ABSTRACT
In the vertebrate ventral spinal cord, p2 progenitors give rise to two interneuron subtypes: excitatory V2a interneurons and inhibitory V2b interneurons. In the differentiation of V2a and V2b cells, Notch signaling promotes V2b fate at the expense of V2a fate. Later, V2b cells extend axons along the ipsilateral side of the spinal cord and express the inhibitory transmitter GABA. Notch signaling has been reported to inhibit the axonal outgrowth of mature neurons of the central nervous system; however, it remains unknown how Notch signaling modulates V2b neurite outgrowth and maturation into GABAergic neurons. Here, we have investigated neuron-specific Notch functions regarding V2b axon growth and maturation into zebrafish GABAergic neurons. We found that continuous neuron-specific Notch activation enhanced V2b fate determination but inhibited V2b axonal outgrowth and maturation into GABAergic neurons. These results suggest that Notch signaling activation is required for V2b fate determination, whereas its downregulation at a later stage is essential for V2b maturation. Accordingly, we found that a Notch signaling downstream gene, her15.1, showed biased expression in V2 linage cells and downregulated expression during the maturation of V2b cells, and continuous expression of her15.1 repressed V2b axogenesis. Our data suggest that spatiotemporal control of Notch signaling activity is required for V2b fate determination, maturation and axogenesis.
INTRODUCTION
In the vertebrate central nervous system (CNS), interneurons transmit information from peripheral sensory neurons to motoneurons. Along the dorsoventral axis, the ventral interneurons V0, V1, V2 and V3 are generated in the ventral region of the neural tube (Jessell, 2000; Wilson and Maden, 2005). V2 interneurons are a type of ventral interneuron. V2 progenitors (p2) in the medial ventricular zone differentiate into excitatory V2a interneurons and inhibitory V2b interneurons (Batista et al., 2008; Briscoe et al., 2000; Karunaratne et al., 2002; Kimura et al., 2006). V2a and V2b interneurons form a neural network with other interneurons (V0 and V1) and motoneurons and are involved in locomotor activity (Ramirez-Jarquin and Tapia, 2018).
Notch signaling is a signal-transduction system that is commonly used throughout animal development. In vertebrate CNS development, Notch signaling is required for the maintenance of a stem or progenitor cell pool (Siebel and Lendahl, 2017). Genes in the hairy and enhancer of split (Hes) family, Hes1 and Hes5, are transcriptional repressors and downstream effectors of Notch signaling. Hes1 and Hes5 critically contribute to the maintenance of mammalian neural progenitor cells by repressing the expression of proneuronal genes, such as Ascl1, at the direction of Notch signaling (Engler et al., 2018; Imayoshi and Kageyama, 2014). Hes6 is another member of the Hes family of genes, and it promotes neurogenesis by inhibiting the function of Hes1 and Hes5 (Fior and Henrique, 2005; Koyano-Nakagawa et al., 2000). Hes6 physically interacts with the Hes1 protein and inhibits its transcriptional function (Koyano-Nakagawa et al., 2000). Conversely, Hes5 is negatively regulated by Hes6 at the transcriptional level (Fior and Henrique, 2005).
In the Notch signaling pathway, the different combinations of the Hes family of genes control cell fate determination, proliferation and stem cell maintenance spatiotemporally. During the development of V2 interneurons, different combinations of Notch ligands and receptors regulate p2 maintenance and V2a and V2b cell fate determination; moreover, Notch signaling activation is required for V2b interneuron differentiation (Kimura et al., 2008; Okigawa et al., 2014). However, the genes in the Hes family that are involved in V2a and V2b interneuron development remain unknown. After fate determination, V2b interneurons extend axons along the ipsilateral side of the spinal cord (Goulding, 2009). These observations are inconsistent with those of previous studies, which reported that the activation of Notch signaling in mature neurons inhibits neurite outgrowth (Berezovska et al., 1999; Sestan et al., 1999; Shi et al., 2011). In addition, whether Notch signaling is involved in the maturation of V2b neurons, which express the inhibitory neurotransmitter GABA, remains unknown.
Here, we have examined the function of Notch signaling in V2b axon outgrowth and maturation into GABAergic neurons. We used and characterized the neuronal-specific driver line deltaAdmc72 (Okigawa et al., 2014) because the systemic activation or suppression of Notch lacks spatial specificity. Non-neural tissue defects affect neuronal development; for example, caudal primary motoneuron axonal outgrowth depends on somite differentiation (Gray et al., 2001), and Notch signaling regulates the differentiation of non-neuronal tissues, such as somites or blood cells. (Siebel and Lendahl, 2017). Moreover, we explored genes in the Hes family that act downstream of Notch signaling and found that hes6 and her15.1 (a possible Hes5 homolog in zebrafish) are potential mediators of V2 cell development. Our findings provide additional details regarding Notch signaling-dependent V2 fate determination and maturation, as well as insights into how a complex neuronal circuit is established by the temporal regulation of Notch signaling.
RESULTS AND DISCUSSION
deltaAdmc72a-Gal4FF-dependent gene induction was observed in V2A and V2B interneurons
We used the deltaAdmc72a line, an enhancer trap line in which Gal4FF is inserted at the deltaA locus to control Notch activity in the V2 linage. The deltaAdmc72a heterozygous zebrafish line was crossed with Tg(UAS:mRFP) zebrafish, and spatiotemporal deltaAdmc72a-driven induction of gene expression in the V2 linage was analyzed (Fig. S1A-D) (Okigawa et al., 2014). We performed immunostaining for a V2a-specific transcription factor, Vsx2 (Kimura et al., 2008), in TgBAC(vsx1:GFP); deltaAdmc72a; Tg(UAS:mRFP) embryos. Vsx2 was expressed in one cell of the paired GFP-expressing cells in TgBAC(vsx1:GFP); deltaAdmc72a; Tg(UAS:mRFP) embryos at 24 h post fertilization (hpf), as shown previously (Kimura et al., 2008; Okigawa et al., 2014). deltaAdmc72a-driven mRFP expression was observed in both GFP/Vsx2 double-positive V2a cells and the GFP-positive V2b cells (over 94% of the total of V2a and V2b, n=217; Fig. S1A,B). We performed fluorescence recovery after photobleaching (FRAP) analysis. mRFP fluorescence was recovered at 16 somite stage (ss), 19 ss and 24 hpf, but the recovery rate at 24 hpf was less than it was at 16 ss and 19 ss (Fig. S1C,D). We also performed in situ hybridization using the mRFP probe and confirmed that mRFP mRNA was expressed in V2 linage interneurons of deltaAdmc72a; Tg(UAS:mRFP) embryos at 24 hpf (Fig. S1E). These data indicate that deltaAdmc72a-driven induction of genes occurred in both V2a and V2b interneuron cell lineages; this induction was attenuated at 24 hpf but not at 16 ss and 19 ss.
deltaAdmc72a-driven Notch signaling modulation affected V2A and V2B fate determination
To activate Notch signaling in V2 cell lineages, deltaAdmc72a and Tg(UAS:Notch1a intracellular domain (UAS:NICD)) (Scheer and Campos-Ortega, 1999) were crossed, and UAS-GAL4-dependent NICD induction was observed from 16 ss to 24 hpf, which is when p2 to V2a/b differentiation occurs. However, at 36 hpf, NICD induction was found to be attenuated in the V2 linage (Fig. S2A). In addition, intrinsic deltaA expression was enhanced by NICD induction at 16 ss and 19 ss but not at 24 hpf (Fig. S2B). These results cannot be explained by lateral inhibition, as observed in neural plates (Appel and Eisen, 1998). One of the Notch ligands, Delta-like 4, is positively regulated by Notch signaling in endothelial cells (Caolo et al., 2010). deltaAdmc72a-driven NICD might positively control intrinsic deltaA expression in the V2 linage.
In deltaAdmc72a; Tg(UAS:NICD), the expression of vsx2- and V2b-specific transcription factors tal1 and gata3 (Batista et al., 2008) was examined by in situ hybridization and antibody staining. deltaAdmc72a-mediated Notch signaling activation increased the number of V2b interneurons and decreased the number of V2a interneurons (Fig. 1A,D,E). However, the total number of V2 interneurons did not differ between sibling controls and deltaAdmc72a; Tg(UAS:NICD) embryos (Fig. 1D,E).
deltaAdmc72a-mediated changes in Notch signaling affected V2a and V2b cell differentiation. (A-C) Lateral view of vsx2 and gata3 expression in each line. (D,F,H) The immunostaining against Tal1 or Vsx2 in the neural tube in each line. The white arrowheads indicate V2b or V2a cells. (E,G,I) Data analysis of D,F,H.
deltaAdmc72a-mediated changes in Notch signaling affected V2a and V2b cell differentiation. (A-C) Lateral view of vsx2 and gata3 expression in each line. (D,F,H) The immunostaining against Tal1 or Vsx2 in the neural tube in each line. The white arrowheads indicate V2b or V2a cells. (E,G,I) Data analysis of D,F,H.
To reduce Notch signaling in V2 cell lineages, we generated two UAS-mediated effector lines, Tg(UAS;Xsu(H)DN) and Tg(UAS:hMAMLDN), using the Tol2 transposon system (Kawakami et al., 2000); in these lines, the expression of dominant-negative forms of the Xenopus laevis Suppressor of hairless [XSu(H)DN] (Wettstein et al., 1997) and human master-mind-like Pro103stop (hMAMLDN) (Kitagawa et al., 2001) was regulated by UAS. XSu(H)DN lacks DNA-binding ability and inhibits intrinsic Su(H) function by forming nonfunctional transcription complexes (Wettstein et al., 1997). hMAMLDN can form transcription complexes with NICD and Su(H), but it lacks transactivating activity and inhibits intrinsic Maml function (Kitagawa et al., 2001). The number of V2a interneurons was increased, and the number of V2b interneurons was decreased in both deltaAdmc72a; Tg(UAS:XSu(H)DN) and deltaAdmc72a; Tg(UAS:hMAMLDN) embryos (Fig. 1B,C,F-I). The total number of V2 interneurons was increased in deltaAdmc72a; Tg(UAS:hMAMLDN) embryos but not in deltaAdmc72a; Tg(UAS:Xsu(H)DN) embryos. This difference may have been caused by the different degrees of Notch signaling suppression achieved by Xsu(H)DN and hMAMLDN. Consistently, we showed that inhibition of Notch signaling accelerates differentiation of the V2 linage from p2, the maintenance of which is regulated by Notch signaling (Okigawa et al., 2014). Taken together, our data indicate that deltaAdmc72a-driven Notch signaling modulation affects V2a/V2b cell fate more effectively than it regulates maintenance of V2 progenitor cells.
deltaAdmc72a-driven Notch signaling modulation did not affect non-neuronal tissue development
We examined whether deltaAdmc72a-driven Notch signaling modulation affects the development of other tissues, such as somites. The segmentation of somites in deltaAdmc72a; Tg(UAS:NICD) and deltaAdmc72a; Tg(UAS:hMAMLDN) embryos was found to be normal at 24 hpf compared with that of sibling control embryos (normal somites; deltaAdmc72a; Tg(UAS:NICD), n=16, 100%; deltaAdmc72a; Tg(UAS:hMAMLDN), n=8, 100%; Fig. S3A,B, upper panels). In addition, Notch signaling regulates the formation of neural crest cell-derived melanophores (Cornell and Eisen, 2000). However, deltaAdmc72a; Tg(UAS:NICD) and deltaAdmc72a; Tg(UAS:hMAMLDN) embryos did not show a reduction in the number or an abnormal appearance of melanophores, at least up to 2 dpf [normal pigmentation: deltaAdmc72a; Tg(UAS:NICD), n=6, 100%; deltaAdmc72a; Tg(UAS:hMAMLDN), n=8, 100%; Fig. S3A,B, bottom panels]. These results suggest that deltaAdmc72-mediated Notch signaling modulation does not affect these non-neuronal tissues.
GABAergic neuron differentiation was affected by deltaAdmc72a-driven notch activation
V2b interneurons are inhibitory, and the vast majority of V2b interneurons are GABAergic (Batista et al., 2008). Tal1 is required for the acquisition of GABAergic characteristics of V2b cells after differentiation (Andrzejczuk et al., 2018). To assess the effect of continuous Notch signaling activation on the GABAergic characteristics of V2b, we generated Tg(tal1:EGFP) embryos. In Tg(tal1:EGFP); deltaAdmc72a; Tg(UAS:NICD) embryos, V2b fate transformation was enhanced (Fig. 2A,B). However, the percentage of GABAergic V2b interneurons was decreased in deltaAdmc72a; Tg(UAS:NICD) embryos (Fig. 2C). This result indicates that the continuous activation of Notch signaling promotes tal1-positive V2b fate, but it might interfere with the acquisition of GABAergic characteristics of V2b cells.
Notch activation affects GABAergic neuron differentiation. (A) Lateral view of the immunostaining for GABA in the neural tube of the trunk region of Tg(tal1:EGFP); deltaAdmc72a; Tg(UAS:NICD) embryos at 24 hpf. The white arrowheads indicate V2b cells. The white arrows indicate GABA-negative V2b cells. (B,C) Data analysis of A.
Notch activation affects GABAergic neuron differentiation. (A) Lateral view of the immunostaining for GABA in the neural tube of the trunk region of Tg(tal1:EGFP); deltaAdmc72a; Tg(UAS:NICD) embryos at 24 hpf. The white arrowheads indicate V2b cells. The white arrows indicate GABA-negative V2b cells. (B,C) Data analysis of A.
Forced Notch activation in V2B cells inhibited axon outgrowth
Our data indicate that Notch signaling promotes V2b fate over V2a fate. However, previous reports showed that forced Notch signaling activation affects neuronal cell morphology (Sestan et al., 1999). Thus, we investigated the effect of deltaAdmc72a-driven Notch signaling activation on V2b axonal projections at 24 hpf. To analyze V2b axon outgrowth at the single-cell level in vivo, we injected UAS:NICD and UAS:Lifeact-mCherry (which was used for visualizing the shape of the cells) expression constructs into deltaAdmc72a; Tg(tal1:EGFP) embryos at the one-cell stage. In control embryos, V2b cells exhibited a long axonal projection that reached to the caudal region of the neural tube (Fig. 3A, left panels; n=7). Conversely, deltaAdmc72a-driven Notch signaling activation reduced the length of V2b axonal extensions (Fig. 3A, right panels; n=15). Statistical analysis indicated a significant reduction in axonal length induced by forced Notch signaling activation compared with that of the control (Fig. 3B).
Forced Notch activation affected V2b neuronal outgrowth. (A) Lateral view of V2b axons in the deltaAdmc72a; Tg(tal1:EGFP) trunk region at 24 hpf. The white dotted lines indicate V2b cell shapes. Arrowheads indicate V2b cell bodies. (B) Data analysis of A.
Previous studies have shown that Notch signaling promotes zebrafish V2b interneuron, KA′ interneuron and mouse dILB interneuron differentiation at the expense of V2a, motoneuron progenitor and dILA differentiation, respectively (Kimura et al., 2008; Mizuguchi et al., 2006; Okigawa et al., 2014; Shin et al., 2007). However, Notch signaling activation inhibits neuronal outgrowth in the mouse neocortex, hippocampus and cerebral cortical neurons, and in chick commissural neurons (Berezovska et al., 1999; Sestan et al., 1999; Shi et al., 2011). These observations raise the question of how Notch signaling activity affects V2b axogenesis. Recently, Pinto-Teixeira et al. (2018) showed that transient Notch signaling is active in two consecutive neuronal divisions, and that study found the synchronization of their projection patterns in Drosophila. This provides evidence of the importance of transient Notch activity for the coupling of neurogenesis and for neuronal network formation; however, the manner in which the dendrites of each subtype are properly organized remains unknown. Here, we show that Notch signaling activity must be transient; high Notch activity is required for the V2b neuron fate, whereas its attenuation is required for neuronal outgrowth. Therefore, spatiotemporal control of Notch signaling might be an essential molecular mechanism that plays a central role in neuronal differentiation and subsequent neuronal outgrowth, and contributes to the establishment of a complex neuronal circuit. Further studies are needed to clarify whether similar transient Notch activity is required for KA′ and dILB neurite outgrowth.
her15.1, but not hes6, might be a downstream effector of Notch signaling in V2B differentiation
To further explore the downstream genes of Notch signaling activation in V2b cell lineages, we investigated the expression of Notch signaling downstream targets, the Hes family of genes (Salazar and Yamamoto, 2018), during V2 interneuron development. A very low level of expression of hes15.1 was detected in p2 (data not shown) and V2 neuron pairs at 16 ss, which is when V2a and V2b differentiation was starting (66% of V2 neuron pairs were double-negative for her15.1 expression, n=56; Fig. 4A and Fig. S4A). Expression of her15.1 became obvious in one of the V2 neuron pairs at 19 ss (80% of V2 neuron pairs showed biased her15.1 expression, n=100; Fig. 4A and Fig. S4A). However, her15.1 expression was decreased at 24 hpf (74% of V2 neuron pairs were double negative for her15.1 expression, n=170; Fig. 4A and Fig. S4A). Moreover, her15.1 was ectopically induced in V2 interneurons by deltaAdmc72a-driven Notch activation at 24 hpf; in deltaAdmc72a; Tg(UAS-NICD) embryos, 95% of V2 paired cells showed her15.1 positivity in both cells (n=68), whereas 74% of V2 neuron pairs in sibling controls were her15.1 negative, and 26% of neuron pairs showed biased expression (n=155) (Fig. 4B, Fig. S4A,D). In addition, her15.1 was co-expressed with the V2b marker gene tal1 (Fig. 4C). Next, we overexpressed Myc-tagged wild-type (WT) and truncated mutant (ΔWRPW) her15.1 in the V2 lineage. her15.1WT induced the tal1 in both V2 neuron pairs (Fig. 4D, 86%, n=7). On the other hand, her15.1ΔWRPW, which lacks the interaction motif for the co-repressor Groucho (Fischer and Gessler, 2007), did not affect biased tal1 expression (Fig. 4D, 93%, n=15). These results suggest that her15.1 is a transiently expressed Notch signaling downstream target gene that can convert the V2a lineage to the V2b lineage in a repressive function-dependent manner. Next, we examined Her15.1 function in axon outgrowth. deltaAdmc72a-driven her15.1WT overexpression in the V2 linage of interneurons significantly inhibited axon outgrowth compared with that of the sibling control at 24 hpf (Fig. 4E,F; control: n=36, her15.1 overexpression: n=17). These results suggested that her15.1 function is involved in V2b interneuron differentiation from p2, but its downregulation is required for V2b interneuron morphological maturation.
her15.1 was expressed in V2b interneurons under Notch signaling. (A) Statistical analysis of her 15.1 expression patterns in V2 neuron pairs. The expression patterns correspond to the expression criteria detailed in Fig. S4A. (B) Representative her15.1 expression patterns in V2 neuron pairs. (C) Expression of her15.1 and tal1 in V2 neuron pairs at 19 ss. (D) her15.1WT, but not her15.1ΔWRPW, exhibited induced tal1 expression in both V2 neuron pairs. (E) her15.1WT overexpression affected axon outgrowth. The magenta dotted lines show the shape of V2 linage interneurons. (F) Data analysis of E. (G) Hypothetical model of V2a and V2b interneuron differentiation and maturation. Notch signaling is transiently activated in the V2b progenitor and induces her15.1. her15.1 might activate downstream genes and direct to V2b fate. After V2b fate determination, Notch signaling is downregulated and the V2b cell matures.
her15.1 was expressed in V2b interneurons under Notch signaling. (A) Statistical analysis of her 15.1 expression patterns in V2 neuron pairs. The expression patterns correspond to the expression criteria detailed in Fig. S4A. (B) Representative her15.1 expression patterns in V2 neuron pairs. (C) Expression of her15.1 and tal1 in V2 neuron pairs at 19 ss. (D) her15.1WT, but not her15.1ΔWRPW, exhibited induced tal1 expression in both V2 neuron pairs. (E) her15.1WT overexpression affected axon outgrowth. The magenta dotted lines show the shape of V2 linage interneurons. (F) Data analysis of E. (G) Hypothetical model of V2a and V2b interneuron differentiation and maturation. Notch signaling is transiently activated in the V2b progenitor and induces her15.1. her15.1 might activate downstream genes and direct to V2b fate. After V2b fate determination, Notch signaling is downregulated and the V2b cell matures.
We also found that hes6 showed transient and biased expression in the V2 lineage. At 16 ss, hes6 was strongly expressed in p2 cells (100%, n=36; Fig. S4B); expression was typically observed in both cells of a pair of V2 cells (double positive; 51%) or in one cell in a pair of V2 cells (biased; 39%), whereas no hes6 expression was observed in a small number of neuron pairs (double negative; 10%) (n=77, Fig. S4C,D). At 19 ss, the percentage of cells in the double-positive population was decreased (39%, n=115 neuron pairs), whereas that of the biased subset was unchanged (39%, n=100; Fig. S4C,D). Conversely, the percentage of double-negative samples was increased (36%, n=100 neuron pairs) at 19 ss (Fig. S4C,D). At 24 hpf, the percentage of the biased hes6 expressed subset was diminished, and over 90% of V2 neuron pairs were double negative (n=176 neuron pairs; Fig. S4C,D). Furthermore, in a biased hes6-expressing V2 pair, hes6 was complementarily expressed with the V2b marker gene tal1 at 19 ss (Fig. S4F). These data suggest that the high hes6 expression observed in p2 is inherited by daughter cells but is maintained only transiently in V2a interneurons and rapidly disappears in V2b interneurons. However, hes6 expression was hardly affected by Notch signaling activation, suggesting that the expression of hes6 is controlled by a mechanism other than Notch signaling during V2 interneuron differentiation (Fig. S4D,E).
How does the biased expression of hes6 and her15.1 fit into previous V2 differentiation models? We propose a model for hes6 and her15.1 involvement during V2 development (Fig. 4G). Our data suggest that hes6 expression in the V2 lineage is independent of Notch signaling. hes6 may be induced initially by proneural bHLH proteins, such as ASCL-family molecules, but not by the Notch pathway in p2 progenitors (Koyano-Nakagawa et al., 2000). The V2 progenitors also express gata2 and lhx3, and the expression of gata2 in V2a interneurons and of lhx3 in V2b interneurons is rapidly downregulated (Batista et al., 2008). Lhx3 positively regulates V2a interneuron differentiation though Vsx2/Chx10 expression (Clovis et al., 2016). As Hes5 suppresses lhx3 expression in hESC-derived pMN-like progenitors (Ben-Shushan et al., 2015), Her15.1 might also suppress lhx3 in cells that will potentially enter the V2b lineage. The negative-feedback loop between hes6 and her15.1 may contribute to the refinement of V2a and V2b segregation, as evidenced by the hes5/hes6 circuitry of negative cross-regulation observed during chick and Xenopus neurogenesis (Fior and Henrique, 2005; Koyano-Nakagawa et al., 2000). Ascl1 reportedly promotes GABAergic neuron specification and neurite outgrowth (Fode et al., 2000; Parras et al., 2002). Hes5 inhibits proneuronal genes, including Ascl1 (Engler et al., 2018). Therefore, forced ectopic Notch-activated her15.1 expression might block GABAergic neuron specification and axon growth. Further studies are needed to examine the roles and regulation of hes6 and her15.1 during V2 interneuron development in detail.
MATERIALS AND METHODS
Fish maintenance and transgenic lines
Zebrafish were raised and maintained under standard conditions (Westerfield, 1995) with the approval of the Chiba University Institutional Animal Care and Use Committee. Zebrafish embryos were obtained from the natural spawning of wild-type adults or identified carriers of deltaAdmc72, TgBAC(vsx1:GFP), Tg(tal1:GFP), Tg(UAS:NICD), Tg(UAS:Xsu(H)DN) and Tg(UAS:hMAMLDN). For the deltaAdmc72 line, we only used heterozygous mutants.
Generation of transgenic lines
Two plasmids with two Tol2 transposition sites flanking the UAS:hMAMLDN-HA or UAS:Xsu(H)DN-Myc sequence of DNA, which were to be integrated, were constructed. These Tol2-UAS:hMAMLDN-HA or Tol2-UAS:Xsu(H)DN-Myc constructs were injected with Tol2 transposase mRNA into one-cell stage embryos, which were then raised to adulthood. Subsequently, the embryos were crossed with a wild-type line, and F1 embryos were collected. The F1 embryos were raised to adulthood, and transgene carriers were identified by PCR using the following primers: UAS, hMAMLDN F, 5′-GACTGGGTGCTCAGGTAGTGGTTGT-3′ and R, 5′-GTGCAGGGCGAAGGTGTGTT-3′; UAS, Xsu(H)DN F, 5′-GACTGGGTGCTCAGGTAGTGGTTGT-3′ and R, 5′-TTAGCACAGATAAGGCGGAATACACTTT-3′. Subsequently, Tg(UAS:hMAMLDN)chi10 and Tg(UAS:Xsu(H)DN)chi11 fish were isolated.
For the generation of Tg(tal1:GFP), a promoter sequence of the tal1 gene (approximately 7.2 kb in length) was amplified from the zebrafish genomic DNA using the following primers: 5′-CAAGTGACATATCACCGGAATCTCAAT-3′ and 5′-CAGAAAATGTACAAAGTGGTATC-3′. The amplified tal1 promoter, GFP and SV40 poly(A) were subcloned into a Tol2-based plasmid, and transgenic fish were generated as described previously (Urasaki et al., 2006).
Whole-mount in situ hybridization and antibody staining
Whole-mount in situ hybridization and whole-mount fluorescent in situ hybridization were performed as described previously (Lauter et al., 2014; Mizoguchi et al., 2017; Mizoguchi et al., 2011). For tyramide-signaling-amplification-based detection, Cy3-tyramide (AAT Bioquest) and iFluor 488 tyramide (AAT Bioquest) were used. All probes used here have been published previously: gata3 (Neave et al., 1995), vsx2 (Kimura et al., 2006), tal1 (Qian et al., 2007), her15.1 (Raya et al., 2003), hes6 (Kawamura et al., 2005) and mRFP (Campbell et al., 2002).
Whole-mount antibody staining was performed as described previously (Mizoguchi et al., 2011, 2017) using polyclonal or monoclonal antibodies against Vsx2 (1:500) (Kimura et al., 2008), Tal1 (1:10) (Qian et al., 2007), GABA (1:500) (rabbit polyclonal; Merck Millipore, #ABN131), mRFP (1:400) (rabbit polyclonal; MBL, #PM0005), GFP (1:400) (mouse monoclonal, GF200; Nakarai Tesque, #04363-24; or rabbit polyclonal; MBL, #598) and Myc (1:400) (mouse monoclonal, 9E10; Wako; or rabbit polyclonal, A-14; Santa Cruz Biotechnology). Fluorescently labeled anti-mouse (1:400), anti-guinea pig (1:400) or anti-rabbit IgG (1:400) (Thermo Fisher Scientific, #A-11029; Jackson ImmunoResearch, #715-495-150, #706-495-148, #711-545-152, #711-515-152) antibodies were used as secondary antibodies. For anti-GABA staining, embryos were fixed with 4% paraformaldehyde (PFA)/0.05% GTA/5 mM MgSO4/1×PBS/0.1% Triton X-100. For staining with other antibodies, embryos were fixed with 4% PFA/1× PBS. Confocal imaging was performed with an SP8 confocal microscope (Leica) using a 40× lens. The fluorescence intensity was measured by Fiji software (Schindelin et al., 2012).
DNA injection
To generate the UAS:NICD construct, Notch1a ICD was cut with BamHI and NotI from pCS2-Notch1a ICD, and then subcloned into the p5E-UAS vector. For the generation of UAS:Myc-her15.1WT and UAS:Myc-her15.1ΔWRPW, her15.1WT and her15.1ΔWRPW fragments were amplified from zebrafish cDNA and subcloned into pCS3MT. PCR was performed with the following primers: her15.1WT F, 5′-GGGGAATTCAATGGCTCCTGTGTATATGACTGAATAC-3′ and R, 5′-GGGCTCGAGCTACCAGGGTCTCCAGAGCGGAGCG-3′; her15.1ΔWRPW F, 5′- GGGGAATTCAATGGCTCCTGTGTATATGACTGAATAC-3′ and R, 5′-GGGCTCGAGCTAGAGCGGAGCGTGCGCCCGCGGCTCCTGC-3′. Then, Myc-tagged her15.1WT and her15.1ΔWRPW were cut with ClaI and NotI, and subcloned into a p5E-UAS vector.
The UAS:NICD (25 pg) and UAS:Lifeact-mCherry (Mizoguchi et al., 2016) (25 pg) constructs were injected into Tg(tal1:GFP); deltaAdmc72a animals at the one-cell stage. UAS:Myc-her15.1 WT and UAS:MT-her15.1 ΔWRPW constructs were injected into TgBAC(vsx1:GFP); deltaAdmc72a animals at the one-cell stage.
Statistical analysis
Statistical analysis was performed by GraphPad Prism (*P values indicated in the figure; n.s., not significant; unpaired, two-tailed t-test). Data are mean±s.d. with individual data points shown; n in each graph indicates the sample size.
Acknowledgements
We thank Drs. M. Kitagawa, N. Sheer, C. Kintner and Z. Wen for providing plasmid vectors and an antibody; the Zebrafish National BioResource Project in Japan for providing wild-type RIKEN zebrafish strains; and H. Matsuo, M. Kajikawa, A. Shirota, R. Inaba, K. Matsuura, Y. Akatsu, N. Fujiwara, C. Tsukamoto and A. Higaki for technical assistance.
Footnotes
Author contributions
Conceptualization: T.M., M. Itoh; Methodology: T.M., M. Itoh; Validation: T.M.; Formal analysis: T.M.; Investigation: T.M., M.F., M. Iihama, X.S., S.F., S.K., S.O.; Resources: S.-i.H.; Data curation: T.M., M.F., M. Iihama, X.S., S.F., S.K., S.O., M. Itoh; Writing - original draft: T.M.; Writing - review & editing: T.M., M. Itoh; Project administration: M. Itoh; Funding acquisition: T.M., M. Itoh.
Funding
This research was supported by a Grant-in-Aid for Young Scientists (B) (JP25840068 and JP16K18547) and a Grant-in-Aid for Scientific Research (C) (JP19K06454) from the Japan Society for the Promotion of Science (JSPS) (to T.M.), by a Grant-in-Aid for Scientific Research on Innovative Areas (Multi-Dimensional Fluorescence Live Imaging of Cellular Functions and Molecular Activities; JP25113703) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) (to T.M.), by a Grant-in-Aid for Scientific Research (B) (JP18H02568) from the Japan Society for the Promotion of Science (to M. Itoh), by a Grant-in-Aid for Scientific Research on Innovative Areas (Brain Protein Ageing and Dementia Control; grant no. JP15H01551) from the Ministry of Education, Culture, Sports, Science and Technology (to M. Itoh) and by a Daiichi Sankyo research grant (to M. Itoh).
References
Competing interests
The authors declare no competing or financial interests.