β-Adrenergic Pathway Induces Apoptosis through Calcineurin Activation in Cardiac Myocytes*

Apoptosis of cardiac myocytes is one of the causes of heart failure. Here we examine the mechanism by which the activation of β-adrenergic receptor induces cardiomyocyte apoptosis. Terminal deoxynucleotide transferase-mediated dUTP nick end labeling and DNA ladder analyses revealed that isoproterenol (Iso) induced the apoptosis of cardiac myocytes of neonatal rats through an increase in intracellular Ca2+ levels. The Iso-induced cardiomyocyte apoptosis was strongly inhibited by the L-type Ca2+ channel antagonist nifedipine and by the calcineurin inhibitors cyclosporin A and FK506. Iso reduced the phosphorylation levels of the proapoptotic Bcl-2 family protein Bad and induced cytochrome c release from mitochondria to the cytosol through calcineurin activation. Infusion of Iso increased calcineurin activity by ∼3-fold in the hearts of wild-type mice but not in the hearts of transgenic mice that overexpress dominant negative mutants of calcineurin. Terminal deoxynucleotide transferase-mediated dUTP nick end labeling analysis revealed that infusion of Iso induced apoptosis of cardiac myocytes and that the number of apoptotic cardiomyocytes was significantly less in the hearts of the transgenic mice compared with the wild-type mice. These results suggest that calcineurin plays a critical role in Iso-induced apoptosis of cardiac myocytes, possibly through dephosphorylating Bad.

Heart failure is the final clinical manifestation of a variety of human heart diseases, including idiopathic dilated cardiomyopathy, hypertrophic cardiomyopathy, and coronary artery disease. In these pathologic conditions, cardiomyocytes undergo apoptosis, and a loss of cardiomyocytes is postulated to cause heart failure (1)(2)(3). However, the mechanism by which cardiomyocytes fall into apoptosis is still unknown.
Recently, accumulating evidence has suggested that various factors such as angiotensin II, tumor necrosis factor-␣, and oxidative stress induce apoptosis of cardiac myocytes (4 -6). An adrenergic receptor agonist, norepinephrine, has been reported to be elevated in the plasma of heart failure patients (7) and to induce cardiomyocyte apoptosis (8,9). It has recently been reported that isoproterenol (Iso) 1 induces apoptosis of cardiac myocytes in vivo through ␤-adrenergic receptors (10) and that norepinephrine-induced cardiomyocyte apoptosis is suppressed by protein kinase A inhibitors (8 -10). Selective overexpression of heterotrimeric GTP-binding proteins ␣, which transmit signals from ␤-adrenergic receptors to adenylyl cyclase, has also been reported to induce cardiomyocyte apoptosis in transgenic mice hearts (11). These results suggest that activation of protein kinase A through ␤-adrenergic receptors induces apoptosis of cardiac myocytes during the development of heart failure. Activation of ␤-adrenergic receptors increases intracellular Ca 2ϩ levels through voltage-dependent Ca 2ϩ channels. Elevation of cytosolic Ca 2ϩ has been reported to induce apoptosis in some cell types through activation of a Ca 2ϩ -dependent phosphatase calcineurin (12)(13)(14). Although calcineurin has recently attracted a great attention as a novel regulator of cardiomyocyte hypertrophy (15), it remains unknown whether calcineurin is involved in ␤-adrenergic stimulation-induced cardiomyocyte apoptosis.

EXPERIMENTAL PROCEDURES
Materials-Iso and nifedipine were obtained from Sigma, ionomycin was from Calbiochem, and cyclosporin A was from Wako Chemical Industries, Ltd. (Osaka, Japan). FK506 was the kind gift of Fujisawa Pharmaceutical Co., Ltd. (Osaka, Japan). Anti-Bad and anti-phosphospecific-Bad(serine 136) antibodies were purchased from New England Biolabs, Inc. (Beverly, MA). Anti-Bcl-2 and anti-Bcl-xL monoclonal antibodies were from Transduction Laboratories (Lexington, KY), and anti cytochrome c polyclonal antibody was from Santa Cruz Biotechnology, Inc.
Cell Culture-Primary cultures of cardiac myocytes were prepared from the ventricles of 1-day-old Wistar rats as described previously (16). Cardiomyocytes were plated onto 60-mm plastic culture dishes at a field density of 1 ϫ 10 5 cells/cm 2 and cultured in Dulbecco's modified Eagle's medium with 10% fetal calf serum. Immunocytochemical study revealed that more than 90% of cells were cardiac myocytes.
Terminal Deoxynucleotide Transferase-mediated dUTP Nick End Labeling (TUNEL)-Cardiomyocytes plated on a cover glass were fixed with 4% paraformaldehyde solution for 30 min at room temperature. After a rinse with phosphate-buffered saline, the samples were first incubated with phalloidin-rhodamine for 1 h and with TUNEL reaction mixture containing terminal deoxynucleotidyl transferase and fluorescein isothiocyanate-dUTP. In tissues, the 3-m thick paraffin sections were deparaffinized by immersing in xylene, rehydrated, and incubated in phosphate-buffered saline with 2% H 2 O 2 to inactivate endogenous peroxidases. Next, the sections were incubated with proteinase K (20 g/ml), washed in phosphate-buffered saline, and incubated with terminal deoxynucleotidyl transferase for 90 min and fluorescein isothiocyanate-dUTP for 30 min at 37°C using an apoptosis detection kit (Takara Biomedical). The sections were stained with diaminobenzine for 10 min at room temperature, washed in phosphate-buffered saline, and mounted for light microscopic observations. The number of TUNEL-positive cardiac myocytes was determined by counting 3 ϫ 10 5 cardiac myocytes. All morphometric measurements were performed by at least two independent individuals in a blinded manner.
Agarose Gel Electrophoresis for DNA Fragmentation-Cells (4 ϫ 10 5 ) were lysed in 200 l of lysis buffer (10 mM Tris-HCl (pH 7.4), 10 mM EDTA, 0.5% Triton X-100) followed by incubation with 40 g of RNase (Roche Molecular Biochemicals) for 1 h at 37°C and 100 g of proteinase K (Roche Molecular Biochemicals) for 1 h at 37°C, and only fragmented DNA was extracted. The pellet was resuspended in TE buffer (10 mM Tris-HCl (pH 7.4), 1 mM EDTA) and treated with DNase-free RNase (Roche Molecular Biochemicals) for 1 h at 37°C. DNA was ethanol-precipitated and finally resuspended in distilled water. The fragmented DNA was electrophoretically fractionated on 1.5% agarose gel and stained with ethidium bromide.
To detect cytochrome c, cells were suspended in a buffer (20 mM HEPES (pH 7.5), 10 mM KCl, 1.5 mM MgCl 2 , 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride) for 3 min on ice, homogenized by 10 strokes in a Dounce homogenizer, and centrifuged at 15,000 rpm for 15 min. The supernatant was the cytosol fraction, and the pellet was resolved in lysis buffer as the membrane fraction.
Generation of Dominant Negative (dn) Calcineurin Transgenic Mice-A cDNA encoding human calcineurin A (CnA) was obtained from a T cell gt 10 library using oligonucleotides as hybridization probes (12). ⌬CnA lacking the autoinhibitory and the calmodulin binding domains was constructed by polymerase chain reaction to introduce a stop codon after N407. The catalytically inactive calcineurin mutant (dn calcineurin) was obtained from ⌬CnA by mutating the histidine at position 160, a calcineurin active site, to glutamine (14). ⌬CnA has been reported to prevent Bad redistribution as a trans-dominant negative mutant of CnA induced by an intracellular Ca 2ϩ release agent (14). Hemagglutinin-tagged dn calcineurin was subcloned into the ␣-myosine heavy chain promoter-containing expression vector between the lamin A untranslated region (12) and simian virus 40 poly(A). The linearized DNA was injected into pronuclei of eggs from BDF1 mice, which were transferred into the oviducts of pseudopregnant ICR mice. The transgene was identified by polymerase chain reaction with transgene-specific primers and by Southern blot analysis using a 32 P-labeled simian virus 40 intron sequence. Two lines of 16 -18-week-old dn calcineurin transgenic mice and wild-type littermate mice were used in the present study. All protocols were approved by the guidelines of the University of Tokyo.
Calcineurin Phosphatase Assays-Determination of the calcineurin activity of hearts was performed as described previously (12). Briefly, murine hearts were homogenized in 100 l of lysis buffer (50 mM Tris-HCl (pH 7.5), 1 mM EDTA, 0.5% Tween-20, 0.5 mg/ml bovine serum albumin, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml leupeptin), and after cell debris had been removed by centrifugation at 10,000 ϫ g for 5 min at 4°C, calmodulin-binding calcineurin was separated from calmodulin and non-calmodulin-binding calcineurin with an ultrafree-MC microcentrifuge filtertube (100,000 NMWL filter unit; Millipore Co.), and the supernatant was used for the phosphatase assay. A calcineurin substrate, glutathione S-transferase-RII peptide, which was fixed to glutathione-cellulose beads, was first phosphorylated by protein kinase A in the presence of [␥Ϫ 32 P]ATP (12). The 32 P-labeled RII peptide was incubated in the extracts with 50 l of phosphatase buffer (100 mM Tris-HCl (pH 7.4), 1 mM MnCl 2 , 0.1 mM CaCl 2 , 0.5 mg/ml bovine serum albumin, 100 mM calmodulin, 0.5 mM dithiothreitol) for 30 min at 30°C. 500 nM okadaic acid was added to the reactions to suppress endogenous protein phosphatases PP1 and PP2A. Reactions were stopped by adding 500 l of stop buffer (10% trichloroacetic acid, 0.1 M sodium phosphate, 10 mg/ml bovine serum albumin). After centrifugation, the released [ 32 P]phosphate was determined by Cherenkov methods.
Statistical Analysis-All results are expressed as mean Ϯ S.E. Mul-tiple comparisons among three groups were carried out by 2-way analysis of variance and Fisher's exact test for post hoc analyses. A value of p Ͻ 0.05 was considered significant statistically.

Iso Induces Cardiomyocyte Apoptosis through an Increase in Intracellular Ca 2ϩ
Levels-To elucidate the mechanism by which ␤-adrenergic stimulation induces cardiomyocyte apoptosis, cultured cardiac myocytes of neonatal rats were incubated with a ␤-adrenergic receptor agonist, Iso, and subjected to TUNEL analysis. Less than 5% of cardiac myocytes were TUNEL-positive in the serum-free culture condition. Incubation with Iso for 48 h increased the number of TUNEL-positive cells in a dose-dependent manner (Fig. 1). A calcium ionophore, ionomycin (1 M), which causes sustained elevation of intracellular Ca 2ϩ concentration, also increased the number of TUNEL-positive cells (Fig. 1). Many nuclei of these TUNELpositive cells were condensed and fragmented (data not shown), suggesting that Iso induces apoptosis of cardiac myocytes.
Iso Induces Cardiomyocyte Apoptosis through the Ca 2ϩ -Calcineurin Pathway-Activation of ␤-adrenergic receptors has been reported to increase intracellular Ca 2ϩ levels by increasing Ca 2ϩ influx through voltage-dependent Ca 2ϩ channels (17,18). When cardiomyocytes were treated with the selective Ltype Ca 2ϩ channel antagonist nifedipine (1 M) for 1 h, Isoinduced cardiomyocyte apoptosis was significantly suppressed ( Fig. 2A), suggesting that the Ca 2ϩ influx through L-type Ca 2ϩ channels plays a critical role in Iso-induced cardiomyocyte apoptosis. Elevation of cytosolic Ca 2ϩ has been reported to induce apoptosis in some cell types through activation of a Ca 2ϩ -dependent phosphatase calcineurin (12)(13)(14). We thus examined whether calcineurin is also involved in Iso-induced cardiomyocyte apoptosis. Treatment with the calcineurin inhibitors cyclosporin A (CsA) (1 M) and FK506 (10 g/ml) strongly suppressed Iso-induced apoptosis of cardiac myocytes (Fig. 2, A  and B). To confirm the role of calcineurin in Iso-induced apoptosis, we examined DNA fragmentation by gel electrophoresis. When cardiac myocytes were exposed to Iso (50 M) for 48 h, extracted genomic DNA showed prominent ladder formation characteristic of apoptosis. The Iso-induced DNA ladder formation was almost completely suppressed by treatment with either CsA or FK506 (Fig. 2C). These results suggest that the Ca 2ϩ -calcineurin pathway plays a critical role in ␤-adrenergic receptor-induced apoptosis in cardiac myocytes.
Iso Dephosphorylates Bad through the Ca 2ϩ -Calcineurin Pathway-Apoptosis is determined by the relative balance be- tween proapoptotic molecules such as Bad and Bax and antiapoptotic molecules such as Bcl-2 and Bcl-xL (19 -21). Bad promotes cell death by inhibiting Bcl-2 and Bcl-xL function. Phosphorylation of Bad at Ser-136 by protein kinase B (22) or mitogen-activated protein kinase cascades (21) promotes cell survival. Recently, Ca 2ϩ -mobilizing agents have been reported to dephosphorylate Bad by activating calcineurin and to enhance Bad heterodimerization with Bcl-xL, leading to apoptosis (14). We thus examined the role of Bad in Iso-induced apoptosis of cardiomyocytes. Western blot analysis using antiphospho-Bad antibody revealed that Bad was highly phosphorylated in control cardiac myocytes. Phosphorylation levels of Bad were transiently reduced at 2 h and 4 h after addition of Iso (50 M) (Fig. 3A). A calcium ionophore, ionomycin (1 M), also decreased phosphorylation levels of Bad in cardiac myocytes (Fig. 3A). When the Ca 2ϩ influx and calcineurin activation were inhibited by treatment with nifedipine (1 M) and FK506 (10 g/ml), respectively, Iso-induced dephosphorylation of Bad was abolished (Fig. 3A). These results suggest that the Ca 2ϩ -calcineurin pathway is necessary for Iso-induced dephos-phorylation of Bad in cardiac myocytes. On the other hand, expression levels of Bcl-2 and Bcl-xL were not changed by Iso stimulation (Fig. 3B).
Iso Induces the Release of Cytochrome c from Mitochondria to the Cytosol through the Ca 2ϩ -Calcineurin Pathway-A growing body of evidence indicates that cytochrome c release from mitochondria to the cytosol, which is tightly regulated by the Bcl-2 family proteins, induces activation of caspases, leading to apoptosis (23)(24)(25). We thus examined the subcellular distribution of cytochrome c in cardiomyocytes in the absence or presence of Iso. In unstimulated cardiomyocytes, cytochrome c existed abundantly in the membrane fraction and slightly in the cytosol fraction. From 8 h after exposure to Iso (50 M), the cytosol fraction of cytochrome c was significantly increased, and the increase of cytochrome c in the cytosol continued until 24 h (Fig. 4A). To determine whether calcineurin is involved in the Iso-induced release of cytochrome c, the cells were pretreated with CsA (1 M) and FK506 (10 g/ml). Both calcineurin inhibitors strongly reduced the release of cytochrome c from mitochondria to the cytosol induced by Iso (Fig. 4B). These results suggest that calcineurin plays an important role in the Iso-induced release of cytochrome c from mitochondria to the cytosol in cardiac myocytes.
Iso Induces Less Apoptosis in the Hearts of dn Calcineurin Transgenic Mice-We further examined the role of calcineurin in cardiomyocyte apoptosis of in vivo heart using transgenic mice that overexpress dominant negative mutants of calcineurin (dn calcineurin). The calcineurin activity was increased ϳ3-fold by infusion of Iso (20 mg/kg) in the hearts of wild-type mice but not of the transgenic mice (Fig. 5A). We next evaluated Iso-induced cardiomyocyte apoptosis in the heart using the TUNEL method. There was no TUNEL-positive cardiomyocyte in the hearts of wild-type mice before infusion of Iso. Infusion of Iso (20 mg/kg) induced apoptosis of many cardiac myocytes of wild-type mice (7.2 Ϯ 2.5 of 10 5 cardiac myo-cytes) (Fig. 5, B and C). The number of TUNEL-positive cells was significantly smaller in the hearts of the transgenic mice (1.5 Ϯ 1.3 of 10 5 cardiac myocytes) and wild-type mice administered with FK506 (1 mg/kg/day, intramuscular) for 3 days (1.3 Ϯ 1.5 10 5 cardiac myocytes) than in the hearts of wild-type mice (Fig. 5B). These results suggest that calcineurin is also involved in ␤-stimulant-induced cardiomyocyte apoptosis of in vivo heart. DISCUSSION Many lines of evidence have suggested that activation of the sympathetic nervous system is observed in patients with heart failure and exerts deleterious effects on human hearts (7,26,27). Activation of the sympathetic nervous system provides the rationale for the use of ␤-adrenergic receptor antagonists to treat heart failure. In fact, therapeutic interventions by ␤-adrenergic receptor antagonists not only improve cardiac contractility but also improve the prognosis of heart failure (28 -30). In the present study, we examined the mechanism by which ␤-adrenergic stimulation induces injury of cardiomyocytes.
Apoptosis has been demonstrated to occur in the myocardium in a variety of pathological situations. The number of apoptotic cardiomyocytes is increased in the myocardium obtained from heart failure patients (1,2). Furthermore, prominent cardiomyocyte apoptosis was observed in the hearts of transgenic mice overexpressing GTP-binding proteins ␣, which transmit signals from ␤-adrenergic receptors to adenylyl cyclase, and treatment with ␤-adrenergic receptor antagonists prevented cardiomyocyte apoptosis in mice (31). These observations suggest that ␤-adrenergic stimulation may induce deterioration of cardiac function by inducing cardiomyocyte apoptosis during the development of heart failure. Stimulation of ␤-adrenergic receptors activates adenylate cyclase, which increases intracellular cAMP. The cAMP-dependent protein kinase A activates L-type Ca 2ϩ channels, leading to a significant increase in cytosolic Ca 2ϩ levels. In the present study, Iso increased the number of TUNEL-positive cells. A calcium ionophore, ionomycin, which causes a sustained increase in intracellular Ca 2ϩ concentration, increased the number of TUNELpositive cells as much as did maximum Iso stimulation. Pretreatment with the L-type Ca 2ϩ channel antagonist nifedipine and the selective calcineurin inhibitors CsA and FK506 strongly inhibited the Iso-induced increase in TUNEL-positive cells and Iso-induced DNA ladder formation. Taken together, these results suggest that the Ca 2ϩ -calcineurin pathway plays a critical role in ␤-adrenergic receptor-induced cardiomyocyte apoptosis.
The Bcl-2 family proteins are important regulators of cell death in mammalian cells. Apoptosis is determined by the relative balance between proapoptotic molecules such as Bad and Bax and antiapoptotic molecules such as Bcl-2 and Bcl-xL. Phosphorylated Bad is sequestered in the cytosol by 14 -3-3 proteins and is inactivated, thus promoting cell survival. Dephosphorylated Bad promotes cell death at least in part through heterodimerization with the antiapoptotic proteins Bcl-2 and Bcl-xL. The present study demonstrated that Iso as well as a calcium ionophore transiently reduced phosphorylation levels of Bad and that Iso-induced dephosphorylation of Bad was abolished when the Ca 2ϩ -calcineurin pathway was inhibited by calcineurin inhibitors and an L-type Ca 2ϩ channel antagonist. Furthermore, we examined the subcellular distribution of cytochrome c in cardiomyocytes in the presence or absence of Iso. Iso significantly induced the release of cytochrome c from mitochondria to the cytosol, and the Iso-induced release of cytochrome c was reduced by pretreatment with calcineurin inhibitors. Bcl-2 and Bcl-xL function in the mitochondrial membrane as antiapoptotic molecules by preventing the release of cytochrome c from mitochondria (24,25). When Bad is dephosphorylated and translocated from the cytosol to mitochondria, Bad has been reported to form a complex with Bcl-2 and induce the release of cytochrome c (14). These results and observations collectively suggest that calcineurin plays an important role in the Iso-induced release of cytochrome c from mitochondria to the cytosol, possibly through dephosphorylating Bad in cardiac myocytes.
It has been reported that the calcineurin inhibitors prevent the development of cardiac hypertrophy and cardiomyopathy in rodent models (32). In the present study, to elucidate whether calcineurin is involved in the apoptosis of cardiac myocytes in vivo, we generated dn calcineurin transgenic mice under the control of cardiac myocyte-specific ␣-myosin heavy chain promoter. It has been reported that transfected ⌬CnA induces the redistribution of Bad from the cytosol to mitochondria and the translocation of another calcineurin substrate, nuclear factor of activated T cells, from the cytosol to the nucleus. In contrast, dn calcineurin, a trans-dominant inhibitory mutant of calcineurin, prevented the Ca 2ϩ -induced redistribution of Bad and nuclear factor of activated T cells and reduced apoptosis, indicating that dn calcineurin was effectively suppressing endogenous calcineurin function (14). Infusion of Iso increased the calcineurin activity and the number of TUNEL-positive cells in the heart. Iso-induced increases in the number of TUNEL-positive cells as well as in calcineurin activity were significantly less in the hearts of the transgenic mice or FK506treated mice than in the hearts of wild-type mice. These results suggest that stimulation of ␤-adrenergic receptors also induces cardiomyocyte apoptosis through calcineurin in vivo.
Our results collectively suggest that the Ca 2ϩ -calcineurin pathway plays a critical role in the progression of heart failure by regulating cardiomyocyte apoptosis and that inhibition of the Ca 2ϩ -calcineurin pathway may be effective in the treatment of heart failure. To the contrary, De Windt et al. (33) recently reported that calcineurin activation prevents apoptosis of cardiac myocytes and that calcineurin transgenic mice hearts significantly increased TUNEL-positive cells of noncardiomyocytes but not of cardiomyocytes. The reason for the discrepancy is unknown at present; however, it may come from the different stimuli. De Windt et al. (33) have reported that the activation of calcineurin protects cardiomyocytes against apoptosis induced by 2-deoxyglucose and staurosporine in vitro and by ischemia/reperfusion in vivo. We also observed that ischemia/reperfusion induced more apoptosis in the hearts of transgenic mice expressing dn calcineurin than in the hearts of wild-type mice. 2 It has been reported that calcineurin can activate opposing pathways that either suppress or induce apoptosis in the same cells (34). Further studies are necessary to elucidate in what situations calcineurin induces or suppresses cardiomyocyte apoptosis.
FIG. 5. Iso induces less apoptosis in the hearts of dn calcineurin transgenic mice. A, transgenic mice overexpressing dn calcineurin in the heart were created using the promoter of the ␣-myosin heavy chain gene. To determine calcineurin activity, the lysates of murine hearts were subjected to phosphatase assay using the 32 P-labeled RII peptides. Note that Iso activated calcineurin in the hearts of wild-type mice (wild) but not in the hearts of dn calcineurin transgenic mice (Tg). *, p Ͻ 0.05 versus control wildtype mice. B, at 24 h after Iso infusion (20 mg/kg), the number of TUNEL-positive cardiomyocytes was counted and is presented per 10 5 cardiomyocyte nuclei. There were significantly fewer TUNELpositive cells in the hearts of dn calcineurin transgenic mice or FK506-administered wild-type mice (wildϩFK506) than in the hearts of wild-type mice. *, p Ͻ 0.05 versus wild-type mice infused with Iso. C, representative histological sections from the hearts of Iso-infused transgenic mice. A TUNEL-positive cell is indicated by an arrow.