Synergistic Induction of Neutrophilic Inflammatory Programs by Staphylococcus aureus and Cigarette Smoke in Airway Epithelial Cells

Abstract

Staphylococcus aureus (SA) colonization and cigarette smoking are both implicated in the pathogenesis of chronic airway disease, yet their combined effects on epithelial responses remain unclear. We investigated transcriptomic changes in human bronchial epithelial cells (BEAS-2B) following co-exposure to SA and cigarette smoke extract (CSE). RNA sequencing revealed that combined SA+CSE co-exposure was associated with a marked increase in differentially expressed genes, compared with single exposures. Functional enrichment and network analyses identified significant activation of pathways related to neutrophil migration, extracellular matrix remodeling, and inflammatory cascades, including TNF and IL-17 signaling. Key hub genes, notably CCL20, CXCL1, CXCL8, and IL-24, showed marked synergistic upregulation, which was validated by quantitative RT-PCR. These findings suggest that SA and cigarette smoke co-exposure is associated with a transcriptomic profile suggestive of neutrophilic inflammation. The involvement of IL-24 and IL-17 signaling suggests potential pathways linking bacterial colonization and smoking to airway inflammation and remodeling.

INTRODUCTION

Airway epithelial cells serve as a crucial barrier against exposure to environmental factors, such as allergens, irritants, bacteria, and viruses in the bronchi. In addition, they play a pivotal role in immune responses by secreting various cytokines. However, chronic exposure to these stressors can disrupt epithelial junctions, promote allergic sensitization, or cause persistent changes to cellular structure and function (1). Consequently, chronic airway diseases and tissue remodeling may develop.

Staphylococcus aureus (SA) is a gram-positive bacterium well-known for acute infection, but it is also a common commensal organism that colonizes the skin and upper airways (2). SA secretes various virulence factors, including toxins, proteases, and staphylococcal enterotoxin (SEs), which can impair airway epithelial integrity, alter permeability, increase susceptibility to allergic sensitization, and induce type 2 (T2) inflammation. Exposure to SA has been associated with eosinophilic inflammation, T2 cytokine production, and reduced lung function, including fixed airflow obstruction (3,4,5,6).

Cigarette smoke (CS) exposure is a well-known risk factor for fixed airflow obstruction. Notably, however, the interaction between CS and SA appears to drive a distinct pathology. In a community-population study, IgE sensitization to SE was more common in current smokers, and the IgE level positively correlated with smoking pack-years in life (7). A recent longitudinal study conducted before and after a smoking cessation program found that smoking was positively associated with nasal SA load and inversely associated with antibacterial immune responses (8). Mechanistically, CS exposure can increase the permeability of human bronchial epithelium, thereby promoting allergen penetration and sensitization (9). Furthermore, concomitant exposure to CS and SA has been reported to increase biofilm formation and invasion of the airway epithelium, inducing alterations in bacterial virulence and secretome (10,11,12,13). These changes may compromise airway epithelial integrity, induce neutrophilic inflammation, and contribute to chronic airway disease and airway remodeling (14).

Despite these findings, the specific transcriptomic landscapes distinguishing the individual and synergistic effects of SA and CS exposure on airway epithelial cells remain poorly characterized. Thus, this study aimed to investigate the transcriptomic alterations induced by SA, cigarette smoke extract (CSE), and their combination, in order to elucidate the molecular mechanisms underlying airway inflammation and remodeling.

MATERIALS AND METHODS

CSE preparation

The CS extraction protocol was optimized based on previous published methods (15). Briefly, a 25 ml pipette fitted with a variable flow pump (C.B.S Scientific, Solana Beach, CA, USA) was filled with 10 ml ice-cold PBS and connected to one reference cigarette (3R4F, Lexington, KY, USA) without a filter. One cigarette was fixed horizontally to be burned, and the mainstream smoke was aspirated at a constant flow rate (one cigarette per 8 min). Smoke from 10 cigarettes was passed through 10 ml of solvent to collect CSE; and this concentration was defined as 100% (one cigarette per 1 ml). The 100% CSE was then filtered through a 0.2 μm filter, aliquoted, and stored at −80°C until use.

SA culture supernatant preparation

SA Rosenbach strain (American Type Culture Collection [ATCC] 29213) was used to obtain the culture broth. BD Bacto™ Tryptic Soy Broth (TSB) served as the basal culture medium. A single colony of SA was inoculated into 10 ml of TSB and cultured with shaking at 200 rpm for over 16 h. The culture was then diluted into fresh TSB or TSB containing 10% CSE, adjusted to an OD of 0.1–0.15 and incubated at 37°C for 8 h. Following incubation, the culture was centrifuged and the supernatant was filtered through a 0.2 μm filter for use.

Cell culture and treatment

BEAS-2B cells (ATCC, Rockville, MD, USA) were cultured in BEGM media (BEGM Bullet kit No. CC-3170; Lonza, Basel, Switzerland) at 37°C with 5% CO₂. Cells were seeded in 12-well plates at a density of 1×10⁵ cells/ml. Upon reaching 90% confluence, cells were treated for 24 h with one of the following: medium only (Control), CSE (1:10,000 dilution), SA culture supernatant (1:1,000 dilution), or combined (SA+CSE) supernatant (1:1,000 dilution of SA supernatant derived from 10% CSE-exposed SA culture) (Fig. 1A). To determine the optimal working dilution, we performed a dose-response titration (ranging from 1:10 to 1:1,000) based on IL-6 and IL-8 secretion levels. The 1:1,000 dilution was selected as the maximum non-stimulatory concentration for the SA supernatant alone, where inflammatory cytokine levels remained comparable to the control. In contrast, the combined SA+CSE supernatant retained significant pro-inflammatory activity at this dilution. Total RNA was extracted 24 h post-treatment.

Figure 1. Experimental design and transcriptomic landscapes of airway epithelial cells. (A) Schematic diagram of the experimental design; CSE was prepared by bubbling smoke from 3R4F reference cigarettes into PBS. SA was cultured in TSB or TSB supplemented with 10% CSE to generate normal and CSE-primed SA supernatants, respectively. BEAS-2B cells were divided into 4 groups: Control (media only), CSE (1:10,000 dilution of CSE), SA (1:1,000 dilution of normal SA supernatant), and SA+CSE (1:1,000 dilution of CSE-primed SA supernatant). After 24 h of exposure, total RNA was extracted for transcriptomic profiling (RNA-Seq) and qRT-PCR validation. (B) Venn diagram showing the overlap of DEGs among CSE, SA, and combined treatment groups (0: up-regulated, 0: down-regulated). (C) Hierarchical clustering heatmap of DEGs across all experimental groups.

RNA sequencing and data processing

Total RNA was isolated using TRIzol reagent (Ambion, San Francisco, CA, USA; Thermo Fisher Scientific, Waltham, MA, USA). RNA quality and quantity were assessed using the TapeStation4000 System (Agilent Technologies, Amstelveen, The Netherlands) and the ND-2000 Spectrophotometer (Thermo Fisher Scientific, Dover, DE, USA), respectively. Libraries were prepared using the CORALL RNA-Seq V2 Library Prep Kit (LEXOGEN, Inc., Vienna, Austria), which included mRNA isolation with the Poly (A) RNA Selection Kit (LEXOGEN, Inc.), cDNA synthesis and shearing, Illumina index addition, and PCR enrichment. Library quality and concentration were evaluated using the TapeStation HS D1000 Screen Tape (Agilent Technologies) and a StepOne Real-Time PCR System (Life Technologies, Inc., Carlsbad, CA, USA). Paired-end 100 bp sequencing was performed on a NovaSeq 6000 (Illumina, Inc., San Diego, CA, USA).

Quality control of raw sequencing data was performed using FastQC (16). Adapter sequences and low-quality reads were removed using fastp (17). Trimmed reads were then mapped to the reference genome using STAR (18). Read quantification was processed using Salmon (19). Read counts were normalized based on the trimmed mean of M-values + counts per million normalization method using EdgeR (20).

Differential expression gene (DEG) analysis

Only genes with a mean count value greater than 2 were retained for subsequent analysis. Differential protein-coding genes were screened based on fold change and statistical significance. The screening criteria were set to p-value <0.05 and |log₂fold change|>2. The expression patterns of DEGs across different groups and samples were visualized using hierarchical cluster analysis.

Gene Ontology (GO) and functional enrichment analyses of the DEGs

GO enrichment analysis, along with Kyoto Encyclopedia of Genes and Genomes (KEGG) and Reactome pathway enrichment analyses, was performed using g:Profiler software (https://biit.cs.ut.ee/gprofiler/gost) for the identified DEGs in each group (Supplementary Fig. 1) (21). GO analysis was conducted to identify enriched terms associated with molecular functions, biological processes, and cellular components. Additionally, biological pathway analysis was carried out using the KEGG and Reactome databases to elucidate the functional pathways involving these DEGs.

Protein-protein interaction (PPI) and gene-gene interaction (GGI) networks analysis

PPI network analysis was performed using the STRING database (http://string-db.org/) for the DEGs. The minimum required interaction score was set to 0.4 (medium confidence), and networks were visualized using Cytoscape software (Version 3.10.3). To identify hub genes, CytoHubba plug-in was used. The GeneMANIA database (https://genemania.org/) was utilized to construct GGI networks, as well as to analyze co-expression and enrichment pathways interacting with hub genes (22). GeneMANIA further demonstrates interactions such as co-expression, co-localization, pathways, predicted physical interactions, shared protein domains, and genetic interactions. Each interaction type was displayed in the network visualization.

Quantitative RT-PCR (qRT-PCR) validation

qRT-PCR was used to confirm the differential expression of selected genes. Extracted total RNA was reverse transcribed into cDNA using reverse transcriptase (Transcriptor First Strand cDNA Synthesis Kit; Roche, Penzberg, Germany). The resulting cDNA was analyzed on the Roche LightCycler 480 system using SYBR Green dye (LightCycler 480 SYBR Green I Master; Roche). Primer sequences were designed using Primer-Blast (https://www.ncbi.nlm.nih.gov/tools/primer-blast/index.cgi), and certain primers were sourced from OriGene Technologies. GAPDH was used as the reference gene, and relative gene expression was calculated using the 2^−ΔCt method.

Statistical analyses and visualization

Data mining and graphic visualization were performed using ExDEGA (Ebiogen Inc., Seoul, Korea) and GraphPad Prism 10.01 software (GraphPad Software, San Diego, CA, USA). Pearson correlation analysis was used to evaluate the relationship between replicates within the same group. Principal component analysis (PCA) was utilized to distinguish between groups, and Student’s t-test was used for comparison between the 2 groups. A p-value of <0.05 was considered statistically significant.

RESULTS

Differential gene expression

Pearson correlation analysis confirmed high reproducibility within treatment groups (r>0.95) (Supplementary Fig. 1A). PCA and hierarchical clustering revealed distinct separation among 3 primary groups: media/CSE, SA, and SA+CSE (Supplementary Fig. 1B). Given the minimal separation observed between the media and CSE groups, the CSE group was selected as the reference control for all subsequent differential gene expression analyses.

Differential gene expression analysis identified distinct profiles across the treatment groups (Fig. 1B, Supplementary Tables 1, 2, 3). Compared to the CSE group, the combined SA+CSE treatment yielded 190 DEGs (155 upregulated and 35 downregulated). This represents a 6.8-fold increase compared to the SA group (28 DEGs; 27 upregulated, 1 downregulated), suggesting a potential synergistic effect at the transcriptional level. Furthermore, comparison of the SA+CSE group to the SA group identified 47 DEGs (42 upregulated and 5 downregulated, Fig. 1C), indicating additional transcriptional alterations associated with combined exposure.

GO enrichment analysis

GO enrichment analysis highlighted distinct molecular functions, biological processes and cellular components affected by SA and CSE treatment in BEAS-2B cells. In the SA group (vs. CSE group), enriched GO terms were primarily associated with stress response regulation, apoptotic processes, the cell cycle, and immune system processes (Fig. 2A, Supplementary Table 4). In contrast, compared to the CSE group, the SA+CSE group showed altered genes involved in response to external stimuli, including extracellular matrix (ECM) composition and organization, cell junction assembly, and TLR signaling pathways (Fig. 2A, Supplementary Table 5).

Figure 2. Functional enrichment and pathway analysis of synergistic responses. (A) Gene Ontology (BP, MF, CC) analysis of DEGs in (i) SA (vs. CSE), (ii) SA+CSE (vs. CSE), and (iii) SA+CSE (vs. SA), (B) KEGG and Reactome pathway analysis highlighting the enrichment of inflammatory signaling cascades, (i) CSE vs. SA, (ii) CSE vs. SA+CSE, and (iii) SA vs. SA+CSE.

BP, biological process; MF, molecular function; CC, cellular component.

Notably, comparison between the SA+CSE and SA groups revealed further modulation of GO terms liked to inflammatory responses, such as neutrophil migration and ECM remodeling. These findings suggest enhanced inflammatory responses and microenvironmental changes associated with the SA+CSE co-exposure (Fig. 2A, Supplementary Table 6).

KEGG and Reactome pathway analysis

KEGG and Reactome pathway enrichment analyses were performed to explore the molecular mechanisms underlying the effects of SA and CSE exposure. Compared to the CSE group, the SA group exhibited significant enrichment in inflammation-related pathways, including TNF signaling, IL-17 signaling, and their downstream mechanisms (Fig. 2B, Supplementary Table 4).

The SA+CSE treatment was associated with significant enrichment of multiple inflammatory signaling pathways, notably TNF, IL-17, NF-κB, TLR, and ECM-related pathways (vs. CSE group, Fig. 2B, Supplementary Table 5). Furthermore, compared to the SA group, the SA+CSE group showed enhanced activation of pathways related to inflammation (IL-10, IL-17 signaling, cytokine-cytokine receptor interaction, and TLR cascades) and tissue remodeling (ECM-receptor interaction and ECM organization) (Fig. 2B, Supplementary Table 6). This pattern suggests that combined exposure may be associated with enhanced inflammatory and tissue remodeling–related responses in airway epithelial cells.

Key targets involved in the SA+CSE treatment group identified by PPI and GGI networks

To investigate the key targets associated with pathways in the SA+CSE group, the 47 DEGs identified from the SA+CSE vs. SA comparison were analyzed using the STRING database to construct a PPI network (Fig. 3A).

Figure 3. Identification of hub genes via protein-protein interaction networks. (A) STRING network analysis and (B) GGI network analysis of the combined treatment group (vs. SA group).

The top 10 hub genes (CCL20, CXCL1, CXCL8, IL-1α IL-1β, IL-6, IL-24, TGF-β2, SERPINB2 and TLR4) were identified using the CytoHubba plugin in Cytoscape software. GeneMANIA was used to construct a GGI network to predict potential interactions between 10 hub genes and other targets. The GGI network analysis revealed a complex interplay between the hub genes and their interacting partners. Functional enrichment analysis further indicated significant enrichment of biological processes related to neutrophil migration (false discovery rate [FDR]=1.22e-19), positive regulation of cell-cell adhesion (FDR=2.58e-4), regulation of T cell proliferation (FDR=5.26e-3), regulation of tissue remodeling (FDR=8.03e-3), regulation of T cell-mediated immunity (FDR=1.67e-2), regulation of epithelial cell apoptotic process (FDR=1.67e-2), and regulation of epithelial-to-mesenchymal transition (EMT, FDR=2.44e-2) (Fig. 3B).

Verification of hub genes using qRT-PCR

To validate the transcriptomic findings, we conducted qRT-PCR on the identified hub genes across the different treatment groups (Fig. 4). The SA+CSE treatment significantly increased the mRNA levels of all hub genes compared to the CSE and single-exposure groups. These findings corroborate the RNA-Seq data and support a synergistic effect of SA and CSE co-exposure on the regulation of genes involved in inflammation, immune responses, and ECM remodeling in BEAS-2B cells.

Figure 4. Validation of key hub genes. Synergistic upregulation was confirmed by qRT-PCR in human bronchial epithelial cells. (A) CXCL1; (B) CCL20; (C) IL-1α; (D) IL-1β; (E) IL-6; (F) IL-8; (G) TLR4; (H) TGF-β2; (I) SERPINB2; (J) IL-24.

ns, no significance.

^*p<0.01, ^**p<0.001, ^***p<0.001, ^****p<0.0001.

DISCUSSION

This study investigated transcriptomic changes and cellular mechanisms underlying airway inflammation and remodeling associated with co-exposure to SA culture supernatant and CSE in human bronchial epithelial cells (BEAS-2B). Our findings demonstrate that exposure to supernatant from CSE-primed SA induces a robust inflammatory response and ECM alterations. Transcriptomic analyses, supported by GO, KEGG, and Reactome pathway enrichments, revealed alterations in epithelial cell junctions, ECM remodeling, and the activation of inflammatory cascades including TNF, IL-17, NF-κB, and TLR pathways. Network analysis identified pivotal hub genes such as CCL20, CXCL1, CXCL8, IL-1α, IL-1β, IL-6, IL-24, TGF-β2, SERPINB2, and TLR4, which were validated by qRT-PCR. These findings provide insights into molecular pathways associated with SA-CS interactions at the airway epithelial level and may contribute to understanding the progression of chronic airway diseases.

Previous studies have shown that SA exposure in bronchial epithelial cells stimulates inflammation and modulates epithelial cell adhesion, partly through neutrophil itaconate production and IL-8 secretion (23,24). Conversely, CS is known to activate pro-inflammatory pathways and disrupt cellular homeostasis, potentially leading to cell death and lung injury (25,26). While SA and CS individually contribute to airway pathology, their interaction may be particularly harmful. CS can compromise the epithelial barrier, increasing susceptibility to bacterial virulence, which in turn may exacerbate CS-induced damage. Evidence suggests that CS can alter the airway microbiome, favoring pathogenic bacteria and enhancing the virulence of SA (10,27). Consistent with these observations, our findings suggest that SA, in the context of CS exposure, is associated with enhanced expression of pro-inflammatory mediators compared to single exposures. This interaction may contribute to increased airway inflammation and remodeling, potentially influencing the severity and persistence of chronic airway diseases.

Notably, the transcriptomic profile of the SA+CSE group suggests a shift toward a neutrophilic phenotype. Coordinated changes in chemokines, cytokines (CXCL1, CXCL8), TLRs, and the specific enrichment of the IL-17 signaling pathway support a pattern consistent with sustained neutrophilic inflammation (28,29). Persistent neutrophil infiltration is a hallmark of severe, often steroid-resistant, airway disease and is associated with tissue damage (30). Furthermore, these inflammatory pathways are linked to airway remodeling, promoting the activation of fibroblasts and ECM deposition, particularly when TGF-β2 signaling is dysregulated (31,32). Together, these findings suggest that the SA+CSE combination may create a microenvironment that promotes both persistent inflammation and airway remodeling.

A key finding of this study is the marked increase of IL-24 in the SA+CSE group. IL-24 is a cytokine with pleiotropic effects, exhibiting both pro and anti-inflammatory properties, depending on the context (33). While in vivo studies in asthmatic mouse models have linked epithelial IL-24 and airway remodeling, its role in SA and CS co-exposure has been unexplored (34,35). Our GGI network analysis suggests a potential link between IL-24 and genes involved in neutrophil migration and EMT. These findings suggest that IL-24 may serve as a candidate biomarker or therapeutic target in SA- and CS-related chronic airway diseases or their exacerbations. Further investigation is warranted to determine if IL-24 plays a causal role in disease severity or reflects disease activity.

This study has several limitations. First, the sample size was relatively small, although consistency across replicates was supported by high Pearson correlation. Second, the BEAS-2B cell line may not fully recapitulate the complex in vivo environment, which involves 3D structural organization and diverse cell-cell interactions. Third, the use of a single SA strain may limit the generalizability. Fourth, the identified mechanisms require further validation using primary human bronchial epithelial cells and in vivo animal models to explore translational relevance. Finally, although the induction of chemoattractants (CXCL1, CXCL8, and CCL20) suggests neutrophil involvement, functional studies are warranted to confirm these phenotypic effects in vivo.

Despite these limitations, this study provides a comprehensive transcriptomic analysis of bronchial epithelial cells exposed to the combined effects of SA and CS. This unbiased approach offers new insights into gene regulation in a multi-factorial environment. By identifying key pathways, including IL-17 signaling, and molecules such as IL-24, our findings characterize early transcriptional changes that may contribute to the transition from acute inflammation to chronic airway remodeling.

In conclusion, our findings suggest that the combined presence of SA and CS can exert a synergistic inflammatory effect on airway epithelial cells, amplifying the expression of key pro-inflammatory cytokines and chemokines. This interaction may contribute to neutrophilic airway inflammation and promote airway remodeling. These findings provide a basis for future research into targeted therapies for patients with chronic airway diseases associated with smoking and bacterial colonization.

Abbreviations

ATCC

American Type Culture Collection

CS

cigarette smoke

CSE

cigarette smoke extract

DEG

differentially expressed gene

ECM

extracellular matrix

EMT

epithelial-to-mesenchymal transition

FDR

false discovery rate

GGI

gene-gene interaction

GO

Gene Ontology

KEGG

Kyoto Encyclopedia of Genes and Genomes

PCA

principal component analysis

PPI

protein-protein interaction

qRT-PCR

quantitative real-time polymerase chain reaction

SA

<italic>Staphylococcus aureus</italic>

SE

staphylococcal enterotoxin

T2

type 2

TSB

BD Bacto™ Tryptic Soy Broth

SUPPLEMENTARY MATERIALS

Supplementary Table 1

List of top 10 DEGs (SA vs. CSE)

Supplementary Table 2

List of top 10 DEGs (SA+CSE vs. CSE)

Supplementary Table 3

List of top 10 DEGs (SA+CSE vs. SA)

Supplementary Table 4

GO enrichment, KEGG and Reactome pathway analysis between CSE and SA groups

Supplementary Table 5

GO enrichment, KEGG and Reactome pathway analysis between CSE and SA+CSE groups

Supplementary Table 6

GO enrichment, KEGG and Reactome pathway analysis between SA and SA+CSE groups

Supplementary Figure 1

Quality control of RNA sequencing data. (A) Pearson’s correlation matrix within groups, (B) PCA plot of the 4 groups.