Search

In vertebrates, the periodic formation of somites from the presomitic mesoderm (PSM) is driven by the molecular oscillator known as the segmentation clock. The hairy-related gene, hes6/her13.2, functions as a hub by dimerizing with other oscillators of the segmentation clock in zebrafish embryos. Although hes6 exhibits a posterior–anterior expression gradient in the posterior PSM with a peak at the tailbud, the detailed mechanisms underlying this unique expression pattern have not yet been clarified. By establishing several transgenic lines, we found that the transcriptional regulatory region downstream of hes6 in combination with the hes6 3′UTR recapitulates the endogenous gradient of hes6 mRNA expression. This downstream region, which we termed the PT enhancer, possessed several putative binding sites for the T-box and Ets transcription factors that were required for the regulatory activity. Indeed, the T-box transcription factor (Tbx16) and Ets transcription factor (Pea3) bound specifically to the putative binding sites and regulated the enhancer activity in zebrafish embryos. In addition, the 3′UTR of hes6 is required for recapitulation of the endogenous mRNA expression. Furthermore, the PT enhancer with the 3′UTR of hes6 responded to the inhibition of retinoic acid synthesis and fibroblast growth factor signaling in a manner similar to endogenous hes6. The results showed that transcriptional regulation by the T-box and Ets transcription factors, combined with the mRNA stability given by the 3′UTR, is responsible for the unique expression gradient of hes6 mRNA in the posterior PSM of zebrafish embryos.

In mammals, cerebellar neurons are categorized as glutamatergic or GABAergic, and are derived from progenitors that express the proneural genes atoh1 or ptf1a, respectively. In zebrafish, three atoh1 genes, atoh1a, atoh1b, and atoh1c, are expressed in overlapping but distinct expression domains in the upper rhombic lip (URL): ptf1a is expressed exclusively in the ventricular zone (VZ). Using transgenic lines expressing fluorescent proteins under the control of the regulatory elements of atoh1a and ptf1a, we traced the lineages of the cerebellar neurons. The atoh1(+) progenitors gave rise not only to granule cells but also to neurons of the anteroventral rhombencephalon. The ptf1a(+) progenitors generated Purkinje cells. The olig2(+) eurydendroid cells, which are glutamatergic, were derived mostly from ptf1a(+) progenitors in the VZ but some originated from the atoh1(+) progenitors in the URL. In the adult cerebellum, atoh1a, atoh1b, and atoh1c are expressed in the molecular layer of the valvula cerebelli and of the medial corpus cerebelli, and ptf1a was detected in the VZ. The proneural gene expression patterns coincided with the sites of proliferating neuronal progenitors in the adult cerebellum. Our data indicate that proneural gene-linked neurogenesis is evolutionarily conserved in the cerebellum among vertebrates, and that the continuously generated neurons help remodel neural circuits in the adult zebrafish cerebellum.

The lateral line system displays highly divergent patterns in adult teleost fish. The mechanisms underlying this variability are poorly understood. Here, we demonstrate that the lateral line mechanoreceptor, the neuromast, gives rise to a series of accessory neuromasts by a serial budding process during postembryonic development in zebrafish. We also show that accessory neuromast formation is highly correlated to the development of underlying dermal structures such as bones and scales. Abnormalities in opercular bone morphogenesis, in endothelin 1-knockdown embryos, are accompanied by stereotypic errors in neuromast budding and positioning, further demonstrating the tight correlation between the patterning of neuromasts and of the underlying dermal bones. In medaka, where scales form between peridermis and opercular bones, the lateral line displays a scale-specific pattern which is never observed in zebrafish. These results strongly suggest a control of postembryonic neuromast patterns by underlying dermal structures. This dermal control may explain some aspects of the evolution of lateral line patterns.

The cerebellum is important for the integration of sensory perception and motor control, but its structure has mostly been studied in mammals. Here, we describe the cell types and neural tracts of the adult zebrafish cerebellum using molecular markers and transgenic lines. Cerebellar neurons are categorized to two major groups: GABAergic and glutamatergic neurons. The Purkinje cells, which are GABAergic neurons, express parvalbumin7, carbonic anhydrase 8, and aldolase C like (zebrin II). The glutamatergic neurons are vglut1+ granule cells and vglut2high cells, which receive Purkinje cell inputs; some vglut2high cells are eurydendroid cells, which are equivalent to the mammalian deep cerebellar nuclei. We found olig2+ neurons in the adult cerebellum and ascertained that at least some of them are eurydendroid cells. We identified markers for climbing and mossy afferent fibers, efferent fibers, and parallel fibers from granule cells. Furthermore, we found that the cerebellum-like structures in the optic tectum and antero-dorsal hindbrain show similar Parvalbumin7 and Vglut1 expression profiles as the cerebellum. The differentiation of GABAergic and glutamatergic neurons begins 3 days post-fertilization (dpf), and layers are first detectable 5 dpf. Using anti-Parvalbumin7 and Vglut1 antibodies to label Purkinje cells and granule cell axons, respectively, we screened for mutations affecting cerebellar neuronal development and the formation of neural tracts. Our data provide a platform for future studies of zebrafish cerebellar development.

Newborn neurons migrate extensively in the radial and tangential directions to organize the developing vertebrate nervous system. We show here that mutations in zebrafish trilobite (tri) that affect gastrulation-associated cell movements also eliminate tangential migration of motor neurons in the hindbrain. In the wild-type hindbrain, facial (nVII) and glossopharyngeal (nIX) motor neurons are induced in rhombomeres 4 and 6, respectively, and migrate tangentially into r6 and r7 (nVII) and r7 (nIX). In all three tri alleles examined, although normal numbers of motor neurons are induced, nVII motor neurons are found exclusively in r4, and nIX-like motor neurons are found exclusively in r6. The migration of other neuronal and nonneuronal cell types is unaffected in tri mutants. Rhombomere formation and the development of other hindbrain neurons are also unaffected in tri mutants. Furthermore, tangential neuronal migration occurs normally in the gastrulation mutant knypek, indicating that the trilobite neuron phenotype does not arise nonspecifically from aberrant gastrulation-associated movements. We conclude that trilobite function is specifically required for two types of cell migration that occur at different stages of zebrafish development.

The broad diversity of neurons is vital to neuronal functions. During vertebrate development, the spinal cord is a site of sensory and motor tasks coordinated by interneurons and the ongoing neurogenesis. In the spinal cord, V2-interneuron (V2-IN) progenitors (p2) develop into excitatory V2a-INs and inhibitory V2b-INs. The balance of these two types of interneurons requires precise control in the number and timing of their production. Here, using zebrafish embryos with altered Notch signaling, we show that different combinations of Notch ligands and receptors regulate two functions: the maintenance of p2 progenitor cells and the V2a/V2b cell fate decision in V2-IN development. Two ligands, DeltaA and DeltaD, and three receptors, Notch1a, Notch1b, and Notch3 redundantly contribute to p2 progenitor maintenance. On the other hand, DeltaA, DeltaC, and Notch1a mainly contribute to the V2a/V2b cell fate determination. A ubiquitin ligase Mib, which activates Notch ligands, acts in both functions through its activation of DeltaA, DeltaC, and DeltaD. Moreover, p2 progenitor maintenance and V2a/V2b fate determination are not distinct temporal processes, but occur within the same time frame during development. In conclusion, V2-IN cell progenitor proliferation and V2a/V2b cell fate determination involve signaling through different sets of Notch ligand–receptor combinations that occur concurrently during development in zebrafish.

Despite a number of reports on transgenic zebrafish, there have been no reports on transgenic zebrafish in which the gene is under the control of a promoter of zebrafish origin. Neither have there been reports on transgenic zebrafish in which the gene is under the control of a tissue-specific promoter/enhancer. To investigate whether it is possible to generate transgenic zebrafish which reliably express a reporter gene in specific tissues, we have isolated a zebrafish muscle-specific actin (alpha-actin) promoter and generated transgenic zebrafish in which the green fluorescent protein (GFP) reporter gene was driven by this promoter. In total, 41 GFP-expressing transgenic lines were generated with a frequency of as high as 21% (41 of 194), and GFP was specifically expressed throughout muscle cells in virtually all of the lines (40 of 41). Nonexpressing transgenic lines were rare. This demonstrates that a tissue-specific promoter can reliably drive reporter gene expression in transgenic zebrafish in a manner identical to the control of the endogeneous expression of the gene. Levels of GFP expression varied greatly from line to line; i.e., fluorescence was very weak in some lines, while it was extremely high in others. We also isolated a zebrafish cytoskeletal beta-actin promoter and generated transgenic zebrafish using a beta-actin-GFP construct. In all of the four lines generated, GFP was expressed throughout the body like the beta-actin gene, demonstrating that consistent expression could also be achieved in this case. In the present study, we also examined the effects of factors which potentially affect the transgenic frequency or expression levels. The following results were obtained: (i) expression levels of GFP in the injected embryo were not strongly correlated to transgenic frequency; (ii) the effect of the NLS peptide (SV40 T antigen nuclear localization sequence), which has been suggested to facilitate the transfer of a transgene into embryonic nuclei, remained to be elusive; (iii) a plasmid vector sequence placed upstream of the construct might reduce the expression levels of the reporter gene.

F-spondin is a secreted protein expressed at high levels by the floor plate cells. The C-terminal half of the protein contains six thrombospondin type 1 repeats, while the N-terminal half exhibited virtually no similarity to any other protein until recently, when a Drosophila gene termed M-spondin was cloned; its product was found to share two conserved domains with the N-terminal half of F-spondin. We report the molecular cloning of four zebrafish genes encoding secreted proteins with these conserved domains. Two are zebrafish homologs of F-spondin, while the other two, termed mindin1 and mindin2, encode mutually related novel proteins, which are more related to the Drosophila M-spondin than to F-spondin. During embryonic development, all four genes are expressed in the floor plate cells. In addition to the floor plate, mindin1 is expressed in the hypochord cells, while mindin2 is expressed in the sclerotome cells. When ectopically expressed, Mindin proteins selectively accumulate in the basal lamina, suggesting that Mindins are extracellular matrix (ECM) proteins with high affinity to the basal lamina. We also report the spatial distribution of one of the F-spondin proteins, F-spondin2. F-spondin2 is localized to the thread-like structure in the central canal of the spinal cord, which is likely to correspond to Reissner's fiber known to be present in the vertebrate phylum. In summary, our study has defined a novel gene family of ECM molecules in the vertebrate, all of which may potentially be involved in development of the midline structure.

In vertebrate embryos, spinal motor neurons project through segmentally reiterated nerves into the somites. Here, we report that zebrafish secondary motor neurons, which are similar to motor neurons in birds and mammals, depend on myotomal cues to navigate into the periphery. We show that the absence of myotomal adaxial cells in you-too/gli2 embryos severely impairs secondary motor axonal pathfinding, including their ability to project into the somites. Moreover, in diwanka mutant embryos, in which adaxial cells are present but fail to produce cues essential for primary motor growth cones to pioneer into the somites, secondary motor axons display similar pathfinding defects. The similarities between the axonal defects in you-too/gli2 and diwanka mutant embryos strongly suggest that pathfinding of secondary motor axons depends on myotome-derived cues, and that the diwanka gene is a likely candidate to produce or encode such a cue. Our experiments also demonstrate that diwanka plays a central role in the migration of primary and secondary motor neurons, suggesting that both neural populations share mechanisms underlying axonal pathfinding. In summary, we provide compelling evidence that myotomal cells produce multiple signals to initiate and control the migration of spinal nerve axons into the somites.

Islet-1 (Isl1) is a member of the Isl1 family of LIM-homeodomain transcription factors (LIM-HD) that is expressed in a defined subset of motor and sensory neurons during vertebrate embryogenesis. To investigate how this specific expression of isl1 is regulated, we searched for enhancers of the isl1 gene that are conserved in vertebrate evolution. Initially, two enhancer elements, CREST1 and CREST2, were identified downstream of the isl1 locus in the genomes of fugu, chick, mouse, and human by BLAST searching for highly similar elements to those originally identified as motor and sensory neuron-specific enhancers in the zebrafish genome. The combined action of these elements is sufficient for completely recapitulating the subtype-specific expression of the isl1 gene in motor neurons of the mouse spinal cord. Furthermore, by direct comparison of the upstream flanking regions of the zebrafish and human isl1 genes, we identified another highly conserved noncoding element, CREST3, and subsequently C3R, a similar element to CREST3 with two CDP CR1 recognition motifs, in the upstream regions of all other isl1 family members. In mouse and human, CRESTs are located as far as more than 300 kb away from the isl1 locus, while they are much closer to the isl1 locus in zebrafish. Although all of zebrafish CREST2, CREST3, and C3R activate gene expression in the sensory neurons of zebrafish, CREST2 of mouse and human does not have the sequence necessary for sensory neuron-specific expression. Our results revealed both a remarkable conservation of the regulatory elements regulating subtype-specific gene expression in motor and sensory neurons and the dynamic process of reorganization of these elements whereby each element increases the level of cell-type specificity by losing redundant functions with the other elements during vertebrate evolution.

Bilateral symmetric tissues must interpret axial references to maintain their global architecture during growth or repair. The regeneration of hair cells in the zebrafish lateral line, for example, forms a vertical midline that bisects the neuromast epithelium into perfect mirror-symmetric plane-polarized halves. Each half contains hair cells of identical planar orientation but opposite to that of the confronting half. The establishment of bilateral symmetry in this organ is poorly understood. Here, we show that hair-cell regeneration is strongly directional along an axis perpendicular to that of epithelial planar polarity. We demonstrate compartmentalized Notch signaling in neuromasts, and show that directional regeneration depends on the development of hair-cell progenitors in polar compartments that have low Notch activity. High-resolution live cell tracking reveals a novel process of planar cell inversions whereby sibling hair cells invert positions immediately after progenitor cytokinesis, demonstrating that oriented progenitor divisions are dispensable for bilateral symmetry. Notwithstanding the invariably directional regeneration, the planar polarization of the epithelium eventually propagates symmetrically because mature hair cells move away from the midline towards the periphery of the neuromast. We conclude that a strongly anisotropic regeneration process that relies on the dynamic stabilization of progenitor identity in permissive polar compartments sustains bilateral symmetry in the lateral line.