Transcription factor EB (TFEB) improves ventricular remodeling after myocardial infarction by inhibiting Wnt/β-catenin signaling pathway

Introduction

Acute myocardial infarction (AMI) is caused by hypoxia and ischemia, with high morbidity and mortality (Bai et al., 2021). AMI often leads to heart failure (HF), which is the major risk patients have to face (Bahit, Kochar & Granger, 2018; Velagaleti et al., 2008). Therefore, effective treatment is needed to reduce the size of myocardial infarction (MI), preserve left ventricular (LV) function, and prevent HF in patients with AMI. After AMI, extensive myocardial injury, impaired myocardial contractility, continuous activation of the neuroendocrine system, and remodeling of extracellular matrix (ECM) occur in the heart, which results in left ventricular remodeling (LVR). Adverse left ventricular remodeling affects cardiac function and increases the risk of HF (Azevedo et al., 2016). Nowadays, myocardial remodeling is recognized as a complex process in response to cardiac overload and loss of functional myocardium, resulting in structural and functional disorders of the myocardium (Van der Bijl et al., 2020).

The post-AMI cardiac response can be divided into three phases: (1) a proinflammatory phase (days 1–3 after MI) aimed at removing cellular debris from the ischemic infarct zone. In the ischemic environment, myocardial cells undergo anaerobic metabolism, cell membrane is unstable, and cells undergo apoptosis, autophagy and necrosis (Fu et al., 2020; Heusch & Gersh, 2017). Neutrophils and macrophages lead to destruction of the extracellular collagen matrix (ECM) and enlargement of the infarct area, thereby changing the shape of the ventricle, and the infarct myocardium becomes thinner and dilated. (2) During the repair period (4–7 days after MI), acute inflammatory mechanisms are down-regulated and myocardial injury is alleviated, while wound healing and scar formation are performed to prevent cardiac rupture. After the inflammatory response, fibroblasts are directed to the infarct area, where they produce new collagen matrix and form scar tissue (Bhatt, Ambrosy & Velazquez, 2017; Gabriel-Costa, 2018; Sutton & Sharpe, 2000; Xie, Burchfield & Hill, 2013). (3) At the mature stage (7 days after MI), the non-infarcted cardiomyocytes became hypertrophic and the ECM changed. The cardiac ECM is a highly organized structural and functional protein network that surrounds cardiomyocytes and generates a cellular scaffold that maintains LV shape (Iyer et al., 2014). If inflammation and fibrosis-related signals are continuously activated, it may lead to adverse left ventricular remodeling (Ong et al., 2018; Ruparelia et al., 2017; Westman et al., 2016).

After MI, fibroblasts will proliferate and differentiate into myofibroblasts. Myofibroblasts express the contractile proteins α-smooth muscle actin (α-SMA) and embryonic smooth muscle myosin, exhibit an extended endoplasmic reticulum, and secrete abundant matrix proteins to generate collagen scars (Frangogiannis, Michael & Entman, 2000). Fibroblasts early activation and late remodeling is important for cardiac function (Van Nieuwenhoven & Turner, 2013). Adverse fibrosis will lead to myocardial stiffness, diastolic and systolic dysfunction, and eventually the development of HF (Prabhu & Frangogiannis, 2016).

TFEB is one member of the microphthalmia-associated transcription factor E family and is actively involved in many cellular biological and pathological processes (Rehli et al., 1999). TFEB studies showed that under starvation or stress conditions, TFEB dephosphorylated and accumulated in the nucleus, recognized and bound to genes sequence, upregulated genes expression at the transcriptional level involved in lysosome and autophagosome biogenesis, as well as phagocytosis and various immune responses (Godar et al., 2015; Ma et al., 2015; Medina et al., 2011; Palmieri et al., 2011; Settembre et al., 2011). Until now, whether TFEB was involved in the fibrosis process after MI remains unclear. Thus, we designed this study to systematically evaluate the impacts of TFEB on the pathology processes of autophagy, ventricular remodeling, and fibrosis after MI.

The Wnt signaling is involved in cell proliferation and differentiation progress, and is necessary for cardiac myocyte formation (Chen et al., 2001; Moon et al., 2002). Extracellular Wnts bind to transmembrane receptor complexes. The destruction complex consisting of β-catenin, Axin, APC (adenomatous polyposis coil), CK1 (casein kinase1) and GSK3 (glucogen synthase kinase 3 beta), has been relocated to the plasma membrane. The β-catenin is free from the destruction complex and accumulates within the cytosol. And this leads to stabilisation and translocation of β-catenin into the nucleus, where it binds to T-cell factor (TCF) and the lymphoid enhancer factor (LEF), and activates various target genes (Clevers, 2006; Moon et al., 2004). Studies show that the Wnt signaling pathway is involved in the fibroblast activation and proliferation during cardiac fibrosis (Duan et al., 2012; Honda et al., 2013; Sullivan & Black, 2013).

Here, we showed that TFEB, a transcription factor related to autophagy and lysosomal biogenesis, is involved in the progression of fibrosis. We also found that the inhibition effect of TFEB in fibrosis was mediated by the formation TFEB- β-catenin-TCF/LEF1 complex, which changed the gene expression profile of β-catenin.

Materials and Methods

Animals and MI modeling

We used male mice to establish a stable MI model considering that estrogen has shown a protective effect against pathological hypertrophic remodeling in pressure-overload (Cao et al., 2011; Patten et al., 2004). Two-month-old wild-type male C57BL/6 mice (20-30g) were purchased from the Animal Experimental Centre of Jicui Yaokang Biotechnology Co. LTD. (Jiangsu, China). All mice were fed in the SPF animal laboratory at the Animal Center of Sun Yat-Sen University, Guangzhou, China, with free access to standard laboratory food and water. 142 mice were used to construct MI models and 130 mice were included. 12 mice were excluded because of technical failure. 130 mice were randomized into five groups: the sham group, AAV9-TFEB group, AAV9-NC group, AAV9-shTFEB group, and the AAV9-shNC group, with random allocation software by Dr. Cong Liu. One, three, five, nine and six mice of each group died after MI. The AAV9-NC and AAV9-sh-NC group served as controls of the AAV9-TFEB and AAV9-sh-TFEB groups, respectively. Every mouse was pretreated with myocardial multipoint injection of Adeno-associated virus 9 (AAV9) or normal saline for two weeks before the MI modeling. After 14 days of feeding, the anterior descending branch of the mice’s left coronary artery (LAD) was permanently ligated to set up the MI model. The surgery was conducted under the anesthetization of 50 mg/kg intraperitoneally injected pentobarbital sodium (Sigma). For mice in the sham group, the chest was surgically opened without LAD ligation. Echocardiography was performed every time point. The mice were anesthetized and sacrificed by cervical dislocation at different time points after MI or sham operation. Hearts were collected and stored at −80 °C for the next step measurements. The flowchart of this experiment is illustrated in Fig. 1A. Each test was repeated with at least three mice in each group. Each test was performed by the same person.

Results

TFEB affected the differentiation of fibroblasts and extracellular matrix synthesis and transformation in vivo

Two to four days after modeling, fibroblasts activated by the stimulation of inflammatory cytokines began to proliferate and produce ECM (Kurose, 2021). As revealed by the Sirius red staining, AAV9-shTFEB group had a higher concentration of collagen I and III than the AAV9-shNC group (Fig. 2A).

The effects of TFEB on cardiac fibroblasts during fibrosis model in vitro

TGF-ß1 was used to simulate fibroblast proliferation and its differentiation into myofibroblasts. CFs cells were co-incubated with TGF- β1 of different concentrations including 2 ng/mL, 5 ng/mL, eight ng/mL, 10 ng/mL, and 20 ng/mL for different duration including 6, 12, 24, 36, and 48 h. CCK-8 results indicated that low concentration TGF- β1 promoted CFs proliferation, while high concentrations weakened the promotion effect (Fig. 3A). Western blot showed that α-SMA elevated obviously with 5ng/ml TGF- β1 (Fig. 3B). Ad-TFEB group showed lower proliferation rate than the Ad-Gfp group in the CCK-8 test after 12 h of co-incubation with 5ng/ml TGF- β1 (Fig. 3C). Transwell and Wound healing results showed the Ad-TFEB group had a lower cell migration than the Ad-Gfp group (Figs. 3D–3E and Fig. S5). Western blot results indicated that 12-hour TGF- β1 stimulation increased collagen I expression. The Si-TFEB group had a higher collagen I expression than the Si-NC group (Fig. 3F).

TFEB relocated to nucleus and made connection with Wnt pathway by β-catenin-TCF/LEF1 complex

Previous studies have provided that the Wnt signaling is involved in cell proliferation and differentiation progress, and is necessary for cardiac myocyte formation (Chen et al., 2001; Moon et al., 2002). Immunofluorescent staining suggested that TFEB didn’t affect the concentration of β-catenin. High resolution confocal microscopy analysis showed that colocalization between TFEB and β-catenin in the nucleus. By inhibiting the β-catenin-TCF/LEF1 complex formation, PNU-74654 decreased the concentration of β-catenin in the nuclear (Fig. S7A and Figs. 4A–4B). Lamin-B1 and GAPDH were used as an internal control of nucleus and cytoplasm in Western blot. IF and Western blot showed that TFEB relocated to nuclear in fibrosis model (Figs. S6A–S6B). According to results from CCK8, Transwell, wound healing and western blot assays, the enhancements of cell migration, proliferation and collagen I expression by inhibiting TFEB was prevented by PNU-74654 (Figs. 4C–4F and Fig. S7B). TFEB exerts its anti-fibrotic effect probably by inhibiting Wnt signaling pathway, which has been shown to promote fibrosis (Fig. 5).

Discussion

Cardiac repair after MI consists of three phases: a pro-inflammatory phase, an anti-inflammatory repair or proliferative phase, and a maturation phase (Ong et al., 2018; Ruparelia et al., 2017a; Westman et al., 2016). This inflammatory process is usually destructive and leads to excessive death of surviving cardiomyocytes, affecting the final infarct size (Zhao et al., 2001). Increased TFEB transcription activated by metformin and cilostazol could protect against I/R injury by regulating autophagy, lysosome, and apoptosis (Li, Xiang & Xu, 2019; Wang et al., 2019). Javaheri et al. (2019) noticed that macrophage-specific over-expression of transcription factor EB (M ϕ-TFEB) expression could improve ventricular function after IR injury, and TFEB in macrophages played a role in ventricular remodeling after MI by mediating the inflammatory response. In this study, we verified that TFEB affected left remodeling after MI, demonstrating that TFEB alleviated infarction extension and protected the systolic function of the heart (Figs. 1C–1F). WGA staining showed that TFEB down-regulation was associated with severe cardiac hypertrophy (Fig. 1G). This implies the possibility of using TFEB to protect cardiac function in human MI patients.

As the final stage of MI repair, the maturation phase is associated with remodeling of the ECM which commonly lasts for several months. Scar maturation is a process intertwining the reduction in infarct fibroblast numbers (Frangogiannis, Michael & Entman, 2000), the differentiation of fibroblast into myofibroblast, the apoptosis of activated fibroblasts, the and expression of matrix-specific proteins (Fu et al., 2018). The purpose of scarring is to prevent myocardial rupture and deterioration of partially restricted cardiac function, with thinning and dilation of the infarcted area and hypertrophy of the other areas (Ong et al., 2018). Thus, LV remodeling predicts a poor clinical prognosis. The main pathological features of ventricular remodeling include extensive fibrosis, pathological cardiomyocyte hypertrophy, and cardiomyocyte apoptosis. The balance between excessive synthesis and degradation of myocardial fibrotic collagen is critical for maintaining myocardial ECM homeostasis. Two to four days after injury, fibroblasts activated by the stimulation of inflammatory cytokines began to proliferate and produce ECM (Kurose, 2021). The transformation from fibroblasts to myofibroblasts mainly occurred four to seven days after MI (Fu et al., 2018). Myofibroblasts are characterized by the extensive endoplasmic reticulum, the expression of α-smooth muscle actin (α-SMA), and the synthesis of matricellular proteins (Hinz, 2010). Thus, we detected the myofibroblasts by α-SMA staining. In our study, TFEB inhibited fibroblast differentiation into myofibroblasts as early as three days after MI (Fig. 3A). TFEB inhibited collagen I expression in the fibrosis model in vivo (Fig. 3F). Four weeks after MI, TFEB over-expression decreased collagen III synthesis in mice (Fig. 2A). In the context of myocardial fibrosis, TFEB inhibited cell migration and collagen I concentration at the cellular level (Figs. 3D–3F). Unveiling the impacting mechanisms of TFEB in the fibrosis process requires further investigations.

Previous studies have provided that the Wnt signaling is involved in cell proliferation and differentiation progress, and is necesssary for cardiac myocyte formation (Chen et al., 2001; Griffin et al., 2022; Moon et al., 2002; Zhang et al., 2022). In MI model, inhibition of Wnt signaling was shown to reduce collagen concentration and improve cardiac function (Barandon et al., 2011; Fan et al., 2018; Laeremans et al., 2011). The Wnt proteins have been mainly implicated in the promotion of cardiac fibroblast proliferation and collagen expression process (Laeremans et al., 2011).

In recent years, many studies have reported the role of Wnt signaling in cardiac fibrosis in various animal models (Deb, 2014). Classical Wnt/ β-catenin signaling leads to epicardial fibrosis in allogeneic heart grafts, and increased activation of β-catenin and TCF/LEF is observed in transplanted human hearts (Ye et al., 2013). Acute ischemic cardiac injury can up-regulate Wnt1 expression, which is initially expressed in the epicardium and subsequently expressed by cardiac fibroblasts in the injured area, and Wnt1 induces proliferation of cardiac fibroblasts and expression of profibrotic genes (Duan et al., 2012; Von Gise & Pu, 2012). The use of Dishevelled (DVL) to inhibit the overexpression of GSK-3 β and activate canonical and non-canonical Wnt signaling pathways can induce spontaneous myocardial fibrosis and cardiac hypertrophy (Malekar et al., 2010). Meanwhile, Wnt pathway antagonists have also been used to alter the prognosis of fibrosis. secreted frizzled-related protein (sFRP) is a soluble protein with a structure highly homologous to the Fz receptor of Wnt signaling and is a commonly used antagonist of Wnt pathway (Cruciat & Niehrs, 2013; Sklepkiewicz et al., 2015). In myocardial infarct-related studies, inhibition of the Wnt pathway using sFRP1, sFRP2, or sFRP4 was shown to reduce fibrosis and improve cardiac function (Barandon et al., 2011; Fan et al., 2018; He et al., 2010; Laeremans et al., 2011; Matsushima et al., 2010). Mice lacking sFRP1 exhibit increased expression of Wnt ligands, β-catenin, α-SMA, and collagen (Sklepkiewicz et al., 2015). Expression of sFRP2 in cardiac fibroblasts activates Wnt/ β-catenin signaling and promotes proliferation and expression of ECM genes in fibroblasts (Lin et al., 2016). In contrast, SFRP2-deficient mice produced less collagen and had less fibrosis in cardiac fibroblasts after MI (Kobayashi et al., 2009). By inhibiting GSK-3 β in cardiac fibroblasts and activating the classical Wnt pathway, fibrogenesis in the infarcted heart can be promoted (Lal et al., 2014). In addition, the development of post-inflammatory fibrosis was successfully prevented by the administration of sFRP2 in a mouse model of autoimmune myocarditis (Blyszczuk et al., 2017). In a model of myocardial fibrosis induced by aortic coarctation, inhibition of β-catenin in cardiac fibroblasts reduced interstitial fibrosis without changing the number of activated cardiac fibroblasts (Xiang, Fang & Yutzey, 2017).

Another significant finding of this study is the identification of TFEB as an important regulator of the Wnt signaling. TFEB is a well-known master regulator of autophagy and lysosomal biogenesis processes (Du et al., 2022; Evans et al., 2022; Godar et al., 2015; Ma et al., 2015). In the process of fibrosis modeling, TFEB and β-catenin were observed to colocalize within nucleus in fibrosis modeling (Fig. 4A). By inhibiting the β-catenin-TCF/LEF1 complex formation, PNU-74654 decreased the concentration of β-catenin in the nuclear (Fig. 4B). Our findings indicated that the concentration of β-catenin was not influenced by TFEB. Moreover, PNU-74654 was observed to counteract the increase in cell migration, proliferation, and collagen I expression that resulted from the inhibition of TFEB (Figs. 4A–4F). The nuclear localized TFEB forms a TFEB- β-catenin-TCF/LEF1 complex to induce the transcription of genes, that being distinct from previously known regulated by the β-catenin-TCF/LEF1 complex (Kim et al., 2021).The very first possibility is that the inhibition effect of TFEB in fibrosis was mediated by the formation TFEB- β-catenin-TCF/LEF1 complex, which changed the gene expression profile of β-catenin (Fig. 5). It is possible that the influence of TFEB on fibroblasts is mediated via the Wnt pathway. Our study provides evidence that TFEB’s anti-fibrotic activity is likely achieved through the suppression of the Wnt signaling pathway, a mechanism known to foster fibrosis However, more studies may be needed to clarify other pathways through which TFEB may influence.

Supplemental Information