Adiponectin Ameliorates Lung Ischemia-Reperfusion Injury Through SIRT1-Pink1 Signaling-Mediated Mitophagy in Type 2 Diabetic Rats


 BackgroundDiabetes mellitus (DM) is a key contributing factor to the poor survival in lung transplantation recipients. Mitochondrial dysfunction is recognized as a critical mediator in the pathogenesis of diabetic lung ischemia-reperfusion (IR) injury. The protective effects of adiponectin have been demonstrated in our previously study, but the underlying mechanism remained unclear. Here we demonstrated an important role of mitophagy in the protective effect of adiponectin during diabetic lung IR injury.Methods High-fat diet-fed streptozotocin-induced type 2 diabetic rats as recipients were exposed to adiponectin with or without administration of the SIRT1 inhibitor EX527 following lung transplantation. To unravel the mechanisms underlying the action of adiponectin, rat pulmonary microvascular endothelial cells were transfected with SIRT1 small-interfering RNA or Pink1 small-interfering RNA and then subjected to in vitro diabetic lung IR injury.ResultsMitophagy was impaired in the diabetic lung subjected to IR injury, accompanied by increased oxidative stress, inflammation, apoptosis, and mitochondrial dysfunction. Adiponectin induced mitophagy and attenuated subsequent diabetic lung IR injury by improving lung functional recovery, suppressing oxidative damage, diminishing inflammation, decreasing cell apoptosis, and preserving mitochondrial function. However, both inhibitors of mitophagy and knockdown of Pink1 suppressed mitophagy, and reduced the protective action of adiponectin. Furthermore, we demonstrated that APN affected Pink1 stabilization via the SIRT1 signaling pathway, and knockdown of SIRT1 suppressed Pink1 expression and compromised the protective effect of adiponectin.ConclusionThese data demonstrated that adiponectin attenuated reperfusion-induced oxidative stress, inflammation, apoptosis and mitochondrial dysfunction via activation of SIRT1-Pink1 signaling-mediated mitophagy in diabetic lung IR injury.


Introduction
Lung transplantation remains de nitive therapy for end-stage respiratory failure with no other options [1]. However, approximately 15% of the recipients undergoing lung transplantation experience graft complications due to lung ischemia reperfusion (IR) injury [2]. The International Society of Heart and Lung Transplantation Registry data show that diabetes mellitus (DM) is an independent risk factor for mortality at both 1 and 5 years after lung transplantation, whereas an unexpectedly high prevalence of undiagnosed DM in patients awaiting lung transplantation [3]. Lung transplantation recipients may already be at increased risk of DM prior to transplantation [4]. For example, patients with chronic obstructive lung disease (COPD) may already be taking glucocorticoids. Enormous efforts have been made to explore the rescue strategy of lung IR injury in diabetic state. Our previous study showed DM aggravated lung IR injury, and mitochondrial dysfunction was a pivotal factor in this process [5]. Hence, preservation of mitochondrial function is signi cant in management of patients with pretransplant DM who undergoing lung transplantation.
Mitochondrial homeostasis can be regulated in several ways. Moderation and elimination of reactive oxygen species (ROS), and mitochondrial autophagy (mitophagy) have all been demonstrated to be critical mediators of mitochondrial homeostasis [6].
Mitophagy is crucial for mitochondrial quality control, which selectively removes dysfunctional or damaged mitochondria via autophagy, thereby preventing excessive reactive oxygen species (ROS) production and release of mitochondrial pro-apoptotic factors [7]. The role of mitophagy in IR injury remains controversial, and both adaptive and detrimental effects have been reported [8,9]. It has been suggested that mitochondrial dysfunction was found in diabetic complications due to the decreased levels of mitophagy, leading to the accumulation of damaged mitochondria and exacerbating the injury to tissues [10].
Furthermore, the occurrence of mitophagy in diabetic lung IR injury remain unclear. We believe that a better understanding of how mitophagy is regulated during diabetic lung IR injury, which is essential in order to developing a new therapeutic strategy to attenuates pulmonary dysfunction in diabetic lung transplantation recipients.
Silent information regulator 1 (SIRT1) is a nicotinamide adenine dinucleotide (NAD+)-dependent deacetylase that mediate protective effect against IR injury, possibly through maintenance of mitochondrial function [11]. Of note, SIRT1 has been proposed to induce mitophagy contributing to attenuation of mitochondrial dysfunction and subsequent IR injury [12]. We have previously demonstrated normalizing or activating lung SIRT1 signaling conferred a pulmonary protective effect [13]. The question of whether lung SIRT1 could be a master regulator of mitophagy in diabetic lung IR injury represented a logical extension of the above studies.
Adiponectin (APN), as an adipocytokine secreted by adipocytes, can modulates insulin responsiveness, maintains mitochondrial homeostasis [14]. Plasma adiponectin levels are decreased in insulin resistance, obesity, and type 2 diabetic patients [15]. Restoring diabetes-induced hypoadiponectinemia by adiponectin is correlate with decreased insult of IR, as well as favorable functional recovery after IR injury [16]. Previous studies from our lab have indicated that adiponectin protected against lung IR injury in type 2 diabetic rat, but the potential mechanisms remain poorly de ned [17]. Accordingly, the aims of this study were to identify whether adiponectin treatment ameliorated diabetic lung IR injury by regulating mitophagy to protect against mitochondrial dysfunction via activating the SIRT1 signaling pathway. Establishment of a type 2 diabetic rat model High-fat diet-fed streptozotocin-induced type 2 diabetic rat model was induced as described previously [5,13]. The rats with fasting plasma glucose above 11.1 mmol/L 72 hours after STZ injection were considered as diabetic. Next, we tested the glucose tolerance of various groups by conducting the intraperitoneal glucose tolerance test (IPGTT) and oral glucose tolerance test (OGTT) to con rm the successful establishment of the type 2 diabetic rat model. The standard laboratory chow-fed rats were studied as the nondiabetic controls.

Lung Transplantation
Orthotopic left lung transplantation using the cuff technique procedure was carried out as described previously [18,19]. Brie y, donor rats were anesthetized with sodium pentobarbital (30 mg/kg) administered intraperitoneally, intubated with 12-gauge catheter and ventilated with 40% oxygen (balance nitrogen) at a tidal volume of 10 ml/kg with 2 cm H 2 O positive end-expiratory pressure (PEEP). After heparinization, the donor left lung was ushed with 20 ml of low-potassium dextran solution at 4°C at a perfusion pressure of 20 cm H 2 O through the pulmonary artery. The left lung was clipped and attached to a cuff tube, and preserved at 4°C in the perfusion solution for 2 hours.
The recipient rats were anesthetized and ventilated in the same manner as the donor rats. After a left thoracotomy, the left pulmonary arteries, bronchi and left pulmonary veins were conjugated between donors and recipients by the cuff technique.
During the lung transplantation, the tidal volumes were regulated to 6 mL/kg and restored to 10 mL/kg immediately after reperfusion. The recipients were extubated after recovery from anesthesia. The recipient rats were treated with 0.125% ropivacaine by local in ltration analgesia every 12 hours. The recipients using type 2 diabetic rats were studied as diabetic lung transplantation. All rats were positioned on a heating pad to maintain body temperature and sodium pentobarbital was used to maintain anesthesia. The body temperature of individual rats was measured by a rectal thermometer and maintained between 37°C and 39°C.

Histological analysis
The lung tissues were xed in paraformaldehyde and embedded in para n. 5-μm thickness sections were prepared and stained with hematoxylin and eosin. The degree of the lung injury was assessed for airway epithelial cell damage, neutrophil in ltration, hemorrhage, interstitial edema, and hyaline membrane formation, and by 2 pathologists in a blinded manner.

Enzyme-linked immunosorbent assay
Serum concentrations of interleukin-6 and tumor necrosis factor-α were measured by enzyme-linked immunosorbent assay kits (R&D Systems, Minnesota, USA) according to the manufacturer's protocols.
Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay Lung parenchymal cell apoptosis was detected by TUNEL using an In Situ Cell Death Detection kit (Roche Molecular Biochemicals, Mannheim, Germany) according to the manufacturer's protocols.

Western blot analysis
Western blot analysis was carried out as described previously [5,13].

Measurement of mitochondrial membrane potential
Mitochondrial membrane potential was assessed by a JC-1 staining kit (Sigma Aldrich, St. Louis, MO) as described previously [5].

In Vitro Study
The culture and identi cation of rat pulmonary microvascular endothelial cells (PMVECs) Rat PMVECs were isolated and cultured using the "tissue" method as previously described [22]. Brie y, the rats were euthanized by exsanguination, and the lungs were removed by sterile techniques. The visceral pleura was removed and peripheral lung tissue was cut into small pieces (<1 mm 3 ) in medium M199 containing 20% fetal bovine serum and 50 μg/mL endothelial cell growth supplement. Then, the fragments were placed in a 25 cm 2 culture ask upside down adding penicillin-streptomycin (100 U/ml) in a 5% CO 2 , at 37°C. After 60 hours of culture, the tissues were removed and M199 was changed. PMVECs were identi ed according to the results of immunocytochemistry staining of CD31 and lectin binding.

Simulated IR
In vitro ischemia-reperfusion was performed as previously described [22]. PMVECs were placed in a sealed incubator and preventilated with 95% O 2 and 5% CO 2 for 2 hours.

Simulated cold storage
The sealed incubator was placed in 4˚C, and M199 culture medium was immediately replaced with low-potassium dextran solution (pH 7.2-7.4) with gas insu ation stoppage.

Simulated implantation
After 2 hours of simulated cold storage, the incubators were kept at room temperature and sealed for 1 hour to simulate the transplantation period.
For simulated type 2 diabetic reperfusion, high glucose-high fat (HG/HF) M199 containing 15 mM glucose and saturated FFA palmitate (16:0; 500 μM) was used to simulate pathophysiology condition of diabetic state, while normal M199 culture medium was used as a control [23]. For adiponectin pretreatment, adiponectin (2 μg/mL) was used immediately after reperfusion [24]. Gas concentra tions in the incubator were monitored with a gas analyzer (S/N 32590; Datex Ohmeda, Helsinki, Finland).

Immuno uorescent staining
PMVECs were xed with 4% paraformaldehyde and permeabilized with 0.2% Triton X-100, blocked with 5% bovine serum albumin, and subsequently incubated with primary antibodies. After washing with PBS, cells were exposure to secondary antibodies conjugated to an Alexa uorophore. Fluorescent images were obtained with a laser scanning confocal microscope. Five randomly selected elds from one coverslip were included to calculate an average, and experiments were repeated independently at least 3 times.

Cell viability
The effect of high glucose on the cell viability for PMVECs was evaluated by performing WST-8 assay using Cell Counting Kit-8 (CCK8, Dojindo Laboratories, Japan), and the absorbance was assessed at 450 nm using a microplate reader (Bio-Rad iMark; Bio-Rad Laboratories, Hercules, CA, USA).

Apoptosis Assay
Annexin V-7-AAD apoptosis assay kit (Biotech, China) was used to measure apoptosis of PMVECs by ow cytometry following the manufacturer's instructions.

Mitochondrial membrane potential and mitochondrial ROS
Mitochondrial membrane potential was measured by exposing PMVECs to JC-1 molecular probes (Invitrogen, Calif, USA) following the manufacturer's instructions. Mitochondrial membrane potential was expressed as the ratio of red to green uorescence areas.
To assess mitochondrial ROS production, cells were incubated with MitoSOX (Life Technologies, USA). Mitochondrial ROS generation was visualized by uorescence microscopy.

Measurement of mitochondrial morphology
Page 6/20 The mitochondrial disruption was evaluated by the Flameng score [25] as follows: 0, structures are normal and particles are intact; 1, structures are normal and particles are lost; 2, mitochondria are swollen but matrices are clear; 3, cristae are broken and matrices are concentrated; 4, cristae are extensively destroyed and the membranes are ruptured.

Statistical analysis
The data are presented as the means ± standard deviation (SD). Statistical testing was examined using Prism software package version 5.0 (GraphPad Software, La Jolla, CA, USA). Statistical signi cance was evaluated by one-way ANOVA followed by Tukey's post hoc test for multiple comparisons among the groups or using a 2-tailed Student t test for unpaired observations. A value of P < 0.05 was considered to indicate a statistically signi cant difference. In vitro experiments were repeated at least 3 times, and 8 independent experiments for the in vivo study.

Characterization of diabetic animals
As showed in Fig.1 A, B, compared with the non-diabetic rats, diabetic rats showed signi cantly impaired OGTT and IPGTT (P < 0.05), demonstrating that the type 2 diabetic model were successfully developed.
Diabetes reduces lung IR-induced mitophagy, APN restored diabetic lung IR-reduced mitophagy via SIRT1 We have reported that impaired lung SIRT1 signaling associated with type 2 diabetic conditions was further attenuated by IR injury in warm lung IR model [13]. Similar tendencies were also observed in diabetic lung transplantation model ( Fig. 1C and I). We also found that APN treatment restored the expression and activity level of SIRT1 (P < 0.05). Furthermore, the EX527 treatment decreased SIRT1 activity and increased the acetylation of FoxO1 without in uencing SIRT1 expression. (P < 0.05, Fig. 1C, D and I).
The expression of mitophagy-related proteins were evaluated to assess the state of mitophagy. The conversion of LC3-I to LC3-II is a hallmark of autophagosome formation[26]. One paradigm for mitophagy involves the well-established Pink1 (PTEN-induced kinase)-Parkin pathway that tags mitochondria for degradation [27]. As present in Fig. 1C, the level of LC3-II in the mitochondrial fraction were decreased as well as Pink1 and parkin in the DM+ Sham group, (P < 0.05). The expression level of proteins was also attenuated in the DM+IR group compared with the Con+IR group (P < 0.05), suggesting mitophagy was suppressed in lung IR injury under type 2 diabetic conditions. Treatment with APN could improve the levels of mito-LC3-II, an effect that was accompanied by increases in Pink1 and Parkin (P < 0.05), however, inhibition of SIRT1 treatment attenuated these effect (P < 0.05).
SIRT1 signaling pathway-mediated mitophagy participated in APN-mediated pulmonary protection of diabetic lung IR injury To con rm the bene ts of APN are closely related to mitophagy, autophagy inhibitor 3-MA were used. As showed in Fig.2 A, the PaO 2 /FiO 2 ratio in the Con+Sham group and the DM+Sham group showed no difference (P > 0.05). The PaO 2 /FiO 2 ratio during reperfusion at 24 hours were decreased in diabetic rats subjected to lung IR injury (P < 0.05), but APN administration signi cantly increased the PaO 2 /FiO 2 ratio (P < 0.05). However, inhibition of SIRT1 abolished the effect of APN on the PaO 2 /FiO 2 ratio (P < 0.05), and 3-MA administration also attenuated the effect of APN on the PaO 2 /FiO 2 ratio (P < 0.05). Increases in wet weight-to-dry weight ratio were correlated with the decreases in PaO 2 /FiO 2 ratio (Fig.2 B).
As presented in Fig.2 C, grafts in the Con+IR group showed lung edema, increased alveolar damage, hemorrhage, hyaline membrane formation, and accompanied by in ammatory in ltrates. The lung injury scores in the DM+IR group were greater than that in the Con+IR group (P < 0.05). APN treatment attenuated the histologic changes (P < 0.05), while inhibition of SIRT1 aggravated the histologic changes (P < 0.05). The bene ts of APN on the histologic changes were also reversed by 3-MA (P < 0.05).
SIRT1 signaling pathway-mediated mitophagy played an essential role in anti-apoptotic, anti-in ammatory and anti-oxidative effects of APN following diabetic lung IR injury The apoptotic index of lung grafts was determined by TUNEL assay (Fig.3 A). The number of apoptosis cells in the Con+Sham group and the DM+Sham group showed no difference (P > 0.05). The DM+IR group exhibited a higher apoptosis index compared with the Con+IR group (P < 0.05). The number of apoptosis cells in the lung grafts were signi cantly reduced in the DM+IR+A group (P < 0.05), but the antiapoptotic effects of APN were mitigated in the DM+IR+A+S group (P < 0.05). The antiapoptotic effect of APN were also abolished in the DM+IR+A+M group (P < 0.05). JC-1, a uorescent dye highly sensitive to any small changes in mitochondrial membrane potential (Δψm), exhibiting the opposite trend as TUNEL assay (Fig.3 C). Production of cytokines (IL-6, and TNF-a), oxidative stress (MDA) showed the same trend as the apoptotic index, the antioxidative capacity (activities of SOD) exhibited the opposite trend compared with the previous levels (Table 1).

Identi cation of PMVECs
To further elucidate the mechanisms underlying the action of APN on diabetic lung IR injury, we used PMVECs to establish the in vitro model of diabetic transplanted lungs [22]. The con uent PMVECs showed a cobblestone appearance and microvascular structures (Fig.4 A). The cells were identi ed by CD31 antigen and FITC-conjugated lectin expression, which are classical and highly speci c markers of PMVECs (Fig.4 B and C).
In order to explore favorable levels of glucose to mimic the pathophysiology condition of diabetic state, PMVECs were exposed to glucose at a nal concentration of 10, 15 and 30 mM with saturated FFA palmitate in cultures [23,28]. As shown in Fig.4 D, there were no effect in the cultures exposed to 10 mM glucose for 24 h, whereas 30 mM glucose caused severe cell death. Therefore, 15 mM glucose was used because it may mimic the glucose levels of the diabetic state without insulin treatment in the present study and also was commonly used in previous studies [28].
Alterations in mitochondrial ROS production, Δψm and mitophagy in PMVECs subjected to diabetic IR injury LC3-II staining was used to markers for autophagosome, TOMM20 was used to markers for mitochondria, whereas a combination of LC3-II and TOMM20 were used to delineate mitophagy. The extent of mitophagy was examined by the number of LC3-II puncta on mitochondria per cell. As present in Fig.5 A, mitophagy was decreased in the HG/HF group compared with the Con group (P < 0.05), and mitophagy was also decreased in the HG/HF +SIR group (P < 0.05). These data further con rmed that HG/HF-inhibited mitophagy in PMVECs subjected to IR injury. As present in Fig.5 C, the levels of SIRT1 were decreased as well as Pink1 (P < 0.05, the HG/HF group compared with the Con group). The expression levels of proteins were also attenuated in the HG/HF+SIR group compared with the Con+SIR group (P < 0.05). Mitochondrial dysfunction and accumulation of damaged mitochondria may contribute to ROS generation and cell death. We measured the levels of mitochondrial ROS in PMVECs by MitoSOX (red). As showed in Fig.5 G, mitochondrial ROS levels were increased in the HG/HF group (P < 0.05 , and the levels were increased in the HG/HF +SIR group (P < 0.05). HG/HF also induced mitochondrial depolarization following IR injury, as evidenced by diffuse JC-1 green staining (P < 0.05, Fig.5 E). The rate of PMVECs apoptosis showed the same trend as mitochondrial ROS levels (Fig.5 I).
APN upregulated Pink1-dependent mitophagy via SIRT1 signaling pathway in PMVECs subjected to diabetic IR injury.
Next, we explored some insights into the regulation of mitophagy in our experimental models, and PMVECs infected with Pink1 siRNA or SIRT1 siRNA. The expression of SIRT1 mRNA and protein were dramatically reduced by all 3 SIRT1 siRNA compared with that in the negative control, with B more effective than A and C (Fig.6 A, C). Similar changes were observed in Pink1 siRNA, with A more effective than B and C (Fig.6 B, D). We therefore chose SIRT1 siRNA B, and Pink1 siRNA A for further analysis.
As showed in Fig.6 E, APN treatment signi cantly enhanced mitophagy in PMVECs exposed to diabetic IR injury compared with that found in the untreated PMVECs (P < 0.05). Having demonstrated that APN signi cantly increased mitophagy, we next determined the molecular mechanisms mediating this effect in diabetic IR injury. The western blot assays showed that APN treatment notably enhanced the expression levels of Pink1 in PMVECs exposed to diabetic IR conditions (P < 0.05, Fig.6 G). As shown in Fig.6 E, gene silencing of Pink1 markedly abolished the effect of APN on upregulation of mitophagy, suggesting that Pink1 pathway plays a central role in the action of APN on mitophagy. We next investigated the mechanism underlying the induction of Pink1-dependent mitophagy by APN, and the SIRT1 signaling pathway was evaluated because previous studies have identi ed SIRT1 as the upstream mediator of Pink1 [29]. As show in Fig.6 G, western blotting analysis demonstrated that both SIRT1 and Pink1 were repressed by diabetic IR injury, APN treatment upregulated the levels of SIRT1 as well as Pink1 in PMVECs (P < 0.05). Gene silencing of SIRT1 caused a decline in Pink1 expression (P < 0.05), whereas gene silencing of Pink1 had little effect on SIRT1 expression (P > 0.05), suggesting that Pink1 expression was modulated by the SIRT1 signaling pathway.
Interestingly, we also found gene silencing of SIRT1 markedly abolished the effect of APN on upregulation of mitophagy (P < 0.05, Fig.6 E).

APN conferred protective effects of mitochondrial function through SIRT1-Pink1-dependent mitophagy in PMVECs subjected to diabetic IR injury
We next examined the effects of mitophagy on APN-mediated cytoprotection in PMVECs subjected to IR injury. As showed in Fig.7 A, APN treatment restored Δψm, allowing JC-1 uptake by mitochondria (red staining) (P < 0.05), while either knocked down SIRT1 or Pink1 expression attenuated the effect APN on mitochondrial JC-1 (P < 0.05). As showed in Fig.7 B, APN treatment reduced mitochondrial ROS (P < 0.05,), while knocked down SIRT1 expression attenuated the effect APN on mitochondrial ROS (P < 0.05), knocked down Pink1 expression also mitigated the effect APN on mitochondrial ROS (P < 0.05). Next, we examined the mitochondrial morphological changes in PMVECs under transmission electron microscope. As showed in Fig.7 E, the PMVECs in the HG/HF+SIR group exhibited mitochondrial edema, cristae rupture, matrix concentration and mitochondrial membrane rupture. APN treatment attenuated mitochondrial morphological changes, and the Flameng scores in the HG/HF+SIR+A group were lower than that in the HG/HF+SIR group (P < 0.05). Mitophagosome formation were also found in the HG/HF+SIR+A group. However, knocked down Pink1 expression attenuated the effect APN on mitochondrial morphological changes (P < 0.05), and knocked down SIRT1 expression also mitigated the effect of APN on mitochondrial morphological changes (P < 0.05). To investigate the bene cial action of APN-modulated mitophagy on PMVECs, 3-MA was used in APN-treated. As showed in Fig.7 G, APN notably reduced the rate of PMVECs apoptosis caused by diabetic IR injury (P < 0.05), whereas either knocked down SIRT1 or Pink1 expression abolished the effect of APN (P < 0.05). Of note, the bene ts of APN were closely related to mitophagy as the rescue effects of APN were reversed by 3-MA treatment (P < 0.05).

Discussion
The involvement and modulation of mitophagy in the pathogenesis of diabetic lung IR injury remains poorly understood. Data presented in this study made several important observations. First, our study demonstrated DM aggravated lung IR injury during lung transplantation in vivo and in vitro. Second, our results indicated that type 2 diabetes severely impaired mitophagy following lung transplantation. Third, we exhibited evidence that, by upregulating SIRT1-Pink1-dependent mitophagy, APN might ameliorate mitochondrial dysfunction, thus reduced ROS production and in ammation, contributing to cell survival and ultimately preserving lung function during diabetic lung IR injury.
The incidence of diabetes continues to rise worldwide, and the patients with DM pretransplant have poor survival compared with nondiabetic population in lung transplantation recipients [3]. In this study, we developed a new diabetic lung transplantation model by using type 2 diabetic rat as recipient, which simulated the events associated with clinical lung transplantation recipients with pretransplant DM. Concomitant with our previous report of warm diabetic lung IR model, we found that DM aggravated lung IR injury in transplantation model [5,13].
Hyperglycemia lead to overproduction of reactive oxygen species that then potentially provoking mitochondrial dysfunction, and damaged mitochondria may predispose cells to free radical generation and eventual cell death [30]. ROS and various factors released upon cell death potentially provoke in ammation [31]. In response to dysfunctional mitochondria, mitochondrial quality control mechanisms can be activated to allows autophagosomes to selectively eliminate impaired mitochondria known as mitophagy [7]. Mitophagy is critical for mitochondrial quality control and cell survival [6]. Upon cellular stress, excessive damaged mitochondria result in overproduction of cellular ROS from the abnormal electron transport chain, which may induce further injury to mitochondria, forming a vicious cycle eventually leading to irreversible cell apoptosis [32]. In this case, mitophagy is activated allowing elimination of damaged mitochondria. Mitophagy is also closely related to mitochondrial biogenesis [33]. The degradation contents caused by mitophagy, with the resultant components (amino acids and lipids) being recycled and reused during mitochondrial biogenesis [34]. Thus, eliminated mitochondria are continuously substituted for new ones. The best characterized mitophagy pathway so far is Pink1-Parkin pathway. When a subset of mitochondria become damaged and depolarized, Pink1 is accumulation on the mitochondrial outer membrane, and Parkin is then recruited and translocated to mitochondria to initiate mitophagy [35]. Mitochondrial dysfunction is a critical factor in lung IR injury [36]. Increasing lines of evidence demonstrated that a protective role of Pink1-Parkin mediated mitophagy against IR injury [37]. More importantly, convincing evidence showed mitophagy through Pink1-Parkin degradation mechanisms was suppressed in type 2 DM, which contribute to the accumulation of dysfunctional mitochondria [38,39]. Contrary observations have been reported in mice fed with high-fat diet, Tong et al observed that mitophagy activated during the initial phase of high-fat diet consumption was both Atg7and Parkin-dependent in cardiomyocytes. However, Tong et al also found mitophagy was activated but insu cient for the maintenance of mitochondrial function during the early phase of diabetic cardiomyopathy [40]. Our current study demonstrated mitophagy is suppressed in both high-fat diet-fed streptozotocin induced type 2 diabetic rats and high glucose-high fat medium in vitro. The mitophagy in this study was different from seen in the work of Tong, which was most likely caused by the difference in experimental model. However, a common nding in different models and in clinical settings highlights its importance of mitophagy in diabetic complication. Δψm is global index of mitochondrial function. Pink1 de ciency under diabetic conditions enhanced Drp1-dependent mitochondrial fragmentation, which leaded to mitochondrial membrane depolarization [41]. The surveillance of mitochondrial quality control allows these injured organelles to be recycled by mitophagy  (Table 1). All these might play a pivotal role in the development and progression of diabetes and its complications.
Our previous studies have shown that mitochondrial dysfunction caused by DM leading to aggravate lung IR injury [5,13]. Data presented in current study showed mitophagy was suppressed under diabetic conditions accompanied with mitochondrial membrane depolarization, increased mitochondrial ROS, mitochondrial damage, in ammation and cell apoptosis during lung IR injury. Together this study implied the relationship between the mechanisms underlying the diabetic state aggravating lung IR injury and impaired mitophagy.
SIRT1 has been proposed to maintain mitochondrial homeostasis through regulation of ROS, mitophagy and mitochondrial biogenesis [42]. Recent studies have highlighted the importance of SIRT1 on the treatment of IR injury [11]. We recently assessed the role of SIRT1 in H 2 S's protective effect against lung IR injury in type 2 diabetic rats [13]. APN has been well demonstrated to exert a modulatory effect on mitochondrial biogenesis and mitophagy [16,43]. In previous studies we found that APN protected against lung IR injury in type 2 diabetic rat, the potential intracellular mechanisms by which APN exerts its protective effects remain to be elucidated [17]. Of interest, David et al observed APN protected against IR injury via SIRT1 signaling [44]. Thus, we paid much attention on the relationship between APN and SIRT1. Consistent with our previous report of warm diabetic lung IR model, data present in this study showed that lung SIRT1 signaling was dramatically downregulated under type 2 diabetic conditions, and it was further attenuated by ischemia-reperfusion injury accompany with impaired mitophagy during diabetic lung transplantation.
Our study further found that by enhanced clearance of damaged mitochondrial via mitophagy, APN suppressed mitochondrial depolarization, reduced mitochondrial ROS generation, attenuated mitochondrial injury, decreased in ammation, and diminished apoptosis. An accumulating body of evidence suggested that SIRT1 induced mitophagy, contributing to attenuation of mitochondrial dysfunction and subsequent IR injury [11]. SIRT1 modulates the expression of FoxO1, which promotes mitophagy via Pink1-Parkin pathway to protect against mitochondrial dysfunction under diabetic condition [45]. SIRT1-FoxO1 signaling maintain mitochondrial homeostasis by mediating mitophagy [46]. In current study, we found a change in the Pink1-dependent mitophagy signaling pathway with or without administration of APN, and demonstrated that APN treatment induced pathological improvements accompanied by the upregulation of Pink1-dependent mitophagy during diabetic lung IR injury. We found that APN restore the deacetylation of FoxO1 by SIRT1. As SIRT1 acts upstream of Pink1, we also explored mechanisms underlying the effects of APN on upregulating mitophagy under diabetic lung IR conditions. Our study demonstrated that SIRT1, signaled by APN, contributed to the Pink1-dependent mitophagy, eventually inducing pro-survival signals for the diabetic reperfused lung and mitochondria. However, the molecular motifs that directly control SIRT1-Pink1 interaction during diabetic lung IR remains to be evaluated.
The current study raised additional questions that we have yet to resolve. First, we did not measure the levels of APN in plasma. Hypoadiponectinemia caused by diabetes is correlated with increased risk of IR injury [47]. Second, since APN can exert a modulatory effect on autophagic ux, we cannot formally exclude the possibility that increases in general autophagy may also participates in the protective effect of APN during diabetic lung IR injury. In view of the fact that Wang et al demonstrated APN restored diabetes-induced autophagic ux arrest, thus ameliorates diabetic IR injury [16]. Third, although we showed mitophagy was reduced during diabetic lung IR injury, it remains unknown how mitophagy in uences mitochondrial biogenesis under these conditions. Given that both APN and SIRT1 contribute to mitochondrial biogenesis, additional studies are needed to elucidate the potential roles of the mitochondrial biogenesis pathways in the APN treatment in diabetic lung IR injury. Fourth, we developed a high glucose-high fat medium-induced type 2 diabetic model in vitro and a high-fat diet-fed streptozotocin-induced type 2 diabetic model in vivo to mimic the pathophysiology condition of type 2 DM patients [23], but whether these models completely substitutes for diabetes requires further detailed elucidation.

Conclusion
Since patients with pretransplant DM who underwent lung transplantation frequently increased risk of mortality [3], interventions to attenuate diabetic lung IR injury may contribute to the development of a novel therapeutic strategy for diabetic patients following lung transplantation. Our ndings have demonstrated mitophagy was speci cally increased by APN during diabetic lung IR injury, resulting in the selectively removal of damaged mitochondria, preservation of mitochondrial function, inhibition of ROS, in ammation and apoptosis. The potential mechanisms the roles of APN might involve the upregulation of mitophagy through the activation of SIRT1-Pink1 pathway.

Data availability
The datasets supporting the conclusions of this article are included within the article.