Introduction

Chronic rhinosinusitis (CRS) is an inflammation that primarily affects the mucosa of paranasal sinuses and lasts for at least 12 weeks. It is associated with morbidity of approximately 8% in China, 11% in Europe, and 13% in the USA1,2,3. CRS can be traditionally classified into two types based on the presence or absence of nasal polyps (NP): CRS with NP (CRSwNP) and CRS without NP. The latter is predominantly characterized by neutrophilic inflammation, while CRSwNP is associated with type II inflammation characterized by eosinophilia4,5. Neutrophils have been suggested to play an important role in the pathogenesis of CRSwNP, but the exact mechanism remains unrecognized5,6. CRSwNP is typically considered a more severe form of CRS due to its high recurrence rate and association with asthma7. The main treatments for CRSwNP are glucocorticosteroids and surgery; however, a significant number of patients do not achieve satisfactory results, possibly due to the presence of coexisting neutrophilic inflammation8, which is also observed in both asthma and CRSwNP9,10.

The mechanism underlying neutrophil infiltration in CRSwNP is an ongoing topic of research. Neutrophils in the peripheral circulation are recruited to local inflammatory sites by diverse chemokines, such as chemokine (C-X-C motif) ligand (CXCL) 1, CXCL2, CXCL5, and IL-811,12. CXCL1, CXCL2, and CXCL8 are derived from fibroblasts and mast cells in response to corresponding stimuli13,14. IL-8 is secreted by nasal mucosal epithelial cells when they are stimulated by inflammation or foreign bodies, making it an important source of neutrophil chemokines15,16. In addition, eosinophils produce Charcot–Leyden crystals, which promote IL-8 production in NPs and facilitate neutrophil chemotaxis17. While previous studies suggest that neutrophil recruitment plays a crucial and irreplaceable role in the pathophysiology of CRSwNP18, the regulatory mechanisms and biomarkers associated with neutrophil infiltration remain to be fully recognized.

To gain a deeper understanding of neutrophils in the pathophysiology of CRSwNP, we investigated neutrophil infiltration levels and pertinent genes in CRSwNP using three gene datasets obtained from the gene expression omnibus (GEO) database. Furthermore, we employed weighted gene coexpression network analysis (WGCNA) and a protein–protein interaction (PPI) network, which revealed four hub genes associated with neutrophils and their biological functions. Our findings can be used to identify potential targets for the diagnosis and treatment of CRSwNP.

Results

Integrative and immune cell infiltration analyses

A total of 54 NP and 37 control tissue samples were collected from the GSE136825, GSE36830, and GSE72713 datasets after batch effect correction (Fig. 1A, Supplementary Fig. 1). Relative to controls, patients with CRSwNP showed higher infiltration levels of immune cells, including T cells, B lineage, monocytic lineage, myeloid dendritic cells, and neutrophils (Fig. 1B, Supplementary Fig. 2). The increase in neutrophil infiltration levels in patients with CRSwNP was further validated using different databases (Fig. 1C). These results suggested that neutrophil infiltration was markedly different between NP and control tissue samples, indicating that neutrophils may play a crucial role in the pathology of CRSwNP.

Figure 1
figure 1

Integrative and immune cell infiltration analysis. (A) Principal component analysis (PCA) of gene expression profile (GSE136825, GSE36830, and GSE72713) after correcting batch effect. (B) Comparison of immune cell infiltration between control subjects and CRSwNP through MCPcounter. (C) Comparison of neutrophil infiltration between control subjects and CRSwNP through MCPcounter, ssGSEA, Quantiseq, and Cibersort (Software: R version 4.3.1. URL: http://www.R-project.org).

Full size image

WGCNA and neutrophil-associated gene identification

To investigate the main regulators of neutrophil infiltration in CRSwNP, we selected 800 genes associated with neutrophils to construct a gene coexpression network, followed by calculating average linkage and correlation coefficients and performing cluster analysis. A scale-first network was successfully constructed with β = 4 as the soft threshold power (Fig. 2A). We also employed the dynamic hybrid cut method to build a hierarchical clustering tree. Each leaf represented a gene, and each branch represented a module comprising genes with similar expression levels. We then merged functionally equivalent modules into one large module, resulting in seven modules (Fig. 2B). The blue module (MEblue) exhibited a more significant relationship with neutrophil infiltration level in the CRSwNP group (Fig. 2C). Therefore, the MEblue with 116 neutrophil-associated genes was considered as the hub module. Functional enrichment analyses revealed that genes in the MEblue were significantly enriched in the regulation of phosphatidylinositol 3-kinase (PI3K) activity, cell activation involved in immune response, leukocyte activation involved in immune response, and regulation of vascular endothelial growth factor (VEGF) receptor signaling pathway (Fig. 2D). These findings indicated that the 116 genes may serve as the main regulators of neutrophil infiltration in CRSwNP.

Figure 2
figure 2

Gene co-expression network analysis and identification of neutrophil- associated genes in CRSwNP. (A) Construction of the scale-first network with the soft threshold power β = 4. (B) Establishment of co-expressed gene modules based on dynamic hybrid cut method. (C) Correlations between module eigengene and neutrophil infiltration. The blue module (MEblue) indicates a positive relationship with the neutrophil infiltration level in CRSwNP. (D) Function enrichment of the 116 genes in the blue module (Software: R version 4.3.1. URL: http://www.R-project.org).

Full size image

Neutrophil-associated gene identification and validation

To narrow down neutrophil-associated genes in CRSwNP, a PPI network was established using STRING. The 116 neutrophil-associated genes in CRSwNP formed a PPI network containing 66 proteins and 83 interaction edges (Fig. 3A). A highly interconnected module (Cluster 1, MCODE score = 3.333) including four genes in the PPI network was identified as hub genes in CRSwNP (Fig. 3A, Table 1). The expression levels of these genes, namely intercellular cell adhesion molecule-1 (ICAM1), interleukin-1β (IL-1β), TYRO protein tyrosine kinase-binding protein (TYROBP), and B-cell lymphoma 2-related protein A1 (BCL2A1), were found to be considerably upregulated in patients with CRSwNP compared to those in control subjects (Fig. 3B). Correlation analyses revealed that the expression of the hub genes was positively correlated with the infiltration level of neutrophils in patients with CRSwNP (Fig. 3C, Supplementary Tables 1, 2). These results indicated that ICAM1, IL1β, TYROBP, and BCL2A1 may act as positive regulators of neutrophils and represent neutrophil-associated genes in CRSwNP.

Figure 3
figure 3

Protein–protein interaction (PPI) analyses of neutrophils associated genes. (A) PPI network of neutrophils associated genes containing 66 proteins and 83 interaction edges in CRSwNP (Software: STRING. URL: https://string-db.org/). (B) The expression levels of hub genes by Wilcoxon analysis between control subjects and CRSwNP. (C) Correlations between hub genes and the infiltration level of neutrophils by Spearman correlation analysis in CRSwNP ((B–D) Software: R version 4.3.1. URL: http://www.R-project.org).

Full size image
Table 1 Identification of hub genes in the PPI network.
Full size table

Neutrophil-associated gene validation

The relative gene expression levels of BCL2A1, ICAM1, IL-1β, and TYROBP were validated by RT-qPCR using NP tissues from patients with CRSwNP and nasal mucosal samples from healthy controls. The expression levels of BCL2A1, ICAM1, IL-1β, and TYROBP were significantly higher in patients with CRSwNP than in control subjects (P < 0.05, Fig. 4A–D), suggesting that these neutrophil-associated genes positively regulate CRSwNP development. In order to further clarify the relationship between BCLA1, ICAM1, IL-1β and TYROBP and neutrophils, we correlated the PCR results with the percentage of neutrophils in the peripheral blood of the patients. The results showed that the mRNA expression levels of BCLA1, ICAM1, IL-1β and TYROBP in nasal polyps were positively correlated with the percentage of neutrophils in the peripheral blood (Fig. 4E–H).

Figure 4
figure 4

The gene expression levels of hub genes and correlation with the proportion of neutrophils in CRSwNP and control samples. The mRNA relative expression levels of the BCL2A1 (A), ICAM1 (B), IL-1β (C) and TYROBP (D) in CRSwNP (n = 10) and control samples (n = 5) by RT-qPCR and statistical significance by Student’s t-test; Correlations of relative mRNA expression levels of BCL2A1 (E), ICAM1 (F), IL-1β (G) and TYROBP (H) with the percentage of neutrophils in peripheral blood by Pearson correlation analysis. *p < 0.05, **p < 0.01.

Full size image

Discussion

A mixed eosinophilic–neutrophilic inflammation has been reported in 35.8% Chinese patients with CRSwNP19. Previous studies have mostly reported a geographic prevalence of neutrophilic inflammation in case of patients with CRSwNP, predominantly in Asians; however, recent studies have also observed this phenomenon in Western countries6,20. Numerous studies have shown that neutrophils could serve as key diagnostic and therapeutic targets in CRSwNP6,20,21. Nonetheless, the regulators of neutrophil infiltration in CRSwNP remain unknown. Therefore, the objective of this study was to identify hub genes associated with neutrophil infiltration and explore their roles in patients with CRSwNP.

In the present work, we analyzed gene expression levels in 54 NP tissue samples and 37 nasal mucosal samples, which were extracted from public database. CRSwNP patients showed significantly higher neutrophil infiltration levels. Using WGCNA, PPI network construction, and correlation analysis, we identified four hub genes, BCL2A1, ICAM1, IL-1β, and TYROBP, which closely associated with neutrophil infiltration in CRSwNP and validated using GEO database and RT-qPCR. Functional enrichment analysis indicated that the biological processes associated with CRSwNP mainly involved regulation of VEGF receptor signaling pathway, leukocyte aggregation, and leukocyte activation involved in immune response. We believe that these four genes can serve as potential targets for the diagnosis and treatment of refractory CRSwNP.

The elevation of neutrophil infiltration levels in patients with CRSwNP is reportedly associated with poor glucocorticoid response and surgical prognosis20. An earlier machine learning algorithm study found that neutrophil infiltration is a vital risk factor for postoperative recurrence in patients with CRSwNP21. Kim et al. reported that neutrophils negatively affect surgical outcomes in patients with CRSwNP in Asia22. These findings align with our results, supporting that neutrophil infiltration plays a chief role in CRSwNP development.

Herein we identified four hub genes (BCL2A1, ICAM1, IL-1β, and TYROBP) that were closely related to neutrophil infiltration in CRSwNP. BCL2A1, a key anti-apoptotic gene, plays a crucial role in the regulation of cell death and the survival of differentiating and mature neutrophils23. However, data on the expression of BCL2A1 in NP remain scarce. In this study, bioinformatics analysis and RT-qPCR results confirmed the upregulation of BCL2A1 expression levels in patients with CRSwNP. BCL2A1 can be induced by TNF-α and is a target gene for NF-κB24,25, the expression of which was found to be upregulated by Valera et al. in the NP group relative to that in the control group26. ICAM1, a cell surface glycoprotein, belongs to the immunoglobulin superfamily, and it participates in neutrophil transmigration during the inflammatory process. Wang et al. observed that neutrophils exhibited upregulated expression of ICAM1 upon LPS and TNF-α stimulation, resulting in a significant increase in neutrophil adhesion and aggregation27. Higher expression level of ICAM1 in patients with CRSwNP has been documented28, but no study has yet linked it to the role of neutrophils in NP. Altogether, these findings suggest that BCL2A1 and ICAM1, as neutrophil-associated genes, plays vital role in CRSwNP; nevertheless, future studies are warranted to investigate its mechanism of action.

IL-1β is a strictly regulated proinflammatory cytokine that is critical for host defense in response to injury and infection. It is principally produced by neutrophils, macrophages, and monocytes29. Several studies have reported increased expression of IL-1β in neutrophilic CRSwNP19,30, which is consistent with our findings. At present, IL-1β blocking therapy is approved for various chronic inflammatory diseases, such as rheumatoid arthritis and other autoimmune diseases. Regarding TYROBP, limited studies have explored its association with neutrophil infiltration. TYROBP, also known as KARAP/DAP12, is primarily expressed in lymphoid and myeloid cells and serves a key regulator of immune response. Spahn et al. first confirmed that TYROBP facilitates the transendothelial migration of neutrophils and mediates noninfectious tissue injury using a mouse model of pulmonary ischemia–reperfusion injury31. However, the underlying mechanism remains unclear. Therefore, it is necessary to elucidate the mechanism of IL-1β and TYROBP in the pathogenesis of CRSwNP, which may provide a new treatment option for CRSwNP.

To summarize, we identified the role of neutrophil infiltration in the pathophysiology of CRSwNP by integrating gene expression datasets of CRSwNP and neutrophils and performing WGCNA and PPI network analysis. Moreover, we identified and validated four hub genes, namely BCL2A1, ICAM1, IL-1β, and TYROBP, using clinical samples of NPs through RT-qPCR. Our findings provide a new research direction for understanding the pathogenesis of CRSwNP; nevertheless, further studies are warranted to verify the role of BCL2A1, ICAM1, IL-1β, and TYROBP in CRSwNP.

Materials and methods

Dataset acquisition and preparation

Figure 5 depicts our study protocol. We selected microarray and high-throughput sequencing datasets from the GEO database for nasal tissues of patients with CRSwNP: GSE13682532, GSE3683033, and GSE7271334. Derived from the GPL20301 platform (Illumina HiSeq 4000), GSE136825 includes NP tissue samples from 42 patients with CRSwNP and nasal mucosal samples from 28 healthy controls. GSE36830 is derived from the GPL570 platform (Affymetrix Human Genome U133 Plus 2.0 Array) and comprises NP tissue samples from six patients with CRSwNP and nasal mucosal samples from six healthy controls. GSE72713, based on Illumina HiSeq 2000, includes NP tissue samples from six patients with CRSwNP and nasal mucosal samples from three healthy controls. The details of each dataset are shown in Table 2. The ComBat function of the R “SVA” package35,36 was employed to remove batch effects across GSE136825, GSE36830, and GSE72713. This study adheres to the data access strategy of each database.

Figure 5
figure 5

Flowchart of the study.

Full size image
Table 2 The information of GEO datasets.
Full size table

Immune cell infiltration analysis in CRSwNP

Base on the gene expression profile, we utilized MCPcounter37, ssGSEA38, Quantiseq39, and Cibersort40 to calculate and compare immune cell infiltration levels in controls and patients with CRSwNP.

WGCNA

We performed WGCNA using the “WGCNA” R package to identify genes associated with NP. A total of 800 neutrophil-associated genes were collected from gene set enrichment analysis (M7789, M7790, M7792, and M7794), and these genes were used to generate WGCNA modules. The adjacency matrix was transformed into a topological overlap matrix. Genes with similar expression patterns were assigned to the same gene module based on topological overlap matrix-based dissimilarity measures, with a minimum module size of 30 and cut height of 0.25, using average linkage hierarchical clustering. Subsequently, we assessed the correlation between module eigengenes and neutrophil infiltration levels. We identified the module with the higher correlation with neutrophil infiltration levels and evaluated its related biological processes using the clusterProfiler R package.

Candidate hub gene selection and model construction

A PPI network related to genes in the blue model was constructed using STRING (minimum required interaction score = 0.400), with hidden disconnected nodes (http://string-db.org/). Cytoscape was used for visualization. Subsequently, hub genes were evaluated using cytoscape molecular complex detection (MCODE) based on topology (module criteria: degree = 2, node score = 0.2, k-score = 2, and max depth = 100). Wilcoxon analysis was applied to evaluate differences in mRNA levels of hub genes between patients with CRSwNP and control groups in the GEO database. The correlation between hub gene expression and neutrophil infiltration level was determined using Spearman correlation analysis.

Patient recruitment

We collected tissue samples from five control subjects and 10 patients with CRSwNP who underwent nasal endoscopic surgery at the Department of Otorhinolaryngology, Yuhuangding Hospital. All patients with CRSwNP were enrolled according to the European Position Paper on Rhinosinusitis and NP 202041. The exclusion criteria included patients with comorbidities other than NP (e.g. hypertension, diabetes, and allergic diseases), history of corticosteroid or antibiotic treatment in the three months prior to surgery, and postoperative pathological results not indicating NP. Control subjects were individuals undergoing septoplasty for nasal septum deviation or endoscopic optic nerve decompression due to traumatic optic neuropathy. This study was approved by the Institutional Ethics Committee of Yantai Yuhuangding Hospital, and all patients provided informed consent.

RNA extraction and real-time quantitative PCR (RT-qPCR)

Total RNA was isolated from NP tissue and nasal mucosal samples using TRIzol reagent (SparkJade, China), according to manufacturer instructions. Reverse transcription was performed using 1 μg RNA and a cDNA synthesis kit (Vazyme, China). RT-qPCR was performed using SYBR Green qPCR Mix kit (SparkJade, China). GAPDH was used as the internal reference to normalize cDNA concentrations within each sample, and the 2−ΔΔCt method was applied to analyze relative mRNA expression levels. Student’s t-test was employed to compare differences between the groups. Primers used for RT-qPCR are listed in Table 3.

Table 3 Sequences of primers.
Full size table

Ethics approval and consent to participate

All procedures performed in the present study involving human participants were in accordance with the Declaration of Helsinki (1964) and its subsequent amendments. The study was approved by the Institutional Ethics Committee of Yantai Yuhuangding Hospital. Written informed consent was obtained from each patient.