Skip to main content
Download PDF

Identification and analysis of neutrophil extracellular trap-related genes in periodontitis via bioinformatics and experimental verification

BMC Oral Health volume 25, Article number: 1294 (2025) Cite this article

Abstract

AbstractSection Background

Emerging evidence highlights the significant role of neutrophil extracellular traps (NETs) in periodontitis, though the precise mechanisms remain insufficiently understood. This study intends to investigate the comprehensive effects of NET-related genes (NRGs) on periodontitis by bioinformatic analysis.

AbstractSection Methods

The microarray datasets were obtained from the Gene Expression Omnibus (GEO) database. Differentially expressed NRGs (DE-NRGs) were identified and functionally annotated. Then, machine learning algorithms were exploited to screen hub NRGs, and a predictive model was constructed based on these hub NRGs. Moreover, the expression level of CXCR4, one of the hub NRGs, was experimentally validated.

AbstractSection Results

Eighty-three DE-NRGs were identified and mainly correlated with multiple periodontitis-related pathways. Then, a diagnostic NRG signature based on 7-hub NRGs (LPAR3, CXCR4, F3, MAPK7, KCNN3, SYK, and HIF1A) was constructed using two different machine learning algorithms. The diagnostic NRG signature demonstrated favorable predictive efficacy, with an AUC of 0.929 in the training and 0.936 in the validation cohorts. The mouse periodontitis model verified that CXCR4 and HIF1A was markedly upregulated in periodontitis tissues.

AbstractSection Conclusion

This study reveals that NRGs hold great potential as a robust and promising parameter for assessing periodontitis diagnosis. Targeting NRGs could represent a potential direction for future research into periodontitis treatment.

AbstractSection Clinical trial number

Not applicable.

Peer Review reports

Introduction

Periodontitis is a chronic, multifaceted infectious condition that gradually deforms the supportive structures surrounding the teeth [1, 2]. This disease is highly prevalent, impacting nearly 50% of individuals worldwide [3]. Periodontitis is characterized by gingival inflammation and irreversible alveolar bone loss. While being a chronic inflammatory disease, periodontitis is an ideal model for understanding the immune response to pathogens. Previous research has shown that pathogenic microorganisms like Porphyromonas gingivalis (P. gingivalis) and Aggregatibacter actinomycetemcomitans can disrupt the symbiotic state of dental biofilm, causing dysbiosis and contributing to disease development [4]. Inflammation associated with periodontitis can lead to the activation of immune pathways that are critical for controlling bacterial infections. Moreover, pioneering evidence indicates a mutually reinforcing relationship between periodontitis and systemic diseases like diabetes, cardiovascular disease, and multiple sclerosis, thereby underscoring the importance of understanding the immune response in both local and systemic contexts [2, 5,6,7]. Therefore, it is urgent to further explore the underlying mechanisms of periodontitis pathogenesis to reduce the burden of periodontal disease and its mutually affecting systemic diseases.

Neutrophil extracellular traps (NETs) are complex structures composed of DNA, histones, and antimicrobial proteins that are expelled by neutrophils, forming a net-like arrangement. These structures are associated with a variety of human diseases, encompassing malignant tumors, infectious diseases, atherosclerosis, and autoimmune disorders, including rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) [8, 9]. Recent in vitro and in vivo studies have linked NETs to periodontitis, in addition to their key role in systemic diseases [10,11,12,13]. Evidence indicates that NETs formation begins with the recognition of stimuli like P. gingivals, a key pathogen in the onset and advancement of periodontitis [13]. Our prior research demonstrated that P. gingivals induce oxidative stress in high glucose conditions via the Ca2+-PKC-MEK-ERK-NADPH oxidase-ROS pathway, which in turn promotes NETs formation [14]. Recently, NETs have been shown to induce mucosal inflammation and tooth-supporting bone destruction by upregulating IL-17/Th17 responses, as evidenced by both experimental studies and clinical validation [10]. Nevertheless, the detailed role of NETs in periodontitis remains elusive.

With the large number of RNA-profiling datasets and sophisticated bioinformatics analytic pipelines available, multiple NETs-related signatures have been developed in several human diseases [15, 16]. These bioinformatic-based signatures have shown diagnosis or prognosis power and translational potentials. Thorough and extensive investigations into the relationship between periodontitis and neutrophil extracellular traps (NETs) are still lacking. Recognizing the critical role of NETs in the onset and advancement of periodontitis, we aimed to establish a novel diagnostic signature associated with NETs for periodontitis through an integrative bioinformatics strategy that incorporated machine learning techniques. The expression of a NETs-related gene (NRG) of interest was further experimentally validated in the periodontitis model.

Materials and methods

Data collection and processing

Figure 1 displays the flow chart of this study. This study utilized microarray datasets of periodontal diseases (GSE10334, GSE16134, GSE23586, GSE173078) sourced from the GEO database (https://www.ncbi.nlm.nih.gov/geo/) [17,18,19,20,21]. Data preprocessing was performed using the Robust Multi-array Average (RMA) normalization method implemented in the “affy” and “limma” packages in R, and P-values were adjusted for multiple hypothesis testing using the Benjamini-Hochberg procedure to control the false discovery rate (FDR). Furthermore, we employed the ComBat algorithm from the “sva” package in R to remove batch effects prior to data integration. Detailed information concerning these datasets is listed in the Supplementary Table 1. The NETs gene sets included in this study were sourced from prior study (Supplementary Table 2) [22,23,24,25,26].

Fig. 1
figure 1

The workflow of this study

Full size image

Identification of differentially expressed NRGs

In the GSE10334 dataset, differentially expressed genes (DEGs) were discerned between normal tissues and those affected by periodontitis [27]. The identification of DEGs was carried out using the ‘limma’ package, with an adjusted P-value cutoff set at 0.05. Subsequently, the identified DEGs were compared with the gene sets associated with NETs to isolate the differentially expressed genes related to NETs between the normal and periodontitis tissues.

Functional enrichment analysis

The genes identified as differentially expressed between healthy and periodontitis-affected tissues were subjected to functional enrichment analysis. This analysis included Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment assessments, which were conducted using the ‘clusterProfiler’ package. Additionally, Gene Set Enrichment Analysis (GSEA) was utilized to evaluate and compare the biological processes associated with the differentially expressed NRGs.

Signature generated from machine learning-based integrative approaches

To establish a consensus on highly precise and consistent genes associated with NETs, we employed two machine learning algorithms to evaluate the previously recognized DEGs. LASSO regression serves as a linear regression technique that incorporates a regularization component aimed at mitigating overfitting and facilitating feature selection by diminishing the coefficients of less critical features to zero [28]. Conversely, the support vector machine recursive feature elimination (SVM-RFE) algorithm leverages weighted vectors obtained from SVM to systematically remove features, thus enhancing classification accuracy among distinct groups [29]. Furthermore, we employed two complementary machine learning algorithms—LASSO regression (10-fold cross-validation) and SVM-RFE (Five-fold cross-validation) and a fixed random seed to ensure reliable feature selection and control for model complexity. By intersecting the highest-ranked genes identified through the LASSO and SVM-RFE methodologies, we successfully pinpointed seven gene signatures pertinent to NETs. Subsequently, the risk score and diagnostic efficacy of these seven NRG signatures were further validated using the training cohort (GSE10334) and validation cohort (GSE16134).

Unsupervised clustering of periodontitis patients

To conduct unsupervised clustering of patients with periodontitis, we utilized the ‘ConsensusClusterPlus’ package [30]. We applied agglomerative clustering employing a Spearman correlation distance metric, with 80% of the samples resampled across ten iterations. The ideal number of clusters was ascertained through the analysis of the empirical cumulative distribution function plot. Following this, we executed a principal component analysis (PCA) plot utilizing the ‘stats’ package in R.

Immune cell infiltration analysis

The ssGSEA algorithm was employed to analyze RNA-seq data from normal and periodontitis tissues, inferring the relative proportions of immune-infiltrated cells [31]. Additionally, two more algorithms, CIBESORT [32] and MCPcounter [33] were employed to assess the percentage of immune-infiltrated cells and to corroborate the presence of neutrophils within periodontitis tissues.

Establishment of periodontitis murine model and micro-CT scanning

The Institutional Animal Care and Use Committee (IACUC) of Nanjing Medical University conducted a thorough review and approved the animal experimentation protocols (Approval ID: IACUC-2301021). All animal experiments comply with the ARRIVE guidelines and carried out in accordance with the National Research Council’s Guide for the Care and Use of Laboratory Animals. Male C57BL/6 mice (7–8 weeks) were obtained from the Animal Core Facility of Nanjing Medical University (Nanjing, China). The ten mice were randomly assigned to one of two groups (5 mice per group): the control group and the periodontitis group. Mice in the control group received an oral inoculation of 2% carboxy-methylcellulose (CMC), while those in the periodontitis group were orally administered 109 colony-forming units (CFU) of P. gingivalis W83, suspended in 2% CMC. Following the oral inoculation, the mice were subjected to a fasting regimen and deprived of water for 2 h. After 8 weeks, the mice were euthanized via the intraperitoneal injection of an overdose of sodium pentobarbital (100–150 mg/kg), and their maxillae were excised. Subsequently, the specimens underwent formalin fixation for 24 h before being prepared for micro-computed tomography (CT) scanning using a vivaCT80 system (SCANCO Medical, Switzerland). The maxilla and dental structures were reconstructed in three-dimensional space employing SkyScan software (Bruker).

Hematoxylin-eosin and immunohistochemistry staining

Specimens of periodontitis were subjected to formalin fixation and subsequently embedded in paraffin, followed by sectioning into four µm-thick slices for staining purposes. Upon confirmation of the pathological diagnosis by a senior pathologist at our institute, immunohistochemical (IHC) staining of the tissue sections was executed per established protocols. The slides underwent deparaffinization, and heat-induced epitope retrieval (HIER) was performed utilizing EDTA for 15 minutes at a temperature of 95℃. The sections were then incubated with the primary antibody, CXCR4 (#ab181020, Abcam, 1:200). Following the reaction with secondary antibodies and the application of a 3,3’-Diaminobenzidine (DAB) detection kit (#ab64238, Abcam), the sections were counterstained with Mayer’s hematoxylin. The stained sections were subsequently visualized and photographed using an upright microscope (Lecia).

RNA isolation, quantitative reverse transcription-quantitative PCR

The procedures for total RNA extraction and quantitative reverse transcription polymerase chain reaction (qRT-PCR) assays were conducted following previously established protocols [34, 35]. The primers used were listed as follows: mouse CXCR4: Forward: 5’-CCTCCTCCTGACTATACCTGACTTC-3’, Reverse: 5’-ACACCACCATCCACAGGCTATC-3’; mouse HIF1A: Forward 5’-CCTGCACTGAATCAAGAGGTTGC-3’, Reverse 5’-CCATCAGAAGGACTTGCTGGCT-3’; mouse MPO: Forward: 5’-AGGGCCGCTGATTATCTACAT-3’, Reverse: 5’-CTCACGTCCTGATAGGCACA-3’; mouse ELANE: Forward: 5’-CCTTGGCAGACTATCCAGCC-3’; Reverse: 5’-GACATGACGAAGTTCCTGGCA-3’; mouse GAPDH: Forward: 5’-CTCATGACCACAGTCCATGC-3’, Reverse: 5’-CACATTGGGGGTAGGAACAC-3’.

Statistical analysis

The statistical evaluations carried out in this study utilized R software (version 4.1.2). For the differential analysis of continuous variables, the student’s t-test was applied for data that followed a normal distribution, while the Kruskal-Wallis and Wilcox. test was employed for data that did not adhere to normality. For multiple-comparison analyses (e.g., DEGs, GO/KEGG, GSVA, immune infiltration), adj. P < 0.05 was used to control the false discovery rate. All P values were analyzed as two-tailed, with a threshold of P < 0.05 deemed statistically significant.

Results

Identification of DE-NETs in periodontitis

Figure 1 illustrates the study’s flow chart. To clarify the involvement of NETs in the advancement of periodontal disease, we analyzed the expression levels of 137 NRGs in tissues affected by periodontitis in comparison to healthy controls, utilizing the GSE10334 dataset. As displayed in Figs. 2A and 10,409 genes were differentially expressed between the control and diseased samples. After intersecting the DEGs with NRGs, a total of 83 genes were identified. Among them, 38 genes were significantly upregulated in the periodontitis tissues, whereas 45 genes were significantly downregulated (Fig. 2B). The detailed differently expressed NRGs were listed in Supplementary Table 3. Moreover, to explore the internal correlation between these differently expressed NRGs, a correlation analysis was carried out. The relationship among these differently expressed NRGs was shown in Fig. 2C. Furthermore, the chromosomal locations of the 83 differently expressed NRGs were analyzed and displayed in the Fig. 2D.

Fig. 2
figure 2

Identification and characterization of differentially expressed NRGs in periodontitis. (A) Volcano plot displaying differentially expressed genes between normal gingival and periodontitis samples. Red and blue dots represent significantly upregulated and downregulated genes, respectively. (B) Heatmap illustrating the expression patterns of 83 differentially expressed NET-related genes (NRGs). (C) Correlation matrix depicting the pairwise relationships among the 83 differentially expressed NRGs. Positive correlations are shown in blue and negative correlations in red. (D) Circos plot showing the chromosomal locations of the 83 differentially expressed NRGs across the human genome

Full size image

Functional enrichment analysis of differently expressed NRGs

To explore the potential biological functions and pathways of these differentially expressed NRGs, GO enrichment analyses focusing on the Biological Process (BP) category were conducted. GO analysis revealed that differently expressed NRGs were predominantly enriched in pathways related to cytokine-mediated signaling pathway, leukocyte migration, myeloid leukocyte activation, neutrophil activation, neutrophil migration, positive regulation of interleukin-8 production, regulation of inflammatory response, and et al. (Fig. 3A, B). GSEA analysis revealed that differently expressed NRGs were mainly enriched in defense response, humoral immune response, cytokine-cytokine receptor interaction, interleukin 4 and interleukin 13 signaling, and inflammatory response, corroborating the reliability of GO enrichment results (Fig. 3C, D). These pathways exhibited a strong association with inflammation, bone resorption, and immune activation, thereby reinforcing the significant involvement of NETs in the onset and advancement of periodontal disease.

Fig. 3
figure 3

Functional enrichment analysis of differentially expressed necroptosis-related genes (NRGs) in periodontitis. A, B. Gene Ontology (GO) enrichment analysis of differentially expressed NRGs, highlighting significantly enriched Biological Process (BP) terms between healthy and periodontitis gingival samples. The dot plot visualizes gene ratio, gene count, and adjusted p-values (A). And illustrating functional relationships among significantly enriched terms based on gene overlap and semantic similarity (B). C, D. Gene Set Enrichment Analysis (GSEA) based on the MSigDB_c5_GO gene sets and MSigDB_c2_ Curated gene sets, identifying significantly enriched pathways associated with NRGs in periodontitis compared to normal controls

Full size image

Construction and validation of periodontitis predictive model

Following the identification and functional enrichment analysis of differently expressed NRGs in periodontitis, we developed a predictive model for periodontitis using two machine learning algorithms. As illustrated in Fig. 4A, B, LASSO regression analysis was utilized to obtain 16 genes from 83 DE-NETs as potential periodontitis biomarkers. The SVM-RFE algorithm identified 19 periodontitis-related NRGs based on the smallest error (highest accuracy) (Fig. 4C, D). By intersecting the 16 NRGs identified through the LASSO algorithm with the 19 NRGs filtered via the SVM-RFE algorithm, we identified 7 hub NRGs for periodontitis patients (Fig. 4E). A set of seven genes associated with NETs, specifically LPAR3, CXCR4, F3, MAPK7, KCNN3, SYK, and HIF1A, was identified for the development of a predictive model for periodontitis through the application of LASSO regression analysis. The regression coefficients corresponding to these genes were recorded as follows: -0.946891053 for LPAR3, 0.375294426 for CXCR4, -0.046654034 for F3, -0.105432455 for MAPK7, 0.239100560 for KCNN3, -0.078897325 for SYK, and 0.106394952 for HIF1A (Fig. 4F).

Fig. 4
figure 4

Construction and validation of a 7-gene necroptosis-related predictive model using machine learning algorithms. A, B. Identification of 16 candidate differentially expressed NRGs via LASSO regression with 10-fold cross-validation. the optimal λ value was indicated. C, D. Selection of 19 key differentially expressed NRGs using Support Vector Machine-Recursive Feature Elimination (SVM-RFE) algorithm. Five-fold cross-validation error rates for different numbers of features. the top 19 features yielded the highest performance cross-validation accuracy across feature subsets. E. Venn diagram showing the 7 overlapping core NRGs identified by both LASSO and SVM-RFE algorithms. F. LASSO coefficient profiles of the selected genes incorporated into the predictive model. G-J. Distribution of risk scores based on the 7-gene model in the training cohort (GSE10334) (G) and validation cohort (GSE16134) (H); statistical significance was evaluated using Wilcox. Test. Receiver operating characteristic (ROC) curves showing the predictive performance of the 7-gene model in the training set (I) and validation set (J)

Full size image

Then, the risk scores of all samples in the periodontitis and healthy control groups were calculated and compared. Interestingly, as shown in Fig. 4G, the risk score of the periodontitis group was significantly higher than the healthy control group (P<0.0001). The ROC curve analysis indicated an AUC of 0.929 for the predictive model in the training cohort (GSE10334) (Fig. 4I). We further validated the model using an independent cohort (GSE16134) and an additional combined cohort (GSE23586 and GSE173078). As expected, the risk scores were significantly higher in periodontitis samples (P < 0.0001 and P = 0.011, respectively), and the model demonstrated robust predictive performance with AUCs of 0.936 and 0.852 (Fig. 4H, J; Supplementary Fig. 1). In summary, our predictive model for periodontitis, which is founded on the expression levels of seven NRGs, demonstrated strong and consistent predictive capability across independent cohorts.

Stratification of periodontitis patients according to differently expressed NRGs

To enhance our understanding of periodontal disease, we performed a cluster analysis employing the consensus clustering approach, which was founded on the expression profiles of 83 differentially expressed NRGs. As illustrated in Fig. 5A-D, the Delta Area Plot revealed that the most significant increase in the relative change of the CDF area occurred at k = 3, beyond which the gain plateaued—indicating diminishing returns in clustering stability. This approach is widely accepted for evaluating cluster robustness and was used to guide the choice of k = 3. Thus, based on the consensus clustering algorithm, the periodontal disease samples were divided into three clusters, namely the C1 cluster (n = 68), C2 cluster (n = 78), and C3 cluster (n = 37). Principal component analysis (PCA) revealed the expression patterns of differently expressed NRGs across the three periodontitis subtypes (C1 ~ 3) (Fig. 5E). The functional enrichment among these three different clusters was performed using the GSVA algorithm (Supplementary Fig. 2). We examined the expression patterns of the seven hub NRGs (LPAR3, CXCR4, F3, MAPK7, KCNN3, SYK, and HIF1A) across the three clusters. As shown in Fig. 5G, H, the expression of most hub NRGs (6 of 7) was significantly different between the three different clusters except for the HIF1A. Moreover, the risk score of NETs in different clusters was also investigated. As displayed in Fig. 5F, among the three periodontitis clusters, cluster C1 showed the lowest risk score, whereas cluster C2 displayed the highest risk score, which further confirmed the success of our three clustering subgroups.

Fig. 5
figure 5

Identification of three distinct NET-related gene clusters across periodontitis samples. (A) Unsupervised consensus clustering matrix and optimal clusters. (B) Consensus cumulative distribution function (CDF) curves for k = 2 to 7. C, D. Delta area plot showing the relative change in the area under the CDF curve, indicating the most stable clustering at k = 3 (C). Tracking plot demonstrating sample assignments across different k values, highlighting cluster stability (D). E. Principal component analysis (PCA) plot depicting the separation of samples into three clusters based on NRG expression profiles. F. Comparison of risk scores among the three clusters, indicating significant differences Statistical analysis was performed using the Kruskal–Wallis test. G, H. Heatmap (G) and violin plot (H) showing differentially expression landscape of the 7 NRGs among the three clusters by using Kruskal-Wallis test. Ns: none significance; *:P < 0.05; **:P < 0.01; ***: P < 0.001

Full size image

Variations of immune characteristics, inflammatory mediators and lipid metabolic pathways level between the clusters

Emerging evidence indicates that NETs are crucial in the innate immune system and may expedite the progression of various diseases, such as periodontitis, by altering the immune microenvironment. Therefore, we further determine the immune characteristics between the three clusters by leveraging CIBERSORT, ssGSEA, as well as MCPcounter algorithms. As illustrated in Fig. 6A and Supplementary Fig. 3B, the results of ssGSEA immune cell infiltration analysis revealed that most immune-related cells, including T cells, CD8+ T cells, B cells, NK cells, macrophages, dendritic cells, and neutrophils, were significantly higher in the cluster C2 as compared with cluster C1 or C3. Moreover, analogous immune cell infiltration was observed in the results, which were identified via CIBERSORT and MCPcounter algorithms (Fig. 6A and Supplementary Fig. 3A, C). Moreover, we further confirmed that the abundance of neutrophils between these different periodontitis clusters for its critical role in the NETs. The neutrophil infiltration was notably elevated in clusters C2 and C3 in contrast with cluster C1 (Fig. 6B-D). Furthermore, Fig. 6E displayed the relationship between 7 hub NRGs and the correlations between immune-related cells, 7 hub NRGs, and the three different periodontitis clusters. Our results suggested that patients of cluster C2 were correlated with higher inflammatory and immune responses, and the production of NETs may be more. The correlation analysis revealed that several hub NRGs, particularly CXCR4, SYK, and HIF1A, were significantly associated with multiple immune cell types, including neutrophils, macrophages, B cells, and T cell subsets. Notably, KCNN3 exhibited negative correlations with cytotoxic and NK cells (Supplementary Fig. 4). These results suggest that NRG expression is closely linked to the immune microenvironment in periodontitis.

Fig. 6
figure 6

Comparison of the immune infiltration landscape between the three distinct clusters. A. Heatmap showing the immune cell infiltration profiles between the three clusters using the CIBERSORT (22 immune cell types), ssGSEA (24 immune cell types), and MCP counter algorithms (10 immune cell types), respectively. B-D. Comparison of the infiltration level of neutrophils between clusters based on CIBERSORT (B), ssGSEA (C), and MCPcounter (D) algorithms. Statistical significance assessed using the Kruskal–Wallis test. E. Correlation analysis among 24 immune cell types, 7 differentially expressed NRGs, and the three clusters. Top: pairwise Pearson correlations between 24 immune cells. Bottom left: pairwise Pearson correlations between NRGs. Network: Mantel test-based associations between NRGs, immune cells, and cluster subtypes (C1–C3), with line thickness and color representing correlation strength and significance

Full size image

As shown in Supplementary Fig. 5A, the three clusters displayed distinct expression patterns of representative inflammatory mediators, including IL1B, IL6, MMP9, NLRP3, and NOS2, indicating distinct inflammatory profiles among the clusters. Notably, Cluster 2 and Cluster 3 showed significantly higher expression of IL1B, IL6, and MMP9 compared to cluster 1, indicating a stronger pro-inflammatory profile (Supplementary Fig. 5B). To further characterize the metabolic distinctions among the clusters, we conducted GSVA (Gene Set Variation Analysis) of hallmark metabolic pathways. The results (Supplementary Fig. 5C, D) revealed that: Cluster 2 showed notably lower scores in oxidative phosphorylation and fatty acid degradation, but higher activity in glycosphingolipid biosynthesis pathways. Cluster 3 was characterized by elevated scores in glycerophospholipid metabolism, indicating distinct metabolic reprogramming. These findings indicate that the clusters differ not only in immune infiltration, but also in inflammatory signaling and metabolic profiles, indicating underlying biological heterogeneity in periodontitis.

Validate the upregulated expression of CXCR4 and HIF1A in periodontitis via qRT-PCR and immunohistochemistry

CXCR4 is involved in the immune response of periodontitis and is associated with NETs in periodontitis [36]. We speculate whether CXCR4 plays a role in the progression of periodontitis, and for this, we verified its expression in various publicly available periodontitis datasets and the mouse periodontitis model. Firstly, the samples from training cohort (GSE10334), validation cohort (GSE16134), as well as GSE23586 dataset were bisected into periodontitis group and healthy control group. Then the mRNA expression level of CXCR4 between the periodontitis group and healthy control group was further measured in the three different periodontitis datasets. Consistent with previous results, the CXCR4 expression level was notably overexpressed in the periodontitis tissues as compared with their normal counterparts (Fig. 7A-C). Then, to further validate the upregulation of CXCR4 in periodontal disease, a classical periodontal disease mice model that could largely mimic the progression of human periodontal disease was established. As shown in Fig. 7D, E, the images of micro-CT, as well as H&E staining, proved the successful establishment of the model. Moreover, RT-qPCR results also revealed that CXCR4 and HIF1A were significantly upregulated in the periodontitis tissues (Fig. 7F, Supplementary Fig. 6A). Meanwhile, we evaluated the relationship between CXCR4, HIF1A and several NETs markers (ELANE and MPO) via RT-qPCR assay in the mouse periodontal disease tissues. The results suggested that the expression level of CXCR4 and HIF1A were positively correlated with ELANE and MPO (Fig. 7G, H, Supplementary Fig. 6B, C). Furthermore, the protein abundance of CXCR4 in periodontitis tissues was determined via immunohistochemical staining. In line with the results of the former RT-qPCR assay, the expression level of CXCR4 was also significantly elevated in the periodontitis tissues as compared to normal gingival samples (Fig. 7I, J). Overall, we observed that CXCR4 and HIF1A were markedly overexpressed in periodontal disease tissues, and positively associated with the abundance of NETs markers (ELANE and MPO), which was in line with the results of bioinformatic analysis.

Fig. 7
figure 7

RT-qPCR and immunohistochemical staining of CXCR4 in periodontitis. A-C. Comparison of the CXCR4 mRNA expression between periodontitis and healthy gingival samples in GSE10334 (A), GSE16134 (B), and GSE23586 (C) datasets. Statistical significance was evaluated using Wilcox. ns: none significance; *: P < 0.05; **: P < 0.01; ***: P < 0.001. D. The flow diagram of RT-qPCR and immunohistochemical staining experimental validation. E. Representative 3D reconstruction of Micro-CT image, bucco-palatal sagittal slices of Micro-CT images, and H&E staining of maxillary alveolar bone surrounding the maxillary second molars in healthy control and periodontitis samples. F. The mRNA expression level of CXCR4 in healthy control and periodontitis samples was determined using RT-qPCR assay. Students’ t test. G, H. The correlation between ELANE (G)/MPO (H) and CXCR4 mRNA levels was compared. Pearson correlation coefficient. I, J. Immunohistochemical staining (I) and quantitative data (J) of CXCR4 in healthy control and periodontitis samples. Students’ t test. ns, none significance; *, P < 0.05; **, P < 0.01; ***, P < 0.001

Full size image

Discussion

Periodontitis ranks among the most prevalent diseases globally. This condition not only results in the loss of teeth due to the resorption of alveolar bone but is also significantly associated with numerous systemic diseases, such as diabetes [5,6,7, 37, 38]. Research indicates that NETs could play a role in the advancement of periodontitis; however, the exact underlying mechanisms remain incompletely elucidated. Here, in this study, we identified seven candidate diagnostic hub NRGs, as well as their relationship with immune cells in periodontitis via comprehensive bioinformatic analyses. Moreover, a hub gene of interest, CXCR4, was further experimental validated in murine periodontitis models.

Neutrophils release web-like formations composed of DNA, histones, and antimicrobial proteins, known as NETs, in reaction to infection or inflammatory stimuli [39, 40]. The process of NETs formation is a specialized form of cell death that allows neutrophils to release their DNA and antimicrobial proteins to combat infections. However, excessive NETs formation can also contribute to inflammatory diseases and tissue damage since it could also trigger pro-inflammatory activity and prolong the inflammatory response [8, 41]. As previously stated, periodontitis is characterized as a chronic, multifactorial inflammatory condition. In addition to the gradual deterioration of connective tissue attachments and alveolar bone, it also has detrimental effects on aesthetics, chewing function, and overall systemic health [1, 2]. Kaneko et al. have demonstrated that the circulating levels of NETs in the blood were positively correlated with advanced periodontitis [42]. Moreover, the concentration of NETs was dramatically downregulated after periodontal management [43]. The results indicate that NETs may serve as a potential biomarker for periodontitis; nevertheless, the precise function of NETs in the context of periodontitis remains predominantly unclear.

Consequently, we conducted an extensive bioinformatic analysis of RNA-seq to investigate the involvement of NRGs in periodontitis. After variation analysis, a total of 83 differentially expressed NRGs were identified between normal gingival tissues and periodontitis tissues. Functional enrichment analysis revealed significant enrichment in immune-regulated pathways, including cytokine-mediated signaling, neutrophil activation, positive regulation of interleukin-8 production, and inflammatory regulation. Previous studies have shown that type I interferon (IFN) promotes plasmacytoid dendritic cell (pDC) activation through NET production, contributing to the pathogenesis of Mycobacterium tuberculosis [44]. Furthermore, NETs contribute to inflammatory bone loss by upregulating IL-17/Th17 responses in periodontitis [10]. In addition, IL-8 promotes neutrophil recruitment and activation, potentially triggering excessive NET release [45]. The functional enrichment analysis in our study were in line with the previous research and suggested significant involvement of IL-8, IL-17/Th17 signaling, and neutrophil activation pathways closely linked to NET formation. These findings suggest that NET dysregulation may mediate inflammation-driven tissue damage in periodontitis.

Although periodontitis has been widely investigated and positively treated, the outcome of periodontitis is still unsatisfactory. To further promote the accuracy of periodontitis’ early diagnosis, several machine learning algorithms including LASSO and SVM-RFE were performed. Using machine learning, we identified 7 hub genes (hub-NRGs) associated with periodontitis: CXCR4, KCNN3, HIF1A, F3, SYK, MAPK7, and LPAR3. Mounting evidence has suggested that these genes are highly correlated to the NETs. Among them, CXCR4 has shown the ability to enhance the NETs formation which further triggers the incidence of skin inflammation, sepsis-induced acute lung injury, and environment-driven allergic asthma [46,47,48]. The activation of KCNN3 (gene encoding SK3 channel) and mitochondrial reactive oxygen species (mtROS) has been reported to mediate calcium-activated NOX-independent NETosis [49]. Moreover, McInturff et al. have revealed that the mTOR modulates NET formation by regulating the posttranscriptional expression of HIF1A [50], which was further confirmed in recent studies [51, 52]. SYK could promote the excessive formation of NETs through several different pathways, such as PSGL-1/SYK/Ca2+/PAD4 and SYK-ERK-NF-kappaB [53, 54]. Genetic knockout of LPAR3 in mice may increase NET formation, indicating LPAR3’s protective role in sepsis [55]. Our correlation analysis revealed that several NRGs, such as CXCR4, SYK, and HIF1A, were positively associated with neutrophils, macrophages, and B cells, suggesting their potential roles in modulating immune cell infiltration in periodontitis. These associations support previous reports linking these genes to immune cell recruitment and activation. However, the precise mechanisms through which these genes influence specific immune populations warrant further investigation using single-cell or spatial transcriptomics.

To assess the predictive capacity of the NRGs, a risk model comprising seven NRGs was developed utilizing LASSO regression within the training dataset. The resulting NRGs risk score was found to be considerably elevated in tissues affected by periodontitis when contrasted with healthy tissues within both the training and validation datasets. The ROC curve illustrates the model’s high predictive performance, exhibiting values of 0.929 for the training dataset and 0.936 for the validation dataset. This study provides a comprehensive and integrative analysis of the expression patterns and diagnostic relevance of NRGs in periodontitis. Moreover, aside from its implications for periodontal disease, the NETs risk model demonstrates significant applicability to a range of other inflammatory conditions. For example, the NRGs risk model established for osteoarthritis exhibited commendable predictive performance [56]. In addition, Tao and colleagues established a NET-score model specifically designed for anti-neutrophil cytoplasmic antibody-associated glomerulonephritis (ANCA-GN), which effectively identifies patients at high risk [57]. The investigations revealed that predictive models utilizing NETs exhibit a commendable level of accuracy in diagnosing periodontitis as well as other inflammatory conditions.

Consensus clustering is categorized as an unsupervised algorithm designed to discern potential groupings by analyzing the inherent characteristics of the data [30]. This method identified three distinct molecular subtypes in the periodontitis dataset. Among the three periodontitis clusters, cluster C1 showed the lowest risk score and cluster C2 displayed the highest risk score, which further confirms the success of our three clustering subgroups. Furthermore, our study revealed clear biological heterogeneity among the identified periodontitis subtypes. Cluster 2 and Cluster 3 demonstrated the highest immune cell infiltration, especially neutrophils, along with elevated expression of pro-inflammatory mediators such as IL1B, IL6, and MMP9, suggesting a highly inflammatory and immune-active phenotype potentially associated with enhanced NETs formation. In contrast, Cluster 1 showed relatively low immune and inflammatory activity. Metabolically, Cluster 2 exhibited suppressed oxidative phosphorylation and fatty acid degradation but upregulated glycosphingolipid biosynthesis, while Cluster 3 showed enhanced glycerophospholipid metabolism. These findings highlight distinct immune-metabolic profiles across clusters, suggesting that immune activation and metabolic reprogramming may jointly contribute to periodontitis heterogeneity and progression.

As mentioned above, CXCR4 can enhance the formation of NETs in multiple diseases. Moreover, in this study, CXCR4 was identified as a hub NRG in periodontitis with the highest LASSO coefficients. We validated the expression level of CXCR4 and its relationship with NETs biomarkers (ELANE and MPO) in a mouse periodontitis model. Bioinformatic analysis revealed a significant elevation of CXCR4 in periodontitis tissues compared to normal tissues. Moreover, the expression of CXCR4 in periodontitis mice was also markedly correlated with the NETs biomarkers, ELANE and MPO. These findings are in line with recent reports highlighting the role of CXCR4 in inflammatory diseases [58, 59]. In addition, we also focused on HIF1A, another representative hub gene in our model. HIF1A, a key transcription factor activated under hypoxic and inflammatory conditions, has been previously implicated in modulating inflammatory response in periodontitis [60, 61]. Our analysis revealed a significant upregulation of HIF1A in periodontitis tissues, and was positively correlated with NETs-associated markers ELANE and MPO, suggesting that HIF1A may also participate in promoting NETs formation in the context of periodontitis. Together, these results underscore the potential regulatory role of both CXCR4 and HIF1A in periodontitis, supporting their value as important targets in diagnostic or therapeutic strategies.

There are certain limitations inherent to our study. Firstly, the study did not incorporate additional datasets for the identification of differentially expressed genes (DEGs), and the application of more sophisticated bioinformatics techniques and machine learning approaches could potentially lead to a more precise identification of key genes. Secondly, the data utilized were sourced from retrospective studies, which may be subject to selection bias. Future longitudinal studies aimed at evaluating the fluctuations in NRG expression over time in individuals suffering from periodontitis could shed light on their involvement in disease progression and the possibility of reversibility. Thirdly, the relatively limited sample sizes of the available public datasets and the inherent complexity of machine learning algorithms may still pose a risk of overfitting that cannot be entirely ruled out. Future studies using larger, multicenter cohorts and prospective validation will be essential to further confirm the robustness and generalizability of our predictive model. In addition, our model despite showed strong predictive performance, it was not directly compared with traditional diagnostic markers. Future studies are needed to benchmark its added value in clinical settings. Lastly, while the expression levels of CXCR4 were experimentally validated and CXCR4 was included due to its reported role in neutrophil migration and NET formation, it is not exclusively expressed by neutrophils. CXCR4 is also expressed by stromal cells and lymphocytes, which may outnumber neutrophils in gingival tissue samples. This lack of cell-type specificity may confound the interpretation of bulk RNA-based immune profiling and NRG-based risk scoring. Future studies employing single-cell RNA sequencing or spatial transcriptomics will be necessary to resolve the cellular origin of CXCR4 expression and better clarify its specific role in the context of periodontitis.

Conclusion

In summary, we identified seven candidate diagnostic hub NETs-related genes via a systematic bioinformatics pipeline and revealed their correlations with the immune microenvironment in periodontitis. This study offers novel perspectives on the dysregulation of NETs that contribute to the pathogenesis of periodontitis, as well as identifying potential biomarkers for diagnostic forecasting.

Data availability

The relevant datasets (GSE10334, GSE16134, GSE23586, GSE173078) utilized in this study were publicly available and obtained from Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/). Further inquiries can be directed to the corresponding authors.

References

  1. Slots J. Periodontitis: facts, fallacies and the future. Periodontol 2000. 2017;75(1):7–23.

    Article  PubMed  Google Scholar 

  2. Sanz M, Del Marco A, Jepsen S, Gonzalez-Juanatey JR, D’Aiuto F, Bouchard P, et al. Periodontitis and cardiovascular diseases: consensus report. J Clin Periodontol. 2020;47(3):268–88.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Trindade D, Carvalho R, Machado V, Chambrone L, Mendes JJ, Botelho J. Prevalence of periodontitis in dentate people between 2011 and 2020: A systematic review and meta-analysis of epidemiological studies. J Clin Periodontol. 2023;50(5):604–26.

    Article  PubMed  Google Scholar 

  4. Suvan J, Leira Y, Moreno Sancho FM, Graziani F, Derks J, Tomasi C. Subgingival instrumentation for treatment of periodontitis. A systematic review. J Clin Periodontol. 2020;47(Suppl 22):155–75.

    Article  PubMed  Google Scholar 

  5. Ozmeric N, Elgun S, Kalfaoglu D, Pervane B, Sungur C, Erguder I, et al. Interaction between hypertension and periodontitis. Oral Dis. 2024;30(3):1622–31.

    Article  PubMed  Google Scholar 

  6. Lalla E, Papapanou PN. Diabetes mellitus and periodontitis: a Tale of two common interrelated diseases. Nat Rev Endocrinol. 2011;7(12):738–48.

    Article  CAS  PubMed  Google Scholar 

  7. Sheu JJ, Lin HC. Association between multiple sclerosis and chronic periodontitis: a population-based pilot study. Eur J Neurol. 2013;20(7):1053–9.

    Article  PubMed  Google Scholar 

  8. Mutua V, Gershwin LJ. A review of neutrophil extracellular traps (NETs) in disease: potential Anti-NETs therapeutics. Clin Rev Allergy Immunol. 2021;61(2):194–211.

    Article  CAS  PubMed  Google Scholar 

  9. Lee KH, Kronbichler A, Park DD, Park Y, Moon H, Kim H, et al. Neutrophil extracellular traps (NETs) in autoimmune diseases: A comprehensive review. Autoimmun Rev. 2017;16(11):1160–73.

    Article  CAS  PubMed  Google Scholar 

  10. Kim TS, Silva LM, Theofilou VI, Greenwell-Wild T, Li L, Williams DW et al. Neutrophil extracellular traps and extracellular histones potentiate IL-17 inflammation in periodontitis. J Exp Med. 2023;220(9).

  11. Onuora S. NETs implicated in periodontitis-associated bone loss. Nat Rev Rheumatol. 2023;19(8):463.

    Article  CAS  PubMed  Google Scholar 

  12. Oliveira SR, de Arruda JAA, Schneider AH, Bemquerer LM, de Souza RMS, Barbim P, et al. Neutrophil extracellular traps in rheumatoid arthritis and periodontitis: contribution of PADI4 gene polymorphisms. J Clin Periodontol. 2024;51(4):452–63.

    Article  CAS  PubMed  Google Scholar 

  13. Magan-Fernandez A, O’Valle F, Abadia-Molina F, Munoz R, Puga-Guil P, Mesa F. Characterization and comparison of neutrophil extracellular traps in gingival samples of periodontitis and gingivitis: A pilot study. J Periodontal Res. 2019;54(3):218–24.

    Article  CAS  PubMed  Google Scholar 

  14. Tong Y, Xin Y, Fu L, Shi J, Sun Y. Excessive neutrophil extracellular trap formation induced by Porphyromonas gingivalis lipopolysaccharide exacerbates inflammatory responses in high glucose microenvironment. Front Cell Infect Microbiol. 2023;13:1108228.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Li Q, Chen W, Li Q, Mao J, Chen X. A novel neutrophil extracellular trap signature to predict prognosis and immunotherapy response in head and neck squamous cell carcinoma. Front Immunol. 2022;13:1019967.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Zhang Y, Guo L, Dai Q, Shang B, Xiao T, Di X et al. A signature for pan-cancer prognosis based on neutrophil extracellular traps. J Immunother Cancer. 2022;10(6).

  17. Demmer RT, Behle JH, Wolf DL, Handfield M, Kebschull M, Celenti R, et al. Transcriptomes in healthy and diseased gingival tissues. J Periodontol. 2008;79(11):2112–24.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Papapanou PN, Behle JH, Kebschull M, Celenti R, Wolf DL, Handfield M, et al. Subgingival bacterial colonization profiles correlate with gingival tissue gene expression. BMC Microbiol. 2009;9:221.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Chen Y, Wang H, Yang Q, Zhao W, Chen Y, Ni Q, et al. Single-cell RNA landscape of the osteoimmunology microenvironment in periodontitis. Theranostics. 2022;12(3):1074–96.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Abe D, Kubota T, Morozumi T, Shimizu T, Nakasone N, Itagaki M, et al. Altered gene expression in leukocyte transendothelial migration and cell communication pathways in periodontitis-affected gingival tissues. J Periodontal Res. 2011;46(3):345–53.

    Article  CAS  PubMed  Google Scholar 

  21. Kim H, Momen-Heravi F, Chen S, Hoffmann P, Kebschull M, Papapanou PN. Differential DNA methylation and mRNA transcription in gingival tissues in periodontal health and disease. J Clin Periodontol. 2021;48(9):1152–64.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Wu J, Zhang F, Zheng X, Zhang J, Cao P, Sun Z, et al. Identification of renal ischemia reperfusion injury subtypes and predictive strategies for delayed graft function and graft survival based on neutrophil extracellular trap-related genes. Front Immunol. 2022;13:1047367.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Zheng WC, Lin F, Qiu QR, Wu YP, Ke ZB, Chen SH, et al. Identification of neutrophil extracellular traps (NETs)-related molecular clusters in prostate cancer: implications for predicting biochemical recurrence. Int Immunopharmacol. 2025;146:113908.

    Article  CAS  PubMed  Google Scholar 

  24. Chen N, He D, Cui J. A neutrophil extracellular traps signature predicts the clinical outcomes and immunotherapy response in head and neck squamous cell carcinoma. Front Mol Biosci. 2022;9:833771.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Shang B, Hu Z, Xie R, Wu J, Qu W, Zhang W, et al. Predictive value of neutrophil extracellular traps in neoadjuvant chemotherapy for Muscle-Invasive bladder cancer. Mol Carcinog. 2025;64(2):305–16.

    Article  CAS  PubMed  Google Scholar 

  26. Aierken Y, Tan K, Liu T, Lv Z. Prognosis and immune infiltration prediction in neuroblastoma based on neutrophil extracellular traps-related gene signature. Sci Rep. 2025;15(1):5343.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43(7):e47.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Tibshirani R. Regression shrinkage and selection via the Lasso. J Roy Stat Soc: Ser B (Methodol). 1996;58(1):267–88.

    Article  Google Scholar 

  29. Gold C, Sollich P. Model selection for support vector machine classification. Neurocomputing. 2003;55(1–2):221–49.

    Article  Google Scholar 

  30. Wilkerson MD, Hayes DN. ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking. Bioinformatics. 2010;26(12):1572–3.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Bindea G, Mlecnik B, Tosolini M, Kirilovsky A, Waldner M, Obenauf AC, et al. Spatiotemporal dynamics of intratumoral immune cells reveal the immune landscape in human cancer. Immunity. 2013;39(4):782–95.

    Article  CAS  PubMed  Google Scholar 

  32. Newman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015;12(5):453–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Becht E, Giraldo NA, Lacroix L, Buttard B, Elarouci N, Petitprez F, et al. Estimating the population abundance of tissue-infiltrating immune and stromal cell populations using gene expression. Genome Biol. 2016;17(1):218.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Wu Y, Chang X, Yang G, Chen L, Wu Q, Gao J, et al. A physiologically responsive nanocomposite hydrogel for treatment of head and neck squamous cell carcinoma via Proteolysis-Targeting chimeras enhanced immunotherapy. Adv Mater. 2023;35(12):e2210787.

    Article  PubMed  Google Scholar 

  35. Wu Y, Wang Y, Diao P, Zhang W, Li J, Ge H, et al. Therapeutic targeting of BRD4 in head neck squamous cell carcinoma. Theranostics. 2019;9(6):1777–93.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Qiu W, Guo R, Yu H, Chen X, Chen Z, Ding D et al. Single-cell atlas of human gingiva unveils a NETs-related neutrophil subpopulation regulating periodontal immunity. J Adv Res. 2025;72:287–301.

  37. Yadalam PK, Arumuganainar D, Natarajan PM, Ardila CM. Artificial intelligence-powered prediction of AIM-2 inflammasome sequences using Transformers and graph attention networks in periodontal inflammation. Sci Rep. 2025;15(1):8733.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Yadalam PK, Chatterjee S, Natarajan PM, Ardila CM. Comparison of light gradient boosting and logistic regression for interactomic hub genes in Porphyromonas gingivalis and Fusobacterium nucleatum-induced periodontitis with alzheimer’s disease. Front Oral Health. 2025;6:1463458.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Papayannopoulos V. Neutrophil extracellular traps in immunity and disease. Nat Rev Immunol. 2018;18(2):134–47.

    Article  CAS  PubMed  Google Scholar 

  40. Wigerblad G, Kaplan MJ. Neutrophil extracellular traps in systemic autoimmune and autoinflammatory diseases. Nat Rev Immunol. 2023;23(5):274–88.

    Article  CAS  PubMed  Google Scholar 

  41. Herre M, Cedervall J, Mackman N, Olsson AK. Neutrophil extracellular traps in the pathology of cancer and other inflammatory diseases. Physiol Rev. 2023;103(1):277–312.

    Article  CAS  PubMed  Google Scholar 

  42. Kaneko C, Kobayashi T, Ito S, Sugita N, Murasawa A, Nakazono K, et al. Circulating levels of carbamylated protein and neutrophil extracellular traps are associated with periodontitis severity in patients with rheumatoid arthritis: A pilot case-control study. PLoS ONE. 2018;13(2):e0192365.

    Article  PubMed  PubMed Central  Google Scholar 

  43. White P, Sakellari D, Roberts H, Risafi I, Ling M, Cooper P, et al. Peripheral blood neutrophil extracellular trap production and degradation in chronic periodontitis. J Clin Periodontol. 2016;43(12):1041–9.

    Article  CAS  PubMed  Google Scholar 

  44. Kotov DI, Lee OV, Fattinger SA, Langner CA, Guillen JV, Peters JM, et al. Early cellular mechanisms of type I interferon-driven susceptibility to tuberculosis. Cell. 2023;186(25):5536–e55535522.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Luo Y, Fraser L, Jezykowski J, Gupta NA, Miethke AG, Taylor SA et al. Interleukin 8-CXCR2-mediated neutrophil extracellular trap formation in biliary Atresia associated with neutrophil extracellular trap-induced stellate cell activation. Hepatology. 2024. Epub ahead of print.

  46. Radermecker C, Sabatel C, Vanwinge C, Ruscitti C, Marechal P, Perin F, et al. Locally instructed CXCR4(hi) neutrophils trigger environment-driven allergic asthma through the release of neutrophil extracellular traps. Nat Immunol. 2019;20(11):1444–55.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Liu D, Xiao M, Zhou J, Wang P, Peng J, Mao W, et al. PFKFB3 promotes sepsis-induced acute lung injury by enhancing NET formation by CXCR4(hi) neutrophils. Int Immunopharmacol. 2023;123:110737.

    Article  CAS  PubMed  Google Scholar 

  48. Chen J, Bai Y, Xue K, Li Z, Zhu Z, Li Q, et al. CREB1-driven CXCR4(hi) neutrophils promote skin inflammation in mouse models and human patients. Nat Commun. 2023;14(1):5894.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Douda DN, Khan MA, Grasemann H, Palaniyar N. SK3 channel and mitochondrial ROS mediate NADPH oxidase-independent NETosis induced by calcium influx. Proc Natl Acad Sci U S A. 2015;112(9):2817–22.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  50. McInturff AM, Cody MJ, Elliott EA, Glenn JW, Rowley JW, Rondina MT, et al. Mammalian target of rapamycin regulates neutrophil extracellular trap formation via induction of hypoxia-inducible factor 1 alpha. Blood. 2012;120(15):3118–25.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. Li M, Weng L, Yu D, Yang G, Hao J. Increased formation of neutrophil extracellular traps induced by autophagy and identification of autophagy-related biomarkers in systemic lupus erythematosus. Exp Dermatol. 2024;33(1):e14881.

    Article  CAS  PubMed  Google Scholar 

  52. Burczyk G, Cichon I, Kolaczkowska E. Itaconate suppresses formation of neutrophil extracellular traps (NETs): involvement of hypoxia-inducible factor 1alpha (Hif-1alpha) and heme oxygenase (HO-1). Front Immunol. 2022;13:864638.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  53. Xu Q, Shi M, Ding L, Xia Y, Luo L, Lu X, et al. High expression of P-selectin induces neutrophil extracellular traps via the PSGL-1/Syk/Ca(2+)/PAD4 pathway to exacerbate acute pancreatitis. Front Immunol. 2023;14:1265344.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  54. Lu CH, Li KJ, Wu CH, Shen CY, Kuo YM, Hsieh SC et al. The FcgammaRIII engagement augments PMA-stimulated neutrophil extracellular traps (NETs) formation by granulocytes partially via cross-talk between Syk-ERK-NF-kappaB and PKC-ROS signaling pathways. Biomedicines. 2021;9(9).

  55. Pei S, Xu C, Pei J, Bai R, Peng R, Li T, et al. Lysophosphatidic acid receptor 3 suppress neutrophil extracellular traps production and thrombosis during sepsis. Front Immunol. 2022;13:844781.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Luan T, Yang X, Kuang G, Wang T, He J, Liu Z, et al. Identification and analysis of neutrophil extracellular Trap-Related genes in osteoarthritis by bioinformatics and experimental verification. J Inflamm Res. 2023;16:3837–52.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. Tao M, He Y, Li L, Li Y, Liao W, Nie H, et al. Identification and validation of immune-associated NETosis subtypes and biomarkers in anti-neutrophil cytoplasmic antibody associated glomerulonephritis. Front Immunol. 2023;14:1177968.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  58. Kim AR, Bak EJ, Yoo YJ. Distribution of neutrophil and monocyte/macrophage populations induced by the CXCR4 inhibitor AMD3100 in blood and periodontal tissue early after periodontitis induction. J Periodontal Res. 2022;57(2):332–40.

    Article  CAS  PubMed  Google Scholar 

  59. Nagashima H, Shinoda M, Honda K, Kamio N, Hasuike A, Sugano N, et al. CXCR4 signaling contributes to alveolar bone resorption in Porphyromonas gingivalis-induced periodontitis in mice. J Oral Sci. 2017;59(4):571–7.

    Article  PubMed  Google Scholar 

  60. Zhou M, Hu H, Han Y, Li J, Zhang Y, Tang S, et al. Long non-coding RNA 01126 promotes periodontitis pathogenesis of human periodontal ligament cells via miR-518a-5p/HIF-1alpha/MAPK pathway. Cell Prolif. 2021;54(1):e12957.

    Article  CAS  PubMed  Google Scholar 

  61. Wang Y, Lee BH, Yang Z, Ho TJ, Ci H, Jackson B, et al. Chewing-Activated TRPV4/PIEZO1-HIF-1alpha-Zn axes in a rat periodontal complex. J Dent Res. 2025;104(4):398–407.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

Not applicable.

Funding

This work is financially supported, in whole or in part, by the National Natural Science Foundation of China (82403833), Natural Science Foundation of Jiangsu Province (BK20240517), Postdoctoral Fellowship Program of China Postdoctoral Science Foundation (GZC20240739), Jiangsu Funding Program for Excellent Postdoctoral Talent (2024ZB381), China Postdoctoral Science Foundation (2025T180495, 2024M751492), The Natural Science Foundation of the Jiangsu Higher Education Institutions of China (24KJB320010), Jiangsu Province Capability Improvement Project through Science, Technology and Education-Jiangsu Provincial Research Hospital Cultivation Unit (YJXYYJSDW4), Jiangsu Provincial Medical Innovation Center (CXZX202227), Anhui Province Engineering Research Center for Dental Materials and Application (2024AMCD02), Program for Excellent Sci-tech Innovation Teams of Universities in Anhui Province (2023AH010073).

Author information

Author notes

  1. Miao Yu and Zhenqi Ye contributed equally to this work.

Authors and Affiliations

  1. Department of Oral and Maxillofacial Surgery, The Affiliated Stomatological Hospital of Nanjing Medical University, Jiangsu, 210029, P.R. China

    Yaping Wu

  2. Department of General Dentistry, The Affiliated Stomatological Hospital of Nanjing Medical University, Jiangsu, 210029, P.R. China

    Xiang Wu

  3. State Key Laboratory Cultivation Base of Research, Prevention and Treatment for Oral Diseases, Jiangsu, 210029, P.R. China

    Miao Yu, Zixin Ye, Yaping Wu & Xiang Wu

  4. Jiangsu Province Engineering Research Center of Stomatological Translational Medicine, Nanjing Medical University, Jiangsu, 210029, P.R. China

    Miao Yu, Yaping Wu & Xiang Wu

  5. Department of Stomatology, The First Affiliated Hospital of Wannan Medical College, Wuhu, 241001, P.R. China

    Zhenqi Ye

Authors
  1. Miao Yu
    View author publications

    Search author on:PubMed Google Scholar

  2. Zhenqi Ye
    View author publications

    Search author on:PubMed Google Scholar

  3. Zixin Ye
    View author publications

    Search author on:PubMed Google Scholar

  4. Yaping Wu
    View author publications

    Search author on:PubMed Google Scholar

  5. Xiang Wu
    View author publications

    Search author on:PubMed Google Scholar

Contributions

M. Yu: Study conceptualization, data analysis, visualization, writing-original draft, writing review, and editing. ZQ. Ye: Data analysis, visualization, writing-original draft and methodological discussions. ZX. Ye: Data analysis and methodological discussions. Y. Wu: data analysis, visualization and writing-original draft. X. Wu: Study conceptualization, supervision, project administration, writing-original draft.

Corresponding authors

Correspondence to Yaping Wu or Xiang Wu.

Ethics declarations

Ethics approval and consent to participate

The Institutional Animal Care and Use Committee (IACUC) of Nanjing Medical University conducted a thorough review and approved the animal experimentation protocols (Approval ID: IACUC-2301021). All animal experiments comply with the ARRIVE guidelines and arried out in accordance with the National Research Council’s Guide for the Care and Use of Laboratory Animals and the American Veterinary Medical Association (AVMA) Guidelines for the Euthanasia of Animals (2020).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Yu, M., Ye, Z., Ye, Z. et al. Identification and analysis of neutrophil extracellular trap-related genes in periodontitis via bioinformatics and experimental verification. BMC Oral Health 25, 1294 (2025). https://doi.org/10.1186/s12903-025-06663-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12903-025-06663-2

Keywords

BMC Oral Health

ISSN: 1472-6831

Contact us