Abstract
BACKGROUND: This study aimed to investigate the role of integrated stress response (ISR)-related biomarkers in periodontitis (PD). METHODS: Transcriptomic data related to PD were obtained from public databases. A bioinformatics approach combined with machine learning techniques was used to identify ISR-associated molecular markers involved in PD pathogenesis and to validate their expression patterns. Pathway enrichment analyses and immune landscape characterization were performed to elucidate the molecular mechanisms of these markers in PD progression. Single-cell RNA sequencing (scRNA-seq) was employed to resolve cellular heterogeneity and examine the expression patterns of candidate biomarkers. Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) assays were conducted to validate the expression profiles. RESULTS: BTG2, DERL3, FOS, HSPA13, and YOD1 were identified as potential PD biomarkers. Among them, BTG2, DERL3, FOS, and HSPA13 were co-enriched in the "osteoclast differentiation" pathway. DERL3 showed the strongest positive correlation with plasma cells and the strongest negative correlation with resting dendritic cells (|cor| > 0.3, P < 0.05). scRNA-seq analysis highlighted T cells as the key population. During T cell differentiation, BTG2 expression initially increased, then decreased, followed by a subsequent rise in the mid-to-late stages; DERL3 expression exhibited a transient increase before returning to baseline; and FOS expression increased gradually throughout the process. RT-qPCR results confirmed that the expression levels of BTG2, DERL3, FOS, and HSPA13 were significantly upregulated, while YOD1 expression was downregulated in the PD group (P < 0.05), which was consistent with the database-predicted patterns. CONCLUSION: This study integrated bulk and single-cell RNA-seq analyses to identify BTG2, DERL3, FOS, HSPA13, and YOD1 as PD biomarkers, with T cells as the central cell type, providing novel diagnostic insights for PD.