Tirzepatide Regulates Pacemaker Function by Modulating cAMP and Calcium Dynamics in Human Sinoatrial Node Cells

Dual glucagon-like peptide-1 receptor (GLP1R) and glucose-dependent insulinotropic polypeptide receptor (GIPR) agonists, have revolutionized type 2 diabetes mellitus (T2DM) clinical management¹. However, their administration has been linked to an increase in resting heart rate (HR)¹, raising cardiovascular concerns in patients with prevalent comorbidities such as heart failure (HF). Conversely, these agents may offer therapeutic benefits for conditions characterized by pathological bradycardia, including sick sinus syndrome. Understanding the mechanisms by which these drugs modulate HR is therefore essential. To bridge the knowledge gap, we investigated the dual GLP1R/GIPR agonist tirzepatide at the cellular and 3D level using human induced pluripotent stem cell-derived sinoatrial node (iPSC-SAN) and cardiac organoid (SAN-CO) (Figure A).

Figure. Tirzepatide enhances human iPSC-derived sinoatrial node (iPSC-SAN) cells and organoid pacemaker activity through cAMP-mediated calcium cycling.

(A) Schematic overview of the experimental strategy, illustrating the use of iPSC-SAN cells and 3D organoids to model human SAN pacemaker activity, test pharmacological interventions, and determine the mechanism of tirzepatide (tirz)-induced heart rate modulation. Where applicable, numbers printed within or adjacent to each bar indicate the sample size (n) for that group. (B) RT-qPCR data analysis demonstrating the expression of canonical SAN markers short stature homeobox 2 (SHOX2), ISL LIM homeobox 1 (ISL1), and T-box transcription factor 18 (TBX18) in iPSC-SAN cells relative to ventricular myocytes (VM). Data are presented as mean ± SEM (n = 4 independent differentiations from 2 iPSC lines, for both iPSC-SAN and VM groups), analyzed using non-parametric statistics (Mann-Whitney U). (C) Representative traces and summary data of multielectrode array (MEA) recordings showing spontaneous beating rate and corrected field potential duration (FPDc) of SAN-COs and VM-COs. Data are presented as mean ± SEM (n = 10–13 wells), analyzed using non-parametric statistics (Mann-Whitney U). (D-E) MEA at baseline and after exposure to dobutamine (100 nM) and ivabradine (10 nM and 100 nM) were recorded for functional validation of SAN organoids. Dobutamine increased beating rate, whereas ivabradine decreased beating rate after 10 min of treatment. To assess the effects of tirzepatide on HR modulation, baseline MEA recordings were acquired, followed by tirzepatide application for 15 min, with or without 24-h pretreatment of GLP-1 receptor (GLP1R) antagonist (Exendin(9–39) amide, 1 μM) and GIP receptor (GIPR) antagonist (human GIP (3–30) amide, 1 μM). Tirzepatide (100 nM) significantly increased beating rate, whereas combined GLP-1R and GIPR antagonism blocked tirzepatide’s chronotropic effect. In separate acute experiments, subsequent addition of propranolol reduced beating rate in the presence of tirzepatide in a concentration-dependent manner. Data are presented as mean ± SEM (n = 6–8 wells) and were analyzed using non-parametric statistics (Kruskal–Wallis test with Dunn’s multiple-comparisons post hoc test). (F–G), GLP1R and GIPR are preferentially expressed in human atrial myocytes assessed from published transcriptomic data (PMID: 32971526) of the adult human atrial tissues, as depicted by the UMAP and the bubble plot. CM: cardiomyocytes. (H) Immunofluorescence staining of iPSC-SAN cells showing SHOX2 (green) and DAPI (blue) (left), and co-localization of phospholamban (PLN; green) with GLP-1R (red, upper) or GIPR (red, lower) (right). (I) Fluorescence resonance energy transfer (FRET)-based cAMP biosensor assays using CUTie biosensors detecting compartmentalized cAMP changes following tirzepatide treatment. Representative images of iPSC-SAN cells expressing CUTie sensors localized to the plasma membrane (PM, AKAP79-targeted), and sarcoplasmic reticulum (SR, AKAP18δ-targeted) and cytosol. (J) Time course of normalized FRET responses (R/R₀) in iPSC-SAN cells expressing the CUTie sensors localized to SR, PM, and cytosol after the exposure to 100 nM tirzepatide or tirzepatide in the presence of combined GLP-1R/GIPR antagonism (Exendin(9–39) amide + human GIP (3–30) amide, 1 μM each). Bar graph shows summary data for the maximal increase in the FRET ratio response to treatments. Data are presented as mean ± SEM (n = 5–8 cells). Statistical comparisons were performed using two-way ANOVA followed by Holm–Šídák multiple-comparisons correction. (K) Proximity ligation assay (PLA) demonstrating co-localization of GLP1R with PLN; and GIPR with PLN in SR nanodomains of iPSC-SAN cells; and negative control. Negative control was used with GLP1R, GIPR or PLN antibody alone. Data are presented as mean ± SEM (n = 6–16 cells) and were analyzed using non-parametric statistics (Kruskal–Wallis test with Dunn’s multiple-comparisons post hoc test). (L) Representative confocal line-scan images of iPSC-SAN cells treated with vehicle or 100 nM tirzepatide. Superplot analysis shows increased beating rate and shortened Ca²⁺ decay time constant following tirzepatide treatment. Each color denotes an independent differentiation (biological replicate). Small dots represent individual cell measurements (n = 19–20 cells); large dots represent the mean of each biological replicate (n = 3 independent differentiations per group). Data are presented as mean ± SEM. Given the hierarchical structure of the data (cells nested within biological replicates), statistical comparisons were performed using a linear mixed-effects model (LMM) with treatment as a fixed effect and biological replicate as a random effect, estimated using restricted maximum likelihood (REML). (M) Western blot analysis of PLN phosphorylation at Ser16 in vehicle and tirzepatide-treated iPSC-SAN cells. Cells were treated with 100 nM tirzepatide or vehicle for 15 min before harvesting for western blot analysis. Representative blots show phosphorylated PLN (Ser16) under control conditions, tirzepatide, and tirzepatide with combined GLP-1R/GIPR blockade (Exendin(9–39) + GIP(3–30)). Representative blots for GAPDH (~37 kDa), total Phospholamban (total PLN, ~10 kDa), and phosphorylated Phospholamban (p-PLN, ~10 kDa). GAPDH served as the loading control. Quantification of p-PLN levels normalized to GAPDH or total PLN across treatment groups. Data are presented as mean ± SEM (vehicle: n = 11, tirzepatide: n = 11, tirzepatide + GLP-1R/GIPR blockade: n = 4, independent differentiations) and were analyzed using non-parametric statistics (Kruskal–Wallis test with Dunn’s multiple-comparisons post hoc test). (N) Enriched pathways that were upregulated after GLP-1 stimulation, analyzed from the phosphoproteomics dataset of the porcine SAN following GLP-1 treatment (PMID: 38832935). (O) Phosphorylation levels of proteins with significant changes, particularly those involved in calcium homeostasis. Data were obtained from a published phosphoproteomics dataset⁵ and reanalyzed to highlight GLP1-response (PMID: 38832935). (P) Schematic showing the hypothesized pathway through which tirzepatide acts on pacemaker cells to elevate heart rate.

Differentiated iPSC-SAN cells were generated using our established protocols² and confirmed by canonical SAN markers (Shox2, Isl1 and Tbx18) (Figure B). iPSC-SAN cells were combined with iPSC-cardiac fibroblasts (iPSC-CFs; 9:1 ratio) to generate SAN-CO, with spontaneous beating rates of ~100 bpm, significantly higher than ventricular cardiomyocyte COs (VM-CO) (Figure C). SAN-COs were used to evaluate the chronotropic effects of tirzepatide and pharmacological modulators. Functional integrity was confirmed by expected responses to dobutamine and ivabradine (Figure D–E). Tirzepatide increased SAN-CO spontaneous beating rate by 8.59 ± 1.88% at 100 nM, consistent with the clinical HR increase. Co-administration of the GLP-1R antagonist (Exendin(9–39) amide) and the GIPR antagonist (human GIP (3–30) amide) (24 h pretreatment) attenuated the tirzepatide-induced increase in beating rate, supporting a direct role for GLP1R and GIPR signaling and SAN activity. In contrast, acute propranolol treatment blunted tirzepatide-induced chronotropic effects, indicating sensitivity to β-adrenergic blockade.

To explore the molecular basis of the effect, we examined the expression of GLP1R and GIPR in SAN cells. Public transcriptomic datasets³ confirmed preferential GLP1R and GIPR expression in adult human atrial cardiomyocytes (Figures F–G). Immunofluorescence confirmed GLP1R and GIPR expression in iPSC-SANCs alongside the SAN marker, Shox2 (Figure H).

To elucidate the molecular mechanism underlying our functional findings, we used iPSC-SANCs for cell type-specific interrogation of signaling pathways. As GLP1R and GIPR are G_αs- coupled receptors, tirzepatide may increase cyclic adenosine monophosphate (cAMP) to mediate its positive chronotropic effects. We performed fluorescence resonance energy transfer (FRET) experiments using the CUTie biosensor⁴ targeted to plasma membrane (PM), sarcoplasmic reticulum (SR), and cytosol (Figure I). FRET signal at the PM and SR significantly increased upon the application of 100 nM tirzepatide, indicating a rise in cAMP at these compartments (Figure J). Notably, the FRET change at the SR was significantly more pronounced than at the PM, suggesting a key role for tirzepatide-induced SR cAMP signaling in driving the increase in HR observed with tirzepatide. To confirm receptor specificity and exclude non-specific effects, GLP1R/GIPR antagonists blocked the tirzepatide-induced FRET response, demonstrating that the compartmentalized cAMP elevation is dependent on dual incretin receptor activation.

To investigate the spatial organization of GLP1R and GIPR relative to the SR, we next performed a proximity ligation assay (PLA) using phospholamban (PLN) as an SR marker. Both PLA (Figure K) and high-resolution confocal imaging (Figure H) demonstrated spatial co-localization of GLP1R with PLN and GIPR with PLN in iPSC-SAN cells, suggesting potential crosstalk between these receptors and SR calcium signaling. To assess whether the localized cAMP surge and receptor proximity to PLN affected calcium dynamics, we measured calcium transients in SAN cells treated with vehicle or tirzepatide (Figure L). Tirzepatide increased beating rate, and shortened decay tau, suggesting faster Ca²⁺ cycling.

Since phosphorylated PLN loses its inhibitory effects, enhancing calcium reuptake into the SR, we then assessed PLN phosphorylation level using western blot analysis (Figure M). Tirzepatide increased the phosphorylation of PLN at Serine-16, a protein kinase A (PKA) phosphorylation site, which aligns with prior reports on GLP1-mediated modulation of Ca²⁺ handling in the porcine SAN⁵. We also found GLP-1R/GIPR inhibition produced a substantial suppression of tirzepatide-induced PLN Ser16 phosphorylation. These findings indicate that tirzepatide-induced enhancement of SR Ca²⁺ handling and SAN automaticity is primarily mediated through coordinated GLP1R and GIPR signaling.

Moreover, we reanalyzed a phosphoproteomics dataset⁵ and identified upregulated pathways and proteins involved in Ca²⁺ homeostasis following GLP1 treatment (Figures N–O). Enhanced Ca²⁺ cycling at systolic and diastolic phases (RYR2, PLN, ATP2A2), suggested a coordinated remodeling of Ca²⁺ handling underlying increased SAN automaticity.

In summary, we demonstrated that tirzepatide, a dual GLP1R/GIPR agonist, acts directly on iPSC-SAN cells by elevating cAMP levels at SR nanodomains (Figure P). The localized surge in SR cAMP activates PKA, leading to phosphorylation of PLN at Serine-16, which enhances Ca²⁺ cycling and increases beating frequency at the iPSC-SAN cellular and 3D organoid levels. Our findings show that tirzepatide directly impacts SAN pacemaker function by modulating intracellular receptor signaling and Ca²⁺ dynamics. Clinically, this mechanism provides a critical molecular explanation for the positive chronotropic effects observed with dual GLP1R/GIPR agonists, which may help optimize GLP1R/GIPR therapy while minimizing potential cardiovascular risks in vulnerable patient populations.

All experiments were performed using previously established, de-identified human induced pluripotent stem cell lines, the use of which does not constitute human subjects research. The data supporting the findings of this study are available from the corresponding author upon reasonable request. Study materials used in this study are available from the corresponding author upon reasonable request and subject to material transfer agreements.

Non-standard Abbreviations and Acronyms

cAMP

cyclic adenosine monophosphate

FRET

fluorescence resonance energy transfer

GIPR

glucose-dependent insulinotropic polypeptide receptor

GLP-1

glucagon-like peptide-1

GLP1R

glucagon-like peptide-1 receptor

HF

heart failure

HR

heart rate

iPSC

induced pluripotent stem cell

iPSC-CF

induced pluripotent stem cell-derived cardiac fibroblast

iPSC-SAN

induced pluripotent stem cell-derived sinoatrial node

PKA

protein kinase A

PLA

proximity ligation assay

PLN

phospholamban

PM

plasma membrane

SAN

sinoatrial node

SAN-CO

sinoatrial node cardiac organoid

SR

sarcoplasmic reticulum

T2DM

type 2 diabetes mellitus

VM-CO

ventricular myocyte cardiac organoid