Therapeutic effect of rosuvastatin and propylthiouracil on ameliorating high-cholesterol diet-induced fatty liver disease, fibrosis and inflammation in rabbit

Abstract

This study tested the hypothesis that high-cholesterol diet (HCD)-induced fatty liver disease could be ameliorated by rosuvastatin (Ros) and propylthiouracil (PTU) therapy. Thirty-two Zealand rabbits were equally divided into group 1 (sham-control), group 2 (HCD for 8 weeks), group 3 [HCD-Ros (20 mg/kg/day administration after 4-week HFD for 4 weeks)], group 4 [HCD-PTU (0.1% PTU in drinking water) with treatment course as group 3]. Liver weight, fibrosis, collagen deposition area, and serum levels of AST/ALT were highest in group 2, lowest in group 1, and significantly higher in group 4 than group 3 (all P<0.0001). The levels of inflammatory (TNF-α/NF-κB/IL-1ß/IL-6/MMP-9/VCAM-1/PAI-1/TLR-4, MyD88/IL-12/IFN-γ), oxidative stress (NOX-1/NOX-2/oxidized protein), apoptotic (Bax/cleaved-capase-3/PARP), fibrotic (Smad-3/TGF-ß), and mitochondria-damaged (cytosolic-cytochrome-C) proteins showed an identical pattern, whereas antiapoptotic (Bcl-2), mitochondrial-integrity (mitochondrial-cytochrome-C) and antioxidative (SIRT1/SIRT3) biomarkers exhibited an opposite pattern to fibrosis among the four groups (all P<0.0001). The cellular expressions of inflammatory (Kupffer/CD14/CD44), α-fetoprotein-positively stained biomarkers, apoptotic nuclei and fat cells displayed an identical pattern to fibrosis (all P<0.0001). In conclusion, Ros-PTU therapy attenuated liver fibrosis, inflammatory reaction and generation of oxidative stress and fatty liver after HCD challenge in rabbits.

Introduction

Metabolic syndrome (MetS) is increasingly common worldwide, reflecting rising obesity rates in children and adults, and may overtake smoking as the leading risk factor for coronary artery disease (CAD) [1-3]. MetS is also is an independent risk factor for the development of nonalcoholic fatty liver disease (NAFLD), suggesting that NAFLD may be a hepatic manifestation of MetS [1-7]. The prevalence of NAFLD is also increasing worldwide and is a common cause of chronic liver disease, which can progress to cirrhosis and liver failure [8-12]. Presentation ranges from simple steatosis to non-alcoholic steatohepatitis (NASH) [1-12]. NAFLD and NASH are the most common causal etiologies of chronic liver disease with well-known detriment to patients’ health-related quality of life [9,10]. It is clear that patients with NAFLD are at increased risk of cardiovascular events, which are the leading cause of death in this population [13]. Furthermore, there is mounting evidence that an increased risk of hepatocellular carcinoma (HCC) exists in NAFLD patients, even outstripping other etiologies in some high-income countries [14-17]. Clinical observation studies have also established that the hepatic manifestation of MetS can predispose patients to HCC in the absence of cirrhosis or advanced fibrosis [16].

Although the link between MetS and NASH/NAFLD is well established and the impact of NASH/NAFLD on unfavorable clinical outcomes has been extensively investigated [14-18], the pathogenesis and mechanisms underlying MetS-caused NASH/NAFLD is not well understood [1]. Furthermore, an effective treatment for NASH/NAFLD is not yet available [12].

Statins have unique anti-inflammatory properties [18] that suppress the production of reactive oxygen species (ROS) [19,20] and oxidant/free radicals [21,22]. Statins inhibit plaque formation and attenuate plaque burden in human coronary arteries [23] and rabbit aortas [24]. We have previously demonstrated that propylthiouracil (PTU), a thiouracil-derivative used to treat hyperthyroidism, ameliorates monocrotaline-induced pulmonary artery hypertension in rodent, mainly through suppressing inflammation and enhancing nitric oxide (NO) production [25]. Accordingly, this study tested the hypothesis that statin and PTU therapy could attenuate high-cholesterol diet (HCD)-induced fatty liver disease and fibrosis of liver parenchyma, mainly through reducing inflammation and the generation of oxidative stress, as well as upregulating anti-oxidant and anti-oxidative stress signaling.

Materials and methods

Ethics

All animal experimental protocols and procedures were approved by the Institute of Animal Care and Use Committee at Kaohsiung Chang Gung Memorial Hospital (Affidavit of Approval of Animal Use Protocol No. 2014021801) and performed in accordance with the Guide for the Care and Use of Laboratory Animals [The Eighth Edition of the Guide for the Care and Use of Laboratory Animals (NRC 2011)].

Animals were housed in an Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC)-approved animal facility in our hospital (IACUC protocol no. 101008) with controlled temperature and light cycles (24°C and 12/12 light cycles).

Experimental protocol for high-fat diet (HCD)-induced nonalcoholic fatty liver disease (NAFLD)

The purpose of this experimental model of HCD-induced fatty liver was to mimic the clinical setting of hypercholesterolemia induced NAFLD. The procedure and protocol for HCD-induced fatty liver have recently been described [26]. In detail, thirty-two New Zealand white rabbits weighing 2.2 to 2.5 kg were purchased from BioLasco Technology (Taiwan). They were categorized into four diet groups: (1) Control diet (n=8); (2) 2% high-cholesterol diet [HCD only; n=8; purchased from Test Diet (Richmond, IN)]; (3) 2% HCD plus rosuvastatin administration (20 mg/kg/day; n=8); and (4) 2% HCD plus PTU (0.1% PTU in drinking water; n=8).

Rosuvastatin and PTU treatment began after 4 weeks of HCD feeding and was continued for 4 weeks. The rationale for this PTU therapy regimen was based on our previous report [25,26]. Briefly, the 0.1% PTU in 100 ml water was equivalent to 0.27 mg per 100 ml of drinking water, which was the average daily water consumption for each animal in our study. The safety and efficacy of this therapy regimen were assessed in our previous work [25,26]. Additionally, the therapy regimen of rosuvastatin was based on our previous report [26,27]. After total 8 weeks of study, rabbits were sacrificed and livers were harvested for further analyses.

Biochemical examinations

Rabbit sera were analyzed for blood sugar level, thyroid function (free T4) and lipids, including total cholesterol (TC), high density lipoprotein cholesterol (HDL-C), low density lipoprotein cholesterol (LDL-C) and triglycerides [26] as well as serum level of serum levels of aspartate transaminase (AST) and alanine transaminase (ALT).

Western blot analysis

The procedure and protocol for Western blot analysis were based on our recent reports [28-30]. In detail, equal amounts (50 mg) of protein extracts were loaded and separated by SDS-PAGE using acrylamide gradients. After electrophoresis, the separated proteins were transferred electrophoretically to a polyvinylidene difluoride (PVDF) membrane (GE, UK). Nonspecific sites were blocked by incubation of the membrane in blocking buffer [5% nonfat dry milk in T-TBS (TBS containing 0.05% Tween 20)] overnight. The membranes were incubated with the indicated primary antibodies [Caspase 3 (1:1000, Cell Signaling), Poly (ADP-ribose) polymerase (PARP) (1:1000, Cell Signaling), matrix metallo), tuproteinase (MMP)-9 (1:3000, Abcam), nuclear factor (NF)-κB (1:1000, Abcammor necrosis factor (TNF)-α (1:1000, Cell Signaling), interleukin (IL)-1b (1:1000, Cell Signaling), intercellular adhesion molecule (ICAM)-1 (1:1000, Abcam), plasminogen activator inhibitor (PAI-1) (1:1000, Abcam), toll-like receptor (TLR)-4 (1:500, Novus Biologicals), myeloid differentiation factor 88 (MyD88) (1:1000, Abcam), IL-6 (1:750, Abcam), IL-12 (1:500, Abcam) and interferon (IF)-γ (1:1000, Abcam), SIRT1 (1:4000, Abcam ), SIRT3 (1:500, Abcam), Bax (1:1000, Abcam), cytosolic cytochrome (c-Cyt) C (1:1000, BD), mitochondrial cytochrome (mCyt) C (1:1000, BD), NOX-1 (1:1500, Sigma), NOX-2 (1:750, Sigma), γ-H2AX (1:1000, Cell Signaling), andactin (1:1000, Millipore)] for 1 hour at room temperature. Horseradish peroxidase-conjugated anti-rabbit immunoglobulin (IgG; 1:2000, Cell Signaling, Danvers, MA, USA) was used as a secondary antibody for one-hour incubation at room temperature. The washing procedure was repeated eight times within one hour. Immunoreactive bands were visualized by enhanced chemiluminescence (ECL; Amersham Biosciences, Amersham, UK) and exposed to Biomax L film (Kodak, Rochester, NY, USA). For quantification, ECL signals were digitized using Labwork software (UVP, Waltham, MA, USA).

Immunohistochemical (IHC) and immunofluorescent (IF) staining

The procedure and protocol for IF staining have been detailed in our previous reports [28-30]. For IHC and IF staining, rehydrated paraffin sections were first treated with 3% H₂O₂ for 30 minutes and incubated with Immuno-Block reagent (BioSB, Santa Barbara, CA, USA) for 30 minutes at room temperature. Sections were then incubated with primary antibodies specifically against r-H2AX (1:500, Abcam), Kupffer cell (Monoclonal Mouse Anti-Rabbit Macrophage, Clone RAM11, 1:400, Dako), α-fetoprotein (1:50, Abcam), CD44 (1:500, GeneTex), TUNLE assay (In Situ Cell Death Detection Kit, POD, Roche), and CD14 (1:50, Santa Cruz), while sections incubated with the use of irrelevant antibodies served as controls. Three sections of liver from each rabbit were analyzed. For quantification, three randomly selected high-power fields (HPFs) (200× or 400× for IHC and IF studies, respectively) were analyzed in each section. The mean number of positively-stained cells per HPF for each animal was then determined by summation of all numbers divided by 9.

Assessment of oxidative stress

The procedure and protocol for evaluating protein expression of oxidative stress have been detailed in our previous reports [28-30]. The Oxyblot Oxidized Protein Detection Kit was purchased from Chemicon, Billerica, MA, USA (S7150). DNPH derivatization was carried out on 6 μg of protein for 15 minutes according to the manufacturer’s instructions. One-dimensional electrophoresis was carried out on 12% SDS/polyacrylamide gel after DNPH derivatization. Proteins were transferred to nitrocellulose membranes that were then incubated in the primary antibody solution (anti-DNP 1:150) for 2 hours, followed by incubation in secondary antibody solution (1:300) for 1 hour at room temperature. The washing procedure was repeated eight times within 40 minutes. Immunoreactive bands were visualized by enhanced chemiluminescence (ECL; Amersham Biosciences, Amersham, UK) which was then exposed to Biomax L film (Kodak, Rochester, NY, USA). For quantification, ECL signals were digitized using Labwork software (UVP, Waltham, MA, USA). For oxyblot protein analysis, a standard control was loaded on each gel.

Histological quantification of liver fibrosis and collagen deposition area

Masson’s trichrome staining and Sirius red staining were used for assessing liver fibrosis and collagen deposition areas, respectively. Three serial sections of liver in each animal were prepared at 4 µm thickness by Cryostat (Leica CM3050S). The integrated area (µm²) of fibrotic area and collagen deposition area on each section were calculated using the Image Tool 3 (IT3) image analysis software (University of Texas, Health Science Center, San Antonio, UTHSCSA; Image Tool for Windows, Version 3.0, USA). Three randomly selected high-power fields (HPFs) (100×) were analyzed in each section. After determining the number of pixels in each fibrotic and collagen deposition area per HPF, the numbers of pixels obtained from three HPFs were summated. The procedure was repeated in two other sections for each animal. The mean pixel number per HPF for each animal was then determined by summating all pixel numbers and dividing by 9. The mean integrated area (µm²) of fibrosis and collagen deposition in liver per HPF was obtained using a conversion factor of 19.24 (1 µm² represented 19.24 pixels).

Statistical analyses

Quantitative data are expressed as means ± SD. Statistical analyses were performed by ANOVA, followed by Bonferroni multiple-comparison post hoc test. SAS statistical software for Windows version 8.2 (SAS institute, Cary, NC) was utilized. A P value of less than 0.05 was considered statistically significant.

Results

Lipid profile and thyroid function 8 weeks after HCD administration (Table 1)

Table 1.

Biochemical parameters, thyroid function and body weight by the end of 8 weeks after HCD administration (n=8 for each group)

Variables	Control diet	HCD	HCD-Ros	HCD-PTU	p-value
Total cholesterol (mg/dL)	59 ± 15a	1855 ± 547b	760 ± 94c	923 ± 167d	<0.0001
LDL (mg/dL)	12 ± 3a	528 ± 144b	284 ± 43c	389 ± 34d	<0.0001
TG (mg/dL)	66 ± 29a	247 ± 54b	168 ± 24c	104 ± 24d	<0.001
Sugar (mg/dL)	161 ± 34	156 ± 40	147 ± 37	153 ± 26	0.642
Free T4 (ng/dL)	1.43 ± 0.29	1.42 ± 0.25	1.39 ± 0.33	1.30 ± 0.30	0.684
Body weight (kg)	3.48 ± 0.10	3.38 ± 0.21	3.24 ± 0.27	3.20 ± 0.25	0.426

Data are expressed as mean ± SD. HCD = high-cholesterol diet; TG = triglyceride; Ros = rosuvastatin; PTU = propylthiouracil.

^aindicate significance (at 0.05 level).

^bindicate significance (at 0.05 level).

^cindicate significance (at 0.05 level).

^dindicate significance (at 0.05 level).

The results in Table 1, which were adapted with permission from Lin PY, et al. [26], expressed the blood levels of total cholesterol, low-density lipoprotein (LDL), triglyceride (TG), blood sugar and free T4, as well as body weight at the end of study before sacrifice of animals. At the end of the study period (i.e. at the 8^th week after HCD feeding), plasma levels of free T4 and sugar were not different among the four groups, suggesting that the 0.1% PTU in drinking water did not influence thyroid function of the rabbits. The final body weight also was similar among the four groups. However, the serum levels of total cholesterol, LDL and TG were highest in group 2 (i.e. CHD only), lowest in group 1 (i.e. control group), and significantly higher in group 4 (i.e. HCD + PTU) than in group 3 (i.e. HCD + rosuvastatin).

Liver morphology and weight and histopathological findings of liver parenchyma at the 8^th week after HCD feeding (Figure 1)

Figure 1.

Liver morphology and weight and histopathological findings of liver parenchyma at the 8^th week after HCD feeding. A-D: Illustrating the grossly anatomical feature of liver among the four groups. The remarkably fatty liver in HCD group than in other groups was clearly identified. E: Total liver weight, *vs. other groups with different symbols (†, ‡, §), P<0.001. F-I: Illustrating the H&E light microscopy (100×) for identification of inflamed fatty liver (i.e. steatohepatitis with bright color). Note that the triglyceride droplet vacuoles in the tissue cells of the liver were clearly recognized. J: Analytical result of the fat-deposit area in liver, *vs. other groups with different symbols (†, ‡, §), P<0.0001. Scale bars in right lower corner represent 100 µm. All statistical analyses were performed by one-way ANOVA, followed by Bonferroni multiple comparison post hoc test (n=8 for each group). Symbols (*, †, ‡, §) indicate significance (at 0.05 level). SC = sham control; HCD = high-cholesterol diet; Ros = rosuvastatin; PTU = propylthiouracil. HPF = high-power field.

The grossly anatomical feature showed that as compared with group 1, severe fatty liver was clearly identified in group 2 that was markedly reversed in groups 3 and 4. Additionally, liver weight was significantly higher in group 2 than in groups, 1, 3 and 4, significantly higher in group 4 than in groups 1 and 3, and significantly higher in group 1 than in group 3. Furthermore, H&E light microscopy showed that there was accumulation of triglyceride droplet vacuoles in the tissue cells of the liver in group 2 animals, suggesting that inflamed liver with fatty occurred in the animals. However, this situation was identified to be reversible in groups 3 and 4. Importantly, analytical results showed that the area of inflamed liver with fatty (i.e. steatohepatitis) exhibited an identical pattern to liver weight among the four groups.

Fibrosis and collagen deposition areas at the 8^th week after HCD feeding (Figure 2)

Figure 2.

Fibrosis and collagen deposition areas in liver parenchyma at the 8^th week after HCD feeding. A-D: Illustrating microscopic finding (100×) of Masson’s trichrome staining for identification of fibrosis area (blue). E: Analytical result of fibrosis area, *vs. other groups with different symbols (†, ‡, §), P<0.0001. F-I: Illustrating microscopic finding (100×) of Sirius red staining for identification of collagen deposition area (pink). J: Analytical result of collagen deposition area, *vs. other groups with different symbols (†, ‡, §), P<0.0001. Scale bars in right lower corner represent 100 µm. All statistical analyses were performed by one-way ANOVA, followed by Bonferroni multiple comparison post hoc test (n=8 for each group). Symbols (*, †, ‡, §) indicate significance (at 0.05 level). SC = sham control; HCD = high-cholesterol diet; Ros = rosuvastatin; PTU = propylthiouracil. HPF = high-power field.

Masson’s trichrome staining showed that the fibrosis area in liver parenchyma was highest in group 2, lowest in group 1, and significantly higher in group 4 than in group 3. Additionally, Sirius red staining demonstrated that the collagen deposition area showed an identical pattern to fibrosis among the four groups.

The protein expression of inflammatory biomarkers at the 8^th week after HCD feeding (Figure 3)

Figure 3.

The protein expression of inflammatory biomarkers in liver parenchyma at the 8^th week after HCD feeding. A: Protein expression of plasminogen activator inhibitor (PAI)-1, *vs. other groups with different symbols (†, ‡, §), P<0.0001. B: Protein expression of intercellular adhesion molecule (ICAM)-1, *vs. other groups with different symbols (†, ‡, §), P<0.0001. C: Protein expression of interleukin (IL)-1ß, *vs. other groups with different symbols (†, ‡, §), P<0.0001. D: Protein expression of tumor necrosis factor (TNF)-α, *vs. other groups with different symbols (†, ‡, §), P<0.0001. E: Protein expression of toll like receptor (TLR)-4, *vs. other groups with different symbols (†, ‡, §), P<0.0001. F: Protein expression of myeloid differentiation factor 88 (MyD88), *vs. other groups with different symbols (†, ‡, §), P<0.0001. G: Protein expression of IL-6, *vs. other groups with different symbols (†, ‡, §), P<0.0001. H: Protein expression of 1L-12, *vs. other groups with different symbols (†, ‡, §), P<0.0001. I: Protein expression interferon gamma (IF-γ), *vs. other groups with different symbols (†, ‡, §), P<0.0001. All statistical analyses were performed by one-way ANOVA, followed by Bonferroni multiple comparison post hoc test (n=8 for each group). Symbols (*, †, ‡, §) indicate significance (at 0.05 level). SC = sham control; HCD = high-cholesterol diet; Ros = rosuvastatin; PTU = propylthiouracil.

The protein expression of PAI-1, ICAM-1, IL-1ß, TNF-α, TLR-4, MyD88, IL-6, IL-12 and INF-γ, nine inflammatory biomarkers, were highest in group 2, lowest in group 1 and significantly higher in group 4 than in group 3.

The protein expression of apoptotic, anti-apoptotic and oxidative stress biomarkers at the 8^th week after HCD feeding (Figure 4)

Figure 4.

Protein expression of apoptotic, anti-apoptotic, DNA-damaged and oxidative stress biomarkers in liver parenchyma at the 8^th week after HCD feeding. A: Protein expression of cleaved caspase 3 (c-Casp 3), *vs. other groups with different symbols (†, ‡, §), P<0.0001. B: Protein expression of mitochondrial Bax, *vs. other groups with different symbols (†, ‡, §), P<0.0001. C: Protein expression of cleaved poly (ADP-ribose) polymerase (c-PARP), *vs. other groups with different symbols (†, ‡, §), P<0.0001. D: Protein expression of γ-H2AX, *vs. other groups with different symbols (†, ‡, §), P<0.0001. E: Protein expression of NOX-1, *vs. other groups with different symbols (†, ‡, §), P<0.0001. F: Protein expression of NOX-2, *vs. other groups with different symbols (†, ‡, §), P<0.0001. G: Protein expression of oxidized protein, *vs. other groups with different symbols (†, ‡, §), P<0.0001. (Note: left and right lanes shown on the upper panel represent protein molecular weight marker and control oxidized molecular protein standard, respectively). M.W = molecular weight; DNP = 1-3 dinitrophenylhydrazone. All statistical analyses were performed by one-way ANOVA, followed by Bonferroni multiple comparison post hoc test (n=8 for each group). Symbols (*, †, ‡, §) indicate significance (at 0.05 level). SC = sham control; HCD = high-cholesterol diet; Ros = rosuvastatin; PTU = propylthiouracil.

The protein expressions of cleaved caspase3, Bax and cleaved PARP, three indicators of apoptosis, were highest in group 2, lowest in group 1, and significantly higher in group 4 than group 3. Additionally, the protein expressions of NOX-1, NOX-2 and oxidized protein, three indices of oxidative stress, also showed an identical pattern to apoptosis among the four groups. Additionally, the protein level of γ-H2AX, a marker for DNA damage, exhibited an identical pattern to apoptotsis.

The protein expressions of fibrotic, anti-fibrotic, mitochondrial damage, mitochondrial-integrity and anti-oxidative stress biomarkers at the 8^th week after HCD feeding (Figure 5)

Figure 5.

Protein expressions of fibrotic, anti-fibrotic, mitochondrial damaged, mitochondrial-integrity and anti-oxidative stress biomarkers in liver parenchyma at the 8^th week after HCD feeding. A: Protein expression of phosphorylated (p)-Smad3, *vs. other groups with different symbols (†, ‡, §), P<0.0001. B: Protein expression of transforming growth factor (TGF)-ß, *vs. other groups with different symbols (†, ‡, §), P<0.0001. C: Protein expression of p-Smad1/5, *vs. other groups with different symbols (†, ‡, §), P<0.0001. D: Protein expression of bone morphogenetic protein (BMP)-2, *vs. other groups with different symbols (†, ‡, §), P<0.0001. E: Protein expression of cytosolic cytochrome C (cyt-Cyto C), *vs. other groups with different symbols (†, ‡, §), P<0.0001. F: Protein expression of mitochondrial cytochrome C (mit-Cyto C), *vs. other groups with different symbols (†, ‡, §), P<0.0001. G: Protein expression of SIRT1, *vs. other groups with different symbols (†, ‡, §), P<0.0001. H: Protein expression of SIRT3, *vs. other groups with different symbols (†, ‡, §), P<0.0001. All statistical analyses were performed by one-way ANOVA, followed by Bonferroni multiple comparison post hoc test (n=8 for each group). Symbols (*, †, ‡, §) indicate significance (at 0.05 level). SC = sham control; HCD = high-cholesterol diet; Ros = rosuvastatin; PTU = propylthiouracil.

The protein expression of phosphorylated (p)-Smad3 and TGF-ß, two indicators of fibrosis, were highest in group 2, lowest in group 1, and significantly higher in group 4 than in group 3. On the other hand, the protein expression of p-Smad1/5 and BMP-2, two indicators of anti-fibrosis, showed an opposite pattern to fibrosis.

The protein expression of cytosolic cytochrome C, an indicator of mitochondrial damage, exhibited an identical pattern to fibrosis among the four groups. On the other hand, the protein expression of mitochondrial cytochrome C, an index of mitochondrial integrity, displayed an opposite pattern to fibrosis among the four groups. Consistently, the protein expressions of SIRT1 and SIRT3, two indicators of oxidative stress suppression, exhibited an identical pattern to mitochondrial cytochrome C among the four groups.

Inflammatory cellular expressions at the 8^th week after HCD feeding (Figure 6)

Figure 6.

Inflammatory cellular expressions in liver parenchyma at the 8^th week after HCD feeding. A-D: Illustrating microscopic finding (400×) of immunofluorescent (IF) staining for identification of CD14+ cells (green). E: Analytical results of number of positively-stained CD14 cells, *vs. other groups with different symbols (†, ‡, §), P<0.0001. F-I: Illustrating microscopic finding (400×) of IF staining for identification of CD44+ cells (green). J: Analytical results of number of positively-stained CD44 cells, *vs. other groups with different symbols (†, ‡, §), P<0.0001. Scale bars in right lower corner represent 20 µm. All statistical analyses were performed by one-way ANOVA, followed by Bonferroni multiple comparison post hoc test (n=8 for each group). Symbols (*, †, ‡, §) indicate significance (at 0.05 level). SC = sham control; HCD = high-cholesterol diet; Ros = rosuvastatin; PTU = propylthiouracil.

The IF microscopy showed that the expressions of CD14+ and CD44+ cells, two further indices of inflammation, were highest in group 2, lowest in group 1, and significantly higher in group 4 than in group 3.

The liver inflammatory and damage biomarkers at the 8^th week after HCD feeding (Figure 7)

Figure 7.

Liver inflammatory and damage biomarkers at the 8^th week after HCD feeding. A-D: Illustrating microscopic finding (200×) of immunohistochemical (IHC) staining for identification of Kupffer+ cells (gray). E: Analytical results of number of positively-stained Kupffer cells, *vs. other groups with different symbols (†, ‡, §), P<0.0001. F-I: Illustrating microscopic finding (200×) of IHC staining for identification of α-fetoprotein+ cells (gray). J: Analytical results of number of positively-stained α-fetoprotein cells, *vs. other groups with different symbols (†, ‡, §), P<0.0001. Scale bars in right lower corner represent 50 µm. K: Serum level of aspartate transaminase (AST), *vs. other groups with different symbols (†, ‡, §), P<0.0001. L: Serum level of alanine transaminase (ALT), *vs. other groups with different symbols (†, ‡, §), P<0.0001. All statistical analyses were performed by one-way ANOVA, followed by Bonferroni multiple comparison post hoc test (n=8 for each group). Symbols (*, †, ‡, §) indicate significance (at 0.05 level). SC = sham control; HCD = high-cholesterol diet; Ros = rosuvastatin; PTU = propylthiouracil.

The IHC microscopy showed that the expressions of Kupffer+ cells, an indicator of inflammation, were highest in group 2, lowest in group 1, and significantly higher in group 4 than in group 3. Additionally, IHC stain exhibited that the number of α-fetoprotein positively-stained cells displayed an identical pattern to Kupffer cells among the four groups, suggesting that synthesis of α-fetoprotein in hepatocytes implicated a situation of intrinsic response to liver damage. Furthermore, serum levels of AST and ALT, two indices of liver function, also exhibited an identical pattern of Kupffer cells among the four groups.

Microscopy for apoptotic nuclei at the 8^th week after HCD feeding (Figure 8)

Figure 8.

Microscopy for DNA-damaged marker and apoptotic nuclei at the 8^th week after HCD feeding. A-D: Illustrating microscopic finding (200×) of TUNEL assay for identification of apoptotic nuclei (gray). E: Analytical results of number of apoptotic nuclei, *vs. other groups with different symbols (†, ‡, §), P<0.0001. Scale bars in right lower corner represent 50 µm. All statistical analyses were performed by one-way ANOVA, followed by Bonferroni multiple comparison post hoc test (n=8 for each group). Symbols (*, †, ‡, §) indicate significance (at 0.05 level). SC = sham control; HCD = high-cholesterol diet; Ros = rosuvastatin; PTU = propylthiouracil.

IHC microscopy demonstrated that the cellular expression of apoptotic nuclei (i.e., by TUNEL assay) an indicator of DNA-damage, was highest in group 2, lowest in group 1, and significantly higher in group 4 than group 3.

Discussion

This study investigated the impact of HCD on inducing NAFLD in an experimental model and provided several striking implications. First, an experimental model of NAFLD was successfully created by HCD, providing a platform to survey the underlying mechanism of NAFLD. Second, inflammation and the generation of oxidative stress were found to be the main factors involved in the development of NAFLD. Third, Ros-PTU therapy effectively suppressed HCD-induced NAFLD, highlighting that this regimen may have therapeutic potential in clinical setting for NAFLD patients.

Previous studies [17,31,32] have shown that NAFLD reflects a process of chronic inflammatory disease and can thus present as a range of severity between simple steatosis to NASH [1-12]. An important finding in the present study was that H&E staining, which identifies lipid accumulation (i.e., inflamed fatty liver implicated steatohepatitis) showed that fat cell distribution was much higher in HCD animals compared to control diet animals. The degree of inflammation according to cellular and protein levels was also substantially increased in HCD animals compared to control diet animals. Accordingly, HCD-induced lipid accumulation plays a crucial role in the initiation and propagation of inflammation in this setting. Furthermore, histopathology showed that the specialized macrophage infiltrations (i.e. Kupffer+ cells in liver parenchyma/walls of sinusoids) in liver verified that inflammation had occurred in HCD animals. Therefore, our findings corroborate the findings of previous studies [17,31,32].

Another important finding was that the mitochondrial and DNA damage markers were markedly increased in the HCD group compared to the control diet group. Furthermore, fibrotic (i.e. Masson’s trichrome stain) and collagen-deposition (Sirius red stain) areas, as well as the protein expressions of apoptosis and fibrosis, were remarkably higher in the HCD group than in the control diet group. Additionally, the generation of oxidative stress was significantly higher in HCD animals compared to control diet animals. Interestingly, copious studies have previously identified that inflammation always elicits oxidative stress and vice versa [28,29,33]. Thus, these two processes coexist in conditions involving mitochondrial dysfunction, DNA damage, cell apoptosis/death and, ultimately, organ damage [28,29,33]. In this way, our findings reinforce those of previous studies [28,29,33] and help explain why fibrosis, cell apoptosis and DNA/mitochondrial damage were markedly upregulated in HCD animals compared to control diet animals.

Currently, the treatment of NAFLD/NASH is both uncertain and ineffective [9,10,12]. The most important finding in the present study was that inflammation, oxidative stress, apoptosis, mitochondrial/DNA damage and liver fibrosis as well as NAFLD/NASH in HCD animals were all significantly reduced by PTU treatment, and then significantly reduced further by rosuvastatin treatment. Growing clinical [11-13] and experimental [34] studies have shown that statins can improve fatty liver, NAFLD and NASH. The underlying mechanisms of statin treatment in protecting the endothelial cell/vascular wall against atherosclerosis, cells from apoptosis/death and tissue/organs from damage have been established as mainly through suppressing inflammation [18], inhibiting the generation of ROS [19,20] and reducing the production of oxidant/free radicals [21,22]. These findings are supported by previous studies [11-13,18-22,34] and provide explanation for the improved outcomes observed in HCD animals. However, we have previously shown that PTU can suppress inflammation, oxidative stress, smooth muscle proliferation and arterial atherosclerosis, and enhance nitric oxide (NO) production [25,26]. In this way, the results of our previous investigations [25,26] support the findings of our present study, in that PTU treatment suppressed the molecular-cellular perturbations of inflammation, oxidative stress, mitochondrial/DNA damage and liver fibrosis in HCD animals.

Study limitations

Although the findings in the present study are promising, our study has limitations. First, without a HCD group treated by rosuvastatin plus PTU, we were unable to investigate any potential synergistic effect of combining rosuvastatin and PTU for HCD animals. Second, this study only provided tissue and serum (such as aspartate transaminase and alanine transaminase) levels of liver damage biomarkers without imaging investigations by abdominal ultrasound or magnetic resonance imaging (MRI) for evaluating the degree of liver parenchymal disease in living animals.

In conclusion, the present study demonstrated that HCD induced NAFLD/NASH and liver fibrosis in rabbits, and found a therapeutic role for PTU and rosuvastatin for protecting the liver from HCD-induced damage. Rosuvastatin-PTU may therefore be an alternative management strategy for NAFLD, which is a rapidly emerging worldwide health problem.

Disclosure of conflict of interest

None.