Click here to close
Hello! We notice that you are using Internet Explorer, which is not supported by Xenbase and may cause the site to display incorrectly.
We suggest using a current version of Chrome,
FireFox, or Safari.
???displayArticle.abstract???
The homeobox gene otx2 is a key regulator of positional identity in vertebrates, however its downstream target genes and mechanism of action are not known. We have analyzed otx2 function during formation of the Xenopus cement gland, an organ that expresses otx2. The cement gland forms at early neurula from extreme anteriorectoderm and corresponds to the chin primordium of mammals. Previous studies (Blitz, I. and Cho, K. (1995) Development 121, 993-1004; Pannese, M., Polo, C., Andreazzoli, M., Vignali, R., Kablar, B., Barsacchi, G. and Boncinelli, E. (1995) Development 121, 707-720) showed that misexpressed otx2 could activate cement gland formation. However, it was not clear whether this was a direct effect of otx2 or a secondary consequence of other tissues induced by otx2. In this study we ask whether otx2 activity is spatially and temporally restricted in the ectoderm and whether cement gland-specific genes are direct targets of otx2. In order to control the timing of otx2 activity, we constructed a dexamethasone-inducible otx2 protein (otx2-GR) by fusion with the ligand-binding domain of the glucocorticoid receptor. We conclude first, that regionally restricted factors regulate otx2 activity since otx2-GR is able to activate the cement gland markers XCG and XAG only in ventrolateral ectoderm, and never in the neural plate. Second, we show that temporal responsiveness of the ectoderm to otx2-GR is limited, beginning only at mid-gastrula but continuing as late as tailbud stages. Third, we show that otx2-GR activates expression of the cement gland differentiation marker XCG in the absence of protein synthesis, identifying a direct target of otx2. otx2-GR can also activate expression of the endogenous otx2 gene, defining an autoregulatory loop. Fourth, we show that otx2-GR is sufficient to overcome the inhibitory effects of retinoic acid on cement gland formation, indicating that this effect is caused by failure to express otx2. Corroboratively, we show that otx2 autoactivation is prevented by retinoic acid. Together, these findings suggest that otx2 directly controls cement gland differentiation, and that spatial and temporal modulation of otx2 activity limits cement gland formation to the front of the embryo.
Fig. 1. An otx2-glucocorticoid receptor fusion activates hormone-dependent ectopic XCG expression in embryos and explants.
(A) Experimental scheme. Albino embryos were injected with otx2-GR RNA at the two cell stage. Injected embryos or blastula (stage 8 or 9) animal caps were incubated alone or with dexamethasone (dex) until embryos reached hatching (stage 32). Arrows indicate the stage treatment was initiated, triangles indicate the stage of harvest.
(B) Ectopic cement gland induction in whole embryos. Embryos were injected in one blastomere with 100 pg otx2-GR RNA mixed with 60 pg lacZ RNA. Embryos were incubated without (a) or with (b) dex beginning at early gastrula (stage 10.5) and until stage 32, when they were stained for lacZ (light blue) and for XCG expression by in situ hybridization (black). In addition to endogenous XCG (white open arrowhead), ectopic XCG was observed in the ventrolateral ectoderm (black arrowheads) and at lower frequencies in dorsal epidermis (white arrowheads). Lateral view, anterior to the center.
(C) Ectopic cement gland induction in animal caps. Embryos were injected in both blastomeres with 50 pg otx2-GR RNA. Animal caps dissected at stage 9 were left untreated or transferred to dex when control embryos reached the stages indicated. XCG expression was analyzed by in situ hybridization when control embryos reached stage 32. (a) untreated; (b) +dex st.11.5; (c) +dex st. 14; (d) +dex st. 20.
(D) Northern blot analysis of otx2-GR RNA levels in animal caps. Embryos were injected in both blastomeres with 50 pg MT-otx2-GR RNA. Animal caps were dissected at stage 8 and pools of ten caps were analyzed by northern blot for otx2 expression when control embryos reached stage 8.5 (lane 1), stage 10 (lane 2), stage 12 (lane 3), or stage 20 (lane 4). Ethidium bromide-stained 28S rRNA is a loading control.
(E) Western blot analysis of inducible protein levels in animal caps. Explants prepared as in D were harvested in groups of ten when sibling embryos reached stage 8.5 (lane 1), stage 10 (lane 2), stage 12 (lane 3), or stage 20 (lane 4). Two cap equivalents of protein were analyzed by western blot analysis using an anti-myc antibody. Ponceau S staining indicated that all lanes were loaded equally (data not shown).
Fig. 2. otx2 expression correlates with insensitivity of the cement gland to retinoic acid. Albino embryos were left untreated (a, e) or incubated in retinoic acid (RA) beginning at stage 10.5 (b, f), stage 11.5 (c, g), or stage 12.5 (d, h) and until harvest at stage 15 for analysis by in situ hybridization. (a-d) XCG expression. Black open arrowheads indicate XCG expression in the cement gland anlage. Anterior view, dorsal up. (e-h) otx2 expression. White open arrowheads indicate otx2 expression in the cement gland anlage. Anterior view, dorsal up.
Fig. 3. otx2-GR can activate retinoic acid-insensitive XCG expression in whole embryos.
(A) Experimental scheme. Albino embryos were injected with otx2-GR RNA at the two-cell stage. Injected embryos were incubated alone or in dexamethasone (dex) and/or retinoic acid (RA) for 5 hours beginning at blastula (stage 9) or mid-gastrula (stage 11). Arrows indicate the stage treatment was initiated, triangles indicate the stage of harvest.
(B) XCG induction in whole embryos. Embryos injected in one blastomere with 50 pg otx2-GR RNA were left untreated or incubated in dex and/or RA from stage 9 to 11 (a, b) or stage 11 to 15 (c-f). XCG expression was analyzed by in situ hybridization. Open arrowheads indicate endogenous XCG, solid arrowheads mark ectopic XCG. Anteroventral view. (a) untreated; (b) +dex st. 9; (c) untreated; (d) +dex st. 11; (e) +RA st. 11; (f) +RA+dex st. 11.
(C) Western blot analysis of whole embryos. Embryos injected with 50 pg MT-otx2-GR RNA in both blastomeres and prepared as in B were lysed in pools of five for western blot analysis with an anti-myc antibody. One embryo equivalent of protein was analyzed per lane. Ponceau S staining indicated that all lanes were loaded equally (data not shown). Lane 1, untreated; lane 2, +RA st. 9; lane 3, +dex st. 9; lane 4, +RA+dex st. 9; lane 5, untreated; lane 6, +RA st. 11; lane 7, +dex st. 11; lane 8, +RA+dex st. 11.
Fig. 4. otx2-GR cannot activate cement gland marker expression or an otx2 autoregulatory loop until early/mid-gastrula.
(A) Experimental scheme. Wild-type embryos were injected with otx2/3-GR RNA at the two-cell stage. Blastula (stage 8 or 9) animal caps were isolated from injected embryos and incubated without or with dexamethasone (dex) for 5 hours beginning at stage 8 or stage 9, for 5 hours or 20 hours beginning at early/mid-gastrula (stage 11), or for 12 hours beginning at tailbud (stage 20). Arrows indicate the stage treatment was initiated, triangles indicate the stage of harvest.
(B) XCG, XAG and otx2 expression in animal caps. Embryos were injected in both blastomeres with 50 pg otx2/3-GR RNA. Animal caps were dissected at stage 8 or 9, and pools of ten explants were left untreated or were incubated in dex for 5 hours immediately after dissection (lanes 1-4), or for 5 hours (lanes 5, 6) or 20 hours (lanes 7, 9) beginning when control embryos reached stage 11, or for 12 hours (lane 8) when controls embryos reached stage 20. XCG, XAG, and otx2 expression were analyzed by RT-PCR using EF-1a as a control. Since cement gland markers are expressed at very high levels in the mature cement gland compared to their initial expression levels, samples were assayed at high number of PCR cycles (lanes 1-6) or low number of PCR cycles (lanes 7-9) in order to remain in the linear range of amplification at the stage assayed. Early expression is not detectable at low cycle numbers. Identical results were obtained in two independent experiments. Lane 1, untreated; lane 2, +dex st. 8; lane 3, untreated; lane 4, +dex st. 9; lane 5, untreated; lane 6, +dex st. 11; lane 7, untreated; lane 8, +dex st. 20; lane 9, +dex st. 11.
Fig. 5. otx2 regulation of target gene expression.
(A) Experimental scheme. Wild-type embryos were injected with otx2/3-GR RNA at the two-cell stage. Blastula (stage 9) animal caps were isolated from injected embryos and incubated alone or in dexamethasone (dex) and/or cycloheximide (CHX) or retinoic acid (RA) for 5 hours or for 20 hours beginning at early/mid-gastrula (stage 11). Arrows indicate the stage treatment was initiated, triangles indicate the stage of harvest.
(B) XCG, XAG and otx2 expression in animal caps. Embryos were injected in both blastomeres with 50 pg otx2/3-GR RNA. Animal caps were dissected at stage 9, and pools of ten explants were left untreated or were incubated in dex without or with CHX or RA for 5 hours (lanes 1-6) or 20 hours (lanes 7-10) beginning when control embryos reached stage 11. XCG, XAG and otx2 expression were analyzed by RT-PCR using EF-1a as a control. Since cement gland markers are expressed at very high levels in the mature cement gland compared to their initial expression levels, samples were assayed at high number of PCR cycles (lanes 1-6) or low number of PCR cycles (lanes 7-10) in order to remain in the linear range of amplification at the stage assayed. Early expression is not detectable at low cycle numbers. Comparable results were obtained with cycloheximide in eight independent experiments and by in situ hybridization (three independent experiments), and with RA in four independent experiments. Lane 1, untreated; lane 2, +dex; lane 3, +CHX; lane 4, +CHX+dex; lane 5, +RA; lane 6, +RA+dex; lane 7, untreated; lane 8, +dex; lane 9, +RA; lane 10, +RA+dex.
(C) Western blot analysis of cycloheximide-treated explants. Embryos were injected in both blastomeres with 50 pg MT-otx2-GR RNA and animal caps dissected at stage 9. When control embryos reached late gastrula (stage 12.5), caps were treated without or with dex and/or CHX for 3 hours, a period of time encompassed by treatments described in (B), and were lysed in groups of ten for western blot analysis with an anti-myc antibody. Two cap equivalents of protein were analyzed per lane. Ponceau S staining indicated that all lanes were loaded equally (data not shown). Lane 1, untreated; lane 2, +CHX; lane 3, +dex; lane 4, +CHX+dex.
