Influence of sex on intracellular calcium homoeostasis in patients with atrial fibrillation

Abstract

Aims: Atrial fibrillation (AF) has been associated with intracellular calcium disturbances in human atrial myocytes, but little is known about the potential influence of sex and we here aimed to address this issue.

Methods and results: Alterations in calcium regulatory mechanisms were assessed in human atrial myocytes from patients without AF or with long-standing persistent or permanent AF. Patch-clamp measurements revealed that L-type calcium current (ICa) density was significantly smaller in males with than without AF (-1.15 ± 0.37 vs. -2.06 ± 0.29 pA/pF) but not in females with AF (-1.88 ± 0.40 vs. -2.21 ± 0.0.30 pA/pF). In contrast, transient inward currents (ITi) were more frequent in females with than without AF (1.92 ± 0.36 vs. 1.10 ± 0.19 events/min) but not in males with AF. Moreover, confocal calcium imaging showed that females with AF had more calcium spark sites than those without AF (9.8 ± 1.8 vs. 2.2 ± 1.9 sites/µm2) and sparks were wider (3.0 ± 0.3 vs. 2.2 ± 0.3 µm) and lasted longer (79 ± 6 vs. 55 ± 8 ms), favouring their fusion into calcium waves that triggers ITIs and afterdepolarizations. This was linked to higher ryanodine receptor phosphorylation at s2808 in women with AF, and inhibition of adenosine A2A or beta-adrenergic receptors that modulate s2808 phosphorylation was able to reduce the higher incidence of ITI in women with AF.

Conclusion: Perturbations of the calcium homoeostasis in AF is sex-dependent, concurring with increased spontaneous SR calcium release-induced electrical activity in women but not in men, and with diminished ICa density in men only.

Keywords: Afterdepolarizations; Calcium sparks; Ryanodine receptor phosphorylation; Sarcoplasmic reticulum calcium release; Transient inward current.

Graphical abstract

Figure 1

Effect of sex on the L-type calcium current amplitude and properties. (A) Representative I_Ca recordings from four patient groups of male (left traces) and female patients (right traces), without AF (blue traces) and with AF (red traces). (B) Mean I_Ca densities in the four groups. (C) Current–voltage curves for male and female patients with no AF or with AF. (D) Superimposed I_Ca-tracings are shown on the left and time constants for fast I_Ca inactivation are shown on the right. Values were analysed and corrected for the clinical factors marked as confounders in Table 1 using a linear regression model. P-values are given for significant differences between bars. Number of patients is given for each bar.

Figure 2

Effect of sex on the I_TI frequency and spontaneous membrane depolarizations. (A) Representative recordings of I_TI currents from the same patients as in Figure 1 divided into male (left traces) and female patients (right traces). (B) Mean I_TI frequencies. Values in (A)–(B) were analysed and corrected for the clinical factors marked as confounders in Table 1 using a linear regression model. (C) Representative recordings of I_TI amplitudes in a male and a female patient with AF. Mean amplitudes are shown on the right. (D) Spontaneous membrane depolarizations recorded in male and female patients with AF. The mean frequency is shown on the right. Statistical significance was determined in (C)–(D) using an unadjusted regression model. P-values are given for significant differences between bars. Number of patients is given for each bar.

Figure 3

Effects of sex on calcium spark frequency and properties. (A) Images of human atrial myocytes from patients without (no AF) and with AF. Calcium spark sites are indicated with circles and calcium signals for each site are shown below. (B) Density of spark sites. (C) Sparks per site. (D) Spark duration at half maximum. (E) Spark width at half maximum (F) Spark amplitude. Statistical significance was determined using an unadjusted linear regression model. P-values are given for significant differences between bars. Number of patients is given for each bar.

Figure 4

Effect of sex on SR calcium load and uptake. (A) Representative caffeine-induced transient inward NCX currents (top) and their time-integral (bottom) from a male and a female patient without AF (left) and with AF (right). (B) Caffeine releasable SR calcium load estimated from the time-integral of the caffeine-induced current. Values are corrected for the clinical factors marked as confounders in Table 1 using a linear regression model. (C) SERCA2a protein expression. (D) NCX-1 protein expression. Densitometry quantification of protein levels is shown in the upper panels and representative western blots in the lower panels. Protein levels were normalized to α-actinin. P-values are given for differences between bars. Statistical significance was determined using an unadjusted linear regression model. Number of patients is given for each bar.

Figure 5

Effect of sex on the expression and distribution of Csq-2. (A) Representative western blots of Csq-2 and α-actinin. (B) Densitometry quantification of Csq-2expression normalized to α-actinin. The number of patients is indicated for each bar. P-values are given above bars. (C) Overlay of fluorescently labelled RyR2 (in green) and Csq-2 (in red). (D) Csq-2/RyR2 intensity ratios measured at different distances from the sarcolemma (given below bars in µm). Atrial myocytes from No AF patients had higher ratios than those from AF for females (P =0.02) but not for males (P =0.06). Statistical significance was determined using an unadjusted linear regression model. Statistical differences between pairs of AF and noAF are indicated with ***P<0.001, **P<0.01, *P=0.05. Number of experiments is given in parentheses.

Figure 6

Effect of sex on the density and phosphorylation of the RyR2. (A) Immunoflourescent labelling of the RyR2. (B) Mean density of RyR2 clusters. (C) Overlay of total RyR2 (in green) and s2808 phosphorylated RyR2 (in red). (D) Mean s2808 phosphorylated RyR2 measured as the fluorescence intensity ratio (s2808/RyR2) for all RyR2 clusters at different distances from the sarcolemma (given below bars in µm). Statistical significance was determined using an unadjusted linear regression model. Significant differences between pairs of bars are indicated with ***P<0.001, **P<0.01. Number of experiments is given in parentheses.

Figure 7

A_2AR inhibition or treatment with beta-blockers reduces the I_TI frequency in females with AF. (A) Effect of A_2AR activation with CGS21680 on the I_TI frequency. (B) Prevention of A_2AR activation by treating myocytes with ADA reduces the I_TI frequency in females with AF to levels observed in patients without AF. (C) Treatment of patients with beta-blockers (B-Block) reduces the I_TI frequency in female patients with AF to levels in patients without AF. Data were analysed using an unadjusted regression model. P-values for paired comparison of pharmacological treatment and control are given above bars. Number of patients is given for each bar.