Abstract
β3-adrenergic receptors (AR) have recently been identified in mammalian hearts and shown to be up-regulated in heart failure (HF). β3-AR stimulation reduces inotropic response associated with an inhibition of L-type Ca2+ channels in normal hearts; however, the effects of β3-AR activation on Ca2+ channel in HF remain unknown. We compared the effects of β3-AR activation on L-type Ca2+ current (ICa,L) in isolated left ventricular myocytes obtained from normal and age-matched rats with isoproterenol (ISO)-induced HF (4 months after 340 mg/kg s.c. for 2 days). ICa,L was measured using whole-cell voltage clamp and perforated-patch recording techniques. In normal myocytes, superfusion of 4-[-[2-hydroxy-(3-chlorophenyl)ethylamino]propyl]phenoxyacetate (BRL-37,344; BRL), a β3-AR agonist, caused a dose-dependent decrease in ICa,L with maximal inhibition (21%, 1.1 ± 0.2 versus 1.4 ± 0.1 nA) (p < 0.01) at 10–7 M. In HF myocytes, the same concentration of BRL produced a proportionately greater inhibition (31%) in ICa,L (1.1 ± 0.2 versus 1.6 ± 0.2 nA) (p < 0.05). A similar inhibition of ICa,L was also observed with ISO (10–7 M) in the presence of a β1- and β2-AR antagonist, nadolol (10–5 M). Inhibition was abolished by the β3-AR antagonist (S)-N-[4-[2-[[3-[3-(acetamidomethyl)phenoxy]-2-hydroxypropyl]amino]ethyl]phenyl]benzenesulfonamide (L-748,337; 10–6 M), but not by nadolol. The inhibitory effect of BRL was attenuated by a nitric-oxide synthase (NOS) inhibitor, NG-nitro-l-arginine methyl ester (10–4 M), and was prevented by the incubation of myocytes with pertussis toxin (PTX; 2 μg/ml, 36°C, 6 h). In conclusion, β3-AR activation inhibits L-type Ca2+ channel in both normal and HF myocytes. In HF, β3-AR stimulation-induced inhibition of Ca2+ channel is enhanced. These effects are likely coupled with PTX-sensitive G-protein and partially mediated through a NOS-dependent pathway.
β1-, β2-, and β3-adrenergic receptors have been found to be present in mammalian hearts and shown to modulate cardiac contractility by a variety of mechanisms. β1- and β2-AR stimulation of L-type Ca2+ channel are mediated by a cAMP/protein kinase A-signaling mechanism and coupled with Gs proteins (Skeberdis et al., 1997; Xiao et al., 1999; Zhang et al., 2001). β2-AR also couples with Gi protein (Xiao et al., 1999). Recently, β3-AR was identified in mammalian hearts, including human, dog, rat, and guinea pig (Gauthier et al., 1996; Kitamura et al., 2000; Cheng et al., 2001; Dincer et al., 2001). β3-AR stimulation inhibits cardiac contractility via a Gi protein pathway and by a mechanism coupled with the nitric-oxide synthase (NOS) system (Gauthier et al., 1996, 1998; Seppet, 2003). The negative inotropic effect of β3-AR stimulation is associated with alterations of action potentials (Gauthier et al., 1996) and decreased Ca2+ transient (Kitamura et al., 2000). BRL-37,344, a selective β3 agonist, inhibits L-type Ca2+ channels and attenuates intracellular Ca2+ transients in canine ventricular myocytes with an associated dose-dependent decrease in contractility (Cheng et al., 2001). A similar effect on basal ICa,L is partly abolished by NG-nitro-l-arginine methyl ester (l-NAME) (Au and Kwan, 2002).
Recently, several studies have reported that β3-ARs are up-regulated in the failing human heart (Moniotte et al., 2001), in the canine models of HF (Cheng et al., 2001), as well as in diabetic rat hearts (Dincer et al., 2001). HF is associated with selective down-regulation of β1-AR and a marked increase in Gi protein (Cheng et al., 2001; Moniotte et al., 2001). The exaggerated β3-AR/Gi signaling may cause alteration in the regulation of Ca2+ channel, thus contributing to contractile dysfunction. However, the role and mechanism of β3-AR activation on ICa,L in HF have not been defined.
Accordingly, the purpose of this study was to compare the effects of β3-AR stimulation on cardiac ICa,L in LV myocytes of normal rats and rats infused with ISO as a model of HF and to determine the underlying cellular mechanism. Our results indicate that β3-AR activation inhibits L-type Ca2+ channel in both normal and HF myocytes. In HF, β3-AR stimulation-induced inhibition of ICa,L was enhanced. These effects are likely coupled with a pertussis toxin (PTX)-sensitive Gi protein and partially mediated through a NOS-dependent pathway. The current findings extend our knowledge regarding the impaired β-adrenergic regulation of L-type Ca2+ channel in HF and provide valuable new insight into the cellular mechanism of the progression of functional impairment in HF.
Materials and Methods
Experimental Heart Failure Model. This investigation conforms to the Guide for the Care and Use of Laboratory Animals, published by the U.S. National Institutes of Health (NIH Publication No 85-23, revised 1985).
As previously described (Teerlink et al., 1994; Suzuki et al., 1998; Kong et al., 2004), HF in the rat model was induced by ISO injections with some modification (4 months after 340 mg/kg s.c., for 2 days). Briefly, male Sprague-Dawley rats (200–250 g) received two subcutaneous injections of 340 mg/kg ISO-HCl at 24-h intervals. The mortality rate was approximately 40 to 48% within 48 h. The control group of rats received the same amount of sterile saline. Animals were housed and fed under identical conditions.
Four months after the injection protocol, survivors (n = 16) and sham-injected rats (n = 20) were lightly anesthetized with intraperitoneal Ketamine HCl (50 mg/kg) and Xylazine (10 mg/kg). Then, hemodynamic measurements were obtained using a micro-tip pressure transducer (Millar Instruments, Houston, TX), inserted into the LV through the carotid artery to verify the presence of HF in the ISO-treated rats. Heart and lung weight was obtained after the study. Consistent with previous reports (Rona et al., 1959; Teerlink et al., 1994), the hearts of ISO-injected rats displayed large infarct-like necrosis involving more than one-third to half of the LV extending to the adjacent area of the interventricular septum and right ventricle. Diffuse subendocardial necrosis was also observed. To obtain a high yield of viable isolated myocytes, as previously described, with some modification (Rona et al., 1959; Pfeffer et al., 1979), the gross lesions of infarct-like necrosis area were measured. Briefly, after Langendorff perfusion with an enzymatic buffer, the heart was weighed, and LV was separated. Then the infarct-like necrosis area was carefully excised from the LV and weighed separately. Calculation of the ratio of the weights between the necrosis areas to LV was used as an approximate estimation of infarct size.
Isolation of LV Myocytes. Myocytes were enzymatically dissociated by Langendorff perfusion as previously described (Suzuki et al., 1998). With our well established technique, more than an 80% yield of viable myocytes was obtained from both control and ISO-treated rats. The cells were used within 10 h.
Electrophysiological Measurement. Membrane calcium current was recorded at 22 to 23°C with the whole-cell patch-clamp technique as previously described (Hamill et al., 1981; Zhang et al., 2001). An Axopatch 200A amplifier (Axon Instruments, Foster City, CA) was interfaced with a 12-bit A/D-D/A converter (Digidata 1200; Axon Instruments). PClamp software (PClamp 6.02; Axon Instruments) was used for data acquisition and analysis. Data were filtered by a 5-kHz low-pass filter and digitized at 5 kHz.
After stabilization, a drop of cell pellet containing the isolated myocytes was placed in a perfusion chamber (0.5-ml volume) mounted on the stage of an inverted microscope (IMT 2-F3; Olympus, Herndon, VA) and continuously superfused at a constant rate of 2 ml/min. Only quiescent rod-shaped cells with clear cross-striations were studied. Borosilicate glass micropipette (O.D. 1.6 mm) was pulled with a two-step puller (Model PP-83; Narishige, Tokyo, Japan) and heat-polished with a microfuge (MT-83; Narishige). The tip resistances were 1.5 to 2.5 MΩ when filled with pipette (internal) solution.
Liquid junction potentials (<5 mV) were corrected before the pipette touched the cell. After formation of GΩ-seal, the electrode capacitance was compensated electronically. Then the cell membrane was ruptured by gentle suction to establish whole-cell configuration.
In a subgroup, a perforated-patch recording technique was used (Horn and Marty, 1988) in which nystatin stock was added to the internal solution (final nystatin concentration of 100–200 μg/ml). The development of electrical access was monitored by the appearance of a capacitive current. The access resistance was <20 MΩ.
The membrane capacitance and series resistance were compensated to minimize the duration of the capacitive transient. The membrane capacitance was measured before compensation with a 10-mV depolarizing step from a holding potential of –80 mV to –70 mV and integrated the area under the current transient calculated with the following formula: Cm = τcI0/ΔVm[1 – (I∞/I0)], where Cm is the membrane capacitance, τc is the decay time constant of membrane capacitance, I0 is the maximum capacitance current value, ΔVm is the amplitude of the voltage step, and I∞ is the amplitude of steady-state current. The membrane capacitance was used as an index to normalize ICa,L for cell size. The series resistance was calculated as Rs = ΔVm/I0, where Rs is the series resistance, and compensated at about 80%.
Myocytes were voltage clamped at –80 mV. Ca2+ currents were elicited by stepping up the membrane voltage from a holding potential to 0 mV testing potential for 200 ms at 12-s intervals. To avoid contamination by fast sodium channel activation and to reduce the run-down of ICa,L, a brief prepulse (60-ms duration) was applied to –40 mV before approaching the test potential. The average peak ICa,L run-down was about 10 to 20% for 30 min after initial measurement. Most (80%) of the run-down occurred within the initial 8 to 10 min. Thus, the window of time between 10 and 30 min after the initial recording was chosen to measure ICa,L with respect to drug effects (Xiao and Lakatta, 1993). ICa,L was measured by the standard method as the difference between peak inward current and the current at the end of a 200-ms pulse. For current-voltage relations, test potentials were from –35 to +60 mV at 5-mV increments and 0.1 Hz.
Solutions. The compositions of the pipette solution and recording bath solution were chosen to allow isolation of ion flow through the Ca2+ channel by blocking other ionic currents. Initially, the myocytes were superfused with a modified Tyrode's solution containing 137 mM NaCl, 5.4 mM KCl, 1.2 mM MgSO4, 15 mM glucose, 10 mM HEPES, and 1.5 mM CaCl2. The pH was adjusted to 7.4 with NaOH at 20–22°C. After formation of giga-Ω seal, superfusion buffer was changed to a patch-recording bath solution, i.e., Na+-K+-free Tyrode's solution, in which NaCl was substituted by tetraethylammonium chloride, and 50 μM tetrodotoxin was added to eliminate sodium current and KCl replaced by CsCl and 3 mM 4-aminopyridine to abort the potassium current. The solution was gassed with 100% O2. The internal solution for the pipette contained 140 mM cesium aspartate, 1.0 mM MgCl2, 3 mM Na2ATP, 0.4 mM GTP, 10 mM EGTA, and 5 mM HEPES. The pH was adjusted to 7.2 (with titrated CsOH).
For perforated patch recording, nystatin stock solution (10 mg/ml in acidified methanol) was prepared on each day and added to the internal solution at a final concentration of 100 to 200 μg/ml (Horn and Marty, 1988). The pipette was dipped in nystatin-free internal solution for 2 s and then back-filled with nystatin internal solution.
Drugs. BRL 37,344 was obtained from Tocris (Ballwin, MO). L-748,337 was a gift from Merck Research Laboratories (Rahway, NJ). Nadolol, isoproterenol, ICI-118,551 (ICI), l-NAME, nifedipine, and PTX were obtained from Sigma-Aldrich (St. Louis, MO).
Statistical Analysis. Data are presented as mean ± S.E.M. Statistical comparisons were performed with Student's t test or analysis of variance. A p value of <0.05 was considered significant. Prism 3.0 (GraphPad Software) was used for the concentration-ICa,L relationship nonlinear regression analysis. As previously described by Robberecht et al. (1983) and Lands et al. (1967), the Hill equation may allow us to model cooperativity between multiple receptor sites on each cardiomyocyte with respect to β3-AR agonist (BRL) binding. Thus, data were fitted with the Hill equation. The best fit by the Hill equation was also compared with a fit by the one-site competition equation.
Results
Verification of Experimental HF. The general hemodynamic and ICa,L features in the ISO-treated rats are presented in Table 1. LV end-diastolic pressure increased 5-fold, and LV dP/dtmax and LV dP/dtmin were significantly decreased. The rate of LV relaxation slowed as indicated by a significant increase in the time constant of isovolumic LV pressure decay (τ, 185%) (p < 0.05).
All ISO-treated animals had clear evidence of HF (anorexia, edema, and pulmonary congestion). In the ISO-injected rats, the total infarction area was about 43 ± 3%. There was no significant change in body weight (596 ± 8 versus 601 ± 9 g) (p = N.S.), whereas the heart weight (2.25 ± 0.03 versus 1.68 ± 0.03 g) (p < 0.05), calculated ratio of LV to body weight (2.53 ± 0.04 versus 1.86 ± 0.04 g/kg) (p < 0.05), and calculated ratio of wet lung to body weight (4.85 ± 0.09 versus 2.89 ± 0.08 g/kg) (p < 0.05) were all significantly increased in ISO-injected rats. In the HF rat myocytes, the membrane capacitance was significantly increased, whereas the current density was significantly lower than that of the normal myocytes (62%) (p < 0.01), indicating an absolute reduction of ICa,L.
In addition, the response of ICa,L to β-AR stimulation in HF myocytes was significantly attenuated. As shown in Fig. 1, in the normal myocytes, in response to the exposure to ISO (10–7 M), ICa,L was doubled (119 ± 9%) (p < 0.01, n = 11). However, in HF myocytes, ISO only caused half the increase in ICa,L (55 ± 4%) (p < 0.05, n = 11). The current was blocked by nifedipine (5 × 10–6 M), a Ca2+ channel blocker, consistent with the characteristics of ICa,L. These findings demonstrated the existence of established HF in this model.
Increased Inhibition of ICa,L to β3-AR Stimulation in HF Myocytes. The effects of direct β3-AR stimulation with BRL, the most potent β3-AR agonist in rats (Gauthier et al., 1999), on ICa,L in normal and HF myocytes are summarized in Table 2 and displayed in Fig. 2. Superfusion of BRL (10–7 M) caused significant decreases in peak ICa,L in normal myocytes (21%, 1.1 ± 0.1 versus 1.4 ± 0.1 nA, p < 0.01, n = 7). In HF myocytes, BRL caused a greater relative and absolute decrease in ICa,L (31%, 1.1 ± 0.2 versus 1.6 ± 0.2 nA, p < 0.05, n = 7). The absolute decrease in ICa,L was greater in HF myocytes (0.3 ± 0.03 nA) versus (0.5 ± 0.06 nA) (p < 0.05). In the HF myocytes, the membrane capacitance was significantly increased. After normalization of the membrane capacitance, the BRL-induced decreases in ICa,L remained statistically different in both normal myocytes (21%, 5.9 ± 0.4 versus 7.5 ± 0.5 pA/pF) (p < 0.01, n = 7) and HF myocytes (31%, 2.9 ± 0.4 versus 4.1 ± 0.3 pA/pF) (p < 0.05, n = 7). Figure 2, C and D demonstrates current-voltage relations for the response of ICa,L to BRL in normal and HF myocytes. BRL caused no change in the voltage dependence of peak ICa,L amplitude in normal and HF myocytes.
The inhibitory effects of BRL persisted after washout. To eliminate the influence of “run-down”, in a subgroup, we used nystatin perforated patch recording to examine the effect of BRL. A similar result was observed (data not shown). This is consistent with the findings of Au and Kwan (2002).
In a subgroup, we further examined the ICa,L response to ISO (10–7 M) in the presence of a β1- and β2-AR antagonist, nadolol (Nad, 10–5 M). In contrast to ISO alone, we observed an inhibition of ICa,L in both normal and HF myocytes. Compared with normal myocytes, this inhibition was enhanced in HF myocytes (29 ± 2 versus 20 ± 2%) (p < 0.05, n = 5).
Concentration-Dependent Inhibition of ICa,L by BRL. Concentration-response curves of ICa,L to BRL in normal and HF rat myocytes are compared in Fig. 3. The maximal response of ICa,L to BRL was significantly enhanced in HF myocytes. The dose-response curve was shifted downward. The half-maximal inhibition concentration (IC50) was 1.1 nM for HF myocytes and 1.2 nM for normal myocytes.
The Effects of β1- and β2-AR Antagonist, β2-AR Antagonist, and β3-AR Antagonist on BRL-Induced Inhibition in ICa,L. To determine the potential mechanism, the myocytes were preincubated with a β1- and β2-AR antagonist, Nad (10–5 M), or a β3-AR antagonist, L-748,337 (10–6 M), for 20 min, and BRL was given in the presence of Nad or L-748,337. To further exclude BRL action through β2-AR, in a subgroup of three normal rats, the myocytes were preincubated with ICI-118,551, a β2-AR antagonist (10–7 M), for 20 min, and BRL was given in the presence of ICI.
As shown in Fig. 4, after using Nad to block β1- and β2-AR, BRL-induced decreases in ICa,L still persisted [(5.5 ± 0.5 versus 7.0 ± 0.6 pA/pF) (p < 0.05, n = 3) and (3.3 ± 0.4 versus 4.5 ± 0.4 pA/pF) (p < 0.05, n = 3)] in normal and HF myocytes, respectively. In a subgroup of normal rats, after preincubation with the β2-AR antagonist ICI, the BRL-induced reductions in ICa,L remained unaffected (6.5 ± 0.5 versus 8.1 ± 0.6 pA/pF) (p < 0.05, n = 4). In contrast, as shown in Fig. 5, after preincubation with a β3-AR blocker, l-748,337, BRL-induced inhibition in ICa,L was abolished [(7.2 ± 0.5 versus 7.4 ± 0.4 pA/pF) (n = 4) and (4.6 ± 0.6 versus 4.7 ± 0.6 pA/pF) (n = 4, p = N.S.)] in normal and HF cells, respectively, indicating that BRL inhibited ICa,L through β3-AR, not β1- and β2-AR.
The Effect of Gi Protein Blockade on BRL-Induced Inhibition in ICa,L. To define the role of Gi protein in β3-AR inhibition of ICa,L, myocytes were pretreated with PTX (2 μg/ml, 36°C, 6 h). The adequacy of the complete blockage of inhibitory Gi protein in PTX-treated cells was routinely verified by the loss of the ability of acetylcholine (10–5 M) to reverse the stimulatory effect of ISO on ICa,L consistent with our past report (Zhang et al., 2001) that an adequate blockade of Gi with this concentration and incubation time was achieved in the current study. PTX-treated myocytes were compared with myocytes that had been kept at 36°C in the absence of PTX for an equal amount of time. As shown in Fig. 6, after PTX preincubation of the myocytes to block inhibitory Gi protein, there was no significant change in baseline ICa,L; whereas, BRL-induced inhibition in ICa,L was prevented both in normal cells (8.1 ± 0. 7 versus 8.2 ± 0.5 pA/pF) (p = N.S., n = 4) and in HF cells (5.3 ± 1.0 versus 5.3 ± 0.9 pA/pF) (p = N.S., n = 4).
The Effect of NOS Pathway Blockade on BRL-Induced Inhibition in ICa,L. We further examined the role of NO signaling in β3-AR inhibition of ICa,L by preincubating the myocytes with an NOS blocker, l-NAME (10–4 M), for 20 min, and BRL was given in the presence of l-NAME. Pretreatment of myocytes with l-NAME caused no significant changes in baseline ICa,L but significantly altered myocyte ICa,L response to BRL. In untreated myocytes, BRL caused about 21 and 31% decreases of ICa,L in both normal and HF myocytes, respectively (Table 2 and Figs. 2 and 3). However, comparing BRL-caused changes of peak ICa,L in myocytes without l-NAME treatment, in the presence of l-NAME, BRL-induced ICa,L inhibition was significantly attenuated and produced only about a 12% and an 11% decrease in ICa,L in both normal (6.8 ± 0.3 versus 5.9 ± 0.4 pA/pF) (p < 0.05, n = 5 and 7) and HF myocytes (4.1 ± 0.1 versus 2.9 ± 0.4 pA/pF) (p < 0.05, n = 4 and 7), respectively (Tables 2, 3 and Fig. 7).
Discussion
The present study demonstrates that β3-AR stimulation with BRL causes an enhanced inhibition of ICa,L in LV myocytes of rats with HF. These effects were coupled with PTX-sensitive inhibitory Gi protein and partly mediated through a NOS-dependent mechanism.
Effects and Possible Mechanism of BRL on ICa,L. β-AR activation modulating cardiac ICa,L plays an important role in the positive inotropic response to β-AR stimulation. β1- and β2-AR stimulation of L-type Ca2+ channel are mediated by a cAMP/protein kinase A-signaling mechanism and coupled with Gs proteins (Skeberdis et al., 1997; Xiao et al., 1999; Zhang et al., 2001). β2-ARs also have the ability to activate nonclassical signaling pathways and link to inhibitory G proteins (Gi), suggesting a function distinct from the β1-AR subtype (Zhang et al., 2001; Xiao et al., 2003). Recent observations indicate a more complex β-AR-mediated regulation of myocardial inotropism by catecholamines. In addition to β1- and β2-AR, a third β-AR, β3-AR, initially found to be widely expressed in fat tissues (Krief et al., 1993), was also found to exist in mammalian hearts and to modulate cardiac contractile function (Gauthier et al., 1996; Kitamura et al., 2000; Cheng et al., 2001; Dincer et al., 2001). Stimulation of β3-AR negatively modulates cardiac function through the Gi protein pathway and is coupled with the NOS system (Gauthier et al., 1996, 1998; Seppet, 2003). The negative inotropic effect of β3-AR stimulation was associated with alterations of myocardial electrophysiology and Ca2+ signaling (Gauthier et al., 1996; Kitamura et al., 2000; Cheng et al., 2001).
In the present study, using freshly isolated cardiomyocytes, we found that stimulation of β3-AR with BRL resulted in a dose-dependent inhibition in ICa,L with maximum inhibition (21%) at a concentration of 10–7 M in normal cells. This effect was completely abolished by a highly-selective β3-AR antagonist but not by the β1- and β2-AR blockade or β2-AR blockade, indicating that the alteration of ICa,L following BRL superfusion was due to β3-AR stimulation but not mediated by β1-AR or β2-AR activation.
Furthermore, as in previous studies (Gauthier et al., 1998; Kitamura et al., 2000; Varghese et al., 2000), we found that in the presence of l-NAME, the BRL-induced negative response of ICa,L was significantly reduced, indicating involvement of the NO pathway in β3-AR-mediated action. However, our current observation indicates that the NO pathway may not be fully responsible for the altered LV myocyte ICa,L response to BRL since we found that the ICa,L response to BRL was only partially inhibited by pretreatment myocytes with a NOS inhibitor. In the presence of l-NAME, myocyte ICa,L remained significantly reduced, indicating β3-AR activation is not mediated exclusively through the NO pathway.
Our study indicates that β3-AR-induced inhibition of ICa,L is mediated by Gi, since pretreatment of PTX completely prevented BRL-induced ICa,L responses in both normal and HF myocytes. However, this is not consistent with the past report of Gauthier et al. (1996). In their study, it was shown that in human ventricular strips treated with PTX, the effect of BRL on contractility was attenuated but not completely suppressed. This discrepancy may be due to species difference or incomplete blockade of Gi by PTX since a lower level of exposure to PTX (0.5 μg/ml for 2 h) was used in that study. It has been reported that PTX treatment for 3 to 5 h at 5 μg/ml is required to completely inactivate human cardiac Gi (Brown and Harding, 1992). Therefore, our current study indicated that the enhanced ICa,L response to β3-AR stimulation in rat CHF myocytes may be coupled to Gi through both NO-dependent and NO-independent mechanisms.
These results are consistent with previous observations of cardiac functional response to β3-AR stimulation made in normal hearts and cardiomyocytes of several species, including that of the human, dog, and guinea pig (Gauthier et al., 1996; Kitamura et al., 2000; Cheng et al., 2001). Gauthier and colleagues first demonstrated that in the human heart, stimulation of β3-AR with a BRL resulted in dose-dependent, negative inotropic effects, which were associated with decreased action potential amplitude and reduced action potential duration (Gauthier et al., 1996). However, in their study, β3-AR transcripts were not detected in the rat ventricular strips by using a reverse transcription-polymerase chain reaction assay. The failure to detect functional cardiac β3-AR expression may be attributed to the use of nonspecific primers for rat β3-AR mRNA (Gauthier et al., 1999). Recently, by using reverse transcription-polymerase chain reaction, polyacrylamide gel electrophoresis, and Western blot analysis, Dincer et al. (2001) clearly demonstrated the presence of functional β3-AR in rat hearts and showed significantly increased β3-AR mRNA and protein levels in streptozotocin-induced diabetic rat hearts. Barbier and colleagues also detected transcripts and cell surface expression of β3-AR in rat hearts. They further found that LV β3-AR density significantly increased in female rats with treadmill training for 8 weeks (Barbier et al., 2004).
A novel finding in the present study is that the inhibition of ICa,L with β3-AR was enhanced after HF. In HF myocytes, the basal ICa,L was significantly decreased, and its response to β-AR stimulation (ISO 10–7 M) was also markedly blunted (55 versus 119% increment) (p < 0.05). However, the inhibitory response to BRL (10–7 M) was significantly enhanced (31 versus 21%).
This is consistent with our past observation made in a pacing-induced canine HF model (Cheng et al., 2001). In that study, we assessed the direct effects of BRL on cardiomyocytes isolated from the dogs before and after pacing-induced advanced HF. In these studies, we removed the effects of extracardiac factors. We clearly demonstrated that compared with normal myocytes, BRL caused a much greater decrease in ICa,L in HF myocytes. We further found that β3-AR stimulation produces direct inhibition of myocyte contraction, relaxation, and [Ca2+]i transient. In pacing-induced HF, these inhibitions were increased. These responses were coupled to Gi protein. However, our observations do not agree with the findings by Moniotte et al. (2001), who reported a similar up-regulation of β3-AR expression but blunted negative inotropic response to BRL in failing human myocardium. This discrepancy might result from several factors, such as species difference in β3-AR (Gauthier et al., 1999), variation in severity of HF, or the effect of pharmacotherapy on the failing human heart.
There is a documented different specificity for β3-AR across species and even tissues. It is possible that the intracellular coupling of β3-AR may be different in the failing rat and human heart. It is also possible that the severity of HF may contribute to the different finding. Our rat model of HF was associated with a 40 to 48% mortality, which is much higher than would be expected in a population of human HF patients. The model of HF is also different. We examined a single LV myocyte obtained from ISO-induced failing rat hearts, whereas Moniotte et al. (2001) studied ventricular strips obtained from dilated or ischemia-caused failing human hearts. The use of ventricular strips of multiple cell preparation versus single LV myocyte preparation may induce quantitatively different results due to the additional variable of extracellular matrix and remodeling, although qualitative differences are less likely. We feel that the most likely explanation for the discrepancy in our finding is that the human group received multiple cardiovascular specific medications for HF. A standard regimen for HF patients consists of significant neurohormonal blockade with β-AR blockers, angiotensin-converting enzyme inhibitors, and aldosterone antagonists. It would not be surprising if this regimen alters β3-AR expression or function. Chronic treatment with cardiovascular specific medications (such as β-AR blockers and angiotensin-converting enzyme inhibitors) may alter the primary defect in the contractile properties of the myocyte itself as well as the abnormalities of extracardiac factors in HF, thus, further modifying LV and cardiomyocyte functional response to β3-AR stimulation in HF (Spinale et al., 1995, 1998; Gunja-Smith et al., 1996; Bristow, 2000; Cohn et al., 2000; Lohse et al., 2003). It is possible that the intracellular coupling of β3-AR may be also altered by these cardiovascular specific medications in the failing human hearts. Although we studied a rat model of HF (ISO-induced cardiomyopathy) that reproduces many of the functional and neurohormonal features of clinical HF, we cannot be certain that our results apply generally to HF of other causes. Nevertheless, the findings of Moniotte et al. (2001) and our past reports all demonstrated a similar, potentially detrimental, functional consequence with β3-AR activation in HF.
The mechanism(s) for the BRL-induced enhanced decrease in ICa,L for HF cells versus normal cells are unclear. We speculate that an increase in β3-AR density on the membrane of HF cells may contribute to our current findings.
Since β3-ARs are activated at higher catecholamine concentrations than β1- and β2-ARs (Lafontan, 1994), and β3-ARs are relatively resistant to chronic, agonist-induced desensitization processes (Liggett et al., 1993), they could be involved in HF. Although the β3-AR density per myocyte has not been measured in HF previously, it is possible that the enhanced inhibition of ICa,L in LV myocytes of rats with HF may reflect the presence of an increase in the number of β3-ARs per cell. Further studies are currently underway to elucidate this point.
The enhanced response to β3-AR stimulation in HF may also be related to an altered signal transduction. Although, the intracellular pathway coupling β3-AR stimulation is incompletely characterized, it has been reported that β3-AR stimulation decreases cardiac contractility through activation of the NOS pathway (Kitamura et al., 2000; Varghese et al., 2000). In HF, the NO-cGMP signaling may be altered (Mohan et al., 1996), thereby altering HF myocyte response to β3-AR stimulation. Consistent with previous studies, we found that in the presence of l-NAME, BRL-induced inhibition with ICa,L was largely attenuated, indicating an involvement of the NO pathway in β3-AR-mediated action. In addition, we also found that the enhanced ICa,L response to β3-AR stimulation in HF myocytes couples to Gi. Thus, an up-regulation of Gi in HF may also contribute to the enhancement of the inhibitory effect of β3-AR. The activation of Gi also has the potential to couple β3-AR to other important signaling pathways such as the mitogen-activated protein kinase (Soeder et al., 1999). Clearly, up-regulation of cardiac β3-AR-mediated inhibitory pathways is responsible for the enhanced BRL-induced inhibition of ICa,L in HF. However, the exact contribution of up-regulation of cardiac β3-AR versus increased levels of Gi is unclear. Further studies are needed to fully characterize the intracellular pathway coupling β3-AR stimulation.
In conclusion, we found that β3-AR activation inhibits L-type Ca2+ channel and decreases ICa,L in both normal and HF myocytes. In rats with ISO-induced HF, β3-AR activation-induced inhibition of ICa,L was enhanced. These effects are likely to be partially mediated through a NOS-dependent mechanism and coupled with a PTX-sensitive inhibitory G protein. This finding is possibly an underlying mechanism for the impaired inotropic response to β-AR stimulation in the failing heart.
Acknowledgments
We acknowledge the expertise and assistance of Drs. J. Mu and R. L. Wu with regard to the patch-clamp technique, the computer programming of Dr. Ping Tan, the technical assistance of Ellen Tommasi and Michael Cross, and the editorial review and administrative support of Amanda Burnette. We thank Merck Research Laboratories (Rahway, NJ) for providing L-748,337.
Footnotes
-
This study was supported in part by grants from the National Institutes of Health (HL45258 and HL53541) and the American Heart Association (9640189N).
-
doi:10.1124/jpet.105.089672.
-
ABBREVIATIONS: AR, adrenergic receptor; NOS, nitric-oxide synthase; BRL, 4-[-[2-hydroxy-(3-chlorophenyl)ethyl-amino]propyl]phenoxyacetate (BRL-37,344); ICa,L, L-type Ca2+ current; l-NAME, NG-nitro-l-arginine methyl ester; HF, heart failure; LV, left ventricular; ISO, isoproterenol; PTX, pertussis toxin; L-748,337, (S)-N-[4-[2-[[3-[3-(acetamidomethyl)phenoxy]-2-hydroxypropyl]amino]ethyl]phenyl]benzenesulfonamide; ICI, (±)-1-[2,3-(dihydro-7-methyl-1H-inden-4-yl)oxy]-3-[(1-methylethyl)amino]-2-butanol (ICI-118,551); Nad, nadolol; NO, nitric oxide.
-
↵1 Current affiliation: First Department of Internal Medicine, Mie University School of Medicine, Tsu City, Japan.
-
↵2 Current affiliation: Department of Internal Medicine and Pathophysiology, Nagoya City University Graduate School of Medical Sciences, Mizuho-cho, Mizuho-ku, Japan.
- Received May 16, 2005.
- Accepted August 30, 2005.
- The American Society for Pharmacology and Experimental Therapeutics