Type III Transforming Growth Factor-β Receptor Drives Cardiac Hypertrophy Through β-Arrestin2–Dependent Activation of Calmodulin-Dependent Protein Kinase IINovelty and Significance
The role of type III transforming growth factor-β receptor (TβRIII) in the pathogenesis of heart diseases remains largely unclear. Here, we investigated the functional role and molecular mechanisms of TβRIII in the development of myocardial hypertrophy. Western blot and quantitative real time-polymerase chain reaction analyses revealed that the expression of TβRIII was significantly elevated in human cardiac hypertrophic samples. Consistently, TβRIII expression was substantially increased in transverse aortic constriction (TAC)– and isoproterenol-induced mouse cardiac hypertrophy in vivo and in isoproterenol-induced cardiomyocyte hypertrophy in vitro. Overexpression of TβRIII resulted in cardiomyocyte hypertrophy, whereas isoproterenol-induced cardiomyocyte hypertrophy was greatly attenuated by knockdown of TβRIII in vitro. Cardiac-specific transgenic expression of TβRIII independently led to cardiac hypertrophy in mice, which was further aggravated by isoproterenol and TAC treatment. Cardiac contractile function of the mice was not altered in TβRIII transgenic mice; however, TAC led to significantly decreased cardiac contractile function in TβRIII transgenic mice compared with control mice. Conversely, isoproterenol- and TAC-induced cardiac hypertrophy and TAC-induced cardiac contractile function impairment were partially reversed by suppression of TβRIII in vivo. Our data suggest that TβRIII mediates stress-induced cardiac hypertrophy through activation of Ca2+/calmodulin-dependent protein kinase II, which requires a physical interaction of β-arrestin2 with both TβRIII and calmodulin-dependent protein kinase II. Our findings indicate that stress-induced increase in TβRIII expression results in cardiac hypertrophy through β-arrestin2–dependent activation of calmodulin-dependent protein kinase II and that transforming growth factor-β and β-adrenergic receptor signaling are not involved in spontaneous cardiac hypertrophy in cardiac-specific transgenic expression of TβRIII mice. Our findings may provide a novel target for control of myocardial hypertrophy.
Pathological cardiac hypertrophy is usually induced by pressure overload and sustained β-adrenergic receptor (β-AR) activation. Cardiac hypertrophy is a strong predictor of subsequent cardiovascular events, such as diastolic dysfunction and arrhythmias, which ultimately results in heart failure.1,2 Understanding the key molecular events that mediate pathological hypertrophy is essential for developing new therapeutic strategies to prevent morbidity and mortality associated with cardiac hypertrophy.
Type III transforming growth factor-β (TGF-β) receptor (TβRIII, also known as β-glycan) is the most abundant TGF-β coreceptor in many cell types.3,4 TβRIII is an 851 amino acid (aa) proteoglycan, comprising a large 766 aa extracellular domain, a single-pass hydrophobic transmembrane region, and a short 42 aa cytoplasmic domain.5 Loss or reduced expression of TβRIII has been shown to be an important regulator of cell migration, invasion, cell growth, and angiogenesis in cancer3–8; however, the role of TβRIII in the development of human heart diseases has not been investigated.
TβRIII is classically thought to function as a coreceptor, presenting TGF-β superfamily ligands to their respective signaling receptors. However, TβRIII has the potential to increase or decrease TGF-β signaling depending on the cell type.3 Recently, new insights have been gained into the structure and function of cytoplasmic TβRIII domain, which have suggested its essential role in ligand-dependent and ligand-independent functions through interactions with β-arrestin2 and Gα-interacting protein–interacting protein, C terminus.7,9–12
To date, there is no information on the expression levels of TβRIII in normal or pathological human cardiac tissues. Moreover, the role of TβRIII, as well as its downstream effectors, has not been yet studied in mature cardiomyocytes. Accordingly, in this study, we demonstrate that TβRIII expression is increased in human cardiac hypertrophy tissues. Concomitantly, we show that TβRIII is an important regulator in transverse aortic constriction (TAC)– and isoproterenol-induced cardiac hypertrophy in mouse models.
Detailed and expanded methodology is provided in the online-only Data Supplement.
Human Heart Samples From Patients With Cardiac Hypertrophy
The human study conformed to the principles outlined in the Declaration of Helsinki. The study protocol was approved by the ethics committee of First Affiliated Hospital of Nanjing Medical University (No. 2014-SRFA-128). Written informed consent was obtained from all patients. The tissues were obtained from 14 individuals undergoing heart surgery (8 cardiac hypertrophic samples and 6 noncardiac hypertrophic samples to serve as control). The samples were stored at −80°C until further analyses. Information about the human cardiac samples is listed in Table S1.
Generation of TβRIII Transgenic Mice
TβRIII transgenic (Tg) mice were generated by Cyagen Bioscience Inc (Suzhou, China). Briefly, the mouse TβRIII (GenBank: NM_011578.3) cDNA provided by GeneChem Co (Shanghai, China) was amplified by polymerase chain reaction and cloned into the region between the 5.4-kb mouse α-myosin heavy chain (α-MHC) promoter and the Simian virus 40 (SV40) polyadenylation sequence via SalI sites. The orientation of the inserted TβRIII cDNA was confirmed by sequencing. This construct was microinjected into the fertilized eggs of FBV mouse strain. The founders (F0) of TβRIII-Tg lines were backcrossed with C57BL/6 mice for 1 to 5 generations. The positive offspring were identified using polymerase chain reaction with the following oligos (5′–3′): transgene polymerase chain reaction forward, ACGTAAACGGCCACAAGTTC and reverse, GATCTTGAAGTTCACCTTGATGC and internal control polymerase chain reaction forward, TCTTAGCTCTGCTCTCCGGT and reverse, CACTGGCTGAGGAAGGAGAC.
In Vivo Adeno-Associated Viral 9 Carrying shRNA for TβRIII Infection
Adeno-associated viral (AAV) 9–mediated cardiac-specific gene knockdown has been previously demonstrated.13,14 Green fluorescent protein–tagged AAV9-mediated shRNA for TβRIII (5′-GGGAGGTTCACATCCTAAA-3′; AAV9-shRNA) and green fluorescent protein-tagged AAV9-scramble control (5′-GTTC TCCG AACG TGTC ACGT-3′) were synthesized from Biowit Technologies (Shenzhen, P.R. China). The method to infect mouse heart with virus-mediated shRNA delivery system was previously described by our group.15 Briefly, after the mice were anesthetized and a thoracotomy was performed, the chest was opened via the fourth intercostal space. The ascending aortic artery and main pulmonary artery were clamped and then AAV9-shRNA or AAV9-scramble control (1×109 pfu in 100-μL saline) was injected into the left ventricular (LV) cavity through the tip of the heart using a 30-gauge syringe. The occlusion of arteries remained for 10 s after AAV9 injection. Mice in sham group underwent the same procedures and received 100-μL saline. After 7 days of AAV9 infection, mice received either isoproterenol infusion for 2 weeks or TAC for 3 weeks.
Statistical analysis was performed with SPSS for Windows software version 11.5 (SPSS Inc, Chicago, IL). Comparisons between the 2 groups were determined by a Student t test. Multiple group comparison was performed using 1-way ANOVA with the Bonferroni post hoc test. Comparison between the 2 distinct independent variables, a 2-way ANOVA test was performed. A value of P<0.05 was considered statistically significant.
Increased Expression of TβRIII in Human and Mouse Cardiac Hypertrophic Samples
We initially examined TβRIII expression profiles in human cardiac hypertrophic specimens. The expression levels of TβRIII mRNA and protein were, respectively, increased by 1.88-fold (Figure 1A) and 1.56-fold (Figure 1B and 1C) in human cardiac hypertrophic samples compared with noncardiac hypertrophic samples. Accordingly, the levels of hypertrophic biomarkers including atrial natriuretic peptide, β-MHC (Figure 1A through 1C), and brain natriuretic peptide were significantly elevated (Figure 1A) in human cardiac hypertrophic samples.
Consistently, TβRIII protein levels of the LV tissue were increased gradually during the time-course experiment in either the Alzet pump application of isoproterenol or the TAC-induced mouse cardiac hypertrophy (Figure 1D and 1E; Figure S1 shows the specificity of TβRIII antibody). Moreover, immunofluorescence staining clearly showed that TβRIII was expressed in mouse LV tissue (Figure 1F, a). Line-scanning analysis of the fluorescence intensity suggested that TβRIII was mainly present in the vicinity of the plasma membrane of mature cardiomyocytes (Figure 1F, b). These results suggested that an increased expression of TβRIII may play a role in the development of cardiac hypertrophy.
We found that the expression levels of TβRIII in primary cultured neonatal mouse cardiomyocytes (NMCMs) were upregulated by isoproterenol in a dose- and time-dependent manner (Figure S2). We then examined whether manipulating the expression levels of TβRIII in NMCMs affects isoproterenol- or angiotensin II–induced hypertrophic phenotype. Our data clearly show that gain-of-function of TβRIII led to cardiac hypertrophy phenotype without isoproterenol stimulation, whereas loss-of-function of TβRIII attenuated isoproterenol- or angiotensin II–induced cardiomyocyte hypertrophy (Figure S2).
Cardiac-Specific Overexpression of TβRIII Recapitulates the Myocardial Hypertrophy Phenotype
We then examined whether a TβRIII transgenic mouse model carrying the cardiac-specific α-MHC promoter (CS-TβRIII-Tg) could recapitulate the cardiac hypertrophic phenotype (Figure 2A). Three months after CS-TβRIII-Tg, expression levels of TβRIII were specifically increased in mouse hearts but not in kidneys, suggesting the specificity and successful establishment of the CS-TβRIII-Tg mouse model (Figure 2B and 2C). Relative to wild-type (WT) mouse, the levels of TβRIII and β-MHC proteins were, respectively, increased ≈2.7-fold and ≈2.4-fold in heterozygous mouse hearts (Figure 2C). The expression levels of TβRIII, atrial natriuretic peptide, brain natriuretic peptide, and β-MHC mRNA were remarkably increased in 3-month-old CS-TβRIII-Tg mice compared with age-matched WT mice (Figure 2D). Cardiac dimensions were monitored in conscious mice using serial echocardiography. CS-TβRIII-Tg mice showed increased LV wall thickness and LV mass (Figure 2E; Table S4). Comparative analysis of histology (Figure 2F), ratios of heart weight/body weight, and heart weight/tibia length (Figure 2G and 2H) between 3-month-old CS-TβRIII-Tg mice and age-matched WT mice revealed that cardiac-specific overexpression of TβRIII resulted in spontaneous cardiac hypertrophy. Moreover, isoproterenol infusion at a rate of 10 mg/kg per day or TAC produced more severe cardiac hypertrophy in CS-TβRIII-Tg mice than in either WT mice or CS-TβRIII-Tg mice (isoproterenol: Figure 2I through 2K; TAC: Figure 2L through 2N). Notably, after 3 months, there was no alteration of cardiac function in CS-TβRIII-Tg mice; however, cardiac contractility declined dramatically in these mice 3 weeks post TAC operation or 2 weeks of isoproterenol stimulation, as indicated by the reduced ejection fraction and fractional shortening (Figure 2O and 2P). In addition, Masson staining showed that isoproterenol and TAC stimulation resulted in more severe cardiac fibrosis in CS-TβRIII-Tg mice than their age-matched WT mice (Figure 2Q and 2R). Taken together, our data suggest that cardiac overexpression of TβRIII was sufficient to induce cardiac hypertrophy and that elevated expression of TβRIII accelerated stress-induced injury of heart function.
Cardiac AAV9-Mediated TβRIII Gene Silencing Attenuates Stress-Induced Myocardial Hypertrophy
We reasoned that if overexpression of TβRIII promoted myocardial hypertrophy, the knockdown of endogenous TβRIII should prevent stress-induced cardiac hypertrophy. We, therefore, used cardiac-targeted AAV9 vector-mediated specific shRNA (AAV9-shTβRIII) to knockdown TβRIII gene in mouse heart. Immunofluorescence staining data suggested that the heart tissue was effectively infected by either AAV9-shTβRIII-green fluorescent protein or AAV9-scramble-green fluorescent protein (Figure 3A). Western blot revealed that the expression levels of TβRIII in the LV significantly decreased on infection of hearts with AAV9-shTβRIII in a time-dependent manner (Figure 3B) but not with AAV9-mediated scramble shRNA (data not shown). According to these results, 1 week after infection, mice received isoproterenol infusion (10 mg/kg per day) for 2 weeks or TAC operation for 3 weeks. Echocardiographic examination showed that isoproterenol-induced increase in hypertrophic indices in control mice were greatly decreased in AAV9-shTβRIII–infected mice (Figure 3C, a; Table S5). Moreover, TAC-induced cardiac hypertrophy and impairment of cardiac function were partially rescued by AAV9-shTβRIII infection (Figure 3C, b; Table S6). The ratios of LV mass/body weight and LV mass/tibia length were significantly greater in isoproterenol- or TAC-induced cardiac hypertrophy mice than in sham-operated mice, whereas these indices of isoproterenol- or TAC-induced cardiac hypertrophy were greatly decreased in AAV9-shTβRIII–mediated cardiac-specific knockdown mice compared with the mice infected with AAV9-mediated scramble shRNA (Figure 3D through 3G). Furthermore, 12 weeks after TAC operation, the ejection fraction and fractional shortening declined remarkably in both the WT and the AAV9-mediated scramble shRNA-infected mice; importantly, this TAC-induced impairment of cardiac contractility was partially rescued by AAV9-shTβRIII–mediated cardiac-specific knockdown of TβRIII in mice (Figure 3H and 3I). Consistently, isoproterenol- or TAC-induced increase in the expression levels of β-MHC protein and atrial natriuretic peptide, brain natriuretic peptide, and β-MHC mRNA were significantly decreased by AAV9-shTβRIII–mediated cardiac-specific knockdown of TβRIII (Figure 3J through 3L). These results further confirm that TβRIII plays an important role in stress-induced cardiac hypertrophy.
TβRIII-Mediated Activation of CaMKII Is Linked by β-Arrestin2
Cytoplasmic domain of TβRIII interacts with the scaffolding protein, β-arrestin2, and the consequence of which affects proliferation, mobility, and migration in various cancer cell types.4,16 Mangmool et al17 revealed that β-arrestins are essential for calmodulin-dependent protein kinase II (CaMKII) activation by isoproterenol. We, therefore, examined whether β-arrestin2 served as a scaffold for either TβRIII or CaMKII in NMCMs. The data showed that TβRIII complexed with β-arrestin2 (Figure 4A), and β-arrestin2 also formed a complex with CaMKII (Figure 4B). However, our Co-immunoprecipitation data revealed that there was no direct interaction between TβRIII and CaMKII (Figure S3). Interestingly, the magnitude of these complexes was, respectively, enhanced by isoproterenol stimulation and by transient overexpression of TβRIII in NMCMs (Figure 4C). Conversely, isoproterenol-induced increase in the formation of TβRIII/β-arrestin2 (Figure 4C, a) and β-arrestin2/CaMKII (Figure 4C, b) complexes was remarkably decreased by knockdown of TβRIII in NMCMs. Western blot analysis showed that phosphorylated CaMKII (p-CaMKII) levels were significantly increased by isoproterenol stimulation, an event that was greatly blunted by knockdown of TβRIII in NMCMs (Figure 4D). Interestingly, transient overexpression of TβRIII independently increased expression levels of p-CaMKII and β-MHC in NMCMs (Figure 4D). We next examined whether the scaffold role of β-arrestin2 was required for TβRIII-mediated CaMKII activation and cardiac hypertrophy by knocking down β-arrestin2 (efficiency shown in Figure S4) in TβRIII overexpressed NMCMs. Consistently, knockdown of β-arrestin2 in NMCMs greatly reduced p-CaMKII and β-MHC levels (Figure 4E). KN-93, an inhibitor of CaMKII, could abrogate the cell surface enlargement (Figure 4F), elevated ratio of protein/DNA ratio (Figure 4G) and up-regulation of β-MHC protein (Figure 4H) caused by TβRIII overexpression in NMCMs. These results suggested that the scaffold role of β-arrestin2 between TβRIII and CaMKII was likely necessary for TβRIII-mediated cardiomyocytes hypertrophy.
Consistent with the in vitro study results, immunofluorescence staining revealed that the colocalizations of TβRIII/β-arrestin2 and β-arrestin2/CaMKII were enhanced in CS-TβRIII-Tg mouse heart tissue (Figure 5A and 5B). Co-immunoprecipitation data also demonstrated that both formations of TβRIII/β-arrestin2 and β-arrestin2/CaMKII complexes were significantly increased in LVs of CS-TβRIII-Tg mice compared with WT mice (Figure 5C). Accordingly, expression levels of p-CaMKII were significantly elevated in LVs of 3-month-old CS-TβRIII-Tg mice compared with age-matched WT mice (Figure 5D).
We next examined whether isoproterenol-stimulation or TAC operation affected the abundance and localization of TβRIII/β-arrestin2 and β-arrestin2/CaMKII complexes in WT hearts. The fluorescent intensity, an indicator of the abundance of TβRIII expression in the heart tissue, was greatly enhanced by either isoproterenol or TAC induction. Most importantly, the merged images demonstrated that there was an increased fluorescent intensity near the plasma membrane compartment of cardiomyocytes in response to isoproterenol or TAC, reflecting that isoproterenol or TAC promoted the formation of TβRIII/β-arrestin2 (Figure 5E) and β-arrestin2/CaMKII complexes (Figure 5F). Co-immunoprecipitation data further confirmed the important role of TβRIII in stress-induced cardiac hypertrophy, given that isoproterenol- or TAC-induced increase in the magnitude of TβRIII/β-arrestin2 and β-arrestin2/CaMKII complexes was greatly decreased by AAV9-mediated specific knockdown of TβRIII in the heart tissue (Figure 5G and 5H). Furthermore, isoproterenol- or TAC-induced elevation of p-CaMKII levels was substantially attenuated by cardiac-specific inhibition of TβRIII (Figure 5I and 5J). These data together suggested that TβRIII interfered with the process of cardiac hypertrophy through the interaction with β-arrestin2 and activation of CaMKII.
TβRIII-Induced Spontaneous Cardiac Hypertrophy Is Independent to TGF-β1 and Smad2/3
We next examined whether an increased TGF-β1 and Smad2/3 are involved in spontaneous cardiac hypertrophy in 3-month-old CS-TβRIII-Tg mice using Western blot analysis. The data show that expression levels of TGF-β1 (Figure 6A) and ratio of phosphorylated Smad2/3 (p-Smad2/3)/total Smad2/3 (T-Smad2/3; Figure 6B) were not altered in LV tissues dissected from the hearts of 3-month-old CS-TβRIII-Tg mice compared with age-matched mice. Furthermore, the expression levels of TGF-β1 in TAC- or isoproterenol-induced hypertrophic heart tissues were significantly elevated, although isoproterenol- or TAC-induced increase in TGF-β1 levels was not attenuated by cardiac-specific knockdown of TβRIII (Figure 6C and 6D). These results suggest that spontaneous cardiac hypertrophy observed in CS-TβRIII-Tg mice was most likely because of ligand-independent signaling.
β-ARs Are Not Involved in Cardiac TβRIII Overexpression-Induced Hypertrophy
We first detected whether there is a direct interaction between TβRIII and β1-AR or between TβRIII and β2-AR. Co-immunoprecipitation data show that TβRIII did not assemble to form a complex with either β1-AR (Figure 7A) or β2-AR (Figure 7B) in mouse LV tissues. Moreover, the downstream signaling of β-AR, cAMP, and protein kinase A levels in LV tissues, dissected from 3-month-old CS-TβRIII-Tg mouse hearts (Figure 7C and 7D), were not altered compared with age-matched WT mice. Interestingly, isoproterenol-induced increase in cAMP and protein kinase A levels in LV tissues of age-matched WT mice were not attenuated by AAV9-shTβRIII–mediated cardiac-specific knockdown (Figure 7E and 7F). These results strongly suggest that manipulating the expression levels of TβRIII did not affect the downstream signaling of β-AR in overexpression-induced spontaneous cardiac hypertrophy.
We demonstrated, for the first time, that (1) increased expression of TβRIII in the cardiomyocytes lead to cardiac hypertrophy independent to the downstream signaling of TGF-β1 and β-AR; (2) cardiac-specific overexpression of TβRIII accelerates stress-induced cardiac hypertrophy and impairment of cardiac function; (3) β-arrestin2, respectively, links TβRIII and CaMKII and is required for TβRIII-mediated activation of CaMKII and cardiac hypertrophy. These findings may provide a novel therapeutic strategy for preventing cardiac hypertrophy.
To date, investigations on the role of TβRIII in the heart have been focused on noncardiomyocytes. Townsend et al10 indicated that endocardial cell epithelial–mesenchymal transformation required TβRIII. We have previously shown that TβRIII was a direct target for microRNA-21, and the transient overexpression of full-length TβRIII plasmid negatively regulated TGF-β pathway by preventing the formation of TβRI/TβRII complex in cardiac fibroblasts.18,19 A report by Hermida et al20 also demonstrated that exogenous administration of P144, a soluble synthetic peptide from extracellular TβRIII domain, exhibited inhibitory effects on collagen synthesis in cardiac fibroblasts. More recently, we reported that simvastatin alleviates cardiac fibrosis, in the case of myocardial infarction mouse model, via upregulation of TβRIII expression in cardiac fibroblasts, through an enhanced interaction between TβRIII and Gα-interacting protein–interacting protein, C terminus to inhibit mitogen-activated protein kinase signaling.21 However, no experimental investigation was conducted to clarify whether TβRIII is expressed in cardiomyocytes and whether altered expression of TβRIII in cardiomyocytes may affect cardiac function. In the present experimental model, TβRIII was overexpressed specifically in cardiomyocytes both in vivo and in vitro, whereas the elevated expression of TβRIII in cardiomyocytes leading to cardiac hypertrophy there was no cardiac fibrosis found in CS-TβRIII-Tg mice, albeit stresses accelerated cardiac fibrosis in these mice. This paradox phenomenon with our previous results may attribute to following factors: (1) overexpression of TβRIII in different cell types may trigger diverse intracellular signaling, as evident in that TβRIII coupled with Gα-interacting protein–interacting protein, C terminus in fibroblast cells to inhibit mitogen-activated protein kinases21 and that TβRIII linked with β-arrestin2 in cardiomyocytes to activate CaMKII; (2) the effects of TβRIII may also depend on experimental models; with regard to pathological processes, myocardial infarction model is distinct to TAC or isoproterenol stimulation; (3) the data generated, from 3-month-old to 1-year-old CS-TβRIII-Tg mouse (data not shown), revealed that while developing cardiac hypertrophy, there was no concomitant cardiac fibrosis. Collectively, the experimental evidences provided in this study do not support the notion that TβRIII has profibrotic effects in hearts when TβRIII overexpressed in cardiomyocytes. However, the mechanisms by which TAC and isoproterenol accelerated cardiac fibrosis in CS-TβRIII-Tg mice may involve in complicated intracellular signaling and need further investigation.
Importantly, we found that mRNA and protein expression levels of TβRIII were significantly increased in the patients’ hypertrophic heart tissue. Consistently, the expression levels of TβRIII in mouse hearts were greatly increased by either isoproterenol stimulation or TAC operation. These results together suggest that TβRIII is a critical regulator for stress-induced cardiac hypertrophy. The use of these animal models in our study allowed the minimization of confounding factors for dissecting the potential molecular mechanisms by which TβRIII regulates cardiac hypertrophy.
It seems that deletion of TβRIII by the disruption of its exon 2 or exon 3 resulted in embryonic lethality for the majority of the null mice or defects in heart development.22,23 These mouse models support essential and nonredundant roles for TβRIII in heart development. In this study, we used cardiac-specific overexpression of TβRIII mouse to determine whether gain-of-function may lead to cardiac hypertrophy and may affect cardiac function. Although CS-TβRIII-Tg mice clearly showed cardiac hypertrophic phenotype, echocardiography did not show cardiac dysfunction in 3-month-old to 1-year-old CS-TβRIII-Tg mouse compared with age matched WT littermate controls (data not shown). However, 3 weeks after TAC operation, CS-TβRIII-Tg mice developed exaggerated cardiac hypertrophy and displayed a strong decrease in ejection fraction and fractional shortening compared with TAC-treated WT mice. Finally, our results show that cardiac-specific AAV9-mediated knockdown of TβRIII partially rescued isoproterenol- or TAC-induced cardiac hypertrophy and preserved ejection fraction and fractional shortening in TAC-treated mice. These results demonstrate that TβRIII is a potential regulator of the development of cardiac hypertrophy and heart failure.
Notably, the highly conserved cytoplasmic domain of TβRIII between species has been shown to greatly affect TβRIII function.4 Specifically, identified sequences within the cytoplasmic tail of TβRIII associate with β-arrestin2, a scaffolding protein.3,12 β-arrestins were originally identified as regulators of G-protein–coupled receptors, which bind to activated receptors, targeting them for internalization and desensitization.24 Surprisingly, a series of studies using a variety of cell types showed that via interaction with β-arrestin2, TβRIII regulates several signaling pathways, including both Smad-dependent and Smad-independent signaling.7,11,25 Collectively, there seems to be considerable cell type specificity in the mechanisms of the TβRIII/β-arrestin2 interaction resulting signaling output. Our data show that TβRIII influences CaMKII activation via interaction with β-arrestin2 in mouse cardiomyocytes.
CaMKII seems to play an important pathogenic role in the development of cardiac hypertrophy and heart failure.26 β-arrestin2 scaffolds CaMKII and is required for isoproterenol-stimulated CaMKII activation in mouse heart.17 Moreover, knockdown of β-arrestin2 prevented cardiac myocyte hypertrophy on β-AR activation.27 We show here that β-arrestin2 serves as a scaffolding protein between TβRIII and CaMKII, suggesting that these proteins are assembled within a TβRIII/β-arrestin2/CaMKII complex. More importantly, isoproterenol- or TAC-induced increase in expression level of TβRIII is paralleled with cardiac hypertrophy and with the elevated magnitudes of TβRIII/β-arrestin2 and β-arrestin2/CaMKII complexes, as well as with the enhanced activation of CaMKII. Moreover, overexpression of TβRIII both in vitro and in vivo resulted in a spontaneous phenotype of cardiac hypertrophy. Conversely, knockdown of TβRIII, both in vitro and in vivo, greatly attenuated stress-induced activation of CaMKII and cardiac hypertrophy. The novel finding of this study implies that β-arrestin2–dependent CaMKII activation is most likely required for stress-induced TβRIII-mediated development of cardiac hypertrophy. Specifically, we identified that TβRIII, a multifunctional sensor, serves as upstream signaling for β-arrestin2–dependent activation of CaMKII.
It is well documented in several studies that TGF-β or β-AR signaling play crucial role in cardiac hypertrophy.28,29 Therefore, we tested whether TβRIII overexpression-induced spontaneous hypertrophy is also linked to downstream signaling of TGF-β1 because TβRIII functions as a coreceptor, presenting ligands to their respective signaling receptors. Our data demonstrate that downstream signaling of TGF-β1 are most likely not involved cardiac hypertrophy because of CS-TβRIII-Tg, because neither TGF-β1 nor p-Smad2/3 levels, in LV tissues, were enhanced in these mice compared with WT mice. Furthermore, the stimulatory effects of TAC or isoproterenol on expression levels of TGF-β1 were not blunted by cardiac-specific AAV9-mediated knockdown of TβRIII, which further confirmed that manipulating TβRIII levels in the hearts does not affect TGF-β1 expression.
β-arrestin2 is known to mediate translocation of CaMKII and Epac to agonist occupied β1-AR complex and then brings these molecules in close proximity to the location of cAMP generation by adenylate cyclase. cAMP directly binds to and stimulates Epac, leading to CaMKII activation via a Rap-PLC-PKC mechanism.17 β-arrestins also scaffold β-ARs and regulate their internalization and desensitization30; however, our data show that there is no direct interaction between TβRIII and β-ARs in mouse hearts. More importantly, the levels of cAMP and protein kinase A were virtually the same in LV tissues between 3-month-old CS-TβRIII-Tg mice and age-matched WT mice. TAC- or isoproterenol-induced increase in cAMP and protein kinase A levels was not decreased by cardiac-specific knocking down of TβRIII. These results together suggest that downstream signaling of β-AR does play a role in TβRIII-induced cardiac hypertrophy, under the conditions without stimulation by the stresses.
However, this study does not provide details as to how TβRIII expression was upregulated in experimental cardiac hypertrophic models and human cardiac hypertrophic specimens. At the transcriptional level, TβRIII expression was positively regulated by dexamethasone, aldosterone, and hydrocortisone.4 Similarly, we also found that angiotensin II, as an inducer for cardiac hypertrophy, was also capable of upregulating TβRIII protein expression in cardiac myocytes.
In summary, this study provides compelling experimental evidence that TβRIII induces cardiac hypertrophy through the β-arrestin2–dependent activation of CaMKII. These results aid our understanding of the mechanisms of cardiac hypertrophy, and interference of TβRIII expression in cardiomyocytes may be a potential approach for the prevention or therapy of myocardial hypertrophy.
The novel mechanisms by which TβRIII drives cardiac hypertrophy provide new mechanistic insights into the cardiac hypertrophy. Discovery of a small molecule to inhibit TβRIII-initiated signaling pathway may be a new strategy for preventing and curing prohypertrophic stresses-induced cardiac hypertrophy and dysfunction.
Sources of Funding
This work was supported, in part, by the National Basic Research Program of China (973 program, 2012CB517803/2014CB542401), and the National Nature Science Foundation of China (31100826/81370244/81270340/81330004), and Heilongjiang Chang Jiang Scholar Candidates Program for Provincial Universities (2013CJHB004).
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.116.07420/-/DC1.
- Received February 26, 2016.
- Revision received March 7, 2016.
- Accepted June 22, 2016.
- © 2016 American Heart Association, Inc.
- Blobe GC,
- Schiemann WP,
- Pepin MC,
- Beauchemin M,
- Moustakas A,
- Lodish HF,
- O’Connor-McCourt MD
- You HJ,
- Bruinsma MW,
- How T,
- Ostrander JH,
- Blobe GC
- You HJ,
- How T,
- Blobe GC
- Tan M,
- Walwyn WM,
- Evans CJ,
- Xie CW
- Mythreye K,
- Blobe GC
- Chen W,
- Kirkbride KC,
- How T,
- Nelson CD,
- Mo J,
- Frederick JP,
- Wang XF,
- Lefkowitz RJ,
- Blobe GC
- Pan Z,
- Sun X,
- Shan H,
- Wang N,
- Wang J,
- Ren J,
- Feng S,
- Xie L,
- Lu C,
- Yuan Y,
- Zhang Y,
- Wang Y,
- Lu Y,
- Yang B
- Mangmool S,
- Shukla AK,
- Rockman HA
- Hermida N,
- López B,
- González A,
- Dotor J,
- Lasarte JJ,
- Sarobe P,
- Borrás-Cuesta F,
- Díez J
- Stenvers KL,
- Tursky ML,
- Harder KW,
- Kountouri N,
- Amatayakul-Chantler S,
- Grail D,
- Small C,
- Weinberg RA,
- Sizeland AM,
- Zhu HJ
- Compton LA,
- Potash DA,
- Brown CB,
- Barnett JV
- Lee NY,
- Kirkbride KC,
- Sheu RD,
- Blobe GC
- Berthouze-Duquesnes M,
- Lucas A,
- Saulière A,
- Sin YY,
- Laurent AC,
- Galés C,
- Baillie G,
- Lezoualc’h F
- Vidal M,
- Wieland T,
- Lohse MJ,
- Lorenz K
- Ferguson SS,
- Downey WE III.,
- Colapietro AM,
- Barak LS,
- Ménard L,
- Caron MG
Novelty and Significance
What Is New?
We report here, for the first time, that there is an enhanced expression of TβRIII in both the human and the mouse cardiac hypertrophy and that elevated expression of TβRIII in the heart independently lead to cardiac hypertrophy.
Our findings disclose a novel regulatory complex that is composed of TβRIII, β-arrestin2, and CaMKII, where the scaffolding role of β-arrestin2 is required for TβRIII-mediated activation of CaMKII and cardiac hypertrophy.
What Is Relevant?
Manipulation of TβRIII expression levels in heart affects distinct prohypertrophic stresses–induced cardiac hypertrophic phenotype, where an increased TβRIII expression promotes cardiac remodeling and the impairment of cardiac function; conversely, stresses-induced cardiac hypertrophy and declined cardiac function are significantly attenuated by downregulation of TβRIII.
Increased expression of TβRIII leads to cardiac hypertrophy via β-arrestin2–dependent activation of CaMKII pathway; modulation of their levels may provide an attractive therapeutic target for controlling myocardial hypertrophy.