A total of 24 male 8-week-old C57BL/6 mice (18–22 g) were purchased from Vital River Laboratory Animal Technology Co. Ltd (Beijing, China) and randomly divided into four groups, with 6 mice in each group. These were the following: Negative control (NC), TUDCA, CIH and CIH+TUDCA groups. Animals in NC group were kept in normal air (21% O₂), while the CIH model was established according to Lai et al (29) with minor modification: Mice were housed in a sealed cabin, the air was controlled by an automated nitrogen/air switch pump. The concentration of O₂ was gradually reduced to 7%, stabilized for 60 sec, then increased to a peak of 21%, maintained for 60 sec, this cycle was repeated over 8 h (from 8:30 to 16:30) during the animals' diurnal sleep period. Mice in CIH+TUDCA group were treated with TUDCA 100 mg/kg/day by intraperitoneal injection, start from 2 days before CIH until the end of the experiment. Mice in TUDCA group were intraperitoneal injected with 100 mg/kg/day at the same schedule but did not undergo CIH treatment. Eight weeks later, the mice were anesthetized with an intraperitoneal injection of pentobarbital sodium (50 mg/kg; Chengdu Xiya Chemical Technology Co., Ltd., Chengdu, China), eyeballs were removed, blood samples were collected, animals were sacrificed by exsanguination as previously reported (30), and the the livers were harvested. The care and disposal procedures of animal were strictly subject to the provisions of the Animal Ethics Committee of China Medical University (Shenyang, China).

Serum alanine aminotransferase (ALT), aspartate aminotransferase (AST) and alkaline phosphatase (ALP) were determined by commercial kits (Nanjing Jiancheng Bioengineering Institute, Jiangsu, China) following the user's instructions. Liver tissues were homogenized, centrifuged and total liver lysate were collected. Reactive oxygen species (ROS) activity was measured by ROS assay kit (Nanjing Jiancheng Bioengineering Institute). The level of tumor necrosis factor α (TNF-α) and interleukin-1β (IL-1β) in liver were determined by ELISA kits purchased from Wuhan Boster Biological Technology, Ltd. (Wuhan, China).

Liver tissues was fixed with formalin, embedded in paraffin and cut into 5-µm sections. After dehydrated with grade alcohols, the sections were stained with hematoxylin and eosin terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining was performed using a commercial kit (KeyGEN, Nanjing, China) according to the manufacturer's instruction. Briefly, the sections was incubated with 1% Triton X-100 for 8 min at room temperature, then, endogenous peroxidase activity was blocked by 3% H₂O₂. Followed incubation with Proteinase K solution, the sections were stained with TUNEL reaction mixture at 37°C for 60 min. Then TUNEL positive cells were labeled with streptavidin-horse radish peroxidase (HRP) and developed with 3,3′-diaminobenzidine. Finally, the sections were counterstained with hematoxylin.

The liver tissues were dissolved by radioimmunoprecipitation assay buffer (Beyotime Institute of Biotechnology, Haimen, China), total protein was isolated and the protein concentration was determined using a bicinchoninic acid Method-Protein Quantitative Reagent kit (Beyotime). After that, tan equal quantity of protein was fractionated using 8, 10 or 13% SDS-PAGE and transferred onto polyvinylidene difluoride membranes (EMD Millipore, Bedford, MA, USA). Then, the membranes were blocked by 5% (w/v) fat-free milk and immunoblotted with primary antibodies (dilution and suppliers were listed in Table I) 4°C overnight. After washing, the membranes was incubated with HRP-conjugated secondary antibodies (goat anti-rabbit IgG, 1:5,000; goat anti-mouse IgG, 1:5,000; Beyotime Institute of Biotechnology) at 37°C for 45 min. Next, the specific protein bands was visualized using enhanced chemiluminescence reagent (Qihai Biotec, Shanghai, China) and the gray density was analyzed using Gel-Pro Analyzer software version 3.0 (Media Cybernetics, Inc., Silver Spring, MD, USA).

Data are expressed as the mean ± standard deviation. Differences between two groups were analyzed by Student's t-test. P<0.05 was used to indicated a statistically significant difference. Statistical analysis was performed using SPSS version 18.0 (SPSS, Inc., Chicago, IL, USA).

As shown in Fig. 1A, histology analysis of the liver tissues revealed normal hepatic architecture with no obvious lobular or portal inflammation, steatosis or necrosis in NC and TUDCA group. In contrast, mice exposed to CIH showed liver injury with discernible swelled, disordered hepatocytes and infiltrated inflammatory cells, however, these pathological changes were evidently ameliorated by the treatment of TUDCA. In order to evaluate the effect of TUDCA on CIH-induced liver function injury, the serum levels of liver enzymes ALT, AST and ALP were determined. As shown in Fig. 1B, the levels of ALT and AST in CIH group was significantly elevated when compare to the NC group (71.96±18.64 vs. 27.38±6.36 U/l, P<0.01; 213.06±56.24 vs. 75.18±18.97 U/l, P<0.01), but these were significantly decreased in CIH+TUDCA group when compared to CIH group (38.41±8.64 vs. 71.96±18.64 U/l, P<0.01; 109.39±30.26 vs. 213.06±56.24 U/l, P<0.01). There were no significant changes in the serum ALP level in CIH or CIH+TUDCA groups. The above results indicate that TUDCA alleviated CIH-induced liver pathological changes and protected liver function.

The activity of ROS in liver was detected to evaluate the role of TUDCA on CIH-induced liver oxidative stress (Fig. 2A). Exposed to CIH-induced a 1.8-fold increase in ROS activity in the liver when compared to the NC group, while treatment with TUDCA decreased by 30% of the ROS activity when compared to CIH group. We further measured the levels of proinflammatory factors in livers to detect the anti-inflammatory effect of TUDCA, as shown in Fig. 2B and C, in the CIH group, there was a significant elevation in both TNF-α (515.00±112.83 vs. 50.26±11.89 pg/mg, P<0.01) and IL-1β (122.21±32.82 vs. 44.06±7.50 pg/mg, P<0.01) level when compared to the NC group. However, these levels were strongly decreased in the CIH+TUDCA group when compared to the CIH group (TNF-α, 130.75±38.80 vs. 515.00±112.83 pg/mg, P<0.01; IL-1β, 83.60±17.58 vs. 122.21±32.82 pg/mg). Thus, treatment with TUDCA suppressed CIH-induced oxidative stress and inflammatory reactions in liver.

After eight weeks exposure to CIH, the presence of hepatocytes undergoing apoptosis was determined by TUNEL staining. As shown in Fig. 3A, TUNEL-positive apoptotic cells could obviously seen in the liver of CIH group, as expected, treatment with TUDCA distinctly reduced the apoptotic cell number. Additionally, the expression of two important biochemical markers of apoptosis, cleaved caspase-3 and cleaved caspase-9 were evaluated using western blot (Fig. 3B and C). Exposure to CIH significantly upregulated the protein levels of the two makers (P<0.01), while these changes were evidently restored by the treatment of TUDCA. These results suggested that TUDCA protected the liver from CIH-induced apoptosis.

To delineate the role of TUDCA on CIH-induced ER stress, western blot was used to evaluate the protein levels of the molecular chaperone 78 kDa glucose-related protein (GRP78) and three major transducers of unfolded protein response (UPR), eukaryotic translation initiation factor 2 α kinase 3 (PERK), activating transcription factor (ATF)6 and inositol-requiring enzyme (IRE1), that are known to be specifically induced by ER stress. As shown in Fig. 4, CIH increased the protein levels of GRP78, PERK, cleaved ATF6 and p-IRE1 in liver (P<0.01); however, these proteins were significantly reduced in the CIH+TUDCA group when compared to the CIH group (P<0.05). With the changed level of GRP78 and three UPR transducers, we further investigated the signal molecules of ER stress-associated apoptotic pathways (Fig. 5). There was an evident increase of C/EBP homologous protein (CHOP), cleaved caspase-12, p-eukaryotic translation initiation factor 2 subunit α (eIF2α) and p-janus kinase (JNK) in CIH group when compared to the NC group (P<0.01). However, these changes of protein levels were all significantly reversed by the treatment of TUDCA (P<0.05). Therefore, the present findings demonstrated that TUDCA suppressed CIH-induced ER stress and ER stress-associated apoptosis of the liver.