Exosomes from adipose‐derived mesenchymal stem cells alleviate liver ischaemia reperfusion injury subsequent to hepatectomy in rats by regulating mitochondrial dynamics and biogenesis

Abstract Hepatic ischaemia reperfusion injury (HIRI) is a major factor leading to liver dysfunction after liver resection and liver transplantation. Adipose‐derived mesenchymal stem cells (ADSCs) have potential therapeutic effects on HIRI. Exosomes derived from ADSCs (ADSCs‐exo) have been widely studied as an alternative of ADSCs therapy. Thus, the aim of this study was to evaluate the potential protective effect and related mechanism of ADSCs‐exo on HIRI subsequent to hepatectomy. Rats were randomly divided into four groups: Sham, I30R+PH, ADSCs and ADSCs‐exo group. After 24 h of reperfusion, liver and serum of the rats were immediately collected. ADSCs‐exo improved liver function, inhibited oxidative stress and reduced apoptosis of hepatocytes in HIRI subsequent to hepatectomy in rats. ADSCs‐exo significantly promoted the recovery of mitochondrial function, markedly increased the content of ATP in the liver tissue, and improved the ultrastructure of mitochondria in hepatocytes. Moreover, ADSCs‐exo significantly increased the expression of OPA‐1, MFN‐1 and MFN‐2 proteins related to mitochondrial fusion, while DRP‐1 and Fis‐1 mRNA and protein expression associated with mitochondrial fission were significantly decreased after the treatment with ADSCs‐exo. In addition, ADSCs‐exo significantly increased the expression of PGC‐1α, NRF‐1 and TFAM genes and proteins related to mitochondrial biogenesis. ADSCs‐exo improves liver function induced by HIRI subsequent to hepatectomy in rats and maintains mitochondrial homeostasis by inhibiting mitochondrial fission, promoting mitochondrial fusion and promoting mitochondrial biogenesis. Therefore, ADSCs‐exo may be considered as a potential promising alternative to ADSCs in the treatment of HIRI subsequent to hepatectomy.


| INTRODUC TI ON
Hepatic ischaemia reperfusion injury (HIRI) is the main factor contributing to liver dysfunction after liver resection and liver transplantation, which may further aggravate the liver surgical injury and increase postoperative morbidity and mortality. 1 How to alleviate HIRI is an urgent problem to be solved in the medical research, and novel treatments are urgently required. Hepatic ischaemia results in disruption of energy supply to hepatocytes, leading to the death of hepatocytes. 2 During reperfusion, a large amount of reactive oxygen species (ROS) and inflammatory factors further aggravate HIRI, promoting the occurrence of hepatocyte necrosis and apoptosis. 3 The liver needs plenty of mitochondria to provide sufficient energy for its multiple physiological functions, since they are the major organelles responsible for the ATP generation through oxidative phosphorylation and for the maintenance of the homeostasis in the cellular environment. Recently, several studies confirmed that mitochondrial damage is one of the main factors leading to liver dysfunction in HIRI and fluoride-induced liver injury. 4,5 ROS induces mitochondrial oxidative damage and reduce the production of ATP in mitochondria. Disruption of mitochondrial homeostasis causes mitochondrial fragmentation, leading to an inflammatory response that further aggravates ischaemia reperfusion injury. [6][7][8] Mitochondrial biogenesis is involved in mitochondrial quality control and must be well coordinated to properly regulate cellular adaptation in response to mitochondrial damage. 9 Mitochondrial dynamics regulates mitochondrial distribution and morphology through fusion and fission processes. 10 Interventions targeting factors involved in mitochondrial homeostasis yielded beneficial results in the treatment of HIRI.
Therefore, the reduction of mitochondrial injury may represent an ideal strategy to protect against HIRI.
Mesenchymal stem cells (MSCs) are adherent fibroblast-like pluripotent cells possessing self-renewal ability and multi-directional differentiation potential under a specific environment, thus, they are widely investigated in the research of regenerative medicine and gene-based therapy. 11,12 The advantages of adipose derived mesenchymal stem cells (ADSCs) include abundant sources, ease of obtaining, fat collection less harmful for the body, low immunogenicity and strong proliferation ability in vitro culture. 13 ADSCs can prevent ischaemia reperfusion injury in the kidney, liver, heart and brain. [14][15][16][17] Studies indicated that MSCs perform biological functions and therapeutic effects on adjacent cells mainly by a paracrine mechanism. 18,19 Exosomes, with a diameter of 30-150 nm, are released by cells into the extracellular environment, and they represent the main bioactive vesicles of MSCs acting on surrounding cells. 20,21 MSCsderived exosomes have a protective effect on liver fibrosis, HIRI and drug-induced liver injury. [22][23][24] A previous study showed that MSCs protect against retinal ischaemia by improving mitochondrial function. 25 However, it is not yet clear whether ADSCs-exosomes (ADSCs-exo) possesses the ability to ameliorate HIRI subsequent to hepatectomy. Therefore, the present study investigated the protective effect of ADSCs-exo on HIRI subsequent to hepatectomy in a rat model and explored the mechanism of action of ADSCs-exo on mitochondrial dynamics and mitochondrial biogenesis.

| Animals
Male Sprague-Dawley rats (6-8 weeks, weighing 200-250 g) were purchased from Liaoning Changsheng biotechnology. All rats were fed with a unified standard chow, had free access to food and drinking water and were housed under standard conditions. All experimental protocols were approved by the Institutional Animal Care and Use Committee of the Northeast Agricultural University (NEAUEC 2019 05 08).

| Experimental protocol
Twenty-four rats were randomly divided into four groups and treated as follows: 1. Sham group (n = 6): Only an incision was made in the midline of the abdomen, and no clip was placed on the portal triad of the left and median lobes. Rats were treated with an intravenous injection of 600 μl PBS through the tail vein.
2. I30R+PH group (n = 6): A microvascular clip was placed on the portal triad of the left and median lobes to produce a 70% ischaemia for 30 min, and the left lobe was removed. Rats were treated with an intravenous injection of 600 μl PBS through the tail vein.
3. ADSCs group (n = 6): Rats were subjected to the same surgery as I30R+PH group. After surgery, they were treated with an intravenous injection of 2 × 10 6 ADSCs in 600 μl PBS through the tail vein.
4. ADSCs-exo group (n = 6): Rats were subjected to the same surgery as I30R+PH group. After surgery, they were treated with an intravenous injection of 100 μg ADSCs-exo in 600 μl PBS through the tail vein.
The median lobe and blood samples were collected at 24 h after reperfusion in each group.

| Surgical procedure
After a 12 h fast, rats were anaesthetized by isoflurane inhalation. A midline abdominal incision was performed to expose the portal triad of the left and median liver lobes, which was occluded with a noninvasive microvascular clamp for 30 min, resulting in 70% hepatic ischaemia, while the remaining 30% blood supply to the hepatic lobes prevented the congestion in the portal vein and gastrointestinal tract. The left hepatic lobe was ligated using a 3/0 silk thread and resected during the blockage of the blood flow.

| Isolation and culture of ADSCs
The isolation and culture of ADSCs were performed as previously described. 15 Briefly, the skin of the rats was incised and separated from the subcutaneous groyne fat to isolate the adipose tissue. The adipose tissue was washed three times with PBS, cut into small pieces and digested in PBS containing 1 mg/ml (0.1%) collagenase type I under gentle agitation at 37°C for 50 min, and subsequently centrifuged at 300 × g for 10 min. The undigested adipose tissue was removed and an equal volume of L-DMEM (HyClone) containing 10% foetal bovine serum (FBS, Clark) was added to terminate the digestion. The mixture was filtered through a 75 μm cell strainer to remove additional undigested tissue fragments and centrifuged at 300 × g for 5 min. Then, the pellet was collected and incubated with an erythrocyte lysate for 5 min. The collected ADSCs were seeded in flasks containing L-DMEM supplemented with 10% FBS, 2 mM L-glutamine, 100 u/ml penicillin and 100 μg/ml streptomycin and incubated in a humidified incubator (Galaxy 170S, Eppendorf) at 37°C under 5% CO 2 . ADSCs were used and the phenotype was evaluated at passage 3-5.

| Identification of ADSCs
Adipose-derived mesenchymal stem cells at the passage 3-5 were

| Preparation of ADSCs-exo
Adipose-derived mesenchymal stem cells at the passage 3-5 were cultured until reaching 80% confluence, and then, they were washed three times with PBS. The serum-containing medium was replaced with serum-free medium, and the cells were cultured for other 24 h.
Then the supernatant was collected and subject to sequential centrifugation: 300 × g for 15 min, 2000 × g for 25 min, filtered through a 0.22 μm filter and centrifuged again at 10,000 × g for 50 min. The supernatant was transferred into a new ultracentrifugation tube (Beckman, 355618) and centrifuged at 100,000 × g for 90 min using a Ti-50.2 rotor (Beckman). The pellet was resuspended in PBS and centrifuged at 100,000 × g for 90 min. Finally, the pellet containing exosomes was resuspended in PBS and stored at −80°C until further use.

| Measurement of liver function and ATP content
Blood samples were centrifuged at 1800 × g for 15 min to collect the serum that was stored at −80°C until further use. The levels of serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), alkaline phosphatase (ALP), total protein (TP) and lactate dehydrogenase (LDH) were determined using an automatic biochemical analyzer (BioTek). The ATP content in the liver tissue was measured using an ATP assay kit (Beyotime Biotechnology) according to the manufacturers' instructions.

| Determination of oxidative stress
The 10% liver tissue homogenate solution was obtained by homogenization in cold normal saline and consequent centrifugation at 600 × g for 10 min at 4°C. The obtained supernatant was used to measure malonaldehyde (MDA), superoxide dismutase (SOD), glutathione peroxidase (GSH-px) and catalase (CAT) using the corresponding assay kits from Nanjing Jiancheng Bioengineering Institute, Nanjing, China. The ROS level in the liver tissue was detected by DHE staining.

Caspase-9 activities
Liver tissue was homogenized using a lysis buffer included into the Caspase-3 and Caspase-9 activity assay kit (Beyotime). The supernatants were used to detect their activity according to the manufacturer's instructions. Protein concentration was determined using the Bradford assay kit (Beyotime).

| TUNEL staining
The median lobe of the liver was fixed in 4% paraformaldehyde and embedded in paraffin using standard methods. The paraffin block was then cut into 5μm sections, which were dewaxed and rehydrated. Apoptotic cells were detected using the In Situ Cell Death Fluorescein Detection Kit (Roche Molecular Biochemicals) according to the manufacturer's instructions. The images were captured using a Nikon Ti-U microscope.

| Real-time quantitative PCR (RT-qPCR)
The total RNA from liver tissue was extracted using Trizol reagent (Invitrogen) and converted into cDNA using the PrimeScript RT reagent kit with gDNA Eraser (Takara) according to the manufacturer's protocol. The RT-qPCR was performed using a LightCycler 480 I (Roche) following to the manufacturer's instruction. Relative mRNA expression was calculated by 2 −ΔΔCt method and normalized to β-actin. The primer sequences were designed and synthesized by Sangon Biotech and listed in Table 1.

| Western Blot analysis
Liver proteins were extracted by RIPA lysis buffer and western blot was performed as previously described. 26 The primary antibodies used were the following: rabbit anti-PGC-1α (1:1000 dilution,

| Statistical analysis
Statistical analysis was performed using SPSS 22.0 software. Oneway ANOVA was used for comparison among groups and LSD post analysis was performed. Results were presented as mean ± standard deviation. A value of p < 0.05 was considered statistically significant.

| Identification of ADSCs-exo
The exosomes were successfully isolated from the culture supernatant of ADSCs, and their typical cup-shape morphology was observed by TEM (40,000×) (Figure 2A). Particle Metrix Analyzer showed that the average size of ADSCs-exo was 142.3 nm ( Figure 2B). Western blots results ( Figure 2C) showed that the exosomal markers CD9, CD81 and TSG101 were expressed in exosomes. These results indicated that exosomes were successfully isolated from rat ADSCs.

| Treatment with ADSCs-exo improves liver function affected by I30R+PH
To clarify the effect of ADSCs-exo on liver function after HIRI subsequent to hepatectomy in rats, serum ALT, AST, TBIL, ALP, LDH and TP levels were measured ( Figure 3A Figure 3F).

| Treatment with ADSCs-exo reduces oxidative stress due to I30R+PH
The level of MDA in the liver of the I30R+PH group was significantly increased (p < 0.01) and SOD, CAT, GSH-px levels were markedly reduced  Figure 4E). Indeed, DHE fluorescence intensity ( Figure 4F) in the I30R+PH group was significantly increased compared with that in the Sham group (p < 0.01). However, DHE fluorescence intensity was markedly reduced after the treatment with ADSCs and ADSCs-exo compared with that in the I30R+PH group (p < 0.01).

| Treatment with ADSCs-exo increases ATP content after I30R+PH
The ATP content ( Figure 4G) in the liver of the I30R+PH group was significantly reduced compared with its content in the Sham group (p < 0.01). However, its content significantly increased after ADSCs and ADSCs-exo treatment (p < 0.05 vs. I30R+PH group).

| Morphology of the liver tissue after treatment with ADSCs-exo
Transmission electron microscopy images showed a normal ultrastructure of the hepatocytes in the Sham group ( Figure 5A,E).
Nevertheless, hepatocytes from the I30R+PH group ( Figure 5B,F) showed a severe ultrastructural damage with nuclear shrinkage, and a decrease in the number of mitochondria, which were also swollen. The endoplasmic reticulum in the I30R+PH group was severely expanded and the glycogen granules were decreased from the hepatocytes. Treatment with ADSCs ( Figure 5C,G) and ADSCs-exo ( Figure 5D,H) alleviated the ultrastructural changes of hepatocytes observed above.

| Treatment with ADSCs-exo inhibits apoptosis due to I30R+PH
The expression of Bcl-2 mRNA ( Figure 5J) and protein ( Figure 5K,M) in the I30R+PH group was markedly decreased when compared with the expression in the Sham group (p < 0.01). However, Bcl-2 mRNA and protein expression in the ADSCs and ADSCs-exo group were significantly increased when compared with their expression in the I30R+PH group (p < 0.01 or p < 0.05). The expression of Bax mRNA ( Figure 5I) and protein ( Figure 5K,L), as well as Caspase-3 ( Figure 5N) and Caspase-9 activity ( Figure 5O) in the I30R+PH group were significantly However, the rate was remarkably reduced after ADSCs-exo and ADSCs treatment (p < 0.01 or p < 0.05) ( Figure 5P,Q).

| Treatment with ADSCs-exo inhibits excessive mitochondrial fission and promotes mitochondrial fusion after I30R+PH
The results indicated that DRP-1 and Fis-1 mRNA ( Figure 6A

| Treatment with ADSCs-exo promotes mitochondrial biogenesis after I30R+PH
The expression of PGC-1α, TFAM and NRF-1 mRNA and protein were markedly decreased at 24 h after HIRI subsequent to hepatectomy (p < 0.01 vs. Sham group) ( Figure 7A-G). However, both the above mRNA and protein expression were significantly increased in ADSCs and ADSCs-exo group when compared with the I30R+PH group (p < 0.01 or p < 0.05).

| DISCUSS ION
The present study revealed at first that the treatment with ADSCs and ADSCs-exo could significantly reduce HIRI subsequent to Some studies found that the pathogenesis of HIRI mainly includes the overproduction of ROS, intracellular calcium overload, and inflammatory response. The continuous release of ROS disrupts the antioxidant system, leading to oxidative stress. 27 Thus, ROS scavenging is essential for attenuating HIRI. 28 A recent study confirmed that the ADSCs-exo exert an antioxidant effect by suppressing ROS production. 29 The present study found that ADSCs-exo treatment significantly decreased the levels of oxidative damage indicators such as MDA and ROS in the liver, while markedly increased the content of antioxidant enzymes SOD, CAT and GSH-px after HIRI subsequent to hepatectomy. Thus, the present results demonstrated that ADSCs-exo protected against oxidative stress after HIRI subsequent to hepatectomy in rats.
The mechanism inducing hepatocyte death caused by HIRI typically involves apoptosis. Mitochondria are involved in metabolic and energy for cells, and regulate apoptosis. Mitochondrial injury caused by excessive ROS reduces mitochondrial membrane potential and down-regulate ATP synthesis, subsequently leading to cell apoptosis. 30 Ischaemia and hypoxia induce mitochondrial swelling and cytochrome C release into the cytoplasm, which further promotes the caspase cascade reaction and stimulates apoptosis. 31 The regulation of hepatocyte apoptosis might provide a therapeutic method to cure HIRI. One study demonstrated that exosomes from human-induced pluripotent stem cell-derived mesenchymal stromal cells protect against liver injury and inhibit apoptosis in HIRI. 32 Additionally, ADSCs-exo protect H9c2 cells from H 2 O 2 induced apoptosis and reduce apoptosis in cardiac ischaemia reperfusion injury. 33 In the present study, the treatment with ADSCs-exo decreased apoptosis by inhibiting the activity of Caspase-3 and Caspase-9, decreasing Bax mRNA and protein expression and increasing Bcl-2 mRNA and protein expression. These results indicated that ADSCs-exo inhibited apoptosis, thus, they were beneficial to alleviate HIRI subsequent to hepatectomy in rats.
Mitochondria homeostasis determines cell fate during ischaemia reperfusion injury. 34 In this study, the morphology of hepatocytes revealed that the damage of mitochondria was significantly reduced after ADSCs-exo treatment. As regard the ATP production our results were consistent with those of previous studies, 35,36 since the ATP content in the liver of the I30R+PH group was significantly lower than that in the Sham group, which might be due to a direct consequence of impaired mitochondrial dynamics. ADSCs-exo treatment reversed the decrease of ATP content caused by HIRI subsequent to hepatectomy. Thus, our hypothesis was that mitochondrial injury and functional recovery might be associated with ADSCs-exo regulation of mitochondrial dynamics balance.
Mitochondria are highly dynamic organelles that continuously undergo two opposite processes such as fission and fusion to maintain mitochondrial function under physiological conditions. 37 Thus, mitochondrial fission and fusion are used to evaluate the morphology, quantity and abundance of mitochondria in liver diseases. 38 Excessive mitochondrial fission induces mitochondrial fragmentation, eventually leading to apoptosis. 39  mechanisms regulating the quality and quantity of mitochondria and PGC-1α is regulating it. PGC-1α promotes the increase of NRF-1 expression and interacts with NRF-1 to regulate mitochondrial biogenesis. 46 Therefore, the induction of PGC-1α expression in the liver is a critical procedure for the energy metabolism pathway for the production of ATP and for maintaining the homeostasis of the mitochondrial environment. NRF-1 is a transcription factor that regulates five respiratory complexes synthesis and TFAM. 47 TFAM is a transcription factor necessary to regulate the transcription and replication of mitochondrial DNA. 48 However, persistent hypoxia and ischaemia impairs mitochondrial biogenesis and inhibits protein expression in the PGC-1α-NRF-1-TFAM pathway. 49 Recent studies suggested that the Quantification of PGC-1α (E), NRF-1 (F) and TFAM (G) protein expression in four groups. Results are presented as mean ± SD, * p < 0.05, ** p < 0.01 versus the Sham group. # p < 0.05, ## p < 0.01 versus the I30R+PH group enhancement of mitochondrial biogenesis protects against HIRI. 50,51 Our results showed that the expression of PGC-1α, NRF-1 and TFAM were significantly decreased after HIRI subsequent to hepatectomy, but these indicators were markedly increased after ADSCs-exo treatment. These results indicated that ADSCs-exo had the ability to enhance mitochondrial biogenesis and protected against mitochondrial dysfunction induced by HIRI subsequent to hepatectomy.
In summary, our study is the first demonstrating that ADSCsexo protects against HIRI subsequent to hepatectomy by reducing mitochondrial fission, promoting mitochondrial fusion and improving mitochondrial biogenesis. Thus, ADSCs-exo treatment might be considered as a potential novel and promising therapeutic approach to combat HIRI subsequent to hepatectomy, providing a basis for the potential clinical application of ADSCs-exo in liver surgery.

ACK N OWLED G EM ENTS
The authors thank Hongbin Wang who designed this study and revised the manuscript. The authors thank the Heilongjiang Key Laboratory for Laboratory Animals and Comparative Medicine.

CO N FLI C T O F I NTE R E S T
The authors declare that there are no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.