Targeting of phospholamban by peroxynitrite decreases β-adrenergic stimulation in cardiomyocytes

Abstract

Aims

Peroxynitrite production increases during the pathogenesis of numerous cardiac disorders (e.g. heart failure). However, limited studies have investigated the mechanism through which peroxynitrite exerts anti-adrenergic effects. Thus, the purpose of this study is to investigate the contribution of phospholamban (PLB), a critical excitation–contraction coupling protein, to the peroxynitrite-induced dysfunction.

Methods and results

Isolated myocytes from wild-type (WT, CF-1) and PLB knockout (PLB^−/−) mice were stimulated at 1 Hz, and myocyte shortening and Ca²⁺ transients were simultaneously recorded. PLB phosphorylation was measured via western blot. Myocytes were superfused with isoproterenol, a β-adrenergic agonist, and SIN-1, a peroxynitrite donor. SIN-1 superfusion dramatically decreased isoproterenol-stimulated Ca²⁺ transients and myocyte shortening in WT myocytes. These effects were inhibited upon addition of the peroxynitrite decomposition catalyst, FeTPPS. Surprisingly, SIN-1 had no functional effect on β-adrenergic-stimulated PLB^−/− myocytes. Western blot analyses revealed that SIN-1 significantly decreased isoproterenol-stimulated PLB_Ser16 phosphorylation. Experiments with the protein phosphatase inhibitor, okadaic acid, alleviated the SIN-1-induced functional effects and the decrease in PLB phosphorylation.

Conclusions

The peroxynitrite donor SIN-1 decreases β-adrenergic stimulation by reducing PLB_Ser16 phosphorylation via protein phosphatase activation. This peroxynitrite-induced decrease in PLB phosphorylation may be a key mechanism in the β-adrenergic dysfunction observed in many cardiomyopathies.

1. Introduction

The process of excitation–contraction coupling is responsible for contraction in the cardiomyocyte.¹ Following the cardiac action potential, L-type Ca²⁺ channels open to facilitate Ca²⁺ entry into the cell, triggering the opening of sarcoplasmic reticulum (SR) Ca²⁺ release channels (RyR) and the discharge of additional Ca²⁺ from the SR. This Ca²⁺ subsequently activates the myofilaments, resulting in myocyte contraction. Relaxation is primarily mediated by the SR Ca–ATPase/phospholamban (SERCA/PLB) complex, which serves to reuptake Ca²⁺ into the SR.

Phospholamban plays a critical role in excitation–contraction coupling by regulating SERCA uptake of Ca²⁺ into the SR, and is important in determining SR Ca²⁺ load and thus contractility.² Under basal conditions, PLB remains in a dephosphorylated state and inhibits SERCA uptake of Ca²⁺. PLB inhibition of SERCA can be relieved when PLB is phosphorylated, allowing greater uptake of Ca²⁺ into the SR. For example, activation of the β-adrenergic receptor signalling cascade leads to positive inotropic and lusitropic effects,³ which result mainly from the phosphorylation of PLB at its protein kinase A (PKA)-dependent serine 16 (Ser16) site.⁴ Additionally, PLB can also be phosphorylated at its Ca²⁺/calmodulin kinase II (CaMKII)-dependent threonine 17 (Thr17) site. There are several protein phosphatases that serve to dephosphorylate PLB, including protein phosphatase 1 (PP1) and protein phosphatase 2a (PP2a).⁵

Peroxynitrite (ONOO⁻) is formed upon the reaction of nitric oxide (NO) and superoxide $\left({\mathrm{O}}_2^{-}\right)$. Studies have shown that peroxynitrite is increased in many cardiomyopathies including ischemia/reperfusion injury, sepsis, and heart failure and is detrimental to cardiac function.⁶−¹² The negative effects of peroxynitrite have been confirmed in normal hearts during basal stimulation.¹³−¹⁵ Additionally, the peroxynitrite donor, SIN-1, was shown to have anti-adrenergic effects in isolated cardiomyocytes.^16,17 However, studies have also demonstrated positive inotropy with peroxynitrite.¹⁸−²⁰ This discrepancy may be due to the proposed biphasic nature of peroxynitrite, where peroxynitrite produces positive inotropic effects at low concentrations, but negative inotropic effects at high concentrations. Although the biphasic nature of peroxynitrite in the myocardium has been characterized, most studies have not investigated the mechanism(s) of peroxynitrite-induced β-adrenergic hyporesponsiveness. Therefore, despite this role for peroxynitrite in the modulation of cardiac contractility, little is known regarding the mechanism(s) underlying the effect(s) of peroxynitrite.

Therefore, the objective of this study is to evaluate critically the role of PLB in the peroxynitrite-induced β-adrenergic hyporesponsiveness. We hypothesize that peroxynitrite ultimately targets PLB and selectively decreases PKA-dependent Ser16 phosphorylation via activation of protein phosphatases.

2. Methods

2.1. Cardiomyocyte isolation

Ventricular myocytes were isolated as described previously (see Supplementary material online, available at http://www.science-direct.com).²¹ This investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85−23, revised 1996) and was approved by the Institutional Laboratory Animal Care and Use Committee.

2.2. Measurement of peroxynitrite release rate

Electron paramagnetic resonance (EPR) spectroscopy with 1-hydroxy-3-carboxy-2,2,5,5-tetramethylpyrrolidine hydrochloride (CP-H; Alexis, Lausen, Switzerland) was used to measure the rate of peroxynitrite release from SIN-1 under our experimental conditions as previously described (see Supplementary material online).²²

2.3. Simultaneous measurement of cellular Ca²⁺ transients and myocyte shortening

Ca²⁺ transients and myocyte shortening were measured in isolated myocytes exposed to various experimental solutions (control, ISO, ISO+SIN-1, etc.) as previously described (see Supplementary material online).²¹ All measurements were recorded at room temperature (22°C).

2.4. Western blot for phosphorylated PLB

Whole hearts were perfused with the various experimental solutions (control, ISO, ISO+SIN-1, etc.) using a Langendorff apparatus, homogenized, and analysed via western blot as previously described (see Supplementary material online).²¹ Membranes were probed using either a custom antibody to pentameric PLB (Zymed, San Francisco, CA, USA) or phosphorylated PLB_Ser16 (Cyclacel, Dundee, UK) or phosphorylated PLB_Thr17 (Cyclacel).

2.5. Solutions and drugs

Normal Tyrode control solution consisted of (in mmol/L): NaCl (140), KCl (4), MgCl₂ (1), CaCl₂ (1), Glucose (10), and HEPES (5); pH 7.4 adjusted with NaOH and/or HCl. Isoproterenol was used as a non-specific β-adrenergic agonist (ISO; Sigma, St. Louis, MO, USA). 3-Morpholinosydnonimine (SIN-1; Alexis) was used as a peroxynitrite donor. Okadaic acid (Sigma) was used as a PP1/PP2a inhibitor. 5,10,15,20-Tetrakis-[4-sulfonatophenyl]-porphyrinato-iron[III] (FeTPPS; Calbiochem, La Jolla, CA, USA) was used as a specific peroxynitrite decomposition catalyst. Forskolin (Sigma) was used as an adenylate cyclase activator. All solutions were made fresh daily.

2.6. Statistics

Data are presented as mean ± SEM. Statistical significance (P < 0.05) was determined between groups using an ANOVA (followed by Neuman–Keuls test) for multiple groups or a Student's paired t-test for two groups.

3. Results

3.1. Peroxynitrite production by SIN-1

Using EPR spectroscopy with CP-H, we determined the rate of peroxynitrite release by 200 μmol/L SIN-1 under our experimental conditions (normal Tyrode control solution, 22°C). At this concentration, SIN-1 released 18 nmol L⁻¹min⁻¹ of peroxynitrite over the same time course as our functional/biochemical experiments.

3.2. Negative inotropic effect of SIN-1 on WT myocyte function

We examined the effect of SIN-1 (200 μmol/L) on the maximal β-adrenergic response in isolated WT myocytes. Ca²⁺ transients and myocyte shortening were simultaneously recorded in response to various experimental conditions, as shown in Figure 1A. Upon reaching steady state in our normal Tyrode control solution (Ca²⁺ transient, 0.8 ± 0.2 ΔF/F₀; shortening, 3.8 ± 0.8 μm), perfusion with a maximal dose of ISO (1 μmol/L) produced a large increase in Ca²⁺ transient amplitude and shortening (Ca²⁺ transient, 1.4 ± 0.2 ΔF/F₀; shortening, 20.2 ± 3.6 μm) in WT myocytes (n = 10/5 hearts). After reaching steady state, co-infusion of 1 μmol/L ISO with 200 μmol/L SIN-1 significantly reduced β-adrenergic-stimulated Ca²⁺ transients and myocyte shortening in WT myocytes (Ca²⁺ transient, 1.1 ± 0.2 ΔF/F₀; shortening, 14.8 ± 3.4 μm, P < 0.05 vs. ISO). Ca²⁺ transient and myocyte shortening amplitudes were used to determine the % Δ from control (Figure 1B) and the % Δ from ISO. SIN-1 reduced β-adrenergic stimulation in WT myocytes regardless of whether myocytes were co-infused with ISO+SIN-1 prior to or after superfusion with ISO alone.

Figure 1.

SIN-1 decreases β-adrenergic stimulation in WT myocytes. (A) Individual, steady-state shortening (top) and Ca²⁺ transient (bottom) traces representing the anti-adrenergic effect of SIN-1 in a WT myocyte. (B) Pooled data (mean ± SEM) expressed as a % Δ from control demonstrating the effect of ISO (1 μmol/L, clear bar) and ISO+SIN-1 (200 μmol/L, filled bar) on shortening (top) and Ca²⁺ transients (bottom) in WT myocytes; *P < 0.05 vs. ISO. (C) FeTPPS inhibits the anti-adrenergic effect of SIN-1 in WT myocytes. Pooled data (mean ± SEM) expressed as % Δ from ISO demonstrating the effect of SIN-1 on β-adrenergic-stimulated shortening (top) and Ca²⁺ transients (bottom) in WT (clear bar, B) and WT+FeTPPS (filled bar) myocytes. *P < 0.01 vs. WT. (D) The anti-adrenergic effect of SIN-1 is not due to ISO oxidation. Pooled data (mean ± SEM) expressed as % Δ from ISO demonstrating the effect of SIN-1 on β-adrenergic-stimulated shortening (top) and Ca²⁺ transients (bottom) in WT (clear bar, C) vs. ISO washout (WO) in WT myocytes (filled bar). *P < 0.01 vs. WT.

Perfusion of WT myocytes (n = 18/4 hearts) with SIN-1 during sub-maximal ISO stimulation (0.01 μmol/L) also produced a significant reduction in Ca²⁺ transient amplitude and myocyte shortening (Ca²⁺ transient, −12 ± 2%; shortening, −13 ± 5% Δ from ISO, P < 0.05 vs. ISO). However, the anti-adrenergic effect of SIN-1 was not as great during sub-maximal ISO stimulation compared with maximal ISO (Ca²⁺ transient, −17 ± 4%; shortening, −28 ± 5% Δ from ISO). Additionally, perfusion with 200 μmol/L SIN-1 alone (n = 10/5 hearts) had no effect on contractility (Ca²⁺ transient, 0.7 ± 0.2 ΔF/F₀; shortening, 5.2 ± 1.0 μm) compared with WT myocyte function in normal Tyrode control solution alone (Ca²⁺ transient, 0.7 ± 0.3 ΔF/F₀; shortening, 5.0 ± 0.8 μm).

3.3. SIN-1-induced β-adrenergic hyporesponsiveness results from peroxynitrite

As SIN-1 is not a straightforward donor of peroxynitrite, we sought to confirm the causal species responsible for the anti-adrenergic effects of SIN-1. Upon reaching steady state in our normal Tyrode control solution (Ca²⁺ transient, 1.4 ± 0.2 ΔF/F₀; shortening, 3.0 ± 0.3 μm), perfusion with a maximal dose of ISO (1 μmol/L) produced a large increase in Ca²⁺ transient amplitude and myocyte shortening (Ca²⁺ transient, 4.9 ± 0.3 ΔF/F₀; shortening, 14.2 ± 2.2 μm) in WT myocytes (n = 21/3 hearts). Co-infusion of WT myocytes with 1 μmol/L ISO+200 μmol/L SIN-1 and 10 μmol/L FeTPPS, a specific peroxynitrite decomposition catalyst,²³ alleviated the anti-adrenergic effects of SIN-1 during maximal β-adrenergic stimulation (Ca²⁺ transient, 4.6 ± 0.3 ΔF/F₀; shortening, 13.8 ± 2.2 μm) compared with WT myocytes not treated with FeTPPS. This effect is shown in Figure 1C, which shows the % Δ from ISO, and thus implicates peroxynitrite as the causal species. Additionally, perfusion with 1 μmol/L ISO+10 μmol/L FeTPPS alone did not have a significant effect on the myocyte response to ISO (Ca²⁺ transient, 3.7 ± 0.8 ΔF/F₀; shortening, 9.2 ± 2.8 μm) compared with the myocyte response to 1 μmol/L ISO alone (Ca²⁺ transient, 3.6 ± 0.7 ΔF/F₀; shortening, 10.2 ± 2.8 μm).

The possibility also exists for peroxynitrite to oxidize catecholamines into inactive aminochromes.^15,24,25 Therefore, experiments were performed simulating the oxidation of ISO by SIN-1. Following a steady-state response to 1 μmol/L ISO, WT myocytes (n = 5/3 hearts) were perfused with control solution (normal Tyrode) in order to washout the effects of ISO (see Supplementary material online, Figure S1). However, washout only resulted in a slight decrease in Ca²⁺ transient amplitude and myocyte shortening over a 5 min period (Ca²⁺ transient, −3 ± 1%; shortening, −3 ± 3%Δ from ISO), while the maximal effect of SIN-1 generally occurred within 3 min and was much greater than that produced by ISO washout alone (Figure 1D; Ca²⁺ transient, −17 ± 4%; shortening, −28 ± 5% Δ from ISO, P < 0.01 vs. ISO washout). Additionally, experiments using forskolin (10 μmol/L), a direct activator of adenylate cyclase, were used to verify that peroxynitrite was not primarily affecting targets upstream of adenylate cyclase, including ISO oxidation. Perfusion with 1 μmol/L forskolin induced an increase in myocyte Ca²⁺ transients and shortening, similar to the effect seen with 1 μmol/L ISO, but upon co-infusion with 1 μmol/L forskolin+200 μmol/L SIN-1, a significant anti-adrenergic effect remained (data not shown).

3.4. Effect of SIN-1 on β-adrenergic-stimulated PLB^−/− myocyte function

In our previous study, PLB was identified as a potential target of SIN-1 signalling.¹⁶ Therefore, we examined the effect of SIN-1 on the β-adrenergic response in isolated PLB^−/− myocytes. PLB^−/− myocytes (n = 15/7 hearts) showed the typical enhanced basal contractility (Ca²⁺ transient, 1.9 ± 0.3 ΔF/F₀; shortening, 10.2 ± 1.2 μm), as shown in Figure 2A. Thus, PLB^−/− myocytes exhibited a reduced β-adrenergic responsiveness to 1 μmol/L ISO (Ca²⁺ transient, 2.3 ± 0.8 ΔF/F₀; shortening: 15.2 ± 2.4 μm) compared with WT, as has been demonstrated previously.²⁶ After reaching steady state, co-infusion with 1 μmol/L ISO+200 μmol/L SIN-1 had surprisingly little effect on the PLB^−/− myocytes (Ca²⁺ transient, 2.3 ± 0.8 ΔF/F₀; shortening, 15.6 ± 3.0 μm). Ca²⁺ transient and myocyte shortening amplitudes were used to determine the % Δ from control (Figure 2B) and the % Δ from ISO. SIN-1 did not affect β-adrenergic responsiveness in PLB^−/− myocytes regardless of whether myocytes were co-infused with ISO+SIN-1 prior to or after superfusion with ISO alone. Thus, the anti-adrenergic effect of SIN-1 was significantly greater in WT compared with PLB^−/− myocytes (Figure 2C; Ca²⁺ transient, −17 ± 4 and −3 ± 3%; shortening, −28 ± 5 and −2 ± 4% Δ from ISO, P < 0.05), and therefore indicates that PLB is a significant target in this peroxynitrite signalling pathway. Additionally, perfusion with 200 μmol/L SIN-1 alone (n = 7/4 hearts) had no significant effect on contractility (Ca²⁺ transient, 1.1 ± 0.2 ΔF/F₀; shortening, 11.6 ± 4.0 μm) compared with PLB^−/− myocyte function in normal Tyrode control solution alone (Ca²⁺ transient, 1.2 ± 0.2 ΔF/F₀; shortening, 12.2 ± 4.0 μm).

Figure 2.

SIN-1 has no effect in PLB^−/− myocytes. (A) Individual, steady-state shortening (top) and Ca²⁺ transient (bottom) traces representing the lack of an anti-adrenergic effect of SIN-1 in a PLB^−/− myocyte. (B) Pooled data (mean ± SEM) expressed as % Δ from control demonstrating the effect of ISO (1 μmol/L, clear bar) and ISO+SIN-1 (200 μmol/L, filled bar) on shortening (top) and Ca²⁺ transients (bottom) in PLB^−/− myocytes. P > 0.05. (C) The anti-adrenergic effect of SIN-1 is greater in WT compared with PLB^−/− myocytes. Pooled data (mean ± SEM) expressed as % Δ from ISO demonstrating the effect of SIN-1 on β-adrenergic-stimulated shortening (top) and Ca²⁺ transients (bottom) in WT (clear bar, data derived from Figure 1C) and PLB^−/− (filled bar) myocytes. *P < 0.01 vs. WT.

3.5. Effect of SIN-1 on PLB phosphorylation

After identifying PLB as a target for peroxynitrite signalling, we further investigated its role by examining the effect of SIN-1 on PKA-dependent PLB_Ser16 phosphorylation in WT hearts (n = 5 hearts/group). As expected, 0.1 μmol/L ISO caused a large increase in PLB_Ser16 phosphorylation [Figure 3; 266 ± 29 AU (arbitrary units)] compared with control (66 ± 13 AU). However, upon perfusion with 0.1 μmol/L ISO + 200 μmol/L SIN-1, a significant decrease in PLB_Ser16 phosphorylation was observed (163 ± 18 AU). This reduction in PLB_Ser16 phosphorylation is likely responsible for the anti-adrenergic effects of SIN-1, as dephosphorylated PLB will likely re-associate with SERCA, thus reducing the SR Ca²⁺ load and myocyte contractility. No differences were observed in PLB_total between groups or in phosphorylation at the CaMKII-dependent Thr17 site with 0.1 μmol/L ISO or 0.1 μmol/L ISO + 200 μmol/L SIN-1 (data not shown). This result with PLB_Thr17 phosphorylation should be expected as CaMKII would not have sufficient time in which to be activated and phosphorylate this site during the time course of our experiments.²⁷

Figure 3.

SIN-1 decreases β-adrenergic-stimulated PLB_Ser16 phosphorylation. (A) Representative western blot specific for PLB_Ser16 phosphorylation (10, 20 μg protein loading) and pentameric PLB_total (2, 4 μg protein loading). (B) Pooled data (mean ± SEM) for PLB_Ser16 phosphorylation (P-PLB_Ser16) with control (NT, clear bar), ISO (0.1 μmol/L, black bar), or ISO+SIN-1 (200 μmol/L, grey bar). Data displayed as arbitrary units (AU). *P < 0.0001 vs. Cont, **P < 0.01 vs. ISO.

3.6. Effect of PP1 and PP2a inhibition on myocyte function

We studied the SIN-1-induced decrease in PLB_Ser16 phosphorylation further by examining the functional effects of protein phosphatase inhibition in WT myocytes. We repeated the same functional experiments described earlier. However, this time WT myocytes were pre-incubated with okadaic acid, an inhibitor of PP1 and PP2a activity. Upon inhibition with 5 μmol/L okadaic acid (n = 13/3 hearts), we observed a significant increase in basal contractility (Ca²⁺ transient, 1.6 ± 0.2 ΔF/F₀; shortening, 7.8 ± 1.4 μm) compared with WT myocytes not pre-incubated with okadaic acid (Ca²⁺ transient, 0.8 ± 0.2 ΔF/F₀; shortening, 3.8 ± 0.8 μm, P < 0.05 vs. WT + okadaic acid). The increased basal contractility resulted in a reduction in the β-adrenergic reserve compared with normal WT myocytes (Ca²⁺ transient, 1.8 ± 0.2 ΔF/F₀; shortening: 14.2 ± 2.2 μm). Pre-incubation with 5 μmol/L okadaic acid, however, did alleviate the anti-adrenergic effects of SIN-1 (Ca²⁺ transient, 1.7 ± 0.3 ΔF/F₀; shortening, 14.0 ± 2.6 μm). We reduced the concentration of okadaic acid to 1 μmol/L (n = 18/3 hearts) in order to decrease its effect on basal function (Ca²⁺ transient, 0.6 ± 0.1 ΔF/F₀; shortening: 4.2 ± 1.0 mm; Figure 4A), and preserve the response to ISO (Ca²⁺ transient: 1.4 ± 0.2 ΔF/F₀; shortening, 15.4 ± 2.8 μm) compared with WT myocytes not pre-incubated with okadaic acid. Upon perfusion with 1 μmol/L ISO+200 μmol/L SIN-1, we observed only minimal changes in Ca²⁺ transient amplitude, and cell shortening in myocytes pre-incubated with 1 μmol/L okadaic acid (Ca²⁺ transient, 1.3 ± 0.2 ΔF/F₀; shortening, 15.2 ± 3.0 μm, P < 0.05 vs. WT okadaic acid). The preventative effect of okadaic acid pre-incubation is shown in Figure 4B, which shows the % Δ from ISO. These data indicate that SIN-1 exerts anti-adrenergic effects via activation of protein phosphatases.

Figure 4.

Okadaic acid prevents the effects of SIN-1 in WT myocytes. (A) Individual, steady-state shortening (top) and Ca²⁺ transient (bottom) traces representing the lack of an anti-adrenergic effect of SIN-1 in a WT myocyte pre-incubated with okadaic acid (1 μmol/L). (B) Pooled data (mean ± SEM) expressed as % Δ from ISO demonstrating the effect of SIN-1 on β-adrenergic-stimulated shortening (top) and Ca²⁺ transients (bottom) in WT (clear bar, data from Figure 1C) and WT + okadaic acid (filled bar) myocytes. *P < 0.05 vs. WT.

3.7. Effect of PP1 and PP2a inhibition on PLB serine 16 phosphorylation

Upon observing that protein phosphatase inhibition alleviated the functional effects of SIN-1 in WT myocytes, we sought to examine the effect of PP1 and PP2a inhibition on PLB_Ser16 phosphorylation. We repeated the same biochemical experiments described earlier. This time, however, hearts were perfused with 1 μmol/L okadaic acid in order to inhibit protein phosphatase activity, prior to perfusion with 0.1 μmol/L ISO or 0.1 μmol/L ISO + 200 μmol/L SIN-1 (n = 4 hearts/group). As expected, pre-treatment with okadaic acid alone led to a slight increase in basal PLB_Ser16 phosphorylation. Upon perfusion with 0.1 μmol/L ISO + 200 μmol/L SIN-1, however, the decrease in PLB_Ser16 phosphorylation was alleviated compared with ISO alone (139 ± 4 vs. 134 ± 7 AU), as seen in % Δ from ISO shown in Figure 5B. No differences were observed in PLB_total. These data further implicate protein phosphatases in the anti-adrenergic effects of the peroxynitrite donor, SIN-1.

Figure 5.

Okadaic acid (OA) prevents the SIN-1-induced decrease in PLB_Ser16 phosphorylation. (A) Representative western blot (20 μg protein loading) specific for PLB_Ser16 phosphorylation and pentameric PLB_total (4 μg protein loading). (B) Pooled data (mean ± SEM) expressed as % Δ from ISO demonstrating the effect of SIN-1 on β-adrenergic-stimulated PLB_Ser16 phosphorylation (P-PLB_Ser16) with (filled bar) and without (clear bar, data derived from Figure 4) protein phosphatase inhibition. *P < 0.05 vs. WT.

4. Discussion

Few studies have examined the mechanism(s) underlying the effects of peroxynitrite on β-adrenergic responsiveness in the mammalian myocardium. There are many studies demonstrating both positive and negative effects of peroxynitrite on myocardial contractility,¹³−²⁰ but the majority of these studies addressed effects on basal contractility and not on β-adrenergic responsiveness. Our study, however, demonstrates that peroxynitrite (18 nmol L⁻¹ min⁻¹), produced via SIN-1, serves to reduce β-adrenergic responsiveness in murine cardiomyocytes. Upon perfusion with SIN-1, a subsequent decrease in β-adrenergic-stimulated Ca²⁺ transient amplitude and myocyte shortening was observed in WT myocytes. The peroxynitrite decomposition catalyst, FeTPPS, reversed the anti-adrenergic effects of SIN-1. We examined the excitation–contraction coupling protein PLB as a potential target in the peroxynitrite pathway and observed that SIN-1 was without effect in PLB^−/− knockout myocytes. A further examination of PLB detected decreased β-adrenergic-stimulated PLB_Ser16 phosphorylation upon perfusion with SIN-1. Therefore, it is likely that PLB is an end target of peroxynitrite signalling and serves to alter Ca²⁺ handling in the cardiomyocyte. Additional experiments using okadaic acid, a protein phosphatase inhibitor, suggest that peroxynitrite activates protein phosphatases that ultimately lead to the anti-adrenergic effects of SIN-1. These findings serve to clarify the effects of peroxynitrite production on cardiac contractility and provide a potential mechanism for the observed β-adrenergic hyporesponsiveness.

4.1. SIN-1 reduces β-adrenergic stimulation in WT myocytes

Previous studies have demonstrated the anti-adrenergic effects of the peroxynitrite donor, SIN-1, in isolated cardiomyocytes.^16,17 We confirmed these results in murine cardiomyocytes, in that SIN-1 induced a decrease in Ca²⁺ transient amplitude and myocyte shortening during the maximal β-adrenergic stimulation (Figure 1). Additionally, SIN-1 had no effect on basal contractility, which indicates that 200 μmol/L SIN-1 exclusively modulates the β-adrenergic responsiveness.

SIN-1 is considered to be a peroxynitrite donor, as it breaks down to form nitric oxide (NO) and superoxide $\left({\mathrm{O}}_2^{-}\right){.}^{28-30}$ Nitric oxide and superoxide will subsequently couple to form peroxynitrite (ONOO⁻). However, SIN-1 has a complex chemistry and other reactive species may be formed.³¹ In our studies, peroxynitrite was confirmed as the causal species, as the peroxynitrite decomposition catalyst, FeTPPS, alleviated the anti-adrenergic effects of SIN-1 (Figure 1).

Many reports in the literature also suggest that it is possible for catecholamines to be oxidized into aminochromes by peroxynitrite.^15,24,25 ISO washout experiments, however, demonstrated that this was not the case. The washout of ISO, which simulated ISO oxidation, resulted in only a slight decrease in myocyte contractility, whereas SIN-1 produced a strong anti-adrenergic effect (Figure 1). Additionally, experiments with forskolin, an activator of adenylate cyclase, showed that the primary effect of SIN-1 was not mediated via targets upstream of adenylate cyclase, as SIN-1 still produced anti-adrenergic effects. Thus, in our experimental setting, it is highly unlikely that the main effect of SIN-1 is exerted via oxidation of ISO, but by a direct effect on myocyte Ca²⁺ handling. Further, it is unlikely that the β-adrenergic receptor itself or any β-adrenergic-activated G-coupled proteins are targeted by SIN-1, as the anti-adrenergic effects of SIN-1 were still present upon the direct activation of adenylate cyclase with forskolin. We therefore decided to investigate the role of PLB in this peroxynitrite-induced β-adrenergic hyporesponsiveness.

4.2. SIN-1 and PLB^−/− myocyte function

Although we observed the anti-adrenergic effects of SIN-1 in WT myocytes, we saw no effect of SIN-1 on PLB^−/− myocytes (Figure 2). This would indicate that SIN-1, and thus peroxynitrite, exerts anti-adrenergic effects by targeting the excitation–contraction coupling protein PLB.

4.3. SIN-1 reduces PLB serine 16 phosphorylation

We investigated the role of PLB further by examining its PKA-dependent Ser16 phosphorylation site as a potential mechanism for the functional effects of SIN-1. Studies have shown that there is a reduction in PLB_Ser16 phosphorylation in heart failure,³²−³⁴ potentially resulting in reduced Ca²⁺ sensitivity of the SERCA pump and abnormal Ca²⁺ handling.³⁵ Examination of this phosphorylation site revealed an increase in PLB_Ser16 phosphorylation with ISO (Figure 3). This is to be expected as ISO activates the β-adrenergic pathway, resulting in PKA activation. However, SIN-1 induced a significant reduction in β-adrenergic-stimulated PLB_Ser16 phosphorylation. These data provide a potential mechanism for the functional effects of SIN-1, as decreased PLB_Ser16 phosphorylation would increase the interaction of PLB with SERCA, thus reducing SR Ca²⁺ load and myocyte contractility. This reduction in PLB_Ser16 phosphorylation may be a key component of the contractile dysfunction observed in heart failure, and would serve to reduce the affinity of the SERCA pump for Ca²⁺.

4.4. Protein phosphatase inhibition alleviates anti-adrenergic effects of SIN-1

Our current data show that peroxynitrite exposure leads to a reduction in PLB_Ser16 phosphorylation. Additionally, one study has demonstrated an interaction between peroxynitrite and protein phosphatase activity in erythrocytes,³⁶ but no studies have investigated a direct link in cardiomyocytes. We therefore decided to investigate alterations in protein phosphatases as a potential mechanism for the functional effects of SIN-1, as Neumann et al.³⁷ demonstrated an increased PP1 activity in preparations from failing human hearts vs. non-failing hearts using phosphorylated PLB as the substrate. Inhibition of protein phosphatase activity with okadaic acid (1 μmol/L) resulted in no significant changes in basal contractility or in response to ISO compared with normal WT myocytes. Protein phosphatase inhibition, however, alleviated the anti-adrenergic effect of SIN-1 in myocytes pre-incubated with okadaic acid (Figure 4), providing a potential mechanism for the reduction in PLB_Ser16 phosphorylation.

4.5. Protein phosphatase inhibition prevents the SIN-1-induced decrease in PLB serine 16 phosphorylation

The inhibition of protein phosphatase activity not only prevented the functional effect of SIN-1 in WT myocytes, but also alleviated the SIN-1-induced decrease in PLB_Ser16 phosphorylation. Pre-incubation with okadaic acid (1 μmol/L) resulted in a slight increase in PLB_Ser16 phosphorylation. However, upon perfusion with SIN-1, no decrease in ISO-stimulated phosphorylation was observed in hearts perfused with okadaic acid (Figure 5).

4.6. Physiological relevance of SIN-1 concentration

The concentration of SIN-1 (200 μmol/L) used in this study was determined to release peroxynitrite at a rate of 18 nmol L⁻¹ min⁻¹ under experimental conditions. In terms of physiological relevance, myocardial peroxynitrite injury is often associated with inducible nitric oxide synthase (iNOS, NOS2) expression.³⁸ We have previously shown that acute inhibition of NOS2 in failing human myocytes increased the Ca²⁺ transient and myocyte shortening amplitude during β-adrenergic stimulation.³⁹ This same functional phenomenon was observed in our current study (i.e. peroxynitrite decreased the β-adrenergic response). Also, NOS2 inhibition had no effect on basal function in failing human myocytes. Once again the same phenomenon was observed in our current study. Thus, we believe that the concentration of SIN-1 is relevant under pathophysiological conditions (i.e. heart failure) and may explain the mechanism responsible for the reversible, NOS2-induced β-adrenergic hyporesponsiveness. Additionally, moderately high concentrations of peroxynitrite (>25 μmol/L) have been shown to have effects on basal contractility,¹⁵ whereas extremely high concentrations of peroxynitrite (>200 μmol/L) have been shown to induce a state of rigour,⁴⁰ and neither of these effects were observed in our study.

In conclusion, the peroxynitrite donor, SIN-1, reduces β-adrenergic stimulation by ultimately targeting PLB. SIN-1 exerts anti-adrenergic effects by reducing PKA-dependent PLB_Ser16 phosphorylation via activation of protein phosphatases. A functional effect on β-adrenergic stimulation is yielded through a disruption in Ca²⁺ handling. As SIN-1 has also been shown to reduce cAMP levels in cardiomyocytes,¹⁶ future studies will address the effects of SIN-1 on adenylate cyclase activity, cAMP levels, and PKA activity. Previous studies have also shown that peroxynitrite can directly inactivate SERCA.^41,42 Additionally, peroxynitrite has been demonstrated to affect other excitation–contraction coupling proteins, including troponin I, RyR, and the Na⁺/Ca²⁺ exchanger.^40,43,44 The effects of peroxynitrite, however, may not have been observed in PLB^−/− myocytes due to their hyperdynamic contractile state, and because of concentration and time-dependent effects. Therefore, a direct effect of peroxynitrite on any of the aforementioned excitation–contraction coupling proteins cannot be completely ruled out.

In many cardiomyopathies, including heart failure, nitric oxide production is increased because of the expression of NOS2.^39,45 Additionally, superoxide production is increased via NADPH and/or xanthine oxidoreductase.^46,47 Elevated nitric oxide and superoxide production could lead to the formation of high levels of peroxynitrite. Furthermore, the expression of NOS2 by itself may lead to peroxynitrite formation and myocardial injury.^38,39,48 Thus, our current observation provides a plausible mechanism for the diminished β-adrenergic responsiveness observed in heart failure, as peroxynitrite formation and protein phosphatase activity have been shown to be increased,¹⁰−^12,37,49 while PLB_Ser16 phosphorylation was shown to be decreased.³²−³⁴ Therefore, this peroxynitrite signalling cascade could be a key pathway in the decreased PLB phosphorylation and resulting dysfunction observed in heart failure and other cardiomyopathies.