Melatonin attenuates lung ischemia-reperfusion injury through SIRT3 signaling-dependent mitophagy in type 2 diabetic rats

Abstract

Background: Lung ischemia-reperfusion injury (LIRI) remains the major cause of primary lung dysfunction after lung transplantation. Diabetes mellitus (DM) is an independent risk factor for morbidity and mortality following lung transplantation. Mitochondrial dysfunction is recognized as a key mediator in the pathogenesis of diabetic LIRI. Melatonin has been reported to be a safe and potent preserving mitochondrial function agent. This study aimed at investigating the potential therapeutic effect and mechanisms of melatonin on diabetic LIRI. Methods: High-fat-diet-fed streptozotocin-induced type 2 diabetic rats were exposed to melatonin, with or without administration of the SIRT3 short hairpin ribonucleic acid (shRNA) plasmid following a surgical model of ischemia-reperfusion injury of the lung. Lung function, inflammation, oxidative stress, cell apoptosis, and mitochondrial function were examined. Results: The SIRT3 signaling and mitophagy were suppressed following diabetic LIRI. Treatment with melatonin markedly induced mitophagy and restored SIRT3 expression. Melatonin treatment also attenuated subsequent diabetic LIRI by improving lung functional recovery, suppressing inflammation, decreasing oxidative damage, diminishing cell apoptosis, and preserving mitochondrial function. However, either administration of SIRT3 shRNA or an autophagy antagonist 3-methyladenine (3-MA) suppressing mitophagy, and compromised the protective action of melatonin. Conclusion: Data indicated that melatonin attenuates diabetic LIRI through activation of SIRT3 signaling-mediated mitophagy.

Keywords:

• Acute lung injury
• lung ischemia-reperfusion injury
• melatonin
• mitophagy
• SIRT3
• type 2 diabetes mellitus

Introduction

Acute lung injury is commonly encountered in hospital and outpatient settings and remains a leading cause of patient morbidity and mortality. Many medical conditions, such as pulmonary embolism, acute respiratory distress syndrome (ARDS), cardiopulmonary bypass surgery, and lung transplantation can cause ischemia reperfusion–induced lung injuries.¹ The global prevalence of diabetes mellitus (DM) is significantly increasing across all age groups.² Emerging data suggest the lung is a target of diabetic injury, and DM is a significant risk factor for mortality at both 1 and 5 years after lung transplantation.³ Previously, we reported that DM exacerbated lung ischemia-reperfusion injury (LIRI) in type 2 diabetic rats, and mitochondrial dysfunction play a central role in diabetic LIRI.^4,5 To this end, preservation of mitochondrial function under type 2 DM represents a potential therapeutic target for treatment of LIRI.

Mitophagy, specifically responsible for the degradation of damaged mitochondria, is crucial for mitochondrial quality control.⁶ Mitophagy is mediated by two pathways including Parkin-dependent pathway and mitophagy receptor-dependent pathway.⁷ Prolonged hyperglycemia under the diabetic condition contribute to an over-generation of reactive oxygen species (ROS) by the mitochondria, which in turn may result in the development of diabetic complications.⁸ Nonetheless, mitophagy signaling is impaired in diabetes in spite of extensive mitochondrial dysfunction, contributing to the accumulation of dysfunctional mitochondria and exacerbating the damage to tissue.⁹ It is worthwhile noting that mitophagy plays a crucial role in the maintenance of mitochondrial function during diabetic ischemia reperfusion injury.¹⁰ Our previous study showed that DM impaired mitophagy following lung transplantation.⁵ However, the occurrence of mitophagy in warm diabetic LIRI is still not completely unknown.

Melatonin (N-acetyl-5-methoxytryptamine), is synthesized and secreted nocturnally by the pineal gland.¹¹ Convincing evidence demonstrated melatonin has protective effects against a variety of acute lung injuries especially in LIRI.¹² However, the effect and mechanism of melatonin treatment on LIRI under type 2 diabetic conditions are still unknown. SIRT3 is nicotinamide adenine dinucleotide (NAD⁺)-dependent enzymes, which is mainly located in the mitochondria.¹³ In previous studies we found lung SIRT3 signaling was impaired in diabetic LIRI model, while normalizing SIRT3 signaling conferred pulmonary protective effect.⁴ SIRT3 is documented to protect mitochondrial function through regulation of mitophagy and has been demonstrated to regulate a variety of cellular processes in both physiological conditions and pathological stress insults.¹⁴ Moreover, melatonin is able to modulate SIRT3 to preserved mitochondrial function effectively.¹⁵ However, it is currently unknown whether mitophagy could be regulated by melatonin in diabetic LIRI and its specific relationship with SIRT3 signaling in this circumstance.

We herein report that mitochondrial alterations in diabetic LIRI, mitochondrial abnormalities were associated with decreased mitophagy. Melatonin treatment improved mitochondrial function and less lung injury via the restoration mitophagy. Importantly, this mechanism appears a functional role for melatonin in preserving mitochondrial function by SIRT3 signaling-dependent mitophagy in diabetic LIRI experiments. These rescue mechanisms point the way toward a potential therapeutic target for the treatment of type 2 diabetic patients with acute lung injury induced by ischemia reperfusion injury.

Materials and methods

The study was conducted in accordance with the Basic & Clinical Pharmacology & Toxicology policy for experimental and clinical studies.¹⁶

Animals

Pathogen-free male Sprague–Dawley weighing about 200 to 250 g were obtained from the animal care facility of Harbin Medical University. All animal experiments were performed under the Guidelines on National Institutes of Health Guidelines for the Care and Use of Laboratory Animals (NIH Publication No. 85–23, revised, 1996). The animal care and experimental protocols were approved by the Ethical Board of the Second Affiliated Hospital of Harbin Medical University (Harbin, China; No. SYDW2021-100).

Type 2 diabetic rat model

High-fat diet-fed streptozotocin-induced type 2 diabetic rat model was established as described previously.^4,5,17 Diabetes was defined as a fasting blood glucose level above 11.1 mmol/L 72 h after STZ injection. Rats fed the standard laboratory chow were used as the nondiabetic controls.

Rat LIRI model

Lung ischemia-reperfusion model was established as described previously.^4,17 Briefly, the rats were anesthetized with sodium pentobarbital (30 mg/kg) administered intraperitoneally, intubated through a tracheostomy and ventilated with a tidal volume of 10 ml/kg, at a positive end-expiratory pressure of 2 cm H₂O with 40% oxygen. The respiratory rate of 45–55 breaths/minute was adjusted to maintain arterial carbon dioxide tension (PaCO₂) at 35 to 45 mmHg. Their right femoral vein and artery were cannulated with 24-gauge catheters for intravenous injection, blood gas analysis, and blood pressure monitoring. The rats underwent a left thoracotomy. The left hilum was injected with 50 IU heparin and clamped with a non-colliding micro clip to induce ischemia for 90 min with the tidal volume was adjusted to 6 ml/kg. The clip was removed for reperfusion and the ventilation was restored to the initial tidal volume with a positive end-expiratory pressure of 2 cm H₂O, followed by suturing the thoracic cavity. During the observation period, animals were infused with pentobarbital sodium and rocuronium bromide to maintain stable anesthesia and muscle relaxation. After reperfusion for 4 h reperfusion, the animals were sacrificed. Rats in the sham groups underwent the same procedure except for the left lung hilum occlusion.

Experimental groups

The rats were randomly assigned to the following groups (as depicted in Figure 1): sham group (Con + Sham), LIRI group (Con + IR), DM + sham group (DM + Sham), DM + LIRI group (DM + IR), DM + LIRI + melatonin-treated group (DM + IR + M), DM + LIRI + melatonin and SIRT3 shRNA treated group (DM + IR + M + S), and DM + LIRI + melatonin + 3-methyladenine (3-MA) treated group (DM + IR + M + 3MA). SIRT3 shRNA plasmid or scrambled shRNA plasmid was injected into the rat’s lung through 16-gauge oral catheter at a dose of 2 mg/kg 2 days before the surgery as our previously described.⁴ Melatonin (20 mg/kg) was orally administered after the establishment of DM model for 1 week.¹⁸ 3-MA (15 mg/kg) was intraperitoneally administered 30 min before LIRI.⁵

Figure 1. A schematic illustration of the experimental protocols. The left lung ischemia for 90 min and reperfusion for 4h. STZ, streptozotocin; 3-MA, 3-methyladenine.

The SIRT3 shRNA expression vector

The rat SIRT3 was treated as the target gene and shRNA oligonucleotides used pcDNATM6.2-GW/EmGFP-miR as expression vector were designed and synthesized by Gene Pharma (Shanghai, China) as our previously described.⁴ The target-specific sequence was 5′-AATTCGGGTCCTTTGTATCAGCTACGGTTTTGGCCACTGACTGACCGTAGCTGATACAAAGGACCCA-3′ (sense) and

5′-GCCCAGGAAACATAGTCGATGCCAAAAC CGGTGACTGACTGGCATCGACTATGTTTCCTGGGTGGCC-3′ (antisense).

Real-time quantitative PCR

Real-time PCR was performed as described previously.^4,17 Amplification was conducted using the following primers: 5′-GGAAAGTCGCACAGGAGATCATC-3′ (forward) and 5′-AGGTCCCTGGTCAGCCTTAACA-3′ for rat SIRT3. The following thermal cycling protocol was used 95 °C for 3 min, followed by 40 cycles of amplification at 95 °C for 30 s and 62 °C for 40 s.

Western blot analysis

Mitochondrial protein or tissue proteins were separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred onto polyvinyl difluoride membranes. After blocking, the membranes were incubated with primary antibodies against SIRT3 (2627, 1:1000, Cell Signaling Technology), Parkin (2132, 1:1000, Cell Signaling Technology), LC3-II (3868, 1:1000, Cell Signaling Technology), p62 (23214, 1:1000, Cell Signaling Technology), PGC-1α (MA5-32563, 1:1000, Thermo Fisher Scientific), GADPH (8884, 1:1000, Cell Signaling Technology), VDAC (4661, 1:1000, Cell Signaling Technology) at 4 °C overnight. Then, the membranes exposed to the corresponding horseradish peroxidase-conjugated secondary antibody (1:5000, Zhongshan Golden Bridge Biotechnology, Beijing, China) for 2 h. After being washed, the bound antibodies were detected with horseradish peroxidase (HRP)-conjugated secondary antibodies and visualized using an enhanced chemiluminescence reagents. The levels of target protein were quantified by densitometric analysis using Image J software.

Wet-to-dry weight (W/D) ratios and bronchoalveolar lavage fluid (BALF) protein concentrations

The upper section of lung tissues was dissected and weighed immediately. The lung tissues were desiccated at 80 °C for 72 h and weighed. The left lung was lavaged three times with 10 ml/kg of saline. The protein concentrations in BALF samples were measured using a BCA Protein Assay Kit (Beyotime, China).

Histological examination and scoring

The lung tissues were fixed in 4% paraformaldehyde and paraffin-embedded. The tissue sections (5 μm) were stained with hematoxylin and eosin (HE) and examined under a light microscope (BX41, Olympus, Tokyo, Japan), followed by photo imaged. The degrees of airway epithelial cell damage, interstitial edema, neutrophil infiltration, hemorrhage and hyaline membrane formation in individual sections were evaluated as the lung injury score (LIS) normal = 0; minimal change = 1; mild change = 2; moderate change = 3; and severe change = 4.¹⁹ The lung injury was evaluated by 2 pathologists in a blinded manner.

Measurements of inflammatory and oxidative mediators

The levels of plasma tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β) and interleukin-10 (IL-10) were determined by Enzyme-linked immunosorbent assay (ELISA) using specific kits (R&D Systems, Minneapolis, MN, USA). In addition, the levels of myeloperoxidase (MPO) activity in lung tissue samples, malondialdehyde (MDA), superoxide dismutase (SOD) activity were measured using commercial kits (Nanjing Jiancheng, Nanjing, China) according to the manufacturers’ protocols.

Mitochondrial permeability transition pores (MPTPs) and mitochondrial membrane potential

MPTP opening was determined using a purified mitochondrial MPTP detection kit (GMS10095.3, Genmed Scientifics Inc, USA). Increased MPTP opening after more severe damage caused the mitochondria to swell, resulting in additional rapid decreased absorbance at 540 nm. The fluorescent signals were measured in a fluorescence microplate reader (Infinite M200 Pro, Tecan, Switzerland) at 540 nm. Mitochondrial membrane potential was assessed by a JC-1 staining using a kit from Sigma Aldrich (St. Louis, MO) as described previously.^4,5

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. Cells with red nuclear staining were considered positive and all of the cells with DAPI (4′,6-diamidino-2-phenylindole) staining. The apoptosis index was expressed as the ratio of the number of apoptotic nuclei to the total number of nuclei counted.

mtDNA (mitochondrial DNA) quantification

The whole DNA was isolated from BALF and culture media by using the DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocols. mtDNA levels were measured by SYBR-green dye-based RT-PCR assay. Rat NADH dehydrogenase 1 gene (mtDNA): forward 5′-CGCCTGACCAATAGCCATAA-3′ (forward), and 5′- ATTCGACGTTAAAGCCTGAGA-3′. The following thermal cycling protocol was used 95 °C for 3 min, followed by 40 cycles of amplification at 95 °C for 10 s and 55 °C for 30 s. Concentration of BALF mtDNA were converted to copy number via a DNA copy number calculator. mtDNA copy number was estimated according to the following formula: c = Q *V_DNA/V_PCR * 1/V_ext. C means the concentration of mtDNA. V_DNA means the total volume of DNA solution obtained from extraction, and 100 μl in this study; V_PCR means the volume of DNA solution for RT-PCR, and 1 μl in this study; V_ext means the volume of used for extraction, and 100 μl in this study.

Transmission electron microscopy (TEM)

One cubic millimeter fresh tissue was fixed in 2.5% glutaraldehyde at 4 °C overnight. The microstructural damages in the lungs of rats were evaluated by TEM. Mitochondria and osmiophilic multilamellar bodies in type II alveolar epithelial cells were imaged under an electron microscope. The degrees of mitochondrial injury were assessed with the Flameng score.²⁰ The scores corresponding to the criterion were listed as follows: 0, normal structures with intact particles; 1, normal structures with loss of particles; 2, swollen mitochondria but clear matrices; 3, broken cristae and concentrated matrices; 4, extensively destroyed cristae and ruptured membranes.

Statistical analysis

Data are expressed as median (interquartile range). Statistical comparisons of data among groups were compared by the non-parametric Kruskal-Wallis test. All statistical analyses were performed using Graph Pad Prism software version 5.0 (GraphPad Software, San Diego, CA, USA) and a P-value of <0.05 was considered statistically significant.

Results

Melatonin regulated mitophagy activity through SIRT3 signaling during diabetic lung ischemia reperfusion injury

Initially, we detected SIRT3 expression by RT-PCR and western blot to evaluate whether SIRT3 could be silenced by the designed SIRT3 shRNA. As our previously report,⁴ compared with control group, the expression level of SIRT3 protein was significantly decreased in SIRT3 shRNA group (p < 0.05, Figure 2A). The mRNA expression of SIRT3 showed the same trend as the expression level of SIRT3 protein (Figure 2C).

Next, we measured the SIRT3 protein in both diabetic and nondiabetic rats by Immunoblotting analysis. As shown in Figure 2D, concomitant with our previous experiment of warm diabetic LIRI model, lung SIRT3 was significantly downregulated in type 2 diabetic rats, and it was further attenuated by LIRI. Melatonin treatment markedly restored the impaired lung SIRT3 expression caused by diabetic LIRI (p < 0.05, compared with the DM + IR group), while the lung SIRT3 expression was decreased in DM + IR + M + S group (p < 0.05, compared with the DM + IR + M group).

To investigate whether melatonin regulates diabetic lung ischemia-reperfusion induced mitophagy through SIRT3 signaling, the protein levels of mitochondrial LC3-II, Parkin, mitochondrial p62 were examined to assess mitophagy activation. As exhibited in Figure 2D, the Parkin protein levels were decreased in diabetic rats (p < 0.05, compared with the Con + Sham group), the expression level of Parkin attenuated in the DM + IR group (p < 0.05, compared with the Con + IR group). Melatonin treatment dramatically enhanced Parkin expression (p < 0.05, compared with the DM + IR group), while the effect of which was accentuated by SIRT3 knockdown. (p < 0.05, compared with the DM + IR + M group). The conversion of LC3-I to LC3-II is biochemical hallmarks of autophagy activation.²¹ The p62 protein located on mitochondria were dependent on Parkin, and Parkin recruited p62 to mitochondria during mitophagy.^14,22 Similar changes were observed in mitochondrial LC3-II, and mitochondrial p62 exhibited opposite changes compared with the previous levels. Thus, the elevated mitochondrial LC3-II and decreased p62 expression with administration of melatonin supports the hypothesis that melatonin enhances Parkin-mediated mitophagy through SIRT3 signaling during diabetic LIRI. Similar changes were observed in PGC-1α expression compared with the previous levels.

These results indicated Parkin-mediated mitophagy was decreased in response to DM during LIRI, melatonin inhibited diabetic LIRI-mediated mitophagy downregulation through SIRT3 signaling.

Melatonin improved pulmonary function through SIRT3-dependent mitophagy following diabetic LIRI

To determine the effect of melatonin on diabetic LIRI via regulation of mitophagy, autophagy inhibitor 3-MA was used. As presented in Figure 3A, there were no significant between-group differences with respect to PaO₂/FiO₂ at baseline. The PaO₂/FiO₂ in the Con + Sham group and the DM + Sham group showed no difference (p > 0.05). The PaO₂/FiO₂ in the DM + IR group was significantly lower than that in the Con + IR group during reperfusion at 240 min (p < 0.05). Melatonin treatment markedly increased PaO₂/FiO₂ (p < 0.05, compared with the DM + IR group). However, either SIRT3 knockdown or administration 3-MA abolished the effect of melatonin on the PaO₂/FiO₂ (p < 0.05). The wet weight-to-dry weight ratio showed opposite changes compared with the previous levels (Figure 3B).

Figure 2. Melatonin regulated mitophagy activity through SIRT3 signaling following diabetic lung ischemia reperfusion injury. (A) The downregulation of levels of SIRT3 were evaluated by western blots (n = 4 in each group). (B) SIRT3 expression. (C) mRNA expression of SIRT3 (n = 4 in each group). (D) Representative blots. (E) SIRT3 expression. (F) Parkin expression. (G) LC3-II expression. (H) p62 expression. LC3-II, microtubule-associated protein 1 light chain 3 beta. (I) PGC-1α expression. (**P < 0.05 versus Con group, *P < 0.05 versus Con + Sham group, ^{^}P < 0.05 versus Con + IR group, ^#P < 0.05 versus DM + IR group, ^&P < 0.05 versus DM + IR + M group; n = 8 in each group).

As presented in Figure 3C, the grafts in the Con + IR group exhibited leukocyte infiltration, interstitial thickening, intra-alveolar hemorrhage and edema in the alveolar septa and spaces. The lung injury score was higher in the DM + IR group (p < 0.05, compared with the Con + IR group). Administration of melatonin reduced the lung injury scores (p < 0.05, compared with the DM + IR group), and this improvement was attenuated in the DM + IR + M + S group. The lung injury scores in the DM + IR + M + 3MA group were higher than that in the DM + IR + M group (p < 0.05). The concentrations of proteins in BALF exhibited the same trend as the lung injury score (Figure 3E).

These results indicated melatonin might protect the lung function by promoting mitophagy via SIRT3 during diabetic LIRI.

Melatonin alleviated inflammatory, oxidative stress through SIRT3-dependent mitophagy following diabetic LIRI

As shown in Figure 4A, the levels of serum IL-1β and TNF-α were significantly increased in the DM + IR group (p < 0.05, compared with the Con + IR group). Melatonin treatment markedly decreased the levels of serum IL-1β and TNF-α (p < 0.05, compared with the DM + IR group), whereas the effect of which was mitigated by SIRT3 knockdown (p < 0.05, compared with the DM + IR + M group). Administration of 3-MA also increased the levels of serum IL-1β and TNF-α (p < 0.05, compared with the DM + IR + M group). IL-10 showed the opposite changes compared with the previous levels (Figure 4B). The level of MPO, an indicator of neutrophil infiltration exhibited the same trend. Similar changes were observed in level of MDA, an indicator of the oxidative stress response (Figure 4F). Moreover, the antioxidative capacity (activities of SOD) showed the opposite changes compared with the previous levels (Figure 4E).

Figure 3. Melatonin improved pulmonary function through SIRT3-dependent mitophagy following diabetic LIRI. (A) Arterial blood gas analysis. T0–T4 represent the following time points: baseline, end of ischemia, 120 and 240 min after reperfusion. (B) Wet/dry weight ratio. (C) Histologic analysis of lung tissues (scale bars, 100 μm). (D) Lung injury score. (E) Protein concentrations in the BALF. BALF, bronchoalveolar lavage fluid; PaO₂/FiO₂, partial pressure of arterial oxygen (PaO₂)/fraction of inspired oxygen (FiO₂). (*P < 0.05 versus Con + Sham group, ^{^}P < 0.05 versus Con + IR group, ^#P < 0.05 versus DM + IR group, ^&P < 0.05 versus DM + IR + M group; n = 8 in each group).

These results collectively indicated that the effects of melatonin on anti-inflammatory and anti-oxidative actions by promoting mitophagy might be partially mediated by SIRT3 signaling.

Melatonin attenuated cell apoptosis through SIRT3-dependent mitophagy following diabetic LIRI

The apoptosis in lung grafts of was evaluated by TUNEL assay. As shown in Figure 5, the apoptotic index was significantly increased in the DM + IR group compared with apoptosis levels in the Con + IR group (p < 0.05). The percentages of apoptotic cells in DM + IR + M group were significantly lower than that in the DM + IR group (p < 0.05), but the antiapoptotic effect of melatonin was abolished either in the DM + IR + M + S group or the DM + IR + M + 3MA group.

Figure 4. Melatonin alleviated inflammatory, oxidative stress through SIRT3-dependent mitophagy following diabetic LIRI. (A) Serum concentrations of interleukin-1β (IL-1β). (B) Serum concentrations of interleukin-10 (IL-10). (C) Serum concentrations of TNF-α. (D) Lung concentrations of MPO. (E) Lung concentrations of SOD. (E) Lung concentrations of MDA. IL, interleukin; TNF-a, tumor necrosis factor-a; MPO, myeloperoxidase; MDA, malonaldehyde; SOD, superoxide dismutase. (*P < 0.05 versus Con + Sham group, ^{^}P < 0.05 versus Con + IR group, ^#P < 0.05 versus DM + IR group, ^&P < 0.05 versus DM + IR + M group; n = 8 in each group).

These results indicated that the effects of melatonin on anti-apoptotic by promoting mitophagy might be partially mediated by SIRT3 signaling.

Melatonin limited mitochondrial damage and dysfunction through SIRT3-dependent mitophagy following diabetic LIRI

As shown in Figure 6, the images of mitochondria morphology by electron microscopy showed mitochondrial structural integrity in the Con + IR group was damaged, while the mitochondrial Flameng scores of the DM + IR groups were greater than the scores of the Con + IR group (p < 0.05). Melatonin treatment significantly reduced the mitochondrial Flameng scores (p < 0.05, compared with the DM + IR group), whereas the effect of which was attenuated by SIRT3 knockdown (p < 0.05, compared with the DM + IR + M group). Similar, administration of 3-MA also partially abolished these effects (p < 0.05, compared with the DM + IR + M group). MPTP and JC-1 exhibited the same changes compared with the previous levels. mtDNA level was significantly increased in the DM + IR group compared with those in the Con + IR groups in BALF (p < 0.05). Melatonin treatment markedly reversed the elevated mtDNA level in BALF caused by diabetic LIRI (p < 0.05, compared with the DM + IR group). Of note, the benefits of melatonin on mtDNA level in BALF were also reversed by SIRT3 knockdown or 3-MA (p < 0.05, compared with the DM + IR + M group). These results suggested that the protective effects of melatonin against mitochondrial damage and dysfunction by promoting mitophagy might be partially mediated by SIRT3 signaling.

Figure 5. Melatonin attenuated cell apoptosis through SIRT3-dependent mitophagy following diabetic LIRI. (A) Representative in situ detection of lung parenchymal cell apoptosis by TUNEL staining (scale bars, 500 μm). (B) Percentage of TUNEL-positive nuclei. TUNEL, Terminal deoxynucleotidyl transferase dUTP nick end-labeling. (*P < 0.05 versus Con + Sham group, ^{^}P < 0.05 versus Con + IR group, ^#P < 0.05 versus DM + IR group, ^&P < 0.05 versus DM + IR + M group; n = 8 in each group).

Figure 6. Melatonin limited mitochondrial damage and dysfunction through SIRT3-dependent mitophagy following diabetic LIRI. (A) Mitochondrial ultrastructure. Mitochondria (arrows) in type II alveolar epithelial cells were imaged with transmission electron micrographs (scale bars, 5 μm). (B) Flameng score. (C) Determination of MPTP opening. (D) Determination of mitochondrial membrane potential. (E) BALF mtDNA measurements. MPTP, mitochondrial permeability transition pores; MMP, mitochondrial membrane potential (*P < 0.05 versus Con + Sham group, ^{^}P < 0.05 versus Con + IR group, ^#P < 0.05 versus DM + IR group, ^&P < 0.05 versus DM + IR + M group; n = 8 in each group).

Discussion

Here we found that in experimental diabetic LIRI model: (1) diabetic LIRI downregulated the SIRT3 signaling and suppressed mitophagy. (2) melatonin effectively preserved mitochondrial function via inducing the SIRT3-dependent mitophagy, thus ameliorating LIRI under type 2 diabetic conditions.

Melatonin is secreted by the pineal gland, with various therapeutic effects such as antioxidant, preserving mitochondrial function. Interestingly, lower melatonin secretion was independently associated with a higher risk of developing type 2 DM and its complications.²³ Therefore, melatonin has been encouraged to be administered in the treatment of ischemia reperfusion in diabetes.²⁴ We present data to show that melatonin protected against diabetic LIRI and that SIRT3 signaling-dependent mitophagy played a key role in this process. As melatonin membrane receptors (Mel1a receptor and Mel1b receptor) participate in different signaling pathways,²³ further study is required to evaluate the role and mechanism of melatonin membrane receptors in its protective action in diabetic LIRI.

Although the lung is one of the least studied organs in diabetes, numerous studies demonstrate that it is an inevitable target of diabetic complications.^8,25 In diabetic patients, the risk of an exaggerated ischemia reperfusion injury is more serious, associated with worse outcomes than in the nondiabetic population.²⁶ LIRI remains a major factor for the early mortality of lung transplantations. Based on previous data of ours showing that DM aggravated LIRI and mitochondrial dysfunction played a key role in this process.^4,5 Diabetic hyperglycemia result in over-generation of ROS that then may induce injury of mitochondrial DNA, respiratory complex proteins and membrane lipids. All these may contribute to a catastrophic feedforward cycle of oxidative stress and eventually cell death.²⁷ Thus, maintaining mitochondrial quality is critical for cell homeostasis in diabetic complication.²⁸ Clinical and experimental studies have shown that compared to individuals without diabetes, individuals with diabetes exhibit increased vulnerability to organ ischemia reperfusion.²⁹ Impaired lung mitochondrial function is the key elements of LIRI.³⁰ Hyperglycemia exacerbated mitochondrial dysfunction, which contributes substantially to development of LIRI.^4,5 Our data identify the levels of MPTP and MMP, which were regarded as the markers of mitochondrial function. The level of MMP was decreased in diabetic state following LIRI surgery in comparison to non-diabetic state, and MPTP opening was also further increased in diabetic state following LIRI surgery. A plausible explanation could be that the DM induced mitochondrial dysfunction may make lung vulnerable to ischemia reperfusion injury. PGC-1α regulates mitochondrial biogenesis but also has effects on mitochondrial functions beyond biogenesis.^31,32 Moreover, ample evidence has indicated PGC-1α also in the regulation of mitophagy.³³ More robust data concerning the PGC-1α functions as an upstream regulator of SIRT3,³⁴ while suppressing SIRT3 decreased PGC-1α expression and mitochondrial function.³⁵ In our current study, we found melatonin treatment markedly restored the impaired lung PGC-1α expression caused by diabetic LIRI through SIRT3 signaling, which indicate melatonin preserving mitochondrial function partially mediated by SIRT3 signaling.

Mitophagy is critical for mitochondrial quality control, which allows autophagosomes to selectively eliminate damaged mitochondria.⁶ Abnormal mitochondria overwhelm the capacity of mitophagy at the stage of ischemia reperfusion injury, which may result in damaged mitochondria accumulate.^36–38 The accumulation of damaged mitochondria can lead to neighboring normal mitochondria injury by propagating injurious signals and inducing MPTP opening and mitochondrial dysfunction, which may contribute to excessive ROS production, and ultimately activation of cell apoptosis.³⁹ Various experimental models of ischemia reperfusion injury have shown beneficial effect of mitophagy on cell survival.^40–42 mtDNA are released by damaged mitochondria during LIRI contributing to primary graft dysfunction and lung injury.⁴³ Increased mtDNA is a consequence of impaired mitophagy.⁴⁴ The data presented in the current study showed that mtDNA level in BALF was significantly increased in diabetic state following LIRI surgery in comparison to non-diabetic state. Melatonin attenuated mtDNA level through SIRT3 signaling-dependent mitophagy following diabetic LIRI. Mitophagy can be regulated by several pathways, and Parkin-dependent pathway is one of the best characterized pathways mediating mitophagy.⁴⁵ Intriguingly, evidence is accumulating for the relationship between impaired mitophagy and mitochondrial dysfunction under diabetic condition.⁴⁶ In particular, recent studies have demonstrated mitophagy through Parkin degradation mechanisms is impaired under type 2 diabetic condition, which contribute to the accumulation of damaged mitochondria.⁴⁷ Concomitant with our previously diabetic lung transplantation model, we present data to show that Parkin-mediated mitophagy was suppressed associated with increased MPTP opening, decreased mitochondrial membrane potential, and mitochondrial damage cell death was increased in diabetic LIRI setting, which implied the relationship between the mechanisms underlying the diabetic state aggravating LIRI and impaired mitophagy, and the detailed mechanisms remains to be elucidated.

It has been well established that melatonin could preserve mitochondrial function in diabetic state in multiple organs.^48,49 Melatonin presumably enters mitochondria through oligopeptide transporters.⁵⁰ Measurement of the subcellular distribution of melatonin showed that the concentration of melatonin in mitochondria significantly exceeds that in blood.⁵¹ Moreover, melatonin also modulates Parkin-mediated mitophagy activity.⁹ Herein, we observed melatonin treatment effectively rescued Parkin-mediated mitophagy and protected against diabetic LIRI. We further found that the pulmonary protective effect against diabetic LIRI of melatonin mediated mitophagy was compromised by 3-MA, a broad autophagy inhibitor. Taken together, these results provided evidence that melatonin exerts protective effects against diabetic LIRI by increasing mitophagy.

SIRT3, a nicotinamide adenine dinucleotide (NAD⁺)-dependent deacetylase, is mainly localized in mitochondria.¹³ SIRT3 modulates global mitochondrial lysine acetylation, and preserving mitochondrial function.⁵² We and others also found that SIRT3 signaling might represents a promising strategy to attenuated diabetic ischemia reperfusion insult.^4,15 We found that lung SIRT3 signaling was dramatically decreased under type 2 diabetic condition, and it was further downregulated by the LIRI. Intriguingly, accumulating evidence suggests that melatonin can act as a SIRT3 inducer in multiple systems.^15,53 Moreover, it was reported that a member of the Forkhead box, subgroup O (FoxO) transcription factors FoxO3, regulates PINK1 transcription.⁵⁴ SIRT3 is able to upregulate FoxO3a dependent gene expression via interacts with the daf-16 homolog Foxo3a in the mitochondria.⁵⁵ The network of SIRT3-FoxO3-PINK1-Parkin was demonstrated to control mitophagy.⁵⁶ Here, we found that melatonin treatment can rescue the impaired SIRT3 caused by diabetic LIRI. We also found SIRT3 inhibition impaired the Parkin-mediated mitophagy activity of melatonin and attenuated its protective action against diabetic LIRI. These data suggested that SIRT3-Parkin-mediated mitophagy reducing inflammation, oxidative stress, apoptosis and preserving mitochondrial function, which played a key role in the protective actions of melatonin in diabetic LIRI.

The present study has several limitations. First, we did not measure the levels of melatonin in plasma. Several studies found that circulating levels of melatonin are decreased in both patients with type 2 DM and diabetic animal models, which might contribute to diabetic complications.⁵⁷ Second, high fat-diet-fed streptozotocin-induced type 2 diabetic model was used to simulated pathophysiology condition of type 2 DM patients in present study, but whether these models are truly representative of diabetes requires further detailed elucidation. Third, although we demonstrate that melatonin enhanced LC3-II in the mitochondrial fraction, indicating that melatonin promotes mitophagy, we cannot formally exclude the possibility that increases in general autophagy may also contribute to the protective effect of melatonin during diabetic LIRI.

Our work supports the theory suggesting that melatonin treatment protected against diabetic LIRI by attenuating inflammation, oxidative stress, apoptosis and preserving mitochondrial function via SIRT3-mediated mitophagy. It also illustrates the potential for a personalized medicine approach where the presence of melatonin may identify as SIRT3 enhancement therapy for treating acute lung injury induced by ischemia reperfusion injury in patients with type 2 DM.

Disclosure statement

The authors declare that they have no conflict of interest.

Data availability statement

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

Additional information

Funding

This study was supported by National Natural Science Foundation of China (No. 82001488), Heilongjiang Postdoctoral Fund (No. LBH-Z19172), and The Postgraduate Research Innovation Fund of Harbin Medical University (No. YJSCX2020-44HYD).