Effects of XIAP on high fat diet-induced hepatic steatosis: a mechanism involving NLRP3 inflammasome and oxidative stress

XIAP is down-regulated in the livers of both HFD-fed mice and palmitate-treated cultured hepatocytes

To determine whether fat accumulation in the liver affects XIAP expression in vivo, mice were fed either an HFD or a standard chow diet (SCD) as a control. As shown in Figure 1A, HFD mouse livers had significantly lower expression levels of both XIAP mRNA and protein compared to the controls. Immunohistochemical (IHC) assay further demonstrated the down-regulation of XIAP in HFD-fed mouse livers (Figure 1B). To determine whether fat accumulation in the liver affects XIAP expression in vitro, both primary hepatocytes and AML-12 cells were incubated with 250 μM PA for 0, 4, 8, 12, or 24 h. As expected, both XIAP mRNA and protein expression levels were significantly decreased in the PA-treated cells compared to controls, and this decrease occurred in a time-dependent manner (Figure 1C). These results suggest that XIAP likely participates in the regulation of HFD-induced fatty liver.

XIAP knockdown aggravates metabolic syndrome, liver inflammation, and hepatic steatosis in HFD-fed mice

In order to elucidate the role of XIAP in hepatic steatosis and its complications, XIAP knockdown mice were generated by in vivo siRNA transfection and then fed an HFD for 12 weeks. Prior to this, the efficiency of XIAP siRNA in vivo in high fat diet induced fatty liver were confirmed using qPCR and western blotting assay at 1^st, 4^th, 8^th, and 12^th week during the period of in vivo siRNA transfection, as indicated in Supplementary Figure 1. Accordingly, as shown in Figure 2A, the recorded body weights at 0, 4, 8, and 12 weeks were significantly increased in both ItsiRNA/HFD and ItsiXIAP/HFD mice compared to SCD-fed mice, suggesting that a long-term HFD leads to dramatic weight fluctuations in mice. In addition, knockdown of XIAP in vivo may result in a further increase in mouse body weight, as previous studies have demonstrated that a prolonged HFD significantly contributed to the development of metabolic disorder [3, 10].

To identify the effect of XIAP on HFD-induced systemic metabolic syndrome and hepatic inflammation, fasting blood glucose and insulin levels in 0- and 12-week-old mice were measured. As illustrated in Figure 2B, HFD-fed mice had significantly increased fasting blood glucose and insulin levels compared to controls. Interestingly, ItsiXIAP/HFD mice had higher blood glucose and insulin levels than did the ItsiRNA/HFD group, indicating that XIAP knockdown may have promoted the development of HFD-induced metabolic imbalance in mice. This result is supported by the up-regulation of liver injury-associated indicators in these mice. As shown in Figure 2C, there was a significant increase in the liver wet weight in HFD-fed mice compared to SCD-fed mice, which was due in part to excessive lipid deposition. Moreover, there was a remarkable difference in liver weight between the ItsiXIAP/HFD and ItsiRNA/HFD groups, suggesting that XIAP knockdown itself was capable of altering the weight of livers in HFD-fed mice. To determine the degree of hepatic injury in both the HFD- and SCD-fed mice, livers from these mice were subjected to Hematoxylin and Eosin (H&E) staining and Sirius Red staining (Figure 2D). Higher levels of steatosis, necrosis, and inflammation, as well as higher ballooning and NAFLD activity (NAS) scores (Figure 2E), were observed in ItsiXIAP/HFD and ItsiRNA/HFD mice compared to SCD-fed mice, suggesting that long-term HFD ingestion contributed to the development of inflammatory infiltration in liver and hepatic steatosis. Notably, mice injected with XIAP-siRNA exhibited more serious pathological changes and had a higher risk of developing NAFLD than did mice injected with ItsiRNA/HFD. This indicates that XIAP expression may help to stabilize and recover metabolic balance in HFD-induced fatty livers. Furthermore, mice fed an HFD had increased AST, ALT, and AKP levels and serum endotoxin content, suggesting that an HFD promoted hepatic steatosis in these mice and may be associated with differential regulation of XIAP.

To determine whether XIAP activity is essential for the regulation of metabolic balance, glucose tolerance tests (GTT) and insulin tolerance tests (ITT) were used to measure hepatic lipid metabolism indicators in both HFD- and SCD-fed mice As indicated in Figure 3A, ItsiXIAP/HFD and ItsiRNA/HFD mice had higher levels of glucose tolerance and insulin resistance than did the SCD-fed control group. Notably, siXIAP-treated mice exhibited higher glucose levels and dramatic shifts in area under curve (AUC) scores, suggesting that knockdown of XIAP in vivo may impair glucose tolerance and insulin sensitivity, further exacerbating metabolic disorders. This hypothesis was supported by the presence of lipid deposition in the liver as measured by oil-red staining (Figure 3B), as well as changes in adipocyte area (Figure 3C), triglyceride (TG) and total cholesterol (TC) levels in the liver or serum (Figure 3D), and the alteration of glycosylated hemoglobin (HbA1c) (Figure 3E). In addition, XIAP knockdown resulted in a significant up-regulation in the expression of genes associated with fatty acid synthesis and uptake (p-ACCα, FAS, SCD1, CD36, FATP1, and FABP1) and a significant down-regulation in the expression of genes related to fatty acid β-oxidation (CPT-1α and PPAR-α) in the livers of mice fed an HFD (Figure 3F and 3G). These data suggest that XIAP plays a vital role in the regulation of HFD-stimulated metabolic imbalance.

Previous studies have shown that XIAP activation was able to suppress the caspase-8/NLRP3 signaling-triggered inflammation response, which was induced significantly by prolonged stimulation [19, 20]. To determine whether NLRP3 expression was altered and how XIAP activity affects NLRP3-mediated inflammatory responses in HFD-induced NAFLD, the expression levels of involved proteins were measured. As shown in Figure 4A and Table 1, IHC detection of caspase-8 in the liver sections showed a significant increase in caspase-8 expression in HFD-treated mice compared to the SCD group. This indicated that the up-regulation of caspase-8 expression contributed to the development of NAFLD in HFD-fed mice. Immunoblotting analysis revealed that NLRP3, mature IL-18, and mature IL-1β were more highly expressed in the fatty livers of HFD-fed mice as compared to the SCD-fed controls (Figure 4B and 4C). Moreover, the mRNA expression levels of NLRP3, ASC, and caspase-1 were increased in HFD-fed mice compared to controls. These results indicate that the activation of NLRP3 signaling is involved in the development of NAFLD

XIAP knockdown exacerbates HFD-induced oxidative stress in fatty livers

Previous studies have shown that the long-term intake of excess fat associated with an HFD contributes to the development of obesity, diabetes, cardio-cerebrovascular disease, and nonalcoholic steatohepatitis through the up-regulation of oxidative stress levels [1–7]. In order to better understand the role that XIAP plays in the HFD-induced development of fatty liver, the level oxidative stress in the liver was measured. As expected, the HFD-induced increase in oxidative stress levels was further elevated in the livers of mice with XIAP knockdown. There was an increase in ROS generation, malondialdehyde (MDA) production, H₂O₂ and O₂^- content, and XO and iNOS activity, as well as a reduction in SOD, TAC, CAT, GST, and GSH-Px activity in ItsiXIAP/HFD mice compared to ItsiRNA/HFD and ItsiRNA/SCD mice (Figure 5A). Notably, the impairment of hepatic Nrf2 signaling caused by prolonged HFD ingestion was further aggravated in XIAP knockdown mice. This was accompanied by the functional loss of Nrf2, SOD1, NQO-1, HO-1, and GCLM activities in these mice compared to the corresponding control groups (Figure 5B and C). These data were also supported by IHC analysis, which showed that the protein expression levels of Nrf2, HO-1, and SOD1 expression were down-regulated in ItsiXIAP/HFD mice compared to ItsiRNA/HFD mice (Table 1). These results demonstrate that XIAP knockdown exacerbates the effects of HFD-induced NAFLD by triggering an increase in oxidative stress (Figure 5D).

XIAP knockdown promotes PA-induced lipometabolic disturbance in vitro

To further define the role of XIAP in HFD-induced hepatic steatosis, primary hepatocytes and AML-12 cells were used to evaluate the differential expression of oxidative stress indicators upon treatment with PA in vitro. Both AML-12 cells and primary hepatocytes were transfected with either XIAP siRNA or the corresponding control vectors and then treated with 250 μM PA for 24 h to establish the in vitro model. As shown in Figure 6A, XIAP expression was downregulated in both PA-treated AML-12 cells and hepatocytes, indicating that XIAP knockdown efficiency was good. Similar to the in vivo results, XIAP knockdown resulted in a significant upregulation of fatty acid synthesis and uptake indicators, including ACCα, FAS, and SCD1, as well as a sharp increase in lipid deposition in PA-treated AML-12 cells and hepatocytes (Figure 6B and 6C). Moreover, changes in fatty acid β-oxidation indicators, such as CPT-1α, PPARα, CD36, FATP1, and FABP1, indicated that they may play a role in the development of lipometabolic disturbances in PA-treated cells. These results suggested that PA treatment could alter lipid homeostasis in liver-related cell lines, promote fatty acid synthesis and uptake, and suppress fatty acid β-oxidation, ultimately increasing lipid accumulation in cells. Additionally, activation of NLRP3 signaling was observed in PA-treated AML-12 cells and hepatocytes. As shown in Figure 7A and 7B, there were significant increases in both NLRP3 mRNA and protein levels, indicating an increase in the activation of NLRP3 signaling in PA-treated cells. Taken together, these data suggest that NLRP3 signaling triggered inflammation-associated lipometabolic disturbance and induced lipid deposition in vitro.

XIAP knockdown facilitates PA-induced oxidative stress in vitro

In vivo experiments have shown that XIAP knockdown promoted HFD-induced oxidative stress injury and NLRP3 signaling-related lipid deposition in the liver of mice, further stimulating the development of NAFLD. As a result, the alteration of oxidative stress levels in PA-treated cells was investigated. DCFH-DA analysis revealed that ROS production was increased in PA-treated AML-12 cells and hepatocytes compared to the control groups (Figure 8A), suggesting that PA treatment alone is able to promote the generation of ROS in vitro. In addition, 24-h PA treatment reduced the viability of AML-12 cells and hepatocytes (Figure 8B). The functional loss of XIAP in PA-treated cells resulted in both an increase in oxidative stress, as evidenced by an increase in MDA production and H₂O₂ and O₂^- content, as well as a decrease in the antioxidant activities ofSOD, TAC, CAT, GST, and GSH (Figure 8C). These observations suggest that PA-induced oxidative stress injury contributes to the imbalance of lipid metabolism in cells. In addition, the activation of Nrf2 signaling was significantly suppressed in PA-treated cells (Figure 8D). Reduction in the mRNA and protein expression of antioxidants, such as NQO-1, HO-1, and SOD1, was more pronounced in these cells compared to the siRNA/PA groups, suggesting that PA promoted oxidative stress injury in cells by downregulating antioxidant production. These results indicate that PA-induced imbalances in cellular lipid metabolism are associated with oxidative stress injury and the suppression of antioxidant production.

Enhancing XIAP expression reversed PA-induced oxidative stress injury in hepatic cells

To explain the role that XIAP plays in PA-induced oxidative stress more comprehensively, an XIAP overexpression cell model was constructed. Primary hepatocytes and AML-12 cells were transfected with pUNO1/mXIAP or a corresponding control vector and then incubated with 250 μM PA for 24 h to establish the in vitro model. As shown in Figure 9A, hepatocytes and AML-12 cells treated with 50 ng or 100 ng of the vectors exhibited higher expression of both XIAP mRNA and protein compared to the 0 ng loading control group. Both mRNA and protein expression levels of ACCα, FAS, and SCD1 were down-regulated in cells overexpressing XIAP (Figure 9B), indicating that XIAP activation helps maintain lipid metabolism balance in PA-treated cells through the regulation of fatty acid synthesis and uptake genes. Moreover, cells overexpressing XIAP had a decrease in both the lipid content (TC and TG levels) (Figure 9C) and the expression of fatty acid synthesis and uptake genes (ACCα, FAS, SCD1, CD36, FATP1, and FABP1), as well as an up-regulation in the expression of fatty acid β-oxidation genes (CPT-1α and PPAR-α) compared to control cells (Figure 9D). Consistent with previous studies [19, 20], cells overexpressing XIAP had a reduction in NLRP3 signaling activation, and inflammation-associated dyslipidemia induced by PA treatment was eliminated in these cells (Figure 10A and 10B). These results suggest that XIAP is involved in the regulation of the NLRP3 inflammasome-mediated inflammatory response in PA-treated hepatic cells.

Similar to the in vivo results, overexpression of XIAP resulted in both an increase in the generation of antioxidants and a decrease in the production of ROS without affecting cell viability (Figure 11A–11C), suggesting that XIAP regulates the PA-induced increase in oxidative stress injury in cells by regulating antioxidant production. Moreover, both Nrf2 activation and the expression of down-stream indicators were significantly increased in PA-treated cells compared to the EV group, indicating that XIAP plays an essential role in the regulation of Nrf2 activity in response to PA challenge (Figure 11D). Taken together, these results suggest that the expression of XIAP directly affects the oxidative stress response and antioxidant levels in hepatic cells under PA stress.

Study approval and ethics statement

All study protocols associating with animals were approved by the Institutional Animal Care and Use Committee in Guangzhou University of Chinese Medicine and Chongqing Key Laboratory of Medicinal Resources in the Three Gorges Reservoir Region, School of Biological and Chemical Engineering, Chongqing University of Education. The animals received humane care according to the criteria outlined in the Guide for the Care and Use of Laboratory Animals prepared by the National Academy of Sciences and published by the National Institutes of Health in 1996. The methods used in this work also were in accordance with the Regulations of Experimental Animal Administration issued by the Ministry of Science and Technology of the People's Republic of China (http://www.most.gov.cn).

Animals and experiment design

Male mice (C57BL/6; wild type) at the age of 6-8 weeks, weighing 18–25 g, were purchased from the Beijing Vital River Laboratory Animal Technology Co., Ltd (Beijing, China). A total of 210 mice were used in this study. Prior to all experiments, the mice were allowed to adapt to the environment for 7 days. The mice were maintained in a constant temperature (controlled by GREE air conditioner), humidity and pathogen-free-controlled environment (25 °C ± 2 °C, 50% ± 3 %) with a standard 12 h light/12 h dark cycle, abundant food and water (pathogen-free) in their houses. After one-week acclimatization, mice with nonalcoholic fatty liver disease model were constructed by feeding with high fat diet (HFD) fodder (20% kcal protein, 60 kcal% fat and 20% kcal carbohydrate, Cat#: D12492;Research Diets, New Brunswick, NJ, USA) for 12 weeks until experimental mice were sacrificed for further study. In addition, mice that were fed a standard chow diet (SCD)(20% kcal protein, 10 kcal% fat and 70% kcal carbohydrate, Cat#: D12450H; Research Diets, New Brunswick, NJ, USA)as control group. To generate in vivo knockdown mice, XIAP siRNA in vivo transfection (ItsiXIAP) was performed by injecting with a XIAP siRNA (Generay Biotech Co.Ltd., Shanghai, China). A scrambled siRNA was used as a negative control (ItsiRNA). All mice were randomly divided into 4 groups: (1) ItsiRNA/SCD; (2) ItsiXIAP/SCD; (3) ItsiRNA/HFD and (4) ItsiXIAP/HFD. The XIAP siRNA or scrambled siRNA (50 ug) was diluted in 50 uL of Entranster^TM-in vivo RNA transfection reagent (Engreen Biosystem Co., Ltd., Beijing, China) and 10% glucose(Cat#: G6172, Macklin Inc., Shanghai, China) mixture according to the manufacturer’s instructions. A volume of 50 uL siRNA or siXIAP mix was continuously subjected to mice by tail vein every 10 days for a total of 12 weeks according to the reagent’s instructions. The generation of in vivo knockdown mice preparation and protocol was performed in accordance with our previous work [46]. The body weight of all mice was recorded every 4 weeks. After SCD or HFD feeding for 12 weeks, blood was sampled from removing the eyeball for the plasma analysis. Subsequently, the liver tissues were carefully and quickly removed for weight measurement, and/or in part dissected for RT-PCR, western blot, immunohistochemical assay or other biochemical analysis.

Biochemical analysis

For glucose tolerance test (GTT) and insulin tolerance tests (ITT) test, all mice associating with this regard were fasted for 8 h to confirm the correction of physiological response. For GTT test, mice were orally gavaged with glucose (2 g/kg body weight) (Cat#: D810588, Macklin Inc., Shanghai, China). Then, the concentration of blood glucose of tail venous blood at 0 min, 30 min, 60 min, 90 min and 120 min after glucose treatment was examined using commercial blood glucose test strips (ACCU-CHEK^®, Roche Diabetes Care GmbH, Shanghai, China). For insulin resistance detection, mice were treated with an intraperitoneal injection of insulin (1 U/kg body weight, Sigma Aldrich). The GTT and ITT analysis protocol were used in accordance with our previous study [46]. Subsequently, blood samples were collected from the tail vein at 0 min, 30 min, 60 min, 90 min and 120 min post-injection for measurement of glucose levels. Serum AST (glutamic-oxalacetic transaminase), ALT (glutamic-pyruvic transaminase), AKP (alkaline phosphatase), serum insulin and TG (triglyceride), TC (total cholesterol) in liver or serum were next detected using commercially-available detection kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). The serum endotoxin was also determined using mouse endotoxin, ET ELISA Kit (Nanjing SenBeiJia Biological Technology Co., Ltd., Nanjing, China). The glycosylated hemoglobin (HbA1c) of mice was further detected using commercial mouse HbA1c ELISA Kit (Nova Lifetech Inc.).

Cells culture and treatment

Primary hepatocytes were isolated from the liver of 6-8-week-oldmale mice using the collagenase perfusion method as previously described [25, 26]. In brief, after the liver tissues were digested with a perfusion of collagenase type IV solution (Gibco™, Thermo Fisher Scientific, USA), the liver was excised, minced and filtered using a 70-μm cell strainer (BD Falcon, USA). The resulting solution was centrifuged to harvest hepatocytes. Then, the primary hepatocytes were separated and purified via Percoll solutions (Sigma Aldrich). Next, the purified hepatocytes were cultured in collagen-coated plates in DMEM (Gibco Corporation, Gaithersburg, MD, USA) supplemented with 10% fetal bovine serum (Gibco Corporation, Gaithersburg, MD, USA) and 1% penicillin-streptomycin(Gibco Corporation, Gaithersburg, MD, USA). The normal mice cultured AML-12 cells were purchased from ATCC (American Type Culture Collection), and then cultured at 37 °C, 5% CO₂ in DMEM/F12 medium (Gibco Corporation, Gaithersburg, MD, USA) containing 10% fetal bovine serum, 1×10⁵ U/L streptomycin (Gibco Corporation, Gaithersburg, MD, USA). Then, confluent cultures were passaged by trypsinization (Hyclone) and maintained in 12-well, 24 well or 96-well plates (Corning, Shanghai, China) at a density of 2×10⁵/ml or 1×10⁴/ml for next 3 days. Subsequently, to calculate the inflammatory response and explore the activity of XIAP signaling, XIAP specific siRNA (5’-AATAGTGCCACGCAGTCTA-3’) and pUNO1 bearing the mouse XIAP plasmid (InvivoGen, HongKong, China), and corresponding negative control sequence or empty vector were subjected to cells using Lipofectamine 3000 Reagent (Invitrogen, Thermo Fisher Scientific, USA). For the effects detection of XIAP signaling, the cells were next treated with 250 μM palmitate (PA) (Sigma) for 24 h. Then, the cells were collected for the further study. The WST-8 assay [2-(2-Methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2 Htetrazolium Sodium Salt] was used to examine the 24 h siRNA and/or PA co-treated cells viability by using Cell Counting Kit 8 (CCK-8) (Solarbio, Beijing, China) according to the manufacturer instructions.

Oxidative stress analysis in vivo and in vitro

Assay kits for detection of malondialdehyde (MDA) (Cat#: A003-1), superoxide dismutase (SOD) (Cat#: A001-3), catalase (CAT) (Cat#: A007-1-1), and glutathione (GSH) (Cat#: A005) were obtained from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). Hydrogen peroxide (H₂O₂) (Cat#: S0038), O₂^- (Cat#: S0060), inducible nitric oxide synthase (iNOS) (Cat#: S0025) and total antioxidant capacity (TAC)(Cat#: S0116) levels were tested using corresponding kits from commercially available kits (Beyotime, Nanjing, China). Xanthine oxidase (XO) (Cat#: K710, Biovision, San Francisco, USA) and glutathione S-transferase (GST) (Cat#: ABIN773241, Abnova) were also introduced into this regard to determine the corresponding levels of indicators. In addition, 2’,7’-Dichlorofluorescein diacetate (DCFH-DA) stain for ROS measurement reagents (Cat#: S0033) was purchased from Beyotime (Nanjing, China).

Real-time quantitative PCR and immunoblotting assay

In this regard, TRIzol product (Cat#:15596026, Invitrogen, Thermo Fisher Scientific, USA) was used to harvest the total RNA samples from cells or liver tissues. Specifically, 1 μg of total RNA extraction was reverse transcribed using the M-MLV-RT system (Invitrogen^TM, Shanghai, China). The program was performed at 42 °C for 1 h and terminated by deactivation of the enzyme at 70 °C for 10 min. PCR were conducted using SYBR Green (Bio-Rad) in BIO-RAD CFX96 touch q-PCR detection systems (Bio-Rad). Pre-denatured products for amplification was performed at 94 °C for 55s; followed by 45 cycles at 95 °C for 30s, 57 °C for 30s and 73 °C for 30s; followed by 95 °C for 10s, 65 °C for 45s, and 40 °C for 60s. In brief, fold induction values were calculated according the 2^-ΔΔCt expression, where ^ΔCt represents the differences in cycle threshold number between the target gene and GAPDH, and ^ΔΔCt represents the relative change in the differences between control and treatment groups. The primer sequences for RT-PCR were produced by Sangon Biotech (Shanghai) Co., Ltd., and indicated in Table 2. For immunoblotting analysis, cells or tissues were homogenized into 10% (wt/vol) hypotonic lysis buffer (pH 8.0, 1mM EDTA, 50 μg/ml aprotinin, 25mM Tris-HCl, 5 μg/ml leupeptin, 1mM Pefabloc SC, 4mM benzamidine, 5 μg/ml soybean trypsin inhibitor) to yield a homogenate. Next, the final liquid supernatants were harvested by centrifugation at 13500 rpm for 30 min. Protein concentration was measured using Pierce^TM Rapid Gold BCA Protein Assay Kit (Thermo, USA) with bovine serum albumin as a standard. The total protein extraction will be performed for western blot analysis. Equal amounts of total protein of tissues were subjected to 10% or 12% SDS-PAGE and then transferred to a 0.45 μM PVDF membrane (Millipore Company, USA) followed by immunoblotting using the primary antibodies. The primary antibodies used in this study included anti-GAPDH (K106389P, dilution 1:5000, Solarbio, China), anti-XIAP (ab21278, dilution 1:1500, Abcam), anti-ACC (3676, dilution 1:1000, Cell Signaling Technology), anti-p-ACC (11818, dilution 1:1000, Cell Signaling Technology), anti-Fas (ab82419, dilution 1:1250, Abcam), anti-SCD1 (2794, dilution 1:1500, Cell Signaling Technology), anti-Caspase-8 (ab25901, dilution 1:1000, Abcam), anti-NLRP3 (ab214185, dilution 1:1000, Abcam), anti-ASC (sc-365611, dilution 1:300, Santa Cruz Biotechnology), anti-Caspase-1 (ab1872, dilution 1:1000, Abcam), anti-Nrf2 (ab62352, dilution 1:1000, Abcam), anti-HO-1 (ab13248, dilution 1:1000, Abcam), anti-SOD1 (ab13498, dilution 1:1000, Abcam), anti-NQO-1 (ab28947, dilution 1:1000, Abcam), anti-IL-18 (ab71495, dilution 1:1000, Abcam) and anti-IL-1β (#31202, dilution 1:1000, Cell Signaling Technology). Then, the membranes were blocked with 5% skim milk (Difco^TM Skim Milk, BD, USA)in 1×TBS buffer (Cat#: T1080, Solarbio, Beijing, China) containing 0.1% Tween-20 (1247ML100, BioFROXX, Germany) (TBST) for 1 h and incubated with the primary antibodies overnight (4°C). Membranes were rinsed in TBST and incubated with HRP-conjugated anti-rabbit or anti-mouse secondary antibodies (Abcam) for 2 h at room temperature. Bands were visualized by Pierce^TM ECL Western Blotting Substrate (Thermo Scientific, China) and exposed to Kodak (Eastman Kodak Company, USA) X-ray film. Corresponding protein expression will be determined as grey value (Version 1.4.2b, Mac OS X, ImageJ, National Institutes of Health, USA) and standardized to housekeeping genes (GAPDH) and expressed as a fold of control.

Histological assay

Fixed white adipose tissue (WAT) and liver sections were embedded in paraffin; sections were stained to visualize the pattern of lipid accumulation and the inflammatory status of tissues. Four fields from each WAT section were analyzed to obtain the mean cell area [46, 47]. Briefly, the mice liver and white adipose tissues were fixed with 4% paraformaldehyde (Cat#: P1110, Solarbio, China), embedded in paraffin (Cat#: P100928, Aladdin, Shanghai, China), and sectioned transversely. Thin sections were stained with hematoxylin and eosin (H&E) (Hematoxylin and Eosin Staining Kit, Yeasen, Shanghai, China) and sirius red staining according to a standard histopathological processes. All sections were detected by 3 histologists without knowledge of the treatment procedure. The criteria regarding the evaluation of NAFLD associated histopathological changes were followed to the Kleiner or the Brunt criteria [48]. To visualize lipids accumulation, liver tissues were frozen in Tissue-Tek OCT (Tissue-Tek, Sakura Finetek, USA) and sections were then stained with Oil Red O Stain Kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) for 10 min. After being rinsed with 60% isopropyl alcohol (Cat#: I811925, AR, Macklin Inc., Shanghai, China), the tissue sections were re-stained with hematoxylin. For immunohistochemistry, embedded sections were deparaffinized before administration with primary antibodies at 4 °C overnight. The tissues were subjected to immunohistochemical (IHC) staining for the measurement of XIAP (ab21278, dilution 1:300, Abcam), Caspase-8 (ab25901, dilution 1:250, Abcam), Nrf2 (ab62352, dilution 1:250, Abcam), HO-1 (ab13248, dilution 1:250, Abcam) or SOD1 (ab13498, dilution 1:250, Abcam) expression. All the histological procedure was performed in accordance with the standard procedures as indicated in reagent specifications. All the images were captured using optical microscope (Olympus, Japan). Four fields from each WAT section were analyzed to obtain the mean cell area. The evaluation of immunohistochemical (IHC) and adipocyte size were determined using Image J software derived from National Institutes of Health (NIH), USA. The number of positive cells per high power field (HPF) for quantification of IHC staining was counted using Image J. Antibody staining was assessed and scored using the “Quick Score method” [49]. Briefly, the proportion of positive cells was estimated and scored on a scale of 1 to 6; 0–4% = 1, 5–19% = 2, 20–39% =3, 40–59% = 4, 60–79% = 5, and 80–100% = 6.

Statistical analysis

Results were expressed as mean ± standard error of the mean (SEM) from three independent experiments. Comparisons between groups were analyzed by one-way ANOVA followed by Dunnett's multiple comparisons test. GraphPad Prism Software (version 7.0 for Mac OS X Snow Leopard; Graph Pad Software, Inc., San Diego, CA) was used for the analysis. A p-value less than 0.05 will be considered significant.