Sacubitril/valsartan Decreased Atrial Fibrillation Susceptibility by Inhibiting Angiotensin II-induced Atrial Fibrosis through p-Smad2/3, p-JNK, and p-p38 Signaling Pathways


 Background: Sacubitril/valsartan (SAC/VAL), combined inhibitors of angiotensin receptor (AT-R) and neprilysin receptor, prevents Angiotensin Ⅱ (AngⅡ) from binding AT1-R and blocks degradation of natriuretic peptides. Despite its efficacy in reducing ventricular fibrosis and preserving cardiac functions, which has been extensively demonstrated in myocardial infarction or pressure overload models, few studies have been conducted to determine whether SAC/VAL could attenuate atrial fibrosis and decrease atrial fibrillation (AF) susceptibility. Methods: Sprague-Dawley rats were divided into three groups after implantation of an osmotic pump preloaded with AngⅡ and received corn oil, VAL, or SAC/VAL treatment for 28 days. Electrophysiological study was performed to determine inducibility and duration of AF. Echocardiography was performed before and 28 days after osmotic mini-pump implantation to evaluate cardiac functions. Enzyme-linked immunosorbent assay was perfomed to detect the serum concentration of atrial natriuretic peptide (ANP), N-terminal pro-brain natriuretic peptide (NT-proBNP) and AngⅡ. Masson staining was performed to determine the extent of interstitial fibrosis. Immunohistochemical, and immunofluorescence staining as well as transwell and MTT assay were also performed. Western blot analysis was perfomed to figure out how SAC/VAL exerts its anti-fibrosis effects in atriums.Results: After 28 days of AngⅡ stimulation, rats in SAC/VAL group exhibited reduced reduced extent of atrial fibrosis, inhibited proliferation, migration, and differentiation of atrial fibroblasts and decreased susceptibility to atrial fibrillation. We further found that SAC/VAL exerted its effect on AngⅡ-induced atrial fibrosis by inhibiting the p-Smad2/3, p-JNK, and p-p38 MAPK signaling pathways in vivo. Conclusions: Our study provided experimental evidence for the inhibition of atrial fibrosis and reduced susceptibility to AF by SAC/VAL. These results emphasize the importance of SAC/VAL in the prevention of AngⅡ-induced atrial fibrosis and may help to enrich the options for atrial fibrillation pharmacotherapy.


Background
As one of the most frequently encountered rhythm disturbances in the clinical setting, atrial brillation (AF) aggrandizes stroke risk and worsens the prognosis [1,2]. Currently, available treatment options for AF fail to achieve satisfactory therapeutic bene ts as the complex mechanisms behind this disease, including structural and electrical remodeling and autonomic nervous system dysfunction [3][4][5][6]. Notable among these mechanisms is atrial structural remodeling, which is perhaps a vital link of all the AF mechanisms. The development of atrial structural remodeling is strongly associated with atrial brosis [7]. Studies in this regard have identi ed that therapies that target atrial brosis would be of great clinical importance in treating AF [8][9][10].
Angiotensin-II (Ang ) is a major mediator of the renin-angiotensin-aldosterone system (RAAS) and plays an essential role in the formation of atrial brosis [11,12]. Upon binding to its type 1 receptor (AT 1 -R), Ang exerts pro-brotic effects and eventually forms an arrhythmogenic substrate, which initiates and perpetuates AF [13]. The higher the Ang level, the more severe the extent of atrial brosis and the increased incidence of atrial brillation [14]. Growing evidence from animal studies has a rmed that blocking RAAS using angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) reverse atrial brosis and decrease atrial arrhythmia inducibility [15,16].
Sacubitril/valsartan (SAC/VAL), the rst representative of a rst-in-class drug, combined inhibitors of angiotensin receptor and neprilysin (neutral endopeptidase, NEP) receptor, prevents Ang from binding to AT 1 -R and blocks degradation of natriuretic peptides (NPs) [17]. The PARADIGM-HF trial has evaluated its e cacy in reducing mortality and hospitalization among patients with heart failure with reduced ejection fraction (HFrEF) [18]. Previous experimental studies have demonstrated that SAC/VAL deterred the ventricular structural remodeling and dysfunction post-myocardial infarction (MI). Recently, Chang et al.
used a rabbit MI and HF model to demonstrate SAC/VAL could preserve heart systolic function and avert MI-induced electrophysiologic remodeling by reducing phosphorylated expression calmodulin-dependent protein kinase II (p-CaMKII) [19]. Vaskova et al. showed that downregulation of miR-181a exosomes could also be one of the mechanisms by which SAC/VAL improves cardiac function and reduces myocardial brosis in rats with chronic MI [20]. Protein kinase G (PKG) signaling and nuclear factor kappa-light-chainenhancer of activated B cells (NF-κB)/nucleotide-binding oligomerization domain (NOD)-like receptor protein 3 (NLRP3) signaling pathways are also involved in the attenuation of ventricular brosis by SAC/VAL [21,22]. However, all of these published data on the signaling pathways through which SAC/VAL exerts its anti brotic effects are heterogeneous. Moreover, it is still unclear that the particular mechanisms behind these protective roles of SAC/VAL on AF. Therefore, in this study, we aim to use a rat All SD rats were anesthetized with iso urane (5% induction; 2-3% maintenance). An osmotic pump (model 2ML1, Alzet 2004, USA) preloaded with Ang was aseptically implanted in rats subcutaneously to induce atrial brosis by the continuous release of Ang (750 ng/kg/min) for 28 days. Later, all rats were assigned to three groups to receive corn oil, valsartan (VAL, 48 mg/kg body weight, n = 6), or sacubitril/valsartan (SAC/VAL, 60 mg/kg body weight, n = 6) for 28 days. All drugs were dissolved in corn oil and administrated by gavage. Systolic blood pressure (SBP), diastolic blood pressure (SBP), and mean blood pressure (MBP) were measured at the end of experiment by the tail-cuff method as previously described [23].

Electrophysiological study
The electrophysiological study (EPS) was performed as previously described before the rats were sacri ced [24]. Brie y, rats were anesthetized with iso urane (5% induction; 2-3% maintenance) and placed on the heated pad to maintain body temperature at 37 °C. An electrode catheter (1.1 F, Science) was introduced into the right atrium via the right jugular vein then released a train of electrical stimuli at a pacing cycle length of 100 ms to test the susceptibility to atrial arrhythmias. Electrical stimulation was performed ten times in the same manner, and the number of successful AF induction within these ten times was recorded. The duration of AF is determined by the time from the end of the burst pacing to the rst sinus P wave after the atrial rhythm. AF in this study was de ned as a rapid and irregular atrial rhythm with an irregular R-R interval lasting more than 1000 ms.

Echocardiography
Before and 28 days after osmotic mini-pump containing Ang implantation, rats were examined with Mmode echocardiography (Vevo 770, Visual Sonics, Inc., Toronto, Ontario, Canada). Inhalational anesthesia with 2% iso urane and a heated pad were adopted during the process of image acquisition.
Left atrial diameter (LAD), LV end-diastolic diameter (LVEDD), and ejection fraction (EF) were measured and averaged over three cardiac cycles. Mitral valve ow was assessed by using pulsed-wave Doppler ultrasonography. The early (E) and atrial (A) peaks are measured and averaged over three cardiac cycles.

Enzyme-linked immunosorbent assay
The serum concentration of atrial natriuretic peptide (ANP), N-terminal pro-brain natriuretic peptide (NT-proBNP) and Ang were detected by ELISA according to the manufacturer's instructions. Kits with Catalog E02A0204, E02A0493, and E02N0008 were purchased from Shanghai BlueGene Biotech Co., Ltd.. The optical density (OD) at 450 nm was measured by an ELISA reader. The results were expressed as pg/ml or ng/ml.
2.6. Histological, immunohistochemical, and immuno uorescence staining Left atrial tissue samples from rats were xed with 4% paraformaldehyde, embedded in para n, transversely cut into 5 µm thickness. Masson staining was performed as previously described to highlight the bers, and the extent of interstitial brosis was determined by brosis area / total administration area × 100%. Immunohistochemical staining were performed using antibodies against collagen I (Abcam, ab34710) and collagen III (Abcam, ab7778) as previously described. Immuno uorescence staining were performed using the methods described previously with antibodies against α-smooth muscle actin (α-SMA, Abcam, ab5694) and vimentin (Abcam, ab137321). Images acquisition and analysis were used by Image-Pro Plus.

Cell migration and proliferation evaluation
Rat atrial broblasts (AFBs, Cat NO.: CP-R074) were purchased from Procell Life Science&Technology Co.,Ltd (Wuhan, China) for cell proliferation and migration evaluation. The AFBs were divided into three groups and pretreated with dimethyl sulfoxide (MedChem Express, control group), valsartan (10 µmol/L, MedChem Express, VAL group), and LBQ657 + valsartan (10 µmol/L, 1:1 ratio, MedChem Express, SAC/VAL group) for 24 h respectively. AFBs in these three groups were then stimulated with Ang (100 nmol/L) for another 24 h. Transwell assay was performed to evaluate cell migration. The nonmigratory cells remaining in the upper chamber were removed, and the membranes containing AFBs were xed and then stained with crystal violet. Five elds at × 200 magni cation were randomly selected under the light microscope to count the cells migrating to the lower compartment. Cell proliferation was quanti ed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium Bromide (MTT) assay (Sigma, St. Louis, MO, USA). The OD value at a wavelength of 490 nm of each well was detected using a spectrophotometer (BioTek, Elx800, Winooski, VT, USA).

Western Blotting Analysis
RIPA lysis buffer containing protease and phosphatase inhibitors were added to the proteins extracted from the left atrium of rats and quanti ed using the BCA protein analysis kit, as described previously. The blotting. Gel Imaging System (Tanon 5200) was used to detect the bands and the results were quanti ed by densitometry software (Image-Pro Plus).

Statistical Analysis
All data were represented as or percentage and analyzed using GraphPad Prism (version 6.0).
One-way ANOVA followed by Tukey's post-hoc was used to determine statistical signi cance for multiple groups. A P value < 0.05 was considered statistically signi cant.

Sacubitril/valsartan attenuated Ang -induced atrial brosis
To determine the effects of SAC/VAL on atrial brosis, the osmotic pump containing Ang was implanted subcutaneously into rats to generate the atrial brosis model. Pathological assessment using Masson's staining and immunohistochemistry of collagen I and collagen III deposition were evaluated at day 28 after the procedure. In the saline group, Masson's staining revealed signi cant LA brosis in response to Ang infusion (Fig. 1). Immunohistochemistry (IHC) staining showed increased deposition of collagen I and collagen III (Fig. 1). While both SAC/VAL or VAL alone signi cantly restored histological changes and decreased collagen deposition in the atriums of Ang stimulated rats. The quantitative morphometric analysis demonstrated that the extent of brosis and collagen I and collagen III depositions were signi cantly reduced in the rats from both VAL and SAC/VAL group compared to saline-treated rats (P < 0.05). Further analysis showed that the brotic extent and collagen deposition were signi cantly lower with SAC/VAL treatment than with VAL, suggesting that SAC/VAL possesses more potent anti brotic capacity over VAL. 3.2. Sacubitril/valsartan inhibit the proliferation, migration, and differentiation of atrial broblasts We next used the MTT assay and the Transwell test to verify the effect of SAC/VAL on the migration and proliferation of atrial broblasts. Ang promoted more remarkable cell migration and proliferation in the control group. As illustrated in Figs. 2B and 2C, the migration and proliferative rate of AFBs was signi cantly reduced after SAC/VAL or VAL alone administration. Compared with the VAL group, SAC/VAL had a stronger inhibitory effect on AFBs migration and proliferation. These results indicated that SAC/VAL might play roles in the function of AFBs in vitro.
To further con rm the protective effects of SAC/VAL, the immuno uorescence assay was carried out (Fig. 3). Results clearly revealed the increased expression of α-SMA and vimentin in atriums from salinetreated rats after Ang stimulation. While in SAC/VAL and VAL treated rats, the distribution and expression of both α-SMA and vimentin were decreased (Fig. 3). Furthermore, a more obvious attenuation was observed in the SAC/VAL group. This result was consistent with the ndings obtained by Masson's staining and IHC staining.

Sacubitril/valsartan preserved atrial and ventricular functions
We performed echocardiography to assess the systolic and diastolic function of the heart, as well as the left atrial dimension on the day before the procedure and the 28-day after the procedure. No difference was found in baseline parameters between all three groups ( Fig. 4C-4F). After 28 days of Ang infusion, rats in the saline group developed marked LA enlargement, but not in SAC/VAL or VAL treated rats. The LVEDD and LVEF were increased, and the MV E/A ratio was decreased in saline-treated rats post-Ang stimulation. Such changes were corrected in the SAC/VAL, or VAL alone treated rats (p < 0.05) (Fig. 4). No difference was found in the abovementioned parameters between the two treated regimen. These ndings were corroborated that previous clinical trials had demonstrated that SAC/VAL could improve HFrEF patients' heart function. As expected, both VAL and SAC/VAL reduced Ang -induced systolic, diastolic, and mean blood pressure increases compared with saline, but no difference in the two drugs' anti-hypertensive effects was observed. The serum concentrations of ANP, NT-proBNP, and Ang were measured at the end of the experimental period. The level of ANP was elevated signi cantly in SAC/VAL group as expected (Fig. 5A). In addition, SAC/VAL therapy resulted in a signi cant reduction of serum NT-proBNP compared to the control and VAL group (Fig. 6C). This result was consistent with changes in left ventricular systolic function. We also observed a tendency for an increase in the level of Ang of SAC/VAL treatment, but there was no statistical signi cance (Fig. 6B).

Sacubitril/valsartan decreased atrial arrhythmias inducibility
We next tested whether marked atrial brosis post-Ang stimulation could lead to a lower threshold for inducibility of AF. We performed in vivo electrophysiologic study (EPS), as mentioned previously [24]. In the whole process of EPS, all of the rats did not show a spontaneous arrhythmia. Saline treated rats showed signi cantly increased vulnerability to AF, as evidenced by an increased number of AF episodes and lengthier AF duration compared to VAL alone and SAC/VAL treated rats after two rounds of 30 s burst pacing (Fig. 5, p < 0.05). While the SAC/VAL group were more resistant to the AF-provoking effect of Ang than VAL alone treated rats (p < 0.05). These results demonstrated that SAC/VAL plays a protective role against the vulnerability to AF during in vivo EPS.

Sacubitril/valsartan inhibited Ang -induced Smad2/3, p38 and JNK activation
To explain the mechanism of protective effects of SAC/VAL against Ang induced atrial brosis, the relative protein expression of p-Smad2/3 and non-Smad proteins (ERK1/2, p-ERK1/2, p38 MAPK, p-p38 MAPK, JNK, and p-JNK) were detected using western blot. As depicted in Fig. 7, after 28 days of Ang infusion, the relative expressions of p-Smad2/3, p-ERK1/2, p-p38 MAPK, and p-JNK were signi cantly increased in rats treated with saline. Both SAC/VAL and VAL could signi cantly reduce the expression of the abovementioned proteins. Compared with the saline group, SAC/VAL caused about a 27% decrease in the relative expression of p-Smad2/3, a 28% decrease in the relative expression of p38, and a 25% decrease in the relative expression of p-JNK more than VAL (p < 0.05). However, no signi cant difference was observed between the relative expression of p-ERK in both SAC/VAL and VAL groups. The level of t-ERK1/2, t-p38, and t-JNK did not exhibit a marked difference among the three groups as well.

Discussion
The brous scar in atriums, acting as electrical barriers, uncouples the well-connected syncytium and increases heterogeneous electrical conduction [25]. This interaction between atrial structural remodeling and electrical remodeling allows for AF initiation and perpetuation [26]. However, as the most notable feature of atrial remodeling, atrial brosis mechanisms have not yet been fully elucidated. A growing body of evidence has suggested that atrial brosis might be a potential therapeutic target for AF [8][9][10]. In the present paper, we manifested that SAC/VAL could prevent Ang induced atrial brosis as demonstrated by less distorted LA architecture, lower deposition of collagen and , and attenuated distribution and expression of both α-SMA and vimentin. We also demonstrated that the susceptibility to atrial arrhythmias was signi cantly decreased following SAC/VAL treatment. We further veri ed that these protective roles of SAC/VAL could be attributable to the inhibition of p-Smad2/3, p-p38 MAPK, and p-JNK pathways.
Currently, therapeutic regimens, including ARBs, have been recommended for AF prevention by their capabilities to moderate the atrial structural and electrical remodeling through inhibition of RAAS in addition to their anti-hypertensive effects [27]. Likewise, active NPs have also been proven to inhibit brotic responses to prevent structural and electrical remodeling, as evidenced by the reduced risk of postoperative AF occurrence after administration of recombinant human atrial natriuretic peptides (ANP) or brain natriuretic peptides (BNP) [28,29]. The same effects can be achieved by inhibiting NEP to increase endogenous NPs [30,31]. SAC/VAL is an agent with dual inhibitory effects on ACE and NEP. The latter enzyme is responsible for the degradation of three NPs -ANP, BNP, and C-type natriuretic peptide (CNP). All these three NPs, upon binding to their receptors, are capable of elevating the cGMP level and inhibiting Ang induced DNA synthesis in both cardiomyocytes and AFBs [32]. Previous basic studies have substantively focused on the ventricle, and results from these studies shown that SAC/VAL has an inhibitory effect on ventricular brosis and thus preserves systolic cardiac function [21,33,34]. A recent study used the rabbit MI model demonstrated that SAC/VAL could also ameliorate post MI electrophysiologic remodeling and alleviate ventricular tachyarrhythmia inducibility. However, a few studies have been conducted to determine whether SAC/VAL could attenuate atrial brosis and decrease AF inducibility. In our study, we found that SAC/VAL did reduce AF susceptibility by inhibiting atrial brosis, which were consistent with two recent studies regarding the impact of SAC/VAL on AF [30,31].
We also found that VAL had less evident effects on the prevention of atrial brosis and reduction of atrial arrhythmias inducibility than SAC/VAL. Previous study reported that simultaneous inhibition of NEP and AT1-R with sacubitril and valsartan has been shown to reduce Ang -induced collagen synthesis in rat AFBs to a greater extent than either compound alone [33]. Moreover, we veri ed the ability of SAC/VAL to inhibit the proliferation, migration, and differentiation of atrial broblasts both in vivo and in vitro. We also observed that elevated ANP levels in the SAC/VAL group were accompanied by a signi cant reduction in the extent of atrial brosis as well as the proliferation, differentiation, and migration of AFBs. Indeed, endogenous ANP released by cardiomyocytes inhibited atrial broblasts proliferation and collagen deposition [35].
To date, the role of SAC/VAL in suppressing ventricular brosis is well-de ned, but the mechanism behind it remains enigmatical by the current evidence, not to mention the mechanism underlying the inhibition of atrial brosis by SAC/VAL [20][21][22]. Only one study focusing atrial electrical remodeling demonstrated that SAC/VAL could reduce AF susceptibility by inhibiting the calcineurin/nuclear factor of activated T cell (NFAT) pathway [30]. In our study, we nally looked at signaling pathways to give our answer as to how SAC/VAL exerts its anti-brosis effects in atriums. Atrial brosis necessitates involving RAAS, a crucial player in the pathogenesis of cardiac remodeling [36]. Ang is a signi cant mediator of this system and renders manifold downstream cytokines active through multiple signaling pathways, including the mitogen-activated protein kinase (MAPK) pathway. In addition to directly contributing to brosis, activation of the MAPK signaling pathway stimulates the secretion of TGF-β1, while TGF-β1 has also been reported to implicate in the development of atrial brosis, which promotes the secretion and differentiation of atrial broblasts via the typical Smad-dependent pathway and noncanonical Smadindependent pathways[37]. A previous study has suggested that SAC/VAL possesses the capacity to ameliorate Ang induced brosis by inhibiting TGF-β dependent brotic processes in streptozotocininduced diabetic mice model [34]. Hence, to clarify the mechanisms of the protective effects of SAC/VAL, we detected the relative expression of Smad proteins and MAPK proteins. In our study, we observed that Ang stimulation signi cantly enhanced the expression of p-Smad2/3 in saline-treated rats, while both VAL and SAC/VAL treatment reduced the level of p-Smad2/3. When comparing the latter two groups, it was found that SAC/VAL reduced the expression of p-Smad2/3 more signi cantly. The results suggested that decreasing the phosphorylation of Smad2/3 may the mechanism by which SAC/VAL exerts its role in antagonizing Ang -induced brosis. In addition to the above canonical pathways, previous studies have demonstrated that SAC/VAL could inhibit the phosphorylation of JNK and p38MAPK in experimental models of diabetic cardiomyopathy [38]. Our study also found the pronouncedly high level of p-JNK, p-p38MAPK, and p-ERK1/2 in the Ang induced AF model. After 4 weeks of treatment with VAL or SAC/VAL, these proteins' expression levels declined dramatically. Moreover, the expression levels of p-JNK and p-p38MAPK, not p-ERK1/2, were signi cantly lower in the SAC/VAL group compared to the VAL group, which may explain the reason for the superiority of SAC/VAL over VAL in attenuating atrial brosis and reducing AF inducibility.

Conclusion
In summary, our study revealed part of the mechanisms by which SAC/VAL inhibits atrial brosis and decreases AF inducibility, which is bene cial for enriching AF treatment regimens. We also demonstrated that SAC/VAL is superior to VAL in reducing atrial structural remodeling and attenuated AF susceptibility, possibly in part through inhibition of the p-Smad2/3, p-p38 MAPK, and p-JNK pathways.      quanti cation (left) using anti-phospho-Smad2/3, anti-phospho-ERK1/2, anti-ERK1/2, anti-phospho-p38, anti-p38, anti-phospho-JNK, and anti-JNK. GAPDH was used as loading control. Data are expressed as the mean ± SD (n= 6 samples per group). *P < 0.05 vs another group; ns, not signi cant. p-Smad2/3,