Angiocrine signals in charge: Ets1 shapes ventricular compaction

This editorial refers to ‘Ets1-regulated endothelial-secreted factors promote compact myocardial growth and contribute to the pathogenesis of ventricular non-compaction’ by L. Wang et al., https://doi.org/10.1093/cvr/cvaf264. Left ventricular non-compaction (LVNC) is characterized by excessive trabeculation and failure of the compact myocardium to form properly. Clinically, it spans a broad spectrum—from incidental anatomic variants to severe systolic dysfunction, arrhythmias, and progressive heart failure. Although genetic mutations account for some cases, LVNC also occurs across a wide range of neuromuscular and multisystem disorders,1 highlighting its biological complexity. In this issue of CVR, Wang et al. deliver a compelling and mechanistic study showing how the endothelial transcription factor Ets1 drives compact myocardial growth by coordinating two essential angiocrine pathways—one endocardial, one coronary2 (Figure 1). Together with their recent demonstration that loss of PRDM16 in the mouse model resulted in LV-specific dilation, dysfunction, and biventricular non-compaction,3 their work provides valuable mechanistic insight into the intercellular signalling networks required for building a mature, functional LV wall. Ets1 regulates angiocrine factor expression in the endocardium and coronary endothelium to induce compact myocardial development. Ets1 expression in endocardial endothelial cells (endocardial, left panel) regulate precise expression of endocardial angiocrine signals: Ets1 → Dlk1/Dll4 → Notch1 → Nrg1. In coronary endothelial cells (coronary EC, right panel), Ets1 expression regulates a different set of angiocrine factors: Hmcn1, Slit2, and Col18a1. Both pathways promote cardiomyocyte (CM) proliferation and the formation of a compact left ventricular (LV) myocardium. Endothelial/endocardial Ets1 KO (eKO) mouse models display deviations in the precise level of angiocrine factors expression, as indicated by red arrows in centre panel. Using an endothelial-specific Ets1 knockout (Tie2Cre;Ets1fl/fl; eKO), the authors previously linked Ets1 loss to coronary vascular defects and ventricular non-compaction.4 In this new study, they deliver a remarkably comprehensive analysis, combining single-cell RNA sequencing, RNAscope, immunostaining, analysis of ChIP-seq datasets, explant assays, and recombinant ligand rescue experiments. Their central conclusion is compelling: Ets1 orchestrates two parallel endothelial signalling systems, each indispensable for compact-zone cardiomyocyte proliferation. Disruption of Ets1 in either the endocardium or the coronary endothelium compromises compact myocardial growth, revealing that these endothelial populations supply non-overlapping mitogenic cues. Single-cell analysis reveals considerable heterogeneity within the embryonic endocardium, including clusters with distinct metabolic states, proliferative signatures, and ligand-receptor expression patterns. In Ets1-deficient hearts, two clusters show strong upregulation of Dlk1 together with reduced Dll4 expression. ChIP evidence suggests that Dll4 is a direct transcriptional target of Ets1, establishing perturbed Notch ligand balance as a primary defect. Functional assays strengthen this model: recombinant Dlk1 suppresses Notch1 activity in E9 explants, lowering Notch1 intracellular domain levels. Because Notch drives Nrg1 expression, a key mitogen for compact-layer cardiomyocyte proliferation,5–7 Ets1 loss reduces both Notch signalling and Nrg1 expression. These results define a coherent developmental signalling cascade—Ets1 → Dlk1/Dll4 → Notch1 → Nrg1 → compact myocardial proliferation—that operates early (E9.5), before any measurable myocardial thinning is apparent. This places endothelial Ets1 as a key player in the control of endocardial–myocardial communication. Complementing the endocardial pathway, Wang et al.4 identify an entirely new angiocrine programme in coronary endothelial cells. scRNA-seq reveals broad transcriptional remodelling in Ets1-deficient coronary ECs, including altered metabolic profiles and shifts in arterial/venous identity. Most striking is the consistent downregulation of three secreted ECM-associated mitogens—Hmcn1, Slit2, and Col18a1—across coronary subpopulations, each supported as a direct Ets1 target by ChIP-seq analysis. Explant assays show that recombinant Slit2-C, Col18a1-N, and Hmcn1-C each boost cardiomyocyte proliferation, confirmed by FUCCI imaging. Importantly, these secreted factors act independently of the Notch/Nrg1 axis, revealing a second, mechanistically distinct endothelial programme essential for compact myocardial expansion. This positions the coronary endothelium as an active developmental signalling hub. The involvement of ROBO receptors (Slit2), integrin–ECM adhesion (Col18a1), and matrix scaffolding (Hmcn1) expands the conceptual framework of ventricular wall development far beyond classical pathways. To dissect lineage-specific roles, the authors use Nfatc1Cre for endocardial deletion and PdgfbCreERT2 for coronary endothelial deletion. Both models impair compact-layer thickening, though less severely than the broader Tie2Cre knockout. These results confirm that each endothelial population is independently required and that their combined signalling underlies the non-compaction phenotype. This dual dependency provides a compelling explanation for the heterogeneity of human LVNC presentations and underscores the coordinated need for multiple inputs during the development of a compact myocardial wall. Perhaps the most exciting result, with translational implications, is that recombinant Slit2-C, Col18a1-N, Hmcn1-C, and/or Nrg1 each restore proliferation in Ets1-deficient ventricular explants. While validating both signalling axes, the possibility of developing diagnostic tests or targeted therapies that improve cardiomyocyte proliferation may be applicable to inducing or measuring regenerative capacity. Beyond mechanistic advances, the high-resolution single-cell atlas generated here is a major community resource. By mapping proliferative endocardial clusters, immature cardiomyocyte states, and intermediate myocardium populations, the authors have generated a valuable community resource. This dataset offers a foundation for future spatial transcriptomic studies, developmental lineage tracing, and targeted perturbation experiments. Furthermore, the interplay between endothelial cells and other non-myocyte populations—fibroblasts, immune cells—remains underexplored. LVNC may ultimately prove to be a multicellular network disorder, demanding system-level interventions. Wang et al.4 present a transformative study revealing how endothelial Ets1 coordinates two essential angiocrine programmes that drive compact myocardial expansion. The dual discovery of an endocardial Notch–Nrg1 pathway and a novel coronary endothelial ECM-based signalling axis provides a comprehensive framework linking endothelial dysfunction to LVNC. The identification of three previously unrecognized coronary endothelial mitogens—each able to rescue proliferative deficits—adds significant conceptual and translational impact. Overall, this work advances our understanding of ventricular development and opens new avenues for therapeutic exploration in LVNC and related disorders of impaired myocardial growth. Figure 1 was created with BioRender.com. This work was supported by National Heart, Lung, and Blood Institute at the National Institutes of Health via R01HL177546, R01 HL168900 and R01HL165794. No new data were generated or analysed in support of this research.