The present study utilized a total of 26 4-week-old male C57BL/6 J ApoE^−/− mice, weighing 16.30±0.45 g (Model Animal Research Center of Nanjing University, Nanjing, China). All mice were bred in a specific pathogen-free facility and maintained in a 12-h light-dark cycle at a controlled temperature (22±2°C) and humidity (55±5%) and fed ad libitum. The mice were divided into the following three groups, which were given different diets: Standard chow diet (SC group, n=8); high-fat high-cholesterol diet (HFHC group, n=8); and HFHC diet supplemented with BBR (BBR group, n=10). C57BL/6J ApoE^−/− mice receiving a HFHC diet serves as a typical model of obesity-induced NASH (13). The chow diet from Research Diets, Inc. (New Brunswick, NJ, USA; cat. no. D10012 G) contained 15.8% fat, 20.3% protein, and 63.9% carbohydrate. The HFHC diet feed (cat. no. D12079B; Research Diets, Inc.) contained 41% fat, 17% protein, 43% carbohydrate and 0.21% cholesterol. BBR (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) was administered at a dosage of 0.2 g/kg daily, starting from the sixth week to the end of the experiment. Food intake was measured 3 times/week and individual body weight was measured weekly over a 12-week period. At the end of the experiments, the animals fasted for 12 h, and following anesthesia with 3% pentobarbital (Sigma-Aldrich; Merck KGaA) and sacrifice via cervical dislocation. Blood samples were rapidly obtained by retro-orbital sampling, as previously described (14). Liver tissues were harvested and samples were fixed with 4% formaldehyde at room temperature for 24 h, whilst the remaining liver was snap-frozen in liquid nitrogen until required. All animal protocols were reviewed and approved by the Ethics Committee of Hangzhou Normal University Affiliated Hospital (Hangzhou, China).

Serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), triglyceride and cholesterol levels were measured using a Hitachi 7180 automatic biochemical analyzer (Hitachi Ltd., Tokyo, Japan) according to the manufacturer's protocol. Serum NE and α1-AT levels were examined using a Mouse Neutrophil Elastase/ELA2 DuoSet ELISA or α 1 Antitrypsin ELISA kit following the manufacturers' instructions (DY4517-05, R&D Systems, Inc., Minneapolis, MN, USA; ab205088, Abcam, Cambridge, UK).

Formalin-fixed liver tissues were embedded in paraffin, sectioned (5 µm thickness), and stained with hematoxylin and eosin as previously described (15). Frozen liver sections were cut to 10 µm thickness, fixed with 10% paraformaldehyde for 5 min at room temperature and stained with Oil Red O for 10 min at room temperature. Histological scoring was performed by two specialists blinded to the experimental design and data. Scoring ranges were as follows: Degree of steatosis (0–3), lobular inflammation (0–3), hepatocyte ballooning (0–2). In addition, the NAFLD activity score (NAS) was scored according to the NASH clinical research network scoring system (16).

Immunohistochemical staining was performed as previously described (6). The primary antibodies used were as follows: Anti-NE (1:200; cat. no. orb1614; Biorbyt Ltd., Cambridge, UK), anti-C-X-C chemokine receptor type 4 (CXCR4; 1:50; cat. no. ab124824, Abcam) and anti-C-X-C motif chemokine 12 (CXCL12; 1:50; cat. no. sc-28876; Santa Cruz Biotechnology, Inc., Dallas, TX, USA). Samples were then incubated with the anti-rabbit (1:2,000; cat. no. ab97051; Abcam) secondary antibody for 1 h at room temperature under gentle agitation. Samples were subsequently visualized under an Eclipse 80i light microscope (Nikon Corporation, Tokyo, Japan) and representative images were captured for analysis.

Hepatic mRNA levels were analyzed by RT-qPCR using an ABI Prism 7900 HT Fast Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc., Waltham, MA, USA). Total RNA was isolated from liver tissues using TRIzol (Invitrogen; Thermo Fisher Scientific, Inc.) followed by DNAse treatment to remove genomic DNA. cDNA was synthesized using 2 µg of total RNA with Superscript^® II Reverse Transcriptase (Invitrogen; Themo Fisher Scientific, Inc.) at 42°C for 50 min. Amplification reactions were performed using a SYBR Select Master Mix (AmpliTaq^® Fast DNA Polymerase, Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. PCR conditions were as follows: 95°C for 10 min, followed by 40 cycles at 95°C for 10 sec and 60°C for 1 min. PCR products were verified by melt curve analysis (17). Mouse β-actin or GAPDH were used as the reference genes to normalize for differences in the amount of total RNA in each sample. Relative gene expression levels were analyzed using the 2^−ΔΔCq method (18). PCR murine primers were obtained from Invitrogen (Thermo Fisher Scientific, Inc.) and their sequences were as follows: β-actin forward (F), 5′-CTGGCTCCTAGCACCATGAA-3′ and reverse (R), 5′-CGCAGCTCAGTAACAGTCCG-3′; GAPDH F, 5′-AACAGCAACTCCCACTCTTC-3′ and R, 5′-CCTGTTGCTGTAGCCGTATT-3′; α1-AT F, 5′-GAGCATTGGCACAGCGTTTG-3′ and R, 5′-AAGCGATGGTTGGATGTCAGC-3′; CXCR4 F, 5′-CTACAGCAGCGTTCTCATCC-3′ and R, 5′-TTTCAGCCAGCAGTTTCCTT-3′; CXCL12 F, 5′-CGACTTCTCAGGCAGGTGA-3′ and R, 5′-GCCATTCCCATAGCATTCAT-3′; phosphatidylinositol-3 kinase (PI3K) F, 5′-AATGTGCCCTCTTTCGTTGT-3′ and R, 5′-TGAATGGTGACTGGCTGACT-3′; interleukin (IL)-1 F, 5′-CTCTTGCCTGTCATCCCAAC-3′ and R, 5′-ACCATCTGTTCCCAATACGG-3′; IL-6 F, 5′-CGGAGAGGAGACTTCACAGAG-3′ and R, 5′-CATTTCCACGATTTCCCAGA-3′; IL-8 F, 5′-GGATGGGAACAACGATAGGA-3′ and R, 5′-AGAACGTGGCGGTATCTCTG-3′; protein kinase B (AKT) F, 5′-ACTCATTCCAGACCCACGAC-3′ and R, 5′-ACAATCTCCGCACCATAGA-3′; nuclear factor-κB (NF-κB) F, 5′-GCTCAAGATCTGCCGAGTAAA-3′ and R, 5′-GTCCCGTGAAATACACCTCAA-3′; tumor necrosis factor (TNF)-α F, 5′-AAGGGAGAGTGGTCAGGTTG-3′ and R, 5′-TCTGTGAGGAAGGCTGTGC-3′; phospholipase C-β (PLC-β) F, 5′-GCAGGTCCAAGTGTTGATTG-3′ and R, 5′-TTCTTCTCCGCTCAGGTAGC-3′; and monocyte chemoattractant protein-1 (MCP-1) F, 5′-TCTCTCTTCCTCCACCACCAT-3′ and R, 5′-GCTCTCCAGCCTACTCATTGG-3′. Each experiment was performed in duplicate and repeated three times.

Liver tissue (60 mg) was homogenized in RIPA buffer (P0013B; Beyotime Institute of Biotechnology, Haimen, China), supplemented with protease inhibitor cocktail (Sigma-Aldrich; Merck KGaA). Samples (50 µg protein per lane) were separated by 10% SDS-PAGE and electrotransferred onto polyvinylidene difluoride membranes. Following blocking with 5% nonfat milk, membranes were probed with anti-NE antibody (1:200) overnight at 4°C. The membranes were then incubated with the secondary antibody for 1 h at room temperature, followed by SuperSignal™ West Pico Chemiluminescent detection (34077; Thermo Fisher Scientific, Inc.). Protein band intensities were quantified using Quantity One software 4.62 (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

The Neutrophil Elastase 680 FAST imaging agent (PerkinElmer, Inc, Waltham, MA, USA) was used to detect NE in vivo at the end of the experiment according to the manufacturer's protocol. This agent is based on a peptide (PMAVVQSVP) that is specifically recognized by NE. Mice were administered with 4 nmol of the NE imaging agent intravenously via the tail vein, and 4 h later an in vivo imaging system (Molecular Imaging software version 5.3.5; Carestream, CA, USA) was used to capture the fluorescence and detect NE activity.

Statistical analysis was conducted using GraphPad Prism (version 5; GraphPad Software, Inc., La Jolla, CA, USA). All data are presented as the mean ± standard deviation. The statistical significance of differences between groups was assessed using an unpaired Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

The effect of BBR on weight gain and liver metabolic features in C57BL/6J ApoE^−/− mice receiving a HFHC diet was observed (Fig. 1). Mice fed a HFHC diet had significantly increased body weights, associated with hypertriglyceridemia and hypercholesterolemia, compared with regular chow-fed mice in the SC group (P<0.05; Fig. 1A). This weight gain was also accompanied by a significantly higher relative liver:body weight ratio (liver index) in the HFHC group compared with the SC group (P<0.01; Fig. 1B). Furthermore, mice in the HFHC group had significantly enhanced increased ALT and AST levels, indicative of liver injury, compared with the SC group (P<0.01; Fig. 1E and F).

BBR supplementation markedly reduced body weight, serum total cholesterol and triglyceride levels in ApoE^−/− mice, though the results were not statistically significant compared with the SC group (Fig. 1A, C and D). Thus, treatment with BBR did not notably impact the global metabolic profile of the mice. However, the liver index and serum ALT and AST levels were significantly lower in the BBR group compared with the HFHC group, demonstrating a hepatoprotective role of BBR against liver injury (P<0.05; Fig. 1E and F).

Compared with mice in the SC group, significant pathological findings of NASH, including marked steatosis (P<0.01), ballooning (P<0.05), lobular inflammation (P<0.01) and NAS (P<0.01) were observed in the HFHC group (Fig. 2). However, no significant fibrosis was observed in the mouse liver tissues.

BBR-treated mice had a lower steatosis score compared with HFHC-fed mice, although this difference was not statistically significant (Fig. 2B). Lobular inflammation and ballooning scores were significantly lower in the BBR group compared with the HFHC group (P<0.05; Fig. 2C and D). Indeed the overall NAS score indicated that BBR treatment had a protective effect, being significantly lower in the BBR group compared with the HFHC group (P<0.05; Fig. 2E). These data indicate that inflammation is significantly associated with HFHC diet-induced liver damage, and that treatment with BBR significantly ameliorates this inflammation and liver injury.

Feeding a high-fat diet to mice induces an increase in neutrophil recruitment to tissues (19), which may initiate inflammatory signaling pathway cascades. Particularly, NE serves an important role in host defense and inflammation (20), whereas its counterpart, α1-AT, protects tissues from serine protease-induced damage (8). To validate whether the expressions of α1-AT and NE were altered in the present model, ELISA assays were performed. The results demonstrated that serum α1-AT levels were significantly lower in the HFHC group compared with the SC or BBR groups (P<0.05; Fig. 3A), whereas NE levels and the NE/α1-AT ratio were significantly higher in the HFHC group compared with the SC and BBR groups (P<0.05; Fig. 3B and C).

Histopathological and immunoblot examinations of liver specimens demonstrated that NE protein expression was significantly elevated in the HFHC group compared with the SC group (P<0.05), and BBR treatment significantly reduced NE expression in liver tissues (P<0.05) (Fig. 3D and E). It was also observed that NE enzymatic activity was markedly associated with NASH (the HFHC group) and inhibited by BBR treatment (Fig. 3F). In contrast to NE, α1-AT protein expression in the HFHC group was significantly lower compared with that in the SC group (P<0.05), and α1-AT levels were significantly upregulated in the BBR group compared with the HFHC group (P<0.05) (Fig. 3G). This NE/α1-AT balance is consistent with the reduced inflammatory foci and ALT levels observed in the ApoE^−/− mice following BBR treatment.

Localization of inflammatory cells to the liver is an essential step in the progression of NASH (21). The active form of NE is externalized during neutrophil activation at inflammatory sites, thus helping to promote inflammatory and immune responses (20,22). The present study investigated whether NE elevation affected the expression of inflammatory chemokines and cytokines. The HFHC diet stimulated the expression of TNF-α, MCP-1, IL-6, CXCR7 and CXCL11 in the mouse liver compared with mice fed a standard diet (data not shown). However, although BBR treatment reduced the expression of these cytokines and chemokines, no statistically significant differences were observed between the HFHC and BBR group (data not shown).

Previous studies have suggested that the CXCR4/CXCL12 signaling pathways serves a key role in maintaining neutrophil homeostasis (23). Liver mRNA levels of CXCR4 and CXCL12 were significantly lower in the BBR group compared with the HFHC group (P<0.05; Fig. 4A and B). Immunochemistry staining yielded similar results for the protein levels of CXCR4 and CXCL12 (Fig. 4C). Gene expression analysis found no significant differences in the expression of mitogen-activated protein kinase 14, extracellular signal-regulated kinase 1/2 or and B-cell lymphoma 2-associated agonist of cell death between the HFHC and BBR groups (data not shown). However, RAS, PLC-β, PI3K, AKT, NF-κB, IL-1 and IL-8 expression was significantly attenuated by BBR treatment (P<0.05; Fig. 4D-J).