Effects of increased systolic Ca²⁺ and phospholamban phosphorylation during β-adrenergic stimulation on Ca²⁺ transient kinetics in cardiac myocytes

Abstract

Previous studies demonstrated higher systolic intracellular Ca²⁺ concentration ([Ca²⁺]_i) amplitudes result in faster [Ca²⁺]_i decline rates, as does β-adrenergic (β-AR) stimulation. The purpose of this study is to determine the major factor responsible for the faster [Ca²⁺]_i decline rate with β-AR stimulation, the increased systolic Ca²⁺ concentration levels, or phosphorylation of phospholamban. Mouse myocytes were perfused under basal conditions [1 mM extracellular Ca²⁺ concentration ([Ca²⁺]_o)], followed by high extracellular Ca²⁺ (3 mM [Ca²⁺]_o), washout with 1 mM [Ca²⁺]_o, followed by 1 μM isoproterenol (ISO) with 1 mM [Ca²⁺]_o. ISO increased Ser¹⁶ phosphorylation compared with 3 mM [Ca²⁺]_o, whereas Thr¹⁷ phosphorylation was similar. Ca²⁺ transient (CaT) (fluo 4) data were obtained from matched CaT amplitudes with 3 mM [Ca²⁺]_o and ISO. [Ca²⁺]_i decline was significantly faster with ISO compared with 3 mM [Ca²⁺]_o. Interestingly, the faster decline with ISO was only seen during the first 50% of the decline. CaT time to peak was significantly faster with ISO compared with 3 mM [Ca²⁺]_o. A Ca²⁺/calmodulin-dependent protein kinase (CAMKII) inhibitor (KN-93) did not affect the CaT decline rates with 3 mM [Ca²⁺]_o or ISO but normalized ISO's time to peak with 3 mM [Ca²⁺]_o. Thus, during β-AR stimulation, the major factor for the faster CaT decline is due to Ser¹⁶ phosphorylation, and faster time to peak is due to CAMKII activation.

primary regulation of intracellular Ca²⁺ concentration ([Ca²⁺]_i) uptake in the sarcoplasmic reticulum (SR) is performed by the SR Ca²⁺-ATPase (SERCA) (2). SERCA's function is modulated by the regulatory protein, phospholamban (PLB), which inhibits SERCA function (26). For example, transgenic mice with overexpression of PLB result in decreased SERCA activity and SR Ca²⁺ uptake, which results in a slower Ca²⁺ decline rate (19). The PLB-mediated inhibition of SERCA can be relieved by its dissociation. An increase in Ca²⁺ levels will cause this dissociation (20). Thus, increasing the peak amplitude of the [Ca²⁺]_i transient will increase SERCA activity, which results in faster [Ca²⁺]_i decline rates (3). PLB is also a major phosphoprotein in the myocyte. PLB can be phosphorylated on Ser¹⁶ through protein kinase A (PKA) or on Thr¹⁷ through Ca²⁺/calmodulin-dependent protein kinase (CAMKII). Phosphorylation at either site also results in the dissociation of PLB from SERCA. In fact, phosphorylation at these sites with β-adrenergic (β-AR) stimulation in PLB overexpression mice normalizes the Ca²⁺ decline compared with the wild type (WT) (19). Furthermore, PLB knockout mice result in increased SERCA activity and SR Ca²⁺ uptake, which results in faster Ca²⁺ decline rates (25). Interestingly, there was no further acceleration of the Ca²⁺ decline due to β-AR stimulation in these myocytes (24). In addition, a previous study in mice that had mutated PLB, which could not be phosphorylated (both Ser¹⁶ and Thr¹⁷ mutated to alanine), had no lusitropic effect during β-AR stimulation (5). Thus, PLB is an important mediator of Ca²⁺ decline during β-AR stimulation. However, the faster [Ca²⁺]_i decline with β-AR stimulation may not only be due to PLB phosphorylation but the higher systolic Ca²⁺ levels as well. Unlike these mentioned studies, we matched Ca²⁺ transient peaks to best determine if the increase in systolic Ca²⁺ and/or PLB phosphorylation contributed to the lusitropic effect during β-AR stimulation.

Because lowering the [Ca²⁺]_i is the initiating event that permits relaxation, clearly defining the processes contributing to Ca²⁺ resequestation in the SR is important. Relaxation of the cardiac muscle is governed by processes at the myofilament level (17, 18), and duration of activation is critically impacted by Ca²⁺ transient kinetics. Thus, governing of Ca²⁺ transient decline by the SR may have important clinical ramifications since relaxation is altered in many cardiomyopathies such as heart failure (31, 39).

Therefore, our purpose is to determine the major factor in the faster [Ca²⁺]_i decline during β-AR stimulation (i.e., the high systolic levels of [Ca²⁺]_i or PLB phosphorylation). We hypothesize that the high systolic [Ca²⁺]_i levels seen during β-AR stimulation or high extracellular Ca²⁺ ([Ca²⁺]_o) will be near the maximal velocity (V_max) of the SERCA pump. Thus, matched systolic [Ca²⁺]_i levels with high [Ca²⁺]_o and β-AR stimulation will exhibit similar Ca²⁺ decline rates in the initial decline phase. In addition, PLB phosphorylation (i.e., β-AR stimulation) would then result in a faster Ca²⁺ transient decline during the final phase because of the leftward shift in the Michaelis constant at low [Ca²⁺]_i levels. To test this hypothesis, each myocyte will be superfused with high [Ca²⁺]_o (3 mM) and with the β-AR agonist isoproterenol (ISO) to obtain data under these conditions with matched levels of Ca²⁺ transient amplitude.

β-AR stimulation also results in positive chronotropy. It is known that increasing frequency of stimulation also results in faster Ca²⁺ decline rates. However, this is independent of PLB phosphorylation and occurs via a separate mechanism (34). Hence the chronotropic effect (i.e., increasing pacing frequency) was not investigated. To our knowledge, no one has measured [Ca²⁺]_i decline rates comparing high [Ca²⁺]_o and ISO in the same myocyte using matched Ca²⁺ transient amplitudes. A previous study examined the effects of ISO and high [Ca²⁺]_o in the same trabeculas (7). However, this study was specifically focused on these interventions on developed pressure. There was some examination of Ca²⁺ decline, in which they observed a faster Ca²⁺ decline with ISO vs. 20 mM [Ca²⁺]_o, but the amplitudes were not matched.

Our data suggest that, during β-AR stimulation, Ser¹⁶ phosphorylation is the major factor that drives the faster acceleration of the Ca²⁺ transient decline. Also, the faster Ca²⁺ transient time to peak is the result of CaMKII activation.

MATERIALS AND METHODS

Cardiomyocyte Isolation

Ventricular myocytes were isolated from mice (C57BL/6) (Jackson Laboratories, Bar Harbor, ME) as previously described (22). Briefly, the heart was cannulated and hung on a Langendorff apparatus. It was then perfused with Ca²⁺-free Tyrode solution for 4 min. The solution was then switched to Tyrode solution containing Liberase Blendzyme II (0.077 mg/ml) (Roche Applied Science, Indianapolis, IN). After 3–5 min, the heart was taken down, the ventricles were minced, and myocytes were dissociated by trituration. Subsequently, the myocytes were filtered, centrifuged, and resuspended in Tyrode solution containing 200 μmol/l Ca²⁺. Myocytes were used within 6 h of isolation. All of the animal protocols and procedures were performed in accordance with National Institutes of Health guidelines and approved by the Institutional Laboratory Animal Care and Use Committee at The Ohio State University.

Measurement of myocyte Ca²⁺ transients.

Ca²⁺ transient measurements were performed as previously described (22). Briefly, myocytes were loaded at room temperature with fluo 4-AM (10 μmol/l; Molecular Probes, Eugene, OR) for 30 min, and then another 30 min were allowed for intracellular deesterification. The solution for deesterification was Tyrode solution containing 200 μmol/l Ca²⁺. The instrumentation used for cell fluorescence measurements was a Cairn Research Limited (Faversham, UK) epifluorescence system. [Ca²⁺]_i was measured by fluo 4 epifluorescence with excitation at 480 ± 20 nm and emission at 535 ± 25 nm. The illumination field was restricted to collect the emission of a single cell. Data were expressed as ΔF/F₀, where F is the fluorescence intensity, and F₀ is the intensity at rest. Myocytes were stimulated at 1 Hz via platinum electrodes connected to a Grass Telefactor S48 stimulator (West Warwick, RI).

Western blot analysis.

Homogenized myocytes were used to measure specific phosphorylation at Ser¹⁶ and Thr¹⁷ (Badrilla, Leeds, UK) with phosphospecific antibodies and normalized to calsequestrin (ABR, Golden, CO) via Western blot analysis, as previously described (37).

Ca²⁺ uptake rate.

SR Ca²⁺ uptake rate was determined as described previously (32) by converting the RT₅₀ (time to 50% of the Ca²⁺ transient decline) into a tau. This value was then used to fit a declining single exponential curve, which was then converted to a total cytosolic Ca²⁺ ([Ca]_T) using well-characterized Ca²⁺-buffering parameters. Next, we took the derivative over time (d[Ca]_T/dt). Finally, this derivative was fit into the Hill equation, and the uptake rate was calculated at 300 nm (roughly the RT₅₀ [Ca²⁺]_i value).

Solutions and drugs.

Normal Tyrode (NT) solution consisted of (in mmol/l): 140 NaCl, 4 KCl, 1 MgCl₂, 1 CaCl₂, 10 glucose, and 5 HEPES, pH 7.4 adjusted with NaOH or HCl. ISO (1 μmol/l, a nonselective β-AR agonist) was prepared fresh each day. KN-93 (1 μmol/l, CAMKII inhibitor; EMD Chemicals, Gibbstown, NJ) was prepared each day from frozen aliquots. Zinterol (Zint, 100 nmol/l, β₂-AR agonist; Tocris Biosciences, Ellisville, MO) All chemicals were from Sigma (St. Louis, MO) except where indicated.

Experimental protocol.

Our basic experimental protocol consisted of the myocyte first being perfused with control solution [NT with 1 mM Ca²⁺ concentration ([Ca²⁺])] until steady state was reached, and the myocyte was then superfused with the 3 mM [Ca²⁺] NT solution, which resulted in increased systolic Ca²⁺ levels. After steady state was reached, the solution was then switched back to control solution (1 mM [Ca²⁺] NT), resulting in the washout of 3 mM [Ca²⁺]_o and the Ca²⁺ transient amplitude returning back to basal levels (∼3 min). The myocyte was then perfused with 1 μM ISO (with 1 mM [Ca²⁺]_o), which increased myocyte systolic Ca²⁺ levels.

Statistics.

Data were presented as means ± SE. Differences between groups were evaluated for statistical significance (P < 0.05) by ANOVA for multiple groups or paired Student's t-tests for two groups.

RESULTS

To compare [Ca²⁺]_i decline rates of comparable [Ca²⁺]_i with low and high PLB phosphorylation, a NT solution made with 3 mM [Ca²⁺]_o was used to achieve similar [Ca²⁺]_i peaks as during β-AR stimulation with ISO. Variability of Ca²⁺ handling between cells being exposed to 3 mM [Ca²⁺]_o and ISO was minimized by superfusing each myocyte with both solutions.

Ca²⁺ transient kinetics experimental protocol.

Shown in Fig. 1A is a representative experiment showing Ca²⁺ transients over time with the various solutions as explained in materials and methods.

Fig. 1.

Experimental protocol. A: representative time plot of the experimental protocol. B: pooled data (means ± SE) of maximum Ca²⁺ transient amplitude with 1 mM extracellular Ca²⁺ concentration ([Ca²⁺]_o, open bar), 3 mM [Ca²⁺]_o (black bar), or isoproterenol (ISO, gray bar). C: pooled data (means ± SE) of Ca²⁺ transient decline measured as the time to 50% relaxation (RT₅₀). P < 0.05 vs. 1 mM [Ca²⁺]_o (*) and vs. 1 and 3 mM [Ca²⁺]_o (**) (n = 21 cells/10 hearts).

Shown in Fig. 1B, perfusing the myocytes with 3 mM [Ca²⁺]_o and ISO resulted in a significantly increased maximal peak systolic [Ca²⁺] (1 mM [Ca²⁺]_o: 1.6 ± 0.1; 3 mM [Ca²⁺]_o: 3.6 ± 0.2; ISO: 3.9 ± 0.2 ΔF/F_o; n = 21 cells/10 hearts). The majority of myocytes reached a higher maximal peak systolic [Ca²⁺] with ISO compared with 3 mM [Ca²⁺]_o (although not significant). Shown in Fig. 1C, perfusion with 3 mM [Ca²⁺]_o and ISO also resulted in a significantly faster Ca²⁺ transient decline rate measured as RT₅₀ (1 mM [Ca²⁺]_o: 225 ± 7; 3 mM [Ca²⁺]_o: 154 ± 5; ISO: 127 ± 3 ms). Interestingly, the ISO Ca²⁺ transient decay rate was significantly faster than the 3 mM [Ca²⁺]_o.

Ca²⁺ transient kinetics with matched amplitudes.

Although the maximum systolic Ca²⁺ levels were not statistically different between 3 mM [Ca²⁺]_o and ISO, the slightly higher peak with ISO may have resulted in the faster Ca²⁺ transient decline. Thus, for better comparisons of Ca²⁺ transient decline rates, we used matched Ca²⁺ transient amplitude levels. Representative matched individual [Ca²⁺]_i transient traces are shown in Fig. 2A. Shown in Fig. 2B are the matched peak values of 3 mM [Ca²⁺]_o (3.5 ± 0.2 ΔF/F_o) and ISO (3.5 ± 0.2 ΔF/F_o).

Fig. 2.

Matched Ca²⁺ transient peaks and phospholamban (PLB) phosphorylation with 3 mM [Ca²⁺]_o and ISO. A: representative trace of matched intracellular Ca²⁺ concentration ([Ca²⁺]_i) amplitudes with 3 mM [Ca²⁺]_o (black) or 1 μM ISO (gray). B: pooled data (means ± SE) of matched Ca²⁺ transient amplitude with 3 mM [Ca²⁺]_o (black) or ISO (gray) (n = 21 cells/10 hearts). CaT, Ca²⁺ transient. C: PLB Ser¹⁶ phosphorylation with 3 mM [Ca²⁺]_o or ISO (means ± SE) (n = 4). CSQ, calsequestrin; AU, arbitrary units. D: PLB Thr¹⁷ phosphorylation with 3 mM [Ca²⁺]_o or ISO (means ± SE). *P < 0.05 vs. 3 mM [Ca²⁺]_o (n = 4 myocyte homogenates).

Similar results were observed when we switched the order of solutions. That is, after initial equilibration at the basal condition of 1 mM [Ca²⁺]_o, the myocytes were first superfused with ISO, washed out (∼15 min), and then perfused with 3 mM [Ca²⁺]_o (data not shown). Thus, the effects of these interventions on Ca²⁺ transient kinetics are independent of the order of the solutions.

We also measured the Ser¹⁶ and Thr¹⁷ phosphorylation under similar experimental conditions (i.e., myocyte perfusion with 3 mM [Ca²⁺]_o or ISO for 3 min, which is a similar time point in which we matched the Ca²⁺ transient amplitudes). Shown in Fig. 2C, myocytes perfused with ISO had significantly increased Ser¹⁶ phosphorylation compared with myocytes perfused with 3 mM [Ca²⁺]_o [4.4 ± 0.5 vs. 0.9 ± 0.3 arbitrary units (AU) (n = 4)]. However, there was no difference in Thr¹⁷ phosphorylation between ISO and 3 mM [Ca²⁺]_o [1.5 ± 0.6 vs. 1.7 ± 0.6 AU (Fig. 2D)]. Thus, with our experimental protocol and matching Ca²⁺ transient amplitudes, the only difference between ISO and 3 mM [Ca²⁺]_o is the Ser¹⁶ phosphorylation levels.

Using matched-amplitudes Ca²⁺ transient data, we explored the effects of 3 mM [Ca²⁺]_o and ISO on Ca²⁺ decline by analyzing the time it takes the Ca²⁺ transient to decline by 25% (RT₂₅), 50% (RT₅₀), 75% (RT₇₅), and 90% (RT₉₀) from its peak amplitude. As shown in Fig. 3A, at each time point, the Ca²⁺ decline with ISO was significantly faster compared with 3 mM [Ca²⁺]_o. We further analyzed the effects of ISO and 3 mM [Ca²⁺]_o on the Ca²⁺ transient decline by examining various time intervals. By breaking the decline into intervals, we can better determine at which point in the [Ca²⁺]_i transient PLB phosphorylation will result in different decline rates compared with matched systolic [Ca²⁺]_i levels with 3 mM [Ca²⁺]_o. Thus, we divided the declining Ca²⁺ transient into intervals: RT_50–25, RT_75–50, and RT_90–75. For example, we subtracted the RT₂₅ from the RT₅₀ to get the RT_50–25 interval. This is the time it took for Ca²⁺ to decline 25% from the peak amplitude to the 50% point. Data are shown in Fig. 3B. The RT_50–25 interval was the only interval significantly different between 3 mM [Ca²⁺]_o and ISO. Thus, the faster Ca²⁺ transient decline with ISO compared with 3 mM [Ca²⁺]_o occurs only in the first 50%.

Fig. 3.

Effects of 3 mM [Ca²⁺]_o and ISO on Ca²⁺ transient decline with matched peaks. A: pooled data (means ± SE) of Ca²⁺ decline with 3 mM [Ca²⁺]_o (black) or ISO (gray). RT₂₅, RT₇₅, and RT₉₀, the time it takes the Ca²⁺ transient to decline by 25, 75, and 90%, respectively, from its peak amplitude. B: pooled data (means ± SE) of Ca²⁺ transient decline time intervals with 3 mM [Ca²⁺]_o or ISO. RT_50-25, the time for Ca²⁺ to decline 25% from the peak amplitude to the 50% point; RT_75-50, the time for Ca²⁺ to decline 50% from the peak amplitude to the 75% point; RT_90-75, the time for Ca²⁺ to decline 75% from the peak amplitude to the 90% point. *P < 0.05 vs. corresponding 3 mM [Ca²⁺]_o (n = 21 cells/10 hearts).

We also investigated the effects of 3 mM [Ca²⁺]_o and ISO on diastolic Ca²⁺ levels. We observed no difference in diastolic Ca²⁺ values between 1 mM [Ca²⁺]_o and 3 mM [Ca²⁺]_o (103% of 1 mM [Ca²⁺]_o). There was a slight but significant decrease in diastolic Ca²⁺ values with ISO compared with 1 mm [Ca²⁺]_o (91% of 1 mM [Ca²⁺]_o). We believe these results further strengthen our argument that Ser¹⁶ phosphorylation results in a greater Ca²⁺ decline, which results in a decreased diastolic [Ca²⁺]_i. We further analyzed the relationship between the systolic Ca²⁺ levels and RT₅₀. Shown in Fig. 4 are the maximal steady-state Ca²⁺ transient amplitudes plotted against their respective RT₅₀. Myocytes with low PLB phosphorylation are shown in black (NT with 1 and 3 mM [Ca²⁺]_o). Consistent with previous studies (3), the higher the peak [Ca²⁺]_i, the faster the rate of decline. Myocytes with ISO (i.e., high Ser¹⁶ phosphorylation) are shown in gray. In addition to the 1 μM ISO (1 mM [Ca²⁺]_o) group, we were also able to obtain smaller peak Ca²⁺ transient amplitudes by perfusing myocytes with 1 μM ISO and 0.25 mM [Ca²⁺]_o. We then analyzed the slope of these lines. Our data show that the phosphorylation of Ser¹⁶ resulted in a weaker correlation (slope of −14.3 ± 3.4) compared with myocytes with low phosphorylated PLB (slope of −38.3 ± 2.8). Thus, our data suggest that the PLB phosphorylation is more effective at increasing the rate of [Ca²⁺]_i decline and is much less dependent on peak Ca²⁺ transient amplitudes. Thus, with β-AR stimulation (and PLB phosphorylation), the cardiomyocyte is less dependent on the peak [Ca²⁺]_i levels for the rate of Ca²⁺ decline.

Fig. 4.

Effects of PLB Ser¹⁶ phosphorylation on the relationship between maximal steady-state Ca²⁺ transient amplitude and decline. Individual values of peak [Ca²⁺]_i vs. RT₅₀ with low (black, r² = 0.77) and high (gray, r² = 0.37) Ser¹⁶ phosphorylation.

We further investigated this point by grouping the myocytes that had a higher maximum response to ISO together (n = 14) and the myocytes that had a higher response to 3 mM [Ca²⁺]_o together (n = 7). Shown in Fig. 5A, when grouping the data, the myocytes that had a higher maximum peak systolic Ca²⁺ with ISO (4.0 ± 0.2 vs. 3.5 ± 0.2 ΔF/F_o, P < 0.05) also had a faster Ca²⁺ transient decline rate (126 ± 3 vs. 161 ± 7 ms, P < 0.05). It is unknown if the major factor in the faster Ca²⁺ decline is the higher systolic Ca²⁺ levels or PLB Ser¹⁶ phosphorylation. However, the myocytes that a higher maximum peak systolic Ca²⁺ with 3 mM [Ca²⁺]_o (4.1 ± 0.1 vs. 3.7 ± 0.1 ΔF/F_o, P < 0.05) had a slower Ca²⁺ transient decline rate (141 ± 3 vs. 130 ± 5 ms, P < 0.05) (Fig. 5B). We also calculated Ca²⁺ uptake rates with 3 mM [Ca²⁺]_i and ISO. Shown in Fig. 5C, ISO had a greater Ca²⁺ uptake rate at 300 nm [Ca²⁺]_i (roughly the [Ca²⁺]_i at the RT₅₀) vs. 3 mM [Ca²⁺]_i (540 ± 16 vs. 468 ± 28 μM/s, P < 0.05). Thus these data suggest that PLB Ser¹⁶ phosphorylation is the major factor responsible for the faster Ca²⁺ transient decline rate during β-AR stimulation.

Fig. 5.

Grouped Ca²⁺ transient amplitudes and declines. A: pooled data (means ± SE) of Ca²⁺ transient amplitude (left) and decline (right) in myocytes that had a higher response to ISO (n = 14). B: pooled data (means ± SE) of Ca²⁺ transient amplitude (left) and decline (right) in myocytes that had a higher response to 3 mM [Ca²⁺]_o (n = 7). C: pooled data (means ± SE) of calculated Ca²⁺ uptake rate at 300 nM [Ca²⁺]_i (*P < 0.05) vs. corresponding 3 mM [Ca²⁺]_o (n = 21 cells/10 hearts).

Effects of β₂-AR stimulation on Ca²⁺ transient decline rates with 3 mm [Ca²⁺]_o and Zint.

The functional response to β-AR stimulation is primarily regulated through β₁-AR and β₂-AR receptors. β₁-AR stimulation leads to high PLB Ser¹⁶ phosphorylation by PKA. Interestingly, β₂-AR stimulation does not increase Ser¹⁶ PLB phosphorylation (23). To further test the contribution of Ser¹⁶ phosphorylation on Ca²⁺ decline, we performed our experimental protocol but substituted Zint, a β₂-AR, agonist, for ISO. Matched Ca²⁺ transient amplitudes for 3 mM [Ca²⁺]_o (2.2 ± 0.2 ΔF/F_o) and Zint (2.2 ± 0.2 ΔF/F_o) (n = 5 cells/3 hearts) were analyzed. Figure 6A displays representative matched Ca²⁺ transient traces of 3 mM [Ca²⁺]_o and Zint. Shown in Fig. 6B are the Ca²⁺ transient decline rates (RT₂₅, RT₅₀, RT₇₅, and RT₉₀). We also analyzed the time intervals between 25–50%, 50–75%, and 75–90% (Fig. 6C), as described above. The data in Fig. 6, B and C, show that the Ca²⁺ transient decline is not significantly different at any time points or between intervals. Thus, with matched [Ca²⁺]_i, β₂-AR stimulation does not result in faster Ca²⁺ decline.

Fig. 6.

Effect of 3 mM [Ca²⁺]_o and β-adrenergic type 2 receptor (β₂-AR) agonist zinterol (Zint) on Ca²⁺ transient decline. A: representative traces of matched [Ca²⁺]_i amplitudes with 3 mM [Ca²⁺]_o (white) and Zint (gray). B: pooled data (means ± SE) of Ca²⁺ decline with 3 mM [Ca²⁺]_o (white checkered) or Zint (gray checkered). C: pooled data (means ± SE) of Ca²⁺ transient decline time intervals with 3 mM [Ca²⁺]_o or Zint. *P < 0.05 vs. corresponding 3 mM [Ca²⁺]_o (n = 3 hearts/5 cells).

Effects of CAMKII inhibition on Ca²⁺ transient decline rates with 3 mm [Ca²⁺]_o and ISO.

β-AR stimulation and high [Ca²⁺]_o can lead to CAMKII activation and PLB Thr¹⁷ phosphorylation. Although we observed no difference in the Thr¹⁷ phosphorylation status with 3 mM [Ca²⁺]_o and ISO, we wanted to determine if Thr¹⁷ phosphorylation played a role in the Ca²⁺ transient decline. Hence, we repeated our experiments in the presence of the CAMKII inhibitor KN-93 (1 μm/l).

KN-93 resulted in no significant changes in maximum steady-state Ca²⁺ transient amplitudes (Fig. 7A) or RT₅₀ (Fig. 7C) during superfusion with 3 mM [Ca²⁺]_o.

Fig. 7.

Effects of the Ca²⁺/calmodulin-dependent protein kinase inhibitor KN-93. Pooled data (means ± SE) of 3 mM [Ca²⁺]_o ± KN-93 on maximal steady-state Ca²⁺ transient amplitude (A) and corresponding RT₅₀ (C). Pooled data (means ± SE) of ISO ± KN-93 on maximal steady-state Ca²⁺ transient amplitude (B) and corresponding RT₅₀ (D) (−KN-93: n = 21 cells/10 hearts, +KN-93: n = 15 cells/4 hearts).

We also investigated the effect of ISO with KN-93 on maximum steady-state amplitudes (Fig. 7B) and RT₅₀ (Fig. 7D). KN-93 did not have a significant effect on the response to ISO.

Figure 8A displays representative matched Ca²⁺ transient traces during superfusion with 3 mM [Ca²⁺]_o and ISO in the presence of KN-93. Shown in Fig. 8B are the Ca²⁺ transient decline rates (RT₂₅, RT₅₀, RT₇₅, and RT₉₀) in the presence of KN-93. Results with and without KN-93 were similar, that is, the Ca²⁺ transient decline at each time point was faster with ISO superfusion compared with 3 mM [Ca²⁺]_o.

Fig. 8.

Effects of 3 mM [Ca²⁺]_o and ISO on Ca²⁺ transient decline in the presence of KN-93. A: representative traces of matched [Ca²⁺]_i amplitudes with 3 mM [Ca²⁺]_o + KN-93 (black) and ISO + KN-93 (gray). B: pooled data (means ± SE) of Ca²⁺ decline with 3 mM [Ca²⁺]_o + KN-93 (black) or ISO + KN-93 (gray). C: pooled data (means ± SE) of Ca²⁺ transient decline time intervals with 3 mM [Ca²⁺]_o + KN-93 or ISO + KN-93. *P < 0.05 vs. corresponding 3 mM [Ca²⁺]_o (n = 15 cells/4 hearts).

We also analyzed the time intervals with KN-93 between 25–50%, 50–75%, and 75–90%, as described above. The data in Fig. 8C show that only the Ca²⁺ transient decline in the first 50% is significantly different. Thus, CAMKII inhibition (and PLB Thr¹⁷ phosphorylation) had no effect on the ISO- or 3 mM [Ca²⁺]_o-induced Ca²⁺ decline. Thus, these data suggest that Ser¹⁶ is the major phosphorylation site responsible for the faster Ca²⁺ transient decline during β-AR stimulation.

Effects of 3 mm [Ca²⁺]_o and ISO on Ca²⁺ transient time to peak.

Ca²⁺ transient time to peak, the time it takes for the myocyte to reach its maximum systolic level, was compared with 3 mM [Ca²⁺]_o vs. ISO (± KN-93). Shown in Fig. 9A, ISO reached its peak significantly faster than 3 mM [Ca²⁺]_o (71 ± 2 vs. 79 ± 2 ms, P < 0.05). However, in the presence of KN-93, the time to peak with 3 mM [Ca²⁺]_o and ISO is similar (78 ± 3 vs. 76 ± 3 ms) (Fig. 9B). Therefore, the faster time to peak during β-AR stimulation with ISO is likely because of the activation of CAMKII.

Fig. 9.

Effects of 3 mM [Ca²⁺]_o and ISO on Ca²⁺ transient time to peak. A: pooled data (means ± SE) of Ca²⁺ transient time to peak with 3 mM [Ca²⁺]_o (black) or ISO (gray). B: pooled data (means ± SE) of Ca²⁺ transient time to peak with 3 mM [Ca²⁺]_o + KN-93 or ISO + KN-93. *P < 0.05 vs. 3 mM [Ca²⁺]_o (−KN-93: n = 21 cells/10 hearts, +KN-93: n = 15 cells/4 hearts).

DISCUSSION

Our data show that, when we compared matched systolic [Ca²⁺]_i levels with ISO and 3 mM [Ca²⁺]_o, PLB Ser¹⁶ phosphorylation is the major factor for the faster Ca²⁺ decline. Interestingly, this faster decline was only observed during the first 50% of the decline. β-AR stimulation also resulted in a faster Ca²⁺ transient time to peak, which was prevented by CAMKII inhibition.

β-AR stimulation.

Stimulation of the β-AR pathway results in the classical positive inotropic and lusitropic effect (4), that is, in terms of Ca²⁺ handling, an increase in systolic [Ca²⁺]_i levels with a faster time to peak and decline. In mouse myocytes, the majority (>93%) of the Ca²⁺ decline is the result of SR uptake via SERCA (2). SERCA is regulated by PLB. Ca²⁺ binding to SERCA results in the dissociation of PLB from SERCA, which will result in enhanced activation. The higher [Ca²⁺]_i levels during β-AR stimulation should result in further dissociation of PLB from SERCA and faster Ca²⁺ decline. It has been observed that, with higher systolic Ca²⁺ levels, there is a faster Ca²⁺ transient decline (3). PLB is also a phosphoprotein and a key end target of the β-AR pathway. Thus, PKA phosphorylates PLB on Ser¹⁶ (21), and CAMKII phosphorylates PLB on Thr¹⁷. Phosphorylation of PLB also results in dissociation of PLB from SERCA. Hence, it is unknown which is the major factor accounting for the faster Ca²⁺ transient decline during β-AR stimulation: PLB phosphorylation or the greater systolic Ca²⁺ levels.

High extracellular Ca²⁺ vs. ISO.

We investigated Ca²⁺ transient kinetics with ISO and 3 mM [Ca²⁺]_o. ISO and 3 mM [Ca²⁺]_o increased peak systolic Ca²⁺ levels and increased the Ca²⁺ decline rate (Fig. 1). Interestingly, the decline rate with ISO was faster compared with 3 mM [Ca²⁺]_o. To be assured that the faster decline observed with ISO was the result of PLB phosphorylation and not the small difference in systolic Ca²⁺ levels, we examined matched Ca²⁺ transient amplitudes (Fig. 2).

Using the matched amplitudes, we investigated the time it took the Ca²⁺ transient decline to reach 25, 50, 75, and 90% decline from its peak (Fig. 3A). Our data show that, at each time point, the Ca²⁺ decline rate was faster with ISO. We further analyzed these effects by examining various time intervals of the Ca²⁺ transient decline. We measured the interval from 25–50%, 50–75%, and 75–90% decline from its peak (Fig. 3B). Interestingly, our data show that ISO resulted in a faster decline rate during the initial 50% of the decline. These results do not match our hypothesis that PLB phosphorylation would result in a faster decline during the final phase of the decline. Therefore, we now propose that systolic Ca²⁺ levels in mouse myocytes, even during β-AR stimulation, do not reach the V_max of the pump, but, during the first 50% of the decline, the dissociation constant between 3 mM [Ca²⁺]_o and ISO is at its greatest difference at this point. Furthermore, we were able to show that, at 300 nm [Ca²⁺]_i, ISO has a greater SR Ca²⁺ uptake rate compared with 3 mM [Ca²⁺]_i (Fig. 5C). Thus, ISO and, more specifically, Ser¹⁶ phosphorylation, accelerates the SR Ca²⁺ uptake at [Ca²⁺]_i levels at our RT₅₀.

PLB phosphorylation with high extracellular Ca²⁺ and ISO.

As mentioned, β-AR stimulation leads to the phosphorylation of Ser¹⁶ by PKA. It has also been shown that β-AR stimulation and increased [Ca²⁺]_o can activate CAMKII and PLB Thr¹⁷ phosphorylation (28, 35). Thus, it has been suggested that Thr¹⁷ phosphorylation is also involved in the faster Ca²⁺ decline rates. Our data show that there was a large increase in Ser¹⁶ phosphorylation with ISO compared with 3 mM [Ca²⁺]_o; however, Thr¹⁷ phosphorylation levels were similar (Fig. 2). Because our Ca²⁺ transient amplitudes were matched, the only difference between our two datasets is the level of Ser¹⁶ phosphorylation. Thus, we believe our faster Ca²⁺ decline with ISO compared with 3 mM [Ca²⁺]_o is the result of Ser¹⁶ phosphorylation.

Effect of PLB Ser¹⁶ phosphorylation on the relationship between systolic Ca²⁺ levels and [Ca²⁺]_i decline rates.

We also examined the relationship between maximum systolic Ca²⁺ levels and the Ca²⁺ transient decline rate measured as RT₅₀ (Fig. 4). Our data show that, with low Ser¹⁶ phosphorylation, as systolic Ca²⁺ levels increased there was a direct relationship for a faster Ca²⁺ transient decline rate. This is consistent with a previous study (3). There was a similar tendency when PLB Ser¹⁶ was phosphorylated. However, when we examined the slopes of each line, the ISO value was considerably less than the 3 mM [Ca²⁺]_o value. Thus, the ISO group is less dependent upon Ca²⁺ transient peak. In addition, the r² value for the ISO data indicated that we did not have a good fit. Thus, there is less of a correlation between Ca²⁺ transient amplitude and RT₅₀ in the ISO group. Our data suggests that the Ca²⁺ transient decline rate with PLB Ser¹⁶ phosphorylation is much less dependent on systolic Ca²⁺ levels. We further analyzed this by grouping together the myocytes that had a higher response to 3 mM [Ca²⁺]_o and grouping together the myocytes that had a higher response to ISO (Fig. 5). Interestingly, even though the 3 mM [Ca²⁺]_o group had a significantly higher Ca²⁺ transient amplitude, this group had a significantly slower Ca²⁺ decline rate. Taken together, our data suggest that Ser¹⁶ phosphorylation is the major factor responsible for the faster Ca²⁺ transient decline rate with β-AR stimulation.

Effects of β₂-AR stimulation on Ca²⁺ transient decline.

To further investigate the role of Ser¹⁶ phosphorylation on accelerating Ca²⁺ decline, we tested if β₂-AR stimulation would change Ca²⁺ transient decline rates. β₂-AR stimulation is known to increase Ca²⁺ transients via local regulation of Ca²⁺ entry, but, because of its coupling with G_i and its compartmentalization, β₂-AR stimulation does not phosphorylate Ser¹⁶ PLB to the extent of β₁-AR (23). Thus, if Ser¹⁶ phosphorylation is the driving force behind the accelerated Ca²⁺ decline, β₂-AR stimulation should not result in faster Ca²⁺ decline with matching Ca²⁺ transient peaks. Our data demonstrate that this was indeed the case. That is, matched Ca²⁺ transient peaks from 3 mM [Ca²⁺]_o and Zint had similar Ca²⁺ decline rates. Thus, this provided additional evidence that Ser¹⁶ phosphorylation is the driving force behind the accelerated Ca²⁺ decline.

Effects of CAMKII inhibition on Ca²⁺ transient decline.

Because we did observe Thr¹⁷ phosphorylation with ISO and 3 mM [Ca²⁺]_o, we examined if this played any role in Ca²⁺ decline. The CAMKII inhibitor KN-93 had no effect on the maximal amplitude of the Ca²⁺ transient or the decline rate with ISO or 3 mM [Ca²⁺]_o (Fig. 7). Although not significant, CAMKII inhibition blunted the Ca²⁺ transient amplitude with β-AR stimulation, consistent with previous data (6). We also examined the effects of Ca²⁺ transient decline with matched Ca²⁺ transient amplitudes during ISO and 3 mM [Ca²⁺]_o perfusion with KN-93. We observed a similar effect, that is, even with CAMKII inhibition, ISO resulted in a faster Ca²⁺ decline during the first 50% (Fig. 8). Thus, under our experimental conditions, Thr¹⁷ phosphorylation does not play a role in enhancing the Ca²⁺ transient decline rate. Previous studies have found, for CAMKII-mediated PLB Thr¹⁷ phosphorylation, there must be concomitant phosphatase inhibition or acidosis (28, 35). Because we did not inhibit phosphatases with 3 mM [Ca²⁺]_o, we would get minimal phosphorylation. During β-AR stimulation, there is an inhibition of the phosphatases (29). However, this phosphorylation step occurred much slower (∼5 min) than Ser¹⁶ phosphorylation. Given that we were using Ca²⁺ transients that had matched amplitudes, we picked ISO-stimulated Ca²⁺ transients before the peak was reached (∼2–3 min). With the short perfusion time and the lack of phosphatase inhibition or acidosis, CAMKII did not phosphorylate PLB and hence had no effect on ISO-stimulated Ca²⁺ transient decline. Furthermore, similar to our conclusions, these authors showed that the lusitropic effect of β-AR stimulation correlated with PLB Ser¹⁶ phosphorylation.

Other factors that may contribute to Ca²⁺ decline could be the Na⁺/Ca²⁺ exchanger (NCX) or troponin (Tn) I phosphorylation. NCX contributes to the Ca²⁺ transient decline (2). Studies have found that β-AR stimulation is able to enhance NCX current (30). However, this is controversial, since others did not observe this effect (10). Under our experimental conditions, we believe that NCX does not play a role in the effects of β-AR stimulation on Ca²⁺ transient decay. NCX plays a minor role in the Ca²⁺ decline rates in mouse myocytes (<5%); thus, this effect, if any, would be minor at best. Furthermore, with the use of the NCX knockout mouse, there was no difference in Ca²⁺ transient decline rates with β-AR stimulation in WT compared with knockout myocytes (14). In larger species (e.g., rabbit, human, etc.) NCX plays a greater role in the Ca²⁺ transient decline rate (∼30%) (2). Consequently, there may be a species difference on the role of NCX and regulation of Ca²⁺ transient decay during β-AR stimulation. However, in rabbit myocytes during β-AR stimulation, the SR accounts for ∼90% of the Ca²⁺ removal (36). Thus, the effect of NCX would still be minor. Therefore, during β-AR stimulation, we believe the major stimulus for the increased acceleration of the Ca²⁺ transient decline is PLB Ser¹⁶ phosphorylation.

Phosphorylation of TnI during β-AR stimulation desensitizes the myofilaments to Ca²⁺ and results in a greater disassociation of Ca²⁺ from TnC. This may contribute to the faster Ca²⁺ decline observed with ISO. However, a previous study has shown that TnI phosphorylation does not play a role in Ca²⁺ decline in an unloaded preparation (i.e., isolated myocytes) (24). Thus, we do not believe TnI phosphorylation plays a role in the faster Ca²⁺ decline with ISO under our experimental conditions.

Ca²⁺ transient time to peak with high extracellular Ca²⁺ and ISO.

We also measured the time it took the Ca²⁺ transient to reach its peak point (i.e., time from stimulation to peak). We compared the effects of ISO and 3 mM [Ca²⁺]_o on Ca²⁺ transient time to peak. ISO perfusion resulted in a significantly faster Ca²⁺ transient time to peak, and this effect was prevented by CAMKII inhibition (Fig. 9). A previous study also observed that, with matched SR Ca²⁺ loads and L-type Ca²⁺ current, ISO also resulted in a faster rate of [Ca²⁺]_i rise and implicated the ryanodine receptor (RyR) (11). Studies have shown that CAMKII can phosphorylate the RyR (12, 15, 27). Additionally, studies have shown that, during β-AR stimulation, CAMKII phosphorylates the RyR to increase its activity (6, 8). Our results with KN-93 may seem surprising in that CAMKII can affect Ca²⁺ transient upstroke but not decline. However, a previous study found that there was a time course disparity in Thr¹⁷ phosphorylation and RyR phosphorylation by CAMKII (16). That is, with increasing frequency of stimulation, PLB Thr¹⁷ phosphorylation did not change in the first 40 s and then linearly increased for 5 min. RyR Ser²⁸¹⁴, the CaMKII site (although controversial), reached maximum phosphorylation after 90 s of increasing the frequency of stimulation. Although these measurements were performed using a different protocol than ours (force frequency vs. β-AR stimulation), there does appear to be a different time dependence to PLB and RyR CaMKII phosphorylation. Our data using β-AR stimulation are consistent with these results in that we observed an apparent effect of CaMKII on RyR function but not PLB function after 3 min. Thus, β-AR stimulation will activate CAMKII to likely increase RyR activity, which results in a faster Ca²⁺ transient time to peak. The faster time to peak may also be because of faster SR Ca²⁺ uptake. Nevertheless, a previous study found that transgenic mice overexpressing SERCA2a did not result in a faster Ca²⁺ transient time to peak (13). Our data are consistent with these observations, since our faster time to peak with ISO was prevented by CaMKII inhibition with KN-93. Yet, KN-93 had no effect on Ca²⁺ transient decline rates. Thus, if increased SERCA activity was responsible for the faster time to peak, this effect would not have been prevented by KN-93.

Our study also has implications for other mediators of PLB phosphorylation in addition to the β-AR/PKA pathway [i.e., phosphatases, SERCA activators via glutathionylation (1), nitroxyl-induced nitrosylation of PLB (9), reactive nitrogen species, etc.]. For example, nitric oxide (NO) is an important regulator of cardiac contraction (38). We have previously shown that neuronal NO synthase (NOS1) signaling targets PLB to modulate myocyte contraction (37). We have shown that NOS1 knockout myocytes had decreased PLB Ser¹⁶ phosphorylation, depressed Ca²⁺ transient amplitude, and a slower Ca²⁺ transient decline. It was unknown if the slower [Ca²⁺]_i decline in these myocytes was because of the reduced PLB Ser¹⁶ phosphorylation and/or blunted Ca²⁺ transient amplitude. Our current results would suggest that the NOS1-mediated PLB Ser¹⁶ phosphorylation is the major factor on Ca²⁺ transient decline rather than the amplitude of the Ca²⁺ transient.

Some limitations of our study include only using one mouse strain. However, our previous work has shown no difference in force decline during β-AR stimulation in three different mouse strains (33). Thus, we believe that our observations are valid for most (if not all) mouse strains. β-AR stimulation also increases heart rate. Increasing pacing frequency by itself also accelerates the Ca²⁺ decline. However, this is independent of PLB phosphorylation and a separate mechanism (34). Thus, we did not investigate the effects of pacing. In addition, we only used one ISO concentration. We believe using lower ISO concentrations will produce the same qualitative results, but not as great quantitatively. Furthermore, we believe that performing these experiments will not change our conclusion that PLB Ser¹⁶ is the major factor for the faster Ca²⁺ transient decline during β-AR stimulation.

In conclusion, we have shown that the major factor for the faster Ca²⁺ decline during β-AR stimulation is PLB Ser¹⁶ phosphorylation, which only occurs during the initial 50%. β-AR stimulation also results in a faster Ca²⁺ transient time to peak that was dependent on CAMKII.

GRANTS

This research was supported by the American Heart Association (Established Investigator Award 0740040N, P. M. L. Janssen) and the National Heart, Lung, and Blood Institute (RO1HL-071893, T. R. Shannon, K02HL-094692, and R01HL-079283, M. T. Ziolo).

DISCLOSURES

No conflicts of interest are declared by the authors.