Single cell transcriptomic analysis reveals transcriptome differences of different cells between eosinophilic chronic rhinosinusitis with nasal polyps and non-eosinophilic chronic rhinosinusitis with nasal polyps

Abstract

Background: Chronic rhinosinusitis (CRS) with nasal polyps is a heterogeneous chronic inflammatory disease generally divided into two phenotypes including eosinophilic CRS with nasal polyps (eCRSwNP) and non-eosinophilic CRS with nasal polyps (neCRSwNP). However, its pathogenesis remains largely unknown. The aim of this study was to explore mechanistic differences between eCRSwNP and neCRSwNP using a bioinformatics approach.

Methods: We comprehensively analyzed single-cell RNA sequencing data from 3 healthy controls and 6 patients with CRSwNP (including 3 with eCRSwNP and 3 with neCRSwNP) to explore the heterogeneity and potential mechanisms of CRSwNP.

Results: Cluster analysis based on differential gene expression delineated 14 cell clusters. The eCRSwNP group exhibited a markedly higher prevalence of glandular cells and a notable reduction in fibroblasts, myoepithelial cells, and secretory cells compared to patients with neCRSwNP. Functional enrichment analysis of differentially expressed genes revealed the activation of pathways such as IL2-STAT5 signaling and the inhibition of apoptotic pathways in eCRSwNP compared to neCRSwNP. Significant differences in the metabolic profiles of epithelial cell subpopulations were observed between eCRSwNP and neCRSwNP. Furthermore, there were notable discrepancies in the numbers and functionality of immune cells between eCRSwNP and neCRSwNP. The CD4+Th2 cell subsets were found to be significantly enriched in eCRSwNP. The highest number of cellular communications from type 2 conventional dendritic cells (cDC2) to CD4+Th2 cells was found in CRSwNP, where the ICAM1-CD226 pathway from cDC2 to CD4+Th2 was significantly upregulated in eCRSwNP. In addition, eCRSwNP was mainly infiltrated with tissue-resident macrophages, whereas neCRSwNP was mainly infiltrated with monocyte-derived macrophages.

Conclusions: Our study provides new insights into the heterogeneity, molecular mechanisms, and biomarkers of CRSwNP, contributing to improved diagnostic and therapeutic options for this condition.

Fig 1. Single-cell transcriptome landscape of human nasal mucosa in control and CRS.

(A-B) UMAP plot of single-cell transcriptome profiles. Colors indicate clusters (A) and cell types (B). (C) Tissue prevalence of major cell clusters estimated by Ro/e score. (D) Heat map plot of marker genes in each cell types. (E) Heat map plot showing the most enrichment GO biological process terms of marker genes in each cell types. (F) Heat map plot of top5 transcription factors in each cell types.

Fig 2. Analysis of metabolic differences in epithelial cell subtypes.

(A-B) UMAP plot of epithelial cell subtypes. Colors indicate clusters (A) and cell types (B). (C) Tissue prevalence of major cell clusters estimated by Ro/e score. (D) Dot plot of top5 marker genes in each epithelial cell subtypes. (E) Dot plot showing specific metabolic pathway score in different epithelial cell subtypes. (F) Volcano plot showing the differential metabolic pathways in secretory cells. (G) Box plot showing the score of Nitrongen metabolism and Thiamine metabolism. *** p < 0.001.

Fig 3. CD4+ Th2 cells are specific in eCRSwNP.

(A) UMAP plot of CD4+ T cell subtypes. Colors indicate cell types. (B) Dot plot of top5 marker genes in each CD4+ T cell subtypes. (C) Tissue prevalence of major cell clusters estimated by Ro/e score. (D) Bar plot showing the most enrichment GO biological process terms of marker genes in CD4+ Th2 cells. (E) Venn plot showing the overlapped genes of CD4+ Th2 markers and ferroptosis genes. (F) Violin plot of PPARG and SOCS1 in CD4+ T cell subtypes.

Fig 4. Communication between DCs and CD4+ T cells.

(A) UMAP plot of DC cell subtypes. Colors indicate cell types. (B) Dot plot of top10 marker genes in each DC cell subtypes. (C) Tissue prevalence of major cell clusters estimated by Ro/e score. (D) Signaling pathways with significant differences in the overall information flow of each sample groups. (E) Differential number of interactions in DCs and CD4+ T cells. (F) Chord diagram showing up-regulated ligand-receptor pairs in DCs and CD4+ T cells. (G) Violin plot showing expression of ligand-receptor pairs in DCs and CD4+ T cells.

Fig 5. Developmental trajectory of macrophages.

(A-C) UMAP plot of macrophage subtypes. Colors indicate clusters and cell types. (D) Tissue prevalence of major cell clusters estimated by Ro/e score. (E) Dot plot of marker genes in each macrophage subtypes. (F) Pseudo-time plots of macrophage. Color-coded according to states, clusters and pseudo-time value. (G) Heat map showing of differential pseudo-time genes. (H) Line chart showing the C1QA, C1QB, C1QC, HRH1 and FN1 relative expression according by pseudo-time. (I) Bar plot of pseudo-time states in different groups. (J) Dot plot of top5 marker genes in each pseudo-time states. (K) Violin plot of S100A8, CCL18 and EGER in macrophage subtypes.