Integrinβ3 Mediates the Protective Effect of Soluble Receptor for Advanced Glycation End-products During Myocardial Ischemia/Reperfusion

Soluble receptor for advanced glycation end-product (sRAGE) was reported to protect myocardial ischemia/reperfusion (I/R) injuries via interacting with AGEs (the ligands of RAGE). Besides, sRAGE was also reported to interact with myocardial cell membranes to protect myocardial I/R injuries. However, the specic molecular of the interaction between sRAGE and myocardial cell membrane is not clearly dened. Integrins were the major adhesion receptors expressed on myocardial cells. The present study showed sRAGE affected the expression of integrinβ3 around integrinβ1 to β5 in I/R treated cardiomyocytes. Consequently, it was supposed that integrinβ3 might be related to the protective effects of sRAGE on myocardial ischemia-reperfusion injuries. The results in this study showed that the protective effects of sRAGE on cardiac function, cardiac infracting size and apoptosis in mice were cancelled by cilengitide (the inhibitor of integrinβ3). Myocardial apoptosis and autophagy decreased by sRAGE in I/R treated cardiomyocytes were showed to be increased by the inhibition of integrinβ3. Mechanistically, the inhibition of integrinβ3 decreased the phosphorylation of Akt and STAT3 but did not affect the phosphorylation of FAK, PTEN, and MAPKs (P38 MAPK, JNK, Erk) in OGD/R and sRAGE treated cardiomyocytes. In addition, the phosphorylation of STAT3 was signicantly downregulated by the inhibition of Akt (LY294002,10μM) in OGD/R and sRAGE treated cardiomyocytes. The present study demonstrated integrinβ3 mediated the protective effects of sRAGE on myocardial I/R injuries through Akt-induced STAT3 signaling pathway thus may provide a novel molecular therapy strategy for ameliorating I/R injury.


Introduction
Cardiovascular diseases are the leading cause of death worldwide. Myocardial ischemia/reperfusion (I/R) injuries and aggravation of tissue injury after blood perfusion of the ischemic myocardium are considered risk factors for cardiac function impairment. [1] , [2] I/R injury is characterized by the outbreak of reactive oxygen species (ROS), [3] mitochondrial dysfunctions, apoptosis, autophagy, and increased infarct area [3,4] Previous studies have shown that soluble receptors for advanced glycation end-products (sRAGE) have a protective role in I/R injury by inhibiting myocardial apoptosis and autophagy. [5] sRAGE is produced by proteolytic cleavage or alternative RNA splicing in the extracellular area of the cell surface receptor.
[6] It acts as a natural antagonist to RAGE signaling for isolating RAGE ligands, such as AGEs to inhibit RAGE-dependent cellular responses. [7] However, whether sRAGE depended on RAGE or other membrane receptors on cells to protect I/R myocardium remains unclear.
Integrins, which have been reported as a major adhesion receptor on the myocardium,[8] can promote mitochondrial function through STAT3 and lead to cell survival. [9] Previous studies have shown that RAGE can interact with integrins to mediate leukocyte adhesion. [10] Also, sRAGE might interact with integrins to induce the protective effects on cardiomyocytes during I/R injuries.
Integrinβ3 promotes cardiomyocyte proliferation and reduces hypoxia-induced apoptosis. [11] Yet, the role of integrinβ3 in sRAGE protecting cardiomyocytes from I/R injuries has not been reported so far. In this study, we investigated the relationship between integrinβ3 and sRAGE in the I/R mice model.

Animals
C57bL/6 adult male mice, 8 ~ 10 weeks old, were purchased from the Experimental Animal Laboratory, Capital Medical University. All the animals were housed in an environment with a temperature of 22 ± 1 ºC, relative humidity of 65 ± 5%, and a light/dark cycle of 12/12 hr, having free access to food and water.
All animal studies (including the mice euthanasia procedure) were done in compliance with the regulations and guidelines of Capital Medical University institutional animal care and conducted according to the AAALAC (Association for Assessment and Accreditation of Laboratory Animal Care) and the IACUC (Institutional Animal Care and Use Committee) guidelines. The present study was approved by the Ethics Committee of Capital Medical University.

I/R operation
The mice were subjected to I/R operation as previously described. [23] Brie y, the hearts of iso uraneanesthetized mice were treated with 6 − 0 silk thread 30min after coronary artery occlusion and 24h reperfusion, after which the hearts were exposed through the fourth intercostal space. [5] All animals were anesthetized by pentobarbital sodium (50mg·kg − 1 ) and then euthanized by cervical dislocation. The hearts were collected and stored in the − 80°C refrigerator for further experiments.

2,3,5-triphenyl tetrazolium chloride (TTC) staining
Hearts were immediately removed after re lling with 1%TTC (Sigma-Aldrich St. Louis. MO. USA) solution. [25] Hearts were incubated in 1%TTC solution at 37°C for 3-5min and xed in 4% (wv − 1 ) paraformaldehyde for 5min. They were then cut vertically and photographed with a digital camera. Image J software (NIH) was used to quantify the area of infarction and normal sections. The data were expressed as the percentage of the infarction area and total area.

Immuno uorescence staining
Myocardial tissue was xed with 10% buffered formalin and embedded in para n, after which it was cut into 4µm thick slices. The sections were incubated in 0.1% (vv − 1 ) Triton X-100 for 8min and blocked with normal horse serum for 30min at room temperature. The sections were then incubated with cleaved- ImageJ software (NIH) was used to calculate the positive area. TUNEL assay TUNEL staining was performed in accordance with the instructions from TMR Red (Roche, Mannheim, Germany). [27] Tissues were para n-embedded and cut into 4µm thick slices. The myocardial cells were washed twice with PBS and xed for 15min at room temperature. The slices or cells were incubated in 0.1% (VV-1) Triton X-100 for 8min and sealed with 3% (WV-1) bovine serum for 1h at room temperature.
Each section was incubated with TUNEL conjugated dUTP at 37°C for 60min. After staining with α-actin and 4 ', 6-diamino-2-phenylindoles (DAPI), sections were stored in a refrigerator at -20°C. Images were obtained using an Olympus BX51 uorescence microscope (Olympus America Inc., Center Valley, PA, USA) at 20× magni cation. Image J (NIH) was used to calculate the number of TUNEL-positive cells and the total cells in each eld.
Isolation and culture of the myocardium Neonatal rat ventricular cardiomyocytes were isolated from the hearts of 1-to 2-day-old Sprague-Dawley rats with 0.25% trypsin as previously described.
[28] Rats were purchased from Viton Lever Laboratory Animal Technology Co., LTD. (Beijing, China). Cell pellets were resuspended in DMEM/F12 medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin and cultured in an incubator of 5%CO 2 and 21%O 2 at 37°C for 1.5h to remove broblasts. Then, 1×10 6 cells/well were cultured in DMEM/F12 medium containing 0.3g/l glutamine, 4.5g/l glucose, 10% FBS, and 1% penicillin/ streptomycin at an atmosphere of 5% CO 2 and 21%O 2 at 37°C for 48h. After the cardiomyocytes adhered to the wall, the subsequent operation for further experiments was carried out.

Adenovirus transfection and OGD/R model simulation in cardiomyocytes
The recombinant adenovirus vector was purchased from Hanheng Biotechnology Co Ltd (Shanghai, China). Myocardial cells were transfected by recombinant adenovirus expressing GFP or a sequence of the extracellular domain of human RAGE for 24h before stimulation with I/R injury. Cells were then treated with an "ischemia buffer" (pH 6.3), which included 118 NaCl, 24 NaHCO 3 , 1.0 NaH 2 PO 4 , 2.5 CaCl 2 ·2H2O, 1.2 MgCl 2 , 20 sodium lactate, 16 KCl, and 10 deoxyglucose (mM) for 2h in 1%O 2 and 5%CO 2 at 37°C (ischemia). [29] Subsequently, DMEM/F12 was added to the cells in an atmosphere of 5%CO 2 and 21%O 2 at 37°C (reperfusion). In the control groups, cells were incubated with DMEM/F12 in an atmosphere of 5%CO 2 and 21%O 2 at 37°C. In the I/R and sRAGE group, cilengitide (1µM) [30] was added to the medium 30 min before and during the ischemia and reperfusion.
Recombinant sRAGE protein (900µM) [31](Aidi Bo biological Ltd. Beijing. China) was then applied to the medium 30minutes before I/R stimulation. After reperfusion, cardiomyocytes were xed with 4% (wv − 1 ) paraformaldehyde for 15min, and the nucleus was stained by DAPI. Twenty images were collected using confocal uorescence microscopy (TCS SP8 MP, Leica, Buffalo Groove, Illinois, USA) to observe LC3 spots in cells.

Real-time PCR
Real-time polymerase chain reaction (real-time PCR) was performed as previously described. [32] Approximately 1-2µg of total mRNA was reverse-transcribed to complementary DNA (cDNA) using Go Script TM reverse transcription system (S1000 Thermal Cycler, California, USA). Expression of integrins was analyzed via q-PCR using SYBR-Green PCR Master Mix (RR420A; TaKaRa Bio) with a 7500 Real-Time

Western blot
Cardiac proteins of the same amounts (40µg) were separated by SDS-PAGE (6-15%) and transferred to PVDF membranes. After being blocked by 5% skim milk at room temperature for 1 hour, the membranes balanced for 20min at room temperature and incubated with a horseradish peroxidase-conjugated secondary antibody anti-rabbit IgG (1:2000; 7074; Cell Signaling Technology) for 1 hour. Speci c bands were exposed to ultra ltration HRP substrate (WBKLS0500, Millipore, Billerica, MA, USA). Fluor Chem FC3 (Protein Simple, Wallingford, CT) was used to capture the images. The density was determined by Image J software (NIH). Loading differences were controlled by probing the blot with an antibody against GAPDH.

Statistical analysis
Data were expressed as the mean ± SEM. SPSS V25.0 (SPSS Inc., Chicago, IL, USA) was used for statistical analysis. A nonparametric test was used when data deviated from a normal distribution. Signi cance between three groups or more was determined by the parametric test of one-way ANOVA. The least signi cant difference test (LSD) was used to evaluate the signi cant difference between the two groups. A p < 0.05 was considered statistically signi cant.

sRAGE upregulated the expression of integrinβ3 after I/R treatment
To decide which subunit of integrins was in uenced by I/R and sRAGE, integrin β1, β2, β3, β4, and β5 in cardiomyocytes were detected by real-time PCR. The results showed that I/R decreased relative integrinβ3 mRNA levels, which were reversed by sRAGE in cardiomyocytes (p < 0.05) (Fig. 1c); these data were consistent with Western blot results, which revealed that I/R decreased integrinβ3 protein expression, which was reversed by sRAGE treatment (p < 0.05) (Fig. 1f and 1g). However, I/R or sRAGE had no effects on the expression of other integrins (Fig. 1a, b, d, and e). These results suggested that integrinβ3 might be involved in the antagonistic effects of sRAGE on I/R injuries in cardiomyocytes.

The inhibition of integrinβ3 abolished the antagonistic effects of sRAGE on myocardial apoptosis after I/R injury
The results from the TUNEL assay, which was used to determine myocardial apoptosis in the marginal zone of myocardial infarction, showed that cilengitide increased the ratio of TUNEL-positive cells from 0.0462 ± 1.23% (I/R and sRAGE treated mice) to 0.3140 ± 2.88% (sRAGE, I/R, and cilengitide treated mice) (n = 6-8, p < 0.05, Fig. 3a, c), which was remarkably decreased by sRAGE in I/R-treated hearts. In addition, cleaved-caspase3 immuno uorescent staining was used to detect the distribution of apoptotic cardiomyocytes. The results showed that after integrinβ3 was inhibited by cilengitide, the relative cleaved-caspase3 positive area was increased from 11.13%±0.83% (I/R and sRAGE treated mice) to 37.33% ±0.88% (sRAGE, I/R, and cilengitide treated mice) (n = 6-8, p < 0.05, Fig. 3b, d). These results indicated that integrinβ3 mediates the antagonistic effects of sRAGE on myocardial apoptosis during I/R injuries.

The inhibition of integrinβ3 diminished the antagonistic effects of sRAGE on cardiomyocytes apoptosis after I/R injuries
Cardiomyocytes were subjected to hypoxia, and a TUNEL assay was used to determine the hypoxiainduced cardiomyocytes apoptosis. It was observed that after integrinβ3 was inhibited by cilengitide, the percentage of TUNEL-positive cells was signi cantly increased from 9.95%±0.01% (I/R and sRAGE treated cardiomyocytes) to 24.51%±0.02% (sRAGE, I/R, and cilengitide treated cardiomyocytes) (n = 6-10, p < 0.05, Fig. 4a, b). Meanwhile, Western blotting was performed to detect the expression of cleaved-caspase3. After integrinβ3 was inhibited by cilengitide, the expression of cleaved-caspase3 was signi cantly increased from 0.89 folds (I/R and sRAGE treated cardiomyocytes) to 1.76 folds (sRAGE, I/R, and cilengitide treated cardiomyocytes) (n = 3, p < 0.05, Fig. 4c, d). These results con rmed that integrinβ3 in uenced the antagonistic effects of sRAGE on cardiomyocytes' apoptosis during hypoxia/reperfusion stimulation.

The inhibition of integrinβ3 abolished the antagonistic effects of sRAGE on autophagy in myocardial cells after I/R injury
After cardiomyocytes were infected with mRFP-GFP-LC3 adenovirus for at least 24 hours, autophagy ow was detected. Yellow and red dots represented autophagosomes and autolysosomes, respectively. [12] Image J (NIH) was used to analyze the number of spots. The results revealed that after integrinβ3 was inhibited by cilengitide, the number of autophagosomes increased from 16.5 ± 0.76% (I/R and sRAGE treated cardiomyocytes) to 26.33 ± 1.52% (sRAGE, I/R and cilengitide treated cardiomyocytes) (n = 6, p < 0.05, Fig. 5a, b) and the number of autolysosomes were not remarkably changed by the inhibition of integrinβ3 in I/R and sRAGE treated cardiomyocytes (n = 6-10, Fig. 5a, c). Besides, Western blot was performed to detect the expression of autophagy relative proteins. These results showed that after integrinβ3 was abolished by cilengitide, autophagy relative proteins such as LC3-II were activated from 1.19 folds (I/R and sRAGE treated cardiomyocytes) to 1.47 folds (sRAGE, I/R and cilengitide treated cardiomyocytes) (n = 5-6, p < 0.05, Fig. 5d, f) and Beclin-1 was activated from 1.11 folds (I/R and sRAGE treated cardiomyocytes) to 1.37 folds (sRAGE, I/R, and cilengitide treated cardiomyocytes) (n = 3-4, p = 0.2, Fig. 5d, g). Meanwhile, after integrinβ3 was inhibited by cilengitide, the expression of p62 was downregulated from 1.62 folds (I/R and sRAGE treated cardiomyocytes) to 1.14 folds (sRAGE, I/R, and cilengitide treated cardiomyocytes) (n = 4-7, p < 0.05, Fig. 5d, e). Brie y, this data suggested that integrinβ3 mediated the reductive effect of sRAGE on excessive autophagy in I/R-treated cardiomyocytes.
AKT and STAT3 signalings are downstream molecules of integrinβ3 induced by sRAGE To demonstrate the intracellular signaling of integrinβ3 induced by sRAGE in I/R treated cardiomyocytes, the potential molecules were detected after integrinβ3 was inhibited by cilengitide. Brie y, the expression of integrinβ3 was showed to be decreased from 1.75 folds (I/R and sRAGE treated cardiomyocytes) to 0.48 folds (sRAGE, I/R, and cilengitide treated cardiomyocytes) (n = 6-8, p < 0.05, Fig. 6a, b) in OGD/R treated cardiomyocytes after integrinβ3 was inhibited by cilengitide, suggesting that cilengitide inhibited the expression of integrinβ3 in cardiomyocytes during I/R inducement. In addition, phosphorylated STAT3 was downregulated from 3.7 folds (I/R and sRAGE treated cardiomyocytes) to 0.73 folds (sRAGE, I/R, and cilengitide treated cardiomyocytes) (n = 5-6, p < 0.05, Fig. 6a, d), which had been proved to be increased from 0.47 folds (I/R treated cardiomyocytes) to 3.7 folds (I/R and sRAGE treated cardiomyocytes) (n = 5-6, p < 0.05, Fig. 6a, d) by sRAGE in I/R treated cardiomyocytes. However, no changes in the expression of phosphorylated FAK, phosphorylated PTEN, and MAPKs (JNK, Erk, P38) were observed among the groups in cardiomyocytes (n = 3-6, p > 0.05, Fig. 6a, c, e to i). Additionally, the expression of phosphorylated Akt was decreased from 0,67 folds (I/R and sRAGE treated cardiomyocytes) to 0.42 folds (sRAGE, I/R, and cilengitide treated cardiomyocytes) (n = 4, p < 0.05, Fig. 6e, j). To sum up, these results suggested that FAK, PTEN, and MAPKs were not the downstream pathway of sRAGE-integrinβ3 in cardiomyocytes during I/R injury, while STAT3 and Akt might be involved in the effects of integrinβ3 mediated function of sRAGE during cardiac I/R injuries.
To further explore the interaction between Akt and STAT3, LY294002 (an inhibitor of Akt) was adopted in I/R and sRAGE treated cardiomyocytes. The results showed that the phosphorylated STAT3 was downregulated after Akt was inhibited in I/R and sRAGE treated cardiomyocytes (n = 4, p < 0.05, Fig. 6k to m). These results suggested that Akt induced the activation of STAT3 signaling in I/R and sRAGE treated cardiomyocytes.

Discussion
The present study demonstrated that integrinβ3 mediated the protective effects of sRAGE on myocardial I/R injuries through Akt-induced STAT3 signaling pathway, which might be used as a potential molecular therapy strategy for ameliorating I/R injury in hearts.
Integrins are widely expressed on cell membranes, including cardiomyocytes. [13] The integrin family contains many subunits, including integrin β1, 2, 3, 4, and 5. [14] This study showed that integrinβ3 was markedly decreased by I/R, which was reversed by sRAGE in cardiomyocytes (Fig. 1). Previous studies have revealed that the expression of integrinβ3 was upregulated by hypoxia inducement in H9C2 cells (cell line derived from embryonic BD1X rat heart tissue). [15] However, the results in this study revealed that oxygen and glucose deprivation/reperfusion reduced integrinβ3 expression, which was reversed by sRAGE. Due to the previous results that sRAGE protected hearts from ischemia/reperfusion injuries, it was supposed that integrinβ3 might be related to the protective effects of sRAGE on I/R cardiomyocytes. sRAGE was reported to protect cardiac function and the heart infarction size induced by I/R.[16] The same as the previous reports, the results in this study showed that sRAGE protected cardiac function and the infarction size in I/R treated hearts. However, the protective effects of sRAGE were reversed after integrinβ3 was inhibited by cilengitide (Fig. 2), which suggested that integrinβ3 might mediated the antagonistic effects of sRAGE on I/R injuries in hearts. In addition, sRAGE was also reported to inhibit apoptosis induced by I/R in hearts besides the cardiac function and the heart infarction size. [5] Did integrinβ3 mediate the effects of sRAGE on apoptosis in the hearts as well?
Then,myocardial apoptosis in mice was examined by TUNEL assay and cleaved-caspase3 immuno uorescent staining. The results showed that myocardial apoptosis was decreased by sRAGE in I/R treated mice, which was increased after integrinβ3 was inhibited in I/R and sRAGE treated hearts (Fig. 3). These results suggested that integrinβ3 mediated the suppressive effects of sRAGE on apoptosis during I/R injuries in mice hearts. Then the question arising was, did sRAGE interact with integrinβ3 on myocardial cells directly?
Consequently, cardiomyocytes were isolated and treated with sRAGE and cilengitide (the inhibitor of integrinβ3). The results showed that myocardial apoptosis induced by I/R and inhibited by sRAGE were reversed by the inhibition of integrinβ3 in I/R and sRAGE treated cardiomyocytes (Fig. 4), which con rmed that sRAGE interacted with integrinβ3 on myocardial cells directly to inhibit apoptosis in I/R treated cardiomyocytes. In addition, previous studies have reported that the inhibition of excessive autophagy resisted apoptosis induced by I/R in cardiomyocytes, [5] which contributed to the inhibiting effects on myocardial ischemia-reperfusion injuries. [17] Then, did the effects of sRAGE on apoptosis which mediated by integrinβ3 in cardiomyocytes also mediated via autophagy? Therefore, autophagosomes and autolysosomes were detected by mRFP-GFP-LC3 virus transfection in cardiomyocytes. The results showed that the number of autophagosomes was increased, the expression of LC3-II and Beclin-1 were upregulated and P62 was downregulated after integrinβ3 was inhibited in I/R and sRAGE treated cardiomyocytes (Fig. 5). All the results suggested that integrinβ3 mediated the antagonistic effects of sRAGE on excessive autophagy in myocardial cells. The present study veri ed the assumption that integrinβ3 mediated the suppressive effects of sRAGE on excessive autophagy in I/R myocardium. Then, what was the intracellular signaling pathway of integrinβ3 to mediate the antagonistic effects of sRAGE on ischemia/reperfusion injuries in cardiomyocytes?
Integrin-FAK signaling was reported to enhance mitochondrial function [9] and decrease cell death[18] via STAT3 signaling. In this study, phosphorylation of STAT3 but not FAK was decreased after integrinβ3 was inhibited by cilengitide in I/R and sRAGE treated cardiomyocytes (Fig. 6a to d). Mechanistically, Akt signaling [19], MAPKs [20], JNK signaling [21] and PTEN [11] were reported to be affected by the activation of integrin in non-small cell lung cancer cells, human osteoarthritis chondrocytes, vascular endothelial cells, and cardiomyocytes, respectively. Nevertheless, the results in this study showed that the phosphorylation of PTEN, p38 MAPK,JNK, and Erk1/2 were not affected by the inhibitor of integrinβ3 in I/R and sRAGE treated cardiomyocytes (Fig. 6e to i). Interestingly, the phosphorylation of Akt was signi cantly upregulated after sRAGE administration in I/R treated cardiomyocytes, while it was downregulated after integrinβ3 was inhibited (Fig. 5b, f). Meanwhile, the phosphorylation of STAT3 was downregulated by the inhibitor of Akt (LY294002) in I/R and sRAGE treated cardiomyocytes, which suggested that STAT3 was the downstream molecule of Akt pathway. Based on these ndings, the present study concluded that integrinβ3 mediated the protective effects of sRAGE during I/R in cardiomyocytes via Akt and STAT3 signaling pathways.
In summary, the present study demonstrated that integrinβ3 mediated the suppressive effects of sRAGE on I/R-induced apoptosis and autophagy via the Akt and STAT3 signaling pathway in myocardial cells, which contributed to the antagonistic effects of sRAGE on cardiac function and the infarct size in hearts during I/R injuries.

Declarations
Ethics approval and Consent to participate All animal studies (including the mice euthanasia procedure) were done in compliance with the regulations and guidelines of Capital Medical University institutional animal care and conducted according to the AAALAC (Association for Assessment and Accreditation of Laboratory Animal Care) and the IACUC (Institutional Animal Care and Use Committee) guidelines. The study was reported in accordance with ARRIVE guidelines. This study did not involve human participants.

Consent for publication
This work did not involve the identi cation images or other personal or clinical details of participants. The authors of this paper consent for publication.

Availability of data and materials
The data that support the ndings of this study are not openly available and are available from the rst author or the corresponding author upon reasonable request.