In search of more efficacious and safe pharmacological treatments for atrial fibrillation (AF), atria-selective antiarrhythmic agents have been promoted that target ion channels principally expressed in the atria. This concept allows one to engage antiarrhythmic effects in atria, but spares the ventricles from potentially proarrhythmic side effects. It has been suggested that cardiac small conductance Ca2+-activated K+ (SK) channels may represent an atria-selective target in mammals including humans. However, there are conflicting data concerning the expression of SK channels in different stages of AF, and recent findings suggest that SK channels are upregulated in ventricular myocardium when patients develop heart failure. To address this issue, RNA-sequencing was performed to compare expression levels of three SK channels (KCNN1, KCNN2, and KCNN3) in human atrial and ventricular tissue samples from transplant donor hearts (no cardiac disease), and patients with cardiac disease in sinus rhythm or with AF. In addition, for control purposes expression levels of several genes known to be either chamber-selective or differentially expressed in AF and heart failure were determined. In atria, as compared to ventricle from transplant donor hearts, we confirmed higher expression of KCNN1 and KCNA5, and lower expression of KCNJ2, whereas KCNN2 and KCNN3 were statistically not differentially expressed. Overall expression of KCNN1 was low compared to KCNN2 and KCNN3. Comparing atrial tissue from patients with AF to sinus rhythm samples we saw downregulation of KCNN2 in AF, as previously reported. When comparing ventricular tissue from heart failure patients to non-diseased samples, we found significantly increased ventricular expression of KCNN3 in heart failure, as previously published. The other channels showed no significant difference in expression in either disease. Our results add weight to the view that SK channels are not likely to be an atria-selective target, especially in failing human hearts, and modulators of these channels may prove to have less utility in treating AF than hoped. Whether targeting SK1 holds potential remains to be elucidated.

No abstract

The healthy heart adapts continuously to a complex set of dynamically changing mechanical conditions. The mechanical environment is altered by, and contributes to, multiple cardiac diseases. Mechanical stimuli are detected and transduced by cellular mechano-sensors, including stretch-activated ion channels (SAC). The precise role of SAC in the heart is unclear, in part because there are few SAC-specific pharmacological modulators. That said, most SAC can be activated by inducers of membrane curvature. The lectin LecA is a virulence factor of Pseudomonas aeruginosa and essential for P. aeruginosa-induced membrane curvature, resulting in formation of endocytic structures and bacterial cell invasion. We investigate whether LecA modulates SAC activity. TREK-1 and Piezo1 have been selected, as they are widely expressed in the body, including cardiac tissue, and they are "canonical representatives" for the potassium selective and the cation non-selective SAC families, respectively. Live cell confocal microscopy and electron tomographic imaging were used to follow binding dynamics of LecA, and to track changes in cell morphology and membrane topology in human embryonic kidney (HEK) cells and in giant unilamellar vesicles (GUV). HEK cells were further transfected with human TREK-1 or Piezo1 constructs, and ion channel activity was recorded using the patch-clamp technique. Finally, freshly isolated cardiac cells were used for studies into cell type dependency of LecA binding. LecA (500 nM) binds within seconds to the surface of HEK cells, with highest concentration at cell-cell contact sites. Local membrane invaginations are detected in the presence of LecA, both in the plasma membrane of cells (by 17 min of LecA exposure) as well as in GUV. In HEK cells, LecA sensitizes TREK-1, but not Piezo1, to voltage and mechanical stimulation. In freshly isolated cardiac cells, LecA binds to non-myocytes, but not to ventricular or atrial cardiomyocytes. This cell type specific lack of binding is observed across cardiomyocytes from mouse, rabbit, pig, and human. Our results suggest that LecA may serve as a pharmacological tool to study SAC in a cell type-preferential manner. This could aid tissue-based research into the roles of SAC in cardiac non-myocytes.