Intrinsic Differences between the Open Chromatin Regions of Oral and Epidermal Keratinocytes.

Abstract

To investigate and characterise the differences between the open chromatin regions of oral and epidermal keratinocytes. Human immortalised oral epithelial cell lines (HIOECs) were used as the standard model for oral keratinocytes, and primary normal human epidermal keratinocytes (NHEKs) were chosen as the model for epidermal keratinocytes. Assay for transposase accessible chromatin using sequencing (ATAC-seq) and H3K27ac chromatin immunoprecipitation sequencing (ChIP-seq) were used to evaluate the dynamic changes in open chromatin regions and active enhancers during oral keratinocyte differentiation. In silico prediction and dual-luciferase assays were used to evaluate the enriched motifs and maintain enhancer activity in specific enriched HIOECs. Integration and comparison of HIOEC ATAC-seq with NHEK ATAC-seq were used to identify oral keratinocyte-enriched open chromatin regions along with key motifs governing differential enhancer activity. The genomic regulatory elements and GWAS overlap algorithm was used to compare the annotation rate of HIOEC-overlapped craniofacial enhancers with other craniofacial enhancers for orofacial cleft-associated variants. During the differentiation of HIOECs, 14933 open chromatin regions became more accessible. Grainyhead-like (GRHL) and Krüppel-like factor (KLF) motifs were overrepresented in maintaining HIOEC-specific activity. Compared with NHEKs, 16161 open chromatin regions were uniquely accessible in HIOECs. Within these regions, the C/EBP motif governed HIOEC-specific enhancer regulating SOX2 and PITX2, which enhanced oral keratinocyte wound healing. When intersected with human craniofacial super-enhancers, open chromatin regions in HIOECS can better annotate the common variants associated with orofacial cleft. The intrinsic differences between the open chromatin regions of human oral and epidermal keratinocytes are directly maintained by a set of transcription factors.