(A) Plot of the moderated log2 fold change (mondoa-mo vs. mondoa-mis injected embryos) over averaged normalized counts. The most downregulated gene nsdhl is indicated. (B) The main steps in the conversion of acetyl coenzyme A (acetyl-CoA) to cholesterol and pregnenolone. (C, D) Transcript levels of nsdhl (C) and ef1a (D) of embryos injected with water (white bars) as a control or with the glucose analog 2-deoxy-D-glucose (2-DG; black bars) in the presence (+) or absence (-) of mondoa and mlx mRNA (n = 9). (E) Embryos injected with a splice-site MO against nsdhl (nsdhl-mo) show a severe developmental delay and arrest at ~50% epiboly, when uninjected controls were at bud stage (~10 hpf). (F) Quantiﬁcation of the experiments in (E). Percentages of ‘affected’ embryos: uninjected (0/84, 0%), nsdhl-mo injected (n = 136/141, 96.5%). (G) Injection of nsdhl-mo into 1 k cell stage embryos led to arrest at 95% epiboly (n = 11/11, 100%), when uninjected control embryos (n = 15/15, 100%) were at bud stage (10 hpf). (H,I) Percentage of epiboly progression under different treatments determined via no tail (ntl) WISH. (H) Uninjected control embryos (n = 30) were ﬁxed at the 75% epiboly stage along with the treated embryos. Treatment with 20 µM pregnenolone (P5) led to a partial rescue of the mondoa morphants (n = 19), which achieved 48% epiboly (untreated morphants: 38% epiboly; n = 25). Control and mondoa-mo morphant results are reproduced from Figure 2E, as these experiments were carried out in parallel. (I) P5 treatment also led to a partial rescue of the nsdhl morphants (n = 30, 74.6% epiboly compared to 59.0% in untreated morphants, n = 50). Asterisks label the blastoderm margin. Scale bar: 0.2 mm. Error bars represent SEM; *, p≤0.05; **, p≤0.01; ***, p≤0.001.