A single cell transcriptional roadmap of human pacemaker cell differentiation

Abstract

Each heartbeat is triggered by the sinoatrial node (SAN), the primary pacemaker of the heart. Studies in animal models have revealed that pacemaker cells share a common progenitor with the (pro)epicardium, and that the pacemaker cardiomyocytes further diversify into ‘transitional’, ‘tail’, and ‘head’ subtypes. However, the underlying molecular mechanisms, especially of human pacemaker cell development, are poorly understood. Here, we performed single cell RNA sequencing (scRNA-seq) and trajectory inference on human induced pluripotent stem cells (hiPSCs) differentiating to SAN-like cardiomyocytes (SANCMs) to construct a roadmap of transcriptional changes and lineage decisions. In differentiated SANCM, we identified distinct clusters that closely resemble different subpopulations of the in vivo SAN. Moreover, the presence of a side population of proepicardial cells suggested their shared ontogeny with SANCM, as also reported in vivo. Our results demonstrate that the divergence of SANCM and proepicardial lineages is determined by WNT signaling. Furthermore, we uncovered roles for TGFβ and WNT signaling in the branching of transitional and head SANCM subtypes, respectively. These findings provide new insights into the molecular processes involved in human pacemaker cell differentiation, opening new avenues for complex disease modeling in vitro and inform approaches for cell therapy-based regeneration of the SAN.

Retinoic acid signaling modulation guides in vitro specification of human heart field-specific progenitor pools

“…8c). We also identified a small cluster 14 which was enriched in CPNE5 and IGFBP5 transcripts, and could correspond to a transitional population of nodal CMs occurring during in vitro differentiation of hiPSC 29,30 (Fig. 5d, Supplementary Fig.…”

“…Cells then were subjected to modified cardiac differentiation protocol described above. Modulations of BMP, RA and WNT signaling cues were described previously to favor either myocardial or epicardial lineages 29 and thus we decided to use low levels of WNT signaling inhibition with concomitant addition of BMP4 and RA. Briefly, cells were re-plated in CDM-BSA media supplemented with 20 ng/ml BMP4, 8 ng/ml bFGF, 10 μg/ml insulin, 2.5 μM IWP2, 1 μM RA with addition of 0.5% Penicillin/Streptomycin, 1% FCS, 10 μM of Rock Inhibitor Y- 27632.…”

Single Cell Multi-Omics of an iPSC Model of Human Sinoatrial Node Development Reveals Genetic Determinants of Heart Rate and Arrhythmia Susceptibility

“…To test for concordance with other SAN differentiation methodologies, we projected scRNA-seq data from a recently published hiPSC SAN differentiation 19 into our UMAP and found very close agreement in cluster assignments between the two protocols, such that 94% of SAN head cells mapped to either our SAN head cluster (C2) or to the narrow interface between our SAN head and the nearest SAN tail clusters (C2/C3 interface); 90% of SAN tail cells mapped to our SAN tail clusters (C0+C3); 82% of transitional cells mapped to our transitional clusters (C1+C4), 64% of sinus venosus cells mapped to our sinus venosus clusters (C5+C6+C7+C8), and 97% of proepicardial cells mapped our proepicardial cluster (C9) (Extended Data Fig 3a). Similarly, to test whether primary PCs from adult human hearts were also represented in our culture system, we projected cells from a recent single cell multi-omics study of human SAN tissue 29 into the SANCM UMAP space.…”

“…One preprocessed scRNA-seq datasets was downloaded from the Gene Expression Omnibus (GEO) (accession code GSE189782 19 ) and processed using Seurat as described in the studies. Processed snRNA-seq data of human SAN from donors with normal conduction parameters [PMID: 37438528] was downloaded from the Heart Cell Atlas (https://www.heartcellatlas.org).…”

“…Existing protocols to generate hiPSC-SAN cells use a variety of methods, including selection from mixed populations of hiPSC-derived cardiomyocytes 14 , directed differentiation using exposure to morphogens and growth factors 11,15 , overexpression of transcription factors [16][17][18] , and directed differentiation using small molecules. Recent work has shown that PC subtypes can be discerned from hiPSC-based differentiations using single cell (sc)-RNA sequencing and clustering analysis 19 , raising the possibility of using in vitro systems to model SAN development, and to develop datasets that can be integrated with human genetic data to yield insights into SND and atrial fibrillation.…”