IADR Abstract Archives
Transcriptional Signature and Master Regulator Identification During Experimental Periodontitis
Objectives: By identifying the complete gene repertoire and transcription factors (TFs) involved in the pathogenesis of periodontitis, this study aimed to identify differentially expressed genes and, after their functional analysis, build a gene network of master regulators.
Methods: A ligature-induced mice model of periodontitis model was used. Non-periodontitis mice were used as control. Alveolar bone loss was characterized by micro-computed tomography and histomorphology. Total RNA was extracted from the periodontally affected or healthy mucosa and then subjected to RNA-Seq analysis. Transcripts were considered differentially expressed (DE) when a fold-change >1.0 and p-value <0.05 was obtained. EnrichR was used to identify GO terms for DE genes, and CEMiTool to identify co-expressed modules. High-quality TF-gene regulations from the DoRoThea, TRRUST, and RegNetwork databases were used, which were filtered with a normalized gene counts matrix. Interactions arising from TF-genes with averaged counts >10 were kept or removed otherwise, resulting in two context-specific regulatory networks -control and periodontitis- which were compared using LoTo. Using the CEMiTool, the seed DE-genes from the most representative modules were selected to identify master regulator genes associated with the periodontitis phenotype. Finally, the first and second neighbors from the seed genes were selected, keeping only TFs and removing loosely connected nodes until all remaining TFs had high indegree and outdegree.
Results: In periodontitis, 406 DE genes enriched in immuno-inflammatory response GO terms were identified. There were thirteen co-expression modules, two entirely related to immune response. The GRN analysis identified that the master regulators related to periodontitis immune response were SP1, ETS1, CEBPA, SPI1, E2F1, FOSL, and NFE2L2.
Conclusions: The combination of transcriptomic and GRN analysis allows identifying potential therapeutic targets and molecular biomarkers of periodontitis which will be validated through mechanistic studies.
Methods: A ligature-induced mice model of periodontitis model was used. Non-periodontitis mice were used as control. Alveolar bone loss was characterized by micro-computed tomography and histomorphology. Total RNA was extracted from the periodontally affected or healthy mucosa and then subjected to RNA-Seq analysis. Transcripts were considered differentially expressed (DE) when a fold-change >1.0 and p-value <0.05 was obtained. EnrichR was used to identify GO terms for DE genes, and CEMiTool to identify co-expressed modules. High-quality TF-gene regulations from the DoRoThea, TRRUST, and RegNetwork databases were used, which were filtered with a normalized gene counts matrix. Interactions arising from TF-genes with averaged counts >10 were kept or removed otherwise, resulting in two context-specific regulatory networks -control and periodontitis- which were compared using LoTo. Using the CEMiTool, the seed DE-genes from the most representative modules were selected to identify master regulator genes associated with the periodontitis phenotype. Finally, the first and second neighbors from the seed genes were selected, keeping only TFs and removing loosely connected nodes until all remaining TFs had high indegree and outdegree.
Results: In periodontitis, 406 DE genes enriched in immuno-inflammatory response GO terms were identified. There were thirteen co-expression modules, two entirely related to immune response. The GRN analysis identified that the master regulators related to periodontitis immune response were SP1, ETS1, CEBPA, SPI1, E2F1, FOSL, and NFE2L2.
Conclusions: The combination of transcriptomic and GRN analysis allows identifying potential therapeutic targets and molecular biomarkers of periodontitis which will be validated through mechanistic studies.