3 Epiblast ablation by SOX2 knockout does not impair early development of bovine extra-embryonic membranes

Early pre-implantation development in ungulates is characterised by two lineage differentiation events. The first one precedes blastocyst formation and gives rise to the inner cell mass (ICM) and the trophectoderm (TE), the first extra-embryonic lineage. The second event occur around blastocyst hatching and differentiates the ICM into epiblast (the lineage from which the embryonic disc and the embryo proper derive) and the hypoblast (a second extra-embryonic lineage). The hypoblast migrates along the TE during early conceptus elongation, both forming the extra-embryonic membranes (EEMs). Although TE cells are known to grow autonomously, it is unclear whether properly formed EEMs (i.e. a TE layer internally lined by a hypoblast layer) can develop and grow without the embryo proper. To solve that question we have generated epiblast-devoid bovine embryos by ablating SOX2, a transcription factor potentially required for epiblast development, using CRISPR technology. Bovine in vitro matured oocytes were microinjected with Cas9-encoding mRNA and a single-guide RNA against SOX2 (C+G group, containing knockout [KO] embryos) or with Cas9-encoding mRNA alone (C group, formed by wild-type [WT] embryos) as microinjection control. Microinjected oocytes were fertilized in vitro and allowed to develop up to Day 8 blastocyst in conventional culture or Day 12 in a posthatching system. Day 8 and 12 embryos were fixed, subjected to immunohistochemistry (IHC) to detect the development of specific lineages, and finally genotyped by miSeq to identify KO embryos in C+G group. The rate of blastocyst development was similar between microinjection groups (36.1 ± 1.4 vs 24.4 ± 3.3% for C and C+G groups, t-test P > 0.05) and ~55% (40/72) of Day 8 blastocysts in C+G group were KO. SOX2 KO showed no SOX2+ cells in IHC and displayed comparable total number of cells than WT (71 ± 6.3 vs 50.3 ± 6.2 for WT and KO, t-test P > 0.05). In vitro development from blastocyst to Day 12 embryos was similar between groups (80.6 ± 1 vs 80.6 ± 2.8% for C and C+G groups, t-test P > 0.05) and ~47% (15/32) of Day 12 embryos in C+G group were KO. The IHC analysis confirmed the ablation of SOX2 in KO embryos and observed no expression of NANOG (a later epiblast specific marker) in SOX2 KO embryos. Complete hypoblast migration rate (evidenced by hypoblast SOX17+ cells completely covering the TE layer, CDX2+) was unaffected by epiblast ablation (7/31 vs 7/15 for WT and KO, chi-squared P > 0.05) and Day 12 embryo diameter—a proxy of EEMs proliferation—was also unaltered in epiblast-devoid Day 12 embryos (843 ± 70 vs 631 ± 68 µm for WT and KO, t-test P > 0.05). In conclusion, SOX2 is required for bovine epiblast formation, and the early development of EEMs is uncoupled from that of the epiblast. As epiblast-devoid conceptuses are unable to form an embryonic disc and hence a fetus, these results highlight the need of observing embryonic disc formation to infer proper development along early conceptus elongation.