The Need for an Atlas of Zebrafish Development
Over the past two decades zebrafish have been established as a genetically tractable model system to investigate many different aspects of vertebrate development. The strengths of this model system stem from a unique combination of embryological manipulability and the optical clarity of the early embryo and larvae, which allows, by simple optical inspection, the visualization of cell biological events within an in vivo context. This is further enhanced by the ability to apply invertebrate-style forward genetics to questions of vertebrate development. While the credentials of the system as a developmental model have long been recognized, it is the application of the zebrafish to a different aspect of biology that has generated a surge in popularity and greatly increased zebrafish laboratory use. Many investigators have now turned to directly modeling human disease states in zebrafish, realizing that many of the same strengths that made it a compelling model for the study of development also allow it to complement existing mammalian disease models (Lieschke and Currie 2007).
Numerous human disease conditions are now being modeled in zebrafish, from addiction syndromes to carcinoma. Remarkably, in the vast majority of instances the results of these studies indicate that the zebrafish models share a similar disease pathogenesis with humans (Darland and Dowling 2001; Langenau et al. 2003; Amsterdam et al. 2004; Carvan et al. 2004; Lockwood et al. 2004; Berghmans et al. 2005; Shepard et al. 2005; Patton et al. 2005). Zebrafish have also come of age as a reverse genetic model system. The ability to specifically knock down and mutate individual loci by a number of different techniques has allowed function of human disease gene orthologues to be directly examined in zebrafish. Antisense oligonucleotides or morpholinos have been utilized very effectively to knock down specific gene function during the embryonic and early larval period. Targeted lesion detection (Wienholds et al. 2003) has produced mutant models of a number of human diseases. This strategy has identified p53 mutant zebrafish that have a predisposition to cancer, rag1 deficient zebrafish with immunodeficiency (Wienholds et al. 2002, Berghmans et al. 2005), and has also been utilized to generate mutations in the APC gene that predispose to colorectal cancer (Hurlstone et al. 2003).
These approaches possess the ability to revolutionize the way that zebrafish biologists approach disease modeling in this system. The commitment of the Wellcome Trust’s Sanger Centre to producing a fully sequenced and integrated high-quality genome sequence for the zebrafish has enabled the development of many of the new reverse genetic techniques available to zebrafish researchers as well as facilitating the cloning of randomly generated mutations. Analysis of the zebrafish genome has demonstrated a considerable level of synteny with the human genome, and direct orthologues of human genes can nearly always routinely be located within the zebrafish genome. An ambitious new project that aims to mutate every gene within the zebrafish genome using a combination of the approaches outlined above has also been recently proposed. There is a possibility that a percentage of mutant phenotypes produced by the zebrafish knockout consortium will be phenotyped in a systematic way and catalogued in a database.
Another recent development has raised the stakes considerably for zebrafish as a human disease model. The application of chemical genetics to zebrafish biology has allowed for the first time the simultaneous analysis of a mutant phenotype representative of a human disease and the screening for drugs that will influence this phenotype in vivo (Peterson et al. 2004). The ability to screen in vivo using zebrafish allows the drug discovery process to leapfrog a number of the initial hurdles of toxicity and specificity that plague drug discovery employing in vitro models.
The unique features of zebrafish biology, such as external fertilization, high fecundity, ease of adding putative drugs to larval fish, make this the only fully integrated vertebrate model system for drug discovery. Collectively, these recent advances have heralded a watershed period for researchers utilizing zebrafish as a tool for modeling human disease and indicate a further expansionary phase in its use as a model system given that many of its strengths complement existing mammalian models so fully. However, despite the advances in many areas of zebrafish biology and the generation of sophisticated models of human disease, one area has substantially lagged behind. There is a lack at this current point in time of a detailed anatomical reference for zebrafish. While the first 24 hours of zebrafish development have been well studied, comparatively little is known about the development of larval, juvenile and adult anatomy. Such knowledge is critical for the analysis of developmental processes extending later in development and the accurate modeling of human disease in zebrafish, as these periods are the most likely to generate information that is relevant to human disease pathogenesis. Therefore the aim of this book is to provide an anatomical account of the development of the zebrafish from embryo to adult. Our hope is that this resource, in conjunction with the online three-dimensional (3D) anatomical resource FishNet (http://www.FishNet.org.au, discussed in Chapter 2) from which the images in this book are derived, can be utilized as a standard reference text on which to base further analyses that will investigate aspects of normal and perturbed zebrafish development.
Existing Anatomical Resources for Zebrafish
The major online resource for zebrafish is the zebrafish information network ZFIN (http://www.zfin.org). ZFIN curates information on tools, publications and laboratories involved in zebrafish research, as well as information about the zebrafish strains and DNA stocks held by the Zebrafish International Resource Center (ZIRC). Importantly, ZFIN also contains the first online collection of anatomical images comprising of nearly 80 sections covering the first 5 days of development. Other anatomical resources are scarce for zebrafish and piecemeal in their representation of anatomy. Several specialized zebrafish anatomy atlases exist in book form, including atlases of adult and embryonic brain anatomy. Necessarily these types of atlases are limited by showing only two-dimensional (2D) images in restrictive views. For instance the adult brain atlas (Wullimann et al. 1996), despite its high degree of anatomical complexity, shows 2D brain structure in complete isolation from any other tissues. The embryonic brain atlas is also limited in scope, showing only transverse views of a very limited number of stages (Mueller and Wulliman 2005). Other embryonic atlases and staging series have been published, which collectively approximate the resources currently online at ZFIN (Kimmel et al. 1995; Schilling 2002). All of these resources rely on 2D imaging.
Several online organ-specific 3D atlases of development have now been developed for zebrafish. A 3D atlas of zebrafish vasculature created by microangiography in the Weinstein laboratory shows the formation of the vasculature from 1 to 7 days in a series of 3D renderings (http://uvo.nichd.nih.gov/atlas.html; Isogai et al. 2001; Kamei et al. 2004). This site allows the user to view a series of images at each stage showing the vasculature in 3D utilizing confocal microscopy of specimens labeled using microangiography to chart the forming vasculature. This series of images represents the first example of a 3D developmental model of any organ system in zebrafish. However, the lack of ability to simultaneously visualize both the vasculature and the tissue through which it forms limits the usefulness of the atlas. Furthermore, the use of confocal microscopy to generate 3D models is limited to stages at which the laser light can suitably penetrate the tissue, and necessarily this is restricted to the earliest stages of zebrafish development.
A high resolution 3D anatomical atlas of the zebrafish developing brain is also being developed within the group of Steve Wilson at UCL in London (http://www.zebrafishbrain.org). This atlas relies on the optical clarity of the zebrafish embryo, coupled with a host of transgenic and antibody labelings to highlight discrete neuroanatomical structures within the developing brain. It uses confocal microscopy exclusively to generate highly detailed 3D interactive models of embryonic and larval brain development. Although this will be a unique and important neuroanatomical resource this approach will necessarily be restricted to relatively early stages of zebrafish brain development and will not be applicable to juvenile and adult zebrafish.
A growing repository of 2D section information is being deposited at the Penn State Atlas project (http://zfatlas.psu.edu), which, when completed, will be a significant data set. Altho...