Resolvin D1 Inhibits Endothelial Permeability and Mitochondrial Damage Following Cardiac Ischemia–Reperfusion in Diabetic Mice

Purpose Resolvin D1(RvD1), a metabolite derived from docosahexaenoic acid, plays important therapeutic roles in inammatory diseases. However, the role of RvD1 in diabetic myocardial ischemia/reperfusion (IR) is still unknown. Diabetic mice was established by high-fat diet and streptozotocin, RvD1 was pretreated by intraperitoneal injection for 3 days, followed by myocardial IR. To evaluate the effects of RvD1 on chronic cardiac remodeling, RvD1 was administered for another 2 weeks after IR. The effects of RvD1 following myocardial IR injury were measured, including severity of infarct size, regional inammation, cardiac function, as well as permeability of cultured endothelial monolayer. Mitochondrial reactive oxygen species (mito-ROS) and mitochondrial membrane potential (MMP) were determined by MitoSOX and JC-1.


Introduction
Diabetes mellitus (DM) is a chronic metabolic disorder characterized by insulin resistance and dysfunction of pancreatic beta cells [1]. DM is highly associated with myocardial infarction and heart failure. Obesity, hyperglycemia, and hyperlipidemia are the most common metabolic disorders identi ed in DM and are established cardiovascular risk factors [2,3]. Diabetic cardiomyopathy is one of the most common complication of DM, which is recognized as a microvascular disease. It has been reported that diabetic cardiomyopathy was associated with higher morbidity and mortality in diabetic patients [4,5].
Meanwhile, diabetic heart is more susceptible to myocardial ischemia and reperfusion (IR) and that the cardioprotective effects of ischemic and pharmacological pre-conditioning are compromised in the presence of diabetes [6][7][8]. Moreover, diabetes induced cardiac dysfunction is associated with endothelial dysfunction that leads to cardiovascular complications [4]. Furthermore, endothelial cells dysfunction represents a crucial step in the development of endothelial cardiomyopathy with diabetes [9,10].Thus, new therapeutic targets prevent the diabetic heart against IR damage are urgently needed.
Resolvin D1 (RvD1), one of the most crucial molecules in specialized pro-resolving mediators (SPMs), which has been proven to play important roles in anti-in ammatory and pro-resolution [11,12]. Recently, Xia et al have found that RvD1 was remarkably decreased in plasma sample of T2DM patients [13].
Interestingly, another two studies also reported that RvD1 ameliorated streptozotocin-induced diabetes by its anti-in ammatory action, which suggested that RvD1 has potent insulin-sensitizing actions, and might be implicated in the pathogenesis of diabetes [14,15]. However, it remains unknown whether RvD1 affects vascular permeability in diabetic IR. In the present study, we explored the therapeutic effect of RvD1 on cardiac microvascular damage and explored potential protective mechanisms in diabetic I/R mice.

Methods And Materials
Myocardial ischemia and reperfusion mice model The animal experiment according to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1985). All experimental procedures involving animals were approved by the Ethics Committee of the Animal Care and the Ethics Committee of Sichuan University. For diabetic mice model, 8-10 weeks old male C57BL/6J mice were fed with high fat diet for 4 weeks and then given a single intraperitoneal injection of low-dose STZ (90 mg/kg, Sigma). One week after STZ injection, mice with 12-hours fasting blood glucose level ≥ 11.1 mmol/L were considered to have developed diabetes [16,17]. For myocardial ischemia reperfusion model, the left anterior descending (LAD) coronary artery was ligated with a 7-0 suture for 1 hour and then reperfusion for 24 hours. For the sham group, the similar surgical procedure was performed, while the LAD coronary artery was not ligated. RvD1 (Cayman Chemical) was administered to a dose of 50ug/Kg/day by intraperitoneal injection.

Echocardiography
Echocardiography analysis was performed 14 days after the IR surgery. Mice were anaesthetized with iso urane, and cardiac function was assessed by an echocardiographic imaging system (GE Vivid 7 equipped with a 12-MHz imaging transducer). Measurements were performed at the midpapillary level from well-aligned M-mode images from the parasternal short axis view, left ventricular ejection fraction (EF%) and fractional shortening (FS%), interventricular septum (IVS), left ventricularend diastolic dimension (LVID: d), left ventricularend-systolic dimension, (LVID: s), left ventricular-posterior-wall (LVPW), end-diastolic volume (EDV), and end-systolic volume (ESV) were measured.

TTC staining and Evans blue staining
Following myocardial IR (1 hour/24 hours), 2,3,5-Triphenyltetrazolium chloride (TTC) (Sigma-Aldrich) and Evans blue (Sigma-Aldrich) dye were applied for the measurement of myocardial infract size. Brie y, at the end of reperfusion, 1 mL 1% Evans blue was injected through abdominal aorta after re-ligating the LAD. The heart was then extracted and sectioned once it hardened at −20 °C, then cut into slices (2-3 mm). The slices were incubated in the TTC staining solution for 15 minutes at 37°C. Subsequently, the slices were then xed in 4% formaldehyde for 1 hour. Generally, the area stained with Evans blue represented the non-IR myocardium, whereas the unstained area was the I/R myocardium, the red area was at risk, and the white area was infarct size. Images were captured using a camera and the area of the infarcted myocardium in percentage of risk area was analyzed by ImageJ software.

Vascular permeability analysis
Vascular permeability was evaluated by Evans blue staining as reported [18][19][20]. In brief, following myocardial IR, at 3 hours before termination of the experiment, Evans blue dye (20 mg/kg) was injected intravenously, the mice were then euthanized and immediately subjected to aorta perfusion with phosphate buffer saline (PBS). The hearts were then excised, and Evans blue dye was eluted in formamide for 18 hours at 70°C. The absorbance of Evans blue dye at 620 nm was measured by Spectra Max 250 spectrophotometer (Molecular Devices, Sunnyvale, CA, USA) Cell culture Human cardiac microvascular endothelial cells (HCMECs) were purchased from Shanghai Huzhen Biotechnology. HCMECs were cultured in DMEM containing10% fetal bovine serum and 1% penicillin/streptomycin in an incubator with 5% CO2 at a temperature of 37°C. The cells were cultured with a nal concentration of 30 mM glucose as a high glucose concentration, 5.5 mM glucose condition served as control. Permeability across endothelial cell monolayer was measured in transwell units (with polycarbonate lter, 0.4 µm pore; Corning Costar) under following conditions in the presence or absence of RvD1 (100 ng/ml), (1) Hypoxic conditions (5% CO2 and 94% N2, 1% O2,) used fresh DMEM with 1% FBS to mimic ischemic conditions for 12 hours, reoxygenation at the 5% CO2 incubator and with DMEM containing10% fetal bovine serum and 1% penicillin/streptomycin for another 12 hours. H2O2 (400 μM) for 12 hours, LPS (500 ng/ml) for 12 hours. FITC-dextran (0.5 mg/ml, average molecular mass 40,000; Sigma) as a uorescent indicator was added into the upper-chamber for 2 hours before the end experiments. And 100 μl sample was taken from the lower-chamber and the uorescence of FITC-dextran was measured (485/535 nm, absorption/emission wavelengths) by Spectra Max 250 spectrophotometer (Molecular Devices, Sunnyvale, CA, USA).

Immuno uorescence
At the nal of the cellular experiments, remove the medium, and phosphate buffered saline (PBS) washed for 3 times, then xed in 4% paraformaldehyde, permeabilized with 1% Triton X-100 in PBS, 5% bovine serum albumin in PBS was used to block cells for 15 min. A primary antibody against VE-cadherin for overnight and a secondary antibody were used for immunostaining 30 minutes, DAPI was used for staining the nucleus for 15 minutes, images were captured by a confocal microscope (N-STORM & A1, NiKON).

MitoSOX staining
HCMECs were cultured were seeded in 24-well plates with a density of 5 × 10 4 /ml. The cells were cultured with a nal concentration of 30 mM glucose as a high glucose concentration, 5.5 mM glucose condition served as control. RvD1 (100 ng/ml) was pretreaed for 12 hours, followed by 400 μM H2O2 for 4 hours, then cells were washed three times with warm Hank's Balanced Salt Solution (HBSS). MitoSOX-Red (4 μM, ThermoFisher Scienti c) dye was added to each well, incubated for 30 minutes in the dark at 37°C, then washed three times with warm HBSS. Finally, uorescence intensity was determined by Celigo imaging cytometer, images were captured by a confocal microscope (N-STORM & A1, NiKON).

Mitochondrial transmembrane potential (MMP) assay
Mitochondrial membrane potential detection was used JC-1 (MedChemExpress Cat. No.: HY-K0601). According to the manufacturer's instructions, JC-1 (2 μM) was added to per well, and incubated at 37°C for 20 minutes, followed by centrifuged for 3 minutes at 400g, washed cells twice with PBS, and nally added 500 μL PBS to suspend cells. JC-1 uorescence was measured with a ow cytometer.

Immunostaining analysis
Immunostaining for heart sections, hearts were excised from mice after euthanasia, xed in 4% paraformaldehyde overnight, embedded in para n, and sectioned. The sections were then depara nized, rehydrated, retrieved for antigens, and subjected to immunohistochemical analysis using against CD45 (Abcam), CD68(Abcam), followed by staining with a secondary antibody (goat anti-rabbit antibody).
Images were captured by ZEISS IX83.For Masson staining, heart sections were depara nized, stained with Masson's composite staining solution and a bright-green staining solution and dehydrated in absolute ethanol. Acetic acid was used for washing. Finally, the sections were subjected to transparency in xylene and then mounted with neutral gum for analysis.
Quantitative real-time reverse transcription-polymerase chain reaction(qRT-PCR) The total RNA was extracted from tissues or cell lysates using Trizol (Invitrogen), after extracting using chloroform and precipitating with isopropanol, the RNA was washed twice using 75% ethanol, and the concentration of RNA was measurement after dissolved in RNase-free water. Subsequently the RNA was reversely transcribed by the PrimeScript RT Reagent Kit (Takara). qRT-PCR assay was performed on the CFX96TM Real-time PCR Detection System (BioRad) using EvaGreen Supermix Kit (Bio-Rad). Relative mRNA expression was normalized to GAPDH. The qPCR primer sequences were as following Heart tissues were harvested and quickly xed in 2.5% glutaraldehyde for 1 hours, exposed to 1% osmium tetroxide, then sectioned to 60 nm and mounted on Cu-grids contrasted with uranyl acetate and lead citrate, and analyzed using a JEM-1400 electron microscope.

Statistical analysis
All experimental data were presented as means ± SD. The difference between 2 groups of variables was compared by the 2-tailed, unpaired T-test. For data analysis of more than 2 groups, one-way ANOVA test was used. A value of P < 0.05 was considered to be statistically signi cant.

Results
RvD1 pretreatment protects against vascular hyperpermeability after myocardial IR injury in diabetic mice After successful established diabetic mice model. DM+IR+ RvD1 group were pretreated RvD1 for 3 days.
After myocardial ischemia reperfusion operation. An assay with Evans blue dye and TTC staining was performed to determine myocardial infarct size after IR, TTC staining showed that pretreatment with RvD1 signi cantly reduced the infarct size by 12% compared with I/R group in diabetic mice. As shown in (Fig. 2C, 2D). Vascular leakage was measured by Evn's blue extravasation after myocardial IR, as the results presentation, Evan's blue increased after I/R injury in diabetic mice, however, the effect was inhibited by RvD1treatment (Fig. 2E, 2F). These data suggest that RvD1 provided myocardial protection against I/R injury by inhibiting vascular permeability in diabetes.

RvD1 pretreatment inhibited in ammatory responses in diabetic I/R injury
The TEM images of endothelial barrier demonstrated that endothelial cells from the diabetic I/R heart appeared swollen with increased lysosomal-like inclusions and the integrity of endothelial barrier was also damaged, while pretreated with RvD1 could alleviated endothelial barrier damage (Fig. 4A, 4B). The in ammation in response to diabetic IR was evaluated by immunohistochemical detection of the density of in ammatory cells (CD45 and CD68) in hearts (Fig. 3A, 3B), gene expression of in ammatory cytokines (IL-1β, IL-6 and MCP-1) in diabetic IR hearts (Fig. 3C). AS the results showed that there was a robust increase in the density of leukocytes CD45 and macrophages CD68 in vehicle-treated group. Above all, RvD1 pretreatment therapy reduced the density of in ammatory cells, and also signi cantly downregulated expression of IL-1β, IL-6 and MCP-1 versus vehicle treated group.

RvD1 pretreatment improved cardiac performance and alleviated chronic cardiac remodeling
To study the in uence of RvD1 treatment on chronic cardiac remodeling, RvD1 was administered via intraperitoneal injection (50 μg/kg/day) for another 2 weeks after I/R. Echocardiography was performed at 2 weeks after IR. Mice with IR exhibited signi cantly decline in EF and FS (Fig. 5A). Comparison with untreated mice, RvD1 treated mice had a signi cantly improvement of EF and FS. To assess effect of RvD1 on cardiac brosis, Masson staining was performed to evaluate cardiac remodeling after IR injury (Fig. 5B, 5C), the results showed that brosis was increased after IR injury, while RvD1 treatment alleviated cardiac brosis at 2 weeks after IR in diabetes.

RvD1 improved permeability of endothelial cells stimulated with different pathological conditions
To assess the effect of RvD1 on endothelial permeability, we used transwell device to grow HCMECs into a monolayer in the condition of high glucose. Cells were subjected to 12 hours hypoxia following by reoxygenation of 12 hours (H/R). Meanwhile various stimuli, including H2O2 or LPS were added to the culture medium with or without the pretreatment of RvD1 for another 2 hours. The dextran-FITC of lower chamber was etected to determine endothelial permeability. Fluorescence intensity of FITC in lower chamber was signi cantly higher after H/R (Fig. 6B), H2O2 (Fig. 6C) and LPS (Fig. 6D) stimulation, but it was attenuated with RvD1 treatment. Q-PCR analysis showed that RvD1 treatment signi cantly suppresses H/R-induced expression of in ammatory genes MCP-1, IL-6, and IL-1β mRNA expression in high glucose condition (Fig. 6E). Taken together, these ndings suggested that RvD1 attenuate HR, H2O2 and LPS-induced endothelial permeability and decrease H/R-induced in ammation in high glucose condition.

RvD1-mediated microvascular protection was regulated by mitochondrial damage
Elevated ROS levels have been shown to have important roles in regulating endothelial barrier. First, Mito-SOX probe was used to examine the cellular level of ROS cultured with high glucose DMEM. As the results presented that H2O2 signi cantly increased mito-ROS, while RvD1 treatment decreased mito-ROS level in high glucose condition (Fig. 7A, 7B). Second, as mitochondrial transmembrane potential (MMP) plays a key role in mitochondrial homeostasis through selective elimination of dysfunctional mitochondria [21], JC-1 was used to detect the MMP. The owcytometry results revealed RvD1 pretreatment signi cantly mitigated mitochondrial membrane potential drop induced by H2O2 in high glucose condition (Fig. 7C).

RvD1 inhibited endothelial permeability with preservation of VE-cadherin
As VE-cadherin-mediated cell-cell adhesion organizes the endothelial junctions and maintains the barrier function, which is central in permeability changes. Thus, we investigated the effect of RvD1 on the expression levels of VE-cadherin. We used immuno uorescence and western blots examined the cellular level of VE-cadherin cultured with high glucose DMEM after H/R. Immuno uorescence and western blot analysis showed that RvD1 treatment signi cantly increased the level of VE-cadherin after H/R (Fig. 8A,  8B). Meanwhile, VE-cadherin was also detected in the diabetic heart tissue after I/R, the results also revealed that VE-cadherin was decreased in diabetic heart tissue after IR, which was reversed by RvD1 pretreatment (Fig. 8C).
Therefore, these evidences suggested that the effect of RvD1 on the endothelial permeability may be mediated by regulating the expression of VE-cadherin.

Discussion
Numerous studies report that high glucose environments aggravate myocardial vulnerability to ischemia/reperfusion injury [22][23][24]. For the rst time, we provide evidence that RvD1, an important member of endogenous specialized pro-resolving mediators, provides cardio-protection from myocardial ischemia reperfusion injury in diabetic mice by restricting endothelial permeability. First, RvD1 pretreatment signi cantly decreased myocardial content of Evans blue and reduced infarct size in diabetic IR mice. Second, RvD1 pretreatment of was associated with suppression of regional in ammatory responses, alleviation of cardiomyocytes death, consequently treated with RvD1 improved cardiac performance and restricted cardiac remodelling. Finally, RvD1 pretreatment protected endothelial cell monolayer against endothelial leakage induced by hypoxia/re-oxygenation, H2O2 or Lipopolysaccharide, effects that are associated with preservation of VE-cadherin.
Resolvin D1, derived from the omega-3 fatty acids docosahexaenoic acid, reveals remarkable potentiality in resolving in ammation-related diseases such as abdominal aortic aneurysm formation, sepsis [12,[25][26][27]. Recently, Xia et al have found that RvD1 was remarkably decreased in Type 2 diabetes mellitus (T2DM) patients' plasma samples versus healthy controls, indicating that RvD1 may play important role in the pathogenesis of T2DM [13]. While, in our present study, RvD1 treatment alleviated the endothelial permeability in diabetic I/R mice without in uencing the fasting glucose.
Several studies reported that RvD1 prevented LPS-induced polymorphonuclear leukocytes recruitment to endothelial cells, and attenuated endothelial barrier permeability [28][29][30]. Moreover, stabilization of the vasculature by RvD1 maybe another potential molecular mechanism underlying the protection of RvD1 on diabetic IR heart. It is well-known that VE-cadherin is a component of endothelial cell-to-cell adherens junctions, and is an important determinant of the barrier function of the vascular endothelium [31]. On the other hand, a burst of reactive oxygen species sourced from mitochondria at the beginning of the reperfusion, aggravate myocardial injury through a number of different mechanisms [32,33]. Therefore, mitochondrial damage determines myocardial damage during IR and subsequent remodeling [34][35][36]. RvD1 has already been proven to reduce reactive oxygen species levels, improve mitochondrial morphology and function, promote mitochondrial DNA repair in primary retinal cells with diabetic retinopathy [37]. In diabetes, high blood glucose and free fatty acids induce ECs mitochondrial ssion, with high levels of reactive oxygen species (ROS), leading to blunted cell growth and altered cell-adhesion molecule expression on the cell surface [38]. In our study, the results were consistent with those studies that RvD1 pretreatment not only reduced cardiomyocytes apoptosis, but also subsequently treatment with RvD1 also attenuated pathological cardiac remodeling after IR in diabetic mice. At the same time, the results indicated that RvD1 deceased mito-ROS generation stimulated with H2O2 in high glucose condition. Mitochondrial transmembrane potential (MMP) is also an important parameter of the mitochondrial function which is considered as an indicator of cell health [21]. During apoptosis, MMP decreases is associated with the opening of the mitochondrial permeability pores and loss of the electrochemical gradient. In the present study, we found that RvD1 prevented H2O2-induced MMP reduction. While, further studies are needed to better understand the underlying mechanisms of RvD1 in regulating mitochondrial function in cardiovascular diseases.
In summary, RvD1 attenuated myocardial microvascular damage following diabetic cardiac IR injury, which presented with reduction of infarct size, suppression of in ammatory response, and decreased cardiomyocyte apoptosis. The underlying molecular mechanism is partially through preservation of VEcadherin and modulating mitochondria function. Therefore, RvD1 treatment may be a promising therapeutic drug to protect diabetic heart against ischemia and reperfusion.