The sirtuin 6 prevents angiotensin II-mediated myocardial fibrosis and injury by targeting AMPK-ACE2 signaling

Sirtuin 6 (SIRT6) is an important modulator of cardiovascular functions in health and diseases. However, the exact role of SIRT6 in heart disease is poorly defined. We hypothesized that SIRT6 is a negative regulator of angiotensin II (Ang II)-mediated myocardial remodeling, fibrosis and injury. The male Sprague-Dawley rats were randomized to Ang II (200 ng/kg/min) infusion with an osmotic minipump and pretreated with recombinant plasmids adeno-associated viral vector (AAV)-SIRT6 (pAAV-SIRT6) or pAAV-GFP for 4 weeks. Ang II triggered downregulated levels of SIRT6 and angiotensin-converting enzyme 2 (ACE2) and upregulated expression of connective tissue growth factor (CTGF) and proinflammatory chemokine fractalkine (FKN), contributing to enhanced cardiac fibrosis and ultrastructural injury. Reduced levels of phosphorylated pAMPK-α, increased myocardial hypertrophy and impaired heart dysfunction were observed in both Ang II-induced hypertensive rats and ACE2 knockout rats, characterized with increases in heart weight and left ventricular (LV) posterior wall thickness and decreases in LV ejection fraction and LV fractional shortening. More importantly, pAAV-SIRT6 treatment strikingly potentiated cardiac levels of pAMPKα and ACE2 as well as decreased levels of CTGF, FKN, TGFβ1, collagen I and collagen III, resulting in alleviation of Ang II-induced pathological hypertrophy, myocardial fibrosis, cardiac dysfunction and ultrastructural injury in hypertensive rats. In conclusion, our findings confirmed cardioprotective effects of SIRT6 on pathological remodeling, fibrosis and myocardial injury through activation of AMPK-ACE2 signaling and suppression of CTGF-FKN pathway, indicating that SIRT6 functions as a partial agonist of ACE2 and targeting SIRT6 has potential therapeutic importance for cardiac fibrosis and heart disease.


INTRODUCTION
Heart failure is growing epidemic with high morbidity and mortality at an international scale [1,2]. The sirtuin SIRT6, a conserved nicotinamide adenine dinucleotide-dependent protein deacetylase, plays critical roles in the regulation of cardiovascular functions, energy metabolism and aging [3][4][5]. Both single-and double-SIRT6 knockout mice develop significant cardiac hypertrophy, fibrosis and dysfunction [6]. Myocardial

Research Paper
Oncotarget 72303 www.impactjournals.com/oncotarget samples from patients with heart failure showed markedly reduced SIRT6 levels, indicating an important role of the SIRT6 signaling in heart disease [6]. SIRT6 is highly expressed in heart and is an important modulator of cardiovascular functions in health and diseases [3,5,7].
Increasing evidence indicates the crosstalk between the SIRT signaling and the renin-angiotensin system (RAS) [6,7]. Physiological, pharmacological and clinical studies have demonstrated that activation of the RAS is a key mediator of the progression of hypertension and heart failure [2,[8][9]. Angiotensin (Ang) II, a main biological peptide of RAS, plays an essential role in myocardial interstitial fibrosis by enhancing activation of connective tissue growth factor (CTGF)-fractalkine (FKN) signaling, contributing to increased myocardial hypertrophy, dysfunction and injury [10,11]. ACE2 is considered as a potential therapeutic target of the RAS for the treatment of hypertension and heart failure and converts Ang II into Ang-(1-7), which, by virtue of its actions on the Mas receptor, opposes the molecular and cellular effects of Ang II [12][13][14]. As a critical enzyme in the metabolism of Ang II, ACE2 serves to directly balance the levels of Ang II and Ang (1)(2)(3)(4)(5)(6)(7). Absence of ACE2 triggers greater increases in cardiac CTGF-FKN signaling in ACE2-deficient mice in response to chronic Ang II infusion, leading to exacerbation of the Ang II-induced myocardial hypertrophy and ultrastructure injury [10]. Intriguingly, ACE2 expression is found to be regulated by SIRT1 and adenosine 5'-monophosphate-activated protein kinase (AMPK) activation results in increased ACE2 expression [15]. However, the interaction between the SIRT6 and AMPK/ACE2 signaling in heart disease is poorly defined. In the present study, we hypothesized that SIRT6 is a negative regulator of Ang II-mediated myocardial remodeling and fibrosis by targeting AMPK/ ACE2 signaling. We evaluated the effects of Ang II and ACE2 deficiency on SIRT6 levels and regulatory roles of SIRT6 in AMPK/ACE2 signaling and Ang II-mediated heart disease by the use of ACE2-mutant and Ang IIinfused hypertensive rats.

Downregulation of SIRT6 and ACE2 levels and enhanced myocardial hypertrophy are obversed in Ang II-induced hypertensive rats
We firstly evaluated the effects of Ang II infusion on cardiac SIRT6 and ACE2 expression and myocardial hypertrophy. Ang II infusion led to marked increases in systolic blood pressure levels ( Figure 1A) and downregulation of myocardial SIRT6 levels ( Figure 1C) and ACE2 ( Figure 1D) in the recombinant plasmids adenoassociated viral vector (AAV)-GFP (pAAV-GFP)-treated hypertensive rats associated with increased levels of plasma Ang II ( Figure 1B). Double-immunofluorescence staining analysis further confirmed downregulated levels of SIRT6 and ACE2 ( Figures 2B, 2D) in Ang II-mediated hypertensive rats when compared with wildtype (WT) control rats. In addition, histomorphometric analysis ( Figures 3A, 3B) and Wheat germ agglutinin (WGA) staining ( Figures 3C, 3D) demonstrated enhanced myocardial hypertrophy in Ang II-mediated hypertensive rats along with obvious increases in heart weight (HW; Figure 3E) and the ratio of HW and body weight (BW) ( Figure 3F).

ACE2-deficient rat developes downregulation of SIRT6 and p-AMPKα levels, increased myocardial fibrosis and impaired heart function
We next evaluated the regulatory roles of ACE2 deficiency in levels of SIRT6 and phosphorylated AMPKα (pAMPKα) and pathological hypertrophy, fibrosis and cardiac function using ACE2 knockout (KO) rats. Loss of ACE2 resulted in reduced levels of SIRT6 (Figures 4D, 4E) and pAMPKα ( Figure 4F) as well as marked increases in FKN expression ( Figure 4G) and plasma Ang II levels ( Figure 4H) in ACE2KO hearts. Histomorphometric analysis ( Figure 5A), WGA ( Figure 5B, 5E) and picrosirius red staining ( Figure 5C) and echocardio-graphic assessment ( Figure 5D) revealed greater cardiac hypertrophy, increased myocardial fibrosis and impaired systolic function in ACE2-deficient rats characterized with decreases in left ventricular (LV) ejection fraction (LVEF; Figure 5I) and LV fractional shortening (LVFS; Figure 5J). These changes were associated with marked increases in HW ( Figure 5F), HW/BW ratio ( Figure 5G) and LV posterior wall thickness (LVPWT; Figure 5H) in ACE2-null hearts.

SIRT6 treatment prevented Ang II-induced myocardial hypertrophy and dysfunction by activation of AMPK/ACE2 signaling
Compared with pAAV-GFP-treated hypertensive rats, administration of pAAV-SIRT6 significantly prevented Ang II-mediated pressor response at post 4-week treatment ( Figure 1A) and myocardial hypertrophy ( Figure 7C) in hypertensive rats with upregulated cardiac levels of ACE2 ( Figure 1D) and pAMPKα ( Figures  6A,6B) . These changes were associated with marked decreases in HW ( Figure 3E), HW/BW ratio ( Figure 3G) and LVPWT ( Figure 7D) and augmentation of LVEF ( Figure 7E) and LVFS ( Figure 7F) in pAAV-SIRT6treated hypertensive hearts. However, administration of pAAV-SIRT6 did not affect systolic blood pressure levels at post 1-week treatment ( Figure 1A). Our data clearly demonstrate cardioprotective roles of SIRT6 in controlling cardiac remodeling and dysfunction via activation of the AMPK/ACE2 signaling.

DISCUSSION
In this study, we demonstrate a model of crosstalk between SIRT6 and ACE2 signaling in controlling values (D) illustrating myocardial levels of SIRT6 and ACE2 in rats. In immunofluorescence images, the red color represents SIRT6, the green color represents ACE2, and blue color represents DAPI stained nuclei. WTC=wildtype control; DAPI, 4',6-diamidino-2-phenylindole. n = 4. ** P < 0.01 compared with WTC group; ## P < 0.01 compared with Ang II+pAAV-GFP group.
Oncotarget 72306 www.impactjournals.com/oncotarget  Ang II-mediated pathological hypertrophy, myocardial remodeling, fibrosis, cardiac dysfunction and injury by enhancing p-AMPKα expression and preventing CTGF/FKN signaling. Increasing evidence suggests that sustained activation of CTGF-FKN signaling is associated with the pathogenesis of fibrotic disorders by regulating diverse cellular processes such as proliferation, migration, survival, and collagen deposition [10, [16][17][18]. CTGF has been revealed to enhance the levels of inflammatory chemokines FKN [19], which is a newly identified membrane-bound chemokine and plays a vital role in the transition from compensated ventricular remodeling hypertrophy to heart failure through the G protein-coupled chemokine receptor CX3CR1 [11]. We have previously Oncotarget 72308 www.impactjournals.com/oncotarget demonstrated that loss of ACE2 leads to greater increases in the Ang II-mediated myocardial inflammation, fibrosis and adverse myocardial injury via the activation of the CTGF-FKN and matrix metalloproteinase signaling, and pre-treatment with the CTGF-neutralizing antibody and recombinant ACE2 significantly abrogates Ang IIinduced upregulation of superoxide production and FKN levels in cardiofibroblasts [10,20]. Recent studies have also showed the cross-talk action between the FKN and Ang II signaling in cardiovascular remodeling and fibrosis [11,16,21]. SIRT6 binds to nuclear factor-κlight chain enhancer of activated B cells to negatively regulate their target gene transcription, including FKN [22,23]. In this work, we revealed that Ang II infusion promoted cardiac levels of proinflammatory chemokine FKN. These changes were linked with increases in plasma Ang II levels and myocardial fibrosis and impaired heart function. In contrast, SIRT6 treatment prevented Ang II-mediated hypertension, pathological hypertrophy, myocardial fibrosis, dysfunction and injury by upregulation of ACE2 levels and suppression of the FKN signaling.
The mammalian genome encodes 7 sirtuin isoforms (SIRT 1 to SIRT7) that vary in their tissue specificity, subcellular localization, enzymatic activity, and targets [22,[24][25]. SIRT6 is a chromatin-associated enzyme involved in deacetylating histone 3 at Lys9 (H3K9) and H3K56, thereby regulating gene expression, cellular metabolism, and the inflammatory response [22]. Previous studies demonstrated that SIRT6 expression is downregulated in human failing hearts and mice hearts subjected to transverse aortic constriction (TAC) or chronic infusion of hypertrophic agonists isoproterenol or Ang II, which may play a pivotal role in the etiology of hypertension [6]. In contrast, SIRT6 overexpression has been shown to activate the AMPK [5]. Intriguingly, AMPK activation ablates Ang II-induced endoplasmic reticulum stress and hypertension [26] and results in increased ACE2 expression [15]. ACE2 maps to a defined quantitative trait locus on the X chromosome in three different rat models of hypertension and targeted disruption of ACE2 in mice results in a severe cardiac contractility defect and increased Ang II levels [2,27]. In chronic heart failure patients, recombinant human ACE2 (rhACE2) effectively metabolized Ang-(1-10) and Ang II into Ang-(1-9) and Ang-(1-7), respectively. Myocardial Ang II levels in explanted human hearts with dilated cardiomyopathy were elevated and Ang II was effectively converted to Ang-(1-7) by rhACE2 [28], indicating ACE2 being an important negative regulator of the RAS and Ang II-mediated hypertension. In this work, we demonstrated that administration of pAAV-SIRT6 partially prevented Ang II-mediated pressor response at collagen I (E), collagen III (F) and FN1 (G) were measured in rats by Western blotting analysis. WTC= wildtype control; AMPK, adenosine 5'-monophosphate-activated protein kinase; CTGF, connective tissue growth factor; FKN, fractalkine; FN1, fibronectin 1. n=5 for each group. ** P < 0.01, compared with WTC group; # P < 0.05; ## P < 0.01 compared with Ang II+pAAV-GFP group. www.impactjournals.com/oncotarget post 4-week treatment, but not at post 1-week treatment, in hypertensive rats with upregulated cardiac levels of ACE2 and pAMPKα, indicating roles of AMPK/ACE2 signaling in hypotensive effects of SIRT6. The deacetylase activity of SIRT6 is decreased in Ang II-induced hypertrophic cardiomyocytes and TAC-induced hypertrophic hearts; while overexpression of wild-type SIRT6 but not its catalytically inactive mutant, attenuates AngII-induced cardiomyocyte hypertrophy [7]. Cardiac-specific overexpression of SIRT6 protected mice from pressure overload and agonist isoproterenol-induced hypertrophy [6,22]. In this work, we revealed that both Ang II infusion and genetic ACE2 deletion reduced SIRT6 levels and developed pathological hypertrophy and impaired cardiac function. These findings indicate that SIRT6 deficiency is associated with the development of cardiac hypertrophy and heart failure.
SIRT6 was reported to blunt the expression of pro inflammatory and profibrotic genes [23]. In this study, we also demonstrated that SIRT6 treatment significantly blocked cardiac expression of proinflammatory chemokine FKN and profibrotic factors such as TGFβ1, CTGF, collagen I and collagen III. For the reason that cardiac hypertrophy and inflammation are associated with myocardial fibrosis and injury, it is conceivable to think that SIRT6 is an antifibrotic and antiinflammatory sirtuin whose upregulation may help to impede the development of fibrosis-and inflammation-related diseases. SIRT6mutant mice spontaneously developed concentric cardiac hypertrophy with significantly increased HW/ BW ratio, reduced left ventricular internal diameter, reduced fractional shortening as well as increased cardiomyocyte size and interstitial fibrosis [6]. In contrast, SIRT6 overexpression has been shown to promote AMP/ ATP and then activate the AMPK-forkhead box O3α axis and further initiated the downstream antioxidant-encoding gene expression (manganese superoxide dismutase and catalase), thereby decreasing cellular levels of oxidative stress and mediating cardioprotective roles in the ischemic heart [5]. Intriguingly, SIRT1 is found to regulate ACE2 expression by activation of AMPK signaling. AMPK activation by 5-amino-4-imidazole-carboxamide riboside (AICAR) results in increased cellular ratio of NAD+ to NADH and increased ACE2 expression [15]. SIRT1, in  ultrastructural changes were observed in WT control (A), pAAV-GFP-(B) and pAAV-SIRT6-treated hypertensive rats (C) by transmission electron microscope analysis with×4200 and ×17500 magnification. Compared with WT control rats, severe myocardial ultrastructure injury was observed in pAAV-GFP-treated hypertensive rats, characterized with disruption or dissolution of myocardial myofilaments, myofilaments arranged irregularly and loosely (arrow), and vacuolar degenerational and swollen mitochondrias (star). Notably, myocardial ultrastructure injury were alleviated in hypertensive rats in response to pAAV-SIRT6 treatment.
Oncotarget 72311 www.impactjournals.com/oncotarget presence of a possible unknown co-factor, binds to the promoter region of ACE2 and this binding is promoted by AICAR. Intriguingly, the AICAR-induced ACE2 expression is inhibited by an inhibitor of SIRT1, providing strong evidence to the SIRT1 mediated transcriptional regulation of ACE2 under conditions of energy stress [15]. Local RAS exists in the cardiovascular tissues, and ACE2, a negative regulator of the RAS, is present in various celltypes including the cardiofibroblasts, cardiomyocytes and coronary microcirculation [2,9,29]. ACE2 plays a critical role in the control of cardiac physiology and its altered expression is linked to major pathophysiological changes of the cardiovascular system and fibrosis-related heart disorder [9,30,31]. In this study, we found that SIRT6 treatment remarkably potentiated phosphorylated AMPKα levels and subsequently enhanced ACE2 expression, thereby being responsible for attenuation of pathological hypertrophy, cardiac dysfunction, myocardial fibrosis, Figure 10: SIRT6 is a negative regulator of Ang II-mediated myocardial remodeling, fibrosis and injury. SIRT6 serves as a key regulator of Ang II-mediated hypertension, pathological hypertrophy, myocardial fibrosis and injury in hypertensive rats. Both Ang II infusion and ACE2 deficiency lead to reduction of SIRT6 levels, greater myocardial hypertrophy and impaired heart function. Administration of pAAV-SIRT6 upregulates cardiac levels of p-AMPK and ACE2 and prevents the activation of CTGF-FKN signaling, contributing to the amelioration of Ang II-mediated myocardial hypertrophy, fibrosis, dysfunction and ultrastructure injury in hypertensive rats.
Oncotarget 72312 www.impactjournals.com/oncotarget and injury in hypertensive rats. These observations confirmed cardioprotective effects of SIRT6 on myocardial remodeling, dysfunction and fibrosis through normalization of AMPK-ACE2 signaling.
In summary, both ACE2 deficiency and Ang II infusion block cardiac expression of SIRT6 and enhance the susceptibility to pathological remodeling, dysfunction and injury. Ang II promotes activation of CTGF-FKN pathway and increases myocardial fibrosis, resulting in adverse ultrastructure deterioration. In contrast, treatment with SIRT6 prevents Ang II-mediated pathological hypertrophy, myocardial fibrosis, impaired heart function and ultrastructural injury in hypertensive rats. These results indicate cardiac protective roles of SIRT6 in the prevention of myocardial remodeling, fibrosis, cardiac dysfunction and injury via activation of AMPK-ACE2 signaling and suppression of the CTGF-FKN pathway and support the notion that SIRT6 functions as a partial agonist of ACE2. Targeting SIRT6 and ACE2 has potential therapeutic importance for controlling myocardial remodeling, fibrosis and injury. Future studies are required to more precisely clarify roles of cross-talk between SIRT6 and ACE2 in cardiac injury and related heart diseases.

Experimental animals and protocols
The twelve-week-old male ACE2KO (Ace2 -/y ) rats were provided from H&D Biotechnology Inc (Wuhan, China) and backcrossed into a pure Sprague-Dawley (SD) rat background by the use of transcription activator-like effector nuclease (TALEN) and CRISPR/ Cas9 technologies as before [32]. DNA sequence chromatograms of polymerase chain reaction (PCR) products from WT and ACE2KO rats illustrated the 13 bp deletion in the homozygous sample ( Figure 4). The male SD rats were used to the normal control or randomized to either Ang II (200 ng/kg/min) or saline (control) infusion with an osmotic minipump (model 2004, Alzet Corp, Palo Alto, CA) for 4 weeks as before [10,31]. The recombinant pAAV-human SIRT6 (pAAV-SIRT6) were purchased from Shanghai Genechem Co.,Ltd.
(Shanghai, China) by the use of GV388 BamHI/BamHI vector. The hypertensive SD rats were treated by a rat tail-intravenous injection of pAAV-SIRT6 with the dose of 5 × 10 10 or pAAV-GFP at 1 week prior to 4-week Ang II infusion. Systolic blood pressure levels of rats were measured non-invasively using the tail-cuff method. All experiments were approved and performed in accordance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No.85-23, revised 1996) and the Animal Research Ethics Committee at Shanghai Jiao Tong University School of Medicine.

Echocardiography
Transthoracic echocardiography for rats was performed and analyzed in a blinded manner as described previously using a Vevo 2100 high-resolution imaging system [2,33]. Rats were anesthetized with 0.75% isoflurane for the duration of the recordings. M-mode images were obtained for measurements of LV endsystolic diameter (

WGA, picro-sirius red staining and confocal laser scanning analyses
Wheat Germ Agglutinin (WGA), H&E and Picrosirius red (PSR) staining were carried out according to manufacturer's instructions as previously described [14,20]. PSR staining of heart section was used to assess for myocardial interstitial fibrosis. Confocal laser scanning microscopy was used to test the colocalization SIRT6 of with ACE2 protein in rat hearts by using multi-fluorescent method. Species-appropriate secondary antibodies were conjugated to different fluorophores containing CY3 (Red) and fluorescein isothiocyanate (FITC; Green) in separate channels. Upon nuclear stain and mount in antifade medium containing 4',6'-diamidino-2-phenylindole (DAPI, blue), immunofluorescence images were acquired using a confocal laser scanning microscope (TCS SP2; Leica Microsystems AG, Wetzlar, Germany).

Transmission electron microscope array
For transmission electron microscope (TEM) analysis, samples of heart tissues were immediately cut into small pieces and immersed in 2.5% glutaraldehyde as described previously [10,20]. The myocardial ultrastructure of rat was observed on a conventional scanning transmission electron microscope (Philips, CM120, Amsterdam, Holland).

Statistical analysis
Values are represented as mean±SEM. All statistical analyses were performed with SPSS software (Version 16.0; Chicago, IL) either by Student's t test or by ANOVA followed by the Student-Newman-Keuls test for multiplecomparison testing as appropriate. A value of P < 0.05 was considered to indicate statistically significance.

Author contributions
Z.Z.Z, J.C.Z. and H.Y.J. conceived the study design, participated in the acquisition of data and drafted the manuscript. Z.Z.Z., Y.W.C., Q.C., Y.L.X., R.X. and B.S. carried out the experiments and participated in the acquisition of data, analysis and interpretation. Q.H.S., and L.X.C. have been involved in analyzing the data and provided critical advice. All authors read and approved the final manuscript.