Physically interacting cell sequencing of mouse embryonic endoderm at E7.5, E8.5, and E9.5

Cells continuously communicate with the neighboring cells during development. Direct interaction of different cell types can induce molecular signals dictating lineage specification and cell fate decisions. The current single-cell RNAseq (scRNAseq) technology cannot study cell contact dependent (or niche specific) gene expression due to the loss of spatial information. To overcome this issue and determine cell contact specific gene expression during embryogenesis, we performed RNA sequencing of physically interacting cells (PICseq) and assessed alongside our single cell transcriptomes (scRNAseq) derived from developing mouse embryos between embryonic day (E) 7.5 and E9.5. Analysis of PICseq data identifies an interesting suite of gene expression signatures depending on neighboring cell types. For instance, neural progenitor (NP) cells expressed Nkx2-1 when interacting with definitive endoderm (DE) and DE cells expressed Gsc when interacting with NP. Based on the identified cell contact specific genes, we devised a means to predict the neighboring cell types from individual cell transcriptome. We further developed spatial-tSNE to show the pseudo-spatial distribution of cells in a 2-dimensional space. In sum, we suggest an approach to study niche specific gene regulation during embryogenesis. Overall design: Mouse FOXA2Venus embryos were collected between embryonic days (E) 7.5 and 9.5. The E9.5 embryos were dissected in order to enrich the sample with gut endoderm cells. Multiplet cells were isolated from FOXA2Venus mouse embryos. Cells were sorted using a BD FACS Aria III. FSC/SSC gates were used to define a homogeneous population, FSC-H/FSC-W gates were used to include multiplets, remove singlets and dead cells were excluded based on DAPI inclusion.