The β-adrenergic receptor (βAR) signaling system is one of the most powerful regulators of cardiac function and a key regulator of Ca2+ homeostasis. We investigated the role of βAR stimulation in augmenting cardiac function and its role in the activation of Ca2+/calmodulin-dependent kinase II (CaMKII) using various βAR knockouts (KO) including β1ARKO, β2ARKO, and β1/β2AR double-KO (DKO) mice. We employed a murine model of left anterior descending coronary artery ligation to examine the differential contributions of specific βAR subtypes in the activation of CaMKII in vivo in failing myocardium. Cardiac inotropy, chronotropy, and CaMKII activity following short-term isoproterenol stimulation were significantly attenuated in β1ARKO and DKO compared with either the β2ARKO or wild-type (WT) mice, indicating that β1ARs are required for catecholamine-induced increases in contractility and CaMKII activity. Eight weeks after myocardial infarction (MI), β1ARKO and DKO mice showed a significant attenuation in fractional shortening compared with either the β2ARKO or WT mice. CaMKII activity after MI was significantly increased only in the β2ARKO and WT hearts and not in the β1ARKO and DKO hearts. The border zone of the infarct in the β2ARKO and WT hearts demonstrated significantly increased apoptosis by TUNEL staining compared with the β1ARKO and DKO hearts. Taken together, these data show that cardiac function and CaMKII activity are mediated almost exclusively by the β1AR. Moreover, it appears that β1AR signaling is detrimental to cardiac function following MI, possibly through activation of CaMKII.
- Ca2+/calmodulin-dependent kinase II
- heart failure
- knockout mice
- pressure-volume relations
β-adrenergic receptors (βARs) play a critical role in the maintenance of cardiac homeostasis (10). Regulation of these homeostatic mechanisms may in part be due to differences in the signaling pathways that are activated by the two major βAR subtypes in the heart (11, 49). Although both β1 and β2AR subtypes can couple to Gs proteins to activate adenylyl cyclase and increase cAMP levels (41), β2ARs are also able to couple to Gi, which may promote cell survival signals and protect the heart during heightened sympathetic nervous system activity (23, 48, 54). However, to what degree β1AR promote detrimental signaling to cause pathological cardiac dysfunction is not well understood.
Evidence to support a pathogenic role for β1ARs in heart failure comes from studies showing that overexpression of β1ARs in mice causes early hypertrophy and interstitial fibrosis followed by marked cardiac dysfunction (18). In contrast, modest overexpression of β2ARs (60- to 100-fold over endogenous levels) results in enhancement of cardiac function without deterioration of cardiac function (30, 34). Moreover, in isolated myocytes, β2AR stimulation protects against stress-induced apoptosis (12, 51), whereas in adult mice lacking β2ARs chronic isoproterenol (Iso) stimulation results in enhanced myocyte apoptosis and increased mortality compared with mice lacking β1ARs (39). In addition, in a rodent model of myocardial infarction (MI)-induced heart failure, selective activation with a β2AR agonist improves cardiac performance without promoting apoptosis (1). Although these data support a concept that β2AR signaling is cardioprotective, it has been suggested that opposing the protective effect of β2AR signaling is β1AR mediated activation of Ca2+/calmodulin-dependent kinase II (CaMKII) (55).
CaMKII belongs to a family of multifunctional protein kinases that regulate Ca2+ homeostasis and are essential for normal cardiac function (5, 7, 9). Enhanced βAR stimulation has been shown to activate apoptotic signaling pathways via PKA-independent stimulation of L-type Ca2+ channels, resulting in the activation of CaMKII (7, 55), which may participate in the deterioration of cardiac function in the failing heart (52). Precisely which βAR subtype is responsible for activation of CaMKII in the heart has not been fully elucidated.
To test the individual contribution of β1 and β2ARs in determining normal left ventricular (LV) function and whether βAR subtypes are important for the development of cardiac dysfunction, we studied knockout (KO) mice lacking either β1ARs (β1AR knockout, β1ARKO), β2AR (β2AR knockout, β2ARKO) or both β1ARs and β2ARs (β2AR, and β2AR knockouts, DKO). In this study we identify the specific contribution of the β1AR on LV contractile function and CaMKII activation and the role the β1AR plays in the development of cardiac dysfunction following MI.
Experimental animals and surgical preparation.
The animals in this study were handled according to animal welfare regulations and protocols approved by the authors' Institutional Review Boards. Genetically engineered, 8- to 12-wk-old β1ARKO, β2ARKO, DKO, and C57/B6 wild-type (WT) mice were used for this study (43). Mice were anesthetized with a mixture of ketamine (100 mg/kg) and xylazine (2.5 mg/kg). Under a dissecting microscope Nissho Optical T-240 (Labtek, Campbell, CA) animals were placed in the supine position on a heated operation board and a midline cervical incision was made to expose the trachea. Following successful endotracheal intubation, the cannula was connected to a rodent ventilator. At study termination, hearts were removed and chambers were separated and weighed. The entire LV (infarct and noninfarct zone) was divided in half between the base and apex and used for histology and biochemical assays, respectively.
Hemodynamic measurement and P-V loops.
In vivo pressure-volume (P-V) analysis was performed as previously described (19, 40). Briefly, after bilateral vagotomy, the chest was opened and the pericardium was dissected to expose the heart. A 7-0 suture ligature was placed around the transverse aorta to manipulate loading conditions. A 1.4-Fr pressure-conductance catheter (Millar Instruments, Houston, TX) was inserted into the right carotid artery and advanced retrograde into the LV to record hemodynamics. A polyethylene-50 catheter was inserted into the right external jugular vein for Iso infusion. Steady-state pressure and volume measurements were recorded at baseline and after 3-min Iso infusion (20 pg·g−1·min−1). Pressure and volume measurements were obtained during the increase in the afterload generated by gently pulling on the suture to transiently constrict the aorta. Subsequently, parallel conductance (Vp) was determined by 10-μl injection of 15% saline into the right jugular vein to establish the offset due to the conductivity of structure to the blood pool. The derived Vp was used to correct the P-V loop data. Data were recorded digitally at 1,000 Hz and analyzed with pressure volume analysis software (PVAN data analysis software version 3.3; Millar Instruments) as previously described (40).
Western blotting analysis of CaMKII activation and PLB phosphorylation.
To evaluate the activation of CaMKII and phospholamban (PLB) phosphorylation, LV tissue was homogenized in Nonidet P-40 lysis buffer containing 20 mM Tris (pH 7.4), 137 mM NaCl, 1% Nonidet P-40, 20% glycerol, 10 mM PMSF, 1 mM Na3VO4, 10 mM NaF, 2.5 μg/ml aprotinin, and 2.5 μg/ml leupeptin. Protein concentration of sample was assayed with Bio-Rad protein assay reagent, and 100 μg of protein was denatured by heating at 95°C for 5 min prior to resolution by SDS-PAGE. Total CaMKII and pCaMKII (autophosphorylation site at Thr-286) were detected by using anti-CaMKII (1:2,000, Santa Cruz) and anti-phosphorylated CaMKII at Thr-286 (1:2,000, Cell Signaling), respectively. PLB phosphorylation was detected with the phosphorylation site-specific antibodies recognizing phospho-PLB (Ser-16) or phospho-PLB (Thr-17) (1:2,000, Santa Cruz). Total PLB was detected with polyclonal antibody against PLB (1:2,000, Santa Cruz). Immunoblots were revealed by enhanced chemiluminescence (ECL, Amersham Biosciences).
Myocardial infarction was performed as previously described (15). A left thoracotomy was performed in the fourth intercostals space at the left sternal border and MI was produced by ligation the left anterior descending (LAD) coronary artery with an 8-0 prolene suture at the site of the vessels' emergence past the tip of the left atrium. The chest cavity was closed, the mouse was gradually weaned off the respirator, and the endotracheal tube was removed. The animal was placed in a cage on a heating pad until fully conscious.
Transthoracic echocardiography was performed with a linear 30-MHz transducer (Vevo 770 High Resolution Imaging System, VisualSonics, Toronto, ON, Canada). LV end-diastolic and end-systolic dimensions (LVEDD and LVESD, respectively), fractional shortening (FS), interventricular septal wall thickness, posterior wall thickness, and heart rate were acquired in conscious mice placed on a heated pad as previously described (20, 35, 40).
CaMKII activity assay.
The enzyme activity was assayed in a total assay volume of 50 μl with use of a CaMKII peptide (281-291) as the enzyme substrate. The assay contained 50 mM Tris·HCl (pH 7.4), 382.7 μM CaMKII substrate (281-291, Calbiochem), 0.3 mM ATP ([γ-32P]ATP = 1 μCi per assay), 10 mM magnesium acetate, 0.2 mM EDTA, ±5 μM calmodulin, ±250 μM CaC12 and an indicated amount of tissue fraction with varying protein concentrations from 0 to 70 μg. The reaction was initiated by adding enzyme, incubated at 30°C for 10 min, and terminated by adding 20 μl of 300 mM. EDTA and 32P incorporation in the substrate were determined as previously described (24).
Histological analysis and determination of infarct size.
After termination of the experiment, hearts were excised and the LV was transversely dissected into two halves. The portion from the base to the midcavity was flash frozen and stored for subsequent biochemical studies. The portion of LV from midcavity to apex was fixed in 10% buffered formalin, embedded in paraffin, sectioned, and stained with Masson's trichrome. Infarct size was determined by studying serial sections and calculating the mean percentage of epicardial and endocardial circumference that was scar as previously described (15).
Myocardial membranes were prepared by homogenization of the base of the LV as described (35, 40). Total βAR density was determined by incubation of 20 μg of membranes with a saturating concentration (200 pmol/l) of 125I-labeled cyanopindolol and 20 mmol/l alprenolol to define nonspecific binding. Assays were conducted at 37°C for 60 min and then filtered over GFIC glass fiber filters (Whatman), then washed and counted in a gamma counter as previously described (35, 40).
In situ TUNEL staining.
Heart specimens were fixed with 10% neutral buffered formalin, embedded in paraffin, and sectioned at 5-μm thickness. DNA fragmentation was detected in situ in deparaffinized sections using the ApopTag Kit (Intergene) according to manufacturer's protocol as previously described (35). The total number of nuclei was determined by manual counting of DAPI-stained nuclei in six random fields per section (magnification, ×200). All terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL)-positive nuclei were counted in each section.
Values are shown as means ± SE. The significance of differences between baseline and Iso stimulation within same genetic species and between Pre (before MI) and 8-wk post-MI echocardiographic values within group was evaluated by Student's t-test. To evaluate differences hemodynamic parameters among groups from P-V loops, multigroup comparisons were made by use of a one-way ANOVA and Tukey's test. Also, ANOVAs with Bonferroni correction for multiple comparisons were used to evaluate the morphometric and serial echocardiographic measurements and CaMKII activity. P < 0.05 was considered significant.
Catecholamine stimulation of cardiac inotropy and chronotropy is mediated by β1ARs.
To test which βAR subtype is responsible for the inotropic and chronotropic response of the heart, we administered Iso to the β1ARKO, β2ARKO, DKO, and WT mice and measured contractile function by P-V loop analysis. Basal ejection fraction as measured by PV loops and peak rate of pressure rise (dP/dtmax) was similar among the different genotypes (Table 1, Fig. 1). In contrast, in response to Iso only the WT and β2AR KO mice showed an increase in the slope of the end-systolic P-V relation, indicating increased contractility, whereas the β1AR KO and DKO showed no significant response to catecholamine infusion (Table 1, Fig. 1). Thus the presence of β1ARs are necessary for an Iso induced increase in heart rate, dP/dtmax, maximal power, and end-systolic elastance by 40–65%, with no response observed to Iso infusion in the β1ARKO and DKO mice (Fig. 2). No changes in diastolic parameters are found for the different groups.
Stimulation of β1AR activates CaMKII signaling pathway.
To determine the βAR subtype in the heart that is responsible for the activation of CaMKII, we measured the extent of CaMKII phosphorylation on the autophosphorylation site Thr-286 in hearts from the different βARKO mice. After 5 min of Iso infusion, CaMKII activity significantly increased in WT and β2AR KO as shown by the increased phosphorylation of the cytoplasmic CaMKIIδC isoform (Fig. 3A). In contrast, stimulation with Iso in β1ARKO and DKO mice did not affect CaMKII phosphorylation in the heart (Fig. 3A). To determine whether the differential activation of CaMKII by the βAR subtypes is translated to downstream signaling pathways, we measured the level of phosphorylation of the sarcoplasmic reticulum protein PLB, a known effector of PKA and CaMKII signaling (52).Since PKA and CaMKII can independently phosphorylate PLB at Ser-16 and Thr-17, respectively, we examined the site-specific phosphorylation of PLB in response of βAR stimulation. Following Iso infusion, phosphorylation at both Ser-16 (PKA site) and Thr-17 (CaMKII site) were significantly increased only in WT and β2ARKO mice and were unchanged from basal levels in the β1ARKO and DKO mice (Fig. 3B). These data show that in the in vivo heart CaMKII activation and the subsequent enhancement of downstream signaling are mediated solely by stimulation of β1AR.
Mice lacking β1ARs have preserved cardiac function after MI.
To determine whether the two βAR subtypes play a differential role in the development of postinfarct cardiac dysfunction, we subjected 12-wk-old β1ARKO, β2ARKO, DKO, and WT mice to MI by ligation of the LAD coronary artery. Serial echocardiographic measurements were obtained in the WT and the various βARKO mice prior to LAD ligation and at 4 and 8 wk post-MI. Histological analysis of myocardial scar via Masson's trichrome staining revealed similar infarct size for all four groups (WT: 33 ± 3%, β1ARKO: 29 ± 3%, β2ARKO: 29 ± 2%, and DKO: 33 ± 3%, P = not significant, Fig. 4A). Despite the similar infarct size between the four groups, β1AR, β2AR, DKO, and WT mice showed differing severity in the subsequent development of cardiac dysfunction as demonstrated by M-mode echocardiography (Fig. 4B, Table 2). At baseline, β1ARKO and DKO mice showed reduced cardiac function compared with either WT or β2ARKO mice, reflecting the contribution of β1ARs to the maintenance of normal cardiac function (FS of 47 and 45% for β1ARKO and DKO, respectively, vs. 64% for both β2ARKO and WT mice). Eight weeks post-MI, WT and β2ARKO mice showed progressive LV enlargement as measured by the percent change in LVEDD and LVESD associated with significant fall in %FS (Fig. 4, C–E). In contrast, both β1ARKO and DKO mice showed an attenuated decline in cardiac function after MI as demonstrated by smaller changes in % FS and LV dimensions compared with preinfarction (Fig. 4, C–E).
Downregulation of βARs only in the WT and β2ARKO mice following MI.
Since downregulation of βARs is a hallmark of failing myocardium, we measured βAR density in sham and MI hearts from the four groups of mice. Sham WT and sham β2ARKO hearts had similar βAR densities (40 and 35 fmol/mg protein, respectively) consistent with the known data that the majority of βARs in the myocardium are of the β1AR subtype (Fig. 5A). Following MI, βAR downregulation was observed only in the WT and β2ARKO mice, demonstrating that β1AR is the primary receptor subtype that undergoes downregulation in the failing heart (Fig. 5A).
Presence of β1AR contributes to elevated CaMKII activity following MI.
CaMKII is a downstream target of β1AR signaling and recent studies have shown that CaMKII inhibition can substantially reduce myocyte apoptosis and maladaptive remodeling from excessive βAR activation (52, 55). To examine whether in the failing heart the activation of CaMKII was dependent on a specific βAR subtype, we evaluated CaMKII activity in each subgroup after either sham surgery or MI. A significant increase in Ca2+-dependent CaMKII activity was seen in WT and β2ARKO hearts 8 wk post-MI (Fig. 5B). In contrast, no increase in CaMKII activity was found in the β1ARKO and DKO post-MI hearts (Fig. 5B). As expected, there were no differences in Ca2+-dependent CaMKII activity in the sham-operated controls (Fig. 5C). As a positive control, Iso infusion in normal WT mice significantly increased CaMKII activity (Fig. 5, B and C).
Increased apoptosis in WT and β2ARKO mice within the infarct border zone following MI.
To test whether divergent signals by the individual βAR subtypes would contribute to differences in the development of apoptosis, TUNEL staining was performed within the border zone and remote regions from the infarcted tissue. The number of TUNEL-positive nuclei was significantly higher within the border zone of the infarct in the β2ARKO and WT hearts compared with that in the β1ARKO and DKO hearts (P < 0.004) (Fig. 6A). TUNEL staining in the remote zones distant from the infarcted tissue showed no differences in the level of apoptosis between the groups (Fig. 6B).
In our present study using specific βARKO mice, we demonstrate that the β1AR specifically contributes to the increase in LV function and CaMKII activation following acute adrenergic stimulation. We show that β1ARs contribute to normal contractility at baseline, and it is the β1AR that transduces the catecholamine signal to augment heart rate and contractility. In addition, it is the β1AR that uniquely transmits the intracellular cascade that leads to the activation of CaMKII and CaMKII-dependent downstream signaling. Moreover, it appears that β1ARs exert a predominantly adverse influence on postinfarction remodeling as shown by the worsening of cardiac function after myocardial injury, activation of myocardial CaMKII activity and increased apoptosis in the infarct border zone. In contrast, mice lacking β1ARs (i.e., β1AR KO and DKO mice) show a significantly attenuated decline in cardiac function post-MI that is not associated with enhanced CaMKII activity or increased in border zone apoptosis.
Several studies have reported cardiac responses after catecholamine stimulation as measured by changes in heart rate, blood pressure, and LV dP/dt in βAR KO mice (13, 26, 42–44). Although LV dP/dtmax provides useful information on relative changes in contractile function, it has limited sensitivity when groups of different animals are compared and therefore analysis of P-V relations is considered the most rigorous approach in the evaluation of cardiac contractility (21, 25, 36). Using this more sophisticated hemodynamic evaluation we clearly show the importance of β1ARs in augmenting contractility in response to catecholamine infusion.
Our data suggest that one of the potential mechanisms contributing to the β1AR-mediated deterioration in post-MI cardiac function could be due to enhanced activation of CaMKII. Recent studies suggest that activation of CaMKII may promote adverse cardiac remodeling, since overexpression of CaMKII can induce cardiomyopathy and inhibition of CaMKII by transgenic overexpression of an inhibitory peptide blocks cardiac deterioration post-MI (52, 53). At least two mechanisms have been recently identified that could potentially contribute to this pathological role of CaMKII following myocardial injury: 1) phosphorylation of the class IIa histone deacetylase (HDAC) 4 (6), and 2) phosphorylation of the cardiac ryanodine receptor (RYR2) (2). In the heart, is has recently been shown that CaMKII signals specifically to HDAC4 to promote its phosphorylation, nuclear export, and derepression of HDAC target genes that are involved in cardiac hypertrophic response (53). The action of CaMKII to phosphorylate the RYR2 and enhance SR diastolic Ca2+ leak (2, 22) and to phosphorylate Na+ channels (46) likely contributes to the contractile dysfunction and arrhythmogenesis in the failing heart (4, 32, 33). Indeed, KO of CaMKIIδ in mice results in attenuated LV chamber enlargement and cardiac dysfunction after pressure overload (31). These morphological remodeling changes were associated with decreased Ca2+ spark frequency and RYR2 Ca2+ leak in the CaMKII KO mice (22).
Previous studies have shown relationship between the level of CaMKII activity and apoptosis (47), which is consistent with our present data showing increased apoptosis in the hearts with enhanced CaMKII activity. Furthermore, β1AR-mediated development of apoptosis can be prevented with transgenic overexpression of a CaMKII inhibitor (50). When these mice are crossed with PLB-deficient mice, which have increased susceptibility to catecholamine-induced apoptosis, inhibition of CaMKII abolished the resistance to apoptosis (50). Apoptosis has been implicated in cardiac ischemic injury and is likely involved in the transition from cardiac hypertrophy to decompensated heart failure, particularly after MI-induced LV dysfunction (45). Moreover, apoptosis of cardiac myocytes appear to peak during the first several days following MI in the ischemic area with little apoptosis found in remote regions of the myocardium from the injury (8, 28). These findings are consistent with our study showing increased apoptosis in the ischemic border zones and not in the remote areas of the myocardium. Importantly, we show that mice lacking β1AR reveals significantly less apoptosis, suggesting that β1AR activation promotes robust apoptosis. Similar data have been reported in cultured adult myocytes (14, 56). Although we demonstrate significant less apoptosis in the border zone of the β1AR KO and DKO mice, our study was not designed to test whether this would translate into reduced infarct size since infract size was matched across the four groups. Indeed, a number of studies have shown beneficial effects of reducing cardiomyocyte apoptosis on ventricular function in a variety of heart failure models (17).
Prior studies have suggested that stimulation of β2ARs may provide a protective mechanism in the heart by activating PI3K/Akt signaling (56), in part through their ability to switch coupling from Gs to Gi (16). It has been proposed that βAR-Gi cardioprotective signaling protects against myocyte apoptosis, while simultaneously attenuating detrimental β2AR -Gs inotropism (54). However, since the level of apoptosis and cardiac dysfunction was similar in both β1ARKO and DKO mice, our data support an alternate concept for the in vivo heart during pathological stress. That is, in response to myocardial injury β2AR provide little in the way of cardioprotection, whereas β1AR markedly stimulate detrimental signaling pathways.
Recently, Palazzesi et al. (37) assessed the role of βARs in the hypertrophic response to pressure overload. Consistent with our data, basal contractility in hearts of double βAR KO mice was similar to WT mice (37). Although response to catecholamine infusion was not performed, they showed that the hypertrophic response was not attenuated in banded double βAR KO mice, consistent with the concept that Gq-coupled receptors are the primary G protein-coupled receptors involved in the development of cardiac hypertrophy following pressure overload (3).
β1AR account for majority of the βARs on the cardiac myocytes and are significantly downregulated in chronic human heart failure (41). Only hearts from WT and β2AR KO mice showed marked reduction in plasma membrane receptor levels after MI consistent with β1AR downregulation. In contrast, we did not observe any change in the receptor levels in the β1ARKO mice following MI, suggesting little downregulation of β2AR as seen in humans. Since we did not measure the level of β3ARs in the βAR KO mice, we cannot exclude a possible contribution of β3ARs to the attenuated adverse remodeling seen in the DKO mice. However, in transgenic mice with cardiac overexpression of human β3ARs, selective β3AR stimulation resulted in an increase in contractility that was entirely Gs dependent (27), suggesting that signaling via the β3AR is unlikely to account for the observed phenotype. Finally, our data are consistent with a number of studies using β1 selective and nonselective β-antagonists and KO mice with regard to their effect on LV remodeling following pressure overload (26, 29, 38). However, in this study we demonstrate the importance of CaMKII activation as a potential mechanism for the adverse postinfarct remodeling and further show the relative contribution of both βAR subtypes in this response.
In summary, we demonstrate that cardiac contractility and CaMKII is mediated almost exclusively by β1AR under basal and catecholamine-stimulated conditions. Moreover, we show that β1AR signaling promotes detrimental cardiac remodeling following MI. Although multiple mechanisms are likely to be involved in the development of heart failure, our study supports the concept that β1ARs activate CaMKII leading to enhanced apoptosis, both of which can potentially contribute to cardiac dysfunction. Contrary to the previous reports of the protective effects of β2AR signaling, our findings from the β1ARKO and DKO suggest that β2AR do not actively contribute to the antiapoptotic components during the development of heart failure following MI. Taken together, these findings show that the β1AR has a predominant influence on cardiac function in response to adrenergic stimulation and the pathological remodeling that follows MI.
This work was supported by National Heart, Lung, and Blood Institute Grant HL-61558 (H. A. Rockman).
- Copyright © 2009 American Physiological Society