Multimodal single-cell transcriptomics with patient-specific iPSC-derived airway organoids as a drug screening approach for cystic fibrosis with nonsense mutations

Cystic fibrosis arises from loss-of-function mutations in the CFTR gene, disrupting epithelial ion homeostasis and impairing airway mucus clearance. While missense mutations typically lead to minor conformational alterations that can be rectified with pharmacological interventions, nonsense mutations pose a more significant therapeutic challenge. In this research, we established an in vitro model of cystic fibrosis (CF) employing patient-specific induced pluripotent stem cells (iPSCs) that contain the CFTR-S308X nonsense mutation. We generated CF patient-specific iPSC-derived 3D airway organoids (AOs) that exhibit phenotypic characteristics akin to those of CF, including airway epithelial cells, basal cells, and goblet cells. By employing a single-cell RNA sequencing (scRNA-seq) approach, we analyzed the cellular composition of CFTR-S308X-mutated CF-AOs to explore the potential molecular pathogenesis. Multimodal scRNA-seq analyses were conducted using CellChat to evaluate intercellular communication and RNA velocity to assess transcriptomic dynamics. The CellChat analysis indicated that the TGF-β and EGF signaling pathways may facilitate inflammatory interactions among goblet cells and other cellular subpopulations. The RNA velocity analysis suggested that TNF-α-mediated processes, including autophagy, renin secretion, and relaxin signaling, were upregulated. Deconvolution of bulk RNA sequencing data from patients with CF expressing nonsense mutations corroborated the persistent activation of the TGF-β, TNF-α, and EGF-driven inflammatory pathways in CFTR-S308X-AOs, particularly within the goblet cell subpopulation. Utilizing multimodal single-cell transcriptomics with the ASGARD framework, we found that nintedanib, N-acetylcysteine, and losartan were associated with attenuation of the TNF-α pathway in CF with nonsense mutations. This CF-3D airway organoid model recapitulates aspects of pathogenesis at the single-cell level and serves as a platform for evaluating therapeutics targeting CFTR nonsense mutations.