Transcriptome analysis shows ambiguous phenotypes of murine primitive endoderm‐related stem cell lines

Abstract

Primitive endoderm (PrE)‐related cell lines (XEN, pXEN and nEnd cells) show key features of the PrE. By transcriptome analysis, we show: (a) Compared to embryonic stem cells, PrE‐related cell lines are less in vivo like, although early nEnd cells are most similar to the PrE. (b) These cell lines show post‐PrE features of parietal (XEN and pXEN cells) or visceral (nEnd cells) endoderm, likely driven by Tgf‐β and Wnt/Activin signaling, respectively. (c) pXEN and nEnd cell lines additionally show pre‐PrE features. Hence, neither pXEN nor nEnd cell cultures represent a distinct in vivo entity. Rather, their properties are compatible with mixed and hybrid phenotypes. Our findings indicate that pre‐PrE, PrE and early post‐PrE phenotypes result from different niches, which need to be better understood to derive cell lines that distinctly represent the early stages of the extraembryonic endoderm.

A pendulum of induction between the epiblast and extra-embryonic endoderm supports post-implantation progression

“…S13B " type="url"/> ). Finally, insulin has been reported to limit the initial specification of 2D cultured PrE-/VE-like cells termed nEND ( Anderson et al, 2017 ; Zhong and Binas, 2019 ). Complementing the N2B27 medium with additional insulin or with the PI3K inhibitor ZSTK474 did not prevent Gata6 + cell specification and pro-amniotic-like cavity formation.…”

“…To more finely assess the state and reflected stage of the cells within the fully developed EpiCs, we performed additional single-cell RNA sequencing after 0, 24 and 64 h in plain N2B27 medium. We also included controls in the form of naive ESCs (2i/Lif), which are XEN cells that are thought to best reflect the parietal endoderm ( Lin et al, 2016 ; Zhong and Binas, 2019 ), and of Matrigel-embedded ESCs, which form rosette-like cells undergoing lumenogenesis in the presence of serum but in the absence of XEn-like cells ( Bedzhov and Zernicka-Goetz, 2014 ). We visualized single-cell distribution using uniform manifold approximation and projection (UMAP), and identified 10 distinct clusters ( Fig.…”

Chemical‐based primary human hepatocyte monolayer culture for the study of drug metabolism and hepatotoxicity: Comparison with the spheroid model

“…Raw sequencing data were acquired from either the GEO (http://www.ncbi.nlm.nih.gov/geo/) or EBI (http://www.ebi.ac.uk) database; the detailed information is listed in Table S2. To prepare the count matrix, raw data were processed according to the study by Zhong et al 23 with the hg38 genome and GRCh38.98 annotation library for gene mapping and annotation, respectively. The batch effect was removed by Limma, and gene expression was normalized by the DESeq2 “RLD” function.…”