Abstract
Aldosterone, as a major product of renin–angiotensin–aldosterone system (RAAS), determines multiple pathophysiological processes in cardiovascular diseases. The excess inflammatory response is one of the key profiles in aldosterone-mediated cardiac remodeling. However, the potential mechanisms of aldosterone/inflammatory signaling were still not fully disclosed. The present study aimed to investigate whether TIR-domain-containing adapter-inducing interferon-β (Trif) participated in the aldosterone-induced cardiac remodeling, and to explore potential molecular mechanisms. Trif knockout mice and their littermates were osmotically administrated with aldosterone (50 μg/kg per day) for 21 and 42 days. The cardiac structural analysis, functional parameters, and mitochondrial function were measured. Aldosterone dose- or time-dependently increased the levels of TRIF in primary mouse cardiomyocytes or mouse heart tissues. Trif deficiency protected against aldosterone-induced cardiac hypertrophy, fibrosis and dysfunction. Moreover, Trif deficiency also suppressed aldosterone-induced cardiac inflammatory response and mitochondrial injuries. Mechanistically, overexpression of cardiac microRNAs (miR)-34a reversed the cardiac benefits of Trif deficiency in aldosterone-treated mice. Taken together, Trif/miR-34a axis could provide a novel molecular mechanism for explaining aldosterone-induced cardiac hypertrophy, fibrosis and functional disorders.
Introduction
Cardiac pathophysiological remodeling is one of the major characteristics in the development of heart failure. There has been a sustained effort to develop novel strategies to protect against cardiomyopathy. Among them, agents that target the renin–angiotensin–aldosterone system (RAAS) are widely prescribed to combat heart failure [1]. Angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and mineralocorticoid receptor antagonists have significant therapeutic benefits in alleviating the morbidity and mortality of heart failure [2,3]. Although clinical trials and animal experiments have provided a variety of molecular findings, the underlying signaling pathways are still not fully understood. Therefore, it is necessary to disclose the mechanisms that RAAS affects cardiac structural and functional remodeling and the remaining challenges.
Aldosterone, as a prognostic and mediator of heart diseases, was significantly increased in patients from early cardiac injuries to more advanced heart failure [4]. Specific binding to the mineralocorticoid receptor (MR), aldosterone-activated multiple pathways to impair cardiac biological processes independently of blood pressure effects. Experimental findings showed that aldosterone-salt diet-induced rat cardiac fibrosis [5]. Aldosterone-salt-treated animals also exhibited progressive inflammatory phenotypes during cardiac hypertrophy [6,7]. In contrast, administration of spironolactone, a first-line of mineralocorticoid receptor antagonist, reduced the risk of mortality from any cause in patients with NYHA class III or IV heart failure [8]. Mechanistically, the inhibition of aldosterone/mineralocorticoid receptor binding activities blocked the downstream molecular signaling, including pro-inflammation, oxidative stress, and calcium accumulation [9–11]. Among them, abnormal regulation of microRNAs (miRs) could determine the pathophysiological process of aldosterone in many tissues. MiRs, such as miR-21, miR-24 and miR-204, mediated tissue injuries and dysfunction in aldosterone-treated animals [12–14]. All these potential mechanisms supported the therapeutic benefits of pharmacological MR antagonism in treating heart failure.
Meanwhile, RAAS has been reported to activate Toll-like receptors (TLRs) as well as its downstream signaling cascade, such as myeloid differentiation primary response (MyD) 88 (MyD88) and TIR-domain-containing adapter-inducing interferon-β (Trif). In Ang II-infused mouse models, TLRs/MyD88/Trif pathways controlled the process of cardiac remodeling [15]. Trif deficiency attenuated Ang II-induced cardiac structural remodeling with down-regulation of cardiac expression of inflammatory genes [16]. According to multiple studies, activation of Trif stimulated inflammation and fibrosis in different tissues [17,18]. Due to the wide distribution in different organs and types of cells, Trif exhibited complicated physiological effects. However, it was unclear how Trif signaling affected RAAS-mediated cardiac remodeling. Therefore, it is necessary to specifically address the pathophysiological roles of cardiac Trif and potential molecular mechanisms in aldosterone-mediated cardiac biology.
To delineate the role of Trif in aldosterone-mediated cardiac remodeling, our present study utilized the genetic Trif knockout mouse model to disclose the potential molecular mechanisms. Our findings will link aldosterone and Trif signaling in cardiac remodeling, and provide evidence that therapeutic benefits of Trif inhibition in protecting against abnormal RAAS-induced heart failure.
Materials and methods
Reagents
The Hematoxylin and Eosin solution was purchased from Sigma chemicals (Sigma, St. Louis, U.S.A.). The Sirius Red (SR) stain kit was purchased from Abcam (#ab150681, Cambridge, U.K.). Anti-TRIF (#ab13810), anti-TGF-β1 (#ab92486), anti-phosphorylated (phos)-SMAD3 (#ab52903), anti-SMAD3 (#ab40854), anti-phos-IFN regulatory factor (IRF) 3 (IRF3) (#ab76493), IRF3 (#ab68481), and anti-Tubulin (#ab59680) antibodies were from Abcam (Cambridge, U.K.). Anti-atrial natriuretic peptide (ANP) and anti-brain natriuretic peptide (BNP) antibodies were from Novus (Novus Biologicals, CO). IFN-α antibody was purchased from Invitrogen (Thermo Fisher, MA). Measurement kit for cellular adenosine triphosphate (ATP) level was from Sigma chemicals (#213-579-1). Mouse kits for TNF-α (#MTA00B), IL-1β (#MLB00C), and nitric oxide (#KGE001) were from R&D (R&D Systems, MN). The adeno-associated viruses (AAVs) encoding miR-34 mimic and control with cardiac-specific α myosin-heavy chain (Myh6) promoter, and lentivirus encoding miR-34a sponge and control were a gift from Dr. Qian Ni (Tianjin University).
Primary cardiomyocyte isolation and treatment
Two to three days old neonatal mice were killed for isolating primary cardiomyocytes as previous procedure [19]. A total of 1 × 106 mouse primary cardiomyocytes were treated with different concentrations (0, 10−6, 10−5, 10−4 mol/l) of aldosterone for 24 h.
Animal experiment
All mouse experimental protocols were approved by the Institutional Animal Use and Care Committee at the Harbin Medical University (Harbin, China). Male Trif knockout mice and their littermates (the Jackson Laboratory), aged 8–10 weeks, were subcutaneously administrated with aldosterone (50 μg/kg per day, dissolved in saline) for 21 or 42 days by osmotic pumps. For mouse experiments of cardiac miR-34a overexpression, Trif knockout mice were tail-vein injected with 1 × 109 AAV encoding miR-34a mimic or control with cardiac-specific Myh6 promoter for 14 days, then treated with aldosterone (50 μg/kg per day) by using osmotic pump for 21 days. Systolic arterial blood pressure was measured at 0, 7, 21 days after infusion with aldosterone by tail-cuff using the Visitech-2000 system. For mose experiments of cardiac miR-34a inhibition, C57BL/6J mice were locally injected with 1 × 1012 lentivirus endocind miR-34a sponge or control with cardiac-specific Myh6 promoter for 14 days, then treated with aldosterone (50 μg/kg per day) by using osmotic pump for 21 days. Mice were peritoneally injected with dormicum (6.25 mg/kg) for euthanasia, then the blood and hearts were collected for further analysis. All animal experiments took place in the Shufeng Li’s lab at Harbin Medical University.
Echocardiography analysis of cardiac function
All mice underwent echocardiographic measurement by using the Vevo 2100 system (Vevo). Echocardiographic parameters included heart rate, ejection fraction (EF), fractional shortening (FS), left ventricular anterior wall (LVAW), left ventricular internal dimension (LVID), left ventricle posterior wall thickness (LVPW), and left ventricle mass (LV mass).
Histological analysis of cardiac tissues
Heart tissues were fixed in 4% paraformaldehyde for 48 h and embedded in paraffin. Five-micrometer-thick paraffin sections were prepared and stained with Hematoxylin and Eosin (H&E) solution. To investigate the histological changes, the cardiac images were observed under a light microscope (magnification ×100, Nikon, Tokyo). The average cardiomyocyte size was calculated in 20 cells/field (10 fields per mouse). For analyzing cardiac fibrosis, hydrated cardiac sections were stained with 0.1% SR F3B and 1.3% saturated aqueous solution of picric acid to evaluate the type IV collagen collection. The relative expression of collagen was measured by density analysis in captured images (magnification ×200, 10 fields per mouse). The calculation of histological analysis was done in a blinded fashion by two individuals.
Western blot analysis
Fifty micrograms of protein were run on 10% SDS/PAGE, and then transferred to a 0.22-μm polyvinylidene difluoride membrane (PVDF, Amersham Biosciences). The membrane was blocked for 1.5 h at room temperature with 10% milk in phosphate-buffered saline/0.05% Tween 20. The membrane was incubated overnight at 4°C with anti-TRIF, anti-TGF-β1, anti-SMAD3, anti-IRF3, or anti-Tubulin antibody and secondary antibodies (Cell Signaling, Danvers, MA). The protein expression was visualized using enhanced chemiluminescence reagents (Bio-Rad, Hercules, CA). The amounts of the proteins were analyzed using ImageJ analysis software version 1.38e.
Real-time quantitative PCR analysis
Total RNA in heart tissues or primary cardiomyocytes was extracted by using TRIzol (Invitrogen, Shanghai) according to the manufacturer’s protocol. Reverse transcription was performed using the Superscript III Reverse Transcription System (Invitrogen), and real-time PCR analysis was performed using SYBR Green (Applied Biosystems, Alameda, CA). The sequence of primers were listed as following: Trif, forward 5′-AGGACAAACGCCGGAACTTTT-3′, reverse 5′-GCCGATAGTCTGTCTGTTCTAGT-3′; Vdac1, forward 5′- TGCCTCCCACATATGCTGATC-3′, reverse 5′-TCCGCCAGGGATGTCTT-3′; Pgc1α, forward 5′-GGACATGTGCAGCCAAGACTCT-3′, reverse 5′-CACTTCAATCCACCCAGAAAGCT-3′; Nrf1, forward 5′-GACCCAAACTGAACACATGGC-3′, reverse 5′-GTGCTCAGTGTACGTGGCTA-3′; Tgf-β1, forward 5′-CGCCTCTATGAGAAAACC-3′, reverse 5′-GTAACGCCAGGAATTGT-3′; Col6α1, forward 5′-GATGAGGGTGAAGTGGGAGA-3′, reverse 5′-CAGCACGAAGAGGATGTCAA-3′; MMP12, forward 5′-ACATTTCGCCTCTCTGCTGATGAC-3′, reverse 5′-CAGAAACCTTCAGCCAGAAGAACC-3′; Anp, forward 5′-GCCGGTAGAAGATGAGGTCA-3′, reverese 5′-AGCTGGATCTTCGTAGGCTC-3′; Bnp, forward 5′-GGCCTCACAAAAGAACACCC-3′, reverese 5′-TTACAGCCCAAACGACTGAC-3′; Gapdh, forward 5′-AGGAGCGAGACCCCACTAAC-3′, reverse 5′-GATGACCCTTTTGGCTCCAC-3′. The relative expression was normalized to Gapdh. For miRNA analysis, 200 ng RNA was reversed using the TaqMan MicroRNA Reverse Transcription Kit (Thermo Fisher), and real-time PCR analysis was performed using TaqMan Assay (Applied Biosystems). The relative expression of miR-34a was normalized to sno202.
Mitochondrial activity measurement
Cardiac ATP was measured using an ATP measurement kit (Molecular Probes, Carlsbad) [20]. The endogenous basal oxygen consumption was measured with a Clark electrode in a water-jacketed chamber connected to a circulating water bath (Hansatech, Norfolk, U.K.). The mitochondrial citrate synthase activity was measured by using Citrate Synthase Activity Assay Kit (Abcam, #ab119692).
Flow cytometry analysis of infiltrated immune cells
The cells from hearts were isolated by enzyme digestion [21]. Briefly, left ventricle was collected, and cut into small strips. Then the tissue was placed in a digestion buffer containing 1.5 mg/ml of Collagenase A (Roche) and 0.4 mg/ml DNaseI (Roche) in HBSS plus 5% fetal bovine serum and 10 mM HEPES. After two-round digestion, the cells were extracted by centrifuge. For toxic T analysis, the cells were stained with anti-Cd45 and anti-Cd3 antibodies. For macrophage analysis, the cells were stained with anti-Cd11b and anti-F4/80 antibodies. The result was recorded by calculating the positive cells in 105 heart cells.
Statistical analysis
Data were presented as mean ± SD. Student’s t test was used for comparing two groups, and one-way ANOVA was used for comparing multiple groups. GraphPad Prism 5 (GraphPad, San Diego, CA) was used to analyze the statistical significance between sets of data. Differences were considered to be significant at P<0.05.
Results
Cardiac TRIF participates in aldosterone-induced cardiac dysfunction
Aldosterone, one of the end products of RAAS signaling, has multiple biological effects in cardiovascular remodeling [4]. In pathophysiological process, aldosterone causes inflammation leading to fibrosis and remodeling in the heart and vasculature [5–7]. TLR4 is involved in aldosterone-induced cardiac and kidney injuries [22]. Therefore, we further explored the possible role of TRIF in aldosterone-mediated cardiac remodeling. In primary mouse cardiomyocytes, aldosterone dose-dependently increased expression of Trif (Figure 1A) and TRIF protein levels (Figure 1B,C). Furthermore, mice were treated with aldosterone for 0, 3, 7, and 21 days. As Figure 1D–F showed, aldosterone also time-dependently increased cardiac expression of Trif (Figure 1D) and TRIF protein levels (Figure 1E,F). Chronic higher level of aldosterone impaired cardiac function, including reduction in EF and FS [23]. Interestingly, as Figure 1G,H showed, the increasing level of cardiac Trif was negatively correlated with EF (r = −0.7032, P<0.01) and FS (r = −0.8323, P<0.001). These results indicated up-regulation of cardiac TRIF might modulate aldosterone-induced cardiac dysfunction.
The aldosterone-induced TRIF level was positively correlated with cardiac dysfunction
(A–C) A total of 1 × 106 mouse primary cardiomyocytes were treated with different concentration (0, 10−6, 10−5, 10−4 mol/l) of aldosterone (Aldo) for 24 h. Real-time PCR analysis of Trif level (A), and immunoblot analysis of TRIF (B) and quantitative analysis of relative protein expression (C). (D,E) Male C57BL/6N mice were treated with aldosterone (50 μg/kg per day) by using osmotic pump for 0, 3, 7 and 21 days. Real-time PCR analysis of Trif level (D), and immunoblot analysis of TRIF (E) and quantitative analysis of relative protein expression (F). (G,H). Echocardiographic analysis of mouse cardiac EF and FS. Correlation between cardiac gene level of Trif and EF (G) and FS (H). Results were shown as mean ± SEM, and n=5–7 independent experiments or 5 mice/group. *P<0.05, **P<0.01, ***P<0.001.
(A–C) A total of 1 × 106 mouse primary cardiomyocytes were treated with different concentration (0, 10−6, 10−5, 10−4 mol/l) of aldosterone (Aldo) for 24 h. Real-time PCR analysis of Trif level (A), and immunoblot analysis of TRIF (B) and quantitative analysis of relative protein expression (C). (D,E) Male C57BL/6N mice were treated with aldosterone (50 μg/kg per day) by using osmotic pump for 0, 3, 7 and 21 days. Real-time PCR analysis of Trif level (D), and immunoblot analysis of TRIF (E) and quantitative analysis of relative protein expression (F). (G,H). Echocardiographic analysis of mouse cardiac EF and FS. Correlation between cardiac gene level of Trif and EF (G) and FS (H). Results were shown as mean ± SEM, and n=5–7 independent experiments or 5 mice/group. *P<0.05, **P<0.01, ***P<0.001.
TRIF deficiency protects against aldosterone-induced cardiac hypertrophy, fibrosis, inflammatory response, and mitochondrial dysfunction
To identify the role of TRIF in aldosterone-induced cardiac remodeling, we next treated TRIF knockout mice with aldosterone for 21 days. As shown in Figure 2A,B, aldosterone-induced cardiac enlargement (Figure 2A), and increased the ratio of heart weight/tibia length (Figure 2B, P<0.001). Consistently, aldosterone also increased the average size of cardiomyocytes (Figure 2C,D, P<0.001). However, TRIF deficiency attenuated aldosterone-induced cardiac hypertrophy, heart weight/tibia length ratio and average cell size (Figure 2A–D, P<0.01). A variety of studies have demonstrated that cardiac remodeling was affected by several essential factors, including ANP, BNP, and myosin heavy chain (MHC) [24–26]. Therefore, we detected the expression of these markers in heart tissues. As showed in Supplementary Figure S1A–C, Trif deficiency effectively decreased aldosterone-induced cardiac levels of ANP, BNP, and MHC. Furthermore, aldosterone induced more severe cardiac injuries in 42 days (Supplementary Figure S2A–C), but TRIF deficiency effectively reversed aldosterone-induced cardiac hypertrophy. Tissue fibrosis is one of key characters in cardiac remodeling [27]. SR) stainin showed aldosterone obviously accumulated collagen deposit in hearts (Figure 2C,E, P<0.001), whereas TRIF deficiency reversed this up-regulation (Figure 2C,E, P<0.01). Then, we measured the fibrotic signaling. As showed in Figure 2E–G, aldosterone significantly increased the expression of TGF-β1 and phos-SMAD3 in wild-type mice (P<0.001). TRIF deficient mice exhibited an effective reduction of fibrotic protein levels (Figure 2E–G, P<0.001). Furthermore, cardiac functional measurement showed TRIF deficiency increased the levels of EF and FS, but reduced left ventricle wall thickness and left ventricle mass, as compared with aldosterone-treated wild-type mice for 21 or 42 days (Table 1 and Supplementary Table S1).
Trif deficiency protects against aldosterone-induced mouse cardiac hypertrophy and fibrosis
Trif knockout and wild-type mice were treated with aldosterone (50 μg/kg per day) by using osmotic pump for 21 days. (A) Representative image of hearts. (B) The ratio of heart weight/tibia length. (C–E) Cardiac slides were stained with H&E or SR staining solution (C), quantitative analysis of average cardiomyocyte size (D), and relative density of collagen (E). Scale bar = 100 μm. (F–H) Immunoblot analysis of TGF-β1, phos-SMAD3, and SMAD3 (F), and quantitative analysis of the relative density of TGF-β1/Tubulin (G) and phos-SMAD3/SMAD3 (H). Results were shown as mean ± SEM, and n=5–8 mice/group. *P<0.05, **P<0.01, ***P<0.001.
Trif knockout and wild-type mice were treated with aldosterone (50 μg/kg per day) by using osmotic pump for 21 days. (A) Representative image of hearts. (B) The ratio of heart weight/tibia length. (C–E) Cardiac slides were stained with H&E or SR staining solution (C), quantitative analysis of average cardiomyocyte size (D), and relative density of collagen (E). Scale bar = 100 μm. (F–H) Immunoblot analysis of TGF-β1, phos-SMAD3, and SMAD3 (F), and quantitative analysis of the relative density of TGF-β1/Tubulin (G) and phos-SMAD3/SMAD3 (H). Results were shown as mean ± SEM, and n=5–8 mice/group. *P<0.05, **P<0.01, ***P<0.001.
. | WT-PBS . | KO-PBS . | WT-Aldo . | KO-Aldo . |
---|---|---|---|---|
Heart rate (beats/min) | 471 ± 341 | 460 ± 221 | 499 ± 59 | 479 ± 171 |
EF% | 75.54 ± 9.332 | 77.32 ± 11.762 | 50.11 ± 10.34 | 69.32 ± 6.322 |
FS% | 48.12 ± 5.432 | 49.12 ± 9.012 | 32.54 ± 9.03 | 43.12 ± 6.021 |
LVAW;d (mm) | 0.87 ± 0.113 | 0.85 ± 0.213 | 1.34 ± 0.26 | 0.96 ± 0.142 |
LVAW;s (mm) | 1.12 ± 0.203 | 1.05 ± 0.193 | 1.68 ± 0.33 | 1.35 ± 0.211 |
LVID;d (mm) | 3.64 ± 0.092 | 3.32 ± 0.182 | 4.42 ± 0.38 | 3.80 ± 0.451 |
LVID;s (mm) | 2.11 ± 0.161 | 2.22 ± 0.171 | 2.79 ± 0.31 | 2.33 ± 0.191 |
LVPW;d (mm) | 0.69 ± 0.213 | 0.63 ± 0.173 | 1.14 ± 0.27 | 0.82 ± 0.291 |
LVPW;s (mm) | 1.09 ± 0.232 | 1.11 ± 0.122 | 1.56 ± 0.25 | 1.23 ± 0.181 |
LV mass (mg) | 114.32 ± 9.333 | 113.44 ± 15.033 | 181.99 ± 26.42 | 132.07 ± 21.232 |
. | WT-PBS . | KO-PBS . | WT-Aldo . | KO-Aldo . |
---|---|---|---|---|
Heart rate (beats/min) | 471 ± 341 | 460 ± 221 | 499 ± 59 | 479 ± 171 |
EF% | 75.54 ± 9.332 | 77.32 ± 11.762 | 50.11 ± 10.34 | 69.32 ± 6.322 |
FS% | 48.12 ± 5.432 | 49.12 ± 9.012 | 32.54 ± 9.03 | 43.12 ± 6.021 |
LVAW;d (mm) | 0.87 ± 0.113 | 0.85 ± 0.213 | 1.34 ± 0.26 | 0.96 ± 0.142 |
LVAW;s (mm) | 1.12 ± 0.203 | 1.05 ± 0.193 | 1.68 ± 0.33 | 1.35 ± 0.211 |
LVID;d (mm) | 3.64 ± 0.092 | 3.32 ± 0.182 | 4.42 ± 0.38 | 3.80 ± 0.451 |
LVID;s (mm) | 2.11 ± 0.161 | 2.22 ± 0.171 | 2.79 ± 0.31 | 2.33 ± 0.191 |
LVPW;d (mm) | 0.69 ± 0.213 | 0.63 ± 0.173 | 1.14 ± 0.27 | 0.82 ± 0.291 |
LVPW;s (mm) | 1.09 ± 0.232 | 1.11 ± 0.122 | 1.56 ± 0.25 | 1.23 ± 0.181 |
LV mass (mg) | 114.32 ± 9.333 | 113.44 ± 15.033 | 181.99 ± 26.42 | 132.07 ± 21.232 |
Data are means ± SEM.
1P<0.05.
2P<0.01.
3P<0.001, as compared with WT-Aldo group.
Aldosterone is a strong inducer of cardiovascular inflammation, with the help of TLR4 signaling [22]. To this end, we analyzed the downstream signaling of TLR4 cofactor TRIF in aldosterone-treated mice. TRIF could activate IRF3 in a phosphorylation-dependent manner, and transduce the inflammatory signaling [28]. Consistently, aldosterone remarkably increased the expression of cardiac phos-IRF3 (Figure 3A,B), downstream IFN-α (Figure 3C and Supplementary Figure S2D, P<0.001), and inflammatory cytokines TNF-α (Figure 3D and Supplementary Figure S2E, P<0.001), IL-1β (Figure 3E and Supplementary Figure S2F, P<0.001), and nitric oxide (Figure 3F and Supplementary Figure S2G, P<0.001) on days 21 and 42. In contrast, cardiac deficiency of TRIF significantly decreased aldosterone-induced phos-IRF3 and the production of inflammatory cytokines (Figure 3A–F). Furthermore, we measured the local expression of cardiac IFN-α by immunofluorescence staining. As shown in Supplementary Figure S3A,B, IFN-α was highly localized in aldosterone-treated WT mice (P<0.001), but TRIF deficiency suppressed this up-regulation (P<0.001). Flow cytometry analysis further supported TRIF deficiency attenuated aldosterone-induced cardiac inflammatory response by decreasing the percentage of F4/80+ macrophages (Supplementary Figure S3C, P<0.001) and Cd3+ toxic T cells (Supplementary Figure S3D, P<0.01). Previous studies have demonstrated that aldosterone can elevate blood pressure, which contributes to cardiac remodeling [29,30]. As shown in Supplementary Figure S4, aldosterone obviously increased systolic blood pressure on days 7 and 21 (P<0.01). However, TRIF deficiency had no significant effects on blood pressure in WT or KO mice.
Trif deficiency attenuates aldosterone-induced cardiac inflammatory response and mitochondrial dysfunction
(A,B) Immunoblot analysis of IRF-3 (A), and quantitative analysis of the relative density of IRF-3/Tubulin (B). (C–F) Cardiac lysates were measured by kits for IFN-α (C), TNF-α (D), IL-1β (E), and nitric oxide (F) levels. (G) The mitochondrial gene levels of Vdac1, Pgc1α, and Nrf1 in hearts. (H–J) Extracted cardiac mitochondria were measured for relative ATP level (H), oxygen consumption (I), and citrate synthase activity (J). Results were shown as mean ± SEM, and n=5–8 mice/group. *P<0.05, **P<0.01, ***P<0.001.
(A,B) Immunoblot analysis of IRF-3 (A), and quantitative analysis of the relative density of IRF-3/Tubulin (B). (C–F) Cardiac lysates were measured by kits for IFN-α (C), TNF-α (D), IL-1β (E), and nitric oxide (F) levels. (G) The mitochondrial gene levels of Vdac1, Pgc1α, and Nrf1 in hearts. (H–J) Extracted cardiac mitochondria were measured for relative ATP level (H), oxygen consumption (I), and citrate synthase activity (J). Results were shown as mean ± SEM, and n=5–8 mice/group. *P<0.05, **P<0.01, ***P<0.001.
Mitochondrial biogenesis and activities determine cardiac metabolism and the response to foreign stimuli. Ibarrola et al. reported that aldosterone impaired cardiac mitochondrial function in human myocardial biopsies [31]. Similarly, present study also found aldosterone decreased the levels of mitochondrial biogenesis genes, including Vdac1, Pgc1α, and Nrf1 (Figure 3G), but TRIF deficiency could significantly reverse this down-regulation (Figure 3G, P<0.05). Furthermore, aldosterone also inhibited cardiac mitochondrial activities, including ATP production (Figure 3H, P<0.01), oxygen consumption (Figure 3I, P<0.001) and citrate synthase activity (Figure 3J, P<0.001). However, TRIF deficiency effectively maintained the mitochondrial activities in aldosterone-treated mice (Figure 3H–J, P<0.05). Taken together, these results determined TRIF played a critical role in aldosterone-induced cardiac remodeling.
Cardiac miR-34a mediates the physiological role of aldosterone/TRIF in cardiac remodeling
Amounts of studies have found miR-34a existed crucial roles in cardiac aging, proliferation, and immune response [32–34]. However, it was unknown whether miR-34a participated in aldosterone-induced cardiac remodeling. Here, we found aldosterone dose-dependently increased the expression of miR-34a in primary mouse cardiomyocytes (Figure 4A), and time-dependently stimulated miR-34a levels in mouse hearts (Figure 4B). More importantly, aldosterone downstream inflammatory molecules, such as IFN-α, TNF-α, and IL-1β, dose-dependently increased miR-34a levels in cardiomyocytes (Supplementary Figure S5A–C). The incremental up-regulation of cardiac miR-34a was also positively correlated with Trif levels (Figure 4C, r = 0.8232, P<0.001). More importantly, the up-regulated level of miR-34a was obviously decreased in aldosterone-treated TRIF deficient mice (Figure 4D, P<0.01).
The elevated level of cardiac miR-34a closely correlates with Trif expression in aldosterone-treated mice
(A) A total of 1 × 106 mouse primary cardiomyocytes were treated with different concentrations (0, 10−6, 10−5, 10−4 mol/l) of aldosterone (Aldo) for 24 h. Real-time PCR analysis of miR-34a level. (B,C) Male C57BL/6N mice were treated with aldosterone (50 μg/kg per day) by using osmotic pump for 0, 3, 7, and 21 days. Real-time PCR analysis of miR-34a level (B), and the correlation between miR-34a level and cardiac Trif levels (C). (D) Trif knockout and wild-type mice were treated with aldosterone (50 μg/kg per day) by using osmotic pump for 21 days. The cardiac expression of miR-34a. Results were shown as mean ± SEM, and n=5 independent experiments or 5 mice/group. *P<0.05, **P<0.01, ***P<0.001.
(A) A total of 1 × 106 mouse primary cardiomyocytes were treated with different concentrations (0, 10−6, 10−5, 10−4 mol/l) of aldosterone (Aldo) for 24 h. Real-time PCR analysis of miR-34a level. (B,C) Male C57BL/6N mice were treated with aldosterone (50 μg/kg per day) by using osmotic pump for 0, 3, 7, and 21 days. Real-time PCR analysis of miR-34a level (B), and the correlation between miR-34a level and cardiac Trif levels (C). (D) Trif knockout and wild-type mice were treated with aldosterone (50 μg/kg per day) by using osmotic pump for 21 days. The cardiac expression of miR-34a. Results were shown as mean ± SEM, and n=5 independent experiments or 5 mice/group. *P<0.05, **P<0.01, ***P<0.001.
To identify the role of miR-34a in aldosterone-induced cardiac remodeling, we next utilized the AAV to specifically overexpress cardiac miR-34a in TRIF-deficient mice. As shown in Supplementary Figure S6, the decreased cardiac miR-34a was obviously reversed in TRIF-deficient mice treated with AAV encoding miR-34a mimic (P<0.001). Interestingly, replenishment of cardiac miR-34a abolished the cardiac benefits, including enlargement of cardiomyocyte size (Figure 5A,B, P<0.05), induction of collagen deposits (Figure 5A,C, P<0.01), and impaired cardiac function (Figure 5D,E, P<0.05), as compared with aldosterone-treated TRIF-deficient mice. The expression fibrotic genes, such as Tgf-β1, Col6α1, and Mmp12, were remarkably increased in miR-34a mimic-treated mice (Figure 5F, P<0.05). Besides, the mitochondrial protection of TRIF deficiency was also partially abolished after miR-34a overexpression. In details, overexpression of cardiac miR-34a decreased ATP production (Figure 5G, P<0.05), oxygen consumption (Figure 5H, P<0.05) and citric synthase activity (Figure 5I, P<0.05).
MiR-34a modulates Trif-mediated cardiac remodeling in aldosterone-treated mice
Trif knockout mice were administrated with 1 × 109 AAV encoding miR-34a mimic or control with cardiac-specific Myh6 promoter for 14 days, then treated with aldosterone (50 μg/kg per day) by using osmotic pump for 21 days. (A–C) Cardiac slides were stained with H&E or SR staining solution (A), and quantitative analysis of average cardiomyocyte size (B) and relative density of collagen (C). Scale bar = 100 μm. (D,E) Echocardiographic analysis of mouse cardiac EF (D) and FS (E). (F) The fibrotic gene levels of Tgf-β1, Col6α1 and Mmp12. (G–I) Extracted cardiac mitochondria were measured for relative ATP level (G), oxygen consumption (H), and citrate synthase activity (I). Results were shown as mean ± SEM, and n=6 mice/group. *P<0.05, **P<0.01, ***P<0.001.
Trif knockout mice were administrated with 1 × 109 AAV encoding miR-34a mimic or control with cardiac-specific Myh6 promoter for 14 days, then treated with aldosterone (50 μg/kg per day) by using osmotic pump for 21 days. (A–C) Cardiac slides were stained with H&E or SR staining solution (A), and quantitative analysis of average cardiomyocyte size (B) and relative density of collagen (C). Scale bar = 100 μm. (D,E) Echocardiographic analysis of mouse cardiac EF (D) and FS (E). (F) The fibrotic gene levels of Tgf-β1, Col6α1 and Mmp12. (G–I) Extracted cardiac mitochondria were measured for relative ATP level (G), oxygen consumption (H), and citrate synthase activity (I). Results were shown as mean ± SEM, and n=6 mice/group. *P<0.05, **P<0.01, ***P<0.001.
On the contrast, specific inhibition of cardiac miR-34a attenuated aldosterone-induced cardiac hypertrophy (Supplementary Figure S7A,B, P<0.01). SR staining showed miR-34a sponge decreased cardiac collagen accumulation in aldosterone-treated mice (Supplementary Figure S7A,C, P<0.01). Similarly, cardiac functional measurement showed miR-34a sponge increased the levels of EF and FS, as compared with aldosterone-treated mice (Supplementary Figure S7D,E). MiR-34a sponge also improved cardiac mitochondrial activities, including ATP production (Supplementary Figure S7F, P<0.05), oxygen consumption (Supplementary Figure S7G, P<0.01), and citrate synthase activity (Supplementary Figure S7H, P<0.05). All these findings supported cardiac miR-34a partially determined aldosterone/TRIF-mediated cardiac remodeling.
Discussion
Abnormal activation of RAAS, including a higher level of aldosterone, has been demonstrated to trigger an inflammatory process in cardiovascular diseases [4]. In aldosterone-treated rats, there was a significant induction of inflammatory response in hearts [6,7]. A recent study found deletion of MR, known as the aldosterone receptor, markedly reduced the cardiac inflammation, fibrosis, and functional disorders in deoxycorticosterone acetate/salt-treated animals [35]. The present study aimed to identify the role of TRIF, a critical inflammatory regulator, in aldosterone-induced cardiac remodeling and the molecular mechanisms. Through utilizing TRIF-deficient mouse model, we found TRIF determined aldosterone-induced cardiac remodeling, including cardiac hypertrophy, functional disorders, fibrosis, inflammatory response, and mitochondrial injuries. Mechanistically, the incremental increasing miR-34a was a potential mediator in aldosterone/TRIF-induced heart damages. Cardiac special overexpression of miR-34a abolished the cardioprotective profiles in TRIF-deficient mice, whereas miR-34a inhibition could attenuate aldosterone-induced abnormal cardiac remodeling. Taken together, TRIF/miR-34a axis mediated aldosterone-induced cardiac remodeling.
Multiple factors are contributed to the process of cardiac remodeling leading to cardiac functional disorders and further heart failure. Excess activation of adrenergic signaling and/or RAAS are major molecular pathways triggering cardiovascular changes. The circulating levels of aldosterone were closely correlated with the increasing incidence of cardiovascular diseases [36]. In patients with severe heart failure, there were extremely higher levels of circulating aldosterone [8,37]. Beygui et al. found that plasma aldosterone levels were quickly increased to more than two-fold after myocardial infarction, and precisely predicted the therapeutic outcomes [38].In obese mice, plasma levels of aldosterone were also obviously elevated [39,40]. Blockage of aldosterone signaling had therapeutic benefits in protecting against heart failure and related cardiovascular complications [8,37].
Although it was well-established that aldosterone closely linked to cardiac remodeling, mechanisms by how aldosterone affected cardiac pathophysiological process remained unclear. One of the key contributors was aldosterone-induced cardiac inflammatory response. Aldosterone triggered excess production of reactive oxygen species (ROS) and inflammatory cytokines, whereas blockage of TLR 4 could reverse cardiac hypertrophy, dysfunction, and inflammatory response [22]. As one of necessary adaptors in TLR4 complex, TRIF is another critical mediator of signaling cascades that directly influences multiple cellular processes [15]. For example, stimulation of TRIF activates NF-κB and MAPK pathways that trigger transcription of numerous inflammatory cytokines [41,42]. Previous studies have reported TRIF-mediated inflammatory cascades led to oxidative stress, cardiac hypertrophy, and cellular apoptosis after myocardial infarction [43]. Present echocardiographic results showed cardiac TRIF was a critical mediator in aldosterone-induced cardiac hypertrophy. As an inflammatory transducer, this finding supported the close links between excessive inflammatory response and cardiac hypertrophy. Pathological hypertrophy is often characterized as impaired myocardial vascularization, unfavorable changes in the extracellular matrix composition, and fibrosis [44]. The abnormal inflammatory response, such as increased levels of TNF-α and soluble TNF-α receptors, serves as prognostic markers in patients with heart hypertrophy [45]. In present study, TRIF deficiency not only significantly reduced aldosterone-induced cardiac inflammation, but also improved cardiac hypertrophy.
A variety of studies have demonstrated that aldosterone and the MR antagonists modulate cardiac hypertrophy [46–48]. Aldosterone also directly promoted cardiomyocyte hypertrophy and induced the expression of fetal-type cardiac genes in vitro in a manner dependent on activation of the MR. In present study, the expression of cardiac hypertrophic markers, including ANP, BNP, and MHC, was significantly decreased in Trif KO mice treated with aldosterone. Furthermore, aldosterone has been demonstrated to influence cardiac remodeling independent of its impact on systemic blood pressure, whereas cardiac fibrosis could be prevented by treatment with non-antihypertensive doses of spironolactone [44,49]. These reported findings were consistent to our present results. The Trif KO had no significant effect on blood pressure, but suppressed cardiac expression of fibrotic genes and protein. Therefore, although our results had not provided a completely new mechanism in MR activation-induced myocardial hypertrophy, we first addressed the novel role of aldosterone/Trif/miR-34a signaling in cardiac remodeling.
Previous studies have also demonstrated that aldosterone could regulate the cardiac expression of miR levels. In neonatal rat cardiomyocytes, aldosterone-induced cell hypertrophy concurred with up-regulation of miR-23a level [50]. Aldosterone down-regulated the expression of miR-208b, which suppressed β-myosin heavy chain transcription [51]. Syed et al. found miR-21 deficiency exacerbated aldosterone-mediated cardiac abnormal remodeling and dysfunction [12]. In the present study, we first provided evidence that cardiac miR-34a was also increased in aldosterone-treated primary mouse cardiomyocytes and heart tissues. MiR-34a is one of the most widely studied miRNAs in mammalian cells due to its critical role in the pathophysiology of multiple phenotypes such as solid tumors, cardiac injury, and tissue inflammation [32,34,52]. In several cardiovascular diseases, miR-34a exacerbated cardiac aging, suppressed cardiac repair and regeneration following myocardial infarction, and positively correlated with the incidence of cardiovascular diseases [32,33,53]. Mechanistically, miR-34a stimulated tissue inflammation through repressing transcriptional factors, such as Kruppel-like factor (KLF) 4 and sirtuin (SIRT) 1 [34,54]. Consistently, our present findings showed that overexpression of cardiac miR-34a abolished the anti-inflammatory effects of TRIF deficiency, and maintained the pro-inflammatory profiles of aldosterone in hearts. It will be meaningful to maintain cardiac structure and function through blocking TRIF/miR-34a axis.
The present study first addressed the existence of cardiac TRIF/miR-34a signaling, which mediated the pathophysiological role of aldosterone in cardiac remodeling. However, the detailed molecular interaction between these two targets was still unclear. Our finding supposed TRIF downstream inflammatory factors were possible mediators. In a future study, it needs to fully investigate the underlying mechanism.
In conclusion, the present study supported TRIF was a potential therapeutic target for cardiac pathophysiological remodeling. The TRIF/miR-34a axis could provide a novel molecular mechanism for explaining aldosterone-induced cardiac hypertrophy, fibrosis, and dysfunction.
Clinical perspectives
Aldosterone, as a major product of RAAS, deteriorates the pathophysiological process of myocardial injuries through affecting multiple signaling in clinical studies. Among them, TRIF signaling-mediated cardiac inflammatory response is one of key factors in determining cardiac remodeling. Aldosterone obviously increased the cardiac levels of TRIF, which was closely correlated with cardiac dysfunction. However, it was unknown whether or how cardiac TRIF mediated aldosterone-induced cardiac remodeling.
Aldosterone antagonists have significant therapeutic benefits in patients with cardiac disorders. Therefore, we explored the roles of TRIF in aldosterone-induced cardiac injuries by utilizing TRIF genetic-deficient mice. TRIF deficiency protected against aldosterone-induced cardiac remodeling. Mechanistically, miR-34a participated in aldosterone/TRIF signaling, and specific overexpression of cardiac miR-34a abolished the cardiac benefits of TRIF deficiency in aldosterone-treated mice.
Cardiac TRIF/miR-34a provided an explanation for the pharmacological benefits of aldosterone antagonists in cardiac injuries, and both cardiac TRIF and miR-34a might become the potential therapeutic targets.
Competing Interests
The authors declare that there are no competing interests associated with the manuscript.
Funding
This work was supported by the National Natural Science Foundation of China [grant number 81800362]; and the Young and Middle-aged Innovation Science Foundation from Second Affiliated Hospital of Harbin Medical University [grant number KYCX2018-24].
Author Contribution
S.j.L. conducted the mouse experiments, analyzed data, and discussed the results. W.C. conducted the primary cell experiments and analyzed data. B.W. and E.Z. conducted the osmotic pump implantation. J.X. discussed the results. S.f.L. guided all procedures, discussed the results, and wrote the manuscript.
Acknowledgements
The authors thank Dr. Qian Ni (Tianjin University) for gifting the viruses.
Abbreviations
- AAV
adeno-associated virus
- ANP
atrial natriuretic peptide
- ATP
adenosine triphosphate
- BNP
brain natriuretic peptide
- EF
ejection fraction
- FS
fractional shortening
- IRF
IFN regulatory factor
- LVID
left ventricular internal dimension
- MHC
myosin heavy chain
- miR
microRNA
- MR
mineralocorticoid receptor
- MyD88
myeloid differentiation primary response 88
- Myh6
cardiac-specific α myosin-heavy chain
- phos
phosphorylated
- RAAS
renin–angiotensin–aldosterone system
- SR
Sirius Red
- TLR
Toll-like receptor
- Trif/TRIF
TIR-domain-containing adapter-inducing interferon-β