Small ubiquitin-related modifier (SUMO)ylation of SIRT1 mediates (-)-epicatechin inhibited- differentiation of cardiac fibroblasts into myofibroblasts

Abstract Context (-)-Epicatechin (EPI) is a crucial substance involved in the protective effects of flavanol-rich foods. Previous studies have indicated EPI has a cardioprotective effect, but the molecular mechanisms in inhibition of cardiac fibrosis are unclear. Objective We evaluated the effect of EPI in preventing cardiac fibrosis and the underlying molecular mechanism related to the SIRT1-SUMO1/AKT/GSK3β pathway. Materials and methods Cardiac fibrosis mice model was established with transaortic constriction (TAC). Male C57BL/6 mice were randomly separated into 4 groups. Mice received 1 mg/kg/day of EPI or vehicle orally for 4 weeks. The acutely isolated cardiac fibroblasts were induced to myofibroblasts with 1 µM angiotensin II (Ang II). The cardiac function was measured with the ultrasonic instrument. Histological analysis of mice’s hearts was determined with H&E or Masson method. The protein level of fibrosis markers, SUMOylation of SIRT1, and AKT/GSK3β pathway were quantified by immunofluorescence and western blot. Results EPI treatment (1 mg/kg/day) could reverse the TAC-induced decline in LVEF (TAC, 61.28% ± 1.33% vs. TAC + EPI, 74.00% ± 1.64%), LVFS (TAC, 28.16% ± 0.89% vs. TAC + EPI, 37.18% ± 1.29%). Meantime, we found that 10 µM EPI blocks Ang II-induced transformation of cardiac fibroblasts into myofibroblasts. The underlying mechanism of EPI-inhibited myofibroblasts transformation involves activation of SUMOylation of SIRT1 through SP1. Furthermore, SUMOylation of SIRT1 inhibited Ang II-induced fibrogenic effect via the AKT/GSK3β pathway. Conclusion EPI plays a protective effect on cardiac fibrosis by regulating the SUMO1-dependent modulation of SIRT1, which provides a theoretical basis for use in clinical therapies.


Introduction
Cardiac fibrosis, which can broadly be defined as the pathological transformation of fibroblasts into myofibroblasts, is a common pathophysiological process in cardiac remodelling and heart failure (Shah and Mann 2011;Braunwald 2013;Travers et al. 2016). Recently, there has been a potential treatment for inhibiting cardiac fibrosis by suppressing myofibroblasts transformation. (-)-Epicatechin (EPI) is an important chemical compound that exerts protective effects on the cardiovascular system, which is extracted from many flavanol-rich substances, including cocoa and green tea (Quine and Raghu 2005;Yamazaki et al. 2008Yamazaki et al. , 2010Taub et al. 2012). Continuous EPI treatment had a blood pressure-lowering effect in young male borderline hypertensive rats (Kluknavsky et al. 2020). Treatment with EPI has capable of reducing myocardial fibrosis in high glucose-stimulated cardiac fibroblasts (Garate-Carrillo et al. 2021). The above studies focussed primarily on the role of EPI in cardiac function; however, the specific action of EPI on cardiac fibrosis remains unclear, and the exact mechanism of the EPI regulating pathway in cardiac fibroblasts needs to be determined. SIRT1, a member of the sirtuin family of nicotinamide adenine dinucleotide (NAD þ )-dependent histone deacetylases, is not only involved in regulating oxidative stress and inflammatory but also played a protective role in organ fibrosis (Haigis and Sinclair 2010;Zeng et al. 2017;Shaikh et al. 2019). The activation of SIRT1 could prevent cardiac fibrosis induced by isoproterenol via the regulation of epithelial-to-mesenchymal transition . The expression of SIRT1 is controlled at both the transcriptional and post-transcriptional regulation levels; however, the post-translational mechanisms that regulate SIRT1 stability remain unclear.
Therefore, we aimed to evaluate the protective effect of EPI on cardiac fibrosis and elucidate the exact mechanism. Through establishing the cardiac fibrosis model in vivo and in vitro, we demonstrated an anti-fibrotic effect of EPI on cardiac fibrosis and the role of SUMO1-dependent modulation of SIRT1 participated in EPI inhibiting cardiac fibrosis.

Animal model and treatment
Male C57BL/6 mice weighing between 22-28 g and aged 8 weeks were fed through a freely available diet for 7 days of acclimatization under normal room temperature and humidity conditions. For all in vivo studies, male C57BL/6 mice were used with at least five mice per genotype. Mice were then randomly separated into sham/control, EPI (1 mg/kg/day), TAC, and TAC þ EPI (1 mg/kg/day) groups. The dose selection of EPI was confirmed by a previous study (Li et al. 2018). Tribromoethyl alcohol (20 mg/kg, i.p., MCE, USA) was used to anaesthetize mice. TAC mouse model was conducted as a previous study (Qin et al. 2015). Intragastric administration of EPI was conducted in mice for 28 days. Animal experiments were approved by the Institutional Animal Care and Use Committee of First Affiliated Hospital, Harbin Medical University (approval No. 2020099), and complied with the Guide for the Use and Care of Laboratory Animals published by the National Institutes of Health (NIH Publication No. 85-23, revised 1996).

Echocardiographic measurement
Echocardiography of cardiac structure and function was determined as previously reported (Luo et al. 2015). Echocardiographic measurements were conducted with ultrasonic echocardiographic instruments (Visualsonic Vevo 1100, Canada). The cardiac parameters of five mice were measured in each group after 28 days of TAC or treatment.

Histological analysis
Histological analysis of mouse hearts was determined with H&E or Masson method. The heart tissues were fixed in 4% paraformaldehyde for 2 days and then embedded in paraffin and sliced into 4 lm sections. H&E staining kit (Solarbio, Beijing, China) and Masson's trichrome stain kit (Solarbio, Beijing, China) were used to measure cardiomyocyte area and cardiac fibrosis. The myocardial sections were observed by microscope (Carl Zeiss Co. Ltd., USA). Collagen deposition was evaluated with collagen volume fraction (CVF). The heart weight/body weight ratio (HW/ BW) index was calculated using the mouse heart weight to body weight ratio.

Immunohistochemistry assays (IHC)
Mouse heart tissue sections were deparaffinized in xylene series and rehydrated through ethanol series. Sections were subjected to antigen retrieval in citrate buffer before the addition of SIRT1 primary antibody (Proteintech, Wuhan, China) overnight at 4 C.
The following day, the sections were incubated with horseradish peroxidase (HRP)-conjugated anti-mouse antibody and stained with diaminobenzidine (DAB). Thereafter, incubated with the 3,3 0 -diaminobenzidine solution for 10 sec. Images were taken with the microscope (Carl Zeiss Co. Ltd., USA).

Cell culture
Cardiac fibroblasts (CFs) and cardiomyocytes (CMs) were acutely isolated from neonatal SD rats (1-3-day-old). CFs, CMs, and H9C2 cells were cultured under routine conditions (37 C, 5% CO 2 ). The cells were divided into different groups and subjected to experimental procedures at 80% confluence. According to the experimental group, CFs, CMs, and H9C2 cells were treated with 1 mM Ang II, 10 mM EPI, and 0.05 mM MTM (SP1 inhibitor) for 24 h.

Co-immunoprecipitation (Co-IP)
CFs were lysed with the buffer containing a protease inhibitor after washing. The supernatants of cell lysates were obtained with centrifuging at 15,000 rpm at 4 C for 15 min, incubated with SIRT1 polyclonal antibody at 4 C for 3 h. Then, 30 mL magnetic beads (Biomake, TX, USA) were added to the samples at 4 C for 12 h. Co-IP proteins were eluted from the magnetic beads and immunoblotted with antibodies against SIRT1 or SUMO1.

Immunofluorescence assay
Briefly, CFs were stained with a-SMA antibody for 12 h. Cells were washed with PBS. Then, the secondary antibody (Molecular Probes, Invitrogen) was added for 1 h. DAPI was added for 3 min. Immunofluorescence was detected by microscope (Carl Zeiss, USA). The magnification is Â200.

Statistical analysis
Results were shown as the mean ± SEM and analysed with SPSS 20.0 statistical software. Difference comparisons were determined using one-way ANOVA and considered statistically significant when p< 0.05.

EPI prevents cardiac fibrosis
To examine whether EPI protects the heart against cardiac fibrosis, we first established the TAC mouse models. Then intragastric administration of EPI was conducted in mice for 28 days after established TAC models. Echocardiography was used to monitor cardiac function (Figure 1(A)). LVEF, LVFS, LVIDs, LVIDd, LVPWs, LVPWd, IVSs, and IVSd were detected (Figure 1(B-I)). Compared to the control group, TAC decreased LVEF, LVFS, and increased LVIDs, LVIDd, which were ameliorated by EPI treatment.
The protective effect of EPI on the TAC-induced cardiac structure of mice was determined. The results showed that collagen content was increased in the heart tissues of the TAC group, which was significantly reduced by EPI treatment ( Figure  1(J-L)). In addition, the administration of EPI substantially reduced HW/BW index in the TAC group (Figure 1(M)).

EPI blocks Ang II/TGF-b1/SMAD3-induced myofibroblasts transformation
To examine whether EPI protects CFs against Ang II-induced myofibroblasts transformation, we used primary cultured neonatal rat cardiac fibroblast for in vitro models of cardiac fibrosis. Cells were incubated using Ang II or/and EPI for 24 h. As shown in Figure 2(A,B), the a-SMA-positive area was significantly increased in the Ang II-treated group, and EPI inhibited Ang IIinduced fibroblast transformation.
To further evaluate the protective effect of EPI on cardiac fibrosis, protein of a-SMA, Collagen I/III (COL I/III), TGF-b1 and P-SMAD3 were detected by western blot (Figure 2(C-H)). As we expected, Ang II increased the protein expression of a-SMA and COL I/III, TGF-b1 and P-SMAD3. EPI efficiently inhibited these changes, suggesting that EPI could mitigate the myofibroblasts transformation and collagen synthesis induced by Ang II.

EPI activates SUMOylation of SIRT1
The SUMOylation of SIRT1 played a vital role in regulating cellular physiological functions (Han et al. 2016). To examine EPI whether protected the heart through SUMOylation of SIRT1, we firstly detected the SIRT1 and SUMO1 expression in mice heart tissues by IHC assay. Meanwhile, cells were incubated through Ang II or/and EPI for 24 h.
The IHC assay shows that SIRT1 and SUMO1 protein are expressed at lower levels in EPI and TAC þ EPI groups, whereas Control and TAC groups are weakly detected of SIRT1 and SUMO1 in heart tissues ( Figure 3(A-D)). We then examined whether SIRT1 is indeed SUMOylated in CFs. As shown in Figure 3(E-G), the Western Blot bands represented SUMOylated SIRT1, and EPI increased the SUMOylation of SIRT1. In addition, for further confirmation of the effect of EPI on SIRT1-SUMO1, the protein expression of SIRT1-SUMO1 was detected in H9C2, CMs, and CFs that incubated with EPI (Figure 3(H)). As shown in Figure 3(I-N), the results indicated that the protein expression of SIRT1-SUMO1 was increased in the EPI treatment group. The data indicated that EPI could increase SIRT1-SUMO1 protein expression in both H9C2, CMs and CFs.

SIRT1-SUMO1 participates in EPI-induced activation of AKT/GSK3b
We explored whether EPI affected the activation of AKT and GSK3b. As indicated in Figure 4(A-C), EPI increased the protein of P-AKT and P-GSK3b in CFs. These data suggested that EPI might activate the SIRT1-SUMO1/AKT/GSK3b pathway to inhibit cardiac fibrosis.
We explored whether SIRT1-SUMO1 indeed participated in regulating of AKT/GSK3b pathway. As shown in Figure 4(D-H), we used the small interfering RNAs approach to knockdown SIRT1, which blocked SIRT1 and SUMO1 proteins increased by EPI. As we know, SIRT1 could regulate the activation of GSK3b (Yang et al. 2020). We tested whether knockdown of SIRT1 could affect activation of AKT/GSK3b induced by EPI. As we expected, the knockdown of SIRT1 significantly decreased the protein level of P-AKT and P-GSK3b. These data illustrated that EPI-induced activation of AKT and GSK3b was mediated by SIRT1-SUMO1.
GSK3b, which was also required for EPI-inhibited myofibroblasts transformation.
Transcription factor SP1 is considered as a constitutive activator of housekeeping genes. EPI was known to regulate activity of transcription factor SP1 (Feng et al. 2019). As indicated in Figure 6(A-D), EPI could increase protein level of SP1 in both H9C2, CMs and CFs. EPI increased expression of SP1, SIRT1, SUMO1, P-AKT and P-GSK3b were consistent with above observation, MTM (SP1 inhibitor) completely blocked effect of EPI ( Figure 6(E-J)). Furthermore, CFs were preincubated with MTM which showed an obvious increase of the expression levels of a-SMA and COLI/III in CFs (Figure 7(A-F)). These data suggested that SP1 was required for EPI-induced SIRT1-SUMO1 activation.

Discussion
A diet rich in flavanols has a protective effect on cardiovascular health and function as an antioxidant and anti-inflammatory (Yamazaki et al. 2010;Wu et al. 2017;Kang et al. 2020;Leyva-Soto et al. 2021). EPI is extracted from the flavanol-rich plant and has been validated as a dietary supplement in mitigating vascular inflammation and oxidative stress. EPI treatment has been confirmed that can protect the heart against myocardial ischaemia induced-cardiac injury and protect the cardiovascular through reducing arginase expression and increasing NOS expression (Li et al. 2018;Ortiz-Vilchis et al. 2018;MacRae et al. 2019). Furthermore, EGCG treatment could ameliorate cardiac hypertrophy and fibrosis (Cai et al. 2013;Muhammed et al. 2018). In our present study, EPI was verified to improve heart function and reduce HW/BW ratio to protect the heart against cardiac fibrosis by the TAC mouse model. Furthermore, H&E and Masson staining results show the cardiovascular protective effect of EPI. The immunohistochemical analysis results show that the SIRT1 protein expression in the hearts of the TAC mice is upregulated by EPI treatment. EPI treatment decreased the protein level of cardiac fibrosis biomarkers such as COLI, COLIII, and a-SMA. Therefore, EPI blocked TAC or Ang IIinduced myofibroblasts transformation and collagen synthesis. SIRT1 could inhibit oxidative stress, inflammatory and attenuate cardiac fibrosis . SIRT1 was indeed activated by SUMOylation to stabilize its protein levels (Han et al. 2016). The SUMOylation of SIRT1 is involved in inhibition of cellular senescence and acted as a cardioprotective factor against MI/R injury (Yang et al. 2007;Han et al. 2016;Chen et al. 2019). In our study, we found that SIRT1 and SUMO1 participated in EPI-inhibited myofibroblasts transformation and protein levels were upregulated by EPI. The Co-IP results show that SIRT1 could be SUMOylated, and SUMOylation of SIRT1 participated in EPI-inhibited myofibroblasts transformation. Therefore, we found that EPI-inhibited myofibroblasts transformation might involve activation of SUMOylation of SIRT1. The magnification is Â200, scale bar ¼ 50 lm. Data are shown as mean ± SEM (n ¼ 3-6). ÃÃ p< 0.01 compared with control group; # p < 0.05, ## p < 0.01 compared with Ang II group; && p< 0.01 compared with Ang II þ EPI group; ᭝ p< 0.05, ᭝᭝ p < 0.01 compared with Ang II þ EPI þ Si-SIRT1 group.   6). Ã p < 0.05 compared with control group; # p < 0.05, ## p < 0.01 compared with Ang II group; & p < 0.05, && p < 0.01 compared with Ang II þ EPI group.
In addition, the transcription factor SP1 can function in the transcriptional regulation of many genes in positive or negative ways (Sch€ afer et al. 2003;Chou et al. 2005). Previous studies have revealed that oxidation of SP1 was an essential element for the transcriptional regulation of SUMO2 and SUMO3 (Sang et al. 2011). Our research showed that EPI increased protein expression of SP1, SIRT1-SUMO1 in CFs. We also observed that MTM blocked the activation of SIRT1-SUMO1 induced by EPI. Furthermore, MTM blocked the inhibition of EPI on myofibroblasts transformation and collagen synthesis in vitro. The results indicated that SP1 participated in EPI-induced SIRT1-SUMO activation.
Previous studies found that GSK3b mediated the regulation of several signalling proteins (MacAulay and Woodgett 2008; Zhou et al. 2016). GSK3b also participated in activating CFs and regulated cardiac remodelling function (Liu et al. 2016). AKT/GSK3b pathway played a crucial role in regulating cell proliferation and apoptosis (Dai et al. 2016). Furthermore, activating SIRT1 and AKT signalling could prevent cardiac fibrosis in post-myocardial infarction (Jia et al. 2019). The western blot results show that knockdown SIRT1 significantly decreased the protein level of P-AKT and P-GSK3b and blocked EPI-inhibited myofibroblasts transformation in CFs. These data indicated that SIRT1-SUMO1 was involved in EPI-induced activation of AKT/GSK3b, which was also required for EPI-inhibited myofibroblasts transformation in CFs. In our study, Ang II did not affect SIRT1-SUMO1/ AKT/ GSK3b pathway. The previous study indicated that Ang II could induce myofibroblasts transformation through the TGFb1-SMAD3 pathway (Zhou et al. 2017). Moreover, the activation of GSK3b could inhibit the TGFb1-SMAD3 pathway (Lal et al. 2014). So, our results indicate that EPI-induced activation of AKT/GSK3b inhibits the Ang II/TGFb1-SMAD3 pathway for reducing myofibroblasts transformation. Our data indicated that EPI not only increased expression of SUMO1, but also increased the expression of SIRT1. The follow-up mechanism needs to be determined.

Conclusions
Our study indicates that EPI can prevent cardiac fibrosis. SUMO1-dependent modulation of SIRT1 mediates EPI-inhibited cardiac fibrosis. These findings will supply a new agent and mechanism of action for treating cardiac fibrosis in the future.

Disclosure statement
The authors declare that there are no competing interests.

Data availability statement
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Funding
The study was funded by the National Natural Science Foundation of China [No. 81870169, 81400250, 81900366], the Natural Science Foundation of Heilongjiang Province [H2017040], Postdoctoral initiation fund of Heilongjiang Province [LBH-Q19032].