Carvedilol inhibits cADPR- and IP₃-induced Ca²⁺ release

Abstract

Spontaneous Ca²⁺ waves, also termed store-overload-induced Ca²⁺ release (SOICR), in cardiac cells can trigger ventricular arrhythmias especially in failing hearts. SOICR occurs when RyRs are activated by an increase in sarcoplasmic reticulum (SR) luminal Ca²⁺. Carvedilol is one of the most effective drugs for preventing arrhythmias in patients with heart failure. Furthermore, carvedilol analogues with minimal β-blocking activity also block SOICR showing that SOICR-inhibiting activity is distinct from that for β-block. We show here that carvedilol is a potent inhibitor of cADPR-induced Ca²⁺ release in sea urchin egg homogenate. In addition, the carvedilol analog VK-II-86 with minimal β-blocking activity also suppresses cADPR-induced Ca²⁺ release. Carvedilol appeared to be a non-competitive antagonist of cADPR and could also suppress Ca²⁺ release by caffeine. These results are consistent with cADPR releasing Ca²⁺ in sea urchin eggs by sensitizing RyRs to Ca²⁺ involving a luminal Ca²⁺ activation mechanism. In addition to action on the RyR, we also observed inhibition of inositol 1,4,5-trisphosphate (IP₃)-induced Ca²⁺ release by carvedilol suggesting a common mechanism between these evolutionarily related and conserved Ca²⁺ release channels.

Introduction

Calcium (Ca²⁺) is a highly versatile intracellular signal that regulates a wide range of cellular functions, including exocytosis, contraction, metabolism, transcription, fertilization and cell proliferation¹. Sources for Ca²⁺ signaling include both influx through channels at the plasma membrane and Ca²⁺ mobilization from Ca²⁺ storage organelles. A key question is how the ubiquitous Ca²⁺ signaling ion can control cellular processes with a high degree of specificity. One potential solution is the realization that there are multiple Ca²⁺ mobilizing messengers targeting specific Ca²⁺ release channels in distinct organelles. Three molecules have satisfied all the criteria for assignment as Ca²⁺ mobilizing second messengers. These are IP₃, and the pyridine nucleotide metabolites, cADPR and NAADP ². The first two molecules target Ca²⁺ release channels in the cell’s largest Ca²⁺ storage organelle, the endoplasmic reticulum (ER), whilst NAADP releases Ca²⁺ from organelles of the endolysosomal system likely through two-pore channels (TPCs) ^{3, 4}.

IP₃R and RyR channels are the evolutionarily related principal sarco-endoplasmic Ca²⁺ release channels mediating Ca²⁺ mobilization from this organelle in response to various stimuli ⁵. These channels are the main mediators of Ca²⁺-induced Ca²⁺release (CICR) ⁶, an autocatalytic mechanism by which cytoplasmic Ca²⁺ activates the release of Ca²⁺ from internal stores. This mechanism contributes to the globalization of intracellular Ca²⁺ signals in cells including propagating Ca²⁺ waves, since in the absence of such mechanisms buffering mechanisms greatly restrict the spatial diffusion of Ca²⁺ signals making them inherently local. IP₃ and cADPR are thought to evoke openings of channels by sensitizing them to Ca²⁺ as a co-agonist ⁶. Two principal modes for triggering CICR have been proposed. The first is a cytosolic mode whereby an increase in cytosolic Ca²⁺ may activate Ca²⁺ release channels. The second is a luminal mode whereby an increase in intraluminal Ca²⁺ concentrations trigger the opening of Ca²⁺ release channels. The latter is associated with the phenomenon of spontaneous Ca²⁺ release from ER/SR^7–9, which has been proposed as an important mechanism underlying various cardiac arrhythmias ¹⁰.

cADPR was first identified as a Ca²⁺ mobilizing molecule in sea urchin eggs and homogenates ¹¹. Previous work had indicated that NAD was enzymatically converted to a Ca²⁺ mobilizing agent in sea urchin eggs and homogenates ¹². Pharmacological studies showed that cADPR targeted RyRs but not IP₃Rs based on cross-desensitization with RyR activators and inhibition by RyR blockers ¹³. Furthermore, both divalent cations and the RyR pharmacological activator caffeine, potentiate Ca²⁺ release by cADPR ¹⁴. Subsequent studies in many mammalian cell types support the link between cADPR and RyRs and, in cardiac cells, cADPR promotes the production of Ca²⁺ sparks and regulates contractility ^{15 16}. An excess of cADPR may be pro-arrhythmic ¹⁶, and an inhibitor of the enzyme that synthesizes cADPR is proposed as a novel anti-arrhythmic drug ¹⁷. In summary, cADPR is considered a second messenger that acts by sensitizing RyRs to CICR ¹⁸.

It has long been recognized that spontaneous Ca²⁺ release occurs during SR Ca²⁺ overload, a process also known as store-overload induced Ca²⁺ release (SOICR). This is a luminal mechanism whereby as the ER (or SR) fills with Ca²⁺ when the intra-luminal Ca²⁺ concentration or amount reaches a certain level or threshold, the opening of RyRs is triggered ¹⁰. Such mechanisms have been suspected from studies on spontaneous Ca²⁺ release from the SR ^7–9, but Chen and his colleagues demonstrated a drug that could block the process ¹⁹ and also pinpointed the molecular mechanism by which luminal Ca²⁺ triggers SOICR ¹⁰. Such spontaneous Ca²⁺ release may lead to propagating Ca²⁺ waves, essential for activation of eggs at fertilization, but potentially fatal in cardiac myocytes since they underlie arrhythmias ²⁰. Chen and his colleagues proposed that, uniquely amongst β-blockers, carvedilol suppresses SOICR itself, in addition to its β-blocking action ¹⁹. Luminal amino-acid residues of the RyR have been identified as critical for SOICR, and these include the E4872 residue, which is highly conserved not only in RyR between species but also in IP₃Rs ¹⁰. The proposal is that the RyR itself acts as a luminal Ca²⁺ sensor, rather than an accessory or interacting protein such as calsequestrin, as SOICR can persist in calsequestrin-null mice ²¹. The E4872Q mutation in a mouse model suppressed Ca²⁺ waves and ventricular tachyarrhythmias (VTs) ¹⁰.

It has been shown that, compared to 14 other known β-blockers, carvedilol was the only one effective in inhibiting SOICR ¹⁹. This unique inhibitory action of carvedilol on SOICR needs further investigation, as this activity would be contributing to its β-blocking action. Single-channel recordings in lipid bilayers revealed that carvedilol has an impact on the gating properties of RyR2, reducing the duration and increased the frequency of channel openings. In this study, we have utilized the sea urchin egg homogenate preparation ¹² which robustly express both IP₃R and RyR Ca²⁺ release mechanisms on microsomal vesicles ²². Furthermore, ER-derived microsomes in sea urchin egg homogenates display spontaneous Ca²⁺ release due to Ca²⁺ overload, which can be effected by both IP₃Rs and RyRs ⁹. This allowed us to examine the effect of carvedilol and analogs on SOICR mediated by the two principal ER-based Ca²⁺ release mechanisms. Although the interaction of IP₃ with IP₃Rs has now been elucidated at the molecular level ²³, the mechanism by which cADPR activates RyRs is not well understood, with accessory cADPR binding proteins proposed ^24–26. However, since cADPR is known to cause Ca²⁺ release by modulating RyRs ^{13, 14}, it allows us to probe the role of SOICR in the mechanism by which cADPR activates RyRs. We find that carvedilol suppresses cADPR-evoked Ca²⁺ release, and also attenuates IP₃-mediated Ca²⁺ release. A common mechanism by which carvedilol inhibits a conserved luminal Ca²⁺ gate for both Ca²⁺ release mechanisms is suggested.

Materials and Methods

Sea urchin egg homogenate preparation

Gamete shedding was stimulated by intracoelomic injections of 0.5 M KCl solution and sea urchin eggs from Lytechinus pictus were collected. Collection was carried out in artificial sea water (ASW). ASW: 435 mM NaCl, 10 mM KCl, 40 mM MgCl₂, 15mM MgSO₄, 11 mM CaCl₂, 2.5 mM NaHCO₃, 7 mM Tris base, 13 mM Tris-HCl, pH 8.0. Eggs were de-jellied by passage through 100 μM nylon mesh (Millipore), and then washed 4 times in Ca²⁺ -free ASW, with the first two washes containing 1 mM EGTA. Eggs were subsequently washed in intracellular-like medium, glucamine intracellular medium (GluIM). GluIM: 250 mM potassium gluconate, 250 mM N-methyl-D-glucamine, 20 mM Hepes (acid), and 1 mM MgCl₂, pH 7.2 (pH adjusted with glacial acetic acid). Eggs were homogenized with a glass Dounce tissue homogenizer in ice-cold GluIM supplemented with 2 mM MgATP, 20 U/mL creatine phosphokinase (CPK), 20 mM phosphocreatine (PCr), Complete ™EDTA-free Protease Inhibitor tablets were from Roche. Homogenate (50%, v/v) was then centrifuged at 13,000 x g at 4 °C for 10 s. The supernatant were aliquoted into 0.5 ml portions and stored at -80 °C.

Fluorimetry to measure Ca²⁺ release

To measure Ca²⁺ changes, egg homogenate was diluted gradually 20-fold over 3 h in GluIM at 17°C supplemented with the ATP regenerating system (to facilitate optimal Ca²⁺ loading of the stores), fluorescent dyes were added (Fluo-3, 3 μM final) and fluorescence changes monitored in a 600 μl aliquot (2.5% v/v) in a stirred micro-cuvette in a Perkin Elmer LS-50B fluorimeter.

Fluo-3 was excited at 506 ± 3 nm and 526 ± 2 nm emission recorded.

The Fluo-3 fluorescence signal was converted to absolute Ca²⁺ concentrations by sequential addition of 0.5 mM EGTA and 10 mM CaCl₂ to determine F_min and F_max, using the following equation:

$$\left[\text{C}{\text{a}}^{2+}\right]=\left[{\text{K}}_{\text{d}}\times \left(\text{F}-{\text{F}}_{\text{min}}\right)\right]/\left({\text{F}}_{\text{max}}-{\text{F}}_{\text{min}}\right)$$

where [Ca²⁺] is the extravesicular Ca²⁺ concentration, F is the fluorescence at any time, F_min and F_max are the fluorescence in the “absence” of Ca²⁺ and presence of saturating Ca²⁺, respectively. At the end of each experiment, F_min and F_max were empirically determined by the sequential addition of 500 μM EGTA and 10 mM Ca²⁺, respectively (6 μL of 50 mM EGTA; 6 μL of 1 M CaCl₂).

For TPEN studies of effects on the kinetics of Ca²⁺ release, fluorescence (in units (U)/s) was normalized to the resting fluorescence (F₀) to account for variability and therefore expressed as U.F₀/s.

Source of reagents

cADPR and IP₃ were purchased from Sigma-Aldrich and LC Laboratories, respectively. Fluo-3 (K⁺ salt) was purchased from Invitrogen and TPEN from Sigma-Aldrich. Carvedilol was purchased from Abcam Biochemicals. VK-II-86, the carvedilol analog, was synthesized as described previously¹⁹.

Statistical analysis

Two data sets were compared using Student’s t test and multiple groups using ANOVA and Dunnett’s multiple comparisons post test. Data were paired when needed and statistical significance assumed at P<0.05. Graphs were constructed and the following statistical conventions observed: P<0.05 (*), P<0.01 (**), P<0.001 (***), P<0.0001 (****) using Graphpad Prism. Representative fluorimetric traces were plotted as raw fluorescence (RFU) against time.

Results

To gain further insight into the mechanism of cADPR-induced Ca²⁺ release, we examined the effect of carvedilol on Ca²⁺ release mechanisms in in sea urchin egg homogenates. cADPR-induced Ca²⁺ release was monitored in the presence of vehicle DMSO or carvedilol. A sub-maximal concentration of cADPR (0.3 μM) for Ca²⁺ release was determined by establishing a concentration-response curve of cADPR (EC₅₀ ~ 0.25 μM; Fig. 2b). Preincubation with carvedilol for 2 minutes resulted in a concentration-dependent inhibition of cADPR-induced Ca²⁺ release with an approximate IC₅₀ of 30 μM (Fig. 1b).

Fig. 2. Carvedilol inhibits cADPR-induced Ca²⁺ release.

(a) Summarized representative fluorimetric traces. (b) DMSO or 30 μM of carvedilol were preincubated for 2 minutes before the addition of different cADPR concentrations. Statistical significance was determined using a Student’s t test comparing control (n = 9) and carvedilol (n = 9) (b). Data were fit as a sigmoidal concentration-response (b). Data are represented as mean ± SEM.

Fig. 1. Carvedilol suppresses cADPR-induced Ca²⁺ release.

(a) Different concentrations of carvedilol or DMSO were preincubated for 2 minutes with a subsequent addition of a sub-maximal concentration of cADPR (0.3μM) and the raw control response of cADPR was 72 ± 5 nM (n = 9). Summarized representative fluorimetric traces shown in (a) and data fit as a sigmoidal concentration-response shown in (b). Statistical significance was determined with a Dunnett’s test in comparison to control response without carvedilol (b). Data are represented as mean ± SEM.

Generating a cADPR concentration-response curve in the presence or absence of a sub-maximal concentration of carvedilol revealed that carvedilol reduced the maximal response to cADPR without substantially changing the EC₅₀ for cADPR (Fig. 2b), i.e. was likely a non-competitive blocker. The EC₅₀ of cADPR with DMSO (and 95% confidence interval) was 250 nM (153–391 nM) and in the presence of carvedilol it was 410 nM (297–564 nM) (P>0.05).

Next we examined the time course of inhibition by carvedilol. Carvedilol (30 μM) was preincubated with homogenate for 10, 30, 60 and 120 seconds. Our results show that carvedilol suppressed cADPR-induced Ca²⁺ release maximally within 10 seconds, and there was no further inhibition at longer incubations (Fig. 3).

Fig. 3. Carvedilol suppresses cADPR-induced Ca²⁺ release rapidly.

Effect of 30 μM carvedilol after preincubation with carvedilol for 10 s, 30 s, 60 s and 100 s is shown. Uncalibrated control ∆[ Ca²⁺ ] was 76 ±4 nM (n = 9). Statistical significance was determined using Dunnett’s test versus 0 μM of carvedilol. Data are represented as mean ±SEM.

In order to investigate whether the effect of carvedilol was unique to cADPR- induced Ca²⁺ release and the RyR, we examined the effects of carvedilol on IP₃-induced Ca²⁺ release. The proposed SOICR trigger of RyRs has been proposed to be conserved in IP₃Rs ¹⁰. Carvedilol also suppressed IP₃-induced Ca²⁺ release but the extent of inhibition of IP₃ was less than that for cADPR (Fig. 4).

Fig. 4. Carvedilol inhibits IP3-induced Ca²⁺ release.

(a) Different concentrations of carvedilol were preincubated for 2 minutes before the addition of sub-maximal concentrations of either IP₃ or cADPR (0.3 μM), which were determined by establishing dose-response curves. Raw control ∆[ Ca²⁺ ] with cADPR was 72 ±4 nM (n = 4–7) and with IP3 94 ±1 nM (n = 4–7).(b) Graph summarizing the results. Right Y axis applies for IP3, whereas left Y axis for cADPR. Statistical significance was determined using Student’s t test and data were fit as a sigmoidal concentration-response. Data are represented as mean ±SEM.

Caffeine is also a commonly used pharmacological RyR agonist ²⁷. Caffeine has also shown to release Ca²⁺ in sea urchin egg homogenates and eggs by activating RyRs ^{13, 28}. We compared the effects of carvedilol on cADPR- and caffeine-induced Ca²⁺ release. Different concentrations of carvedilol were preincubated for 2 minutes prior to the addition of 10 mM caffeine¹³ or 0.3 μM cADPR (Fig. 5a). Carvedilol suppressed caffeine-induced Ca²⁺ release with an IC₅₀ indistinguishable from to cADPR with IC₅₀s of 45 μM (for cADPR) and 51 μM for caffeine. The differences were not statistically significant.

Fig. 5. Carvedilol suppresses caffeine-induced Ca²⁺ release.

Different concentrations of carvedilol were preincubated for 2 minutes before the addition of sub-maximal concentrations of either caffeine (10 mM) or cADPR (0.3 μM). Raw control ∆[Ca²⁺ ] with caffeine was 80 ±5 nM (n = 4–7) and with cADPR 85 ±3 (n = 4–7). Statistical significance (P>0.05) was determined using Student’s t test and data were fit as a sigmoidal concentration-response. Summarized representative fluorimetric traces are shown in (a). Data are represented as mean ±SEM.

VK-II-86 is a carvedilol analog but with minimal β-blocking activity, demonstrating that the pharmacophore for SOICR inhibition can be separated from that for β-blockade ^{19, 29}. Like Carvedilol, VK-II-86 inhibited cADPR-induced Ca²⁺ release in a concentration-dependent manner (IC₅₀ of 52 μM; Fig. 6b), demonstrating that the β-blocking activity of carvedilol was dispensable for blocking Ca²⁺ release in homogenates.

Fig. 6. VK-II-86 suppresses cADPR-induced Ca²⁺ release.

(a) Different concentrations of VK-II-86 or DMSO were preincubated for 2 minutes with a subsequent addition of a sub-maximal concentration of cADPR (0.3 μM) and the raw control response of cADPR was 75 ±4 nM (n = 9). Summarized representative fluorimetric traces shown in (a) and data fit as a sigmoidal concentration-response shown in (b). Statistical significance was determined with a Dunnett’s test in comparison to control response without carvedilol (b). Data are represented as mean ±SEM.

To investigate the proposed luminal mechanism of action of carvedilol, the membrane-permeant low affinity Ca²⁺ chelator TPEN (N,N,N’,N’- tetrakis(2-pyridylmethyl)ethylenediamine) was employed to allow the manipulation of luminal Ca²⁺ as previously shown ³⁰. Different concentrations of TPEN (or ethanol vehicle) were preincubated with homogenates for 2 minutes prior to the addition of 0.3 μM cADPR (Fig. 7a). We found that TPEN inhibited cADPR-induced Ca²⁺ release, both in terms of the magnitude and kinetics of response but have a greater impact on kinetics (estimated carvedilol IC₅₀ of 9 μM and 663 μM for kinetics (Fig. 7c) and magnitude respectively (Fig. 7b). The inhibition by TPEN is consistent with a perturbation of luminal Ca²⁺ as previously observed for IP₃ ³⁰. However, the effect of TPEN and carvedilol are dissimilar, as the kinetic data may indicate an additional site of action for carvedilol, and which at high concentrations almost completely inhibits Ca²⁺ release in response to cADPR.

Fig. 7. TPEN inhibits cADPR-induced Ca²⁺ release.

(a) Different concentrations of TPEN (or vehicle ethanol) were preincubated with egg homogenate for 2 minutes prior to the addition of 0.3 μM cADPR.

(b) TPEN reduces the magnitude of cADPR-evoked Ca²⁺ release.

(c) TPEN affects the kinetics of cADPR-evoked Ca²⁺ release. Fluorescence (in units (U)/s) was normalized to the resting fluorescence (F₀) to account for variability and therefore expressed as U.F₀/s. Statistical significance was determined using Student’s t test and data were fit as a sigmoidal concentration-response. Data are represented as mean ±SEM.

Discussion

In the present study we have demonstrated that carvedilol inhibits cADPR-induced Ca²⁺ release in sea urchin egg homogenates. cADPR has been characterized as a second messenger regulating the sensitivity of RyRs to activation by Ca²⁺ in sea urchin eggs ^{13, 14}, and widely in plant and animal cells ². However, the mechanism of action of cADPR, from sea urchin egg to cardiac cells, remains undefined ³¹. A possible mechanism is for cADPR to activate the RyR indirectly by binding to a protein that is part of the RyR macromolecular complex or to a protein that either translocates itself to the RyR or influences a signalling pathway that regulates RyR activity ²⁶. Another possible mechanism is that cADPR stimulates an increase in the level of SR Ca²⁺, perhaps with the involvement of SERCA ³¹, resulting in a Ca²⁺ overload response. Due to the recent elucidation of the action of carvedilol actions on RyRs, we tested the effect of this drug on cADPR-induced Ca²⁺ release in sea urchin egg homogenates. We found that carvedilol was an effective inhibitor of cADPR-induced Ca²⁺ release. Carvedilol caused a substantial inhibition (82%) of cADPR-induced Ca²⁺ release (Fig. 1). This is unlikely to be due to blocking the cADPR binding site since carvedilol inhibition is likely non-competitive, and carvedilol also inhibits caffeine and also IP₃-induced Ca²⁺ release. It is known that although cADPR, caffeine and ryanodine can release Ca²⁺ by activating RyRs, a selective inhibitor of cADPR, 8-NH₂-cADPR does not block Ca²⁺ release by caffeine or ryanodine implying action at different but interacting sites ³². Therefore we propose also that carvedilol and cADPR act at distinct sites.

VK-II-86 also was found to suppress cADPR-induced Ca²⁺ release and this is consistent with the notion that the effect of carvedilol on cADPR-evoked Ca²⁺ release is independent of its β-blocking activity.

It has been proposed that carvedilol inhibits spontaneous Ca²⁺ release from the SR by a mechanism dependent on a luminal trigger site on RyRs. Amino acid residues important for such triggering have been identified and include a cluster of negatively charged residues at the C-terminal part of the predicted inner helix of the RyR ¹⁰. These include the residue E4872 conserved between RyRs, including that for sea urchins (Fig. 8). Importantly, acidic residues at the putative luminal sensor are also conserved in sea urchin IP₃Rs. We have shown for the first time that carvedilol also inhibits IP₃-induced Ca²⁺ release in urchin homogenates (Fig. 4), although to a lesser degree (47% versus >80%) than that for either cADPR or caffeine-induced Ca²⁺ release. Since IP₃-mediated Ca²⁺ signaling has also been implicated in arrhthymogenesis ³³, this may also be pertinent to the clinical effects of carvedilol. Our novel finding that carvedilol can suppress IP₃-induced Ca²⁺ release also suggests that carvedilol may be a potential therapeutic agent for treating other IP₃R-associated disorders in addition ³⁴.

Fig. 8. Key amino acid residues for the proposed luminal Ca²⁺ sensor of RyRs and IP₃Rs are conserved in sea urchins.

Alignment of amino acid sequences of the pore and inner helix region of RyR2 and IP3R2 from human (Hs) and sea urchin (Sp; Strongylocentrotus purpuratus). The acidic nature of residue E4872 (in orange) is conserved between RyR2 and IP3R2. Sequences (HsRyR2: Q92736; SpRyR2: XP_011670306; HsITPR2: Q14571; SpITPR2: XP_011682423) were aligned using Clustal Omega.

A luminal Ca²⁺ regulation site is consistent with the inhibition of a low affinity membrane-permeant Ca²⁺ chelator, TPEN. Not only do IP₃Rs and RyRs share the conserved acidic residues, but they also share sensitivity to TPEN ³⁰ (and this study). Moreover, since TPEN by modulating luminal Ca²⁺ similarly inhibits Ca²⁺ release by both IP₃ and cADPR in a similar way to carvedilol, a common mechanism based on a luminal mechanism is suggested. We previously have proposed that TPEN does not simply reduce the electrochemical gradient for Ca²⁺ because ionomycin-evoked Ca²⁺ release is relatively insensitive to TPEN ³⁰. Again, this implies a modulatory role of TPEN on Ca²⁺ release channels by lowering luminal Ca²⁺, whereas carvedilol has a similar effect by blocking the luminal Ca²⁺ sensor on Ca²⁺ release channels.