IP3 receptor-dependent Ca2+ release modulates excitation-contraction coupling in rabbit ventricular myocytes

Timothy L. Domeier, Aleksey V. Zima, Joshua T. Maxwell, Sabine Huke, Gregory A. Mignery, Lothar A. Blatter


Inositol 1,4,5-trisphosphate (IP3) receptor (IP3R)-dependent Ca2+ signaling exerts positive inotropic, but also arrhythmogenic, effects on excitation-contraction coupling (ECC) in the atrial myocardium. The role of IP3R-dependent sarcoplasmic reticulum (SR) Ca2+ release in ECC in the ventricular myocardium remains controversial. Here we investigated the role of this signaling pathway during ECC in isolated rabbit ventricular myocytes. Immunoblotting of proteins from ventricular myocytes showed expression of both type 2 and type 3 IP3R at levels ∼3.5-fold less than in atrial myocytes. In permeabilized myocytes, direct application of IP3 (10 μM) produced a transient 21% increase in the frequency of Ca2+ sparks (P < 0.05). This increase was accompanied by a 13% decrease in spark amplitude (P < 0.05) and a 7% decrease in SR Ca2+ load (P < 0.05) and was inhibited by IP3R antagonists 2-aminoethoxydiphenylborate (2-APB; 20 μM) and heparin (0.5 mg/ml). In intact myocytes endothelin-1 (100 nM) was used to stimulate IP3 production and caused a 38% (P < 0.05) increase in the amplitude of action potential-induced (0.5 Hz, field stimulation) Ca2+ transients. This effect was abolished by the IP3R antagonist 2-APB (2 μM) or by using adenoviral expression of an IP3 affinity trap that buffers cellular IP3. Together, these data suggest that in rabbit ventricular myocytes IP3R-dependent Ca2+ release has positive inotropic effects on ECC by facilitating Ca2+ release through ryanodine receptor clusters.

  • inositol 1,4,5-trisphosphate
  • Ca2+ sparks

cardiac excitation-contraction coupling (ECC) occurs via Ca2+-induced Ca2+ release (CICR), where Ca2+ influx through voltage-gated (L-type) Ca2+ channels triggers massive release of sarcoplasmic reticulum (SR) Ca2+ through ryanodine receptors (RyRs) (6). The amplification of CICR produces the global cytosolic Ca2+ signal that activates the contractile filaments and initiates myocyte contraction. While the RyR is clearly the primary intracellular Ca2+ release channel that mediates CICR during cardiac ECC, the inositol 1,4,5-trisphosphate (IP3) receptor (IP3R) Ca2+ release channel is also expressed in cardiac myocytes (21, 23), and recent evidence suggests that activation of IP3R may modulate ECC. In atrial myocytes, application of IP3 increases SR Ca2+ release events through IP3R (Ca2+ “puffs”) (45) and facilitates SR Ca2+ release through RyR clusters (Ca2+ “sparks”) (16, 19, 45). Furthermore, with the use of the Gq protein-linked agonist endothelin-1 (ET-1) to induce IP3 production (14, 30), IP3R activation was shown to elevate diastolic Ca2+ concentration ([Ca2+]), enhance the amplitude of the Ca2+ transient during ECC, and increase the propensity for arrhythmogenic spontaneous Ca2+ release events (16, 19, 45). Notably, these effects are absent in atrial myocytes isolated from IP3R type 2-knockout mice (16), confirming observations made with pharmacological inhibition of the IP3R.

While IP3R-dependent Ca2+ release is well established in the atrial myocardium, its presence and significance in the ventricular myocardium remain controversial. Ventricular myocytes express IP3R, but at lower levels than atrial myocytes (17). IP3R have also been shown to colocalize with calmodulin-dependent protein kinase II at the nuclear envelope (3) and mediate excitation-transcription coupling (40). Although most evidence for IP3R-dependent Ca2+ release in ventricle is negative, recent investigations have suggested that this signaling pathway may exert positive inotropic (36) and arrhythmogenic (27) effects during ventricular ECC. Here we address this controversy, using ventricular myocytes isolated from rabbit hearts and test the hypothesis that IP3R-dependent Ca2+ signaling facilitates SR Ca2+ release during ECC. We find that both type 2 and type 3 IP3R (IP3R-2, IP3R-3) are expressed in ventricular myocytes. In these cells, activation of IP3R facilitates SR Ca2+ release through RyR release clusters and enhances the Ca2+ transient during ECC. Therefore, the IP3 signaling pathway may modulate ventricular ECC under conditions in which IP3/IP3R signaling is altered, such as during neurohormone release, ischemia-reperfusion (2), or heart failure (1).

A previous account of this work has been presented in abstract form (8).


Solutions and chemicals.

Normal Tyrode solution contained (in mM) 135 NaCl, 4 KCl, 2 CaCl2, 1 MgCl2, 10 d-glucose, and 10 HEPES; pH 7.4. The internal cell solution for permeabilized myocytes contained (in mM) 100 K+ aspartate, 15 KCl, 5 KH2PO4, 5 MgATP, 0.35 EGTA, 0.12 CaCl2, 0.75 MgCl2, 10 phosphocreatine, 10 HEPES, and 0.03 fluo-4 pentapotassium salt (Molecular Probes-Invitrogen, Carlsbad, CA), with 5 U/ml creatine phosphokinase and 8% dextran (Mr: 40, 000); pH 7.2 with KOH. Free [Ca2+] and [Mg2+] of this solution were 150 nM and 1 mM, respectively. In some experiments, the CaCl2 content of the internal solution was adjusted to achieve a final free [Ca2+] of 100 nM. Unless noted otherwise, all chemicals were purchased from Sigma-Aldrich (St. Louis, MO).

Myocyte isolation.

Rabbit ventricular myocytes were isolated as previously described (4). Briefly, rabbits were anesthetized with pentobarbital sodium (50 mg/kg), and hearts were excised, mounted on a Langendorff apparatus, and retrogradely perfused with nominally Ca2+-free Tyrode solution (5 min) followed by Ca2+-free Tyrode solution containing collagenase (10–20 min) at 37°C. The left ventricular free wall was sliced and further digested in collagenase solution (5–15 min) at 37°C. Digested tissue was minced, filtered, and washed. Isolated cells were kept at room temperature (22–24°C) until experimental procedures. All protocols were approved by the Animal Care and Use Committee of Loyola University Medical Center.

Western blot analysis of IP3R and RyR.

Atrial and ventricular rabbit myocytes were suspended in Ca2+-free Tyrode solution immediately after isolation. An SDS-protease inhibitor mix (P-8340, Sigma) was added (final SDS concentration ∼2%), and the cells were frozen in liquid nitrogen. Protein was quantified with a modified Lowry assay (Pierce, Rockford, IL), and 20–70 μg of protein per lane was subjected to 6% SDS-PAGE and transferred to nitrocellulose membranes overnight. Primary antibodies used were anti-RyR (MA3-925; Affinity BioReagents, Golden, CO) and custom antibodies specific for IP3R-2 (T2C) (28) and IP3R-3 (T3C). Both IP3R antibodies are directed to the COOH terminus of the receptor (29). Protein bands were visualized with the SuperSignal West Dura Kit (Pierce), and signals were quantified with the UVP EpiChemi3 imaging system and LabWorks 4.6 software. Some blots were stripped with RESTORE (Pierce) and reused after confirmation of primary antibody removal. The type 3 antibody T3C does not cross-react with the rat IP3R-2, as judged by Western blot analysis of sf9 insect cells expressing the IP3R-2 baculoviral construct (data not shown).

Construction and adenoviral infection of IP3 affinity trap.

The IP3 affinity trap consists of the ligand binding domain of the rat type 1 IP3R. PCR was used with the following primers, sense CGGGTCGACTCGAGCCACCATGTCTGACAAAATGTCTAGTTTCC and antisense GCCGTCGACTCGAGCTACTTTCGGTTGTTGTGG, to amplify the 589 amino-terminal residues encompassing the ligand binding region of the rat type 1 IP3R (IP3R-1) (20). PCR products were digested with SalI and ligated into similarly digested pCMV-5, and expression was verified in COS-1 cells by transient transfection and Western blotting with a IP3R-1-specific amino-terminal antibody (T1NH) (data not shown). The insert harboring the IP3 binding region was excised with XhoI and ligated into XhoI-digested pShuttle-CMV (Stratagene, La Jolla, CA). The adenovirus was created with a commercially available kit, the AdEasyTM XL adenoviral vector system. The bacterial cell line BJ5183-AD-1, pretransformed with the plasmid pAdEasy-1, was used for in vivo homologous recombination with pShuttle-CMV-Affinity Trap. The pAdEasy-1-Affinity Trap insert-containing plasmid was transformed into DH5α and produced in bulk. Purified AdEasy-1-Affinity Trap plasmid was used to transfect AD-293 cells for virus amplification. Affinity TrapAdV virus was plaque-purified, amplified, CsCl gradient-purified, and stored at −80°C. Freshly isolated rabbit ventricular myocytes were infected (multiplicity of infection: 1) with IP3AT-AdV or the β-galactosidase-AdV (control) for 2 h, followed by 24-h short-term culture at 37°C.

Confocal microscopy of SR Ca2+ release events.

Elementary SR Ca2+ release events (Ca2+ sparks) were studied in permeabilized myocytes to allow for direct application of membrane-impermeant reagents. Myocytes were attached to laminin-coated coverslips and permeabilized with 0.005% saponin for 30 s, followed by application of the internal cell solution. Fluo-4 was excited by the 488-nm line of the argon ion laser, and emission was collected at >500 nm. Confocal (Radiance 2000/MP, Bio-Rad) images were acquired in the line scan mode at 3 ms per scan with a pixel size of 0.12 μm. The scan line was positioned in the cytosol along the longitudinal axis of the cell. Fluo-4 fluorescence emission (F) was normalized to baseline fluorescence emission (F0), and changes of intracellular [Ca2+] ([Ca2+]i) are presented as changes of ΔF/F0 (where ΔF = F − F0). All experiments were performed at room temperature. Ca2+ spark frequency {quantified as number of sparks per second and 100 μm of scanned distance [sparks·(100 μm)−1·s−1]} amplitude, spatial width (full-width half-maximum), and duration (full-duration half-maximum) were calculated with the SparkMaster algorithm (25). SR Ca2+ load was assessed with rapid application of 10 mM caffeine and was quantified as the amplitude (ΔF/F0) of the caffeine-induced Ca2+ transient.

Calcium transient measurements.

Myocytes were attached to laminin-coated coverslips, loaded with 5 μM indo-1 AM in the presence of 0.05% Pluronic F-127 for 25 (freshly isolated cells) or 35 (cultured cells) min, washed for 20 min in Tyrode solution to allow for deesterification of the dye, and then superfused continuously (1 ml/min) with Tyrode solution. Action potentials were triggered at 0.5 Hz by electrical field stimulation with a pair of platinum electrodes. The electrical stimulus was set at a voltage ∼50% greater than threshold to induce myocyte contraction. Indo-1 was excited at 360 nm, and emission signals at 410 nm (F410) and 485 nm (F485) were simultaneously sampled at 100 Hz (pCLAMP10; Axon Instruments/Molecular Devices, Sunnyvale, CA). Changes in [Ca2+]i are expressed as changes of the ratio R = F410/F485. Background fluorescence at each wavelength was subtracted before calculation of F410/F485 (R). Ca2+ transients were sampled during periods of 20-s illumination every 5 min, except where noted otherwise, to minimize photobleaching. The background-corrected indo-1 ratio signal (R) was stable throughout experimental protocols, and no change in diastolic [Ca2+] ([Ca2+]diast) or Ca2+ transient amplitude was observed during control pacing. [Ca2+]diast was determined as minimum R signal before an action potential-induced Ca2+ transient. Ca2+ transient amplitude was defined as ΔR = Rpeak − Rdiast. Only healthy myocytes exhibiting robust control Ca2+ transients (freshly isolated myocytes: ΔR = 0.3–1; cultured myocytes: ΔR = 0.15–0.6) were used for analysis. To assess SR Ca2+ load, myocyte pacing was ceased and the perfusion solution was rapidly switched to Tyrode solution containing 10 mM caffeine to empty SR Ca2+ stores. The amplitude of the caffeine-induced Ca2+ transient was used as an index of SR load. All experiments were performed at room temperature.

Recordings of RyR channel currents.

SR vesicles isolated from cat ventricle were incorporated into planar lipid bilayers, and RyR channel activity was measured as described previously (46). For recordings of Cs+ currents through RyRs, the cis (cytosolic)- and trans (luminal)-chambers contained (mM) 400 CsCH3SO3, 0.1 CaCl2, and 20 HEPES; pH 7.3. Free [Ca2+] in the cis-chamber was adjusted to 3 μM by addition of an appropriate amount of EGTA. IP3 (10 μM) was added to the cis-chamber. Single-channel currents were recorded with an Axopatch 200B amplifier (Axon Instruments/Molecular Devices) at a holding potential of −20 mV. Currents were filtered at 1 kHz and sampled at 5 kHz. The mean open probability (Po) of the channels was used as an index of activity. Po values were calculated from the 50% threshold analysis with pCLAMP software (Axon Instruments/Molecular Devices).

Data analysis and statistics.

Statistical comparisons were made with paired or unpaired Student's t-tests. Data are presented as individual observations or as means ± SE obtained from n individual cells.


In this study we tested the hypothesis that IP3R-dependent Ca2+ signaling facilitates SR Ca2+ release during ECC in rabbit ventricular myocytes. Previous investigations have reported that cardiac myocytes express predominantly IP3R-2, and to a lesser extent IP3R-3 (23). Therefore, total protein was isolated from rabbit myocytes and immunoblotted with antibodies against these receptors (Fig. 1A). Western blot analysis showed that atrial and ventricular myocytes express both IP3R-2 and IP3R-3, with expression of each type being three- to fourfold higher in atrial myocytes (Fig. 1B). Expression of the RyR was similar between the two tissues.

Fig. 1.

Ryanodine receptor (RyR) and inositol 1,4,5-trisphosphate (IP3) receptor (IP3R) expression in atrial and ventricular myocytes. A: immunoblots of total protein isolated from atrial (At) and ventricular (Ve) myocytes of the same rabbit using antibodies for RyR, type 2 IP3R (IP3R-2), and type 3 IP3R (IP3R-3). B: summary data showing relative expression of IP3R-2 and IP3R-3 between atrial (open bars) and ventricular (gray bars) myocytes (n = 4). Data in B are normalized to expression in ventricular myocytes and are shown as means ± SE. *Significant difference from ventricular myocytes, P < 0.05

We used saponin-permeabilized ventricular myocytes and the Ca2+ indicator fluo-4 to study the effects of IP3R activation on cytosolic Ca2+ sparks, the elementary SR Ca2+ release events through RyR release clusters. Under control conditions spontaneous Ca2+ sparks occurred at a frequency of 17.1 ± 2.0 sparks·(100 μm)−1·s−1 (n = 9). Direct application of IP3 (10 μM) produced an increase in Ca2+ spark frequency (Fig. 2, A and B, and Fig. 3C; 121 ± 4.0% of control, n = 9, P < 0.05) that returned to baseline values after 5–10 min (Fig. 2, A and B). Application of IP3 also produced a gradual decrease in spark amplitude (86.9 ± 3.9% of control after 7 min, n = 9; P < 0.05) without affecting either duration or spatial width of release events. The increase in Ca2+ spark frequency caused a reduction of SR Ca2+ content, which was experimentally observed as a decrease in the amplitude of the caffeine-induced Ca2+ transient following 5 min of IP3 application (Fig. 2C; 92.8 ± 2.0% of control, n = 5, P < 0.05). Thus initial IP3 application increases Ca2+ spark frequency (Fig. 2A, 1 min), which eventually leads to a decrease in SR Ca2+ load (Fig. 2C). This decrease in SR Ca2+ load then feeds back on SR Ca2+ release and decreases Ca2+ spark frequency (Fig. 2, A and B, 5 min) and amplitude (Fig. 2B).

Fig. 2.

IP3 increases the frequency of Ca2+ sparks in ventricular myocytes. A: confocal line scan images of saponin-permeabilized myocytes with fluo-4 showing cytosolic Ca2+ sparks under control conditions (left) and 1 min (center) and 5 min (right) after addition of 10 μM IP3. Fluorescence [change in fluorescence emission from baseline (ΔF/F0)] profiles of Ca2+ sparks are presented at bottom and correspond to regions denoted by black bars at left. B: summary data of effects of IP3 addition on Ca2+ spark frequency and amplitude. Data are normalized to mean of values preceding IP3 addition (t = −4 min to t = 0). C: example traces of SR Ca2+ load as assessed by rapid application of 10 mM caffeine under control conditions (left) and 5 min after addition of 10 μM IP3 (right). IP3 application decreased the amplitude of the caffeine Ca2+ transient to 92.8 ± 2.0% of control (n = 5; P < 0.05). *Significant difference from mean value before IP3 addition, P < 0.05.

Fig. 3.

IP3R inhibition prevents effects of IP3 on Ca2+ sparks. Confocal line scan images showing Ca2+ sparks under control conditions (A and B, left), after addition of IP3R inhibitors 2-aminoethoxydiphenylborate (2-APB, 20 μM; A, center) or heparin (0.5 mg/ml; B, center), and after subsequent addition of 10 μM IP3 (A and B, right). C: summary data showing maximum effects of IP3 (n = 9), 2-APB + IP3 (n = 4), and heparin + IP3 (n = 3) on Ca2+ spark frequency. 2-APB + IP3 and heparin + IP3 data were normalized to spark frequency in the presence of the respective blockers alone (100%). *Significant difference from control (100%), P < 0.05.

In a previous investigation of cat atrial myocytes, the combined application of IP3 and tetracaine (used to eliminate RyR-dependent Ca2+ release) revealed IP3R-mediated elementary Ca2+ release events in the form of Ca2+ puffs (45). When 10 μM IP3 was applied with 1 mM tetracaine to rabbit ventricular myocytes, neither Ca2+ sparks nor Ca2+ puffs were detected (data not shown), suggesting that any IP3R-mediated Ca2+ release in these cells is below the detection threshold of the confocal microscopy used here.

Because we were unable to directly detect IP3R-mediated Ca2+ release, we used pharmacological blockers to test whether the effects of IP3 on RyR-mediated Ca2+ release (Ca2+ sparks) were dependent on IP3R activation. In the presence of the IP3R inhibitor 2-aminoethoxydiphenylborate (2-APB, 20 μM; Fig. 3, A and C) the effect of IP3 on Ca2+ spark frequency was almost completely inhibited, whereas heparin (0.5 mg/ml; Fig. 3, B and C) partially prevented the increase in Ca2+ spark frequency. Both inhibitors alone caused a small and statistically nonsignificant increase in spark frequency [2-APB: 7 ± 12% (n = 4); heparin: 4 ± 2.0% (n = 3)].

To test whether IP3 could affect RyR properties directly, we exposed RyRs incorporated into lipid bilayers to IP3. As shown in Fig. 4, P o of single RyR channels was unchanged by 10 μM IP3 added to the cis-chamber. In four experiments, Po of RyRs was 0.211 ± 0.035 under control conditions and 0.195 ± 0.040 in the presence of IP3. IP3 did not affect the mean open time or single-channel current amplitude of RyR channels in these experiments (Fig. 4).

Fig. 4.

IP3 does not affect RyRs incorporated into lipid bilayers. A: representative RyR current recordings under control conditions (left) and after addition of IP3 (10 μM) to the cis side of the chamber (right). Recordings were made at a holding potential of −20 mV with cis Ca2+ concentration ([Ca2+]) = 3 μM; c = closed, o = open state of the channel. B: amplitude histogram for the channel shown in A. C: average RyR open probability (Po) under control conditions and in the presence of 10 μM IP3 (n = 4 channels).

Experiments on permeabilized myocytes were performed with an internal solution with a [Ca2+] of 150 nM, which produced a baseline spark frequency of 17.1 ± 2.0 sparks·(100 μm)−1·s−1 (n = 9), as mentioned above. When [Ca2+] was reduced to 100 nM, basal Ca2+ spark frequency decreased to 7.2 ± 1.0 sparks·(100 μm)−1·s−1 (n = 4). Under these conditions, exposure to IP3 caused a similar increase in Ca2+ spark frequency (to 124 ± 3% of control), suggesting that the effects of IP3 are independent of baseline spark frequency.

To determine the role of IP3R-dependent signaling during ECC in intact myocytes we used the Gq protein-linked agonist ET-1 to induce IP3 production. Consistent with results from rat neonatal and adult cat ventricular myocytes (30), resting rabbit ventricular myocytes expressing the FIRE-1 IP3 sensor showed an increase in fluorescence resonance energy transfer (FRET) after application of 100 nM ET-1, indicative of an increase in [IP3] (n = 3; data not shown). We loaded myocytes with indo-1 and monitored Ca2+ transients during electrical pacing (0.5 Hz, field stimulation). During control pacing, cells displayed stable diastolic [Ca2+] and Ca2+ transient amplitude (20-min pacing: [Ca2+]diast = 99 ± 0.6% of t = 0; Ca2+ transient amplitude: 106 ± 6.2% of t = 0; n = 7). The application of 100 nM ET-1 caused a brief decrease in Ca2+ transient amplitude (75 ± 8.9% of control after 2.5 min, n = 5; data not shown) followed by an increase that reached maximum (138 ± 11.6% of t = 0; n = 13; P < 0.05) after 15 min of ET-1 exposure (Fig. 5A). In contrast to results from atrial myocytes (16, 19, 45), application of 100 nM ET-1 to ventricular myocytes did not result in an increase in [Ca2+]diast (15 min: R = 95 ± 0.7% of control, n = 13). During 20 min of pacing in the presence of ET-1, spontaneous Ca2+ transients or other arrhythmogenic events were typically not observed. To determine whether the ET-1-induced increase in Ca2+ transient amplitude was mediated by IP3R we pretreated myocytes with the IP3R antagonist 2-APB. Application of ET-1 (100 nM) in the presence of 2-APB (2 μM) caused an initial decrease in Ca2+ transient amplitude that was identical to that observed under control conditions (73 ± 3.5% of control after 2.5 min, n = 5; data not shown). However, the Ca2+ transient then remained depressed below baseline for the remainder of the recording period (Fig. 5, B and C). Application of 2-APB alone caused an increase in Ca2+ transient amplitude (127 ± 7.7% of control after 15 min, n = 9), presumably because of nonspecific effects of the compound.

Fig. 5.

IP3R inhibition prevents endothelin-1 (ET-1)-induced increase in Ca2+ transient amplitude. A and B: example indo-1 Ca2+ transients of electrically paced (0.5 Hz) ventricular myocytes before (t = 0) and after the addition of 100 nM ET-1 in the absence (A) and presence (B) of the IP3R inhibitor 2-APB (2 μM). Ca2+ transients were normalized (R/R0) to average diastolic ratio F410/F485 (R) at t = 0 (R0). [Ca2+]i, intracellular [Ca2+]. C: summary data of experiments shown in A and B; Ca2+ transient amplitude is normalized to value before ET-1 addition (t = 0). ET-1: n = 13, 2-APB + ET-1: n = 6. *Significant difference between ET-1 and 2-APB + ET-1, P < 0.05.

Changes in SR Ca2+ load can directly affect Ca2+ transient amplitude. We therefore assessed SR Ca2+ load and found the amplitude of the caffeine-induced Ca2+ transient to be similar before and 15 min after ET-1 addition (Fig. 6, A and B), as well as before and after the combined application of 2-APB and ET-1 (100 ± 1.0% of control, n = 3). Thus it appears that the effects of ET-1 were to increase fractional SR release (i.e., amplitude of the Ca2+ transient normalized to SR content; Fig. 6C) via IP3R activation, independent of total SR Ca2+ content.

Fig. 6.

Sarcoplasmic reticulum (SR) Ca2+ load is unaffected by ET-1. A: example traces of twitch Ca2+ transients followed by measurement of SR Ca2+ load as assessed by rapid application of 10 mM caffeine to the same cell under control conditions (left) and 15 min after addition of 100 nM ET-1 (right). Ca2+ transients were normalized (R/R0) to average diastolic R at t = 0 (R0). B and C: summary data showing effect of ET-1 on caffeine-induced Ca2+ transient amplitude (B) and fractional SR Ca2+ release (C) (n = 6). *Significant difference from control, P < 0.05.

2-APB and other IP3R antagonists associate with nonspecific effects that complicate interpretation of experimental results (26, 31). Therefore, we employed an additional, alternative approach to manipulate IP3-dependent signaling utilizing expression of an IP3 affinity trap that binds to and buffers intracellular IP3. Cellular function of the IP3 affinity trap was initially characterized in COS-1 cells, which exhibit robust IP3-dependent signaling in response to agonists such as ATP (30). In these cells, expression of the IP3 affinity trap eliminated (100 nM ATP) or blunted (1 μM ATP) the ATP-induced Ca2+ transients that were observed in mock-transfected control cells (data not shown; n = 5). Freshly isolated rabbit ventricular myocytes were infected with an adenovirus encoding either the IP3 affinity trap or the β-galactosidase (β-Gal) reporter control and short-term cultured for 24 h. IP3 affinity trap- and β-Gal-infected myocytes were then loaded with indo-1 and paced at 0.5 Hz. Both groups of cells displayed similar [Ca2+]diast (IP3 affinity trap: R = 1.18 ± 0.02; β-Gal: R = 1.19 ± 0.03; n = 12), Ca2+ transient amplitude (IP3 affinity trap: ΔR = 0.39 ± 0.04; β-Gal: ΔR = 0.34 ± 0.03; n = 12), and SR Ca2+ load (caffeine-induced Ca2+ transient amplitude; IP3 affinity trap: ΔR = 1.08 ± 0.09; β-Gal: ΔR = 0.84 ± 0.07; n = 7). Upon application of 100 nM ET-1, β-Gal-infected myocytes responded similarly to freshly isolated control myocytes, with an initial decrease in Ca2+ transient amplitude followed by an increase that reached a maximum at 15 min of ET-1 exposure (Fig. 7, A and C). However, similar to results seen with pharmacological inhibition of the IP3R (Fig. 5), application of 100 nM ET-1 to IP3 affinity trap-infected myocytes resulted in a decrease in Ca2+ transient amplitude that did not return to baseline (Fig. 7, B and C). To ensure that IP3 affinity trap-expressing cells were able to respond to positive inotropic agonists with an increase in Ca2+ transient amplitude, the β-adrenergic agonist isoproterenol was applied (500 nM, 2 min). This agonist, which exerts its effects independent of the IP3 signaling pathway (6), produced a similar, robust increase in Ca2+ transient amplitude in both IP3 affinity trap- and β-Gal-infected cells [IP3 affinity trap: ΔR = 223 ± 9.4% of control (n = 5); β-Gal; ΔR = 211 ± 25.5% of control (n = 7)]. In summary, rabbit ventricular myocytes express IP3R that can contribute to positive inotropic effects on ECC during stimulation of specific neurohumoral pathways.

Fig. 7.

Expression of the IP3 affinity trap prevents ET-1-induced increase in Ca2+ transient amplitude. Example indo-1 Ca2+ transients of electrically paced (0.5 Hz) ventricular myocytes infected with the β-galactosidase (β-Gal) reporter control (Control, A) or the IP3 affinity trap (B) before (t = 0) and after addition of 100 nM ET-1. Ca2+ transients were normalized (R/R0) to average diastolic R at t = 0 (R0). C: summary data of experiments shown in A and B; Ca2+ transient amplitude is normalized to value before ET-1 addition (t = 0). Control (β-Gal): n = 12, IP3 affinity trap: n = 12. *Significant difference from control, P < 0.05.


The IP3/IP3R signaling pathway is integral to intracellular Ca2+ release in numerous cell types and tissues (5). In cardiac myocytes, however, the role of this signaling pathway is much less defined, particularly because Ca2+ release via the more abundantly (∼100-fold) expressed RyRs overwhelms that released by IP3Rs (6). Nevertheless, IP3R-dependent signaling is purported to influence many cardiac processes, including gene transcription (40) and ECC, and seems to play a functional role in certain cardiac pathologies (39). Positive inotropic and arrhythmogenic effects on ECC are frequently observed in atrial myocytes (19, 45), which express IP3R at greater levels than ventricular myocytes (17). Here, Western blot analysis of IP3R-2 and IP3R-3 protein showed that in rabbit myocardium expression of these receptors is ∼3.5-fold greater in atrial than ventricular myocytes, a difference similar to but not as substantial as that observed for IP3R-2 in rat (∼6-fold) (17). Similar to effects observed in atrial myocytes (45), application of IP3 to permeabilized ventricular myocytes produced an increase in cytosolic Ca2+ spark frequency and an associated decrease in SR Ca2+ content (Fig. 2). This was a consequence of IP3R-dependent facilitation of Ca2+ release from clusters of RyRs, because the increase in spark frequency was prevented by inhibitors of IP3R (Fig. 3) and IP3 did not directly affect properties of RyR incorporated into lipid bilayers (Fig. 4). In ventricular myocytes we were unable to directly resolve IP3R-dependent Ca2+ release (Ca2+ puffs). Ca2+ puffs reflect the summation of many individual IP3R release events (or Ca2+ “blips”) (22), and in atrial myocytes Ca2+ puffs are ∼80% smaller in amplitude than Ca2+ sparks (45), i.e., near the signal-to-noise detection threshold of confocal microscopy. Therefore, considering the difference in IP3R expression between atrial and ventricular myocytes, it is not entirely surprising that Ca2+ puffs were not observed here.

To study the role of the IP3/IP3R signaling pathway during ECC in intact ventricular myocytes we applied 100 nM ET-1, which has been shown to induce near-maximal IP3 production (14) and produce IP3-dependent effects on ECC in atrial myocytes (16, 19, 45). As reported in previous investigations (7, 32), ET-1 produced biphasic negative and positive inotropic effects on the amplitude of the Ca2+ transient, with brief initial inhibition followed by a secondary increase that reached maximum after 15 min. This positive inotropic effect was a consequence of augmented fractional SR Ca2+ release and was dependent on IP3/IP3R signaling, because it was prevented by pharmacological inhibition of the IP3R (Fig. 5) as well as by buffering cellular IP3 (Fig. 7). The mechanism(s) underlying ET-1-induced positive inotropy in ventricular myocytes is controversial, with numerous cellular targets being implicated in the response, including L-type Ca2+ channels (11, 15, 33, 37), K+ channels (12), as well as Na+/Ca2+ (42, 43) and Na+/H+ (13) exchange. However, these results were observed in response to concentrations of ET-1 (0.1–20 nM) that may have been insufficient to induce maximal IP3 production, and have therefore been suggested to be dependent on the parallel diacylglycerol-PKC signaling cascade (11, 24, 37, 43). To address this issue we tested the effect of 100 nM ET-1 in the presence of the PKC inhibitor bisindolylmaleimide VII (100 nM, 20 min). These experiments revealed that during PKC inhibition ET-1 increased the amplitude of the Ca2+ transient during ECC (151 ± 25% of control, n = 6), suggesting that under our experimental conditions the observed changes in Ca2+ signaling were most likely due to IP3 signaling independent of PKC.

It is now established that Ca2+ signaling in cardiac myocytes is composed not only of global Ca2+ signals (e.g., the Ca2+ transient that activates the contractile filaments during ECC) but also of spatially restricted Ca2+ signals within signaling microdomains. Accordingly, while IP3R-dependent Ca2+ release may not represent a primary mechanism of global Ca2+ release in cardiac myocytes, it may regulate a variety of cellular processes by altering microdomain [Ca2+]. Po of the cardiac RyR is regulated by cytosolic [Ca2+] (41). Thus a rather straightforward mechanism by which IP3R-dependent Ca2+ release may modulate ECC is to increase microdomain Ca2+ in the vicinity of RyR release clusters, sensitize RyRs to CICR, and increase the efficiency of ECC. This appears to underlie the effects of IP3R-dependent Ca2+ release in atrial myocytes, where IP3R and RyR colocalize near subsarcolemmal SR Ca2+ release sites (19). In these cells, IP3R-dependent Ca2+ release increases RyR Ca2+ spark frequency, aids propagation of CICR from the cell periphery to the cell center, and increases the global Ca2+ transient during ECC (16, 19, 45). However, we cannot rule out more complex associations between IP3R and RyR, such as interactions that may occur in macromolecular signaling complexes and phosphorylation events that occur downstream of IP3R activation (1, 3, 34). In ventricular myocytes, increasing evidence suggests nuclear localization and function of IP3R (3, 18, 40, 44). However, our data indicate that IP3R activation increases the frequency of cytosolic Ca2+ sparks (Figs. 2 and 3), suggesting that IP3Rs also localize near RyRs in the SR, where they facilitate SR Ca2+ release during ECC (Figs. 5 and 7). Interestingly, there appear to be species differences regarding the role of IP3R-dependent Ca2+ release during ECC. Recent evidence from rat ventricular myocytes showed that ET-1- and IP3R-dependent Ca2+ signaling produced positive inotropic and arrhythmogenic effects (27). In contrast, in this study ET-1 caused only an increase in Ca2+ transient amplitude in rabbit ventricular myocytes without arrhythmogenic Ca2+ release, whereas in cat ventricular myocytes IP3-dependent effects were entirely absent (45). These differences are likely due to the diversity in cellular Ca2+ handling among these species (summarized in Ref. 6). During a twitch >90% of the Ca2+ entering the cytoplasm is provided by SR Ca2+ release in rat ventricular myocytes, whereas this fraction is ∼70% in rabbit and only ∼50% in cat ventricular myocytes. These differences in SR Ca2+ dependence correlate with the observed cellular effects of IP3-dependent Ca2+ signaling during ECC. Rat ventricular myocytes, which depend almost entirely on SR Ca2+ release and operate near Ca2+ overload even under physiological conditions, appear to be most susceptible to arrhythmogenic Ca2+ signals. Consistent with this notion, IP3R-dependent Ca2+ signals have been shown to elicit arrhythmogenic Ca2+ release in this species (27). Rabbit myocytes, which represent a species with an intermediate dependence on SR Ca2+, show positive inotropic effects of IP3R-dependent Ca2+ release, but proarrhythmic effects are absent. Finally, in cat ventricular myocytes, which are the least dependent on SR Ca2+ release during a twitch among the species compared, no IP3R-dependent Ca2+ release could be detected. This is also consistent with effects on Ca2+ sparks in response to IP3 application. Spark frequency increased in rabbit ventricular myocytes, but not in cat ventricular myocytes (45).

Alterations in cardiac IP3/IP3R-dependent signaling have been reported under many pathological conditions, such as after excessive neurohumoral agonist release, in obesity (9), during ischemia-reperfusion (2), and in heart failure (1, 10, 35). Indeed, the pathology associated with several of these conditions is purported to be arrhythmogenesis mediated by IP3/IP3R-dependent Ca2+ signaling (19, 27, 38, 45). While we did not observe proarrhythmogenic events during activation of IP3R-dependent signaling in this investigation, we do show that this signaling pathway modulates ventricular SR Ca2+ release during ECC, and may thus serve as an important adaptive mechanism under a variety of physiological and pathological conditions.


This work was supported by National Institutes of Health Grants HL-80101 (to L. A. Blatter and G. A. Mignery), MH-53367 (to G. A. Mignery), HL-62231 (to L. A. Blatter), T32-HL-07692 (to T. L. Domeier), and F32-HL-090211 (to T. L. Domeier) and American Heart Association Grants AHA0550170Z (to L. A. Blatter), AHA0530309Z (to A. V. Zima), and AHA0725724Z (to T. L. Domeier).


The authors thank Junaid Ahsan and Karl Hench for myocyte isolation and Drs. Julie Bossuyt and Andreas Rinne for assistance with adenoviral infections.


  • 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.


View Abstract