부산대학교 도서관

Single cell mRNA sequencing of the whole organ has become a popular technique to reveal rare types and subtypes of previously characterized cells as well as to distinguish and characterize gene expression of previously unknown cell types. Unsupervised clustering can reveal tens or even hundreds of variable genes that characterize cell types. Variation in gene expression is often observed within one cell type, and sometimes cannot be biologically explained without mapping of mRNA on tissue. In this study we aim to (i) map the majority of cell types of human lung, (ii) describe variability in their gene expression and (iii) relate this gene expression to cellular location and neighborhoods. Using three different spatial transcriptomics approaches, we mapped epithelial cell states of airways and submucosal gland, and defined cell type-unrelated gene expression variability along proximo-distal axis, including potential regulators and co-regulators of such cell states in the mesenchymal and immune cell niches. In addition, we mapped rare cell types, such as subtypes of neuroendocrine cells, ionocytes and tuft (brush) cells, revealing tracheal preference for ionocytes, and distal airways for GHRL-positive neuroendocrine cells. Finally, we used the created map as a reference for the diseased tissue from patients with stage II COPD and revealed perturbed cell states and COPD-specific imbalance of cell types, affecting immune and AT0 clusters.