A transcriptional atlas of endothelial cell zonation along the pulmonary vascular tree

Background: The pulmonary vasculature exists as a series of branching vessels that are on gradients of size, oxygenation and pressure. Single cell transcriptomics have provided key insights into the different populations that make up the vasculature but the transcriptomic gradients remain largely undescribed. Methods: We applied a method of endothelial enrichment and deep single cell RNA sequencing to create a high resolution, transcriptomic dataset from the developing mouse lung. We developed an analytical framework to assign vessel-size scores and categorize individual endothelial cells (EC) and mural cells along a continuum of vessel sizes. We delineated a continuum of proximal arterial through distal venous EC states by uncovering transcriptional signatures associated with vessel size, spanning micro- to macrovascular zones. Our data recapitulated previously established zonally defined signaling axes, including Cxcl12 and Cxcr4 in arterioles, and identified localization of disease relevant markers such as Esr2. This vessel-size informed framework was robust across species and revealed how spatial EC heterogeneity underlies key processes in lung development and injury. Results: We generated a robust endothelial cell enriched scRNAseq dataset from the neonatal mouse lung, with deep sequencing coverage. Within this dataset, we defined transcriptomic signatures of macrovascular populations– pulmonary artery (PAEC) and vein (PVEC) endothelial cells– as well as for the microvascular capillary 1 (Cap1) population, incorporating canonical markers genes (e.g. Eln, Vwf, Tmem100, Scn7a). Many of these markers exhibited gradient -like expression patterns extending from arteries or veins toward Cap1, while others displayed polarized expression throughout the Cap1 cluster itself, with subsets exhibiting either artery- or vein-associated signatures. This analytical framework was successfully applied to published human lung datasets across developmental stages, demonstrating cross-species and temporal relevance. Conclusions: By linking scRNA-seq profiles with tissue context, we reconcile molecular signatures with anatomical structure of the pulmonary vasculature, enabling assignment of each individual cell to vessels with defined size. These findings provide a comprehensive transcriptional map of pulmonary endothelial cells and associated mural populations across the vascular continuum, offering valuable insights into spatial inferences and mechanistic insights within single cell RNA sequencing data sets that may help pave the way for targeted therapeutic strategies to treat pulmonary vascular diseases and expansion to tissues outside the lung.