Edaravone promotes viability of random skin flaps via activating PI3K/Akt/mTOR signalling pathway‐mediated enhancement of autophagy

Abstract

Random skin flap transplantation is a commonly used technique. However, ischemia and ischemia–reperfusion injury always impair its therapeutic effectiveness through acclerating oxidative stress, apoptosis and suppressing angiogenesis. To survive, cells rely on mediating autophagy, DNA repair, immunoregulation to resist these cellular injuries. Thus, mediating autophagy may affect the survival of random skin flaps. The edaravone (EDA), a oxygen radicals scavenger, also possesses autophagy mediator potential, we investigated the effects of EDA on skin flap survival and its autophagy‐related mechanisms. In vivo, mice were administered EDA or saline intraperitoneally for 7 days postoperatively. We found that EDA ameliorated the viability of random skin flaps, promoted autophagy and angiogenesis, attenuated apoptosis and oxidative stress. In vitro, mouse umbilical vascular endothelial cells (MUVECs) were administered EDA or 3‐methyladenine (3‐MA, an autophagy inhibitor) or rapacymin (Rapa, an autophagy activator) at the beginning of oxygen glucose deprivation (OGD). We found that EDA promoted cell viability, activated autophagy, enhanced angiogenesis, alleviated apoptosis and oxidative stress. On one hand, 3‐MA reversed the effects of EDA on cell viability, oxidative stress and apoptosis via inhibiting autophagy. On the other hand, Rapa had the similar effects of EDA. Furthermore, EDA‐induced autophagy was mediated through downregulating PI3K/Akt/mTOR signalling pathway. The findings showed that EDA ameliorated viability of random skin flaps by promoting angiogenesis, suppressing oxidative stress and apoptosis, which may be mediated by autophagic activation through downregulating PI3K/AKT/mTOR signalling pathway.

1. INTRODUCTION

Random skin flap transplantations have been extensively applied in clinical medicine to treat and repair cutaneous defects, refractory wounds and large tumour resections, etc. ¹ , ² There is no axial vessels existing in random skin flaps and its blood supply is only from the pedicle bed, so that the length‐to‐width ratio of random skin flaps should not exceed 1.5–2.0, ³ which is a serious limitation for the clinical application of such types of skin flaps. Thus, there is a high likelihood that the distal parts of random skin flaps will suffer from necrosis to varying degrees if the length‐to‐width ratio of skin flaps exceed 1.5–2.0 because of under ischemic conditions. ⁴ , ⁵ Furthermore, ischemia injury is an inevitable pathological process during random skin flap transplantation. ⁶ It is commonly acknowledged that inflammation, oxidative stress and apoptosis are the main mechanisms of ischemia‐referfusion injury, which are important causes of necrosis. ⁷ Hence, it is of a great significance to seek a solution to prevent random skin flaps from necrosis.

Edaravone (EDA) is publicly a type of oxygen radicals scavenger and has advantages that it could ameliorate cerebral ischemia, ⁸ enhance the angiogenesis in retinal vein occlusion, ⁹ refractory wounds ¹⁰ and myocardial infarction. ¹¹ Generally, during 24–48 h after flap transplantation, recipient veins and donor veins begin to inosculate. Subsequently, 2–4 days after flap transplantation, the blood reperfusion can be observed and newborn vessels gradually become mature and possess basal functions. Finally, during 5–7 days postoperatively, angiogenesis completes on the whole, thus parabiosis flaps or avulsion flaps survive. ¹² Moreover, Zhang YD et al reported that EDA could attenuate the skin flaps ischemia–reperfusion injury in rats and its mechanism was suppressing oxidative stress and inflammation. ¹³ Therefore, various evidences suggested that it was of a great significance to investigate the effective roles of EDA on promoting the survival of random skin flaps.

Autophagy is a significant cellular organelle and macromolecular materials degradation process for regulating cellular signalling, maintaining cellular homeostasis, and promoting cell survival in the pathological environments such as ischemia, oxygen deficiency, inflammation and oxidative stress et al. ¹⁴ , ¹⁵ Many studies had found that mediating autophagy could promote the survival of random skin flaps. ¹⁶ , ¹⁷ , ¹⁸ However, when ischemia occurs, the roles and changes in autophagy in survival zones, intermediate zones and necrosis zones are still unknown. In survival zones, autophagy is maintained at the normal level to ensure the proper function of metabolism. ¹⁹ In necrosis zones, severe ischemia causes metabolism cease and cell death, subsequently autophagy disappears because all the cell lose its energy. ²⁰ Remarkably, in intermediate zones, autophagy is the most complex and in dynamic changes. At the beginning of ischemia, autophagy level increases rapidly and markedly to cause autophagic death in order to attenuate harmful stress response. ²¹ In the middle of ischemia, high autophagy level increases the consumption of cellular energy and accelerates cellular death. ²² At this important point, if intermediate zones of skin flaps are rescued properly, the survival rate can increase significantly. However, if skin flaps ischemia continues without treatments, intermediate zones are doomed to become necrosis zones eventually. ²³ Therefore, enhancing the ischemia tolerance of intermediate zones by mediating autophagy is significant for promoting the survival of random skin flaps. To clarity this scientific theory, we hypothesized that EDA could promote the survival rate of random skin flaps by activating autophagy. Hence, we performed this research to verify our hypothesis.

2. MATERIALS AND METHODS

2.1. Reagents and chemicals

EDA (purity = 99.59%), 3‐Methyladenine (3‐MA, purity = 99.83%) and Rapamycin (Rapa, purity = 99.77%) were purchased from MedChem Express (New Jersey, US). Primary antibodies against transforming growth factor‐β1 (TGF‐β1), vascular endothelial growth factor (VEGF) and matrix metalloproteinase 9 (MMP9) were purchased from Wanleibio (Shenyang, China). Heme oxygenase‐1 (HO‐1), NAD(P)H quinone oxidoreductase 1 (NQO1) and β‐actin antibodies were purchased from ABclonal Technology (Wuhan, China). LC3, Beclin1, p62 and Atg5 antibodies were purchased from Affinity Biosciences (OH, US). Cleaved‐Caspase3, Bcl‐2, Bax, p‐PI3K, PI3K, p‐AKT, AKT, p‐mTOR and mTOR antibodies were purchased from Cell Signalling Technology (Danvers, MA). RPMI 1640 medium was purchased from Gibco (Grand Island, NY) and fetal bovine serum was purchased from Animal Blood Ware (Shanghai, China).

2.2. Establishing animal model of dorsal random skin flaps

Sixty healthy BALB/c mice (male and female, 18.0–25.0 g) were obtained from Lanzhou Veterinary Research Institute. Otherwise, the animal experiments were approved by the Animal Welfare and Ethics Committee of Lanzhou University Second Hospital (D2021‐256). All the mice were separately fed a standard and light–dark alternative animal experimental room. Before the operation, 2.0% pentobarbital sodium was intraperitoneally injected into mice (45.0 mg/kg) for anesthetization. Then, a full‐thickness random skin flap (7.0 mm × 50.0 mm) was elevated on the back of each mouse. ²⁴ Next, the bilateral sacral axial arteries that provide blood supply to the random skin flap were cut off completely. Finally, the elevated flap was stitched in place with 6–0 monofilament nylon immediately.

2.3. Investigating the most suitable effective dose of EDA

To investigate the most suitable effective dose of EDA in vivo, forty mice were divided into control group, EDA (10.0 mg/kg, 20.0 mg/kg and 40.0 mg/kg once daily) groups and papaverine group (positive control, 10.0 mg/kg once daily) randomly (n = 8 per group). On the first day after the operation, mice in EDA group were injected intraperitoneally with EDA (10.0, 20.0 and 40.0 mg/kg once daily) until all the mice were sacrificed. Mice in papaverine group were injected intraperitoneally with 10.0 mg/kg papaverine. Mice in control group were injected intraperitoneally with an equivalent dosage of PBS solution. All the mice were sacrificed with overdose of chloral hydrate intraperitoneal injection 7 days postoperatively and skin flap samples of intermediate zones were harvested.

2.4. General evaluation and calculation of random skin flaps survival

Evaluation indicators include colour, dermatoglyph and texture, secretion, swelling, acupuncture and bleeding reaction. On the 7th day postoperation, images of skin flaps were taken by the Canon Digital Camera to calculate the survival rate by Image J software. The survival rate of the random skin flaps was calculated by the following formula: survival rate = survival area/total area×100%. The survival skin flaps were presented with ruddy, soft and haired. The necrotic skin flaps were presented with pale, swollen, hard, dark, contracted and hairless.

2.5. H&E staining and Masson's trichrome staining

On the 7th day postoperation, eighteen 5.0 mm × 5.0 mm skin flap samples (three samples per group) of the intermediate zones were harvested after sacrifice. After being fixed in 4% paraformaldehyde for at least 24 h, all the samples were embedded in paraffin, sectioned transversely and then stained with H&E staining and Masson's trichrome staining agents. Under the light microscope (Olympus, Tokyo, Japan), changes of organisational structures were assessed on H&E stained slides and the arrangement and content of collagen were assessed on Masson's trichrome stained slides. Nine random fields from three random slides were selected for photography and analysis.

2.6. Terminal deoxynucleotidyl transferase‐mediated dUTP nick end labeling (TUNEL) staining

Skin flap samples from the intermediate zone were fixed in 4% formalin, embedded in paraffin. The terminal deoxynucleotidyl transferase‐mediated dUTP nick end labeling (TUNEL) assay was performed according to the manufacturer's instructions (Servicebio, Wuhan, China). Nine random fields from three random slides were selected for photography and analysis.

2.7. Biochemical assays

Assay testing kits of malondialdehyde (MDA) and superoxide dismutase (SOD) were purchased from Nanjing Jiancheng (Nanjing, China). Samples were homogenised, centrifuged and measured according to the manufacturer's instructions respectively.

2.8. Immunohistochemistry (IHC)

Skin flap samples from the intermediate zone were deparaffinised and rehydrated by xylene and graded series of ethanol respectively. Next, sodium citrate buffer (pH = 6.0) and 3% hydrogen peroxide solution were applied to retrieve the antigen and block endogenous peroxidase. After blocking by 5% bovine serum albumin (BSA) for 1 h, the slides were incubated with the following primary antibodies: CD31 (1:50), VEGF (1:100), MMP9 (1:100), LC3B (1:100) and Atg5 (1:100) overnight in the wet box at 4°C. Then, all the slides were incubated with the secondary antibody for 50 min. Finally, diaminobenzidine (DAB) detection kit was applied to the slides. CD31 positive vessel density counting: Nine random fields from three random slides were counted by three senior pathologists independently. Otherwise, positive cell area rate of VEGF, MMP9, LC3B and Atg5 was calculated by Image J software on IHC stained slides.

2.9. Cell culture and oxygen glucose deprivation (OGD)

Mouse umbilical vascular endothelial cells (MUVECs), which were presented by Cardiovascular Medicine of Lanzhou University Second Hospital, were cultured in RPMI 1640 medium (supplemented with 10% fetal bovine serum and 1% penicillin and streptomycin) with 5% CO₂ at 37°C. MUVECs were exposed to OGD and glucose‐free and serum‐free medium to mimic the ischemic and hypoxic conditions. To investigate the most suitable duration of OGD, cells were placed in a hypoxia chamber containing 0% O₂ and 100% CO₂ for 0, 0.5, 1.0, 1.5, 2.0, 2.5 and 3.0 h respectively.

2.10. Cell viability assay

Cell counting kit (CCK‐8) assay was applied to assess the cell viability. Firstly, the number of mouse vascular endothelial cells suspension was accounted and planted into 96‐well plates. Secondly, after incubating for 24 h at the incubator (37°C, 5% CO₂), each well was added and co‐incubated with 10 μL CCK‐8 regent at the incubator (37°C, 5% CO₂) for 4 h. Finally, we applied the microplate reader (Bio‐Rad, USA) to assess the absorbance at a wavelength of 450 nm.

2.11. Flow cytometry

To detect the apoptotic cell rate, flow cytometry was performed by the Annexin V‐FITC/PI Apoptosis Detection Kit (Bipsharp, China) according to the instructions. BECKMAN COULTER CytoFLEX was applied to measure and analyse the data.

2.12. Hoechst 33258 staining

MUVECs were stained with Hoechst 33258 to evaluate nuclear morphological changes after OGD. After exposure to OGD, cells were stained with Hoechst 33258 (5.0 mg/mL, Solarbio) for 30 min. Afterwards, stained cells were washed 3 times with PBS and observed under a fluorescent microscope. Random fields at 400× magnification were selected for photography.

2.13. ROS measurement

To detect reactive oxygen species (ROS) generation level, ROS Detection Kit (Bipsharp, Shanghai, China) was performed according to the instructions. ROS probe was dissolved in basal RPMI 1640 medium to 10 μM and added to the cells. After the incubation for 30 min at 37°C, cells were washed with basal RPMI 1640 medium 3 times and then observed under a fluorescent microscope. Random fields at 400× magnification were selected for photography.

2.14. Quantitative real‐time polymerase chain reaction (qRT‐PCR)

The total RNA of skin flap samples of intermediate zones and cells was extracted by the Tizol reagent, and then cDNA was reversely transcribed applying the Evo M‐MLV RT Mix Kit (Accurate Biotechnology, Hunan, China). Primers were designed and synthesised by Tsingke Biotechnology (Beijing, China). The following primer sequences are:

LC3B 5′‐CCACCAAGATCCCAGTGATTAT‐3′ (forward) and 5′‐TGATTATCTTGATGAGCTCGCT‐3′ (reverse);

Beclin1 5′‐GTCAGCTCTCGTCAAGGCG‐3′ (forward) and 5′‐CGCCTTAGACCCCTCCATTC‐3′ (reverse);

p62 5′‐AACACCAAGAGCTCGGACAG‐3′ (forward), and 5′‐ TGGAGCTCCCCATGTCCATA‐3′ (reverse);

Atg5 5′‐TGCATCAAGTTCAGCTCTTCCT‐3′ (forward), and 5′‐CTGGGTAGCTCAGATGCTCG‐3′ (reverse);

PI3K 5′‐AAACAAAGCGGAACCTATTG‐3′ (forward) and 5′‐TAATGACGCAATGCTTGACTTC‐3′ (reverse);

AKT 5′‐TGCACAAACGAGGGGAATATAT‐3′ (forward) and 5′‐CGTTCCTTGTAGCCAATAAAGG‐3′ (reverse);

mTOR 5′‐CTGATCCTCAACGAGCTAGTTC‐3′ (forward), and 5′‐ GGTCTTTGCAGTACTTGTCATG‐3′ (reverse);

β‐actin 5′‐CATCCGTAAAGACCTCTATGCCAAC‐3′ (forward), and 5′‐ATGGAGCCACCGTCCACA‐3′ (reverse).

Next, the qRT‐PCR reactions were performed by the SYBR Green qPCR Mix (Monad Biotechnology Company) on the Bio‐Rad CFX96 (BIO‐RAD, USA). According to the protocol of qRT‐PCR, the parameters were set as followed: Reverse transcription (1 cycle) at 95°C for 30 s, denaturation (1 cycle) at 95°C for 10 s and annealing (40–45 cycles) at 60°C for 30 s, followed by extension at 72°C for 30 s. Finally, the mRNAs of LC3, Beclin1, p62, Atg5, PI3K, Akt and mTOR were normalised to the mRNA of β‐actin.

2.15. Western blotting analysis

The skin flap samples of intermediate zones and cells were homogenised in the lysis buffer with the inhibitors of protease and phosphatase. The protein concentration of each sample was determined by the BCA assay. Then, the total protein was separated by 12% (w/v) gel electrophoresis and blotted to 0.45 μm polyvinylidene difluoride membranes (Solarbio, Beijing, China). Subsequently, the membranes were incubated with the following primary antibodies: TGF‐β1 (1:500), MMP9 (1:1500), VEGF (1:1000), HO‐1 (1:1000), NQO1 (1:1000), Cleaved‐caspase3 (1:1000), Bcl‐2 (1:1000), Bax (1:1000), LC3II (1:500), Beclin 1 (1:1000), p62 (1:1000), Atg5 (1:1000), PI3K (1:1000), p‐PI3K (1:1000), AKT (1:1000), p‐AKT (1:1000), mTOR (1:1000), p‐mTOR (1:1000) and β‐actin (1:10000). Afterwards, the membranes were incubated with HRP‐conjugated immunoglobulin G secondary antibody (1:10000) for 1 h and visualised by the ECL Plus kit (Biosharp, Shanghai, China). Finally, Image J software was applied to quantify the relative protein expression levels.

2.16. Transmission electron microscope (TEM)

Cell samples were fixed with 10% glutaraldehyde overnight and then fixed with 1% osmic acid for 2 h. After washing with PBS, samples were rehydrated by graded series of ethanol and acetone. Next, immersed and embedded in ethoxyline resin, ultrath slides were made by the ultramicrotome. Finally, the slides were double stained by uranyl acetate and lead citrate, and then measured by transmission electron microscope (HITACHI, Tokyo, Japan).

2.17. Immunofluorescence staining (IF)

Cell samples were fixed with paraformaldehyde and then were punched by Triton. After blocking by 5% BSA for 1 h, slides were incubated with the following primary antibodies: LC3B (1:100) overnight in the wet box at 4°C. Then, all the slides treated with the secondary antibody and DAPI staining.

2.18. Statistically analysis

All the experiments were performed at least three times and the data were expressed as means ± standard deviation (SD). GraphPad Prism 8.0 software was applied to compare the data between two groups by independent sample t‐test. P < .05 was considered statistically significant.

3. RESULTS

3.1. EDA ameliorates viability of random skin flaps

EDA promoted viability of random skin flaps in a dose‐dependent manner and 20.0 mg/kg EDA had the best therapeutic effect (Figure 1A‐C). Thus, 20.0 mg/kg was determined as the following EDA concentration. On day 3 after operation, the necrosis was observed in the distal parts of the flaps, which presented with pale, swollen, hard, dark, contracted and hairless (Figure 2A). However, there was no significant difference between EDA group and control group on day 3 after operation (89.41 ± 4.33% and 86.77 ± 6.75%, P > .05, Figure 2B). On day 7 after operation, the necrosis spread to the proximal end (Figure 2C). Compared with control group, EDA group showed a significantly high flap survival percentage (81.13 ± 8.65% and 62.39 ± 12.53%, P < .01, Figure 2D). HE results showed that EDA inhibited the inflammatory cell infiltration and hyperemia (Figure 2E). Moreover, the inner sides of the flaps were photoed. Compared with EDA group, the distal parts of the flaps in control group were swollen, edematous and bruised with more observed venous blood congestion (Figure 2F).

FIGURE 1.

Dose‐dependent effect of EDA on promoting survival of random skin flaps. (A) Therapeutic effects of different concentrations of EDA and papaverine on random skin flaps on day 7 after operation. (B) Histogram of survival percentage of random skin flaps on day 7 after operation. (C, scale bar: 100 μm) HE staining showed flaps of the five groups on day 7 after operation. Data were shown as means ± SD. *P < .01 versus Control group (n = 8 per group).

FIGURE 2.

EDA ameliorates viability of random skin flaps. (A,C) Macroscopic photographs of random skin flaps on days 3 and 7 after operation. (B,D) Histogram of survival percentage of random skin flaps on day 3 and 7 after operation. (E, scale bar: 100 μm) HE staining of flaps on day 7 after operation. (F) Macroscopic photographs of the inner face of flaps. Data were shown as means ± SD. *P < .05 and **P < .01 versus Control group (n = 10 per group).

3.2. EDA promotes the formation of microvessels and array of collagenous fibre in random skin flaps

IHC for CD31 was performed to quantify microvessel density (MVD) in flaps. The number of CD31 positive vascular endothelial cells in EDA group was more than that in control group (140.87 ± 6.71/mm² and 91.23 ± 9.32/mm², P < .01, Figure 3A,B). To evaluate the array and density of collagenous fibre in random skin flaps, Masson's trichrome was performed. Compared with control group, collagenous fibre was more orderly, closely and evenly arranged in EDA group (Figure 3C).

FIGURE 3.

EDA promotes the formation of microvessels and array of collagenous fibre in random skin flaps. (A, scale bar: 50 μm) IHC of CD31 positive cells in flaps on day 7 after operation. (B) Histogram of CD31 positive vessel density. (C, scale bar: 100 μm) Masson staining of flaps on day 7 after operation. Data were shown as means ± SD. *P < .05 and **P < .01 versus Control group (n = 10 per group).

3.3. EDA promotes angiogenesis in random skin flaps

To evaluate the level of angiogenesis level in random skin flaps, IHC of expression levels of VEGF and MMP9 was performed. EDA significantly promoted VEGF and MMP9 expression compared with that in control group (P < .01, Figure 4A‐D). Furthermore, western blotting was performed to evaluate the levels of TGF‐β1, VEGF and MMP9 in random skin flaps (Figure 4E). Compared with control group, the protein expression of TGF‐β1, VEGF and MMP9 was significantly upregulated in EDA group (P < .05, Figure 4F).

FIGURE 4.

EDA promotes angiogenesis in random skin flaps. (A, scale bar: 50 μm) IHC of VEGF expression in flaps. (B) Histogram of VEGF positive area rate. (C, scale bar: 50 μm) IHC of MMP9 expression in flaps. (D) Histogram of MMP9 positive area rate. (E) Western blotting of TGF‐β1, VEGF and MMP9 in the control and EDA groups. (F) Histograms of expression levels of TGF‐β1, VEGF and MMP9 in the two groups. Data were shown as means ± SD. *P < .05 and **P < .01 versus Control group (n = 10 per group).

3.4. EDA attenuates apoptosis in random skin flaps

TUNEL staining was performed to evaluate cell apoptosis in random skin flaps. The number of apoptotic cells was significantly decreased in EDA group compared with that in control group (P < .01, Figure 5A,B). Furthermore, western blotting was performed to evaluate the levels of Cleaved‐Caspase3, Bcl‐2 and Bax in random skin flaps (Figure 5C). Compared with control group, the protein expression of Cleaved‐Caspase3, Bcl‐2 and Bax was significantly upregulated in EDA group (P < .01, Figure 5D).

FIGURE 5.

EDA attenuates apoptosis in random skin flaps. (A, scale bar: 100 μm) TUNEL staining of flaps on day 7 after operation. (B) Histogram of TUNEL positive area rate. (C) Western blotting of Cleaved‐Caspase3, Bcl2 and Bax in the control and EDA groups. (D) Histograms of expression levels of Cleaved‐Caspase3, Bcl‐2 and Bax in the two groups. Data were shown as means ± SD. *P < .05 and **P < .01 versus Control group (n = 10 per group).

3.5. EDA ameliorates oxidative stress in random skin flaps

Oxidative stress in random skin flaps was evaluated by content analysis of MDA and SOD, western blotting of HO‐1 and NQO1 expression. The content of MDA and SOD was significantly higher in EDA group than that in control group (P < .05, Figure 6A,B). Furthermore, western blotting results showed that compared with control group, the protein expression of HO‐1 and NQO1 increased significantly in EDA group (P < .05, Figure 6C,D).

FIGURE 6.

EDA ameliorates oxidative stress in random skin flaps. (A) Histogram of the MDA content measured by the TBA method. (B) Histogram of SOD activities evaluated by the WST‐1 method. (C) Western blotting of HO‐1 and NQO1 in the control and EDA groups. (D) Histograms of expression levels of HO‐1 and NQO1 in the two groups. Data were shown as means ± SD. *P < .05 and **P < .01 versus Control group (n = 10 per group).

3.6. EDA upregulates autophagy in random skin flaps

Autophagy markers including LC3, Beclin1, Atg5 and p62, which were detected by IHC, qRT‐PCR and western blotting. Both the transcription and expression levels of LC3II, Beclin1 and Atg5 were significantly higher in EDA group compared with that in control group (P < .05, Figure 7A‐G). On the other hand, the transcription and expression level of p62 was significantly lower in EDA group compared with that in control group (P < .05, Figure 7E‐G).

FIGURE 7.

EDA upregulates autophagy in random skin flaps. (A, scale bar: 50 μm) IHC of LC3B expression in flaps. (B) Histogram of LC3B positive area rate. (C, scale bar: 50 μm) IHC of Atg5 expression in flaps. (D) Histogram of Atg5 positive area rate. (E) Transcripts of LC3B, Beclin1, p62 and Atg5 genes detected by qPCR. (F) Western blotting of LC3, Beclin1, p62 and Atg5 in the control and EDA groups. (G) Histograms of expression levels of LC3, Beclin1, p62 and Atg5 in the two groups. Data were shown as means ± SD. *P < .05 and **P < .01 versus Control group (n = 10 per group).

3.7. EDA promotes viability of MUVECs exposed to OGD

CCK‐8 results showed that compared with normal group, cell viability of MUVECs was decreased to 50% after 2.5 h OGD exposure (Figure 8A). Thus, 2.5 h was determined as the following OGD exposure duration. Moreover, CCK‐8 results also showed that EDA promoted viability of MUVECs exposed to OGD in a dose‐dependent manner and 20 μM EDA had the best therapeutic effect (P < .05, Figure 8B). Thus, 20 μM was determined as the following EDA concentration.

FIGURE 8.

EDA promotes viability of MUVECs exposed to OGD. (A) The effects of different durations of OGD on the survival rate of MUVECs detected by CCK‐8. (B) The effects of different concentrations of EDA on the survival rate of MUVECs exposed to 2.5 h OGD. Data were shown as means ± SD. *P < .05 and **P < .01 versus Control group.

3.8. EDA activates autophagy in MUVECs exposed to OGD

To evaluate whether EDA activates autophagy in MUVECs exposed to OGD, we combined EDA with rapamycin (an effective activator of autophagy) and 3‐MA (an effective inhibitor of autophagy) to perform TEM for observing autophagosomes, western blotting for LC3, Beclin1, Atg5 and p62 and IF for LC3. Compared with OGD group, autophagosomes in EDA group and rapamycin group were more numerous while autophagosomes in 3‐MA group were fewer (Figure 9A). Moreover, compared with rapamycin group, autophagosomes in EDA + rapamycin group significantly increased, and compared with 3‐MA group, that in EDA + 3‐MA group significantly increased.

FIGURE 9.

EDA activates autophagy in MUVECs exposed to OGD. (A) TEM of autophagosomes in MUVECs in the seven groups (magnification, ×6000, ×12 000). Black arrows indicating autophagosomes with typical double‐membrane structures containing organelle remnants. (B,D) Western blotting of LC3II, Beclin1, p62 and Atg5 in the control, OGD, EDA, 3‐MA, Rapa, EDA + 3‐MA and EDA + Rapa groups. (C,E) Histograms of expression levels of LC3II, Beclin1, p62 and Atg5 in the seven groups. (F, scale bar: 100 μm) IF of LC3B expression in MUVECs. Data were shown as means ± SD. *P < .05 and **P < .01 versus Control group. #P < .05 and ##P < .01 versus OGD group. &P < .05 and &&P < .01 versus 3‐MA group. +P < .05 and ++P < .01 versus Rapa group.

As for LC3, Beclin1 and Atg5, results showed that compared with OGD group, protein expression levels of LC3, Beclin1 and Atg5 were significantly increased in EDA group and rapamycin group (P < .05, Figure 9D,E), while that in 3‐MA group were significantly decreased (P < .05, Figure 9B,C). And compared with rapamycin group, protein expression levels of LC3, Beclin1 and Atg5 in EDA + rapamycin group significantly increased, and compared with 3‐MA group, that in EDA + 3‐MA group significantly increased. As for p62, results were contrary to that of LC3, Beclin1 and Atg5. In addition, the results of IF also supported the results mentioned above (Figure 9F).

3.9. EDA promotes viability of MUVECs exposed to OGD through activating autophagy

CCK‐8 results showed that compared with OGD group, rapamycin significantly promoted viability of MUVECs while 3‐MA significantly decreased viability of MUVECs (P < .05, Figure 10A). Furthermore, compared with rapamycin group, EDA + rapamycin group significantly promoted viability of MUVECs. And compared with 3‐MA group, EDA + 3‐MA group significantly decreased viability of MUVECs.

FIGURE 10.

(A) EDA promotes viability of MUVECs exposed to OGD through activating autophagy. The survival rate of MUVECs detected by CCK‐8 in the seven groups. Data were shown as means ± SD. *P < .05 and **P < .01 versus Control group. #P < .05 and ##P < .01 versus OGD group. &P < .05 and &&P < .01 versus 3‐MA group. +P < .05 and ++P < .01 versus Rapa group.

3.10. EDA attenuates OGD‐induced MUVECs apoptosis through activating autophagy

The results of flow cytometry, hoechst staining and western blotting showed that EDA attenuated OGD‐induced MUVECs apoptosis through activating autophagy. Flow cytometry data showed that compared with OGD group, the apoptosis rate in EDA group and rapamycin group were significantly decreased by 12.91% and 11.13% respectively (P < .01, Figure 11A), however, the apoptosis rate in 3‐MA group were significantly increased by 32.67%. Furthermore, compared with 3‐MA group, EDA + 3‐MA group significantly increased by 28.04% (P < .01, Figure 11B). Hoechst 33258 staining data also confirmed similar outcomes of flow cytometry, showing that EDA could attenuate OGD‐induced MUVECs apoptosis through activating autophagy (P < .01, Figure 11C,D). Moreover, the protein expression levels of Cleaved‐Caspase3, Bcl‐2 and Bax after EDA treatment were analysed by western blotting. As for Cleaved‐caspase3 and Bax, results showed that compared with OGD group, protein expression levels of Cleaved‐Caspase3 and Bax were significantly decreased in EDA group and rapamycin group, while that in 3‐MA group were significantly increased (P < .05, Figure 11E‐H). And compared with rapamycin group, protein expression levels of Cleaved‐Caspase3 and Bax in EDA + rapamycin group significantly decreased, and compared with 3‐MA group, that in EDA + 3‐MA group significantly increased (P < .05, Figure 11E‐H). As for Bcl‐2, results were contrary to that of Cleaved‐Caspase3 and Bax.

FIGURE 11.

EDA attenuates OGD‐induced MUVECs apoptosis through activating autophagy. (A) Annexin V/PI double staining was performed to evaluate the rate of apoptosis of MUVECs in the seven groups. (B) Histograms of the apoptosis rate of MUVECs detected by flow cytometry in the seven groups. (C, scale bar: 100 μm) Hoechst 33258 staining was performed to evaluate the rate of apoptosis of MUVECs in the seven groups. Blue arrows indicating damaged nuclei with irregular shapes and strong fluorescence intensity. (D) Histograms of the apoptosis rate of MUVECs detected by Hoechst 33258 staining in the seven groups. (E,G) Western blotting of Cleaved‐Caspase3, Bcl‐2 and Bax in the seven groups. (F,H) Histograms of expression levels of Cleaved‐Caspase3, Bcl‐2 and Bax in the seven groups. Data were shown as means ± SD. *P < .05 and **P < .01 versus Control group. #P < .05 and ##P < .01 versus OGD group. &P < .05 and &&P < .01 versus 3‐MA group. +P < .05 and ++P < .01 versus Rapa group.

3.11. EDA attenuates OGD‐induced MUVECs oxidative stress through activating autophagy

The results of ROS measurement and western blotting showed that EDA attenuated OGD‐induced MUVECs oxidative stress through activating autophagy. Compared with OGD group, the results revealed that the level of ROS were dramatically decreased in EDA group and rapacymin group while that in 3‐MA group was dramatically increased (P < .01, Figure 12A,B). Furthermore, compared with rapamycin group, the level of ROS in EDA + rapamycin group significantly decreased (P < .01, Figure 12A,B), and compared with 3‐MA group, the level of ROS in EDA + 3‐MA group significantly increased (P < .01, Figure 12A,B). As for western blotting, compared with OGD group, protein expression levels of HO‐1 and NQO1 were significantly increased in EDA group and rapamycin group (P < .01, Figure 12E,F), while that in 3‐MA group were significantly increased (P < .01, Figure 12C,D). And compared with rapamycin group, protein expression levels of HO‐1 and NQO1 in EDA + rapamycin group significantly increased (P < .01, Figure 12E,F), and compared with 3‐MA group, that in EDA + 3‐MA group significantly decreased (P < .01, Figure 12C,D).

FIGURE 12.

EDA attenuates OGD‐induced MUVECs oxidative stress through activating autophagy. (A, scale bar: 100 μm) ROS measurement was performed to assess the degree of oxidative stress in MUVECs in the seven groups. (B) Histograms of the ROS generation level of MUVECs detected by ROS measurement in the seven groups. (C,E) Western blotting of HO‐1 and NQO1 in the seven groups. (D,F) Histograms of expression levels of HO‐1 and NQO1 in the seven groups. Data were shown as means ± SD. *P < .05 and **P < .01 versus Control group. #P < .05 and ##P < .01 versus OGD group. &P < .05 and &&P < .01 versus 3‐MA group. +P < .05 and ++P < .01 versus Rapa group.

3.12. EDA activates autophagy by downregulating PI3K/Akt/mTOR signalling pathway

qRT‐PCR and western blotting were performed to evaluate the effect of EDA on PI3K, Akt and mTOR phosphorylation levels in MUVECs exposed to OGD. Compared with OGD group, EDA administration significantly decreased the phosphorylation levels of PI3K, Akt and mTOR in MUVECs (P < .05, Figure 13A‐C).

FIGURE 13.

EDA activates autophagy by downregulating PI3K/Akt/mTOR signalling pathway. (A) Transcripts of PI3K, AKT and mTOR genes in the control, OGD, EDA groups detected by qPCR. (B) Western blotting of p‐PI3K, PI3K, p‐AKT, AKT, p‐mTOR and mTOR in the three groups. (C) Histograms of expression levels of p‐PI3K, PI3K, p‐AKT, AKT, p‐mTOR and mTOR in the three groups. Data were shown as means ± SD. *P < .05 and **P < .01 versus Control group. #P < .05 and ##P < .01 versus OGD group.

4. DISCUSSION

Ischemia and subsequent ischemia–reperfusion injury is inevitable during flap transplantations, organic transplantations, trauma and stroke, etc, and can cause organs harm or even necrosis. Impaired angiogenesis process, excessive apoptosis and oxidative stress have been reported to play pivotal roles in this pathological condition. ²⁵ To fight against these adverse factors, cells must mediate its autophagic level to maintain cellular homeostasis. Therefore, mediating autophagy is a method to promote the survival of random skin flaps, which has been supported by previous studies. ¹⁷ , ¹⁸ As a classical oxygen radicals scavenger, EDA has also been confirmed to possess the capability to mediate autophagy. Hence, we performed studies in vivo and vitro to investigate the role of EDA in promoting the survival of random skin flaps and its autophagy‐related mechanisms (Figure 14).

FIGURE 14.

Proposed scheme for the underlying mechanisms of EDA improving viability of random skin flaps. EDA promotes angiogenesis, inhibits oxidative stress and apoptosis by activating autophagy via downregulating PI3K/Akt/mTOR signalling pathway, and subsequently improves viability of random skin flaps. In addition, 3‐MA reversed the effects of EDA and Rapa had a similar effects of EDA.

We demonstrated that EDA ameliorated viability of random skin flaps by activating autophagy through the downregulation of PI3K/Akt/mTOR signalling pathway, and subsequently promoting angiogenesis, inhibiting oxidative stress and apoptosis. To take a further step, this research clarified that activating autophagy might be the key therapeutic target for treating random skin flaps necrosis, and oxygen radicals scavengers would be the promising drug candidates.

EDA could effectively reduce the necrosis area of random skin flaps, which was also confirmed by HE results. HE results showed that EDA could inhibit hyperemia and keep the structural integrity of random skin flaps. EDA could significantly promote angiogenesis of random skin flaps, and its possible mechanisms were that EDA upregulated the protein expressions of TGF‐β1, VEGF and MMP9, as well as promoted cell viability of MUVECs. Many studies ²⁶ , ²⁷ , ²⁸ had confirmed that TGF‐β1, VEGF and MMP9 are inevitable in the biological process of angiogenesis. Studies ²⁹ , ³⁰ had also confirmed that regulated by VEGF, angiopoietin and fibroblast growth factor, endothelial cells can start to proliferate and migrate to the capillary formation, inducing angiogenesis, which is significant for the proper wound healing. Otherwise, MMP9 can be secreted by fibroblasts, macrophages, neutrophils and epithelial cells to degradate collagen and extracellular matrix, thus contributing to angiogenesis. ³¹ , ³² , ³³ Furthermore, exposed to TGF‐β1, VEGF and MMP9, MUVECs adopt specific phenotypes such as tip MUVECs and stalk MUVECs. ³⁴ Tip MUVECs migrate into the avascular area with the help of their filopodia, thus forming as the anchor of stalk MUVECs. ³⁵ Subsequently, anchored by tip MUVECs, stalk MUVECs start to proliferate for growing sprouts. ³⁶ Collectively, EDA could significantly promote angiogenesis of random skin flaps and promoted cell viability of MUVECs.

EDA could ameliorate viability of random skin flaps by suppressing excessive oxidative stress. And its underlying mechanisms were that EDA upregulated the protein expressions of HO‐1 and NQO1, increased the activity of SOD, as well as decreased the content of ROS and MDA (Figure 6 and Figure 12). The HO‐1, NQO1 and SOD are protective proteins against oxidative stress, and MDA is a common product of oxidative stress. ³⁷ , ³⁸ HO‐1 functions as an important protein of cell adaptation to oxidative stress by generating carbon monoxide, biliverdin and free iron. ³⁹ NQO1 is a type of cytosolic flavoprotein, which can transfer electrons, ⁴⁰ and it can fight against excessive oxidative stress by reducing quinones to hydroquinones. ⁴¹ Otherwise, ROS plays an important role in oxidative stress and ROS retention leads to cell damage through lipid peroxidation. ²⁵ , ⁴² To sum it up, EDA could ameliorate viability of random skin flaps and MUVECs by suppressing excessive oxidative stress.

Moreover, the administration of EDA alleviated apoptosis of random skin flaps and MUVECs, which was supported by the upregulation of Bcl‐2 and downregulation of Cleaved‐Caspase3 and Bax (Figure 5 and Figure 11). Bax, a proapoptotic effector protein, is activated and accumulated in the mitochondria outer membrane (MOM), resulting in mitochondrial swelling. ⁴³ Subsequently, Cleaved‐Caspase3 is released from the damaged mitochondria and acted as an apoptotic executor in apoptosis. ¹⁸ Bcl‐2 is a type of anti‐apoptosis protein and prevents apoptosis from such mechanisms as inhibiting the function of proapoptotic gene p53, controling transcription factors and regulating the permeabilization of MOM. ⁴⁴ Taken together, we found that the administration of EDA alleviated apoptosis of random skin flaps and MUVECs.

There are strong interactions among autophagy, oxidative stress and apoptosis. ROS is an important product of excessive oxidative stress, which leads to the damage of proteins, DNA and lipid, activating autophagy. ⁴⁵ In addition, NO is another main product of oxidative stress, which activates autophagy by disturbing the crosstalk between Beclin1 and Bcl‐2. ²¹ Generally, autophagy activation inhibits apoptosis by removing damaged mitochondria which are prone to induce apoptosis. ⁴⁶ Otherwise, because some Atgs such as Atg5, Atg7 and Atg9 possess the identified domain of Caspases, Caspases activation blocks autophagy through cleaving Atgs. ⁴⁷

EDA ameliorated viability of random skin flaps by activating autophagy through the downregulation of PI3K/Akt/mTOR signalling pathway, and subsequently promoting angiogenesis, inhibiting oxidative stress and apoptosis.

Autophagy is key for the promotions of cellular energy metabolism, cellular organs renewal and survival of mammal cells. ¹⁴ , ¹⁵ PI3K/Akt/mTOR is a vital signalling pathway of regulating autophagy, which can suppress the level of autophagy via downregulating the expression of autophagy‐related proteins, participating in the processes of cellular surviving, proliferation and pathology. ¹² , ⁴⁸ Once autophagy activates, the protein expressions of autophagosomal indicators such as LC3II, Beclin1 and Atg5 increase, while the autophagic degradation marker p62 expression decreases. ⁴⁹ , ⁵⁰

To support the role of EDA‐induced autophagy in vitro, 3‐MA, a classical autophagy inhibitor, and Rapa, a classical autophagy activator, were used alone and together with EDA respectively. Our study showed that 3‐MA administration reduced the viability, and worsened apoptosis as well as oxidative stress of MUVECs exposed to OGD while Rapa administration increased the viability, and attenuated apoptosis as well as oxidative stress of MUVECs exposed to OGD. Furthermore, EDA‐mediated promotion of the viability of MUVECs exposed to OGD was reversed by 3‐MA administration while it was not be encouraged by rapacymin administration, which demonstrated that adequate autophagy achieved better results than excessive autophagy. Taken together, we verified our hypothesis that EDA could promote the survival rate of random skin flaps by activating autophagy.

Previous studies had showed that activating autophagy could promote the survival of random skin flaps. ²⁴ , ⁵¹ However, compared with those similar studies, our study had some advantages. Firstly, most of the similar studies applied length: width = 1.5:1 random skin flaps model, however, this kind of random skin flaps model is difficult to suffer from ischemia injury and has limited research values. In our study, we adopted length: width = 7: 1 random skin flaps instead, which had more research values. Secondly, the main weakness of the previous similar studies was that they all lack of cell research. Our study uncovered that EDA attenuated apoptosis and oxidative stress by activating autophagy through downregulating PI3K/Akt/mTOR signalling pathway in vitro research. Hence, our findings were more precise and convincing.

However, this study still had a few limitations. Firstly, we did not evaluate the long‐term therapeutic effects of EDA. And in further study, we will prolong the observation period postoperatively. Secondly, mouse is not the most suitable animal for random skin flaps model. Thirdly, other common cells of the skin flaps such as fibroblasts and keratinocytes should be applied in the vitro experiments. And in the future, on one hand, we plan to establish the random skin flaps model on rats or pigs and extend the observation duration in vivo. On the other hand, we plan to apply fibroblasts or keratinocytes and try to observe effects through the knockdown or overexpression of autophagy‐related genes in vitro.

5. CONCLUSION

The study initially reports that EDA can ameliorate viability of random skin flaps by promoting angiogenesis, suppressing oxidative stress and apoptosis in vivo and vitro, which may be mediated by autophagic activation through downregulating PI3K/AKT/mTOR signalling pathway.

AUTHOR CONTRIBUTIONS

Minglei Bi: Designed the study; Edited the final text; Performed the experiments; Wrote the manuscript. Yonghong Qin: Edited the final text; Liangtao Zhao: Performed the experiment; Xuanfen Zhang: Designed the study; Edited the final text.

FUNDING INFORMATION

This work is supported by the Donated Scientific Research Program of Lanzhou University (071100152) and Cuiying Scientific and Technological Innovation Program of Lanzhou University Second Hospital (CY2019‐BJ17).

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest and all authors consent for publication.

ETHICS STATEMENT

The animal experiments were approved by the Animal Welfare and Ethics Committee of Lanzhou University Second Hospital (D2021‐256).

Bi M, Qin Y, Zhao L, Zhang X. Edaravone promotes viability of random skin flaps via activating PI3K/Akt/mTOR signalling pathway‐mediated enhancement of autophagy. Int Wound J. 2023;20(8):3088‐3104. doi: 10.1111/iwj.14184

DATA AVAILABILITY STATEMENT

All the data and material in this article are available from the authors.