Irisin protects against endothelial injury and ameliorates atherosclerosis in apolipoprotein E-Null diabetic mice

Accepted Manuscript Irisin Protects against Endothelial Injury and Ameliorates Atherosclerosis in Apolipoprotein E-Null Diabetic Mice Junyan Lu, MD, Guangda Xiang, MD, PhD, Min Liu, MD, Wen Mei, MD, Lin Xiang, MD, Jing Dong, MD PII:

S0021-9150(15)30166-0

DOI:

10.1016/j.atherosclerosis.2015.10.020

Reference:

ATH 14320

To appear in:

Atherosclerosis

Received Date: 4 September 2015 Revised Date:

1 October 2015

Accepted Date: 14 October 2015

Please cite this article as: Lu J, Xiang G, Liu M, Mei W, Xiang L, Dong J, Irisin Protects against Endothelial Injury and Ameliorates Atherosclerosis in Apolipoprotein E-Null Diabetic Mice, Atherosclerosis (2015), doi: 10.1016/j.atherosclerosis.2015.10.020. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Irisin Protects against Endothelial Injury and Ameliorates Atherosclerosis in Apolipoprotein E-Null Diabetic Mice

RI PT

Junyan Lu, MD; Guangda Xiang*, MD, PhD; Min Liu, MD; Wen Mei, MD; Lin Xiang, MD; Jing Dong, MD Department of Endocrinology, Wuhan General Hospital of Guangzhou Command, Wuluo Road 627, Wuhan 430070, Hubei Province, China. *Corresponding author: Guangda Xiang, Department of Endocrinology, Wuhan General Hospital of Guangzhou Command, Wuluo Road 627, Wuhan 430070, Hubei Province, China.

SC

Phone number: +86 13517275283 E-mail: [email protected] Fax number: +86 02750772191 Short title Irisin and atherosclerosis

TE D

M AN U

Key Words: irisin, atherosclerosis, endothelium, diabetes mellitus, AMPK Words count: 4622 Figure number: 7 Table number: 1 Conflict of Interest: None.

Junyan Lu, E-mail: [email protected] Min Liu, E-mail: [email protected]

EP

Wen Mei, E-mail: [email protected]

AC C

Lin Xiang, E-mail: [email protected] Jing Dong, E-mail: [email protected]

1

ACCEPTED MANUSCRIPT

Abstract

TE D

M AN U

SC

RI PT

Objective: The circulating irisin increases energy expenditure and improves insulin resistance in mice and humans. The improvement of insulin resistance ameliorates atherosclerosis. Therefore, we hypothesized that irisin alleviates atherosclerosis in diabetes. Methods: Endothelial function was measured by acetylcholine-induced endotheliumdependent vasodilation using aortic rings in apolipoprotein E-Null (apoE-/-) streptozotocin-induced diabetic mice. Atherosclerotic lesion was evaluated by plaque area and inflammatory response in aortas. In addition, the endothelium-protective effects of irisin were also further investigated in primary human umbilical vein endothelial cells (HUVECs) in vitro. Results: The in vivo experiments showed that irisin treatment significantly improved endothelial dysfunction, decreased endothelial apoptosis, and predominantly decreased atherosclerotic plaque area of both en face and cross sections when compared with normal saline-treated diabetic mice. Moreover, the infiltrating macrophages and T lymphocytes within plaque and the mRNA expression levels of inflammatory cytokines in aortas were also significantly reduced by irisin treatment in mice. The in vitro experiments revealed that irisin inhibited high glucose-induced apoptosis, oxidative stress and increased antioxidant enzymes expression in HUVECs, and pretreatment with LY294002, L-NAME, AMPK-siRNA or eNOS-siRNA, attenuated the protection of irisin on HUVECs apoptosis induced by high glucose. In addition, the in vivo and in vitro experiments showed that irisin increased the phosphorylation of AMPK, Akt and eNOS in aortas and cultured HUVECs.

AC C

EP

Conclusions: The present study indicates that systemic administration of irisin may be protected against endothelial injury and ameliorated atherosclerosis in apoE-/- diabetic mice. The endothelium-protective action of irisin was through activation of AMPK-PI3K-Akt-eNOS signaling pathway. Irisin could be therapeutic for atherosclerotic vascular diseases in diabetes. Key Words: Irisin, endothelium, atherosclerosis, diabetes mellitus, AMPK

2

ACCEPTED MANUSCRIPT

Background

EP

TE D

M AN U

SC

RI PT

Cardiovascular disease (CVD) is the leading complication of diabetes and accounts for up to 70% of mortality in patients with diabetes [1,2]. Atherosclerosis, which constitutes the most important contributor to cardiovascular disease, involves an ongoing inflammatory process and oxidative stress in arterial wall. The inflammatory factors and oxidative stress promote macrophages and T lymphocytes infiltration, foam cell formation, extracellular matrix degradation and vascular smooth muscle proliferation leading to atherosclerotic plaque formation. Emerging data support the concept that endothelial injury is an early marker of atherosclerotic vascular disease [3], and has been shown to contribute to atherogenesis [4-6]. Approaches that protect against endothelial injury may be therapeutic for atherosclerosis. Irisin is a cleaved and secreted fragment of the plasma membrane protein fibronectin type III domain-containing protein 5 (FNDC5), and mainly secreted by skeletal muscle [7]. The circulating irisin converts white fat into the more thermogenic beige fat [7], a process known as white fat “browning”. Overexpression of this myokine was sufficient to promote energy expenditure and alleviate insulin resistance in C57/BL mice [7]. Actually, some studies showed that circulating irisin levels were significantly lower in type 2 diabetic patients [8,9]. Moreover, one study showed that plasma irisin levels are negatively associated with homocysteine levels [10], and another study revealed that decreased plasma levels of irisin seem to be associated with urinary albumin excretion in type 2 diabetes [11]. Recently, our and other studies suggested that circulating irisin concentrations are positively associated with endothelial function in type 2 diabetes [11,12], and in non-diabetic patients [13]. Furthermore, one in vitro study showed that irisin can attenuate endothelial cells apoptosis induced by high glucose [14]. Taken together, these data indicate that irisin may be associated with angiopathy in diabetes. However, the potential role of irisin on atherosclerosis has not been investigated. Therefore, we hypothesized that irisin can protect against endothelial

AC C

injury and ameliorate atherosclerosis in diabetic conditions. In the present study, we assessed whether irisin has endothelium-protective and anti-atherosclerotic effects in in vivo and in vitro experiments in diabetic conditions and investigated the possible mechanisms-involved.

Materials and methods

The experiments conformed to the National Institutes of Health Guidelines for the Use of Laboratory Animals. All animal experimental protocols were approved by the animal ethics committee of the Wuhan General Hospital of Guangzhou Command (reference number: WZEA2012008). Thirty nine male 4-week-old apoE-/- mice (The Jackson Laboratory, USA) were housed in a specific–pathogen-free environment with unrestricted access to water. The mice were fed with a high-fat diet (HFD: 45% kcal fat, 35% kcal carbohydrates, and 20% kcal protein) from 4-week-old to the end of the study. After 4 weeks of high-fat diet feeding, a total of twenty six mice were rendered diabetic by five daily intraperitoneal injections of 3

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

streptozotocin (STZ, Sigma, US) at a dose of 50mg/kg. The other mice received the vehicle citrate buffer alone (non-diabetic, n=13). Diabetic animals were further divided into 2 groups (13 mice in each group) matching serum glucose and body weight, then treated with irisin (Phoenix Pharmaceuticals, US) or normal saline (NS), respectively. For diabetic + irisin group, each mouse received tail vein injection of irisin (2µg per mouse in a total of 100µL NS) twice a week, and for both diabetic + NS and non-diabetic groups, each mouse received an equivalent volume of NS by tail vein injection for 12 weeks. The serum glucose and body weight were followed weekly for the 3 groups. After 4 weeks or 12 weeks of irisin or NS intervention, the mice were anesthetized by intraperitoneal injection of pentobarbital sodium (60mg/kg body weight) and euthanized for endothelial function assessment (n=3 mice in each group at 4th week) or for blood tests and histological examination (at 12th week) (n=10 mice in each group), respectively. The serum concentrations of insulin and irisin were measured with ELISA kits (insulin: EMD Millipore Corporation, USA; irisin: Phoenix Pharmaceuticals, USA). All procedures were performed as the manufacturers’ instructions. Measurements were performed in duplicate.

EP

TE D

Glucose and insulin tolerance test The intraperitoneal glucose tolerance test (IPGTT) and insulin tolerance test (ITT) was performed before and after irisin or NS treatment in 3 groups (n = 13 mice before treatment and n=10 mice after treatment), respectively. For IPGTT, mice were fasted overnight, followed by an intraperitoneal injection of 2 g/kg glucose. For ITT, mice were fasted for 5 h, followed by an intraperitoneal injection of 0.075 units/kg insulin. Blood samples were obtained by tail-bleeding and blood glucose level was checked by a portable glucose meter (Johnson & Johnson, USA) before and after glucose or insulin injection, respectively. The incremental area under the curve (AUC) over the basal value of IPGTT during the time period of 120 min was calculated using the trapezoidal rule and used as an index of glucose tolerance. The incremental area below the basal value of ITT during the time period of 90 min was calculated using the trapezoidal rule and used as an index of insulin tolerance.

AC C

Endothelial function assessment in mice The thoracic aortas were cleaned of connective tissue under the stereo-microscope (Olympus, Japan) and cut into 4 mm rings immediately after euthanasia. Aortic rings were mounted on wire hooks, suspended in organ chambers containing Krebs buffer kept at 37°C and aerated with 95% O2. The wire hooks were connected to force transducers (JH-2, Chengdu TME Technology Co., Ltd, China) to record changes of tension by a Data Acquisition System (BL-420S, Chengdu TME Technology Co., Ltd, China). After equilibration for 1h at a preload tension of 0.5g, the rings were precontracted with norepinephrine (NE, 10-6 mmol/L). Once a steady-state was achieved, vasodilation responses were evaluated by cumulative concentration–response curves to acetylcholine (ACh, 10−9 to 10−4 mmol/L) and sodium nitroprusside (SNP, 10−9 to 10−4 mmol/L). In ex vivo experiments, thirty apoE-/- male mice (4-week-old, weighting 18-20g) were 4

ACCEPTED MANUSCRIPT

RI PT

divided into 6 groups with 5 mice in each group. After the mice were euthanized, the aortic rings were prepared and the aortic rings in each group were then incubated in the following medium: (1) Normal glucose group (NG): glucose 5.5mmol/L in Krebs' solution, (2) High glucose group (HG): glucose 20mmol/L in Krebs' solution, (3) Irisin group (Irisin): irisin 100ng/ml in Krebs' solution with glucose 20mmol/L, (4) Compound C group (Compound C, an AMPK inhibitor): Compound C 500µmol/L + irisin 100ng/ml in Krebs' solution with glucose 20mmol/L, (5) LY294002 group (LY294002, a PI3K inhibitor): LY294002 10µmol/L + irisin 100ng/ml in Krebs' solution with glucose 20mmol/L, (6) L-NAME group (L-NAME, an eNOS inhibitor): L-NAME 500µmol/L + irisin 100ng/ml in Krebs' solution with glucose 20mmol/L. The endothelium-dependent and endothelium-independent vasodilation was measured.

M AN U

SC

Quantification of atherosclerotic plaques To evaluate the atherosclerotic plaque in apoE-/- mice, two complementary approaches were used (en face and cross sections analyses). After irisin or NS treatment for 12 weeks, whole aortas (n=3 in each group) were dissected and stained with Oil Red O (Sigma, USA) for plaque en face area analysis. The extent of atherosclerosis was expressed as the percent of the lesion area extending from the ascending aorta to the iliac bifurcation. To quantify luminal cross-sectional area involved by atherosclerotic plaque, 8 sections obtained every 20µm from the abdominal aorta (n=7 in each group) were stained with hematoxylin and eosin, and analyzed by Image-Pro Plus 6.0 (Media Cybernetics, USA).

AC C

EP

TE D

Cell culture Primary HUVECs were purchased from ScienCell (USA). Cells were seeded at a density of 5×103/cm2 flask in M199 (Gibco, USA), supplemented with 20% fetal bovine serum (FBS, Gibco, USA), 10µg/ml heparin, 1% penicillin/streptomycin, and 50µg/ml endothelial cell growth factor (ECGF, Sigma, USA). The cultures were maintained at 37°C with a 5%CO2 atmosphere and the media were refreshed every third day. Endothelial cells of the third to fifth passages were used for experiments. Cells were starved in M199 containing 1% FBS for 20h to synchronize before cytokines treatment. Gene silencing A synthetic small interfering RNA (siRNA) was used to knockdown the expression of adenosine monophosphate-activated protein kinase (AMPK) and endothelial nitric oxide synthase (eNOS) in HUVECs. The siRNA oligonucleotides against AMPK (AMPK-siRNA) eNOS (eNOS-siRNA) and scrambled control (Scr-siRNA) were transfected into cells by using TransMessenger Reagent (Qiagen, Canada) according to the manufacturer’s instructions. A successful knockdown of AMPK and eNOS genes was determined by RT-PCR and Western blot analyses. Analysis of endothelial cell apoptosis in vivo and in vitro Endothelial cell apoptosis in aortas was detected by double-stain with terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling (TUNEL) and anti-CD31, and 5

ACCEPTED MANUSCRIPT

RI PT

HUVECs apoptosis was evaluated by TUNEL single-staining. HUVECs were divided into 6 groups and pretreated with inhibitors against AMPK (Compound C, 500µmol/L), PI3K (LY294002, 10µmol/L) or eNOS (L-NAME, 500µmol/L) for 30min before irisin (100ng/ml) was added. The cells were then cultured for 48h with media containing 5.5mmol/L glucose (NG group) or 20mmol/L glucose (HG group). Moreover, to verify the role of AMPK or eNOS in endothelial cells apoptosis, HUVECs were transfected with Scr-siRNA, AMPK-siRNA or eNOS-siRNA for 24h, followed by cultured in normal (5.5mmol/L) or high (20mmol/L) glucose for 48h in the presence or absence of irisin (100ng/ml). After treatment, the cells were double-stained with Annexin V-FITC/PI (BD, USA) and analyzed by flow cytometry.

M AN U

SC

Western blot analyses Proteins extracted from the aortas of apoE-/- mice and cultured HUVECs were separated by SDS-PAGE and transferred to nitrocellulose membranes. After blocking with 5% bovine serum albumin (BSA) for 1h at room temperature, the membranes were incubated with primary antibodies including P-AMPKα, AMPK, P-Akt, Akt, P-eNOS (Serine 1177), eNOS, Bcl-2 and Bax (all primary antibodies were purchased from CST, USA) overnight at 4°C. The membranes were washed several times in TBST and then incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG for 1h at room temperature. The specific reactions were revealed with the enhanced chemiluminescence detection system (Perkin-Elmer, USA).

TE D

Real-time PCR (RT-PCR) To determine the expression levels of inflammatory cytokines in aortas and antioxidant enzymes in cultured HUVECs, the interleukin (IL)-10, IL-6, tumor necrosis factor-alpha

AC C

EP

(TNF-α), intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1) in aortas, and monocyte chemoattractant protein-1 (MCP-1), glutathione peroxidase type 1 (GPX-1), catalase (CAT) and superoxide dismutase (SOD) in HUVECs were measured by quantitative RT-PCR. Total RNA was extracted from thoracic aortas (n=7 mice in each group) or cultured HUVECs with Trizol reagent (Sigma, USA) according to the manufacturer’s instructions. The mRNA was then reverse-transcribed to cDNA using iScript™ cDNA Synthesis kit (Bio-Rad, USA). Relative changes in mRNA levels among groups were determined with 2∆∆Ct method. Detection of reactive oxygen species (ROS) generation in HUVECs Intracellular ROS generation was detected by using the peroxide-sensitive fluorescent probe 2’, 7’-dichlorofluorescin diacetate (DCFH-DA). HUVECs were seeded in 96-well plates. As described above, HUVECs were divided into 6 groups and treated with compound C, LY294002, L-NAME or irisin for 48h, and then the cells were incubated with DCFH-DA (10µmol/L per well) for 30min at 37°C. After that, the cells were washed with phosphate buffered saline (PBS) and imaged by fluorescence microscope. The fluorescence of DCFH-DA was measured by fluorescence plate reader Victor3TM (Perkin-Elmer, USA). Statistical analysis 6

ACCEPTED MANUSCRIPT Data are expressed as mean ± SD. Differences between 2 groups were tested with unpaired Student’s t test. Data of multiple groups were compared using one-way analysis of variance (ANOVA) followed by least significant difference (LSD) t-test. Statistical significance was defined as p < 0.05. All analyses were performed with SPSS 19.0 (IBM, USA).

Results

SC

RI PT

The effects of irisin on metabolic characteristics of ApoE-/- mice Previous study showed that even moderately increased levels of circulating irisin potently increase energy expenditure, reduce body weight and improve diet-induced insulin resistance [7]. Thus, the effects of irisin on metabolic characteristics were detected in mice in the present study. To explore the effects of irisin on insulin sensitivity in STZ-induced diabetic mice, IPGTT and ITT was performed before and 12 weeks after irisin or NS treatment. Firstly, to determine the half-life of circulating irisin, 10 male 10-week old C57BL/6 mice were purchased, then each mouse received tail vein injection of irisin (2µg per mouse in a total of

M AN U

100µL NS) for one time according to literature [7] and the circulating irisin levels were measured with ELISA kits (irisin: Phoenix Pharmaceuticals, USA) at day 0, 1, 2, 3, 4, 5, and day 6 by tail venous. Since the serum concentrations decreased 21% at day 3, and 60% at day 4, such, we injected irisin twice a week for mice (Supplementary Figure 1, A). So far, receptor(s) for irisin has not been identified yet. According to the results (Supplementary Figure 1, A), the plasma half-life of intravenously injected irisin is surprisingly long, longer

TE D

than 4 days, despite the findings showing irisin is a bioactive molecule probably acting via specific receptors. Such, further study for its receptor(s) are needed. Secondly, before irisin or NS treatment, diabetes displayed a significant decreased glucose and insulin tolerance when compared with non-diabetes (p < 0.05) (Supplementary Figure 1, B, C, Supplementary Figure

EP

2, A and B). After 12 weeks of irisin or NS treatment, irisin treated mice showed increased glucose and insulin tolerance when compared with mice treated with NS, but which was still

AC C

lower than non-diabetic mice (p < 0.05) (Supplementary Figure 1, D, E, Supplementary Figure 2, C and D). Thirdly, compared with non-diabetic mice, fasting insulin was significantly decreased before irisin or NS treatment, and further decreased at 12th week in diabetic mice (p <0.05), but no difference was observed in fasting insulin between the diabetic groups at both two time points (Supplementary Figure 1, E and F). As shown in Table 1, irisin treated mice also exhibited decreased fasting plasma glucose and glycosylated hemoglobin (HbA1c), but had no notable effects on body weight at the end of study (p > 0.05 vs. diabetic + NS group). Irisin improved endothelial dysfunction in ApoE-/- diabetic mice Endothelial dysfunction is an early marker of atherosclerotic vascular disease [3]. We next examined the effects of irisin on endothelial function at 4th week after irisin or NS treatment in mice. The results showed that diabetes decreased the endothelium-dependent vasodilation in response to ACh, while irisin treatment markedly improved the vasodilation in diabetic 7

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

condition (82.43 ± 4.79 % vs. 61.54 ± 4.62% vasodilation at 10-4 mmol/L ACh, 5 aortic rings from each mice, p < 0.05 vs. diabetic + NS group) (Fig. 1, A). In contrast, the endothelium-independent vasodilation in response to SNP did not differ among the 3 groups (Fig. 1, B). To explore the possible mechanisms-involved, proteins extracted from thoracic aortas after 12 weeks of irisin or NS treatment were analyzed by Western blot. The eNOS phosphorylation was significantly lower in diabetic group than that in non-diabetic group, and irisin treatment apparently increased eNOS phosphorylation compared with the diabetic + NS (Fig. 1, C and D). Nitric oxide (NO), produced by eNOS, plays a pivotal role in regulating endothelial cell function, apoptosis, and exhibits athero-protective effects [3]. The serum concentrations of NO decreased in diabetic mice and irisin treatment significantly increased the NO levels after irisin or NS treatment (Supplementary Figure 3, A), which are consistent with eNOS phosphorylation levels in aortas of 3 groups. The phosphorylation of eNOS can be regulated by several kinases [15,16]. Thus, we examined the changes in the levels of PKA, AMPK, and Akt in aortas. Irisin treatment for 3 months increased AMPKα and Akt phosphorylation when compared with the diabetic + NS group, and did not influence the protein expression of PKA (Fig. 1, C and D). To determine whether the changes in the levels of AMPK, Akt and eNOS phosphorylation are responsible for the improvement of endothelial function in irisin treatment group, we performed ex vivo intervention experiments of irisin, high glucose, compound C, LY294002 and L-NAME on aortic rings from apoE-/- male mice. The concentration-dependent vasodilation in response to ACh was significantly decreased in the HG group compared with NG group (38.56 ± 3.63% vs. 95.52 ± 3.14 % vasodilation at 10-4 mmol/L ACh, n=10 aortic rings in each group, p < 0.01 vs. NG group), while irisin can attenuate the decrease of endothelial function induced by high glucose (87.43 ± 3.79 % vs. 38.56 ± 3.63% vasodilation at 10-4 mmol/L ACh, n=10 aortic rings in each group, p < 0.05 vs. HG group), which was still lower than that in NG group (95.52 ± 3.14 % vasodilation at 10-4 mmol/L ACh, p < 0.05). The protective effect of irisin on endothelial dysfunction was abolished by compound C or LY294002 or L-NAME (Fig.1, E). In contrast, the vascular endothelium-independent vasodilation did not differ among different groups (Fig.1, F). Irisin decreased endothelial cell apoptosis in vivo and in vitro Apoptosis of endothelial cells has been suggested to play a role in atherosclerosis [17]. Thus, we tested the apoptosis of endothelial cells in vivo and in vitro in the present study. TUNEL staining with double-stain was applied to identify the apoptotic cells in aortic cross sections of apoE-/- mice treated with irisin or not (Fig. 2, A). The percentage of TUNEL-positive cells within the endothelium was significantly increased in diabetic mice, and irisin treatment significantly decreased apoptotic endothelial cells (22.51 ± 1.92% in diabetic + irisin group vs. 42.83 ± 3.07% in diabetic + NS group, p < 0.01) (Fig. 2, B). As expected, the electron microscopy also showed apoptotic profiles with karyopyknosis in some of endothelial cells and disconnection from the underlying internal elastic lamin (IEL) in diabetic + NS group, 8

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

while most of endothelial cells covered well on the surface of luminal wall in diabetic + irisin group (Fig. 2, C). For the in vitro experiments, first of all, we investigated the concentration-dependent and time-dependent responses for irisin in the phosphorylation of eNOS (Ser1177) by Western blot, and the results showed that the best concentration of irisin for eNOS activation was 100ng/ml, and the best time was 30 minutes (Supplementary Figure 4). Using this intervention condition, the in vitro experiments were performed. TUNEL staining exhibited that exposure of endothelial cells to high glucose (20mmol/L) for 48h resulted in a significant increase in apoptosis when compared with NG group (25.96 ± 3.46% vs. 6.23 ± 1.61%, p < 0.01 vs. NG group) (Fig. 3, A and B), and irisin reduced high glucose-induced endothelial cell apoptosis (20.60 ± 2.79%, p < 0.05 vs. HG group). The protective role of irisin on endothelial apoptosis was blocked by the addition of compound C, LY294002 or L-NAME. Moreover, the suppression of AMPK and eNOS expression by siRNA significantly restrained the effect of irisin on endothelial apoptosis compared with Scr-siRNA transfection (p < 0.05) (Fig. 4, A, B, D and E). In addition, Western blot analysis of apoptotic proteins (Bcl-2 and Bax) in cultured endothelial cell is consistent with these findings (Fig. 3C, 4C and 4F). Taken together, these data suggest that irisin has protective effect of endothelial cell damage induced by diabetes, which may be mediated by the AMPK-PI3K-Akt-eNOS signaling pathway.

AC C

EP

TE D

Irisin stimulated AMPK-PI3K-Akt-eNOS signaling in cultured endothelial cells Our in vivo and ex vivo experiments showed that AMPK-PI3K-Akt-eNOS signaling was involved in the protective effect of irisin on endothelium. We next explored the effects of irisin on AMPK-PI3K-Akt-eNOS pathway in vitro. Firstly our results showed that high glucose inhibited Akt and eNOS phosphorylation, and irisin (100ng/ml) increased Akt and eNOS phosphorylation in high glucose condition in HUVECs (Supplementary Figure 5, A and B), and the stimulatory effects of irisin on Akt and eNOS phosphorylation were blocked by LY294002 or L-NAME. Furthermore, the concentrations of NO in culture media by different treatment were consistent with the changes of eNOS phosphorylation (Supplementary Figure 3, B). The levels of PKA phosphorylation did not change among different treatment groups (Supplementary Figure 5, C). Secondly, PI3K activity can be regulated by AMPK in endothelial cells [20,21]. We found that irisin stimulated the phosphorylation of AMPKα, Akt and eNOS against high glucose (Fig. 5, A, B and C), and knock-down of the AMPK by siRNA can inhibit these effects of irisin. These findings indicated that irisin could activate AMPK-PI3K-Akt-eNOS signaling pathway in endothelial cells. Irisin alleviated atherosclerotic lesion formation in ApoE-/- diabetic mice Endothelial injury is the early pathophysiologic change of atherosclerosis, approaches that protect against endothelial injury may be therapeutic for atherosclerosis. Therefore, we investigated whether irisin alleviate atherosclerosis besides protecting vascular endothelium. For this purpose, en face and cross sections analyses were performed in atherosclerotic 9

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

plaque. By Oil Red O staining (Fig. 6, A), we found that diabetes significantly increased atherosclerotic plaque area, and plaque area of en face sections were significantly decreased in irisin-treated mice when compared with the diabetic + NS group (22.57 ± 2.17% vs. 35.09 ± 2.38%, p < 0.05) (Fig. 6, B). Similarly, as compared with NS treatment, irisin decreased plaque area of cross sections (16.87 ± 1.85% vs. 33.46 ± 5.77%, p < 0.05 vs. diabetic + NS group) (Fig. 6, C and D). Endothelial injury decreases NO production, leading to inflammatory lesion in aortas [18, 19]. In our in vivo study, irisin protected against endothelial injury and increased serum NO levels in diabetic mice. We next questioned whether irisin could reduce the inflammatory lesion in aortas. Therefore, identification of macrophages and T lymphocytes in atherosclerotic plaque was performed with anti-CD68 and anti-CD90 staining, respectively (Fig. 7, A). Quantification of CD68- and CD90-positive area revealed that diabetes significantly increased inflammatory infiltration within plaque, and irisin treatment significantly reduced inflammatory infiltration within plaque when compared with diabetic + NS group (CD68: 35.37 ± 3.59% vs. 50.02 ± 5.73%, p < 0.05; CD90: 22.52 ± 4.43% vs. 34.09 ± 4.08%, p < 0.05, vs. diabetic + NS) (Fig. 7, B). In addition, the inflammatory cytokines in aortas were assessed by quantitative RT-PCR. As shown in Figure 7, C, irisin treatment significantly down-regulated the mRNA expression levels of IL-6, TNF-α, ICAM-1, VCAM-1 and MCP-1, but not that of IL-10 (p > 0.05).

AC C

EP

TE D

Irisin inhibited oxidative stress in vivo and in vitro Our previous studies have revealed that high glucose increased ROS generation in endothelial cells, resulting in endothelial injury [20-23]. In the present study, we found that irisin protected against endothelial injury in vivo and in vitro. Therefore, we investigated the effect of irisin on oxidative stress. The results showed that high glucose-induced reduction in total antioxidant capacity was significantly improved by irisin treatment in mice (Supplementary Figure 6, A). In addition, a 2.85-fold increase in superoxide production was observed in aortic ring segments isolated from apoE-/- diabetic mice, and treatment with irisin significantly prevented aortic superoxide overproduction (Supplementary Figure. 6, B). We also measured intracellular ROS in cultured endothelial cells incubated with high glucose in the presence or absence of irisin. High glucose-induced ROS generation in endothelial cells was dramatically increased compared with NG group (p < 0.01), whereas irisin treatment markedly inhibited it (p < 0.01 vs. HG group) (Supplementary Figure 7, A and B). Moreover, the inhibitory effect of irisin on ROS generation was abolished by compound C or LY294002 or L-NAME. On the other hand, the mRNA expression levels of GPX-1, CAT and SOD were reduced in high glucose-treated endothelial cells, and irisin treatment significantly up-regulated antioxidant enzymes expression (p < 0.05), and the effect could also be blocked by compound C or LY294002 or L-NAME (Supplementary Figure 7,C).

Discussion 10

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

In the present study, we have demonstrated for the first time that systemic administration of irisin protected against endothelial injury and ameliorated atherosclerosis induced by diabetes in the apoE-/- mice. Such benefit is related to an activation of AMPK-PI3K-Akt-eNOS signaling pathway and inhibition of oxidative stress by irisin. Thus, irisin can be considered as a potential therapeutic for diabetic angiopathy. Atherosclerosis is considered as an inflammatory disease [24]. Inflammatory process participates in all stages of atherosclerosis [24,25]. The endothelium plays a crucial role in regulating inflammatory process [26]. Endothelial injury has been shown to promote atherosclerosis initiation and progression [27]. Numerous studies have evaluated interventions to improve endothelial dysfunction and many of these are known to limit atherosclerotic vascular disease [28-30]. The present study showed that diabetes reduced endothelium-dependent vasodilation and increased endothelial apoptosis in apoE-/- mice, while treatment with irisin predominantly improved endothelial dysfunction and reduced endothelial apoptosis. Moreover, our data demonstrated that irisin decreased the plaque area and the infiltrating macrophages and T lymphocytes in the plaques, and down-regulated the mRNA expression of inflammatory cytokines in the aortas. Also, our in vitro experiments suggested that irisin decreased endothelial cell apoptosis induced by high glucose. Moreover, the irisin concentration 100 ng/ml used for the cell culture experiments is very similar to that of plasma levels of mice (Table 1), indicating that endogenous irisin circulating in the blood has a very significant role in regulating endothelial function and development of atherosclerotic lesions. Taken together, these results suggested that irisin could protect against endothelial injury, consequently ameliorate inflammatory reaction and atherosclerosis. Numerous studies have shown that inflammation results in insulin resistance, and the improvement of insulin sensitivity alleviates endothelial dysfunction and atherosclerosis [31-34]. In this study, glucose and insulin tolerance tests in STZ-induced apoE-/- diabetic mice fed by high fat diet displayed impairment in both glucose tolerance and and insulin sensitivity when compared with non-diabetic mice, and irisin treatment for 12 weeks significantly improved insulin and glucose tolerance when compared with diabetic without irisin, suggesting that irisin can improve insulin resistance in STZ-induced apoE-/- diabetic mice. As a result, the fasting glucose and HbA1c in apoE-/- diabetic mice significantly decreased in irisin treatment group, which is consistent with the previous study in insulin resistance mice [7]. Therefore, the improvement of plasma glucose and insulin sensitivity can partially explain the endothelium-protective as well as anti-atherosclerotic effects of irisin. However, our ex vivo intervention experiments on aortic rings from apoE-/- male mice showed that irisin can reverse the impairment of endothelium-dependent vasodilation induced by high glucose, indicating that irisin may improve endothelial dysfunction by direct effect on endothelium independent of directly ameliorating diabetic control. We next questioned whether oxidative stress involved in the endothelium-protective and anti-atherosclerotic effects of irisin in diabetic conditions. Many studies have revealed that 11

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

exposure of endothelial cells to high glucose apparently increases the generation of ROS [35, 36], which inhibits production of NO by blocking eNOS phosphorylation and activates the apoptotic pathways implicated in endothelial cells. This contributes to endothelial damage and accelerates atherosclerosis. Moreover, hyperglycemia blunted the defense of antioxidant enzymes, thus allowing excessive intracellular ROS accumulation resulting in endothelial damage. Our results showed that high glucose decreased eNOS phosphorylation, NO production and increased ROS generation, consequently endothelial injury and atherosclerosis were deteriorated, and irisin can partially reverse these reactions. We found that irisin prevented ROS overproduction in aortic ring segments and suppressed high glucose-induced ROS generation in HUVECs. Furthermore, irisin up-regulated the mRNA expression of antioxidant enzymes (GPX-1, CAT and SOD) in HUVECs, consequently endothelial injury and atherosclerosis were ameliorated. At least in part, we can conclude that irisin improves diabetes-induced endothelial dysfunction and apoptosis through eNOS-NO-ROS pathway. The molecular signal pathways for the protective effect of irisin on artery are unclear. It has been proven that AMPK-PI3K-Akt-eNOS pathway plays an important role in prevention of endothelial injury [20,21,37]. Therefore, we examined the contribution of the AMPK-PI3K-Akt- eNOS pathway to the protective effects of irisin in high glucose condition. Firstly, irisin stimulated phosphorylation of AMPK, Akt and eNOS in aortas of apoE-/- mice, and consequently serum NO concentrations increased in irisin treated mice. Secondly, the improvement of endothelial function of irisin from ex vivo intervention experiments on aortic rings was abolished by addition of inhibitors for AMPK, Akt and eNOS. Thirdly, irisin reversed the decreased phosphorylation of AMPK, Akt and eNOS, and production of NO in cultured endothelial cells, and the effect of irisin on eNOS phosphorylation and NO production could also be blocked by compound C, LY294002 and L-NAME. Fourthly, knockdown of eNOS expression by siRNA restrained the protective effect of irisin on endothelial apoptosis. These results suggested that the AMPK-PI3K-Akt-eNOS pathway is involved in the endothelium-protective effects of irisin. However, Very recently, Haibo Song et al. found that treating HUVECs with irisin (20nmol/L) had no effect on Akt signaling [14]. Possibly the concentration of irisin was not enough to activate Akt signaling in HUVECs. Some limitations should be mentioned here. Firstly, the relationship between cardiovascular disease and diabetes has been mechanistically demonstrated in a recent report focusing on intracellular calcium [38]. However, extracellular calcium ions in endothelium were not measured in the present study. Secondly, we did not explore the Akt/mammalian target of rapamycin (mTOR) pathway, because some studies showed that Akt/mTOR pathway was involved in cardiovascular diseases [39,40]. Thirdly, it is established that calcium/calmodulin-dependent kinase IV (CaMKIV) and adrenergic system are associated with the regulation of vascular tone [41,42]. In the present study, we did not investigate the effect of irisin on the association between CaMKIV, adrenergic system and vascular tone. Fourthly, we did not measure blood pressure or serum lipids, such the changes of blood pressure and serum lipids before and after irisin treatment could not be evaluated. Finally, we can not provide a better imaging of the endothelial layer (i.e. using en face preparations of the 12

ACCEPTED MANUSCRIPT vessels). Therefore, some further studies are needed. In summary, by studying apoE-/- diabetic mice treated with irisin, our results showed that irisin may be play a crucial role in protecting against endothelial injury and alleviating atherosclerosis. The in vivo, ex vivo and in vitro experiments established that the protective effects of irisin on endothelium are through the activation of AMPK-PI3K-Akt-eNOS

Abbreviations

apoE, apolipoprotein E eNOS, endothelial nitric oxide synthase

M AN U

HUVEC, human umbilical vein endothelial cell IPGTT, intraperitoneal glucose tolerance test

SC

Ach, acetylcholine AMPK, adenosine monophosphate-activated protein kinase

RI PT

signaling pathway. Our study suggests that irisin could be therapeutic for the treatment of atherosclerotic vascular diseases in diabetes mellitus.

ITT, insulin tolerance test L-NAME, L-nitro-arginine methyl ester PI3K, phosphatidyl inositol 3-kinase Scr-siRNA, scrambled small interfering RNA

NS, normal saline

TE D

siRNA, small interfering RNA SNP, sodium nitroprusside

EP

Funding

AC C

This work was supported by National Natural Science Foundation of China (81370896) and Natural Science Foundation of Hubei Province (number 2011CDA002 and number 2013CFB432).

Acknowledgments

We thank Professor Changhua Wang, from the Department of Pathophysiology, Medical College of Wuhan University, for excellent technical assistance.

Conflict of Interest None. 13

ACCEPTED MANUSCRIPT

Author’s contributions Guangda Xiang performed research design; Junyan Lu and Min Liu conducted the animal experiment and the data analysis; Wen Mei and Lin Xiang performed the in vitro experiment;

Author detailed

RI PT

Jing Dong conducted the manuscript writing.

M AN U

SC

Department of Endocrinology, Wuhan General Hospital of Guangzhou Command, Wuluo Road 627, Wuhan 430070, Hubei Province, China

References

G. L. Booth, M. K. Kapral, K. Fung, J. V. Tu, Recent trends in cardiovascular complications among men and women with and without diabetes. Diabetes Care 29(2006)32-37.

2.

S. R. Preis, S-J. Hwang, S. Coady, M. J. Pencina, R. B. Sr. D'Agostino, P. J. Savage, et al., Trends in all-cause and cardiovascular disease mortality among women and men with and without diabetes mellitus in the Framingham Heart Study, 1950 to 2005. Circulation 119(2009)1728-1735.

3.

S. Verma, M. R. Buchanan, T. J. Anderson, Endothelial Function Testing as a Biomarker of Vascular Disease. Circulation 108(2003)2054-2059.

4.

P. J. Kuhlencordt, R. Gyurko, E. Han, M. Scherrer-Crosbie, T. H. Aretz, R. Hajjar, et al., Accelerated atherosclerosis, aortic aneurysm formation, and ischemic heart disease in apolipoprotein E/endothelial nitric oxide synthase double-knockout mice. Circulation 104(2001)448-454.

6.

EP

AC C

5.

TE D

1.

J. W. Knowles, R. L. Reddick, J. C. Jennette, E. G. Shesely, O. Smithies, N. Maeda, Enhanced atherosclerosis and kidney dysfunction in eNOS(-/-)Apoe(-/-) mice are ameliorated by enalapril treatment. J Clin Invest 105 (2000) 451-458.

R. De Caterina, P. Libby, H. B. Peng, V. J. Thannickal, T. B. Rajavashisth, M. A. Jr. Gimbrone, et al., Nitric oxide decreases cytokine-induced endothelial activation: Nitric oxide selectively reduces endothelial expression of adhesion molecules and proinflammatory cytokines. J Clin Invest 96(1995) 60-68. 14

ACCEPTED MANUSCRIPT P. Boström, J. Wu, M. P. Jedrychowski, A. Korde, L. Ye, J. C. Lo, et al., A PGC1-α-dependent myokine that drives brown-fat-like development of white fat and thermogenesis. Nature 481(2012)463-468.

8.

J-J. Liu, M. D. S. Wong, W. C. Toy, C. S. Tan, S. Liu, X. W. Ng, et al., Lower circulating irisin is associated with type 2 diabetes mellitus. J Diabetes Complications 27 (2013) 365-369.

9.

Y-K. Choi, M-K. Kim, K. H. Bae, H. A. Seo, J. Y. Jeong, W. K. Lee, et al., Serum irisin levels in new-onset type 2 diabetes. Diabetes Res Clin Pract 100 (2013) 96-101.

10.

R. Alis, F. Sanchis-Gomar, H. Pareja-Galeano, A. Hernández-Mijares, M. Romagnoli, V. M. Víctor, et al., Association between irisin and homocysteine in euglycemic and diabetic subjects. Clin Biochem 47(2014) 333-335.

11.

H-H. Wang, X-W. Zhang, W. Chen, Q-X. Huang, Q-Q. Chen, Relationship between serum irisin levels and urinary albumin excretion in patients with type 2 diabetes. J Diabetes Complications 29(2015) 384-389.

12.

L. Xiang, G. Xiang, L. Yue, J. Zhang, L. Zhao, Circulating irisin levels are positively associated with endothelium-dependent vasodilation in newly diagnosed type 2 diabetic patients without clinical angiopathy. Atherosclerosis 235(2014) 328-333.

13.

N. Hou, F. Han, X. Sun, The relationship between circulating irisin levels and endothelial function in lean and obese subjects. Clin Endocrinol (Oxf)(2015) in press

14.

H. Song, F. Wu, Y. Zhang, Y. Zhang, F. Wang, M. Jiang, et al., Irisin Promotes Human Umbilical Vein Endothelial Cell Proliferation through the ERK Signaling Pathway and Partly Suppresses High Glucose-Induced Apoptosis. PLoS One 9 (2014) e110273

15.

Y. C. Boo, G. Sorescu, N. Boyd, I. Shiojima, K. Walsh, J. Du, et al., Shear stress stimulates phosphorylation of endothelial nitric-oxide synthase at Ser1179 by Akt-independent mechanisms: role of protein kinase A. J Biol Chem 277(2002) 3388-3396.

17.

SC

M AN U

TE D

EP

AC C

16.

RI PT

7.

B. J. Michell, Z. P. Chen, T. Tiganis, D. Stapleton, F. Katsis, D. A. Power, et al., Coordinated control of endothelial nitric-oxide synthase phosphorylation by protein kinase C and the cAMP-dependent protein kinase. J Biol Chem 276(2001)17625-1768. M. M. Birck, A. Saraste, P. Hyttel, M. Odermarsky, P. Liuba, P. Saukko, et al. Endothelial cell death and intimal foam cell accumulation in the coronary artery of infected hypercholesterolemic minipigs. J Cardiovasc Transl Res 6(2013) 579-587.

15

ACCEPTED MANUSCRIPT K. A. Lindstedt, M. J. Leskinen, P. T. Kovanen, Proteolysis of the pericellular matrix: A novel element determining cell survival and death in the pathogenesis of plaque erosion and rupture. Arterioscler Thromb Vasc Biol 24(2004)1350-1358.

19.

M. M. Kockx, Apoptosis in the atherosclerotic plaque: quantitative and qualitative aspects. Arterioscler Thromb Vasc Biol 18(1998)1519-1522

20.

C. F. García-Prieto, H. Pulido-Olmo, G. Ruiz-Hurtado, M. Gil-Ortega, I. Aranguez, M. A. Rubio, et al. Mild caloric restriction reduces blood pressure and activates endothelial AMPK-PI3K-Akt-eNOS pathway in obese Zucker rats. Vascul Pharmacol 65-66(2014) 3-12.

21.

C. F. García-Prieto, F. Hernández-Nuño, D. Del. Rio, G. Ruiz-Hurtado, I. Aránguez, M. Ruiz-Gayo, et al. High-fat diet induces endothelial dysfunction through a down-regulation of the endothelial AMPK-PI3K-Akt-eNOS pathway. Mol Nutr Food Res 59 (2015) 520-532.

22.

M. Liu, G. Xiang, J. Lu, L. Xiang, J. Dong, W. Mei, TRAIL protects against endothelium injury in diabetes via Akt-eNOS signaling. Atherosclerosis 237(2014) 718-724.

23.

G. Xiang, J. Pu, L. Yue, J. Hou, H. Sun, Apha-lipoic acid can improve endothelial dysfunction in subjects with impaired fasting glucose. Metabolism 60 (2011) 480-485.

24.

K. J. Williams, I. Tabas, Atherosclerosis--an inflammatory disease. N Engl J Med 340(1999)1928.

25.

P. Libby, Inflammation in atherosclerosis. Nature 420(2002) 868-874.

26.

A. Lerman, A. M. Zeiher, Endothelial function: cardiac events. Circulation 111(2005) 363-368.

27.

J. Davignon, P. Ganz, Role of endothelial dysfunction in atherosclerosis. Circulation 109(2004)(23 Suppl 1): III27-32.

29.

TE D

EP

AC C

28.

M AN U

SC

RI PT

18.

R. Hambrecht, A. Wolf, S. Gielen, A. Linke, J. Hofer, S. Erbs, et al., Effect of Exercise on Coronary Endothelial Function in Patients with Coronary Artery Disease. N Engl J Med 342(2000) 454-460. C. B. Treasure, J. L. Klein, W. S. Weintraub, J. D. Talley, M. E. Stillabower, A. S. Kosinski, et al., Beneficial Effects of Cholesterol-Lowering Therapy on the Coronary Endothelium in Patients with Coronary Artery Disease. N Engl J Med 332(1995) 481-487.

16

ACCEPTED MANUSCRIPT A. Prasad, T. Tupas-Habib, W. H. Schenke, R. Mincemoyer, J. A. Panza, M. A. Waclawin, et al. Acute and Chronic Angiotensin-1 Receptor Antagonism Reverses Endothelial Dysfunction in Atherosclerosis. Circulation 101(2000) 2349-2354.

31.

G. Arcaro, A. Cretti, S. Balzano, A. Lechi, M. Muggeo, E. Bonora, et al., Insulin causes endothelial dysfunction in humans: Sites and mechanisms. Circulation 105(2002) 576-582.

32.

A. A. Lteif, K. Han, K. J. Mather, Obesity, insulin resistance, and the metabolic syndrome: determinants of endothelial dysfunction in whites and blacks. Circulation 112(2005) 32-38.

33.

D. S. Bell, J. H. ÓKeefe, Lowering the triglyceride/high-density lipoprotein cholesterol and its association with the beneficial impact of pioglitazone on coronary atherosclerosis in the PERISCOPE study is likely due to lowering insulin resistance. J Am Coll Cardiol 58(2011) 778.

34.

E. E. Mulvihill, J. M. Assini, J. K. Lee, E. M. Allister, B. G. Sutherland, J. B. Koppes, et al., Nobiletin attenuates VLDL overproduction, dyslipidemia, and atherosclerosis in mice with diet-induced insulin resistance. Diabetes 60(2011) 1446-1457

35.

L. Quagliaro, L. Piconi, R. Assaloni, L. Martinelli, E. Motz, A. Ceriello, Intermittent high glucose enhances apoptosis related to oxidative stress in human umbilical vein endothelial cells: the role of protein kinase C and NAD(P)H-oxidase activation. Diabetes 52(2003) 2795-2804

36.

T. Inoguchi, P. Li, F. Umeda, H. Y. Yu, M. Kakimoto, M. Imamura, et al., High glucose level and free fatty acid stimulate reactive oxygen species production through protein kinase C--dependent activation of NAD(P)H oxidase in cultured vascular cells. Diabetes 49(2000) 1939-1945

37.

F. M. Ho, W. W. Lin, B. C. Chen, C. M. Chao, C. R. Yang, L. Y. Lin, et al. High glucose-induced apoptosis in human vascular endothelial cells is mediated through NF-kappaB and c-Jun NH2-terminal kinase pathway and prevented by PI3K/Akt/eNOS pathway. Cell Signal 18(2006) 391-399

AC C

EP

TE D

M AN U

SC

RI PT

30.

38. G. Santulli, G. Pagano, C. Sardu, W. Xie, S. Reiken, S. L. D'Ascia, et al. Calcium release channel RyR2 regulates insulin release and glucose homeostasis. J Clin Invest 125 (2015) 1968-78 39. Q. Yuan, Z. Chen, G. Santulli, L. Gu, Z. G. Yang, Z. Q. Yuan, et al. Functional role of Calstabin2 in age-related cardiac alterations. Nature Sci Rep 4(2014)7425

17

ACCEPTED MANUSCRIPT 40. G. Santulli, H. Totary-Jain. Tailoring mTOR-based therapy: molecular evidence and clinical challenges. Pharmacogenomics14(2013)1517-26 41. G. Santulli, B. Trimarco, G. Iaccarino. G-protein-coupled receptor kinase 2 and hypertension: molecular insights and pathophysiological mechanisms. High Blood Press Cardiovasc Prev 20 (2013) 5-12

EP

TE D

M AN U

SC

RI PT

42. G. Santulli, E. Cipolletta, D. Sorriento, C. Del Giudice, A. Anastasio, S. Monaco, et al. CaMK4 Gene Deletion Induces Hypertension. J Am Heart Assoc 1(2012) e001081

Figure Legends

AC C

Figure 1. Irisin improved endothelial dysfunction in apoE-/- diabetic mice. (1) ApoE-/- mice were treated with irisin or normal saline (NS) for 4 weeks. Aortic rings were collected for endothelial function assessment after euthanasia. A: Endothelium-dependent vasodilation to acetylcholine (ACh). B: Endothelium-independent vasodilation to sodium nitroprusside (SNP). (The data represent mean ± SD. n = 3 mice, 15 aortic rings/group, *p < 0.05 vs. Non-diabetic or diabetic + NS). (2) C: Western blot analyses of protein from the thoracic aortas of mice (n=7 mice in each group) at 12th week after irisin or NS treatment. D: Relative PKA, AMPK, Akt and eNOS protein levels in the aortas for Western blot (The data represent mean ± SD. n = 7 mice in each group, *p < 0.05, **p < 0.01). (3) E and F: Aortic rings from apoE-/- mice without treatment (5 mice in each group) were incubated in different media. Endothelium-dependent vasodilation to Ach (E) and endothelium-independent vasodilation to SNP (F) was detected. NG indicates normal glucose group (5.5mmol/L); HG, High glucose group (20mmol/L) (The data represent mean ± SD. n=10 aortic rings/group, 18

ACCEPTED MANUSCRIPT *p < 0.05, compared with other groups, **p < 0.01, compared with NG group). **

Figure 2. Irisin decreased endothelial apoptosis in apoE-/- diabetic mice. ApoE-/- mice were treated with irisin or normal saline (NS) for 12 weeks. The endothelial

RI PT

apoptosis was evaluated by TUNEL with double staining or electron microscopy at 12th week. A: representative images of TUNEL staining in sections from thoracic aortas (14 sections in each group). Co-stained for apoptosis (TUNEL; green), endothelial cells (ECs) (CD-31; red), and nuclei (DAPI; blue). Arrows indicate the TUNEL-positive cells. Magnification × 400.

SC

Bar=20µm. B: the percentage of TUNEL-positive cells per total nuclei within the endothelium. Values are mean ± SD. *p < 0.05, **p < 0.01. C: electron microscopic examination of ECs. ECs present on the luminal wall of abdominal aortas from apoE-/- mice. Arrow shows apoptotic endothelial cell detached from internal elastic lamin (IEL).

M AN U

Magnification×1500. Bar=20µm.

TE D

Figure 3. Irisin attenuated high glucose-induced endothelial cell apoptosis via AMPK-PI3K-Akt-eNOS signaling pathway. Apoptotic of HUVECs cultured in high glucose with compound C (AMPK inhibitor), LY294002 (LY; PI3K inhibitor), L-NAME (eNOS inhibitor) or irisin were determined by TUNEL staining (A, Magnification×400. Bar=20µm) and Western blot for the expressions of apoptotic proteins (C) as well as the the percentage of apoptotic cells were analyzed. Values are mean ± SD (each experiment repeated 6 times). *p < 0.05, **p < 0.01.

AC C

EP

Figure 4. Reduction of high glucose-induced apoptosis by irisin was blocked by AMPK or eNOS gene silencing. The apoptosis of cultured HUVECs in high glucose with or without irisin after knockdown of AMPK or eNOS expression by small interfering RNA (siRNA) (Scr-siRNA indicates scramble siRNA). A and D: HUVECs apoptosis detected by Annexin V-FITC/PI staining (A: siRNA for eNOS, D: siRNA for AMPK). B and E: the percentage of apoptotic cells (B: after knockdown for eNOS, E: after knockdown for AMPK). C and F: the expressions of apoptotic proteins by Western blot analysis (C: after knockdown for eNOS, F: after knockdown for AMPK).Values are mean ± SD (each experiment repeated 6 times). *p < 0.05, **p < 0.01. Figure 5. Irisin stimulated AMPK-PI3K-Akt-eNOS signaling in endothelial cells cultured with high glucose. HUVECs were cultured in media containing normal glucose (NG, 5.5mmol/L) or high glucose (HG, 20mmol/L) with or without irisin after knockdown of AMPK expression by small interfering RNA (siRNA) (Scr-siRNA indicates scramble siRNA). AMPK (A), Akt (B) and eNOS (C) phosphorylation were detected by Western blot. Values are mean ± SD (each experiment repeated 6 times). *p < 0.05, **p < 0.01. 19

ACCEPTED MANUSCRIPT

RI PT

Figure 6. Irisin alleviated atherosclerotic plaque formation in apoE-/- diabetic mice. ApoE-/- mice were treated with irisin or normal saline (NS) for 12 weeks. Atherosclerotic plaque was evaluated with en face and cross sections analyses. A: Representative en face plaque images of whole aortas (3 mice in each group) stained with Oil red O. Bar=1mm. B: quantification analysis of en face plaque (Image-Pro Plus 6.0). C: Representative images of abdominal aortas cross sections (7 mice in each group) stained with hematoxylin and eosin, showing atherosclerotic plaque (with arrows). D: quantification analysis of aortas sections (Image-Pro Plus 6.0). Magnification×200. Bar=50µm. Values are mean ± SD. *p < 0.05, **p < 0.01.

TE D

M AN U

SC

Figure 7. Irisin reduced inflammatory infiltration within plaque and inflammatory cytokines in aortas of apoE-/- diabetic mice. ApoE-/- mice were treated with irisin or normal saline (NS) for 12 weeks. Atherosclerotic lesion was evaluated by inflammatory infiltration and mRNA expression of inflammatory cytokines. A: Macrophages and T lymphocytes in atherosclerotic plaque detected by anti-CD68 or anti-CD90 staining (Immunohistochemistry), respectively. B: quantification of positive area for anti-CD68 or anti-CD90 (brown) within plaque. Magnification×400. Bar=20µm. C: relative mRNA expression levels of interleukin-10 (IL-10), IL-6, tumor necrosis factor-α (TNF-α), intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1) and monocyte chemoattractant protein-1 (MCP-1) were measured by quantitative RT-PCR in the abdominal aortas (7 mice in each group) and expressed as fold of the mean value in control group. Values are mean ± SD. *p < 0.05, **p < 0.01.

AC C

EP

Table 1. Metabolic characteristics of apoE-/- mice at the end of study

Body weight (g) Plasma glucose (mmol/L) HbA1c (%) Insulin (ng/ml) Irisin (ng/ml)

Non-diabetic (n = 10)

28.1 ± 1.8 12.35 ± 0.74 5.8 ± 0.7 1.95 ± 0.24 105.3 ± 6.9

Diabetic + NS (n = 10)

Diabetic + irisin (n = 10)

25.6 ± 2.5 28.50 ± 2.45 13.5 ± 1.3 0.42 ± 0.11 72.1 ± 10.6

26.2 ± 2.3 21.48 ± 2.75* 10.1 ± 1.1* 0.52 ± 0.12 189.6 ± 15.2**

HbA1c indicates glycosylated hemoglobin. Values are mean ± SD and analyzed by ANOVA. *p < 0.05, **p < 0.01, compared with diabetic + NS group.

p < 0.05,

p < 0.01, compared with other two groups.

20

ACCEPTED MANUSCRIPT

B

Figure 1

C

(n=3 mice in each group)

100

100 Vasodilation (%)

* *

40

40

AMPK

Non-diabetic

20

*

20

P-AMPK

Diabetic + irisin

0

Non-diabetic Diabetic + NS Diabetic + irisin

P-Akt

Diabetic + NS

SC

60

RI PT

*

PKA

60

9

8

7

6

5

4

SNP (-lg mol/L) 0 4

E

0.3 0

Relative P-Akt/Akt

2 1.5 1 0.5 0

**

** *

1.5

**

0.5 0 2 1.5 1

**

80

* *

60

NG

*

40

**

**

*

HG Irisin

60 NG HG

40

Irisin Compound C

**

20 ** **

0.5

Compound C LY294002

20

LY294002 L-NAME

L-NAME

0 0

100

80

*

1

F

TE D

RelativeP-AMPK/AMPK

0.6

Relative P-eNOS/eNOS

RelativeP-PKA/PKA

0.9

**

eNOS

100

Vasodilation (%)

2

1.2

EP

D

8 7 6 5 ACh (-lg mol/L)

AC C

9

Akt P-eNOS

M AN U

Vasodilation (%)

(n=3mice in each group)

Diabetic + irisin

P-PKA

80 80

Non-diabetic Diabetic + NS

Vasodilation (%)

A

9

8 7 6 5 4 ACh (-lg mol/L)

0 9 8 7 6 5 4 SNP (-lg mol/L)

ACCEPTED MANUSCRIPT

Figure 2 A

B CD31

DAPI

Merge

*

C

AC C

EC

EP

Non-diabetic

IEL

Diabetic + NS

40 **

Diabetic + irisin

EC EC

*

30

20

10

0

TE D

Diabetic + irisin

M AN U

Diabetic + NS

SC

Apoptotic Endothelial Cells (%)

RI PT

Non-diabetic

TUNEL

(n=10)

(n=10)

(n=10)

ACCEPTED MANUSCRIPT

A

TUNEL

DAPI

B

Merge

RI PT

NG

HG

** *

40

*

**

30 20 10 0

+ -

+ -

+ + -

+ + + -

+ + + -

+ + +

*

C Bcl-2/Bax Ratio

M AN U AC C

EP

TE D

HG + irisin + Compoubd C

HG + irisin + L-NAME

50

SC

Glucose(5.5mmol/L) Glucose (20mmol/L) Irisin (100ng/ml) Compound C(500µmol/L) LY294002(10µmol/L) L-NAME(500µmol/L)

HG + irisin

HG + irisin + LY294002

Apoptotic Cells (%)

Figure 3

2.5 2 1.5 1 0.5 0

** *

*

Bcl-2 Bax β-actin Glucose(5.5mmol/L) Glucose (20mmol/L) Irisin (100ng/ml) Compound C(500µmol/L) LY294002(10µmol/L) L-NAME(500µmol/L)

+ -

+ -

+ + -

+ + + -

+ + + -

+ + +

ACCEPTED MANUSCRIPT

80 60 40 20

HG+Scr-siRNA+IrisinHG+eNOS-siRNA+Irisin

0

HG+Scr-siRNA

SC

+ + +

M AN U

TE D

E NG+Scr-siRNA

+ + +

+ + -

+ + -

Glucose(5.5mmol/L) Glucose (20mmol/L) Scr-siRNA eNOS-siRNA Irisin(100ng/ml) Annexin V-FITC

D

*

HG+AMPK-siRNA+Irisin

Apoptotic Cells (%)

EP HG+Scr-siRNA+Irisin

AC C

Propidium Iodide

80 60

**

40

**

**

β-actin + + +

**

**

**

2.5

+ + +

+ + -

+ + -

Glucose(5.5mmol/L) Glucose (20mmol/L) Scr-siRNA eNOS-siRNA Irisin(100ng/ml)

2 1.5 1 0.5 0

20

Bcl-2

0

Bax

Glucose(5.5mmol/L)

+

-

-

-

Glucose (20mmol/L)

-

+

+

+

+

+

+

-

AMPK-siRNA

-

-

-

+

Irisin(100ng/ml)

-

-

+

+

Scr-siRNA Annexin V-FITC

**

**

2.5 2 1.5 1 0.5 0

Bcl-2 Bax

F

100

**

Bcl-2/Bax Ratio

HG+Scr-siRNA

*

**

Bcl-2/Bax Ratio

NG+Scr-siRNA

C

100 Apoptotic Cells (%)

Propidium Iodide

A

B

RI PT

Figure 4

β-actin Glucose(5.5mmol/L) Glucose (20mmol/L) Scr-siRNA AMPK-siRNA Irisin(100ng/ml)

+

-

-

-

+ -

+ + -

+ + +

+ + +

ACCEPTED MANUSCRIPT

Figure 5 B **

**

0.9 0.6 0.3

**

1 0.5

M AN U

P-Akt

P-AMPK

TE D

+ + +

EP

+ + +

Akt Glucose(5.5mmol/L) Glucose (20mmol/L) Scr-siRNA AMPK-siRNA Irisin(100ng/ml)

AC C

+ + -

+ + -

**

1.5

0

0 AMPK Glucose(5.5mmol/L) Glucose (20mmol/L) Scr-siRNA AMPK-siRNA Irisin(100ng/ml)

2

**

RI PT

**

SC

Relative P-Akt/Akt

Relative P-AMPK/AMPK

1.5 1.2

C

2.5

+ + -

+ + -

+ + +

Relative P-eNOS/eNOS

A

2.5

**

**

**

2 1.5 1 0.5 0

P-eNOS

+ + +

eNOS Glucose(5.5mmol/L) Glucose (20mmol/L) Scr-siRNA AMPK-siRNA Irisin(100ng/ml)

+ + -

+ + -

+ + +

+ + +

ACCEPTED MANUSCRIPT

Figure 6 A

Non-diabetic

B Diabetic + NS

Diabetic + irisin

50

**

**

Diabetic + NS

Diabetic + irisin

AC C

Non-diabetic

30 20 10 0

(n=3)

(n=3)

(n=3)

D Aortic Plaque Area Cross Sections(%)

C

EP

TE D

M AN U

SC

Aortic Plaque Area en Face Sections(%)

RI PT

40

50

**

**

40 30 20 10 0

(n=7)

(n=7)

(n=7)

ACCEPTED MANUSCRIPT

Figure 7 60 Diabetic + NS

Diabetic + irisin

**

CD90

45 30 15 (n=7 )

(n=7 )

IL-10

IL-6

6 1.2

4.8

3

*

2.4

*

30 15

VCAM-1

2 *

(n=7)

(n=7)

1.6

3 *

2.4

0.4

1.5

0.6

1.2

0.4

0.6

0

0

0

0

0

0

0.8

3

1.8 1.2

2.4

Diabetic + NS (n = 7 mice)

(n=7)

MCP-1

3.6

4.5

*

**

0

(n=7 )

ICAM-1

TNF-α

45

AC C

Relative mRNA Levels

1.6

EP

C

TE D

M AN U

0

*

RI PT

CD68 Positive Area /Plaque Area(%)

60 CD68

*

*

B

CD90 Positive Area /Plaque Area(%)

Non-diabetic

SC

A

1.2

1.8

0.8

1.2

*

Diabetic + irisin (n =7 mice)

ACCEPTED MANUSCRIPT

Highlights 1. Our data provide both in vivo and in vitro evidence supporting the protective

effects of irisin against diabetes-induced endothelial damage.

mice.

RI PT

2. Irisin treatment ameliorates atherosclerosis in apolipoprotein E-null diabetic

3. The endothelium-protective action of irisin was through activation of

AC C

EP

TE D

M AN U

SC

AMPK-PI3K-Akt-eNOS signaling pathway