Differential DNA methylation and mRNA transcription in gingival tissues in periodontal health and disease

Abstract

Aim: We investigated differential DNA methylation in gingival tissues in periodontal health, gingivitis, and periodontitis, and its association with differential mRNA expression.

Materials and methods: Gingival tissues were harvested from individuals and sites with clinically healthy and intact periodontium, gingivitis, and periodontitis. Samples were processed for differential DNA methylation and mRNA expression using the IlluminaEPIC (850 K) and the IlluminaHiSeq2000 platforms, respectively. Across the three phenotypes, we identified differentially methylated CpG sites and regions, differentially expressed genes (DEGs), and genes with concomitant differential methylation at their promoters and expression were identified. The findings were validated using our earlier databases using HG-U133Plus2.0Affymetrix microarrays and Illumina (450 K) methylation arrays.

Results: We observed 43,631 differentially methylated positions (DMPs) between periodontitis and health, and 536 DMPs between gingivitis and health (FDR < 0.05). On the mRNA level, statistically significant DEGs were observed only between periodontitis and health (n = 126). Twelve DEGs between periodontitis and health (DCC, KCNA3, KCNA2, RIMS2, HOXB7, PNOC, IRX1, JSRP1, TBX1, OPCML, CECR1, SCN4B) were also differentially methylated between the two phenotypes. Spearman correlations between methylation and expression in the EPIC/mRNAseq dataset were largely replicated in the 450 K/Affymetrix datasets.

Conclusions: Concomitant study of DNA methylation and gene expression patterns may identify genes whose expression is epigenetically regulated in periodontitis.

Keywords: epigenetics; gingivitis; pathogenesis; periodontitis; transcriptomics.

Figure 1.

Scatter plot of the -log10 (FDR) of differential gene expression and methylation. The red dotted lines indicate FDR = 0.05 in each case. The red boxed gene names are the 12 differential genes between periodontitis and health that are specified in the section.

Figure 2.

Spearman correlation plots between methylation and gene expression levels for each of the 12 genes in the main dataset (RNA-Seq & EPIC chip data).

Figure 3.

Principal Component Analysis examining the influence of sex, race/ethnicity or age on methylation- and gene expression levels in the main dataset.

Figure 4.

Spearman correlation plots between methylation and gene expression levels for 11 of the 12 genes in the validation dataset (450K methylation data / Affymetrix Microarray gene expression data).

Figure 5.

Significantly enriched GO:CC, GO:MF, and Reactome pathways with a Bonferroni corrected p-value ≤0.1. For each pathway, the adjusted p-value, the −log10 of the adjusted p-value, and involved genes are presented.