Activation of the 15-lipoxygenase pathway in aspirin-exacerbated respiratory disease

Abstract

Background: Aspirin-exacerbated respiratory disease (AERD) is characterized by asthma, chronic rhinosinusitis with nasal polyps (CRSwNP), and an intolerance of medications that inhibit cyclooxygenase-1. Patients with AERD have more severe upper and lower respiratory tract disease than do aspirin-tolerant patients with CRSwNP. A dysregulation in arachidonic acid metabolism is thought to contribute to the enhanced sinonasal inflammation in AERD.

Objective: Our aim was to utilize an unbiased approach investigating arachidonic acid metabolic pathways in AERD.

Methods: Single-cell RNA sequencing (10× Genomics, Pleasanton, Calif) was utilized to compare the transcriptional profile of nasal polyp (NP) cells from patients with AERD and patients with CRSwNP and map differences in the expression of select genes among identified cell types. Findings were confirmed by traditional real-time PCR. Lipid mediators in sinonasal tissue were measured by mass spectrometry. Localization of various proteins within NPs was assessed by immunofluorescence.

Results: The gene encoding for 15-lipooxygenase (15-LO), ALOX15, was significantly elevated in NPs of patients with AERD compared to NPs of patients with CRSwNP (P < .05) or controls (P < .001). ALOX15 was predominantly expressed by epithelial cells. Expression levels significantly correlated with radiographic sinus disease severity (r = 0.56; P < .001) and were associated with asthma. The level of 15-oxo-eicosatetraenoic acid (15-Oxo-ETE), a downstream product of 15-LO, was significantly elevated in NPs from patients with CRSwNP (27.93 pg/mg of tissue) and NPs from patients with AERD (61.03 pg/mg of tissue) compared to inferior turbinate tissue from controls (7.17 pg/mg of tissue [P < .001]). Hydroxyprostaglandin dehydrogenase, an enzyme required for 15-Oxo-ETE synthesis, was predominantly expressed in mast cells and localized near 15-LO⁺ epithelium in NPs from patients with AERD.

Conclusions: Epithelial and mast cell interactions, leading to the synthesis of 15-Oxo-ETE, may contribute to the dysregulation of arachidonic acid metabolism via the 15-LO pathway and to the enhanced sinonasal disease severity observed in AERD.

Keywords: 15-lipoxygenase; 15-oxo-eicosatetraenoic acid; AERD; ALOX15; CRSwNP; Chronic sinusitis; aspirin-exacerbated respiratory disease; hydroxyprostaglandin dehydrogenase; nasal polyps.

Figure 1.. scRNA-seq analysis of cells from nasal polyps from CRSwNP and AERD patients.

scRNA-seq analysis of combined nasal polyp cells from CRSwNP and AERD patients identified 26 cell clusters (A). Clusters were annotated based upon expression of canonical genes: KRT5 (basal epithelial cells), KRT8 (apical epithelial cells), LTF (glandular epithelial cells), FOXJ1 (ciliated epithelial cells), COL1A2 (fibroblasts), ACKR1 (endothelial cells), CD79A (B cells), TRBC2 (T cells), HLA-DRA (myeloid), and TPSAB1 (mast cells) (B). Clusters containing the same cell types were combined together resulting in 10 unique clusters (C). Scaled expression of key genes involved in arachidonic acid metabolism and signaling via the 5-LO (D) and COX (E) pathways.

Figure 2.. 15-oxo-ETE, but not 5-oxo-ETE, was elevated in nasal polyps.

Levels of 5-oxo-ETE and 15-oxo-ETE were measured by HPLC-MS/MS in sinonasal tissues. Representative traces from a NP sample illustrating the relative abundance and mass per charge (m/z) ratios of 5-oxo-ETE and 15-oxo-ETE (A-C). Levels of 5-oxo-ETE (D) and 15-oxo-ETE (E) were quantified from sinonasal tissue samples from healthy control, CRSsNP, CRSwNP, AERD patients. 15-oxo-ETE was significantly elevated in NP tissues from both CRSwNP and AERD patients. Dot plots illustrate individual data points with solid lines representing the mean ± SEM. Statistical significance (p<0.05) was determined by Kruskal-Wallis test with post-hoc analysis using the Dunn’s test for multiple comparison; **p<0.01 and ***p<0.001.

Figure 3.. ALOX15 was elevated in ciliated and apical epithelial cells and correlated with eosinophils in AERD and CRSwNP nasal polyps.

Violin plots of normalized ALOX15 gene expression in all NP cells (A), in all epithelial cells (apical, basal, ciliated, and glandular subsets) (B), and in individual cell clusters (C) from AERD and CRSwNP patients. ALOX15 expression was highest in ciliated and apical epithelial cells and, in these clusters, was elevated in AERD compared CRSwNP NP (C). ALOX15 expression was measured in separate sinonasal tissue samples by qRT-PCR, and was highest in NP from AERD patients (D). Increased expression of ALOX15 significantly correlated with expression of CLC (E) and with protein levels of Eosinophilic Cationic Protein (ECP) (F) among matched sinonasal tissues. Dot plots illustrate individual data points with solid lines representing the mean ± SEM. Statistical significance (p<0.05) was determined by Wilcoxon rank sum (A-C), by Kruskal-Wallis test with post-hoc analysis using the Dunn’s test for multiple comparison (D), or Spearman’s rank correlation (E-F); *p<0.05, ***p<0.001, ****p<0.0001.

Figure 4.. 15-HETE was not elevated in CRSwNP or AERD nasal polyps.

15-HETE was detected in sinonasal tissues by HPLC-MS/MS. Representative plots from a NP sample illustrating the peak and m/z of 15-HETE (A-B). Quantification of 15-HETE in sinonasal samples found no significant difference in levels of 15-HETE among the patient groups. (C). Statistical significance (p<0.05) was determined by Kruskal-Wallis test with post-hoc analysis using the Dunn’s test for multiple comparison. Dot plots illustrate individual data points with solid lines representing the mean ± SEM.

Figure 5.. HPGD was predominantly expressed by mast cells in AERD and CRSwNP nasal polyps.

HPGD gene expression level superimposed on tSNE of NP cells from AERD and CRSwNP patients (A) with scaled expression levels showing HPGD highest in mast cells (B). HPGD expression was measured in separate sinonasal tissue samples by qRT-PCR, and was elevated in NP from both CRSwNP and AERD patients compared to control (C). HPGD expression correlated with expression of TPSB2 in matched sinonasal tissue samples (D). NP were stained by immunofluorescence using no primary control (E) or with HPGD (green) and tryptase (red) in CRSwNP (F) and AERD (G) NP. The majority of HPGD⁺ cells detected also stained positive for the mast cell marker tryptase. Dot plots illustrate individual data points with solid lines representing the mean ± SEM. Statistical significance (p<0.05) was determined by Kruskal-Wallis test with post-hoc analysis using the Dunn’s test for multiple comparison (C) or Spearman’s rank correlation (D); *p<0.05 and **p<0.01.

Figure 6.. ALOX15-expressing epithelial cells and HPGD-expressing mast cells co-localized within AERD and CRSwNP nasal polyps.

Gene expression of HPGD and ALOX15 across patients is positively correlated in CRSwNP and AERD NP as measured by scRNA-seq (A) or by qRT-PCR (B) in matched sinonasal tissue samples (B). NP were stained by immunofluorescence using no primary control (C) or with HPGD (green) and 15-LO (red) in CRSwNP (F) and AERD (G) NP. HPGD⁺ mast cells were localized in close proximity to 15-LO⁺ epithelial cells in NP. Statistical significance (p<0.05) was determined by Spearman’s rank correlation (A-B).

Figure 7.. ALOX15 gene expression correlated with sinonasal disease severity.

ALOX15 gene expression was measured by qRT-PCR in control UT, CRSsNP UT, CRSwNP NP, and AERD NP and analyzed according to the patient’s asthma status (A), degree of asthma severity (B), Lund-Mackay score (C), and number of sinus surgeries (D). Dot plots illustrate individual data points with solid lines representing the mean ± SEM. Statistical significance (p<0.05) was determined by Kruskal-Wallis test with post-hoc analysis using the Dunn’s test for multiple comparison (A, B) or Spearman’s rank correlation (C); *p<0.05, **p<0.01, and ***p<0.001.