FGF6 promotes cardiac repair after myocardial infarction by inhibiting the Hippo pathway

Abstract Objectives Myocardial infarction (MI) commonly occurs in patients with coronary artery disease and have high mortality. Current clinical strategies for MI still limited to reducing the death of myocardial cells but failed to replace these cells. This study aimed to investigate the role of fibroblast growth factor 6 (FGF6) in enhancing the proliferative potential of cardiomyocytes (CMs) after ischemic injury via the Hippo pathway. Materials and Methods Expression of FGF6 protein was analysed in mice with MI induced by ligation of the left anterior descending coronary artery. Activation of the Hippo pathway and the proliferation potential were examined in ischemic CMs, treated with FGF6 protein or transfected with an adeno‐virus carrying FGF6 sh‐RNA. Immunofluorescence staining and western blotting were performed to assess the relationship between FGF6 and the Hippo pathway. Results We found that FGF6 expression was significantly increased in the MI mouse model. Knockdown of FGF6 synthesis resulted in poorer heart function after MI. By contrast, treatment with recombinant human FGF6 protein improved heart function, reduced infarct size, and promoted cardiac repair. Additionally, FGF6 restrains the activation of the Hippo pathway and subsequently promotes nuclear accumulation of YAP. This was largely counteracted by treatment with extracellular signal‐regulated kinase 1/2 (ERK1/2) inhibitor U0126. Conclusion FGF6 inhibits the Hippo pathway via ERK1/2, and facilitates nuclear translocation of YAP, and thereby promotes cardiac repair after MI.


| INTRODUCTION
MI is one of the most common cardiac vascular diseases worldwide.
Due to that MI causes loss of many CMs, consequently, morbidity and mortality rates following MI remain high. 1 Interventional therapy is the most common and standard care for patients with MI. Although the number of patients with fatal MI has declined, interventional therapy cannot replace cell death upon MI. 2,3 As a results, the loss of CMs eventually leads to a decline in left ventricular (LV) function and contributes to an increase in the prevalence of heart failure. 4,5 The inability to replace lost CMs is the largest obstacle in the treatment of MI. To overcome this, tremendous interest in strategies that increase the number of CMs by stimulating re-entry of CMs into the cell cycle, survival of CMs via inhibition of apoptosis, and differentiation of CMs from stem cells have been explored. 6,7 Nevertheless, current clinical strategies for MI reduce the death of myocardial cells, but fail to replace these cells. 7 Therefore, novel strategies to promote cell cycle re-entry in CMs must be developed.
Regenerative therapies are conceptually different, because they aim to rebuild the myocardium rather than salvage tissue. 8 In contrast to organs such as the liver and skin, the heart possesses only a minimal regenerative capacity. [9][10][11] It lacks a progenitor cell population, and CMs exit the cell cycle shortly after birth and few of these cells re-enter the cell cycle after injury. 12,13 Thus, any loss of CMs is essentially irreversible and can lead to or exaggerate heart failure, which is a major public health problem. 14,15 New therapeutic options are urgently needed; however, regenerative therapies in cardiovascular medicine are only in their infancy.
The fibroblast growth factor (FGF) family consists of 23 members, which can promote cell proliferation. 16 For example, FGF6 is a critical component of the muscle regeneration process in mammals. 17 It belongs to a family of cytokines that controls cell proliferation, cell differentiation, and morphogenetic events. Several lines of evidence demonstrate that FGF6 exhibits a restricted expression profile and is predominantly expressed in the myogenic lineage. 18,19 Floss et al. found that FGF6 can completely restore experimentally damaged skeletal muscle, 20 leading us to speculate that it may elicit similar in the heart, which is a muscle tissue too. To investigate the effect of FGF6 on MI, we first examined its expression in the heart after permanent ligation of the left anterior descending coronary artery and confirmed that FGF6 expression was increased in mice with MI. Then, we investigated the therapeutic effect of FGF6 on MI and found that FGF6 alleviated MI by promoting cardiac repair.
Acting as a major downstream effector of the Hippo signalling pathway, YAP plays a critical role in controlling organ size. 21 Accumulating evidence demonstrates the detrimental role of the Hippo/YAP pathway in regulating the division of CMs and the regeneration of damaged myocardium through multiple transcriptional mechanisms. 22,23 Accordingly, in this study, we demonstrated that the cardioprotective effect of FGF6 is mainly mediated by the Hippo/YAP pathway and that FGF6 triggers cell cycle re-entry of CMs and promotes myocardial repair via this pathway. Our observations raise the possibility that targeting FGF6 can facilitate CMs cell cycle re-entry and promote cardiac repair after injury in mammals. University and performed in accordance with the Guide for the Care and Use of Laboratory Animals. Animals were housed at 23 ± 2 C with humidity of 50 ± 5%. Mice were randomly divided into five groups (n >5 for each group): sham group, MI group, MI + FGF6 group, MI+ Ad-cTNT-sh-FGF6 group, and MI + FGF6 + verteporfin (VP) group. MI was established in male mice. The adult mice received tracheal intubation and were ventilated with 3% isoflurane for induction and 2% isoflurane for maintenance of anaesthesia. Then, the left anterior descending coronary artery was ligated with 7-0 prolene suture, and 25 μL Ad-cTNT-sh-FGF6 (8.69 Â 10 10 PFU/mL) or Ad-cTNT-sh-Con was injected into the myocardium following MI. After completion of the surgery, the chest was closed. Mice were then warmed for several minutes until recovery. The mice in the recombinant human FGF6 protein (10 μg/kg/day, R&D, 238-F6-025) treatedgroup were intraperitoneally administered into mice once a day for 2 weeks, while mice in the PBS group were injected with PBS buffer. VP (20 mg/kg/day, MCE, 129497-78-5) was intraperitoneally administered into mice once a day for 1 week, other groups was injected corn oil once a day for 1 week before sacrifice. After 2 weeks of MI, the mice were anaesthetized and sacrificed for collection of heart tissues and follow-up examinations.

| Western blot and antibodies
The equal amounts (30 μg) of protein lysed from heart tissue and cardiomyocytes, were separated by SDS-PAGE and then transferred onto polyvinylidene fluoride (PVDF) membranes (Merck Millipore, IPVH00010).
Next, membranes were blocked with 5% BSA in Tris-buffered saline con- β-Actin (CST, 4970) was used as the internal reference to normalized protein expression levels.

| TTC stain
After MI, the animals were anaesthetized and intubated, and the chest was opened. Hearts were excised, and LVs were sliced into 1-mm-thick cross sections. The heart sections were then incubated with a 1% TTC (Aladdin, T130066-5 g) solution at 37 C for 15 min.
The infarct area (white), and total LV area from both sides of each section were measured using ImageJ software (NIH), and the values obtained were averaged. The percentages of each section infarct area were multiplied by the weight of the section and then totalled from all sections. Infarct area/LV were expressed as percentages.

| Echocardiographic analysis
Echocardiographic examination was performed after MI or Sham surgery for 2 weeks. The mice were firstly anaesthetized with isoflurane (1-1.5% for maintenance) mixed in 1 L/min O 2 via a facemask, and meanwhile maintained normal breathing. Next, the parameters of cardiac function were evaluated by long-axis M-mode echocardiography using small animal ultrasound system (Vevo 2100, Canada) with a linear 30-MHz transducer as described in detail. The LV ejection fraction (EF) and fractional shortening (FS), the indicators of cardiac function, were calculated and averaged from at least three consecutive cardiac cycles. All of these measurements were performed by a single experienced technician in a blinded manner.

| Immunofluorescence
Heart section (5 μm) were subject to deparaffinization and rehydration followed by antigen retrieval by heating the slides in 10 mM

| EdU assay
For NRCMs, the no glucose medium was replaced with no glucose medium containing 10 μM EdU (Beyotime, C0075) 3 h after OGD, and cells were fixed 3 h later. For EdU assay adult mice cardiomyocytes, a dose of EdU (50 mg/kg/day, Beyotime, ST067) per animal was injected intraperitoneally once a day for 3 days. For EdU staining, the heart sections and cells were incubated with Click™ EdU 555 Imaging Kit reagents (Beyotime, C0075) to reveal EdU incorporation according to the manufacturer's instructions.

| Histological analysis
Heart tissue was fixed in 4% paraformaldehyde solution for 24 h and embedded in paraffin after dehydration using 70%-100% ethanol.
The heart paraffin blocks were cut transversely into 5 μm sections.
After deparaffinization, the heart sections subjected to Masson's trichrome staining were performed according to instruction book. To

| Real-time quantitative PCR
Total RNA was extracted from heart tissue and cardiomyocytes using TRIzol reagent (Takara Bro Inc, 9108), as described by the manufacturer's instructions. The RNA samples (1 ng) was reversely transcribed to cDNA by the Hiscript ® III Reverse Transcriptase kit (Vazyme, R223-01).

RT-qPCR analysis was performed on a QuantStudio™ 3 Real-Time PCR
Detection System using ChamQ Universal SYBR qPCR Master Mix (Vazyme, Q711-02) with specific primers. The relative expression levels of each gene were quantitated using the 2 ÀΔΔCT method and normalized to the amount of endogenous glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The sequences of specific primers used for RT-PCR in this study are listed in Table S1.

| FGF6 expression is increased after MI
To explore the change in FGF6 expression after MI, we first examined the level of FGF6 in hearts of mice subjected to permanent ligation of the left anterior descending coronary artery for 2 weeks or to a sham operation. Western blotting showed that FGF6 expression was significantly increased after MI ( Figure 1A), especially, FGF6 was mainly expressed at infarct area (IA) rather than border zone (BZ) or remote area (RA) ( Figure S1A). Meanwhile, RT-PCR analysis demonstrated that MI significantly increased the mRNA level of FGF6 ( Figure 1B). Consistently immunofluorescence staining showed that FGF6 expression was upregulated in hearts after MI and from CMs ( Figure 1C), and this results was further confirmed it ( Figure S1B). Similarly, in vitro immunofluorescence staining of cardiac troponin T (c-TNT) the CMs marker, and FGF6 demonstrated that OGD mainly elevated expression of FGF6 in primary neonatal rat cardiac myocytes (NRCMs), but not in other cell type ( Figure 1D). At the same time we found FGF6 was markedly up-regulated in NRCMs after 6 h of oxygen-glucose deprivation (OGD) treatment as well ( Figure 1E). In addition, immunofluorescence results and western blotting demonstrated that OGD mainly elevated expression of FGF6 in NRCMs, not in cardiac fibroblasts. Nevertheless, the expression of FGF6 was pretty low in both cells under normal conditions ( Figure 1F). In parallel, the mRNA level of FGF6 significantly increased after OGD treatment ( Figure 1G). Thus, FGF6 was specifically upregulated in CMs after MI.

| FGF6 contributes NRCMs cell cycle re-entry by inhibiting the Hippo pathway
The Hippo/YAP pathway plays a critical role in cardiac repair. Thus, we investigated whether the cardioprotective effect of FGF6 is dependent on Hippo signalling. Immunoblotting showed that OGD decreased the protein level of YAP and increased phosphorylation of YAP ( Figure 5A).
Meanwhile, OGD also elevated the phosphorylation levels of MST1 and LATS1, other components of the Hippo pathway. By contrast, FGF6 treatment largely counteracted OGD induced activation of Hippo signalling, as reflected by decreased phosphorylation of YAP and its upstream kinases LATS1 and MST1. In addition, the elevated level of p-YAP under OGD conditions was further increased by si-FGF6 treatment, and the total YAP protein level was the lowest between all the groups ( Figure 5A). To our interesting, the mRNA level of YAP did not significantly differ between the groups ( Figure 5D), indicating that FGF6 affects the level of YAP by blocking its degradation. Overall, these results indicate that FGF6 increases the level of YAP protein in CMs by inhibiting OGD induced activation of the Hippo pathway.
To investigate the role of YAP in ischemic heart repair, the level of YAP was measured in the nuclear fraction. The nuclear level of YAP was lower in the OGD treated group than in the control group and was further decreased by si-FGF6 treatment, while FGF6 treatment significantly counteracted the negative effect of OGD on nuclear accumulation of YAP ( Figure 5B). Moreover, immunofluorescence staining showed that FGF6 treatment promoted nuclear translocation of YAP ( Figure 5C). Next, we analysed the CTGF, a target gene of YAP, by RT-PCR to confirm the transcriptional regulatory effect of YAP. As expected the mRNA level of CTGF correlated with nuclear accumulation of YAP ( Figure 5D). In addition, western blotting provided evidence that FGF6 treatment promoted expression of CTGF and CYR61 via YAP ( Figure 5A).
In summary, these results suggest that FGF6 inhibits the Hippo pathway and subsequently facilitates nuclear accumulation of YAP to activate downstream effector, and thereby promotes CMs cell cycle re-entry under ischemic conditions.

| FGF6 promotes cardiac repair by inhibiting the Hippo pathway in vivo
To confirm the influence of FGF6 on myocardial injury in vivo, wild-type mice were subjected to MI by permanently ligating the  Figure 6A). Next, we sought to determine the subcellular distribution of YAP in response to different treatments after MI. Western blot analysis of fractionated CMs protein samples ( Figure 6B). Immunofluorescence staining showed that nuclear YAP accumulation was decreased in MI mice and further reduced by knockdown of FGF6, but was remarkably elevated by

| DISCUSSION
In this study, we revealed that FGF6 protects against MI injury by inhibiting the Hippo pathway. FGF6 promotes nuclear accumulation of YAP and thereby, facilitates CMs cell cycle re-entry and cardiac repair ( Figure 9). These findings provide new clues and ideas for developing potential methods to treat MI and for promoting myocardial repair.
During MI, many CMs die after oxygen deprivation, resulting in progression of life-threatening heart failure. 24 Although CMs could proliferate to generate new CMs in the adult heart, their slow turnover rate is insufficient to compensate for the significant cell loss after MI. 25,26 Accordingly, we are exploring ways to improve the efficiency of CMs proliferation. Several lines of evidence indicate that CMs are not all equivalent and their proliferation is not randomly distributed.
For instance, mononucleated CMs were reported to have a greater proliferative potential than multinucleated CMs, while pathological enlargement of cells reduces the likelihood of cell proliferation. 27,28 In this study, we found that FGF6 increased expression of Cyclin D1 and Cyclin E1, which are important to promote the transition from G1 to S phase in CMs. [29][30][31] FGF6 expression is stimulated after skeletal muscle injury and can promote skeletal muscle regeneration, whereas interference of FGF6 can induce a severe regeneration defect with enhanced fibrosis and myotube degeneration. 32 There is emerging evidence that the Hippo pathway is a key player in a wide range of biological processes during growth and development of tissues and organs. 12 Elevated expression of YAP, which is negatively regulated by the Hippo pathway, contributes to cardiac repair. Under normal conditions, YAP is phosphorylated and thus sequestered in the cytoplasm by 14-3-3 and inhibited. When the Hippo pathway is inactivated, YAP is dephosphorylated and F I G U R E 8 FGF6 inhibits the Hippo pathway via ERK1/2. (A) Western blot was performed to determine the protein levels of p-ERK1/2, ERK1/2, p-MST1, MST1, p-LATS1, LATS1, p-YAP and YAP in NRCMs in different group. Quantitative analysis of expressions of p-MST1, MST1, p-LATS1, LATS1, p-YAP, YAP, CTGF and CYR61. n >3 per group. (B) Western blot analysis the subcellular localization of YAP in NRCMs in different group. n = 3 per group. Quantitative analysis of expressions of YAP in nuclear extracts. (C) Representative immunofluorescence (green for YAP, blue for DAPI) analysis subcellular localization of YAP in NRCMs in different group. n = 5 per group. Data represent means ± SEM. Two-tailed Student's t test. *p <0.05, **p <0.01, ***p <0.001 F I G U R E 9 Schematic illustration of the protective effects of FGF6 on cardiomyocytes under ischemic conditions. Myocardial infarction would cause the loss of a huge number of cardiomyocytes. In this study, we revealed that FGF6 restrains the Hippo pathway via ERK, and promotes YAP nuclear translocation then increase cardiomyocytes cell cycle re-entry and cardiac repair translocates into the nucleus, where it binds to TEAD transcriptional co-activators and thereby stimulates gene expression, which contributes to cell survival, migration, and proliferation. 21,35 In this study, we demonstrated that FGF6 treatment elevated expression and nuclear accumulation of YAP in ischemic CMs, without altering the YAP mRNA level, suggesting that degradation of YAP was restrained.
We noticed that FGF6 could activate the mitogen-activated protein kinase signalling pathway and promote activation of ERK1/2. ERK1/2 is one of the mitogen-activated protein kinases 27,34 the key components of the reperfusion injury salvage kinase pathway, and plays an important role in protecting the myocardium from lethal ischemia-reperfusion injury. Constitutive activation of mitogenactivated protein kinase 1 (CaMEK) promotes ERK1/2 expression, which is expected to protect the heart against ischemia-reperfusion injury. 25,27 The relationship between ERK1/2 and the Hippo pathway has been reported in tumour diseases. 36 By using U0126, a specific ERK1/2 inhibitor, we showed that the cardioprotective effect of FGF6 via blockade of the Hippo pathway was primarily mediated by ERK1/2. Considering that treatment with VP, a specific YAP inhibitor, did not significantly change ERK1/2 activity in vitro or in vivo, we speculated that ERK1/2 functions upstream of the Hippo pathway.
However, further exploration are still needed.
MI causes loss of a huge number of CMs. It is clinically treated mainly using interventional therapy, which cannot prevent cell death.
Exploration of alternative drugs or strategies would be extremely valuable for management of MI patients. Meanwhile a growing number of results suggest that a strategy to promote CMs proliferation is most favourable to protect the heart during the process of regeneration. [37][38][39][40] Our study suggests that FGF6 is a powerful regulator to promote CMs cell cycle re-entry and improve cardiac function after MI. And this study also demonstrates that FGF6 attenuates activation of the Hippo pathway via ERK1/2 and progression of cardiac ischemic injury. Although the effect of FGF6 on MI injury requires further investigation and subsequent clinical translation our study presents novel evidence that FGF6 is an effective candidate drug to treat MI.