Gutkovich YE et al. (2010), Xenopus Meis3 protein lies at a nexus downstrea...

XB-ART-40704

Dev Biol 2010 Feb 01;3381:50-62. doi: 10.1016/j.ydbio.2009.11.024.

Show Gene links Show Anatomy links

Xenopus Meis3 protein lies at a nexus downstream to Zic1 and Pax3 proteins, regulating multiple cell-fates during early nervous system development.

Gutkovich YE , Ofir R , Elkouby YM , Dibner C , Gefen A , Elias S , Frank D .

Abstract
In Xenopus embryos, XMeis3 protein activity is required for normal hindbrain formation. Our results show that XMeis3 protein knock down also causes a loss of primary neuron and neural crest cell lineages, without altering expression of Zic, Sox or Pax3 genes. Knock down or inhibition of the Pax3, Zic1 or Zic5 protein activities extinguishes embryonic expression of the XMeis3 gene, as well as triggering the loss of hindbrain, neural crest and primary neuron cell fates. Ectopic XMeis3 expression can rescue the Zic knock down phenotype. HoxD1 is an XMeis3 direct-target gene, and ectopic HoxD1 expression rescues cell fate losses in either XMeis3 or Zic protein knock down embryos. FGF3 and FGF8 are direct target genes of XMeis3 protein and their expression is lost in XMeis3 morphant embryos. In the genetic cascade controlling embryonic neural cell specification, XMeis3 lies below general-neuralizing, but upstream of FGF and regional-specific genes. Thus, XMeis3 protein is positioned at a key regulatory point, simultaneously regulating multiple neural cell fates during early vertebrate nervous system development.

PubMed ID: 19944089
Article link: Dev Biol

Species referenced: Xenopus laevis
Genes referenced: egr2 fgf3 fgf8 foxd1 foxd3 hesx1 hoxb3 hoxd1 id3 isl1 meis3 myc myt1 nes neurod1 neurog2 nrp1 pax3 snai2 sox15 sox2 tub tubb2b zic1 zic2 zic5
Morpholinos: meis3 MO1 pax3 MO1 zic1 MO1

Article Images: [+] show captions

	Fig. 1. XMeis3 knockdown inhibits primary neuron and neural crest marker expression, but does not inhibit expression levels of Sox and Zic family genes or upstream regulators of neural crest. Two-cell stage albino embryos were injected unilaterally into the animal hemisphere of one blastomere with 10–20 ng of XMeis3 MO. All embryos are injected on the right side, viewed dorsally and are oriented anterior (top), posterior (bottom); the red arrows mark the dorsal midline in A–D. (A) N-tubulin expression is inhibited in 91% of the embryos (n = 100/110). (B) XMyt1 expression is inhibited in 100% of the embryos (n = 12/12). (C) FoxD3 expression is inhibited in 83% of the embryos (n = 116/139). (D) Slug expression is inhibited in 75% of the embryos (n = 70/93). (E) Zic1 expression is normal in 96% of the embryos (n = 26/27). (F) Zic2 expression is normal in 96% of the embryos (n = 23/24). (G) Zic5 expression is normal in 91% of the embryos (n = 91/100). (H) Sox2 expression is normal in 95% of the embryos (n = 20/21). (I) SoxD expression is normal in 96% of the embryos (n = 26/27).(J) Pax3 expression is normal in 85% of the embryos (n = 28/33). 15% of the embryos had minor expression reductions. (K) c-Myc expression is normal in 83% of the embryos (n = 25/30). 17% of the embryos had minor expression reductions. (L) Id3 expression is normal in 84% of the embryos (n = 32/38). 16% of the embryos had minor expression reductions. In all experiments, Krox20 expression inhibition is assayed as a positive control for XMeis3 MO activity. In the described experiments, Krox20 expression is inhibited in 94% of the embryos (n = 190/202). Expression of the proneural gene (not shown), Xngnr1 is inhibited in 77% of the embryos (n = 48/62). For Sox, Zic, Pax3, c-Myc and Id3, overall gene expression levels are quite normal despite morphological perturbations on the XMeis3 MO injected side.
	Fig. 2. Ectopic expression of the HoxD1 antimorph protein knockdown inhibits primary neuron and neural crest markers in addition to hindbrain markers. (A) Two-cell stage albino embryos were injected unilaterally into the animal hemisphere of one blastomere with 100 pg of RNA encoding the HoxD1 antimorph protein. All embryos are injected on the right side, viewed dorsally and are oriented anterior (top), posterior (bottom). (a) Krox20 expression is inhibited in 87% of the embryos (n = 83/95). (b) N-tubulin expression is inhibited in 97% of the embryos (n = 57/59). (c) Slug expression is inhibited in 85% of the embryos (n = 44/52). (d) FoxD3 expression is inhibited in 87% of the embryos (n = 52/60). (e) Pax3 expression is normal in 100% of the embryos (n = 16/16). (f) Zic1 expression is normal in 93% of the embryos (n = 13/14). XMeis3 expression (not shown) is normal in 100% of the embryos (n = 16/16). (B) Ectopic HoxD1 expression rescues n-tubulin expression in XMeis3 morphant embryos. The dashed red lines mark the dorsal midline. (a) Uninjected control embryo. (b) 18 ng XMeis3 MO, n-tubulin expression (right side) was inhibited in 68% (n = 22) of the embryos. (c) 18 ng XMeis3 MO, 0.8 ng HoxD1 RNA, n-tubulin expression was highly rescued (left side) in 75% (n = 12) of the embryos.
	Fig. 3. Inhibition of Zic protein activity by ectopic expression of the Zic5 dominant-negative (DN) protein or the Zic1 MO inhibits hindbrain, primary neuron, and neural crest marker expression. (A) Two-cell stage albino embryos were injected unilaterally into the animal hemisphere of one blastomere with 200 pg of Zic5 DN protein encoding RNA. All embryos are injected on the right side, viewed dorsally and are oriented anterior (top), posterior (bottom). (a) XMeis3 expression is inhibited in 84% of the embryos (n = 42/50). (b) FoxD3 expression is inhibited in 91% of the embryos (n = 74/81). (c) N-tubulin expression is inhibited in 83% of the embryos (n = 91/109). (d) Slug expression is inhibited in 86% of the embryos (n = 37/43). (B) One-cell stage embryos were injected in the animal hemisphere with 0.4 ng of Zic5 dominant-negative protein encoding RNA. Total RNA was isolated from pools of seven neurula stage 16 embryos in each group. Semi-quantitative RT-PCR analysis was performed with the markers: XMeis3, Krox20, n-tubulin, Xngnr1, Slug, FoxD3, and EF1α. In all shown experiments, EF1-alpha serves as a control for quantitating RNA levels in the different samples. RT-PCR was performed on total RNA isolated from normal embryos (not shown in this experiment). (C) One-cell stage embryos were injected in the animal hemisphere with 0.4 ng of Zic5 dominant-negative protein encoding RNA. Total RNA was isolated from pools of eight gastrula stage 11.5 embryos in each group. Semi-quantitative RT-PCR analysis was performed with the markers: HoxD1 and EF1α. -RT-PCR was performed on total RNA isolated from normal embryos (not shown in this experiment). (D) One-cell stage embryos were injected in the animal hemisphere with 25, 30, 35, or 40 ng of Zic1 MO. RNA was isolated from pools of ten neurula stage 18 embryos in each group. Semi-quantitative RT-PCR analysis was performed with the markers: Krox20, HoxB3, NeuroD, n-tubulin, Slug, FoxD3, SoxD, and EF1α. -RT-PCR was performed on total RNA isolated from normal embryos.
	Fig. 4. Ectopic XMeis3 expression rescues the Zic protein knock down phenotypes. (A) One-cell stage embryos were separately injected in the animal hemisphere with 20 ng of Zic1 MO and/or 0.2 ng of XMeis3 RNA. At neurula stage 16, embryos were scored for neural fold inhibition and fixed. 71% of the Zic1 morphant embryos (n = 21) had severely inhibited neural fold closure. In the Zic1 MO/XMeis3 coinjected group, only 23% of the embryos (n = 40) had severely inhibited neural fold closure. Only those embryos with rescued neural folds were taken for Krox20 expression analysis. (a) In the control group, 6/6 embryos stained positive for Krox20 expression; (b) In the Zic1 MO group, 1/6 stained positively. (c, d) In the elongated embryos from the Zic1 MO/XMeis3 injected group, 5/7 stained positively. (B) One-cell stage embryos were separately injected in the animal hemisphere with 20 ng of Zic1 MO and/or 0.2 ng of XMeis3 RNA. Total RNA was isolated from pools of seven gastrula stage 11 embryos in each group. Semi-quantitative RT-PCR analysis was performed with the markers: HoxD1 and EF1-alpha. -RT-PCR was performed (not shown) on total RNA isolated from normal embryos. (C) One-cell stage embryos were separately injected in the animal hemisphere with 20 or 30 ng of Zic1 MO and/or 0.7 or 1.4 ng of XMeis3 RNA. Total RNA was isolated from pools of six to twelve neurula stage 16 embryos in each group. Semi-quantitative RT-PCR analysis was performed with the markers: Krox20, n-tubulin, Nrp1, and EF1-alpha. Nrp1 is a panneural marker. -RT-PCR was performed on total RNA isolated from normal embryos. (D) One-cell stage embryos were separately injected in the animal hemisphere with 45 ng of Zic1 MO and/or 0.9 ng of XMeis3 RNA. Total RNA was isolated from pools of six late neurula stage 20 embryos in each group. Semi-quantitative RT-PCR analysis was performed with the markers: Krox20, FoxD1, and EF1-alpha. -RT-PCR was performed on total RNA isolated from normal embryos. (E) One-cell stage embryos were separately injected in the animal hemisphere with 20 ng of Zic1 MO and/or 0.1–0.2 ng of XMeis3 RNA. (a) At neurula stage 16, embryos were scored for neural fold inhibition and fixed. (b) 89% of the Zic1 morphant embryos (n = 18) had extremely inhibited neural fold closure. (c, d) In the Zic1 MO/XMeis3 coinjected group, 58% of the embryos (n = 57) had fairly normal neural fold closure, versus 11% in the Zic1 morphant group (b). (F) Two-cell stage albino embryos were injected unilaterally into the animal hemisphere of one blastomere with 200 pg of Zic5 DN protein and/or 0.4 ng of HoxD1 or 0.5 ng of XMeis3 encoding RNA. Significant morphological rescue was also observed in the co-injected embryos. All embryos are oriented, right side/injected side, dorsal view, anterior (top), posterior (bottom). The dashed red lines mark the dorsal midline. (a) At neurula stage 17, embryos were scored for Krox20 expression. (b) Krox20 was expressed in only 7.5% of the Zic5 DN injected embryos. (c) Krox20 was expressed highly in 38% of the XMeis3/Zic5 DN co-injected embryos (n = 50). (d) Krox20 was expressed highly in 20% of the HoxD1/Zic5 DN co-injected embryos (n = 50). (G) One-cell stage embryos were separately injected in the animal hemisphere with 0.35 ng of Zic5 DN RNA and/or 0.3 ng of XMeis3 RNA. (a) At neurula stage 17, embryos were scored for n-tubulin expression (b). In the Zic5 DN injected group, 69% of the embryos had medium to undetectable levels of n-tubulin expression (n = 21). (c, d) In the Zic5 DN/XMeis3 injected group 79% of the embryos had medium to strong levels of n-tubulin expression (n = 24). In this same experiment, inhibition of Krox20 expression by the Zic5 DN protein was also rescued by ectopic XMeis3 protein levels (not shown).
	Fig. 6. Pax3 knockdown inhibits primary neuron and hindbrain markers in addition to neural crest marker expression. Two-cell stage albino embryos were injected unilaterally into the animal hemisphere of one blastomere with 20 ng of Pax3 MO. All embryos are viewed dorsally and are oriented anterior (top), posterior (bottom). In each embryo trio panel, the control embryo is on the left, Pax3 mis-match MO injected (mm) center, and the Pax3 MO (MO) on the right. The red arrows mark the dorsal midline. (A) In the shown experiment, XMeis3 expression was inhibited in 76% of the Pax3 morphant embryos (n = 21). Expression was normal in uninjected controls (91%, n = 23) and embryos injected with a 5 bp mm Pax3 MO (90%, n = 22). In all experiments, XMeis3 expression is inhibited in 83% of the morphant embryos (n = 38). (B) In the shown experiment, Slug expression is inhibited in 95% of the morphant embryos (n = 21). Slug expression was normal in uninjected controls (85%, n = 20) and embryos injected with a 5-bp mm Pax3 MO (76%, n = 17). In all experiments, Slug expression is inhibited in 93% of the morphant embryos (n = 55). FoxD3 expression is inhibited in 79% of the morphant embryos (n = 33, not shown). (C) In the shown experiment, n-tubulin expression was inhibited in 78% of the embryos (n = 18) Expression was normal in uninjected controls (81%, n = 21) and embryos injected with a 5-bp mm Pax3 MO (75%, n = 24). In all experiments, n-tubulin expression is inhibited in 90% of the morphant embryos (n = 58). (D) In the shown experiment, Krox20 expression is inhibited in 95% of the morphant embryos (n = 21). Expression was normal in uninjected controls (76%, n = 21) and embryos injected with a 5-bp mm Pax3 MO (74%, n = 19). In all experiments, Krox20 expression is inhibited in 93% of the morphant embryos (n = 133). (E) In the shown experiment, c-myc expression was normal in morphant embryos (86%, n = 22). Expression was normal in uninjected controls (88%, n = 24) and embryos injected with a 5 bp mm Pax3 MO (77%, n = 22). In all experiments, c-Myc expression is normal in 92% of the Pax3 morphant embryos (n = 38). (F) Sox2 expression was normal in morphant embryos (76%, n = 23). Expression was normal in uninjected controls (91%, n = 23) and embryos injected with a 5-bp mm Pax3 MO (82%, n = 22).
	Fig. 7. Ectopic Xngnr1 expression rescues n-tubulin expression in XMeis3 morphant embryos. Two-cell stage albino embryos were co-injected unilaterally into the animal hemisphere of one blastomere with 10–20 ng of XMeis3 MO and/or 0.8 ng of RNA encoding the Xngnr1 protein. All embryos are injected on the right side, viewed dorsally and are oriented anterior (top), posterior (bottom). Whole mount in situ hybridization of the n-tub marker is shown. A. N-tubulin expression is normal in 100% of the uninjected control embryos (n = 22/22). B. N-tubulin expression is highly inhibited in 92% of the XMeis3 morphant embryos (n = 11/12). As a control for XMeis3 MO activity, Krox20 expression was examined (not shown); its expression was inhibited in 90% of the embryos (n = 9/10). (C) Injection of Xngnr1 RNA induces ectopic n-tubulin expression in 87% of the control embryos (n = 20/23). (D) Injection of Xngnr1 RNA induces high ectopic n-tubulin expression levels in 100% of the XMeis3 morphant embryos (n = 18/18). Krox20 was not ectopically activated in these embryos (not shown).
	Fig. 8. FGF3/8 genes lie downstream to XMeis3 protein. (A) FGF3 and FGF8 are XMeis3 direct-target genes. One-cell stage embryos were injected in the animal hemisphere with 250 pg of inducible XMeis3-GR encoding RNA. In all experimental groups, eighteen animal cap explants were respectively removed from embryos at blastula stages 8–9. Explants from each group were grown in cyclohexamide and/or dexamethasone (materials and methods) to stage 12 and total RNA was isolated. RT-PCR analysis was performed with the markers: Krox20, FGF3, FGF8 and EF1α. For controls, RT-PCR and -RT-PCR was performed on total RNA isolated from five normal embryos. (B) XMeis3 knockdown inhibits FGF3 expression. Two-cell stage embryos were injected unilaterally into the animal hemisphere of one blastomere with 7–10 ng of XMeis3 MO. All embryos are viewed dorsally and are oriented anterior (top), posterior (bottom). As designated by the red arrows, in comparison to control embryos (left panel) FGF3 expression in the hindbrain is inhibited on the injected side of the shown embryos (right panel). FGF3 expression was inhibited in 77% of the XMeis3 morphant embryos (n = 177). (C) Ectopic FGF8a expression rescues neural folding and convergent extension in XMeis3 morphant embryos. One-cell stage embryos were injected unilaterally into the animal hemisphere of one blastomere with 19 ng of XMeis3 MO followed by injection of 15 pg of a CMV-vector driving zygotic FGF8a expression. Embryos were cultured until stage 18, and fixed for analysis. All embryos are viewed dorsally and are oriented anterior (top), posterior (bottom). Left panel: uninjected control, center panel: XMeis3 morphant, a shorter embryo with poor neural folding, right panel: XMeis3 MO + FGF8a, the embryo is longer with more distinct neural folds. (D) One-cell stage embryos were separately injected in the animal hemisphere with 15 ng of XMeis3 MO and/or 15–30 pg of FGF8a encoding DNA vector. At neurula stage 16, embryos were fixed and examined for expression of the anterior neural marker. The red arrows emphasize the XANF-1 expression region. 82% of the morphant embryos (n = 28) had expanded XANF-1 expression. In the XMeis3MO/FGF8a coinjected group, only 20% of the embryos (n = 44) had expanded XANF-1 expression, whereas most embryos had fairly normal expression levels. (E) One-cell stage embryos were separately injected in the animal hemisphere with 18 ng of XMeis3 MO and/or 20–30 pg of FGF8a encoding DNA vector. At neurula stage 17, embryos were fixed and examined for expression of the n-tubulin marker. 96% of the morphant embryos (n = 26) had highly reduced n-tubulin expression. In the XMeis3MO/FGF8a coinjected group, 93% of the embryos (n = 55) had rescued-ectopic levels of n-tubulin expression, similar to embryos injected with the FGF8a vector alone.
	Supplemental Figure 1. The XMeis3 MO inhibits neuron and neural crest cell fates in later tadpole stage embryos. A. Islet-1 (Xisl-1) expression is inhibited in XMeis3 morphant embryos in comparison to controls embryos (upper panel), 7.5 ng XMeis3 MO (center panel) or 10 ng XMeis3 MO (lower). Xisl-1 expression was highly inhibited in over 85% of the morphant embryos (n = 69). XMeis3 MO was injected at the one-cell stage. At tadpole stages, Xisl-1 marks a subset of cranial neurons (upper panel). Note that Xisl-1 expression in the heart region of morphant embryos is fairly normal despite the strong neural phenotype. B. Tanabin/nestin expression is inhibited in XMeis3 morphant embryos. At tadpole stages, tanabin/nestin is a marker for cranial nerve cells (upper panel). Expression is ablated in morphant embryos (lower panel). C. Melanocytes are lost in XMeis3 morphant embryos. Compare controls embryos (left panel), to morphants, 10 ng XMeis3 MO (center panel) or 12.5 ng XMeis3 MO (right panel). Melanocyte formation was highly inhibited in over 90% of the morphant embryos (n = 50). XMeis3 MO was injected into one blastomere at the two-cell stage.
	Supplemental Figure 2. mRNA rescue of morphant phenotypes. A. Ectopic expression of XMeis3 mRNA rescues FoxD3 expression in XMeis3 morphant embryos. Injection of the XMeis3 MO (23 ng) strongly inhibited FoxD3 expression in 74% of the morphant embryos (n = 27). In embryos co-injected with 0.2.6 ng of XMeis3 RNA, expression was expressed at fairly high levels in 97% of the embryos (n = 34). In additional experiments, as determined RT-PCR detection, n-tubulin and Krox20 expression was also rescued in morphant embryos. B. Ectopic expression of Zic1 mRNA rescues n-tubulin and Krox20 expression in Zic1 morphant embryos. Injection of the Zic1 MO (15 ng) inhibited n-tubulin expression in 68% of the morphant embryos (n = 19) and Krox20 in 72% of the embryos (n = 18). In embryos co-injected with 300 pg of Zic1 RNA, n-tubulin and Krox20 expression was expressed at high levels in over 70% of the embryos (n = 27 for each group). C. Ectopic expression of Pax3 mRNA rescues FoxD3 and Krox20 expression in Pax3 morphant embryos. Injection of the Pax3 MO (33 ng) inhibited FoxD3 expression in 56% of the morphant embryos (n = 9) and Krox20 in 71% of the embryos (n = 7). In embryos co-injected with 7550 pg of Pax3 RNA, FoxD3 was expressed at high levels in over 75% of the embryos (n = 17) and Krox20 was expressed at high levels in over 55% of the embryos (n = 18).