Male Sprague-Dawley rats (n=41; 6 weeks old), weighing 160-200 g, were purchased from Sino-British SIPPR/BK Lab Animal Ltd. (Shanghai, China). The animals were housed in a pathogen-free environment with standard conditions of temperature (27±2°C), humidity (50±5% RH) and a 12-h light/dark cycle. All of the animals were provided with 24-h free access to food and water. All procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and was approved (approval no. 00100097) by the Medical Ethics Committee of Naval Medical University (Shanghai, China).

ADSCs were isolated from the inguinal fat. Briefly, 5 rats were euthanized by intraperitoneal injection of sodium pentobarbital (100 mg/kg). Following confirmation of death by lack of pulse, breathing and corneal reflex, the abdomen and back of rats were disinfected using 75% alcohol and placed under an aseptic environment. The skin from both sides of the abdomen in the groin region was removed using scissors to expose fat pads. Fat pads were clamped using an Allis tissue clamp and removed using ophthalmic scissors. A total of 4-5 ml of fat tissue from each rat was obtained under sterilized conditions, and was digested with collagenase type I (0.25%) for 50 min at 37°C with shaking using a laboratory shaker. Following digestion, high sugar DMEM containing 10% FBS (both from Life Technologies; Thermo Fisher Scientific, Inc.) and standard concentrations of penicillin (10 U/ml) and streptomycin (100 µg/ml; both from Shanghai Yeasen Biotechnology Co., Ltd.) was added to the fat mixture followed by filtration through a cell strainer. Subsequently, following appropriate dilution, the supernatant was centrifuged at 1,000 × g for 10 min at 25°C to collect cells that were further cultured in 10 cm² cell culture plates. Following the third passage of cells, surface markers CD73, CD90, CD105, CD34, CD45 and CD11b were identified using marker identification kit Oricell^®; cat. no. RAXMX-09011; [Saiye (Guangzhou) Biotechnology Co., Ltd.] according to the manufacturer's protocol, and a flow cytometer (BD LSR II; BD Biosciences).

The third generation of ADSCs was cultured at 37°C and 5% CO₂ in a petri dish with complete medium, and after 80% confluency was reached, the medium was replaced with serum-free medium that was collected after 48 h. The collected serum-free medium was centrifuged at 400 × g for 15 min at 4°C to remove floating cells. The supernatant was collected and centrifuged at 10,000 × g for 15 min at 4°C to remove cellular debris. Exosomes were collected through ultracentrifugation at 100,000 × g for 13 h at 4°C. Subsequently, the exosomes were gently washed with PBS and collected through centrifugation at 14,000 × g for 10 min at 4°C. Collected exosomes were observed under transmission electron microscopy (TEM) and surface markers including TSG101 (1:1,000; product code ab83), CD9 (1:1,000; product code ab92726) and CD63 (1:1,000; product code ab193349; all from Abcam) were identified using western blot analysis according to the manufacturer's protocol.

A total of 36 SD rats were divided into three groups: Sham-operated rats (Sham), PBS-treated rats for I/R injury (PBS + I/R) and exosome-treated rats for I/R injury (ADSCs-exo + I/R). Rats were fasted 12 h prior to the experiment. For the surgical procedure, rats were anesthetized by intraperitoneal (i.p.) injection of 1% pentobarbital sodium (45 mg/kg of body weight). Following midline laparotomy, the portal vein and portal artery were clamped to induce 70% hepatic ischemia. A total of 50 µl exosomes (30 µg) were administered via the portal vein before induction of ischemia in rats included in the ADSCs-exo + I/R group while a similar volume of PBS was administered in rats in the PBS + I/R group (Fig. 1). Ischemia was maintained for 60 min followed by reperfusion. During experimental procedures, the rats were placed on a 37°C warmed surface. Following 2 and 6 h of reperfusion, rats (n=6, for each time-point) were sacrificed using 100 mg/kg intraperitoneal injection of pentobarbital sodium and liver and serum samples were collected for further examination. The death of the rats was confirmed by lack of pulse, breathing and corneal reflex. For the sham group, the same procedure was followed without clamping of hepatic vessels.

Collected blood was centrifuged at 1,000 × g for 15 min at 4°C to obtain serum that was analyzed for quantification of alanine aminotransferase (ALT), aspartate aminotransferase (AST) and lactate dehydrogenase (LDH) using commercial kits (cat. no. ALT000A, AST000S and LDH000S for ALT, AST and LDH, respectively; from Beijing Autobio Co. Ltd.) following the manufacturer's protocol on Auto-Analyzer (TBA-120FR; TOSHIBA).

The levels of malondialdehyde (MDA; cat. no. MBS727531; MyBioSource, Inc.), superoxide dismutase (SOD; cat. no. MBS266897; MyBioSource, Inc.) and ROS (cat. no. LS-F9759; LifeSpan BioSciences, Inc.) were determined in liver homogenates using a commercial testing kits according to the manufacturer's protocol.

The expression levels of interleukin (IL)-1β (product code ab255730; Abcam) and TNF-α (product code ab236712; Abcam) were detected in rat serum using respective ELISA kits.

The amount of cAMP in serum was quantified using cAMP Complete ELISA kit (product code ab133051; Abcam). Furthermore, prostaglandin E2 (PGE2) was also quantified in serum using PGE2 ELISA kit (product code ab133021; Abcam).

Following 2 and 6 h of reperfusion, collected liver tissue samples were fixed in 4% formaldehyde solution for 12 h at 4°C. The tissue was dehydrated and embedded in paraffin (23). Embedded tissues were sliced into 5-µm thick sections. For necrosis evaluation, hematoxylin and eosin (H&E) staining was performed according to the manufacturer's protocol (Beyotime Institute of Biotechnology). Briefly, hematoxylin staining solution was added on specimens for 5 min at room temperature followed by washing with tap water for 1-2 min. The slides were dipped in 1% hydrochloric acid in 70% ethanol for 10 sec followed by washing with tap water for 10 min. Eosin staining solution was added on specimens for 30 sec and directly immersed in 95% ethanol followed by absolute ethanol for 10 sec. Necrotic areas were quantified using ImageJ software (version 1.51j8; National Institutes of Health). TdT-mediated dUTP nick-end labeling (TUNEL) staining was also performed to observe apoptotic cells in liver samples using a commercial kit (cat. no. 11684817910; Roche Diagnostics), following the manufacturer's protocol. Briefly, TUNEL reaction mixture (50 µl) was added onto specimens and placed at 37°C for 1 h in a dark humidified box. After washing with PBS (three times for 3 min each), 50 µl converter-peroxidase was added to specimens for 30 min at 37°C for 1 h in a dark humidified box followed by DAB staining for 30 min at 25°C. Stained samples were observed using light microscope (Olympus CX41; Olympus Corporation). The percentage apoptotic cells was counted using the following formula: (Number of apoptotic cells/total number of cells) ×100.

For immunohistochemical analysis, liver samples were fixed using 4% paraformaldehyde (in normal saline) for 10 min at room temperature, and further sliced into 30-µm-thick sections. Following washing with PBS, permeabilization was performed using washing buffer with 0.3% H₂O₂ and 0.5% Triton X-100. Sections were incubated with respective primary antibodies including ERK1/2 (1:1,000; product no. 4695), p-ERK1/2 (1:2,000; product no. 4370), GSK-3β (1:1,000; product no. 5676) and p-GSK-3β (1:1,000; product no. 9322; all from Cell Signaling Technology, Inc.) at 4°C overnight after blocking non-specific binding sites with 3% BSA (Shanghai Yeasen Biotechnology Co., Ltd.) for 1 h at room temperature. Samples were washed and incubated in secondary antibodies (HRP-goat anti-mouse IgG, cat. no. GB23301 for ERK1/2 and GSK3β and HRP-goat anti-rabbit IgG, cat. no. GB23303 for p-ERK1/2 and p-GSK3β; both from Servicebio) at 1:200 dilution for 1 h at room temperature followed by staining with DAB (0.05%) for 10 min at room temperature. Images were analyzed to calculate the positive expression of phosphorylation of extracellular receptor kinase (ERK) and glycogen synthase kinase-3b (GSK-3β). Positive area <10% was graded as (-), positive area between 10-30% was graded as (+) while positive area >30% was graded as (++).

Following 6 h of reperfusion, electron microscopic analysis of liver tissue was performed to observe subcellular modification. Liver samples were first fixed in a 1% osmic acid fixative solution for 3 h at 4°C. Fixed samples were dehydrated with graded ethanol solution followed by ethanol/acetone solutions and 90% acetone. Following washing with PBS, dehydrated samples were embedded in acetone: OCT compound (2:1) for 4 h. Tissues were finally embedded in 100% embedding medium and dried (at 37°C overnight, 45°C for 12 h and 60°C for 24 h). Dried samples were sliced into 50-nm sections and observed under a transmission electron microscope (Crossbeam 550 FE; Carl Zeiss AG) following staining with 2% uranyl acetate and lead citrate at room temperature for 10 min. Embedding medium and staining solutions were provided by Microscopy Core Facility of Westlake University.

Protein expression levels of ERK1/2, phosphorylated (p)-ERK1/2, GSK-3β, p-GSK-3β, Bcl-2 and Bax were evaluated using western blotting as per standard protocol. The liver lysate was obtained by homogenizing the liver tissue in RIPA lysis buffer (cat. no. P0013B; Beyotime Institute of Biotechnology) followed by centrifugation at 12,000 × g at 4°C for 15 min. A total of 40 µg of protein sample, following quantification using a BCA kit, were separated using 10% SDS-PAGE and transferred to PVDF membranes. Non-specific binding of antibodies was blocked by incubation in 5% BSA for 32 h at 4°C. Blots were further incubated at 4°C overnight with respective primary antibodies: ERK1/2 (1:1,000; product no. 4695), p-ERK1/2 (1:2,000; product no. 4370), GSK-3β (1:1,000; product no. 5676), p-GSK-3β (1:1,000; product no. 9322; all from Cell Signaling Technology, Inc.), Bcl-2 mouse monoclonal antibody (1:1,000; 60178-1-Ig; ProteinTech Group, Inc.), Bax rabbit monoclonal antibody (1:1,000; cat. no. 60267-1-Ig; ProteinTech Group, Inc.) and GAPDH (1:1,000; product no. 5174; Cell Signaling Technology, Inc.) followed by incubation with goat anti-rabbit IgG HRP-linked secondary antibody (1:2,000; cat. no. 7074; Cell Signaling Technology, Inc.) for 1.5 h at 25°C. Blots were washed with TBS with 0.05% Tween-20 (TBST) thrice and a chemiluminescence reaction was achieved with an ECL reagent (cat. no. 34076; Thermo Fisher Scientific, Inc.) for visualization. Images were captured using an image analysis system (model GIS-1000; Tanon Science and Technology Co., Ltd.). Western blot images were further analyzed using ImageJ software (version 1.51j8; National Institutes of Health) to determine the relative expression of proteins.

In addition to immunohistochemistry and western blotting, RT-qPCR was also performed to analyze the expression of ERK1/2 and GSK-3β. Total RNA was extracted using TRIzol LS and TRIzol (Invitrogen, Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Subsequently, the concentration and purity of obtained RNA were determined by Nano-Drop spectrophotometer (Thermo Fisher Scientific, Inc.), and if the OD260/280 was in the range of ~1.8-2.0 this was deemed as acceptable. Then, the expression level of genes was examined by RT-qPCR with the SYBR Green I (Takara Bio, Inc.) dye detection method according to the manufacturer's protocol. The thermocycling conditions were as follows: Denaturation at 95°C for 5 min followed by annealing at 60°C for 15 sec and extension at 72°C for 25 sec (40 cycles). A total of 1 µg of total RNA from each sample was reverse-transcribed using Takara reverse transcription kit (cat. no. 639505, Takara Bio, Inc.) following the manufacturer's protocol. β-actin was used as a loading control. The primers were as follows: β-actin forward, 5′-CCC GCG AGT ACA ACC TTC T-3′ and reverse, 5′-CGT CAT CCA TGG CGA ACT -3′; ERK1/MAPK3 (248 bp) forward, 5′-TCC GCC ATG AGA ATG TTA TAG GC-3′ and reverse, 5′-GGT GGT GTT GAT AAG CAG ATT GG-3′; ERK2/MAPK1 (84 bp) forward, 5′-GGT TGT TCC CAA ATG CTG ACT -3′ and reverse, 5′-CAA CTT CAA TCC TCT TGT GAG GG-3′; and GSK-3β (200 bp) forward, 5′-TAT GGT CTG CAG GCT GT G TG-3′ and reverse, 5′-CCG AAA GAC CTT CGT CCA A-3′. Finally, the cycle threshold (Cq) of each sample was detected using ABI 7900 thermocycler (Applied Biosystems; Thermo Fisher Scientific, Inc.), and the relative expression of genes was calculated using 2^−ΔΔCq method (24). Each experiment was repeated three times.

All data were presented as the mean ± standard deviation (SD) (n=6). Comparison between different groups was performed using one-way ANOVA followed by Tukey's post hoc test. Statistical analysis was performed using GraphPad Prism software, version 8.0.2 (GraphPad Software, Inc.). P<0.05 was considered to indicate a statistically significant difference.

Isolated ADSCs were identified using surface markers. Fig. 2A reveals CD73, CD90 and CD105 on cell surfaces that are positive markers for ADSCs. Negative expression of CD34, CD45 and CD11b was also observed for ADSCs as revealed in Fig. 2B. Furthermore, isolated exosomes from ADSCs were also confirmed using surface markers such as TSG101, CD9 and CD63 through western blot analysis as revealed in Fig. 2C. The morphology and size of exosomes were evaluated by TEM, and the images obtained revealed spherical vesicular shape with a size range of 80-150 nm (Fig. 2D). Isolation of exosomes is a critical step and several methods have been reported. All methods have advantages and disadvantages; however, no consensus exists between researchers on a single method. Ultracentrifugation is a commonly employed method for isolation of exosomes and was used in the present study as it does not significantly affect the protein and RNA content of exosomes (25).

Serum concentrations of AST, ALT and LDH were quantified for three groups as revealed in Fig. 3. The PBS + I/R rats exhibited higher concentrations of all markers indicating hepatic damage following 2 and 6 h of perfusion. In the case of exosome-treated rats, AST, ALT and LDH concentrations were higher than those of the sham group; however, their concentrations were significantly lower than the PBS + I/R group (Fig. 3). These results demonstrated that exosome treatment could prevent hepatocyte damage induced by hepatic I/R injury.

TEM was performed to observe subcellular structures of hepatocytes following 6 h reperfusion. Hepatocytes in the sham group exhibited normal features of nuclei, mitochondria and endoplasmic reticulum (Fig. 4A). The PBS + I/R group exhibited typical structural changes of I/R injury including swelling of mitochondria with visible cristae, dilation of hepatic sinuses, swelling of the endoplasmic reticulum, presence of Kupffer cells and the nuclei of hepatocytes exhibited pyknosis (Fig. 4B). In the ADSCs-exo + I/R group, nuclei were slightly swollen, the endoplasmic reticulum was slightly widened, while no noticeable swelling was observed in mitochondria (Fig. 4C). Collectively, ADSCs-exo treatment was found to attenuate the I/R injury effect on subcellular components.

The protective effect of ADSCs-exo against I/R injury was further evaluated using necrosis and apoptosis. H&E-stained images of liver samples of all groups following 2 and 6 h of reperfusion are presented in Fig. 5A. The sham group exhibited no significant liver tissue necrosis (Fig. 5A and B). The PBS + I/R group exhibited a significantly higher necrotic area which was evidence of I/R injury (Fig. 5B). In addition, with ADSCs-exo treatment, a significant reduction in the necrotic area was observed which indicated the efficacy of ADSCs-exo in protecting against I/R injury (Fig. 5A and B). Similarly, the PBS + I/R group exhibited a higher number of apoptotic cells as determined by TUNEL staining of liver sections following 2 and 6 h of reperfusion (Fig. 5C and D). ADSCs-exo pre-treatment produced a significant reduction in apoptotic cells as revealed in Fig. 5C and D. Collectively, it was demonstrated that ADSCs-exo pre-treatment produced a significant reduction in liver tissue necrosis and apoptosis caused by I/R injury.

For oxidative stress evaluation, the concentration of SOD and ROS activity in liver homogenates were quantified. A lower concentration level of SOD was observed in the PBS + I/R group following 2 and 6 h of reperfusion compared with the sham group (Fig. 6A). This reduced level of SOD also increased ROS in the PBS + I/R group (Fig. 6B). Regarding ADSCs-exo pre-treatment, a significant increase in SOD was revealed to reduce the oxidative stress by reducing ROS (Fig. 6A and B). Reduction in ROS may protect hepatocytes from damage induced by I/R injury. Furthermore, MDA was also quantified as a marker for lipid peroxidation. Higher MDA was observed in the PBS + I/R group as compared with the ADSCs-exo-treated rats indicating that ADSCs-exo could reduce lipid peroxidation (Fig. 6C). In summary, ADSCs-exo treatment caused a reduction in oxidative stress and consequently a reduction in lipid peroxidation.

IL-1β and TNF-α (inflammatory cytokines) in serum are markers of neutrophil activation and reported to be increased in I/R injury (26). The PBS + I/R group exhibited a higher concentration of IL-1β and TNF-α than the ADSC-exo-treated rats following 2 and 6 h of reperfusion as revealed in Fig. 7. The sham group showed lower values for inflammatory markers as compared with both the PBS + I/R and ADSCs-exo + I/R groups (Fig. 7). ADSCs-exo treatment could reduce inflammatory accumulation of inflammatory cytokines, thus, protecting the liver damage from such mediators.

ERK1/2 and GSK-3β were evaluated using immunohistochemical analysis as these pathways have been reported to attenuate ischemic injury by reducing apoptosis and oxidative stress (27,28). As shown in Fig. 8A, phosphorylation of ERK1/2 and GSK-3β were revealed to be increased in the ADSCs-exo-treated rats. Furthermore, p-ERK1/2 and p-GSK-3β upregulation was also confirmed by western blot analysis following 2 h of reperfusion which revealed a higher ratio of p-ERK/ERK and p-GSK-3β/GSK-3β in the ADSCs-exo-treated rats compared with the PBS + I/R group (Fig. 8B and C). In the sham group lower expression of p-GSK-3β was due to lack of I/R injury. ADSCs-exo caused phosphorylation of ERK1/2 and GSK-3β in the liver contributing to a protective effect against damage.

For further validation of results, RT-qPCR was also performed which revealed significantly higher expression levels of ERK1 and GSK-3β in the ADSCs-exo-treated rats compared with the PBS + I/R group. However, no significant increase was observed in the expression of ERK2 (Fig. 9).

Relative expression of Bax and Bcl-2 is important to determine apoptotic behavior. As revealed in Fig. 10A, higher Bcl-2 expression was observed in the ADSCs-exo-treated group compared with the PBS + I/R group.

PGE2 levels in the sham, PBS + I/R and ADSCs-exo + I/R groups are presented in Fig. 10B, which revealed a significantly higher concentration of PGE2 in the ADSCs-exo + I/R group after 6 h of reperfusion compared to both the sham and PBS + I/R groups. Furthermore, a similar behavior was observed in the case of cAMP levels (Fig. 10C).