The present study was conducted according to the Experimental Animal Regulations of the People's Republic of China and the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and the Guide for the Care and Use of Laboratory Animals of the USA (22), ensuring that all animals received humane care. The present study was approved by the Ethics Committee of Wuhan University (Wuhan, China).

A total of 18 adult male Sprague-Dawley rats (age, 8 weeks; body weight, 250±10 g) were purchased from the Experimental Animal Culture Center of Hubei Centers for Disease Control (Hubei, China), and were maintained in the Animal Experimental Center of Zhongnan Hospital of Wuhan University (Wuhan, China). The rats were fed standard chow and water, and housed under standard experimental conditions (temperature: 20–25°C, humidity: 50–70%) under a 12 h light/dark cycle. To simulate DCD liver transplantation, 30 min of warm ischemia was conducted in livers (n=12, CS and HMP groups) in situ, followed by a rewarming period of 15 min prior to being connected to the isolated perfused rat liver device (IPRL). To investigate the protective effects of HMP on the DCD livers, the extent of graft injury in healthy livers (non-DCD, no IRI) was compared with DCD livers preserved by either CS or HMP (described below). Following graft preservation, to simulate the period of rewarming during re-implantation, all the livers were left untouched on a petri dish at room temperature for ~15 min prior to reperfusion (23).

Then, in order to analyze reperfusion injury, livers were re-perfused in the isolated perfusion rat liver model for 1 h following graft preservation within the CS and HMP groups, and the same is true for the control group. For this purpose, the following experimental groups were selected: i) Control group (n=6), livers without warm ischemia and subsequent underwent 1 h reperfusion prior to sample collection; ii) CS group (n=6), DCD livers that underwent CS for 3 h, followed by 1 h reperfusion in vitro and the iii) HMP group (n=6), in which DCD livers were connected to the HMP system. HMP was performed via the portal vein for 3 h, followed by 1 h reperfusion in vitro.

Rats were anesthetized via an intraperitoneal injection of 1% sodium pentobarbital (30 mg/kg; Sinopharm Chemical Reagent Co., Ltd., Shanghai, China). A midline incision was conducted to provide entry into the abdominal cavity. The liver was carefully separated from the attached round ligament. Subsequently, the common bile duct was cannulated using a guided epidural tube (Jiangsu Changfeng Medical Industry, Co., Ltd. (Jiangsu, China) to collect bile during reperfusion. In the experimental groups, DCD was induced by hypoxia via an incision of the diaphragm without portal clamping prior to heparinization, as described below. The onset of in situ-warm ischemia was determined to be the point of the cardiac arrest, which was maintained for 30 min at 29±1.5°C. The control group did not experience the warm ischemia and the other operations were the same as the experimental groups. Subsequently, the hepatic artery was ligated and then 2 ml saline containing 100 IU heparin (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) was injected via the right iliac vein. A total of 20 ml (4°C) histidine-tryptophan-ketoglutarate (HTK; Dr. Franz Köhler Chemie GmbH, Bensheim, Germany) solution was used to rinse the liver via the portal vein, which was cannulated using a self-made polyethylene (PE) catheter (outer diameter, 2.1 mm, inner diameter, 1.8 mm) (24). Finally, the liver was harvested and a PE catheter (internal diameter, 3 mm) was used to cannulate the suprahepatic inferior vena cava for collection of hepatic effluent.

Following the modeling procedure, the liver was stored in the HTK solution at 0–4°C for 3 h to maintain a static state without any treatment. Following preservation, the CS group was simulated the period of rewarming for a 15 min and then connected to the IPRL, which is the reperfusion device used to assessment of IRI severity against livers (described below).

HMP and IPRL was performed as previously described (25,26) with certain modifications. The HTK solution served as a perfusate for HMP; the devices employed for perfusion were maintained in an ice water solution (0–4°C). Following collection of the livers, the samples from the HMP group were connected to the perfusion device and perfused via the portal vein. The perfusion flow was maintained at 0.23 ml/min/g (23). The perfusate was oxygenated with air and recirculated for 3 h (perfusate volume, 100 ml). Following preservation, the caudate lobe of the rat livers of the CS and HMP groups was ligated and harvested to obtain samples prior to reperfusion. The remaining rat liver tissues in the three groups were then weighed and connected via the portal vein to the IPRL system for reperfusion. IPRL is an in vitro system that simulates physiological and pathophysiological conditions of reperfusion in liver transplantation and is often used as a tool for the assessment of IRI severity against livers (27).

The same perfusion device was used for HMP, as well as for the 1 h reperfusion period (n=12, CS and HMP groups). A detailed description of the HMP and IPRL system is given in as described in a previous study (26). Krebs-Henseleit buffer (Macgene™ M&C Genetechnolgy, Beijing, China) with 4% dextran was used as a perfusate for reperfusion (28). The temperature of the perfusate was maintained at 36–37°C during reperfusion and oxygenated to maintain PO₂ >500 mmHg under the effect of mixed gas (95% oxygen and 5% carbon dioxide). The flow of portal venous perfusion was maintained at 3 ml/min/g (29) and recirculated for 1 h (perfusate volume, 250 ml).

During reperfusion, which was performed for 1 h, intrahepatic resistance (IHR) was recorded by the portal pressure and portal flow and the perfusate was collected per 15 min. Intrahepatic resistance was calculated according to the following formula: Intrahepatic resistance (mmHg/ml/min/g liver)=portal pressure (10.3 mm Hg)/portal flow (ml/min/g liver) (30). Hepatic effluent was obtained from the perfusion fluid via the PE catheter every 15 min. Samples were centrifuged at 14,000 × g and 4°C for 5–10 min and the supernatant was collected and stored at −80°C prior to the determination of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) activities. The enzyme activities were assessed using ALT (cat. no. C009-2), AST (cat. no. C010-2) and LDH assay kits (cat. no. A020-2; all Nanjing Jiancheng Bioengineering Institute), according to the manufacturer's protocol. As the density of bile was equal to the water (23,31), bile production was measured at 60 min by weighing the guided epidural tube in which bile was collected from the common bile duct. Then, the bile flow was gravimetrically estimated and expressed as µl/h/g liver (26).

Frozen liver samples were lysed with 0.05 M Tris. HCl (Beijing Biotopped Science & Technology Co., Ltd., Being, China) extraction buffer on ice. The cell lysates were centrifuged at 4°C 12,000 × g for 10 min. The resulting cell lysates were used to assess the SOD activity and MDA content. To measure total SOD activity, a SOD kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) was employed according to the manufacturer's protocols. MDA content was assessed with an MDA assay kit (Nanjing Jiancheng Bioengineering Institute) based on the products of membrane lipid peroxidation, which are important indicators of oxidative damage during hepatic IRI.

The hepatic concentration of ATP served as an indicator of the energy status of grafts following 1 h of ex situ reperfusion. The ATP content of the tissue was measured using an ATP assay kit (Nanjing Jiancheng Bioengineering Institute) according to the manufacturer's protocols. ATP levels were expressed as µmol/g protein.

Following reperfusion, about 0.25 g liver samples were fixed with 10% buffered formalin for 24 h at room temperature (pH=7.2; cat. no. G2161; Beijing Solarbio Science & Technology, Co., Ltd., Beijing, China), embedded in paraffin and cut into 5-µm sections for histological analysis via hematoxylin-eosin staining (H&E; hematoxylin staining for 5–15 min and eosin staining for 1–3 min; all performed at room temperature). Sections were analyzed under a confocal microscope (magnification, ×200; Nikon A1R/A1; Nikon Corporation, Tokyo, Japan) and images were obtained. A total of 6 fields of view per section were randomly selected for the assessment of liver damage. Numerical assessment of liver damage was conducted according to the histological criteria for assessment of liver damage (32).

Following reperfusion, a portion of the livers were fixed with 10% buffered formalin for 24 h at room temperature (pH=7.2; cat. no. G2161; Beijing Solarbio Science & Technology, Co., Ltd., Beijing, China), and then embedded in paraffin, sliced, dewaxed (Dewaxing was routinely performed at 60°C for 20 min, and immediately xylene 1–3 for 10 min, respectively. However certain sections prepared on the day could be treated at 60°C for 3–4 h), and hydrated conventionally using an ethanol gradient (from high to low). The tissues were cut into 4-µm sections and mounted on glass slides. After 30 min of blocking at room temperature with 5% bovine serum albumin (Beijing Solarbio Technology Co., Ltd., Beijing, China), the samples were incubated with rabbit anti-Nrf2 antibody (1:1,000; cat. no. 16396-1-AP; Wuhan Sanying Biotechnology, Wuhan, China) overnight at 4°C. The samples were then incubated for 1 h at room temperature with a horseradish peroxidase-conjugated goat anti-rabbit IgG (1:3,000; cat. no. GB23303; Wuhan Goodbio Technology Co., Ltd., Wuhan, China). Subsequently, the samples were incubated with 3,3′-diaminobenzidine chromogen (DAB; Maixin-Bio Ltd.) at room temperature for 5 min, and then blocked on a coverslip. Any brown and yellow staining was considered to indicate positive Nrf2 expression, as visualized under a light microscope (magnification, ×200, Leica DM2000; Leica Microsystems GmbH, Wetzlar, Germany). A total of 5 visual fields were randomly selected for analysis.

Total RNA was isolated from the liver specimens using TRIzol^® reagent (Thermo Fisher Scientific Inc., Waltham, MA, USA) according to the manufacturer's protocols. RNA was reverse transcribed into cDNA using the Easy Script One-Step gDNA Removal and cDNA Synthesis Super Mix (Beijing TransGen Biotech, Co., Ltd., Beijing, China). Total RNA (1 µg), Random primer (0.1 µg/µl), 2X ES Reaction Mix (10 µl), RI Enzyme Mix (1 µl), gDNA Remover (1 µl) were employed; the solution was made up to 20 µl with water (RNase-free). RT reactions were performed under the following conditions: 42°C for 1 h and 75°C for 5 min. The primers were synthesized by Shanghai ShineGene Molecular Biotech, Inc. (Shanghai, China; Table I). qPCR analysis was performed with the SYBR^® Select Master Mix (Applied Biosystems; Thermo Fisher Scientific, Inc.) in a StepOnePlus Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.) to determine the mRNA expression levels of Nrf2, HO-1, NQO1, GST-1, GCL and β-actin. The thermocycling conditions were as follows: 50°C for 2 min, 95°C for 10 min, followed by 40 cycles of 95°C for 10 sec and 60°C for 30 sec. Expression levels were normalized to β-actin, which was measured on the same plate; the differences were calculated via the 2^−ΔΔCq method (33,34).

Apoptosis was determined via a TUNEL assay (One-Step TUNEL Apoptosis assay kit, Beyotime Institute of Biotechnology, Haimen, China) according to the manufacturer's protocols. Briefly, 4-µm thick paraffin sections, which contained the liver tissues were deparaffinized and hydrated, then treated with proteinase K for 20 min and subsequently incubated with a mixture of fluorescent labeling solution and TdT enzyme at 37°C for 1 h in a humidified atmosphere. The samples were washed in 1XPBS and mounted in mounting media containing DAPI at room temperature for 10 min without the light. Blue ray was chosen as the exciting light, the wavelength is 420–485 nm. GFP was excited and emitted 515 nm green fluorescence. Liver cells, which expressed GFP emitted green fluorescent. The total hepatocytes and TUNEL-positive cells were detected in 4–5 randomly selected fields (magnification, ×200) for each liver section using a fluorescence microscope (Olympus X71; Olympus Corporation, Tokyo, Japan). The apoptotic rate was calculated according to the formula: Number of TUNEL positive cells/number total cells × % (35,36). The TUNEL positive cells were calculated by three different authors.

Prior to and following reperfusion, all collected liver tissues were rapidly dissected within 5 min and stored at −80°C for cryopreservation. In order to extract the total proteins, the liver tissue was thawed and homogenized in radioimmunoprecipitation assay buffer containing a protease inhibitor (cat. no. G2002; Wuhan Servicebio Co., Ltd., Wuhan, China) and then centrifuged at 20,000 × g for 10 min at 4°C. Following collection of the supernatants and measuring the total protein concentration, 30 mg protein, which was calculated by the bicinchoninic acid kit (cat. no. G2026; Wuhan Servicebio Co., Ltd., Wuhan, China) was separated by 10% SDS-PAGE and transferred to polyvinylidene difluoride membranes (EMD Millipore, Billerica, MA, USA). The membranes were blocked in 5% skimmed milk for 1 h at room temperature. The blots were then incubated in 4°C for 12 h with the following antibodies: Rabbit anti-Nrf2 antibody (1:1,000; cat. no. 16396-1-AP; Wuhan Sanying Biotechnology), rabbit anti-HO-1 antibody (1:2,000; cat. no. 27282-1-AP Wuhan Sanying Biotechnology), rabbit anti-NQO1 antibody (1:200; cat. no. bs-23407R Beijing Biosynthesis Biotechnology Co., Ltd., Beijing, China), rabbit anti-Toll-like receptor 4 (TLR4) antibody (cat. no. 19811-1-AP, 1:1,000; Wuhan Sanying Biotechnology), rabbit anti-B-cell lymphoma 2 (Bcl-2) antibody (1:1,000; cat. no. 10435-1-AP; Wuhan Sanying Biotechnology), rabbit anti-Bcl-2-associated X (Bax) antibody (1:1,000; cat. no. 60267-1-Ig; Wuhan Sanying Biotechnology) and anti-β-actin antibody (1:3,000; ProteinTech Group, Inc., Chicago, IL, USA). Following incubation for 1 h at room temperature with horseradish peroxidase-conjugated goat anti-rabbit IgG (1:3,000; cat. no. GB23303; Wuhan Goodbio Technology Co. Ltd., Wuhan, China), the proteins were detected using an enhanced chemiluminescence reagent (cat. no. G2020; Wuhan Servicebio Co., Ltd., Wuhan, China) followed by exposure to X-ray film. Quantification of protein bands was performed using ImageJ v1.42q software (National Institutes of Health, Bethesda, MA, USA).

All data were analyzed using SPSS 16.0 statistical software for Windows (SPSS, Inc., Chicago, IL, USA) by one-way analysis of variance with Tukey's post-hoc test. All results are presented as the mean ± standard deviation (n=6 for each experiment). P<0.05 was considered to indicate a statistically significant difference.

The activities of ALT and AST enzymes in the perfusate were used to assess the severity of IRI to DCD livers. In the present study, the levels of liver enzymes ALT and AST in the perfusate of the CS group increased significantly compared with in the control group (P<0.01; Fig. 1A and B, respectively). The HMP group exhibited significantly decreased levels of ALT and AST compared with in the CS group (P<0.01; Fig. 1A and B). Additionally, the intrahepatic resistance of livers in the control group was low during reperfusion; however, in the CS group, it was significantly higher compared with in the HMP group at each analyzed time point (P<0.01; Fig. 1C). In addition, bile production in the CS group was significantly lower compared with in the control group (P<0.01; Fig. 1D). Furthermore, bile production decreased in the HMP group; however, no significant difference was observed compared with in the CS group (P>0.05). Visual and numerical analyses of the histological images indicated that there were few observable abnormalities in the livers of the control group (Fig. 2). Conversely, in the CS group, histological analysis revealed significant congestion of the hepatic sinusoid, vacuolar degeneration and necrosis (P<0.01; Fig. 2A-C). Additionally, significantly attenuated liver damage was observed in the HMP group compared with in the CS group (P<0.05; Fig. 2A-C).

The present study analyzed the typical biochemical markers of oxidative stress to further investigate the effects of HMP on the IRI livers. Oxidative stress has been considered to be an important factor leading to IRI. The SOD activities and expression levels of MDA of the tissues were presented in Fig. 3A and B. Compared with the control group, SOD activity in the CS group was significantly decreased (P<0.01), and the expression levels of MDA were significantly increased (P<0.01). The results of the present study indicated that HMP treatment may significantly decrease the MDA expression levels (P<0.01) and significantly increase the SOD activity (P<0.01); opposite trends to those observed in the CS group. These results suggested that HMP may attenuate oxidative stress. In addition, the ATP content in the liver was also measured to analyze liver function. As presented in Fig. 3C, within the CS group, the ATP content was significantly lower compared with the control group (P<0.01); however, HMP treatment significantly increased the ATP content compared with in the CS group. (P<0.01).

Oxidative stress can also induce hepatic inflammation and apoptosis (37,38). Therefore, the present study investigated the expression levels apoptosis-associated proteins in the liver (Fig. 3D). The ratio of Bcl-2 to Bax is also important in determining susceptibility of cells to apoptosis (39). The overexpression of Bax revealed that apoptosis was promoted in response to a death-inducing signal, which suggested its role as an apoptosis agonist (40). Bcl-2 overexpression has been associated with heterodimerization with Bax and reduced levels of apoptosis (41). In addition, TLR4 has been reported to be an important marker of inflammation (42). As presented in Fig. 3E, the ratio of Bcl-2 to Bax was significantly downregulated in the CS group and significantly upregulated in the HMP group (P<0.01). The protein expression levels of TLR4 were significantly increased in the CS group and significantly decreased in the HMP group (P<0.01; Fig. 3F). Furthermore, the present study revealed that, compared with the control group, the rate of apoptosis in the CS group was significantly increased (P<0.01; Fig. 4); however, treatment with HMP was associated with significantly reduced rates of apoptosis.

Nrf2 serves an important role in the main defense mechanisms induced by cellular oxidative stress (10,11). Therefore, the present study investigated whether HMP conferred protection against IRI to rat DCD livers via alterations in Nrf2 expression. The results reveled that compared with the control group, the expression levels of Nrf2 in the CS group were significantly decreased (P<0.05); however, the HMP group exhibited a significant increase compared with in the CS group (P<0.01; Fig. 5A and B). In addition to the expression levels of Nrf2 protein, NQO1 protein expression levels in the CS and HMP groups demonstrated opposing expression patterns (P<0.05; Fig. 5A and B). Notably, despite the significant increase in HO-1 protein expression levels in the CS group compared with the control, the HMP group exhibited a significant decrease compared with the CS group (P<0.01; Fig. 5A and B).

Reperfusion injury due to toxic reactive oxygen species generated upon reintroduction of blood flow and oxygen supply to ischemic tissues is the main cause of DCD liver injury (43). Therefore, the expression levels of proteins in the presence or absence of reperfusion were investigated in the present study to determine whether HMP may activate the Nrf2-ARE signaling pathway.

The results of the present study revealed that the expression levels of Nrf2, NQO-1 and HO-1 were all been significantly upregulated during reperfusion in both the CS and HMP groups (P<0.05; Fig. 5C and D). In addition, the induction of Nrf2 (0.392±0.137 vs. 0.912±0.122; P<0.0001) and NQO-1 (0.342±0.057 vs. 1.076±0.102; P<0.0001) expression in the presence of IRI was significantly higher within the HMP group compared with the induction of Nrf2 (0.300±0.079 vs. 0.529±0.075; P=0.007) and NQO-1 (0.287±0.104 vs. 0.575±0.078; P<0.0001) expression in the CS group. All this suggested that the effects of HMP on oxidative stress may occur via activation of the Nrf2-ARE signaling pathway (P<0.05; Fig. 5C and D).

Additionally, the mRNA expression levels of Nrf2, HO-1, NQO1, GST-1, and GCL in the liver were investigated via RT-qPCR (Fig. 6). The results of the present study revealed that except for HO-1, the mRNA expression levels of the other ARE-regulated genes were significantly enhanced in the HMP group compared with the CS group (P<0.05; Fig. 6). This suggested that additional regulatory elements may be involved in the activation of HO-1 in the CS group; however, the activation of Nrf2 and other ARE-regulated genes, as well as the inhibition of oxidative stress markers, including MDA in the HMP group, suggested that HMP may be responsible for the increase in antioxidative ability of cells by activating the Nrf2-ARE signaling pathway.

Furthermore, previous studies have confirmed that Nrf2 may be activated in endothelial cells by steady laminar flow (19,20). Thus, immunohistochemistry was conducted to assess whether steady laminar flow provided by HMP may induce the expression levels of Nrf2 (Fig. 7). The results revealed that within the HMP group, Nrf2 expression was significantly increased in the periportal regions of livers when compared with the CS group (P<0.01), This suggested that Nrf2 in hepatocytes and liver sinusoidal endothelial cells around periportal regions may be induced by steady laminar flow.