Diosgenin regulates adipokine expression in perivascular adipose tissue and ameliorates endothelial dysfunction via regulation of AMPK

Accepted Manuscript Title: Diosgenin regulates adipokine expression in perivascular adipose tissue and ameliorates endothelial dysfunction via regulation of AMPK Author: Yan Chen Xiaoshan Xu Yuanyuan Zhang Kang Liu Fang Huang Baolin Liu Junping Kou PII: DOI: Reference:

S0960-0760(15)30021-2 http://dx.doi.org/doi:10.1016/j.jsbmb.2015.07.005 SBMB 4443

To appear in:

Journal of Steroid Biochemistry & Molecular Biology

Received date: Revised date: Accepted date:

20-5-2015 7-7-2015 12-7-2015

Please cite this article as: Y. Chen, X. Xu, Y. Zhang, K. Liu, F. Huang, B. Liu, J. Kou, Diosgenin regulates adipokine expression in perivascular adipose tissue and ameliorates endothelial dysfunction via regulation of AMPK, Journal of Steroid Biochemistry and Molecular Biology (2015), http://dx.doi.org/10.1016/j.jsbmb.2015.07.005 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.

Diosgenin regulates adipokine expression in perivascular adipose tissue and ameliorates endothelial dysfunction via regulation of AMPK Yan Chen1, Xiaoshan Xu2, Yuanyuan Zhang 1, Kang Liu2, Fang Huang2, Baolin Liu2,*,

National Key Laboratory of Natural Products, Jiangsu Key Laboratory of TCM Evaluation

cr

1

ip t

Junping Kou1,*

and Translational Research, Department of Complex Prescription of TCM, China

Department of Pharmacology of Chinese Material Medica, China Pharmaceutical University,

M

639 Longmian Road, Nanjing 211198, China

* Corresponding authors:

d

Prof. Baolin Liu

Department of Pharmacology of Chinese Material Medica, China Pharmaceutical University,

te

2

an

2

us

Pharmaceutical University, 639 Longmian Road, Nanjing 211198, China

Ac ce p

639 Longmian Road, Nanjing 211198, China Tel & Fax: +86-25-86185127

E-mail: [email protected] Dr. Junping Kou 1

Jiangsu Key Laboratory of TCM Evaluation and Translational Research, Department of

Complex Prescription of TCM, China Pharmaceutical University, 639 Longmian Road, Nanjing 211198, China Tel & Fax: +86-25-86185158 E-mail: [email protected] or [email protected] 1

Page 1 of 43

Abstract Perivascular adipose tissue (PVAT) has been recognized as an active contributor to vascular function due to its paracrine effects on cells contained within vascular wall. The

ip t

present study was designed to investigate the effect of diosgenin on adipokine expression in PVAT with emphasis on the regulation of endothelial function. Palmitic acid (PA) stimulation

cr

induced inflammation and dysregulation of adipokine expression in PVAT. Diosgenin treatment

us

inhibited IKKβ phosphorylation and downregulated mRNA expressions of proinflammatory cytokines/proteins including tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), monocyte

an

chemoattractant protein (MCP-1), and inducible nitric oxide synthase (iNOS), while reduced

M

gene expressions for adiponectin, PPARγ, and arginase 1 (Arg-1) were reversed by diosgenin treatment. Diosgenin enhanced AMPK phosphorylation under basal and inflammatory

d

conditions in PVAT, whereas knockdown of AMPK by SiRNA diminished its modulatory effect,

te

indicating that diosgenin inhibited inflammation in an AMPK-dependent manner. We prepared

Ac ce p

conditioned medium from PA-stimulated PVAT to induce endothelial dysfunction and found that pre-treatment of PVAT with diosgenin effectively restored the loss of ACh-induced vasodilation and increased eNOS phosphorylation in rat aorta. High-fat diet feeding in rats induced inflammation in PVAT and the impairment of endothelium-dependent vasodilation, whereas these alterations were prevented by oral administration of diosgenin at doses of 20 and 40 mg/kg. In conclusion, the obtained data showed that diosgenin ameliorated inflammation-associated

adipokine

dysregulation,

and

thereby

prevented

endothelial

dysfunction. Our findings would shed a novel insight into the potential mechanism by which diosgenin protected endothelial function against inflammatory insult.

2

Page 2 of 43

Keywords Diosgenin; Perivascular adipose tissue; AMPK; Inflammation; Endothelial dysfunction

ip t

 Diosgenin modulates adipokine expression in PVAT against inflammation.  The anti-inflammatory activity of diosgenin in PVAT depends on AMPK.

cr

Protecting PVAT function by diosgenin prevents endothelial dysfunction

us



an

1. Introduction

M

Perivascular adipose tissue (PVAT) is the adipose tissue surrounding the vasculature directly, thus allowing for easy access for a large number of mediators that modulates vascular

d

function. Adipocytes and stromal cells contained within PVAT produce adipocytokines such

te

as adiponectin and tumor necrosis factor- (TNF-α) that act in a paracrine or an endocrine

Ac ce p

fashion to control metabolic and endothelial functions [1]. It is known that PVAT can protect vessel function by attenuating vasoconstriction through different mechanisms, mainly including secreting adipocyte-derived relaxing factor (ADRF) and stimulating the generation of NO by endothelium [2-4]. Although the identity of ADRF is still unclear, adiponectin has been suggested to be one of the vasodilators [6]. In obesity, accumulation of lipids in PVAT stimulates the innate immunity defense leading to the recruitment of monocytes into the adipose tissue and subsequently induces dysregulation of adipocytokine production, leading to endothelial dysfunction [7]. AMP-activated protein kinase (AMPK) is a crucial regulator of energy metabolic homeostasis and emerging evidences demonstrate its anti-inflammatory action in vessel and adipose tissue [8, 9]. In a recent work, to be noticed, we also found that 3

Page 3 of 43

pharmacological activation of AMPK beneficially regulated adipocytokine expression in PVAT against inflammatory insult and ameliorated endothelial dysfunction. These findings demonstrate the role of AMPK activation in the regulation of PVAT/endothelial functions

ip t

[10]. Diosgenin, a well-known steroid sapogenin derived from plants, is pivotal in reducing the

cr

risks of acquiring cardiovascular disorders. Previous investigations have shown that important

us

biomarkers of the cardiovascular system related to endothelial dysfunction, inflammatory and oxidative stress are affected by diosgenin [11-13]. We previously found that diosgenin

an

ameliorated endothelial dysfunction by increasing insulin-mediated nitric oxide (NO)

M

production through IKKβ/IRS-1 pathway [14]. Moreover, studies support the potential of diosgenin in the management of chronic inflammation in adipose tissues involved in the

d

pathogenesis of obesity-related insulin resistance [15, 16]. Based on the potential therapeutic

te

value of diosgenin in obesity-associated cardiovascular disease, it is tempting for us to know

Ac ce p

the action of diosgenin in the regulation of PVAT function and the potential contribution to the improvement of vascular endothelial homeostasis. In the present paper, we investigated the effect of diosgenin on AMPK activation and adipokine expression in PVAT with emphasis on the regulation of endothelial function and found that diosgenin ameliorated endothelial dysfunction by inhibiting inflammation in PVAT. These findings would provide novel information regarding the potential mechanism of diosgenin in the management of

obesity-associated cardiovascular diseases.

2. Materials and Methods 2.1. Reagents 4

Page 4 of 43

Diosgenin (98% in purity) was purchased from Shanxi Huike Plant Co. Ltd (Xian, China) and dissolved in DMSO (0.1% v/v). Resveratrol (98% in purity) was obtained from Nanjing Zelang Medical Technology Co., Ltd. (Nanjing, China). AICA riboside (AICAR) and

ip t

compound-C were from Sigma (St. Louis, MO, USA). Palmitic acid (PA, Sinopharm Chemical Reagent Co., Ltd. Shanghai, China) was dissolved in ethanol as stock solution and

cr

then was diluted with 10% FFA-free BSA. All the primers were obtained from Sangon Biotec

us

(Shanghai, China). The following antibodies were obtained from Cell Signaling Technology, Inc. (Beverly, MA, USA): anti-AMPKα (#2532s), anti-phospho-AMPKα (T172) (#2531s),

an

anti-eNOS (49G3) (#9586) and anti-phospho-eNOS (Ser1177) (#9571). anti-phospho-IKK

M

(Y199) (BS4320), and anti-IKK (F182) (BS1407), Goat Anti-Rabbit IgG (H+L) HRP

d

(BS13278) and GAPDH (AP0063) were from Bioworld Technology (St. Paul, MN, USA).

te

2.2. Animals

Ac ce p

Sprague-Dawley rats (180-220 g) were supplied by the Laboratory Animal Center of Nanjing Qinglongshan. Animals were kept in a 12-h dark-light cycle and fed standard rodent chow ad libitum in accordance with the Provisions and General Recommendation of Chinese Experimental Animals Administration Legislation. For the high-fat diet feeding, Rats were given the high-fat diet (HFD, 60% kcal fat, 20% kcal carbohydrates, 20% kcal protein) for 8 weeks simultaneously with administration of diosgenin (20, 40 mg/kg) or resveratrol (20 mg/kg) by oral gavage every day. All procedures described were approved by the Animal Ethics Committee of School of Chinese Materia Medica, China Pharmaceutical University.

2.3. Serum analysis in HFD rats 5

Page 5 of 43

Overnight fasting-rats were anesthetized with diethyl ether and blood was collected from the orbital sinus. Contents of TNF-α and adiponectin in serum were determined with ELISA kits (Jiancheng Bioengineering Institute, Nanjing, China), following the manufacturer's

ip t

protocols.

cr

2.4. Perivascular adipose tissue-derived conditioned medium (CM) preparation

us

Rats (200-250 g) were anesthetized with diethylether and killed by cervical dislocation. Perivascular adipose tissues (PVAT) around the aorta were isolated and chopped into small

an

pieces. Equal amount of PVAT was pretreated with diosgenin (0.1, 1 and 10 µM), or

M

resveratrol (50 μM) individually in the presence or absence of AMPK inhibitor compound C (25 μM) and simultaneously stimulated with PA (100 μM) for 2 h. After the treatment, the

d

PVAT was washed with PBS twice to remove the treated agents, and cultured in fresh DMEM

te

for another 22 h. The medium was collected as conditioned medium (CM). For the collection

Ac ce p

of CM from HFD-fed rats, PVAT were isolated and directly incubated in DMEM for 24 h without any treatment.

2.5. Endothelium-dependent relaxation assessment Endothelium-dependent relaxation was assayed as we previously described [10]. In brief,

the prepared aortic ring was suspended in an organ bath containing 10 mL K-H solution maintained at 37°C, pH 7.4, and continuously aerated with 95% O2 and 5% CO2. After reaching the base tension, the contractive ability of the vessel was examined by contractive response to 60 mM KCl, while the functionality of vascular endothelium was confirmed by relaxation to 10 μM acetylcholine (ACh, Sigma, St. Louis, MO, USA)

The aortic ring 6

Page 6 of 43

(relaxation≥80%) was treated with CM for 0.5 h. After washing, the aortic ring was pre-contracted by 1 μM phenylephrine and the endothelium-dependent relaxation was induced by cumulative addition of ACh (0.001-10 μM). The relaxation was expressed as a percentage

ip t

of the phenylephrine-induced contraction. For the assessment in the aorta from HFD-fed rats,

cr

all procedures were as same as described above, but without pretreatment with CM.

us

2.6. RNA extraction and real-time quantitative PCR

PVAT from normal rats were pretreated as described in section 2.4. The total RNA was

an

obtained by Total RNA Extraction Reagent (SunShineBio, Nanjing, China). RNA was

M

converted into cDNA using the TransScript First-strand cDNA Synthesis Super Mix (TransGen Biotech, Beijing, China). The synthesized cDNA was used for PCR amplification

d

with the primers previously reported [10]. Quantitative PCR was performed using SsoFastTM

te

EvaGree Supermix (Bio-Rad, Hercules, California, USA) with the Bio-Rad iQ5 sequence

Ac ce p

detection system (Bio-Rad, Hercules,California, USA). Two protocols were used for qPCR of different genes. One protocol: 40 cycles (30 s at 95°C and 10s at 54°C) after an initial activation step for 10 min at 95°C. The other protocol: 40 cycles (30 s at 95°C and 10 s at 52°C) after an initial activation step for 10 min at 95°C. Primer sequences were shown in Supplementary Table. The mRNA level of individual genes was normalized and presented as a ratio to β-actin and calculated using the ΔΔCT method.

2.7. Western blot analysis Western blot analysis was performed as previously described [17]. For protein analysis, PVAT or aorta was lyzed in RIPA plus PMSF, and then the protein concentration of samples 7

Page 7 of 43

was determined by Bicinchoninic Acid (BCA) Protein Assay kit (Biosky Biotechnology Corporation, Nanjing, China). Equal amount of protein was separated by 10% SDS–polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride

ip t

(PVDF) membranes (0.45 μm, Millipore Co., Ltd.). After blocked with 5% non-fat milk for 2 h at room temperature, the PVDF membrane was incubated with primary antibody overnight

cr

at 4°C. Afterwards, the PVDF membrane was incubated with the secondary antibody at room

us

temperature for 2 h. The membranes were developed with enhanced chemiluminescence (ECL) visualized with an enhanced chemiluminescence detection system (ChemiDocXRS,

an

Bio-rad), and then quantitated by densitometry with Image-Pro Plus 6.0 (IPP 6.0) software.

d

above without any in vitro treatment.

M

The protein expression in the PVAT or aorta from HFD-fed rats was detected in the same way

te

2.8. Cell culture and AMPK small interfering RNA transfection

Ac ce p

3T3-L1 preadipocytes were purchased from the cell bank of the Chinese Academy of Sciences (Shanghai, China). Cells (2×105/ml) were cultured in six-well plates in DMEM containing 25 mM glucose at 37°C in a humidified atmosphere containing 5% CO2. 3T3-L1 preadipocytes were transfected with small interfering RNAs for AMPKα (sc-45313; Santa Cruz Biotechnology) to knockdown levels of endogenous AMPKα and negative control siRNA(sc-37007; Santa Cruz Biotechnology) using siRNA transfection reagent(sc-29528; Santa Cruz Biotechnology) in transfection medium (sc-36868; Santa Cruz Biotechnology) for 6 hours, according to the manufacturer’s instructions. After additional 48 hour culture the efficiency of siRNA-mediated AMPK knockdown was confirmed by western blotting.

8

Page 8 of 43

2.9. Statistical analysis The results were shown as mean ± SD. Data were analyzed by Student’s t-test or one-way ANOVA followed by Newman-Keuls test. Differences were considered statistically

ip t

significant at P < 0.05.

cr

3. Results

us

3.1. Diosgenin regulated AMPK activation in PVAT

In view of the important role of AMPK in the regulation of lipid metabolism, we first

an

investigated the influence of diosgenin on AMPK activity in PVAT. As shown in Fig.1A,

M

diosgenin treatment increased basal AMPK activity at concentrations ranging from 0.1 to 10 μM, indicated by enhanced AMPK phosphorylation. Meanwhile, we also observed the

d

regulation in PVAT exposed to PA insult. PA challenge reduced AMPK phosphorylation,

te

wherease this change was prevented by pre-treatment with diosgenin (Fig.1B). Meanwhile,

Ac ce p

AICIR, an AMPK agonist at 500 M, exhibited similar activities. These results showed that diosgenin enhanced AMPK activation under both basal and inflammatory conditions.

3.2. Diosgenin inhibited inflammation and modulated adipokine expression in PVAT PA stimulation triggered inflammation in PVAT, as we found an increase in IKKβ

phosphorylation. Diosgenin effectively reduced IKKβ phosphorylation, indicative of its anti-inflammatory activity. Co-treatment with AMPK inhibitor compound C attenuated the inhibitory effect of diosgenin (10 μM), suggesting the potential involvement of AMPK in the action (Fig.2A). Resveratrol, taken as a positive control, showed a similar regulation as diosgenin.

Meanwhile, we examined the effect of diosgenin on adipokine expression 9

Page 9 of 43

implicated in inflammation. As shown in Fig.2B and 2C, mRNA expressions for proinflammatory cytokines, including TNF-α, IL-6 and MCP-1, were enhanced, while expressions for adiponectin and PPARγ were downregulated, but these alterations were

ip t

reversed by treatment with diosgenin. In addition, diosgenin enhanced arginase 1(Arg-1) expression with downregulation of iNOS expression, and thereby resulted in an increased ratio

us

cr

of Arg-1 to iNOS (Fig.2D).

3.3. AMPK silencing blocked the anti-inflammatory effect of diosgenin in adipocytes

an

AMPK inhibitor compound C abrogated the inhibitory effects of diosgenin on

M

inflammation, suggesting the involvement of AMPK. To confirm this, we transfected 3T3- L1 cells with AMPK1/2 siRNA to knockdown AMPK expression. As shown in Fig.3A, silencing

d

AMPK significantly decreased the inhibitory effect of diosgenin on PA-mediated IKK

te

phosphorylation (approximately from 87% to 49%). Diosgenin treatment modulated altered

Ac ce p

production of TNF- and adiponectin upon PA challenge, but this action was also blocked by knockdown of AMPK (Fig.3C and 3D).These results indicated that diosgenin inhibited inflammation in a manner dependent on AMPK.

3.4. Treatment of PVAT with diosgenin restored the loss of ACh-induced vasodilation To investigate the influence of PAVT on vessel function, we incubated PVAT with PA and

collected the medium as conditioned medium (CM) to stimulate the rat aorta. As shown in Figure.3A, PA stimulation in PVAT impaired vasodilation in response to ACh, as the ACh-induced vasodilation at concentrations of 1, 10 μM were reduced from 83% to 38% (1 μM of ACh) and from 92% to 41% (10 μM of ACh), respectively. Pre-treatment of PVAT with 10

Page 10 of 43

diosgenin prevented the changes and effectively restored the loss of ACh-induced vasodilation in a concentration-dependent manner (Fig.4A). Meanwhile, we also found that treating PVAT with diosgenin effectively normalized eNOS phosphorylation in the aorta subjected to CM

ip t

challenge (Fig.4B). AMPK inhibitor compound C blocked the actions of diosgenin in the regulation of vasodilation and eNOS phosphorylation, suggesting the involvement of AMPK in

us

cr

diosgenin’s action.

3.5. Oral administration of diosgenin regulated AMPK and IKK phosphorylation in PVAT of

an

HFD fed rats

M

We further investigated the action of diosgenin in high fat diet (HFD)-fed rats. Long term-HFD feeding increased IKK phosphorylation in PVAT, whereas AMPK phosphorylation

d

was reduced. Oral administration of diosgenin effectively prevented the loss of AMPK

te

phosphorylation (Fig.5A) and reduced IKK phosphorylation (Fig.5B), indicating its

Ac ce p

anti-inflammatory activity in PVAT. Resveratrol, an AMPK activator as positive control, also exhibited an inhibitory effect on IKK phosphorylation. Meanwhile, we also observed that HFD led to an elevated level of TNF-α in the blood, while the level of adiponectin was reduced, but these alterations were reversed by oral administration of diosgenin at 20 mg/kg (Fig.5C and 5D).

3.6. Oral administration of diosgenin improved endothelium-dependent vasodilation in HFD-fed rats. HFD feeding impaired endothelium-dependent vasodilation, as we observed that ACh-mediated vasodilation was reduced in the aorta from HFD-fed rats. Oral administration of 11

Page 11 of 43

diosgenin effectively improved vascular function, as ACh-induced vasodilation (10 μM) was restored from 41% to 83% (20 mg/kg) and 70% (40 mg/kg) respectively by diosgenin treatment (Fig.6A). In line with the improvement of vasodilation, oral administration of diosgenin also

ip t

significantly enhanced eNOS phosphorylation in the aorta (Fig.6B). As a positive control,

us

cr

resveratrol showed a similar activity as diosgenin.

3.7. Oral administration of diosgenin improved vasodilation impaired by stimulation with

an

PVAT-CM from HFD-fed rats

M

Although no significant body gain was observed in HFD-fed rats, long term HFD-feeding increased PVAT deposit around the rat aorta, whereas the increased PVAT mass was reduced in

d

diosgenin-treated rats (Fig.7A). To know the direct influence of PVAT on vessel function, we

te

prepared CM by incubating PVAT from HFD-fed rats and then observed its influence on

Ac ce p

vasodilation in aorta from normal rats. HFD-feeding-derived CM impaired ACh-induced vasodilation in normal rat, as the percentages of relaxation induced by ACh (10 μM) were reduced from 90% to 39% (10 μM of ACh). Whereas, the alteration was prevented by diosgenin administration in HFD-fed rats, as the impaired ACh-induced vasodilation was restored to 68% (diosgenin, 20 mg/kg) and 57% (diosgenin, 40 mg/kg), respectively.

4. Discussion To investigate the paracrine regulation of arterial tone by PVAT, researchers may pay more attention on the regulation of vessel muscle, trying to identify adipocyte-derived relaxation factors (ADRF), which work somewhat in an endothelium-independent fashion[18, 19]. In 12

Page 12 of 43

fact, the endothelial homeostasis plays an important role in the regulation of vessel function and endothelial dysfunction predicts the development of cardiovascular diseases independently of other known risk factors [20]. In the present study, we prepared an ex vivo

ip t

model of PVAT/endothelial dysfuntion by treating rat aorta with PVAT-derived conditioned medium and successfully observed the beneficial effects of diosgenin on adipokine expression

cr

and further implications in endothelial function.

us

In addition to regulating energy metabolism, AMPK exerts anti-inflammatory activity, and this action has been documented to be implicated in the normalization of adipose and

an

endothelial functions [8, 21]. In view of the important role of AMPK, we first investigated the

M

action of diosgenin in the regulation of AMPK activation in PVAT and found it increased basal AMPK activity by enhancing phosphorylation at 0.1-10 M (Fig.1A). We also found

d

that PA stimulation reduced AMPK phosphorylation in PVAT, consistent with the previous

te

event [22]. However, this change was prevented by diosgenin (Fig.1B). The potency of

Ac ce p

diosgenin 10 M was similar to that of AMPK agonist, AICAR at 500 M. These results demonstrated the positive regulation of AMPK activity by diosgenin, which also presented the first evidence of diosgenin on AMPK activation. In obesity, recruited macrophages trigger inflammation in adipose tissue, leading to the

dysfunction of adipose tissue. Diosgenin positively regulated AMPK activity, and therefore we wanted to know whether this action contribute to the inhibition of inflammation. PA challenge induced inflammation in PVAT, evidenced by enhanced IKKβ phosphorylation. Diosgenin attenuated IKKβ phosphorylation, suggesting the inhibition of IKKβ/NF-κB inflammatory signalling. AMPK inhibitor compound C blocked the action of diosgenin, indicative of the involvement of AMPK (Fig.2A). PVAT is susceptible to inflammation, and 13

Page 13 of 43

as a result, we observed enhanced gene expressions for TNF-α, IL-6 and MCP-1, while the expressions for PPARγ and adiponectin, which exert anti-inflammatory effects, were downregulated. Diosgenin reversed the alteration of adipokine expression (Fig. 2B and 2C),

ip t

indicating that its anti-inflammatory activity was implicated in the beneficial regulation of adipokine expression. In obesity, an alteration of macrophage polarization (M1/M2) can be

cr

found in adipose tissue, and the alteration is towards M1 macrophages [23]. High level of

us

Arg-1 expression in contrast to low level of iNOS is considered to be an important indicator for M2 polarization, because this change facilitates the secretion of anti-inflammatory

an

cytokines, such interleukin-10 (IL-10) [21]. AMPK is shown to promote the polarization

M

towards M2 macrophage [24].Consistent with the regulation of AMPK, diosgenin restored Arg-1 expression and downregulated iNOS expression in PVAT subjected to PA insult, and

d

thereby increased the ratio of Arg-1 to iNOS (Fig.2D). Because there are more infiltrated

te

macrophages in adipose tissue and the change of Arg-1 and iNOS expression was a result

Ac ce p

from macrophages, the beneficial regulation of M2/M1 by diosgenin should contribute to blocking the inflammatory cross-talk between recruited macrophages and adipocytes in PVAT. For further identification of the role of AMPK in the anti-inflammatory activity of

diosgenin, we observed the activity of diosgenin in adipocytes. AMPKα knockdown using siRNA diminished the inhibitory effect of diosgenin on IKKβ activation (Fig.3A) and blocked its beneficial regulation of TNF-α and adiponectin production (Fig. 3B and 3C), further identifying the role of AMPK in the anti-inflammatory activity of diosgenin. Meanwhile, we should note that despite “silencing” of AMPK attenuated the inhibitory effect of diosgenin on IKKβ phosphorylation and TNF-α production, however this action is not complete, suggesting that other mechanisms might be involved in its anti-inflammatory action in PVAT. In fact, 14

Page 14 of 43

diosgenin is a multifunctional compound and its action in the regulation of MAPK and Akt have been documented to inhibit inflammation in the vessel [14, 25], In addition, sirtuin 1(SIRT1), a protein deacetylase interacting with AMPK, has been proven to contribute to

ip t

inhibiting inflammation in PVAT [10]. Thus, it might be inferred that other modulations, more than AMPK, could be implicated in the anti-inflammatory action of diosgenin in PVAT.

cr

Adipocyte-derived adiponectin promotes NO production in the endothelium and

us

enhances PPARγ action for the improvement of endothelial function [26, 27]. Given the functional interaction between PVAT and vasculature, we expected that the beneficial

homeostasis

through

paracrine

influences.

Endothelial

dysfunction

is

M

endothelial

an

regulation of adipokine expression in PVAT should contribute to the maintenance of

characterized by the impairment of endothelium-dependent vasodilation, and therefore we

d

pretreated PVAT with diosgenin and collected the medium as conditioned medium (CM) to

through

regulation

of

adipokine

secretion

in

PVAT.

ACh-induced

Ac ce p

vasodilation

te

treat rat aorta, aiming to observe the influence of diosgenin on endothelium-dependent

endothelium-dependent vasodilation is mediated through eNOS activation and subsequent NO production. Pretreatment of PVAT with PA reduced eNOS phosphorylation and impaired ACh-mediated vasodilation, indicating the association between dysregulation of adipokine expression and endothelial dysfunction. In this study, we demonstrated that diosgenin treatment in PVAT effectively enhanced eNOS phosphorylation and improved the vasodilation in response to Ach (Fig. 4A and 4B). These results proved our hypothesis that the beneficial regulation of PVAT function by diosgenin contributed to the amelioration of endothelial dysfunction. Above-observed results were derived from experiments ex vivo, we needed more 15

Page 15 of 43

evidence in vivo to confirm the regulation. Therefore we further observed the effects of diosgenin by focusing on PVAT/endothelial function in HFD-fed rats. Adiponectin is an adipocyte-derived relaxation factor with anti-inflammatory activity [28]. Long term HFD

ip t

feeding in rats led to reduced serum adiponectin, while the level of pro-inflammatory cytokine TNF-α was elevated. Oral administration of diosgenin restored the loss of adiponectin with

cr

downregulation of TNF-α level, suggesting its beneficial regulation of adipose tissue function.

us

Although no significant body weight gains were observed in high-fat fed rats (data not shown), HFD feeding led to an increase fat deposit in PVAT (Fig.7A). Similar to the finding ex vivo,

an

we observed reduced AMPK phosphorylation in PVAT, while IKKβ phosphorylation

M

increased (Fig. 5A and 5B), indicating the converse relation between inflammation and AMPK activities in PVAT. Oral administration of diosgenin reversed these alterations, and as

d

an expected result, we observed increased eNOS activity and the improvement of vasodilation

te

in response to ACh in the aorta (Fig. 6A and 6B). To address the influence of PVAT on vessel

Ac ce p

function, we isolated PVAT from HFD-fed rats to prepare conditioned medium (CM) for the treatment of the aorta from normal rats. Orally given diosgenin effectively prevented the alteration of CM and thus alleviated the loss of endothelium-dependent vasodilation induced by CM (Fig.7B), and thereby finally confirmed that diosgenin’s anti-inflammatory effect in PVAT contributed to the amelioration of endothelial dysfunction. Recent evidences for the involvement of PVAT in the development of vascular disorders

[29, 30] shed some new insights on the management of cardiovascular diseases. Previously, some studies have demonstrated the modulatory effect of diosgenin on chronic inflammation in adipose tissue[15, 16] and impaired endothelium-dependent vasodilation [14, 31], whereas few data are available to correlate the influence of diosgenin on adipose tissues to that on 16

Page 16 of 43

endothelium. We presented here, for the first time, that the protective effect of diosgenin on endothelium is at least resulting from its action on PVAT, and activation of AMPK in adipocytes accounts for the underlying mechanism. Subsequent studies might focus on

ip t

whether modulation of PVAT by diosgenin contributes to the regulation of other cells contained within arterial wall such as adventitial fibroblasts and vascular smooth muscle cells,

cr

for these cells constitute a critical pathogenetic component of vascular remodeling during the

us

progression of vascular diseases [32]. On the other hand, since AMPK has become the focus of researches as a novel therapeutic target in cardiometabolic diseases [33], further

an

investigations of the effect of diosgenin on AMPK activation in other related cells are

M

desirable.

In conclusion, our works ex vivo and in vivo show that diosgenin modulates adipokine

d

expression in PVAT by inhibiting inflammation via positive regulation of AMPK, and thereby

te

ameliorates endothelial dysfunction. The proposed mechanism of diosgenin has been shown

Ac ce p

in Fig.8. The findings suggest that the regulation of adipose function might be a potential target through which diosgenin improves vascular function.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this

paper.

Acknowledgements This study was supported by the National Natural Science Foundation of China (No.81274131), a Project Funded by the Priority Academic Program Development of Jiangsu 17

Page 17 of 43

ip t

Higher Education Institutions.

T. Szasz, R.C. Webb, Perivascular adipose tissue: more than just structural support,

us

[1]

cr

References

Clinical Science 122 (2012) 1-12.

R.P. Brandes, The fatter the better? Perivascular adipose tissue attenuates vascular

an

[2]

M

contraction through different mechanisms, British journal of Pharmacology 151 (2007) 303-304.

G. Fésüs, G. Dubrovska, K. Gorzelniak, et al., Adiponectin is a novel humoral

d

[3]

M.G. Ortega, P. Stucchi, R.R. Guzmán, et al., Adaptative nitric oxide overproduction in

Ac ce p

[4]

te

vasodilator, Cardiovascular Research 75 (2007) 719-727.

perivascular adipose tissue during early diet-induced obesity, Endocrinology 151 (2010) 3299-3306.

[5]

L. Chang, L. Villacorta, R. Li, et al., Loss of perivascular adipose tissue upon PPARγ deletion in smooth muscle cells impairs intravascular thermoregulation and enhances atherosclerosis, Circulation 126 (2012) 1067-1078.

[6]

M.A.Oriowo, Perivascular adipose tissue, vascular reactivity and hypertension, Medical Principles and Practice 24 (2015) 29-37.

[7]

T.A. Almabrouk, M.A. Ewart, I.P. Salt, et al., Perivascular fat, AMP-activated protein kinase and vascular diseases, British journal of Pharmacology 171 (2014) 595-617. 18

Page 18 of 43

[8]

S. Bijland, S.J. Mancini, I.P. Salt, Role of AMP-activated protein kinase in adipose tissue metabolism and inflammation, Clinical Science 124 (2013) 491-507.

[9]

M.S. Gauthier, E.L. O'Brien, S. Bigornia, et al., Decreased AMP-activated protein

ip t

kinase activity is associated with increased inflammation in visceral adipose tissue and

Biophysical Research Communications 404 (2011) 382-387.

cr

with whole-body insulin resistance in morbidly obese humans, Biochemical and

us

[10] Y. Sun, J. Li, N. Xiao, et al., Pharmacological activation of AMPK ameliorates perivascular adipose/endothelial dysfunction in a manner interdependent on AMPK and

an

SIRT1, Pharmacological Research 89 (2014) 19-28.

M

[11] J.X. Song, L. Ma, J.P. Kou, et al., Diosgenin reduces leukocytes adhesion and migration linked with inhibition of intercellular adhesion molecule-1 expression and NF-κB p65

te

142-149.

d

activation in endothelial cells, Chinese journal of Natural Medicines 10 (2012)

Ac ce p

[12] H.P. Yang, L. Yue, W.W. Jiang, et al., Diosgenin inhibits tumor necrosis factor-induced tissue factor activity and expression in THP-1 cells via down-regulation of the NF-κB, Akt, and MAPK signaling pathways, Chinese journal of Natural Medicines 11 (2013) 608-615.

[13] K.L. Dias, N.A. Correia, K.K. Pereira, et al., Mechanisms involved in the vasodilator effect induced by diosgenin in rat superior mesenteric artery, European journal of Pharmacology 574 (2007) 172-178. [14] K. Liu, W.W. Zhao, X.J. Gao, et al., Diosgenin ameliorates palmitate-induced endothelial dysfunction and insulin resistance via blocking IKKβ and IRS-1 pathways, Atherosclerosis 223 (2012) 350-358. 19

Page 19 of 43

[15] T. Uemura, S. Hirai, N. Mizoguchi, et al., Diosgenin present in fenugreek improves glucose metabolism by promoting adipocyte differentiation and inhibiting inflammation in adipose tissues, Molecular Nutrition & Food Research 54 (2010) 1596-1608.

ip t

[16] S. Hirai, T. Uemura, N. Mizoguchi, et al., Diosgenin attenuates inflammatory changes in the interaction between adipocytes and macrophages, Molecular Nutrition & Food

cr

Research 54 (2010) 797-804.

us

[17] Q. Liu, J.P. Kou, B.Y. Yu, Ginsenoside Rg1 protects against hydrogen peroxide-induced cell death in PC12 cells via inhibiting NF-kappaB activation, Neurochemistry

an

International 58 (2011) 119-125.

M

[18] M. Gollasch, Vasodilator signals from perivascular adipose tissue, British journal of Pharmacology 165 (2012) 633-642.

d

[19] J. Schleifenbaum, C. Köhn, N. Voblova, et al., Systemic peripheral artery relaxation by

Ac ce p

1875-1882.

te

KCNQ channel openers and hydrogen sulfide, Journal of Hypertension 28 (2010)

[20] J. Calles-Escandon, M. Cipolla, Diabetes and endothelial dysfunction: a clinical perspective, Endocrine Reviews 22 (2001) 36-52.

[21] Y. Zhang, J. Qiu, X. Wang, et al., AMP-activated protein kinase suppresses endothelial cell inflammation through phosphorylation of transcriptional coactivator p300, Arteriosclerosis, Thrombosis, and Vascular Biology 31 (2011) 2897-2908.

[22] G.R. Steinberg, B.J. Michell, B.J. van Denderen, et al., Tumor necrosis factor α-induced skeletal muscle insulin resistance involves suppression of AMP-kinase signaling, Cell Metabolism 4 (2006) 465-474. [23] C.N. Lumeng, J.L. Bodzin, A.R. Saltiel, Obesity induces a phenotypic switch in adipose 20

Page 20 of 43

tissue macrophage polarization, The Journal of Clinical Investigation 117 (2007) 175-184. [24] S. Galic, M.D. Fullerton, J.D. Schertzer, et al., Hematopoietic AMPK β1 reduces mouse

ip t

adipose tissue macrophage inflammation and insulin resistance in obesity, The Journal of Clinical Investigation 121 (2011) 4903-4915.

cr

[25] K.W. Choi, H.J. Park, D.H. Jung, et al., Inhibition of TNF-α-induced adhesion molecule

us

expression by diosgenin in mouse vascular smooth muscle cells via downregulation of the MAPK, Akt and NF-κB signaling pathways, Vascular Pharmacology 53 (2010)

an

273-280.

M

[26] K.K. Cheng, K.S. Lam, Y. Wang, et al., Adiponectin-induced endothelial nitric oxide synthase activation and nitric oxide production are mediated by APPL1 in endothelial

d

cells, Diabetes 56 (2007) 1387-1394.

te

[27] W.T. Wong, X.Y. Tian, A. Xu, et al., Adiponectin is required for PPARγ-mediated

Ac ce p

improvement of endothelial function in diabetic mice, Cell Metabolism 14 (2011) 104-115.

[28] D.H. Kim, C. Kim, E.L. Ding, et al., Adiponectin Levels and the Risk of Hypertension A Systematic Review and Meta-Analysis, Hypertension 62 (2013) 27-32.

[29] C.C. Ruan, Q. Ge, Y. Li, et al., Complement-Mediated Macrophage Polarization in Perivascular Adipose Tissue Contributes to Vascular Injury in Deoxycorticosterone Acetate-Salt Mice, Arteriosclerosis, Thrombosis, and Vascular Biology 35 (2015) 598-606. [30] D. Manka, T.K. Chatterjee, L.L. Stoll, et al., Transplanted Perivascular Adipose Tissue Accelerates Injury-Induced Neointimal Hyperplasia Role of Monocyte Chemoattractant 21

Page 21 of 43

Protein-1, Arteriosclerosis, Thrombosis, and Vascular Biology 34 (2014) 1723-1730. [31] J. Manivannan, E. Balamurugan, T. Silambarasan, et al., Diosgenin improves vascular function by increasing aortic eNOS expression, normalize dyslipidemia and ACE

ip t

activity in chronic renal failure rats, Molecular and Cellular Biochemistry 384 (2013) 113-120.

cr

[32] S. Sartore, A. Chiavegato, E. Faggin, et al., Contribution of adventitial fibroblasts to

us

neointima formation and vascular remodeling from innocent bystander to active participant, Circulation Research 89 (2001) 1111-1121.

an

[33] A.K. Wong, J. Howie, J.R. Petrie, et al., AMP-activated protein kinase pathway: a

M

potential therapeutic target in cardiometabolic disease, Clinical Science 116 (2009)

Ac ce p

te

d

607-620.

22

Page 22 of 43

Legends Fig. 1. Diosgenin regulated AMPK phosphorylation in PVAT. (A): PVAT was incubated with diosgenin (dio) for 1 h. (B): PVAT was pretreated with diosgenin for 0.5 h, and then stimulated with PA (100 M) for another

ip t

0.5 h. AMPK phosphorylation was determined by Western blot. AICAR was used as a positive control. The results were expressed as the mean ± SD of three independent experiments. #P<0.05 and ##P<0.01 vs. blank;

cr

P<0.01 vs. control.

us

**

Fig. 2. Diosgenin inhibited IKKβ activation and regulated adipokine expression in PVAT. (A): PVAT was

an

pretreated with diosgenin (dio) or diosgenin plus compound C for 30 min before co-incubated with PA (100

M

M) for another 0.5 h. IKK phosphorylation was assayed by Western blot. (B-D): PVAT was incubated with diosgenin in the presence of PA (100 M) for 2 h and then mRNA expressions were determined by qPCR.

*

P<0.01 vs. blank; P<0.05 and

**

P<0.01 vs. control; $P<0.01 vs. compound C

Ac ce p

treatment.

##

te

independent experiments.

d

Resveratrol (Res) was taken as a positive control. The results were presented as the mean ± SD of three

Fig. 3. AMPK siRNA transfection impaired diosgenin’s ability to suppress inflammation in 3T3-L1 preadipocytes. (A): 3T3-L1 preadipocytes were transfected with AMPK1/2 siRNA or control siRNA. After gene silencing for 48 h, siRNA-mediated AMPK knockdown was confirmed by western blotting. (B): siRNA-transfected adipocytes were pretreated with diosgenin (Dio) for 0.5h, and then stimulated with PA (100 M) for another 0.5h. IKK phosphorylation was assayed by Western blot. The results were presented as the mean ± SD of three independent experiments. (C-D): After transfection, cells were incubated with diosgenin (Dio) in the presence of PA (100 M) for 2h and removed the treated agents, then cells were cultured in fresh medium for another 22 h. The concentrations of adiponectin and TNF- in the supernatant 23

Page 23 of 43

were measured with ELISA Kits. The results were presented as the mean ± SD (n=4). **

##

P<0.01 vs. blank;

P<0.01 vs. control; $P<0.01 vs. AMPK siRNA treatment.

ip t

Fig. 4. Effects of diosgenin on vasodilation and eNOS phosphorylation in CM-treated rat aorta. Conditioned medium (CM) was prepared by stimulating PVAT with PA (100 M) in the presence or absence of diosgenin

cr

(Dio) and compound C (CC). (A): the aortic ring was pretreated with PA for 0.5 h. contracted with

us

phenylephrine (PE, 1μM), and the relaxation was induced by ACh treatment. Resveratrol (Res) was taken as a positive control. The result was expressed as the mean ± SD (n=5). (B): The aorta was exposed to CM for

**

P<0.01 vs. control; $P<0.01 vs. compound C treatment.

M

three independent experiments. ##P<0.01 vs. blank;

an

0.5 h, and eNOS phosphorylation was assayed by Western blot. The data was expressed as the mean ± SD of

d

Fig. 5. Diosgenin regulated AMPK and IKKβ activation in PVAT and serum levels of adiponectin and TNF-.

te

Rats were fed with HFD for 8 weeks with oral administration of diosgenin (Dio). (A-B): AMPK and IKKβ

Ac ce p

phosphorylation in PVAT from HFD-fed rats was determined by western blot. Data were derived from 3 rats. (C-D): TNF-α and adiponectin in serum were determined with ELISA kits. Resveratrol (Res) was taken as a positive control. The results were expressed as the mean ± SD (n=8). **

##

*

P<0.01 vs. blank; P<0.05 and

P<0.01 vs. control.

Fig. 6. Diosgenin improved endothelium-dependent relaxation in aorta from HFD-fed rats and restored eNOS phosphorylation. Rats were fed with HFD for 8 weeks with oral administration of diosgenin (Dio). (A): the aortic ring was pre-contracted with phenylephrine (PE, 1 M), and ACh was used to induce the vasodilation. The results were presented as the mean±SD (n=5). (B): eNOS phosphorylation in the aorta from HFD-fed rats was assayed by Western blot. Data were derived from 3 rats. Resveratrol (Res) was taken as a positive 24

Page 24 of 43

control. ##P<0.01 vs. blank;

**

P<0.01 vs. control.

Fig. 7. Diosgenin reversed the altered vasodilation of aorta from normal rats pretreated with PVAT-CM. Rats

ip t

were fed with HFD for 8 weeks with oral administration of diosgenin (Dio). (A): The mass of PVAT deposit around the aorta from HFD-fed rats; (B): The aortic rings from normal rats were incubated with CM derived

cr

from PVAT of HFD-fed rats and the relaxation was induced by ACh. Resveratrol (Res) was taken as a **

P<0.01 vs. control.

us

positive control. The results were expressed as the mean±SD (n=5). ##P<0.01 vs. blank;

an

Fig. 8. Schematic representation of the anti-inflammatory mechanisms of diosgenin in ameliorating

Primer Sequence

TNF-α

Forward

GAGTAAGGGGATGCAGCTAAGA

Reverse

CAGTTTCAGGGCAAGAAGTACC

Forward

TCTTGGGACTGATGTTGTTGAC

81

Reverse

GGGTGGTATCCTCTGTGAAGTC

Forward

AATGAGTCGGCTGGAGAACTAC

87

Reverse

TCTCTCTTGAGCTTGGTGACAA

Forward

AAGGGGACAACAATGGACTCTA

Reverse

CTACGGGCTGCTCTGAATTAGT

Forward

CAAGAATACCAAAGTGCGATCA

Reverse

CTTCATGTGGCCTGTTGTAGAG

Forward

ATCATGGAAGTGAACCCAACTC

Reverse

TCCAAAACAAGACAAGGTCAAC

Forward

GCAGAAGCACAAAGTCACAGAC

MCP-1

Ac ce p

IL-6

Adiponectin

PPARγ

Arg-1

iNOS

d

Gene

te

Table Primer sequences used in the study

M

PVAT/endothelial dysfunction and the models employed in the study.

Size of Product (bp) 93

110

88

93

90 25

Page 25 of 43

Forward

GACGTTGACATCCGTAAAGACC

Reverse

TGCTAGGAGCCAGGGCAGTA

115

te

d

M

an

us

cr

ip t

TCCTTTTCCTCTTTCAGGTCAC

Ac ce p

β-actin

Reverse

26

Page 26 of 43

Ac ce p

te

d

M

an

us

cr

ip t

Figure 1

Page 27 of 43

Ac

ce

pt

ed

M

an

us

cr

i

Figure 2

Page 28 of 43

Ac

ce

pt

ed

M

an

us

cr

i

Figure 3

Page 29 of 43

Ac ce p

te

d

M

an

us

cr

ip t

Figure 4

Page 30 of 43

Ac

ce

pt

ed

M

an

us

cr

i

Figure 5

Page 31 of 43

Ac ce p

te

d

M

an

us

cr

ip t

Figure 6

Page 32 of 43

Ac ce p

te

d

M

an

us

cr

ip t

Figure 7

Page 33 of 43

Ac ce p

te

d

M

an

us

cr

ip t

Figure 8

Page 34 of 43

Ac ce p

te

d

M

an

us

cr

ip t

black-and-white version of figure 1

Page 35 of 43

Ac

ce

pt

ed

M

an

us

cr

i

black-and-white version of figure 2

Page 36 of 43

Ac

ce

pt

ed

M

an

us

cr

i

black-and-white version of figure 3

Page 37 of 43

Ac ce p

te

d

M

an

us

cr

ip t

black-and-white version of figure 4

Page 38 of 43

Ac

ce

pt

ed

M

an

us

cr

i

black-and-white version of figure 5

Page 39 of 43

Ac ce p

te

d

M

an

us

cr

ip t

black-and-white version of figure 6

Page 40 of 43

Ac ce p

te

d

M

an

us

cr

ip t

black-and-white version of figure 7

Page 41 of 43

Ac ce p

te

d

M

an

us

cr

ip t

black-and-white version of figure 8

Page 42 of 43

Ac

ce

pt

ed

M

an

us

cr

i

Graphical Abstract

Page 43 of 43