Nasal epithelial gene expression and total IgE in children and adolescents with asthma

Abstract

Background: Little is known about nasal epithelial gene expression and total IgE in youth.

Objective: We aimed to identify genes whose nasal epithelial expression differs by total IgE in youth, and group them into modules that could be mapped to airway epithelial cell types.

Methods: We conducted a transcriptome-wide association study of total IgE in 469 Puerto Ricans aged 9 to 20 years who participated in the Epigenetic Variation and Childhood Asthma in Puerto Ricans study, separately in all subjects and in those with asthma. We then attempted to replicate top findings for each analysis using data from 3 cohorts. Genes with a Benjamini-Hochberg-adjusted P value of less than .05 in the Epigenetic Variation and Childhood Asthma in Puerto Ricans study and a P value of less than .05 in the same direction of association in 1 or more replication cohort were considered differentially expressed genes (DEGs). DEGs for total IgE in subjects with asthma were further dissected into gene modules using coexpression analysis, and such modules were mapped to specific cell types in airway epithelia using public single-cell RNA-sequencing data.

Results: A higher number of DEGs for total IgE were identified in subjects with asthma (n = 1179 DEGs) than in all subjects (n = 631 DEGs). In subjects with asthma, DEGs were mapped to 11 gene modules. The top module for positive correlation with total IgE was mapped to myoepithelial and mucus secretory cells in lower airway epithelia and was regulated by IL-4, IL5, IL-13, and IL-33. Within this module, hub genes included CDH26, FETUB, NTRK2, CCBL1, CST1, and CST2. Furthermore, an enrichment analysis showed overrepresentation of genes in signaling pathways for synaptogenesis, IL-13, and ferroptosis, supporting interactions between interleukin- and acetylcholine-induced responses.

Conclusions: Our findings for nasal epithelial gene expression support neuroimmune coregulation of total IgE in youth with asthma.

Keywords: EVA-PR; TWAS; VDKA; childhood asthma; total IgE.

Figure 1. Overview of study design

Footnote: EVA-PR=The Epigenetic Variation and Childhood Asthma in Puerto Ricans study; PIAMA=The Prevention and Incidence of Asthma and Mite Allergy study; VDKA=The Vitamin D Kids Asthma study.

Figure 2. Differentially expressed genes associated with total IgE in all subjects or subjects with asthma

Footnote: (A) Density plot showing the distribution of total IgE in all subjects from discovery (EVA-PR) and replication (PIAMA and GSE65204) cohorts. (B) Volcano plot showing the coefficient and –log₁₀(BH-adjusted P-value) of all the genes in EVA-PR. Genes above the dashed line had BH-adjusted P-value < 0.05 in EVA-PR, which were colored in red or blue if replicated (P-value < 0.05) in at least one cohort in the same direction of association. (C) Density plot showing the distribution of total IgE in subjects with asthma from discovery (EVA-PR) and replication (VDKA and GSE65204) cohorts. (D) Volcano plot showing the coefficient and –log₁₀(BH-adjusted P-value) of all the genes in EVA-PR. Genes above the dashed line had BH-adjusted P-value < 0.05 in EVA-PR, which were colored in red or blue if replicated (P-value < 0.05) in at least one cohort in the same direction of association.

Figure 3. Mapping gene modules to certain cell types in scRNA-seq data

Footnote: (A) UMAP plot showing 14 annotated cell types in lung and airway epithelia collected from healthy donors. (B) Boxplots summarizing scores for the turquoise and yellow modules in the 14 cell types. (C) TSNE plot showing 10 annotated cell types, as well as patients’ polyp status, in nasal airway epithelia collected from chronic rhinosinusitis patients. (D) Boxplots summarizing scores for the turquoise and yellow modules in the 10 cell types. (E) Boxplots summarizing scores for the turquoise and yellow modules in the 10 cell types when split by patients’ polyp status. (F) Hub genes within the turquoise models.