The β2-Adrenergic Receptor Delivers an Antiapoptotic Signal to Cardiac Myocytes Through Gi-Dependent Coupling to Phosphatidylinositol 3′-Kinase
Abstract—Recent studies have shown that chronic β-adrenergic receptor (β-AR) stimulation alters cardiac myocyte survival in a receptor subtype-specific manner. We examined the effect of selective β1- and β2-AR subtype stimulation on apoptosis induced by hypoxia or H2O2 in rat neonatal cardiac myocytes. Although neither β1- nor β2-AR stimulation had any significant effect on the basal level of apoptosis, selective β2-AR stimulation protected myocytes from apoptosis. β2-AR stimulation markedly increased mitogen-activated protein kinase/extracellular signal–regulated protein kinase (MAPK/ERK) activation as well as phosphatidylinositol-3′-kinase (PI-3K) activity and Akt/protein kinase B phosphorylation. β1-AR stimulation also markedly increased MAPK/ERK activation but only minimally activated PI-3K and Akt. Pretreatment with pertussis toxin blocked β2-AR–mediated protection from apoptosis as well as the β2-AR–stimulated changes in MAPK/ERK, PI-3K, and Akt/protein kinase B. The selective PI-3K inhibitor, LY 294002, also blocked β2-AR–mediated protection, whereas inhibition of MAPK/ERK activation at an inhibitor concentration that blocked agonist-induced activation but not the basal level of activation had no effect on β2-AR–mediated protection. These findings demonstrate that β2-ARs activate a PI-3K–dependent, pertussis toxin–sensitive signaling pathway in cardiac myocytes that is required for protection from apoptosis-inducing stimuli often associated with ischemic stress.
Dysregulated apoptosis has been implicated in the pathogenesis and cellular demise associated with many degenerative diseases.1 Apoptosis of cardiac myocytes has been documented as a consequence of ischemic and reperfusion injury in the heart and as a potential contributing factor to cardiac dysfunction associated with mechanical stretch, hemodynamic overload, and chronic heart failure both in animal models and human disease.2 In addition, chronic exposure of myocytes to neuroendocrine factors, such as angiotensin II and norepinephrine (NE), promotes cardiac cell death, in part, through apoptosis.3 4 5 6 7 For NE and isoproterenol (ISO), this effect is primarily attributable to its signaling through β1-adrenergic receptors (ARs) and the subsequent activation of protein kinase A.5 6 7 Because concurrent β2-AR blockade potentiates NE-mediated apoptosis, it has been suggested that β2-AR stimulation antagonizes β1-AR–stimulated apoptosis7 and may, in fact, protect myocytes. That β1- and β2-AR signaling have markedly different roles in apoptosis is made even more attractive by the fact that these receptors couple to different G proteins.8 9 10 The precise downstream signaling events responsible for the apparent opposing effects of β1- and β2-ARs on apoptosis are unknown. In addition, it is unclear what effect β-AR signaling has on cardiac myocyte apoptosis caused by stimuli other than chronic β1-AR stimulation.
In this study, we examined the effects of selective β1- and β2-AR stimulation on apoptosis in cultured neonatal cardiac myocytes exposed to hypoxia or the reactive oxygen species generated by H2O2. We show that selective β2-AR stimulation prevented changes in cell morphology and nuclear fragmentation characteristic of apoptosis, whereas β1-AR did not. Both the mitogen-activated protein kinase/extracellular signal–regulated protein kinase (MAPK/ERK) and phosphatidylinositol 3-kinase (PI-3K)/Akt pathways have been implicated in intracellular signaling associated with cell survival in cardiac myocytes and other cells.11 12 13 14 Both β1- and β2-AR stimulation increased MAPK/ERK activation. β2-AR stimulation led to increased PI-3K activity and activation of its downstream target, Akt/protein kinase B (PKB), which was significantly greater than that observed with β1-AR stimulation and comparable with that seen with carbachol (CCh), a muscarinic receptor agonist that can also protect cardiomyocytes. The increase in β2-AR–mediated PI-3K activity and Akt activation was inhibited by pertussis toxin (PTX), which also blocked the protective effects of β2-AR stimulation, as did the PI-3K–specific inhibitor LY 294002. In contrast, the mitogen-activated protein/ERK kinase (MEK) inhibitor PD98059 at concentrations sufficient to completely block stimulus-induced MAPK/ERK activity had no effect on protection by β2-AR stimulation. These findings demonstrate that β2-AR can activate a prosurvival signaling pathway mediated through PTX-sensitive PI-3K activation.
Materials and Methods
All cell culture media and additives were purchased from Gibco/BRL and used as described in the Figure⇓ legends and online data supplement (available at http://www.circresaha.org).
Primary Neonatal Myocyte Cultures
Neonatal ventricular myocytes were cultured as previously described,15 16 except that fibroblasts were selectively removed by preplating. Cardiac myocytes were plated at a density of 6.6×104 cells/cm2, cultured for 24 hours, and then switched to serum-free media. Experimental manipulations were started 24 hours after the switch to serum-free conditions, at which time 90% to 95% of the cells stained positively for sarcomeric actin or α-actinin (see the online data supplement). Cells were exposed to hypoxia as described previously.15
Nuclear Fragmentation, TUNEL Staining, DNA Laddering, and Cell Death ELISA
Nuclear fragmentation was detected in fixed (4% paraformaldehyde) cells either by incubating in 10 μmol/L Hoechst 33342 (15 minutes) or by TUNEL staining with a commercially available kit using fluorescein-12-dUTP for detection (Promega Corp). Dead cells were identified by staining cells before fixation with propidium iodide (PI). Eight to 10 fields of ≈50 to 70 cells each were randomly selected from each dish for the determination of total cell number, percent apoptotic nuclei (TUNEL+ or condensed nuclei detected by Hoechst 33342), or percent dead cells (PI+). At least 2 dishes were counted in this manner for each experiment, and at least 3 experiments (separate myocyte preparations) were performed for each manipulation.
Genomic DNA was isolated by proteinase K and phenol-chloroform extraction followed by ethanol precipitation. DNA ladders were detected by size fractionation on 2% agarose gels and staining with Sybr-Gold (Molecular Probes). Cytosolic DNA fragments were detected using a commercially available kit (Cell Death ELISA Plus, Boehringer Mannheim). All measurements were made in triplicate with all results normalized to total cellular protein. Additional details are provided in the online data supplement.
MAPK/ERK and Akt/PKB Western Blot Analysis
Phosphorylation of MAPK/ERK was measured by Western blotting as described previously.17 All data were normalized by reprobing the blot with an antibody to total ERK2 (Santa Cruz Biotechnology, Inc). Akt/PKB phosphorylation was measured using an antibody to phosphoSer473-Akt normalized to total Akt (both antibodies from Cell Signaling Technologies).
PI-3K activity was measured in immunoprecipitates of cardiomyocyte lysates using a p110 antibody that reacts with p110 α, β, γ, and δ (Santa Cruz Biotechnology), as described elsewhere18 19 and in detail in the online data supplement. Quantitation was performed by liquid scintillation counting of the excised portion of the plate corresponding to PI-3–phosphate (PI-P) (determined by comigration with unlabeled standards) (Sigma Chemical Co).
Measurements of cellular cAMP content were performed on clarified supernatants from cells sonicated in 20 mmol/L phosphate buffer (pH 7), 20 mmol/L EDTA, and 1 mmol/L 3-isobutyl-1-methylxanthine and then boiled for 7 minutes. A commercially available spectrophotometric enzyme immunoassay for cAMP was used, and the data were analyzed with regression formulae provided by the manufacturer (Stratagene Cloning Systems, catalogue No. 200020).
All data are presented as mean±SEM. Differences among multiple conditions were determined by ANOVA using a post hoc Tukey’s test. Differences were considered to be significant at a P value of <0.05. The value n represents the number of independent myocyte preparations with each preparation composed of cells pooled from the hearts of at least 50 neonatal rats.
An expanded Materials and Methods section can be found in an online data supplement available at http://www.circresaha.org.
Selective β-AR Stimulation and Neonatal Cardiomyocyte Apoptosis
We first examined the effects of 24 hours of selective β1- and β2-AR stimulation on apoptosis in untreated/normoxic neonatal cardiomyocytes. For these experiments, selective β1-AR stimulation was achieved with NE in combination with the α1-adrenergic receptor-blocker prazosin (Praz) and the β2-AR blocker ICI118,551 (ICI). Selective β2-AR stimulation was achieved with NE/Praz in combination with the β1-AR blocker CGP 20712A (CGP) or with the β2-AR–specific agonist zinterol (ZINT). None of the above treatments had any significant effect on apoptosis in myocytes maintained under untreated/normoxic conditions (Figure 1A⇑). ISO, an agonist with equal affinity for both β1- and β2-ARs, in combination with either ICI (β2-AR blockade) or CGP (β1-AR blockade) also had no significant effect on apoptosis (Figure 1A⇑), nor did the β1-selective agonist dobutamine (untreated, 4.6±0.4% apoptotic [fragmented] nuclei versus 24 hours dobutamine, 5.2±1.1%; P>0.05; n=3).
Both NE/Praz and ZINT did increase intracellular cAMP levels in the myocytes (3.6±0.4- and 2.9±0.6-fold over control, respectively, measured 15 minutes after stimulation) (Figure 1B⇑). Selective β2-AR blockade (ICI) had no effect on NE/Praz-induced intracellular cAMP accumulation but completely blocked ZINT-induced accumulation. On the other hand, selective β1-AR blockade (CGP) completely blocked NE/Praz-induced cAMP accumulation but was without effect on ZINT-induced accumulation. These results demonstrate that the neonatal cardiac myocytes used in the present study were capable of activating both β1- and β2-ARs, at least with respect to cAMP production.
β1- and β2-AR Stimulation and Hypoxia-Induced Apoptosis
We next examined the effects of selective β1- and β2-AR stimulation on cardiac myocytes exposed to 24 hours of hypoxia, a stimulus that is well-documented in neonatal cardiomyocytes to cause cell death through apoptosis.11 15 Figure 2A⇓ shows the images of neonatal cardiomyocyte cultures under normoxia (a through c), hypoxia (d through f), and hypoxia after pretreatment with ZINT (g through i). Cells in these fields were simultaneously stained with Hoechst dye 33342 (a, d, and g) to identify total nuclei and assess changes in their morphology, TUNEL (b, e, and h) to identify nuclei undergoing DNA fragmentation characteristic of apoptosis, and α-actinin (c, f, and i) to identify cardiomyocytes. TUNEL staining is most prominent in hypoxia-treated cells (e), where myocytes undergoing fragmentation are indicated by arrows. Pretreatment with ZINT (g through i) significantly reduced the increase in TUNEL staining and changes in nuclear morphology in myocytes exposed to hypoxia. Figure 2B⇓ shows the quantitative compilation of data derived from multiple fields for various pretreatments and receptor agonists. There was a 2- to 3-fold increase in the percent of fragmented nuclei on exposure to hypoxia (bars 1 and 2). As the example in Figure 2A⇓ illustrates, this occurred predominantly in cells stained positively for α-actinin (ie, myocytes). It was unaffected by preincubation with the β-AR receptor blockers, ICI (bar 3), or CGP (bar 4) alone. Pretreatment with NE/Praz alone also had no significant effect on the percent of fragmented nuclei (bar 5), but the combination of NE/Praz and β1-AR blockade (CGP) completely suppressed hypoxia-induced fragmentation (bar 7). This protective effect of NE/Praz/CGP was β2-AR dependent, because it was blocked by ICI (bar 8). ISO was also protective in both the presence and absence of β1-AR blockade (CGP) (bars 11 and 9, respectively), reducing the percent apoptotic nuclei to that seen in untreated (normoxic) controls (bar 1). The protective effect of ISO was also attenuated by β2 blockade (ICI; bar 10). Not surprisingly, the selective β2-AR agonist ZINT also completely suppressed hypoxia-induced nuclear fragmentation (bar 12), an effect that was completely abolished by β2-blockade (ICI; bar 13) and unaffected by β1-blockade (CGP; bar 14). Together, these results demonstrate that β2-AR stimulation protects cardiomyocytes from the morphological changes and nuclear fragmentation associated with hypoxia-induced apoptosis.
After 24 hours of hypoxia, there were 7.8±3.5% fewer cells attached than in control (normoxia) dishes. Of the cells attached, 3.2±0.5% of the cells were dead (PI+). A similar percentage of dead cells were also seen in control cultures (see Table 1 online, available at http://www.circresaha.org). Although ZINT pretreatment reduced apoptosis to control levels, it did not prevent cell loss or reduce the percent of PI-positive cells.
Next, the ability of β2-AR stimulation to block DNA laddering associated with apoptosis was examined. Figure 3A⇓ shows DNA laddering results for normoxic and hypoxic myocytes with and without ZINT, whereas Figure 3B⇓ shows the results of an ELISA used to detect nucleosomal DNA in the cytosol. Both methods revealed a large increase in fragmented nucleosomal DNA caused by exposure of the cells to hypoxia. The relative increase in apoptosis caused by hypoxia and measured by these assays was greater than that observed with Hoechst staining (Figure 2B⇑), because cells that have progressed to the later stages of apoptosis detach from the dish and are not included in the dye assay. However, these floaters are collected for the analyses shown in Figures 3A⇓ and 3B⇓.
ZINT caused a large and significant reduction in hypoxia-induced nucleosomal fragments that was evident in both assays. For the group of experiments shown in Figure 3B⇑, ZINT blocked hypoxia-induced DNA fragmentation by 78±5.2%, which was completely prevented by β2-AR blockade (ICI). A large reduction in hypoxia-induced DNA fragmentation was also observed in cells preincubated with ISO or ISO+β1-AR blockade (CGP) (Figure 3B⇑) and was β2-AR dependent.
A relatively small percentage of cells in the neonatal cardiomyocyte cultures were not myocytes (eg, α-actinin negative) and were referred to as fibroblasts (Figures 1⇑ and 2⇑ online). Although most, if not all, of the apoptosis occurring in response to hypoxia was confined to cardiomyocytes (Figure 2A⇑),15 the possibility exists that the fibroblasts are responsible for the β2-AR–mediated protection of cardiomyocytes through a paracrine mechanism. This was tested by examining the effect of conditioned media from ZINT-treated pure fibroblast cultures (Figure 2⇑ online) on hypoxia-treated cardiomyocytes in the presence of β2-AR blockade. Fibroblast conditioned media protected the myocytes, but this effect was independent of β2-AR stimulation (Figure 3⇑ online). Thus, although fibroblasts may influence the survival of cardiomyocytes, paracrine stimulation of cardiomyocyte survival by fibroblasts is not involved in β2-AR–mediated protection.
β2-AR Protection From Apoptosis Is Mediated Through a PTX-Sensitive Pathway
β2-ARs couple to both Gs- and Gi-mediated signaling pathways, whereas β1-ARs apparently couple only to Gs.8 9 10 To determine if the protective effect of selective β2-AR stimulation was related to its differential ability to engage Gi signaling pathways, we pretreated cardiomyocytes with PTX to inactivate this pathway. Figure 4A⇓ shows that the reduction in DNA laddering in hypoxic myocytes exposed to ZINT was abolished by pretreatment with PTX. Figure 4B⇓ provides quantitative data using the nucleosomal DNA ELISA, showing that the reduction in nucleosomal fragments detected in hypoxic samples exposed to ZINT is no longer seen in the presence of PTX. To determine if other G protein–coupled receptors known to engage Gi-dependent signaling pathways could also protect cardiomyocytes from hypoxia-induced apoptosis, we pretreated the cells with the muscarinic receptor agonist CCh. As shown in Figure 4B⇓, CCh completely prevented hypoxia-induced apoptosis, and its ability to do so was inhibited by PTX pretreatment. Together, these results indicate that signaling events emanating from Gi proteins coupled to ligand activated receptors, such as the β2-AR, protect cardiomyocytes from apoptosis induced by hypoxia.
Gi-Mediated Protection From Apoptosis Requires PI-3K but Not MAPK/ERK Activation
The MAPK/ERK and PI-3K pathways have been shown to protect cells from apoptosis.13 14 20 21 22 23 To determine if signaling through these pathways was linked to the protective effect of β2-AR stimulation on hypoxia-induced cardiomyocyte apoptosis, we first measured MAPK/ERK and PI-3K activation after stimulation with β-AR agonists and CCh.
MAPK/ERK activation was assessed by measuring agonist-mediated phosphorylation of ERK1 and ERK2 (Figure 5A⇓, left). The relative values determined by scanning were then normalized to total ERK2 protein (shown below the phospho-ERK blot) and plotted as fold increases over basal expression in the graph to the right. The results demonstrate that stimulation of both β1- and β2-AR (ISO+ICI and ISO+CGP, respectively), as well as muscarinic receptors responsive to CCh, increased MAPK/ERK activation at least 10-fold. The increases in MAPK/ERK activation in CCh-treated and ISO+CGP (β2 mode)–treated cells were markedly inhibited by pretreatment with PTX, whereas that of ISO+ICI (β1 mode) was unaffected.
PI-3K activation was measured using a lipid kinase assay to monitor the conversion of PI into PI-P (Figure 5B⇑). Selective β1-AR stimulation (ISO+ICI) caused a small but significant increase in PI-P production, whereas both ZINT and CCh caused even greater PI-P production, which in both cases was completely suppressed by PTX. As a positive control, myocytes were also stimulated with insulin-like growth factor-1 (IGF-1), which had a markedly greater effect in stimulating PI-P than any of the G protein–coupled receptor agonists.
To confirm these findings regarding PI-3K and identify possible downstream targets for PI-3K activity in cardiomyocytes, we also examined the phosphorylation status of Akt/PKB. Figure 5C⇑ shows the results using an antibody specific to phospho(Ser473)-Akt, which paralleled the PI-3K activity results. ISO and ICI caused a small increase in Akt phosphorylation, with the effect of ZINT and CCh being markedly greater. IFG-1 also had a markedly greater effect on stimulating Akt phosphorylation, reflecting its much larger PI-3K response.
To examine the functional significance of the PTX-dependent activation of PI-3K in antiapoptotic signaling, cardiac myocytes were pretreated for 1 hour with the selective PI-3K inhibitor LY294002, exposed to ZINT, and then subjected to hypoxia. Figure 6A⇓ shows the effects of the inhibitor on the phosphorylation of Akt/PKB, an immediate target of PI-3K activation.13 Stimulation of the cells with ZINT caused a large increase in Akt/PKB phosphorylation that was effectively blocked by 1 μmol/L LY294002. At this concentration, LY294002 had no effect on the basal level of nuclear fragmentation observed in untreated/normoxic myocytes or hypoxic myocytes, yet it effectively blocked the ability of either ZINT or CCh to prevent hypoxia-induced nuclear fragmentation. At 10 μmol/L LY 294002, a concentration more commonly used in similar studies, the inhibitor significantly increased hypoxia-induced apoptosis in the absence of ZINT or CCh, suggesting that it may be affecting survival through a mechanism different from that triggered by these survival factors. These results demonstrate that, whatever its source, Gi-dependent PI-3K activation results in a strong prosurvival signal in cardiomyocytes that can effectively disable an apoptotic stimulus.
In contrast to PI-3K inhibition, inhibition of MAPK/ERK activation using the MEK1 inhibitor PD98059 (10 μmol/L) had no significant effect on the ability of ZINT or CCH to protect cells from hypoxia-induced apoptosis (Figure 7B⇓), although it effectively inhibited the increase in ERK-2 phosphorylation caused by these agonists (Figure 7A⇓). At 25 to 50 μmol/L, a concentration far in excess of that needed to block stimulus-induced MAPK/ERK activation, PD98059 did block protection by ZINT and CCh, but it also caused increased apoptosis in untreated/normoxic myocytes (untreated 4.9±0.5% versus 9.2±1.2%; P<0.01, n=4).
β2-AR Stimulation Protects Neonatal Cardiomyocytes From H2O2-Induced Cell Death
Figure 8⇓ shows the effect of β2-AR stimulation (ZINT) on the changes in cellular and nuclear morphology induced by exposure of the cells to hydrogen peroxide (H2O2). Exposure to H2O2 caused extensive rounding up and detachment of the cells from the culture substratum (Figure 8A⇓, panel b) and nuclear fragmentation assessed by staining with the Hoechst dye (Figure 8B⇓). The extent of nuclear fragmentation (42±8.2%) was much greater than that seen with hypoxia, as was cell loss (21.2%). ZINT completely prevented this cell loss (Table 1 online) as well as the rounding of the cells (Figure 8A⇓, panel c) while dramatically reducing the number of cells containing fragmented nuclei. As was the case for hypoxia, these protective effects of β2-AR stimulation were effectively blocked by the PI-3K inhibitor LY294002 (Figures 8A⇓ [panel d] and 8B).
We examined the signaling pathways through which chronic β-AR stimulation affects cardiomyocyte survival using an experimental model of apoptosis in which cultured neonatal rat cardiomyocytes were exposed to hypoxia or H2O2. We demonstrate that signaling from β2-ARs, but not β1-ARs, prevented cardiomyocyte apoptosis caused by these different stimuli. We also show that the selective effect of β2-AR was the result of its ability to couple to Gi proteins and activate a PTX-sensitive signaling pathway that increases intracellular PI-3K activity. The importance of this pathway was reinforced by our finding that stimulation of Gi-coupled cardiac muscarinic receptors was as effective as β2-AR stimulation in activating PI-3K activity and preventing myocyte apoptosis. Taken together, these results show that β2-AR stimulation has a broad antiapoptotic effect attributable to the Gi-dependent engagement of the well-described PI-3K survival pathway.
The results of Communal et al7 show that β2-AR blockade potentiated β1-AR induced apoptosis, suggesting that β2-AR stimulation could protect myocytes from apoptosis caused by chronic β1-AR stimulation. Because selective receptor stimulation was achieved in that study using nonselective agonists coupled with selective receptor blockade, the stimulus for and potential modifier of apoptosis were not independent of one another so that the evidence for a protective role was indirect. We used an independent apoptotic stimulus in combination with a selective β2-AR agonist (ZINT) and showed that receptor stimulation fully protected myocytes against hypoxia-induced apoptosis. This is direct proof of a protective role for β2-AR stimulation and suggests that the receptor may be effective against diverse apoptotic stimuli. In support of this, we also showed that selective β2-AR stimulation prevented peroxide-induced apoptosis and myocyte cell loss (Figure 8⇑). The broad range of protection afforded by the β2-AR is apparently the result of its ability to activate PI-3K, an important cell-survival signaling event observed in many different cell types.13 20 21 22
We have also shown that MAPK/ERK activation plays no role in β2-AR protection against hypoxia-induced cell death (Figure 7⇑). Thus, inhibition of MAPK/ERK with the MEK1 inhibitor PD98059 had no effect on the ability of either β2-AR agonists or CCh to protect cardiomyocytes from hypoxia-induced apoptosis (Figure 7B⇑). The dose of the inhibitor used in this study was at least 5-fold less than that used by others and was chosen as the minimum dose required to block agonist-induced MAPK/ERK activation (Figure 7A⇑). Our results do not exclude a role for the MAPK/ERK signaling pathway in cardiomyocyte survival caused by other stimuli but clearly show that it is not part of the mechanism through which β2-AR agonists or CCh protect neonatal cardiomyocytes from hypoxia-induced cell death.
Given that significant metabolic and biochemical differences exist between neonatal and adult myocytes, there are likely to be important limitations in translating the findings reported here to adult myocytes in the intact heart. Although the relative proportion of β2-ARs in isolated adult and neonatal cardiomyocytes have been reported to be similar, coupling of β2-ARs to downstream events has been reported to be stronger in neonatal cardiomyocytes.24 This observation is consistent with the results we report here, in which the prosurvival response in Figure 2B⇑ with ISO alone suggests a preferential activation of β2-AR signaling. In contrast, a protective effect for ISO in adult cells is only revealed with concomitant β1-AR blockade.7 Another difference noted in this study is the lack of an apoptotic response of neonatal cardiomyocytes to selective β1-AR stimulation, although there was a trend toward increased apoptosis by NE and ISO in the presence of complete β2-AR blockade (Figure 1A⇑). In adult myocytes, β1-AR stimulation causes significant apoptosis, although the absolute extent of apoptosis over basal levels is only about 2-fold.5 7 Other studies have observed ISO-induced apoptosis in neonatal cardiac myocytes, but at 10-fold higher concentrations and 2- to 3-fold longer incubation times than used in this study.6
A recent report on adult cardiomyocytes by Kang et al25 showed that after 24 hours of hypoxia, ≈50% of the cells were PI+ and that only a small fraction of this cell death could be attributed to apoptosis (ie, inhibited by zVAD-fmk or infection with an adenovirus expressing bcl-2). In contrast, we and others15 have observed a significant amount of apoptosis in neonatal cardiomyocytes in response to hypoxia. We also report a significant amount of cell loss as well as a small but constant percentage of cell death attributable to nonapoptotic mechanisms (PI+ cells in Table 1). Although apoptosis in hypoxia-treated myocytes was completely prevented by ZINT, ZINT failed to prevent the unexplained cell loss. In contrast, both apoptosis and total cell loss in response to H2O2 were completely prevented by ZINT. Interestingly, Kang et al25 found that reoxygenation, which would be expected to increase reactive oxygen species such as H2O2, caused cell death primarily through apoptosis. These results suggest that β2-AR agonists, such as ZINT, can prevent only cell death attributable to apoptosis.
In summary, we have shown that signaling from β2-ARs protects neonatal cardiomyocytes from hypoxia- and reactive oxygen species–induced apoptosis. This ability of β2-AR can be traced to its selective coupling to Gi proteins and is shared by the Gi-coupled receptor for the muscarinic agonist, CCh. Gi proteins activate downstream signaling events that trigger, among other things, a PI-3K–dependent cell survival pathway. Inhibition of Gi coupling to PI-3K or of PI-3K itself inhibits the protective action of β2-AR stimulation. These findings indicate that β2-AR signaling protects myocytes from diverse apoptotic stimuli and contributes to the complex role that adrenergic signaling plays in the normal and diseased heart.
This study was funded through the Intramural Research Program of the National Institute on Aging, National Institutes of Health.
- Received June 1, 2000.
- Revision received October 11, 2000.
- Accepted October 11, 2000.
- © 2000 American Heart Association, Inc.
Thompson CB. Apoptosis in the pathogenesis and treatment of disease. Science. 1995;267:1465–1462.
Haunstetter A, Izumo S. Apoptosis: basic mechanisms and implications for cardiovascular disease. Circ Res. 1998;82:1111–1129.
Leri A, Claudio PP, Li Q, Wang X, Reiss K, Wang S, Malhotra A, Katsjura J, Anversa P. Stretch-mediated release of angiotensin II induces myocyte apoptosis by activating p53 that enhances the local rein-angiotensin system and decreases the bcl2-to-bad ratio in the cell. J Clin Invest. 1998;101:1326–1342.
Mann DL, Kent RL, Parsons B, Cooper G IV. Adrenergic effects on the biology of the adult mammalian cardiomyocyte. Circulation. 1992;85:790–804.
Communal C, Singh K, Pimental DR, Colucci WS. Norepinephrine stimulates apoptosis in adult rat ventricular myocytes by activation of the β-adrenergic pathway. Circulation. 1998;98:1329–1334.
Iwai-Kanai E, Hasegawa K, Araki M, Kakita T, Morimoto T, Sasayama S. α-and β-Adrenergic pathways differentially regulate cell type-specific apoptosis in rat cardiac myocytes. Circulation. 1999;100:305–311.
Communal C, Singh K, Sawyer DB, Colucci WS. Opposing effects of β1- and β2-adrenergic receptors on cardiac myocyte apoptosis. Circulation. 1999;100:2210–2212.
Xiao RP, Avdonin P, Zhou Y, Cheng H, Akhter SA, Eschenhagen T, Lefkowitz RJ, Koch WJ, Lakatta EG. Coupling of β2-adrenoreceptor to Gi protein and its physiological relevance to murine cardiac myocytes. Circ Res. 1990;84:43–52.
Xiao RP, Ji X, Lakatta EG. Functional coupling of the β2-adrenoreceptor to a pertussis toxin-sensitive G protein in cardiac myocytes. Mol Pharmacol. 1995;47:322–329.
Matsui T, Ling L, del Monte F, Fukui Y, Franke TF, Hajjar RJ, Rosenzweig A. Adenoviral gene transfer of activated phosphatidylinositol 3′-kinase and Akt inhibits apoptosis of hypoxic cardiomyocytes in vitro. Circulation. 1999;100:2373–2379.
Erhardt P, Schremser EJ, Cooper GM. B-raf inhibits programmed cell death downstream of cytochrome c release from mitochondria by activating the MEK/Erk pathway. Mol Cell Biol. 1999;19:5308–5315.
Kennedy SG, Kandel ES, Cross TK, Hay N. The PI 3-kinase/Akt signaling pathway delivers an anti-apoptotic signal. Genes Dev. 1997;11:701–713.
Sheng Z, Knowlton K, Chen J, Hoshijima M, Brown JH, Chien KR. Cardiotropin I (CT-1) inhibition of cardiac myocyte apoptosis via a mitogen-activated protein kinase-dependent pathway. J Biol Chem. 1997;272:5783–5791.
Zheng JS, Boluyt MO, O’Neill L, Crow MT, Lakatta EG. Extracellular ATP induces immediate early gene expression but not cellular hypertrophy in neonatal cardiac myocytes. Circ Res. 1994;74:1034–1041.
Scheid MP, Lauener RW, Duronio V. Role of phosphatidylinositol 3-OH-kinase activity in the inhibition of apoptosis in hematopoietic cells. Biochem J. 1995;312:159–162.
McGuire TF, Corey SJ, Sebti SM. Lovastatin inhibits platelet-derived growth factor (PDGF) stimulation of phosphatidylinositol 3-kinase activity as well as association of p85 subunit to tyrosine-phosphorylated PDGF receptor. J Biol Chem. 1993;268:22227–22230.
Yao R, Cooper GM. Requirement for phosphatidylinositol 3-kinase in the prevention of apoptosis by nerve growth factor. Science. 1995;267:2003–2006.
Datta SR, Brunet A, Greenberg ME. Cellular survival: a play in three Akts. Genes Dev. 1999;13:2905–2927.
Kuznetsov V, Pak E, Robinson RB, Steinberg SF. β2-adrenergic receptor actions in neonatal and adult rat ventricular myocytes. Circ Res. 1995;76:40–52.
Kang PM, Haunstetter A, Aoki H, Usheva A, Izumo S. Morphological and molecular characterization of adult cardiomyocyte apoptosis during hypoxia and reoxygenation. Circ Res. 2000;87:118–125