Both CRSwNP and asthma are airway inflammatory disorders and have serious effects on quality of life. A great proportion of CRSwNP patients have comorbid asthmatic conditions, which is one of the most challenging phenotypes to treat [3, 12]. However, the phenotypes especially the molecular phenotypes of CRSwNP + AS are not clear. To our knowledge, this is the first study investigating the molecular phenotypes of CRSwNP + AS compared to CRSwNP-alone by whole-transcriptome RNA sequencing. Our study identified distinct type 2-high inflammation patterns and their associated transcriptome features in CRSwNP + AS compared to CRSwNP-alone, which will be helpful in understanding the underlying mechanisms and developing specific molecular biomarkers and personalized therapeutic strategies.
Our findings indicated that patients with CRSwNP + AS had more frequent atopy, increased chance for recurrence, more local eosinophils infiltration and higher level of tissue IgE compared to CRSwNP-alone. We previously showed that high proportion of eosinophils in nasal tissue acted as a reliable prognostic indicator for CRSwNP recurrence [24]. Thus, the local features of nasal tissue were closely associated with the clinical characteristics. Together with the recent finding that asthma is a dominant factor increased the chance of recurrence in CRSwNP [6], comorbid asthmatic conditions might be a strong indicator for CRSwNP recurrence.
Airway type 2 inflammation is mainly mediated by eosinophils, mast cells, Th2 cells, ILC2s and IgE-producing B cells, and high production of cytokines IL-5, IL-4 and IL-13 [25]. More than 80% of Western white patients with CRSwNP are characterized by type 2 inflammation, while less than 50% of CRSwNP cases in East Asia countries show features of type 2 reaction [26, 27]. Our findings that patients with CRSwNP + AS had more severe type 2 inflammation than patients with CRSwNP-alone might more valuable to Chinese CRSwNP patients which characterized by Th2/Th1/Th17 mixed pattern of immune response.
Our RNA sequencing data provides valuable information for exploring the general molecular mechanisms underlying the pathogenesis of CRSwNP. Regardless of the subtypes of CRSwNP, genes and pathways that most likely contribute to the pathogenesis of CRSwNP might be mainly associated with cytokine and chemokine signaling pathway, staphylococcus aureus infection, eicosanoid metabolism and cell adhesion molecules, which were similar to Peng’s findings [28]. The well-known genes or biomarkers closely related to CRSwNP such as CLC, POSTN, CCL18, IL13, TSLP and BPIFA1 were also nicely identified as top DE-mRNAs in both CRSwNP-alone group and CRSwNP + AS group.
The transcriptome signatures of CRSwNP + AS were characterized by groups of differentially expressed genes and their enriched pathways, compared to CRSwNP-alone. We showed that CRSwNP + AS were mainly associated with higher activities of arachidonic acid metabolism, Th2 signaling pathway and fibrinolysis pathway, and lower activity of IL-17 signaling pathway. Beyond that, top significant DE-mRNAs between CRSwNP + AS and CRSwNP-alone also provided important information. In line with the high concentration of tissue IgE in CRSwNP + AS, increased mRNA expression of constant region of heavy chain of IgE (IGHE) was also confirmed by RNA sequencing. Some recently identified biomarkers of CRSwNP such as CST1 [29], were found to be differentially expressed between CRSwNP + AS and CRSwNP-alone. Importantly, as top significant DE-mRNAs, several genes that closely associated with the pathogenesis of asthma such as ITLN1, KCNA3 and CCR10 [30–32] were up-regulated in nasal tissue of CRSwNP + AS compared to CRSwNP-alone. Therefore, the gene expression pattern of upper airway could be influenced by lower airway conditions, which is consistent with the united airway concept.
It has been well demonstrated that alterations in the arachidonic acid pathway play an important role in airway inflammatory conditions like rhinosinusitis, nasal polyps, allergic rhinitis, and asthma [33]. We found a generally enhanced activity of arachidonic acid metabolism in CRSwNP + AS, indicated by the up-regulated expression of PLA2 (PLA2G4A, PLA2G4B, PLA2G4D and PLA2G6) which promoted the release of membrane-bound arachidonic acid [34]. The imbalanced synthesis of eicosanoid which characterized by increased synthesis of cysteinyl leukotrienes (CysLTs) is correlated with the inflammatory pattern and severity of the airway inflammation [33, 35]. Consistent with this, we showed an increased expression of LTC4S which promoted the biosynthesis of CysLTs, indicated a severe inflammatory condition in CRSwNP + AS. The increased expression of PTGS1 and PTGIS may accelerate arachidonic acid convertion to prostacyclins (PGI2) and prostaglandins (PGD2, PGE2) which related to an aggravated airway inflammation and asthmatic conditions [36, 37]. Moreover, the increased expression of ALOX15B in CRSwNP + AS might have an active proinflammatory role [38]. A quantitative metabolomic approach is needed to further profile arachidonic acid metabolites in different subtypes of CRSwNP.
Th2 and Th17 signaling pathways are two major regulators and negatively interrelated in patients with CRSwNP [17]. Our RNA sequencing data revealed higher activities of Th2 signaling pathway and lower activities of IL-17 signaling pathway in CRSwNP + AS than CRSwNP-alone. IL-4, IL-5 and IL-13 are typical Th2 cytokines which reflect the severity of type 2 inflammation, whereas IL-17 is a typical Th17 cytokine. We further showed that the expression of IL5 and IL13 are positively correlated with other indicators of type 2 inflammation (local eosinophils and tissue IgE). Genes that related to arachidonic acid metabolism were also correlated with the expression of type 2 cytokines, which suggested the crucial roles of type 2 inflammation in CRSwNP + AS.
LncRNAs play important roles in various biological processes and are emerging as biomarkers and potential therapeutic targets of human chronic diseases [39–40]. However, very few studies have reported the involvement of lncRNAs in chronic nasal inflammation. Yue et al. found that linc00632 was down-regulated in nasal tissues of allergic rhinitis patients and inhibits IL-13 induced inflammatory cytokine and mucus production [41]. Wang et al. showed that lncRNA XLOC_010280 might regulate the expression of CCL18 and eosinophilic inflammation in eosinophilic CRSwNP [42]. In the present study, the whole-transcriptome sequencing revealed all the dysregulated lncRNAs in both subtypes of CRSwNP, which provided a candidate reservoir for lncRNAs research in CRSwNP.
WGCNA is an effective method of multigene analysis to construct coexpression network, and has successfully been applied for studying mRNAs and lncRNAs to distinguish dysfunctional regulatory subnetworks, select out potentially key genes and predict lncRNA functions [43, 44]. We applied WGCNA to predict functions of DE-lncRNAs and identify hub genes. Common dysregulated lncRNAs in CRSwNP-alone and CRSwNP + AS have very similar predictive functions to common DE-mRNAs, as indicated by genes in the maximal module. Furthermore, we identified LINC01146 as the only one top hub lncRNA which may play key roles in the pathogenesis of CRSwNP. LINC01146 was found to be dysregulated in hepatocellular carcinoma with unverified functions [45]. Further study is needed to validate the function of common dysregulated lncRNAs in CRSwNP.
The key lncRNAs associated with CRSwNP + AS had also been identified. HK3-006 is the only one top hub lncRNA in the module that most positively correlated with phenotypic traits of CRSwNP + AS. HK3-006 may be related to asthma pathway and arachidonic acid metabolism as predicted by its highly coexpressed mRNAs. Several top hub lncRNAs were identified in the module that most negatively correlated with phenotypic traits of CRSwNP + AS. Among them, a new identified lncRNA LINC686 might be most likely associated with IL-17 signaling pathway. Although growing studies show that lncRNAs are involving in the regulation of cytokine signaling and inflammation [46–47], our understanding of lncRNA functions is just at the beginning.
There are some limitations in our study. First, this study did not subgroup CRSwNP + AS and CRSwNP-alone. Both CRSwNP and asthma are heterogeneous disease. One recent study showed that patients with CRSwNP + AS can be grouped into 3 subtypes with different inflammatory status by clinical phenotypes [19]. Thus, future studies to investigate the clinical and molecular phenotypes of subgroups of CRSwNP + AS are needed. Second, we did not evaluate lower airway inflammation such as assessment of inflammatory cells and cytokines in bronchial biopsy or induced sputum, which might be helpful in understanding the association between CRSwNP and asthma. Third, the detection of critical cytokines based on the secreted protein levels should be conducted and verified. Finally, this study performed RNA sequencing on heterogeneous tissue with multiple cell types. Single-cell RNA sequencing allows the analysis of transcriptome from individual cells in nasal tissues [48], which might provide more information for understanding the function of different cell types.