Cardiac Ablation of SOCS3 Aggravates DOCA-Salt-Induced Hypertrophic Remodeling by Activation of Gp130-Dependent Signaling in Mice Open Access

Hypertensive cardiac remodeling is a major risk factor for cardiovascular diseases leading to heart failure. The changes include hypertension, increased cardiomyocyte size, fetal gene activation, protein synthesis and collagen deposition ultimately leading to myocardial hypertrophy and electrical conduction changes in the heart [1]. There is substantial evidence that administration of a synthetic mineralocorticoid derivative, deoxycorticosterone acetate (DOCA), in combination with a high salt intake to animals can induce cardiac remodeling, and DOCA-salt animals mimic most of the changes in human volume-overload induced hypertension [2]. However, the DOCA-salt hypertensive rat model shows a significantly decreased circulating rennin concentration and thus has been regarded as an angiotensin-independent model [3]. This may involve multiple signaling pathways, including oxidase stress, mitogen-activated protein kinases, receptor or nonreceptor-associated tyrosine kinases, inflammation and neurohumoral factors, such as endothelin, vasopressin and sympathetic nerves [2].

In this study, we presented novel evidence demonstrating that ablation of SOCS3 in cardiomyocytes significantly aggravated DOCA-salt-induced cardiac remodeling and inflammation, but overexpression of SOCS3 attenuated these pathological changes. These effects were mostly associated with activation of gp130/STAT3/AKT/ERK, TGF-β/Smad2/3 and NF-kB pathways in the hearts. Thus, our results indicate that SOCS3 plays a critical role in DOCA-salt-induced hypertrophic remodeling, and suggest that SOCS3 is a promising therapeutic target for the treatment of hypertensive cardiac diseases.

SOCS3^flox/flox mice were obtained from Jackson laboratories. To delete the SOCS3 in the myocardium, SOCS3^flox/flox mice were mated with mice expressing Cre under α-myosin heavy chain (α-MHC) promoter (α-MHC-Cre). To induce Cre-dependent recombination, tamoxifen (20 mg/kg body weight, Sigma-Aldrich) was injected intraperitoneally (i.p.) for 5 days over a 3-week duration before any further experimentation. All SOCS3^flox/flox (WT) and SOCS3^flox/flox/α-MHC-Cre (SOCS3cKO) mice used in this study were males (10- to 12-week-old) in a C57BL/6J background. Mice were housed under a 12-h light–dark cycle at 18–25°C and received commercial mice chew and ad libitum. All procedures were performed in accordance with the protocol outlined in the Guide for the Care and Use of Laboratory Animals published by the US National Institute of Health (NIH publication No. 85-23, revised 1996) and approved by the Committee on the Ethics of Animal Experiments of Dalian Medical University.

Male mice (10–12 weeks of age) were allocated randomly to 4 groups for treatment (n=8): only uninephrectomized WT mice; uninephrectomized SOCS3cKO mice; DOCA/salt-WT mice and DOCA/salt-SOCS3cKO mice. Mice were anesthetized with sodium pentobarbital intraperitoneally (50 mg/kg). The left kidney was then uninephrectomized. Part of animals received DOCA from subcutaneous pellets (DOCA, 2.5 mg/d; Innovative Research of America) and 0.9% NaCl with drinking water for 3 weeks. Control mice were only uninephrectomized, but DOCA and saline were not given. WT mice were received eplerenone (Pfizer, Inc., New York, NY, USA), which was mixed into the diets at a dose of 1.25 g/kg of chew). The systolic blood pressure was measured by the tail-cuff system (Softron BP98A; Softron Tokyo, Japan) 3 days before and during each of the three days of treatment as previously described [12].

The mouse SOCS3 complementary DNA was constructed to adeno-associated virus (AAV) vector (pAAV-IRES-ZsGreen). The constructed AAV-SOCS3 or control vector and the packaging vectors (pHelper and pAAV-RC) were transfected into 293 AAV cells to package AAV serotypes 9 (AAV9-SOCS3 and AAV9-control) (Biowit Technologies, Shenzhen, China). For in vivo transduction, 100 µl AVV9-SOCS3 or AAV9-control virus particles (1×10¹² viral genomes/ml) were administered to WT mice by tail vein injection. Three weeks later, mice were subjected to DOCA-salt for additional 3 weeks. Transduction efficiency of in vivo gene transfer by AAV was assessed using fluorescence microscopy and western blot analysis.

Animals were anesthetized with isoflurane (3%) and then underwent M-mode echocardiography at 21 days after DOCA-salt treatment by using a 30 MHz probe (Vevo 770 system; VisualSonics, Toronto, Ontario, Canada) as we described previously [13-15]. The ejection fraction (EF), fractional shortening (FS), left ventricular anterior wall thickness at diastole (LVAW;d), left ventricular posterior wall thickness at diastole (LVPW;d), left ventricular inner diameter at diastole (LVID;d), left ventricular anterior wall thickness at systole (LVAW;s), and left ventricular posterior wall thickness at systole (LVPW;s) and left ventricular inner diameter at systole (LVID;s) were measured digitally on the M-mode tracings and averaged from at least 3 separate cardiac cycles. EF(%)=100*((LV Vol;d - LV Vol;s)/LV Vol;d); LV Vol;d(ul)=7.0/(2.4+LVID;d)* LVID;d³; LV Vol;s(ul)=7.0/(2.4+LVID;s)* LVID;s³ [16].

The hearts were quickly dissected out and rinsed with cool sterile saline. The hearts were fixed in 10% formalin for histological analysis. Heart sections (5 µm) were stained with hematoxylin and eosin (H&E), Masson's trichrome and Wheat germ agglutinin (WGA). Immunohistochemistry was performed with anti-Mac-2 antibody (Santa Cruz, CA, 1: 200 dilution) and anti-SOCS3 antibody (Abcam, Cambridge, UK, 1: 100 dilution). Analysis of SOCS3-positive cells, Mac-2-positive cells, myocyte cross-sectional area, fibrotic area was quantitatively performed with NIH Image 1.61 software as previously described [13-15].

Total RNA was extracted from fresh mouse hearts using Trizol method (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. The first-strand cDNA was generated from 1–2µg of total RNA by RT Enzyme mix on a PCR Thermocycler (S1000 Thermal Cycler, Bio-Rad, USA). The mRNA expressions of atrial natriuretic peptide (ANF), B-type natriureticpeptide (BNP) and β-myosin heavy chain (β-MHC), collagen I, collagen III, IL-1β, IL-6, TNF-α and ICAM-1 were normalized to the expression of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) by the ΔΔCt method [17, 18].

Western blot analysis was performed as described previously [12, 13]. Lysates were isolated from heart tissues and quantified with BCA Protein Assay according to the manufacturer's instructions. Proteins (40-50 µg) were subjected to SDS-PAGE, and PVDF membrane (Bio-Rad) were probed with primary antibodies, including gp130, JAK2, p-JAK2 (Tyr1008), STAT3, p-STAT3 (Tyr705), AKT, p-AKT (Ser473), ERK1/2, p-ERK1/2 (Thr202/Tyr204), TGF-β, Smad2/3, p-Smad2/3, α-SMA, TNF-α, p65, p-p65 (Cell Signaling Technology, Danvers, MA, 1: 800-1000 dilution) and GAPDH (Cell Signaling Technology,1: 3000 dilution). Results of Western blots were quantified by the NIH Image J program.

All data were expressed as mean ± SEM. Differences in mean values were assessed using the Student's unpaired t test or two-way ANOVA for factorial design with two factors. Statistical significance was accepted at p< 0.05.

To determine the functional role of SOCS3 in regulating cardiac hypertrophy and remodeling, we first examined the expression of SOCS3 in the heart. After 3 weeks of DOCA-salt treatment, SOCS3 expression was significantly increased in DOCA-salt treated heart compared with control (Fig. 1A). Moreover, immunostaining indicates that SOCS3-positive area was higher in DOCA-salt heart than in control hearts (Fig. 1B). To elucidate the mechanism by which DOCA-salt upregulates SOCS3, we co-treated mice with mineralocorticoid receptor (MR) inhibitor eplerenone and DOCA-salt for 3 weeks. Administration of eplerenone significantly reduced DOCA-salt-induced upregulation of gp130, p-JAK2, p-STAT3 and the downstream target SOCS3 compared with vehicle control (Fig. 1C). Moreover, DOCA-salt-induced increase in cardiac performance, cardiac fibrosis and heart weight/body weight (HW/ BW) ratio was remarkably attenuated in eplerenone-treated mice compared with vehicle control (Fig. 1D-F), suggesting that DOCA-salt induced SOCS3 expression and cardiac remodeling through MR-gp130 signaling.

To determine whether cardiac-specific deletion of SOCS3 affects cardiac remodeling and contractile performance in vivo. Echocardiography was performed. After 3 weeks of DOCA-salt treatment, no cardiac arrhythmia or sudden death was observed in SOCS3cKO mice. Systolic blood pressure was similar between WT mice and SOCS3cKO mice (Fig. 2A). However, DOCA-salt treatment resulted in increased cardiac performance as indicated by EF% and FS%, and this adaptive response was reduced in SOCS3cKO mice (Fig. 2B-D). In addition, DOCA-saltinduced increase of systolic and diastolic LVAW and LVPW was enhanced in SOCS3cKO mice (Fig. 2E-H). In contrast, systolic and diastolic LVID was markedly decreased in SOCS3cKO mice (Fig. 2 I-J). There was no significant difference in these parameters between WT and SOCS3cKO mice after control treatment (Fig. 2C-J).

DOCA-salt treatment significantly increased the heart size, LV wall thickness, ratios of HW/BW and heart weight/tibia length (HW/TL) in WT mice, and this was further exacerbated in DOCA-salt-treated SOCS3cKO mice (Fig. 3A and 3B). Moreover, a significant increase in cardiomyocyte cross-sectional area and the expression of ANF, BNP, and β-MHC was observed in SOCS3cKO mice as compared with WT mice after DOCA-salt treatment (Fig. 3C and 3D). There was no significant difference in cardiac hypertrophy between control groups (Fig. 3A-D).

Since SOCS3 is a key negative-feedback regulator of gp130-JAK-STAT3 pathway, we then tested whether SOCS3 deficiency enhances gp130 mediated signaling. As expected, DOCA-salt induced activation of gp130, JAK2, STAT3, AKT and ERK1/2 in the hearts of WT mice was further augmented in SOCS3cKO mice (Fig. 3E). Although there was a slight increase in gp130 and activation of their downstream mediators under control condition, no significant statistically difference was observed between WT and SOCS3cKO mice (Fig. 3E).

Masson’s trichrome staining revealed that DOCA-salt treatment significantly increased left ventricular perivascular and interstitial fibrosis as compared with WT mice in control groups, and this was markedly aggravated in DOCAsalt SOCS3cKO mice (Fig. 4A). Likewise, the increase in the mRNA expression of collagen I, collagen III and α-SMA in the DOCA-salt WT mice was also significantly accelerated in DOCAsalt SOCS3cKO mice (Fig. 4B). No significant difference was observed in these parameters between control groups (Fig. 4A and 4B). Moreover, DOCA-salt treatment activated TGF-β and Smad2/3 pathways in the heart of WT mice, which was further enhanced in SOCS3cKO mice (Fig. 4C).

To determine the degree of inflammatory response in the SOCS3cKO hearts, we performed H&E and immunohistochemical staining. DOCA-salt treatment caused a marked increase in the infiltration of perivascular and interstitial proinflammatory cells, including Mac-2-positive macrophages in the DOCA-salt WT mice compared with control, and this increase was further accelerated in DOCA-salt SOCS3cKO mice (Fig. 5A and 5B). Accordingly, the mRNA expression of IL-1β, IL-6, TNF-α and ICAM-1 was significant higher in DOCA-salt SOCS3cKO than in DOCA-salt WT mice (Fig. 5C). Moreover, DOCA-salt-induced increase of TNF-α protein level and p65 phosphorylation were markedly higher in SOCS3cKO than in WT mice (Fig. 5D).

To define the effect of SOCS3 overexpression on the heart in vivo, WT mice were delivered with AAV9-SOCS3 or AAV9-control, and then subjected to DOCA-salt for 3 weeks. AAV9 injection resulted in a higher infection efficiency (Fig. 6A), and an increase of 2-fold SOCS3 protein in the hearts (Fig. 6B). Moreover, SOCS3 overexpression improved DOCA-salt-induced cardiac dysfunction (EF% and FS%) compared with AAV9-control mice (Fig. 6C). Furthermore, DOCA-salt-induced increase in the heart size, HW/BW and HW/TL ratios, cardiomyocyte cross-sectional area and fibrotic area were also attenuated in AAV9-SOCS3 mice (Fig. 7A-D). In addition, AAV9-SOCS3 injection markedly reduced gp130 level and phosphorylation of JAK2 and STAT3 after DOCA-salt treatment compared with AAV9-control groups (Fig. 7E). There was no significant difference in cardiac remodeling and signaling between control groups (Fig. 7A-E).

In this study, we sought to determine the pathophysiological role of SOCS3 in DOCA-salt-induced cardiac remodeling using a SOCS3cKO animal model. We demonstrated that SOCS3 expression was significantly increased in DOCA-salt-treated heart through MR/gp130 signaling. Moreover, DOCA-salt-induced increase of cardiac remodeling were markedly aggravated in SOCS3cKO mice but attenuated in AAV9-SOCS3-injected mice. These effects were mostly associated with activation of gp130/JAK/STAT3/ERK1/2 and AKT, TGF-β/ Smad2/3 and TNF-α/NF-κB pathways.

The SOCS proteins are identified as the target of the JAK/STAT pathway, forming a negative-feedback loop to inhibit signal propagation. They can also be induced by other signaling pathways independent of JAK/STAT signaling. Increasing evidence indicates that the list of inducers of SOCS proteins in the myocardium is growing, such as IL-6, INF-γ, TNF-α, and angiotensin II [4, 19]. In addition, SOCS3 is markedly induced during acute and chronic hypertrophic response following pressure overload [10]. In this study, we found that DOCA-salt-induced SOCS3 expression, activation of gp130/JAK/STAT3, cardiac dysfunction and remodeling were markedly attenuated by MR inhibitor eplerenone (Fig. 1), indicating that MR mediated DOCA-salt-induced SOCS3 expression and cardiac remodeling.

DOCA-salt rats have been reported to mimic most of the changes observed in chronic cardiac remodeling in humans including hypertension, hypertrophy, fibrosis and altered function. Furthermore, inflammation is strongly implicated in inducing cardiac hypertrophy and fibrosis in DOCA-salt animal models [20-22], which were confirmed by our findings (Fig. 2-5). Notably, inhibition of monocyte/macrophage infiltration with fasudil or tranilast markedly attenuated cardiac fibrosis in DOCA-salt rats [23, 24]. Overexpression of SOCS3 markedly reduced DOCA-salt-induced cardiac dysfunction and remodeling (Fig. 6, 7). Conversely, SOCS3 knockout aggravated DOCA-salt-induced response (Fig. 2-5). Thus, SOCS3 plays a critical role in regulating hypertrophic remodeling and inflammation in the hearts induced by DOCA-salt stress.

It is well known that SOCS3 can directly interact with a gp130-JAK1 complex thereby regulating three major downstream pathways: JAK/STAT, Ras/MEK/ERK and PI3K/ AKT, which play different roles in the heart [10, 11]. For example, deletion of SOCS3 in cardiomyocytes induces JAK-STAT-mediated cytokine expression, such as leukemia inhibitory factor (LIF) and G-CSF, thereby preventing acute myocardial infarction-induced left ventricular (LV) remodeling [25]. In contrast, continuous activation of gp130-STAT3 signaling aggravates inflammation and LV rupture after myocardial infarction [26]. These data suggest that gp130-JAK-STAT signaling pathway exerts anti-apoptotic role in cardiomyocytes in response to ischemic stress. Interestingly, in response to hypertrophic stress such as pressure overload, deletion of SOCS3 activates cardiac gp130 signaling including JAK-STAT, ERK1/2 and AKT results in cardiac hypertrophy and contractile dysfunction [11]. Similarly, our results also demonstrated that DOCA-salt-induced activation of gp130/JAK2/STAT3 signaling was enhanced in SOCS3cKO mice (Fig. 3), but inhibited in AAV9-SOCS3-treated hearts (Fig. 7). Overall, these data suggest that inhibition or activation of SOCS3 could be selected depending on the type of diseases. TGF-β and its canonical downstream signaling mediator Smad are central to the development of progression of fibrosis [27]. Interestingly, IL-6/gp130 contributes to lung fibrosis via TGF-β/Smad2/3 signaling [28]. Consistent with this study, our results further confirmed that SOCS3 deletion enhanced gp130-mediated activation of TGF-β/Smad2/3 signaling leading to myocardial fibrosis (Fig. 4). NF-κB pathway is pivotal in the production of inflammatory mediators [29, 30]. Inhibition of NF-κB by Fenofibrate, an activator of PPAR-α, significantly reduces cardiac fibrosis and improve function in DOCA-salt rat hearts [31]. In agreement with these results, our results showed that SOCS3 ablation also increased gp130-mediated activation of TNF-α and NF-κB signaling, causing increased inflammation in the hearts (Fig. 5). Collectively, cardiac deletion of SOCS3 promotes DOCA-salt-induced remodeling through gp130-mediated pathways.

In conclusion: We provide novel evidence demonstrating that SOCS3 plays a critical role in the development of DOCA-salt-induced cardiac remodeling. Ablation of SOCS3 aggravated DOCA-salt-induced cardiac dysfunction and remodeling via activation of multiple signaling pathways, whereas overexpression of SOCS3 ameliorated these effects. Thus, our data identify SOCS3 as a new potential therapeutic target for treating hypertensive heart diseases.

We thank Qing Xu for her excellent technical assistance in echocardiography. This work was supported by grants from China National Natural Science Funds (81330003, 81630009, 81500207, 31400452), International S&T Cooperation Program of China (2014DFA31930), and Chang Jiang Scholar Program (T2011160).

No conflict of interests exists.