November 1, 2005;
Maternal Xenopus Zic2 negatively regulates Nodal-related gene expression during anteroposterior patterning.
During the development of Xenopus laevis, maternal mRNAs and proteins stored in the egg
direct early patterning events such as the specification of the dorsoventral axis and primary germ layers. In an expression screen to identify maternal factors important for early development, we isolated a truncated cDNA for maternal Zic2
(tZic2), encoding a zinc-finger transcription factor. The predicted tZic2 protein lacked the N-terminal region, but retained the zinc-finger domain. When expressed in embryos, tZic2 inhibited head
and axial development, and blocked the ability of full-length Zic2
to induce neural crest genes. Depletion of maternal Zic2
from oocytes, using antisense oligonucleotides, caused exogastrulation, anterior
truncations and axial defects. We show that loss of maternal Zic2
results in persistent and increased expression of Xenopus nodal
-related (Xnr) genes, except for Xnr4
, and overall increased Nodal
signaling. Injection of a Nodal
-short, reduced the severity of head
and axial defects in Zic2
-depleted embryos. Depletion of Zic2
could not restore Xnr expression to embryos additionally depleted of VegT
, a T-domain transcription factor and an activator of Xnr gene transcription. Taken together, our results suggest a role for maternal Zic2
in the suppression of Xnr genes in early development. ZIC2
is mutated in human holoprosencephaly (HPE), a severe defect in brain
hemisphere separation, and these results strengthen the suggestion that increased Nodal
-related activity is a cause of HPE.
[+] show captions
Fig. 2. Depletion of maternal Zic2. (A) Dose-dependent depletion of Zic2 from oocytes of two different donor females. Oligo 10MP was injected at the indicated doses and oocytes were cultured for 24 hours before analysis by RT-PCR. Samples were quantitated relative to uninjected oocytes (Un) from female 1 (black bar). (B) Zic2 is the only Zic gene expressed maternally. RT-PCR of Zic1, Zic2, Zic3 and Zic5 expression in stage 10.5 embryos (Em; stage 14 for Zic5), stage VI oocytes (WO) or animal and vegetal halves of oocytes (An and Vg). (C,D) Representative phenotypes of uninjected (C) and Zic2-depleted (D) embryos at stage 38/39. (E,F) Zygotic Zic2 is expressed in maternal Zic2-depleted embryos. (E) Whole-mount in situ hybridization of Zic2 expression in stage 18 uninjected (two lower left embryos) and maternal Zic2-depleted embryos (three upper right). Although neural tube formation is delayed, Zic2 is expressed anteriorly and at the neural plate border. (F) Gastrula stage series of Zic2 expression in uninjected and Zic2-depleted embryos. (G) Histological sections of embryos from C,D taken through the head (upper panel) or just behind the level of the otic vesicle (lower panel). In each panel, the uninjected embryo is on the left and the Zic2-depleted embryo is on the right. The arrow indicates a duplicated notochord. (H) Real-time RT-PCR analysis of neural markers in control and Zic2-embryos. Relative expression values were normalized to ODC and expressed as a percentage of a stage 22 embryo.
Fig. 5. Increased Xnr expression in Zic2-depleted embryos is restricted to vegetal cells. (A) Xnr5 in situ hybridization at stages 9 (upper panel) and 10.5 (lower panel). Dorsal is oriented towards the top of the figure. Xnr5 is expressed in dorsal vegetal cells of uninjected and Zic2-depleted embryos. Xnr5 expression at stage 10.5 is undetectable in control embryos (0/11) but is present ventrally in Zic2-depleted embryos (8/13; arrow). (B) Expression of Xnr genes and Xnr target genes in dorsal (D) and ventral (V) halves at stage 10.25. Uninjected (Uninj.) 10.25 and Zic2-depleted (Zic2-) 10.25 are intact embryos. –RT, stage 10.25 sample processed in the absence of reverse transcriptase. All genes shown are expressed ventrally. (C) Expression of Xnr genes and Xnr target genes in equatorial (Eq) and vegetal base (Bs) explants.
Fig. 1. Isolation and dominant-interfering activity of a truncated Zic2 cDNA. (A) Upper panel: diagram showing the Xenopus Zic2 protein compared with the tZic2 protein. Lower panel: partial alignment of the amino acid sequences of the full-length and truncated-Zic2 proteins. Two potential starting methionines are boxed. The start of the zinc-finger domain is underlined. (B) In vitro translation of tZic2. Two preparations of tZic2 plasmid were translated in transcription-translation coupled reactions. Luciferase (Luc), ∼50 kDa. (C) Injection of tZic2 causes microcephaly and cyclopia. Uninjected and tZic2-injected (500 pg) embryos at stage 40. (D) Hematoxylin and Eosin stained sections from the embryos in C showing reduced forebrains (large arrowhead) and abnormal or cyclopic eyes (small arrowhead). (E) Real-time RT-PCR analysis of neural crest (Slug, Sox9, Twist) and neural (NCAM) marker expression in animal caps cultured to stage 22. Samples and amounts injected are indicated in the key. Relative expression values; samples were normalized to ODC and expressed as a percentage of a whole stage 22 embryo sample.
Fig. 3. Zic2 regulates Xnr expression during gastrulation. (A) Dose-dependent induction of Xnr5 by injected tZic2 RNA (330 pg, 660 pg, 990 pg) in whole embryos assayed by real-time RT-PCR. Un, uninjected embryo. (B) Expression of Xnrs, mesoderm and endoderm markers in control (Un) and Zic2-depleted (Zic2-; 5 ng oligo 10MP) embryos assayed by real-time RT-PCR at stage 10.25 (left panel) and stage 10.5 (right panel). (C) Increased activin-like activity in Zic2-depleted embryos. Mean relative luciferase units (RLU)=A3-luciferase activity (3× activin response element)/RLTK (Renilla luciferase, thymidine kinase promoter) activity. Error bars show s.d. in a representative experiment; the asterisk indicates statistical significance (P<0.05).
Fig. 4. Rescue of Xnr expression and normal development by injected Zic2 mRNA. (A) Expression of Xnr1, Xnr3 and Xnr5 in uninjected controls (Un), Zic2-depleted embryos and Zic2-depleted embryos (5 ng 10MP) injected with Zic2 mRNA (75 pg into oocytes prior to host transfer). Expression values are relative to the uninjected stage 10.25 sample. (B) Expression of Chordin, Gsc and Xhex in the same samples as A. (C) Examples of embryos from the experiment in A,B shown at two stages, late gastrula (stage 12, left panels) and tailbud (stage 36, right panels). (D) Expression of Xnr1 and Xnr5 in a second rescue experiment, as in A, except that 200 pg of Zic2 mRNA was used.
Fig. 6. Rescue of Zic2-depletion by injection of CerS. (A) Analysis of microcephaly (microceph.) and axial truncations (axial trunc.) in control stage 33 embryos (Un), controls injected with 50 pg CerS mRNA (Un+CerS), Zic2-depleted embryos (Zic2-) or Zic2-embryos injected with 50 pg CerS mRNA (Zic2-, +CerS). Data were pooled from two host-transfer experiments. Bars represent the number of embryos with the indicated phenotype. Axial truncations were present in a subset of embryos with microcephaly. (B) Expression of Xnr1, Xnr5 and Xnr6 in gastrula (stage 10.5) stage embryos analyzed by RT-PCR. (C,C′) Examples of embryos from the experiment in B. Zic2-embryos show both microcephaly and axial truncation; the Zic2-, +CerS embryos represent a normal and a microcephalic embryo (upper and lower embryo, respectively).
Fig. 7. Zic2 antagonizes VegT regulation of Xnr expression. (A) Expression of Xnr1, Xnr3 and Xnr5 in whole embryos at stage 10.5. Un, uninjected; Zic2-, 5 ng 10MP; VegT MO, 18 ng morpholino oligo. (B) Expression of Xnr5 in uninjected and Zic2-depleted embryos in the absence or presence of VegT mRNA (200 pg). Oligos were injected vegetally in A and Zic2 oligo and VegT RNA were injected animally in B. (C) Zic2 does not inhibit VegT in animal caps. Zic2 mRNA (500 pg), VegT mRNA (250 pg) or a combination of the two, were injected at the two-cell stage. Animal caps were dissected at the early gastrula stage and cultured for 2 hours.