SEARCH

SEARCH BY CITATION

Keywords:

  • ROR alpha;
  • cerebellum;
  • gene expression;
  • evolution;
  • zebrafish

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Mouse genetic studies have identified several genes involved in cerebellar development. The mouse mutants staggerer and lurcher are functionally deficient for the retinoid-related orphan receptor alpha (ROR alpha) and glutamate receptor delta2 (Grid2) genes, respectively, and they show similar functional and developmental abnormalities in the cerebellum. Here, we report the cloning and expression pattern of zebrafish ROR alpha orthologues rora1 and rora2, and compare their expression pattern with that of grid2. Expression of rora1 and rora2 is initiated at late gastrula and pharyngula stages, respectively. Both rora1 and rora2 are spatially expressed in the retina and tectum. Expression of rora2 was further observed in the cerebellum, as reported for mammalian ROR alpha. In the adult brain, rora2 and grid2 are coexpressed in brain regions, designated as cerebellar-like structures. These observations suggest an evolutionarily conserved function of ROR alpha orthologues in the vertebrate brain. Developmental Dynamics 236:2694–2701, 2007. © 2007 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Morphological analysis of teleost and mammalian brains reveals both similarities and differences (Butler and Hodos,1996; Meek and Niewenhuys,1997). Among vertebrate brains, the cerebellum is one of the most conserved structures. In the cerebellum of fish and mammals, the dendrites of Purkinje cells and axons of granule cells form synaptic connections within the molecular layer which overlies a layer formed by the cell bodies of the Purkinje cells, and granule cells reside beneath this layer of Purkinje cells. However, there are some distinct morphological differences. For example, the mammalian cerebellum is highly foliated, whereas the cerebellum of teleosts exhibits no such foliation. On the other hand, anatomical and functional similarities to the cerebellum are suggested for some structures of the vertebrate brain, such as the cochlear nucleus. These are referred to as “cerebellar-like structures.” The cerebellar-like structures, including the cerebellum itself, have similar neuronal circuitry. They have a layer of principal neurons, which receive two major inputs: parallel fibers and primary afferents in the apical and basal area of the dendrites, respectively (Devor,2000; Bell,2002). Despite the common features among the cerebellar-like structures, it remains unclear whether they are homologous (Bell,2002). Prior studies investigating these cerebellar-like structures remain inconclusive as to whether similar neuronal circuits originated from multiple regions of the brain across different vertebrate species.

Ataxia is a typical symptom of abnormal cerebellar function, and analysis of mouse mutants that display ataxic movements have played key roles in the identification of genes essential for cerebellar development. One such mutant is staggerer, which lacks the retinoid-related orphan receptor alpha (ROR alpha) gene. Spontaneous (staggerer) or a targeted mutation of ROR alpha gene leads to cerebellar atrophy caused by death of both Purkinje and granule cells (Hamilton et al.,1996; Dussault et al.,1998; Steinmayr et al.,1998). ROR alpha is strongly expressed in both developing and mature Purkinje cells (Hamilton et al.,1996; Nakagawa et al.,1997; Ino,2004) and controls the expression of Sonic Hedgehog (Shh) in Purkinje cells, which in turn promotes proliferation of granule cells (Wallace,1999; Wechsler-Reya and Scott,1999; Dahmane and Ruiz-i-Altaba,1999). Thus, ROR alpha primarily functions in Purkinje cells and indirectly affects the proliferation of granule cells. ROR alpha is also expressed in other brain regions such as layer IV of the cerebral cortex and cartwheel cells of the dorsal cochlear nucleus (Ino,2004; Nakagawa and O'Leary,2003), but its function in these regions is not defined. The lurcher mouse harbors a mutation in the glutamate receptor delta 2 (Grid2) gene, in which there is a single nucleotide substitution that changes a highly conserved alanine to threonine (Zuo et al.,1997). This mutation adversely changes the membrane permeability of Purkinje cells, and induces cell death. In the adult mouse brain, the Grid2 protein is highly expressed in Purkinje cells and is also expressed at lower levels in several hindbrain neurons (Araki et al.,1993; Lomelie et al.,1993; Takayama et al.,1996). Because of the similarities of the staggerer and lurcher mutant phenotypes, functional relationships between ROR alpha and Grid2 have been inferred (Messer et al.,1991; Soha and Herrup,1995; Messer and Kang,2000).

Here, we have discovered that the zebrafish genome contains two ROR alpha orthologues: rora1 and rora2. Analyses of gene expression pattern by whole-mount in situ hybridization showed that expression of rora2, but not rora1 was detected in the cerebellum. A comparative analysis of the expression patterns of rora2 and grid2 in the embryonic and adult zebrafish brains demonstrates that these genes are coexpressed in cerebellar-like structures.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Zebrafish Genome Contains Two ROR alpha Genes

Through mining of the zebrafish genome database, we have identified orthologues of ROR alpha and named these rora1 and rora2. Using a polymerase chain reaction (PCR) -based strategy, we amplified a single rora1 cDNA fragment, whereas two species of rora2 cDNA sequence were amplified. The shorter rora2 cDNA had a deletion of 111 nucleotides (indicated by green letters in Fig. 1A) when compared with the other PCR fragments. The full-length cDNA sequence of rora1 was identified within the NCBI database (accession number BC051158). The full-length cDNA sequence of rora2 (accession number AB298802) was obtained using a cRACE protocol. Comparison of amino acid sequences of the putative ROR proteins showed that Rora2 exhibited a higher amino acid identity to human ROR alpha than Rora1, and Rora1 contains an insertional sequence that is not observed in either human ROR alpha or Rora2 (Fig. 1B). When the amino acid sequence of ROR proteins including beta and gamma of human was aligned, there are amino acid residues conserved among ROR alpha, Rora1, and Rora2, but not in beta and gamma (Fig. 1B). Thus, we concluded that these two genes are genuine zebrafish orthologues of ROR alpha, and rora1 is a diversified gene within the ROR alpha subgroup. A molecular phylogenetic tree supports this conclusion (Fig. 1C). Within the protein-coding region, the position of exon–intron junction was completely conserved among human and zebrafish ROR alpha genes. The 111 base nucleotide sequence, which was not observed in the shorter rora2 cDNA (indicated by green letters in Fig. 1A), corresponded to an exon. This finding suggested that the two species of rora2 cDNA that we identified in our amplified PCR products were derived from endogenous transcripts and are generated by alternative splicing.

thumbnail image

Figure 1. A: The sequence of rora2 (accession no. AB298802), a zebrafish homologue of ROR alpha. Each exon is indicated by alternate colors. The nucleotide sequence indicated by green is deleted by alternative splicing. The sequences underlined by blue arrows were used for cRACE. B: Comparison of amino acid sequence of zebrafish ROR alpha (Rora1 and Rora2) and human ROR proteins. Red and green boxes indicate the zinc finger domain and the ligand-binding domain, respectively. C: The phylogenetic tree of ROR alpha and the related proteins generated by CLUSTALW.

Download figure to PowerPoint

Expression Pattern of rora1 and rora2 during Zebrafish Development

We carried out whole-mount in situ hybridization to examine the spatial expression pattern of rora1 and rora2 during zebrafish development. rora1 expression was detectable from a late gastrula stage (10 hpf; Fig. 2A). The expression at this stage was observed as two spots, which are bilateral and partially fused in the midline. These spots became separated during the early and mid-somite stages (Fig. 2B,C). rora1 expression was observed very weakly at the 14-somite stage in the midbrain region (Fig. 2C). Lateral observation of the in situ specimens revealed that rora1 expression in the midbrain intensified and extended along the anterior–posterior axis as embryogenesis proceeded (Figs. 2, 3B–E). A frontal view of the specimens showed that expression of rora1 in the midbrain was a faint spot within the midline of the 18-somite embryos (Fig. 2D). This expression spread laterally in the prim-5 stage (Fig. 2E), and then separated bilaterally in the 2 days postfertilization (dpf) brain (Fig. 2F). The 3 dpf brains then appeared fused again in the midline (Fig. 2G). We interpret this finding to mean that the temporal and spatial expression pattern of rora1 in the midbrain region indicates either neuronal differentiation or maturation within the developing tectum. Collectively, rora1 was expressed in the eyes and the tectum during zebrafish development. The rora1 expression in these regions also persisted in the 5 dpf larvae (data not shown).

thumbnail image

Figure 2. Expression pattern of rora1 detected by whole-mount in situ hybridization. A–D: Dorsal views of developing brain. E–G: Views from the anterior top. Embryonic expression of rora1 starts in the eye rudiments from the 1-somite stage (10 hours postfertilization [hpf]), and expression in the presumptive tectum was detected from the 14-somite stage (16 hpf).

Download figure to PowerPoint

thumbnail image

Figure 3. A–H: Lateral views of in situ hybridization specimens in which expression of rora1 (A–E) or rora2 (F–H) was detected. A: At the five-somite stage (12 hours postfertilization [hpf]). B: At the 18-somite stage (18 hpf). C,F: Expression of rora1 (C) was observed in the eyes and midbrain, but expression of rora2 (F) was not detected in 24 hpf specimens. D,G: Expression of rora1 (D) and rora2 (G) was similarly observed in the eyes and tectum, but only rora2 expression was observed in the hindbrain of 2 days postfertilization (dpf) specimens (G). H: Expression of rora2 in the cerebellum is observed in 3 dpf specimens.

Download figure to PowerPoint

In the 2 dpf specimens, strong expression of rora2 was observed in a similar pattern to that of rora1 in the eyes and the midbrain region (Fig. 4A). Of interest, the expression of rora2 was also observed in the hindbrain region (Figs. 3G, 4B) where rora1 expression was never detected (Fig. 3D). Expression of rora2 was also observed in the upper rhombic lip of 3 dpf larvae (Figs. 3H, 4D, indicated by arrows). Expression of rora2 in the retina of 3 dpf specimens was weak when compared with that of rora1 and became spatially restricted in the inner nuclear layer (Fig. 4C). The forebrain region of 3 dpf specimens showed weak expression of rora2 (Fig. 4C).

thumbnail image

Figure 4. Expression pattern of rora2 in the larval brain and eyes detected by whole-mount in situ hybridization. A–D: Frontal (A,C) and dorsal (B,D) views of 2 days postfertilization (dpf; A,B) and 3 dpf (C,D) specimens are shown. Frontal and dorsal views of a 3 dpf larva clearly show rora2 expression in the eyes and differentiating Purkinje cells (indicated by arrows in D), respectively.

Download figure to PowerPoint

Similar Expression Pattern of rora2 and glutamate receptor delta2

The spatial expression pattern of rora1 and rora2 in the adult zebrafish brain was examined. When digoxigenin (DIG) -labeled RNA probe of rora1 was hybridized to the sections, no specific staining was observed, suggesting that rora1 is not expressed in the adult brain (not shown). Hybridization of rora2 probe gave distinctive staining patterns as described below, and we noticed that the expression pattern of rora2 in the adult brain is very similar to that of grid2 described previously (Mikami et al.,2004). Thus, we carried out a detailed comparison of expression patterns of rora2 and grid2 in the adult and developing brains of zebrafish. Mikami et al. (2004) reported specific expression of grid2 in the cerebellar Purkinje cells (Fig. 5G,I), cells in a layer of the tectum termed the stratum fibrosum et griseum superficiale (SFGS; Fig. 5G), and also in the crest cells in the medial octavolateralis nucleus (MON; Fig. 5K). In these regions, grid2-positive cells were observed as densely stained dots. Expression of rora2 was similarly observed in these neuronal populations (Fig. 5H,J,L). Our in situ hybridization studies uncovered that grid2 expression was detected in the granular cell layer of the cerebellum and in the periventricular zone of the tectum (Fig. 5G). Expression of grid2 and rora2 was observed in the mammillary body (MC) with strong staining at the boundary of this brain nucleus, and in the caudal part of the periventricular hypothalamus (Hc; Fig. 5E,F). In sections hybridized to grid2 probe, weak staining was observed in the ventral region of the telencephalon (Fig. 5C). Expression of rora2 in the telencephalon was not definitive (Fig. 5D).

thumbnail image

Figure 5. A–L: Expression pattern of grid2, glutamate receptor delta2 (A,C,E,G,I,K) and rora2 (B,D,F,H,J,L) was detected by in situ hybridization in adjacent sections of an adult brain of zebrafish. High magnification views of the boxed regions in A and B are shown in the ventral part of the telencephalon (C,D), the ventral part of the midbrain (E,F), the optic tectum and anterior part of the cerebellum (G,H), the posterior part of the cerebellum (I,J), and the dorsal part of the pons (K,L). Colocalized expression of these two genes was observed in the cerebellar-like structures. Abbreviations are the same as used in Wullimann et al. (1996). Vdm, dorsal nucleus of V; Vv, ventral nucleus of V; PTN, posterior tuberal nucleus; Hc, caudal zone of periventricular hypothalamus; CM, corpus mamillare; SM, stratum marginale; SO, stratum opticum; SFGS, Stratum fibrosum et griseum superficiale; SGC, stratum griseum centrale; PGZ, periventricular gray zone; Val, lateral division of valvula cerebelli; CCe, corpus cerebelli; PL, Purkinje cell layer; LCa, lobus caudalis cerebelli; MON, medial octavolateralis nucleus.

Download figure to PowerPoint

To indirectly investigate the transcriptional regulation of grid2 and rora2, we examined the expression pattern of these genes during development by whole-mount in situ hybridization. Their expression pattern was visualized in specimens of 48, 72, and 120 hours postfertilization (hpf) stages. As shown in Figure 6, expression of both rora2 and grid2 was observed in the telencephalon (indicated by red arrow), the tectum (indicated by black arrow), the upper rhombic lip (indicated by blue box), the hindbrain (indicated by black box), and the retina. These two genes exhibited spatially similar expression patterns in the telencephalon and the tectum (Fig. 6A–D). A significant difference in their expression pattern can be observed in the upper rhombic lip, which is the presumptive cerebellar region. Expression of rora2 was sometimes detected as a very faint spot in the dorsal-most region in some 48 hpf larvae (Fig. 6B), but not in others (Fig. 4A,B). This result suggests that 48 hpf is the temporal period when rora2 expression begins in the upper rhombic lip. The upper rhombic lip also strongly expressed grid2 at this developmental stage (Fig. 6A,C,E). Expression of rora2 in the 72 hpf rhombic lip was observed as a thin stripe (Fig. 6D) that became stronger at 120 hpf (Fig. 6F). In the adult brain, rora2 was strongly expressed in Purkinje cells, whereas strong expression of grid2 was observed in both Purkinje and granule cells of the cerebellum (Fig. 5G,H). Thus, rora2-positive cells in the upper rhombic lip likely corresponds to differentiating Purkinje cells, and grid2 expressing cells in the rhombic lip may include both Purkinje cells and granule cells.

thumbnail image

Figure 6. A–F: Developmental expression pattern of grid2 (A,C,E) and rora2 (B,D,F) in 48 hpf (A,C), 72 hpf (B,D), and 120 hpf (E,F) specimens. Expression of both genes was observed in the telencephalon (red arrows), the optic tectum (black arrows), upper rhombic lip (the cerebellum, blue boxes), and the lower rhombic lip (black boxes) and eyes. Expression of grid2 in the cerebellum is wider than that of rora2 along the anterior–posterior axis. In C, nonspecific staining was observed in surface of the cranial cavity.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Here we have uncovered that the zebrafish genome contains two ROR alpha genes: rora1 and rora2. These two genes exhibited both shared and different expression patterns within the zebrafish nervous system. Coexpression of rora1 and rora2 was observed in the developing eyes and the tectum. Expression of rora1 commenced earlier than that of rora2 and expression of rora2, but not rora1, was detected in the adult brain. Transcripts of rora2 were detected in wider regions of the brain than those of rora1, such as the cerebellum, the telencephalon, and the hindbrain. Blast searches of the genome of Medaka and the puffer fish Tetraodon nigroviridis detected a single locus, which exhibited a high score of sequence identity to ROR alpha. It will be interesting to study the expression pattern of ROR alpha homologue of Medaka fish to gain insights into the evolution of ROR alpha genes.

During preparation of this report, Flores et al. (2007) reported expression patterns of three ROR family genes (rora, rorb, rorc) of zebrafish. One of them (rora) corresponds to rora1, and its spatial and temporal expression pattern described by Flores et al. (2007) was basically the same to that of rora1.

Developmental Anatomy of Zebrafish Brain by rora2 Expression

The development and morphogenesis of the zebrafish cerebellum has remained poorly described to date. Here, we show that rora2 is expressed in Purkinje cells in the adult zebrafish brain. Embryonic expression of ROR alpha in differentiating Purkinje cells of mouse (Nakagawa et al.,1997) suggests that rora2-expressing cells in the upper rhombic lip may correspond to differentiating Purkinje cells of zebrafish. The rora2 expression and the cerebellum generating cells detected by GFP expression in a transgenic line (Koester and Fraser,2001,2006) are similarly localized in 3 dpf larvae. Double labeling of this transgenic line by GFP and rora2 expression will provide further detailed information about morphogenesis of the zebrafish cerebellum.

The fish cerebellum is morphologically divided into three parts, the valvula cerebelli (Va), which is located under the optic tectum, the corpus cerebelli (CCe), and the crista cerebellaris (CC), which receives inputs from the vestibular apparatus (Alonso et al.,1992; Finger,1983; Maeyama and Nakayasu,2000; Meek and Nieuwenhuys,1991). The Va and the CCe are suggested to be related to the flocculus and the vermis of higher vertebrates, respectively (Sarnat and Netsky,1974), but it is controversial as to whether CC is a part of the fish cerebellum or not, and it is also unclear whether it contains Purkinje cells (Brouch et al.,1990; Diaz-Regueria and Anadon,1995; Porteros et al.,1998; Miyamura and Nakayasu,2001). Because expression of rora2 was not observed in CC, our data support the idea that CC does not contain genuine homologues of Purkinje cells.

Detailed histological observations have showed lamination of the upper half of the optic tectum (Meek and Schellart,1978; Nguyen et al.,1999) and researchers have divided this zone into several layers, such as stratum marginale, stratum opticum, SFGS, and stratum griseum centrale (for example, Herrero et al.,1999; Xue et al.,2003). We found that rora2 is coexpressed with grid2 in SFGS (Mikami et al.,2004) in the zebrafish adult brain. Costagli et al. (2002) reported that the reelin gene is expressed in SFGS. We compared expression of reelin and rora2 by dual color in situ hybridization and have found that the layer of reelin-expressing cells is localized between the molecular layer and the rora2-expressing layer (Oomiya and Katsuyama, manuscript in preparation). Similarly laminated gene expression was observed in the superior colliculus of mouse (Dekimoto and Katsuyama, manuscript in preparation). Thus, this laminated gene expression pattern in the dorsal midbrain is evolutionarily conserved.

The cells expressing rora2 in the MON are likely to be the crest cells, which are marked by grid2 expression (Mikami et al.,2004). Because of its cell composition, localization in the brain and function, it has been suggested that MON is homologous to the dorsal cochlear nucleus (DCN) of mammals (Maler and Mugnaini,1993; Montgomery et al.,1995). Expression of ROR alpha gene in the MON of fish (this study) and DCN of mouse (Ino,2004) further supports this homology.

Zebrafish rora Genes Are Not Involved in shh Expression

ROR alpha mutant mice display cerebellar atrophy. ROR alpha is required for proliferation of the granular cells by regulating expression of the mitogenic protein SHH in the developing cerebellum (Dahmane and Ruiz-i-Altaba,1999; Wallace,1999; Wechsler-Reya and Scott,1999; Lewis et al.,2004). Colocalized expression of shh and rora2 was not observed throughout the zebrafish central nervous system from the gastrula stage through 5 dpf. Furthermore, none of the other zebrafish hedgehog family genes (twhh, ihh, dhh) is expressed in the developing cerebellum (Ekker et al.,1995; Avaron et al.,2006; Scholpp et al.,2006). This finding suggests that rora genes do not appear to regulate proliferation of granular cells through regulation of hedgehog expression in the zebrafish cerebellum, and it is likely that this aspect of ROR function has diverged between the teleost and mammalian lineages during vertebrate evolution. The cerebellum of teleosts (and other lower vertebrates) further does not exhibit foliation. SHH signaling is known to regulate the complexity of cerebellar foliation (Corrales et al.,2006). These facts suggest that evolution of regulated SHH expression by ROR alpha may be related to foliation of the cerebellum. Microarray studies have revealed that ROR alpha regulates expression of several genes involved in calcium signaling (Gold et al.,2003), which are essential for morphological maturation of Purkinje cells in the mouse (Ichikawa et al.,2002; Gold et al.,2003; Miyazaki et al.,2004; Boukhtouche et al.,2006). Coexpression of rora2 and grid2 (Wollmuth et al.,2000) in differentiating Purkinje cells of zebrafish hints to the possibility that the regulation of factors involved in calcium signaling is a conserved (and probably an archetypal) function of ROR alpha genes.

Cerebellar-Like Structures and ROR alpha Genes

In the zebrafish brain, rora2 and grid2 exhibited a similar expression pattern. As rora2 encodes a transcription factor, there is a possibility that rora2 may regulate transcription of the grid2 gene. However, two facts indicate that this possibility may be unlikely. First, grid2 expression in the staggerer mutant cerebellum is normal and unaffected (Messer and Kang,2000) and second, the onset of grid2 expression predates rora2 expression in the upper rhombic lip (Fig. 6). However, our observations cannot rule out the possibility that the transcriptional regulation of rora2 may reside downstream of a signaling pathway in which grid2 is involved. As such rora2 is coexpressed with grid2 in multiple sites of the zebrafish brain. It is also possible that rora2 and grid2 are under control of a similar transcriptional regulator.

Strong expression of both rora2 and grid2 was detected in the neurons of the optic tectum, the crest cells of the MON, and Purkinje cells of the cerebellum. They are grouped as “Purkinje-like cells” of “cerebellar-like structures” (Devor,2000; Bell,2002; Mikami et al.,2004). Cerebellar-like structures are designated based on their anatomical similarities to the cerebellum. The principal neurons (Purkinje-like cells) of the cerebellar-like structures receives two major inputs: parallel fibers from the granule cells in their apical dendritic area, and primary afferents of sensory inputs in their basal dendritic area (Devor,2000; Bell,2002). Bell (2002) has suggested that the cerebellar-like structures are not homologues and the similarities of these brain regions could have arisen by convergence. Expression of rora2 and grid2 in these structures suggests a possibility that rora2 and grid2 maybe involved in the generation of the similarities of such cerebellar-like structures. Thus, it will be interesting to investigate whether the function of rora2 and grid2 is essential for development of the characteristics shared by the Purkinje-like cells. Additionally, analysis of the transcriptional regulation of rora2 and grid2 genes may provide crucial insights into the evolution of these cerebellar-like structures.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

cDNA Cloning

Blast searches of the zebrafish genome database (http://www.ensembl.org) using the mouse ROR alpha protein sequence identified two predicted genes, ENSDARG00000001910 and ENSDARG00000019956, and were tentatively designated as rora1 and rora2, respectively. cDNA fragments of these genes were amplified by PCR using the specific primers: GACTAGTCCAAATCTGCACATGCTCCT and GGGATCCTCACTGCCTCAGAATATGTC for rora1, and GCTCTAGATCGAAATAATTCCCTGCAAG and GGAATTCGAATAACTCCTTGTACAAAG for rora2. PCR reactions were carried out using the BD advantage2 PCR system (BD Biosciences Clontech). PCR products were cloned in pGEM easy T/A cloning vectors (Promega Co.) and sequenced. To obtain full-length cDNA sequence of rora2, we carried out cRACE protocol using SMART RACE KIT (Invitrogen). The sequences of the primers used for RACE were indicated by blue arrows in Figure 1A. RACE products were subcloned in pGEM vector and sequenced.

Whole-Mount In Situ Hybridization

The DIG-labeled RNA probes were synthesized using DIG-labeling kit (Roche). Synthesized RNA probes were applied onto ProbeQuant G-50 column (Amersham Biosciences) to remove nucleotides that were not incorporated in the reaction. The embryos and larvae were fixed in 4% paraformaldehyde (PFA) in phosphate buffer saline (PBS), and stored in 100% methanol. After rehydration, the specimens were treated with proteinase K, and post-fixed by 4% PFA in PBS for 1 hr at room temperature (RT). The specimens were prehybridized in hybridization buffer (50% formamide [FA], 5× standard saline citrate [SSC], 0.1% Tween 20, 50 μg/ml heparin, 500 μg/ml yeast RNA pH 6.0) without RNA probe. Embryos were hybridized to the probes at 65°C for longer than 14 hr. Washing to remove excess probe was done in washing buffer 1 (50% FA, 2× SSC, 0.1% Tween 20), at 65°C for 30 min twice, washing buffer 2 (25% FA, 2× SSC, 0.1% Tween 20) at 65°C for 15 min, 2× SSC containing 0.1% Tween 20 at 65°C for 15 min, 0.2× SSC containing 0.1% Tween 20 at 68°C for 30 min twice. Two percent blocking reagent (Roche) in PBS was used for blocking before antibody reaction. The anti-DIG antibody alkaline phosphatase conjugate (Roche) was added to the 2% blocking reagent and the specimens were left in it at 4°C overnight. Free anti-DIG antibodies were removed by washing thoroughly in PBS containing 0.2% Tween 20. After washing by coloring buffer (0.1 M Tris-Cl pH 9.5, 0.1 M NaCl, 50 mM MgCl2, 0.1% Tween 20), coloring reaction was done using BM purple (Roche). The specimens were left in a dark box. The stained specimens were photographed in glycerol.

Section In Situ Hybridization

The adult brains of zebrafish were fixed in cold 4% PFA in 0.1 M phosphate buffer (PB) overnight at 4°C. After dehydration by ethanol series and xylene, the fixed brains were embedded in paraffin. The paraffin blocks were sectioned at 6-μm thickness on a sliding microtome. Sections were air-dried, cleared by xylene, rehydrated in ethanol-down series, and re-fixed in 4% PFA in 0.1 M PB. After 200 mM HCl treatment, the section were washed with TBS (3 min × 2), and acetylated by 0.5% acetic anhydride in 0.1 M Tris-HCl pH 8.0 (10 min). After Proteinase K treatment, the sections were hybridized in hybridization buffer (2× SSC, 10% dextran sulfate, 50 μg/ml fish sperm DNA, 0.02% sodium dodecyl sulfate, 50% formamide, 50 μg/ml tRNA) containing 0.5 μg/ml DIG-labeled RNA probes at 64°C. After hybridization, sections were washed in 2× SSC/50% formamide and 2× SSC at 64°C. Sections were treated with 5 μg/ml RNase A in RNase buffer (0.5 M NaCl, 0.01 M Tris-HCl pH 8.0, 1 mM ethylenediaminetetraacetic acid), and washed in 2× SSC/50% formamide for 30 min at 64°C, in 1× SSC for 15 min at RT, and in TBS for 3 min. After 1 hr incubation in 2% blocking reagent (Roche) in TBS, the sections were treated by alkaline phosphatase-conjugated anti-DIG antibody (Roche) diluted in the blocking buffer for 1 hr at RT. After three washes in TBS for 3 min at RT, the color reaction was carried out using the nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate (NBT/BCIP) system (Roche). After dehydration, the sections were embedded permanently and photographed.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank members of the Division of Anatomy and Developmental Neurobiology, Kobe University Graduate School of Medicine and The Laboratory for Vertebrate Body Axis, CDB RIKEN for their support throughout these studies. We also thank Dr. Raymond Habas of UMDNJ-Robert Wood Johnson School of Medicine and Dr. Yasunori Murakami of Ehime University for critical reading of this manuscript. Y.K. was funded by a grant-in-aid from the Ministry of Education, Culture, Sports, Science, and Technology and Shourei-Kenkyu-Josei of the Hyogo Science and Technology Association.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES