Protective mechanism of SIRT1 on Hcy‐induced atrial fibrosis mediated by TRPC3

Abstract High plasma levels of homocysteine (Hcy) are regarded as a risk factor for atrial fibrillation (AF), which is closely associated with the pathological consequence of atrial fibrosis and can lead to heart failure with a high mortality rate; here, we show that atrial fibrosis is mediated by the relationship between canonical transient receptor potential 3 (TRPC3) channels and sirtuin type 1 (SIRT1) under the stimulation of Hcy. The left atrial appendage was obtained from patients with either sinus rhythm (SR) or AF and used to evaluate the relationship between the concentration of Hcy and a potential mechanism of cardiac fibrosis mediated by TRPC3 and SIRT1. We next performed transverse aortic constriction (TAC) in mouse to investigate the relationship. The mechanisms underlying atrial fibrosis involving TRPC3 and SIRT1 proteins were explored by co‐IP, BLI and lentivirus transfection experiments. qPCR and WB were performed to analyse gene and protein expression, respectively. The higher level of atrial fibrosis was observed in the HH mouse group with a high Hcy diet. Such results suggest that AF patients may be more susceptible to atrial fibrosis and possess a high probability of progressing to hyperhomocysteinemia. Moreover, our findings are consistent with the hypothesis that TRPC3 channel up‐regulation leads to abnormal accumulation of collagen, with the down‐regulation of SIRT1 as an aetiological factor of high Hcy, which in turn predisposes to atrial fibrosis and strongly enhances the possibility of AF.

(HF) due to the accumulation of collagen fibres, such that patients with AF have HF with a preserved ejection fraction (HFpEF) rather than HF with reduced ejection fraction (HFrEF), accounting for 30% of cases. [5][6][7] These findings suggest that the atrial fibrosis is associated with collagen metabolism and is involved in the mechanism of AF. Furthermore, hyperhomocysteinemia is a pathological feature in the aetiology of atrial fibrosis, although the underlying mechanisms remain unclear. 8 Hcy, as a stimulus, can bind to G protein couple receptors (GPCRs) and regulate the phospholipase C (PLC) activation, producing intracellular messengers, 1,2-diacylglycerol (DAG); it recruits protein kinase C (PKC) and subsequently effecting the downstream signalling pathway of cardiac fibrosis. 9 Canonical transient receptor potential receptor 3 (TRPC3) is an indispensable factor in regulating the mechanisms of fibrosis development and in promoting the transition of fibroblasts into myofibroblasts with an adverse influence on the modulation of collagen. 10 Interestingly, TRPC3 is directly activated by PKC phosphorylation. We speculated that Hcy could trigger TRPC3 to mediate the mechanism of atrial fibrosis. On the other hand, sirtuin-1 (SIRT1) appears to function as an anti-fibrotic protein. Recently, the progression of myocardial fibrosis has been shown to simultaneously activate the renin-angiotensin system (RAS). This causes myocardial apoptosis through the TGF-β pathway and controls the aggregation of monocytes and fibroblasts, following with the down-regulation of SIRT1. 11 It remains unclear that whether TRPC3 and SIRT1 can control and modulate the fibrotic system to reciprocally affect cardiac structural remodelling under Hcy stimulation. In an attempt to address this issue, we hypothesize that TRPC3 is a novel regulator of SIRT1 in modulating the TGF-β pathway. In addition, our study aim to elucidate whether SIRT1 is directly involved in the process of TRPC3 mediated atrial fibrosis and its role in fibroblast proliferation and differentiation under high Hcy conditions.

| Animal model
All animal experiments were approved by the SLAC Labomouseory Animal Co. Ltd, Hunan, China. Mice were kept on a 12 hours light/12 hours dark cycle at a room temperature of 20-25°C, with a relative humidity of 40%-70%. Baseline information on male C57B6 mice (n = 60) was detected by transthoracic echocardiography. All experienced mice underwent transverse aortic constriction (TAC) at four weeks of age following randomization. Mice in the high-Hcy (HH) diet group were fed a high-Hcy diet (AIN-76A + 4% methionine with irradiation, ReadyDietech, China) to induce hyperhomocysteinaemia with HF (n = 25), and mice in the NH diet group were fed a normal diet (n = 25). The majority of the mice in the NH and HH groups showed HF signs after 7 weeks of age. The HH+Res group included mice that underwent TAC with a HH diet that were subjected to a single intraperitoneal injection of 20 mg/kg/d resveratrol (Res) for 21 days. 12 The mice in the HH group received the same dose of saline by intraperitoneal injection and underwent sham surgery. These mice were fed a folic acid and high-Hcy mixed diet classed as HH+FC group (Table 2).

| Transthoracic echocardiography
Mice were anaesthetized with 5% isoflurane for transthoracic echocardiography, which was performed using a Vevo2100 imaging system (VisualSonics). Ejection fraction (EF) was regarded as a systolic parameter, and E/A and E/E′ ratios were regarded as diastolic markers via baseline echocardiography. Pulsed-wave Doppler and tissue Doppler were performed to detect the peak ratio of E/A and E/E′ in the three groups at 4,7 and 16 weeks. Left ventricular (LV) end-diastolic volume (EDV) and end-systolic volume (ESV) were obtained by the Simpson method of disks. Ejection fraction was calculated as EF (%) = (EDV − ESV)/EDV × 100% and was used to determine systolic function from images in the parasternal shortaxis view as previously described. 13 Left ventricular end-diastolic diameter (LVEDD) and end-systolic diameter (LVESD) were recorded. Fractional shortening (FS) was evaluated with the following

| Surface electrocardiography
Surface electrocardiography (ECG) was recorded prior to in vivo arrhythmia induction studies for HH, NH and SH mice at 16 weeks of age. 15 The PR interval, QRS dimension, QT interval and RR interval were measured three times and averaged on MedLab6 software (Biological signal acquisition and processing system, Beijing, China).

| Isolation of mouse atrial fibroblast
Mouse atrial fibroblasts were collected and cultured from neonatal C57B6 mice at 0-2 days (Slake, Hunan, CHINA), weighing 3-5 g, in order to identify the relationship between atrial fibrosis and Hcy.
The specific methods for mouse atrial fibroblast culturing are as follows. Firstly, mice were disinfected, and the neonatal heart was rapidly removed. Left and right ventricles with a magnifying glass and ventricular tissue were removed. The myocardial tissue was digested by addition of a trypsin/Collagenase II mixture three times, followed by treatment with a serum-containing medium that inhibited enzyme activity. The cell suspension was centrifuged (500 g, 5 minutes) it, collected and plated. Cells were allowed to adherence for 2 hours in a CO 2 incubator.

| Cell proliferation analysis
Cardiac fibroblasts were cultured in T25 culture flasks

| Protein-protein interaction
GST pull-down: The gene encoding TRPC3-N (1-369aa) and TRPC3-C (659-836aa) was synthesized by Detai Biologics Co., Ltd. HEK 293-T cells were transfected with the above expression constructs to produce and purify recombinant protein for validating an interaction between TRPC3 and SIRT1 by bio-layer interferometry (BLI). Different concentrations of the TRPC3-C or -N peptide were applied in the mobile phase, and the association between the immobilized and flowing proteins was detected. The reference buffer was PBST and 5% DMSO (pH 7.4 PBS, tween 20 0.05% and 5% DMSO v/v).
Statistical analysis was performed using Data Analysis 9.0 software.
The dissociation rate constant (K D ) was obtained by curve fitting of the association and dissociation phases of the sensograms.

| Protein isolation and Western blot
Protein was extracted from heart tissues using a Tissue Homogenizer

| Quantitative reverse transcriptase polymerase chain reaction (RT-qPCR)
Heart tissues were stored in Allprotect Tissue Reagent (Qiagen) for RT-qPCR. Total RNA was isolated from heart tissue ground over liquid nitrogen and extracted using the standard TRIzol method F I G U R E 1 (Continues)

| Statistical analysis
Continuous variables are shown as the mean ± standard deviation. The same group of objects was exposed to multiple interventions, which was necessarily measured by repeated analysis. SPSS 22.0 software was used for the analyses.

| Patients with AF are more often diagnosed with severe cardiac fibrosis in the atrium than SR patients
As shown in Figure 1A-C, the left atrial appendage of AF patients exhibited an increase in TRPC3 and pro-fibrotic proteins, such as F I G U R E 1 High plasma levels of Hcy are closely related to the occurrence of cardiac fibrosis. A-B, The expression levels of TRPC3 and the relevant proteins, such as TGF-β and Collagen-I in AF patients. C, The relative mRNA level. D1-D2, The atrial nucleus in patients with AF is shown with haematoxylin and eosin (HE) staining, Masson's trichrome -stain and immunohistochemistry, as well as statistical results of positive areas and ratio to total area of the fibre (Magnification × 100, Scale bar = 200 μm; n = 4 per group). E, The ultra-structure of atrial tissues was observed by electron microscopy: E1-E3: SR patients, E4-E6: AF patients. Error line indicates mean and standard deviation. *P < .01, **P < .001. Multiple t-tests were performed using step-down bootstrap sampling to control the familywise error rate at 0.05 for 1B, C, and unpaired t-test was used for 1D. Fisher's exact test was used for analysing Table 1 | 495 TGF-β and Col-I. This was accompanied by a down-regulation of SIRT1, which was detected at both the protein and mRNA levels. Figure 1D shows that patients with AF more often presented with atrial fibrosis compared to patients with SR, as detected by IHC and Masson staining experiments. In addition, we observed by electron microscopy that the ultra-structure of atrial tissues was severely damaged in AF patients compared with SR patients ( Figure 1E). For example, the fascicle was ruptured, the crude and fine filaments were loosely arranged, and the mitochondria displayed compensatory enlargement, hyperplasia and disordered crista. The baseline characteristics of AF and SR patients are presented in Table 1, which shows that the morbidity of hyperhomocysteinemia and the QTc prolongation in AF patients exceeded that in SR patients. On the other hand, the incidence of hypertension was similar in AF and SR patients. Intriguingly, the phenomenon of AF combined with HF was commonly observed, accounting for 33.3% of patients with AF, which is consistent with the current statistics of this comorbidity. Figure 2A shows how the Hcy diet was administered to transverse aortic constriction (TAC) mice (n = 25) at 4-16 weeks. TAC mice that F I G U R E 2 Timeline for establishing mouse HFpHR model with a high Hcy diet and echocardiographic parameters for assessing systolic and diastolic function. A, Mice underwent TAC or sham surgery at 4 weeks of age in the HH group (n = 25), which were fed a high-Hcy diet (4% methionine) from 4-18 weeks of age. The NH group (n = 25) also underwent the above TAC operation but were fed a normal diet. Mice that underwent sham surgery and were fed a normal diet were considered the sham (SH) group (n = 10). Repeat echocardiogram (ECG) was performed at 16 and 18 weeks of age. We performed programmed ECG analysis, RT-qPCR and Western blot in the three groups at 18 weeks of age. B1, The value of ejection fraction (EF) in SH, NH and HH mice. B2, The value of interventricular septum thickness (IVS-S received the high-Hcy diet were classed as the HH group and those that received a normal diet as the NH group, with mice undergoing sham surgery regarded as the SH group. To induce AF, we administered a constant dose of acetylcholine (0.5 μL/mg), which was injected into each mouse through the caudal vein. Figure 2B-E reveals echocardiographic results at 4, 7 and 16 weeks of age. The EF was not significantly different at baseline (4 weeks) amongst the three groups ( Figure 2B). The interventricular septal thickness (IVS-S) in HH mice was markedly greater than that in the NH mice at 16 weeks of age (1.914 ± 0.167 in HH vs 1.832 ± 0.134 in NH, P < .001; Figure 2B). The   Figure 2B). The measurements of LVIDd were markedly increased following TAC (NH and HH) was observed, especially in the HH group ( Figure 2B, Table 3). Evolving mitral flow velocity patterns were analysed during different time periods ( Figure 2C). The ultrasonic images at different periods were obtained from a longitudinal section of the heart ( Figure 2D). In addition, increasing posterior wall thickness at end-diastole (PWTd) and posterior wall thickness at end-systole (PWTs) measurements were concurrently noted (Table 3). However, the heart rate (reported here as the RR interval) was not different amongst the three groups (Table 3).  Figure 3B). In addition, the features of ventricular fibrillation (VF) include the disappearance of the QRS wave, which is replaced by different VF waves and were observed in the NH (n = 3/18) and HH (n = 4/17) groups ( Figure 3C,D). Figure 3D illustrates that AF was induced in 10 out of 17 Figure 3D). Despite the lack of statistical significance, the propensity to develop AF with or without feeding a high-Hcy diet was higher in groups that underwent TAC (HH and NH groups) than in the SH group ( Figure 3D).

| Arrhythmia induced by acetylcholine and atrial fibrosis in mice
Interestingly, the duration of induced AF was longer in both the HH and NH groups than in the SH group, especially in the HH group ( Figure 3E).
Next, we studied whether Hcy affects the process of atrial fibro-  Figure 3F). The similar changes were observed for the heart weight-to-tibia length ratio (HW/TL) at 16 weeks of age ( Figure 3G). In addition, Masson's trichrome staining simultaneously affirmed that the HH groups were more vulnerable to fibrosis, compared with the NH group ( Figure 3I). Moreover, transverse sections of atrium and cardiac size showed a greater enlargement in the HH group, compared with the NH group ( Figure 3H,K) and further demonstrated that Hcy could exaggerate the adverse influence of TAC-induced HF. Survival analysis showed that mice in the NH and HH groups exhibited higher mortality, especially after 10 weeks, than that in the SH group, as shown in Figure 3J. F I G U R E 4 TRPC3-KD or SIRT1-OE in mice could effectively alleviate atrial fibrosis. A, C57B6 mice (n = 10) at 1 week of age were injected with purified lentivirus via the caudal vein. Mice underwent TAC and were fed with a high-Hcy diet at 4 weeks of age, with relevant tissues obtained after 16 weeks. B, F, Heart weight-to-tibia length ratio (HW/TL). C, The Masson staining images with magnified local images in atrial tissues. D, The protein levels of TRPC3 and TGF-β measured in the three groups. E, Ten C57B6 mice underwent TAC at 4 weeks of age and were fed with a high-Hcy and folic acid mixed diet during 4-16 weeks (HH+FC group). F, A subset of these animals were injected with Res (20 mg/kg/d) for 21 day (HH+Res group). G, The statistical total area of the fibre ratio. H, I, The protein and mRNA expression levels of SIRT1 and TRPC3. Error line indicates mean and standard deviation. *P < .05 and **P < .01. Mixed model regression with post hoc testing (Tukey adjustment) was used for 4B and 4F. Multiple t-tests were performed using step-down bootstrap sampling to control the familywise error rate at 0.05 for 4C, 4D and 4G-4I

| SIRT1-overexpression and TRPC3-KD mice can efficiently control Hcy-mediated atrial fibrosis
with a high-Hcy diet compared to those mice with a normal diet, there is no statistic significance (Figure 4). According to their echocardiography, the IVS-S at end-systole and LVIDd were efficiently alleviated by the above treatments ( Figure S1A). Meanwhile, the Masson staining of TRPC3-KD+HH and SIRT1-OE+HH mice were dramatically decreased in comparison with the HH+vector mice ( Figure 4C), whilst the EF level Unpaired t-test was used for 5A-5F, and mixed model regression with post hoc testing (Tukey adjustment) was used for 5G, 5H remained at a relatively stable state ( Figure S1B). In addition, transgenic mice displayed no apparent changes in cardiac structure and function ( Figure S1C,D). This result confirmed that despite the influence of the high-Hcy diet and the TAC surgery, the protein level of TRPC3 and the related TGF-β signalling pathway were inhibited in the TRPC3-KD+HH group, whilst SIRT1 was increased. Similar results were observed in the SIRT1-OE+HH group with the lower level of TGF-β ( Figure 4D). In general, the TRPC3-KD and SIRT1-OE models generated by injecting purified lentiviruses exhibited dramatically abrogated Hcy-induced fibrosis in atrial fibroblasts and decreased levels of pro-fibrotic proteins, such as TGF-β and Col-I (P < .001). Figure 4E shows the procedure for intraperitoneal injection of Res 20 mg/kg/d for 21 days from the fourth week, whilst another group were fed a diet containing folic acid and injected with the same dose of saline, respectively regarded as HH+Res or HH+FC mice. Interestingly, regardless of whether folic acid and vitamins were added to the food or not, there was no notable protection from the damage caused by high Hcy levels ( Figure 4F,H). However, Res attenuated the enhancement of TRPC3 and decrease in SIRT1 induced by the high-Hcy diet combined with TAC injury and controlled the TGF-β improvement ( Figure 4F,H). In addition, the mRNA levels of TRPC3 and SIRT1 in the above groups were consistent with their protein levels ( Figure 4I).

| Hcy can promote the proliferation and differentiation of atrial fibroblasts and modulate the relevant protein levels
First, the results revealed that the expression levels of total TRPC3 were significantly increased by the intervention of Hcy in a dose-dependent manner ( Figure 5A), compared with those in the control group (Hcy 0 μmol/L). However, those expression levels were significantly decreased by Pyr-10 treatment, which is characterized by the inhibition of DAG-mediated TRPC3 signalling pathway (P < .001, Figure 5B). These results further confirmed that Hcy could activate the upstream protein of DAG, GPCRs and increase cardiac TRPC3 expression in mouse atrial fibroblast, and that effect was abolished by Pyr-10. Second, we observed that the protective effects of SIRT1 against cardiac remodelling were decreased by Hcy (P < .001). Intriguingly, Res could prevent the Hcy-induced enhancement of TRPC3 level. In addition, the above effects of Hcy on atrial fibroblast were accompanied by a reduction in the expression level of fibrotic proteins, such as TGF-β, which was demonstrated to be a pivotal molecule in fibrosis (P < .001) ( Figure 5B). Finally, when the inhibitor of SIRT1 (Sal) was applied to mice fed a high-Hcy diet, the pro-fibrotic effects on atrial fibroblast were exacerbated ( Figure 5B).
Second, we examined the expression levels of cardiac TRPC3 using purified lentivirus to transfect primary cultured atrial fibroblast. Figure S2A shows protein levels in a dose-dependent manner along with the promotion of TGF-β expression ( Figure 5F), which suggested that TRPC3 can affect the transcription and translation of SIRT1, which further modulates the protein levels of TGF-β. Finally, the results suggest that Hcy could promote the proliferation and differentiation of fibroblasts ( Figure 5G). In contrast, the promotion of fibroblast proliferation by Hcy was blocked in TRPC3-shRNA stable cell lines ( Figure 5H).

| SIRT1 is an TRPC3-interacting partner
Although TRPC3 is well-established to play a central role in transducing cardiac fibrosis signalling, 10 the mechanisms underlying its signal transduction function remain poorly understood. To address this issue, researchers have identified SIRT1 as a modulator of the transcription of TGF-β-dependent genes, which participate in the process of fibrosis. 21,22 In a preliminary experiment, we found that Hcy could increase the level of TRPC3; nevertheless, the protein level of SIRT1 was decreased ( Figure 5A). SIRT1 could be a potential TRPC3interacting partner, as both regulate the process of cardiac fibrosis in synergy. To confirm this hypothesis, we performed an in vitro GST pull-down assay using Flag-TRPC3 expressed in HEK293T cells and purified GST-TRPC3 to further validate the TRPC3-SIRT1 interaction in mammalian cells. In addition, IP experiments were performed to corroborate that Flag-TPRC3 and HA-SIRT1 interact in HEK293T cells ( Figure 6B). Endogenous TRPC3 protein immunoprecipitated F I G U R E 6 The interaction between TRPC3 and SIRT1. A, Co-localization of TRPC3 and SIRT1. Cells are co-stained with antibodies against TRPC3 (Alexa Fluor 488, green) and SIRT1 (Alexa Flour 564, red). Arrow indicates co-localization. B, Interaction of TRPC3 with SIRT1 in mammalian cells. C, Endogenous TRPC3 binds to SIRT1 in MCFs. D, The interaction of endogenous TRPC3 with SIRT1 in MCFs. E, Schematic illustration of the domains of TRPC3 and its association with SIRT1. F, The C-terminal tail (659-836) of TRPC3 interacts with SIRT1. G, H, Analyse of the binding affinity between full-length SIRT1 and TRPC3-C (G) or TRPC3-N (H) by bio-layer interferometry (BLI). All experiments were performed in triplicate with a SIRT1 antibody ( Figure 6C), indicating that TPRC3 and SIRT1 could form a protein complex in cells. Moreover, the interaction between endogenous TRPC3 and SIRT1 was markedly improved by Hcy stimulation ( Figure 6D). Finally, IF staining showed that TRPC3 and SIRT1 co-localized in the cytoplasm of cells ( Figure 6A), suggesting that the binding of these proteins occurs in the cytoplasm.
Next, we determined the functional outcomes of the TRPC3-SIRT1 interaction and evaluated whether TRPC3 affects the functions of SIRT1. We demonstrated that the regulation of fibrosis by SIRT1 and TRPC3 is antagonistic. We found that the stable cell line transfected with TRPC3-OE lentivirus exhibited a low expression level of SIRT1 ( Figure 5F), which suggested that TRPC3 is required for the activa- ( Figure 6F).

| D ISCUSS I ON
With the increasing prevalence of cardiac structure remodelling in various cardiovascular diseases, the pathogenesis of cardiac fibrosis is an important research topic. Here, we used hyperhomocysteinemia to aggravate the progression of atrial fibrosis and ultimately the structural re-entry circuits and local conduction block in the atrium.
It indirectly determines the basic pathogeny of AF. We found that patients with AF were more vulnerable to severe atrial fibrosis, than SR patients ( Figure 1E, in HH mice than in the NH mice ( Figure 3I). This AF occurrence was due to interruption of the continuity of atrial myocytes by excessive extracellular matrix (ECM) deposition and the formation of aberrant electrical circuits, which assisted in understanding the disorder of current conduction in AF ( Figure 3B). In our model, the duration of AF was more easily prolonged in the HH group than in the NH group, which further demonstrated that Hcy is a risk factor for atrial fibrosis-related AF and the increasing risk of AF. 28,29 Amongst the F I G U R E 7 Schematic representation of our working hypothesis. Increased Hcy levels promote the proliferation and differentiation of fibroblasts, which can bind to GPCRs, and increase the protein levels of TPRC3 and through interaction with SIRT1 promote atrial fibrosis. This may occur though the direct activation of TGF-β to induce collagen-I accumulation associated with the pathological mechanism of AF. However, the specific mechanism of AF requires further investigation treatment groups, HF mice were found to manifest cachexia with different degree of death, which is characterized by a reduction in subcutaneous fat, activity and food intake, starting from 7 weeks of age. 30 TRPC3 is considered an indispensable factor in regulating the mechanisms of fibrosis development in mouse cardiomyocytes.   Figure 5H). These results indicated that TRPC3 is critical mediator in the TGF-β signalling pathway and predominantly mediates Hcy-induced maladaptive fibrosis in mouse hearts. Although TGF-β signalling is an important role for cardiac remodelling, 35 the role of TRPC3 in regulating TGF-β and promoting fibrosis remains unknown.
We further studied the potential molecular mechanisms underlying how TRPC3 modulates TGF-β-mediated cardiac fibrosis. It is worth acknowledging that up-regulation of SIRT1, an important cytokine that protects against cardiac remodelling was observed and led to the inhibition of collagen accumulation by inhibiting TGF-β expression. 36,37 The severity of cardiac fibrosis was correlated with a lower expression level of SIRT1 in AF patients and HH mice. However, adding Res, an activator of SIRT1, could effectively attenuate cardiac fibrosis in both animal and cell models ( Figure 4H). Similarly, SIRT1-OE+HH mice efficiently mitigated the extent of atrial fibrosis compared with that in the HH+vector group ( Figure 4A-D). Thus, SIRT1 may play a pivotal role in the mechanism of cardiac fibrosis.
Interestingly, SIRT1 overexpression inhibited the TGF-β excess induction, but no changes were not observed in the protein level of TRPC3 ( Figure 5E). In addition, with the increase of TRPC3 by titrating the amount of ectopic Flag-TRPC3, SIRT1 protein levels were significantly decreased in a dose-dependent manner, following the enhancement of TGF-β ( Figure 5F), which suggested that SIRT1 might be required for the indirect adjustment of TGF-β expression mediated by TRPC3. Indeed, the application of the SIRT1 inhibitor, Salermide, to atrial fibroblasts further enhanced the protein level of TGF-β and Col-I, but no change in the expression levels of TRPC3, compared with HH groups ( Figure 5B). In our study, it was identified SIRT1 as a novel regulator and partner of TRPC3 via direct or indirect binding. Although previous studies have shown that both SIRT1 and TRPC3 participate in mediating TGF-β signalling, 32,38,39 it remains unknown how TPRC3 regulates the localization of SIRT1 and whether these proteins directly interact with each other. In fibroblasts transfected with purified TRPC3-shRNA lentiviruses, the expression levels of SIRT1 are increased with the MOI value in a dose-dependent manner, which indicates that the stability of SIRT1 is directly affected by TRPC3. Conversely, transfection of the SIRT1-OE vector into HEK293 cells did not affect the protein expression level of TRPC3 compared to transfection of a scrambled vector ( Figure 5E). Additionally, the co-IP, BLI and IF results demonstrated the correlation between TRPC3 and SIRT1 in regulating the process of fibrosis ( Figure 6A-D), and moreover, it affirmed their binding domains to the C-terminal (659-836aa) of TRPC3 and SIRT1 ( Figure 6E-H). As further discussed below, TRPC3 can physically interact and subsequently activate SIRT1 in response to TGF-β signalling and play a prominent role in promoting the proliferation and differentiation of fibroblasts.
Recent studies have determined that SIRT1 can modulate the regulation of a variety of cellular processes associated with RAS.
Amongst them, SIRT1 protects the cell from oxidative stress. 40 Combined with the evidence of the interaction between TRPC3 and SIRT1, it indicated that Hcy activates TRPC3-SIRT1 axis by modulating TGF-β activation, driving fibrosis in both cardiomyocytes and cardiac fibroblasts. In this study, we investigated the mechanism of TRPC3 in the progression of atrial fibrosis both in vitro and in vivo. By analysing left atrial appendage specimens, we uncovered that TRPC3 levels were closely correlated with the degree of atrial fibrosis and the incidence of AF. Moreover, patients with hyperhomocysteinemia exhibited increased protein levels of TRPC3 and decreased levels of SIRT1 along with activation of the TGF-β signalling pathway (Figure 7). In agreement with this, our analysis showed that the RNA level of TRPC3 was increased in the AF group, whereas the RNA level of SIRT1 was reduced ( Figure 1C) compared with that in the SR group. Remarkably, we further confirmed that the Hcy group exhibited a marked increase in fibroblast proliferation and enhanced expression of a pro-fibrotic protein, TGF-β, in vitro and in vivo. Interestingly, our experiments also revealed that TRPC3 can directly interact with SIRT1, acting as a negative modulator of SIRT1, leading to TGF-β signalling pathway activation under Hcy stimulation.

| CON CLUS I ON S AND PER S PEC TIVE
Our present study not only illustrates the biochemical function of TRPC3, which binds directly to SIRT1 through its C-terminal (659-836aa) to modulate TGF-β signalling, with the interaction intensified under Hcy stimulation, but also unveils the roles of TRPC3 and SIRT1 in atrial structure remodelling and fibrosis, which reciprocally increase the occurrence of AF accompanied by HF (Figure 7).
Together, our results suggest that TRPC3 may serve as a biomarker for preventing the consequence of atrial fibrosis in AF patients with homocysteinemia, especially as activation of SIRT1 is associated with inhibition of the TGF-β signalling pathway. However, the underlying mechanisms of atrial fibrosis are mediated by TRPC3 and SIRT1 merits further investigation.

ACK N OWLED G EM ENTS
We are grateful to Dr Juxiang Li for careful reading and revision of the manuscript.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

E TH I C A L S TATEM ENT
The authors have no ethical conflicts to disclose.

CO N S ENT S TATEM ENT
Our institution's committee on human research gave approval for this study, and all participates gave informed consent. The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.