QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

doi: 10.1242/10.1242/dev.00364

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in Development Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Karlstrom, R. O. Articles by Schier, A. F. Search for Related Content PubMed PubMed Citation Articles by Karlstrom, R. O. Articles by Schier, A. F.

Genetic analysis of zebrafish gli1 and gli2 reveals divergent requirements for gli genes in vertebrate development

Rolf O. Karlstrom^1,*, Oksana V. Tyurina^1,, Atsushi Kawakami^2,3,, Noriyuki Nishioka^4,5, William S. Talbot⁶, Hiroshi Sasaki⁴ and Alexander F. Schier²

¹ Department of Biology, University of Massachusetts, Amherst, MA 01003, USA
² Developmental Genetics Program, Skirball Institute of Biomolecular Medicine and Department of Cell Biology, New York University School of Medicine, New York, NY 10016, USA
³ Department of Biological Science, University of Tokyo, Tokyo, Japan
⁴ Laboratory for Embryonic Induction, Center for Developmental Biology, RIKEN, Kobe, 650-0047 Japan
⁵ Graduate School of Frontier Biosciences, Osaka University, Osaka 565-0871, Japan
⁶ Department of Developmental Biology, Stanford University School of Medicine, Stanford, CA 94305, USA

View larger version (107K): [in a new window]   Fig. 1. dtr–/–, yot–/–and dtr–/+;yot–/+embryos have defects in body axis formation and expression of Hh target genes in the brain. (A-D) Examination of live 36-hour embryos reveals curled body axes in dtr–/–, yot–/–and dtr–/+;yot–/+mutant embryos. U-shaped somites, indicative of defects in slow muscle cell differentiation, are seen only in yot–/–and dtr–/+;yot–/+embryos, dtr–/–embryos have wild-type somites (insets). (E-H) patched 1 (ptc1) expression is generally reduced in all three genotypes. In situ labeling was performed simultaneously and embryos were developed for the same amount of time in E, F and G. Inset in H shows wild-type sibling developed in same tube as this transheterozygote. (I-L) In all gli mutant embryos, nk2.2 expression is reduced or absent from the anterior pituitary anlage (arrowheads), as well as from different regions of the ventral midbrain and ventral hindbrain. (M-P) Expression of pax6, a gene known to be repressed by Hh signaling, is variably expanded in the MDB (arrowhead) and hindbrain (arrows). Expression of pax6 is expanded across the MDB expression domain of shh (not shown), ptc (E), and nk2.2 (I). All panels show lateral views, anterior to the left. Eyes were removed in E-P. Gene expression is indicated on the left. Di, diencephalon; HB, hindbrain; MB, midbrain; MDB, mid-diencephalon boundary; MHB, midbrain-hindbrain boundary; te, telencephalon.

View larger version (53K): [in a new window]   Fig. 2. Zebrafish gli mutations block Hh signaling. (A) Wild-type expression of the Hh target gene nk2.2 is unaffected by injection of lacZ mRNA. (B,C) nk2.2 expression is regionally absent in dtr and yot/gli2 mutant embryos (arrows). (D) Injection of shh mRNA leads to an expansion of nk2.2 throughout the CNS in wild-type embryos (arrowheads). (E,F) Over expression of shh does not activate nk2.2 expression in defective regions of dtr and yot/gli2 mutants (arrows), but nk2.2 expression is expanded in unaffected regions (arrowheads). All panels show lateral views of 20-somite (19 hour) embryos, anterior to the left, eyes removed. di; diencephalon; HB, hindbrain; MB, midbrain; MDB, mid-diencephalon boundary; MHB, midbrain-hindbrain boundary; te, telencephalon.

View larger version (48K): [in a new window]   Fig. 3. Sequence of zebrafish Gli1 and identification of point mutations in the three dtr alleles. (A) The deduced amino acid sequence of zebrafish Gli1 (zfGli1) aligned with mouse Gli1 (mGli1) and zebrafish Gli2 (zfGli2). The entire coding region of gli1 was sequenced in each of the three ENU-induced dtr alleles (dtrtm276, dtrte370and dtrts269) and point mutations were found for each allele (boxes). The altered amino acid in dtrtm276is shown above the box while nonsense mutations are indicated by red hexagons. Gli2 mutations found in you-too are from Karlstrom et al. (Karlstrom et al., 1999). The five zinc finger regions are indicated by lines and potential sites for phosphorylation by protein kinase A (PKA) are indicated by asterisks. A putative VP-16 activator-like domain is indicated by a blue box. Colored sections indicate regions of homology schematized in C. (B) Sequencing ferograms showing point mutations in the three dtr alleles. In dtrtm276U 1633 is mutated to G, changing tyrosine 440 (UAC: Y) into an aspartic acid (GAC: D). In dtrts269 C 2956 is mutated to U, changing glutamine 881 (CAG: Q) into a stop codon (UAG). In dtrte370C 3073 is mutated to U, changing glutamine 920 (CAG: Q) into a stop codon (UAG). (C) Schematic representation of zebrafish and mouse Gli1 and Gli2 protein sequences showing the positions of the stop codons (arrowheads) in the zebrafish mutant alleles. The position corresponding to the site of cleavage that results in a repressor form of Ci is shown by an arrow. Red boxes indicate regions shared among all three sequences, green boxes indicate sequences shared in mouse and zebrafish Gli1 (with percentage amino acid identity indicated), while gray boxes show sequences shared between zebrafish and mouse Gli2 (with percentage amino acid identity indicated). The zinc finger region is marked by ZnFn. Blue box shows region of homology to the VP-16 activator domain, asterisks indicate potential PKA phosphorylation sites. (D) Cladogram showing similarity of mouse (m), frog (Xn) and zebrafish (zf) Gli sequences. Tree is based on ClustalW alignment of amino acid sequences. A search of zebrafish EST databases and genomic trace sequences using mouse Gli1 sequence did not reveal a sequence more similar than the zebrafish Gli1 sequence shown above.

View larger version (59K): [in a new window]   Fig. 4. Activity of Gli1, Gli2 and mutant Gli proteins in MNS70 cells. (A) Schematic of effector and reporter genes co-transfected into MNS70 cells. Different gli constructs were expressed under the control of a CMV promoter. Luciferase activity is induced in a reporter containing 8xGli protein binding sites from the mouse HNF3ß floor plate enhancer (see Sasaki et al., 1999). (B) pcDNA constructs encoding mouse Gli1 (mGli1) and mouse Gli2 (mGli2) both activate the luciferase reporter. A pcDNA construct encoding full-length zebrafish Gli1 (zfGli1) activates luciferase activity, while pcDNA constructs encoding zebrafish Gli2 (zfGli2) or the dtr/gli1 (tm276, te370, ts269) or yot/gli2 (ty119, ty17) mutations show no activation. When co-transfected with full-length gli1, dtrtm276(but not dtrte370or dtrts269) enhances reporter gene activation by wild-type Gli1. In contrast, co-transfection of gli1 with constructs encoding full-length Gli2 or the C-terminally truncated yot alleles result in the elimination of Gli1 mediated transcriptional activation. Transfection with a pJT4 plasmid encoding Shh activates luciferase activity. Co-transfection with pcDNA-zfgli1 and pJT4-shh has a roughly additive effect on luciferase activity. Co-transfection of pcDNA-gli2 with pJT4-shh reduces the luciferase activity induced by Shh alone. Averaged results of 2 experiments with standard errors. Relative luciferase activities are indicated by bars while protein schematics at top show the sites of the mutations encoded by each gli mutant construct. (C) Western analysis showing Gli proteins produced in cell culture. Asterisks indicate bands of predicted size for each transfected construct.

View larger version (65K): [in a new window]   Fig. 5. gli1 knockdown phenocopies dtr/gli1. (A) Expression of nk2.2 is unaffected by injection of a control morpholino (MO). (B) dtr/gli1 mutations eliminate nk2.2 expression in some regions of the brain (arrow and arrowhead). (C) Injection of a gli1MO into wild-type embryos leads to a loss of nk2.2 expression identical to that seen in dtr/gli1 mutant embryos (compare arrows, see also Table 1). (D) Expression of fkd4 in the medial and lateral floor plate is unaffected by control MO injections. (E) fkd4 expression is extremely reduced in dtr/gli1 mutants (compare bracket and arrow to those in D). (F) fkd4 expression is similarly reduced in lateral floor plate cells after gli1 MO injection (compare bracket and arrow to those in D). (d', e', and f') show cross sections through the trunk at the level of the yolk plug. nc; notochord

View larger version (111K): [in a new window]   Fig. 6. Developmental expression of zebrafish gli1. (A) 80% epiboly. Transcripts for gli1 are first detected in the anterior neural plate (arrowhead) and in pre-somitic mesoderm (arrows). (B) 2-somite stage. In the trunk, both gli1 (left panel) and gli2 (right panel) are expressed in adaxial cells (arrowheads) adjacent to the notochord. gli1, like gli2, is also expressed in paraxial mesoderm, with gli2 expression extending more laterally (arrows). (C) 5-somite stage. gli1 is expressed throughout the anterior neural plate (white arrowhead), in adaxial cells that give rise to slow muscle fibers (black arrowheads), as well as in the tailbud (out of focus). Some patchy expression is present in the developing spinal cord (arrow). (D) 10-somite stage, dorsal view (left) and cross section (right) of the trunk. gli1 expression continues in adaxial cells (arrowheads) and spreads laterally into developing somites (asterisk). gli1 is expressed ventrally in the spinal cord (larger arrow) but not in floor plate cells adjacent to the notochord (smaller arrow). (E-J) Lateral views of the brain, eyes have been removed. (E) 10-somite stage. gli1 is expressed throughout the ventral forebrain, midbrain, hindbrain, and spinal cord (not shown). (F) 20-somite stage. In the brain, gli1 is expressed in ventral regions in a pattern similar to that of ptc1 (see Fig. 3). In the forebrain, gli1 is primarily expressed in the diencephalon, but expression also extends into the ventral telencephalon dorsal to the optic recess (black dot). Expression is now absent in the ventral-most diencephalon, with the exception of a large patch in the posterior part of the developing hypothalamus (arrow). (G,H) 24 hours and 30 hours. gli1 expression continues in the ventral CNS, including in the pre- and postoptic areas on either side of the optic recess (black dot) and in the patch in the posterior hypothalamus (arrow). (I) Expression in the trunk at 30 hours. gli1 is strongly expressed in the spinal cord (arrows) and is more weakly expressed in somites. Cross section through trunk (right) shows spinal cord gli1 expression (larger arrow) is absent from dorsal cells and ventral floor plate cells (smaller arrow). (J) 36 hours. By 36 hours, gli1 is expressed predominantly along the diencephalon/telencephalon border and in the ventral hypothalamus, including the region of the anterior pituitary anlage (arrowhead). gli1 is also expressed in a small patch in the telencephalon (arrow) and in endoderm (white arrow). (K) Expression in the fin bud at 36 hours. Both gli1 (left) and gli2 (right) are expressed in the pectoral fin buds (arrowheads). gli1 expression is more limited than gli2, being predominantly in the posterior and distal mesenchyme, while gli2 is expressed throughout the fin mesenchyme (compare arrowheads). (A-D) and (K) are dorsal views, (E-J) are lateral views. Anterior is to the left in all panels except (A) and (K), where anterior is up. di; diencephalon, FB; forebrain, HB; hindbrain, hy; hypothalamus, MB; midbrain, MDB; mid-diencephalon boundary, MHB; midbrain-hindbrain boundary, nc; notochord, te; telencephalon.

View larger version (119K): [in a new window]   Fig. 7. Hh signaling regulates gli1 expression. (A,B) Over expression of shh in wild type expands gli1 expression dorsally throughout the embryo (compare arrowheads). (C,D) gli1 expression is extremely reduced in Hh signaling-defective smu/smo mutant embryos relative to wild-type siblings, especially in the diencephalon (arrows). Some gli1 expression remains in the ventral spinal cord and hindbrain (arrowheads). (E) Dorsal view of wild-type gli1 expression in a 4-somite stage embryo; treated with ethanol (cyclopamine carrier). (F) In 4-somite stage smu/smo mutants, gli1 expression is reduced in adaxial cells (arrowhead) and is less affected in the developing brain (arrows). (G) Similarly, cyclopamine treatment of wild-type or smu/smo embryos reduces but does not eliminate gli1 expression. All 40 cyclopamine-treated embryos from a smu-/+incross showed the same gli1 labeling pattern, indicating that the smu/smo mutation blocks Hh signaling as completely as cyclopamine, and that maternal smu/smo function is not responsible for low level gli1 expression in smu/smo mutant embryos.

View larger version (112K): [in a new window]   Fig. 8. gli2 MO injection rescues nk2.2 and myoD expression defects in yot/gli2 mutants and reveals a weak activator role for Gli2. (A) gli2 MO injection expands ptc1 expression ventrally in the diencephalon (arrowheads) and causes a minor but consistent overall expansion of ptc1 expression (compare to inset). (B) yot/gli2 mutants have significantly reduced ptc1 expression. (C) gli2MO injections rescue the ptc1 defects seen in yot/gli2 mutants and expand ptc1 expression ventrally (arrowhead). (D) Injection of gli2MOs into wild-type embryos has no effect on nk2.2 expression. (E,F) Injection of a gli2MOs into yot–/–mutant embryos can completely rescue yot-induced defects in nk2.2 expression (compare arrows). (G) gli2MO injection does not affect myoD expression in adaxial cells (arrowheads). (H,I) gli2MO injections partially rescue yot-induced defects in adaxial myoD expression (compare arrowheads). (J-M) Injection of 3-10 ng of gli2MO into embryos from a cross between dtr–/+heterozygous parents (25% dtr–/–mutants expected) resulted in an additional loss of nk2.2 expression in the tegmentum (compare arrows in J,K) and a reduction in adaxial myoD expression (compare arrowheads in L,M) in 60/206 embryos (29%), all of which were dtr–/–mutants as judged by forebrain and hindbrain nk2.2 expression defects. This suggests Gli2 may activate Hh signaling in a small area of the ventral midbrain and in adaxial cells. Control MO injections had no effect on nk2.2 expression in 85/85 embryos from a similar dtr–/+x dtr–/+cross, with 25 embryos (29%) showing the dtr–/–nk2.2 defects (J) and 60 embryos (71%) showing wild-type nk2.2 expression as expected for dtr–/+ and dtr+/+embryos. (A-F,J, and K) are lateral views of the head, eyes removed. (G-I,L, and M) are dorsal views of the tail region. All embryos are at the 20 somite (19 hour) stage. For yot/gli2, embryo genotypes were inferred by myoD expression in adaxial cells, then were verified by PCR (not shown, see Materials and Methods). D and G, E and H, F and I, J and L and K and M show the same individual labeled simultaneously with nk2.2 and myoD.

View larger version (101K): [in a new window]   Fig. 9. Regulation of nk2.1b by gli2, yot/gli2, smu/smo and dtr/gli1. (A) nk2.1b is normally expressed in the anterior/ventral telencephalon (arrowhead) and in the diencephalon (bracket). (B) gli2MO injection into wild-type embryos leads to a dorsal expansion of telencephalic nk2.1b expression (arrowhead), as well as an increase in expression in the hypothalamus (compare brackets). This expansion was seen in 70/72 wild-type embryos injected with 10 ng of gli 2MO. (C) yot–/–embryos have reduced nk2.1b expression in the diencephalon adjacent to the first ventricle (arrow). (D) gli2 MO injection into yot–/–embryos rescues the diencephalic nk2.1b expression defect (compare arrows in C and D, Table 2), and also leads to expanded expression in the telencephalon (compare arrowheads). (E) nk2.1b expression is extremely reduced in smu/smo mutants, with small patches of expression remaining in the diencephalon and telencephalon (arrowhead). (F) Injection of 10 ng of gli2 MO into embryos from a cross of two smu+/–parents resulted in telencephalic nk2.1b expansion (arrowhead) in 89/89 embryos, including 18 smu–/–embryos (20%) and 71 wild-type and heterozygous siblings (80%). This shows that Gli2 repression of this Hh target gene is independent of Hh signaling. No nk2.1b expansion was detected in 49/49 embryos injected with 10 ng of control MO. (G) dtr–/–embryos have reduced nk2.1b expression in the diencephalon adjacent to the first ventricle (arrow) similar to the yot/gli2 phenotype. (H) gli2 MO injection does not rescue diencephalic nk2.1b expression in dtr/gli1 mutants, but does expand nk2.1b expression in the telencephalon (arrowhead). Injection of 3-7 ng of gli2 MO resulted in telencephalic nk2.1b expansion in 64/64 embryos, including 6 embryos (10%) that were clearly homozygous dtr–/–mutants based on diencephalic nk2.1b defects. The remaining 58 siblings (90%) also had expanded telencephalic nk2.1b expression. All panels show 30-hour embryos, lateral views of the forebrain, eyes removed, anterior to the left. All panel pairs show sibling embryos from the same experiment. Dot shows the optic recess, the anterior edge of the border between the diencephalon (di) and telencephalon (te).