Figure 6: Hippo pathway-dependent and -independent mechanisms in lineage segregation.

The results we present here show that Amot becomes enriched in the ICM as cell lineages become progressively established, and that Amot functions to prevent differentiation of the pluripotent lineage. Our results suggest that Amot acts through two routes to prevent ICM differentiation, of which one is dependent and another is independent of the Hippo pathway. The Hippo pathway-independent (Lats1/2–independent) mechanism is likely to involve Amot’s ability to tether Yap to the cytoplasm/membrane, as we find that Yap-binding motifs of Amot are essential for its full activity. But, it can also involve Amot interacting with actin/cell polarity molecules. The Hippo pathway-dependent mechanism is likely to involve Amot’s role in activating Lats1/2 (as shown in²⁴) to allow phosphorylation of Yap, which prevents its transition to the nucleus⁹. In agreement, we find that mutation of the Yap-binding sites does not completely abolish the activity of Amot. Combined, this results in Yap being sequestered from the nucleus and therefore the expression of TE-genes, such as Cdx2, is switched off and the ICM pluripotency maintained. In contrast, in TE cells, most of Amot is degraded and the remaining Amot is sequestered to the apical domain. We hypothesize that in this location Amot is shielded from binding Yap through its interactions with F-actin and/or polarity factors. In agreement with this hypothesis, Amot is reported to associate with actin³² and tight-junction proteins via its PDZ-binding domain in tissue culture cells^14,15,16,17. This suggests that cell polarity is involved in allowing Yap to translocate to the nucleus to act, together with Tead4, to stimulate Cdx2 expression and TE differentiation.