Standard Article

You have free access to this content

Zebrafish as an Experimental Organism

  1. Corinne Houart

Published Online: 25 APR 2001

DOI: 10.1038/npg.els.0002094

eLS

eLS

How to Cite

Houart, C. 2001. Zebrafish as an Experimental Organism. eLS.

Author Information

  1. King's College, London, UK

Publication History

  1. Published Online: 25 APR 2001

History

  1. Top of page
  2. History
  3. Maintenance and Life Cycle
  4. Experimental Manipulations of the Embryo
  5. Zebrafish Genetics and Genomics
  6. The Future of Danio rerio
  7. References

The first steps towards establishing the zebrafish (Danio rerio) as an important model for studying vertebrate development were taken in 1972 by George Streisinger in Eugene, Oregon. Looking for a vertebrate in which to study early development, he chose the zebrafish for possessing three important qualities: the production of a large number of embryos accessible at the earliest stages of development, the optical clarity of the developing fish, and a diploid genome allowing genetic studies. His laboratory isolated stable inbred lines selected for health and the ability to give eggs and sperm by squeezing. From methods of feeding and breeding to those of homozygous diploid by early pressure and gamma-ray mutagenesis, Streisinger and colleagues built up the foundation of what was to become, 15 years later, a very powerful model organism. With the publication in the early 1980s of two important papers on ‘Production of clones of homozygous diploid zebrafish’ in Nature in 1981 (Striesinger et al., 1981) and ‘Induction of mutations by gamma-rays in pregonial germ cells of zebrafish embryos’ in 1983 (Walker and Streisinger, 1983), the zebrafish began to attract other scientists studying vertebrate development. See also History of Developmental Biology

Eisen and Westerfield's laboratories exploited the transparency of the embryos to study the development of motor neurons and their ability to project specifically to form correct muscle innervations (Eisen, 1991; Kimmel et al., 1991). They have been able to develop single-cell injections in embryos and to follow their fate and behaviour in vivo. Kimmel's group showed very quickly that it was possible to isolate efficiently mutants affected in some main developmental processes (Ho and Kane, 1990; Hatta et al., 1991; Halpern et al., 1997). Their characterization gave rise to new insights into vertebrate development, especially in axis formation. Again, the research was made easier by the production of large numbers of optically clear embryos at spawning. See also Motor System Organization, Axon Guidance, and Zebrafish Embryo as a Developmental System

Based on these first successes, two major laboratories decided in the early 1990s to undertake a large-scale mutagenesis project in zebrafish. Their goal was to generate a panel of fish lines carrying recessive point mutations affecting most of the genes involved in vertebrate development. Using ethyl-nitroso-urea (ENU) as a mutagen, they generated more than 2000 mutant lines (Zebrafish Issue, 1996). Concomitantly, an international effort towards mapping the zebrafish genome was undertaken. Thanks to this map, it is now possible to identify the gene modified in any mutant line available. Using that approach, several genes have already been found as key players in vertebrate gastrulation, brain development and midline signalling (Feldman et al., 1998, Lun and Brand, 1998; Reifers et al., 1998; Griffin et al., 1998; Karlstrom et al., 1999), including one-eyed-pinhead, a novel gene from the family of Crypto/Cryptic (Zhang et al., 1998). See also Mutagenesis, Cleavage and Gastrulation in Zebrafish Embryos, Vertebrate Central Nervous System, and Axon Guidance at the Midline

The availability of detailed analyses of any defect at a single-cell level, coupled with the capacity to identify the genes responsible for each phenotype, make possible the association between the genes involved in development and their function. Functional genomics will therefore be the most important approach in zebrafish development for the next decade.

Zebrafish research has progressed from less than 10 international publications per year in the 1970s to more than 400 a year in the 1990s. George Streisinger would have been amazed to see that the zebrafish has now reached a position similar to the mouse as a model system for developmental studies.

Maintenance and Life Cycle

  1. Top of page
  2. History
  3. Maintenance and Life Cycle
  4. Experimental Manipulations of the Embryo
  5. Zebrafish Genetics and Genomics
  6. The Future of Danio rerio
  7. References

Danio rerio is a tropical fish originally found in India. It lives in fresh water and has been a popular aquarium fish for decades. From pet shops to laboratories, the transition went smoothly. The zebrafish requires warm water and is usually kept at 28°C. The water flowing through the tank is sterilized after each passage and recycled into the system. Water quality is an important parameter for the health and well-being of the fish and recycling allows for a minimal use of water supply but requires a complex sterilizing system to ensure good water quality.

Zebrafish reproduce very easily in the aquarium. Each female can produce up to 1000 eggs a week. Embryos grow rapidly and from birth they take less than 3 months to reach sexual maturity. The generous spawning and short generation times are the main qualities that made the zebrafish a popular model system. Moreover, sperm and eggs can be routinely collected by squeezing the adult males and females, allowing in vitro fertilization and storage of frozen sperm. Mutant lines that have not been kept in facilities as live fish can be re-established from samples of sperm stored for years in liquid nitrogen. See also Reproduction in Fish, and Cryopreservation of Cells

In addition to its natural advantages, the zebrafish can also provide pseudodiploid adult females, by putting the eggs under pressure (Westerfield, 1995). Haploid embryos are also available, by in vitro fertilization of the eggs with ultraviolet light (UV)-inactivated sperm (Westerfield, 1995).

Experimental Manipulations of the Embryo

  1. Top of page
  2. History
  3. Maintenance and Life Cycle
  4. Experimental Manipulations of the Embryo
  5. Zebrafish Genetics and Genomics
  6. The Future of Danio rerio
  7. References

As with many other model organisms, the function of a gene during zebrafish development is often assessed by over- or misexpression of the gene using injection of RNA or DNA constructs. As in Xenopus or chicken, the zebrafish embryo has also been used to establish the fate map of different tissues or organs and is suitable for studies in culture systems (Detrich et al., 1999a). However, its strength resides in its transparency, which gives a unique quality to whole-mount in situ (Figure 1c) and antibody staining and most importantly led to the development of refined ablation and transplantation techniques on early embryos. It is these two specific aspects of experimental manipulations that we will review here. See also Microinjection into Xenopus Oocytes, and In Situ Hybridization

thumbnail image

Figure 1. (a) One-day-old zebrafish embryo, lateral view. (b) Dorsal view of the neurons of the pineal gland. (c) Expression pattern of an early neurogenic marker in the neural plate at the end of gastrulation (dorsal view). (d) Primary axonal tracts in a 2-day-old brain. (e) Lateral view of the zebrafish brain showing telencephalic emx-1 gene expression (blue) and transplanted cells in the eye (brown). (f) Rescue of a telencephalon (expressing emx-1 in blue) in a brain mutant after transplant of wild-type anterior neural plate cells (brown). (g) Transplanted kriesler mutant cells cannot contribute to rhombomere 5 and 6 in a wild-type embryo. Brown cells (g) are kreisler cells. Blue cells are rhombomere 3 and 5 krox-20-expressing cells. (g) is adapted from Moens et al. 1998.

The outstanding level of detection of RNA and protein in vivo led to important contributions in major processes of vertebrate development. A couple of examples are the finding that primary axonal tracts (Figure 1d) grow along specific boundaries of gene expression domains (Macdonald et al., 1997), and the analysis of the development of the pineal gland (Masai et al., 1997; Figure 1b). See also Axon Growth

The easy access to any cell in the embryo at any stage of development allowed scientists to address problems of cell commitments and cell identities. Ablations and transplantations of specific subpopulations of cells in the early neural plate of gastrula embryos led to the identification of a signalling centre located at the anterior edge of the plate required for patterning and differentiation of the forebrain (Houart et al., 1998; Figure 1d). This same technique is routinely used to study the effect of genetic deficiencies at a cellular level, by generating clones of wild-type cells in mutant embryos and vice versa (Figure 1e,f). By this mean, Moens et al. 1998 showed that kreisler (kr), a gene previously studied in mouse, is required for the cells to respond to the hindbrain rhombomeres 5 and 6 positional information. Cells carrying a null mutation for kr are unable to be part of rhombomere 5 or 6 in a wild-type embryo (Moens et al., 1998; Figure 1g). By similar mosaic analyses, specific genetic interactions involved in fine regulation between axial (notochord) and paraxial (muscle) mesoderm has been dissected (Amacher and Kimmel, 1998). These last examples also illustrate the wealth of information that can arise by coupling cellular approaches with genetics. See also Vertebrate Central Nervous System: Pattern Formation

Zebrafish Genetics and Genomics

  1. Top of page
  2. History
  3. Maintenance and Life Cycle
  4. Experimental Manipulations of the Embryo
  5. Zebrafish Genetics and Genomics
  6. The Future of Danio rerio
  7. References

Mutagenesis screens

Inspired by the first mutants characterized in Oregon, Nusslein-Volhard's laboratory as well as Wolfgang Dreiver and Mark Fishman's groups undertook, in the early 1990s, a couple of large-scale mutagenesis screens of the zebrafish genome. Their goal was to achieve in a vertebrate what had been done in Drosophila and to collect point mutations in all the principal genes involved in zebrafish development. In less than 5 years they isolated more than 2000 mutant lines covering every developmental process (Figure 2). By complementation tests between lines showing similar phenotypes, it transpired that a few genes have been hit up to 10 times while an important number of others have been mutated only once. The significant level of single hits strongly indicates that despite the numerous lines isolated, the screens have not reached saturation of the genome. The lines carry defects affecting a variety of developing tissue: body axes (+/−10 genes), mesoderm (20), central nervous system (+/−60), heart (+/−20), blood (+/−17), endoderm (+/−10), eyes (+/−10), ears (+/−20), pigments (+/−60) and jaw and gills (+/−20). It has been estimated that these cover 50–90% of the genes involved in each process.

thumbnail image

Figure 2. (a–d) Lateral views of mutants affected in notochord and tail formation (a is wild-type). (e, f) Lateral view of the brain of the acerebellar mutation (f, arrow) compared to wild-type (e, arrow). Alcian blue staining of the jaw in wild-type (g) and a jaw mutant (h). Lateral view of a 5-day-old wild-type (i) and colourless pigment mutant (j) embryos. (k) Wild-type (top, arrow) and heart mutant (bottom, arrow) 5-day-old embryos. (l) Axonal staining (brown) of a forebrain mutant masterblind (left) and wild-type (right) 2-day-old embryos. All panels but (l) are adapted from the zebrafish issue, Development 123 (December 1996).

Since then, a few smaller groups have begun their own screens with the aim of completing the initial collection in their specific field of interest. Most of the mutations missed by the large screens are probably mutations giving more subtle morphological defects and for that reason, the most recent screens are carried out using labelling of embryos with specific antibodies or RNAs targeted on the tissue or organ of interest, increasing the chance of identifying slight modifications in a specific developmental step. One of these screens has been targeted on heart defects and has isolated very successfully an impressive number of new mutant lines, several of which are good models for human heart conditions. The same laboratory also successfully isolated mutations affecting organ formation and functions (Drummond et al., 1998; Dooley et al., 2000). See also Heart Development: Gene Control

Thanks to the two large screens, the zebrafish became an important genetic system for studying vertebrate development. Several of the mutants isolated in these screens are now the subject of intensive studies by other laboratories and already new insights into the mechanisms of dorsoventral and anteroposterior patterning (Barth et al., 1999; Fekany-Lee et al., 2000) have been achieved, as well as information on gastrulation movements (Heisenberg et al., 2000) and midline signalling (Feldman et al., 2000; Karlstrom et al., 1999; Barresi et al., 2000). See also Vertebrate Embryo: Establishment of Left-Right Asymmetry, and Vertebrate Embryo: Neural Patterning

Genetic maps and gene cloning

Despite the success of the mutagenesis approach and the wealth of information obtained from the mutant analyses, zebrafish genetics was lacking one important component to be an ideal model system: the accessibility to the genes responsible for the phenotype. The main disadvantage of the zebrafish is the absence of targeted and insertional mutagenesis (absence of embryonic stem cell in which to perform gene knockout, no transposon like insertional technique). Although scientists are putting considerable effort into establishing these techniques (Detrich et al., 1999b), it will take a few more years before the community can use these approaches efficiently. Meanwhile, it has been necessary to find ways to localize the site of mutation and isolate the modified gene. Concomitantly with the large-scale screens, John Postlethwait took the initiative and began the mapping of the zebrafish genome. He was soon followed by other groups and most of the genome is now covered by a high density of markers. These are of different origins: randomly amplified polymorphic DNA (RAPD), CA repeat, single-strand conformational polymorphism (SSCP), amplified fragment length polymorphism (AFLP) and expressed sequence tags (EST). It has therefore become feasible to identify efficiently markers located at less than a centimorgan from the mutation of interest (Detrich et al., 1999b). To date, approximately 20 mutations have been identified as modifications of a specific gene. All but one have been found by linking the position of the mutation with candidate genes. The two exceptions are one-eyed-pinhead (oep) and miles apart (mil), which have been identified by walking from the nearest markers to the gene. These impressive results have been possible thanks to the availability of zebrafish artificial chromosome libraries. Yeast artificial chromosomes (YACs) contain fragments of 250–500 kb of genomic sequence and bacterial artificial chromosomes (BACs) have inserts of up to 100 kb. By comparing a few BACs containing the nearest marker, Alexander Schier's and Didier Stainier's teams have been able to restrict their choice to a small fragment of DNA and located the gene by rescuing the mutant phenotype and injecting the early embryo with a candidate cDNA. The oep gene encodes a novel EGF-related protein with similarity to the EGF-CFC proteins cripto, cryptic and FRL-1 (Zhang et al., 1998). The mil gene encodes a novel sphingosine-1-phosphate receptor (Kupperman et al., 2000). See also Mutagenesis: Site-specific, Genome Catalogue, Genome Sequence Analysis, and Genome Mapping

The progress in gene mapping also led to the finding that the zebrafish genome carries a partial duplication similar to that in the mammalian genome. This duplication is thought to have occurred early in the teleost lineage. It implies that some of the genes involved in vertebrate development are present in two copies in fish and, in some cases, their properties will probably have evolved so that each copy is now responsible for part of the initial gene function. One can also predict that in rare cases one of the copies will have acquired a new function. In addition to the obvious inconvenience of not having a perfect correlation between fish and mammals, the split of a specific gene function into two gene copies (or paralogues) has the advantage of allowing the dissection of the developmental function of genes involved early in complex processes of development. In such cases, mouse null mutations give early aborted embryos from which very poor information can be drawn, while each paralogue is likely to cover complementary parts of that function and allows the analysis of partial phenotypes. See also Genome Evolution: Overview, Gene Mapping: Comparative, and Evolutionary Developmental Biology: Gene Duplication, Divergence and Co-option

The Future of Danio rerio

  1. Top of page
  2. History
  3. Maintenance and Life Cycle
  4. Experimental Manipulations of the Embryo
  5. Zebrafish Genetics and Genomics
  6. The Future of Danio rerio
  7. References

A tremendous international effort is currently under way to place on the zebrafish genomic map all the known genes as well as all existing mutations. Extensive nonredundant full-length cDNA libraries (containing a unique clone for each specific cDNA sequence) are being engineered and an embryonic expression profile for each cDNA of each given library will be obtained. From 2001–2006, the zebrafish community will see its genetic resources grow exponentially. We can therefore predict an explosion of investigations exploring the many routes from gene to function during development. Despite the lack of efficient targeted mutagenesis and thanks to recent progress in gene technology, the zebrafish will, in the very near future, become an outstanding genetic model system. In addition, we predict that interspecies genetic approaches between mouse and fish will be increasingly taken to assess the evolution of gene function in vertebrate development. See also Experimental Organisms Used in Genetics, Transgenic Zebrafish Production, and The Promise of Whole Genome Sequencing

Glossary
Axis formation

Establishment of the prechordal plate and notochord, main components of the embryonic axial mesoderm, pillar of the body axis in the early embryo.

Complementation test

Genetic cross between a male heterozygous for mutation A and a female heterozygous for mutation B in order to test whether mutation A and B are affecting the same gene. If so, a quarter of the progeny obtained must show a phenotype, if not the progeny is wild-type.

Fate map

Definition of territories in the early embryo that will give rise to a particular tissue in the adult.

Phenotype

Morphological traits of the embryo or adult, consequence of its genetic background.

Point mutation

Modification of one base pair in the coding or noncoding region of a gene which affects its correct transcription.

Rhombomere

Segmental unit of the hindbrain. The hindbrain is composed of a repetition of 7–8 rhombomeres.

Transplantation

Insertion of a few cells from a donor embryo into a host of the same age (homochronic) or of a different age (heterochronic).

Wild-type

Phenotype of the embryo or adult of a given species as found in the wild, in the absence of induced mutations.

References

  1. Top of page
  2. History
  3. Maintenance and Life Cycle
  4. Experimental Manipulations of the Embryo
  5. Zebrafish Genetics and Genomics
  6. The Future of Danio rerio
  7. References
  • Amacher S and Kimmel CB (1998) Promoting notochord fate and repressing muscle development in zebrafish axial mesoderm. Development 125: 13971406.
  • Barresi MJ, Stickney HL and Devoto SH (2000) The zebrafish slow-muscle-omitted gene product is required for Hedgehog signal transduction and the development of slow muscle identity. Development 127: 21892199.
  • Barth KA, Kishimoto Y, Rohr KB et al. (1999) Bmp activity establishes a gradient of positional information throughout the entire neural plate. Development 126: 49774987.
  • Detrich WH, Westerfield M and Zon LI (eds) (1999a) The Zebrafish: Biology. Methods in Cell Biology, vol. 59. San Diego: Academic Press.
  • Detrich WH, Westerfield M and Zon LI (eds) (1999b) The Zebrafish: Genetics and Genomics. Methods in Cell Biology, vol. 60. San Diego: Academic Press.
  • Dooley K and Zon LI (2000) Zebrafish: a model system for the study of human disease. Current Opinion in Genetics and Development 10: 252256.
  • Drummond IA, Majumdar A, Hentschel H et al. (1998) Early development of the zebrafish pronephros and analysis of mutations affecting pronephric function. Development 125: 46554667.
  • Eisen JS (1991) The role of interactions in determining cell fate of two identified motoneurons in the embryonic zebrafish. Neuron 8: 231240.
  • Fekany-Lee K, Gonzalez E, Miller-Bertoglio V and Solnica-Krezel L (2000) The homeobox gene bozozok promotes anterior neuroectoderm formation in zebrafish through negative regulation of BMP2/4 and Wnt pathways. Development 127: 23332345.
  • Feldman B, Gates MA, Egan ES et al. (1998) Zebrafish organizer development and germ-layer formation require nodal-related signals. Nature 395: 181185.
  • Feldman B, Dougan ST, Schier AF and Talbot WS (2000) Nodal-related signals establish mesendodermal fate and trunk neural identity in zebrafish. Current Biology 10: 531534.
  • Griffin KJP, Amacher SL, Kimmel CB and Kimelman D (1998) Molecular identification of spadetail: regulation of zebrafish trunk and tail mesoderm function by T-box genes. Development 125: 33793388.
  • Halpern ME, Hatta K, Amacher SL et al. (1997) Genetic interactions in zebrafish midline development. Developmental Biology 187: 154170.
  • Hatta K, Kimmel CB, Ho RK and Walker C (1991) The cyclops mutation blocks specification of the floor plate of the zebrafish central nervous system. Nature 350: 339341.
  • Heisenberg CP, Tada M, Rauch GJ et al. (2000) Silberblick/Wnt11 mediates convergent extension movements during zebrafish gastrulation. Nature 405: 7681.
  • Ho RK and Kane DA (1990) Cell-autonomous action of zebrafish spt-1 mutation in specific mesodermal precursors. Nature 348: 728730.
  • Houart C, Westerfield M and Wilson SW (1998) A small population of anterior cells patterns the forebrain during zebrafish gastrulation. Nature 391: 788792.
  • Karlstrom RO, Talbot WS and Schier AF (1999) Comparative syntony cloning of zebrafish you-too: mutations in the Hedgehog target gli2 affect ventral forebrain patterning. Genes and Development 13: 388393.
  • Kimmel CB, Hatta K and Eisen JS (1991) Genetic control of primary neurons development in zebrafish. Development supplement 2: 4757.
  • Kupperman E, Au S, Osborne N, Waldron S and Stainier DY (2000) A sphingosine-1-phosphate receptor regulator cell migration during vertebrate heart development. Nature 406: 192195.
  • Lun K and Brand M (1998) A series of no isthmus (noi) alleles of the zebrafish pax2.1 gene reveals multiple signaling events in development of the midbrain-hindbrain boundary. Development 125: 30493062.
  • Macdonald R, Scholes J, Strahle U et al. (1997) The pax protein Noi is required for commissural axon pathway formation in the rostral forebrain. Development 124: 23972408.
  • Masai I, Heisenberg CP, Barth KA, Macdonald R, Adamek S and Wilson SW (1997) Floating head and masterblind regulate neuronal patterning in the roof of the forebrain. Neuron 18: 4357.
  • Moens CB, Cordes SP, Giorgianni MW, Barsh GS and Kimmel CB (1998) Equivalence in the genetic control of hindbrain segmentation in fish and mouse. Development 125: 381391.
  • Reifers F, Bohli H, Walsh EC, Crossley PH, Stainier DY and Brand M (1998) Fgf8 is mutated in zebrafish acerebellar (ace) mutants and is required for maintenance of midbrain-hindbrain boundary development and somitogenesis. Development 125: 23812395.
  • Striesinger G, Walker C, Dover N, Knauber D and Singer F (1981) Production of clones of homozygous diploid zebrafish. Nature 291: 293296.
  • Walker C and Striesinger G (1983) Induction of mutations by γ-rays in pre-gonial germ cells of zebrafish embryos. Genetics 103: 125136.
  • Westerfield M (1995) The Zebrafish Book. Eugene, OR: University of Oregon Press.
  • Zebrafish Issue (1996) Development 123: 1460.
  • Zhang J, Talbot WS and Schier AF (1998) Positional cloning identifies zebrafish one-eyed-pinhead as a permissive EFG-related ligand required during gastrulation. Cell 92: 241251.