Inositol-1,4,5-trisphosphate-dependent Ca²⁺ signalling in cat atrial excitation–contraction coupling and arrhythmias

Abstract

Inositol-1,4,5-trisphosphate (IP₃)-dependent Ca²⁺ release represents the major Ca²⁺ mobilizing pathway responsible for diverse functions in non-excitable cells. In the heart, however, its role is largely unknown or controversial. In intact cat atrial myocytes, endothelin (ET-1) increased basal [Ca²⁺]_i levels, enhanced action potential-evoked [Ca²⁺]_i transients, caused [Ca²⁺]_i transients with alternating amplitudes (Ca²⁺ alternans), and facilitated spontaneous Ca²⁺ release from the sarcoplasmic reticulum (SR) in the form of Ca²⁺ sparks and arrhythmogenic Ca²⁺ waves. These effects were prevented by the IP₃ receptor (IP₃R) blocker aminoethoxydiphenyl borate (2-APB), suggesting the involvement of IP₃-dependent SR Ca²⁺ release. In saponin-permeabilized myocytes IP₃ and the more potent IP₃R agonist adenophostin increased basal [Ca²⁺]_i and the frequency of spontaneous Ca²⁺ sparks. In the presence of tetracaine to eliminate Ca²⁺ release from ryanodine receptor (RyR) SR Ca²⁺ release channels, IP₃ and adenophostin triggered unique elementary, non-propagating IP₃R-dependent Ca²⁺ release events with amplitudes and kinetics that were distinctly different from classical RyR-dependent Ca²⁺ sparks. The effects of IP₃ and adenophostin were prevented by heparin and 2-APB. The data suggest that IP₃-dependent Ca²⁺ release increases [Ca²⁺]_i in the vicinity of RyRs and thus facilitates Ca²⁺-induced Ca²⁺ release during excitation–contraction coupling. It is concluded that in the adult mammalian atrium IP₃-dependent Ca²⁺ release enhances atrial Ca²⁺ signalling and exerts a positive inotropic effect. In addition, by facilitating Ca²⁺ release, IP₃ may also be an important component in the development of Ca²⁺-mediated atrial arrhythmias.

During each heart beat an action potential depolarizes the cell membrane of cardiac myocytes to allow Ca²⁺ entry through voltage-gated Ca²⁺ channels. This relatively small amount of Ca²⁺ entry triggers a massive Ca²⁺-induced Ca²⁺ release (CICR) from intracellular SR Ca²⁺ stores by activating Ca²⁺-sensitive Ca²⁺ release channels (ryanodine receptors, RyRs) in the SR membrane. CICR represents the key step in excitation–contraction (E-C) coupling which provides the necessary amount of cytoplasmic Ca²⁺ to activate the contractile proteins resulting in contraction of the heart. Ca²⁺ release occurs from clusters of RyRs (Blatter et al. 1997) in the form of localized non-propagating elevations of [Ca²⁺]_i, termed Ca²⁺ sparks. Here we use the term Ca²⁺ spark to refer strictly to elementary Ca²⁺ release events from RyRs. Ca²⁺ sparks are the building blocks of Ca²⁺ release and the spatio-temporal summation of these elementary Ca²⁺ release events forms the whole cell [Ca²⁺]_i transient during E-C coupling.

IP₃ is an important activator of a specific class of SR Ca²⁺ release channels, i.e. IP₃ receptors (IP₃Rs). IP₃-dependent Ca²⁺ release represents the main avenue of intracellular Ca²⁺ release in electrically non-excitable cells (Berridge, 1997). In contrast, in cardiac tissue the main pathway of Ca²⁺ release occurs through RyRs, and IP₃Rs are expressed at 1–2 orders of magnitude lower density than RyRs (Perez et al. 1997). Although IP₃-dependent Ca²⁺ release in cardiac tissue was demonstrated early on (Hirata et al. 1984; Fabiato, 1986; Nosek et al. 1986), the role of IP₃ in E-C coupling and cardiac function in the adult mammalian heart has remained highly controversial (Marks, 2000; Bers, 2001; Blatter et al. 2003). There is evidence that IP₃-dependent signalling may be important during development (Rosemblit et al. 1999; Poindexter et al. 2001) and cardiac injury (Mouton et al. 1992; Jacobsen et al. 1996; Woodcock et al. 1997,1998; Harrison et al. 1998; Yamada et al. 2001), or may be relevant to the regulation of specific cellular functions such as propagation of electrical signals in Purkinje fibres, regulation of organellar and nuclear membrane permeability, Ca²⁺-dependent gene transcription, cardiac hypertrophy signalling and cell growth (e.g. Jaconi et al. 2000; for references see Bers, 2001). Atrial tissue expresses functional IP₃-receptors at 6–10 times higher levels than ventricular myocytes and IP₃Rs seem to colocalize with RyRs in the subsarcolemmal space (Lipp et al. 2000; Mackenzie et al. 2002). Although it has been proposed that IP₃-dependent Ca²⁺ signalling plays a direct role in atrial E-C coupling under physiological as well as pathological conditions (see, e.g. Woodcock et al. 1998; Mackenzie et al. 2002), the spatio-temporal organization of IP₃-dependent Ca²⁺ release and the specific mechanisms by which IP₃ signalling modulates Ca²⁺ handling in atrial myocytes is not clear.

Methods

Cell isolation

The procedure for cell isolation was approved by the Institutional Animal Care and Use Committee of Loyola University Chicago, Stritch School of Medicine. Adult mongrel cats of either sex (19 animals were used in this study) were anaesthetized with thiopental sodium (30 mg kg⁻¹i.p.). Following thoracotomy hearts were quickly excised, mounted on a Langendorff apparatus, and retrogradely perfused with collagenase-containing solution at 37°C according to the method previously described (Kockskämper & Blatter, 2002; Sheehan & Blatter, 2003). All experiments were carried out at room temperature (22–24°C).

[Ca²⁺]_i measurements

[Ca²⁺]_i was measured in intact and permeabilized atrial myocytes with fluorescence laser scanning confocal microscopy. Intact atrial myocytes were loaded with the Ca²⁺ indicator fluo-4 by 20min incubation in Tyrode solution containing 20μm fluo-4 acetoxymethyl ester (fluo-4/AM; Molecular Probes, Eugene, OR, USA) at room temperature. Cells were superfused continuously (1 ml min⁻¹) with normal Tyrode solution (composition in mm: NaCl 140; KCl 4; CaCl₂ 2; MgCl₂ 1; glucose 10; Hepes 10; pH 7.4 adjusted with NaOH). Fifteen to twenty minutes was allowed for de-esterification of the dye. [Ca²⁺]_i measurements were performed with a laser scanning confocal microscope (Radiance 2000 MP, Bio-Rad, UK) equipped with a × 40 oil-immersion objective lens (N.A. = 1.3). Fluo-4 (and fluo-3 in permeabilized cells; see below) was excited with the 488nm line of an argon ion laser and fluorescence was measured at wavelengths >515nm. Images were acquired in the linescan mode (3 or 6ms per scan; pixel size 0.3μm). Whole-cell [Ca²⁺]_i transients were obtained by averaging the entire cellular fluorescence signal from the line scanned. [Ca²⁺]_i transients are presented as background-subtracted normalized fluorescence (F/F₀) where F is the fluorescence intensity and F₀ is resting fluorescence recorded under steady-state conditions at the beginning of an experiment. [Ca²⁺]_i transients were evoked by electrical field stimulation (0.5 Hz). Ca²⁺ sparks were detected and quantified in terms of amplitude, spatial width and frequency using an automated detection algorithm (Cheng et al. 1999). Ca²⁺ spark frequencies are expressed as number of observed sparks per second and per 100μm of scanned distance in the confocal linescan mode (sparks s⁻¹ (100μm)⁻¹). Atrial myocytes were permeabilized with saponin (Zima et al. 2003). First, the cells were suspended in a solution containing (mm): potassium aspartate 100; KCl 20; EGTA 0.5; MgCl₂ 0.75; Hepes 10; pH 7.2 (KOH) and placed in the experimental chamber (final volume 50 μl) for 15 min. The cell surface membrane was permeabilized by adding 0.005% (w/v) saponin for 30 s. After 30 s the bath solution was exchanged for a saponin-free internal solution composed of (mm): potassium aspartate 100; KCl 15; KH₂PO₄ 5; MgATP 5; EGTA 0.4; CaCl₂ 0.12; MgCl₂ 0.75; phosphocreatine 10; creatine phosphokinase 5 U ml⁻¹; dextran (M_r: 40 000) 8%; Hepes 10; fluo-3 potassium salt 0.04; pH 7.2 (KOH). Free [Ca²⁺] and [Mg²⁺] of this solution were 100nm and 1mm, respectively (calculated using WinMAXC 2.05, Stanford University, CA, USA).

Drugs

IP₃ and adenophostin were obtained from Calbiochem, and 2-aminoethoxydiphenyl borate (2-APB), ET-1, heparin (M_r 6000) and tetracaine were from Sigma.

Data analysis

Data are presented as the mean ±s.e.m. of n measurements. Statistical comparisons between groups were performed with Student's t test.

Results

Endothelin effects on Ca²⁺ signals in intact atrial myocytes: involvement of IP₃ signalling

Neurohumoral stimuli (α-adrenergic agents, angiotensin II or endothelin) can cause an increase of [IP₃] in atrial cells (Vogelsang et al. 1994). We used endothelin (ET-1), which binds to ET receptors (ET_A receptor subtype), to study the effect of IP₃-dependent Ca²⁺ signalling during E-C coupling in intact adult mammalian atrial myocytes. In electrically stimulated cells ET-1 (100nm; 10min exposure time) caused an increase in diastolic [Ca²⁺]_i by 61 ± 11% (P < 0.05; observed in 93% of the cells tested; n= 14 cells). The amplitude (Fig. 1A) of electrically evoked action potential-dependent [Ca²⁺]_i transients increased by 27 ± 8% (P < 0.05; 71% of the cells). This positive inotropic effect was typically seen after about 4min of exposure to ET-1, and the maximum effect of ET-1 on Ca²⁺ signalling was observed after 6–8 min. Furthermore, ET-1 caused spontaneous [Ca²⁺]_i transients and Ca²⁺ waves (Fig. 1A and C; observed in 43% of the cells tested). A majority of cells developed Ca²⁺ alternans (Figs 1B; 64% of the cells). Ca²⁺ alternans occurred with a delay of 1–2min after the first signs of an ET-1-induced positive inotropic effect and was stable until the end of exposure to ET-1. In unstimulated cells ET-1 increased the frequency of spontaneous RyR-dependent Ca²⁺ sparks (Fig. 1C) from 0.96 ± 0.16 to 3.5 ± 1.1 sparks s⁻¹ (100μm)⁻¹ (P < 0.05; n= 6 cells). This effect of ET-1 was seen within the first minute of application of the agonist and was fully established after 6min of exposure to ET-1. The membrane permeant IP₃R antagonist 2-APB (2–5μm) prevented or abolished all ET-1 effects on [Ca²⁺]_i in atrial cells (Fig. 1D), suggesting that ET-1 effects were mediated by IP₃, presumably through IP₃-dependent Ca²⁺ release.

Figure 1. Effects of endothelin-1 (ET-1) on Ca²⁺ signalling in intact cat atrial myocytes.

Confocal linescan images and spatially averaged [Ca²⁺]_i transients recorded from field-stimulated (0.5 Hz) atrial myocytes. A, ET-1 (100nm) increased diastolic [Ca²⁺]_i and amplitude of electrically evoked [Ca²⁺]_i transients. Spontaneous [Ca²⁺]_i transients and Ca²⁺ waves (marked by the blue triangles in the right panel) occurred between triggered transients in the presence of ET-1. The traces represent [Ca²⁺]_i expressed as normalized changes of fluo-4 fluorescence (F/F₀). B, ET-1 elicited Ca²⁺ alternans. C, confocal linescan images of spontaneous Ca²⁺ sparks (top) and selected subcellular [Ca²⁺]_i traces (F/F₀, averaged over a distance of 1μm marked by the red triangles to the left of the images) under control conditions (left) and after application of ET-1 (right). ET-1 caused a significant increase in Ca²⁺ spark frequency and resting [Ca²⁺]_i. D, the IP₃R inhibitor 2-APB (2μm) prevented the effects of ET-1 on [Ca²⁺]_i.

IP₃-dependent Ca²⁺ signals in permeabilized cardiac myocytes

We tested the effect of the physiological agonist IP₃ directly in saponin-permeabilized myocytes (Fig. 2A). IP₃ (20μm) increased (Fig. 2C) basal [Ca²⁺]_i by 34 ± 4% (P < 0.001; n= 12) and Ca²⁺ spark frequency from 6.4 ± 0.7 to 9.7 ± 1.1 sparks s⁻¹ (100μm)⁻¹ (P < 0.01; n= 12). However, IP₃ did not alter Ca²⁺ spark amplitude or duration. The latter suggests that IP₃ enhanced the probability of an SR Ca²⁺ release unit (cluster of RyRs) to liberate Ca²⁺ but did not change properties of the RyR cluster. Similar to IP₃, adenophostin (5μm), a more potent IP₃ agonist with two orders of magnitude higher affinity not subject to cellular enzymatic degradation, also increased (Fig. 2B and C) basal [Ca²⁺]_i by 33 ± 7% (P < 0.01; n= 6) and raised the Ca²⁺ spark frequency from 5.4 ± 0.5 to 8.9 ± 0.6 sparks s⁻¹ (100μm)⁻¹ (P < 0.01; n= 6). The effects on basal [Ca²⁺]_i and Ca²⁺ sparks were seen rapidly (<1 min) after exposure to the agonists, but required approximately 5min to develop fully. In contrast to atrial myocytes IP₃ failed to change the frequency and properties of Ca²⁺ sparks in cat ventricular cells (n= 6 cells) indicating that the IP₃ effect on Ca²⁺ signalling was specific to atrial myocytes (Fig. 2D).

Figure 2. Effects of IP₃ and adenophostin on [Ca²⁺]_i in saponin-permeabilized atrial and ventricular myocytes.

A, top, confocal linescan images (fluo-4 fluorescence images; a.u., arbitrary fluorescence intensity units) under control conditions and 1 and 5min after exposure to 20μm IP₃. Bottom, local subcellular changes of [Ca²⁺]_i (F/F₀ averaged over 1μm indicated by the red triangles to the left of the images). B, effect of adenophostin (5μm; same experimental conditions as in A. C, average data of the effects of IP₃ and adenophostin on basal [Ca²⁺]_i and Ca²⁺ spark frequency. Statistically different at P < 0.01 (*) and P < 0.001 (**). A–C reflect data obtained from atrial myocytes. D, IP₃ had no effect on basal [Ca²⁺]_i and frequency and properties of Ca²⁺ sparks in permeabilized cat ventricular myocytes, suggesting that the IP₃ effects were specific to atrial myocytes. [Ca²⁺]_i was measured with fluo-3 pentapotassium salt (fluo-3 bath concentration was 40μm).

To assure that the observed effects of IP₃ were indeed due to IP₃-dependent Ca²⁺ release from the SR we applied the known IP₃R antagonists heparin as well as 2-APB. In the presence of heparin (0.5 mg ml⁻¹) IP₃ (50μm) failed to increase basal [Ca²⁺]_i and Ca²⁺ spark frequency (Fig. 3A). Similarly, the action of IP₃ on basal [Ca²⁺]_i and Ca²⁺ sparks was also prevented by 2-APB (5μm; Fig. 3B). In addition, inhibition of IP₃Rs with 2-APB reversed the effects of IP₃ on basal [Ca²⁺]_i and Ca²⁺ spark frequency (Fig. 3C).

Figure 3. Effect of inhibitors of IP₃ receptors.

A, preincubation of permeabilized atrial myocytes with heparin prevented the effect of IP₃ (50μm) on basal [Ca²⁺]_i and Ca²⁺ sparks. B, same experiment as in A but with the IP₃R blocker 2-APB (5μm). C, 2-APB reversed the increase of basal [Ca²⁺]_i and Ca²⁺ spark frequency induced by preceding exposure to IP₃.

Elementary IP₃R-dependent Ca²⁺ release events in cardiac myocytes

Elementary non-propagating IP₃R-dependent Ca²⁺ release events, termed Ca²⁺ blips and puffs, have been observed in non-excitable cells such as oocytes (Parker & Yao, 1996), HeLa (Bootman et al. 1997) and vascular endothelial cells (Hüser & Blatter, 1997). They differ from RyR-mediated Ca²⁺ sparks in amplitude and kinetics, and have not been observed in cardiac myocytes, presumably because they are difficult to discern in the ‘Ca²⁺ noise’ from RyR-dominated Ca²⁺ release. We tested whether IP₃-dependent puff-like events occurred in conditions where Ca²⁺ release via RyRs was blocked. For this purpose permeabilized atrial myocytes revealing spontaneous Ca²⁺ spark activity were treated with the RyR inhibitor tetracaine (Györke et al. 1997). Tetracaine instead of ryanodine was used to block the RyR, because ryanodine locks RyRs into a subconductance state which can lead to depletion of the SR. In control experiments we confirmed the inhibition of RyR activity by tetracaine. Tetracaine (1mm) reduced the open probability of the RyR channel on average by 98% (P < 0.01; n= 3) as measured with single channel recordings from RyRs incorporated into lipid bilayer (data not shown). Tetracaine also blocked spontaneous RyR-mediated Ca²⁺ sparks. Ca²⁺ spark frequency decreased from 6.0 ± 0.5 to 0.2 ± 0.1 sparks s⁻¹ (100μm)⁻¹ (P < 10⁻⁷; n= 12). After eliminating spontaneous Ca²⁺ sparks with tetracaine, permeabilized atrial myocytes were exposed to IP₃ (20μm). Despite inhibition of RyRs, IP₃ caused a significant increase of basal [Ca²⁺]_i by 37 ± 7% (P < 0.001; n= 6). In addition, localized non-propagating [Ca²⁺] elevations appeared. Heparin completely abolished these Ca²⁺ release events. Adenophostin in the presence of tetracaine elicited the same type of elementary Ca²⁺ release events (Fig. 4A, bottom). Figure 4B shows averaged linescan images and [Ca²⁺]_i profiles of RyR-mediated Ca²⁺ sparks (control) and IP₃R-mediated elementary release events elicited with IP₃ and adenophostin, respectively. In summary (Fig. 4C, bottom) the IP₃R-dependent events had amplitudes which were 75–80% smaller than the average Ca²⁺ spark amplitude (see also surface plots of averaged RyR- and IP₃R-mediated elementary release events, top panel of Fig. 4C). On average IP₃R-mediated events were three to four times longer than Ca²⁺ sparks and the rise time was prolonged by approximately a factor of 2. The spatial spread of the two types of release events did not differ significantly. The differences in kinetics become evident when the average amplitudes were normalized (Fig. 4B, right). The first derivative of the [Ca²⁺]_i transient (d(F/F₀)/dt), which approximates the underlying Ca²⁺ flux (Sheehan & Blatter, 2003), revealed that the Ca²⁺ release flux of IP₃-dependent events was clearly smaller. On rare occasions small non-propagating Ca²⁺ release events were observed in the presence of tetracaine alone. These events were approximately three times smaller in amplitude than Ca²⁺ sparks recorded under control conditions, however, they revealed otherwise the same spatial and temporal characteristics (average Ca²⁺ spark properties in tetracaine: amplitude ΔF/F₀= 0.67; duration 26.6ms; width 2.0μm; rise time 17.2ms) as regular Ca²⁺ sparks (compare to tabulated values in Fig. 4C). Thus, these rare events were different from those observed after addition of IP₃ and resulted from the opening of a smaller number of RyRs in a cluster of release channels. In summary, the IP₃-dependent elementary Ca²⁺ release events we observed in permeabilized atrial myocytes were distinctly different from RyR Ca²⁺ sparks and were reminiscent of Ca²⁺ puffs typically observed in non-excitable tissue where IP₃-dependent Ca²⁺ signalling is predominant (Parker & Yao, 1996; Berridge, 1997; Bootman et al. 1997; Hüser & Blatter, 1997).

Figure 4. Elementary Ca²⁺ release events from IP₃Rs in atrial myocytes.

A, confocal linescan images from permeabilized cat atrial myocytes. Top, tetracaine (1mm) abolished spontaneous Ca²⁺ spark activity. In the presence of tetracaine, IP₃ (20μm) caused an increase in basal [Ca²⁺]_i and the occurrence of non-propagating Ca²⁺ release events with significantly different amplitude and kinetics compared to Ca²⁺ sparks. Heparin abolished the elevation of basal [Ca²⁺] and Ca²⁺ release events. Bottom, localized Ca²⁺ release events elicited with adenophostin in the presence of tetracaine. B, left, average linescan images and F/F₀ traces of Ca²⁺ sparks (black), and IP₃- (red) and adenophostin- (blue) mediated Ca²⁺ release events. Right, normalized amplitudes of local [Ca²⁺]_i transients (F/F₀, norm.) and first derivatives (d(F/F₀)/dt). d(F/F₀)/dt serves as a measure of the underlying Ca²⁺ release flux. C, surface plot representation of averaged linescan images of Ca²⁺ sparks and IP₃R-mediated Ca²⁺ release events from B. Bottom, table showing average values of Ca²⁺ release event amplitude (ΔF/F₀), duration (measured at half maximum amplitude), spatial width (full width at half maximum amplitude) and rise time. ‘Control’ indicates RyR-mediated Ca²⁺ sparks recorded in the absence of tetracaine and IP₃R agonists.

Discussion

Atrial myocytes reveal two classes of elementary Ca²⁺ release events

In this study we show that elementary Ca²⁺ release events from RyRs and IP₃Rs coexist in atrial myocytes. Although available quantitative data on the spatio-temporal properties of Ca²⁺ sparks and Ca²⁺ puffs or blips vary considerably (see, e.g. summary in Bootman, 1996), distinct differences in amplitude and kinetics are described. IP₃R-dependent elementary release events have slower rise and decline kinetics, last longer and have a smaller amplitude. The same observation was made in this study where the two types of release events could be compared directly. The IP₃-dependent release events differed from RyR-dependent Ca²⁺ sparks in all of these parameters in the same manner. The duration of the IP₃-dependent events observed in atrial myocytes compares fairly well with Ca²⁺ puffs observed in HeLa (Bootman et al. 1997) and vascular endothelial cells (Hüser & Blatter, 1997) which is in the range of 100–200ms, compared to the duration of a Ca²⁺ spark (<50ms; see, e.g. Bers 2001). Thus, the elementary IP₃R-dependent events recorded from atrial myocytes are clearly reminiscent of Ca²⁺ puffs observed in non-excitable tissues.

Cross-talk between IP₃R- and RyR-dependent Ca²⁺ release: modulation of CICR by IP₃

In atrial myocytes Ca²⁺ release from the SR during E-C coupling occurs primarily through RyRs. We have shown that in cat atrial myocytes inhibition of RyRs reduces the [Ca²⁺]_i-transient amplitude by ∼90% (Kockskämper et al. 2001; Sheehan & Blatter, 2003). Nonetheless, in atrial tissue IP₃-dependent Ca²⁺ release exerts an important modulatory role for Ca²⁺ signalling during E-C coupling by facilitating Ca²⁺ release via RyRs. IP₃-dependent Ca²⁺ release makes also a small direct contribution to the [Ca²⁺]_i transient although in quantitative terms this effect is likely to be small based on the small number of IP₃Rs and the magnitude of IP₃-dependent Ca²⁺ release events. Ca²⁺ affects the behaviour of both types of channels. Ca²⁺ is the primary activator of RyRs, and cytoplasmic as well as lumenal Ca²⁺ change the sensitivity of the release channel to CICR. The open probability (P_o) of the IP₃R type-2 (cardiac) shows a steep Ca²⁺ dependence in the range of 10–100nm, but is rather Ca²⁺-independent at [Ca²⁺] > 100nm (Ramos-Franco et al. 1998), i.e. at [Ca²⁺] encountered in cardiac cells at rest as well as during activation. Type-2 IP₃R has the highest sensitivity to IP₃, suggesting, together with its Ca²⁺-independence at [Ca²⁺] > 100nm, that the cardiac IP₃R functions as a pure IP₃ sensor. This suggests that IP₃Rs are rather unaffected by Ca²⁺ release from RyR. In contrast, release of Ca²⁺ from IP₃Rs can impose critical changes of [Ca²⁺]_i in the microenvironment of the RyRs which facilitates CICR from neighbouring RyR Ca²⁺ release sites. In the present study we found IP₃-dependent release events in both, in subsarcolemmal as well as in deeper regions of the cell, suggesting that IP₃-dependent Ca²⁺ release may affect CICR from both junctional and non-junctional SR of atrial myocytes (Blatter et al. 2003; Sheehan & Blatter, 2003). In summary, IP₃ can exert a positive inotropic effect by enhancing Ca²⁺ release from the SR and contraction in a beat-to-beat fashion. This may indeed represent one of the mechanisms through which neurohumoral agents such as α-adrenergic agonists, angiotensin II or endothelin modulate cardiac Ca²⁺ signalling and contractility.

IP₃-dependent Ca²⁺ signalling and atrial arrhythmias

IP₃-dependent Ca²⁺ signalling has been implied in cardiac arrhythmias due to ischaemia and reperfusion injury, inflammatory processes and developing cardiac failure (see, e.g. Woodcock et al. 1998; Mackenzie et al. 2002). Our data lend direct support to the notion that IP₃-dependent Ca²⁺ release plays a causal role in the genesis of atrial arrhythmias. IP₃ caused spontaneous [Ca²⁺]_i transients and Ca²⁺ waves as well as Ca²⁺ alternans (Fig. 1), all disturbances in Ca²⁺ signalling related to cardiac arrhythmias (see Kockskämper & Blatter, 2002). Our observation that IP₃-dependent Ca²⁺ signalling is pivotal for atrial E-C coupling and is responsible for the pro-arrhythmogenic disturbances of cellular Ca²⁺ homeostasis suggests that therapeutic agents which target the IP₃R and the IP₃/Ca²⁺ signalling cascade may prove beneficial for the prevention and treatment of cardiac arrhythmias (Woodcock et al. 1998).