Recapitulation of zebrafish sncga expression pattern and labeling the habenular complex in transgenic zebrafish using green fluorescent protein reporter gene



Human synuclein family consists of α-, β-, and γ-synucleins. Here, we cloned three genes, sncb, sncga and sncgb from zebrafish. They encode β-, γ1-, and γ2-synucleins, respectively. The zSyn-β, zSyn-γ1, and zSyn-γ2 proteins display 69%, 47%, and 50% identity to human β-synuclein and γ-synuclein, respectively. By reverse transcriptase-polymerase chain reaction, we demonstrated that sncb and sncga mRNA were abundant in brain and eye, while sncgb expression was moderate in brain, kidney, ovary and testis. The 1.8-kb 5′-upstream/promoter region of the sncga gene was sufficient to direct green fluorescent protein (GFP) expression in the central nervous system and cranial ganglions. A transgenic line, Tg(sncga:GFP), was generated and its GFP expression is similar to that of endogenous sncga mRNA. Moreover, this line also labels the habenular complex and the domain of GFP expression is larger in the left than in the right habenula. Thus, this line can be used to study sncga gene regulation and for left–right asymmetry study in zebrafish brain. Developmental Dynamics 238:746–754, 2009. © 2009 Wiley-Liss, Inc.


The synuclein family contains three members (α-, β-, and γ-synucleins) in mammals, and they have a highly conserved α-helical lipid-binding motif at the N-terminal region (Iwai et al.,1995). The α-synuclein was first isolated from the electric fish Torpedo californica as a protein which localized in both synapse and nuclear envelope (Maroteaux et al.,1988). Until now, synuclein proteins have only been found in vertebrates. The α- and β-synuclein proteins are found mainly in brain tissues, while the γ-synuclein is found primarily in the peripheral nervous system and the retina (Brenz Verca et al.,2003). Only α-synuclein has been implicated in the pathogenesis of Parkinson's disease (PD), until recently there has been no evidence to suggest the similar role for other synucleins. In contrast, β-synuclein could inhibit α-synuclein aggregation, suggesting a possible role as an anti-parkinsonian factor (Hashimoto et al.,2001; Bertoncini et al.,2007; Tsigelny et al.,2007). The γ-synuclein is also known as BCSG-1 (breast cancer-specific gene 1; Ji et al.,1997) and has been shown to associate with cancer progression and may be used as a tumor marker (Jia et al.,1999; Iwaki et al.,2004; Li et al.,2004; Ahmad et al.,2007).

Zebrafish (Danio rerio) has been demonstrated as an excellent genetic model for studying vertebrate development and diseases (Penberthy et al.,2002). In addition, the availability to establish green fluorescent protein (GFP) transgenic lines under tissue-specific promoters further aid the study of morphogenetic process during development (Kawakami,2004). In this study, we isolated the synuclein gene family from zebrafish. It consists of three genes, which encode β-, γ1-, and γ2-synucleins. Unlike the fugu synuclein gene family, there is no α-synuclein in the zebrafish genome. Based on the deduced amino acid sequences, these zebrafish synuclein-related genes are identical to three genes, sncb, sncga, and sncgb, which were published very recently (Sun and Gitler,2008). The developmental and tissue-specific expression patterns of these genes in zebrafish were determined by reverse transcriptase-polymerase chain reaction (RT-PCR).

Among synuclein-related genes, only the transcriptional regulation of human α-syn has been studied (Rideout et al.,2003; Tan et al.,2003; Chiba-Falek et al.,2005; Clough and Stefanis,2007), whereas the transcriptional regulation of human β-syn and γ-syn genes remains largely unknown. Here, we characterized the putative promoter region of sncga gene by examining the ability of 5′-upstream region to drive GFP expression in zebrafish embryos. In addition, a stable transgenic zebrafish line showing spatial and temporal GFP distribution similar to endogenous sncga mRNA expression pattern was established under the control of the sncga promoter. Moreover, this line also labels the habenular complex including the left- and right-side habenular nuclei, efferent axons and the interpeduncular nucleus (IPN). Thus, this line will be useful for the study of left–right asymmetry in zebrafish brain as well as for the study of the function of sncga gene in neurons.


Cloning of the synuclein-related genes from zebrafish

To identify zebrafish cDNA related to four fugu synuclein genes (Yoshida et al.,2006), the coding regions of fugu α-synuclein (accession no. DQ177904), β-synuclein (accession no. DQ177905), and γ1-synuclein (accession no. DQ177906) were used to blast the GenBank for related expression sequence tag (EST) sequences and six zebrafish EST clones (CN831461, CN021343, CK363296, EH436891, BI706991, and EH452533) were identified. By sequence assembly and 5′-RACE, three zebrafish synuclein-related cDNAs, designated zSyn-β,zSyn-γ1, and zSyn-γ2 were assembled and deposited in GenBank with the following accession numbers: EU368945, zSyn-β; EU368946, zSyn-γ1; and EU368947, zSyn-γ2. These sequences are identical to those of the published sncb, sncga, and sncgb, respectively (Sun and Gitler,2008).

The alignment of the deduced amino acid sequences of zebrafish, fugu and human synuclein-related proteins is shown in Figure 1A. The zSyn-β protein containing 127 amino acid residues is homologous to fugu and human β-synuclein with 77% and 69% amino acid identity, respectively. The zSyn-γ1 and zSyn-γ2 cDNA encoded proteins of 114 and 111 amino acid residues. The zSyn-γ1 protein exhibited a higher amino acid identity (72%) to fugu γ1-synuclein than to fugu γ2-synuclein (57%) and human γ-synuclein (47%). However, zSyn-γ2 protein displayed 63%, 66% and 50% amino acid identity to fugu γ1-synuclein, γ2-synuclein and human γ-synuclein, respectively. All 3 zebrafish synuclein-related proteins contained 5–7 imperfect repeats of 11 amino acid residues in the N-terminal region. Both zSyn-γ1 and zSyn-γ2 have 7 continuous repeats (amino acids 9-89) with a 4 amino acid linker between repeats IV and V. In contrast, zSyn-β has only 5 repeats (amino acids 9-63) and lacks an 11 amino acids stretch in the hydrophobic region (GVTAVAQKTVE in human α-synuclein). In all γ-synuclein proteins, a stretch of 18 amino acid residues in human β-synuclein or 15 amino acids in human α-synuclein in the C-terminal region are missing (Fig. 1A). Apparently, no sequence homologous to fugu and human α-synuclein can be found in zebrafish. Further phylogenetic tree analysis clusters α-, β-, and γ-synuclein from human, mouse, fugu, and zebrafish into three groups, while fugu α-synuclein is away from human and mouse α-synuclein (Fig. 1B). The zSyn-β protein is clustered with members in the β-synuclein group. Zebrafish γ1-synuclein and γ2-synuclein are clustered with fugu γ1- and γ2-synuclein in the same group.

Figure 1.

Alignment of amino acid sequences and phylogenetic analysis of three zebrafish synucleins with those from other species. A: Multiple alignment of synucleins from fish and human was performed by using CLUSTAL X. The amino acid sequence of three zebrafish synucleins (zSyn-β: EU368945; zSyn-γ1: EU368946; zSyn-γ2: EU368947) were compared with those from fugu (FuSyn-α: NP_001029020; FuSyn-β: NP_001029018; FuSyn-γ1: NP_001029017; FuSyn-γ2: NP_001029019), and human (hSyn-α: NP_000336; hSyn-β: NP_001001502; hSyn-γ: NP_003078). The identical amino acids are highlighted and I–VII imperfect repeats of 11 amino acids are also labeled. Arrows indicate mutations (A30P, A53T and E46K) and C-terminal serine 129 which are found in human α-synuclein implicated in Parkinson's disease. B: Phylogenetic tree analysis of synuclein proteins. Bootstrap values, calculated from 1,000 replicates, are indicated at the nodes. The accession numbers of mouse synuclein proteins are as follows: mSyn-α: NP_033247; mSyn-β: NP_291088; mSyn-γ; NP_035560.

Expression Profiles of Zebrafish synuclein-Related Genes in Adult Tissues and Embryos From Different Developmental Stages

The expression levels of zebrafish synuclein-related transcripts in adult tissues and in embryos from different developmental stages were examined by RT-PCR analysis. As shown in Figure 2A, high levels of the sncb mRNA were detected in the brain and eyes, while less abundantly in the ovary, testis, intestine, and kidney. Similarly, high levels of sncga transcription were detected in the brain and eye, while low level expression was found in the testis and gill. For sncgb, moderate mRNA expression level was detected in the testis, kidney and brain.

Figure 2.

Expression profiles of zebrafish synuclein-related transcripts in various adult tissues and at different developmental stages by reverse transcriptase-polymerase chain reaction (RT-PCR). A,B: RT-PCR was performed with gene-specific primers. β-actin bands were used to normalize the amount of cDNA prepared from different tissues (A) and from different developmental stages (B). The developmental expression profile of each gene was examined in embryos from 4 hours postfertilization (hpf) to 5 dpf.

During embryogenesis, both sncb and sncgb transcripts were detected at 4 hours post fertilization (hpf). The sncb expression level was kept constant afterward, while the sncgb expression level was decreased after 4 days postfertilization (dpf). Moderate sncga expression level was not detected until 1 dpf (Fig. 2B).

Characterization of sncga Promoter by Transient and Stable Transgenic Approaches

The transcriptional regulation of human α-syn has been widely studied (Rideout et al.,2003; Tan et al.,2003; Chiba-Falek et al.,2005; Clough and Stefanis,2007). However, the transcriptional regulation of human β-syn and γ-syn genes remains largely unknown. To identify the promoter region of sncga gene that can recapitulate its central nervous system (CNS) -specific expression, promoter fragments containing 1.8 to 4 kb of the 5′-upstream from the translational start site were amplified by PCR from the bacterial artificial chromosome (BAC) clone (CH211-103O20). These PCR fragments were cloned into a Tol2 expression vector containing the GFP reporter transgene (Kawakami et al.,2004) and co-injected with Tol2 transposase mRNA into 1-cell zygotes. The transcriptional activity of each reporter construct was evaluated by analyzing GFP expression at 2 dpf by fluorescence microscopy. The 3- and 4-kb promoter fragments drive only weak expression of GFP in the brain (data not shown). In contrast, the 1.8-kb construct yields robust GFP expression in the CNS and cranial ganglions, which is consistent with the spatial distribution pattern of sncga mRNA during early embryonic development (Sun and Gitler,2008). In DNA-injected embryos at 48 hpf, GFP expression was clearly shown in the hindbrain, trigeminal ganglion, posterior lateral line ganglion, spinal cord, retina, and lens (data not shown).

Embryos that exhibited mosaic GFP expression were raised to adulthood and out-crossed to AB wild-type strain. GFP-positive F1 embryos were identified, raised to adulthood, and used to generate several transgenic stable lines. Among three transgenic fish lines generated, only the line possessing the highest expression level was used for further analysis and designated Tg(sncga:GFP). The transgene was transmitted according to Mendel inheritance, because 50% F3 fish showing GFP expression was obtained when crossing Tg(sncga:GFP) F2 fish with wild-type. GFP distribution pattern in a live Tg(sncga:GFP) F3 embryo at 3 dpf was shown (Fig. 3A). In addition to the expression in the spinal cord, GFP expression was also observed in the habenula, hindbrain, midbrain, eyes, trigeminal ganglion, vagal ganglion, and posterior lateral line ganglion, which was consistent with endogenous sncga mRNA distribution pattern from whole-mount in situ hybridization (Fig. 3D).

Figure 3.

The expression pattern of green fluorescent protein (GFP) in transgenic zebrafish Tg(sncga:GFP) line. Microinjection of the expression construct into zebrafish embryos at the one- to two-cell stage and generation of transgenic GFP line by means of Tol2-mediated transgenesis were described in the text. A: Images from Tg(sncga:GFP) transgenic line at 3 days postfertilization (dpf). Images of bright field and fluorescence are merged and shown in panel b, while fluorescence images are shown in panel a. sc, spinal cord; nc, notochord. B: Confocal image analysis. Zebrafish larva of transgenic Tg(sncga:GFP) line at 3 dpf were collected and fixed with 4% paraformaldehyde. After treatment of FocusClear, embryos were subjected to high-resolution confocal image analysis. All embryos are shown with the anterior to the left. Embryos in a, c, d, e are shown in lateral view, while embryos in b are shown in dorsal view. White arrowheads indicated habenulo-interpeduncular projection (panel b) and posterior lateral line projection (panels d, e). C: Whole-mount immunostaining of Tg(sncga:GFP) embryos by polyclonal antibody against GFP. Embryos in a and b are shown in lateral view, while embryos in c are shown in dorsal view. The habenulo-interpeduncular projection was labeled by black arrowheads. D: Expression patterns of sncga mRNA in zebrafish embryo at 3 dpf. gcl, retinal ganglion cell layer; gP, posterior lateral line ganglion; gV, trigeminal ganglion; gVIII, statoacoustic ganglion; gX, vagal ganglion; Ha, habenula; hb, hindbrain; mb, midbrain; IN, interneuron; inl, inner nuclear layer; IPN, interpeduncular nucleus; l, lens; LHa, left-side habenula; MHB, midbrain–hindbrain boundary; RB, Rohon-Beard neuron; RHa, right-side habenula.

The Expression Pattern of Transgenic Zebrafish Tg(sncga:GFP) Recapitulates the sncga Expression Pattern Including the Habenular Complex

A region has been called epiphysis previously (Sun and Gitler,2008), but we labeled as habenula (Ha) instead as shown in Figure 3A,D. To clarify this issue, high-resolution confocal image analysis in combination with FocusClear technology (Liu and Chiang,2003) was performed to investigate the accurate structure of this region. From the dorsal view (Fig. 3B, panel b), the left- and right-side habenular nuclei, efferent axons and the interpeduncular nucleus (IPN) are labeled with GFP. Furthermore, immunohistochemical staining by using anti-GFP antibody also revealed the similar structure (Fig. 3C).

We further performed double fluorescence in situ hybridization to investigate whether the endogenous sncga gene is expressed in the habenula. As shown in Figure 4A, gfp-positive cells (green) totally overlap with cells expressing sncga (a′ and a″; Gamse et al.,2003), and leftover (lov; b′ and b″), but partially overlap with cells expressing right on (ron; c′ and c″; Gamse et al.,2003). The domain of gfp expression is larger in the left habenula than in the right. In addition, most gfp-positive cells (green) are localized within cells expressing cerebellum postnatal development associated protein 2 (cpd2) (d′, and d″; Gamse et al.,2003) and cells expressing f-spondin2 (e′ qne e″;Gamse et al.,2003). However, some cpd2-positive and f-spondin2-positive cells are gfp-negative. In addition to the habenula, cells expressing sncga and gfp-positive cells are well colocalized to many regions, such as eyes, midbrain, hindbrain, and posterior lateral line ganglion (Fig. 4B). They also totally overlap in the spinal cord (data not shown). Altogether, the endogenous sncga gene is expressed in the habenula and the expression of gfp in transgenic Tg(sncga:GFP) line recapitulates the endogenous expression pattern of sncga mRNA.

Figure 4.

Double fluorescence in situ hybridizations of habenula-expressing markers, gfp and sncga in Tg(sncga:GFP) fish. A: Expression domains of sncga (a′), lov (b′), ron (c′), cpd2 (d′) and f-spondin2 (e′) along with gfp (a–e) in habenula in Tg(sncga:GFP) fish at 3 days postfertilization (dpf). a″–e″, merged image of a and a′, b and b′, c and c′; d and d′, and e and e′, respectively. B: Double fluorescence in situ hybridization of sncga and gfp in an Tg(sncga:GFP) fish at 3 dpf. Panel a shows the staining of gfp mRNA, whereas panel b shows the staining of sncga mRNA. Panel c shows the merged image, in which gfp and sncga staining completely overlapped. gcl, retinal ganglion cell layer; GFP, green fluorescent protein; gP, posterior lateral line ganglion; Ha, habenula; hb, hindbrain; l, lens; mb, midbrain.

In this study, three genes encoding β-, γ1-, and γ2-synucleins were isolated from zebrafish and they are identical to three genes, sncb, sncga, and sncgb, which have been published very recently (Sun and Gitler,2008). Unlike the fugu synuclein gene family, there is no α-synuclein in the zebrafish genome. Currently, four synuclein genes (α-, β-, γ1-, and γ2-synuclein) are identified in fugu (Yoshida et al.,2006), while three genes (α-, β-, and γ-synuclein) are detected in birds and mammals (Iwai et al.,1995). In mammals, α-synuclein forms the major fibrillar components associated with the pathological lesions of neurodegenerative diseases (Spillantini et al.,1997; Trojanowski and Lee,1998; Shults,2006). The fugu α-, γ1-, and γ2-synucleins could assemble into filaments in vitro with shorter lag times and faster rates than those of human α-Synuclein. In contrast, fugu and human β-synuclein failed to assemble in bulk (Yoshida et al.,2006). A similar case has been reported in zebrafish nogo family that nogo-A is not present in the zebrafish genome (Diekmann et al.,2005). In mammals, the nogo family consists of nogo-A, nogo-B, and nogo-C and only Nogo-A can interfere with axonal regrowth after spinal cord injury (GrandPre et al.,2002). The absence of nogo-A in the zebrafish genome attributes to one of the reasons that zebrafish axons can be regenerated in the fish CNS after damage, unlike mammalian axons that are inhibited by Nogo-A (Diekmann et al.,2005). Currently, the absence and the reason why zebrafish lacks α-synuclein needs further investigation.

In mammals, all three synuclein transcripts are present in the visual system, mainly in the retina and optic nerve (Surguchov et al.,2001). Decreased γ-synuclein mRNA levels were indicated in the optic nerves of rats with episcleral vein cauterization which was used as an animal model of glaucoma (Surgucheva et al.,2002). Similarly, both sncb and sncga mRNAs were detected in the adult zebrafish eyes (Fig. 2) as well as in retinal ganglion cell layer of zebrafish embryos from a transgenic GFP line, Tg(sncga:GFP; Fig. 3). Thus, this line can be used as another animal model for similar study.

Of interest, only the 1.8-kb 5′-upstream/promoter region can drive robust transient GFP expression in the CNS and cranial ganglions, while the 3- and 4-kb promoter fragments drive weak expression of GFP in the brain (data not shown). These data suggest that some repressing elements may be present in the region between 4-kb and 1.8-kb 5′-upstream region. A transgenic GFP line, Tg(sncga:GFP) was generated and the GFP signal was clearly observed in the habenula, hindbrain, midbrain, eyes, trigeminal ganglion, vagal ganglion, posterior lateral line ganglion, and spinal cord (Fig. 3), which was consistent with the endogenous sncga mRNA distribution pattern during early embryonic development assayed by whole-mount in situ hybridization (Fig. 3D). Moreover, the Tg(sncga:GFP) line also labels the habenular complex including the bilateral habenular nuclei, efferent axons and IPN (Fig. 3B).

In addition to this Tg(sncga:GFP) line, other transgenic zebrafish lines also label the habenular complex with GFP, such as et27, et25.2 and et206 (Scott et al.,2007), the ET16 enhancer trap line (Parinov et al.,2004) and a Tg(brn3a-hsp70:GFP) line that expresses GFP under the control of enhancer elements of brn3a gene (Aizawa et al.,2005). All these lines show either a broader GFP expression in the right habenula or an almost symmetric GFP expression in both habenulae. However, the domain of GFP expression in our transgenic line is larger in the left than in the right habenula. Such a different GFP expression pattern in habenular nuclei makes our line potentially useful for the study of left–right asymmetry in the zebrafish brain.

In zebrafish, the epithalamus or dorsal diencephalon, consists of the pineal complex and the habenular complex. Two photoreceptive nuclei, the pineal and parapineal, are present in the pineal complex, while left- and right-side habenular nuclei, efferent axons and the IPN constitute the habenular complex. The pineal gland lies in the middle of the bilateral habenular nuclei and its left-sided parapineal determines the left–right identity of adjacent habenular nuclei (Gamse et al.,2003). The bilateral habenular nuclei project efferent axons to IPN of the ventral midbrain. Currently, the habenular system has becoming an important model for studying brain asymmetries and their development (Concha et al.,2003; Gamse et al.,2003,2005; Halpern et al.,2003; Aizawa et al.,2005,2007; Kuan et al.,2007).

In conclusion, we have cloned and characterized sncb, sncga, and sncgb genes from zebrafish. The 1.8-kb sncga promoter region was identified and characterized and a Tg(sncga:GFP) stable transgenic line was established by Tol2-mediated transgenesis. We found that the distribution of GFP signal in this line recapitulates the endogenous expression pattern of sncga mRNA including the habenular complex. Thus, the 1.8-kb sncga promoter fragment contains all the cis-regulatory elements sufficient for neuron- and habenular-specific expression in vivo. In addition, this line also can be applied for the study of cell-type specific regulation of sncga gene as well as for the study of left–right asymmetry in the zebrafish brain.


Zebrafish Care

Zebrafish embryos were raised at 28.5°C and different developmental stages were determined based on the criteria described in the Zebrafish Book (Westerfield,2000).

Total RNA Isolation and RT-PCR Analysis of Synuclein mRNA

Total RNA was isolated from different developmental stages and various tissues of adult zebrafish, using the RNAzol reagent (Tel-Test, Friendswood, TX) according to the instructions of the manufacturer. After treatment with RQ1 RNase-Free DNaseI (Promega, Madison, WI), 50–100 μg of total RNA was subjected to the first strand cDNA synthesis. PCR amplifications were performed to amplify the 3′-untranslated region (UTR) of each gene with the following primer set: sncb primers (forward primer: 5′-ACA TTC CAA GCC CAA ATC CAG-3′; reverse primer: 5′-ATA TGC CCT TTG GAG GAT TGG-3′), sncga primers (forward primer: 5′-TCT CTC ATG GTG GTA TGG AAG GAG GAG-3′; reverse primer: 5′-GTG TGA GAG GAA GTA CGT GTT GGA GTG), sncgb primers (forward primer: 5′-CCA TGA TGC TCT GCT TCT TGC CTG AGA-3′; reverse primer: 5′-ACC ACA CTT TAC AGA CAT CAG TTG CTC-3′). β-actin was amplified by 5′-GTG CTA GAC TCT GGT GAT GGT GTG-3′ and 5′-GGT GAT GAC CTG ACC GTC AGG AAG-3′ and used as an internal control.

Cloning of the Full-Length cDNAs Encoding Zebrafish synucleins

Full-length cDNA of different zebrafish synucleins were obtained by PCR amplification using gene-specific primers: sncb (forward primer: 5′-GAA GCA CAA GGA CGA AGA AGG TGC-3′; reverse primer: 5′-TGG ATT TGG GCT TGG AAT GTG ACG-3′), sncga (forward primer: 5′-CGA CAC ACT CAG CGT CCA GGA TGG ATG-3′; reverse primer: 5′-TTG GAG CAG AAA CGT AGA GGT GGA GAG-3′), and sncgb (forward primer: 5′-GCA CTG AGC TCC AGC ATG GAT GCA CTG-3′; reverse primer: 5′-TCT CAG GCA AGA AGC AGA GCA TCA TGG-3′). Respective RT-PCR products of different synucleins were cloned into pGEM-T easy vector (Promega) and sequenced by ABI 3730XL DNA sequencer (Applied biosystems, Foster City, CA).

Isolation of Zebrafish 1.8-kb Promoter of sncga Gene

The BAC CH211-103O20 (RZPD, Berlin, Germany) was used as a template to amplify 1.8-, 3-, and 4-kb 5′ upstream fragment of the zebrafish sncga gene using respective PCR primer pair for 1.8-kb 5′ upstream region (forward primer: 5′-CGC ACA CAG GAA CAG CTG ATC ATC TGA TGC-3′; reverse primer: 5′-AGT GTG TCG ATG GAG GGT CTG CAG GAT CTG-3′), 3-kb 5′ upstream region (forward primer: 5′-ATT ACA AAG TCC TGG CTG TGG AGG CAG AGG; reverse primer: 5′-AGT GTG TCG ATG GAG GGT CTG CAG GAT CTG-3′), and for 4-kb 5′ upstream region (forward primer: 5′-GGA AGG GAG ACA CTG TCA TAA CCA GAC TCG-3′; reverse primer: 5′-AGT GTG TCG ATG GAG GGT CTG CAG GAT CTG-3′). These PCR products were then cloned into pGEM-T easy vector (Promega) followed by subcloning into the SalI/ApaI sites of pT2KXIGΔin vector (Kawakami et al.,2004).

Microinjection of Zebrafish Embryos

To obtain stable transgenic fish, one-cell stage embryos were co-injected with 10 ng/μl pSncga-1.4k-gfp plasmid, 5 ng/μl capped Tol2 transposase mRNA and 0.1% phenol red as previously described (Kawakami,2005). The pCS-TP plasmid (Kawakami,2004) encoding the transposase was linearized with NotI and used as the template for in vitro transcription with mMessage mMachine (Ambion, Foster City, CA), according to the manufacturer's instruction.

To generate germline transgenic zebrafish, the injected embryos were raised to adulthood and the 3-month-old fish (F0) were crossed to wild-type fish. F1 embryos were examined under a fluorescence microscope for identification of germline-transmitted F0 founders. Positive F1 embryos were raised to adulthood and were then screened in the same way to estimate copy numbers of the transgene and to establish stable transgenic lines. Three independent lines (F2 and more advanced generations) were obtained, two of which contained a functional single copy of sncga-1.4k-gfp.

Whole-Mount In Situ Hybridization

To synthesize digoxigenin-labeled (Roche, Penzberg, Germany) antisense RNA probes, pGEM-T easy-sncga 3′-UTR was linearized with PstI and transcribed with T7 RNA polymerase. The probes for habenula (pGEM-T easy-lov, pGEM-T easy-ron, pGEM-T easy-cpd2, and pGEM-T easy-f-spondin2) have been described previously (Gamse et al.,2003), which were linearized with NcoI and transcribed with SP6 RNA polymerase. For fluorescein-labeled (Roche) antisense probe, pcDNA3-CMV-gfp was linearized with HindIII and transcribed with SP6 RNA polymerase. Whole-mount in situ hybridization were performed following previously described protocol (Thisse et al.,1993; Chu et al.,2007). For double fluorescence in situ hybridization, the digoxigenin-labeled probes were detected by anti-digoxigenin POD (Roche) with 1:500 dilution, and signals were amplified through tyramide-FITC kit (Perkin Elmer, Boston, MA). After the first staining step, the tyramide working solution was washed away and the embryos were incubated 30 min with 1% H2O2, to inactivate the peroxidase activity of the first antibody. The embryos were then blocked for 1 hour and the fluorescein-labeled probes were detected by anti-fluorescein POD (Roche) with 1:500 dilution, and signals were amplified through tyramide-Cy3 kit (Perkin Elmer).

Whole-Mount Immunostaining

For whole-mount immunostaining, zebrafish larvae from 3 dpf were fixed with 4% paraformaldehyde in PBS for 24 hr, then washed several times in PBS. The following procedures were according to previously described protocol (Elsalini and Rohr,2003) with some modifications. Embryos were incubated in 10% H2O2 in methanol for 10 min to bleach and block endogenous peroxidases, then washed in PBST (PBS+0.1% Triton X-100). Embryos were digested with 50 μg/ml proteinase K in PBST for 30 min at room temperature, then post-fixed in 4% paraformaldehyme for 20 min and washed several times in PBST. Embryos were blocked in PBST containing 2% goat serum and 2% bovine serum albumin (BSA) for 1 hr at room temperature. Polyclonal primary antibody against GFP (Cell Signaling Technology, Inc., Danvers, MA) was used in 1:1,000 dilution in blocking buffer at 4°C overnight. After washing in PBST, biotinylated secondary antibody was used in 1:500 dilution at 4°C overnight. After washing in PBST, embryos were incubated with ABC kit (Vector Lab., Peterborough, England) for 2 hr according to the manufacturer. Embryos were washed again in PBST and then incubated in DAB (0.2 mg/ml PBS) for 20 min. For color detection, added 1 μl 0.3% aqueous H2O2 solution and washed in PBST to stop the reaction. Embryos were post-fixed in 4% paraformaldehyde for 20 min, followed by washing several times in PBST and gradually transferred to 70% glycerol for further analysis by microscope.

High Resolution Confocal Image

Zebrafish embryos were fixed with 4% paraformaldehyde in PBS at 4°C for 24 hr, followed by washing for 3 times in PBS (5 min each time). Embryos were cleared according to Chiang's publication previously (Liu and Chiang,2003). Clear whole-mount embryo by incubation in FocusClear in a small chamber with a spacer ring for 5 min, then mounted with MountClear at room temperature for 30 min. Images were captured with a Zeiss LSM 510 confocal microscope (Carl Zeiss, Jena).


We thank Dr. Y.S. Kuan for helpful discussion and technical suggestions.