Single-cell transcriptome analysis revealed continuous changes in molecular phenotypes during epithelial-to-mesenchymal transition

Many genes involved in epithelial-to-mesenchymal transition (EMT) have been identified to date, but mechanisms contributing to the phenotypic diversity and those governing the coupling between the dynamics of epithelial genes and that of the mesenchymal genes are unclear. In this study, we employed single-cell transcriptome analysis to study the gene expression changes during dynamic E-to-M transition. Dose-dependent response to TGFbeta signal suggested that gene expression patterns between E and M states are rather continuous unlike binary switching between two discrete populations. E and M genes can be classified into several subclusters based on expression correlation and it turned out that some E and M genes are mutually exclusive but others are not. Our data suggests that complex and continuous changes in molecular phenotypes occur during EMT at a single-cell level. 10x Genomic Chromium data set was generated from the MCF10A cells (ATCC) RNA-Seq library was prepared using 10x Chromium Single Cell 3' –end Reagent Kits User Guide (v2 Chemistry). Libraries were sequenced using paired-end sequencing (26bp Read 1 and 98bp Read 2) with a single sample index (8bp) on the Illumina HiSeq 2500.