The potential crosstalk genes and molecular mechanisms between systemic lupus erythematosus and periodontitis

Abstract

Background: Several studies have demonstrated an increased risk of periodontitis (PD) among patients diagnosed with systemic lupus erythematosus (SLE). However, the underlying common mechanism between them remains incompletely understood. Accordingly, the aim of this study is to examine diagnostic biomarkers and potential therapeutic targets for SLE and PD by leveraging publicly accessible microarray datasets and transcriptome analysis.

Method: Datasets pertaining to SLE and PD were retrieved from the Gene Expression Omnibus (GEO) database, and subsequently analyzed for differentially expressed genes (DEGs). Key gene modules were identified through weighted gene co-expression network analysis (WGCNA), and shared genes were obtained by overlapping key genes between DEGs and WGCNA. These shared genes were subsequently subjected to Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses, leading to the establishment of a Protein-Protein Interaction (PPI) network. Random forest (RF) and Least Absolute Shrinkage and Selection Operator (Lasso) regression were employed to identify key hub genes. Receiver operating characteristic (ROC) curves were generated using a new validation dataset to evaluate the performance of candidate genes. Finally, levels of immune cell infiltration in SLE and PD were assessed using CIBERSORTx.

Results: A total of 50 core genes were identified between the genes screened by WGCNA and DEGs. Functional enrichment analysis revealed that these genes are primarily associated with the PI3K-Akt and B-cell receptor signaling pathways. Additionally, using machine learning algorithms and ROC curve analysis, a total of 8 key genes (PLEKHA1, CEACAM1, TNFAIP6, TCN2, GLDC, GNG7, LY96, VCAN) were identified Finally, immune infiltration analysis highlighted the significant roles of neutrophils, monocytes, plasma cells, and gammadelta T cells (γδ T cells) in the pathogenesis of both SLE and PD.

Conclusion: This study identifies 8 hub genes that could potentially serve as diagnostic markers for both SLE and PD, highlighting the importance of VCAN and LY96 in diagnosis. Moreover, the involvement of the PI3K-Akt signaling pathway in both diseases suggests its significant role. These identified key genes and signaling pathways lay the groundwork for deeper comprehension of the interplay between SLE and PD.

Keywords: LY96; PI3K-Akt signaling pathway; VCAN; periodontitis; systemic lupus erythematosus.

FIGURE 1

Research study flowchart.

FIGURE 2

Analysis of differential gene expression. (A, B) Volcano plots illustrated the presence of DEGs in datasets GSE72326 and GSE16134. (C) Overlapping DEGs of GSE72326 and GSE16134.

FIGURE 3

WGCNA of GSE72326 and GSE16134 datasets. (A) Network topology analysis results for dataset GSE72326. (B) Network topology analysis results for dataset GSE16134. (C) Gene modules and clustering results for dataset GSE72326. (D) Gene modules and clustering results for dataset GSE16134. (E, F) Heatmap depicting the correlation between module genes and diseases. (G) Genes overlapping in positively correlated modules.

FIGURE 4

A comprehensive analysis of the shared genes was conducted. (A) GO enrichment analysis. (B) KEGG signaling pathway. (C) PPI network.

FIGURE 5

Screening and validating diagnostic biomarkers for SLE and PD. (A) The thirty most significant genes identified using the RF algorithm. (B, C) Cross-validation plot and coefficient contraction plot of LASSO regression analysis. (D) The intersecting genes selected by the RF and LASSO algorithms. (E, F) The ROC curves of PLEKHA1 and CEACAM1.

FIGURE 6

Analysis of immunoinfiltration. (A, B) The proportion of immune cells between patients with the disease and control groups. (C, D) The relationship between shared central genes and immune cells.