Increased Sarcoplasmic Reticulum Calcium Leak but Unaltered Contractility by Acute CaMKII Overexpression in Isolated Rabbit Cardiac Myocytes
The predominant cardiac Ca2+/calmodulin-dependent protein kinase (CaMK) is CaMKIIδ. Here we acutely overexpress CaMKIIδC using adenovirus-mediated gene transfer in adult rabbit ventricular myocytes. This circumvents confounding adaptive effects in CaMKIIδC transgenic mice. CaMKIIδC protein expression and activation state (autophosphorylation) were increased 5- to 6-fold. Basal twitch contraction amplitude and kinetics (1 Hz) were not changed in CaMKIIδC versus LacZ expressing myocytes. However, the contraction–frequency relationship was more negative, frequency-dependent acceleration of relaxation was enhanced (τ0.5Hz/τ3Hz=2.14±0.10 versus 1.87±0.10), and peak Ca2+ current (ICa) was increased by 31% (−7.1±0.5 versus −5.4±0.5 pA/pF, P<0.05). Ca2+ transient amplitude was not significantly reduced (−27%, P=0.22), despite dramatically reduced sarcoplasmic reticulum (SR) Ca2+ content (41%; P<0.05). Thus fractional SR Ca2+ release was increased by 60% (P<0.05). Diastolic SR Ca2+ leak assessed by Ca2+ spark frequency (normalized to SR Ca2+ load) was increased by 88% in CaMKIIδC versus LacZ myocytes (P<0.05; in an multiplicity-of-infection–dependent manner), an effect blocked by CaMKII inhibitors KN-93 and autocamtide-2–related inhibitory peptide. This enhanced SR Ca2+ leak may explain reduced SR Ca2+ content, despite measured levels of SR Ca2+-ATPase and Na+/Ca2+ exchange expression and function being unaltered. Ryanodine receptor (RyR) phosphorylation in CaMKIIδC myocytes was increased at both Ser2809 and Ser2815, but FKBP12.6 coimmunoprecipitation with RyR was unaltered. This shows for the first time that acute CaMKIIδC overexpression alters RyR function, leading to enhanced SR Ca2+ leak and reduced SR Ca2+ content but without reducing twitch contraction and Ca2+ transients. We conclude that this is attributable to concomitant enhancement of fractional SR Ca2+ release in CaMKIIδC myocytes (ie, CaMKII-dependent enhancement of RyR Ca2+ sensitivity during diastole and systole) and increased ICa.
Ca2+/calmodulin-dependent protein kinase II (CaMKII) is a multifunctional serine/threonine protein kinase that phosphorylates numerous target proteins.1,2 The major cardiac isoform is CaMKIIδ, and the splice variant CaMKIIδC is primarily cytosolic, whereas CaMKIIδB is nuclear because of a nuclear localization sequence.3
During excitation–contraction coupling (ECC), Ca2+ entry, mainly via voltage dependent L-type Ca2+ channels (ICa), triggers sarcoplasmic reticulum (SR) Ca2+ release via ryanodine receptors (RyRs), via Ca2+-induced Ca2+ release.4,5 The resultant increase in intracellular [Ca2+] ([Ca2+]i), causes Ca2+ binding to troponin C, which activates myofilaments, leading to contraction. For relaxation to occur, Ca2+ must be removed from the cytoplasm. SR Ca-ATPase (SERCA) and Na+/Ca2+-exchanger (NCX) are the main mechanisms for Ca2+ removal.4,5 CaMKII can modulate ECC by phosphorylating several important Ca2+-dependent regulatory proteins in heart, including Ca2+ transport proteins, such as RyR and phospholamban (PLB), and possibly L-type Ca2+ channels.2,5
CaMKII is directly associated with RyR and overexpression of CaMKII in transgenic mouse cardiomyocytes increases SR Ca2+ release as shown by increased frequency of spontaneous SR Ca2+ release events (Ca2+ sparks).6,7 Blocking CaMKII (using KN-93) decreases Ca2+ spark frequency dramatically, providing evidence for a direct relationship between CaMKII activity and the increased spark frequency.7 These results in myocytes from CaMKII transgenic mouse hearts were confirmed by Currie et al,8 who showed that the specific CaMKII peptide inhibitor autocamtide-2–related inhibitory peptide (AIP) depresses Ca2+ spark frequency in rabbit hearts because of decreased endogenous CaMKII-dependent RyR phosphorylation. Wehrens et al9 showed using site-directed mutagenesis that CaMKII-dependent phosphorylation of RyR was at Ser2815, rather than at Ser2809 (which they find is a PKA target and phosphorylation causes FKBP12.6 dissociation). Using single-channel measurements in lipid bilayers, Wehrens et al also showed that CaMKII-dependent RyR phosphorylation increased RyR open probability (Po), without alteration of FKBP12.6 association.9
CaMKII may also be involved in the pathogenesis of hypertrophy and heart failure.2 In human heart failure, CaMKII expression are increased.10,11 In neonatal ventricular myocytes, overexpression of CaMKIIδB caused transcriptional activation of atrial natriuretic peptide gene expression (a hypertrophic signaling marker).12 Furthermore, overexpression of the cytoplasmic δC isoform in mouse heart results in profound contractile dysfunction and heart failure.6,7 In our previous studies using these animals, we described major alterations in intracellular Ca2+ handling with marked reductions in Ca2+ transients, SR Ca2+ content, and SERCA, PLB, and RyR protein expression and enhanced NCX function and expression, all of which are typical for heart failure. Most remarkably, however, with respect to the RyR, the frequency of Ca2+ sparks (indicative of diastolic spontaneous SR Ca2+ release events or opening of RyR clusters) was greatly enhanced, demonstrating increased diastolic SR Ca2+ leak despite reduced SR Ca2+ load and diastolic [Ca2+]i7 (which by themselves would normally reduce SR Ca2+ leak).13 We showed that this was most likely attributable to increased CaMKII-dependent RyR phosphorylation increasing RyR openings, because Ca2+ spark frequency could be reduced back to normal levels by blocking CaMKII. Backphosphorylation and subsequent studies using phospho-CaMKII antibodies indeed showed increased RyR phosphorylation in transgenic versus wild type.6,7
Although these results show that CaMKIIδC overexpression can cause heart failure and altered cellular Ca2+ transport, it was unclear how direct effects of acute CaMKII-dependent protein phosphorylation alter Ca2+ handling functionally and with respect to protein expression, especially in the context of possible developmental changes or adaptive responses associated with heart failure induction as reported previously.6,7 Therefore, we have acutely overexpressed CaMKIIδC in ventricular rabbit myocytes and compared these with LacZ-expressing control cells to investigate intracellular Ca2+ handling. We demonstrate that acute CaMKIIδC overexpression enhances SR Ca2+ leak and reduces SR Ca2+ content. However, in acute CaMKIIδC overexpression, we do not see alterations in the protein expression levels or function of NCX and SERCA (in striking contrast to the failing transgenic mice), and twitch contractions and Ca2+ transients are unaltered. This is attributed to an increased fractional SR Ca2+ release (and ICa), which may result from the same CaMKIIδC-dependent enhancement of RyR Ca2+ sensitivity that enhances diastolic SR Ca2+ leak.
Materials and Methods
Generating Adenoviral Vectors and Cardiac Myocyte Isolation
Adenoviral vectors were generated as published previously.14,15 For adenoviral transfection, ventricular myocytes from rabbit hearts (female Chinchilla Bastards; 1.3- to 2.0-kg weight) were isolated using standard procedures16,17 with collagenase B (0.5 mg/mL, Boehringer-Mannheim, Mannheim, Germany) and protease (0.02 mg/mL, Sigma, St Louis, Mo). Cells were plated at a density &4.2×103 rod-shaped cells/cm2 on culture dishes (55 mm) and incubated for 24 hours in supplemented M199 tissue culture medium (Sigma-Aldrich Chemie, Taufkirchen, Germany). All procedures involving animals were performed in compliance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH Publication No. 85-23, revised 1996). Initially, myocytes were transfected with multiplicities of infections (MOIs) of 1, 10, and 100 to test for CaMKIIδC protein overexpression and phosphorylation levels. For functional experiments, myocytes were then transfected with either CaMKIIδC, or LacZ using a MOI of 10 or 100 for 24 hours at 37°C in a humidified incubator (5% CO2, 95% O2). Myocyte volume was calculated from myocytes length×width×40% of width.7
Shortening and Ca2+ Measurements Using Inverted Microscopes
Shortening and [Ca2+]i measurements were performed as reported previously.7,17 Briefly, myocytes were loaded with the Ca2+-sensitive dye indo-1/acetoxymethyl ester to measure diastolic [Ca2+]. Because no difference in diastolic Ca2+ was found, additional experiments used fluo-3/acetoxymethyl ester (10 μmol/L, respectively; Molecular Probes). Excitation wavelengths (360±5 nm for indo and 480±15 nm for fluo) using a 75-W xenon arc lamp on the stage of a Nikon Eclipse TE200-U inverted microscope. Emitted fluorescence was measured using photomultipliers (at 405±15 nm and 485±12.5 nm for indo and 535±20 nm for fluo; IonOptix Corp, Milton, Mass). From the raw fluorescence, indo-1 ratio was calculated (405 nm/485 nm), and for fluo-3, F/F0 was calculated by dividing by the baseline fluorescence F0, after subtraction of the background fluorescence (IonWizard, IonOptix Corp). Myocytes were field-stimulated (voltage 25% above threshold) at 1 Hz and 37°C until steady state.
L-Type Ca2+ Current Measurements
ICa was recorded by voltage-clamp as reported previously.7 Briefly, low-resistance (&2 to 3 MΩ) electrodes were pulled and filled with K-free internal solution containing (in mmol/L) 105 CsCl, 20 HEPES, 5 BAPTA, 1 di-bromo-BAPTA, 1.49 CaCl2, and 5 MgATP resulting in a free [Ca2+] of 100 nmol/L (pH 7.2). Myocytes were superfused with K-free external solution containing (mmol/L) 140 NaCl, 4 CsCl, 5 HEPES, 10 glucose, 1 MgCl2, and 2 CaCl2 (pH 7.4). Current–voltage relationships were established as follows: holding potential was −90 mV; 5 prepulses were applied to 0 mV to ensure equal SR Ca2+ loading; brief Na+ current-inactivating pulses (50 ms, −50 mV) preceded test potentials steps (between −40 mV and +40 mV, 200 ms, in 10 mV steps). ICa facilitation was assessed by repetitive depolarizations to 0 mV after a pause of 1 minute, and amplitudes and kinetics were measured and analyzed (EPC10, Heka Electronics Inc, Lambrecht, Germany).
Ca2+ signals were recorded in fluo-4 loaded myocytes on a laser scanning confocal microscope (Bio-Rad Radiance 2000MP).7 Fluo-4 was excited via an argon laser (488 nm) and emitted fluorescence (F) was collected through a 515 nm long-pass emission filter. [Ca2+]i was calibrated by the equation [Ca2+]i=Kd(F/F0)/(Kd/[Ca2+]i-rest+1−F/F0) with Kd=1100 nmol/L and [Ca2+]i-rest=100 nmol/L.7
Ca2+ sparks were analyzed by a program (IDL 5.3)7 that detects Ca2+ sparks as areas of increased fluorescence with respect to the SD of background fluorescence. We used a Ca2+ spark threshold of 3.8×SD, with human verification. Peaks of Ca2+ sparks were normalized as F/F0, and duration was taken from the full-duration half-maximum (FDHM). Width or spatial size was taken as full-width half-maximum (FWHM). Ca2+ spark frequency (CaSpF) was obtained by averaging the number of sparks in images recorded after 1 Hz stimulation and normalized to cell volume and scan rate as sparks (pL−1s−1), assuming voxel length and width of 0.2 μm and depth of 1 μm.
Solutions and Experimental Protocol
Normal Tyrode’s solution contained (mmol/L) 140 NaCl, 6 KCl, 10 HEPES, 10 glucose, 1 MgCl2, and 2 CaCl2 (37°C). SR Ca2+ load was evaluated by Ca2+ transient amplitudes induced by rapid caffeine (10 mmol/L) application. NCX function was assessed measuring Ca2+ decay 50% relaxation (RT50%) after caffeine application. During this procedure, Ca2+ uptake by SERCA is prevented and other Ca2+ elimination pathways (eg, mitochondrial uniporter, sarcolemmal Ca2+ ATPase, contributing <1% each) can be neglected.5 In a subset of experiments, KN-93 (1 μmol/L) or the membrane permeant AIP (20 μmol/L) was added to the external bath solution or the patch-pipette to inhibit CaMKII. Enough time was allowed for KN-93 and AIP to inhibit CaMKII. In another subset of experiments, tetracaine was used to inhibit SR Ca2+ leak (1 mmol/L).
For shortening-frequency measurements, stimulation frequency was varied stepwise (from 0.5 to 3 Hz), waiting at intermediate frequencies until steady state was reached. For postrest measurements, a rest interval of 30 s was applied measuring steady-state and postrest twitch contraction amplitude. Of note, rabbit myocytes are known to show rest decay of twitches.5
Protein Expression, Phosphorylation Levels, and Immunocytochemistry
Western blot analysis was performed as described previously6,7 using an anti-CaMKIIδ antibody (Santa Cruz), as well as antibodies for SERCA, NCX (Affinity BioReagents), and PLB (Upstate). For phosphorylation levels of CaMKII (Affinity BioReagents), PLB-Thr17, and PLB-Ser16 (Cyclacel) in transfected myocytes, phospho-specific antibodies were used. RyR expression and phosphorylation levels were investigated using antibodies kindly provided by Dr A. Marks (Columbia University, New York).18 For immunohistochemical experiments (epifluorescence), diaminobenzidine staining was performed (picture plus, Zymed) using a hemagglutinin (HA) antibody (Roche) against, which was coexpressed with CaMKII. In parallel, a fluorescent Cy3-conjugated goat anti-mouse IgG (Jackson ImmunoResearch Laboratories Inc) against anti-HA was used for confocal images.
Coimmunoprecipitation and Immunoblotting
Coimmunoprecipitation studies were performed to test for FKBP12.6/RyR2 interaction. CaMKIIδC and LacZ transfected myocytes (5×105) were lysed in 250 μL of lysis buffer containing (in mmol/L) 50 HEPES, pH 7.4, 500 KCl, 1% Triton X-100, and 5 EDTA and supplemented with protease inhibitors (0.2 mmol/L pefabloc SC, 100 nmol/L aprotinin, 1 μmol/L leupeptin, 1 μmol/L pepstatin A, 1 mmol/L benzamidine, 1 μmol/L of calpain inhibitor I, and 1 μmol/L of calpain inhibitor II). After centrifugation for 5 minutes (10 000g, 4°C), cell lysates (1 mg) were suspended in PBS (1 mL). Eight micrograms of anti-RyR antibody (34°C clone, Affinity BioReagents) was added to the samples. After 4 hours of incubation at 4°C, protein G-sepharose beads were added to the samples, incubated for a further 4 hours at 4°C, and washed 3 times with PBS. Afterward, the beads were resuspended in SDS loading buffer and heated at 95°C for 5 minutes, and the eluates recovered by centrifugation were subjected to 4% to 20% linear gradient SDS-PAGE.
Immunoblotting was performed as described previously. Primary antibodies used were rabbit polyclonal anti-FKBP12.6 (1:500, SA-169; raised by Eurogentec, Hestal, Belgium), mouse monoclonal anti-RyR (1:500, C3–33 clone; Affinity BioReagents), and rabbit polyclonal anti–phospho-RyR (1:5000, RyR2-P2809 and RyR2-P2815; generous gifts of Dr A. Marks, Columbia University, New York). Secondary antibody used were donkey anti-rabbit whole Ig (1:10000, Amersham) and donkey anti-mouse affinity-purified IgG (1:500, Affinity BioReagents). Immunoreactive bands were visualized using SuperSignal West Pico Chemiluminescent Substrate (Pierce).
Results are expressed as mean±SEM. Significance (P<0.05) was determined using unpaired Student’s t test or 2-way repeated measurements ANOVA followed by Student–Newman–Keuls test as appropriate. Time constants of [Ca2+]i decline, τCa, were monoexponential least-square fits. Time constants τ1 and τ2 of ICa inactivation were fitted biexponentially.
CaMKII Overexpression in Ventricular Rabbit Myocytes
Figure 1A shows a typical CaMKII Western blot with rabbit myocytes infected at MOI100 (24 hours). Average overexpression of CaMKIIδ was &6-fold that of LacZ controls (P<0.05; n=5). In addition, myocytes showed MOI-dependent (1, 10, 100) increases in CaMKII protein expression, and also in CaMKII phosphorylation status, with maximum phosphorylation increases of &5-fold at MOI100 (P<0.05). Figure 1B shows that expressed CaMKIIδC is localized in the cytosol (no nuclear staining). Myocyte volume (Figure 1C) was not significantly increased in CaMKIIδC (41.6±2.5 pL; n=66) versus LacZ control (35.9±1.8 pL; n=154). Similarly, in voltage-clamped myocytes, membrane capacity was not different between LacZ (112.0±5.1 pF) and CaMKIIδC (113.5±6.9 pF). These results also suggest no alteration in surface to volume ratio.
Twitch Shortening and Ca2+ Transients
Twitch fractional shortening at 1 Hz (Figure 2A) was not altered in CaMKIIδC versus LacZ (3.6±0.2% versus 3.7±0.3% resting cell length; P=0.79). Ca2+ transients were also not significantly decreased (P=0.22; Figure 2B), and there was no change in diastolic [Ca2+]i as measured by indo-1 (diastolic F405/F485 was 0.46±0.02 versus 0.48±0.01; P=0.26). Twitch relaxation and [Ca2+]i decline were not changed, indicating unaltered basal SR Ca2+ ATPase function (particularly because NCX function was also unaltered; see below). However, KN-93 significantly prolonged relaxation (RT80%) at 1 Hz to 225±25 ms versus 187±16 ms (P<0.05). On increasing stimulation frequency from 0.5 to 3 Hz, there tended to be a greater decrease in contraction amplitude in CaMKIIδC-transfected myocytes (18%; n=46) versus LacZ myocytes (7%; n=29; P=NS). Frequency-dependent acceleration of relaxation (FDAR) (Figure 2D) was apparent in both CaMKIIδC and LacZ cells but was significantly enhanced in CaMKIIδC versus LacZ myocytes, consistent with a role for CaMKII in FDAR. The FDAR index τ0.5Hz/τ3Hz=2.14±0.10 in CaMKIIδC versus 1.87±0.10 in LacZ (P<0.05). KN-93 pretreatment partially inhibited FDAR in both myocyte types (not shown).
Figure 3A shows ICa–voltage relationships, where peak ICa was increased in CaMKIIδC versus LacZ by 31% at 0 mV (−7.1±0.5 versus −5.4±0.5 pA/pF; P<0.05). These effects can be significantly reversed by CaMKII inhibition with AIP (−5.8±0.5 pA/pF; P<0.05 versus CaMKIIδC). The ICa inactivation time constants τ1 and τ2 were significantly prolonged in CaMKIIδC versus LacZ myocytes and could be completely reversed by AIP (P<0.05 versus CaMKIIδC; Figure 3B). Repeated depolarization to 0 mV after a 1 minute rest causes a CaMKII-dependent ICa facilitation (Figure 3C) that was enhanced in CaMKIIδC versus LacZ myocytes (P<0.05). The CaMKII inhibitor KN-93 abolished facilitation in CaMKIIδC myocytes (P<0.05) and LacZ myocytes (not shown).
SR Ca2+ Content and NCX Function
Because no significant changes in twitch shortening or Ca2+ transients were found, SR Ca2+ load might be expected to be unchanged. However, SR Ca2+ content measured by caffeine-induced Ca2+ transients was dramatically reduced (by 41%) in CaMKIIδC versus LacZ (310±79 versus 521±120 nmol/L; P<0.05; Figure 4A). To assess NCX function, we measured the half-time of [Ca2+]i decline during caffeine-induced Ca2+ transients. No change was detectable for NCX function in CaMKIIδC versus LacZ (Figure 4B).
During rest, Ca2+ that leaks from the SR is partly taken up by the SR and partly extruded by NCX. In rabbit cardiac myocytes, Ca2+ is predominantly transported out of the cell via NCX, which leads to a gradual decrease in SR Ca2+ content and Ca2+ release during postrest twitches (rest decay). This differs from rest potentiation that is typical in rat or mouse myocytes.5 Figure 4C shows that rest decay was more pronounced in CaMKIIδC (23±4%; P<0.05) versus LacZ (10±4%). This could reflect enhanced SR Ca2+ leak or NCX function in CaMKIIδC myocytes. Given the unaltered NCX and SERCA function in CaMKIIδC cells, enhanced SR Ca2+ leak seems likely.
To test whether SERCA or NCX expression differs between LacZ and CaMKIIδC myocytes, Western blots were performed at different MOIs of 10 and 100 (normalized to GAPDH). SERCA (n=3) and NCX (n=3) protein expression was unaltered.
SR Ca2+ Leak and Ryanodine Receptor Phosphorylation
To more directly assess SR Ca2+ leak, we measured Ca2+ spark frequency. Because SR Ca2+ content is a major determinant of Ca2+ spark frequency and SR Ca2+ content is very low in CaMKIIδC myocytes, spark frequency was normalized to SR Ca2+ content (measured at the same time). Normalized Ca2+ spark frequency was increased by 88% in CaMKIIδC versus LacZ (1.9±0.1 versus 1.0±0.1; P<0.05; Figure 5A), indicating enhanced SR Ca2+ leak at a given SR Ca2+ content. Similarly, fractional SR Ca2+ release during a normal twitch was significantly increased in CaMKIIδC cells (0.41±0.06 versus 0.26±0.03; P<0.05; Figure 5B). Although part of this 58% increase in fractional release may be attributable to the 23% increase in peak ICa, it may also reflect enhanced RyR sensitivity to Ca2+ (especially because the lower SR Ca2+ content by itself would tend to greatly reduce fractional SR Ca2+ release).19,20
The enhanced SR Ca2+ leak and fractional release were associated with significantly increased RyR phosphorylation at Ser2809 and Ser2815 (71±26 and 171±70% respectively, P<0.05; Figure 5C). To investigate whether FKBP12.6 association with RyR2 was altered, coimmunoprecipitation of equal amounts of myocyte lysates with anti-RyR antibody. At equal amount of RyR2 precipitated in each sample, no decrease in FKBP12.6 was observed in cells overexpressing CaMKIIδC (Figure 5D), whereas RyR2 phosphorylation was increased. Thus, the increased phosphorylation of RyR elicited by CaMKII affects RyR function but does not dissociate FKBP12.6 from RyR2, in agreement with others.9
Reversal of SR Ca2+ Leak and CaMKII-Dependent PLB Phosphorylation
The effects of CaMKIIδC overexpression are MOI dependent. Figure 6A and 6B show a dose-dependent increase in SR Ca2+ spark frequency and amplitude. Figure 6C shows that Ca2+ spark frequency can be significantly decreased by the CaMKIIδC inhibitors AIP or KN-93 or by inhibiting RyR gating by tetracaine. Figure 6D shows an alternative SR Ca2+ leak measurement,13 where abrupt RyR block by tetracaine (in Ca2+-free, Na+-free solution) causes [Ca2+]i to decline and SR Ca2+ content to rise (F/F0 decreased from 1.05±0.08 to 0.75±0.06; P<0.05).
In addition, acute CaMKIIδC overexpression at MOI100 resulted in increased PLB-Thr17 phosphorylation (+125±42%, P<0.05) but a slight decrease in PLB-Ser16 phosphorylation (−44±26%; P=NS; Figure 6E). In contrast, PLB protein expression is unchanged (as SERCA and NCX in Figure 4D).
The present study shows for the first time that acute overexpression of CaMKIIδC (24 hours) results in increased Ca2+ leak from the SR and decreased SR Ca2+ load most likely because of RyR phosphorylation. We conclude that the effects of CaMKIIδC overexpression on RyR function and SR Ca2+ leak as observed in transgenic mice previously are mimicked by adenoviral overexpression in myocytes. Most importantly, however, in contrast to chronic CaMKIIδC overexpression in mice, these acute alterations in ECC together with increased ICa do not lead to decreased twitch contractions or Ca2+ transients. Therefore, CaMKIIδC mediated phosphorylation directly increases diastolic RyR opening and enhances ECC efficacy.
Using adenovirus-mediated gene transfer, we elevated CaMKIIδC expression in rabbit cardiac myocytes, and overexpression was specifically in the cytosolic (versus nuclear) compartment, consistent with CaMKIIδC lacking the 11 amino acid nuclear localization sequence in the δB splice variant.3 Whereas our previous results in transgenic CaMKIIδC mice show clear hypertrophy on the whole heart and myocyte level,7 there was no significant increase in myocyte size after 24 hours of overexpression of CaMKIIδC. In CaMKIIδC transgenic mice, the more prolonged overexpression of CaMKIIδC or its possible multimerization with the nuclear CaMKIIδB in vivo could contribute to the otherwise unknown hypertrophic mechanism. Notably, in both transgenic CaMKIIδC mice and the present acute CaMKIIδC overexpressing rabbit myocytes, a major functional finding was increased SR Ca2+ leak associated with enhanced RyR phosphorylation and reduced SR Ca2+ content. It is possible that the increased diastolic Ca2+ leak from the SR may activate Ca2+-dependent hypertrophic signaling pathways.3,21,22
L-Type Ca2+ Current
In the present study, CaMKIIδC overexpression resulted in significantly enhanced peak ICa and also prolonged ICa inactivation parameters τ1 and τ2. Because CaMKII can activate ICa23–25 and we see enhanced ICa facilitation, the 31% increase in peak ICa may reflect a relatively direct CaMKII-dependent regulatory effect on ICa. This interpretation is supported by the observation that acute CaMKII inhibition by KN-93 or AIP blocks both amplitude and inactivation effects. Indeed, both higher ICa amplitude and slowed inactivation are hallmarks of CaMKII-dependent ICa facilitation,5,23 consistent with a common fundamental mechanism. The increased ICa, together with the increases in fractional Ca2+ release from the SR, results in unchanged twitch contraction (at least at low stimulation rates), even in the face of decreased SR Ca2+ content.
SR Ca2+ Content and Contractions
The SR is central in cardiac ECC and CaMKII can accelerate SERCA function via PLB phosphorylation.2–5 Surprisingly, we did not detect altered SERCA function at baseline contraction frequency in CaMKIIδC versus LacZ myocytes. However, we do see modestly enhanced FDAR in the CaMKIIδC versus LacZ myocytes (as in transgenic CaMKIIδC mice).7 FDAR is thought to reflect CaMKII-dependent enhancement of SR Ca2+ uptake (even though it does not require PLB).26 Thus, acute CaMKII appears to have only modest effects on SERCA function here. These modest effects on the rate of [Ca2+]i decline here and in our previous study7 may be because the absolute extent of CaMKII-dependent PLB phosphorylation may be small27 and the increased phosphorylation at PLB-Thr17 may be counterbalanced by less at Ser16, as seen in the present study.
In transgenic CaMKIIδC-overexpressing mouse hearts,7 SR Ca2+ content was also reduced, but that could have been attributable to the enhanced SR Ca2+ leak (and RyR phosphorylation), increased NCX function, or reduced SERCA function that are associated with the heart failure phenotype, as in other heart failure models.28–31 With acute CaMKIIδC expression here, a more modest reduction in myocyte SR Ca2+ content occurs with enhanced SR Ca2+ leak but unaltered NCX and SERCA function and protein expression. This argues strongly in favor of enhanced SR Ca2+ leak causing reduced SR Ca2+ content here. However, the more severe reduction in SR Ca2+ content in heart failure (whether induced by transgenic CaMKII overexpression or otherwise) is attributable not only to enhanced SR Ca2+ leak but also to enhanced NCX function and reduced SERCA function.28–31
In isolated single-channel RyR recordings, CaMKII has been shown to increase cardiac RyR open probability.8,32 At an intermediate level of isolation, CaMKII greatly enhanced Ca2+ spark frequency in permeabilized PLB-KO mouse myocytes (without enhanced SR Ca2+ content).33 In addition, CaMKII is associated with the RyR in the cell,7–9 and can phosphorylate the RyR.7–9,32,34 The cardiac RyR has been reported to be phosphorylated by CaMKII at both Ser2809 and Ser2815 sites.9,32,34 Mark and colleagues have reported that these sites are segregated (CaMKII only at 2815 and PKA only at 2809) and that Ser2809 phosphorylation causes dissociation of FKBP12.6 from RyR and consequent RyR opening, whereas CaMKII-dependent phosphorylation activates RyR without causing FKBP12.6 dissociation.9,18 Likewise, here we do not see FKBP12.6 dissociation from the RyR in myocytes overexpressing CaMKIIδC, despite some increase in RyR phosphorylation at Ser2809. Thus RyR phosphorylation appears to cause enhanced SR Ca2+ leak and reduced SR Ca2+ content in CaMKIIδC versus LacZ myocytes.
A remarkable finding here is that twitch contractions and Ca2+ transients are almost unaffected by the dramatically reduced SR Ca2+ content. This may be attributable in part to the enhanced ICa (as above), but a major factor is probably the sort of autoregulation described previously by Trafford et al35 in the presence of low caffeine concentration (which causes diastolic SR Ca2+ leak and sensitizes the RyR to Ca2+). They showed that altered RyR gating only produces transient changes in Ca2+ transients (but sustained changes in SR Ca2+ content and fractional release). That is, abruptly CaMKII may enhance SR Ca2+ release, but this causes more Ca2+ extrusion (via NCX) and reduces SR Ca2+ content. With the lower SR Ca2+ content the enhanced fractional release only results (in the steady state) in the same amount of SR Ca2+ release. This may be what is happening here with acute CaMKIIδC overexpression. Indeed, Shannon et al36 recently simulated Ca2+ homeostasis mathematically showing that enhanced RyR Ca2+ sensitivity (as by caffeine or phosphorylation) increased SR Ca2+ leak and reduced SR Ca2+ content but enhanced SR fractional release without decreasing the size of the steady-state [Ca2+]i transient (as seen here).
Thus, enhanced RyR Ca2+ sensitivity by itself may contribute substantially to SR Ca2+ unloading on CaMKII overexpression (or in heart failure), without itself being appreciably negatively inotropic. Other factors must be largely responsible for the systolic dysfunction seen in CaMKIIδC transgenic mice or other heart failure models (eg, reduced SERCA function and enhanced NCX function).7,29,31 Indeed, in heart failure, CaMKIIδ is overexpressed,11,37 SR Ca2+ leak is enhanced,31,37 and block of CaMKII in heart failure can greatly enhance SR Ca2+ content without improving systolic function.37 We conclude that CaMKII-dependent enhancement of RyR Ca2+ sensitivity (and thus leak) does not contribute appreciably to systolic dysfunction, but the enhanced diastolic SR Ca2+ leak could possibly increase the propensity for triggered arrhythmias.38
D.M.B. is funded by NIH grants HL-30077 and HL-64724 and J.H.B. by HL-46345 and HL-28143. T.Z. is the recipient of a Scientist Development Grant from the American Heart Association. L.S.M. is funded by the Deutsche Forschungsgemeinschaft (DFG) through an Emmy Noether grant (MA 1982/4−1), by a Young Investigator Award of the GlaxoSmithKline Research Foundation, and by a grant from the Medical Faculty of the University of Göttingen (Anschubfinanzierung).
Original received July 18, 2005; resubmission received October 31, 2005; accepted December 8, 2005.
Zhang T, Brown JH. Role of Ca2+/calmodulin-dependent protein kinase II in cardiac hypertrophy and heart failure. Cardiovasc Res. 2004; 63: 476–486.
Bers DM. Excitation-Contraction Coupling and Cardiac Contractile Force. 2nd ed. Dordrecht, The Netherlands: Kluwer Academic Publishers; 2001.
Zhang T, Maier LS, Dalton ND, Miyamoto S, Ross J Jr, Bers DM, Brown JH. The δC isoform of CaMKII is activated in cardiac hypertrophy and induces dilated cardiomyopathy and heart failure. Circ Res. 2003; 92: 912–919.
Maier LS, Zhang T, Chen L, DeSantiago J, Brown JH, Bers DM. Transgenic CaMKIIδC overexpression uniquely alters cardiac myocyte Ca2+ handling: reduced SR Ca2+ load and activated SR Ca2+ release. Circ Res. 2003; 92: 904–911.
Wehrens XH, Lehnart SE, Reiken SR, Marks AR. Ca2+/calmodulin-dependent protein kinase II phosphorylation regulates the cardiac ryanodine receptor. Circ Res. 2004; 94: e61–e70.
Kirchhefer U, Schmitz W, Scholz H, Neumann J. Activity of cAMP-dependent protein kinase and Ca2+/calmodulin-dependent protein kinase in failing and nonfailing human hearts. Cardiovasc Res. 1999; 42: 254–261.
Hoch B, Meyer R, Hetzer R, Krause EG, Karczewski P. Identification and expression of delta-isoforms of the multifunctional Ca2+/calmodulin-dependent protein kinase in failing and nonfailing human myocardium. Circ Res. 1999; 84: 713–721.
Ramirez MT, Zhao XL, Schulman H, Brown JH. The nuclear δB isoform of Ca/calmodulin-dependent protein kinase II regulates atrial natriuretic factor gene expression in ventricular myocytes. J Biol Chem. 1997; 272: 31203–31208.
Shannon TR, Ginsburg KS, Bers DM. Quantitative assessment of the SR Ca2+ leak-load relationship. Circ Res. 2002; 91: 594–600.
Lehnart SE, Janssen PM, Franz WM, Donahue JK, Lawrence JH, Marban E, Prestle J, Hasenfuss G. Preservation of myocardial function after adenoviral gene transfer in isolated myocardium. Am J Physiol Heart Circ Physiol. 2000; 279: H986–H991.
Schillinger W, Janssen PM, Emami S, Henderson SA, Ross RS, Teucher N, Zeitz O, Philipson KD, Prestle J, Hasenfuss G. Impaired contractile performance of cultured rabbit ventricular myocytes after adenoviral gene transfer of Na+-Ca2+ exchanger. Circ Res. 2000; 87: 581–587.
Wagner S, Seidler T, Picht E, Maier LS, Kazanski V, Teucher N, Schillinger W, Pieske B, Isenberg G, Hasenfuss G, Kögler H. Na+/Ca2+ exchanger overexpression predisposes to reactive oxygen species-induced injury. Cardiovasc Res. 2003; 60: 404–412.
Xiao RP, Cheng H, Lederer WJ, Suzuki T, Lakatta EG. Dual regulation of Ca/calmodulin kinase II activity by membrane voltage and by calcium influx. Proc Natl Acad Sci U S A. 1994; 91: 9659–9663.
Huke S, Bers DM. Frequency-dependent acceleration of relaxation (FDAR) does not require phospholamban phosphorylation by CaM-kinase II in rat cardiomyocytes. Biophys J. 2005; 88: 311a.
Hasenfuss G. Alterations of calcium-regulatory proteins in heart failure. Cardiovasc Res. 1998; 37: 279–289.
Pogwizd SM, Schlotthauer K, Li L, Yuan W, Bers DM. Arrhythmogenesis and contractile dysfunction in heart failure: roles of sodium-calcium exchange, inward rectifier potassium current, and residual beta-adrenergic responsiveness. Circ Res. 2001; 88: 1159–1167.
Hobai IA, O’Rourke B. Enhanced Ca2+-activated Na+-Ca2+ exchange activity in canine pacing-induced heart failure. Circ Res. 2000; 87: 690–698.
Shannon TR, Pogwizd SM, Bers DM. Elevated sarcoplasmic reticulum Ca2+ leak in intact ventricular myocytes from rabbits in heart failure. Circ Res. 2003; 93: 592–594.
Witcher DR, Kovacs RJ, Schulman H, Cefali DC, Jones LR. Unique phosphorylation site on the cardiac ryanodine receptor regulates calcium channel activity. J Biol Chem. 1991; 266: 11144–11152.
Guo T, Zhang T, Brown JH, Bers DM. Effects of CaMKII on cardiac Ca release channels in myocytes. Biophys J. 2004; 86: 241a.
Rodriguez P, Bhogal MS, Colyer J. Stoichiometric phosphorylation of cardiac ryanodine receptor on serine 2809 by calmodulin-dependent kinase II and protein kinase A. J Biol Chem. 2003; 278: 38593–38600.
Ai X, Curran JW, Shannon TR, Bers DM, Pogwizd SM. Ca2+-calmodulin-dependent protein kinase modulates RyR2 phosphorylation and SR Ca2+ leak in a rabbit heart failure. Circ Res. 2005; 97: 1314–1322.
Wu Y, Temple J, Zhang R, Dzhura I, Zhang W, Trimble R, Roden DM, Passier R, Olson EN, Colbran RJ, Anderson ME. Calmodulin kinase II and arrhythmias in a mouse model of cardiac hypertrophy. Circulation. 2002; 106: 1288–1293.