Activation of CaMKII induces a myriad of biological processes and plays dominant roles in cardiac hypertrophy. Caveolar microdomain contains many calcium/calmodulin-dependent kinase II (CaMKII) targets, including L-type Ca2+ channel (LTCC) complex, and serves as a signaling platform. The location of CaMKII activation is thought to be critical; however, the roles of CaMKII in caveolae are still elusive due to lack of methodology for the assessment of caveolae-specific activation. Our aim was to develop a novel tool for the specific analysis of CaMKII activation in caveolae and to determine the functional role of caveolar CaMKII in cardiac hypertrophy. To assess the caveolae-specific activation of CaMKII, we generated a fusion protein composed of phospholamban and caveolin-3 (cPLN-Cav3) and GFP fusion protein with caveolin-binding domain fused to CaMKII inhibitory peptide (CBD-GFP-AIP), which inhibits CaMKII activation specifically in caveolae. Caveolae-specific activation of CaMKII was detected using phosphospecific antibody for PLN (Thr17). Furthermore, adenoviral overexpression of LTCC β2a-subunit (β2a) in NRCMs showed its constitutive phosphorylation by CaMKII, which induces hypertrophy, and that both phosphorylation and hypertrophy are abolished by CBD-GFP-AIP expression, indicating that β2a phosphorylation occurs specifically in caveolae. Finally, β2a phosphorylation was observed after phenylephrine stimulation in β2a-overexpressing mice, and attenuation of cardiac hypertrophy after chronic phenylephrine stimulation was observed in nonphosphorylated mutant of β2a-overexpressing mice. We developed novel tools for the evaluation and inhibition of caveolae-specific activation of CaMKII. We demonstrated that phosphorylated β2a dominantly localizes to caveolae and induces cardiac hypertrophy after α1-adrenergic stimulation in mice.
NEW & NOTEWORTHY While signaling in caveolae is thought to be important in cardiac hypertrophy, direct evidence is missing due to lack of tools to assess caveolae-specific signaling. This is the first study to demonstrate caveolae-specific activation of CaMKII signaling in cardiac hypertrophy induced by α1-adrenergic stimulation using an originally developed tool.
- caveolae microdomain
- calcium/calmodulin dependent kinase II
- L-type calcium channel
cardiovascular diseases are one of the major causes of morbidity and mortality in the Western world (34). Heart failure represents the terminal stage of almost all cardiovascular diseases, including myocardial infarction, hypertensive heart disease, cardiomyopathies, valvular insufficiencies, and congenital heart diseases. Cardiac hypertrophy is one of the predominant risks for the development of heart failure (24), and it is regulated by multiple signaling cascades in which protein kinases play central roles. Subcellular compartmentalization of proteins involved in the signaling process allows multiple biological functions of a small number of adrenergic receptors (ARs).
Caveolae are unique flask-like invaginations of cell membrane rich in cholesterol and sphingolipids, formed by caveolin and cavin proteins (16, 35), and multiple molecules, including receptors, kinases, and ion channels, localize to caveolae. Therefore, they are most likely involved in signal transduction and membrane trafficking (33). However, the determination of the physiological properties of caveolae-specific signaling is difficult due to the lack of useful methods for direct assessment of kinase activation inside caveolae.
Calcium/calmodulin-dependent kinase II (CaMKII) is a serine-threonine (Ser/Thr) kinase with a high number of substrates, including ion channels, calcium-handling proteins, and the components of transcriptional machinery (1). In heart, the activation of CaMKII is observed in ventricles of the experimental models of cardiac dysfunction as well as myocardium of heart failure patients (40, 48). The obtained data indicate that CaMKII plays a key role in the development of cardiac hypertrophy and transition from the adaptive responses to heart failure (41). Cardiac-specific overexpression of CaMKIIδ is sufficient to induce the development of cardiomyopathy, whereas its genetic ablation prevents cardiac hypertrophy or transition to cardiac dysfunction after pressure overload (2, 22, 48). Therefore, CaMKII and its substrates represent promising therapeutic targets for the treatment of heart failure. Of note, location of CaMKII activation determines its biological effects (28).
One of the well-known substrates of CaMKII is the voltage-gated L-type calcium channel (LTCC) complex (8). In cardiomyocytes, Ca2+ influx via LTCC plays the essential role in cellular processes, such as excitation-contraction coupling and regulation of gene expression (4). LTCCs are multisubunit protein complexes, formed by a pore-forming α1C-subunit associated with three auxiliary subunits (β, α2δ, and γ) (9). The auxiliary β-subunit enhances the trafficking of α1-subunit to the plasma membrane in coexpressed cells and modulates channel activation (8). Among four β-subunit isoforms of LTCC, β2-subunit is expressed predominantly in heart (23). Phosphorylation of the β2-subunit by cAMP-dependent kinase (PKA) or CaMKII has been proposed as an activation mechanism mediating the extracellular stimuli (7, 20). LTCC was shown to localize not only to T tubules but also to the plasma membrane microdomains, including caveolae, and it is thought to play different biological roles depending on its localization (3, 5, 38). However, the physiological relevance of CaMKII phosphorylation of LTCC is unknown due to the lack of suitable methods for the analysis of caveolae-specific activation of CaMKII, and, therefore, the development of novel methods is crucial for further studies in this field.
In the present study, we developed a novel method to examine caveolae-specific activation of CaMKII using a fusion protein, caveolin-3 tagged by cytosolic domain of phospholamban (PLN). PLN is a 52-amino acid phosphoprotein, and the cytosolic domain of PLN (cPLN) is flexible and phosphorylatable at two distinct amino acid sites (Ser16, mainly phosphorylated by PKA, and Thr17, specifically phosphorylated by CaMKII) (13, 39, 44). Each phosphorylation can be detected using phosphospecific antibodies and is believed to reflect the activation of corresponding kinases in cytosol (11). We successfully analyzed Thr17 phosphorylation of tagged cPLN using phosphospecific antibody. The method presented here allows us to demonstrate that the LTCC β2-subunit is exclusively phosphorylated in caveolae, forms a positive feedback loop with CaMKII, and plays a pivotal role in the development of cardiac hypertrophy caused by α1-adrenergic stimulation in vivo.
MATERIALS AND METHODS
All animal experiments were approved by the Experimental Animal Care and Use Committee of Osaka University and RIKEN Kobe Branch, and all experimental procedures conformed to the Guide for the Care and Use of Laboratory Animals (8th edition) updated by the United States National Research Council Committee in 2011.
Generation of constructs.
Three constructs investigated in this study were obtained as follows. First, PLN and caveolin-3 (cPLN-Cav3) construct was obtained using cDNA of mouse caveolin-3 (kindly provided by R. G. Parton, University of Queensland, Australia), which was subcloned into pcDNA3.1 vector (Invitrogen). Subsequently, synthesized oligonucleotides corresponding to the amino acid sequence of cytosolic domain of PLN (1–22), harboring a HindIII sequence at its 5′-end and a BamHI sequence at its 3′-end, were inserted in the 5′-site of caveolin-3. The oligonucleotide sequence was as follows: sense, 5′-AGCTTCCTCAGCATGGAAAAAGTGCAATACCTCACTCGCTCGGCTATCAGGAGAGCCTCCACTATTGAAATGCCTG-3′; antisense, 5′-GATCCAGGCATTTCAATAGTGGAGGCTCTCCTGATAGCCGAGCGAGTGAGGTATTGCACTTTTTCCATGCTGAGGA-3′. Caveolin-binding domain (CBD)-green fluorescent protein (GFP) was constructed using cDNA corresponding to the enhanced green fluorescent protein (eGFP) sequence, which was excised from the pEGFP vector (Clontech Laboratories) and subcloned into the pcDNA3.1 vector to obtain eGFP expression. Oligonucleotides corresponding to the CBD (10), harboring a KpnI sequence at its 5′-end and a BamHI sequence at its 3′-end, were synthesized as follows: sense, 5′-CTCAGCAGGCCACCATGGC ACGCAACGTGCCCCCCATCTTCAACGACGTGTACTGGATCGCCTTCTCG-3′; antisense, 5′-GATCCGAGAAGGCGATCCAGTACACGTCGTTGAAGATGGGGGGCACGTTGCGTGCCATGGTGGCCTGCTGAGGTAC-3′. The oligonucleotides were subcloned into pcDNA3.1 vector containing the eGFP sequence. CBD-GFP-autocamtide-2-related inhibitory peptide (AIP) was constructed by removing the stop codon of pcDNA3.1 vector harboring a CBD-GFP by mutation generated by the QuickChange Site-Directed Mutagenesis Kit (Agilent Technologies) according to the manufacturer’s instructions. Oligonucleotides with the sequence corresponding to AIP, which inhibits kinase activity of CaMKII (46), were subcloned into the 3′-site of the modified CBD-GFP. cDNAs of rat LTCC β2a-subunit with CaMKII-phosphorylation-site mutation (cm-β2a:T498A) or with PKA/CaMKII-phosphorylation-site mutation (dm-β2a:S478/479A, T498A) were generated using the QuickChange Site-Directed Mutagenesis Kit (Agilent Technologies) according to the manufacturer’s protocol. The expression vectors harboring cDNA sequences coding for PKAcα, an inhibitory peptide of PKA (PKI), and CaMKIIδ were purchased from Open Biosystems.
Chemicals and reagents.
Chemicals, including isoproterenol (ISO), phenylephrine (PE), forskolin, and methyl-β-cyclodextrin, were purchased from Sigma-Aldrich. Adenoviral vectors carrying cPLN-Cav3, CBD-GFP, CBD-GFP-AIP, wild-type rat LTCC β2a-subunit, cm-β2a, and dm-β2a were generated as previously described (15). Briefly, each cDNA was subcloned into a pACCMVpLpA shuttle vector, which was followed by cotransfection with a pJM17 plasmid in HEK 293 cells to allow homologous recombination.
Cell culture and cardiomyocyte preparation.
Transfection of indicated genes into HEK 293 cells and COS-7 cells was performed using FuGENE6 Transfection Reagent (Promega) following the manufacturer’s protocol. Rat neonatal cardiomyocytes were prepared as previously described (15). Briefly, ventricles from 1- to 2-day-old Wistar rats were excised and digested by 0.1% trypsin (GIBCO/Life Technologies) and collagenase (type IV; Sigma-Aldrich), which was followed by differential adhesion, to remove cardiac fibroblasts, in Dulbecco’s modified Eagle’s medium (Sigma-Aldrich) containing 10% feral bovine serum (GIBCO/Life technologies) and 1% penicillin/streptomycin (Wako). Adult cardiomyocytes were isolated as previously described (29) with minor modification. Hearts were excised from heparinized mice and cannulated for retrograde perfusion using a Langendorff apparatus in Tyrode solution containing Liberase Blendzyme (Roche). After perfusion, cardiac cells were dissociated by mechanical dissection followed by stepwise Ca2+ reintroduction.
Immunoblot analyses were performed as described previously (21). Proteins prepared from cultured cells or hearts were separated on SDS-PAGE and transferred to polyvinylidene difluoride membranes (Millipore). The membrane was blocked with 3% BSA or 2% skim milk for 1 h, followed by incubation with primary antibody including anti-β2-subunit of LTCC (kindly provided by Dr. A. Schwartz, University of Cincinnati), anti-α1C-subunit of LTCC (Alomone Laboratories), anti-α2δ-subunit of LTCC (Alomone Laboratories), anti-sarco(endo)plasmic reticulum Ca2+-ATPase 2a (SERCA2; Santa Cruz Biotechnology), anti-PLN (Abcam), anti-Ser16 phosphorylated PLN (Santa Cruz Biotechnology), anti-Thr17 phosphorylated PLN (Santa Cruz Biotechnology), anti-caveolin-3 (BD Transduction Laboratories), anti-phospho-β2a-subunit of LTCC (T498, generated in our laboratory), or anti-GAPDH (Millipore) antibodies overnight at 4°C. Bound antibodies were visualized using horseradish peroxidase-conjugated secondary antibodies (Santa Cruz Biotechnology) and ECL reagent (Promega). Band densities were measured by ImageQuant LAS 4010 using ImageQuant TL software (GE Healthcare).
Neonatal rat ventricular myocytes or isolated adult mouse cardiomyocytes were fixed with 4% paraformaldehyde and blocked with 3% BSA in Tris-buffered saline with/without 0.1% Triton X. The cells were stained with anti-caveolin-3 (BD Transduction Laboratories), anti-PLN (Abcam), and anti-phospho-LTCC β2a-subunit (T498, originally generated) polyclonal antibodies overnight at 4°C. Alexa Fluor 488-conjugated goat anti-rabbit and Alexa Fluor 546-conjugated goat anti-rabbit antibodies (Molecular Probes) were used as secondary antibodies. Nuclei were stained with Hoechst 33258. The stained cells were observed using a fluorescence microscope (TCS SP5; Leica).
Generation of mutant β2a transgenic mice.
Wild-type β2a transgenic or tetracycline transactivator (tTA) transgenic mice, previously described (30), were crossbred with C57BL6/J for six generations. Inducible LTCC β2a-subunit S478/479A-T498A mutant mice (Accession No. CDB0516T: http://www2.clst.riken.jp/arg/TG%20mutant%20mice%20list.html) was generated as follows; cDNA of nonphosphorylated mutant rat LTCC β2a-subunit (S478/479A, T498A) were generated using the QuickChange Site-Directed Mutagenesis Kit (Agilent Technologies) according to the manufacturer’s protocol. The obtained cDNA was subcloned into the modified murine α-myosin heavy chain (α-MHC) promoter expression vector (kindly provided by J. Robbins, Cincinnati Children’s Hospital Medical Center) to confer tetracycline-regulated expression with concomitant tTA expression (37). The construct was linearized and used for pronuclear microinjections administered to C57BL6/N mice. Mutated or wild-type β2a-subunit overexpressing mice were obtained by crossbreeding with tTA transgenic mice. Genotyping of mutated and wild-type β2a transgenic mice was performed by PCR for tail genomes using the following primers to obtain a 263-bp PCR product: forward, 5′-CACCGTGAACATAACCACAG-3′; reverse, 5′-AGAAGGACACCTAGTCAGAC-3′.
Surgical procedures and echocardiograph analysis.
To analyze hypertrophic response to chronic α131). No wt-DTG, mt-DTG, or tTA mice died during chronic PE stimulation. Briefly, the transverse aortic arch was tied with a 27-gauge needle using 7-0 silk ligature, followed by the removal of the needle to constrict the aorta. Two-dimensional and motion mode transthoracic echocardiography was performed using an iE33 model equipped with a 15-MHz transducer (Philips Electronics) as previously described (21). The investigator was blinded to the identity of analyzed mice.−1·day−1) were surgically inserted in the back of double-transgenic and tTA transgenic 2-mo-old mice subcutaneously. Mice (8–11 wk of age) with the three genotypes were subjected to transverse aortic constriction, as previously described (
Real-time reverse transcription-polymerase chain reaction.
Real-time reverse transcription-polymerase chain reaction was performed as previously described (21). Briefly, total RNA was prepared from isolated hearts or cells using QIAzol Lysis Reagent (Qiagen) according to the manufacturer’s instructions. First-strand cDNA was synthesized using total mRNA and standard oligo(dT) primers (Life Technologies). The expression of hypertrophy-related genes, such as atrial natriuretic factor and brain natriuretic peptide, was quantified by real-time PCR using the SYBR Green kit (Applied Biosystems). The primers used in this study are shown in Table 1.
Histological examination of cell surface area and cardiac fibrosis.
Hearts were excised at the indicated time, rinsed, and mounted in optimal cutting temperature compound (Sakura Finetek). The frozen sections (5 μm thick) were prepared and stained with hematoxylin-eosin (HE) or Masson’s trichrome. Cross-sectional areas of cardiomyocytes were measured on HE-stained slides using ImageJ software (National Institutes of Health). Fibrotic areas were quantified as the percentage of blue-stained area, normalized to the whole area from seven independent images of Masson’s trichrome-stained slides of heart tissue, using ImageJ.
Data are presented as means ± SD. Comparisons between two groups were performed using an unpaired t-test. One-way ANOVA with the Tukey-Kramer test was used for multiple comparisons. Differences were considered statistically significant when probability values were <0.05.
Assessment of PKA activation on cardiomyocyte membrane using a novel fusion protein system.
To assess the activation of PKA and CaMKII in caveolae of cardiomyocytes specifically, we designed a fusion protein, cPLN-Cav3 (Fig. 1A). This fusion protein enabled us to detect PKA activation as phosphorylation at PLN Ser16 and CaMKII activation as the phosphorylation of Thr17 in caveolae. Therefore, we could detect the activation of PKA or CaMKII by immunoblotting, using anti-phospho-Ser16 or -Thr17 antibodies, respectively. We demonstrated that the phosphorylation at Ser16 or Thr17 of cPLN-Cav3 was enhanced in caveolae-deficient HEK 293 cells, together with PKAcα or CaMKIIδ coexpression, respectively (Fig. 1B), indicating the fusion protein is useful for the assessment of the intracellular activation of PKA or CaMKII.
We next examined the activation of PKA in caveolae by transient expression of cPLN-Cav3 in COS-7 cells, known to possess caveolar microdomains. Surprisingly, the phosphorylation of Ser16 in cPLN-Cav3 was enhanced in COS-7 cells in comparison with that in HEK 293 cells, which was further augmented by coexpression with PKAcα, suggesting that PKA activation is stronger and that it is constitutively activated in this microdomain (Fig. 1C). Consistently, the phosphorylation of Ser16 in cPLN-Cav3 was attenuated by coexpression with PKI, a PKA inhibitory peptide, in a dose-dependent manner in COS-7 cells (Fig. 1D).
Finally, we assessed the efficacy of this fusion protein system in neonatal rat cardiomyocytes (NRCMs) using cPLN-Cav3 adenoviral vector. In NRCMs, intrinsic PLN and cPLN-Cav3 expression was detected and their phosphorylation levels significantly increased after forskolin stimulation (Fig. 1E). Immunocytochemical analysis demonstrated that cPLN-Cav3 localized specifically on the membrane (Fig. 1F).
Caveolae-specific activation of CaMKII assessed by cPLN-Cav3 fusion protein system.
We compared the phosphorylation levels of cPLN-Cav3 at Thr17 between caveolin-deficient HEK 293 cells and caveolae-containing COS-7 cells. We showed that the phosphorylation of Thr17 in cPLN-Cav3 was more enhanced in COS-7 cells than in HEK 293 cells, which was further augmented by its coexpression with CaMKII, suggesting that CaMKII is more efficiently and constitutively activated in caveolar microdomain (Fig. 2A). To further elucidate caveolar specificity of this fusion protein system, we generated a fusion protein of AIP, a CaMKII inhibitory peptide, harboring GFP and CBD, which enables the molecule to localize specifically in caveolae (Fig. 2B) (10). In COS-7 cells, the coexpression of CBD-GFP-AIP attenuated phosphorylation of Thr17 in cPLN-Cav3 in a dose-dependent manner (Fig. 2C). In NRCMs, targeted expression of cPLN-Cav3 using adenoviral vector revealed distinct phosphorylation at Thr17 of native PLN after ISO or forskolin treatment (Fig. 2D), suggesting CaMKII activation in caveolae is independent of that in cytosol. Those experiments also indicate that PKA activation by forskolin also induced CaMKII activation, which is consistent with a previous report (43). In addition, phosphorylation of cPLN-Cav3 was upregulated by ISO stimulation in a dose-dependent manner (Fig. 2E). Furthermore, caveolae depletion by treatment with methyl-β-cyclodextrin abolished expression of cPLN-Cav3 in NRCM, suggesting a major proportion of fusion protein would degrade outside caveolae (Fig. 2F). Finally, we assessed caveolae-specific inhibition by CBD-GFP-AIP using the cPLN-Cav3 system in NRCMs. We observed that CBD-GFP-AIP mainly colocalized with caveolin-3 in NRCMs (Fig. 2G), and coexpression of CBD-GFP-AIP significantly reduced the phosphorylation of Thr17 in cPLN-Cav3 compared with those of CBD-GFP (Fig. 2H). Taken together, those results indicate successful development of a tool for assessment of caveolae-specific CaMKII activation.
Caveolae-specific activation circuit between LTCC β2a and CaMKII induces cardiomyocyte hypertrophy.
LTCC β2a is phosphorylated by CaMKII, which modulates LTCC activity, closely related to myocyte contraction and cardiac pathogenesis. To clarify the significance of the phosphorylation of LTCC β2a in caveolae, we generated an adenovirus vector expressing LTCC β2a.
In NRCMs, overexpressed LTCC β2a was shown to be constitutively phosphorylated at Thr498, which was previously reported to represent the site phosphorylated by CaMKII (Fig. 3A). Consistently, phosphorylation was not observed by overexpression of either nonphosphorylatable LTCC β2a mutants. Interestingly, the overexpression of LTCC β2a enhanced caveolae-specific activation of CaMKII assessed by phosphorylation of cPLN-Cav3, suggesting that the modulation of LTCC activity by CaMKII also leads to the activation of CaMKII itself in caveolae (Fig. 3B). Indeed, nonphosphorylatable mutant of LTCC β2a expression failed to caveolae-specific activation of CaMKII (Fig. 3B), suggesting a caveolae-specific activation circuit between LTCC β2a and CaMKII exists in NRCMs. Moreover, caveolae-specific inhibition of CaMKII using simultaneous infection with CBD-GFP-AIP expression vector led to a decrease in the phosphorylation of LTCC β2a at Thr498 (Fig. 3C). Additionally, immunocytochemical analysis revealed that ~80% of phosphorylated-β2a colocalized with caveolin-3 in LTCC β2a-overexpressing NRCMs (Fig. 3D). Finally, physiological relevance of caveolae-specific phosphorylation of LTCC β2a was analyzed. The overexpression of LTCC β2a induced myocyte hypertrophy as demonstrated by atrial natriuretic factor expression and cell surface area (Fig. 3, E and F). Notably, the hypertrophic phenotypes were significantly attenuated by caveolae-specific CaMKII inhibition induced by the simultaneous expression of CBD-GFP-AIP (Fig. 3, E and F).
Generation of transgenic mice with nonphosphorylated mutant of LTCC β2a-subunit.
The cardiac-specific and inducible expression was achieved with a binary α-MHC promoter-based transgene system with FVBN strain. The responder transgene permitted the expression of LTCC β2a in cardiomyocytes only in the presence of the driver transgene encoding the tTA protein. First, we backcrossed both α-MHC tTA transgenic mice and wild-type LTCC β2a responder transgenic mice with C57/B6 mice more than six generations to obtain wild-type LTCC β2a-overexpressing mice (wt-DTG) with the C57/B6 mouse strain. Immunocytochemical analysis of the isolated cardiomyocytes from wt-DTG revealed that phosphorylated LTCC β2a dominantly colocalized with caveolin-3, which is consistent with the results obtained from NRCMs, indicating that LTCC β2a is phosphorylated in caveolae of adult mouse cardiomyocytes (Fig. 4A).
Therefore, we generated nonphosphorylated mutant LTCC β2a-overexpressing mice (mt-DTG) to examine the physiological relevance of LTCC β2a phosphorylation and compared them with that of wt-DTG (Fig. 4B). Two responder lines were obtained, both of which displayed ~2.6-fold increase in the expression of LTCC β2a compared with tTA control mice (Fig. 4, B and C). Additionally, the expression levels of protein induced by transgene were similar between wt-DTG and mt-DTG under coexistence of tTA transgene (Fig. 4, B and C). Cardiomyocytes from wt-DTG showed that the phosphorylation of Thr498 in LTCC β2a was increased, whereas this phosphorylation was abrogated in cardiomyocytes obtained from mt-DTG mice (Fig. 4C). No significant alterations were observed in expression of other calcium cycling-related proteins, including LTCC α1C-subunit, α2δ-subunit, SERCA2a, and PLN (Fig. 4, D and E). Furthermore, no cardiac hypertrophy or systolic function alterations were observed (Fig. 4, F and G).
Lack of phosphorylation of LTCC β2a attenuates cardiac hypertrophy after chronic α1-adrenergic stimulation in mice.
In mice, α1-adrenergic stimulation mediates CaMKII activation and induces cardiac hypertrophy. To elucidate the involvement of LTCC β2a phosphorylation in the process, we assessed cardiac hypertrophic response in wt- and mt-DTG mice 2 wk after chronic PE stimulation. In wt-DTG mice, the heart weight normalized to body weight ratio was shown to be increased compared with that in tTA control, and it was shown to be restored in mt-DTG mice (Fig. 5A). While cardiac systolic function assessed by fractional shortenings as well as diastolic function assessed by mitral flow velocity in echocardiographic analyses were similar between the three investigated genotypes, the increase in cell surface area and hypertrophic molecular marker expression observed in wt-DTG heart after stimulation returned to the control levels in mt-DTG hearts (Fig. 5, B–D). Notably, PE stimulation induced caveolar CaMKII activation in NRCMs (Fig. 5E) as well as phosphorylation of LTCC β2a in wt-DTG hearts, which was not observed in mt-DTG heart (Fig. 5F). Finally, the phosphorylation of Thr498 in LTCC β2a was shown to be enhanced in cardiomyocytes isolated from wt-DTG after PE stimulation, but this enhancement was blocked in cardiomyocytes obtained from mt-DTG hearts (Fig. 5G). Taken together, those results indicate that phosphorylation of LTCC β2a by CaMKII in caveolae mediates enhanced hypertrophic response after PE stimulation in wt-DTG hearts.
In this study, we developed a novel tool for the assessment of the PKA and CaMKII activation in caveolae. Twenty-two amino acids belonging to the cytosolic domain of PLN enabled us to estimate this activation by functioning as a phosphopeptide tag. Using this originally developed methodology, we demonstrated that LTCC β2a phosphorylation occurs exclusively in caveolae, and a positive feedback loop develops between LTCC and CaMKII in membrane microdomain. Additionally, we demonstrated that the phosphorylation of LTCC β2a contributes to the acceleration of cardiac hypertrophy induced by chronic α1-adrenergic stimulation. To the best of our knowledge, the presented results represent the first direct evidence showing the contribution of caveolae-specific signaling to the development of cardiac hypertrophy in vivo.
Caveolae play a central role in mediating signal transduction from membrane to nuclei. The signal activation in caveolae is thought to be distinct from the global intracellular signaling, such as cytosolic signal transduction, and to regulate different cellular function (33). Nikolaev et al. reported that the specific β-adrenergic isoform localized in T tubules or plasma membrane transduces different patterns of activation of global or microdomain PKA using a cAMP-fluorescence resonance energy transfer sensor (32). Additionally, Makarewich et al. reported that caveolae-specific inhibition of LTCC by the inhibitory protein REM in the caveolar microdomain abrogates calcineurin/NFAT signaling in NRCMs (26); however, they failed to show the activation of signaling in caveolae directly because of lack of the approach for direct detection of the activation status of signaling molecules. Here, we have developed a simple and efficient method for the monitoring of kinase activation specifically in caveolae. Our results indicate that this method can be used for future studies deciphering the roles of signaling transduction in cellular microdomains.
There is a previous report that has attempted to estimate kinase activity through the synthetized substrate phosphorylation (12). PLN is an sarco(endo)plasmic reticulum-specific phosphoprotein, composed of 52 amino acids, comprising a highly flexible cytosolic domain and an intramembrane domain (25). The principal function of PLN is to inhibit SERCA2a activity and regulate it through protein-protein interactions. This inhibition is released by β-adrenergic stimulation, regulated through the phosphorylation at Ser16 by PKA and Thr17 by CaMKII (25). The phosphorylation status of PLN represents β-AR signaling-induced PKA or both α-AR and β-AR signaling-induced CaMKII cytosolic activation. We used these unique phosphorylatable characteristics of PLN to develop universally applicable methods for monitoring PKA and CaMKII activation in a specific microdomain. First, pharmacological or genetic activation of PKA or CaMKII was shown to induce the phosphorylation of cPLN-Cav3, which could be detected by a corresponding phosphospecific antibody. Next, the phosphorylation levels were decreased by pharmacological or genetic inhibition. Additionally, caveolae-targeting inhibitory peptide, using CBD, was shown to abolish the phosphorylation of cPLN-Cav3. In contrast, ISO stimulation failed to induce phosphorylation of cPLN-Cav3 at Ser16, despite localization of β1-AR in caveolae in a previous report (3). We speculate that loss of caveolar PKA activation by ISO is due to an inhibitory effect by Gαi signaling activation via β2-AR stimulation, which is also localized in caveolae (19, 36). Taken together, the obtained results indicate that the newly developed fusion protein system can reliably assess caveolae-specific activation of PKA and CaMKII in vitro. Therefore, the phosphorylation of cPLN-Cav3 can faithfully represent kinase activities in caveolin-related microdomains (Fig. 6A).
The activation of CaMKII plays a detrimental role in the development of heart failure, leading to the cardiac hypertrophy and cell death (1). Differential localization of CaMKII activation was demonstrated to lead to distinct intracellular functions, and, for example, the overexpression of cytosolic CaMKIIδC isoform in mice affects excitation-contraction coupling (48), whereas the activation of nuclear CaMKIIδB mediates hypertrophic gene induction (47). Therefore, both induce ventricular dilation and failure. Recently, it was reported that the mitochondrial inhibition of CaMKII attenuates necrotic cell death (18), and the localization of CaMKII activation is closely related to cardiac pathogenesis (28). Deciphering the role of CaMKII in the membrane, as a central platform for signal transduction, is necessary. To the best of our knowledge, the data presented here are the first evidence demonstrating its caveolae-related role. We also revealed that an activation circuit, occurring only in caveolae, between CaMKII and LTCC β2a induces cardiac hypertrophy, which was first demonstrated using this technique.
Phosphorylation of LTCC subunits is considered to play a major role in the modulation of multiple cellular functions, including muscle contractions and cell death. In human heart failure, the increased expression of LTCC β2a and CaMKII activation was frequently observed (1, 17). We previously reported that the overexpression of LTCC β2a induces cardiac hypertrophy and failure in mice (30). Recently, several studies showed that CaMKII- and PKA-mediated phosphorylation of α1C-subunit of LTCC facilitates its activity (8, 45). Physiological importance of the phosphorylation of LTCC β2-subunit has been investigated as well, and remains controversial. Phosphorylatable domain deletion was shown not to alter LTCC activity in mice under physiological conditions (6). In contrast to this, mutated β2-subunit overexpression in cardiomyocytes demonstrates that pathological stimulation of this molecule can lead to cell death (20). Therefore, β2-subunit phosphorylation might play a physiological role in cardiac pathogenesis. Consistent with this scenario, our results indicate that the phosphorylation of upregulated LTCC β2-subunit, which is commonly observed in human heart failure (17), is involved in the development of cardiac hypertrophy.
α1-AR are members of the G protein-coupled receptor family and have been demonstrated to accumulate in myocardial caveolae (14) and to be involved in the development of cardiac hypertrophy. Notably, CaMKII is suggested to be involved in activation and nuclear transition of ERK after α1-AR stimulation to develop cardiomyocyte hypertrophy (9). In contrast to this, the compartmentalization of β-adrenergic signaling is a lot more complex, and the caveolar and noncaveolar localization of β-AR contributes to development of cardiac pathology (42). That may represent one of the reasons for the lack of significant differences in survival time and hypertrophy after chronic β-adrenergic stimulation between two transgenic genotypes (data not shown). Furthermore, transverse aortic constriction was performed on both mutant and wild-type DTGs; however, ~50% of animals died within 2 wk, and hypertrophic response in mice that survived was similar in both groups. The difference of the results obtained after α1- or β-AR stimulation may be due to the exclusive localization of α-AR signaling to caveolae, and, in wild-type β2a transgenic mice, this stimulation induced accelerated hypertrophic responses. Taken together, the obtained results indicate that the phosphorylation of increased levels of LTCC β2a induces exaggerated cardiac hypertrophic response after chronic α1-adrenergic stimulation in mice (Fig. 6B).
The transgenic model we developed shows some limitations. We aimed to develop a phosphospecific antibody for PKA phosphorylation site; however, the antibody was not sufficiently specific, which did not allow us to demonstrate that the phosphorylation of any specific site is responsible for the obtained phenotype. Of note, our results indicate that α1-adrenergic stimulation induces PKA activation in caveolae. A previous report indicated that PKA could be activated by α-AR stimulation in adult rat cardiomyocytes (27). Therefore, the presented transgenic model circumvents those noncanonical phosphorylation effects and allows us to assess the role of total upregulated β2a-subunit phosphorylation. Further investigation should elucidate the role of specific PKA or CaMKII phosphorylation sites of β2a-subunit.
Perspectives and Significance
We developed a simple effective tool that enables us to assess the local activation of PKA and CaMKII in caveolin-related membrane microdomain. Using this method, we demonstrated the existence of a positive feedback loop between LTCC and CaMKII in caveolae that leads to the development of cardiomyocyte hypertrophy. Additionally, this method allowed us to demonstrate that the phosphorylation of LTCC β2-subunit occurs specifically in caveolae. Taken together, this originally developed phosphorylatable peptide tag method shows a great potential for deciphering the local signaling mechanism in the membrane and intracellular microdomain.
This work was partially supported by MEXT/JSPS KAKENHI Grant 23390057 and 26293054 (to H. Nakayama), Takeda Science Foundation, and the Mochida Memorial Foundation for Medical and Pharmaceutical Research to H. Nakayama.
No conflicts of interest, financial or otherwise, are declared by the author(s).
K.T., W.O., S.K., S.M., N.H., S.T., K.M., and H.K. performed experiments; K.T. and S.K. analyzed data; K.T. and H.N. prepared figures; K.T., M.O., M.M., M.A., and Y.F. edited and revised manuscript; M.A. and H.N. interpreted results of experiments; H.K. and H.N. drafted manuscript; H.N. conceived and designed research.
We thank Wakako Okamoto and Chiharu Tottori for secretarial work.
- Copyright © 2017 the American Physiological Society