Epigenetic Alterations Beyond CpG Islands in Periodontitis: In Silico Study of DNA Methylation Data.

OBJECTIVES

To classify differentially methylated CpG sites in periodontitis based on methylation shift direction and location relative to CpG islands.

MATERIAL AND

METHODS

We integrated DNA methylation data from 65 gingival samples (31 periodontitis, 34 healthy) from three GEO datasets. Preprocessing included normalization using R pipelines. Differential DNA methylation analysis was performed with the limma package. CpG sites (adjusted p < 0.05) were further evaluated using an exploratory threshold-crossing framework based on group-mean β-values (cutoff = 0.5) identifying Control low → Periodontitis high and Control high → Periodontitis low shifts, annotated with Illumina 450k references. Top 20 CpG-associated genes were selected by absolute logFC. Gene Ontology (GO) enrichment was conducted using clusterProfiler.

RESULTS

We identified 164,400 differentially methylated CpG sites, including CpGs exhibiting Control low → Periodontitis high and Control high → Periodontitis low shifts, and mapped to gene features. Nearly half of methylation alterations (47.77%) were in OpenSea regions, with additional signals in CpG islands and promoter-proximal regions. GO enrichment analysis revealed that differentially methylated genes in periodontitis were associated with epithelial and organ morphogenesis, including epithelial tube morphogenesis, epithelial cell proliferation, and cell fate commitment. Enriched cellular components were membrane- and vesicle-associated structures, while molecular functions were transmembrane transporter activity, ion channel activity, and DNA-binding transcription factor-related functions. These findings suggest that DNA methylation alterations in periodontitis are linked to epigenetic regulation of epithelial structure and tissue remodeling, with implications for immune-related cellular processes. Among the top 20 genes exhibiting β-threshold-based directionality changes, 13 were significantly enriched in GO terms: 8 in the Control low → Periodontitis high group (DGKA, CHMP6, RUNX2, CDKL1, SLC6A3, CFTR, HLA-C, LRIG1), 5 in the Control high → Periodontitis low group (AP2A2, PARD3, PADI2, FOXK1, RGS1).

CONCLUSIONS

This study provides an exploratory, direction-aware summary framework of DNA methylation in periodontitis, highlighting functional changes beyond CpG islands and identifying candidate genes for further functional validation.