Rad As a Novel Regulator of Excitation–Contraction Coupling and β-Adrenergic Signaling in Heart
Rationale: Rad (Ras associated with diabetes) GTPase, a monomeric small G protein, binds to Cavβ subunit of the L-type Ca2+ channel (LCC) and thereby regulates LCC trafficking and activity. Emerging evidence suggests that Rad is an important player in cardiac arrhythmogenesis and hypertrophic remodeling. However, whether and how Rad involves in the regulation of excitation–contraction (EC) coupling is unknown.
Objective: This study aimed to investigate possible role of Rad in cardiac EC coupling and β-adrenergic receptor (βAR) inotropic mechanism.
Methods and Results: Adenoviral overexpression of Rad by 3-fold in rat cardiomyocytes suppressed LCC current (ICa), [Ca2+]i transients, and contractility by 60%, 42%, and 38%, respectively, whereas the “gain” function of EC coupling was significantly increased, due perhaps to reduced “redundancy” of LCC in triggering sarcoplasmic reticulum release. Conversely, ≈70% Rad knockdown by RNA interference increased ICa (50%), [Ca2+]i transients (52%) and contractility (58%) without altering EC coupling efficiency; and the dominant negative mutant RadS105N exerted a similar effect on ICa. Rad upregulation caused depolarizing shift of LCC activation and hastened time-dependent LCC inactivation; Rad downregulation, however, failed to alter these attributes. The Na+/Ca2+ exchange activity, sarcoplasmic reticulum Ca2+ content, properties of Ca2+ sparks and propensity for Ca2+ waves all remained unperturbed regardless of Rad manipulation. Rad overexpression, but not knockdown, negated βAR effects on ICa and Ca2+ transients.
Conclusion: These results establish Rad as a novel endogenous regulator of cardiac EC coupling and βAR signaling and support a parsimonious model in which Rad buffers Cavβ to modulate LCC activity, EC coupling, and βAR responsiveness.
Cardiac excitation–contraction (EC) coupling is mainly mediated by intermolecular signaling between two types of Ca2+ channels, the voltage-gated L-type Ca2+ channel (LCC), and the ryanodine receptor (RyR) Ca2+ release channel that reside in the plasma membrane and the sarcoplasmic reticulum (SR), respectively. During EC coupling, LCC Ca2+ influx activates a large number of “Ca2+ sparks”1 from clusters of RyRs, via the Ca2+-induced Ca2+ release mechanism.2 Summation of Ca2+ sparks across the cell gives rise to an intracellular Ca2+ transient that signals contractile myofilaments to generate force and movement. Return to the diastolic Ca2+ level, to relax the muscle, is controlled by Ca2+ cycling via the SR Ca2+-ATPase (SERCA) and, to a lesser extent, the sarcolemmal Na+/Ca2+ exchanger (NCX). Albeit controversial, recurrent evidence also suggests the involvement of trigger mechanism other than LCC. In particular, it has been suggested that reverse NCX allosterically activated by LCC current augments the trigger Ca2+ at high membrane voltage.3
Rad (Ras associated with diabetes), a monomeric small G protein that was initially identified by subtractive cloning as genes overexpressed in the skeletal muscle of a subset of patients with type 2 diabetes, is expressed most abundantly in the heart,4 along with its cousin Rem, but not Gem/Kir5 in the RGK family. At the molecular level, Rad comprises multiple functional domains including calmodulin binding and 14-3-3 protein-binding domains, as well as regulatory phosphorylation sites.6,7 A common feature of Rad and other RGK proteins is to bind directly to Cavβ of LCC, a multi-subunit complex consisting of the pore-forming α1 subunit and auxiliary subunits such as Cavβ and α2δ. Among others, Cavβ subunits facilitate the channel complex trafficking to the plasma membrane, increase channel open probability, produce a hyperpolarizing shift of voltage-dependent channel activation, and alter channel gating kinetics.8,9 As such, exogenous overexpression of Gem or Rem suppresses the native LCC current (ICa) in adult or embryonic ventricular myocytes.10,11
Recently, Rad as well as its cousins in the RGK family has emerged as an important regulator of diverse cardiac functions10–14 and a potential therapeutic target for treating heart diseases.10,12 Specifically, dominant negative suppression of Rad prolongs QT intervals and causes ventricular arrhythmias12; conversely, somatic gene transfer of Gem to the heart or focally to the atrioventricular node has been considered a strategy to antagonize LCC activity and thus to stabilize cardiac electrophysiological function.10 In a Rad knockout mouse model, hypertrophic cardiac remodeling becomes exaggerated under pressure overload conditions, via enhanced Ca2+/calmodulin-dependent kinase II signaling.13 A population genetics study has also defined a link between a nonsynonymous single-nucleotide polymorphism within the Rad gene (Radq66p) and patients with congestive heart failure.14 Consistent with Rad being a signaling molecule, Rad undergoes rapid turnover in different physiological and pathophysiological settings: upregulation in diabetic patients4 and downregulation in the failing human heart as well as in a mouse model of cardiac hypertrophy.13 A large, sustained decrease of Rad occurs within hours of acute adrenergic stimulation by phenylephrine in neonatal cardiac myocytes.13
The central hypothesis tested in this study was that, although dispensable to the cascade of cardiac EC coupling, Rad constitutes an endogenous regulator of this process such that up- or downregulation of Rad resets cardiac inotropy. Furthermore, we intended to explore how the Rad signal interplays with β-adrenergic receptor (βAR) stimulation, the most powerful “beat-to-beat” inotropic mechanism in the heart. Our results garnered from confocal, electrophysiological and genetic approaches revealed that Rad imparts bidirectional regulation of cardiac contractility and systolic Ca2+ level, and Rad overexpression negates βAR-mediated ICa and inotropic effects without altering βAR lusitropic effects. Along the way, we demonstrated an asymmetry of bidirectional Rad manipulation such that increasing or decreasing Rad fails to produce opposite effects on kinetic attributes of ICa, the gain function of EC coupling, and responsiveness to βAR stimulation. Mechanistically, these diverse findings were integrated in a parsimonious model in which Rad acts as a buffer of Cavβ subunits that, in turn, constitute a key effector of βAR signaling.
Manipulating Rad Expression in Cardiac Myocytes
All experimental procedures and protocols were approved by the Institutional Animal Care and Use Committee of Peking University, an AAALAC International–accredited institution. Single cardiac myocytes were enzymatically isolated from ventricles of 2- to 3-month-old Sprague–Dawley rats and were then cultured as described previously.15 Cultured myocytes were infected with recombinant adenovirus vectors carrying human Rad (Ad-Rad) or its dominant negative mutant RadS105N (Ad-RadS105N) at a multiplicity of infection of 5, with green fluorescent protein adenovirus (Ad-GFP) applied at the same multiplicity of infection as the control. Recombinant adenovirus for silencing endogenous Rad expression (Rad-shRNA, at a multiplicity of infection of 40) was prepared with the BLOCK-iT Adenoviral RNAi Expression system (Invitrogen). The sequences of the oligonucleotides for Rad RNA interference were as follows: forward, CACCGGCTCAGAGGATGGCGTTTACGAATAAACGCCATCCTCTGAGCC; reverse, AAAAGGCTCAGAGGATGGCGTTTATTCGTAAACGCCATCCTCTGAGCC. Adenoviral vectors containing a scrambled shRNA sequence (Ctrl-shRNA) at the same multiplicity of infection served as the control.
Protein was extracted 48 hours after viral infection and rabbit anti-Rad polyclonal antibody (a kind gift from Ronald Kahn in the Joslin Diabetes Center, Harvard Medical School) was used for Rad detection. GAPDH antibody was purchased from Santa Cruz Biotechnology.
The ICa and NCX current (INCX) were recorded under the whole-cell patch-clamp configuration with an EPC 10 amplifier (HEKA Electronics, Lambrecht, Germany), as described previously.16,17 The pipette-filling solution for ICa recording contained (mmol/L): CsOH 120, aspartic acid 120, Na2ATP 5, HEPES 10, TEA-Cl 20, MgCl2 4.33, pH adjusted to 7.4 with CsOH. Bath solution contained (mmol/L): NaCl 137, KCl 5.4, CaCl2 1.0, MgCl2 1.2, NaH2PO4 1.2, HEPES 20, glucose 10, and TTX 0.02 (pH 7.4). The pipette-filling solution for INCX recording contained (in mmol/L): aspartic acid 80, CsOH 80, TEA-Cl 20, MgCl2 2.5, HEPES 10, EGTA 11, CaCl2 7.5, CsCl 15, NaCl 10, and Na2-ATP 4 (pH=7.2). Cells were first perfused with the solution containing (in mmol/L): NaCl 140, MgCl2 1, glucose 10, HEPES 5, CaCl2 2, BaCl2 1, CsCl 4, nifedipine 0.01, ryanodine 0.005, and quabain 0.02 (pH=7.4). The membrane potential under voltage-clamp was first ramped from a holding potential of −55 mV to +80 mV for 100 ms to inactivate the Na+ current and then ramped down to −110 mV. Once the current reached steady state, the external solution was switched to one added with 5 mmol/L NiCl2. The INCX was defined as the Ni2+-sensitive current.
Measurement of Intracellular Ca2+
Myocytes loaded with Rhod-2 acetoxymethyl ester (Rhod-2 AM, 10 μmol/L, 10 minutes) (Invitrogen) were measured using a Zeiss LSM-510 confocal microscope with a ×40 oil immersion lens (NA 1.3). Laser excitation (543 nm) and emission (>560 nm) were used for detecting Rhod-2 fluorescent signal. To account for the indicator nonlinearity, the F/F0 of Rhod-2 fluorescence was converted into absolute Ca2+ concentration using a method described in the Online Data Supplement, available at http://circres.ahajournals.org. For resting Ca2+ measurement, cells loaded with Indo-1 AM (10 μmol/L, 15 minutes) were excited by UV light, and fluorescence emitted was recorded by a charge-coupled device camera (Luca, Andor Technology, South Windsor, Conn) at 405 nm (F405) and 485 nm (F485) separated via an emission splitter. The cytosolic [Ca2+] was then estimated by R=F405/F485. All experiments were performed at room temperature (22°C to 24°C). Off-line image processing used IDL software (version 6.1, Research Systems Co Inc.)
Data are reported as means±SEM with statistical analysis using unpaired Student t test or 2-way ANOVA, when appropriate. A level of P<0.05 was considered to be statistically significant.
Bidirectional Manipulation of Rad in Cell Models
Because of the lack of pharmacological means to intervene the Rad signal, we resorted to adenoviral gene delivery and RNA interference to either increase or decrease Rad expression in cultured adult rat cardiac myocytes. Ad-Rad infection at a multiplicity of infection of 5 for 48 hours led to ≈3-fold upregulation of Rad protein level compared with that in Ad-GFP infected cells (Figure 1A and 1B). In contrast, Rad protein level was reduced by 70% at 48 hours after Rad-shRNA infection (multiplicity of infection, 40). This relatively high RNA interference efficiency is consistent with the fact that endogenous Rad undergoes rapid turnover as a signaling molecule.13 In a subset of experiments, Rad activity was suppressed by adenoviral expression of a dominant negative mutant, RadS105N. Neither up- nor downregulation of Rad altered myocyte viability (data not shown) or morphology of the transverse tubule (TT) system (Online Figure I), a structure critical to cardiac EC coupling.
Rad Regulation of Ca2+ Transients and Contractility
Using the cell models described above, we first sought to delineate the overall effects of Rad on intracellular Ca2+ transients and contractility. We found that Rad overexpression resulted in a marked reduction in Ca2+ transients and cell shortening elicited by field stimulation at 1 Hz (Figure 1C). On average, the amplitude of Ca2+ transients and the maximal cell shortening were decreased by 42% and 38%, respectively (Figure 1D and 1E). Conversely, a 70% downregulation of endogenous Rad increased the Ca2+ transients and contraction amplitudes by 52% and 58%, respectively (Figure 1D and 1E). These results provide the first direct evidence for the small G protein Rad as a strong regulator of cardiac inotropy. That cardiac contractility is responsive to bidirectional Rad manipulation further implies that Rad and other RGK proteins coexpressed in the heart are not functionally redundant.
To assess the Ca2+ sensitivity of contractile myofilaments, we used the phenomenon of rest potentiation characteristic of rodent cardiac myocytes to vary the magnitudes of the Ca2+ transient and cell shortening. Regardless of Rad manipulation, Ca2+ transients and cell shortening elicited by 1 Hz field stimulation after a 3-minute rest displayed a robust negative staircase (Online Figure III). Analyzing the relationship between peak cell shortening and the corresponding peak Ca2+ transient, revealed no change in myofilament responsiveness to [Ca2+]i in either Rad-overexpressing (Figure 1F) or knockdown cells (data not shown). Hence, the inotropic effect of Rad is mainly attributable to changes in systolic [Ca2+]i levels.
The spatial uniformity and temporal synchrony of Ca2+ transients are important aspects of cardiac EC coupling and its modulation by βAR stimulation.18 Dyssynchronous SR Ca2+ release is also thought to be linked to reduced inotropy and enhanced arrhythmogenesis in heart failure models.19,20 With power spectral analysis applied to spatial Ca2+ profiles during the rise of Ca2+ transients (Figure 2A and 2B), we detected little change in spatial properties (Figure 2C), despite an overall 2.2-fold difference in Ca2+ transient magnitude in the 4 groups. This result indicates that Rad inhibition or enhancement of EC coupling occurs uniformly across the cell.
In characterizing the effects of Rad on Ca2+ handling, we found that bidirectional Rad manipulation does not alter resting Ca2+ levels (Online Figure II, A). Electrophysiological recording of INCX revealed that Rad alteration of LCC Ca2+ influx does not elicit any compensatory change in the capacity of NCX for Ca2+ extrusion (Figure 2G and 2H). Likewise, the capacity for SERCA activity appeared to be intact, evidenced by the lack of changes in the relaxation kinetics of Ca2+ transients (Online Figure II, B). Caffeine-liable SR Ca2+ content (Figure 2D), the rate of occurrence of RyR Ca2+ sparks (Figure 2E) and properties of individual sparks including amplitude, spatial width, and temporal duration (Online Figure IV), all remained unaltered over the range of Rad maneuvers examined. To further address whether Rad perturbs Ca2+ stability, we challenged the cells with up to 20 mmol/L extracellular Ca2+ concentration ([Ca2+]o) and found that the propensity for Ca2+ wave production as a function of [Ca2+]o were independent of Rad manipulation (Figure 2F). These results indicate that Rad modulation of Ca2+ transients occurs with little or no changes in the activities of NCX, RyR and SERCA.
Rad Alteration of ICa Characteristics
To delineate the molecular mechanisms underlying Rad regulation of EC coupling, we measured the ICa response to bidirectional Rad manipulation (Figure 3). Three prominent changes in ICa characteristics were found in cells overexpressing Rad. First, ICa density was markedly diminished over the entire voltage range tested (−50 mV to +50 mV) (Figure 3A and 3B), in good agreement with previous reports.10,12,21 Second, there was a marked depolarizing shift in the voltage-dependent activation of LCCs (Figure 3C), whereas the steady-state voltage-dependent inactivation of the channel remained unchanged (Figure 3D). Specifically, the voltage at half-maximum activation potential (V0.5,act) estimated from Boltzmann fitting of the data were −24.6±2.6 mV (n=13) in the Ad-GFP group and −11.0±1.9 mV in the Ad-Rad group (n=9, P<0.01). Third, time-dependent inactivation of ICa was significantly accelerated (Figure 3E and 3F), rather than being slowed as expected of diminished Ca2+ release-dependent inactivation. These data indicate that Rad regulates both functional channel density and gating properties of LCCs.
In contrast, RNA interference knockdown of endogenous Rad resulted in an enhancement of LCC activity: the peak ICa density at −10 mV was increased by 50% in the Rad-shRNA group (11.21±0.87 pA/pF, n=11) compared to the Ctrl-shRNA group (7.47±0.93 pA/pF, n=11, P<0.01). Similar results were obtained by Rad suppression with a dominant-negative mutant, RadS105N (Online Figure V). Contrary to Rad upregulation, Rad knockdown altered neither voltage-dependent activation (Figure 3C) nor steady-state voltage-dependent (Figure 3D) and time-dependent inactivation (Figure 3E and 3F) of ICa. Taken together, these data showed that Rad exerts a powerful and multifaceted regulatory effect on LCCs native to cardiac myocytes, and that up- and downregulation of Rad differentially modulate the kinetic attributes of ICa, rendering an asymmetry for up- and downregulation of Rad signaling.
Rad Increased EC Coupling Efficiency
Next, we exploited Rad manipulation to probe into intermolecular LCC-to-RyR signaling during EC coupling. For this purpose, we recorded ICa and confocal Ca2+ images simultaneously under whole-cell voltage clamp conditions (Figure 4A), and determined the gain function of EC coupling (ratio of the peak increase of [Ca2+]i over the corresponding peak ICa density) in cells with higher or lower than physiological levels of Rad. Irrespective of Rad manipulation, a bell-shaped voltage dependence for peak [Ca2+]i mirrored a “V”-shaped voltage dependence of its corresponding peak ICa density in all groups (Figure 4B and 4C). Rad upregulation diminished whereas Rad downregulation elevated systolic Ca2+ transients, as was the case in cells under field stimulation.
Two caveats were derived from the gain function analysis. First, Rad depression of ICa was accompanied by a significant increase of the gain such that a given amount of ICa triggered a 287%, 178%, and 77% greater release at −30, −20, and −10 mV, respectively. Second, the gain function in Rad knockdown cells was nearly identical to that in controls (Figure 4D), indicating that the fraction of ICa augmented by Rad knockdown is as efficacious as native ICa in activating RyR-mediated Ca2+ release. Again, up- and downregulation of Rad signaling did not produce opposite effect on the EC coupling efficiency.
Rad Negated βAR Augmentation of ICa and Ca2+ Transients
Data thus far established that Rad acts as an endogenous regulator of cardiac EC coupling by modulating ICa. Because both Rad and βAR signals converge on LCCs, this raises an intriguing possibility that Rad might interact with the classic βAR regulatory mechanism. Indeed, we found that the ability of the βAR agonist isoproterenol (500 nmol/L) to augment ICa was dramatically diminished by Rad overexpression (eg, 218±17% in Ad-GFP versus 144±34% in Ad-Rad, at −10 mV; P<0.05) (Figure 5A). In control cells, Iso caused the voltage-dependent activation of LCCs a typical hyperpolarizing shift of 7 mV; and this was largely abolished in Rad-overexpressing cells (Figure 5C). Concomitantly, βAR enhancement of Ca2+ transients was attenuated by 10% to 80% depending on voltage (Figure 6B and 6C). Intriguingly, not all attributes of βAR signaling were equally affected: Iso acceleration of Ca2+ transients, a hallmark of cardiac βAR signaling, remained intact in cells overexpressing Rad (Figure 6A; Online Figure II, C and D). Furthermore, Iso enhancement of SR content was unaffected by Rad overexpression (Online Figure VI).
For downregulation of Rad, near identical ICa–voltage relationship and voltage dependence of LCC activation were obtained from Rad-shRNA and Ctrl-shRNA groups after Iso stimulation, though basal ICa was significantly elevated (Figure 5B and 5D). That is, the ICa response to βAR signaling is largely unperturbed at reduced levels of Rad. Regarding the gain function of EC coupling (normalized by the SR content), it was decreased by Iso in control cells (Figure 6D), as shown previously,18 but this effect was largely reversed by Rad overexpression (Figure 6E).
The present study provides the first comprehensive characterization of the small G protein Rad in the context of cardiac EC coupling and its modulation by βAR signaling. Major phenotypes of Rad genetic manipulation include bidirectional regulation of ICa, Ca2+ transients and cardiac contractility, and the lack of Rad effects on cardiac lusitropy, myofilament responsiveness, spatial uniformity of EC coupling, and Ca2+ stability in resting cells. Importantly, higher than physiological levels of Rad negate βAR-mediated ICa and Ca2+ responses while sparing the βAR lusitropic response, adding a new layer to modulation of cardiac inotropy. Below we discuss how these diverse findings can be unified in a single parsimonious model that provides new insights into LCC subunit interaction and intermolecular LCC-to-RyR coupling.
New Insights Into Rad regulation of LCCs
The present study identified a number of interesting features of Rad regulation of native LCCs in cardiac myocytes. First, with RNA interference and dominant negative suppression of Rad, we unmasked the role of endogenous Rad in setting basal ICa density in rat cardiac myocytes, in agreement with previous reports.12,21 Second, Rad overexpression causes a large depolarizing shift of the voltage-dependent activation and hastens time-dependent ICa inactivation, in contrast to Yada et al, whose data suggest unaltered voltage-dependent activation by Rad overexpression in guinea-pig ventricular myocytes.12 Furthermore, we showed, for the first time, that Rad overexpression and knockdown are asymmetrical with respect to modulating ICa kinetics. This finding is consistent with previous reports that overexpression of the β2b subunit, a candidate for the endogenous cardiac β subunit, increases LCC density but causes minor or no changes in LCC gating properties.8
Recent studies by Colecraft and others have shown that functionalities of Cavβ include the following: (1) enhancing membrane expression of the α1 subunit; (2) normalizing channel activation kinetics; (3) causing a hyperpolarizing shift of the voltage dependence of channel activation; and (4) conferring distinctive channel inactivation properties. Thus, the aforementioned data can be explained by a parsimonious model in which Rad acts as the buffer of Cavβ, and consequentially, Rad overexpression causes “loss of Cavβ subunit function” and Rad knockdown leads to “gain of Cavβ subunit function.”
What insight can we gain from the striking asymmetry of bidirectional Rad manipulation, as demonstrated in this study? Two models have been proposed for the α1-Cavβ subunit interaction.9,22,23 In the single CaVβ-binding model, one CaVβ binds to the α1 subunit at the endoplasmic reticulum membrane to promote the trafficking of the α1-CaVβ complex to the plasma membrane, where a reversible 1:1 interaction between α1 and CaVβ subunits switches the channel between low- and high-open probability (Po) modes. In the multiple CaVβ-binding model, trafficking and gating-modulation are mediated by different CaVβ subunits: a high-affinity binding of a CaVβ subunit to an α1 subunit promotes the channel complex trafficking to the plasma membrane; another CaVβ subunit binding to an α1 subunit with lower affinity converts the channel from low- to high-Po mode. Both models are consistent with the finding that loss of Cavβ subunit function (as in Rad overexpression) alters the kinetic attributes of ICa, perhaps because of less than unity CaVβ:α1 stoichiometry at the plasma membrane. However, only in the single CaVβ-binding model, excessive Cavβ subunit function (as in Rad knockdown) would not alter the kinetic attributes of ICa because of a saturating 1:1 stoichiometry between α1 and CaVβ subunits at the plasma membrane. Hence, the present data are in favor of the single CaVβ-binding model and argue against the alternative model.
Rad Regulation of EC Coupling Efficiency
Recent studies have suggested a conceptual model which relates the efficiency of unitary LCC currents in triggering the abutting RyRs in a diad to microscopic attributes of LCC (channel number, channel open probability, open duration, unitary current, reopening) and to cooperativity of Ca2+-dependent RyR activation.24,25 Both experimental data24 and theoretical considerations25 have demonstrated that, despite a low LCC-to-RyR coupling fidelity (ie, many LCC openings trigger a RyR Ca2+ spark1,25), there is a “redundancy” of the trigger ICa in a diad: reducing either the number of individual open channel or single channel current alone increased EC coupling gain. Because Rad overexpression decreases LCC density and perhaps open probability as well, the increased gain when ICa decreases with overexpression of Rad could be attributable to reduced redundancy of ICa as the trigger of SR Ca2+ release. Similar reasoning also helps to explain the decreased gain in Iso-stimulated cells, ie, increased redundancy with several-fold increase of ICa. However, the relationship between ICa and the gain function may not be linear, because microscopic attributes of unitary currents (open duration, gating mode, reopening, etc) are likely important determinants, and this might explain the lack of effect on the gain function by Rad knockdown when ICa was moderately increased. Regarding βAR regulation of the gain function, there are at least two components: sensitization of RyR to increase the gain and increase of the ICa redundancy to decrease the gain. As a result, βAR stimulation decreases the gain when ICa is greatly enhanced (eg, this study) or increases the gain when ICa is only moderately augmented.26
Rad As a Novel Regulator of βAR Signaling
A major finding of the present study is that Rad selectively negates certain aspects of βAR signaling in heart cells, indicative of interplay between the Rad and βAR regulatory mechanisms. Mechanistically, activation of βAR signaling increases ICa through protein kinase (PK)A-mediated phosphorylation of both α1 and CaVβ subunits.27,28 In intact cardiac myocytes, agents that increase cAMP cause phosphorylation primarily of CaVβ subunits,29 which harbor a number of consensus sequences for PKA phosphorylation. PKA phosphorylation of Cavβ subunits leads to upregulation of LCC activity and a hyperpolarizing shift of LCC activation.27 Because Rad disrupts βAR signaling onto CaVβ subunits of LCCs, our results on Rad regulation of βAR signaling in heart cells can readily be unified by the above model of α1-CaVβ–Rad interaction. Similar reasoning may also be applicable to the fact that βAR responsiveness is essentially lost in mouse heart overexpressing the α1 subunit.30 Both the present and previous results suggest that Cavβ is a key effector in the manifestation of βAR effect on the LCC.
Rad Versus βAR Regulation of Cardiac EC Coupling
Rad and βAR regulation of cardiac EC coupling are distinct yet complementary in important ways. βAR signaling involves protein kinase A activation, and phosphorylation of a host of effector proteins such as LCCs and contractile myofilaments. Hence, it is dynamic, suitable for strong beat-to-beat regulation as in fight-or-flight situations or during exercise. In contrast, Rad signaling is slow but persistent, consequential to Rad expression. Rad modulation plays a significant role in tuning the “set points” for systolic Ca2+ level and cardiac contractility. Furthermore, βAR, but not Rad, signaling accelerates SR Ca2+ cycling, resulting in a positive lusitropic effect and augmented SR Ca2+ load, the latter is conducive to Ca2+-dependent arrhythmias.31 Although Rad does not perturb Ca2+ stability in electrically unstimulated cells, as shown here, cardiac arrhythmogenesis is seen in Rad manipulated animal models, mainly through affecting action potential duration.12 Finally, it should be noted that both Rad and βAR signaling qualify as regulatory mechanisms because they are dispensable to cardiac EC coupling per se. Neither Rad knockout13 nor β1AR and β2AR double-knockout causes overt abnormalities in laboratory mice without stress.32
In summary, we have demonstrated that the small G protein Rad serves as an important endogenous regulator of cardiac EC coupling and of βAR signal transduction. Mechanistically, Rad modulation of ICa characteristics, EC coupling efficiency and responsiveness to βAR stimulation can largely be explained in a parsimonious model of Rad–Cavβ-α1 interaction and of Cavβ being a key effector for βAR signaling. The present and previous findings thus underscore Rad as a novel therapeutic target for maintaining and rescuing cardiac contractile function in the treatment of heart failure.
We thank Drs Ruiping Xiao and Y. Eugene Chen for critical comments, Iain C. Bruce for manuscript editing, and Quan Du for technical support.
Sources of Funding
This work was supported by grants from the Major State Basic Research Development Program (2007CB512100) and the Natural Science Foundation of China (30630021, 30628009, 30400223, 30671027, and 30570418) and American Heart Association National Career Development Grant 0835237N.
↵*Both authors contributed equally to this work.
Original received March 2, 2009; resubmission received August 31, 2009; revised resubmission received November 9, 2009; accepted November 11, 2009.
Cheng H, Lederer WJ. Calcium sparks. Physiol Rev. 2008; 88: 1491–1545.
Reynet C, Kahn CR. Rad: a member of the Ras family overexpressed in muscle of type II diabetic humans. Science. 1993; 262: 1441–1444.
Finlin BS, Andres DA. Rem is a new member of the Rad- and Gem/Kir Ras-related GTP-binding protein family repressed by lipopolysaccharide stimulation. J Biol Chem. 1997; 272: 21982–21988.
Moyers JS, Bilan PJ, Zhu J, Kahn CR. Rad and Rad-related GTPases interact with calmodulin and calmodulin-dependent protein kinase II. J Biol Chem. 1997; 272: 11832–11839.
Murata M, Cingolani E, McDonald AD, Donahue JK, Marban E. Creation of a genetic calcium channel blocker by targeted gem gene transfer in the heart. Circ Res. 2004; 95: 398–405.
Crump SM, Correll RN, Schroder EA, Lester WC, Finlin BS, Andres DA, Satin J. L-type calcium channel α-subunit and protein kinase inhibitors modulate Rem-mediated regulation of current. Am J Physiol Heart Circ Physiol. 2006; 291: H1959–H1971.
Yada HMM, Shimoda K, Yuasa S, Kawaguchi H, Ieda M, Adachi T, Murata M, Ogawa S, Fukuda K. Dominant negative suppression of Rad leads to QT prolongation and causes ventricular arrhythmias via modulation of L-type Ca2+ channels in the heart. Circ Res. 2007; 101: 69–77.
Chang L, Zhang J, Tseng YH, Xie CQ, Ilany J, Brüning JC, Sun Z, Zhu X, Cui T, Youker KA, Yang Q, Day SM, Kahn CR, Chen YE. Rad GTPase deficiency leads to cardiac hypertrophy. Circulation. 2007; 116: 2976–2983.
Lynch RA, Wagoner L, Li S, Sparks L, Molkentin J, Dorn GW II. Novel and nondetected human signaling protein polymorphisms. Physiol Genomics. 2002; 10: 159–168.
Yang D, Zhu WZ, Xiao B, Brochet DX, Chen SR, Lakatta EG, Xiao RP, Cheng H. Ca2+/calmodulin kinase II-dependent phosphorylation of ryanodine receptors suppresses Ca2+ sparks and Ca2+ waves in cardiac myocytes. Circ Res. 2007; 100: 399–407.
Song LS, Wang SQ, Xiao RP, Spurgeon H, Lakatta EG, Cheng H. β-Adrenergic stimulation synchronizes intracellular Ca2+ release during excitation-contraction coupling in cardiac myocytes. Circ Res. 2001; 88: 794–801.
Song LS, Sobie EA, McCulle S, Lederer WJ, Balke CW, Cheng H. Orphaned ryanodine receptors in the failing heart. Proc Natl Acad Sci U S A. 2006; 103: 4305–4310.
Litwin SE, Zhang D, Bridge JHB. Dyssynchronous Ca2+ sparks in myocytes from infarcted hearts. Circ Res. 2000; 87: 1040–1047.
Altamirano J, Bers DM. Voltage dependence of cardiac excitation–contraction coupling: unitary Ca2+ current amplitude and open channel probability. Circ Res. 2007; 101: 590–597.
Bunemann M, Gerhardstein BL, Gao T, Hosey MM. Functional regulation of L-type calcium channels via protein kinase A-mediated phosphorylation of the β2 subunit. J Biol Chem. 1999; 274: 33851–33854.
Hulme JT, Westenbroek RE, Scheuer T, Catterall WA. Phosphorylation of serine 1928 in the distal C-terminal domain of cardiac CaV1.2 channels during β1-adrenergic regulation. Proc Natl Acad Sci U S A. 2006; 103: 16574–16579.
Muth JN, Yamaguchi H, Mikala G, Grupp IL, Lewis W, Cheng H, Song LS, Lakatta EG, Varadi G, Schwartz A. Cardiac-specific overexpression of the α1 subunit of the L-type voltage-dependent Ca2+ channel in transgenic mice. Loss of isoproterenol-induced contraction. J Biol Chem. 1999; 274: 21503–21506.
Rohrer DK, Chruscinski A, Schauble EH, Bernstein D, Kobilka BK. Cardiovascular and metabolic alterations in mice lacking both β1- and β2-Adrenergic receptors. J Biol Chem. 1999; 274: 16701–16708.