Role of Nitric Oxide in Ca2+ Sensitivity of the Slowly Activating Delayed Rectifier K+ Current in Cardiac Myocytes
Sarcolemmal Ca2+ entry is a vital step for contraction of cardiomyocytes, but Ca2+ overload is harmful and may trigger arrhythmias and/or apoptosis. To maintain the amount of Ca2+ entry within an appropriate range, cardiomyocytes have feedback systems that tightly regulate ion channel activities in response to the changes in intracellular Ca2+ concentration ([Ca2+]i), thereby regulating Ca2+ entry. In guinea pig ventricular myocytes, Ca2+ ionophore, A23187, induced suppression of the L-type Ca2+ currents (ICa,L) and enhancement of the slowly activating delayed rectifier K+ currents (IKs). At a low stimulation rate, ICa,L suppression and IKs enhancement contributed to the A23187-induced APD shortening with a similar magnitude, whereas at a high stimulation rate, IKs enhancement dominantly contributed to APD shortening. IKs enhancement induced by A23187 was attributable to actions of nitric oxide (NO), because they were inhibited by an inhibitor of NO synthase (NOS) and by a NO scavenger. A23187-induced alterations of APD and IKs were strongly suppressed by a NOS3 inhibitor, but barely affected by a NOS1 inhibitor, suggesting that NOS3 was responsible for NO release in this phenomenon. Inhibition of calmodulin (CaM), but not Akt, blocked the enhancement of IKs by A23187. Thus, CaM-dependent NOS3 activation confers the selective Ca2+-sensitivity on IKs. Ca2+-induced IKs enhancement and resultant APD shortening potentially act as a physiological regulatory mechanism of Ca2+ recycling, because they were observed at a physiological range of [Ca2+]i in cardiac myocytes and are induced by physiologically relevant Ca2+ loading, such as digitalis application and rise in extracellular Ca2+ concentration.
To maintain the amount of Ca2+ entry within a narrow range, cardiomyocytes develop feedback systems that tightly regulate ion channel activities and thereby Ca2+ entry in response to the changes in intracellular Ca2+ concentration ([Ca2+]i). The elevation of [Ca2+]i directly regulates Ca2+ entry pathway, the L-type Ca2+ channel.1–3 It also alters the duration of action potential plateau and regulates the duration where the L-type Ca2+ channel can open. The slope of action potential plateau is mainly dependent on the balance between L-type Ca2+ channel currents (ICa,L) and delayed rectifier K+ currents. Delayed rectifier K+ currents consist of two components: the rapidly activating component (IKr) that is sensitive to a class III antiarrhythmic drug E-4031 and the slowly activating component (IKs) that is E-4031-resistant.4 Elevations in [Ca2+]i selectively enhances the IKs component.4,5 The range of [Ca2+]i that regulates IKs (between 10−8 and 10−6 mol/L) is in the range that alters during the cycle of contraction and relaxation of cardiac myocytes.5,6 Thus, the Ca2+ sensitivity of IKs may potentially act as a physiological feedback system to regulate [Ca2+]i.
In a previous article, we showed that IKs enhancement by elevation of [Ca2+]i was inhibited by a calmodulin (CaM) inhibitor, but not by an inhibitor of CaM-dependent kinase II.6 Although this finding suggests that the allosteric regulation of CaM is crucial for Ca2+-sensitive IKs alterations, the detailed underlying mechanism remains unknown. Recently, we found that IKs was enhanced by nitric oxide (NO) via a cGMP-independent mechanism.7 Because allosteric interaction of Ca2+/CaM complex with NO synthase (NOS) is a major mechanism of NOS activation and NO release,8,9 we tested in the present study if NO plays an important role in the Ca2+-sensitive modulation of IKs in cardiac myocytes.
Materials and Methods
The investigation was conducted in accordance with the rules and regulations of the Institutional Animal Care and Use Committee of Tokyo Medical and Dental University.
Quantification of [Ca2+]i
Quantification of [Ca2+]i was performed as described previously.10 (See the Expanded Materials and Methods section in the online data supplement available at http://circres.ahajournals.org.) Briefly, guinea pig ventricular myocytes were loaded with 10 μmol/L indo-1-AM (Dojin), and Ca2+ transients were elicited by field-stimulation at 1 Hz. Cells were excited at 360 nm from a Xenon lamp and the emission bands, 395 to 415 nm and 470 to 490 nm, were separated (W-VIEW system, Hamamatsu Photonics), detected by a high-speed cooled CCD camera (HISCA, Hamamatsu Photonics), and ratio of the intensity of two emission bands was calculated (Aquacosmos software, Hamamatsu Photonics).
Single ventricular myocytes were harvested from hearts of adult guinea pigs (n=41, white Hartrey; Saitama Experimental Animal Supply Co Ltd, Saitama, Japan).11 Action potentials and membrane currents were recorded with an Axopatch 200B amplifier (Axon Instruments) using the perforated configuration of the patch-clamp technique with Amphotericin B (Sigma-Aldrich), in a current-clamp mode and a voltage-clamp mode. Action potentials were elicited by passing depolarizing current pulses (<2 ms in duration at a rate of 1 Hz) of suprathreshold intensity. IKs were elicited by 3.5-second depolarizing pulses to various test potentials between −30 and +50 mV in 10-mV increments at 0.1 Hz from a holding potential of −40 mV, and amplitudes of tail IKs were measured by extrapolating from exponential fits. Compositions of solutions used are described in the online data supplement. All experiments were performed at 36±1°C. The averaged series resistance was 15.7±1.7 MΩ, the capacitance time constant was 2.5±0.3 ms, and the membrane capacitance was 150±13 pF.
Immunoprecipitation and Immunoblotting
Cardiomyocytes were isolated from adult guinea pigs (n=3),12 and immunoprecipitation and immunoblotting were performed as described previously (see online data supplement).13 Briefly, cell lysates were immunoprecipitated with a monoclonal anti-NOS3 antibody (Zymed), followed by immunoblot analysis with an anti-NOS3 antibody (Zymed), an anti-caveolin-3 (Cav-3) antibody (N-18, Santa Cruz Biotechnology), or an anti-CaM antibody (FL-149, Santa Cruz Biotechnology) followed by incubation with a horseradish peroxidase-conjugated anti-mouse IgG (DAKO Japan Co Ltd) or an anti-rabbit IgG (DAKO Japan Co Ltd). Proteins were detected using the advance enhanced chemiluminescence system (Amersham Bioscience).
Chromanol 293 B was supplied by Hoechst. E-4031 was purchased from Eisai Co Ltd, A23187, W7 and W5 from Wako, SH-6 from Merck, digoxin from Nacalai tesque, and all other reagents from Sigma-Aldrich. A23187 was prepared as a 10 mmol/L stock solution in ethanol. Stock solutions of E-4031 (5 mmol/L), W7 (10 mmol/L), W5 (10 mmol/L), S-methylisothiourea (SMTU; 10 mmol/L), N-acetyl-l-cysteine (l-NAC; 100 mmol/L), S-methyl-l-thiocitrulline (SMTC; 5 mmol/L), l-N5-(l-lminoethyl)ornithine (l-NIO; 1 mmol/L), and sodium nitroprusside (SNP; 1 mol/L) were prepared in distilled water, and nisoldipine (10 mmol/L), SH-6 (20 mmol/L), and 2-(4-morpholinyl)-8-phenyl-1-(4H)-benzopyran-4-one hydrochloride (LY-294,002; 60 mmol/L) stock solutions in dimethylsulfoxide. They were diluted in the bath solution to achieve the desired concentration.
All values are presented as mean±SE. Statistical significance was determined by repeated-measure one-way analysis of variance (ANOVA) followed by Scheffé F test. Values P<0.05 are considered to be significant.
Elevation of [Ca2+]i by A23187 Shortened Action Potential Duration and Enhanced IKs
Addition of A23187 to the bath solution increased [Ca2+]i both in the basal level and at the peak of Ca2+ transient (Figure 1A). The basal [Ca2+]i was 0.12±0.01 μmol/L (n=10) in the control state, 0.16±0.03 μmol/L (n=5) in the presence of A23187 at 0.1 μmol/L (P=NS versus control), and 0.22±0.02 μmol/L (n=5) at 2 μmol/L of A23187 (P<0.05 versus control). The peak [Ca2+]i was 0.97±0.03 μmol/L (n=10) in the control state, 1.24±0.03 μmol/L (n=5) at 0.1 μmol/L of A23187 (P<0.05 versus control), and 1.52±0.09 μmol/L (n=5) at 2 μmol/L of A23187 (P<0.05 versus control).
Application of A23187 induced a dose-dependent shortening of action potential duration and enhancement of IKs, which attained a maximal value at a concentration of 2 μmol/L of A23187 (where APD20 shortened by 49%, APD90 by 20%, and IKs amplitude increased by 31%) (see online data supplement). These values are comparable to magnitude of Ca2+-induced enhancement of IKs recorded in the excised giant-patch recording.6 Thus, in the following experiments, we used A23187 at this concentration (2 μmol/L). Once electrical access in the perforated patch configuration was established, there was a slight time-dependent reduction in APD that occurred without drug application. Application of A23187 (2 μmol/L) reversibly shortened both APD20 (▴ in Figure 1B) and APD90 (▵ in Figure 1B) with a significantly greater magnitude than time-dependent reduction in APD without drug application (• and ○ in Figure 1B) (n=5, P<0.01). Treatment with A23187 (2 μmol/L) for 15 minutes shortened APD20 by 49.4±3.3% (Figure 1B, n=5, P<0.01 versus spontaneous APD20 shortening), and APD90 by 19.2±3.6% (Figure 1B, n=5, P<0.01 versus spontaneous APD20 shortening). The tail amplitude of IKs in the control cells exhibited a time-dependent decrease (○ in Figure 1C). Application of A23187 (2 μmol/L) reversibly increased tail IKs by 31.2±2.3% (Figure 1C, n=5, P<0.01 versus spontaneous time-dependent reduction in IKs). Addition of EGTA (10 mmol/L), a Ca2+ chelator, to the pipette solution prevented shortening of APD and enhancement of tail IKs amplitude by A23187 (see online data supplement). These results indicate that the effects of extracellularly applied A23187 on APD and IKs are dependent on the elevation in [Ca2+]i.
Relative Contribution of IKs and ICa,L to Ca2+-Induced APD Shortening
ICa,L amplitude is diminished by elevation in [Ca2+]i via the Ca2+-induced Ca2+ inactivation mechanism, which also induces APD shortening.2,3 To assess relative contribution of Ca2+-induced IKs enhancement and ICa,L inactivation to the APD shortening induced by A23187, we used a IKs channel blocker, chromanol 293B, and/or a ICa,L channel blocker, nisoldipine, at supramaximal concentrations. Action potentials were continuously elicited at a stimulation rate of 0.1, 0.4, 1, and 4 Hz, and after shortening of APD by A23187 (2 μmol/L) had reached a pseudo-steady state, chromanol 293B (30 μmol/L), nisoldipine (3 μmol/L), or a combination of chromanol 293B (30 μmol/L) and nisoldipine (3 μmol/L) was added in the presence of A23187 (2 μmol/L). At 0.1 Hz, chromanol 293B alone or nisoldipine alone partially reversed A23187-induced APD20 and APD90 shortening with a similar magnitude (left panel in Figure 2A): APD20 shortening was reversed by 49.4±3.3% by chromanol 293B and by 48.9±3.1% by nisoldipine, (n=5, P=NS) (Figure 2B), and APD90 shortening was reversed by 54.5±3.4% by chromanol 293B and by 42.7±3.0% by nisoldipine (n=5, P=NS) (Figure 2C). A combination of chromanol 293B and nisoldipine completely reversed A23187-induced APD shortening to the control levels (Figure 2B and 2C). When stimulation rate was increased to 0.4, 1, and then 4 Hz, the magnitude of APD shortening was progressively increased with progressive increases in the fraction of APD shortening reversed by chromanol 293B, leaving the fraction reversed by nisoldipine constant. At 4 Hz, chromanol 293B reversed APD20 shortening by 74.8±3.1% and APD90 shortening by 79.8±3.3%, and nisoldipine reversed APD20 shortening by 23.6±3.1% and APD90 shortening by 20.0±3.1% (right panel in Figure 2A through 2C). Thus, at a low stimulation rate both Ca2+-induced IKs enhancement and Ca2+-induced ICa,L contributed to the Ca2+-induced APD shortening with a similar magnitude. However, at a high stimulation rate, contribution of IKs enhancement was significantly greater than contribution of ICa,L suppression.
Roles of NO in APD Shortening and IKs Enhancement by A23187
Ca2+ is a well-known activator of NOS.8,9 Recently, we reported that NO increased IKs possibly by a mechanism dependent on direct S-nitrosylation of the channel protein,6 which prompted us to examine potential roles of NO in the A23187-induced enhancement of IKs. After A23187 (2 μmol/L) had shortened APD and enhanced IKs, we applied a nonspecific NOS inhibitor, SMTU (1 μmol/L). Effects of A23187 (2 μmol/L) on APD were partially, but significantly reversed by SMTU (Figure 3A), and those on IKs were fully reversed back to the control levels (Figure 3B). Application of a NO scavenger, l-NAC (1 mmol/L), also reversed enhancement of IKs induced by A23187 to the control levels (Figure 3B). Effects of A23187 on the gating properties of IKs were also similar to those by a NO donor, SNP, which provide supporting evidence that A23187 enhances IKs by causing the production of NO (see online data supplement).
NOS3 Is Responsible for A23187-Induced NO Release
Increase in IKs occurs rapidly after application of A23187, which argues against activation of inducible NOS (NOS2). Therefore, we determined whether A23187 activated NOS1 or NOS3 to cause enhancement of IKs. We used 2 different NOS inhibitors, SMTC, which has a significantly lower IC50 for NOS1 (0.31 μmol/L) than for NOS3 (5.4 μmol/L),14 and l-NIO, which has a significantly lower IC50 for NOS3 (0.5 μmol/L) than for NOS1 (3.9 μmol/L).15 Application of SMTC (3 μmol/L) did not alter enhancement of tail IKs amplitude by A23187 (a and c in Figure 3C), whereas l-NIO (1 μmol/L) reversed A23187-enhanced IKs to the initial levels (b and c in Figure 3C). These findings strongly suggest that effects of A23187 on APD and IKs occur via NO released from NOS3.
NOS3 Activation Is Dependent on CaM
NOS3 is activated via at least two distinct pathways, a Ca2+-dependent pathway involving Ca2+-binding protein CaM or a phosphorylation-dependent pathway involving Ser/Thr kinase Akt.16,17 We determined how A23187 activated NOS3 by using a CaM inhibitor (W7) and an Akt inhibitor (SH-6). W7 (10 μmol/L) reversed enhancement of IKs by A23187 back to the initial levels (a in Figure 4A and 4B). A chemically related but inactive analog of W7, W5, did not alter A23187-enhanced IKs (b in Figure 4A and 4B). Neither SH-6 (10 μmol/L) nor LY-294,002 (30 μmol/L), an inhibitor of PI-3 kinase (an upstream activator of Akt), altered A23187-enhanced tail IKs (c and d in Figure 4A and 4B). Similarly, application of W7 (10 μmol/L) reversed shortening of APD20 and APD90 by A23187 (2 μmol/L) (a in Figure 4C and 4D). In contrast, SH-6 (10 μmol/L) application did not reverse shortening of APD20 or APD90 by A23187 (b in Figure 4C and 4D).
In cardiomyocytes, NOS3 localizes into caveolae and interacts with muscle-specific caveolin, Cav-3.18,19 Dissociation of NOS3 from Cav-3 is a mandatory step for NOS3 activation.20 Using cardiac myocyte lysates, we performed immunoprecipitation using a NOS3 antibody to determine its relative associations with Cav-3 or CaM. In the control condition, immunoblot with an anti–Cav-3 antibody or an anti-CaM antibody demonstrated that NOS3 was predominantly associated with Cav-3 but not with CaM (lane 1 in Figure 5A and 5B). In contrast, cells treated with A23187 (2 μmol/L) for 20 minutes reduced association of NOS3 with Cav-3 and increased its association with CaM (lane 2 in Figure 5A and 5B). W7 (10 μmol/L) prevented changes in NOS3 association caused by A23187 (2 μmol/L) (lane 3 in Figure 5A and 5B), whereas SH-6 (10 μmol/L) did not alter changes in NOS3 association caused by A23187 (2 μmol/L) (lane 4 in Figure 5A and 5B). These data indicate that A23187 causes NOS3 to switch its binding partner from Cav-3 to CaM.
Effects of Ca2+ Loading by Digoxin and Rise in Extracellular Ca2+ Concentration on APD and IKs
Finally to examine if physiologically relevant Ca2+-loading also shorten APD and enhance IKs, we examined effects of digoxin and rise in extracellular Ca2+ concentration ([Ca2+]o). Application of digoxin at 0.2 μmol/L to the bath solution did not significantly change basal [Ca2+]i (0.10±0.01 μmol/L in the control state [n=6] and 0.13±0.02 μmol/L in the presence of digoxin at 0.2 μmol/L [n=6]; P=NS), whereas peak [Ca2+]i was increased from 0.98±0.06 μmol/L (n=6) in the control state to 1.47±0.08 μmol/L (n=6) (P<0.05). Digoxin at 2 μmol/L increased both basal [Ca2+]i from 0.10±0.01 μmol/L (n=6) in the control state to 0.20±0.01 μmol/L (n=5) (P<0.05) and peak [Ca2+]i from 0.98±0.06 μmol/L (n=6) to 1.96±0.09 μmol/L (n=5) (P<0.05) (Figure 6A). Digoxin at 0.2 μmol/L shortened APD20 by 28.0±3.9% and APD90 by 12.6±3.1% relative to the control, and digoxin at 2 μmol/L shortened APD20 by 58.8±3.4% and APD90 by 24.0±3.4% relative to the control (Figure 6B and online data supplement). l-NAC (1 mmol/L) partially, but significantly, reversed APD shortening by digoxin (Figure 6B and online data supplement). Digoxin at 0.2 μmol/L increased IKs amplitude by 17.4±3.1%, and digoxin at 2 μmol/L by 37.7±7.1% relative to the control (Figure 6B and online data supplement). IKs enhancement by digoxin was fully reversed by application of l-NAC (1 mmol/L) (Figure 6B and online data supplement).
Rise in [Ca2+]o from the control state (1.8 mmol/L) to 2.4 mmol/L shortened APD20 by 27.2±3.9% and APD90 by 17.4±3.0%, and to 3.0 mmol/L shortened APD20 by 45.4±3.8% and APD90 by 26.7±3.2%, which were partially, but significantly reversed by l-NAC (1 mmol/L) (Figure 6C and online data supplement). Rise in [Ca2+]o from 1.8 mmol/L to 2.4 mmol/L, and to 3.0 mmol/L enhanced IKs by 27.9±8.3% and 43.8±11.2%, respectively, which were fully reversed by l-NAC (1 mmol/L) (Figure 6C and online data supplement).
NO has been shown to play important roles in the physiological regulation of Ca2+ cycling of cardiomyocytes.21 In this study, we demonstrate that Ca2+ regulation of IKs channel is another critical NO-dependent Ca2+ cycling system based on the following findings: (1) a NO donor mimicked Ca2+-induced IKs enhancement; (2) a NOS inhibitor abolished IKs enhancement by Ca2+; (3) a NO scavenger also abolished Ca2+-induced IKs enhancement; and (4) changes in IKs kinetics caused by a NO donor were very similar to those by Ca2+.
To examine the mechanism underlying Ca2+-sensitivity of IKs, we increased [Ca2+]i by using a Ca2+-ionophore, A23187, and recorded whole-cell IKs in the perforated patch configuration. This recording configuration is preferred to study Ca2+ regulation of IKs, because it prevents wash-out of the intracellular signal transduction machinery. The peak [Ca2+]i was increased from 0.97±0.03 μmol/L in the control state to 1.24±0.03 μmol/L by A23187 at 0.1 μmol/L and to 1.52±0.09 μmol/L by A23187 at 2 μmol/L, whereas APD and IKs were barely affected by A23187 at 0.1 μmol/L and maximally changed by A23187 at 2 μmol/L. Thus, the threshold of [Ca2+]i to modulate APD and IKs appears to be between 1.24 and 1.52 μmol/L, which is well within the range where [Ca2+]i physiological varies in cardiac myocytes. The changes in [Ca2+]i reached the stable level at 5 minutes after application of A23187 (see online data supplement), which is similar to the time course of changes in APD and IKs by A23187, further supporting that effects of APD and IKs by A23187 were attributable to changes in [Ca2+]i.
In addition to Ca2+-induced enhancement of IKs, rise in [Ca2+]i causes Ca2+-induced inactivation of ICa,L.13,14 To understand the feedback system of Ca2+ entry, it is important to know relative contributions of Ca2+-induced IKs enhancement and ICa,L inactivation to the Ca2+-induced APD shortening. In the present study, at a low stimulation rate, both IKs enhancement and ICa,L inactivation contribute to the Ca2+-induced APD shortening with a similar magnitude. However, at a high stimulation rate, A23187-induced APD shortening was reversed by 75% to 80% with chromanol 293B and only by 20% to 25% with nisoldipine. Thus, at a high stimulation rate, Ca2+-induced IKs enhancement plays a dominant role in APD shortening. A23187-induced APD shortening was dependent on stimulation rate: the higher the stimulation was, the greater the A23187-induced APD shortening was. This result is consistent with a simulation study showing that IKs underlies rate-dependent adaptation of APD at fast heart rates.22 Fast pacing results in short diastolic intervals that prevent complete deactivation of IKs, resulting in build-up of instantaneous IKs repolarizing current.22 In our study, elevated IKs components at higher [Ca2+]i may cause additional accumulation of the repolarizing current, resulting in the greater APD shortening in the presence of A23187.
We have previously demonstrated that ICa,L is suppressed by NO via a cGMP-dependent pathway.11 However, NO-dependent modulation of ICa,L appears not to play a significant role in the Ca2+-induced ICa,L inactivation in our experiments, because neither a NO scavenger, l-NAC, nor a NOS inhibitor, SMTU, reversed the A23187-induced ICa,L suppression (see online data supplement). Because ICa,L channel inactivation is underlined by direct interaction of Ca2+/CaM with the ICa,L channels2,3 that is situated in a pathway upstream of Ca2+/CaM-dependent activation of NOS3, effects of NO on the ICa,L channels may be masked by a direct effect of Ca2+/CaM. The finding that l-NAC and SMTU reversed A23187-induced APD shortening only partially may further implicate that a fraction of APD shortening caused by ICa,L inactivation is insensitive to NO modulators.
It is also critical to know whether Ca2+-induced IKs enhancement and resultant APD shortening act in physiological and/or pathological conditions of the hearts. To address this question, we used physiologically relevant Ca2+ loading such as digoxin and rise in [Ca2+]o. Digoxin, a selective blocker of Na+/K+-ATPase is known to increase [Ca2+]i by indirectly enhancing activity of Na+-Ca2+ exchanger, and in fact in our study, digoxin increased both basal and peak [Ca2+]i. Both application of digoxin and rise in [Ca2+]o induced shortening of APD and enhancement of IKs. APD shortening was partially reversed by a NO scavenger, l-NAC, and IKs enhancement was fully reversed by l-NAC. Thus, Ca2+-induced IKs enhancement and resultant APD shortening appear to be one of the physiological and pathological regulators of IKs, among other well-known regulators, such as β-adrenergic regulation.23,24
Because Ca2+ enters in cells during the action potential plateau, changes in APD may affect the amount of Ca2+ entry, and thereby [Ca2+]i.25 Indeed, Padmala and Demir reported that the 40% reduction in APD results in 60% reduction in Ca2+ entry through ICa,L, 23% reduction in peak [Ca2+]i, and 18% reduction in diastolic [Ca2+]i.26 Thus, Ca2+-induced APD shortening potentially acts as a negative feedback system to maintain [Ca2+]i within an appropriate range.
Cardiomyocytes express both NOS1 and NOS3.19,21 The finding that l-NIO, a relatively specific NOS3 inhibitor completely abolished IKs enhancement by A23187, whereas SMTC, a relatively specific NOS1 inhibitor, barely affected it, strongly suggests that the activation of NOS3, rather than NOS1, is involved in the IKs enhancement. NOS3 is activated via at least two distinct pathways, a CaM-dependent pathway and an Akt-dependent pathway.17,18 In the present study, Ca2+-dependent IKs enhancement was blocked by a CaM inhibitor, W7, but was barely affected by an Akt inhibitor, SH-6, indicating that a CaM-dependent activation of NOS3 participates in this phenomenon. We have previously reported that IKs is enhanced by intracellular Ca2+, which may be consequences of allosteric interaction of CaM rather than phosphorylation by CaM kinase II, because a CaM inhibitor, W7, but not a CaM kinase II inhibitor, KN62, inhibited the Ca2+-induced IKs enhancement.5 This hypothesis is supported by our immunoprecipitation-immunoblotting data that the elevation of [Ca2+]i by A23187 switched the protein-protein interaction from NOS3-Cav-3 to NOS3-CaM. In cardiomyocytes, NOS1 and NOS3 have opposing actions in Ca2+ cycling: NO released from NOS3 decreases [Ca2+]i by suppressing the ICa,L channel activities, whereas NO released from NOS1 increases [Ca2+]i by activating ryanodine receptor type 2.19,21 The opposing actions of NOS1 and NOS3 in Ca2+ cycling are critically dependent on their spatial confinement with effecter molecules: the ryanodine receptor type 2 and NOS 1 on endoplasmic reticulum and the ICa,L channel and NOS3 on sarcolemma.19,21 The Ca2+-induced enhancement of IKs can be observed in cell-free excised patch configuration,6 implicating that the spatial confinement of NOS3 with the IKs channel also takes place. Interestingly, the α subunit of the IKs channel, KCNQ1, has two potential binding sites for Ca2+-free CaM (apo-CaM), an IQ motif and a Φ motif, in its intracellular C-terminal region.27 Yeast two-hybrid assay also suggests that KCNQ1 associates with apo-CaM.27 Thus, the changes in protein-protein interactions may occur in the membrane-delimited microdomains that involve Cav-3, NOS3, CaM, and KCNQ1. Because of participation of Cav-3, the microdomains are likely to be in the caveola. Although the IKs channel has been suggested to colocalize with the β2-adrenergic receptor that accumulates in caveolae,28 the direct demonstration of localization of components of the IKs channel, KCNQ1 and KCNE1, with the caveolae remains to be shown.
This work was supported in part by Research Grant from the Ministry of Education, Culture, Sports, Science, and Technology of Japan and a research grant from the Takeda Science Foundation. We thank Dr Brian Delisle (University of Wisconsin) for checking English, K. Yoshida for the technical assistance, and A. Sugai for the secretarial services.
Original received July 14, 2004; revision received November 9, 2004; accepted November 17, 2004.
Mori MX, Erickson MG, Yue DT. Functional stoichiometry and local enrichment of calmodulin interacting with Ca2+ channels. Science. 2004; 304: 432–435.
Sanguinetti MC, Jurkiewicz NK. Two components of cardiac delayed rectifier K+ current. Differential sensitivity to block by class III antiarrhythmic agents. J Gen Physiol. 1990; 96: 195–215.
Nitta J, Furukawa T, Marumo F, Sawanobori T, Hiraoka M. Subcellular mechanism for Ca2+-dependent enhancement of delayed rectifier K+ current in isolated membrane patches of guinea pig ventricular myocytes. Circ Res. 1994; 74: 96–104.
Nathan C, Xie Q-W. Regulation of biosynthesis of nitric oxide. J Biol Chem. 1994; 269: 13725–13728.
Garcia-Cardena G, Martasek P, Masters BS, Skidd PM, Couet J, Li S, Lisanti MP, Sessa WC. Dissecting the interaction between nitric oxide synthase (NOS) and caveolin. Functional significance of the NOS caveolin binding domain in vivo. J Biol Chem. 1997; 272: 25437–25440.
Furukawa T, Ito H, Nitta J, Tsujino M, Adachi S, Hiroe M, Marumo F, Sawanobori T, Hiraoka M. Endothelin-1 enhances calcium entry through T-type calcium channels in cultured neonatal rat ventricular myocytes. Circ Res. 1992; 71: 1242–1253.
Zheng Y-J, Furukawa T, Ogura T, Tajimi K, Inagaki N. M phase-specific expression and phosphorylation-dependent ubiquitination of the ClC-2 channel. J Biol Chem. 2002; 277: 32268–32273.
Goligorsky MS, Li H, Brodsky S, Chen J. Relationships between caveolae and eNOS: everything in proximity and the proximity of everything. Am J Physiol. 2002; 283: F1–F10.
Garcia-Cardena G, Oh P., Liu J, Schnitzer JE, Sessa WC. Targeting of nitric oxide synthase to endothelial cell caveolae via palmitoylation: implications for nitric oxide signaling. Proc Natl Acad Sci U S A. 1996; 93: 6448–6453.
Barouch LA, Harrison RW, Skaf MW, Rosas GO, Cappola TP, Kobeissi ZA, Hobai IA, Lemmon CA. Burnett AL, O’Rourke B, Rodoriguez ER, Huang PL, Lima JA, Berkowitz DE, Hare JM. Nitric oxide regulates the heart by spatial confinement of nitric oxide synthase isoforms. Nature. 2002; 416: 337–339.
Michel JB, Feron O, Sacks D, Michel T. Reciprocal regulation of endothelial nitric-oxide synthase by Ca2+-calmodulin and caveolin. J Biol Chem. 1997; 272: 15583–15586.
Walsh KB, Begenisich TB, Kass RS. Beta-adrenergic modulation of cardiac ion channels. Differential temperature sensitivity of potassium and calcium currents. J Gen Physiol. 1989; 93: 841–854.
Marx SO, Kurokawa J, Reiken S, Motoike H, D’Armiento J, Marks AR, Kass RS. Requirement of a macromolecular signaling complex for beta adrenergic receptor modulation of the KCNQ1-KCNE1 potassium channel. Science. 2002; 295: 496–499.
Bouchard RA, Clark RB, Giles WR. Effects of action potential duration on excitation-contraction coupling in rat ventricular myocytes: action potential voltage-clamp measurements. Circ Res. 1995; 76: 790–801.
Yus-Najera E, Santana-Castro I, Villarroel A. The identification and characterization of a noncontinuous calmodulin-binding site in noninactivating voltage-dependent KCNQ potassium channels. J Biol Chem. 2002; 277: 28545–28553.
Dilly KW, Kurokawa J, Terrenoire C, Reiken S, Lederer WJ, Marks AR, Kass RS. Overexpression of beta2-adrenergic receptors cAMP-dependent protein kinase phosphorylates and modulates slow delayed rectifier potassium channels expressed in murine heart: evidence for receptor/channel co-localization. J Biol Chem. 2004; 279: 40778–40787.