Empagliflozin inhibits Na+ /H+ exchanger activity in human atrial cardiomyocytes

Abstract

Aims: Recent clinical trials have proven gliflozins to be cardioprotective in diabetic and non-diabetic patients. However, the underlying mechanisms are incompletely understood. A potential inhibition of cardiac Na⁺ /H⁺ exchanger 1 (NHE1) has been suggested in animal models. We investigated the effect of empagliflozin on NHE1 activity in human atrial cardiomyocytes.

Methods and results: Expression of NHE1 was assessed in human atrial and ventricular tissue via western blotting. NHE activity was measured as the maximal slope of pH recovery after NH₄ ⁺ pulse in isolated carboxy-seminaphtarhodafluor 1 (SNARF1)-acetoxymethylester-loaded murine ventricular and human atrial cardiomyocytes. NHE1 is abundantly expressed in human atrial and ventricular tissue. Interestingly, compared with patients without heart failure (HF), atrial NHE1 expression was significantly increased in patients with HF with preserved ejection fraction and atrial fibrillation. The largest increase in atrial and ventricular NHE1 expression, however, was observed in patients with end-stage HF undergoing heart transplantation. Importantly, acute exposure to empagliflozin (1 μmol/L, 10 min) significantly inhibited NHE activity to a similar extent in human atrial myocytes and mouse ventricular myocytes. This inhibition was also achieved by incubation with the well-described selective NHE inhibitor cariporide (10 μmol/L, 10 min).

Conclusions: This is the first study systematically analysing NHE1 expression in human atrial and ventricular myocardium of HF patients. We show that empagliflozin inhibits NHE in human cardiomyocytes. The extent of NHE inhibition was comparable with cariporide and may potentially contribute to the improved outcome of patients in clinical trials.

Keywords: Empagliflozin; Heart failure; Na+/H+ exchanger 1; SGLT2 inhibitor.

Figure 1

NHE expression in human atrial and ventricular cardiomyocytes. (A) Original western blot for the assessment of NHE1 expression (~95 kD) in human atrial appendages from patients without heart failure (NF SR), with heart failure with preserved ejection fraction in sinus rhythm (HFpEF SR), HFpEF and atrial fibrillation (HFpEF AFib) as well as from explanted end‐stage failing hearts (HTX). (B) Analysis of mean densitometric values normalized to glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) showed a significant reduction of NHE1 expression in atrial appendages from patients with reduced left ventricular ejection fraction (ejection fraction < 40%, N = 4) compared with all other groups, while in patients with HFpEF and atrial fibrillation as well as in patients with end‐stage heart failure, atrial expression of NHE1 is significantly up‐regulated. ^* P < 0.05 vs. NF SR, ^# P < 0.05 vs. HFpEF SR, ^Ϯ P < 0.05 vs. HFpEF AFib. Brown–Forsythe and Welch ANOVA test with two‐stage step‐up procedure of Benjamini, Krieger, and Yekutieli. (C) Original western blot for the assessment of NHE1 expression (~95 kD) in human ventricular myocardium from patients with left ventricular hypertrophy and normal systolic left ventricular function (LVH) and from explanted end‐stage failing hearts (HTX). (D) In ventricular myocardium of end‐stage failing hearts, NHE1 expression is significantly up‐regulated compared with ventricular myocardium from patients with left ventricular hypertrophy (N = 5 vs. 5). Two‐tailed Student's t‐test.

Figure 2

Calibration experiments for the assessment of Na⁺/H⁺ exchanger activity in isolated ventricular cardiomyocytes. (A) Original calibration experiment measuring SNARF1 fluorescence intensity (at 580 and 640 nm emission wavelength) in response to defined pH solutions. (B) Resulting calibration curve applying a non‐linear least squares regression with the means (± SEM) of F₅₈₀/F₆₄₀ ratios plotted against intracellular pH.

Figure 3

Na⁺/H⁺ exchanger (NHE) activity in isolated murine ventricular cardiomyocytes. (A) Original registration of SNARF1 fluorescence (after calibration) before, during, and after NH₄Cl‐dependent acidification in isolated mouse ventricular myocytes. NHE activity was assessed as the maximal slope of pH recovery after NH₄Cl pulse (within rectangle). (B) Averaged pH recordings of the pH recovery phase. (C) Derived maximal slope in ΔpH/min (shown as mean ± standard error of the mean, scatter and spaghetti plots). Compared with vehicle (dimethyl sulfoxide), NHE activity was significantly reduced in the presence of either empagliflozin (Empa, 1 μmol/L) or cariporide (10 μmol/L). The number of mice/cells were 10/12 (dimethyl sulfoxide), 11/17 (empagliflozin), or 5/7 (cariporide). ^* P < 0.05 vs. vehicle, mixed effects analysis with Holm–Sidak's post hoc correction.

Figure 4

Na⁺/H⁺ exchanger (NHE) activity in isolated human atrial cardiomyocytes. (A) Original registrations of SNARF1 fluorescence calibrated to intracellular pH in isolated human atrial cardiomyocytes. NHE activity was assessed as the maximal slope of pH recovery after NH₄Cl pulse. The derived maximal slope (in ΔpH/min) is shown as scatter and spaghetti plots in the presence of (B) either vehicle (veh) or empagliflozin (Empa, 1 μmol/L) and (C) either vehicle or cariporide (10 mmol/L). Compared with veh (dimethyl sulfoxide), NHE activity was significantly reduced in the presence of either Empa or cariporide. The number of patients were 9 (veh) vs. 9 (Empa) and 6 (veh) vs. 6 (cariporide). Wilcoxon matched‐pairs signed‐rank test.