Inhibition of Interleukin-6/glycoprotein 130 signalling by Bazedoxifene ameliorates cardiac remodelling in pressure overload mice.

Abstract The role of IL‐6 signalling in hypertensive heart disease and its sequelae is controversial. Our group demonstrated that Bazedoxifene suppressed IL‐6/gp130 signalling in cancer cells but its effect on myocardial pathology induced by pressure overload is still unknown. We explored whether Bazedoxifene could confer benefits in wild‐type C57BL/6J mice suffering from transverse aortic constriction (TAC) and the potential mechanisms in H9c2 myoblasts. Mice were randomized into three groups (Sham, TAC, TAC+Bazedoxifene, n = 10). Morphological and histological observations suggested TAC aggravated myocardial remodelling while long‐term intake of Bazedoxifene (5 mg/kg, intragastric) attenuated pressure overload‐induced pathology. Echocardiographic results indicated Bazedoxifene rescued cardiac function in part. We found Bazedoxifene decreased the mRNA expression of IL‐6, MMP2, Col1A1, Col3A1 and periostin in murine hearts after 8‐week surgery. By Western blot detection, we found Bazedoxifene exhibited an inhibition of STAT3 activation in mice three hours and 8 weeks after TAC. Acute TAC stress (3 hours) led to down‐regulated ratio of LC3‐Ⅱ/LC3‐Ⅰ, while in mice after long‐term (8 weeks) TAC this ratio becomes higher than that in Sham mice. Bazedoxifene inverted the autophagic alteration induced by TAC at both two time‐points. In H9c2 myoblasts, Bazedoxifene suppressed the IL‐6‐induced STAT3 activation. Moreover, IL‐6 reduced the ratio of LC3‐Ⅱ/LC3‐Ⅰ, promoted P62 expression but Bazedoxifene reversed both changes in H9c2 cells. Our data suggested Bazedoxifene inhibited IL‐6/gp130 signalling and protected against cardiac remodelling together with function deterioration in TAC mice.


| INTRODUC TI ON
Based on data from National Health and Nutrition Examination Survey (NHANES) 2011-2014, 6.5 million American adults suffered heart failure (HF), 1 of which the projected prevalence is expected to increase 46% from 2012 to 2030. 2 While the risk factors of HF vary with geography, hypertension is considered to be highly associated with HF in almost all regions worldwide. 3 Therefore, humans pay more attention to the prevalent magnitude of hypertensive heart disease (HHD). Its cardiovascular sequelae include myocardial hypertrophy, remodelling and the subsequent congestive HF. As a common feature seen in patients with hypertension or aortic stenosis, compensatory myocardial hypertrophy develops to restore the cardiac function in initial stage. 4 However, pressure overload ultimately leads to HF characterized by cardiac dysfunction, structural remodelling and fibrosis in both interstitium and vasculature. [5][6][7] Despite of significant advance in experimental research and clinical treatment to the spectrum of HHD, millions of patients still suffer from a gradual process of HF which accounts for the inordinate mortality. 8 A wealth of information gives priority to angiotensin-converting enzyme inhibitors (ACEIs)/angiotensin receptor blockers (ARBs) by a wide margin in the treatment of cardiac hypertrophy and heart failure but numerous patients are still not free from the cardiac disturbance. Gerber et al 9 reported that from 2000 to 2010 the advance in improving survival kept levelling off and the overall 5-year mortality of HF remained high at about 50% in Olmsted County, Minnesota. Compared with placebo, ARBs and direct renin inhibitors (DRIs) are not beneficial in decreasing the rate of efficacy outcomes including death, cardiovascular death, non-fatal myocardial infarction, HF hospitalization and composite of cardiovascular death or HF hospitalization when HF therapy is not enough. 10 Given the poor prognosis and the staggering social burden HHD and HF exert, a better understanding of the precise mechanism of HHD and HF merits accelerated investigation. 11 Accumulating evidence suggested HHD is associated with inflammation. Levels of pro-inflammatory cytokines have been found elevated in circulating blood in patients with heart failure. 5 As one of the pro-inflammatory cytokines, IL-6 activates JAKs and STAT3 by recognizing its receptor (IL-6R) and co-receptor gp130, all of which play important roles in inflammation. IL-6/gp130 signalling has been implicated in cardiovascular diseases. 12,13 There is ample evidence that IL-6 is elevated during heart failure progression, which highlights IL-6 as a potential target for intervention. 14,15 Studies have shown that infusion with IL-6 leads to concentric cardiac hypertrophy and dysfunction in rats. 16 The mRNA expression of IL-6 in the myocardial tissues was elevated in TAC mice where abrupt pressure overload on left ventricle (LV) was produced. 17,18 Cardiac myocytes produce IL-6 spontaneously, and persistent activation of gp130 is associated with cardiac hypertrophy in mice. 19 However, the role of IL-6 signalling is under debate due to different effects of IL-6 inhibition on setting the trajectory for HHD progression. IL-6 protects the heart from injury or does harm to myocardium depending on the kinetics of the host response. This cytokine could drive protective response in the short term in myocardial infarction but sustained elevated level of IL-6 damages to the heart by resulting in chronic inflammation and fibrotic pathology. 20 Several in vivo studies using IL-6 knockout mice come to a divergence because both the negative 21,22 and positive 23 effects of this genetic manipulation on the development of cardiac remodelling and dysfunction were observed. The existing discrepancies suggest a better understanding of the role of IL-6/gp130 signalling in HHD is needed to elucidate whether regulation of this target could potentially affect the maladaptive cardiac development after an insult of pressure overload.
Previously our group reported that Bazedoxifene could inhibit phosphorylation of STAT3 by targeting IL-6/gp130 interface in pancreatic cancer cells, 24 which implied the potential effect of Bazedoxifene on IL-6/gp130 signalling in cardiovascular diseases.
Here, we found Bazedoxifene preserved cardiac function, and ameliorated myocardial remodelling caused by TAC in mice. Moreover, Bazedoxifene executed an inhibition of p-STAT3 in H9c2 myoblasts stimulated with IL-6 and in pressure overload mice at both the earlier and later stage. Moreover, Bazedoxifene might serve as another potent approach for manipulation of autophagy in vitro and in vivo.

| Animals and reagent
All animal experiments were performed in accordance with the Institutional Animal Use and Care Committee of Tongji Medical College, Tongji Hospital, Huazhong University of Science and Technology. Male wild-type C57BL/6J mice (8 weeks old and ≈25 g) were purchased from the Jackson Laboratory. Mice were randomly assigned to three groups (n = 10 per group for long-term observation and n = 5-6 per group for acute window exploration): (a) sham group; (b) TAC group; (c) TAC+Bazedoxifene (hereinafter referred as BAZ) group. Murine TAC was performed to apply mechanical strain induced by pressure overload derived from aortic constriction on left ventricle as referenced. 25 Mice were anaesthetized by 2% isoflurane mixed with 0.5-1.0 L/min 100% O 2 . PE 90 tube was used to perform endotracheal intubation, then mechanical ventilation around 130 breaths/min and a tidal volume of 0.2-0.3 mL was performed. Mice were placed in a supine position on a warming pad.
H9c2 cells. Our data suggested Bazedoxifene inhibited IL-6/gp130 signalling and protected against cardiac remodelling together with function deterioration in TAC mice.

K E Y W O R D S
Bazedoxifene, cardiac remodelling, interleukin-6, transverse aortic constriction Partial thoracotomy to the second rib was then performed using a chest retractor to retract sternum. The aortic arch was exposed and constricted by a 7-0 silk ligature tied against a 27 1

| Histology and immunohistochemistry
Harvested hearts were embedded in paraffin or stored at −80°C before cutting. The embedded tissues were then sectioned and stained with haematoxylin-eosin, picrosirius red or Massion's trichrome staining according to the manufacturer's protocols. The interstitial fields without any blood vessels and the perivascular fibrosis from the fields containing vasculature were imaged. For immunohistochemistry, we used antibody against IL-6 (D220828, Sangon) to detect the protein expression in LV tissues according to the manufacturer's protocols. Image capture was operated using EVOS FL Auto Imaging System (Life Technologies, ThermoFisher Scientific).

| Echocardiography
Echocardiograms were performed in East Hospital, Tongji University School of Medicine, Shanghai, China. Male wild-type C57BL/6J mice (8 weeks old and ≈25 g) were randomly divided into three groups: (a) sham group, (b) TAC group, (c) TAC+BAZ (n = 4-6 per group). The protocols for gavage was as described above. The echocardiographic parameters were assessed at the indicated time-points and the potential side effect of BAZ on cardiac function was tested in n = 5 healthy mice (Methods S1). Mice were anaesthetized with isoflurane (3% for induction and 2% for maintenance). The gas was mixed in 1 L/min O 2 and administrated via a facemask. Hair on the anterior chest was removed by chemicals and then discarded. All parameters were obtained by an experienced operator and the analysis of raw data was performed on a workstation installed with Vevo 2100 (FUJIFILM VisualSonics) by the same operator who was blind to all the experiment.

| Cell culture and treatment
H9c2 cells were purchased from the American Type Culture Collection and maintained in Dulbecco's modified Eagle's medium (DMEM, KeyGEN BioTECH) high glucose supplemented with 10% foetal bovine serum (FBS, Gibco, ThermoFisher Scientific) and 1% penicillin/streptomycin (Sigma-Aldrich). Cells were plated and cultured for 24 hours in the medium in a humidified 37°C incubator with 5% CO 2 . Prior to treatment Cell treatment in the detection for the transcripts of hypertrophic markers and cell surface area was described as followed.
H9c2 cells were planted in a 6-well plate. Serum starved cells were pre-treated with Bazedoxifene (20 μmol/L) for 1 hour, followed by treatment with IL-6 (25 ng/mL) in the presence of Bazedoxifene for 1 hour. Then, H9c2 cells were cultured with medium containing IL-6 (25 ng/mL) in the absence of Bazedoxifene for 48 hours. After washing with PBS, cells were collected for RNA extraction or fixed with 4% paraformaldehyde (Sigma-Aldrich) for 20 minutes and stained with 0.5% crystal violet (ThermoFisher Scientific) for 30 minutes at room temperature. Excessive crystal violet was discarded, and image capture was operated using EVOS FL Auto Imaging System (Life Technologies, ThermoFisher Scientific) after the plates were dried.
Cell surface area analysis was performed by ImageJ.

| Western blot
The pulverized cardiac tissues and the collected cultured cells were lysed in RIPA lysis buffer containing 1 mmol/L protease inhibitor and 1 mmol/L phosphatase inhibitor. The lysates were centrifuged at 13 800 g for 20 minutes at 4°C, and the supernatant was collected.

| Statistical analysis
Data were expressed as the means ± SEM from triplicated performed experiments. Comparison of multiple groups was analysed by oneway analysis of variance (ANOVA) with Bonferroni's post hoc test.
Statistical significance was defined as P < .05. All statistical analysis was performed with SPSS software (version 22.0). Quantitative assessment of Western blot and relative myocardial fibrosis area was performed by Image J.

| Bazedoxifene attenuated cardiac hypertrophy induced by pressure overload in vivo
To avoid potential bias resulting from different baseline, we chose age-and weight-matched male mice for experiment. Heart tissues were harvested after 4 or 8 weeks of surgery. Gross morphology suggested the surgery of transverse aortic constriction increased the size of heart and Bazedoxifene attenuated this increase at 4 ( Figure 1A,C) and 8 weeks ( Figure 1B,D). We assessed the changes in heart weight (HW). The HW in TAC group (171.00 ± 15.57 mg) significantly increased (P < .001) in mice after 4 weeks of TAC manipulation compared with Sham counterparts (99.07 ± 6.71 mg).
Intriguingly, the heart mass significantly decreased (P < .001) in BAZ group (109.83 ± 7.41 mg) ( Figure 2A). As expected, the heart tissues of mice after 8 weeks of surgery showed a higher mass In addition, longitudinal sections of heart tissues from sacrificed mice after 4 and 8 weeks of TAC were prepared and assessment of myocyte cross-sectional area was performed in each group. After aortic constriction, the transverse cardiomyocyte area was increased than that in Sham group at 4 (P < .01) and 8 weeks (P < .05). Bazedoxifene attenuated this increase significantly at both time-points ( Figure 1E-H, P < .05 at 4 and 8 weeks). To confirm this TAC-induced pathogenesis, we further explored the hypertrophic response at molecular level. It has been reported that brain natriuretic peptide (BNP) is recommended as one of the biomarkers for initial diagnostic tests, and it is suggested to be regarded as exclusion markers according to American Heart Association and European Society of Cardiology. 26 We found the mRNA expression of BNP was increased by 4-(P < .01) and 8-week TAC (P < .001) intervention. BAZ group showed a decreased expression level of BNP at both timepoints ( Figure 1I and J, P < .01 at 4 weeks, P < .05 at 8 weeks). These results suggested administration of Bazedoxifene might alleviate cardiac hypertrophy caused by pressure overload in mice.
Microtubule-associated protein 1 light chain 3 (LC3) is one of the autophagic markers and the role for P62 has been established by many studies in protein turnover through autophagy. We demonstrated the decreased ratio of LC3-Ⅱ/LC3-Ⅰ in heart tissues harvested from the mice after acute pressure overload trigger while the ratio in BAZ mice was higher ( Figure 5D, P < .05). We further noticed the reduced ratio of LC3-Ⅱ/LC3-Ⅰ and increased expression of P62 with IL-6 treatment compared to those in H9c2 cells as control.

| IL-6 contributed to the response to Angiotensin Ⅱ in H9c2 myoblasts
We also found AngⅡ induced the phosphorylation of STAT3 ( Figure 7A  Scatterplot represents means ± SEM including individual data points. *P < .05; **P < .01; ***P < .001, "ns" stands for "None Significance" Given the evidence that AngⅡ is considered as a crucial mediator of HHD [27][28][29] and is the target of developed therapeutic manipulation F I G U R E 4 Bazedoxifene attenuated pressure overload-induced LV fibrosis. LV tissues from mice after 8-week surgery were sectioned and stained with Massion's trichrome (A and B) or picrosirius red (C and D). A-D, scale bar, 200 μm, quantification of relative fibrosis area was performed by Image J, *P < .05 vs sham group; # P < .05 vs TAC group. E-G, mRNA expression profiles of fibrosisrelated genes in murine hearts at 8 wk. The relative abundance of transcripts was quantified and normalized to GAPDH. reported the TAC mice operated using a 27 1/2 gauge needle are expected to develop cardiac hypertrophy within 1-2 weeks and cardiac dilation after 6-8 weeks. Heart of TAC mice has to pump more than normal, and can determine a transient increase in the cardiac contraction, which might be erroneously recorded as an amelioration of his function. Prolonged pressure overload in the long term leads to a gradual maladaptive remodelling, with LV dilatation, tissue scarring and heart failure, which appears evident at 6-8 weeks after surgery. 35 We found that TAC and BAZ had effects on heart mass as well as cardiac hypertrophy already at 4 weeks. The compensatory cardiac hypertrophy at early stage of TAC-induced pathology might account for the reason why TAC negatively effects and Bazedoxifene shows cardioprotective effect on heart function only significantly at 8 weeks after surgery in our study, which sug- in human vascular endothelial cells. 46 In our study, Bazedoxifene led to decreased activation of STAT3 and down-regulated mRNA expression of MMP2 at the later time-point, which might result in a mitigated heart condition rendering down-regulated IL-6 expression as shown in Figure 5A. Additionally, another study shows IL-6-induced STAT3 phosphorylation also promotes the transcription of STAT3, which leads to the accumulation of unphosphorylated STAT3 (u-STAT3) allowing the crosstalk of STAT3 and NF-κB to propagate and amplify the inflammatory response. 47 We did not demonstrate whether this additional transcriptional mechanism involved in our study but it is worth being studied in future. blocks. 48 During autophagosome formation, the nascent LC3 proceeds to become LC3-Ⅰ with latent cytosolic distribution. Lipidation of LC3 is the conversion of cytosolic LC3-Ⅰ to LC3-Ⅱ which is incorporated into the double-membrane structure of autophagosomes and mediates protein incorporation. Therefore, the expression of LC3-Ⅱ/LC3-Ⅰ could be used to evaluate the abundance of autophagosomes. 49,50 P62, known as one of autophagy receptors, harbours both ubiquitin-binding domains and LC3-interacting region conferring the selectivity to cargos for autophagic clearance. However, accumulation of P62 could account for the markedly slowed proteasomal degradation by binding ubiquitinated proteins. 51 Studies found decreased markers of autophagy during an initial (1 week) phase in TAC mice compared with the counterparts 52 and excessive P62 expression is found in aberrant protein aggregates. 53 An opinion suggests that cardiac autophagy at baseline level or short-term activation primes the myocardium to withstand overload pressure but excessive and sustained autophagy might contribute to cardiac malfunction under a hypertrophic insult. 50 Induction of autophagy is considered beneficial in the setting of acute energy stress including fasting, nutrient deprivation and ischaemia-reperfusion. 54 We found the induction of p-STAT3 was coincident with reduction of ratio of LC3-Ⅱ/LC3-Ⅰ in IL-6 stimulated H9c2 cells and in TAC mice at the earlier stage, but at the later stage in TAC mice up-regulation of p-STAT3 was accompanied by increased ratio of LC3-Ⅱ/LC3-Ⅰ.

| D ISCUSS I ON
A review about the understanding of STAT3 in autophagy summarized that different modified or localized STAT3 exhibited pro-or anti-autophagy function in transcription-dependent and transcription-independent manner in various circumstances. 57 Inhibition of STAT3 could lead to induction of lipidation of LC3A resulting in up-regulation of autophagy in glioma cells. 58 However, the mechanisms underlying how STAT3 plays context-dependent role in autophagy is yet to be elucidated. Further studies are required to discern the potential interaction between IL-6 signalling pathway and LC3 or P62 in the nuanced autophagy throughput in pressure overload-induced perturbation.
Recently, another study found circulating IL-6 and tumour necro-

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

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.