Epilepsy-associated Kv1.1 channel subunits regulate intrinsic cardiac pacemaking in mice

The heartbeat originates from the sinoatrial node (SAN), a specialized cluster of pacemaker cells located in the right atrium (Peters et al., 2020). These SAN myocytes spontaneously and rhythmically depolarize to generate action potentials, providing the electrical excitation that initiates each heart contraction (Mangoni and Nargeot, 2008). SAN myocytes exhibit distinctive action potentials that consist of only three sequential phases: slow diastolic depolarization, followed by rapid depolarization, and then repolarization (Zheng et al., 2023; Mangoni and Nargeot, 2008). The rates of repolarization and diastolic depolarization, which are critical determinants of the pacemaker rhythm, are largely regulated by K⁺ currents conducted by several types of K⁺ channels (Aziz et al., 2018; Peters et al., 2020). Repolarization is principally mediated by voltage-gated Kv7.1 and Kv11.1 (ERG) channels and ATP-sensitive Kir6.1 channels, which underlie the I_Ks, I_Kr, and I_KATP, respectively (Aziz et al., 2018; Mangoni and Nargeot, 2008). Additionally, diastolic depolarization is influenced by Ca²⁺-activated K⁺ channels (BK and SK4) and G protein-gated inwardly rectifying potassium (GIRK) channels (Kir3.1 and Kir3.4), which underlie I_KCa and I_KACh currents, respectively (Bettahi et al., 2002; Weisbrod et al., 2016; Lai et al., 2014; Wickman et al., 1998). However, gene expression studies of SAN tissue have revealed the presence of mRNA for numerous other types of K⁺ channels, suggesting they could also play previously unrecognized roles in pacemaking (Marionneau et al., 2005; Leoni et al., 2005). For example, studies of mouse SAN tissue have identified transcripts for most members of the voltage-gated Kv1 channel family, including Kv1.1–Kv1.6, yet their potential contributions to SAN function have remained largely unexplored (Marionneau et al., 2005; Leoni et al., 2005).

Here, we investigated pore-forming Kv1.1 voltage-gated potassium channel α-subunits for a role in cardiac pacemaking. Kv1.1 subunits, encoded by the Kcna1 gene, are important regulators of neuronal excitability in the nervous system where their dysfunction can lead to epilepsy and episodic ataxia (Paulhus et al., 2020; Paulhus and Glasscock, 2023). More recently, Kv1.1 subunits have also emerged as important for normal cardiac physiology, controlling atrial and ventricular action potential repolarization and arrhythmia susceptibility (Glasscock et al., 2015; Si et al., 2019; Trosclair et al., 2021; Glasscock, 2019). In mouse and ferret hearts, the SAN shows varying levels of Kv1.1 transcripts ranging from low to high abundance relative to other K⁺ channels depending on the study (Brahmajothi et al., 1996, 2010; Glasscock et al., 2010; Leoni et al., 2005; Marionneau et al., 2005). Additionally, when mice are subjected to chronic heart rate reduction by administering the HCN channel-blocker ivabradine, the SAN exhibits complex remodeling of ion channel transcripts in the SAN that includes a 30% increase in Kv1.1 mRNA levels (Leoni et al., 2005). The expression of Kcna1 in the SAN suggests that Kv1.1 may play a role in regulating SAN function and cardiac pacemaking, but this has never been examined directly.

To address this gap, in this work, we employed combined electrophysiological and pharmacological approaches in isolated heart tissues and SAN myocytes from Kcna1^−/− and wild-type (WT) mice to test the hypothesis that Kv1.1 subunits are required for normal SAN function. Specifically, we performed Langendorff-perfused isolated heart recordings, multielectrode array (MEA) recordings, and patch-clamp recordings to measure the contribution of Kv1.1 to cardiac pacemaking at the heart, SAN, and SAN myocyte levels, respectively. We also used immunocytochemistry to analyze the presence of Kv1.1 protein in SAN cells. We find that genetic deletion or pharmacological inhibition of Kv1.1 leads to intrinsic bradycardia. These results provide the first evidence that Kv1.1 is critical for normal SAN function, implicating Kcna1 as a potential susceptibility gene for sinoatrial pacemaking dysfunction. Furthermore, our findings illustrate the important principle that low-abundance ion channels that mediate small currents can exert large effects on cardiac pacemaking.

Kcna1 knockout mice (Kcna1^−/−) were generated by targeted deletion of the open reading frame of the Kcna1 gene on chromosome 6, as previously described (Smart et al., 1998). The mice were bred as heterozygotes and maintained on a Tac:N:NIHS-BC genetic background. Mice were housed at ∼22°C, fed ad libitum, and submitted to a 12-h light/dark cycle. All procedures were performed in accordance with the guidelines of the National Institutes of Health, as approved by the Institutional Animal Care and Use Committees of Southern Methodist University and Louisiana State University Health Sciences Center-Shreveport. Genotypes were determined using allele-specific PCR amplification of genomic tail DNA isolated by enzymatic digestion of tail clips using Direct-PCR Lysis Reagent (Viagen Biotech) followed by gel electrophoresis. The primer sequences used for PCR and the resulting amplicon sizes have been described previously (Trosclair et al., 2021).

Isolated heart recordings were performed using 5-month-old age-matched WT (n = 12: 4 male, 8 female; age 149 ± 31 days) and Kcna1^−/− mice (n = 11: 2 male, 9 female; age 155 ± 20 days) of both sexes. 1 h prior to decapitation, each mouse was administered an intraperitoneal (i.p.) injection of heparin sodium (10,000 USP/kg). Immediately after decapitation, the heart and surrounding pericardium were rapidly excised and placed in 5–10°C Krebs–Henseleit bicarbonate (KHB) buffer (in mM: 118 NaCl, 4.7 KCl, 1.25 CaCl₂, 1.20 MgSO₄, 1.20 KH₂PO₄, 25 mM NaHCO₃, and 11 D-glucose; pH 7.4; prepared in 18.5 MΩ water and filtered through a 0.22-μm filter) that was oxygenated with a 95% O₂–5% CO₂ gas mixture. After excess pericardial tissue and fat were trimmed, the heart was mounted on a Langendorff apparatus by cannulation of the aorta with a stainless steel perfusion cannula held in place by silk thread. Each heart was then perfused with KHB and maintained at 37°C by a circulating water bath. A perfusion pressure of 65 mmHg was maintained through the aortic cannula for a period of ∼20 min to remove all blood and allow the heart to recover from the brief period of hypoxia associated with excision. The perfusion column and all feeding lines were double-wall insulated to minimize heat loss and to maintain perfusate temperature, which was continuously monitored by a microprobe thermometer (Physitemp Instruments, Inc) to ensure that 37°C KHB buffer entered the heart through the aortic cannula.

ECG was obtained using an electrode set (140155; Radnoti) composed of an electrode for monophasic action potential recording, positioned at the heart’s apex and spring-loaded to maintain constant contact with the beating heart; and a reference electrode consisting of a flexible platinum wire, which was wrapped around the aorta. ECG signals were amplified with an Animal Bio Amp (ADInstruments) and acquired using PowerLab 8/35 high-performance data acquisition hardware and LabChart software (ADInstruments). After the 20-min equilibration period, continuous ECG activity was recorded for a period of 90 min. ECG waveform characteristics for each heart were analyzed by calculating averages from 2-min recordings sampled during times when the recordings were stable and free of artifacts. ECG waveform components were analyzed using the mouse ECG analysis module of the LabChart software. Briefly, the RR interval was measured as the duration between consecutive R wave peaks, and the PR segment was measured from the end of the P wave to the start of the QRS complex. The QT interval was measured from the beginning of the Q wave until the T wave returned to baseline and heart rate-corrected using the Mitchell algorithm (Mitchell et al., 1998): $QTc=QT/(RR/100)0.5$⁠.

MEA recordings were performed as detailed previously (Kumar et al., 2021). In brief, age- and sex-matched WT (n = 18: 9 males, 9 females; age 49 ± 6 days) and KO mice (n = 13: 6 males, 7 females; age 44 ± 8 days) were euthanized by isoflurane overdose and then the heart was quickly excised and placed in a dish containing complete Tyrode’s solution containing (in mM) 140 NaCl, 1.2 KH₂PO₄, 1.8 CaCl₂.2H₂O, 5.4 KCl, 1 MgCl₂, 5.0 HEPES, and 5.55 glucose. The heart was then dissected to isolate the region of atrial tissue containing the SAN, which was identified as the area corresponding to the SAN artery and bounded by the superior and inferior vena cavae and the crista terminalis. The sinoatrial tissue was then transferred to a recording chamber containing an 8 × 8 planar microelectrode array (150-µm interelectrode distance), which was connected to a 64-channel amplifier (MED 64 System: Alpha Med Science). Spontaneous extracellular field potentials were then recorded in an oxygenated recording Tyrode solution (gassed with 95% O₂, 5% CO₂) maintained at 37°C (pH 7.2–7.4) and containing (in mM) 137 NaCl, 15.5 NaHCO₃, 0.7 NaH₂PO₄, 1.8 CaCl₂.2H₂O, 4 KCl, 1 MgCl₂, and 11.1 glucose. The tissue was perfused by recording Tyrode’s solution at a constant flow rate of 2 ml/min. The data was filtered with a 1 Hz low pass filter and a 1,000 Hz high pass filter. For each tissue preparation, the beat frequency was determined by averaging the field potential measurements during three stable 60-s traces from all 64 microelectrodes using MED64-Mobius software.

SAN cells were isolated from age-matched 6- to 8-wk-old WT and Kcna1^−/− mice. Mice were heparinized (1,000 USP/ml, 200–300 μl i.p.) and euthanized by isoflurane overdose followed by cervical dislocation. The heart was quickly removed and the SAN was dissected in warm complete Tyrode’s solution with heparin (10 USP/ml) (37°C), as described previously (Kumar et al., 2021). The complete Tyrode’s solution was composed of (in mM) 140 NaCl, 5.4 KCl, 1.2 KH₂PO₄, 5.0 HEPES, 5.55 glucose, 1 MgCl₂, and 1.8 CaCl₂ with pH adjusted to 7.4 with NaOH. The SAN tissue was digested in low Ca²⁺/Mg²⁺ Tyrode’s with protease (1.0024 U/2.5 ml) and Liberase TH (0.15 mg/2.5 ml) for 30 min. The low Ca²⁺/Mg²⁺ Tyrode’s solution was composed of (in mM) 140 NaOH, 5.4 KOH, 1.2 KH₂PO₄, 5.0 HEPES, 18.5 glucose, 0.066 CaCl₂, 50 taurine, and 0.015 BSA with pH adjusted to 6.9 with NaOH. After digestion, the cells were dissociated by pipetting and then stored in KB solution composed of (in mM) 25 KCl, 10 KH₂PO₄, 5.0 HEPES, 20 glucose, 20 taurine, 100 K-glutamate, 10 K-aspartate, 2 MgSO₄, 5 creatine, 0.5 EGTA, and 0.015 BSA with pH adjusted to 7.2 with KOH.

Whole-cell patch-clamp recordings were performed at 35°C using borosilicate glass pipette microelectrodes with tip resistances of 2.0–3.0 MΩ when filled with pipette solution. Electrodes were connected to a MultiClamp 700B amplifier (Axon Instruments, Molecular Devices) and electrical signals were digitized with an Axon analog/digital converter (Digidata 1440, Molecular Devices). Data acquisition was performed by Clampfit software (version 11.0.3, Axon Instruments, Molecular Devices). The pipette solution contained (in mM) 5 NaCl, 135 KCl, 10 HEPES, 1 MgCl₂, 0.1 CaCl₂, 10 EGTA, and 4 Mg-ATP with pH adjusted to 7.2 with KOH. The complete Tyrode’s solution used as the bath solution was composed of (in mM) 140 NaCl, 5.4 KCl, 1.2 KH₂PO₄, 5.0 HEPES, 5.55 glucose, 1 MgCl₂, and 1.8 CaCl₂ with pH adjusted to 7.4 with NaOH and 1 nM isoproterenol, as done previously (Clark et al., 2004). The liquid junction potential between the pipette solution and complete Tyrode’s was 4.3 mV as calculated using the JPCalc application in the Clampex software (Molecular Devices). The series resistance was <10 MΩ and compensated for each cell during recording. Spontaneous action potentials were recorded in current clamp mode without current injection. Firing frequency and action potential waveform characteristics were measured in 2-s segments of stable recordings. Diastolic depolarization rate (DDR) was measured as the change in membrane potential during the first 20 ms after the point of the maximum diastolic potential (MDP) + 1 mV (Wilders and Jongsma, 1993). For I_f current recordings in voltage clamp mode, SAN cells were perfused with complete Tyrode’s solution containing 1 mM BaCl₂ to inhibit K⁺ currents. I_f was elicited by 4-s hyperpolarizing voltage steps from −60 mV to −150 mV in 10 mV increments, from a holding potential of −50 mV. For outward K⁺ current recordings in voltage clamp mode, peak outward K⁺ currents were measured as current densities at the end of the 1-s depolarizing voltage steps of +10 mV increments from a holding potential of −50 mV. To control for myocyte-size variability, currents were expressed as densities (pA/pF). The membrane capacitances (Cm) of the cells were 45–55 pF. Activation curves were obtained by plotting peak outward K⁺ current densities against test potential voltage. Dendrotoxin-K (DTX-K; 10 nM; Sigma-Aldrich) was used to selectively block Kv1.1 channels, as done previously (Glasscock et al., 2015; Trosclair et al., 2021; Si et al., 2019). All chemicals used to make the bath and pipette solutions for recordings were obtained from Sigma-Aldrich.

Isolated SAN cells were allowed to settle on laminin-coated chambered slides for at least 2 h followed by fixation with 3.2% paraformaldehyde for 10 min. Cells were permeabilized using 0.1% Triton in phosphate-buffered saline (PBS) and blocked in 10% goat serum for 1 h at room temperature (RT; ∼22°C). Cells were incubated overnight at RT with 1:200 dilution of rabbit polyclonal anti-HCN4 antibody (APC-052; Alomone) and 1:500 dilution of mouse monoclonal anti-Kv1.1 antibody (K20/78; NeuroMab). HCN4 and Kv1.1 immunoreactivities were visualized with 1:1,000 Alexa Fluor 594 and 488 secondary antibodies (Invitrogen), respectively. After three washes with PBS, a coverslip was mounted onto the slides using ProLong Glass Antifade Mountant with NucBlue (Invitrogen).

Tyramide signal amplification (kit B40941; Invitrogen) was used to boost the Kv1.1 immunoreactivity signal according to the manufacturer’s recommended protocol using a 1:500 dilution of mouse anti-Kv1.1 (NeuroMab) and a 1:1,000 tyramide-labeled Alexa Fluor 488 antibody. Anti-HCN4 immunoreactivity was performed the same as described above without tyramide amplification. After three washes with PBS, a coverslip was mounted onto the slides using ProLong Glass Antifade Mountant with NucBlue (Invitrogen).

Images were collected at 40× magnification using an automated microscope (LionHeart FX; BioTek). SAN cell identity was confirmed by HCN4 labeling. Negative control experiments were performed by incubating cells with secondary antibodies only to confirm the absence of a signal. Images were always acquired using the same optimized settings for each fluorophore. Brightness and contrast were minimally adjusted to improve the visualization of acquired images, but these settings were applied equally to each fluorophore for all images. Merged images were not further adjusted. Analysis was performed in a blinded fashion for 10 cells/animal for WT (n = 4) and KO (n = 4) using ImageJ software. Corrected total cell fluorescence (CTCF) was calculated for individual cells from the following equation: CTCF = Integrated density – (selected cell area × mean fluorescence of background reading).

GraphPad Prism 9 software (GraphPad Software, Inc.) was used for statistical analyses. All data are reported as means ± SD. Statistical comparisons between genotypes were performed using unpaired two-tailed Student’s t tests. For comparisons of ECG parameters in isolated hearts, multiple t tests were performed followed by multiple comparisons correction by the Holm–Sidak method using a significance level <0.05. For comparisons of DTX-K effects in single cells, paired two-tailed Student’s t tests were used. For comparisons of corrected total cell fluorescence, the data were not normally distributed so values were log-transformed and a non-parametric Mann–Whitney test was performed on the log-transformed data. P values <0.05 were considered to be statistically significant.

Fig. S1 includes representative images and quantification of HCN4 immunoreactivity in WT and KO SAN cells. The experiments were performed without tyramide signal amplification of HCN4 immunoreactivity.

To test for differences in intrinsic cardiac pacemaking at the whole-heart level due to Kv1.1 deficiency, we performed Langendorff-perfused isolated heart recordings in adult age-matched Kcna1^−/− mice and WT controls. ECGs from Kcna1^−/− hearts (n = 11: 2 males, 9 females) exhibited spontaneous heart rates (HRs) that were significantly reduced by 27% compared with WT hearts (n = 12: 4 males, 8 females; P = 0.0009, unpaired two-tailed Student’s t test; Fig. 1, A and B), indicating that isolated hearts, which are devoid of active autonomic and humoral influences, have intrinsic bradycardia in the absence of Kv1.1. To assess cardiac electrical conduction, we measured ECG waveform components, but no significant differences were identified except for prolonged QT intervals in Kcna1^−/− hearts (P = 0.0005, multiple t tests with Holm–Sidak correction; Fig. 1 C). However, this difference disappeared when the QT intervals were corrected for HR (QT_c; P = 0.37, multiple t tests with Holm–Sidak correction; Fig. 1 C), suggesting that the long QT intervals in Kcna1^−/− hearts were due to their lower intrinsic HR. Structurally, Kcna1^−/− hearts (n = 18: 7 males, 11 females; age 152 ± 22 days) tended to be marginally smaller than WT hearts, but the difference was not significant (n = 17: 6 males, 11 females; age 155 ± 27 days; Fig. 1 D). The marginally smaller heart size was likely due to the smaller overall body sizes of the Kcna1^−/− mice (P = 0.015, unpaired two-tailed Student’s t test) because the two genotypes exhibited comparable heart mass-to-body mass ratios (P = 0.48, unpaired two-tailed Student’s t test; Fig. 1, D–F).

To isolate the SAN and to reduce potential mechano-electric effects on recordings due to feedback from the ventricles, AV node, or Purkinje fibers, we dissected the portion of the right atria containing the SAN from age- and sex-matched Kcna1^−/− (n = 13: 6 males, 7 females) and WT mice (n = 18: 9 males, 9 females) and performed recordings using a MEA. Kcna1^−/− SAN preparations exhibited spontaneous firing rates that were 12% slower than WT (P = 0.0083, unpaired two-tailed Student’s t test; Fig. 2, A and B). Thus, isolated SAN tissue from Kv1.1-deficient mice exhibited a significant reduction of spontaneous firing rate, indicative of intrinsic pacemaking defects.

Since both isolated hearts and regionally dissected SAN tissue from Kcna1^−/− mice exhibited slower heart rhythms, we next tested whether this bradycardia was due to a reduction in the spontaneous firing rate of individual SAN myocytes. To measure this, we freshly isolated mouse SAN myocytes and performed recordings using whole-cell patch clamp electrophysiology. We visually identified the SAN myocytes by their elongated spindle shape and by their spontaneous beating (Fig. 3 A). To confirm the cells were SAN myocytes, we also performed voltage-clamp recordings of HCN channel–mediated hyperpolarization-activated funny current, a hallmark of SAN myocytes (Fig. 3 B).

In current-clamp recordings, Kcna1^−/− myocytes (n = 14 cells from 8 male and 3 female mice) exhibited spontaneous firing rates that were significantly slower than WT SAN myocytes (n = 19 cells from 9 male and 6 female mice) by about 25% (P = 0.0036, unpaired two-tailed Student’s t test; Fig. 3, C and E). Selective pharmacological inhibition of Kv1.1 in WT myocytes using DTX-K (10 nM), a Kv1.1-specific blocker (Wang et al., 1999; Si et al., 2019), reproduced this decrease in firing rate, causing a significant 17% slowing (P = 0.0017, paired two-tailed Student’s t test; Fig. 3, D and F). When we treated Kcna1^−/− myocytes with DTX-K, we observed no significant effects on firing rate (P = 0.12; paired two-tailed Student’s t test; Table 1; and Fig. 3, D and G), demonstrating the specificity of the drug for Kv1.1 subunits. Thus, either genetic or pharmacological inhibition of Kv1.1 slowed SAN myocyte firing.

Given Kv1.1’s established role in regulating action potential repolarization in the atria and ventricles (Glasscock et al., 2015; Si et al., 2019; Trosclair et al., 2021), we next focused on the waveform properties of the spontaneous action potentials in SAN myocytes to determine whether action potential prolongation could be a possible mechanism contributing to the slower firing rate in Kcna1^−/− myocytes. We found that action potential durations (APD) at 50 and 90% of repolarization were significantly prolonged in Kcna1^−/− cells compared to WT myocytes (APD₅₀, P = 0.032; APD₉₀; P = 0.0041; unpaired two-tailed Student’s t test; Table 1). When we treated WT myocytes with DTX-K (10 nM) to pharmacologically inhibit Kv1.1, they also exhibited significantly prolonged APD₅₀ and APD₉₀ (P = 0.012 and 0.00059 for APD₅₀ and APD₉₀, respectively, paired two-tailed Student’s t test) with values similar to those observed for untreated Kcna1^−/− myocytes (Table 1). Additionally, we also measured the MDP and DDR in WT and Kcna1^−/− myocytes, with and without administration of DTX-K. Both MDP and DDR showed no significant differences between genotypes (Table 1). However, in WT myocytes, DDR was significantly decreased in response to DTX-K treatment (Table 1). Thus, genetic or pharmacological ablation of Kv1.1 caused prolongation of action potentials, which is likely the primary mechanism responsible for the observed decrease in firing rate.

To determine whether Kv1.1 subunits contribute to outward K⁺ currents in SAN myocytes which, could affect APD, we performed voltage-clamp recordings utilizing DTX-K to measure peak outward currents in response to 1-s depolarizing voltage steps from −50 to +50 mV (Fig. 4 A). In WT cells (n = 7 cells from four male and one female mice), application of DTX-K channels caused significant reductions of 14–19% in peak outward current amplitudes for holding potentials between +20 and +50 mV, indicative of a small but significant contribution by Kv1.1-containing channels (Fig. 4 B). In contrast, DTX-K had no significant measurable effects on outward currents in Kcna1^−/− cells (n = 5 cells from three male and one female mice) further demonstrating the drug’s specificity for Kv1.1 (Fig. 4, C and D). Hereafter, we refer to this newly identified DTX-K-sensitive current in SAN myocytes as I_Kv1.1. Finally, comparison of outward currents in untreated WT (n = 7 cells from four male and one female mice) and Kcna1^−/− cells (n = 9 cells from three male and two female mice) without DTX-K revealed a significant 27% decrease in Kcna1^−/− cells at a holding potential of +50 mV due to elimination of I_Kv1.1 (Fig. 4 E). Thus, pharmacological or genetic ablation of Kv1.1 significantly attenuated outward repolarizing currents in SAN myocytes.

Previous studies have demonstrated that Kv1.1 transcripts are present in mouse SAN but their corresponding protein expression was not investigated (Glasscock et al., 2010; Leoni et al., 2005; Marionneau et al., 2005). Therefore, we used immunocytochemistry to analyze the presence of Kv1.1 protein in isolated mouse SAN myocytes, the identity of which was confirmed by colabeling with HCN4 antibody as a specific marker of SAN cells. Using immunocytochemistry, we identified Kv1.1-positive immunoreactivity in isolated SAN myocytes from WT mice, but the staining intensity was very weak making it difficult to quantify reliably. No Kv1.1 immunoreactivity was evident in SAN myocytes from KO mice demonstrating the specificity of our anti-Kv1.1 antibody.

To better visualize Kv1.1 expression, we repeated the immunocytochemistry experiments using tyramide signal amplification (TSA) to increase the sensitivity of Kv1.1 detection, as done previously in studies of low-abundance sinoatrial ion channels (Lai et al., 2014). Following TSA to amplify the Kv1.1 signal, the Kv1.1 staining intensity was increased, showing a predominantly diffuse staining pattern that we could not ascribe to any particular subcellular compartment or structure (Fig. 5 A). To quantify the amplified signal, we measured the CTCF, which showed that Kv1.1 immunoreactivity was significantly higher in WT cells (n = 40 cells total from 4 mice; 10 cells/mouse) compared with KO cells (n = 40 cells total from 4 mice; 10 cells/mouse; P = 0.0029, Mann–Whitney U test; Fig. 5 B), which showed some low-level background staining after amplification. We also quantified the non-amplified HCN4 immunoreactivity signal in the same cells and found no significant difference in the degree of HCN4 staining between WT and KO SAN myocytes (P = 0.17, Mann–Whitney U test; Fig. S1), suggesting HCN4 expression is not significantly altered by the absence of Kv1.1. Thus, the Kv1.1 protein was present in SAN myocytes but only at low levels.

We report here for the first time that Kv1.1 is required for normal SAN firing rate and therefore cardiac pacemaking in mice. By performing electrophysiological recordings at the heart, SAN tissue, and single SAN myocyte levels, we showed that genetic or pharmacological abolishment of Kv1.1 function decreases the firing frequency of the SAN and SAN myocytes due largely to the prolongation of action potentials. In addition, we found that Kv1.1 mediates a small outward K⁺ current component (I_Kv1.1) in SAN myocytes that is absent in Kcna1^−/− myocytes. Finally, we also provide the first evidence that mouse SAN myocytes express Kv1.1 protein, which appears to be present in only very small amounts based on immunocytochemical staining results. Nevertheless, despite mediating a small current and having low apparent abundance, Kv1.1 exerts substantial effects on intrinsic SAN pacemaking function, providing proof of the principle that small currents can have large functional effects. These findings establish Kv1.1 as a novel ion channel required for normal sinoatrial automaticity and the first identified member of the Kv1 family of channels to participate in cardiac pacemaking.

Potassium channels can control SAN automaticity by regulating the rate of action potential repolarization (Aziz et al., 2018; Peters et al., 2020). Our study found that Kcna1^−/− SAN myocytes exhibit significant APD prolongation relative to WT myocytes, a defect that was also recapitulated by treating WT myocytes with the Kv1.1-specific blocker DTX-K. Therefore, elongated APD is likely the primary mechanism responsible for the decreased firing rates of spontaneous cardiac action potentials in mice lacking Kv1.1. Furthermore, in previous electrophysiological studies of mouse atrial and ventricular myocytes, Kv1.1 inhibition also led to significant prolongation of APD, further emphasizing the importance of Kv1.1 in mediating cardiac action potential repolarization (Trosclair et al., 2021; Si et al., 2019). Thus, Kv1.1 plays a critical role in controlling cardiac excitability throughout the heart, not only in the SAN, and a lack of Kv1.1 can disturb heart rhythms.

Another way that potassium channels can regulate SAN automaticity is by influencing the rate of diastolic depolarization (Aziz et al., 2018; Peters et al., 2020). In our recordings, Kcna1^−/− SAN myocytes exhibited no significant changes in MDP or DDR. These findings suggest that the diastolic depolarization phase of the action potential cycle does not significantly contribute to the pacemaking deficits in mice lacking Kv1.1. However, we also found that DTX-K administration in WT SAN cells caused moderate slowing of diastolic depolarization without changing the MDP. Therefore, it is possible that Kv1.1 influences DDR, but these effects may be attenuated in Kcna1^−/− SAN myocytes due to compensatory remodeling. Consequently, we cannot exclude the possibility that Kv1.1 inhibition has some minor effects on diastolic depolarization, potentially leading to a reduction in the SAN firing rate. However, we propose that APD prolongation represents the major underlying mechanism.

Although our findings show that the absence of Kv1.1 slows ex vivo and in vitro SAN firing rates, our previous studies have repeatedly shown that Kcna1^−/− mice exhibit relatively normal HR in vivo (Glasscock et al., 2010, 2015; Mishra et al., 2017; Trosclair et al., 2021). Several potential mechanisms could account for the normal HR in Kcna1^−/− mice in vivo, including compensation by neuro-humoral influences, remodeling, or some other factors (MacDonald et al., 2020; Al Kury et al., 2022). In support of a possible autonomic compensatory mechanism, Kcna1^−/− mice exhibit a larger absolute decrease in HR than WT mice in response to the β-adrenergic blocker propranolol (Moore et al., 2014). This augmented HR response to sympathetic blockade in Kcna1^−/− mice suggests a potential increase in resting sympathetic tone that could compensate for the intrinsic pacemaking deficits in Kcna1^−/− mice, providing an explanation for the absence of significant in vivo HR differences in our previous work.

Our study highlights an emerging principle in cardiac pacemaking physiology, which is that even small currents can exert significant effects on heart function. In voltage-clamp recordings, WT myocytes exhibited a small (∼1.5 pA/pF at +50 mV) but significant DTX-K-sensitive outward current, which we termed I_Kv1.1, representing the contribution of Kv1.1-containing channels. Although previous studies have shown expression of Kv1.1 mRNA in the SAN (Brahmajothi et al., 1996, 2010; Glasscock et al., 2010; Leoni et al., 2005; Marionneau et al., 2005), our immunocytochemical studies indicate that the levels of Kv1.1 protein are likely low in WT mouse SAN myocytes, suggesting Kv1.1 has low abundance in the SAN. Regardless, inhibiting Kv1.1 still led to significant slowing of cardiac pacemaking at the organ, tissue, and myocyte levels, underscoring that even minor currents can have major effects.

Similar observations have been noted for BK and HCN4 channels, which both exhibit large effects on pacemaking relative to their current sizes. Like Kv1.1, BK channels are expressed at very low levels in mouse SAN myocytes and require immunohistochemical amplification for detection (Lai et al., 2014). In addition, BK channels mediate only very small outward currents of a magnitude similar to Kv1.1, yet their absence causes greatly reduced SAN firing rates (Lai et al., 2014). HCN4 channels, despite being highly expressed in the SAN, activate mostly outside the physiological voltage range of SAN cells and exhibit slow kinetics making their precise contribution to pacemaking unclear (Marionneau et al., 2005; Lakatta and DiFrancesco, 2009). However, recent investigations into the basis of the funny current (I_f) reveal that their persistent activity allows them to conduct a large fraction of the total charge movement in SAN cells despite small peak current amplitudes and low open probabilities (Peters et al., 2021). Thus, in cardiac pacemaking, the size of the current and the abundance of the channel do not necessarily scale with the magnitude of the observed effect; small currents can yield substantial effects.

An important unresolved question regarding Kv1.1 is its potential contribution to SAN function in humans. While Kv1.1 has been detected in human atria and ventricles (Glasscock et al., 2015; Trosclair et al., 2021), its presence in human SAN remains unknown partly due to the omission of KCNA1 from previous ion channel expression studies (Chandler et al., 2009). Comparative analysis of gene expression studies in mouse and human SAN indicates similarities in the mRNA levels of some ion channels across species, but genes responsible for outward K⁺ currents often exhibit lower expression levels in humans (Marionneau et al., 2005; Chandler et al., 2009). However, despite differences in heart rates, transcriptome analyses reveal that humans and mice still share a conserved basic gene program underlying SAN function, supporting the use of mice as a model for cardiac pacemaking mechanisms (van Eif et al., 2019). In addition, kinetic parameter studies reveal a fundamental similarity in SAN automaticity across species, scaling with body mass and target heart rate (Tagirova Sirenko et al., 2021). Nonetheless, further research will be essential to determine if I_Kv1.1 is present in human SAN myocytes and to what extent it contributes to pacemaking function.

In summary, our study used genetic and pharmacological approaches to establish Kv1.1 as a key regulator of cardiac pacemaking, primarily influencing action potential repolarization. Notably, Kv1.1 stands out as the first functional contributor to SAN function from the Kv1 channel family. Despite its apparently low protein levels and small current amplitudes, Kv1.1 significantly impacts SAN firing rate, underscoring the principle that even small currents can wield substantial effects on the heart. The implications of our findings suggest that KCNA1 should be considered as a potential candidate gene underlying sinus node dysfunction in humans. Furthermore, Kv1.1 may be a viable therapeutic target for addressing sinus rhythm disturbances.

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Jeanne M. Nerbonne served as the editor.

We are grateful to Cathy Proenza at the University of Colorado School of Medicine for providing technical training and assistance with patch-clamp recordings and to Raanju Sundararajan at Southern Methodist University for providing statistical assistance.

This work was supported by the National Institutes of Health (grants R01NS100954, R01NS099188, R01NS129643).

Author contributions: M. Si: Formal analysis, Investigation, Writing—review & editing, A. Darvish: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Resources, Validation, Visualization, Writing—original draft, Writing—review & editing, K. Paulhus: Investigation, Writing—review & editing, P. Kumar: Data curation, Formal analysis, Investigation, Methodology, Validation, K.A. Hamilton: Investigation, Resources, Writing—review & editing, E. Glasscock: Conceptualization, Project administration, Supervision, Writing—original draft, Writing—review & editing.