Phosphorylation of the ryanodine receptor 2 at serine 2030 is required for a complete β-adrenergic response

Abstract

During physical exercise or stress, the sympathetic system stimulates cardiac contractility via β-adrenergic receptor (β-AR) activation, resulting in protein kinase A (PKA)-mediated phosphorylation of the cardiac ryanodine receptor RyR2. PKA-dependent "hyperphosphorylation" of the RyR2 channel has been proposed as a major impairment that contributes to progression of heart failure. However, the sites of PKA phosphorylation and their phosphorylation status in cardiac diseases are not well defined. Among the known RyR2 phosphorylation sites, serine 2030 (S2030) remains highly controversial as a site of functional impact. We examined the contribution of RyR2-S2030 to Ca²⁺ signaling and excitation-contraction coupling (ECC) in a transgenic mouse with an ablated RyR2-S2030 phosphorylation site (RyR2-S2030A^+/+). We assessed ECC gain by using whole-cell patch-clamp recordings and confocal Ca²⁺ imaging during β-ARs stimulation with isoproterenol (Iso) and consistent SR Ca²⁺ loading and L-type Ca²⁺ current (I _Ca) triggering. Under these conditions, ECC gain is diminished in mutant compared with WT cardiomyocytes. Resting Ca²⁺ spark frequency (CaSpF) with Iso is also reduced by mutation of S2030. In permeabilized cells, when SR Ca²⁺ pump activity is kept constant (using 2D12 antibody against phospholamban), cAMP does not change CaSpF in S2030A^+/+ myocytes. Using Ca²⁺ spark recovery analysis, we found that mutant RyR Ca²⁺ sensitivity is not enhanced by Iso application, contrary to WT RyRs. Furthermore, ablation of RyR2-S2030 prevents acceleration of Ca²⁺ waves and increases latency to the first spontaneous Ca²⁺ release after a train of stimulations during Iso treatment. Together, these results suggest that phosphorylation at S2030 may represent an important step in the modulation of RyR2 activity during β-adrenergic stimulation and a potential target for the development of new antiarrhythmic drugs.

Graphical abstract

Figure 1.

ECC gain is decreased in S2030A^+/+ myocytes after β-adrenergic stimulation and matched SR Ca²⁺ loading. (A) Depolarization of voltage-clamped cells from −40 to −25 mV (left) or to 0 mV (right) in control conditions and during β-adrenergic stimulation with 100 nmol/liter Iso (N = 5, n = 11 for both WT and S2030A^+/+ animals). The figure shows representative line profiles of the Ca²⁺ transients (ΔF/F₀; black traces for control and red for Iso), confocal line-scan images, and I_Ca traces for both WT and S2030A^+/+ cells. (B and C) ECC gain was obtained from the ratio of the maximal Ca²⁺ transient amplitude and the peak I_Ca density (ΔF/F₀)/(pA/pF), normalized to the SR Ca²⁺ content. (D) The ECC gain was obtained in conditions of matched SR Ca²⁺ load and I_Ca between all cell groups. The gain was minimally influenced by Iso in WT cells. However, the mutation affected the ECC gain ratio in Iso, mostly because of a decreased Ca²⁺ release during β-adrenergic stimulation (see Table S2). Please note that in this experiment, the SR Ca²⁺ content was experimentally matched between control and Iso. *, P < 0.05, Iso versus control; †, P < 0.05, S2030A versus WT. The whiskers cover the range from 10% to 90%.

Figure 2.

RyR2-S2030 site phosphorylation level is increased by Iso stimulation. (A) Representative blot showing total RyR2 expression levels in WT (N = 8) and S2030A^+/+ (N = 10) animals. (B) Representative blot showing the level of dephosphorylation at RyR2-S2030 site in WT animals (N = 4), before (C, control) and after 100 nmol/liter Iso application. The numbers 1 and 2 identify different animals. The level of the phosphorylation was expressed as the ratio of the dephospho-RyR2-S2030 and the total RyR ODs, both normalized for the GAPDH signal. Please note that a low level of dephosphorylation corresponds to a high level of phosphorylation. (C) Representative blot showing the maximal level of dephosphorylation at the RyR2-S2030 site in WT animals (N = 4) obtained with a high dose of Iso (1 µmol/liter) combined with the phosphatase inhibitor OA (10 µmol/liter). The dephosphorylation levels were normalized to the control (100%). Iso at the concentration used in all experiments (100 nmol/liter) resulted in a decrease in the signal corresponding to an increased level of phosphorylation (69 ± 6% compared with the control). The combination of high Iso concentration and OA further decreased the signal (44.4 ± 4%). The data are expressed as mean ± SEM of all measurements. *, P < 0.05 versus control; **, P < 0.05, Iso versus OA + Iso.

Figure 3.

S2030 site ablation limits the Iso-dependent increase in CaSpF. (A) Confocal line-scan images showing Ca²⁺ sparks in intact cells in control and after 3 min of 100 nmol/liter Iso treatment for WT (N = 4, n = 12), S2030A^+/+ (N = 4, n = 13), and S2808A^+/+ (N = 3, n = 10) myocytes. (B) The basal CaSpF (number of sparks per 100 µm/s) was similar among the animal groups; after application of Iso, the CaSpF in S2030A^+/+ myocytes was significantly lower compared with WT and S2808A^+/+ cells. (C) SR Ca²⁺ content was not significantly different between the mouse models. Please note that unlike in the patch-clamp experiments, the SR Ca²⁺ load here was not matched between control and Iso. *, P < 0.05 versus control; †, P < 0.05 versus WT. The whiskers cover the range from 10% to 90%.

Figure 4.

Maximal SERCA stimulation unmasks a limited cAMP-dependent increase in CaSpF. (A) Representative confocal line-scan images showing Ca²⁺ sparks in β-escin permeabilized cells before and after 5 min in 5 µmol/liter cAMP. (B) Basal CaSpF was similar for WT (N = 5, n = 12) and S2030A^+/+ (N = 4, n = 10; 2.98 ± 0.34 vs. 3.39 ± 0.47 sparks per 100 µm/s, respectively; i). cAMP enhanced SpF in both cell types. SR Ca²⁺ content was increased after treatment with cAMP (ii). (C) Confocal line-scan images showing Ca²⁺ sparks in myocytes pretreated with 100 µg/ml 2D12-Fab (15 min) before and after cAMP application. (D) 2D12-Fab increased basal CaSpF (6.01 ± 0.41 vs. 5.51 ± 0.56 sparks per 100 µm/s, respectively) for WT (N = 6, n = 10) and S2030A^+/+cells (N = 4, n = 10); cAMP further increased CaSpF only in WT cells. *, P < 0.05 versus control; **, P < 0.05 versus 2D12-Fab; †, P < 0.05 versus WT. The whiskers cover the range from 10% to 90%.

Figure 5.

Ca²⁺ spark-to-spark delay shortening is blunted during β-adrenergic stimulation in S2030A^+/+ cells. (A) Representative line-scan image in control conditions (i). A very low concentration of ryanodine (50 nmol/liter) entrains repetitive sparks after binding to only a single monomeric RyR channel subunit in an entire couplon (ii). The spark-to-spark delay and the relative spark amplitudes were measured to investigate Ca²⁺ spark restitution (iii). (B) Histograms of spark-to-spark delays for control WT (328.8 ms, 95% CI, 313.8–348.3 ms; N = 10, n = 36, 937 spark pairs; i), Iso WT (261.8 ms, 95% CI, 254.5–269.4 ms; N = 11, n = 44, 1,783 spark pairs; ii), control S2030A^+/+ (308.3 ms, CI 95%: 291.6–329.2 ms; N = 6, n = 20, 488 spark pairs; iii), and Iso S2030A^+/+ cells (299.6 ms, CI 95%: 286.0–310.9 ms; N = 7, n = 27, 929 spark pairs; iv). Red lines identify the median values.

Figure 6.

Iso does not modify the latency of the first spontaneous Ca²⁺ release in S2030A^+/+ myocytes. (A) Line-scan images were recorded in the end of a train of electrical stimulations at 1 Hz for 30 s. Each cell was recorded in control and after 3 min of 100 nmol/liter Iso treatment. The figure also shows line profiles of the triggered Ca²⁺ transients and the spontaneous Ca²⁺ releases (ΔF/F₀; black traces for the control and red for Iso). The latency was calculated considering the last electrical stimulus and the first spontaneous Ca²⁺ release (arrows). (B) SCaW latency (17.72 ± 3.42 s vs. 8.08 ± 1.16 s), respectively, for control and after Iso stimulation for WT (18.04 ± 1.95 s for control vs. 14.27 ± 2.51 s for Iso in S2030A^+/+). Latency values were not paired, because not all cells displayed SCaW. (C) Occurrence of spontaneous release (percentage of cells that show at least one Ca²⁺ wave; N = 4, n = 16 for WT; N = 4, n = 15 for S2030A^+/+). (D) Comparison of wave speed (WT: N = 4, n = 11; 78.42 ± 5.05 µm/s [control] vs. 99.31 ± 4.03 µm/s [Iso]); S2030A^+/+: N = 4, n = 15; 83.49 ± 5.56 µm/s [control] vs. 85.40 ± 4.64 µm/s [Iso]). In all experiments shown, the SR Ca²⁺ load was not matched experimentally between control and Iso. *, P < 0.05 versus control; †, P < 0.05 versus WT. The whiskers cover the range from 10% to 90%.

Figure 7.

S2030 ablation generates adaptive changes at RyR2-S2808 and S2814 phosphorylation sites in mutant cells. (A) Representative blot showing the phosphorylation of RyR2 at the S2808 and S2814 sites in both WT and S2030A^+/+ cells in control (C) and during Iso stimulation. (B and C) The level of the phosphorylation was expressed as the ratio of the respective phospho- and total RyR ODs, both normalized to the GAPDH signal (N = 8 mice for both WT and S2030A^+/+ cells were used to compare baseline phosphorylation; between these, N = 4 animals were also assessed after Iso application). These data were tested for paired statistical difference. *, P < 0.05 versus control; †, P < 0.05 versus WT. The whiskers cover the range from 10% to 90%.