Cholecystokinin-A signaling regulates automaticity of pacemaker cardiomyocytes

Aims: The behavior of pacemaker cardiomyocytes (PCs) in the sinoatrial node (SAN) is modulated by neurohormonal and paracrine factors, many of which signal through G-protein coupled receptors (GPCRs). The aims of the present study are to catalog GPCRs that are differentially expressed in the mammalian SAN and to define the acute physiological consequences of activating the cholecystokinin-A signaling system in isolated PCs. Methods and results: Using bulk and single cell RNA sequencing datasets, we identify a set of GPCRs that are differentially expressed between SAN and right atrial tissue, including several whose roles in PCs and in the SAN have not been thoroughly characterized. Focusing on one such GPCR, Cholecystokinin-A receptor (CCKAR), we demonstrate expression of Cckar mRNA specifically in mouse PCs, and further demonstrate that subsets of SAN fibroblasts and neurons within the cardiac intrinsic nervous system express cholecystokinin, the ligand for CCKAR. Using mouse models, we find that while baseline SAN function is not dramatically affected by loss of CCKAR, the firing rate of individual PCs is slowed by exposure to sulfated cholecystokinin-8 (sCCK-8), the high affinity ligand for CCKAR. The effect of sCCK-8 on firing rate is mediated by reduction in the rate of spontaneous phase 4 depolarization of PCs and is mitigated by activation of beta-adrenergic signaling. Conclusion: (1) PCs express many GPCRs whose specific roles in SAN function have not been characterized, (2) Activation of the cholecystokinin-A signaling pathway regulates PC automaticity.

differences between heartbeats (RMSSD), and (F) the percentage of successive RR intervals that differ by more than 6 msec (pNN6).Frequency domain analysis was used to calculate the contributions of (G) very low frequency (VLF), (H) low frequency (LF), (I) high frequency (HF) components to (J) the total power.The Mann-Whitney test was used for statistical comparison for all parameters and 'ns' denotes not significant.

Animal Handling and Maintenance
Mouse studies were performed in accordance with IACUC-approved protocols at the University of California, San Francisco.Mouse lines were maintained on C57Bl6.Cckar tm1Kpn (hereafter, Cckar -/-) mice were generated by Alan Kopin et al. and were obtained from Nirao Shah's lab 2 .Cck tm1.1(cre)Zjh (hereafter, Cck Cre ) mice were obtained from Jackson Labs.Gt(ROSA)26Sor tm4(ACTB-tdTomato,-EGFP)Luo (hereafter, ROSA mT/mG mice were generated by Muzumdar et al. and were obtained from Brian Black's lab 3 .

Analysis of Bulk RNA Sequencing Data
To identify G-protein coupled receptors (GPCRs) that were expressed in pacemaker cardiomyocytes (PCs) and right atrial cardiomyocytes (RACMs), respectively, RNA-seq data from neonatal sorted mouse PCs and RACMs were downloaded, and the list of genes expressed in each tissue as defined by average transcripts per million across 3 biological replicates greater than 3 was intersected with an annotated database of from the International Union of Basic and Clinical Pharmacology.For each GPCR identified, a false discovery rate was calculated for differential expression was determined using the method of Benjamini-Hochberg.

Single Cell RNA Sequencing Analysis
Mouse E16.5 SAN s 10x Genomics Mouse single-cell RNA-sequencing files were generated by Goodyer et al. 4 and downloaded from GEO (GSE132658).Adult rabbit single-cell RNA-seq data were generated by Liang et al. 1 and were downloaded as fastq files from the NCBI Sequence Read Archive (PRJNA531288).Fastq files from the rabbit dataset were trimmed using Fastp 5 and aligned to the oryCun2 assembly (UCSC genome browser) using Hisat2 6 .Finally, an expression matrix was constructed for each cell using featureCounts 7 .Both mouse and rabbit datasets were filtered, clustered, and analyzed using Scanpy 8 to identify the cell types expressing both Cckar and Cck.

Quantitative PCR
The SAN and RA regions from 3 adult hearts were manually dissected using anatomical landmarks to delineate the inter-caval region.For each sample, total RNA from SAN and right atrium tissue was extracted using TRIzol reagent (Life technologies, CA, USA) and RNeasy micro kit (QIAGEN)   according to the manufacturer's instructions.0.2 micrograms of total RNA were used to synthesize the first-strand cDNA with Superscript III first-strand synthesis kit (Invitrogen).Real time qPCR was performed using Taqman fast advanced master mix (Thermo Fisher Scientific, USA).GAPDH was used as an endogenous control with the CT method for comparison of SAN and right atrial tissue.A Mann-Whitney test was used for significance with P < 0.05 deemed significant.

Transmitter Implantation and Analysis of Heart Rhythm Data
Adult mice (8-12 weeks old) underwent subcutaneous implantation of telemetry devices (ETA-F10, Data Science International, MN) under anesthesia according to the manufacturer's instructions.
After at least 7 days of recovery, continuous ECG signals were sampled using Ponemah software (DSI) at 1.0 kHz on freely moving and conscious mice for 48 hours per mouse.Maximum heart rate was ascertained after injection of 1.0 g of isoproterenol.To determine the intrinsic heart rate, a separate cohort of mice were injected with 3.0 mg/kg atropine and 10.0 mg/kg propranolol intraperitoneally under anesthesia with 1% isoflurane.Anesthetized ECG signals were recorded with a Power Lab and Animal BioAmp (AD Instruments) and analyzed offline using ChartPro (AD Instruments).
Heart rhythm analysis on data from implanted transmitters was performed offline using Ponemah (Data Sciences International, St Paul, MN).48 hours of heart rhythm data were analyzed for each mouse, with determination of average heart rate and PR interval.To generate heart rate histograms, two 12-hour segments corresponding to low activity (8AM-8PM) and two 12-hour segments corresponding to high activity (8PM-8AM) were analyzed for each mouse.The average heart rate for each minute over the 48hour period was sorted into 10-bpm bins, with the data from low and high activity phases plotted and analyzed separately.To measure the intrinsic heart rate, a 5-minute period of stable heart rate after injection of atropine and propranolol was averaged for each mouse tested.Similarly, maximum heart was

Supplementary Figure 3 .
Heart Rate Variability in WT and Cckar -/-Mice.Poincare plots are shown for a representative WT (A) and (B) Cckar -/-littermate during low activity conditions.The following parameters were determined from EKG tracings acquired during normal rhythm from implanted telemetry devices in freely moving unanesthetized WT (n = 7) and Cckar -/-(n = 9) littermates: (C) RR interval (D) standard deviation of the RR interval (SDNN), (E) Root mean square of successive