effect of CS in the liver, as indicated by the lack of effect on the expression of inflammatory genes. This negative result could be due to the relatively short exposure to CS and further studies with longer exposure should address this aspect. Finally, our study reveals some of the potential cell signaling pathways involved in the pathogenesis of smoking-induced liver injury. In particular, we showed that CS stimulates ERK phosphorylation and decreases AKT activation in both obese and control rats. ERK is induced after acute liver injury and regulates proliferation and biological actions in many liver cell types.36 In hepatocytes, activation of ERK results in stimulation of DNA synthesis.37 The AKT/PKB pathway is an important antiapoptotic factor in hepatocytes and exerts protective effects against cell death.38 Interestingly, AKT-driven cell signaling is altered by CS in other organs.39–42 Based on our results, the apoptotic effects of CS in the liver seem to be caspase-independent. Of note, the deleterious effects of CS in the liver were not observed after a short exposure (5 days). Because acute smoking exposure can cause deleterious effects in other organs and tissues (e.g. heart, lung, and blood vessels), our results raise the possibility that the liver is particularly resistant to the acute effect of smoking.

Our results may have potential pathophysiological implications. First, we provide evidence that CS causes oxidative stress and apoptosis in the liver. This consequence may also play a role in the progression of other liver diseases, such as chronic hepatitis C and primary biliary cirrhosis. Second, we found that CS up-regulates the hepatic expression of genes involved in fibrogenesis such as procollagen α2(I) and TIMP-1. These effects could also contribute to accelerate liver fibrosis in patients with NAFLD. Further studies using different experimental models (e.g., methionine-choline deficient diet) should be performed. Third, we showed that CS impacts the severity of NAFLD, favoring the development of NASH. Because the development of NASH is a prerequisite for fibrosis progression in patients with NAFLD, it is possible that CS exerts a negative impact in the natural history of patients with NAFLD. Prospective clinical studies should investigate this assumption.

In summary, this study shows that CS exacerbates liver injury in a rat model of obesity-related fatty liver, in particular increasing hepatocellular apoptosis and oxidative stress. Further studies should investigate the hepatic effects of longer exposures to CS, as well as better delineate the cellular and molecular mechanisms involved. Importantly, further studies should investigate whether CS also impacts the natural history of patients with obesity-related NAFLD (i.e., development of NASH, more aggressive fibrosis progression, etc.).

Acknowledgements

We thank Cristina Millán for the preparation of histological slides, and Elena Juez for helping in molecular biology techniques.

Ancillary