Fig. 5

Multiple modes of regulation of sox17 and foxa2 by nipbls.

(A–D) Embryos were injected with increasing (3.125–25 pg) amounts of in vitro synthesized sox32 mRNA, and either 5-mis-nipbl-MO1 mixture (0.75 ng each: control) or nipbl-MO1 mixture (0.75 ng each: nipbla/b-MO1). mRNA levels for the sox32-target genes sox17, foxa2, and cxcr4a were quantified at 6.5 hpf by Q-PCR, normalized to ef-1a, and expressed relative to levels in embryos not treated with exogenous sox32 mRNA. As sox32 induces its own expression [38], target gene expression was determined as a function of total sox32 (endogenous and exogenous, both measured directly), and not just the amount of sox32 RNA injected. (A) Top: expected behavior of a sox32 target gene that is not itself affected by nipbls. The same relationship between sox32 level and target gene expression should be observed in control and nipbla/b-morphant embryos. Right: expected behavior of sox32 target gene that is independently acted upon by nipbls (either directly, or because nipbls control the expression of other inputs to the gene). In this case, the relationship between sox32 level and target gene expression will be shifted in nipbla/b-morphant embryos. (B) cxcr4a displays the expected behavior of a sox32 target gene that is not itself affected by nipbls. (C,D) sox17 and foxa2 display the expected behavior of sox32 target genes that are independently sensitive to nipbl function. In panels B–D, data are presented as mean ± S.E.M (n = 3). Dose-response relationships were well fit by straight lines (least-squares regression yielded values of r²>0.95 in all cases). (E, F) To eliminate any indirect effects of Nipbl reduction on gene expression through sox32, endogenous Sox32 protein was removed with a sox32-MO (5 ng) and replaced with sox32-9mis mRNA (10 pg), lacking the MO binding site for sox32-MO, and endodermal gene expression was examined by Q-PCR. (E) Embryos coinjected with sox32-MO and either a wild-type sox32(WT)- or mutated sox32(9mis)-GFP reporter construct show that a 9-base-mutation is sufficient to escape suppression of translation by sox32-MO. Lateral views at 12 hpf. Scale bar: 100 μm. (F) Effects of Nipbl reduction on expression of endodermal genes (sox17, foxa2, and cxcr4a) were examined in control (blue, red), Sox32-deficient (MO; green, orange), and Sox32-restored (MO+mRNA; light blue, pink) embryos at 7 hpf (n = 4, mean ± S.E.M.; * p<0.01 and ** p<0.03 by paired t test). The data show that induction of sox17 and foxa2 by exogenous Sox32 is markedly Nipbl-dependent, whereas induction of a different Sox32 target, cxcr4a, is not.