Dual antiarrhythmic properties of phosphodiesterase-5 inhibitors suppress Ca2+ waves in systolic heart failure

Background: Arrhythmias account for up to half of all deaths in heart failure (HF)(1). Despite therapeutic advances new treatments targeting contractile dysfunction and arrhythmias are urgently required. While there are several mechanisms of ventricular arrhythmia in HF, a crucial trigger at the cardiac myocyte level is increased occurrence of Ca release events from the sarcoplasmic reticulum (SR) causing Ca waves. Waves lead to activation of the Na+-Ca2+ exchanger (NCX) current causing after-depolarisations and eliciting triggered action potentials(2). Ca waves occur when SR content reaches a threshold level(3). Phosphodiesterase-5 (PDE5) inhibitors acutely suppress waves and triggered arrhythmias in control myocytes by reducing SR Ca(4). However whether their benefit extends to HF, where structural and electrophysiological remodelling are present, and following chronic treatment, is undetermined. Aims: Determine effect of both acute and chronic PDE5 inhibition on Ca handling and pro-arrhythmic Ca waves in a sheep model of rapid-pacing induced heart failure. Methods: Young (~18 months age) female welsh mountain sheep were used. HF was induced by rapid pacing via a pacemaker (210 beats/min) implanted 10-14 days prior under isoflurane general anaesthesia (2-3%). After 4 weeks of pacing animals were assigned to either HF, where animals underwent tachypacing to end-stage HF (typically 2-3 wks), or tadalafil-HF, where animals were treated with oral tadalafil 20mg for three weeks while pacing was continued. Left ventricular myocytes were isolated, voltage-clamped, and stimulated at 0.5Hz. Intracellular Ca was measured with Fura-2. Ca waves were induced by raising external Ca to 10mM. To determine threshold SR Ca content to induce waves, caffeine (10mM) was added immediately after a wave, and both wave and caffeine-induced NCX currents integrated. Results: HF increased propensity to Ca waves compared to control myocytes by decreasing threshold SR content (128±16 ctrl vs 68±7 µmol/L HF, p<0.001). In HF cells displaying waves, acute PDE5 inhibitor sildenafil (1µM) reduced wave size in all cells (integrated INCX -36 %, p<0.05) and abolished waves in 3/10 cells by reducing SR content below threshold. In chronic PDE5 inhibitor-treated HF, myocytes were resistant to Ca waves, with only 47% cells exhibiting waves compared with 80% of HF cells (p<0.05, n=19-25 cells). In contrast to acute PDE5 inhibition, this was achieved by elevating threshold SR content to control levels, with waves occurring at 105±15 μmol.l-1 vs 70±7 in HF (p<0.05, n= 8-18 cells). Conclusions: PDE5 inhibition suppresses Ca waves in HF via two separate mechanisms; a reduction in SR Ca content after acute treatment, and via chronic remodelling effects after long-term treatment where threshold SR content is increased. Whether such chronic effects relate to modifications in RyR2 properties or structural changes affecting SR and t-tubule networks require further investigation.