A novel mechanism of diabetic vascular endothelial dysfunction: Hypoadiponectinemia-induced NLRP3 inflammasome activation

    A novel mechanism of diabetic vascular endothelial dysfunction: Hypoadiponectinemia-induced NLRP3 inflammasome activation Jinglong Zhang, Linying Xia, Fen Zhang, Di Zhu, Chao Xin, Helin Wang, Fuyang Zhang, Xian Guo, Yan Lee, Ling Zhang, Shan Wang, Xiong Guo, Chong Huang, Feng Gao, Yi Liu, Ling Tao PII: DOI: Reference:

S0925-4439(17)30053-4 doi:10.1016/j.bbadis.2017.02.012 BBADIS 64691

To appear in:

BBA - Molecular Basis of Disease

Received date: Revised date: Accepted date:

28 November 2016 5 February 2017 9 February 2017

Please cite this article as: Jinglong Zhang, Linying Xia, Fen Zhang, Di Zhu, Chao Xin, Helin Wang, Fuyang Zhang, Xian Guo, Yan Lee, Ling Zhang, Shan Wang, Xiong Guo, Chong Huang, Feng Gao, Yi Liu, Ling Tao, A novel mechanism of diabetic vascular endothelial dysfunction: Hypoadiponectinemia-induced NLRP3 inflammasome activation, BBA - Molecular Basis of Disease (2017), doi:10.1016/j.bbadis.2017.02.012

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 Title A novel mechanism of diabetic vascular endothelial hypoadiponectinemia-induced NLRP3 inflammasome activation

T

Authors

dysfunction:

SC R

IP

Jinglong Zhanga,1, Linying Xia a,1, Fen Zhangb,1, Di Zhuc, Chao Xind, Helin Wanga, Fuyang Zhanga, Xian Guoa, Yan Leea, Ling Zhanga, Shan Wanga, Xiong Guoa, Chong Huanga, Feng Gaoe, Yi Liua,e,**, Ling Taoa,**

D

MA

NU

Affiliations a Department of Cardiology, Xijing Hospital, The Fourth Military Medical University, Xi’an 710032, China; b Department of Hepatic Surgery, Xijing Hospital, The Fourth Military Medical University, Xi’an 710032, China; c Air Force General Hospital of People's Liberation Army, Beijing 100142, China; d Department of Cardiology, The Rocket Force General Hospital of People's Liberation Army, Beijing 100086, China; e Department of Physiology, Fourth Military Medical University, Xi’an, 710032, China.

AC

CE P

TE

Contact information ** Corresponding authors: Ling Tao,MD,PhD Department of Cardiology Xijing Hospital Fourth Military Medical University 15 Changle West Road Xi'an 710032, China. E-mail: [email protected]; Tel: 86-29-84771024; Fax: 86-29-84771024. Yi Liu, MD,PhD Department of Cardiology Xijing Hospital Fourth Military Medical University 15 Changle West Road Xi'an 710032, China. E-mail:[email protected] Tel:86-29-84775183 Fax: 86-29-84771024. 1

Jinglong Zhang, Linying Xia and Fen Zhang contributed equally to this work.

Word Count 3812 1

ACCEPTED MANUSCRIPT Abstract It has been well documented that hypoadiponectinemia is associated with impaired

T

endothelium-dependent vasodilation. However, the exact molecular mechanism which

IP

mediates this process has not been fully described. The current study aimed to

SC R

investigate the role of hypoadiponectinemia-induced NLRP3 inflammasome activation in diabetic vascular endothelial dysfunction and its molecular mechanism.

NU

Male adult adiponectin knockout mice and wild type mice were fed with a high fat diet to establish a type 2 diabetic mellitus model. In addition, human umbilical vein

MA

endothelial cells (HUVECs) were cultured and subjected to high glucose/high fat (HG/HF). The NLRP3 inflammasome activation was increased in type 2 diabetic mice

D

and treatment of diabetic aortic segments with MCC950, a potent selective inhibitor

TE

of NLRP3 inflammasome ex vivo improved endothelial-dependent vasorelaxation. NLRP3 inflammasome activation and vascular endothelial injury were significantly

CE P

increased in APN-KO mice compared with WT mice in diabetes and MCC950 decreased diabetic vascular endothelial dysfunction to comparable levels in APN-KO

AC

mice and WT mice. Adiponectin could decrease NLRP3 inflammasome activation and attenuate endothelial cell injury, which was abolished by NLRP3 inflammasome overexpression. Inhibition of peroxynitrite formation preferentially attenuated NLRP3 inflammasome activation in APN-KO diabetic mice. The current study demonstrated for the first time that hypoadiponectinemia-induced NLRP3 inflammasome activation was a novel mechanism of diabetic vascular endothelial dysfunction.

Key Words: adiponectin ▪ endothelial cells ▪ NLRP3 inflammasome ▪ oxidative/nitrative stress ▪ type 2 diabetic mellitus ▪

2

ACCEPTED MANUSCRIPT 1. Introduction Cardiovascular complications are the leading cause of death for patients with

T

type 2 diabetic mellitus, a disease affecting >94 million people in China[1.2].

IP

Inflammation in endothelial cells plays a critical role in the pathogenesis of vascular

SC R

disease in obesity-related type 2 diabetes[3.4]. It is therefore crucial to clarify the mechanism of inflammatory vascular injury in diabetic mellitus to search for novel

NU

therapeutic strategies.

Adiponectin is an abundant adipocyte-derived plasma protein involved in

MA

myocardial protection and insulin sensitivity[5.6]. The anti-inflammatory and vascular protective effects of adiponectin have been recognized, for example, it has been

D

reported that adiponectin decreases TNF-α-induced ICAM-1 expression and NF-κB

TE

activation in endothelial cells[7.8]. However, many fundamental questions remain unanswered. What’s more, numerous epidemiological studies have shown that plasma

CE P

adiponectin levels were negatively correlated with cardiovascular disease in obesity and diabetes[9.10]. Moreover, several clinical observations and basic studies have that

AC

demonstrated

endothelium-dependent

hypoadiponectinemia vasodilation[11-13].

is

associated

However,

the

with

impaired

exact

molecular

mechanism which mediates this process has not been fully described. Inflammasomes are large multiprotein complexes that consist of caspase-1, apoptosis-associated speck-like protein (ASC), and NLRP (nucleotide-binding oligomerization domain-like receptor with a pyrin domain). The ASC protein bridges the interaction between NLRP and caspase-1, making it essential for inflammasome activation and subsequent interleukin-1β (IL-1β) and IL-18 secretion[14-18]. To date, four additional inflammasome proteins have been identified: NLRP1, NLRP3, NLRC4, and AIM2[16-18]. The NLRP3 inflammasome is the most fully characterized

3

ACCEPTED MANUSCRIPT and has been associated with a wide range of diseases, including infections, auto-inflammatory and autoimmune diseases，and metabolic disorders[19-21]. The

T

NLRP3 inflammasome, activated by endotoxins, K+ channel openers, uric acid and

IP

reactive oxygen species (ROS)[22], plays a crucial role in insulin resistance and the

SC R

pathogenesis of diabetes[23]. Our laboratory has previously demonstrated that the NLRP3 inflammasome is activated in cardiac microvascular endothelial cells (CMECs), but not in cardiomyocytes, playing a critical role in the pathophysiology of

NU

MI/R injury[24]. However, the question of whether the NLRP3 inflammasome also

MA

plays an important role in diabetic vascular endothelial injury has not yet been addressed. It also remains unclear whether the NLRP3 inflammasome activation is

D

involved in hypoadiponectinemia associated diabetic vascular endothelial injury. Our

TE

previous data has indicated that adiponectin can reduce ROS production – an important activator of the NLRP3 inflammasome. Therefore, it is possible that the

CE P

activation of the NLRP3 inflammasome may be linked with loss of adiponectin in diabetes mellitus.

AC

The aims of the current study were: (1) to identify the role of NLRP3 inflammasome activation in diabetic vascular endothelial injury; (2) to determine the role of hypoadiponectinemia in NLRP3 inflammasome activation in diabetes; (3) to investigate the mechanism of hypoadiponectinemia-induced NLRP3 inflammasome activation.

4

ACCEPTED MANUSCRIPT 2. Materials and Methods 2.1. Materials

T

APN knockout mice were provided by professor Xinliang Ma in Thomas Jefferson

IP

University, in USA. They were global knockout. Phenotypic characteristics of male

SC R

APN knockout mice and WT control mice were same and had been previously described[25.26]. Adult male adiponectin knockout mice (APN-KO) and their wild

NU

type littermates (WT) were used in this study. All experiments were performed in accordance with the National Institutes of Health Guidelines on the Use of Laboratory

MA

Animals and were approved by the Fourth Military Medical University Committee on Animal Care. Dihydroethidium (DHE) was from Molecular Probes (Eugene, OR,

D

USA). ELISA kits for IL-1β， IL-18，ICAM-1 and VCAM-1 were purchased from

TE

Nanjing Institute of Jiancheng Bioengineering (Nanjing, China). Antibodies against NLRP3, Caspase-1, Akt and eNOS were from Cell Signaling Technology (Boston,

CE P

USA). MCC950 was pruchased from Med Chem Express. 2.2. Animal care

AC

Male APN knockout (APN-KO) mice and their wild type (WT) C57BL/6 littermates were maintained in a temperature-controlled barrier facility. Mice aged 8 to 9 weeks were used for these studies. Thirty-two APN-KO and thirty-two WT mice were divided into four groups respectively (n=8 for each group):

WT + Normal diet (WT),

APN-KO + Normal diet (APN-KO), WT + High fat diet (WT+HFD), APN-KO + High fat diet group (APN-KO+HFD), WT + High fat diet + EUK134 group (WT+HFD+EUK134),

APN-KO

+

High

fat

diet

+

EUK134

group

(APN-KO+HFD+EUK134), WT+High fat diet+1400W (WT+HFD+1400W), and APN-KO + High fat diet + 1400W (APN-KO+HFD+1400W). Besides WT and APN-KO groups, the type 2 diabetes model was established by feeding the rest of 5

ACCEPTED MANUSCRIPT mice with high-fat diet (D12492, Research Diets, Inc; NJ; USA) containing (kcal) 20% protein, 20% carbohydrate, and 60% fat for 8 weeks. Mice in WT and APN-KO

T

groups were fed with a chow diet (D12450B, Research Diets, Inc; NJ; USA)

IP

containing (kcal) 20% protein, 70% carbohydrate, and 10% fat. Mice were

SC R

anesthetized with 3% isoflurane and sacrificed with an overdose injection of pentobarbital, and serum was harvested and aortic segments were surgically dissected from the heart to the abdominal aortic bifurcation.

NU

2.3. Cell Culture

MA

Experiments on Human umbilical vein endothelial cells (HUVECs) were carried out on the same batch at passages 3-7. After serum starvation for 3 hours, cells were

D

pretreated with globular adiponectin (gAd; 2 μg/mL) for two hours before exposure to

TE

high glucose/high fat (HG/HF) medium, which contained glucose (25mmol/L) and free fatty acid (FFA) palmitate (500μmol/L) for 48 hours. Controls were incubated

CE P

with 5mmol/L glucose and equal concentrations of palmitate-free BSA. Cells were collected after treatment and activation of the NLRP3 inflammasome, endothelial

AC

diastolic function, intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule 1 (VCAM-1) expression were determined as described in detail below.

2.4. Evaluation of vasorelaxation Aortas were placed in cold physiological saline solution (PSS) and aortic segments were cut into 4 rings of 1 mm in length. These aortic rings were inatalled on a multi wire myograph system, which linked to transducers to recorded data. The aortic rings were then stretched to initial length, stable for 60 minutes. The rings were treated with 1μmol/L phenylephrine (PE) before acetylcholine (ACh) and nitroprusside (SNP) were added to test the diastolic function. Diastolic extent of each concentrations were

6

ACCEPTED MANUSCRIPT measured and results were expressed as the percentage of exposure to PE. 2.5. Small Interfering RNA and Plasmid Transfection

T

Small interfering RNA (siRNA) duplexes against NLRP3 were purchased from Santa

IP

Cruz Biotechnology (Santa Cruz, CA). Predesigned NLRP3-specific siRNA or control

SC R

scrambled siRNA (Santa Cruz, Thermo Scientific, respectively) were diluted in 5 % glucose and mixed with jet PEI (polyethyleneimine; Genesee Scientific, San Diego, CA). Each scrambled siRNA contained a sequence that is not predicted to target any

NU

known cellular mRNA. HUVECs (80% confluent) were transfected with

MA

Lipofectamine 2000 Reagent (final siRNA concentration: 100 nmol/L). The NLRP3 overexpression plasmid was purchased from Hanbio Biotechnology ( Shang Hai,

D

CHINA) , and was transfected into HUVECs using LipoFiter.

TE

2.6. Inflammatory cytokine detection by ELISA The levels of inflammatory cytokines (IL-1β, IL-18，sICAM-1 and sVCAM-1) in

CE P

mouse serum and cell culture supernatant were measured using a commercially enzyme-linked immunosorbent assay (ELISA) kit (Elabscience Biotechnology Co,

AC

Wuhan, China).

2.7. Apoptosis assay Apoptosis of HUVECs was assessed rely on caspase-3 activity by using a caspase-3 activity assay kit (Beyotime Institute of Biotechnology, Shanghai, China). Caspase-3 activity was showed as nmol pNA/h/mg protein. 2.8. Western blotting Aortic vessels were surgically dissected and harvested from the heart to the abdominal aortic bifurcation. Collection of arterial and cellular proteins to carry on Western blot analysis. Equal amounts of protein was separated by 10% SDS-PAGE and transferred to a nitrocellulose membranes. Blocking with 5% skim milk before the membrane

7

ACCEPTED MANUSCRIPT was incubated with primary antibodies with a 1:1000 dilution at 4°C overnight. Then the membrane incubated with sheep anti-rabbit or anti-mouse IgG HRP for 60 min at

T

room temperature and visualized using ChemiDocXRS.

IP

2.9. Statistical analysis

SC R

All data were analyzed by prism 5 Graphpad software. Comparison between the two groups was used the statistical method of t test and multiple comparisons between groups were used single factor or two factor analysis of variance statistical methods

NU

followed by Bonferroni post hoc analysis. P<0.05 was considered statistically

MA

significant.

3. Results

D

3.1. The NLRP3 inflammasome activation was increased in type 2 diabetic mice

TE

and treatment of diabetic aortic segments with MCC950, a potent selective inhibitor of NLRP3 ex vivo improved endothelial-dependent vasorelaxation

CE P

The NLRP3 inflammasome activates caspase-1, leading to the maturation of pro-inflammatory cytokines IL-1β and IL-18, thereby playing a crucial role in insulin

AC

resistance and pathogenesis of diabetes[20.22]. To investigate whether the NLRP3 inflammasome was activated during type 2 diabetes, a mouse model was established by feeding the mice a high-fat diet (60% fat) for 8 weeks (Fig.S1A, S1B, Table S1). In these models, NLRP3, caspase-1, IL-1β, and IL-18 levels were evaluated. NLRP3 expression (Fig.1A), caspase-1 activity (Fig.1B), IL-1β content (Fig.1C) and IL-18 content (Fig.1D) were all increased in type 2 diabetic vessels. These data indicated that NLRP3 inflammasome activation may be involved in diabetic vascular injury. To further illustrate the cause and effect, MCC950, a potent, selective, small-molecule inhibitor of NLRP3 inflammasome was used[27.28] to assess the direct effects of NLRP3 inflammasome activation to endothelium-dependent vasorelaxation. Diabetic

8

ACCEPTED MANUSCRIPT aortic segments were incubated with vehicle or MCC950 (15nM) before endothelium-dependent vasorelaxation was evaluated. Diabetic aortic segments

T

displayed a significantly impaired vascular relaxation compared with the control to

IP

ACh (Fig. 1E, maximum relaxation to ACh: 50.1±4.5% vs. 88.4±3.0%, P<0.01).

SC R

However, diabetic aortas exposed to MCC950 displayed improved vascular relaxation (Fig. 1E), maximum relaxation: 76.2±2.7% vs. 50.1±4.5%, P<0.05) compared with vehicle. Vasorelaxation to SNP was comparable among all groups (Fig. 1F). NO

NU

production in response to Ach stimulation was showed in supplementary Figure S2A.

MA

These data together indicated that NLRP3 inflammasome activation was involved in diabetic vascular endothelial injury.

D

3.2. NLRP3 inflammasome knockdown attenuated high glucose/high fat induced

TE

endothelial cell injury

CE P

The expression of NLRP3 was increased in diabetic vessels and NLRP3 inflammasome selective inhibitor, MCC950 could improve endothelial-dependent vasorelaxation, however, aortic segments consists of endothelia cells, smooth muscle

AC

cells and collagen/fibroblasts. To obtain more evidence to clear and definite it is endothelial cells that played an important role in diabetic vascular injury, HUVECs were exposed to HG/HF medium. Consistent with these in vivo experiments, NLRP3 expression (Fig.1G) was increased in HUVECs cultured with HG/HF for forty-eight hours. Caspase-1 activity (Fig.1H), IL-1β content (Fig.1I) and IL-18 content (Fig.1J) were all also increased after incubation. To further confirm the NLRP3 inflammasome play an important role in HG/HF induced endothelia cells injury, expression of NLRP3 inflammasome components were genetically inhibited by siRNA. This approach significantly reduced NLRP3 inflammasome expression by about 53.1% and 9

ACCEPTED MANUSCRIPT significantly inhibited HG/HF-induced NLRP3 inflammasome expression, as evidenced by a decrease in NLRP3 expression (Fig.2A), caspase-1 activity (Fig.2B),

T

IL-1β content (Fig.2C) and IL-18 content (Fig.2D).

IP

Endothelial nitric oxide synthase (eNOS) is responsible for physiologic vascular

SC R

tissue NO production, which plays a key role in endothelium-dependent vasodilatation[29.30]. The activity of eNOS is believed to reflect the function of

NU

endothelial cells. Intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule 1 (VCAM-1) are the main indicators of an inflammatory response,

MA

which can cause direct injury to vessels[31.32]. Our data indicated that siRNA targeting NLRP3 significantly attenuated endothelial cell dysfunction, endothelial cell

D

apoptosis and inflammatory responses induced by HG/HF, as evidenced by

TE

upregulated eNOS phosphorylation (Fig.2E), downregulated Caspase-3 activity

CE P

(Fig.2F), as well as downregulation of ICAM-1 (Fig.2G) and VCAM-1 (Fig.2H) expression. The production of NOx was significantly increased after 24 h of HG/HF incubation and this change was reversed by pre-incubation with NLRP3 specific

AC

siRNA (Fig.S2B). These results further indicated that NLRP3 inflammasome activation was partially responsible for endothelial cell injury, which played an indispensable role in diabetic vascular endohelial injury. 3.3. Adiponectin knockout mice showed significantly increased NLRP3 inflammasome activation and vascular endothelial injury compared with WT mice in diabetes Anti-inflammatory and vascular protective actions of adiponectin have been recognized[33]. However, the exact molecular mechanisms behind these roles were poorly understood. In the current study, we demonstrated that NLRP3 inflammasome activation was partially responsible for diabetic vascular injury. To further research 10

ACCEPTED MANUSCRIPT the causes of NLRP3 inflammasome activation, APN-KO and wild type mice were fed with HF diet to develop type 2 diabetes. Vascular tissue samples from APN-KO

T

mice had significantly increased NLRP3 inflammasome activation compared with WT

IP

mice in type 2 diabetes. Consistent with this, NLRP3 expression (Fig.3A), caspase-1

SC R

activity (Fig.3B), IL-1β content (Fig.3C) and IL-18 content (Fig.3D) were markedly increased in APN-KO mice. These results suggest that hypoadiponectinemia was responsible for NLRP3 inflammasome activation in diabetes. There was no significant

NU

difference in vascular relaxation to acetylcholine (ACh) between APN-KO and WT

MA

mice on ND(Fig.4A). However, compared with WT mice, APN-KO mice demonstrated more serious vascular injury from the high fat diet induced type 2

D

diabetic mellitus, as evidenced by poor endothelium-dependent vasorelaxation(Fig.4A,

TE

maximum relaxation: 40.4±3.2% vs. 50.1±4.5%, p<0.05), downregulated Akt phosphorylation (Fig.4C) and eNOS phosphorylation (Fig.4D), and upregulated

CE P

ICAM-1 (Fig.4E) and VCAM-1 (Fig.4F) content. Vasorelaxation to SNP was comparable among all groups (Fig. 4B). Our data indicated for the first time that

AC

hypoadiponectinemia was responsible for NLRP3 inflammasome activation in diabetes.

3.4. MCC950,a potent selective inhibitor of NLRP3 inflammasome decreased diabetic vascular endothelial dysfunction in adiponectin knockout mice We have observed in vivo that adiponectin knockout mice had significantly increased NLRP3 inflammasome activation and demonstrated more serious vascular endothelial injury compared with WT mice in type 2 diatetes, however, whether NLRP3 inflammasome activation results in vascular endothelial injury remained unknown. To obtain more evidence to test the hypothesis, diabetic aortic segments were incubated with vehicle or MCC950 (15nM), and endothelium-dependent vasorelaxation was

11

ACCEPTED MANUSCRIPT evaluated. Diabetic aortic segments from APN-KO mice exhibited more impaired vascular relaxation compared with the WT mice (Fig.4G, maximum relaxation:

T

40.4±3.2% vs. 50.1±4.5%, p<0.05). It is noteworthy that although endothelial

IP

dysfunction was more serious in diabetic vascular tissue from APN-KO mice than

SC R

tissue from WT mice, treatment with MCC950 improved diabetic vascular endothelial function to comparable levels in APN-KO mice and WT mice(Fig. 4G). Vasorelaxation to SNP was comparable among all groups (Fig.4H). These data

NU

indicated that NLRP3 inflammasome activation was an important cause of diabetic

MA

vascular endothelial dysfunction in APN-KO mice.

D

3.5. Adiponectin decreased NLRP3 inflammasome activation and attenuated

TE

endothelial cell injury

Our research indicated that hypoadiponectinemia was responsible for NLRP3

CE P

inflammasome activation resulting in diabetic vascular endothelial dysfunction. However, whether adiponectin had a direct effect on the activation of NLRP3

AC

inflammasome remains unclear. The role of adiponectin on HG/HF induced NLRP3 inflammasome activation was examined in HUVECs. Adiponectin blocked NLRP3 inflammasome activation induced by HG/HF as exhibited by decreased NLRP3 expression (Fig.5A), caspase-1 activity (Fig.5B), IL-1β content (Fig.5C) and IL-18 content (Fig.5D). Adiponectin also decreased HG/HF induced endothelial cell apoptosis, dysfunction and inflammatory response, as evidenced by increasing eNOS phosphorylation (Fig.5E), decreasing Caspase-3 activity (Fig.5F), as well as decreasing ICAM-1 (Fig.5G) and VCAM-1 (Fig.5H) content. Our results indicated that adiponectin could inhibit the activation of the NLRP3 inflammasome and exerted protective effects in endothelial cells.

12

ACCEPTED MANUSCRIPT 3.6. Vascular protective effect of adiponectin was blocked by NLRP3 inflammasome transgene

T

Our results indicated that adiponectin could inhibit the activation of the NLRP3

IP

inflammasome and exerted protective effects in endothelial cells. Therefore, it was

SC R

possible that adiponectin’s anti-inflammatory and vascular protective effects may be associated with inhibition of NLRP3 inflammasome activation. To address this possibility, HUVECs were transfected with a transgenic NLRP3 inflammasome

NU

overexpressing plasmid. The transgene increased NLRP3 inflammasome expression

MA

upon HG/HF induction, as evidenced by increased NLRP3 expression (Fig.6A) caspase-1 activity (Fig.6B), IL-1β content (Fig.6C) and IL-18 content (Fig.6D).

D

Besides, the transgene increased endothelial cell dysfunction, apoptosis and

TE

inflammatory response (Fig.6E, 6F, 6G, 6H). Adiponectin decreased HG/HF induced endothelial cell dysfunction, apoptosis and inflammatory response as evidenced by

CE P

increasing eNOS phosphorylation (Fig.6E), decreasing Caspase-3 activity (Fig.6F), as well as decreasing ICAM-1 (Fig.6G) and VCAM-1 (Fig.6H) content. However,

AC

adiponectin’s protective effects were abolished by the addition of the NLRP3 transgene (Fig.6E, 6F, 6G, 6H). Our results demonstrated, for the first time that adiponectin decreased endothelial cells injury by direct inhibiting activation of the NLRP3 inflammasome. 3.7. Inhibition of peroxynitrite formation preferentially decreased NLRP3 inflammasome activation in adiponectin knockout mice Our results indicated that hypoadiponectinemia was responsible for NLRP3 inflammasome activation in diabetes. However, the exact mechanism remained unclear. Reactive oxygen species have been shown to play a key role in NLRP3 inflammasome activation triggered in CMECs subjected to SI/R injury[34.35], and

13

ACCEPTED MANUSCRIPT adiponectin exerts its cardioprotective effects through anti-oxidative/nitrative stress pathways[36.37]. NADPH oxidase gp91phox has been reported as the main source of

T

vascular oxidative products and Inducible Nitric Oxide Synthase (iNOS) is the main

IP

source of pathological NO resulting in nitrotyrosine production[38.39]. Our results

SC R

showed that adiponectin suppressed oxidative/nitrative stresses (Fig.S3C, S3D) via inhibiting expression of gp91phox (Fig.S3A) and iNOS (Fig.S3B). In diabetic states, adiponectin levels were reduced and loss of its anti-oxidative/nitrative stress effect

NU

may contribute to activation of NLRP3 inflammasome. Male adult APN-KO mice or

MA

WT mice were fed with a high fat diet to induce type 2 diabetic mellitus, and then treated with either 1400W (2mg/kg), a selective iNOS inhibitor, or EUK134 (5

D

mg/kg), a peroxynitrite scavenger, 4 weeks after the type 2 diabetes model was

TE

established. We found that treatment with 1400W or EUK134 significantly decreased nitrotyrosine content (Fig 7A, 7B) in diabetic vascular tissue and also markedly

CE P

attenuated NLRP3 inflammasome activation, as evidenced by decreased NLRP3 expression (Fig.7C, 7D), caspase-1 activity (Fig.7E, 7F), IL-1β (Fig.7G, 7I) and IL-18

AC

content (Fig.7H, 7J). It is noteworthy that although nitrotyrosine content (Fig.7A,7B) and NLRP3 inflammasome activation (Fig. 7C, 7D, 7E, 7F) were markedly higher in diabetic vascular tissue from APN-KO mice than tissue from WT mice, treatment with 1400W or EUK134 reduced diabetic vascular nitrotyrosine formation (Fig. 7A, 7B) and NLRP3 inflammasome activation (Fig. 7C, 7D, 7E, 7F) to comparable levels in APN-KO mice and WT mice. These results demonstrated that blocking iNOS activity or scavenging peroxynitrite preferentially attenuated diabetic vascular NLRP3 inflammasome activation in APN-KO mice. This further suggested that increased oxidative and nitrative stress in APN-KO mice played a causative role in the activation of the NLRP3 inflammasome in diabetic vascular tissue.

14

ACCEPTED MANUSCRIPT 4. Discussion Several novel observations were made in this study. First, we have provided direct

T

evidence that NLRP3 inflammasome activation was responsible for diabetic vascular

IP

endothelial dysfunction. Cardiovascular complications comprise the main etiology for

SC R

mortality and a high percentage of morbidity in patients with diabetes mellitus[40-42]. Endothelial dysfunction is the initial factor of diabetic vascular injury, defined as

NU

unbalanced relaxation factor and contraction factor produced by endothelial cells and impaired nitric oxide-dependent diastolic function[43]. Endothelial dysfunction in

MA

diabetes has been associated with high glucose, high fat, AGEs, insulin resistance, hypertension, oxidative stress, and the inflammatory response[44.45]. However, the

D

exact molecular mechanism behind the inflammatory response leading to endothelial

TE

dysfunction is poorly understood. Inflammasomes are large multiprotein complexes that consist of caspase-1, apoptosis-associated speck-like protein (ASC), and NLRP.

CE P

The ASC protein bridges the interaction between NLRP and caspase-1, making it essential for inflammasome activation[22.28].

AC

The NLRP3 inflammasome plays an important role in obesity-associated chronic inflammation and insulin resistance in different organs, such as immune effector cells, pancreas, adipose tissue, liver, gut, kidney and skeletal muscle[46]. However, whether the NLRP3 inflammasome plays a critical role in diabetic vasculopathy has not yet been reported. Our results, for the first time demonstrated that NLRP3 inflammasome activation was responsible for diabetic vascular endothelial dysfunction. Our results indicated, for the first time that hypoadiponectinemia resulted in activation of the NLRP3 inflammasome in diabetic mellitus. In obesity and diabetic states, adiponectin levels are significantly decreased and absence of protective effects may contribute to the increased vulnerability of the cardiovascular system observed in

15

ACCEPTED MANUSCRIPT these conditions[47.48]. Numerous epidemiological studies have demonstrated that plasma adiponectin levels were negatively correlated with cardiovascular disease in

T

obese individuals and in patients with diabetes[49.50]. These patients also have

IP

increased concentrations of circulating cytokines, which exert detrimental effects on

SC R

endothelial cell function by increasing the production of reactive oxygen species (ROS). This triggers an inflammatory signaling cascade that results in increased expression of cell adhesion molecules, which enhance leukocyte-endothelium

NU

interactions and smooth muscle proliferation. In metabolic diseases such as diabetes

MA

and obesity, inflammation in endothelial cells plays a key role in the pathogenesis of vascular disease[51.52]. Modulation of vascular inflammation via downregulation of

D

TNF-induced endothelial cell adhesion molecule expression is a well-defined vascular

TE

protective effect of adiponectin. However, the molecular mechanism that links

understood.

CE P

hypoadiponectinemia with endothelial inflammation and dysfunction were poorly

Our current study found that adiponectin knockout mice showed significantly

AC

increased NLRP3 inflammasome activation and vascular endothelial injury compared with WT mice in diabetes. These results, for the first time indicated that levels of circulating adiponectin are inversely correlated with NLRP3 inflammasome activation. A variety of pathogenic factors,such as hypoadiponectinemia caused NLRP3 inflammasome activation, leading to diabetic vascular endothelial dysfunction. In addition, experiments ex vivo showed that adiponectin decreased NLRP3 inflammasome activation and attenuated endothelial cell injury induced by HG/HF. As research continues, the biological functions of adiponectin have been extensively investigated.

Four

major

functions

including

metabolism

regulation,

anti-inflammatory, vasculoprotective, and cardioprotective effects have been

16

ACCEPTED MANUSCRIPT identified[53]. Our current study demonstrated a novel mechanism behind adiponectin anti-inflammatory and vascular protective effects.

T

Finally, these results demonstrate that blocking iNOS activity or scavenging

IP

peroxynitrite preferentially attenuated diabetic vascular NLRP3 inflammasome

SC R

activation in APN-KO mice. This suggested that increased oxidative and nitrative stress in APN-KO mice play a causative role in the activation of the NLRP3 inflammasome in diabetic vascular tissues, making it clear that the mechanism of

NU

hypoadiponectinemia-induced NLRP3 inflammasome activation was oxidative and

MA

nitrative stress. Both the selective iNOS inhibitor 1400W and the peroxynitrite scavenger EUK134 significantly reduced nitrotyrosine content in diabetic vascular

D

tissue and markedly attenuated NLRP3 inflammasome activation. It is noteworthy

TE

that although nitrotyrosine content and NLRP3 inflammasome activation were markedly increased in diabetic vascular tissue from APN-KO mice as compared to

CE P

WT mice, treatment with 1400W or EUK134 reduced diabetic vascular nitrotyrosine formation and NLRP3 inflammasome activation to comparable levels in both

AC

genotypes.

In summary, our results demonstrated that hypoadiponectinemia-induced NLRP3 inflammasome activation was responsible for diabetic vascular endothelial dysfunction and the exact mechanism was oxidative and nitrative stress. These experimental results not only deepen our understanding of the biological functions of adiponectin, but

also

suggest

therapeutic

avenues

for

inhibiting NLRP3

inflammasome activation. This may provide novel vascular protective and anti-inflammatory effects in patients with diabetic mellitus.

17

ACCEPTED MANUSCRIPT 5. Limitations There are some limitations in our research. We demonstrated that

T

hypoadiponectinemia induced NLRP3 inflammasome activation was responsible for

IP

diabetic vascular endothelial dysfunction and the exact mechanism was oxidative and

SC R

nitrative stress and adiponectin suppressed oxidative/nitrative stresses via inhibiting expression of gp91phox and iNOS, however, the direct evidence of adiponectin

NU

regulated gp91phox and iNOS was still undifined. Further studies should be performed to investigate the transcription factor conducting this process. We only

MA

examined the NLRP3 inflammasome activation in the early state of diabetes, did not show NLRP3 inflammasome activation in the late stage of diabetes.We thought that

D

NLRP3 inflammasome activation was increased as the progression of diabetes, further

TE

research should be performed to expain this question. In this research, we established T2DM model by feeding the mice with a high-fat diet (60% fat) for 8 weeks without

CE P

combining with STZ injection. Although this diet composition can probably not be

AC

achieved in real human life, it is the best way existing to induced T2DM in mice.

Conflict of interest None declared.

Acknowledgements This work was supported by Program for National Science Fund for Distinguished Young Scholars of China (Grant No.81225001), National Key Basic Research Program of China (973 Program, Grant No.2013CB531204 ), Key Science and Technology Innovation Team in Shaanxi Province (Grant No.2014KCT-19), Program for Changjiang Scholars and Innovative Research Team in University (Grant No.PCSIRT-14R08), National Science Funds of China(Grants No.81170186, 81470478 and 81400201), Major Science and Technology Project of China 18

ACCEPTED MANUSCRIPT “Significant New Drug Development” (Grant No.2012ZX09J12108-06B), and the fourth military medical university's young talent project(The first level).

References: and Alzheimer's disease in T2DM patients, ENDOCRINE, 53(2016)

IP

help prevent dementia

T

[1]S. Bertram, K. Brixius, C. Brinkmann, Exercise for the diabetic brain: how physical training may 350-363.

SC R

[2]J.E. Villena, Diabetes Mellitus in Peru, Ann Glob Health, 81(2015) 765-775.

[3]C. Lu, X. Wang, T. Ha, Y. Hu, L. Liu, X. Zhang, H. Yu, J. Miao, R. Kao, J. Kalbfleisch, D. Williams, C. Li, Attenuation of cardiac dysfunction and remodeling of myocardial infarction by microRNA-130a are mediated by suppression of PTEN and activation of PI3K dependent

NU

signaling, J. MOL CELL CARDIOL, 89(2015) 87-97.

[4]D.M. Hasan, R.M. Starke, H. Gu, K. Wilson, Y. Chu, N. Chalouhi, D.D. Heistad, F.M. Faraci, C.D. Sigmund, Smooth Muscle Peroxisome Proliferator-Activated Receptor gamma Plays a Critical

MA

Role in Formation and Rupture of Cerebral Aneurysms in Mice In Vivo, HYPERTENSION, 66(2015) 211-220.

[5]M.C. Passos, M.C. Goncalves, Regulation of insulin sensitivity by adiponectin and its receptors in response to physical exercise, HORM METAB RES, 46(2014) 603-608.

D

[6]N. Halberg, T.D. Schraw, Z.V. Wang, J.Y. Kim, J. Yi, M.P. Hamilton, K. Luby-Phelps, P.E. 1961-1970.

TE

Scherer, Systemic fate of the adipocyte-derived factor adiponectin, DIABETES, 58(2009) [7]M. Sawicka, J. Janowska, J. Chudek, Potential beneficial effect of some adipokines positively

CE P

correlated with the adipose tissue content on the cardiovascular system, INT J. CARDIOL, 222(2016) 581-589.

[8]R. Niinaga, H. Yamamoto, M. Yoshii, H. Uekita, N. Yamane, I. Kochi, A. Matsumoto, T. Matsuoka, S. Kihara, Marked elevation of serum M2BP-adiponectin complex in men with coronary artery disease, ATHEROSCLEROSIS, 253(2016) 70-74.

AC

[9]A. Baran, I. Flisiak, J. Jaroszewicz, M. Swiderska, Effect of psoriasis activity on serum adiponectin and leptin levels, Postepy Dermatol Alergol, 32(2015) 101-106. [10]B. Gray, F. Steyn, P.S. Davies, L. Vitetta, Liver function parameters, cholesterol, and phospholipid alpha-linoleic acid are

associated with adipokine levels in overweight and obese adults, NUTR

RES, 34(2014) 375-382. [11]K.C. Tan, A. Xu, W.S. Chow, M.C. Lam, V.H. Ai, S.C. Tam, K.S. Lam, Hypoadiponectinemia is associated with impaired endothelium-dependent vasodilation, J Clin Endocrinol Metab, 89(2004) 765-769. [12]N. Ouchi, M. Ohishi, S. Kihara, T. Funahashi, T. Nakamura, H. Nagaretani, M. Kumada, K. Ohashi, Y. Okamoto, H. Nishizawa, K. Kishida, N. Maeda, A. Nagasawa, H. Kobayashi, H. Hiraoka, N. Komai, M. Kaibe, H. Rakugi, T. Ogihara, Y. Matsuzawa, Association of hypoadiponectinemia with impaired vasoreactivity, HYPERTENSION, 42(2003) 231-234. [13]H. Li, Y. Xiao, H. Liu, X.Y. Chen, X.Y. Li, W.L. Tang, S.P. Liu, A.M. Xu, Z.G. Zhou, Hypoadiponectinemia predicts impaired endothelium-independent vasodilation in newly diagnosed type 2 diabetic patients: an 8-year prospective study, Chin Med J (Engl), 124(2011) 3607-3612. [14]L. Sborgi, F. Ravotti, V.P. Dandey, M.S. Dick, A. Mazur, S. Reckel, M. Chami, S. Scherer, M.

19

ACCEPTED MANUSCRIPT Huber, A. Bockmann, E.H. Egelman, H. Stahlberg, P. Broz, B.H. Meier, S. Hiller, Structure and assembly of the mouse ASC inflammasome by combined NMR spectroscopy and cryo-electron microscopy, Proc Natl Acad Sci U S A, 112(2015) 13237-13242. [15]R. Liu, A.D. Truax, L. Chen, P. Hu, Z. Li, J. Chen, C. Song, L. Chen, J.P. Ting, Expression profile

T

of innate immune receptors, NLRs and AIM2, in human colorectal cancer: correlation with cancer stages and inflammasome components, ONCOTARGET, 6(2015) 33456-33469.

IP

[16]R. Liu, A.D. Truax, L. Chen, P. Hu, Z. Li, J. Chen, C. Song, L. Chen, J.P. Ting, Expression profile of innate immune receptors, NLRs and AIM2, in human colorectal cancer: correlation with cancer

SC R

stages and inflammasome components, ONCOTARGET, 6(2015) 33456-33469. [17]Y. He, M.Y. Zeng, D. Yang, B. Motro, G. Nunez, NEK7 is an essential mediator of NLRP3 activation downstream of potassium efflux, NATURE, 530(2016) 354-357. [18]J.S. Moon, K. Nakahira, K.P. Chung, G.M. DeNicola, M.J. Koo, M.A. Pabon, K.T. Rooney, J.H.

NU

Yoon, S.W. Ryter, H. Stout-Delgado, A.M. Choi, NOX4-dependent fatty acid oxidation promotes NLRP3 inflammasome activation in macrophages, NAT MED, 22(2016) 1002-1012. [19]G. Arbore, E.E. West, R. Spolski, A.A. Robertson, A. Klos, C. Rheinheimer, P. Dutow, T.M.

MA

Woodruff, Z.X. Yu, L.A. O'Neill, R.C. Coll, A. Sher, W.J. Leonard, J. Kohl, P. Monk, M.A. Cooper, M. Arno, B. Afzali, H.J. Lachmann, A.P. Cope, K.D. Mayer-Barber, C. Kemper, T helper 1 immunity requires complement-driven NLRP3 inflammasome activity in CD4(+) T cells, SCIENCE, 352(2016) d1210.

D

[20]J.S. Moon, K. Nakahira, K.P. Chung, G.M. DeNicola, M.J. Koo, M.A. Pabon, K.T. Rooney, J.H.

TE

Yoon, S.W. Ryter, H. Stout-Delgado, A.M. Choi, NOX4-dependent fatty acid oxidation promotes NLRP3 inflammasome activation in macrophages, NAT MED, 22(2016) 1002-1012. [21]J. Garaude, R. Acin-Perez, S. Martinez-Cano, M. Enamorado, M. Ugolini, E. Nistal-Villan, S.

CE P

Hervas-Stubbs, P. Pelegrin, L.E. Sander, J.A. Enriquez, D. Sancho, Mitochondrial respiratory-chain adaptations in macrophages contribute to antibacterial host defense, NAT IMMUNOL, 17(2016) 1037-1045. [22]T. Gicquel, S. Robert, P. Loyer, T. Victoni, A. Bodin, C. Ribault, F. Gleonnec, I. Couillin, E.

AC

Boichot, V. Lagente, IL-1beta production is dependent on the activation of purinergic receptors and NLRP3 pathway in human macrophages, FASEB J., 29(2015) 4162-4173. [23]B. Luo, B. Li, W. Wang, X. Liu, Y. Xia, C. Zhang, M. Zhang, Y. Zhang, F. An, NLRP3 gene silencing ameliorates diabetic cardiomyopathy in a type 2 diabetes rat model, PLOS ONE, 9(2014) e104771. [24]Y. Liu, K. Lian, L. Zhang, R. Wang, F. Yi, C. Gao, C. Xin, D. Zhu, Y. Li, W. Yan, L. Xiong, E. Gao, H. Wang, L. Tao, TXNIP mediates NLRP3 inflammasome activation in cardiac microvascular endothelial cells as a novel mechanism in myocardial ischemia/reperfusion injury, BASIC RES CARDIOL, 109(2014) 415. [25]K. Ma, A. Cabrero, P.K. Saha, H. Kojima, L. Li, B.H. Chang, A. Paul, L. Chan, Increased beta -oxidation but no insulin resistance or glucose intolerance in mice lacking adiponectin, J. BIOL CHEM, 277(2002) 34658-34661. [26]L. Tao, E. Gao, X. Jiao, Y. Yuan, S. Li, T.A. Christopher, B.L. Lopez, W. Koch, L. Chan, B.J. Goldstein, X.L. Ma, Adiponectin cardioprotection after myocardial ischemia/reperfusion involves the reduction of oxidative/nitrative stress, CIRCULATION, 115(2007) 1408-1416. [27]R.C. Coll, A.A. Robertson, J.J. Chae, S.C. Higgins, R. Munoz-Planillo, M.C. Inserra, I. Vetter, L.S. Dungan, B.G. Monks, A. Stutz, D.E. Croker, M.S. Butler, M. Haneklaus, C.E. Sutton, G. Nunez,

20

ACCEPTED MANUSCRIPT E. Latz, D.L. Kastner, K.H. Mills, S.L. Masters, K. Schroder, M.A. Cooper, L.A. O'Neill, A small-molecule inhibitor of the NLRP3 inflammasome for the treatment of inflammatory diseases, NAT MED, 21(2015) 248-255. [28]A. Nalbandian, A.A. Khan, R. Srivastava, K.J. Llewellyn, B. Tan, N. Shukr, Y. Fazli, V.E. Valosin-Containing Protein Myopathy, INFLAMMATION, (2016).

T

Kimonis, L. BenMohamed, Activation of the NLRP3 Inflammasome Is Associated with

IP

[29]Q. Yu, Y. Zhang, C.B. Xu, Apolipoprotein B, the villain in the drama? EUR J. PHARMACOL, 748(2015) 166-169.

SC R

[30]E. Osto, C.M. Matter, A. Kouroedov, T. Malinski, M. Bachschmid, G.G. Camici, U. Kilic, T. Stallmach, J. Boren, S. Iliceto, T.F. Luscher, F. Cosentino, c-Jun N-terminal kinase 2 deficiency protects against hypercholesterolemia-induced endothelial dysfunction and oxidative stress, CIRCULATION, 118(2008) 2073-2080.

NU

[31]S.C. Harwani, J. Ratcliff, F.S. Sutterwala, Z.K. Ballas, D.K. Meyerholz, M.W. Chapleau, F. Abboud, Nicotine Mediates CD161a+ Renal Macrophage Infiltration and Premature Hypertension in the Spontaneously Hypertensive Rat, CIRC RES, (2016).

MA

[32]S.C. Harwani, J. Ratcliff, F.S. Sutterwala, Z.K. Ballas, D.K. Meyerholz, M.W. Chapleau, F. Abboud, Nicotine Mediates CD161a+ Renal Macrophage Infiltration and Premature Hypertension in the Spontaneously Hypertensive Rat, CIRC RES, (2016). [33]Y. Wang, X. Wang, W.B. Lau, Y. Yuan, D. Booth, J.J. Li, R. Scalia, K. Preston, E. Gao, W. Koch,

D

X.L. Ma, Adiponectin inhibits tumor necrosis factor-alpha-induced vascular inflammatory

TE

response via caveolin-mediated ceramidase recruitment and activation, CIRC RES, 114(2014) 792-805.

[34]S. Yue, J. Zhu, M. Zhang, C. Li, X. Zhou, M. Zhou, M. Ke, R.W. Busuttil, Q.L. Ying, J.W.

CE P

Kupiec-Weglinski, Q. Xia, B. Ke, The myeloid heat shock transcription factor 1/beta-catenin axis regulates NLR family, pyrin domain-containing 3 inflammasome activation in mouse liver ischemia/reperfusion injury, HEPATOLOGY, (2016). [35]M. Kalbitz, F. Fattahi, J.J. Grailer, L. Jajou, E.A. Malan, F.S. Zetoune, M. Huber-Lang, M.W.

AC

Russell, P.A. Ward, Complement-induced activation of the cardiac NLRP3 inflammasome in sepsis, FASEB J., (2016). [36]R. Yao, Y. Zhou, Y. He, Y. Jiang, P. Liu, L. Ye, Z. Zheng, W.B. Lau, Y. Cao, Z. Zeng, Adiponectin protects against paraquat-induced lung injury by attenuating oxidative/nitrative stress, EXP THER MED, 9(2015) 131-136. [37]Y. Wang, W.B. Lau, E. Gao, L. Tao, Y. Yuan, R. Li, X. Wang, W.J. Koch, X.L. Ma, Cardiomyocyte-derived adiponectin is biologically active in protecting against myocardial ischemia-reperfusion injury, Am J Physiol Endocrinol Metab, 298(2010) E663-E670. [38]W. Sangartit, P. Pakdeechote, V. Kukongviriyapan, W. Donpunha, S. Shibahara, U. Kukongviriyapan, Tetrahydrocurcumin in combination with deferiprone attenuates hypertension, vascular dysfunction, baroreflex dysfunction, and oxidative stress in iron-overloaded mice, Vascul Pharmacol, (2016). [39]D. Zhu, H. Wang, J. Zhang, X. Zhang, C. Xin, F. Zhang, Y. Lee, L. Zhang, K. Lian, W. Yan, X. Ma, Y. Liu, L. Tao, Irisin improves endothelial function in type 2 diabetes through reducing oxidative/nitrative stresses, J. MOL CELL CARDIOL, 87(2015) 138-147. [40]Z. Eroglu, E. Harman, E. Vardarli, M. Kayikcioglu, A.T. Vardarli, LDLR C1725T Gene Polymorphism Frequency in Type 2 Diabetes Mellitus Patients With Dyslipidemia, J Clin Med

21

ACCEPTED MANUSCRIPT Res, 8(2016) 793-796. [41]M.S. Arslan, E. Tutal, M. Sahin, M. Karakose, B. Ucan, G. Ozturk, E. Cakal, G.Z. Biyikli, M. Ozbek, T. Delibasi, Effect of lifestyle interventions with or without metformin therapy on serum levels of osteoprotegerin and receptor activator of nuclear factor kappa B ligand in patients with

T

prediabetes, ENDOCRINE, (2016). [42]J. Zhu, P. Kahn, J. Knudsen, S.N. Mehta, R.A. Gabbay, Predictive Model for Estimating the Cost

IP

of Incident Diabetes Complications, Diabetes Technol Ther, 18(2016) 625-634.

[43]W.H. Younis, N.H. Al-Rawi, M.A. Mohamed, N.Y. Yaseen, Molecular events on tooth socket

SC R

healing in diabetic rabbits, Br J Oral Maxillofac Surg, 51(2013) 932-936.

[44]Y. Li, Q. Zhou, C. Pei, B. Liu, M. Li, L. Fang, Y. Sun, Y. Li, S. Meng, Hyperglycemia and Advanced Glycation End Products Regulate miR-126 Expression in Cells, J. VASC RES, 53(2016) 94-104.

Endothelial Progenitor

NU

[45]P.M. Seferovic, W.J. Paulus, Clinical diabetic cardiomyopathy: a two-faced disease with restrictive and dilated phenotypes, EUR HEART J., 36(2015) 1718-1727, 1727a. [46]C.J. Tack, R. Stienstra, L.A. Joosten, M.G. Netea, Inflammation links excess fat to insulin

MA

resistance: the role of the interleukin-1 family, IMMUNOL REV, 249(2012) 239-252. [47]A. Schneider, L.M. Neas, D.W. Graff, M.C. Herbst, W.E. Cascio, M.T. Schmitt, J.B. Buse, A. Peters, R.B. Devlin, Association of cardiac and vascular changes with ambient PM2.5 in diabetic individuals, PART FIBRE TOXICOL, 7(2010) 14. cardiometabolic risk

environments: pathophysiological aspects, Horm Mol Biol Clin Investig,

17(2014) 79-87.

TE

D

[48]G. Berg, L. Schreier, V. Miksztowicz, Circulating and adipose tissue matrix metalloproteinases in

[49]R. Li, L.Z. Chen, S.P. Zhao, X.S. Huang, Inflammation Activation Contributes to Adipokine

CE P

Imbalance in Patients with Acute Coronary Syndrome, PLOS ONE, 11(2016) e151916. [50]G. Wang, S. Gao, N. Su, J. Xu, D. Fu, Plasma Adiponectin Levels Inversely Correlate to Clinical Parameters in Type 2 Diabetes Mellitus Patients with Macrovascular Diseases, ANN CLIN LAB SCI, 45(2015) 287-291.

AC

[51]Z. Kobalava, SP 04-1 THE ROLE OF NATRIURETIC PEPTIDES IN THE PATHOGENESIS OF CARDIOVASCULAR DISEASES, J. HYPERTENS, 34 Suppl 1(2016) e377. [52]R.A. Pena-Silva, N. Chalouhi, L. Wegman-Points, M. Ali, I. Mitchell, G.L. Pierce, Y. Chu, Z.K. Ballas, D. Heistad, D. Hasan, Novel role for endogenous hepatocyte growth factor in the pathogenesis of intracranial aneurysms, HYPERTENSION, 65(2015) 587-593. [53]K. Gasbarrino, J. Gorgui, B. Nauche, R. Cote, S.S. Daskalopoulou, Circulating adiponectin and carotid intima-media thickness: A systematic review and meta-analysis, METABOLISM, 65(2016) 968-986.

22

ACCEPTED MANUSCRIPT Figure Legends

MA

NU

SC R

IP

T

Fig.1. Activation of the NLRP3 inflammasome in type 2 diabetic mice vessels and HG/HF incubated endothelial cells, and the effect of MCC950, a potent selective inhibitor of NLRP3 inflammasome on the endothelial function in type 2 diabetic mice. Wild type mice were fed with a high fat diet for 8 weeks to establish a model of type 2 diabetic mellitus. Activation of the NLRP3 inflammasome in type 2 diabetic mice vessels was determined by NLRP3 expression (A), caspase-1 activity (B), IL-1β content (C) and IL-18 content (D). The effects of MCC950 on endothelium-dependent vasorelaxation (E) and endothelium-independent vasorelaxation (F) in type 2 diabetic mice. HUVECs were treated with 25mmol/L glucose and 500μmol/L palmitate for forty-eight hours, and the effects of HG/HF induced NLRP3 inflammasome activation was demonstrated by NLRP3 expression (G), caspase-1 activity (H), IL-1β content (I) and IL-18 content (J). N=8 per group. The results are reported as the mean±SEM, *P<0.05, **P<0.01 vs WT+ND or Con group, #P<0.05, ##P<0.01 vs WT+HFD group. WT+ND: wild type+normal diet; WT+HFD: wild type+high fat diet; Con: control; HG/HF: high glucose/high fat.

AC

CE P

TE

D

Fig.2. The effect of NLRP3 inflammasome knockdown on HG/HF induced endothelial cell injury. HUVECs were transfected with small interfering RNA (siRNA) against NLRP3 to knockdown NLRP3 inflammasome activation as determined by NLRP3 expression (A), caspase-1 activity (B), IL-1β content (C) and IL-18 content (D). Twenty-four hours after transfection, cells were treated with HG/HF for forty-eight hours. Effects of NLRP3 inflammasome knockdown on endothelial cell injury induced by HG/HF was determined by eNOS phosphorylation(E), Caspase-3 activity (F), and expression of soluble intercellular adhesion molecule-1 (sICAM-1) (G) and soluble vascular cell adhesion molecule 1 (sVCAM-1) (H). Results are reported as the mean±SEM of five to seven repeated experiments. *P<0.05, **P<0.01 vs con group, #P<0.05 vs HG/HF group. Con: control; HG/HF: high glucose/high fat. Fig.3. Activation of the NLRP3 inflammasome in type 2 diabetic mice vessels in adiponectin knockout and wild type mice. Male adiponectin knockout (APN-KO) mice and their wild type (WT) littermates (C57BL/6) were fed with a high-fat diet to establish type 2 diabetes. NLRP3 inflammasome activation in vessels from adiponectin knockout and wild type mice was determined by NLRP3 expression (A), caspase-1 activity (B), IL-1β (C) and IL-18 (D) content. N=8 per group. Results are reported as the mean±SEM. *P<0.05, **P<0.01 vs WT+ND group, #P<0.05 vs WT+HFD group. WT+ND: wild type+normal diet; WT+HFD: wild type+high fat diet; APN-KO+ND: adiponectin knockout+normal diet; APN-KO+HFD: adiponectin knockout+high fat diet.

23

ACCEPTED MANUSCRIPT

MA

NU

SC R

IP

T

Fig.4. Endothelial functionin and vascular injury in type 2 diabetic mice and the effect of MCC950, a potent selective inhibitor of NLRP3 inflammasome on the endothelial function in aortic segments dissected from diabetic mice in adiponectin knockout and wild type mice. Male adiponectin knockout (APN-KO) mice and their wild type (WT) littermates (C57BL/6) were fed with a high-fat diet to establish type 2 diabetes. Endothelial functionin and vascular injury were determined by endothelium-dependent vasorelaxation (A), endothelium-independent vasorelaxation (B), and Akt phosphorylation (C), eNOS phosphorylation (D), expression of soluble intercellular adhesion molecule-1 (sICAM-1) (E) and soluble vascular cell adhesion molecule 1 (sVCAM-1) (F). The effects of MCC950 on endothelium-dependent vasorelaxation (G) and endothelium-independent vasorelaxation (H) in type 2 diabetic mice in adiponectin knockout and wild type mice. N=8 per group. Results are reported as the mean±SEM. *P<0.05, **P<0.01 vs WT+ND mice, #P<0.05, ##P<0.01 vs WT+HFD mice. WT+ND: wild type+normal diet; WT+HFD: wild type+high fat diet; APN-KO+ND: adiponectin knockout+normal diet; APN-KO+HFD: adiponectin knockout+high fat diet.

AC

CE P

TE

D

Fig.5. Effects of gAd on HG/HF induced NLRP3 inflammasome activation and endothelial cell injury. HUVECs were pretreated with globular adiponectin domain (gAd) two hours before treatment with HG/HF. The effect of gAd on HG/HF induced endothelial cell NLRP3 inflammasome activation was determined by NLRP3 expression (A), caspase-1 activity (B), IL-1β (C) and IL-18 (D) content. The effect of gAd on endothelial cell injury induced by HG/HF was determined by eNOS phosphorylation (E), Caspase-3 activity (F), and soluble intercellular adhesion molecule-1 (sICAM-1) (G) and soluble vascular cell adhesion molecule 1 (sVCAM-1) (H) expression. Results are reported as the mean±SEM of five to seven repeated experiments. *P<0.05, **P<0.01 vs con, #P<0.05 vs HG/HF group. gAd: globular adiponectin domain; con: control; HG/HF: high glucose/high fat. Fig.6. Effect of gAd or/and NLRP3 inflammasome transgene on endothelial cell injury induced by HG/HF. HUVECs were transfected with a transgenic overexpressing NLRP3 inflammasome plasmid. Twenty-four hours after transfection, cells were pretreated with gAd two hours before treatment with HG/HF for forty-eight hours. Effects of gAd or/and transgene on NLRP3 inflammasome activation were determined by NLRP3 expression (A), caspase-1 activity (B), IL-1β content (C) and IL-18 content (D). Effects of gAd or/and NLRP3 inflammasome transgene on endothelial cell injury induced by HG/HF were determined by eNOS phosphorylation (E), Caspase-3 activity (F), expression of soluble intercellular adhesion molecule-1 (sICAM-1) (G) and soluble vascular cell adhesion molecule 1 (sVCAM-1) (H). Results are reported as the mean±SEM of five to seven repeated experiments. *P<0.05 vs HG/HF group. gAd: globular adiponectin domain; Tg: NLRP3 inflammasome transgene; con: control; HG/HF: high glucose/high fat.

24

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Fig.7. Effects of 1400W and EUK134 on peroxynitrite formation and NLRP3 inflammasome activation in type 2 diabetic mice vessels. Male adult APN-KO mice and WT mice were fed with a high fat diet to induce type 2 diabetic mellitus, and were then treated with either the selective iNOS inhibitor 1400W (2mg/kg), or the peroxynitrite scavenger EUK134 (5 mg/kg), 4 weeks after the type 2 diabetes model was established. 1400W and EUK134 both significantly reduced nitrotyrosine content (A, B) and NLRP3 inflammasome activation as determined by decreased NLRP3 expression (C, D), caspase-1 activity (E, F), IL-1β (G, I) and IL-18 content (H, J) in type 2 diabetic mice vessels. N=8-10 per group. Results are reported as the mean±SEM. *P<0.05, **P<0.01 vs ND group, #P<0.05 vs HFD group. ND: normal diet; HFD: high fat diet; APN-KO: adiponectin knockout.

25

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 1

26

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 2

27

Fig. 3

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

28

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 4

29

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 5

30

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 6

31

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 7

32

AC

CE P

Graphical abstract

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

33

ACCEPTED MANUSCRIPT Highlights

T

NLRP3 inflammasome activation was responsible for diabetic vascular endothelial dysfunction

hypoadiponectinemia-induced

SC R

mediated

AC

CE P

TE

D

MA

NU

Oxidative/nitrative stress inflammasome activation

IP

Hypoadiponectinemia resulted in activation of the NLRP3 inflammasome in diabetic mellitus

34

NLRP3