Transcriptomic Analysis Reveals Genetic Cross-Talk between Periodontitis and Hypothyroidism

Abstract

Background: Aim of this bioinformatics study based on transcriptomic analysis was to reveal the cross-talk between periodontitis (PD) and hypothyroidism (HT).

Methods: The gene expression datasets GSE18152 and GSE176153 of HT and GSE10334, GSE16134, and GSE173078 of PD were downloaded through the Gene Expression Omnibus (GEO) database. Differential Expression Genes (DEG) between cases and controls in each microarray were assessed by using the "limma" (linear models for microarray data) R package (|log2 fold change (FC)| >0 and P-value <0.05). To analyze the cross-talk effect between HT and PD, the intersection of DEG of HT and PD was selected. To investigate the biological function of cross-talk genes, the gene ontology (GO) functional enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were applied. Protein-Protein Interaction (PPI) network was constructed using Cytoscape software. Top 10 cross-talk genes were screened, and the expression values of these 10 genes were extracted. ROC analysis was performed by using the pROC package and GGplot2 package of R language to predict the classification accuracy.

Results: The overlapping DEG between HT and PD were 107 cross-talk genes. The results revealed that developmental process (P-value =1.06E-21) was the most significantly enriched biological process, followed by cell differentiation (P-value =8.49E-18) and immune system process (P-value =6.78E-11). KEGG analysis showed that Complement and coagulation cascades (P-value =2.29E-05), Hematopoietic cell lineage (P-value =2.66E-05), Phospholipase D signaling pathway (P-value =0.034367878) and Chemokine signaling pathway (P-value =0.04946333) were significantly enriched. The top 10 genes with most connections were LCE1B, LCE2B, LCE2A, LCE2C, LCE1C, LCE1F, ITGAM, C1QB, TREM2, and CD19. The AUC values of the two datasets of HT were both greater than 65% (GSE18152 = 81.42%, GSE176153 = 68.75%). AUC values of three datasets of PD were all greater than 60% (GSE10334 = 69.23%, GSE16134 = 73.72%, GSE173078 = 81.6%).

Conclusions: A genetic cross-talk between HT and PD was detected, whereby LCE family genes appeared to play the most important role.