Question: Diabetes is a major health burden worldwide, leading to diabetic cardiomyopathy (DiabCM). One major hallmark of DiabCM is an impaired calcium handling promoting ventricular arrhythmias. This is partly caused by spontaneous calcium releases from the sarcoplasmatic reticulum via ryanodine receptors due to increased phosphorylation by cAMP-dependent kinases. Phosphodiesterases (PDEs) are hydrolyzing enzymes that degrade cAMP. The subtype PDE2 has the unique property to be stimulated by cGMP, thereby shutting down cAMP levels. Accordingly, cGMP signaling downstream of natriuretic peptide receptors has been shown to be beneficial in several heart diseases. Here, we address the impact of CNP/cGMP-mediated PDE2 activation on arrhythmogenic events in cardiomyocytes from mice with DiabCM.

Methods: Mice with heart-specific PDE2 knockout (PDE2 KO) or PDE2 overexpression (PDE2 OE) were injected with 50 mg/kg streptozotocin (STZ) on 5 consecutive days to induce DiabCM. Echocardiography was performed before and 4 weeks after the STZ treatment. At this time point, primary ventricular cardiomyocytes (CM) of PDE2 KO mice were isolated to measure Ca²⁺ sparks (CaSp) and investigate protein expressions by Western Blot. Diabetic hearts from PDE2 OE mice were subjected to ischaemia/reperfusion injuries (I/R) ex-vivo on a Langendorff perfusion system and the number of arrhythmic events occurring during reperfusion was analysed.

Results: After STZ treatment, the mice developed hyperglycemia (blood glucose levels >350 mg/dl).  Echocardiography revealed a reduced cardiac function with significantly lower ejection fraction (EF) and fractional area shortening (FAS) in diabetic PDE2 KO mice compared to diabetic wildtype (WT) mice. Isolated CM of the diabetic hearts showed significant changes in expression and phosphorylation levels of proteins within the β-adrenergic signaling cascade compatible with a pro-arrhythmogenic profile, e.g. higher CaMKII phosphorylation. Interestingly, CaMKII phosphorylation tended to be even higher in diabetic PDE2 KO mice. Moreover, stimulation with catecholamines resulted in a significantly higher number of arrhythmogenic CaSp in DiabCM. Importantly, the simultaneous application of CNP clearly anatgonized the catecholmine-induced increase of CaSp frequency. Of note, genetic deletion of PDE2 or specific PDE2 inhibition with BAY 60-7550 prevented the CNP effect. Quantifying arrhythmia on organ level, isolated diabetic WT hearts showed a slightly higher number of arrhythmic events after I/R compared to non-treated control WT mice. Remarkably, diabetic hearts from PDE2 OE mice were protected from arrhythmia development, showing a significantly lower number of arrhythmias compared to STZ-treated WT mice.

Conclusion: CNP-mediated PDE2 stimulation reduces intracellular pro-arrhythmic triggers and might therefore represent a novel antiarrhythmic strategy in DiabCM. Our data indicate that upregulated PDE2 might protect the diabetic heart from arrhythmic events. Thus, PDE2 activation might represent a novel therapeutic strategy to treat DiabCM induced arrhythmogenesis.