S100A1's single cysteine is an indispensable redox switch for the protection against diastolic calcium waves in cardiomyocytes

Abstract

The EF-hand calcium (Ca²⁺) sensor protein S100A1 combines inotropic with antiarrhythmic potency in cardiomyocytes (CMs). Oxidative posttranslational modification (ox-PTM) of S100A1's conserved, single-cysteine residue (C85) via reactive nitrogen species (i.e., S-nitrosylation or S-glutathionylation) has been proposed to modulate conformational flexibility of intrinsically disordered sequence fragments and to increase the molecule's affinity toward Ca²⁺. Considering the unknown biological functional consequence, we aimed to determine the impact of the C85 moiety of S100A1 as a potential redox switch. We first uncovered that S100A1 is endogenously glutathionylated in the adult heart in vivo. To prevent glutathionylation of S100A1, we generated S100A1 variants that were unresponsive to ox-PTMs. Overexpression of wild-type (WT) and C85-deficient S100A1 protein variants in isolated CM demonstrated equal inotropic potency, as shown by equally augmented Ca²⁺ transient amplitudes under basal conditions and β-adrenergic receptor (βAR) stimulation. However, in contrast, ox-PTM defective S100A1 variants failed to protect against arrhythmogenic diastolic sarcoplasmic reticulum (SR) Ca²⁺ waves and ryanodine receptor 2 (RyR2) hypernitrosylation during βAR stimulation. Despite diastolic performance failure, C85-deficient S100A1 protein variants exerted similar Ca²⁺-dependent interaction with the RyR2 than WT-S100A1. Dissecting S100A1's molecular structure-function relationship, our data indicate for the first time that the conserved C85 residue potentially acts as a redox switch that is indispensable for S100A1's antiarrhythmic but not its inotropic potency in CMs. We, therefore, propose a model where C85's ox-PTM determines S100A1's ability to beneficially control diastolic but not systolic RyR2 activity.NEW & NOTEWORTHY S100A1 is an emerging candidate for future gene-therapy treatment of human chronic heart failure. We aimed to study the significance of the conserved single-cysteine 85 (C85) residue in cardiomyocytes. We show that S100A1 is endogenously glutathionylated in the heart and demonstrate that this is dispensable to increase systolic Ca²⁺ transients, but indispensable for mediating S100A1's protection against sarcoplasmic reticulum (SR) Ca²⁺ waves, which was dependent on the ryanodine receptor 2 (RyR2) nitrosylation status.

Keywords: S100A1; calcium; diastolic calcium waves; intrinsically disordered protein; ryanodine receptor 2.

Figure 1.

S100A1 is endogenously glutathionylated in vivo. A, left: immunoprecipitation of S100A1 from mouse heart tissue demonstrating the specificity of S100A1-IP. A, right: input and S100A1 IP after modified biotin switch assay (mBSA) to detect oxidative posttranslational modification [ox-PTM; reduced with dithiothreitol (DTT)], S-nitrosylation (SNO; reduced with ascorbate), and S-glutathionylation [SSG; reduced by glutaredoxin (Grx) enzymatic reaction]. Yellow overlay indicates that S100A1 is modified by ox-PTM and SSG. Representative from two experiments. B: representative immunoblot of maleimide incorporation after immunoprecipitation of wild-type (WT) and Cys-deficient S100A1 protein from HEK293T cells. Cell lysates were stained with maleimide to label reactive cysteine residues after control, S100A1, C85A, or C85S overexpression. Input (20%) for S100A1 and GAPDH as a loading control (left) is depicted ensuring matching S100A1 overexpression. Representative from two experiments.

Figure 2.

Wild-type (WT) and single cysteine residue 85 (C85)-deficient S100A1 overexpression increase systolic Ca²⁺ transient amplitudes in cardiomyocytes (CMs). A: representative immunoblots from control, S100A1, C85A, and C85S-treated adult rat CMs (ARCMs) 24 h after adenoviral transduction demonstrating equal S100A1 overexpression and viral load. Representative from four experiments. B: representative tracings of electrically stimulated (2 Hz) steady-state Ca²⁺ transients from control, S100A1, C85A, and C85S overexpressing ARCMs. C: quantification of calcium transient amplitude. Overexpression of WT and C85-deficient S100A1 results in increased calcium transients. Note that isoproterenol (Iso) stimulation resulted in a significant increase in calcium transient amplitude in all groups (statistical comparison not shown because of space constraints). D: quantification of diastolic calcium levels. Overexpression of C85A results in enhanced diastolic calcium levels, whereas diastolic calcium levels were lower in C85S-treated cardiomyocytes after Iso stimulation. Note that Iso stimulation results in a significant increase in diastolic calcium levels in control cells, but not in any other group (statistical comparison not shown because of space constraints). Data are given as means ± SE; n = 29, 22, 21, 21, 29, 19, 27, and 25 of control, S100A1, C85A, C85S, control Iso, S100A1 Iso, C85A Iso, and C85S Iso, respectively. Statistical comparison was performed with one-way ANOVA with Sidak’s multiple comparison post hoc test (C and D). *P < 0.05, **P < 0.005, ***P < 0.001, ****P < 0.0001.

Figure 3.

Wild-type (WT) but not single cysteine residue 85 (C85)-deficient S100A1 protects against rapid pacing-induced calcium waves in cardiomyocytes (CMs). A: representative tracings of electrically stimulated Ca²⁺ transients from basal and isoproterenol (Iso)-stimulated (1 µM) control, S100A1, C85A, or C85S-treated adult rat CMs (ARCMs). Cells were electrically stimulated (3 Hz) for 20 s followed by a 20-s resting period and restimulation (3 Hz) for 20 s. B: percentage of cells with spontaneous calcium waves was quantified. Iso stimulation results in an increase in calcium waves in control, C85A, and C85S-treated cardiomyocytes but not after WT-S100A1 overexpression. C: number of calcium waves per cell was quantified. WT-S100A1-treated cardiomyocytes demonstrate fewer calcium waves after Iso stimulation. Iso stimulation results in an increase in calcium waves in control, C85A, and C85S-treated cardiomyocytes but not after WT-S100A1 overexpression (statistical comparison not shown because of space constraints). Data are given as means ± SE; n = 24, 23, 28, 22, 19, 18, 20, and 22 of control, S100A1, C85A, C85S, control Iso, S100A1 Iso, C85A Iso, and C85S Iso, respectively). Statistical comparison was performed with χ² test/Fisher’s exact test (B) or one-way ANOVA with Sidak’s multiple comparison post hoc test (C). §P < 0.05, control Iso vs. S100A1 Iso, ****P < 0.0001.

Figure 4.

Wild-type (WT) and single cysteine residue 85 (C85)-deficient S100A1 overexpression has no impact on ryanodine receptor 2 (RyR2) and phospholamban (PLB) phosphorylation in cardiomyocytes (CMs). A: representative immunoblots for total PLB, PLB Ser-16, and Thr-17, total RyR2, RyR Ser-2808, and Ser-2814 in control, S100A1, C85A-, and C85S-treated adult rat CMs (ARCMs). Quantification shows no differences in RyR2 (B) and PLB (C) phosphorylation levels between control, S100A1, C85A, and C85S-treated CMs, whereas isoproterenol (Iso) stimulation increases protein kinase A (PKA)-dependent phosphorylation of RyR2 and PLB. Data are given as means ± SE; n = 3–4; Statistical comparison was performed with one-way ANOVA with Sidak’s multiple comparison post hoc test.

Figure 5.

Wild-type (WT) and single cysteine residue 85 (C85)-deficient S100A1 interact with the ryanodine receptor 2 (RyR2). A: representative immunoblot for RyR2 and coprecipitating S100A1 protein after RyR2 immunoprecipitation from control, S100A1, C85A-, or C85S-treated neonatal rat cardiomyocytes (NRCMs). Left: input (20%) for S100A1 and GAPDH as loading control. B: representative proximity ligation assay (PLA) images from control, S100A1, C85A-, and C85S-treated NRCMs showing S100A1/RyR2 interaction (red dots). Inset: magnification is twofold. Scale bar represents 20 µm. C: statistical analysis shows a twofold increase of the S100A1/RyR2 binding ratio; n = 3. Data are given as means ± SE. Statistical comparison was performed with one-way ANOVA with Sidak’s multiple comparison post hoc test (C). **P < 0.005, ***P < 0.001.

Figure 6.

Wild-type (WT) S100A1 prevents ryanodine receptor 2 (RyR2) hypernitrosylation during β-adrenergic receptor stimulation. A: representative proximity ligation assay (PLA) images from control, S100A1, C85A-, and C85S-treated neonatal rat cardiomyocytes (NRCMs) showing pan-RyR2 nitrosylation (red dots). Inset: magnification is twofold. Scale bar represents 20 µm. B: quantification of A. RyR2 is hypernitrosylated after isoproterenol (Iso) stimulation, which is prevented after WT, but not after C85-deficient S100A1 protein overexpression; n = 5–9. C: quantification of RyR2-S-nitrosylation (SNO) after S-methyl-l-thiocitrulline (SMLT) stimulation in indicated groups. SMLT reduced RyR2-SNO during Iso stimulation in control cardiomyocytes (CMs); n = 3–4. Data are given as means ± SE. Statistical comparison was performed with one-way ANOVA with Sidak’s multiple comparison post hoc test (B and C). *P < 0.05, **P < 0.005, ****P < 0.0001.

Figure 7.

Proposed model for S100A1 regulation of diastolic ryanodine receptor 2 (RyR2) leakage. During β-adrenergic receptor (βAR) stimulation, increased phosphorylation and nitrosylation renders RyR2 leaky, resulting in diastolic Ca²⁺ release. Overexpression of S100A1 prevents RyR2 nitrosylation and, subsequently, abrogates diastolic Ca²⁺ leak. Overexpression of S100A1 variants, which lack endogenous glutathionylation sites, fails to attenuate RyR nitroslyation and have no protective effect on diastolic Ca²⁺ leakage. Created with a licensed version of BioRender.com.