E. Schollmeier1, M. Trum1, J. Riechel1, S. Lebek1, P. Hegner1, K. Reuthner1, S. Heers1, K. Keller1, Z. Provaznik2, C. Schmid2, L.S. Maier1, M. Arzt1, S. Wagner1

1Department of Internal Medicine II, University Hospital Regensburg, Regensburg, Germany

2Department of Cardiothoracic Surgery, University Hospital Regensburg, Regensburg, Germany

Background: Current guidelines recommend SGLT2 inhibitors for the treatment of heart failure in patients with reduced ejection fraction (HFrEF). SGLT2i, which have been suggested to improve cellular Na and Ca homeostasis in HFrEF, have also been shown an antiarrhythmic effect in post-hoc analyses. In contrast, the mechanisms of action of SGLTi in heart failure with preserved ejection fraction (HFpEF) are unknown, although improvement in clinical outcomes has recently been demonstrated.

Hypothesis: We hypothesized that an increased Na influx due to enhanced late Na current (late INa) can be found in isolated atrial cardiomyocytes of patients with HFpEF, which may be reduced by acute exposure to empagliflozin (Empa) leading to reduced incidence of premature atrial contractions (PACs).

Methods: Atrial tissue samples were obtained from patients undergoing elective cardiac surgery. Total cellular Na influx was demasked upon inhibition of Na/K-ATPase activity and measured with the Na dye Asante Sodium Green-2 (ANG-2) in isolated atrial cardiomyocytes. Late INa was quantified by using whole cell patch clamp technique.  Expression and phosphorylation of cardiac Na channel NaV1.5 and Ca/Calmodulin-dependent protein kinase II (CaMKII) were assessed using western blot and HDAC4 pulldown assays. PACs were measured in electrically field-stimulated (1 Hz) human atrial trabeculae in the presence of isoproterenol (100 nM) and high external Ca (3.5 mM).

Results: Compared to patients without heart failure (NF) Na influx was almost doubled in patients with HFpEF (NF vs HFpEF: 0.21±0.02 vs 0.38±0.04 mmol/L/min (N=7 vs 18); p=0.005, Fig A+B). Interestingly, enhanced Na influx in HFpEF cardiomyocytes was significantly inhibited by exposure to empagliflozin (Fig A+B), and to a similar extent also by direct Na channel block (tetrodotoxin, TTX, 1 µM) or CaMKII inhibition (autocamptide-2 related inhibitor peptide, AIP, 1 µM). In accordance, compared to NF patients, atrial cardiomyocytes of HFpEF patients had an increased late INa, which could be reduced to approximately the level of NF patients by both Empa or AIP (Fig C) suggesting that CaMKII-dependent Na channel regulation may be involved. Consistent with this, CaMKII expression, CaMKII autophosphorylation, CaMKII activity and CaMKII-dependent NaV1.5 phosphorylation were increased in atrial biopsies of HFpEF patients, whereas NaV1.5 protein abundance remained unaltered (Fig D). Moreover, the enhanced frequency of PACs in HFpEF trabeculae was also significantly reduced by Empa (Fig E).

Conclusion: This is the first study to demonstrate increased Na influx in human cardiomyocytes from HFpEF patients potentially by an increased CaMKII-dependent NaV1.5 phosphorylation and consequent increase in late INa.  Importantly, treatment with Empa inhibited Na influx, at least in part by inhibition of late INa, and exerts antiarrhythmic effects. These results are a possible explanations for the cardioprotective effect of SGLT2 inhibitors in HFpEF.