The zebrafish (Danio rerio) motor neurons are accessible for observation from the initial stages of development and are widely used as an experimental model system for the study of the molecular mechanisms involved in axonal pathfinding and neuromuscular junction formation (Eisen et al.,1986; Flanagan-Steet et al.,2005; Jing et al.,2009), and for the analysis of neurodegenerative processes (Boon et al.,2009; Sager et al.). Each somitic hemisegment in the zebrafish embryo is first innervated by three primary motor neurons named rostral (RoP), middle (MiP), and caudal (CaP) (Myers et al.,1986; Westerfield et al.,1986), and by a variable motor neuron (VaP) that develops in half of the somitic segments (Eisen et al.,1990). Cell division of motor neuron precursors ends at approximately 16 h post fertilization (hpf) (Myers et al.,1986). The CaP motor axon exits the spinal cord at ∼18–20 hpf, and a few hours later it is followed by the RoP and MiP axons. Primary motor axons migrate ventrally and establish the first neuromuscular junction with the muscle pioneer cells located at the horizontal myoseptum before choosing cell-type specific pathways to innervate different myotomal territories (Eisen et al.,1986, 1990; Melancon et al.,1997; Myers et al.,1986; Westerfield et al.,1986). In addition to the primary motor neurons, each somitic hemisegment contains approximately 25 secondary motor neurons, which start to exit the spinal cord at 26 hpf, and follow the same axonal trajectories pioneered by the primary motor axons (Myers et al.,1986).
The mnx1 transcription factor is selectively expressed in postmitotic spinal cord motor neurons (Arber et al.,1999; Tanabe et al.,1998). For this reason, the regulatory elements of the mnx1 gene (also known as hb9, hlxb9, scra1, and hoxhb9) can be used as a tool for the temporal and spatial control of protein expression in spinal motor neurons (Flanagan-Steet et al.,2005). However, activation of gene expression in various spinal cord neuron types observed with a 3-kb DNA fragment located upstream of the mnx1 protein coding region (Flanagan-Steet et al.,2005), precludes a widespread use of this promoter region for the manipulation of protein expression in single motor axons beyond the first 24 h of development. For these reasons, we investigated whether a phylogenetically conserved enhancer sequence of 125 base pairs (bp) identified within the mouse mnx1 promoter region (Nakano et al.,2005) could be used to direct gene expression in zebrafish spinal motor neurons.
Sequencing of the 3-kb mnx1 zebrafish promoter (a gift from Meyer) identified a 125-bp fragment starting at position −1946 upstream of the mnx1 protein coding region with 80% homology to the mouse sequence (Nakano et al.,2005). To determine whether the 125-bp sequence directs gene expression in zebrafish spinal motor neurons, three copies of the fragment were cloned in tandem (mnx1-3x125bp) into the p5E entry vector of the Tol2 Kit (Kwan et al.,2007)(see Table 1 for expression vectors used in this study). The mnx1-3x125bp enhancer was placed upstream of prenylated (pren) enhanced green fluorescent protein (EGFP) into the Tol2 destination vector (pDestTol2pA2) using LR recombination, and the new plasmid (mnx1-3x125bp:prenEGFP) injected into wild type zebrafish embryos. At 24 hpf, from 187 normally developed embryos, 173 (93%) of the embryos contained EGFP expressing cells. Embryos were fixed, immunostained with SV2 and znp1 antibodies to label axons and nerve terminals (Brusés, in press; Panzer et al.,2005), and observed by confocal microscopy to determine the pattern of EGFP expression (see Fig 1). At 24 hpf, EGFP was detected in a few primary motor neurons with an average of 11.4 neurons per embryo (Table 2). Examples of somitic hemisegments with a single labeled primary motor neuron are shown in Figure 1. By 48 hpf, the intensity of EGFP labeling was increased and secondary motor neurons were also observed. However, individual primary motor axons and their entire axonal arborization could be observed in some somitic hemisegments (Fig 1, Table 2), indicating that mnx1-3x125bp was sufficient to direct gene expression specifically in spinal motor neurons. EGFP expression in spinal cord interneurons was rarely observed.
Table 1. Components of the Expression Vectors Used in the Study
Expression vectors were generated by combining the elements in p5E, pME, and p3E plasmids into a pDestTol2pA2 vector via LR recombination.
Table 2. Number of Primary Motor Neurons Labeled
Labeled primary motor neurons per embryo
Hemisegments with a single labeled primary motor neurons
Values are expressed as mean ± SEM; n, number of embryos.
11.4 ± 1.6
6.5 ± 1.2
3.1 ± 0.5
1.5 ± 0.3
n = 22
n = 19
mnx1-3x125bp:Gal4-VP16 + 14xUAS:prenEGFP
6.7 ± 1.4
7.9 ± 0.5
3.4 ± 0.7
1.0 ± 0.3
n = 13
n = 26
Tg(mnx1-3x125bp:Gal4-VP16) + 14xUAS:prenEGFP
9.5 ± 1.2
10.1 ± 1.7
3.5 ± 0.7
4.1 ± 0.6
1.5 ± 0.4
0.6 ± 0.3
n = 16
n = 10
However, ectopic EGFP expression was detected in the head of 24 hpf embryos injected with mnx1-3x125bp:prenEGFP including the telencephalon, optic tectum, and eyes (see Fig 1). Ectopic EGFP expression levels were lower than the levels detected in motor neurons and became undetectable at 48 hpf (see Fig 1). Labeled cells outside the spinal cord could not be associated with any particular cell type. By 120 hpf EGFP expression outside the spinal cord was restricted to a small group of cells within the epithalamus (see Fig 1). Ectopic protein expression in the head has also been observed in transgenic mouse embryos expressing lacZ under a similar 3x-125-bp mnx1 enhancer (Nakano et al.,2005), which was abolished when a 313-bp fragment of the mnx1 promoter region (−9040 to −8727) was combined with the 125-bp enhancer sequence, suggesting that the 313-bp sequence suppresses ectopic gene activation (Nakano et al.,2005). Sequence alignment analysis of the zebrafish 3-kb mnx1 promoter identified a 356-bp sequence located 301 bp upstream of the mnx1 125-bp enhancer with 72% homology to the 313-bp sequence reported in mouse. To test whether the 356-bp fragment affects gene expression in zebrafish, a 782-bp DNA fragment (−2603 to −1821) containing both, the 356-bp conserved region and the 125-bp enhancer was cloned upstream of prenEGFP (mnx1-782bp:prenEGFP). Wild type embryos injected with the mnx1-782bp:prenEGFP vector showed EGFP expression in the spinal cord at 24 hpf while ectopic expression in the head was not detected. However, by 72 hpf EGFP expression in motor neurons was substantially decreased and by 120 hpf EGFP expression was undetectable (data not shown). To determine whether the suppression of ectopic EGFP expression was due to the addition of the 356-bp fragment or due to the use of only one copy of the mnx1 125-bp enhancer, the 356-bp DNA fragment was cloned upstream of the mnx1-3x125bp:prenEGFP sequence (mnx1-356bp-3x125bp: prenEGFP) and injected into wild type embryos. Embryos injected with this vector showed ectopic EGFP expression in the head as previously observed with the mnx1-3x125bp:prenEGFP vector, indicating that the transient ectopic EGFP expression outside the spinal cord was likely due to the use of three copies of the mnx1 125-bp.
We then examined whether Gal4-VP16 expression under the mnx1-3x125bp enhancer could be used to drive prenEGFP expression. The Gal4-VP16 sequence was cloned downstream of the mnx1-3x125bp enhancer (mnx1-3x125bp:Gal4-VP16), and the plasmid was co-injected with a vector expressing prenEGFP under a 14xUAS element (14xUAS:prenEGFP) into wild type embryos. An average of 6.7 and 7.9 EGFP labeled primary motor neurons per embryo were observed at 24 hpf and 48 hpf, respectively (Table 2). Figure 2 shows representative confocal images of different types of primary motor neurons. Although secondary motor neurons were also labeled at 48 hpf, single labeled primary motor neurons within a somitic hemisegment could be identified (Table 2). To facilitate the manipulation of protein expression in spinal motor neurons, stable transgenic fish expressing Gal4-VP16 under the mnx1-3x125bp enhancer [Tg(mnx1-3x125bp:Gal4-VP16)] were generated. F1 Tg(mnx1-3x125bp:Gal4-VP16) embryos were injected with the 14xUAS: prenEGFP vector, fixed at 24 hpf and 120 hpf, immunostained with SV2 and znp1 antibodies, and observed under confocal microscopy (see Fig 3). Approximately 10% of the injected embryos contained EGFP expressing motor neurons at 24 hpf with an average of 9.5 primary motor neurons per embryo (Table 2). Primary motor neurons expressing EGFP were also detected in 120 hpf larvae (see Fig 3), indicating that mnx1-3x125 bp enhancer drives EGFP expression in spinal motor neurons beyond embryonic development.
To determine whether the mnx1-3x125bp enhancer directs EGFP expression at sufficient levels to observe motor neurons morphology over time in live embryos and larvae, wild type embryos were injected with the mnx1-3x125bp:prenEGFP vector at the 1-cell stage. At 28 hpf an embryo was embedded in 1% low melting point agarose and observed with a Zeiss LSM510 confocal microscope using an Achroplan 40x/NA0.8 water immersion lens. A z-stack of images was taking at 1 μm intervals every 20 min and each stack was then projected to single plane. Figure 4 shows a CaP motor neuron and magnified images of the growth cone showing extension and retraction of processes over time. Images from the same CaP motor neuron at 48 hpf and 72 hpf show sufficient prenEGFP expression to label the expansion of the axonal arborization within the ventral muscle (see Fig 4). Finally, to examine whether stable transgenic expression of Gal4-VP16 under the mnx1-3x125bp enhancer drives protein expression at sufficient levels for in vivo imaging, F1 Tg(mnx1-3x125bp:Gal4-VP16) embryos were injected at the 1-cell stage with the 14xUAS:prenEGFP vector and observed at 120 hpf. Figure 4 (panel j) shows a projection to a single plane of a z-stack from a CaP motor neuron in which the axonal arborization within the ventral muscle and the dendritic extensions from the cell soma within the spinal cord can be observed.
In summary, the mnx1-3x125bp enhancer directs cell-type specific gene expression that can be used for the analysis of spinal motor neurons. The activity of this promoter beyond embryonic development will facilitate the analysis of gene products involved in neurodegenerative disorders with onsets in postembryonic life. As the mnx1-3x125bp enhancer is only 395-bp long, it can be cloned upstream of a protein coding sequence or it can be combined with other methods using Cre/lox-p recombinase and steroid receptors to provide further temporal control of gene activation (Collins et al., 2010; Matsuda and Cepko,2007).
bovine serum albumin
Developmental Studies Hybridoma Bank
enhanced green fluorescent protein
phosphate buffer saline
MATERIALS AND METHODS
Fish Maintenance and Husbandry
Zebrafish were obtained from the Zebrafish International Resource Center (ZIRC, Eugene, OR), raised, maintained as described in The Zebrafish Book (Westerfield,2000), and staged as previously reported (Kimmel et al.,1995). The zebrafish AB line was used as wild type strain (strain AB-3, stock #5560). The use and manipulation of animals used in this study has been approved by the Institutional Animal Care and Use Committee from the University of Kansas School of Medicine.
This study used att site-specific recombination-based cloning system (MultiSite Gateway technology, Invitrogen, Carlsbad, CA) from lambda phage to build plasmids. The Tol2 Kit from C-B Chien Laboratory was the main source of plasmids for recombination and backbones for novel entry vector development (Hartley et al.,2000; Kwan et al.,2007). Three kinds of entry vectors were used for LR recombination: 5′ entry vector (p5E), middle entry vector (pME), and 3′ entry vector (p3E). The Tol2 Kit plasmid pDestTol2pA2 (Kwan et al.,2007) was used as a destination vector in all LR recombinations using LR clonase II (Invitrogen), and served as a backbone for the expression plasmids carrying an SV40 polyadenylation (pA) signal. The DNA fragment inserted in this plasmid is flanked by Tol2 transposable elements that facilitate DNA insertion (Suster et al.,2009). The Tol2 is a transposable element identified in medaka that has been successfully used to integrate large pieces of DNA into the zebrafish genome by promoting efficient DNA recombination (Kawakami et al.,1998). All sequences were verified by DNA sequencing.
The following entry vectors were generated:
p5E mnx1-3x125bp enhancer: The mnx1 125-bp enhancer located −1946 to −1821 from the start codon was amplified by PCR using 3-kb mnx1 promoter as template (Flanagan-Steet et al.,2005) and three different sets of primers containing unique restriction sites (underlined): primer set 1: forward XhoI, reverse SphI, and HindIII (5′-TCCTCGAGAGTGGTTAGCTGATGA and 5′-ACAAGCTTTTGCATGCCCTCTTATAAGCCTCTTTAA); primer set 2: forward SphI, reverse AvrII, and HindIII (5′-TTGCATGCCGAGAGTGGTTAGCTGA and 5′-ACAAG CTTTTCCTAGGCCTCTTATAAGCCTCTTTAA); and primer set 3: forward AvrII, reverse HindIII (5′-TTCCTA GGCGAGAGTGGTTAGCTGA and 5′-TTAAGCTTCCTCTTATAAGCCTCTTTAA). The PCR products were purified, digested with the corresponding restriction enzymes, and sequentially cloned into the p5E-MCS from the Tol2 Kit (Kwan et al.,2007).
p5E mnx1-356bp-3x125bp enhancer: The 356-bp mnx1 sequence, located −2603 to −2247 upstream of the mnx1 protein coding sequence, was obtained by PCR amplification using the 3-kb mnx1 promoter as template (5′-AAAGGTACCCGAAAATATTTATTGCAAT AAAAGAG, 5′-TAACTCGAGCGCTTTGTCCGCTTCTCA). Cloning was performed via restriction sites KpnI and XhoI (introduced by PCR) upstream of the mnx1 3x125bp enhancer sequence.
p5E mnx1-782bp: The 782-bp mnx1 sequence (−2603 to −1821) which includes the 125-bp enhancer segment and the upstream 356-bp segment was obtained by PCR amplification using the plasmid containing the 3-kb mnx1 promoter as template (5′-AAAA TATTTATTG CAATAAAAGAG, 5′-TCTTATAAGCCTCTT TAA) and inserted into pENTR™5′-TOPO (Invitrogen) via TOPO-TA cloning reaction.
p5E 14xUAS: A DNA fragment containing 14 UAS elements fused to the fish basal promoter E1b derived from the carp β-actin was PCR amplified using pBluescript 14xUAS-E1B-lyn-EGFP (Koster and Fraser,2001) as a template and cloned into pENTR™5′-TOPO (Invitrogen) via TOPO-TA cloning reaction.
p3E 14xUAS-prenEGFP: To create a prenylated-EGFP (prenEGFP) expression cassette, the SV40pA and 14xUAS sequences were cloned upstream of EGFPCAAXpA from the Tol2 Kit (Kwan et al.,2007). The resulting cassette was amplified by PCR to introduce attB2 (GGGGACAGCTTTCTTGTACAAAGTGGAATCCTA GGAGATCCAGACATG) and attB3 (GGGGACAACTTT GTATAATAAAGTTGATTGGAAAAAACCTCCCACACCTC) sites (underlined) at the 5′ and 3′ ends respectively, and the PCR product was recombined via Gateway cloning BP reaction using BP clonase II into Tol2Kit pDONR P2R-P3 (p3E) vector (Kwan et al.,2007).
Generation and Genotyping of Transgenic Zebrafish
To generate a stable transgenic fish (Tg(mnx1-3x125bp:Gal4-VP16) expressing the yeast Gal4 transcriptional activator (Brand and Perrimon,1993; Giniger et al.,1985) fused to a virally derived VP16 activator sequence (Gal4-VP16)(Sadowski et al.,1988) in motor neurons, a Tol2 destination vector (pDestTol2pA2) was created carrying three copies of the 125-bp mnx1 enhancer, followed by the Gal4-VP16 sequence, a SV40 polyadenylation (pA) signal, and flanked by the Tol2 transposable elements (Kawakami et al.,1998). The plasmid was injected into wild type embryos together with in vitro transcribed transposase mRNA. Messenger RNA was synthesized using the mMESSAGE mMachine kit (Ambion, Austin TX). F0 embryos were raised to adulthood and crossed with wild type animals. DNA was extracted from the eggs with the DNeasy kit (Qiagen, Valencia, CA) and used as template to identify F0 animals with germ-line transmission by PCR amplification with primers annealing to the Gal4 sequence (forward, 5′-ATGAAGCTACTGTCTTCTATCG; and reverse, 5′-TGTCTTTGACCTTTGTTACTAC). F1 Tg (mnx1-3x125bp:Gal4-VP16) embryos were injected with a plasmid encoding carrying 14xUAS:prenEGFP and examined at 24 hpf for EGFP expression in spinal motor neurons.
Plasmid injections were carried out with an air-pressured Picospritzer III microinjector (Parker, Cleveland, OH) using glass micro needles pooled with a two stage Kopf 730 vertical puller (Kopf Instruments, Tujunga, CA). Plasmid DNA was prepared using Endotoxin Free Plasmid kits (Qiagen), diluted in injection solution (0.2M KCl, 0.04% phenol red) at a final concentration of 50 ng/μL, and 1–2 nL were injected into the cell of 1-cell stage embryos.
Immunostaining and Confocal Microscopy
Embryos were dechorionated, anesthetized, and sacrificed in ice-cold E3 embryo medium (NaCl, 5 mM; KCl, 0.17 mM; CaCl2, 0.33 mM; and MgSO4, 0.33 mM) containing 0.4% tricaine (MS222, Ethyl 3-aminobenzoate methanesulfonate salt)(Sigma-Aldrich, St Louis, MO), immersed in ice-cold 4% paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA) containing 1% dimethyl sulfoxide (DMSO) in Dulbecco's phosphate buffer saline (PBS), fixed for 3 h at room temperature (RT) followed by 10 min incubation in methanol at −20°C, washed in PBS, and incubated in blocking solution [2% bovine serum albumin (BSA) in PBS] for 1 h at RT. Embryos were then incubated in primary antibodies diluted in blocking solution overnight at 4°C, washed in PBS, incubated with corresponding secondary antibodies conjugated with Cy3 (Jackson ImmunoResearch, West Grove, PA) washed, deyolked, and mount in Prolong Gold (Invitrogen). The following primary antibodies were used: znp1, mouse monoclonal anti-synaptotagmin (Fox and Sanes,2007; Trevarrow et al.,1990) [Developmental Studies Hybridoma Bank (DSHB), Iowa city, IA], and SV2, mouse monoclonal anti-synaptic vesicle protein 2 (Buckley and Kelly,1985) (DSHB). For the labeling of nicotinic acetylcholine receptors, Alexa-546 conjugated α-bungarotoxin (Molecular Probes, Eugine, OR) [10 μg/ml] was added to the blocking solution and incubated overnight at 4°C. Embryos were observed and imaged with a Nikon C1Si confocal (Nikon, Tokyo, Japan) mounted on a Eclipse 90i upright microscope using Nikon lenses (10X/NA0.45 Plan-Apo, 20X/NA0.75 Plan-Apo, Pan-Fluor 40X/NA1.3 oil, a Plan-Apo 60X/NA1.4 oil, and a Apo-TIRF 100X/NA1.49 oil).
Zebrafish embryos and larvae were anesthetized in 0.02% tricaine, embedded in 1% low melting point agarose (SeaPlaque, Lonza, Rockland, ME) dissolved in Hank's solution (NaCl 137 mM, KCl 5.4 mM, Na2HPO4 0.25 mM, KH2PO4 0.44 mM, CaCl2 1.3 mM, MgS04 1.0 mM, NaHCO3 4.2 mM), and placed in a 35 mm tissue culture plastic dish. The dish was filled with Hank's solution and the specimens observed with a Zeiss LSM510 confocal scanning microscope (Carl Zeiss, Thornwood, NY) mounted on a Zeiss Axioplan up-right microscope using an Achroplan 40x/NA0.8 water immersion lens. The specimens were maintained at 26–28°C, the Hank's solution was replaced every hour, and confocal z-stacks images obtained every 20 min for up to 6 h. Specimens observed in consecutive days were removed from the agarose, maintain in a 28.5°C incubator in E3 embryo media, and re-embedded for the imaging session. Spontaneous movements and blood cell circulation at the end of the imaging session were used as indicators of embryo viability.
Analysis of DNA Sequence Homology
The Daio rerio genome (Zv8) was queried with mouse mnx1 313-bp sequence (Nakano et al.,2005), using cross-species megaBLAST algorithm (Altschul SF, 1990). The mnx1 5′ UTR regions of mouse, human, Fugu, and zebrafish were aligned using ClustalX2 to search for phylogenetically conserved sequences. The identified 356-bp and 125-bp fragments in zebrafish were aligned separately with mouse, human, and Fugu mnx1 5′ UTR to determine sequence identity. Identity values were calculated in BioEdit.
The authors thank Xia Zhao for her assistance in maintaining the zebrafish colony, embryo injection, and immunostaining. They also thank Chi-Bin Chien from the University of Utah for providing the Tol2 kit plasmids, and Dirk Meyer from the University of Freiburg for providing a plasmid carrying the mnx1 promoter.