Abnormal Calcium Handling in Atrial Fibrillation Is Linked to Changes in Cyclic AMP Dependent Signaling

Abstract

Both, the decreased L-type Ca²⁺ current (I_Ca,L) density and increased spontaneous Ca²⁺ release from the sarcoplasmic reticulum (SR), have been associated with atrial fibrillation (AF). In this study, we tested the hypothesis that remodeling of 3',5'-cyclic adenosine monophosphate (cAMP)-dependent protein kinase A (PKA) signaling is linked to these compartment-specific changes (up- or down-regulation) in Ca²⁺-handling. Perforated patch-clamp experiments were performed in atrial myocytes from 53 patients with AF and 104 patients in sinus rhythm (Ctl). A significantly higher frequency of transient inward currents (I_TI) activated by spontaneous Ca²⁺ release was confirmed in myocytes from AF patients. Next, inhibition of PKA by H-89 promoted a stronger effect on the I_TI frequency in these myocytes compared to myocytes from Ctl patients (7.6-fold vs. 2.5-fold reduction), while the β-agonist isoproterenol (ISO) caused a greater increase in Ctl patients (5.5-fold vs. 2.1-fold). I_Ca,L density was larger in myocytes from Ctl patients at baseline (p < 0.05). However, the effect of ISO on I_Ca,L density was only slightly stronger in AF than in Ctl myocytes (3.6-fold vs. 2.7-fold). Interestingly, a significant reduction of I_Ca,L and Ca²⁺ sparks was observed upon Ca²⁺/Calmodulin-dependent protein kinase II inhibition by KN-93, but this inhibition had no effect on I_TI. Fluorescence resonance energy transfer (FRET) experiments showed that although AF promoted cytosolic desensitization to β-adrenergic stimulation, ISO increased cAMP to similar levels in both groups of patients in the L-type Ca²⁺ channel and ryanodine receptor compartments. Basal cAMP signaling also showed compartment-specific regulation by phosphodiesterases in atrial myocytes from 44 Ctl and 43 AF patients. Our results suggest that AF is associated with opposite changes in compartmentalized PKA/cAMP-dependent regulation of I_Ca,L (down-regulation) and I_TI (up-regulation).

Keywords: L-type calcium current (ICa,L); atrial fibrillation (AF); cAMP-dependent regulation; patch-clamp; protein kinase A (PKA); transient inward current (ITI).

Figure 1

Effect of protein kinase A (PKA) inhibition on I_Ca,L, I_TI and sarcoplasmic reticulum (SR) Ca²⁺ load. (a) (left): Representative I_Ca,L recordings in myocytes from a patient in sinus rhythm (Ctl) and a patient with atrial fibrillation (AF) before (CON, black trace) and after exposure to the selective PKA inhibitor (H-89, 10 μM, grey trace). (right): Average effect of H-89 in Ctl and AF patients. (b) Mean effect of H-89 on the current-voltage relationship in Ctl and AF patients. (c) (left): Representative traces of spontaneous I_TI recorded in Ctl and AF myocytes before and after exposure to H-89. (right): Mean effects of H-89 on the spontaneous I_TI frequency in Ctl and AF patients. (d) Average effects of H-89 on Ca²⁺ sparks density (left) and frequency (right) before and after exposure to H-89. ROI: region of interest. Significant differences between treatments are indicated with * and between groups with #.

Figure 2

Effects of Ca²⁺/Calmodulin-dependent protein kinase II (CaMKII) inhibition on I_Ca,L, I_TI and Ca²⁺ sparks. (a) (left): Representative I_Ca,L recordings in myocytes from a patient in sinus rhythm (Ctl) and a patient with atrial fibrillation (AF) before (CON, black trace) and after exposure to a selective CaMKII inhibitor (KN-93, 1 μM, grey trace). (right): Average effects of KN-92 (1 μM, the inactive CaMKII inhibitor) and KN-93 in Ctl and AF patients. (b) (left): Representative recordings of spontaneous I_TI before and after exposure to KN-92 and KN-93 in myocytes from in Ctl and AF patients. (right): Average effects of KN-92 and KN-93 on the spontaneous I_TI frequency in AF and Ctl myocytes. (c) Average effects of KN-92 and KN-93 on Ca²⁺ sparks density (left) and frequency (right) before and after exposure to KN-92 and KN-93. Significant differences between treatments are indicated with * and between groups with +.

Figure 3

Effect of β-adrenergic stimulation on I_Ca,L, I_TI and sarcoplasmic reticulum (SR) Ca²⁺ load. (a) (left): Representative I_Ca,L recordings in myocytes from a sinus rhythm (Ctl) patient and a patient with atrial fibrillation (AF), before (CON, black trace) and after exposure of the myocyte to 30 nM isoproterenol (ISO, grey trace). (right): Average effects of ISO in Ctl and AF patients. (b) Mean current-voltage relationship for I_Ca,L in Ctl and AF myocytes before and after exposure to ISO. (c) (left): Representative recordings of I_TI before and after ISO in Ctl and AF. (right): Average effects of ISO on I_TI in Ctl and AF myocytes. (d) (right): Representative recordings of caffeine-induced NCX-currents before and after ISO. (left): Average ISO effect on SR Ca²⁺ load in Ctl and AF. (e) Average ISO effect on time constant (Tau) in Ctl and AF. Significant differences between treatments are indicated with *, between groups with #.

Figure 4

Cytosolic and local cAMP dynamics. (left): representative kinetics of fluorescence resonance energy transfer (FRET) changes (expressed as CFP/YFP ratio) in the ryanodine receptor (RyR2) microdomain recorded in human atrial myocytes from patients in sinus rhythm (Ctl) and with atrial fibrillation (AF), transduced with Epac1-camps-JNC sensor and treated with (a) isoproterenol (ISO, 100 nM) and (b) the non-selective phosphodiesterase inhibitor (IBMX, 100 μM) and the phosphodiesterase type 8 selective inhibitor PF-04957325 (100 nM). (right): average increases in cAMP levels in the cytosol and in different compartments (sarcolemma and RyR2) of living human atrial myocytes, measured as total cAMP produced after exposure (a) to ISO and (b) to the inhibitor IBMX. Significant differences between treatments are indicated with *.

Figure 5

Effect of β-blocker treatment of patients on Ca²⁺-handling. (left): average effects of the treatment of patients with β-blockers (BBs) or without BBs (no BBs) on (a) I_Ca,L, (b) sarcoplasmic reticulum (SR) Ca²⁺ load and (c) I_TI, in patients in sinus rhythm (Ctl) and with atrial fibrillation (AF). (right): Carvediol effects on (d) I_Ca,L, (e) SR Ca²⁺ load and (f) I_TI, in Ctl and AF. Statistical significance was evaluated using a multivariate linear regression model adjusted for the confounding effects of common clinical factors showing a bias between BBs treatment as well as factors suspected to affect Ca²⁺-handling. Significant differences between treatments are indicated with * and between groups with +. Each point represents a patient mean value recorded in myocytes from a total of 53 patients with AF and 104 patients without AF. 23 of the AF patients and 37 of the no AF patients were treated with BBs.