Introduction: Atrial fibrillation (AF) is the most common cardiac arrhythmia and associated with considerable morbidity and mortality. The gap junction protein Connexin 40 (Cx40) is predominantly expressed in atrial cardiomyocytes as well as the conduction system and involved in cell-to-cell conduction of action potentials. Loss-of-function mutations in the Cx40 gene (Cx40A96S mutation) have been detected in patients with AF without structural heart disease. In animal studies, ubiquitous Cx40A96S mutation was associated with a significant prolongation of AF episodes, but also with renin-dependent hypertension. Because this is an independent risk factor for the development of AF, we sought to evaluate a cardiomyocyte-specific Cx40A96S mutation as a murine model of AF.

Methods: We generated two transgenic mouse-lines with either heterozygous ubiquitous (ubi) or cardiomyocyte-specific (cm-s) expression of the Cx40A96S mutation. For assessment of atrial electrical vulnerability electrophysiological investigations were performed in anesthetized mice. Furthermore, we performed electrical activation mapping (EAM) of explanted and Langendorff-perfused hearts using 32 unipolar electrograms recorded from the atrial epicardium. Repetitive blood pressure measurements were performed noninvasively by tail cuff manometry. Hearts were analysed for fibrosis, connexin expression and markers of inflammation.

Results: Baseline parameters and basic electrophysiological characteristics were comparable between ubi, cm-s and wild-type (WT) animals. Atrial electrical stimulation revealed a similar inducibility of AF episodes in all the groups (WT: 70%; ubi: 78%; cm-s: 83%; p=0.5898), but average AF duration was longer in mice expressing the cm-s or ubi mutation with no intergroup difference (WT: 9.3 s; ubi: 26.4 s; cm-s: 26.1 s; p<0.0001 for overall comparison). EAM showed decreased atrial conduction velocities in hearts harboring the Cx40A96S mutation (ubi and cm-s). Mice with cm-s mutation revealed no evidence of renin-dependent hypertension, with comparable blood pressure profiles to WT. To determine the expression and localization of cardiac connexins, we performed immunostaining of cryosectioned atrial tissue and quantitative PCR: cm-s and ubi mutation did not alter Cx40 or Connexin 43 distribution and expression in comparison to WT. Further analyses for atrial inflammation revealed comparable relative gene expression levels of the pro-inflammatory cytokines interleukin 6 (IL-6) and interleukin 1β (IL-1β) in all groups. In addition, macrophage infiltration and accumulation in the myocardium, visualized as CD68 positive cells was unaltered. The Cx40A96S mutation did not cause an increase in atrial fibrosis as quantified by sirius red staining. 

Conclusion: The cardiomyocyte-specific Cx40A96S mutation leads to an increased AF duration mediated by reduced atrial conduction velocities, which is comparable to the ubiquitous mutation model. Therefore, increased perpetuation of AF in mice harboring the Cx40A96S mutation is due to a cell-specific effect in cardiomyocytes, as no changes in blood pressure, atrial fibrosis and inflammation could be detected.