Myricetin Alleviates Pathological Cardiac Hypertrophy via TRAF6/TAK1/MAPK and Nrf2 Signaling Pathway

Myricetin (Myr) is a common plant-derived polyphenol and is well recognized for its multiple activities including antioxidant, anti-inflammation, anticancer, and antidiabetes. Our previous studies indicated that Myr protected mouse heart from lipopolysaccharide and streptozocin-induced injuries. However, it remained to be unclear whether Myr could prevent mouse heart from pressure overload-induced pathological hypertrophy. Wild type (WT) and cardiac Nrf2 knockdown (Nrf2-KD) mice were subjected to aortic banding (AB) surgery and then administered with Myr (200 mg/kg/d) for 6 weeks. Myr significantly alleviated AB-induced cardiac hypertrophy, fibrosis, and cardiac dysfunction in both WT and Nrf2-KD mice. Myr also inhibited phenylephrine- (PE-) induced neonatal rat cardiomyocyte (NRCM) hypertrophy and hypertrophic markers' expression in vitro. Mechanically, Myr markedly increased Nrf2 activity, decreased NF-κB activity, and inhibited TAK1/p38/JNK1/2 MAPK signaling in WT mouse hearts. We further demonstrated that Myr could inhibit TAK1/p38/JNK1/2 signaling via inhibiting Traf6 ubiquitination and its interaction with TAK1 after Nrf2 knockdown in NRCM. These results strongly suggested that Myr could attenuate pressure overload-induced pathological hypertrophy in vivo and PE-induced NRCM hypertrophy via enhancing Nrf2 activity and inhibiting TAK1/P38/JNK1/2 phosphorylation by regulating Traf6 ubiquitination. Thus, Myr might be a potential strategy for therapy or adjuvant therapy for malignant cardiac hypertrophy.


Introduction
Pathological cardiac hypertrophy plays a central role in a variety of cardiovascular diseases and is an independent predictor for heart failure, arrhythmia, and sudden death [1,2]. Many pathological stimuli, such as chronic hypertension, myocardial infarction, and aortic stenosis, could inevitably induce pathological hypertrophy [1,2]. A series of malignant remodeling occurred in pathological hypertrophy, including enlarged cardiomyocyte, accelerated protein synthesis, cell death, aggravated fibrosis, dysregulated Ca 2+ -handling, disturbed mitochondrial function, and reactivated fetal gene expression [1,2]. Multiple signaling pathways, such as mitogen-activated protein kinase (MAPK), NF-κB, and Nrf2/HO-1, were dysregulated and were demonstrated to involve in promoting malignant remodeling. Targeting at these dysregulated signaling might find out strategies for protesting against pathological hypertrophy.
Myricetin (Myr) is a well-known polyphenol and possesses a similar molecular structure with kaempferol, quercetin, morin, and fisetin [3]. Scientific investigations had revealed that Myr possessed multiple pharmacological activities including antioxidant, antiphotoaging, anti-inflammation, antidiabetes, and anticancer [3]. Myr has also been demonstrated to protest against cardiovascular disease in both clinical and basic researches. Myr intake from daily diet was inversely associated with the incidence of cardiovascular diseases [4,5]. In basic researches, Myr treatment protected isoproterenol-(ISO-) induced myocardial infarction via increasing superoxide dismutase (SOD) and catalase (CAT) [6]. Myr exerted a more pronounced capacity than quercetin for preventing STAT1 activation in ischemia/reperfusion [7] and consequently showed more cardiovascular benefits. Myr treatment could also promote coronary dilation via increasing intracellular cGMP without affecting cardiomyocyte systolic and diastolic function [8]. These investigations implied that Myr might be a potential drug for therapy or adjuvant therapy in various cardiovascular diseases, but the underling mechanisms remain to be further explored.
In our laboratory, we have demonstrated that Myr attenuated LPS-induced mouse heart injuries by inhibiting oxidative stress and NF-κB/p65 activity [9]. We further demonstrated that the antioxidant activity of Myr might be associated with enhancing Nrf2/HO-1 pathway in diabetic mouse heart. However, Myr remained to inhibit NF-κB activity after Nrf2 silence in neonatal rat cardiomyocytes (NRCMs) [10]. Thus, some mechanisms independent of Nrf2/HO-1 remain to be investigated in mouse heart. Also, none published studies have investigated the roles and mechanisms of Myr in pressure overload-induced pathological hypertrophy. In this study, we intended to investigate whether Myr could protect mouse heart from pressure overload-induced hypertrophy and clarify the underlying mechanisms. (Proteintech, 16396-1-AP), 4-hydroxynonenal (Abcam, ab46545), SOD1 (Abcam, ab16831), and catalase (Proteintech, 19792-1-AP). Lip 6000 (C0528) and protein A+G Agarose were purchased from Beyotime (Jiangsu, China). Peroxidase-conjugated secondary antibodies were from Jackson ImmunoResearch Laboratories (1 : 10000); fetal calf serum was from HyClone (Waltham, MA, USA).

Animal Models.
All experimental protocols used for animal experiments in this study were approved by the Animal Care and Use Committee of Renmin Hospital of Wuhan University. All surgery procedures were performed according to the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals.

Generation of Cardiac-Specific Nrf2
Knockdown (Nrf2-KD) Mice. Nrf2 shRNA Lentiviral Particles (for mouse) were purchased from Santa Cruz Biotechnology (sc-37049-v). Intramyocardial injection of lentiviral particles was performed according to our previous study [11]. Briefly, mice were anaesthetized with pentobarbital sodium and mechanically ventilated. A horizontal skin incision was made at the 3-4 intercostal space. The heart was smoothly and gently "popped out." Lentiviral particles were injected into the left ventricle with a syringe equipped with a 31-gauge needle. The sham group was injected with scram shRNA Lentiviral Particles. Three different injection points were selected around the left ventricle. After injection, mouse heart was immediately put back and then close muscle and suture the skin. All mice were allowed for shRNA expression for 2 weeks and then were prepared for following experiments.

Pressure
Overload-Induced Hypertrophy Model. Aortic banding (AB) surgery was performed to establish pathological hypertrophy models according to previous depict [11]. Briefly, male mice (C57/BL6, age 8-10 weeks and weight 23.5-27.5 g) were anaesthetized with pentobarbital sodium (50 mg/kg, Sigma) by intraperitoneal injection. After losing toe pinch reflex, open the left chest at the 2-3 intercostal space to expose the aorta with blunting dissection. The aorta was tied with a 27-gauge or 26-gauge needle using a 7-0 silk suture. After ligation, the needle was removed gently to cause aortic constriction. Sham mice went through a similar procedure without aorta ligation.

Animal Groups and Myr Administration.
After three days of AB surgery or sham operation, mice were allocated into different experimental groups randomly. In this study, two independent animal experiments were performed. Firstly, animals were allocated into four groups: sham+normal saline (NS) group (control group, CON), sham+Myr treatment group (Myr), AB+NS (AB) group, and AB+Myr (AB+M) group. Secondly, animals were allocated into scram shRNA+sham+vehicle (NS) (sham+scram+Veh), scram shRNA+sham+Myr (sham+scram+Myr), shRNA+AB+Veh (shRNA+Veh+AB), and shRNA+AB+Myr. Myr were dissolved in NS and were administrated by gastric needle for consecutive 6 weeks with a dose of 200 mg/kg/d, which have been performed according to our previous studies [9,10]. The control groups were treated with NS.

Echocardiographic Analysis.
Cardiac function was measured by echocardiography before mouse sacrifice. Mice were anesthetized by 1.5% isoflurane. Mylab 30CV (ESAOTE S. P. A) equipped with a 15 MHZ linear-array ultrasound transducer was used to assess the left ventricle (LV) dimension at the parasternal short axis. M-mode tracing was used to detect and calculate the following parameters: interventricular septal thickness (IVS) at diastole (IVSd), IVS at systole (IVSs), LV end-diastolic diameter (LVEDd), LV endsystolic diameter (LVEDs), LV end-diastolic posterior wall thickness (LVPWd), LV end-systolic posterior wall thickness (LVPWs), LV ejection fraction (LVEF), and fractional shortening (FS). End-diastole or systole was defined as the phase of the largest or smallest area of the LV, respectively.

Histology Analysis.
After echocardiography measurement, mice were sacrificed by cervical dislocation. Heart weight (HW), lung weight (LW), and tibia length (TB) were recorded to calculate the HW/BW, LW/BW, and HW/TL. Then, mouse hearts were arrested in diastole in 10% KCl and fixed by 10% formalin for 12 h. After dehydration, hearts were embedded in paraffin and cut transversely close to the apex with 4-5 μm thickness. Hematoxylin-eosin (HE) and picrosirius red (PSR) staining were performed to assess the cardiomyocyte cross-sectional area (CSA) and collagen volume in the LV. A digital image analysis system (Image-Pro Plus, version 6.0) was used for image capture and analysis.

Neonatal Rat Cardiomyocyte (NRCM) Culture and
Treatment. Neonatal rat cardiomyocytes (NRCMs) were isolated from 1-3 day old Sprague-Dawley rats according to published protocol [11]. Briefly, NRCMs were isolated in D-hanks buffer containing 0.125% trypsin for repeated digestion with 15 min × 5 times. The harvested cells were resuspended in DMEM/F12 medium supplemented with 15% fetal bovine serum (FBS). A differential attachment method was used to remove cardiac fibroblasts from NRCMs, and then, NRCMs were seeded at a density of 2 × 10 5 cells per well in 6-well plates for protein or mRNA extraction and 1 × 10 4 cells per well in 24-well plates for immunofluorescence staining. NRCMs were incubated with 0.1 mmol/L bromodeoxyuridine (BrdU) in DMEM/F12 medium with 15% FBS for 36 h before the following experiments.
NRCMs were transfected with gene-specific siRNA for Nrf2 silence (SANTACRUZ, SC-156128) or plasmid for TAK1 overexpression using Lipo 6000 transfection reagent according to the manufacturer's instructions. After another 24 h incubation, NRCMs were cultured in serum-free DMEM/F12 for 12 h and then were stimulated with PE (50 μM) or Myr (20 μM). The treatment dosage of Myr was determined according to our previous publication [9].
Immunofluorescence staining was performed to evaluate NRCM hypertrophy. NRCMs were fixed with 4% paraformaldehyde solution for 10 min at room temperature and then washed with PBS for 3 min × 5 times. After permeabilization with 0.1% Triton X-100 in PBS for 15 min, NRCMs were stained with anti-cardiac Troponin T antibody (1 : 100) overnight. In the next day, the secondary antibody, Alexa Fluor 488 anti-mouse IgG, was used to label cTnT in green. The NRCMs were mounted on glass slides with Slow-Fade Gold antifade reagent with DAPI.
To demonstrate that Myr treatment could cause antioxidative stress via partly regulating Nrf2, NRCMs were transfected with specific siRNA for Nrf2 knockdown. After transfection of siRNA for 24 h, NRCMs were incubated with Myr or H 2 O 2 for another 48 hours. And then, NRCMs were harvested for detecting the expression of Nrf2-regulated genes by RT-PCR and antioxidative enzymes by western blots.
2.6. Western Blot Analysis. Mouse heart tissue or NRCMs were lysed in RIPA buffer. To examine NF-κB/P65 and Nrf2 nucleus translocation, nucleus protein was extracted by a commercial kit purchased from Jiancheng Bioengineering Institute (Nanjing, China). Protein concentration was determined by a BCA protein assay kit. 50 μg protein was used for electrophoresis on 10% SDS-PAGE gels and transferred to a polyvinylidene fluoride membrane (Millipore). After blocking with 5% BSA for 1 h, the blots were incubated with corresponding primary antibodies overnight at 4°C. The peroxidase-conjugated secondary antibodies were used to incubate with the blots for 1 h in the next day. The blots were visualized using Bio-Rad ChemiDocTX XRS+. All expressions of proteins were normalized to corresponding GAPDH before relative quantitative calculation.

Quantitative Real-Time PCR (rt-PCR).
Total mRNA was extracted from snap-frozen heart tissue or NRCMs using the TRIzol reagent according to the manufacturer's instructions. Harvested mRNA was spectrophotometrically estimated by A260/A280 and A230/260 relying on the Smartspec Plus Spectrophotometer (Bio-Rad). Total mRNA (2 μg/sample) was converted into cDNA using the Transcriptor First Strand cDNA Synthesis Kit. RT-PCR amplification of target genes in this study was performed using LightCycler 480 SYBR Green 1 Master Mix. All expression levels of genes were normalized to GAPDH before relative quantitative calculation. Primers used in this study are shown in Table 1.  2.9. HEK 293T Culture and Plasmid Transfections. Plasmid of Prk5-HA-Ubiquitin-K63 (K63-Ub) was obtained from addgene, which only expressed ubiquitin with K63, and other lysines were mutated to arginines [12]. The cDNA of TAK1 (NM 145331) was purchased from Vigene Biosciences (China) and cloned into pc-DNA 3.1 system (Invitrogen, USA) for TAK1 overexpression. HEK393T cells were purchased from China Center for Type Culture Collection (Wuhan, China) and prepared according to published protocol [13]. Briefly, HEK293T cells were seeded in 6-well plates with culture medium supplemented with 10% FBS. Cells were transfected with plasmid (2 μg) by Lipo 6000™ transfection reagent when cell density reached the 60-70% area of the culture dish. After 24 h transfection, cells were treated with 50 μM PE or 20 μM Myr for another 30 min. Finally, the cells were washed twice with 4 ml ice-cold PBS and lysed on ice with 70 μl lysis buffer in each well. Six-well plates were swirled on ice for 15 min, and the cell debris was collected with a cell scraper for following experiments.

Immunoprecipitation and Ubiquitination
Assay. Protein lysate from NRCMs or HEK293T cells was immunoprecipitated with anti-Traf6 coupled to protein A+G Agarose (Beyotime, Shanghai, China). Briefly, protein concentration was determined by BCA assay. Sufficient primary monoclonal antibody of Traf6 was added into 200 μg proteins and incubated at 4°C overnight with gentle rotation. The next day, the beads containing complexes were collected by centrifugation at 2500 rpm for 5 min. The containing complexes were washed and boiled in SDS-PAGE loading buffer. The immunoprecipitated proteins were electrophoresed on 10% SDS-PAGE gels and transferred to a polyvinylidene fluoride membrane (Millipore). Blots were incubated with anti-TAK1 and antiubiquitin to determine the Traf6 ubiquitination and interaction between Traf6 and TAK1.
2.11. Statistical Analysis. All data were presented with mean ± s:d. One-way analysis of variance (ANOVA) was used to compare means among groups followed by least significant difference (LSD, equal variances) or Tamhane' s T2 (none equal variances) tests. Image-Pro 6.0 was used for quantitative analysis of western blots and pictures. SPSS 19.0 was used for statistical analysis in this study. p < 0:05 was considered significant.

Myr Treatment Attenuated Pathological Cardiac
Hypertrophy and Fibrosis. As presented in Table 2, pressure overload induced obvious cardiac hypertrophy evidenced by increased HW, HW/BW, and HW/TB ratios compared with CON and Myr groups. However, Myr treatment significantly alleviated cardiac hypertrophy evidenced by decreased HW, HW/BW, and HW/TB. HE staining showed significant enlargement of cardiomyocyte (Figures 1(a) and 1(b)) and overproduction of ANP and BNP after AB surgery ( Figure 1(c)). Besides, the adult α-MHC was downregulated in mouse heart whereas the fetal β-MHC was significantly upregulated compared with CON and Myr groups (Figures 1(c) and 1(d)). Myr treatment significantly blunted these pathological changes induced by pressure overload (Figures 1(a)-1(d)). Cardiac fibrosis is an integrate process in the development of pathological cardiac hypertrophy. In this study, AB induced significant interstitial and perivascular fibrosis compared with CON and Myr groups (Figures 1(e) and 1(f)); however, Myr treatment significantly attenuated cardiac fibrosis and inhibited fibrosis-associated markers' expression compared with the AB group (Figures 1(e)-1(g)). Persistent cardiac hypertrophy and fibrosis directly contributed to cardiac dysfunction and heart failure. After 6 weeks of pressure overload, IVSd, LVEDd, LVEPWd, and LVEDs were significantly increased while EF and FS were markedly decreased compared with the CON group or the Myr group (Table 3). Myr treatment significantly improved cardiac function evidenced by decreased IVSd, LVEDd, LVEPWd, and LVEDs,  as well as increased EF and FS compared with the AB group (Table 3).

Myr
Enhanced the Nrf2/HO-1 and Inhibited MAPK/NF-κB. In our previous study, we have showed that Myr treatment could significantly enhance the Nrf2/HO-1 pathway and block NF-κB nuclei translocation [9,10]. In this study, we also detected the enhanced expression and nuclei translocation of Nrf2 resulted in the activation of the Nrf2/HO-1 pathway (Figures 2(a) and 2(b)), which was downregulated under chronic pressure overload (Figures 2(a) and 2(b)). In addition, chronic pressure overload induced hyperphosphorylation and translocation of NF-κB/p65 (Figures 2(c) and 2(d)), which was also prevented by Myr treatment (Figures 2(c) and 2(d)). Moreover, MAPK signaling (JNK1/ 2, P38, and ERK1/2) was significantly activated under pressure overload (Figures 2(e) and 2(f)); Myr treatment significantly blocked JNK1/2 and P38 overactivation compared with the AB group (Figures 2(e) and 2(f)), but no significant difference of ERK1/2 phosphorylation was found between the AB group and the AB+Myr group (Figures 2(e) and 2(f)).  (Tables 4 and 5). The EF and FS were also significantly improved in the shRNA+AB+Myr group compared to the shRNA+AB group (Table 5). HE and PSR stain-

(c) and 4(d)). Further investigations indicated that
Myr treatment markedly blocked TAK1 excessive phosphorylation, which was the common upstream regulator of p-JNK and p-p38 (Figures 4(c) and 4(d)). According to these results, it was reasonable to deduce that Myr treatment prevented cardiac hypertrophy at least partly through inhibiting the TAK1/MAPK pathway.

Myr Prevented PE-Induced NRCM Hypertrophy and H 2 O 2 -Induced Oxidative Stress.
In in vitro experiment, PE treatment induced significant NRCM hypertrophy, which could be markedly inhibited by Myr treatment (Figures 5(a) and 5(b)). After Nrf2 knockdown, Myr could also partly prevent NRCM from PE-induced hypertrophy (Figures 5(a) and 5(b)). However, after Nrf2 knockdown and TAK1 overexpression in NRCM, Myr could not protect NRCMs from PE-induced hypertrophy (Figures 5(a) and 5(b)). RT-PCR was also performed to examine hypertrophy-associated biomarkers in vitro. Myr treatment could significantly prevent

Discussion
This study demonstrated for the first time that Myr protected mouse heart and NRCM from pressure overload or PEinduced hypertrophy, respectively. Hypertrophic stimuli caused Nrf2/HO-1 downregulation and TRAF6 unregulation and its ubiquitination. Ubiquitinated TRAF6 could interact with TAK1 promoting TAK1/JNK1/2/P38 phosphorylation. Myr treatment significantly inhibited the TRAF6/TAK1/-MAPK pathway and restored Nrf2/HO-1 activity. Moreover,   Traf6/TAK1/MAPK cascade inhibition was independent of Nrf2/HO-1 activity. In addition, Myr treatment prevented Traf6 expression and ubiquitination and also prevented the TRAF6-TAK1 interaction. Previous investigations have presented that Myr was a powerful antioxidant in various cells and diseases [3]. Myr could increase the activity and protein expression of many antioxidases including SOD, GSH/GSSG ratio, catalase, and glutathione. These antioxidases were markedly downregulated in ischemia/reperfusion-induced myocardial injury and deoxycorticosterone acetate-(DOCA-) salthypertensive rats [16,17]. Myr reacted with oxygencentered galvinoxyl radicals 28 times faster than vitamin E [18]. Gene microarray analysis in HepG2 cells suggested that Myr might involve in promoting Nrf2-mediated antioxidant response element (ARE) activation for exerting its antioxidant roles [19]. Further investigations revealed that Myr treatment prevented Nrf2 ubiquitination and protein turnover, promoting Nrf2 expression and kelch-like erythroid cell-derived protein with CNC homology-associated protein 1 modification [19]. Our previous study also demonstrated that Myr treatment attenuated diabetes-associated cardiomyocyte hypertrophy, apoptosis, and interstitial fibrosis via restoring Nrf2/HO-1 pathway activity [10]. Nrf2 has been demonstrated to play important roles in regulating cardiac pathophysiology. Published data have showed that deficiency or downregulation of Nrf2 exacerbated pathological cardiac hypertrophy [20,21], while stimulating Nrf2 expression by genetic or drug treatment strategies could significantly attenuate pathological remodeling [22,23]. Taken together these evidences, Myr-mediated Nrf2 regulation was undoubtedly one of the most important mechanisms for preventing pathological cardiac remodeling. Some clinical investigations have figured out that Myr intake was inversely associated with the morbidity of myocardial infarction or coronary heart disease, but the underlying mechanisms remained to be unclear [4,24].
Our previous investigation showed that Myr treatment could improve LPS-induced cardiac injury by inhibiting IκB/NF-κB (p65) signaling and inflammatory cytokine secretion [9]. In a high glucose-induced NRCM injury model, we further demonstrated that Myr-mediated IκB/NF-κB (p65) inhibition was independent of Nrf2 enhancement [10]. In this study, we observed that Myr treatment inhibited p65 phosphorylation and nuclei translocation in both wild type and Nrf2-knockdown mouse hearts. These studies intensively displayed that Myr possessed potent anti-inflammatory activity. In lipoteichoic acid-(LTA-) treated human gingival  fibroblasts (HGFs), Myr treatment depressed p38 and ERK1/2 activation and also inhibited IκBα degradation [23,25]. In IL-1β-stimulated SW982 synovial cells, Myr significantly reduced JNK and p38 phosphorylation [26]. These results implied that Myr-mediated inflammation reduction might be associated with MAPK signaling regulation. Based on these previous studies, this study detected MAPK signaling and figured out that Myr treatment significantly inhibited p38 and JNK1/2 phosphorylation but showed none significant effects for ERK1/2 phosphorylation. TAK1 is an upstream regulatory protein of p38 and JNK1/2 MAPK [27,28]. TAK1 phosphorylation could contribute to p38 and JNK1/2 MAPK hyperphosphorylation resulted in aggravated pathological cardiac hypertrophy [27,28]. In this study, Myr treatment significantly depressed TAK1 phosphorylation after Nrf2 knockdown in in vivo experiment. Accordingly, we deduced that Myr alleviated cardiac hypertrophy through inhibiting TAK1 activation, which could be regulated by TRAF6 ubiquitination. Ji et al. [29] demonstrated that Traf6 autoubiquitination promoted the interaction of TRAF6 and TAK1 in the process of pressure overload-induced cardiac hypertrophy, and Traf6 deletion or inhibiting Traf6 ubiquitin effectively attenuated pressure overload-induced cardiac hypertrophy [29]. In previous studies, polyphenols have been demonstrated to suppress NF-κB activation through decreasing Traf6 ubiquitination in Hela-T6RZC stable cells and to disrupt the polyubiquitin synthesis in in vitro kinase assay system [15]. Resveratrol, another well-known polyphenolic compound, has also been demonstrated to inhibit LPS-induced p38 and JNK1/2 activation via diminishing TRAF6 ubiquitination [14].
Based on these studies, we investigated whether Myr would show similar function with other polyphenolic compounds to regulate Traf6 ubiquitination. PE induced Traf6 ubiquitination in NRCM accompanied with increased phosphorylation of TAK1, p38, JNK1/2, and ERK1/2. Myr treatment significantly inhibited the TRAF6 expression and its ubiquitination and decreased the activity of TAK1, p38, and JNK1/2. Previous studies have indicated that the TRAF6-associated ubiquitination activation was largely determined by a site-specific nondegradative Lys-63-linked autoubiquitination (K63-Ub) [29][30][31]. So, K63-Ub and TRAF6 plasmids were cotransfected into HEK-293-cells. PE treatment obviously promoted TRAF6 ubiquitination, which was significantly inhibited by Myr treatment. TRAF6 autoubiquitination promoted TRAF6 and TAK1 interaction. TRAF6 and TAK1 plasmids were cotransfected into HEK-293 cells to detect the TRAF6 and TAK1 interaction. Our results showed that PE treatment significantly promoted TRAF6 and TAK1 interaction, which could be markedly blocked by Myr treatment.
Taken together, Myr possessed potent capacity to restore Nrf2/HO-1 activity and to inhibit TAK1/p38/JNK1/2 MAPK signaling via inhibiting TRAF6 autoubiquitination. Thus, Myr might be a potential drug with multiple targets for