cAMP-dependent regulation of HCN4 controls the tonic entrainment process in sinoatrial node pacemaker cells

Figures

• 

  Fig. 1. Generation of HCN4FEA mice and basic characterisation of HCN4FEA-mediated If. a Illustration depicting the HCN4 channel tetramer (inset) and single subunit. The C-terminus contains the C-linker composed of six α-helices (A’–F’) and a cyclic nucleotide-binding domain (autonomic) composed of three α-helices (A–C) and a β-roll between helix A and B. Localisation of the three introduced point mutations is indicated with coloured circles. b C-terminus of the human HCN4 channel (PDB #Q9Y3Q4). Residues Y527 (purple), R669 (orange), and T670 (pink) are highlighted. The R669E and T670A mutations were introduced to completely disrupt cAMP binding. Both residues are located in the loop between the P helix and the β7 sheet of the CNBD and are highly conserved throughout the CNG (cyclic nucleotide-gated) and HCN channel family7,53. c cAMP bound to the CNBD of human HCN4. Residues R669 and T670 are required for cAMP binding. d Activation curves of WT HCN4 and HCN4FEA channels transiently expressed in HEK293 cells in the absence and presence of 100 µM cAMP in the pipette solution. A holding potential of −40 mV was used. cAMP shifts the activation curve of WT HCN4 to more depolarised potentials,…

• 

  Fig. 2. Similar expression and distribution of HCN channels in the WT and HCN4FEA heart. a Gelatin-inflated heart preparation depicting the anatomy of the SAN region (n = 5 biologically independent samples). b Hematoxylin and eosin (H&E, n = 6 WT + 6 HCN4FEA biologically independent samples) and fibrosis stainings (Sirius red/ Fast green, n = 7 WT + 7 HCN4FEA biologically independent samples) of ventricular cross sections. c Representative long-axis echocardiograms (n = 12 WT + 11 HCN4FEA animals). d Sirius red/Fast green stained transverse sections of the central SAN (SAN head, n = 7 WT + 7 HCN4FEA biologically independent samples). e Quantification of fibrosis in the SAN head (n = 7 WT + 7 HCN4FEA biologically independent samples), body (n = 7 WT + 7 HCN4FEA biologically independent samples), and tail (n = 7 WT + 6 HCN4FEA biologically independent samples). f HCN channel expression in SAN, atria (AT), and left ventricle (LV) of WT and HCN4FEA (KI) mice (one blot with pooled tissue from 6 WT + 6 HCN4FEA animals). Tubulin: loading control. g Consecutive sections of the central SAN shown in d demonstrating the transmural distribution of HCN4 (green) and HCN1 (red) in WT and HCN4FEA (n = 3…

• 

  Fig. 3. HCN4FEA mice display bradycardia and sinus dysrhythmia. a, b Telemetric ECG traces obtained from a conscious, freely moving a WT mouse and b HCN4FEA mouse. Insets: Extended ECG traces. c Mean, minimum, and maximum HR determined from 72-h RR intervals (WT: n = 9, HCN4FEA: n = 11; two-sided t-test; mean HR: p = 0.0019, min HR: p = 0.0040, max HR: p = 0.0383). d HR histograms of a WT (light gray) and HCN4FEA mouse (orange) determined from 72-h (12 h/12 h light/dark cycle) ECGs. e Determination of sinoatrial conduction time) with optical imaging measurements. Activation map of a biatrial WT SAN preparation (upper panel) and optical action potentials (OAP) (lower panel) from the leading pacemaker site in the SAN (dark blue) and from the earliest atrial excitation site within the crista terminalis (CT; light blue). SACT is determined as the time difference of activation time points indicated by asterisk. f SACT in WT and HCN4FEA SAN preparations (n = 7 WT + 8 HCN4FEA biologically independent samples; two-sided t-test; p = 0.0029). g Comet-shaped Poincaré plots in HCN4FEA mice (orange). h Frequency domain power spectral density plots obtained from a WT (black) and HCN4FEA (orange)…

• 

  Fig. 4. Long-lasting periods of nonfiring in isolated pacemaker cells of HCN4FEA mice. a–c (right, upper panel) Cartoons of channel variant used in experiments shown in the right panel. (Left panel) long-term perforated-patch-clamp recordings of spontaneous action potentials under baseline conditions in six SAN cells isolated from a WT and b HCN4FEA mice and c WT cells after application of TAT-TRIP8bnano. (Right, lower panel) Magnification of action potential recordings shown in the left panels. d Averaged action potentials obtained from representative perforated-patch-clamp recordings of a WT and HCN4FEA pacemaker cell. e–g firing rate (WT before Iso vs WT Iso: p = 0.000004; HCN4FEA before Iso vs HCN4FEA Iso: p = 0.0035; WT Iso vs HCN4FEA Iso: p = 0.0386), slope of slow diastolic depolarisation (SDD) (WT before Iso vs WT Iso: p = 0.00000003; HCN4FEA before Iso vs HCN4FEA Iso: p = 0.00002; WT Iso vs HCN4FEA Iso: p = 0.4315), and maximum diastolic potential (MDP) under basal conditions (n = 12 WT + 10 HCN4FEA cells) as well as before (n = 10 WT + 7 HCN4FEA cells) and during (n = 10 WT + 7 HCN4FEA cells) application of isoproterenol (100 nM). Values were determined from episodes with…

• 

  Fig. 5. Carbachol induces episodic firing patterns with nonfiring states in both WT and HCN4FEA SAN cells. a, b Representative current-clamp recordings obtained in perforated-patch-clamp mode from a WT and b HCN4FEA cells. Increasing concentrations of carbachol consecutively reduced the firing frequency and caused an episodic firing pattern with nonfiring states upon successive hyperpolarisation. c Cumulative carbachol dose response of WT and HCN4FEA cells (n = 7 WT + 10 HCN4FEA cells; data are presented as mean values ± SEM). d Percentage reduction was calculated by normalising each value to the basal firing rate (n = 7 WT + 10 HCN4FEA cells; data are presented as mean values ± SEM). e Maximum diastolic potential after application of 1000 nM Carbachol (n = 7 WT + 10 HCN4FEA cells). All data were obtained from SAN cells of male WT (black) and HCN4FEA (orange) mice. Boxplots show the median line, perc 25/75, and min/max value; open symbols represent the mean value. Source data are provided as a Source Data file.

• 

  Fig. 6. Nonfiring pacemaker cells can be identified in the SAN network. a Time-lapse of confocal calcium recordings (Fluo-4 AM; fluorescence in arbitrary units [a.u.]) of intact SAN explants of WT, HCN4FEA, and WT after application of TAT-TRIP8bnano (n = 7 WT + 8 HCN4FEA + 4 WT + TRIP8b biologically independent samples). Red arrows indicate subthreshold calcium signals. Images were taken at time points 1–4 during WT calcium transient shown in b. c Number of cells per µm2 displaying subthreshold calcium signals determined from the entire SAN region (n = 7 WT + 8 HCN4FEA + 4 WT + TRIP8b biologically independent samples; Mann–Whitney U test; WT vs HCN4FEA: p = 0.0330; WT vs WT + TRIP8b: p = 0.0040). d (upper panel) Time-lapse of confocal calcium recordings of intact SAN explants of WT, HCN4FEA, and WT after application of TAT-TRIP8bnano (n = 7 WT + 8 HCN4FEA + 4 WT + TRIP8b biologically independent samples). Images were taken at time points 1–4 indicated in the corresponding calcium transients (lower panel). Ca2+ transients were determined from indicated regions of interest (ROI). All experiments were performed using tissue isolated from female animals. Boxplots show the median line, perc…

• 

  Fig. 7. Vagal nerve stimulation in vitro and spontaneous vagal nerve activity in vivo induce overshooting HR responses. a Experimental setup consisting of a Langendorff-perfused heart preparation. VN vagal nerve, SE stimulating electrode, RE ECG recording electrodes. b, c RR tachograms of b WT and c HCN4FEA ECGs before, during (dotted lines), and after vagal nerve stimulation (VNS). d Mean number of sinus pauses in HCN4FEA hearts at different stimulation frequencies during VNS (3 Hz n = 1, 5 Hz n = 2, 10 Hz n = 5, 20 Hz n = 5, 30 Hz n = 6 biologically independent samples). Sinus pauses were not observed in WT hearts. e Activation map of a HCN4FEA biatrial preparation (top) before and (bottom) during VNS. f, g Position of the leading pacemaker in WT and HCN4FEA preparations before and during VNS. IAS: intra-atrial septum; IVC: inferior vena cava. h, i Combined telemetric blood pressure (BP) and ECG recordings. h (top) Arterial BP trace of a WT mouse during a short sequence of three consecutive beats with increasing systolic BP (up sequence; arrow). h (middle) Corresponding RR intervals. h (bottom) Plot of systolic BP (SBP) and corresponding RR intervals for a WT and HCN4FEA mouse (see…

• 

  Fig. 8. Junctional escape rhythm due to chronotropic incompetence of the HCN4FEA SAN. a Telemetric ECG traces from a male HCN4FEA mouse with alternating episodes of sinus rhythm and junctional escape rhythm (JER) rhythm (left) and corresponding RR tachogram (right, top). Inset: During escape rhythm the P wave (red) occurs after the QRS complex. b Number of JER episodes during 3 h in low activity phase (n = 11 HCN4FEA animals). c Optical action potentials (OAP) and surface ECGs recorded from a WT (left) and HCN4FEA (right) Langendorff-perfused heart explanted from female mice. The HCN4FEA heart displays JER. The ventricular signal precedes the atrial signal in beat #3–5. d Activation maps determined from the OAP measurements shown in c. Right, activation maps for beats #1, and 3–5 were similar. Activation map for beat #2: anterograde activation of ventricles and retrograde activation of atria from the junction. e (upper panel) Conduction times determined from d. Numbers indicate atrial (top), atrioventricular (middle), and ventricular (bottom) conduction times in ms. Right, Beat 1: ventricles were activated 10 ms after atrial activation. Beat 3–5: ventricles were activated by a subsidiary…

• 

  Fig. 9. Isorhythmic AV dissociation (IAVD) due to chronotropic incompetence of the HCN4FEA SAN. a, b Telemetric ECG trace (top) and corresponding arterial blood pressure recordings (bottom) from a male HCN4FEA mouse during IAVD. Characteristic flirtatious P waves indicate the presence of two independent pacemakers (top and inset). Concomitantly with the successive decrease in PR intervals, blood pressure dropped (bottom). b (top and inset) The P wave is hidden within the QRS complex, located behind or before the QRS complex. Blood pressure starts to normalise when PR normalises (bottom). c Number of IAVD episodes during 3 h in low activity phase (n = 11 HCN4FEA animals). d OAPs and surface ECG recorded from a female HCN4FEA whole-heart preparation during IAVD. e (left) Activation maps of beat #1 determined from the recordings presented in d showing IAVD ex vivo. Activation maps for beats 1, 2, 3, and 5 were similar. Right: activation map of beat #4. f Ladder diagrams for beats #1–5. Beat #1 and 3: normal activation arising from SAN with regular and constant PR; beat #2, 4, and 5: IAVD with shortened or negative PR indicating that the atria are activated by SAN and ventricles by a…

• 

  Fig. 10. CDR of HCN4 regulates the firing mode of SAN pacemaker cells. a (top) Alternating firing and nonfiring episodes in SAN pacemaker cells. During firing cells hyperpolarise and then depolarise during nonfiring. a (bottom) Expanded view of nonfiring, slowly firing, and fast firing modes from marked sequences. b Illustration of the HCN4 channel. Black ramp indicates cAMP concentration close to the cyclic nucleotide-binding domain (CNBD) which increases from left to right during sympathetic activity. Increased binding of cAMP to the CNBD of HCN4 channels successively switches the activity mode of pacemaker cells from nonfiring (left) to slow firing (middle). Increasing activity of the VN reduces cAMP levels and favours nonfiring. CDR of HCN4 antagonises the response to VN activity and acts synergistically with sympathetic nerve activity. c Tonic entrainment process in the SAN network. Pacemaker cells are connected via gap junctions. Nonfiring cells (gray) are more hyperpolarised and draw tonic currents from neighbouring firing cells, which slightly hyperpolarises neighbouring firing cells and depolarises nonfiring cells. Nonfiring cells either remain nonfiring or begin to fire slowly.…