Single-cell profiling demonstrates the combined effect of wheeze phenotype and infant viral infection on airway epithelial development

Abstract

The development of the airway epithelium in asthma is unclear. We characterized nasal airway epithelial cell (NAEC) developmental phenotypes from children aged 2 to 3 years in an a priori designed nested birth cohort from four mutually exclusive groups of wheezers/nonwheezers and respiratory syncytial virus (RSV)-infected/uninfected in the first year of life. NAECs were differentiated, followed by single-cell RNA sequencing analysis and in vitro RSV infection. Gene expression of NAECs from children with a wheeze phenotype indicated abnormal differentiation and basal cell activation of developmental pathways, plasticity in precursor differentiation, delayed onset of maturation, increased diversity of RSV receptors, and blunted antiviral immune responses to in vitro RSV infection. The most marked changes in differentiation were observed in NAECs from children with both wheeze and RSV in the first year of life. Together, this suggests that airway epithelium in children with wheeze is developmentally reprogrammed and characterized by increased barrier permeability, decreased antiviral response, and altered RSV receptor expression.

Fig. 1.. Graphical outline of research study workflow.

Figure was created in BioRender (T. Hartert, 2025; https://BioRender.com/r41p718).

Fig. 2.. Epithelial subset composition of the developing NAECs (2 to 3 years old) in ALI culture.

(A) An integrated object from all four study groups showing identified epithelial cell subsets. UMAP, Uniform Manifold Approximation and Projection. (B) Dot plot showing conventional markers used to identify epithelial cell subsets. (C) Top markers for each of the 17 clusters identified by differential gene expression analysis. Cluster numbers on top correspond to cluster numbering in (A). Low expression is shown in blue and high expression is shown in orange. (D) An integrated object from all four study groups [wheeze/RSV, wheeze/no RSV, no wheeze/RSV, no wheeze/no RSV (control)] showing cell subset differences by group. (E) An integrated object from all four study groups highlighting cells individually by group. (F) Proportional representation of cells per cluster by study group. From left to right: basal cell clusters, progenitor and precursor clusters, secretory clusters, and ciliated epithelial clusters. Study groups: wheeze/RSV in infancy, wheeze (wheeze/no RSV), RSV (no wheeze/RSV in infancy), and control (no wheeze and no RSV in infancy).

Fig. 3.. NAECs from wheeze study groups show evidence of developmental reprogramming and an increased expansion of epithelial precursor and secretory phenotypes.

(A) LungMAP data mining of human lung tissue development suggests specific epithelial differentiation trajectory in early life. (B) NAECs from wheeze study groups show trend of relative expansion of epithelial precursor and secretory subsets. (C) A diagram showing heterogeneous composition of mature stratified respiratory epithelium. (D) Epithelium from wheeze group samples showed a proportional increase in secretory (club precursor, club, and goblet subsets combined) cells relative to cells from all populations. Left: Individual samples across four groups quantified. Right: Samples combined by wheeze versus nonwheeze groups. *P < 0.05, **P < 0.01 by ANOVA (graph on the left) and t-test (graph on the right). (E) Cell cycle analysis confirms early identity proliferative capacity of basal and progenitor epithelial cell subsets. Cells from all groups are shown in the integrated object. (F) Unsupervised slingshot developmental trajectory inference reveals altered developmental trajectory in wheezers (red lines) compared to nonwheeze controls (blue lines). Black circle denotes basal cell populations as the starting point for developmental trajectories. (G) Interpretation of developmental trajectories in control and wheeze study groups. (G) was created in BioRender (T. Hartert, 2025; https://BioRender.com/r41p718).

Fig. 4.. NAECs from wheeze study groups show early activity of developmental pathways and abnormal activation of basal cells.

(A) Expression of markers of developmental pathways (WNT, Notch, TGF, EGF, and tissue plasminogen system) by cell type and study group. (B) Expression of PAI-2 (SERPINB2) and JAG1 (Notch/Jagged) is increased in wheeze NAECs. (C) Differentially expressed gene (DEG) analysis of basal cell subset comparing RSV and wheeze study groups to controls. False discovery rate–adjusted P values are used. (D) Pathway analysis of up-regulated basal cell DEGs common to control versus RSV and “control versus wheeze/RSV” (left) and up-regulated basal cell DEGs common to control versus wheeze and control versus wheeze/RSV (right). OxPhos, oxidative phosphorylation.

Fig. 5.. NAECs from wheeze study groups show increased diversity of RSV receptors, decreased expression of host antiviral response genes, increased barrier permeability, and increased susceptibility to RSV infection in vitro.

(A and B) NAECs from wheeze groups show increased diversity of RSV receptors. TLR4, Toll-like receptor 4; LDLR, low-density lipoprotein receptor; ICAM-1, intercellular adhesion molecule–1; NCL, nucleolin; HSPG, the gene for perlecan. (C) NAECs from wheeze groups have decreased expression of MX1, a marker of cellular antiviral response. (D) NAECs from wheeze groups have decreased expression of IFNAR1, another antiviral factor. (E) NAECs were differentiated at ALI. TEER permeability was determined at various days post-ALI. n = 5 to 6 in each group and data is shown as mean (SD). *P < 0.05, two-way analysis of variance (ANOVA) with repeated measures and Tukey post hoc test. (F to H) After fully differentiated, NAECs were infected with RSV 01/2-20 (MOI = 3). Twenty-four hours postinfection, cells were harvested and RSV M gene expression (F) or IFNL2 (G) was determined by quantitative polymerase chain reaction (qPCR) and normalized to TATA box–binding protein 1 (TBP1). IFN-β protein levels were measured by enzyme-linked immunosorbent assay (ELISA) in basolateral supernatants of ALI cultures (H). n = 5 to 6 samples from different participants in each group, and data are shown as mean (SD). *P < 0.05 and **P < 0.01, two-way ANOVA with Tukey post hoc test (E) and one-way ANOVA with Kruskal-Wallis post hoc test [(F) to (H)]. n.s., not significant.