Expression and function of Kv1.1 potassium channels in human atria from patients with atrial fibrillation

Abstract

Voltage-gated Kv1.1 channels encoded by the Kcna1 gene are traditionally regarded as being neural-specific with no known expression or intrinsic functional role in the heart. However, recent studies in mice reveal low-level Kv1.1 expression in heart and cardiac abnormalities associated with Kv1.1-deficiency suggesting that the channel may have a previously unrecognized cardiac role. Therefore, this study tests the hypothesis that Kv1.1 channels are associated with arrhythmogenesis and contribute to intrinsic cardiac function. In intra-atrial burst pacing experiments, Kcna1-null mice exhibited increased susceptibility to atrial fibrillation (AF). The atria of Kcna1-null mice showed minimal Kv1 family ion channel remodeling and fibrosis as measured by qRT-PCR and Masson's trichrome histology, respectively. Using RT-PCR, immunocytochemistry, and immunoblotting, KCNA1 mRNA and protein were detected in isolated mouse cardiomyocytes and human atria for the first time. Patients with chronic AF (cAF) showed no changes in KCNA1 mRNA levels relative to controls; however, they exhibited increases in atrial Kv1.1 protein levels, not seen in paroxysmal AF patients. Patch-clamp recordings of isolated human atrial myocytes revealed significant dendrotoxin-K (DTX-K)-sensitive outward current components that were significantly increased in cAF patients, reflecting a contribution by Kv1.1 channels. The concomitant increases in Kv1.1 protein and DTX-K-sensitive currents in atria of cAF patients suggest that the channel contributes to the pathological mechanisms of persistent AF. These findings provide evidence of an intrinsic cardiac role of Kv1.1 channels and indicate that they may contribute to atrial repolarization and AF susceptibility.

Fig. 1

Mice lacking Kv1.1 channels are vulnerable to pacing-induced AF. a Representative simultaneous recordings of surface ECG (lead I) and intracardiac atrial and ventricular electrograms after burst pacing in wild type (WT) and Kcna1-null (KO) mice. In the KO animal shown, pacing produced AF (AF onset) that lasted about 13 s before returning to normal sinus rhythm (AF termination). b During inducible AF in KO animals, the surface ECG exhibited absent P waves and irregular RR intervals, which are shown labeled in milliseconds. c KO mice showed significantly higher incidence of pacing-induced AF than WT controls. *P < 0.05

Fig. 2

Kcna1-null mice exhibit minimal Kv1.x channel remodeling and atrial fibrosis. a Real-time PCR expression analysis of Kv1.x potassium channel α- and β-subunit genes in atria from KO and control WT mice (n = 5 mice per genotype). b Real-time PCR analysis of col6a1 mRNA levels (n = 5 mice per genotype). c Representative samples of atria from WT and KO animals stained with Masson’s trichrome to visualize fibrosis which appears bluish. Images were chosen out of sections from 5 WT mice and 5 KO mice. d Quantification of the percentage of atrial fibrosis between genotypes. *P < 0.05; ns not significant

Fig. 3

Kcna1 expression in mouse cardiomyocytes. a RT-PCR detection of Kcna1 mRNA in isolated atrial (A) and ventricular (V) myocytes from three mice labeled 1–3. Brain tissue (B) was used as a positive control, while no cDNA (N) was used as a negative control. b Quantitative RT-PCR shows that Kcna1 mRNA levels (normalized to Rpl7) are about tenfold more abundant in atrial myocytes than in ventricular (n = 3 mice; P = 0.07, paired t test). Immunostaining of isolated atrial (c) and ventricular (d) myocytes isolated from adult mice and co-stained with antibodies against Kv1.1 and Junctophilin-2 (Jph2), a t-tubule protein. The gain was increased in the ventricular images relative to the atrial images to allow visualization of Kv1.1 immunoreactivity, which is much less intense in ventricular myocytes. The boxed regions in the overlay images are shown magnified 500 % in the bottom panels. Representative images were chosen out of 15 cells (7 atrial and 8 ventricular myocytes) from 2 WT mice and 8 cells (5 atrial and 3 ventricular myocytes) from 1 KO mouse. Scale bars 20 µm

Fig. 4

KCNA1 mRNA and protein expression in human atria. a RT-PCR detection of KCNA1 mRNA in human atria. b mRNA levels (mean ± SEM) of KCNA1 in patients with chronic AF (cAF) and paroxysmal AF (pAF) relative to sinus rhythm (SR). Representative Western blots of Kv1.1 protein with corresponding densitometric quantification (mean ± SEM) of protein levels in patients with pAF (c) and cAF (d) relative to SR. *P < 0.05 vs. SR; ns not significant. Numbers in bars indicate atrial samples

Fig. 5

Kv1.1 immunoreactivity in isolated human atrial cardiomyocytes. Representative confocal images of Kv1.1-positive immunostaining (a, b) and a negative control (c). Kv1.1 immunoreactivity showed a strong fluorescent signal throughout the cell and a tight striated pattern which overlapped with the Z-line. In negative controls, primary antibody was omitted and cells were labeled only with the secondary antibody revealing an absence of nonspecific, background staining

Fig. 6

Dendrotoxin-K sensitive outward current in isolated human atrial myocytes. a Averaged current traces recorded from human atrial myocytes from patients in sinus rhythm (SR) or chronic AF (cAF). Current traces show control measurements at baseline followed by application of 10 nmol/L dendrotoxin-K (DTX-K) to specifically block Kv1.1 channels. The voltage step protocol used for recordings is shown at the top. b Traces show the averaged current difference representing the dendrotoxin-K sensitive component. c Amplitudes of the peak and late phases of the outward K⁺ currents (mean ± SEM) before (control) and after application of DTX-K in atrial myocytes from SR and cAF patients. d Amplitudes of the peak and late phases of the DTX-K sensitive current difference (mean ± SEM) in atrial myocytes from SR and cAF patients. *P < 0.05, **P < 0.01, ***P < 0.001 vs. corresponding means under control conditions using Student’s paired (c) and unpaired (d) t test, respectively. The n values indicate myocyte/patient numbers.