Empagliflozin inhibits increased Na influx in atrial cardiomyocytes of patients with HFpEF

Abstract

Aims: Heart failure with preserved ejection fraction (HFpEF) causes substantial morbidity and mortality. Importantly, atrial remodelling and atrial fibrillation are frequently observed in HFpEF. Sodium-glucose cotransporter 2 inhibitors (SGLT2i) have recently been shown to improve clinical outcomes in HFpEF, and post-hoc analyses suggest atrial anti-arrhythmic effects. We tested if isolated human atrial cardiomyocytes from patients with HFpEF exhibit an increased Na influx, which is known to cause atrial arrhythmias, and if that is responsive to treatment with the SGTL2i empagliflozin.

Methods and results: Cardiomyocytes were isolated from atrial biopsies of 124 patients (82 with HFpEF) undergoing elective cardiac surgery. Na influx was measured with the Na-dye Asante Natrium Green-2 AM (ANG-2). Compared to patients without heart failure (NF), Na influx was doubled in HFpEF patients (NF vs. HFpEF: 0.21 ± 0.02 vs. 0.38 ± 0.04 mmol/L/min (N = 7 vs. 18); P = 0.0078). Moreover, late INa (measured via whole-cell patch clamp) was significantly increased in HFpEF compared to NF. Western blot and HDAC4 pulldown assay indicated a significant increase in CaMKII expression, CaMKII autophosphorylation, CaMKII activity, and CaMKII-dependent NaV1.5 phosphorylation in HFpEF compared to NF, whereas NaV1.5 protein and mRNA abundance remained unchanged. Consistently, increased Na influx was significantly reduced by treatment not only with the CaMKII inhibitor autocamtide-2-related inhibitory peptide (AIP), late INa inhibitor tetrodotoxin (TTX) but also with sodium/hydrogen exchanger 1 (NHE1) inhibitor cariporide. Importantly, empagliflozin abolished both increased Na influx and late INa in HFpEF. Multivariate linear regression analysis, adjusting for important clinical confounders, revealed HFpEF to be an independent predictor for changes in Na handling in atrial cardiomyocytes.

Conclusion: We show for the first time increased Na influx in human atrial cardiomyocytes from HFpEF patients, partly due to increased late INa and enhanced NHE1-mediated Na influx. Empagliflozin inhibits Na influx and late INa, which could contribute to anti-arrhythmic effects in patients with HFpEF.

Keywords: Atrial remodelling; Empagliflozin; HFpEF; Heart failure; SGLT2i; Sodium.

Graphical abstract

Figure 1

Increased Na influx into isolated atrial cardiomyocytes of HFpEF patients. Original registrations of Na influx in isolated human atrial cardiomyocytes following inhibition of Na/K-ATPase by superfusion with K-free solution (A). Na influx was calculated as mean slope of the increase in [Na]_i over time and derived values are illustrated as scatter plot (B). Na influx is almost doubled in isolated cardiomyocytes of HFpEF patients compared to NF patients (P = 0.0078, mixed-effects analysis with Holm-Sidak’s post-hoc correction).

Figure 2

Increased Na_V1.5 phosphorylation in atrial cardiomyocytes of HFpEF patients is associated with increased late I_Na. Na_V1.5 mRNA expression (gene: SCN5A [relative to β-actin as housekeeper]) was similar in NF and HFpEF cardiomyocytes (P = 0.9777, t-test) (A). Mean densitometric data (top) and representative original Western blot (bottom) of Na_V1.5 (B) and pNa_V1.5 (C). Compared to control patients (NF), there was a significant increase in the phosphorylation of Na_V1.5 at S571 in atrial cardiomyocytes from patients with HFpEF (P = 0.0197, t-test), while Na_V1.5 protein levels were unaltered (P = 0.4316, t-test). Original registrations of late I_Na (D) and mean values of late I_Na integrals (E) in atrial cardiomyocytes of HFpEF and NF patients measured by whole-cell patch clamp (voltage protocol is shown in the inset). In line with increased NaV1.5 phosphorylation late I_Na was significantly increased in atrial cardiomyocytes of HFpEF patients compared to NF patients (N = 3 vs. 5, P = 0.0172, t-test). N, number of patients.

Figure 3

Increased CaMKII expression, autophosphorylation, and activity in atrial cardiomyocytes of HFpEF patients. Mean densitometric data (top) and representative original Western blot (bottom) of CaMKII expression (A) and phosphorylation (B) showing a significant increase in CaMKII expression (P = 0.0067, t-test) as well as CaMKII autophosphorylation at T287 (P = 0.0006, t-test) in atrial cardiomyocytes of HFpEF patients compared to NF patients. Accordingly, CaMKII activity assessed by HDAC4 pulldown assay was also enhanced in HFpEF patients (P = 0.0391, t-test) (C). N, number of patients.

Figure 4

Increased Na influx in HFpEF patients is reduced by treatment with empagliflozin, TTX, AIP, and CP. Original registrations (A) and mean values per patient (B) of Na influx in isolated human atrial cardiomyocytes following inhibition of Na/K-ATPase by superfusion with K-free solution. Treatment of cardiomyocytes with empagliflozin resulted in a significant reduction of Na influx into HFpEF cardiomyocytes (P < 0.0001, mixed-effects analysis with Holm-Sidak’s post-hoc correction), similar to the treatment with the late I_Na inhibitor TTX (P = 0.0077, mixed-effects analysis with Holm-Sidak’s post-hoc correction), the CaMKII inhibitor AIP (P = 0.0356, mixed-effects analysis with Holm-Sidak’s post-hoc correction) and the NHE1 inhibitor cariporide (CP [P = 0.0078, mixed-effects analysis with Holm-Sidak’s post-hoc correction]). Numbers at the bottom of the bars represent the respective number of patients analysed.

Figure 5

Increased late I_Na in atrial cardiomyocytes of HFpEF patients is normalized by treatment with empagliflozin as well as AIP. (A) Original registrations of late I_Na and (B) mean values of late I_Na integrals in atrial cardiomyocytes of HFpEF patients treated with vehicle solution, empagliflozin, or the CaMKII inhibitor AIP measured by whole-cell patch clamp (voltage protocol is shown in the inset). Empagliflozin as well as the CaMKII inhibitor AIP significantly reduced late I_Na (P = 0.0187 vs. HFpEF + vehicle, repeated measures one-way ANOVA with Holm-Sidak’s post-hoc correction).