From blastocyst to gastrula: gene regulatory networks of embryonic stem cells and early mouse embryogenesis

Abstract

To date, many regulatory genes and signalling events coordinating mammalian development from blastocyst to gastrulation stages have been identified by mutational analyses and reverse-genetic approaches, typically on a gene-by-gene basis. More recent studies have applied bioinformatic approaches to generate regulatory network models of gene interactions on a genome-wide scale. Such models have provided insights into the gene networks regulating pluripotency in embryonic and epiblast stem cells, as well as cell-lineage determination in vivo. Here, we review how regulatory networks constructed for different stem cell types relate to corresponding networks in vivo and provide insights into understanding the molecular regulation of the blastocyst-gastrula transition.

Keywords: gene regulatory network; mouse embryo; pluripotency; stem cells; systems biology.

Figure 1.

GRNs for Nodal signalling in (a) mouse and (b) sea urchin. (a) A GRN that specifies mesendoderm containing a feedback loop between the NODAL precursor (proNODAL), NODAL, BMP4 and WNT3, as well as a CRIPTO-dependent feedback loop. Red indicates signals that promote mesendoderm formation, green, mesoderm formation, and blue, ectoderm formation. (Adapted from [11].) (b) A GRN for Nodal signalling in sea urchin embryogenesis, illustrating its synergistic relationship with Not in ectoderm-specific gene expression. (Adapted from [12].)

Figure 2.

(a) The ‘classic’ model of pluripotency regulation proposes that a ground state is actively maintained by a core set of TFs in the absence of differentiation signals, created by culture conditions involving inhibition of GSK3 and MAPK signalling. (b) The ‘balance’ model of pluripotency proposes that competition among lineage-specifying TFs regulates pluripotency. Stimulatory (blue) and inhibitory (red) signals from each of the TFs confer multi-lineage differentiation potential, and do not result in commitment to any particular lineage. (Adapted from [49].)

Figure 3.

A GRN for pluripotency in the blastocyst, constructed manually using recently published data from in vivo and in vitro studies [,–69]. The network illustrates how LIF signalling components, MAPK signalling regulators and TFs such as Esrrb, Sall4 and c-Fos influence the dynamics of the core pluripotency network in the blastocyst.

Figure 4.

GRNs for lineage commitment during preimplantation development. (a) A network of gene interactions involved in TE specification, centred on mutual inhibition between Cdx2, Oct4 and Nanog. Hippo signalling components mediate positional cues in the regulation of Tead4 and Cdx2 expression, which in turn regulate the expression of TE-specific genes such as Gata3, Psx1, Hand1, Eomes and Elf5. Concurrently, FGF signalling is thought to regulate a sub-network involving Tcfap2c, Sox2 and Esrrb that inhibits Oct4 expression while stimulating Cdx2. (b) Interactions among genes involved in PrE specification. FGF4 signalling from the Epi stimulates FGFR2-mediated ERK/MAPK signalling in cells of the nascent PrE, resulting in upregulation of Gata6 and Oct4, and inhibition of Nanog. Feedback loops between Nanog, Gata6 and Fgfr2 enforce this dynamic, while Oct4 is thought to be upstream of PrE-specific genes such as Sox7, Sox17, Gata4 and Pdgfra.

Figure 5.

A GRN for lineage commitment in the mouse gastrula. Interactions were inferred from published data [54,58,85,141] among the core pluripotency factors OCT4, NANOG and SOX2, the epigenetic regulators JARID2, SUZ12 and JMJD2 and lineage-specifying factors Sox1 and Brn2 (for neuroectoderm) and Foxa2 and T (for mesendoderm) and Eomes (for endoderm). The direction of the inferred interactions is not fully understood, but such a network suggests how localized changes in the expression of the core pluripotency circuit might influence lineage commitment events. (Adapted from [54].)