The purpose of the present study was to examine the role of Gi2α in Ca2+ channel regulation using Gi2α gene knockout mouse ventricular myocytes. The whole cell voltage-clamp technique was used to study the effects of the muscarinic agonist carbachol (CCh) and the β-adrenergic agonist isoproterenol (Iso) on cardiac L-type Ca2+ currents in both 129Sv wild-type (WT) and Gi2α gene knockout (Gi2α−/−) mice. Perfusion with CCh significantly inhibited the Ca2+ current in WT cells, and this effect was reversed by adding atropine to the CCh-containing solution. In contrast, CCh did not affect Ca2+ currents in Gi2α−/− ventricular myocytes. Addition of CCh to Iso-containing solutions attenuated the Iso-stimulated Ca2+ current in WT cardiomyocytes but not in Gi2α−/− cells. These findings demonstrate that, whereas the Iso-Gsα signal pathway is intact in Gi2α gene knockout mouse hearts, these cells lack the inhibitory regulation of Ca2+ channels by CCh. Therefore, Gi2α is necessary for the muscarinic regulation of Ca2+ channels in the mouse heart. Further studies are needed to delineate the possible interaction of Gi and other cell signaling proteins and to clarify the level of interaction of G protein-coupled regulation of L-type Ca2+ current in the heart.
- Ca2+ current regulation
- gene inactivation
- signal transduction
there are several types of G proteins in the cardiac ventricular myocyte. Each G protein consists of α-, β-, and γ-subunits (4). The structures and functions of the α-subunits most clearly distinguish the Gs, Gi/o, and Gq/11 forms of the G proteins. Gsα is the proximal activator of adenylyl cyclase (AC) and other effector proteins (5). Activation of AC occurs via stimulatory receptor-catalyzed activation of Gs by GTP, resulting in activation of the α-subunit and dissociation of the inhibitory Gβγ complex. β-Adrenergic agonists stimulate AC, increase cAMP concentration, and thereby stimulate cAMP-dependent protein kinase (also called protein kinase A, PKA). The final result is the phosphorylation of effectors such as the L-type Ca2+ channel (21). The group of inhibitory G proteins (Gi), comprised of Gi1, Gi2, and Gi3, is distinguished pharmacologically by the ability of pertussis toxin to transfer the ADP-ribose moiety from NAD to the α-subunit of the Giproteins (17). Muscarinic agonists activate Gi, which inhibits AC and therefore decreases Ca2+ currents. Go (G “other”) is thought to be very similar to Gi (18) and has been shown to regulate muscarinic receptor affinity for agonists in the brain and heart. Goα is also irreversibly inhibited by pertussis toxin. Gqα, a pertussis toxin-insensitive G protein, has been identified in a number of mammalian tissues, including the brain, lung, and heart, and is associated with α1-adrenergic receptor activation (20).
It has been traditionally believed that muscarinic agonists such as carbachol (CCh) bind to muscarinic receptors to promote formation of an activated Gi-GTP complex. On GTP binding, the heterotrimeric G protein dissociates into two moieties, Giα-GTP and Giβγ. Giα inhibits AC (10, 15) and modulates intracellular effectors systems such as Ca2+ channels. In the heart, this classic view has been challenged by the recent discovery that muscarinic inhibition of Ca2+ channels requires the presence of the Goα protein (24). The exact function of Go in the heart has not been well understood (2). In 1997, Valenzuela et al. (24) studied the muscarinic regulation of Ca2+ currents in Goα knockout (Goα−/−) mouse ventricular myocytes. They found that Goα−/− mice have a specific defect in muscarinic regulation of Ca2+ current. These results indicate that the Go protein also plays an important role in the muscarinic regulation of cardiac L-type Ca2+ currents. Thus the role of the Giα protein in the muscarinic regulation of the heart has been called into question.
The purpose of the present study is to determine whether Giis a critical component in the muscarinic regulation of cardiac L-type Ca2+ currents. We hypothesize that Gi, like Go, plays an important role in the muscarinic regulation of Ca2+ currents.
The generation of Gi2α gene knockout mice has been published (22). The Gi2α protein was not detectable in the Gi2α−/− mouse heart preparations (23). Age-matched 129Sv wild-type mice and Gi2α−/− mice were used to isolate cardiac myocytes.
Isolation of calcium-tolerant mouse ventricular myocytes was very similar to our previously described cell isolation from the rabbit heart (8, 9, 26). In brief, mice (8–10 wk of age) were anticoagulated with 1,000 units of heparin and anesthetized with pentobarbital sodium (25 mg) by intraperitoneal injection. Hearts were rapidly excised, cannulated, and perfused for 3 min at a rate of 2.2 ml/min with Ca2+-free Tyrode solution containing (in mM) 136 NaCl, 5.4 KCl, 0.33 NaH2PO4, 1 MgCl2, 10 HEPES, 4 mannitol, 0.6 thiamine-HCl, 10 glucose, and 2 pyruvic acid. The perfusate was switched to Ca2+-free Tyrode solution containing collagenase (1 mg/ml, type L, Sigma; St. Louis, MO) and protease (0.15 mg/ml, type XIV, Sigma), which was recirculated with a peristaltic pump for 5–7 min. The enzymes were washed out for 3 min with 0.1 mM Ca2+ Tyrode solution. In the mouse heart, this technique typically yields between 50 and 70% Ca2+-tolerant rod-shaped cells, which is very similar to the cell yield in the rabbit heart (6, 26).
Ca2+ current measurements.
Whole cell L-type Ca2+ current was measured by using the membrane-ruptured patch configuration (7, 11). To measure Ca2+ currents, ventricular cells were placed in a recording chamber (∼1.0 ml) on the stage of a Diaphot inverted microscope (Nikon). Ca2+ current was recorded using Corning 8161 glass microelectrodes with a tip resistance of 2–4 MΩ when filled with internal solution, which contained (in mM) 96 CsCl, 2 MgCl2, 10 tetraethylammonium chloride, 14 EGTA, 1 CaCl2, 20 HEPES, 0.4 Tris-GTP, and 5 Mg-ATP; pH was titrated to 7.1 with CsOH. The bath solution contained (in mM) 135 CsCl, 10 HEPES, 1.8 CaCl2, 1 MgCl2, 5 glucose, and 0.01 tetrodotoxin and was titrated to a pH of 7.3 with CsOH. Membrane currents were filtered at 1 kHz with an eight-pole Bessel filter (902LPF, Frequency Devices; Haverhill, MA), converted into digital format using an Axolab 1100 data acquisition system with pCLAMP version 5.5 software (Axon Instruments; Burlingame, CA), and digitally stored for later analysis. The whole cell voltage-clamp protocol used in these studies is similar to that described previously (7). Cell membrane potential was held at −80 mV. After a prepulse to −40 mV for 50 ms, Ca2+ currents were elicited by 400-ms depolarizing clamp steps to test potentials ranging from −50 to +50 mV in 10-mV increments. Peak current density was calculated as peak current divided by cell capacitance (in pA/pF). A solenoid-controlled perfusion apparatus was used to change among solutions containing various test agents. A constant flow rate of 1.5 ml/min was used to ensure the exposure of measured cells to the experimental solutions.
The muscarinic agonist CCh, the muscarinic antagonist atropine (ATR), and the β-receptor agonist isoproterenol (Iso) were purchased from Sigma.
Steady-state transmembrane current in response to a ramp deplolarization of 1 mV/ms was used to calculate the capacitance of the cell (C cell) with the use of the relationshipC cell = dQ/dV = (dQ/dt)/(dV/dt), whereQ is charge, V is voltage, and t is time. C cell was used to normalize measured current to total surface area in each cell (27). All experiments were performed at room temperature (23°C).
Muscarinic receptor radiolabeled ligand binding assay.
Individual hearts from wild-type and Gi2α−/− mice (strain 129Sv; 8–10 wk of age) were excised, rinsed in ice-cold Dulbecco's PBS, minced with scissors, and homogenized in 5 ml of 27% (wt/vol) sucrose, 10 mM Tris · HCl (pH 7.5), and 1 mM EDTA (STE buffer) using a glass Dounce homogenizer (20 strokes with loose and 20 strokes with tight pestle). Homogenates were then passed 10 times through a 26-gauge needle. Particulate matter sedimenting between 5 min at 950 g and 60 min at 100,000 g (cardiac membranes) was collected and resuspended in 0.2 ml of ice-cold STE buffer and diluted in 5 mM MgCl2, 1 mM EDTA, and 50 mM Tris · HCl (pH 7.5) (binding buffer) to a concentration of 1 mg protein/ml (ca. 10-fold dilution). Duplicate binding reactions were for 60 min at 22°C in 0.5 ml of binding buffer containing 0.1 mg protein and the indicated concentrations of [N-methyl-3H]scopolamine (85 Ci/mmol, Amersham) and, when present, 1 μM ATR to determine nonspecific binding. Binding reactions were stopped and boundN-methyl-scopolamine was separated from unbound by the bovine γ-globulin/polyethylene glycol coprecipitation method described previously (1, 19). The final membrane pellet was resuspended in 0.4 ml of 0.1 N NaOH, and radioactivity was determined after neutralization with 0.1 ml of 0.4 N HCl in liquid scintillation counter (0.5-ml sample plus 3.0 ml of Parkard Ultima-Flo M scintillation fluid). Under these conditions, binding was proportional to protein concentration and reached equilibrium within 30 min and was stable for up to 90 min.
All values were expressed as means ± SE. Student'st-test was used for paired observations, with a Pvalue of <0.05 indicating a statistically significant difference.
Basal Ca2+ currents are similar in both wild-type and Gi2α knockout mouse cells.
We first examined whether the ventricular myocytes isolated from Gi2α−/− mice had cell sizes and basal Ca2+current properties comparable to wild-type myocytes. Figure1 A shows Ca2+current tracing recordings from a single ventricular myocyte isolated from a wild-type mouse heart. The cell membrane potential was held at −80 mV and clamped to test potentials from −50 to +50 mV with step increments of 10 mV to elicit the Ca2+ current. Figure1 A shows the current tracings from clamping the potential to −40, 0, +20, and +30 mV. Figure 1 B displays the current tracings recorded from a Gi2α−/− mouse myocyte with the same voltage potentials as in Fig. 1 A. The configuration of the Ca2+ current and the time dependence were very similar in both wild-type and Gi2α−/− mouse ventricular myocytes. Figure 1 C shows the current-voltage relation of the L-type Ca2+ currents obtained from these two groups. The data were averaged from 29 wild-type cells (15 mice) and 22 Gi2α−/− cells (14 mice). The peak current density at a test potential of 0 mV was 7.5 ± 0.2 pA/pF (n = 29) in wild-type mice and 7.6 ± 0.3 pA/pF (n = 22) in Gi2α−/− mice (P > 0.05). Figure1 C shows that Ca2+ currents in wild-type and Gi2α−/− mouse ventricular myocytes had similar current characteristics, including the maximum current amplitude, peak current potential, and current reversal potential. In addition, the time to the peak current at a test potential of 0 mV was also similar in wild-type (9.2 ± 0.5 ms, n = 29) and Gi2α−/− mouse ventricular myocytes (8.4 ± 0.5 ms,n = 22, P > 0.05).
Cell sizes between the wild-type and Gi2α knockout mouse hearts were also compared. The cell membrane capacitance was 124.7 ± 5.7 pF in wild-type (n = 29) and 115. 9 ± 6.7 pF in Gi2α−/− (n = 22,P > 0.05) mouse cells. Thus loss of Gi2α expression has no significant effect on the measured basal L-type Ca2+ channel characteristics.
Lack of inhibition of Ca2+ current by a muscarinic cholinergic agonist in Gi2α−/− mice.
To examine the role of the Gi2α protein in the muscarinic regulation of Ca2+ current, we first investigated the inhibitory effects of the muscarinic cholinergic agonist CCh on the wild-type mouse ventricular myocyte Ca2+ current. Figure2, A and B, illustrates the effect of CCh (100 μM) on L-type Ca2+current in the wild-type mouse heart. Figure 2 A shows representative current tracings recorded under control conditions, perfusion with CCh, and perfusion with CCh plus ATR (10 μM). Figure2, A and B, demonstrates that CCh inhibits basal L-type Ca2+ current in wild-type mouse ventricular myocytes and that this inhibition can be reversed by muscarinic receptor blockade with ATR.
In contrast, Fig. 2 C demonstrates that the muscarinic cholinergic regulation of cardiac myocyte Ca2+ current was absent in Gi2α gene knockout myocytes. Neither CCh (100 μM) nor ATR (10 μM) affected the Ca2+ current. Figure2 D displays the time course of current recordings, showing the absence of inhibition in Gi2α−/− mouse Ca2+ current by CCh. These data clearly show that the muscarinic regulation of L-type Ca2+ current by CCh was absent in Gi2α−/− mouse ventricular myocytes.
Figure 3 summarizes the effects of CCh on L-type Ca2+ current from both wild-type and Gi2α−/− mouse ventricular myocytes. In wild-type cells, perfusion of CCh (100 μM) inhibited the L-type Ca2+current from 7.0 ± 0.4 to 5.4 ± 0.3 pA/pF (19 cells from 13 mice, P < 0.01). Perfusion with CCh and ATR (10 μM) blocked the inhibitory effect of CCh, and the current amplitude recovered to 6.5 ± 0.4 pA/pF (n = 17,P < 0.05). On the other hand, CCh did not significantly inhibit the L-type Ca2+ current in Gi2α−/− gene knockout mouse cells (from 7.0 ± 0.6 to 6.9 ± 0.4 pA/pF) (8 cells from 7 mice, P > 0.05). In addition, perfusion of ATR (10 μM) and CCh induced no significant change in L-type Ca2+ current amplitude (6.5 ± 0.3 pA/pF, n = 8, P > 0.05). The small decrease in Ca2+ current amplitude after ATR may be attributed to Ca2+ current rundown. Therefore, these data suggest that Gi expression is necessary for muscarinic cholinergic regulation of L-type Ca2+ current in the heart.
β-Adrenergic agonist-mediated response is intact in Gi2α gene knockout mouse hearts.
We next examined whether the loss of the Gi2α protein interferes with the β-adrenergic agonist-mediated stimulation of Ca2+ current. We compared the responses of L-type Ca2+ currents in ventricular myocytes isolated from either wild-type or Gi2α−/− mice to the β-adrenergic agonist Iso alone and in combination with the cholinergic agonist CCh. Representative examples of wild-type and Gi2α−/− myocyte current amplitudes in response to Iso and CCh are shown in Fig. 4, A and C. Perfusion with 10 μM Iso increased the L-type Ca2+current in wild-type (Fig. 4 A) and Gi2α−/− (Fig. 4 C) mouse ventricular myocytes to a similar extent. This finding indicates that the β-adrenergic agonist-mediated response is not affected in Gi gene knockout mice. Perfusion of CCh (100 μM) in the presence of Iso significantly attenuated the Iso-enhanced Ca2+ current in wild-type mouse ventricular myocytes (Fig. 4, A and B). However, unlike the wild-type mouse ventricular myocyte, perfusion of the same concentration of CCh in the presence of Iso had no significant effect on Iso-stimulated Ca2+ current amplitude in Gi2α−/− myocytes (Fig. 4, C andD).
Figure 5 summarizes the effects of Iso alone and Iso and CCh on L-type Ca2+ current in both wild-type and Gi2α−/− mouse ventricular myocytes. In 11 cells from 9 wild-type mice, L-type Ca2+ current amplitude is increased by Iso from 7.1 ± 0.6 to 9.8 ± 0.8 pA/pF (n = 11, P < 0.01). Addition of CCh to Iso-containing solution reversed the Iso-induced increase in Ca2+ current amplitude to 7.2 ± 0.8 pA/pF (P < 0.01). In Gi2α−/− mice, L-type Ca2+ current amplitude was also increased by Iso from 7.0 ± 0.5 to 10.6 ± 0.7 pA/pF (12 cells from 8 mice,P < 0.01). However, unlike in wild-type cells, subsequent application of CCh did not result in a reduction in the L-type Ca2+ current amplitude (10.4 ± 0.8 pA/pF,n = 12, P > 0.05; Fig. 5).
Gi2α subunit gene knockout does not alter level of muscarinic receptor expression.
To determine the sites and property of muscarinic receptors in Gi2α knockouts, the binding assay of a radiolabeled antagonist ([N-methyl-3H]scopolamine) of the muscarinic receptor to the myocardial membrane protein was performed. Scatchard analysis of the binding of [N-methyl-3H]scopolamine to mouse myocardial membranes showed that the membranes from wild-type mice contained ∼84 fmol/mg membrane protein of specific binding sites with an affinity of 0.3 nM and that the membranes from Gi2α−/− mice contained ∼75 fmol/mg membrane protein of specific binding sites with an affinity of 0.2 nM (Fig.6).
The present study demonstrates that Gi2α is required for the inhibition of L-type Ca2+ currents by muscarinic cholinergic agonists in the heart. Whereas the stimulatory G protein pathway activated by Iso was intact, CCh was not able to inhibit either baseline or Iso-stimulated Ca2+ currents in the Gi2α−/− mouse heart. The targeted disruption of Gi2α gene in the Gi2α−/− mice has been confirmed by ADP ribosylation techniques (23). In the Gi2α−/− mouse, Gi2α protein was not detectable, but levels of Go and other G proteins were unchanged (23). Therefore, the observed absence of muscarinic regulation of L-type Ca2+ current in the Gi2α−/− mouse is attributable to the lack of Gi protein and not to the lack of other cell signaling proteins, such as Go proteins. Although it has been traditionally assumed that Gi is critical for muscarinic regulation of L-type Ca2+ current in the heart, we believe that the present study is the first report to confirm this action of Gi using the specific Gi2α knockout mice model.
Regulation of L-type Ca2+ current by stimulatory G protein-coupled receptors plays an important role in regulating cardiac function. Activation of the channel by the β-adrenergic receptor is a consequence of the activation of AC through Gs. Therefore, cAMP levels rise, PKA is activated, and subsequently the channel is phosphorylated. On the other hand, the mechanism of muscarinic inhibition of the channel is not completely understood (25). In the present study, we found that CCh inhibited the basal (control) Ca2+ current amplitude in wild-type mouse ventricular myocytes. This suggests that Ca2+ channel current may be regulated in vivo by a combination of sympathetic and parasympathetic mechanisms. Therefore, either the use of a β-adrenergic receptor activator such as Iso or a muscarinic cholinoceptor activator such as CCh could affect the Ca2+current amplitude. In the presence of Iso, CCh produced a more profound inhibition of Ca2+ current. This phenomenon has been termed as “accentuated antagonism” (13). More experiments are necessary to elucidate the physiological muscarinic regulation of Ca2+ currents.
The classic view that muscarinic regulation of Ca2+ current requires Gi protein has been challenged by the recent finding that Go protein also plays an important role in muscarinic regulation of cardiac L-type Ca2+ currents (24). The data from the present study, in combination with the findings of Valenzuela et al. (24), suggest that both Gi and Go may be required for muscarinic cholinergic regulation. However, the apparent “dominant negative” effect of both G protein mutations suggests that Gi and Go may interact with their respective G protein-coupled signal pathways in a highly complex fashion.
Further investigation as to which second messengers are involved in the muscarinic regulation of L-type Ca2+ current will help us to understand the mechanisms of this signaling pathway. It has recently been reported that muscarinic cholinergic regulation of cardiac myocyte Ca2+ current is absent in mice with targeted disruption of endothelial nitric oxide (NO) synthase (NOS). Han et al. (13) found that an increase in intracellular cGMP is related to the activity of NOS. This suggests that the Ca2+current is regulated by the NOS and cGMP-dependent protein kinase pathway. In rat ventricular myocytes (3), rabbit sinoatrial nodal cells (14), and atrial ventricular nodal cells (12), muscarinic antagonism of β-adrenergic agonist stimulation of Ca2+ current was reported to depend on activation of constitutively expressed NOS. On the other hand, Vandecasteele et al. (25) recently found that the NO-cGMP pathway does not contribute significantly to the muscarinic regulation of Ca2+ current in human atrial myocytes (25). Although the muscarinic inhibition of AC is impaired in these Gi2α−/− mice (23), we did not examined whether cAMP or NOS/cGMP is involved in the disruption of muscarinic regulation in the Gi2α−/− mouse heart in the present study. Recently, we found in another group of cells that addition of intracellular PKA inhibitor only partially reduced the inhibitory effects of CCh (data not shown). This suggests that after acting on Gi protein, CCh may exert its muscarinic regulatory effects through other signal pathways besides the PKA action.
Our study also shows that loss of muscarinic effect on the L-type Ca2+ current channel in the Gi2α knockout mouse is neither due to the decreasing of receptor binding sites nor the changing of affinity of receptor to its ligand. The data (Fig. 6) presented here indicate that the numbers of receptor binding sites to [N-methyl-3H]scopolamine in the wild-type and Gi2α−/− mouse are similar. The affinity of the receptor to its ligand in the Gi2α−/− mouse myocardial membrane is comparable to that of the wild-type mouse. This suggests that there is no decreasing of the muscarinic receptor binding site in the myocardial membrane of the Gi2α knockout mouse. Loss of the Gi2α subunit by gene knockout does not alter the level of expression of muscarinic receptors in myocardial tissues.
Recently, Jiang et al. (16) also generated Goα gene knockout mice and isolated the ventricular myocytes from these mice. We performed the same experiments as Valenzuela et al. (24). Our preliminary data showed that, similar to their reports, Goα is necessary for muscarinic regulation of Ca2+ current in mouse ventricular myocyte, because muscarinic regulation by CCh is absent in Goα gene knockout mouse ventricular myocytes. Our data confirms findings from Valenzuela et al. and also suggests that both Giα and Goα are necessary for muscarinic regulation of Ca2+ current in the mouse heart. More recently, it has been reported that Gi2α, Gi3α, and Goα are all required for normal muscarinic inhibition of the cardiac Ca2+ channels in nodal and atrial cultured cardiac cells (28). Further studies are needed to determine how these G proteins are involved in cardiac muscarinic regulation and to clarify the level of interaction of G protein-coupled regulation of L-type Ca2+ current in the heart.
The authors thank Trisha Tanabe for assistance in cell isolation and Byron Benedict Waters for help in the preparation of this manuscript.
This work was supported in part by American Heart Association Western States Affiliate Grant 9960052Y, National Institutes of Health Grants HL-02723 and RR-00865, and the Variety Club J. H. Nicholson Endowment.
Address for reprint requests and other correspondence: F. Chen, UCLA School of Medicine, 675 C. E. Young Drive South, Rm. 3754, Los Angeles, CA 90095-7045 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2001 the American Physiological Society