Transcription factor AP-2 is an essential and direct regulator of epidermal development in Xenopus.
Expression of the Xenopus homolog of the mammalian transcription factor AP-2alpha (XAP-2) is activated throughout the animal hemisphere shortly after the midblastula transition, and becomes restricted to prospective epidermis by the end of gastrulation, under the control of BMP signal modulation. Elevated expression in the future neural crest region begins at this time. Ectopic expression of XAP-2 can restore transcription of epidermal genes in neuralized ectoderm, both in ectodermal explants and in the intact embryo. Likewise, loss of XAP-2 function, accomplished by injection of antisense oligonucleotides or by overexpression of antimorphic XAP-2 derivatives, leads to loss of epidermal and gain of neural gene expression. These treatments also result in gastrulation failure. Thus, AP-2 is a critical regulator of ectodermal determination that is required for normal epidermal development and morphogenesis in the frog embryo.
PubMed ID: 11969261
Article link: Dev Biol.
Grant support: HL 42252 NHLBI NIH HHS
Genes referenced: chrd.1 klf6 otx2 tfap2a zic3
Article Images: [+] show captions
|FIG. 1. Early expression of XAP-2. Whole mount in situ hybridizations at stage 10.25 (A) and stage 12.5–13 (B) to antisense XAP-2 probe, stained with BM-purple. In panel A, the arrowhead indicates hybridization to a chordin probe to mark the dorsal lip (magentaphos stain). At the beginning of gastrulation, XAP-2 RNA was found throughout the ectoderm, then becomes excluded from the neural plate by the end of gastrulation, at which time high level expression commences in the cranial neural crest region.|
|FIG. 2. BMP dependence of XAP-2 expression. (A) Effect of chordin expression. Fertilized eggs were injected with 250 pg of RNA encoding chordin, then ectodermal explants were dissected at stage 7/8 and cultured until sibling embryos reached late gastrula (st12.5). RNA was extracted and probed by Northern blot using cDNAs for XAP-2, Zic-r1, or EF1 as a control. Chordin injection inhibited XAP-2 and simultaneously induced expression of the neural plate marker Zic-r1. (B) Requirement for protein synthesis. Embryos were cultured in calcium/magnesium-free medium beginning at early cleavage stage (16/32-cell), and vitelline envelopes removed to facilitate continuous dissociation and dispersion, a procedure which blocks cell-cell communication (Sargent et al., 1986). Dispersed cells were divided into four pools. Protein synthesis was inhibited in two of the pools by treatment in 10 g/ml cycloheximide beginning at the equivalent of stage 7/8 (CHX ). After 15 min, 50 ng/ml recombinant human BMP4 protein (Research Diagnostics, Inc., Flanders, NJ) was added to two pools (BMP ), Ca2 and Mg2 restored to 1 mM each, and the cells cultured until stage 11. RNA was prepared and analyzed by Northern blot, using probes for XAP-2, Dlx3, Msx1, and EF1 as a control for general inhibition of protein synthesis by the CHX treatment. XAP-2 RNA induction by BMP was inhibited by CHX treatment, indicating indirect activation by this signaling pathway.|
|FIG. 3. Rescue of keratin expression in neuralized ectoderm by XAP-2. Fertilized eggs were injected with 250 pg RNA encoding chordin, with or without 100 pg RNA encoding XAP-2. Ectodermal explants were isolated at stage 7–8 and cultured until sibling embryos reached stage 13, then samples were processed for Northern blot analysis. Probes were for type I (XK81) and type II (XK76) embryo-specific epidermal keratins, two epidermal homeodomain genes, Dlx5 and Msx1, and neural-plate (Zic-r1) or EF1 . Chordin strongly inhibited all epidermal gene expression. The keratin genes and Dlx5 were partially restored by addition of XAP-2 RNA. Msx1, however, was not activated, nor was repression of Zic-r1 observed. Note that under these conditions endogenous XAP-2 expression was inhibited, as shown in Fig. 2A and B.|
|FIG. 4. The XK81 gene is a direct target for XAP-2. (A) Fertilized eggs were injected with a mixture of RNAs encoding chordin (500 pg) and glucocorticoid-inducible GRXAP-2 (1 ng). Ectodermal explants were isolated at stage 7–8, and divided into four groups. Two groups were treated with 10 g/ml cycloheximide (CHX ) to inhibit protein synthesis. After 15 min, two groups (DEX ) were treated with 10 M dexamethasone (DEX) to activate the accumulated GRXAP-2 fusion protein. Controls were treated with an equivalent volume of the DEX solvent (ethanol) alone. After culturing to stage 12 equivalent, RNA was isolated and processed for Northern blot analysis. (A) Hybridization to the XK81 keratin probe revealed that the GRXAP-2 fusion protein responded as expected to the DEX administration, resulting in partial recovery of XK81 expression (DEX vs. DEX-), and that this was not prevented by cycloheximide (CHX vs. CHX-). Note that different exposures are shown for XK81 and EF1 probes. (B) Since cycloheximide treatment has a general inhibitory effect on RNA synthesis in Xenopus embryos, phosphorimage analysis was performed to normalize the XK81 expression data to that of the control, EF1 , and the results shown in histogram format. Based on this analysis, induction of XK81 by XAP-2 appeared to have a similar magnitude in the presence or absence of cycloheximide. (C, D) Loss of competence during gastrulation. Fertilized eggs were injected with a mixture of chordin and GRXAP-2 RNAs as in panel A. Ectodermal explants were isolated at stage 7/8, and subsets transferred to medium containing 10 m DEX at the stages indicated, from stage 8.5 through stage 13. Following culture to the equivalent of stage 20 (to minimize the temporal variable in DEX exposure), RNA was isolated for Northern blot analysis. Hybridization with an XK81 keratin probe revealed strong recovery of keratin expression when DEX was added immediately (stage 8.5; note that the uninjected lane was loaded with a 1/10 dilution to facilitate exposure), but that this response rapidly declined for subsequent DEX additions. Hybridization with control EF1 probe showed essentially no differences, and hybridization with a XAP-2 probe indicated that the injected GRXAP-2 RNA (endogenous XAP-2 hybridization not visible at this exposure) was not differentially affected by the DEX treatment. Phosphorimager quantification of Northern blot data are shown in panel D.|
|FIG. 6. Loss-of-function analysis; intact embryos. Fertilized eggs were injected with either (A) 600 pg of ASO281, (B) a mixture of 600 pg ASO281 and 100 pg of XAP-2* RNA, or (C) 600 pg ASO281 with 50 pg each XK81 and XK76, and cultured to stage 20. Control, uninjected embryos are shown in panel D. ASO281 treatment resulted in extensive gastrulation failure in 29/29 of the embryos, with pigmented ectoderm collapsed into a small sac. Co-injection of XAP-2* RNA rescued essentially normal development in about 12/31 of embryos, partially rescued another 9/31, and had little effect on the remainder. An example of each is shown in panel B. Keratin mRNA injection had little if effect; 24/27 were essentially identical to embryos injected with ASO281 alone, although the remaining three showed some a more moderate phenotype. One of the three partially normal embryos is shown in panel C (embryo on the right).|