Distinct expression of SARS-CoV-2 receptor ACE2 correlates with endotypes of chronic rhinosinusitis with nasal polyps

Abstract

Background: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) entry factors, ACE2 and TMPRSS2, are highly expressed in nasal epithelial cells. However, the association between SARS-CoV-2 and nasal inflammation in chronic rhinosinusitis with nasal polyps (CRSwNP) has not been investigated. We thus investigated the expression of SARS-CoV-2 entry factors in nasal tissues of CRSwNP patients, and their associations with inflammatory endotypes of CRSwNP.

Methods: The expression of ACE2 and TMPRSS2 was assessed in nasal tissues of control subjects and eosinophilic CRSwNP (ECRSwNP) and nonECRSwNP patients. The correlations between ACE2/TMPRSS2 expression and inflammatory indices of CRSwNP endotypes were evaluated. Regulation of ACE2/TMPRSS2 expression by inflammatory cytokines and glucocorticoids was investigated.

Results: ACE2 expression was significantly increased in nasal tissues of nonECRSwNP patients compared to ECRSwNP patients and control subjects, and positively correlated with the expression of IFN-γ, but negatively correlated with tissue infiltrated eosinophils, and expression of IL5 and IL13. IFN-γ up-regulated ACE2 expression while glucocorticoid attenuated this increase in cultured nasal epithelial cells. Genes co-expressed with ACE2 were enriched in pathways relating to defence response to virus in nasal tissue. TMPRSS2 expression was decreased in nasal tissues of CRSwNP patients compared to control subjects and not correlated with the inflammatory endotypes of CRSwNP. Glucocorticoid treatment decreased ACE2 expression in nasal tissues of nonECRSwNP patients, but not in ECRSwNP patients, whereas TMPRSS2 expression was not affected.

Conclusion: These findings indicate that ACE2 expression, regulated by IFN-γ, is increased in nasal tissues of nonECRSwNP patients and positively correlates with type 1 inflammation.

Keywords: ACE2; SARS-CoV-2; TMPRSS2; chronic rhinosinusitis with nasal polyps; inflammatory endotype.

Figure 1

Differentially expressed SARS‐CoV‐2 entry factors in nasal tissues. RNA sequencing was performed on nasal tissues from control subjects (n = 19), ECRSwNP patients (n = 16) and nonECRSwNP patients (n = 10). A, Venn diagrams depicting DEGs of ECRSwNP versus control, nonECRSwNP versus control and nonECRSwNP versus ECRSwNP. ACE2, TMPRSS2 and the numbers of DEGs are marked in the corresponding areas. B, Volcano plots illustrating DEGs of nonECRSwNP versus ECRSwNP. C,D, Expression of ACE2 and TMPRSS2 in nasal tissues from inferior turbinate, middle turbinate and uncinate process of control subjects, and nasal polyp tissues from ECRSwNP patients and nonECRSwNP patients. Gene expression was normalized to RPLP0. * P < .05, ** P < .01, *** P < .001. CRSwNP, chronic rhinosinusitis with nasal polyps; ECRSwNP, eosinophilic CRSwNP; nonECRSwNP, noneosinophilic CRSwNP; ACE2, angiotensin‐converting enzyme 2; TMPRSS2, transmembrane protease serine 2; RPLP0, ribosomal protein lateral stalk subunit P0; DEGs, differentially expressed genes; FPKM, fragments per kilo‐base of exon per million fragments mapped; n.s., no significance

Figure 2

Expression of (A) ACE2 and (B) TMPRSS2 detected by Western blot assay in nasal tissues of control subjects, ECRSwNP patients and nonECRSwNP patients. C,D, The intensity of protein bands was quantified by densitometry and normalized to β‐actin. Data are presented as means ± SDs (n = 4 for each group). *P < .05, **P < .01, ***P < .001, Kruskal‐Wallis ANOVA with post hoc Dunn's multiple comparisons test. CRSwNP, chronic rhinosinusitis with nasal polyps; ECRSwNP, eosinophilic CRSwNP; nonECRSwNP, noneosinophilic CRSwNP; ACE2, angiotensin‐converting enzyme 2; TMPRSS2, transmembrane protease serine 2; n.s., no significance

Figure 3

Association between ACE2 expression and differentially infiltrating immune cells in nasal tissue of ECRSwNP and nonECRSwNP patients. A,B, Correlations between ACE2 expression and tissue eosinophils and lymphocytes. ECRSwNP patients are depicted by turquoise dots and nonECRSwNP patients by red dots. C, Cell type deconvolution analysis on gene expression in ECRSwNP and nonECRSwNP patients and control subjects, using xCell. Heatmap depicts the cell type enrichment scores (xCell scores) calculated based on the gene expression data in each sample. *Significantly different between ECRSwNP and nonECRSwNP. D,E, Association between ACE2 expression and xCell scores for eosinophils and dendritic cells, assessed by Spearman correlation analysis. N = 19 for control subjects, n = 16 for ECRSwNP patients and n = 10 for nonECRSwNP patients. CRSwNP, chronic rhinosinusitis with nasal polyps; ECRSwNP, eosinophilic CRSwNP; nonECRSwNP, noneosinophilic CRSwNP; ACE2, angiotensin‐converting enzyme 2; RPLP0, ribosomal protein lateral stalk subunit P0; FPKM, fragments per kilo‐base of exon per million fragments mapped; DC, dendritic cells

Figure 4

Association between ACE2 expression and different inflammatory endotypes of CRSwNP. A, The ACE2 expression in different inflammatory endotypes of CRSwNP. Endotypes of CRSwNP were defined by the expression of IFNG, IL5 and IL17A, respectively. *** P < .001, Kruskal‐Wallis ANOVA with post hoc Dunn's multiple comparisons test. B‐D, RNA sequencing data obtained for nasal tissues from ECRSwNP (n = 16) and nonECRSwNP patients (n = 16) were subjected to Spearman correlation analysis for associations between ACE2 expression and the expression of IFNG, IL5 and IL17A. ECRSwNP patients are depicted by turquoise dots and nonECRSwNP patients by red dots. The cut‐off values of type 1 (IFNG), type 2 (IL5) and type 3 (IL17A) endotypes are indicated by a dashed line. Correlation coefficient and P‐value are shown in the upper right hand corner of each graph. Gene expression was normalized to RPLP0. T, type; Tun, untypeable; ACE2, angiotensin‐converting enzyme 2; IL, interleukin; IFNG, interferon gamma; RPLP0, ribosomal protein lateral stalk subunit P0; FPKM, fragments per kilo‐base of exon per million fragments mapped

Figure 5

Influence of glucocorticoid treatment on the expression of ACE2 and inflammatory cytokines. CRSwNP patients received a 2‐week course of oral glucocorticoid. Nasal polyp tissues were collected before (Baseline) and after glucocorticoid treatment (GC‐treated). The expression of ACE2 (A), IFNG (B), IL5 (C) and IL17A (D) in the tissue was detected by RNA sequencing. N = 13 for ECRSwNP patients, n = 8 for nonECRSwNP patients. *P < .05. CRSwNP, chronic rhinosinusitis with nasal polyps; ECRSwNP, eosinophilic CRSwNP; nonECRSwNP, noneosinophilic CRSwNP; ACE2, angiotensin‐converting enzyme 2; RPLP0, ribosomal protein lateral stalk subunit P0; IL, interleukin; IFNG, interferon gamma; FPKM, fragments per kilo‐base of exon per million fragments mapped; GC, glucocorticoid; n.s., no significance

Figure 6

IFN‐γ‐induced ACE2 expression is attenuated by glucocorticoids. A, Cultured primary nasal epithelial cells were incubated with IFN‐γ (10 ng/mL), IL‐4 (50 ng/mL), IL‐5 (50 ng/mL), IL‐13 (50 ng/mL) or IL‐17 (50 ng/mL) for 24 h, and then analysed for expression of ACE2 by RNA sequencing. N = 3 for IFN‐γ and IL‐13 group, n = 4 for IL‐4, IL‐5 and IL‐17 group. B, Cultured primary nasal epithelial cells (n = 3) were incubated with IFN‐γ (10 ng/mL) ± budesonide (0.4 μg/mL, 0.8 μg/mL) or dexamethasone (0.5 μg/mL, 1.0 μg/mL) for 24 h, and then analysed for expression of ACE2 by real‐time PCR. ACE2, angiotensin‐converting enzyme 2; IL, interleukin; FPKM, fragments per kilo‐base of exon per million fragments mapped; Bud, budesonide; Dex, dexamethasone

Figure 7

Expression and potential function of genes co‐expressed with ACE2 gene in nasal tissue. Co‐expressed genes of ACE2 were assessed in nasal tissues from control subjects, ECRSwNP patients and nonECRSwNP, and performed GO enrichment analysis. A, Bubble chart depicting top 10 significantly enriched GO biological processes by co‐expressed genes of ACE2. B, Expression pattern of ACE2 co‐expressed genes that involved the GO term ‘defence response to virus’. The colour coding of heat maps represents the level of gene expression normalized to control group, calculated based on FPKM. *Significantly increased expression in nonECRSwNP compared to ECRSwNP. CRSwNP, chronic rhinosinusitis with nasal polyps; ECRSwNP, eosinophilic CRSwNP; nonECRSwNP, noneosinophilic CRSwNP; GO, gene ontology; FPKM, fragments per kilo‐base of exon per million fragments mapped