Cell cycle control, DNA damage repair, and apoptosis-related pathways control pre-ameloblasts differentiation during tooth development

Abstract

Background: Ameloblast differentiation is the most critical stepwise process in amelogenesis, and it is controlled by precise molecular events. To better understand the mechanism controlling pre-ameloblasts (PABs) differentiation into secretory ameloblasts (SABs), a more precise identification of molecules and signaling networks will elucidate the mechanisms governing enamel formation and lay a foundation for enamel regeneration.

Results: We analyzed transcriptional profiles of human PABs and SABs. From a total of 28,869 analyzed transcripts, we identified 923 differentially expressed genes (DEGs) with p < 0.05 and Fold-change > 2. Among the DEGs, 647 genes showed elevated expression in PABs compared to SABs. Notably, 38 DEGs displayed greater than eight-fold changes. Comparative analysis revealed that highly expressed genes in PABs were involved in cell cycle control, DNA damage repair and apoptosis, while highly expressed genes in SABs were related to cell adhesion and extracellular matrix. Moreover, coexpression network analysis uncovered two highly conserved sub-networks contributing to differentiation, containing transcription regulators (RUNX2, ETV1 and ETV5), solute carrier family members (SLC15A1 and SLC7A11), enamel matrix protein (MMP20), and a polymodal excitatory ion channel (TRPA1).

Conclusions: By combining comparative analysis and coexpression networks, this study provides novel biomarkers and research targets for ameloblast differentiation and the potential for their application in enamel regeneration.

Fig. 1

Identification of ameloblasts from a developing human incisor for laser capture microdissection. Left panel: H&E staining of a developing human incisor. Right panel: a Magnification of (a) in left panel. Pre-ameloblasts (PAB) were polarized inner enamel epithelial cells that directly contacted the basement membrane and were adjacent to polarized odontoblasts. b Magnification of (b) in left panel. Secretory ameloblasts (SAB) were identified as polarized epithelial cells in direct contact with the enamel matrix. (PAB: preameloblast, SAB: secretory ameloblast, POB: preodontoblast, SOB: secretory odontoblast, D: dentin, E: enamel, SR: Stellate reticulum, SI: Stratum intermedium, P: dental pulp)

Fig. 2

a. Characterization of different stages of micro-dissected human dental epithelial cells. AMLX and AMBN expression levels were up-regulated as cells differentiated, while PCNA expression decreased in SABs compared to PABs. **p < 0.01. PITX2 expression was up-regulated in PABs compared to SABs. b. PCA clustering showed a clear divergence of PABs and SABs

Fig. 3

KEGG pathway analysis. a. KEGG pathway enrichment analysis was performed by Fisher exact test. Significantly enriched KEGG pathways (p < 0.05) are presented. For each KEGG pathway, the bar shows the fold-enrichment of the pathway. b. The expression intensity of genes in the selected KEGG pathway. The italic values on the right of gene names were the fold changes of corresponding genes in PABs compared to SABs

Fig. 4

Gene coexpression modules and assigned module colors. a. Clustering dendrogram of genes with the x-axis corresponding to each gene and the y-axis representing the dissimilarity based on topological overlap of the whole genome throughout ameloblast differentiation. Top color bar: modules were assigned colors corresponding to different dendrogram branches detected by a cluster algorithm. Second color bar: each gene was annotated with DEG information such that significantly up-regulated genes in PABs are marked with red bands and genes down-regulated in PABs marked with green bands. b. Boxplot depicting gene significance defined as -log10(p-value) of each module. The left legend indicates module colors and the number of genes (stars with a number sign) in the corresponding module (for example, 1527 genes in the blue module)

Fig. 5

Gene coexpression networks of modules. Global network visualization of the top two modules “brown” (a) and “blue” (b) that are highly correlated with gene significance. Each node corresponds to a gene, and each edge joining two nodes indicates the connection determined by the adjacency. The nodes in boxes represent DEGs and circles indicate genes without significant differential expression between PABs and SABs. Nodes are also marked with colors to indicate their representative GO biological processes, and grey nodes are those not included in either of the GO terms. The left legend in each network shows colors and corresponding GO terms

Fig. 6

Hub gene selection and gene ontology analysis of brown and blue modules. a and b are the scatterplots between module membership measure (x-axis) and the gene significance of the “brown” and “blue” module. The top 20 genes with both the highest gene significance and membership measure larger than 0.9 were chosen as “hub genes”, and they are annotated with their gene symbols. The grey dashed line in the plot is the threshold for choosing significantly expressed genes, and the threshold value is –log10(0.05). c and d are the hub gene network visualizations for the “brown” and “blue” modules. The colors of the hub genes and their directly connected genes indicate the GO terms from Fig. 5 (a) and (b). e. Barplots represent enriched GO terms of all genes included in the “brown” (left with brown color) and “blue” (right with blue color) networks, and –log2(p-value) represents the relative enrichment of each GO term