Constitutive Phosphodiesterase Activity Restricts Spontaneous Beating Rate of Cardiac Pacemaker Cells by Suppressing Local Ca2+ Releases
Spontaneous beating of rabbit sinoatrial node cells (SANCs) is controlled by cAMP-mediated, protein kinase A–dependent local subsarcolemmal ryanodine receptor Ca2+ releases (LCRs). LCRs activated an inward Na+/Ca2+ exchange current that increases the terminal diastolic depolarization rate and, therefore, the spontaneous SANC beating rate. Basal cAMP in SANCs is elevated, suggesting that cAMP degradation by phosphodiesterases (PDEs) may be low. Surprisingly, total suppression of PDE activity with a broad-spectrum PDE inhibitor, 3′-isobutylmethylxanthine (IBMX), produced a 9-fold increase in the cAMP level, doubled cAMP-mediated, protein kinase A–dependent phospholamban phosphorylation, and increased SANC firing rate by ≈55%, indicating a high basal activity of PDEs in SANCs. A comparison of specific PDE1 to -5 inhibitors revealed that the specific PDE3 inhibitor, milrinone, accelerated spontaneous firing by ≈47% (effects of others were minor) and increased amplitude of L-type Ca2+ current (ICa,L) by ≈46%, indicating that PDE3 was the major constitutively active PDE in the basal state. PDE-dependent control of the spontaneous SANC firing was critically dependent on subsarcolemmal LCRs, ie, PDE inhibition increased LCR amplitude and size and decreased LCR period, leading to earlier and augmented LCR Ca2+ release, Na+/Ca2+ exchange current, and an increase in the firing rate. When ryanodine receptors were disabled by ryanodine, neither IBMX nor milrinone was able to amplify LCRs, accelerate diastolic depolarization rate, or increase the SANC firing rate, despite preserved PDE inhibition–induced augmentation of ICa,L amplitude. Thus, basal constitutive PDE activation provides a novel and powerful mechanism to decrease cAMP, limit cAMP-mediated, protein kinase A–dependent increase of diastolic ryanodine receptor Ca2+ release, and restrict the spontaneous SANC beating rate.
The sinoatrial (SA) node is the primary physiological pacemaker of the heart. The pacemaker action potential (AP) is initiated within the SA node center and then propagates to the atria and ventricle to initiate contraction.1,2 Our recent studies have demonstrated that spontaneous firing of SA node pacemaker cells (SANCs) is controlled by local subsarcolemmal Ca2+ releases (LCRs) from ryanodine receptors (RyR) that occur during the second half of spontaneous diastolic depolarization (DD), just prior to the AP upstroke.3 LCRs activate inward Na+/Ca2+ exchange (NCX) current that accelerates the rate of DD, leading to an earlier occurrence of the subsequent spontaneous AP, ie, to an increase in the beating rate.3 Although LCRs do not require membrane depolarization,4 they are critically dependent on levels of cAMP and cAMP-mediated, protein kinase (PKA)-dependent phosphorylation, both of which are markedly higher in SANCs than in atrial or ventricular myocytes, because of constitutive activation of adenylyl cyclases (ACs).5
The level of cAMP in cells is a result of a balance between synthesis by ACs and degradation by cyclic nucleotide phosphodiesterases (PDEs), which provide the only known mechanism for degrading cAMP.6 The high basal cAMP in SANCs suggests that basal PDE activity and cAMP degradation may be low. However, the efficiency of PDE-dependent control of the basal cAMP level in SANCs has never been tested. Although an increase in the heart beating rate after suppression of PDE activity has been noted in clinical studies, and in studies of isolated hearts,7–9 neither the efficacy of the PDE-dependent control of the spontaneous SANC beating rate nor specific mechanisms have been investigated. The aims of this study were to determine (1) how effectively PDEs degrade cAMP and control spontaneous SANC beating rate in the basal state and (2) the specific mechanisms of this PDE-dependent control of spontaneous firing. We also compared effects of PDE inhibition with effects of β-adrenergic receptor (AR) stimulation, which is thought to be the most potent signaling pathway to increase the spontaneous SANC beating rate.
Our results showed, for the first time, that even in the basal state, PDEs very effectively degraded cAMP, reduced phospholamban (PLB) phosphorylation, and controlled spontaneous SANC beating rate. Surprisingly, these effects markedly exceeded those induced by β-AR stimulation by isoproterenol (ISO). LCRs via RyR were critically involved in the PDE-dependent control of the spontaneous beating rate. When RyR were inhibited by ryanodine, PDE inhibition failed to amplify local Ca2+ release during late DD and failed to increase spontaneous firing of SANCs. Thus, high basal constitutive PDE activity in SANCs coexists with constitutively active ACs, providing a negative feedback on the latter to limit cAMP level. This leads to a suppression of basal LCRs during DD that acts as a brake to keep the basal spontaneous SANC firing under control.
Materials and Methods
An expanded Materials and Methods section is in the online data supplement, available at http://circres.ahajournals.org.
SA Node Cell Preparations and Electrophysiological Recordings
Animals were treated in accordance with the NIH Guide for the Care and Use of Laboratory Animals. Single spindle-shaped spontaneously beating SANCs were isolated from rabbit hearts. Perforated or ruptured patch-clamp techniques were used to record APs or currents from spontaneously beating SANCs. The bath temperature was maintained at 35±0.5°C.
Confocal Imaging of LCRs
SANCs were loaded with fluo-3 acetoxymethyl ester (Molecular Probes, Eugene, Ore). All images were recorded in line scan mode, using confocal microscopy as previously described.4,5
A subset of SANCs was permeabilized with 0.01% saponin as previously described.4,5
The detection of Ser16 PLB phosphorylation was performed in isolated SANCs using a phosphorylation P-Ser-16 PLB antibody (Badrilla) as previously described.5
SA nodal or ventricular cells were homogenized, and cAMP was estimated using the Biotrak cAMP [125I] assay system (RPA 509; Amersham Biosciences).
Data are presented as means±SEM. The statistical significance of effects was evaluated by Student t test or ANOVA where appropriate. A value of P<0.05 was considered statistically significant.
Effects of PDE Suppression on the Level of cAMP and Spontaneous Firing Rate of SANCs
To determine the extent of PDE-dependent control of basal cAMP level in isolated rabbit SANCs, we used a broad-spectrum PDE inhibitor, 3′-isobutylmethylxanthine (IBMX), which produced a 9-fold increase in cAMP, an effect larger than that of a saturating ISO concentration (Figure 1A). In spite of a high basal level of PDE activity, basal cAMP in isolated SANCs was substantially higher than in ventricular myocytes, confirming a prior study.5
Suppression of total PDE activity with IBMX dramatically accelerated SANC spontaneous beating rate (by 55%, Figure 1B), and this effect was reversible after drug washout. Analysis of AP parameters showed that IBMX-induced acceleration was accompanied by a marked increase in the DD rate (62.2±6.8 mV/sec in control versus 116.7±6.7 mV/sec with IBMX; n=7), whereas the maximum diastolic potential (−63.7±2.3 in control versus −64.1±2.8 mV with IBMX; n=7), AP amplitude (97.9±2.7 in control versus 96.8±3.4 mV with IBMX; n=7), and AP upstroke (6.6±0.7 in control versus 6.9±0.7 V/sec with IBMX; n=7) were unchanged. To verify the direct effect of cAMP on SANC spontaneous beating, we used a membrane-permeable cAMP analog, 8-(4-chlorophenylthio)-cAMP (CPT-cAMP), which increased spontaneous beating rate by 36%, confirming a critical role of elevated cAMP to increase the spontaneous SANC beating rate (Figure 1C).
PDE3 Controls SANC Spontaneous Beating Rate in the Basal State
The PDE superfamily consists of 11 families, and PDE1 to -5 are present in the heart.10 To determine which PDE subtypes mediate IBMX-induced increase in the spontaneous SANC beating rate, we used specific inhibitors of different PDE subtypes (see the online data supplement). Figure 2 shows that on average, the increase in the spontaneous beating rate produced by suppression of PDE1, -2, -4, or -5 was relatively small, whereas suppression of PDE3 by the specific PDE3 inhibitor milrinone substantially accelerated spontaneous firing, producing an effect almost equal to the effect of nonspecific PDE inhibitor IBMX. These data strongly suggest that constitutive activation of PDE3 is a major contributor to the total basal PDE activity. The positive chronotropic effect of IBMX markedly exceeded the effect of the saturating concentration of β-adrenergic agonist ISO (P<0.05). Although milrinone-induced acceleration of the beating rate was ≈12%, larger than that produced by ISO, it did not reach statistical significance (Figure 2).
Positive Chronotropic Effects of PDE Inhibition Are Critically Dependent on Local RyR Ca2+ Releases
Our prior work has demonstrated that normal spontaneous beating of SANCs is critically dependent on characteristics of LCRs, which are potently modulated by cAMP mediated PKA-dependent phosphorylation.3,5 To investigate how LCRs are affected by PDE inhibition, we used ryanodine, which locks RyRs in a subconductance open state and depletes the sarcoplasmic reticulum (SR) Ca2+ load. Consistent with our previous results,3 ryanodine inhibition of LCRs and reduction in the firing rate occurred concomitantly. Figure 3 shows that when RyRs were functionally disabled by ryanodine, a suppression of either PDE3 activity with milrinone or total PDE activity with IBMX induced only a minor increase in the spontaneous beating rate. Compared with control, acceleration of the beating rate by IBMX or milrinone in the presence of ryanodine was decreased by ≈14-fold. These results suggest that an essential feature of PDE inhibition–induced acceleration of the rabbit SANC firing rate is to affect Ca2+ release via RyR.
Ca2+ influx via L-type Ca2+ current (ICa,L) is a crucial component of Ca2+ cycling that sustains LCRs in SANCs. To verify whether the aforementioned effects of ryanodine to suppress the milrinone-induced increase in SANC spontaneous beating rate concomitantly suppress ICa,L, we recorded effects of milrinone on ICa,L in the presence of ryanodine. Consistent with our previous data,11 ryanodine pretreatment, per se, had no effect on the average ICa,L amplitude (Figure 4). PDE3 inhibition by milrinone increased ICa,L amplitude by ≈45% (Figure 4A). Moreover, the milrinone-induced increase in the ICa,L amplitude was not affected by ryanodine pretreatment (Figure 4). These data demonstrate that, even when suppression of PDE3 activity with milrinone markedly increased Ca2+ influx via increase of ICa,L amplitude, the positive chronotropic effect of milrinone was markedly suppressed if RyRs were disabled and LCRs were inhibited by ryanodine.
PDE Inhibition Modulates Spatiotemporal Characteristics of LCR
To determine how PDE inhibition specifically modifies spatiotemporal characteristics of LCRs to augment SANC firing, we used confocal microscopy to simultaneously measure LCR and APs in intact SANCs. Figure 5 shows that following PDE3 inhibition, there was a shift in the distribution of LCRs amplitudes (Figure 5B) and spatial widths (Figure 5C) to larger values. Milrinone increased the average LCR amplitude from 0.83±0.07 to 1.05±0.08 ΔF/F0 (n=6, P<0.05) and width from 6.51±0.32 to 8.46±0.21 μm (full width at half maximum; n=6, P<0.01). Suppression of total PDE activity with IBMX also markedly increased average LCR amplitude from 0.78±0.03 to 0.89±0.04 ΔF/F0 (n=4, P<0.05) and size from 5.66±0.56 to 8.32±0.73 μm (full width at half maximum; n=4, P<0.05). These data indicate that following PDE inhibition, LCRs are amplified, in part, by recruitment of additional RyRs to contribute to the local Ca2+ release.
Because the LCR period regulates the spontaneous cycle length and beating rate,4,5 we studied how LCR period is affected by PDE inhibition. During spontaneous beating, regularly occurring AP-induced SR Ca2+ releases depleted SR and inactivated RyR. When the SR content was replenished and RyR recovered from inactivation, LCR occurrence resumed. Therefore, the LCR period was estimated as an interval between the onset of AP-induced Ca2+ transient and the onset of subsequent spontaneous LCR (Figure 5A, inset). Both milrinone (Figure 5D) and IBMX (data not shown) induced a marked decrease in the LCR period, which was highly correlated with a concomitant decrease in the spontaneous cycle length.
Electrogenic Mechanisms of PDE Inhibition–Produced Acceleration of the SANC Spontaneous Beating Rate
It is well recognized that an increase in cAMP in SANCs directly activates the inward hyperpolarization activated current (If), which controls the early phase of the DD.12 To directly probe the If contribution in the PDE inhibition–induced acceleration of spontaneous SANC firing, effects of IBMX and milrinone were compared in the absence and presence of 2 mmol/L Cs+, which effectively blocks If.13 The PDE inhibition accelerated the SANC spontaneous beating rate to almost the same extent in the presence and absence of Cs+ (Figure 6A and 6B). This result was consistent with the idea that If plays a minor, if any, role in the PDE inhibition induced acceleration of the spontaneous SANC beating rate.
To elucidate additional ryanodine-sensitive electrogenic mechanisms involved in PDE inhibition–induced increase in DD rate and spontaneous beating rate, we investigated effects of the broad-spectrum PDE inhibitor IBMX on NCX current and delayed rectifier potassium (IK) current. Because LCRs activate an inward NCX current,3 IBMX-induced amplification of local Ca2+ releases would be expected to augment this current and increase DD rate. When LCRs are inhibited by ryanodine, PDE inhibition–induced increase of NCX current as well as increase in the DD rate would be suppressed. To test this idea, we simulated DD by a voltage ramp protocol and applied IBMX in the absence and presence of ryanodine. IBMX markedly increased the inward current developed during voltage ramp by 107%, from 0.73±0.08 to 1.51±0.20 pA/pF (n=3; Figure 6C and 6D). Ryanodine significantly decreased the inward current, from 0.80±0.10 to 0.52±0.10 pA/pF (P<0.01; n=6), and caused a more than 3-fold reduction in the IBMX-induced amplification of this current. In fact, the IBMX-induced increase in this inward current in the presence of ryanodine did not reach statistical significance (Figure 6C and 6D).
IK current was measured as the amplitude of the outward tail current (IK,tail), and effects of IBMX on IK,tail were studied in the absence and presence of ryanodine. IBMX produced an ≈5-mV shift of IK,tail activation in the negative direction (Figure II in the online data supplement) and an ≈12% increase in IK,tail current amplitude (supplemental Figure III). However, both the shift of IK,tail activation and increase in IK,tail current amplitude were preserved in the presence of ryanodine (supplemental Figures II and III), suggesting that modulation of IK by PDE inhibition is not affected by ryanodine pretreatment.
PDE Inhibition in SANCs Increases PLB Phosphorylation and Accelerates the Kinetics of SR Ca2+ Uptake
Although basal cAMP was high in SANCs, the link of cAMP to the spontaneous beating was not direct but required PKA-dependent phosphorylation of multiple proteins controlling Ca2+ homeostasis, including ICa,L, and proteins controlling SR Ca cycling, ie, RyR and PLB.5 We used the phosphorylation status of PLB as a marker for PKA-dependent protein phosphorylation. Figure 7 shows that PLB phosphorylation at the PKA-dependent serine 16 site was markedly increased by both the broad-spectrum PDE inhibitor IBMX and the PDE3 inhibitor milrinone and was reversed by a specific peptide inhibitor of the PKA catalytic subunit 14-22 amide, PKI.
It is well known that markedly elevated PLB phosphorylation increases the SR Ca2+ pump rate and thus the rate at which Ca2+ is pumped back into SR. This effect results in a more rapid decay of the AP-induced Ca2+ transient.14 In intact SANCs, milrinone markedly decreased 90% decay time of AP-induced Ca2+ transient from 298.8±15.4 to 223.4±9.1 ms (P<0.001; n=6) and increased the average amplitude of AP-induced Ca2+ transient from 1.53±0.10 to 1.65±0.12 F/F0 (P<0.02; n=6), which was consistent with PLB phosphorylation–induced acceleration in SR Ca2+ pumping.
In Permeabilized SANCs, PDE3 Inhibition Modulates LCR Parameters and Increases SR Ca2+ Load
In SR-enriched microsomes of rabbit SA node, PDE3 activity is the major (≈75% of total) PDE activity, suggesting close association of PDE3 with SR.15 To verify direct effects of PDE3 inhibition on the LCR parameters in the absence of spontaneous APs, avoiding concomitant effects of sarcolemmal ionic currents, we permeabilized SANCs with saponin. Similar to its effect in intact SANCs, suppression of PDE3 activity by milrinone markedly increased the likelihood of LCR occurrence and partially synchronized their initiation by increasing number and size of LCRs in “skinned” SANCs (supplemental Figure IVA and IVB). To determine whether this PDE inhibition–induced augmentation of local Ca2+ releases in skinned SANCs was caused, at least in part, by stimulation of SR Ca2+ uptake and increase in the SR Ca2+ load, we applied a pulse of caffeine directly on the SANCs to rapidly empty the SR Ca2+ store. Representative images and average data (supplemental Figure VA and VB) showed that, indeed, following milrinone, there was a significant increase in the SR Ca2+ load by ≈12%.
PDE Inhibition Effects on PLB Phosphorylation and LCR Period Are Tightly Linked to Effects on the Spontaneous Beating Rate
The effects of cAMP PKA-dependent stimulation on the LCR period are likely mediated by an increase in phosphorylation of multiple proteins that regulate cell Ca2+ balance, including L-type Ca2+ channels, PLB phosphorylation, and probably that of other proteins not measured in the present study. Figure 8B shows that the increase of cAMP PKA-dependent protein phosphorylation produced by PDE inhibition or by β-AR stimulation, and indexed by PLB phosphorylation at serine 16, was highly correlated with the concomitant decrease of the LCR period. Thus, interventions that use different pathways to increase the cAMP level (activation of ACs or inhibition of PDEs) are linked to a common mechanism, ie, an increase of cAMP PKA-dependent protein phosphorylation to decrease the LCR period and spontaneous cycle length (Figure 8C). The dependence of the spontaneous cycle length on cAMP/PKA-dependent effects on the LCR period produced by β-AR stimulation or PDE inhibition form a single continuous function.
PDEs Control cAMP Level and SANC Spontaneous Beating in the Basal State
The first novel finding of the present study was that intrinsic PDE activity potently controlled the basal cAMP level and spontaneous beating rate of SANCs. A suppression of the total PDE activity by the broad-spectrum PDE inhibitor IBMX led to a 9-fold increase in the level of cAMP (Figure 1A) and an ≈55% increase in the spontaneous beating rate of SANCs. Thus, high cAMP production by AC in SANCs coexists with high cAMP degradation by PDEs, demonstrating a unique type of balance between former and latter. Moreover, the system acting in this mode can rapidly react to stimuli that alter cAMP without involvement of receptor-dependent mechanisms. Remarkably, effects of total PDE suppression on both cAMP and the spontaneous beating rate exceeded effects of the saturating concentration of β-AR agonist ISO (Figure 2), which may be attributable to the more efficient cAMP degradation by PDEs than cAMP production triggered by β-AR stimulation.
PDE3 Is the Major Constitutively Active PDE That Controls Spontaneous SANC Firing
The second discovery of the present study was that PDE3 was the major PDE subtype that controlled basal spontaneous SANC firing, whereas contribution of other PDEs was relatively small (Figure 2). It is known that PDE1 and PDE2 can hydrolyze both cAMP and cGMP and that PDE3 preferentially hydrolyzes cAMP, whereas PDE4 is specific for cAMP and PDE5 is specific for cGMP.10 Our data are consistent with the species-dependent variation in the activity of different PDE subtypes in the heart. For example, in the murine ventricular cells, PDE4 is the dominant PDE subtype, accounting for 60% of total cAMP hydrolyzing activity.16 In both rabbit SA node and ventricle, PDE3 is the most abundant subtype, accounting for ≈75% of total cAMP-hydrolyzing activity in microsomal fraction (PDEs attached to membranes) and ≈30% in cytosol.15 Importantly, an inhibition of an SR-associated PDE3 subtype is the key factor of PDE3 inhibition–induced augmentation of contractile responses in myocardium.16–18 Immunolabeling patterns of RyR clearly demonstrate a high density of RyR in SANCs beneath sarcolemma.19 The striking effects of milrinone on local subsarcolemmal Ca2+ releases both in intact and permeabilized SANCs (Figure 5 and supplemental Figure IV) could be explained by strategically positioned PDE3 in this region, which could control local cAMP level and PKA-dependent phosphorylation in the vicinity of sarcolemma and SR. Specifically, suppression of PDE3 activity markedly increased PKA-dependent phosphorylation of PLB (Figure 7), amplified SR Ca2+ ATPase pumping rate, and increased SR Ca2+ load (supplemental Figure V) in the subsarcolemmal SR.
PDE-Dependent Regulation of LCRs Provides Control Over Spontaneous Beating of SANCs
The third and probably the most important finding of the present study was the mechanism of the PDE-dependent control of the spontaneous firing of SANCs. LCRs in SANCs, as assessed by line scan3,4 or 2D images,20 are localized to the subsarcolemmal area. LCRs occur during terminal DD and begin as spark-like events; with time, they converge to become wavelets3,20 that propagate locally with a velocity of 156±14 μm/sec and to a distance up to 11.5±1.2 μm,20 which is followed by a steep AP-induced rise in global Ca2+. Whereas confocal images of the present study (Figures 5 and 8⇑) provided only 1 line scan image and showed LCRs in a single location, 2D images show that multiple LCR occur simultaneously throughout SANCs.20 Numeric model simulations show that each LCR activates an inward NCX current equal to ≈0.27 pA that results in a membrane potential response of ≈0.17 mV.21 Experimental measurements of the present study showed that ryanodine-susceptible NCX current was equal to ≈0.28±0.03 pA/pF, which resulted in a total current equal to ≈9 pA (per 32 pF SANCs) and was in a good agreement with numeric simulations of ≈10 pA.21
In the present study, PDE inhibition increased LCR amplitude and size, amplified subsarcolemmal Ca2+ release during DD, and shifted it to earlier times, which could be partially explained by an increased and earlier spontaneous RyR Ca2+ release flux attributable to an increase in PLB phosphorylation, increase in SR Ca2+ ATPase pumping rate, and increase in SR Ca2+ load (Figure 7 and supplemental Figure V). This amplification of subsarcolemmal Ca2+ release during DD led to a 2-fold increase of the inward NCX current (Figure 6D), which “boosted” DD rate and, as a result, increased the spontaneous beating rate (Figures 5 and 8⇑). In the presence of ryanodine, this current was markedly suppressed, and PDE inhibition failed to substantially increase either NCX current (Figure 6D) or the spontaneous SANC beating rate (Figure 3).
Thus, intact RyR function was critically important for PDE inhibition–induced acceleration of SANC spontaneous beating. When RyR local subsarcolemmal Ca2+ release was inhibited by ryanodine, PDE inhibition was unable to increase LCRs, NCX current (Figure 6C and 6D), or spontaneous beating rate (Figure 3). In the presence of ryanodine, the positive chronotropic effect of PDE inhibition was suppressed in spite of preserved PDE inhibition–induced average increase of the ICa,L amplitude (Figure 4) and IK amplitude (supplemental Figure III). Moreover, our data clearly demonstrate that in our experimental conditions, If current likely had, at best, a minor role in IBMX or milrinone-induced acceleration of spontaneous SANC firing.
Different Signaling Pathways Use cAMP-Mediated, PKA-Dependent Modulation of LCRs to Control the Spontaneous SANC Beating Rate
The PDE-dependent control of LCRs and basal spontaneous beating rate uses the very same mechanism that is used by either β-AR stimulation11 or membrane-permeable cAMP analog to accelerate spontaneous SANC firing.5 All of these interventions directly or indirectly (1) elevate level of cAMP; (2) increase cAMP-mediated, PKA-dependent phosphorylation of Ca2+cycling proteins; (3) augment LCRs; (4) amplify NCX current and increase terminal DD rate; and (5) accelerate spontaneous beating rate. Thus, regulation of local RyR Ca2+ release plays a central role in the control of spontaneous SANC firing, regardless of whether intracellular cAMP is increased via activation of AC, as during β-AR stimulation, or by suppression of cAMP degradation, as during PDE inhibition. Intact RyR function is essential for the chronotropic effect of all aforementioned interventions, because inhibition of RyR Ca2+ release by ryanodine prevents increase in the SANC spontaneous beating rate produced by β-AR stimulation,11,22,23 membrane permeable cAMP analog,5 or PDE inhibition (present study).
Previous studies in intact animals, isolated Langendorff heart preparations,7–9 and clinical trails have observed the positive chronotropic effect of PDE inhibition. However, complex effects of PDE inhibitors on both myocardium and blood vessels in vivo precluded a direct demonstration of PDE-dependent effects on the cardiac pacemaker function. Studies of the isolated SA node24 and of isolated SANCs25 have examined the effects of some PDE3 inhibitors on AP parameters24 and sarcolemmal ionic currents,25 respectively. In guinea pig SANCs, amrinone, another PDE3 inhibitor, increased the firing rate by ≈12%,25 compared with ≈47% increase in the rabbit SANC beating rate by milrinone (this study). This difference could be explained, in part, by the different species used and, in part, by the fact that milrinone is regarded as a more potent PDE3 inhibitor.7,8 Moreover, only milrinone effectively inhibits PDE3 activity in SA node–enriched fraction,26 whereas other PDE3 inhibitors have minor effects, despite potent inhibition of PDE3 activity in ventricular tissue.26
In summary, the present data show, for the first time, that basal level of cAMP-mediated, PKA-dependent phosphorylation and basal spontaneous beating rate of SANCs are under a tight restraint of constitutively active PDEs, with PDE3 as the major subtype. This PDE-dependent control was mostly executed through modulation of local RyR Ca2+ release. When PDEs were inhibited, the increase in cAMP activated cAMP-mediated, PKA-dependent phosphorylation of PLB and L-type Ca2+ channel. Acceleration of SR Ca2+ pumping and elevation of SR Ca2+ load increased LCR amplitude and size by partial synchronization and recruitment of additional RyR Ca2+ releases. The earlier occurring and amplified Ca2+ release beneath sarcolemma augmented the inward NCX current and accelerated DD rate and, as a result, spontaneous SANC beating rate. When RyR Ca2+ release was suppressed by ryanodine, efficient PDE-dependent control of the spontaneous beating rate was abolished, in spite of preserved PDE inhibition–induced augmentation of ICa,L and IK. Thus, constitutively active PDEs via reduction of cAMP PKA-dependent protein phosphorylation restrict local RyR Ca2+ release during DD to keep the basal spontaneous SANC firing under control. Suppression of PDE activity is a novel strategy to increase spontaneous beating rate without involvement of β-AR stimulation. It offers a new direction for using PDE-dependent control of SA node beating rate as a tool to adjust cardiac pacemaker function according to the requirements of aging or the diseased heart.
We are grateful to Dr Victor A. Maltsev for providing a program for computer analysis of AP parameters.
Sources of Funding
This work was supported, in part, by Intramural Research Program of the National Institute on Aging, NIH.
Original received August 10, 2007; revision received January 30, 2008; accepted February 4, 2008.
Bleeker WK, Mackaay AJ, Masson-Pevet M, Bouman LN, Becker A. Functional and morphological organization of the rabbit sinus node. Circ Res. 1980; 46: 11–22.
Boyett MR, Honjo H, Kodama I. The sinoatrial node, a heterogeneous pacemaker structure. Cardiovasc Res. 2000; 47: 658–687.
Bogdanov KY, Vinogradova TM, Lakatta EG. Sinoatrial nodal cell ryanodine receptor and Na-Ca exchanger: molecular partners in pacemaker regulation. Circ Res. 2001; 88: 1254–1258.
Vinogradova TM, Zhou Y-Y, Maltsev V, Lyashkov A, Stern M, Lakatta EG. Rhythmic ryanodine receptor Ca2+ releases during diastolic depolarization of sinoatrial pacemaker cells do not require membrane depolarization. Circ Res. 2004; 94: 802–809.
Vinogradova TM, Lyashkov AE, Zhu W, Ruknudin AM, Sirenko S, Yang D, Deo S, Barlow M, Johnson S, Caffrey JL, Zhou YY, Xiao RP, Cheng H, Stern MD, Maltsev VA, Lakatta EG. High basal protein kinase A-dependent phosphorylation drives rhythmic internal Ca2+ store oscillations and spontaneous beating of cardiac pacemaker cells. Circ Res. 2006; 98: 505–514.
Maurice DH, Palmer D, Tilley DG, Dunkerley HA, Netherton SJ, Raymond DR, Elbatarny HS, Jimmo SL. Cyclic nucleotide phosphodiesterase activity, expression, and targeting in cells of the cardiovascular system. Mol Pharmacol. 2003; 64: 533–546.
Vinogradova TM, Bogdanov KY, Lakatta EG. beta-Adrenergic stimulation modulates ryanodine receptor Ca2+ release during diastolic depolarization to accelerate pacemaker activity in rabbit sinoatrial nodal cells. Circ Res. 2002; 90: 73–79.
Bers DM. Regulation of the SR Ca-pump by phospholamban. In: Bers DM. Excitation-contraction Coupling and Cardiac Contractile Force. Boston, Mass: Kluwer; 2001: 164–169.
Bender AT, Beavo JA. Cyclic nucleotide phosphodiesterases: molecular regulation to clinical use. Pharmacol Rev. 2006; 58: 488–520.
Lyashkov AE, Juhaszova M, Dobrzynski H, Vinogradova TM, Maltsev VA, Juhasz O, Spurgeon HA, Sollott SJ, Lakatta EG. Calcium cycling protein density and functional importance to automaticity of isolated sinoatrial nodal cells are independent of cell size. Circ Res. 2007; 100: 1723–1731.
Parsons SP, Vinogradova TM, Lyashkov AE, Spurgeon HA, Stern MD, Lakatta EG, Maltsev AV. Local Ca2+ waves rather than sparks boost the subsarcolemmal Ca2+ rise in late diastole of rabbit sinoatrial node cells: characterization in two dimensions. Biophysical J. 2007; 76a Abstract.
Bogdanov KY, Maltsev VA, Vinogradova TM, Lyashkov AE, Spurgeon HA, Stern MD, Lakatta EG. Membrane potential fluctuations resulting from submembrane Ca2+ releases in rabbit sinoatrial nodal cells impart an exponential phase to the late diastolic depolarization that controls their chronotropic state. Circ Res. 2006; 99: 979–987.
Rigg L, Heath BM, Cui Yi, Terrar DA. Localization and functional significance of ryanodine receptors during β-adrenergic stimulation in guinea-pig sino-atrial node. Cardiovasc Res. 2000; 48: 254–264.