Effects of overexpression of gbx2 and gbx2-ΔNCR on brain development. Embryos were injected with gbx2 mRNA of 5 pg/embryo (low dose) or 30 pg/embryo (high dose) and gbx2-ΔNCR mRNA of 100 pg/embryo and examined at 26 hpf. (A) Head regions of embryos injected with mRNA. Lateral and dorsal views are shown in the top and bottom rows, respectively. di, diencephalon; ey, eye; FB, forebrain; HB, hindbrain; MB, midbrain; ov, otic vesicle. Normal and disrupted isthmuses are marked with solid and open large triangles, respectively. (B) Embryos injected with mRNA were examined at 26 hpf for the expression of brain regional markers shown on the left. Since the activities of gbx2 differed among mRNA preparations, the mRNA used in (A) was also injected throughout the series of experiments shown here. The most striking phenotypes and their proportions are shown in the respective panels. The positions of the MHB are marked with small triangles. Scale bars, 200 µm.

Effects of overexpression of gbx2 and gbx2-ΔNCR on regionalization of the neural plate at the end of epiboly. Embryos were injected with gbx2 mRNA of 5 pg/embryo (low dose) or 30 pg/embryo (high dose) and gbx2-ΔNCR mRNA of 100 pg/embryo mRNA and examined at the bud stage for the expression of brain regional marker genes shown on the left. The most striking phenotypes and their proportions are shown in the respective panels. The positions of the MHB are marked with small triangles. Scale bars, 200 µm.

Effects of transient gbx2 induction on brain formation during development. (A) Schematic structure of heat-inducible gbx2 (hsp70l:gbx2). FT, FLAG tag; pA, polyadenylation signal. (B) Embryos treated with heat shock (HS) were observed at 24 hpf. Half were normal (a, a2), and the remaining embryos showed defects in the isthmus. (b, b2) Embryos showing incomplete isthmic constriction (mild isthmus defect). (c, c′) Embryos completely lacking isthmic constriction (severe isthmus defect). Scale bar, 200 µm. (C) Offspring embryos from a cross between wild-type and hsp70l:gbx2^+/– Tg fish were exposed to HS at the stages specified at the bottom of the graph and scored for the isthmus defects at 24 hpf. (D) Genotyping of embryos exposed to HS at the specified stages. G, genomic sequences amplified for positive control; H, hsp70l:gbx2-derived sequences showing that the embryos harbored the transgene (arrowheads); (M) size marker (500 bp). All normal embryos were negative, and all affected embryos harbored the transgene.

Temporal changes in competence to gbx2 at the MHB region. (A) Offspring embryos from a cross between wild-type and hsp70l:gbx2^+/– Tg fish were exposed to heat shock (HS) at the stages specified on the left and examined at 24 hpf for the expression of six3b, pax2a, and egr2b (left two columns) or shh and fgf8a (right two columns). The expression of six3b and egr2b in the forebrain and hindbrain (r3/r5), respectively, as well as the expression of shh in the hypothalamus, zli, and floor plate, and that of fgf8a in the forebrain were not affected, but pax2a and fgf8a expression at the MHB (solid triangles) was affected. Large open triangles represent complete disruption of MHB expression of pax2a or fgf8a and small triangles indicate incomplete downregulation. When isthmic expression was retained, expression gaps were observed intermediately along the DV axis. The proportions of the defects and total numbers of scored embryos are shown at the bottom in the respective panels. Approximately half of treated embryos were expected to harbor hsp70l:gbx2. (c–h, c′–h′) The numerators and denominators at the top-right show the numbers of hsp70l:gbx2^+/– embryos, determined by genotyping (B), and total embryos, respectively, showing that the MHB defects observed here were indeed due to gbx2 overexpression. Scale bar, 200 µm. (B) Genotyping of embryos exposed to HS at 80% epiboly. See the legend to Fig. 4D for details. pls, PCR product from hsp70l:gbx2 plasmid as a positive control. Embryos showing MHB mis-specification (MHB defect; downregulation of pax2a) alone possessed the transgene, showing that the defects observed at the MHB were due to gbx2 overexpression.

Rapid downregulation of anterior neural genes by gbx2 induction at the bud stage. Offspring embryos from crosses between wild-type fish and hsp70l:gbx2^+/– Tg fish were exposed to heat shock at the bud stage and examined immediately (0 h) or 1 h and 2 h later for the expression of brain regional genes shown at the top. Reduced expression is shown only when it was observed. The proportions of the defects and total numbers of scored embryos are shown in the respective panels. Dorsal views with anterior to the top. Scale bar, 200 µm.

High sensitivity of otx2 in the anterior neural plate to gbx2 overexpression at the bud stage. (A) Changes in the sensitivity of otx2 in the anterior neural plate to gbx2 overexpression during development. Offspring embryos from crosses between wild-type and hsp70l:gbx2^+/– Tg fish were exposed to heat shock (HS) at the stages specified at the left and examined 2 h after HS for otx2 expression. Each panel is composed of a dorsal (left) and lateral view (right), and the positions of the MHB are marked with solid triangles. ‘+’ and ‘’ show intensities of otx2 expression relative to those in normal embryos (+++). The proportions of the defects and total numbers of scored embryos are shown in the respective panels. Scale bar, 200 µm. (B) Genotyping of embryos subjected to heat treatment and in situ hybridization. See the legends to Fig. 4 and Fig. 5 for details. The results show that downregulation of otx2 depends on the presence of hsp70l:gbx2 (arrowhead).

Effects of gbx2 deletion mutants on brain formation. Embryos were injected with mRNA for egfp, ft-gbx2, gbx2-ΔN, gbx2-ΔC, gbx2-HD, or gbx2-ND and examined at the bud stage (10 hpf, left column) and 28 hpf (right column) for the expression of six3b in the forebrain, pax2a at the MHB, and egr2b in rhombomeres 3 and 5 (r3 and r5, respectively) of the hindbrain or for the morphology of live embryos at 28 hpf (middle column). The most striking phenotypes and their proportions are shown in the respective panels. The positions of the MHB are indicated by small triangles. Normal and disrupted isthmuses are shown with solid and open large triangles, respectively. ey, eye; FB, forebrain; HB, hindbrain; MB, midbrain; ov, otic vesicle. Scale bars, 200 µm.

Effects of transcriptionally active and repressive gbx2 on brain formation. (A) Schematic views of constitutively active and repressive Gbx2. The homeodomain of Gbx2 is placed downstream of the EnR domain and VP16 activator domain (VP16-AD)(En-Gbx2 and VP-Gbx2, respectively). (B) Characteristic phenotypes of embryos overexpressing modified gbx2 genes. Embryos were injected with mRNA for egfp (100 pg/embryo), gbx2 (30 pg/embryo), en-gbx2 (30 pg/embryo), or vp-gbx2 (100 pg/embryo) and observed at 26 hpf. (a–d) Lateral views with anterior to the left and dorsal to the top. (a′–d′) Dorsal views with anterior to the left. Normal and disrupted isthmuses are shown with solid and open large triangles, respectively. ey, eye; FB, forebrain; HB, hindbrain; MB, midbrain; ov, otic vesicle. Scale bar, 200 µm.

Effects of modified gbx2 genes on brain regionalization in the neural plate. (A) Embryos were injected with mRNA for egfp (100 pg/embryo), gbx2 (30 pg/embryo), en-gbx2 (30 pg/embryo), or vp-gbx2 (100 pg/embryo) and examined for the expression of brain regionalization genes at the bud stage. The proportions of the defects and total numbers of scored embryos are shown in the respective panels. The positions of the MHB are indicated by small triangles. Ectopic or expanded expression is marked by asterisks. Scale bar, 200 µm. (B) qPCR analysis of the effects of gbx2-related genes on brain patterning genes. The expression of brain patterning genes was examined in triplicate by qPCR analysis. Differences from control expression (egfp) were statistically tested by the Student’s t-test. p < 0.05; p < 0.01.

Effects of overexpression of gbx2 and gbx2-ΔNCR on brain development. Embryos were injected with gbx2 mRNA of 30 pg/embryo (high dose) and gbx2-ΔNCR mRNA of 100 pg/embryo and examined at 28 hpf for the expression of brain regional markers shown on the left. Lateral (a–f) and dorsal (g–l) views are shown. For the embryos expressing gbx2 and gbx2-ΔNCR, typically affected embryos were chosen and subject to whole mount in situ hybridization, and all the embryos examined assumed the expression patterns shown here. Normal and disrupted isthmuses are marked with solid and open large triangles, respectively. di, diencephalon; ey, eye; tec, tectum; tel, telencephalon. Scale bars, 100 µm.

Effects of different-level overexpression of gbx2 on FGF signaling in embryos. Embryos were injected with gbx2 mRNA of 5 pg/embryo (low dose) or 30 pg/embryo (high dose) and examined at the bud stage (A) or 26 hpf (B) for the expression of FGF downstream genes shown on the left. The most striking phenotypes and their proportions are shown in the respective panels. Lateral views with anterior to the top and dorsal to the right (A) or with anterior to the left and dorsal to the top (B). The positions of the MHB are marked with small triangles. Scale bars, 200 µm.

Inducibility of gbx2 during the development of Tg(hsp70l:gbx2) embryos. Offspring embryos from a cross between wild-type and hsp70l:gbx2^+/- Tg fish were exposed to heat shock at the specified stages and immediately examined for gbx2 expression. (A) Whole mount in situ hybridization revealed that gbx2 was always ectopically induced in approximately half of treated embryos throughout the body. (Upper three rows) Dorsal views with anterior to the top. (Bottom row) Lateral views with anterior to the left and dorsal to the top. Scale bars, 200 µm. (B and C) Quantitative analysis of the gbx2 induction. gbx2 mRNA was quantitated by qPCR in heat-treated embryos that were shown by genotyping to possess hsp70l:gbx2 or not (20 individuals each). The quantities of mRNA (coding region and 3′-UTR)(B and C, respectively) are shown relative to that in wild-type bud-stage embryos that were also subjected to HS. (B) The data show that gbx2 is always induced significantly from the shield stage to 18-somite stage. (C) The level of endogenous gbx2 mRNA, which was quantitated by amplification of the 3′-UTR region, was only marginally upregulated by HS.

Transient nature of hsp70l:gbx2 induction. (A) Offspring embryos from a cross between wild-type and hsp70l:gbx2^+/- Tg fish were exposed to heat shock at the bud stage and examined 0 h, 1 h, or 2 h later for gbx2 expression by whole mount in situ hybridization (hours post-heat shock, hph). The expression of endogenous gbx2 is marked with solid triangles. Both the intensity of staining and proportion of stained embryos were significantly reduced in 2 h. Scale bars, 200 µm. (B) Quantitative analysis of the gbx2 mRNA level by qPCR after hsp70l:gbx2 induction. The quantities of gbx2 mRNA are shown relative to those of wild-type embryos that were treated likewise. The levels of endogenous gbx2 mRNA, which was quantitated by amplification of the 3′UTR region, were less than 2% of the total gbx2 mRNA whether or not treated with heat shock (data not shown).

Effects of gbx2-ΔNCR overexpression on FGF signaling in embryos. Embryos were injected with mRNA for egfp and gbx2-ΔNCR (100 pg/embryo) and examined at the bud stage (left two columns) or at 26 hpf (right two columns) for the expression of the FGF downstream genes specified on the left. The most striking phenotypes and their proportions are shown in the respective panels. (a–h) Dorsal views with anterior to the top (a, b, e, f) or to the left (c, d, g, h). (a′–h′) Lateral views with anterior to the top and dorsal to the right (a′, b′, e′, f′) or with anterior to the left and dorsal to the top (c′, d′, g′, h′). Normal and disrupted MHB/isthmuses are marked with solid and open triangles, respectively. Scale bars, 200 µm.

Overexpression of gbx2 and en-gbx2 elicited dorsalization of embryos. Embryos were injected with mRNA for egfp (100 pg/embryo), gbx2 (30 pg/embryo), en-gbx2 (30 pg/embryo), or vp-gbx2 (100 pg/embryo). Lateral views with dorsal to the right and anterior to the top. (a–d) Bud-stage embryos with percentages showing extended morphology. (e–l) Expression of dorsoventral marker genes in 60%-epiboly embryos injected with gbx2-related genes. Ventral expression of eve1 was reduced by gbx2 and en-gbx2, whereas expanded by vp-gbx2. Dorsal expression of chd showed opposite responses to the same genes. Percentages show the embryos with the phenotypes shown in respective panels. Scale bars, 200 µm.