Investigating the genetics of visual processing, function ... - CiteSeerX

Neurogenetics DOI 10.1007/s10048-011-0273-x

REVIEW ARTICLE

Investigating the genetics of visual processing, function and behaviour in zebrafish Sabine L. Renninger & Helia B. Schonthaler & Stephan C. F. Neuhauss & Ralf Dahm

Received: 21 September 2010 / Accepted: 4 January 2011 # Springer-Verlag 2011

Abstract Over the past three decades, the zebrafish has been proven to be an excellent model to investigate the genetic control of vertebrate embryonic development, and it is now also increasingly used to study behaviour and adult physiology. Moreover, mutagenesis approaches have resulted in large collections of mutants with phenotypes that resemble human pathologies, suggesting that these lines can be used to model diseases and screen drug candidates. With the recent development of new methods for gene targeting and manipulating or monitoring gene expression, the range of genetic modifications now possible in zebrafish is increasing rapidly. Combined with the classical strengths of the zebrafish as a model organism, these advances are set to substantially expand the type of biological questions that can be addressed in this species. In this review, we outline how the potential of the zebrafish can be harvested in the context of eye development and visual function. We review recent technological advances used to study the formation of the eyes and visual areas of the brain, visual processing on the cellular, subcellular and molecular level, and the genetics of visual behaviour in vertebrates.

S. L. Renninger : S. C. F. Neuhauss Institute of Molecular Life Sciences, University of Zurich, 8057, Zurich, Switzerland H. B. Schonthaler : R. Dahm Spanish National Cancer Research Centre (CNIO), C/ Melchor Fernández Almagro 3, 28029 Madrid, Spain R. Dahm (*) Department of Biology, University of Padova, Via U. Bassi 58/B, 35121 Padua, Italy e-mail: [email protected]

Keywords Zebrafish (Danio rerio) . Embryonic development . Genetics . Eye . Vision . Visual processing and behaviour . Transgenic techniques . Ocular disease

Introduction The zebrafish as a model organism It is now nearly three decades that the zebrafish (Danio rerio) was first proposed as a model organism to study the genetics of embryonic development and behaviour in vertebrates [1]. Interestingly, already George Streisinger, the founding father of the zebrafish field, had a keen interest in visual behaviour. The unpublished thesis of one of his students contains the outline of most behavioural assays still in use today [2]. Zebrafish research has come a long way since. While in the first two decades experimental approaches were often hampered by a lack of tools or techniques, recent years have seen a tremendous gathering of momentum in zebrafish research. A number of new developments have narrowed the gap to the types of genetic and molecular approaches that can be undertaken with more established model organisms, such as Drosophila or mice. In this review, we will highlight recent key advances that have changed or are likely to soon change zebrafish vision research in a major way. The zebrafish is a vertebrate which has many of the advantageous features normally only found in invertebrate model organisms (reviewed in, e.g. [3]). For instance, it produces a steady supply of large numbers of offspring. Due to its external fertilisation and embryonic development, zebrafish females do not need to be killed in order to obtain embryos. Thus, they can be mated repeatedly with males of the same or different genetic backgrounds to

Neurogenetics

produce several generations of offspring. This is particularly advantageous when working with lines harbouring multiple genetic alterations that are difficult to generate and where individual animals are hence precious. One zebrafish female can lay around 200 eggs each week, which greatly facilitates large-scale approaches, such as drug or mutagenesis screens aimed at discovering genes involved in various aspects of vertebrate development, behaviour or physiology. The fact that zebrafish larvae have a working visual system after 4 days of development (see below) is a significant advantage as it greatly speeds up protocols for discovering genes with roles in the development of the eye and visual functions. Another important strength of this model organism is that zebrafish embryos and larvae are optically translucent— pigmentation of skin melanocytes and the retinal pigment epithelium can be prevented by keeping the developing embryos in 1-phenyl-2-thiourea (PTU), a chemical that blocks melanin synthesis (see [4] for a protocol), or circumvented by using mutants which lack retinal pigment epithelium (RPE) pigmentation, such as the albino mutants (see [5] for references). The transparency of zebrafish embryos and larval stages allows observing the development of internal organs in living animals. Furthermore, by using fluorescent reporters, it permits monitoring gene expression in any region of the developing embryo and in real time. Moreover, zebrafish embryos and larvae are comparatively small. This has three key advantages: Firstly, entire embryos or larvae can be placed under light microscopes and imaged live at high resolution, allowing not only the three-dimensional imaging of entire animals or organs over time [6] but also the visualisation of subcellular details, such as migrating tips of axons during axonal pathfinding, e.g. in the retinotectal projection [7, 8]. Secondly, the fact that the organs in zebrafish are composed of few cells makes it easier to visualise entire organs in vivo. In this context, the newly developed Brainbow technique [9], which allows the simultaneous labelling of large numbers of cells with individual colours (see below), is particularly promising. When applied to the zebrafish brain or neural retina, this technique might, for instance, allow the reconstruction of the neuronal network in an entire eye. Thirdly, their small size also greatly facilitates automated analyses. Zebrafish embryos and larvae fit, for example, into 96-well microtitre plates, allowing their use in largescale screens, such as small compound screens used in drug discovery. Live analyses of embryogenesis are further aided by the speed with which zebrafish develop [10]. Only 24 h postfertilisation (hpf), the anlagen of all major organs, such as the brain and the eyes, are present; on the third day, the larvae hatch and after just 5 days (5 days post-fertilisation, dpf) the fish are able to fend for themselves, swimming,

hunting for food and with all sensory systems, including vision [11, 12], fully functional (see also below). In this context, it is important to point out that zebrafish, unlike mice, are diurnal, have rich colour vision and are a highly visually oriented species whose foraging behaviour depends substantially on functional vision. The zebrafish genome project, which has now reached an approximately sevenfold coverage (see http://www. ensembl.org/Danio_rerio/Info/Index), has made the zebrafish one of the genetically best characterised animal species. The annotated sequence of the zebrafish genome is an important resource for several applications, including the mapping and cloning of mutations from mutagenesis screens, the identification of predicted genes for a reverse genetics approach or evolutionary studies on genome organisation or gene structure. With regard to the latter aspect, it is interesting to note that the zebrafish, as other teleosts, has gone through an additional whole genome duplication following the two rounds of genome duplication at the base of the vertebrate lineage [13]. Approximately 12–24% of these duplicated genes are retained in the genome [14–16], resulting in a significant proportion of mammalian genes having two paralogs in fish [17]. Such paralogous genes may complicate phenotypic analyses when their separation is not complete, resulting in functional redundancy for a given process. Single mutants or knockdowns (see below) may then not show a phenotype. However, paralogous genes can also be advantageous when an essential function early during embryonic development and later function(s) of the mammalian ortholg are separated. In such cases, there is no need to generate inducible knockout models to study late gene functions. There are several examples of such a subfunctionalisation where the ancestral function is distributed to two paralogs [18]. One is the CRALBP gene which is expressed in Müller glia cells and RPE cell in mammals, while in zebrafish, one paralog is only expressed in Müller glia cells and the other only in the RPE [19, 20]. This allows a separate analysis of the functions of the mammalian ortholog of this gene, similar to a cell type-specific conditional knockout. The zebrafish visual system The visual system of the zebrafish develops extraordinarily fast, yielding a functional visual system by about 5 dpf. This is the stage when the yolk deposit is depleted and vision is employed for prey capture. The morphogenesis and structure of the eye are generally conserved among vertebrates, and the zebrafish is no exception (Fig. 1). A solid cell mass evaginates bilaterally from the diencephalon forming the optic lobes at around 10 hpf [21]. These neuroectodermal cells later give rise to the neural retina, the

Neurogenetics

Fig. 1 Morphology of the visual system of a 5-day-old zebrafish larva. GCL ganglion cell layer, INL inner nuclear layer, ONL outer nuclear layer, ON optic nerve, OC optic chiasm. Scale bar, 100 μm

pigment epithelium and the optic stalk. The lens placode is induced at about 16 hpf through contact between the evaginating optic vesicles and the surface ectoderm [22]. Later on, lens placodal cells delaminate from the surface ectoderm to form a spherical mass of cells which will give rise to the characteristic lens fibre cells. Recently, the embryonic development of the zebrafish lens, as well as its adult morphology, has been described in detail [23]. Within the neural retina, ganglion cells are the first cell type to differentiate at around 28 hpf in the ventronasal patch from which differentiation proceeds in a wavelike fashion [24, 25]. Subsequent to initial differentiation, ganglion cells contact higher processing centres via their axonal projections [21, 26] at around 40 hpf. Development of the retina progresses in a stereotyped manner, with rod photoreceptors and Müller glia being the last cells to differentiate [26, 27]. Synaptic maturation of photoreceptor cells starts at 65 hpf after the formation of functional ribbon triads, a characteristic presynaptic structure of the photoreceptor ribbon synapse. However, visual information is not processed until the ribbon synapses of bipolar cells reach maturity at 70 hpf [28]. This rapid morphological maturation of the zebrafish eye is reflected in the early appearance of visual function. Already at 3 dpf, the startle response, an abrupt movement of the larvae, can be elicited by sudden changes in illumination [11, 29]. Shortly thereafter, larval eye movements can be evoked by a surrounding motion stimulus [11, 12]. This optokinetic response (OKR) can be reliably evoked at 5 dpf, shortly before the optomotor response (OMR)—i.e. larvae swimming in the direction of a perceived motion stimulus—is manifested [30, 31]. The emergence of visual behaviour in zebrafish is paralleled by

electrical responses of the retina to light, measured in electroretinograms (ERGs) [32]. Importantly, the contribution of rod photoreceptors to the ERG becomes only apparent starting at 15 dpf [33, 34], indicating that the larval retina is strongly cone-dominant and different from the rod dominated retina of, for instance, mice. In 5-day-old zebrafish larvae, the optic axons project to ten distinct regions in the brain [35]. The most prominent retinofugal target is the optic tectum, a dorsal midbrain structure homologous to the mammalian superior colliculus. The optic tectum is a multilayered structure with four major retinorecipient layers. It likely constitutes the major visual processing centre of the fish brain and has been shown to be essential to coordinate prey capture movements [36]. Despite its name, the zebrafish optic tectum is a multimodal processing centre receiving multiple sensory inputs. Regeneration in the zebrafish eye In some sense, the development of the zebrafish retina is never fully completed since even in the adult retina a circumferential germinal zone of stem cell-like cells remains in the ciliary margin. This zone adds new neurons to the retina throughout the lifetime of the fish [37]. With the exception of rod photoreceptors, all retinal cell types originate from this stem cell niche during zebrafish development. Rod photoreceptors arise from special precursor cells which reside in the inner nuclear layer of the retina and differentiate whilst migrating to the outer retina [38, 39]. Lesions of the zebrafish retina—either via pharmacological, physical or light-induced insults—trigger rapid regeneration of the retina and induce mitotic activity in the inner nuclear layer. In the case of light-induced photoreceptor lesions, inner retinal cells migrate to the outer nuclear layer where they continue dividing and ultimately differentiate into cone photoreceptors [40]. Exclusive damage to the inner retina similarly leads to rapid regeneration of the damaged cell types [41]. In both cases, Müller glia-associated cells of the retina are the source of regeneration [39, 42–44]. They reenter the cell cycle upon damage and, as multipotent stem cells, are able to produce all retinal cell types [41, 44]. It is noteworthy that in the uninjured retina, Müller glia cells constantly proliferate at a low frequency and likely function as rod precursor cells [39, 44]. The zebrafish thus represents an attractive system to understand the molecular orchestration of this retinal regeneration [45–49]. As the retina, the zebrafish lens continues to grow and regenerate throughout the fish’s lifetime. With the recent information about the zebrafish lens proteome [50] and the superb imaging technology available [51, 52], the stage is

Neurogenetics

set to study lens development and differentiation in the zebrafish at molecular resolution.

Genetic analyses in zebrafish Gene targeting and disruption through homologous recombination have been proven enormously useful for our understanding of gene function in development and adult physiology. In zebrafish, homologous recombination was possible in cultured embryonic stem cells [53]. These cells, however, do not successfully contribute to the germline. Nevertheless, recent developments indicate that the gap between mouse and zebrafish genetics may narrow substantially in the near future. Given its many embryological advantages (see above), the zebrafish is therefore set to become the organism of choice in areas where sophisticated genetics is combined with other approaches, such as highresolution in vivo imaging or large-scale screening. This may be a particular advantage for studies aimed at uncovering the genetics of vertebrate eye development and differentiation, studying stem and progenitor cell populations and regeneration processes in the eye, analysing neuronal connectivity in the retina, or testing drugs for ocular diseases with a strong genetic component. Forward genetic approaches Mutagenesis screens have a long tradition in dissecting the genetic underpinnings of developmental pathways. In such a forward genetic approach, mutations are randomly introduced into the genome and mutants are isolated via their phenotypes (Fig. 2). This approach is unbiased and the zebrafish was the first vertebrate where large-scale screening efforts were initiated on chemically mutagenised zebrafish [54, 55]. Alternatively, zebrafish mutants have also been generated through the insertion of retroviral DNA into the genome [56–58]. Since in this type of mutagenesis the mutated loci are tagged by the inserted viral sequence, subsequent identification of the mutation causing the phenotype is greatly facilitated. This offsets the comparatively low mutation rates achieved with this method [59]. Chemical mutagenesis in contrast achieves much higher mutation rates, but subsequent mapping and positional cloning of the mutated locus can be laborious and time-consuming. Chemical mutagenesis screens The original screens focused on morphological, developmental phenotypes that could easily be observed through a dissection microscope. This approach was later complemented by an increasing number of more specific assays. Mainly, the screening for vision mutants has been greatly

aided by behavioural assays. In particular, the optomotor response and the optokinetic response proved to be useful behavioural tools to assess visual function at larval stages [30, 60–62]. In the OMR, the larvae are placed in a Plexiglas lane sitting on top of a computer monitor pointing towards the bottom of the lane. Projection of moving stripes causes sighted larvae to aggregate in one end of the lane following the apparent movement [30, 31, 63]. The OKR is another behaviour evoked by large-scale motion in the surround, e.g. by placing immobilised larvae inside a rotating drum fitted with black vertical stripes or by projecting such stripes [2, 30, 61, 64]. Since both behaviours depend on a motor output, care must be taken to distinguish defects in afferent and efferent pathways, for instance by taking spontaneous motor activity into account [62, 65]. A visual motor behavioural assay (VMR) has been described monitoring larval locomotor responses to changes in illumination [66]. Upon light on and offset, zebrafish larvae show a fast increase in motor activity which returns to normal under constant conditions. Using 96-well plates and automated video-tracking of motor activity, highthroughput screening is easily feasible. Screens exploiting OMR and OKR were very successful in identifying a range of mutants affecting various aspects of vision ranging from photoreceptor degeneration to specific retinal ganglion cell targeting defects [7, 30, 61, 62, 67, 68]. Interestingly, the overlap of identified mutants in independent screens proved to be minimal, indicating that many more genes with functions in eye development and/or visual function are waiting to be isolated. Simple reflex behaviours are robust and quick to assess, which allows for their use in large-scale screens. However, they may not sufficiently probe more complex visual processes since they are largely independent of the optic tectum [69], which, as the major optically innervated brain structure, is expected to play an important role in complex visual processing. One recently developed assay comprises prey capture, which is a more complex visual behaviour that strongly relies on the optic tectum [2, 36]. It can be assessed by monitoring a paramecia culture in the presence of zebrafish larvae using high-speed camera recordings. The decline of paramecia over time then reveals the larvae’s success in prey capture [36, 65]. Whether such prey capture behaviour can be adapted to efficient large-scale screening still needs to be demonstrated. The original forward genetic screens were based on simple visual inspection of unstained larvae. The advent of transgenic technologies now allows for an extension of this approach by scoring for the integrity of transgenically labelled specific structures in the live embryo [70]. Visualisation of distinct cellular structures by transgenic expression of fluorescent markers can substitute for time-

Neurogenetics

Fig. 2 Schematic representation of the procedure of a chemical mutagenesis in zebrafish followed by the breeding scheme of a classical phenotypic screen (a) and the identification and recovery of mutants in a TILLING approach (b). As a first step in both approaches, male fish are mutagenised with ENU—which induces mutations, amongst other cell types, in the spermatogonia—and subsequently mated with wild-type female zebrafish. The resulting fish of the F1 generation are heterozygous for individual mutations (indicated as “m” in coloured circles in this figure). a In the phenotypic screen, these F1 individuals (male and female zebrafish) are mated with wild-type zebrafish in order to obtain F2 families of zebrafish carrying a specific mutation. Following Mendel’s laws, approximately half of the zebrafish in F2 families are heterozygous for a particular mutation (+/m), whereas the other half carries two wildtype alleles of the respective locus (+/+). As the vast majority of ENUinduced mutations are recessive, the heterozygous individuals of the F1 and F2 tend not to show a phenotype. The F3 generation is generated via random matings between siblings of the F2 family. The

F3 is the first generation with individuals that are homozygous for the induced mutations—when two heterozygous individuals of the F2 have been mated—and hence the first generation in which the mutant phenotype of a recessive mutation can be observed (in approximately a quarter of the fish). b In TILLING, the fish of the F1 generation— which carry individual mutations in a heterozygous state—are screened for the presence of mutations in genes of interest. This is generally done by extracting genomic DNA from biopsies, e.g. fin clips, which is sequenced to identify individuals harbouring mutations in (known) genes of interest. As this step of the procedure can be automated, it is possible to screen large numbers of mutagenised genomes. Once an individual with a mutation in a gene of interest has been identified, this fish is outcrossed to a wild-type zebrafish to produce an F2 family carrying this mutation. As in a classical phenotypic screen, incrossing of siblings from F2 families results in F3 fish that are homozygous for the mutation and can be analysed phenotypically. Figure adapted from [173]

consuming histological analysis and speed up screening procedure, as well as allowing in vivo analyses over time.

insertional screen is based on Moloney murine leukaemia virus pseudotyped with the envelope glycoprotein from vesicular stomatitis virus [56, 71, 72]. The virus is microinjected into zebrafish embryos at the blastula stage to enable insertion into primordial germ cells. For unknown reasons, pseudotyped leukaemia viruses have a tropism for 5′ integration in active genes [73]. This renders them highly mutagenic and therefore particularly useful for the purpose of

Insertional screens Insertional mutagenesis offers the advantage of marking the mutated loci, which greatly facilitates subsequent cloning of the affected gene. The only reported large-scale

Neurogenetics

mutagenesis. However, when viral vectors are injected into early embryos, this tropism towards active genes also tends to give rise to mutations in genes which are expressed during early developmental stages and to lead to lower mutation frequencies in those genes that are only active at later stages. Screening insertional mutants for abnormal eye morphology and visual behaviour has identified a diverse set of genes involved in visual system development and function that has previously not been isolated in chemical screens [68]. Recently, retrovirus-mediated insertional mutagenesis has been extended for gain-of-function screens [74]. Besides retroviral insertions, mutagenic potential has also been discussed for transposon insertions [75–77]. Nagayoshi et al. [56] reported mutagenic efficiency for a Tol2-based enhancer trap construct that was comparable to the one of retroviral insertions. Enhancer trapping was previously used in zebrafish to identify temporal and spatial restriction of gene expression [75, 78–80]. A large set of genetic tools (see below) have been developed to modify such Tol2-based constructs and to increase their mutagenic potential. Efficient transposon-mediated insertional mutagenesis therefore offers great potential for future loss-offunction screens as the preparation of the retrotransposon is technically less demanding than the production of highly concentrated pseudotyped retrovirus. Since insertions can be rapidly mapped, the generation of a library of founder fish carrying genetically mapped insertions in all genes is within reach (reviewed in [81]). A community effort to achieve such an ambitious goal is currently discussed. Reverse genetic approaches Forward genetics approaches, i.e. mutagenesis screens, rely on the generation of mutations in random locations in the genome and the subsequent search for individuals displaying a mutant phenotype. This method has the advantage of being unbiased and requires no information on the genes that are mutated. Reverse genetics by contrast requires prior sequence knowledge of the targeted gene. Whilst this may have been a significant disadvantage in the pre-genomics era, the large numbers and high quality of sequenced genomes available today make reverse genetic approaches a method of choice to characterise the still large number of genes for which little functional information is available. In the zebrafish, three techniques are available to study the function of known genes: knockdown via morpholinos, the identification of specific mutations via TILLING and the recently established targeted mutagenesis using sequence-specific Zinc finger nucleases. Morpholino knockdown Morpholinos were the first tools developed to target specific genes in the zebrafish and have therefore been

instrumental to establishing the zebrafish as a model organism to study gene function during embryogenesis. Their biggest advantage is that they can be used to very rapidly assess the function of a given gene during early vertebrate development in vivo. They are thus often employed to test the functions of a novel gene or to confirm or rule out candidate genes for mutants isolated in a mutagenesis screen. Morpholinos are short oligonucleotides (generally 25 nt long) which, like siRNAs, can be designed to be complementary to any desired sequence [82]. In contrast to siRNAs, however, morpholinos are chemically synthesised, not transcribed, and the ribose-phosphate backbone normally present in nucleotides is replaced by morpholine rings connected via phosphorodiamidate moieties (Fig. 3a). This chemical modification renders morpholinos resistant to breakdown by nucleases. Morpholinos are typically designed to bind to a region encompassing or immediately adjacent to the START codon of the target mRNA. This presumably obstructs ribosome assembly and thus prevents the translation of the targeted mRNA into protein (Fig. 3b, left). Alternatively, morpholinos can target splice sites, resulting in incorrect splicing and/or a disruption of mRNA translation (Fig. 3b, right). Morpholinos are generally injected into the zygotes or early blastomeres [82] (Fig. 3c). They distribute in the cytoplasm and are inherited by the daughter cells after each division, preventing the expression of the target gene in the developing embryo. The fact that they are nuclease-resistant ensures that they persist over long periods of time working efficiently during the early stages of development. As the embryo develops, however, more transcripts are generated by the increased number of cells (and hence nuclei), and the morpholinos become progressively diluted until their cytoplasmic concentration drops below the threshold required to keep the target mRNA from being translated into protein. Depending on the targeted gene (e.g. its level and timing of expression) and the amount of morpholino that can be injected without causing off-target or unspecific toxic effects (zebrafish zygotes are large compared to mammalian ones, allowing comparatively large quantities to be injected), a morpholino knockdown typically lasts for between 3 and 5 days of development [83]. In zebrafish, this represents the end of embryogenesis and the time when most organs, including the eye and visual centres of the brain, have taken up their function, respectively. As a consequence, morpholinos can often be used to switch off a gene throughout the full duration of the development of an organ and even be used when trying to assess the function of a gene in early visual behaviour in zebrafish [84]. A major advantage of morpholinos over other reverse genetics techniques is that their use is very fast, simple and comparatively cheap. Designing a morpholino is no more

Neurogenetics Fig. 3 Gene knockdown by morpholino antisense nucleotide injection. a Chemical structure of part of a morpholino oligonucleotide. b Morpholinos are generally designed to bind to either the START codon (ATG), which results in a translational blockage (left), or a splice site (SS), which results in a misspliced mRNA and a defective protein (right). c Overview of the generation of knockdown phenotypes in zebrafish via injection of morpholinos. One-cell stage embryos of the F1 generation are injected with the morpholino and allowed to develop. The morpholino is typically co-injected with a dye (shown here in red) to allow monitoring of the amount injected. Depending on when the targeted gene is required during development, the injected embryos (morphants) show the knockdown phenotype

complex than designing a primer, and they can be readily obtained commercially. An experienced experimenter can inject 100–150 embryos in an hour, generating large numbers of individuals with the knockdown phenotype. This is particularly attractive when many genes are to be targeted, for example in a systematic screen aimed at analysing the functions of a large number of genes. Importantly, there is no need of breeding over several generations to obtain a mutation in a homozygous state. The injected embryos already are the generation that displays the knockdown phenotype. With the zebrafish’s rapid development, this means that it takes only hours to a few days from the beginning of the experiment (injection of the morpholino) to obtaining a phenotype. This time span compares favourably with, for example, the generation of knockout mice where, as a first step, embryonic stem cells need to be targeted, selected and injected into blastocysts (usually not less than 1 month) and at least two generations of mice need to be bred (at least 6 months) before the first homozygous embryos can be obtained. Moreover, the fact that eggs can readily be obtained in very large numbers for injections guarantees an effectively unlimited supply of embryos for analysis. This contrasts significantly with, e.g. mice where heterozygous females have to be killed to obtain the mutant embryos. Moreover, these embryos have to be genotyped prior to analysis. A further very attractive feature of using morpholinos is the possibility to simultaneously inject morpholinos against multiple genes, effectively allowing the generation of double or potentially even triple knockdowns [85]. This can be a significant advantage when having to silence genes

with redundant functions or targeting multiple pathways. As outlined above, this can be done without the need of extensive breeding in order to obtain individuals carrying multiple mutant loci in a homozygous fashion. Naturally, morpholinos can also be injected into mutant backgrounds to assess the additional loss-of-function of other genes in the mutant context. Moreover, by titrating the concentration of the morpholino that is injected it can be possible to mimic an allelic series of phenotypes of varying strength. Cells from an embryo that has been injected with morpholinos can also be transplanted into non-injected embryos (or vice versa) to generate mosaic individuals. This might be advantageous when the targeted gene is widely expressed and its knockdown is early embryonic lethal. Morphant cells can then be transplanted into an area, giving rise to non-essential organs, such as the eyes, which will allow investigating the functions of this gene throughout the development of this organ [20]. Additionally, by controlling the amount of transplanted cells and the location of transplantation, it is possible to generate mosaic organs in which morpholino-injected cells are surrounded by normal cells. Such manipulations allow addressing questions regarding the interplay of cells during development. This approach is facilitated by detailed fate maps of the zebrafish development (see, e.g. [86–88] and references therein). With all their advantages (speed, ease of use and cost) it should, however, be noted that morpholinos are still relatively crude tools to interfere with gene function. The biggest problem, as with all antisense technologies, is off target effects, i.e. the undesired down-regulation of other

Neurogenetics

genes to which the morpholino also binds. This may cause nonspecific phenotypes, unrelated to the functions of the gene of interest, which have to be controlled for. Moreover, the knockdown of a gene will never result in a phenotype as clean and reproducible as a genetic mutation. As a consequence, whilst morpholinos are ideal for studying the function of genes quickly, they are less well suited when detailed analyses of the functions of a gene are desired. In the absence of mutants identified in a mutagenesis screen, zebrafish researchers have two alternatives to disrupt specific genes: TILLING and gene targeting via Zinc finger nucleases. TILLING TILLING (from targeting-induced local lesions in genomes) can be seen as a mixture of forward and reverse genetics approaches. It relies on the random generation of mutations in the genome, e.g. via chemical mutagenesis, as is typically done in a mutagenesis screen. To identify mutations in genes of interest, this is followed by the sequencing of defined regions of these genes in large numbers of individuals (Fig. 2). If null mutations are sought, for instance, the screen may focus on a 5′ exon of the gene and search for premature STOP codons. Thus, whilst TILLING cannot target specific genes, as long as the mutation frequency is high enough and as long as sufficient numbers of individuals are screened, fish carrying a mutation in a gene of interest can be identified. In contrast to classical mutagenesis screens, for which the fish have to be bred to homozygosity (F3 generation) to be able to identify the (recessive) mutation via the mutant phenotype, this is not needed for TILLING experiments where the F1 generation can be analysed directly for the (heterozygous) presence of a mutated allele in the sequencing reaction. An important advantage of TILLING is that this technique is comparatively easy to scale up as only more sequencing reactions are needed to analyse more genes from the same mutated fish. This allows large-scale or even genome-wide approaches to be undertaken which aim at identifying mutants with lesions in defined genes. Importantly, large parts of the procedure lend themselves to automation, which favours large-scale approaches. It should, however, be noted that the comparatively large numbers of fish that have to be generated and screened in order to identify mutations in a gene of interest make this approach prohibitively expensive and time-consuming for most individual laboratories. The generation of collections of TILLING mutants is therefore commonly performed by consortia. A major drawback of TILLING is that the range of mutations that can be identified with this technique is limited. In contrast to homologous recombination—which

can be used to generate a whole range of very sophisticated mutant lines, including the knockout of endogenous genes and the expression of exogenous constructs (knock-in) in an tissue-specific, inducible or even switchable manner or the expression of specific mutations—TILLING mostly results in loss-of-function or hypomorphic alleles that compromise the function of the endogenous gene in cell types that express the gene. Another, albeit generally minor, disadvantage of TILLING is that the random nature of the mutagenesis can result in additional mutations outside of the gene of interest. Whilst most of these mutations will be silent and/or lost when the fish are outbred, it cannot be excluded that observed phenotypes may, at least in part, be caused by additional mutations in genomic regions unrelated to the gene under investigation. The generation of hypomorphic alleles or indeed allelic series via TILLING is, however, also an advantage. Numerous of the most instructive mutants in other species, such as those with behavioural phenotypes in Drosophila, are hypomorphs that could not easily have been identified in approaches relying on homologous recombination. Gene targeting via ZFNs Site-directed mutagenesis via synthetic zinc finger nucleases (ZFNs) represents the latest addition to the repertoire of reverse genetic techniques available for zebrafish, and it is the most powerful in terms of the specificity and sophistication of the genetic manipulations that can be undertaken. In contrast to morpholinos, which act only transiently and are restricted to early developmental stages, ZFN-induced mutations have been shown to be transmitted to subsequent generations. Moreover, they can be designed to specifically target a unique sequence in the genome. Thus, in contrast to TILLING which necessitates a time-consuming screening for (randomly induced) mutations in a gene of interest, ZFNs enable the generation of defined genetic changes in a given locus. Similar to the homologous recombination-based modification of genes, ZFNs can be used, in principle, to generate both “knockout” as well as “knock-in” models. This is a tremendous advance for zebrafish research as it would, for the first time, allow genetic manipulations of unprecedented precision and sophistication to be carried out. Clearly, this method allows the targeting of not only protein-encoding genes but also of any other sequence in the genome, including regions harbouring microRNAs, non-coding RNAs or regulatory sequences. ZFNs are synthetic proteins generated by fusing DNAbinding zinc finger domains to a nonspecific endonuclease domain (reviewed in [89]). Zinc finger domains are naturally occurring protein domains which recognise 3-bp target sequences in DNA. To achieve specific binding to

Neurogenetics

longer—and hence unique—sequences in the genome, ZFNs can be engineered to contain more than one zinc finger domain, each recognising a different 3-bp motif. Using four zinc finger domains in tandem, for instance, will allow the specific recognition of a sequence motif which is 12 bp long. Importantly, ZFNs are only active when bound to DNA as dimers. After binding of two ZFNs to a target site, the nuclease domains dimerise and cleave the DNA backbone, leading to a double-strand break. Such breaks are detected and repaired by the cell’s endogenous double-strand break pathways, i.e. either via non-homologous end-joining (NHEJ) or homologous recombination. The joining of two ends of DNA via NHEJ is often imprecise and can result in mutations in the targeted gene, such as frameshift mutations, which abrogate gene function. This type of repair mechanism can thus lead to a null phenotype (“knockout”). Alternatively, by supplying an excess of exogenous DNA with ends matching the targeted site, the homologous recombination machinery can incorporate these sequences at the desired locus, resulting in a knock-in. The latter approach has so far not been realised in zebrafish (see below). In zebrafish, ZFNs have recently been used to generate mutations in both somatic cells and in the germline by inducing NHEJ [90, 91] (reviewed in [90, 92]). To achieve this, ZFNs were introduced by injecting the respective ZFN-encoding mRNAs into one-cell stage embryos (see also [93] for a protocol). The fish derived from these mutagenised embryos typically carry small insertions or deletion mutation(s) in the gene of interest, and germline transmission was assessed by genotyping their progeny. The first attempts at generating zebrafish mutants with mutations in defined genes via this method and subsequent germline transmission have yielded success rates of up to 30%. This compares very favourably with other methods used to knock out genes in other model organisms, notably homologous recombination-based approaches in mice, which typically do not yield efficiencies of over 5%. Moreover, the authors of one study reported identifying a hypomorphic mutant allele [91]. This suggests that the variable lengths and types of insertions and deletions induced by NHEJ can result in allelic series of mutant phenotypes of varying strength—an important advantage when investigating the function(s) of a gene. The application of ZFNs is currently limited by the availability of validated zinc finger combinations recognising a sequence/gene to be targeted. This situation is addressed by concerted efforts to create collections of ZFNs with characterised binding sites that would cover the entire genome (reviewed in [17, 90, 92]). Importantly, the repertoire of existing (naturally occurring) zinc finger domains can be expanded by mutating domains to generate new ones with novel DNA-binding specificities [89]. In this context, a

recent study is of interest which used the OPEN method (from oligomerized pool engineering) to generate ZFNs to target five different genes in the zebrafish genome, demonstrating that ZFNs resulting in NHEJ-induced insertion/ deletion mutations can be efficiently produced [93]. Moreover, the authors also found one or more potential target sequences for ZFNs designed via OPEN in the first three coding exons of >25,000 transcripts from genes found in zebrafish [94], indicating that genome-wide gene targeting may be achievable in this species. A concern with using ZFNs is the risk of toxic effects, for instance as a result of a cleavage of additional, nontargeted sequences (“off-target” effects) or oncogenic translocations [89]. These problems might, however, be largely circumvented by either one or a combination of approaches, including (1) using obligate heterodimer forms of the endonuclease (reviewed in [90]), (2) using ZFNs with higher numbers of zinc finger domains to increase target specificity [90, 91] (see also [17] for discussion) and (3) by determining the amount of ZFNs needed in a given context to still result in efficient targeting of the locus of interest whilst minimising undesired off-target or general cytotoxic effects. In this context, it is important to point out that already the first two studies which generated ZFN-induced mutations in zebrafish detected very low levels of off-target cleavage and found that a large number of the injected fish developed and bred normally [90, 91], indicating that the method did not have any major adverse consequences. Whilst the usefulness of ZFNs in generating targeted mutations in zebrafish genes of interest have clearly been demonstrated, the method does currently not allow the introduction of exogenous DNA into the zebrafish genome. The crucial next step in making this method significantly more powerful will therefore be to establish efficient and easy-to-use protocols that allow ZFN-induced double-strand breaks to activate the homologous recombination machinery in zebrafish and lead to the incorporation of exogenous DNA. Such ZFNmediated knock-ins would open up a wide range of possibilities to study gene function in this species. This includes inserting transgenes, such as fluorescent reporters, at specific genomic locations to ensure endogenous expression levels, substituting wild-type alleles for alleles associated with disease to mimic human conditions, inducible or switchable alleles to elucidate the functions of genes during defined developmental stages or tissue-specific alleles to assess the role of a gene in a given organ. Preliminary data appear to indicate, however, that gene targeting via homologous recombination might be more difficult to achieve in zebrafish than via NHEJ [94]. Transgenic techniques Gene ablations, either through random mutagenesis or through targeted methods described above, are a great tool

Neurogenetics

to identify new genes or new functions of known genes, including those important for the development of a functional neuronal network. To reveal the complexity of a neuronal network, it is furthermore essential to visualise single components and their connections or to manipulate cell functions. For this, different transgenic techniques are now available in zebrafish that allow efficient, targeted gene expression. DNA can, for example, simply be microinjected into zebrafish eggs. Since microinjected plasmid DNA mainly remains episomal, this gene delivery approach is primarily exploited for mosaic expression analysis. Co-injection of DNA and the I-SceI meganuclease strongly enhances the efficiency of genome integration (reviewed in [95]). To spatially restrict gene delivery, DNA can instead be injected locally and electroporated into cells [96, 97]. More recently, retroviral infection and transposonmediated transgenesis were established as alternative transgenic techniques allowing efficient gene delivery and high germline transmission (reviewed in [98–100]). Transposon-mediated transgenesis using the Tol2 system has thereby emerged as the method of choice for most applications. Different aspects of transgenic techniques have recently been reviewed [101–104]. Here, we will discuss some of the recent improvements and highlight them in the context of vision research. Targeted gene expression Although transgenic techniques now allow the insertion of any genetic construct into the zebrafish genome, it remains challenging to achieve cell-specific expression of a gene of choice. Conceptually, the most direct way is to clone specific promoter sequences of a gene endogenously expressed in the cells of choice and link the transgene to this specific promoter. However, cloning of gene-specific promoters is often time-consuming and expression patterns frequently do not completely reflect the expression of endogenous genes due to missing enhancer and silencer elements. Therefore, an automated high-throughput imaging approach has been developed to yield insights into the specificity of enhancer–promoter interactions in vertebrates and already provides a resource of core promoters for transgenic applications [105]. A recent study has exploited the increasing number of cell-type specifically expressed reporter genes and combined them with other advantages of the zebrafish, in particular its transparency and the small size of its embryos at early larval stages, to study how the neural retina is patterned [106]. In doing so, the authors could show how different members of the FGF family work together at different stages of development to orchestrate the initial patterning and subsequent rearrangement of cell types.

An alternative way of targeted transgene expression in zebrafish is to trap endogenous enhancers (enhancer trap, ET) or genes (gene trap, GT). This is achieved by the random integration of ET or GT constructs into the genome (Fig. 4). In ET and GT screens, various zebrafish lines with distinct expression patterns have been generated [104, 107– 111]. Expression patterns of such lines are, however, rarely restricted to one particular cell type, which is not surprising given the wide expression domains of most genes. To yield a higher sensitivity and flexibility of transgenic reporter constructs a, two-component expression system can be used. Indeed, most recent trapping screens made use of the Gal4/UAS system by inserting the Gal4 gene or a self-reporting Gal4/UAS construct [104, 107, 109]. Such trapped zebrafish lines can be crossed with a variety of different UAS lines, e.g. UAS:gfp, UAS:Kaede or UAS: TeTxLC, to label cells or manipulate cell functions [107]. Additionally, a self-maintaining (Kalooping) construct is available, allowing spatiotemporal genetic fate mapping [112]. The specificity of spatiotemporal transgene expression in ET or GT lines can further be enhanced through the use of three inducible expression systems that have recently been established in zebrafish. One system exploits a chemically inducible variant of Gal4. Here, the Gal4-VP16 is fused to the ligand-binding domain of the ecdysone receptor (EcR) from Bombyx mori [113]. The activity of this chimaeric Gal4 variant (GV-EcR) is induced in the presence of tebufenozide pesticide. Thus, spatially restricted transcriptional activity of GV-EcR can be complemented by the temporal regulation of Gal4-dependent reporter genes expression through tebufenozide administration. Such a spatiotemporal regulation mechanism may become essential in studies in which neuronal cell activity is manipulated (e.g. [114]). In the future, it will be promising to combine modified Gal4 variants of low toxicity [107, 112] and the EcR to further improve effective transgene expression. Alternative systems controlling Gal4 activity in a spatiotemporally restricted manner have been developed in Drosophila. The slit Gal4 expression system or the expression of the Gal4 inhibitor Gal80 are two possibilities (reviewed in [103]). Their use has yet to be reported in zebrafish. A second system that has recently been introduced for spatiotemporal transgene regulation is the LexPR/LexA expression system [115]. Transcriptional activity is mediated by the fusion protein LexPR. LexPR is composed of the DNA-binding domain of the bacterial LexA repressor, the truncated ligand-binding domain of the human progesterone receptor and the activation domain of the human NF-κB/p65 protein. In the presence of the synthetic steroid mifepristone, LexPR binds to a synthetic operator–promoter sequence harbouring LexA-binding sites and drives the expression of

Neurogenetics

Fig. 4 Mechanism of enhancer trapping and gene trapping. a, b Enhancer trap (ET) vectors encode a basal promoter (Pb) upstream of a reporter gene such as fluorescent marker gfp or transcription activator gal4 (green box). Insertion of an ET vector in proximity to an endogenous gene (blue boxes for multiple exons) (a) or into the noncoding region of an endogenous gene (b) allows expression of the reporter gene. Endogenous enhancer (en) elements thereby stimulate transcription of the reporter gene. The distance between the insertion site and endogenous enhancer can be variable (II). c Gene trap (GT)

vectors encode a reporter gene (green box) harbouring a splice acceptor site (SA, red box) at its 5′ end. Insertion of a GT vector into the intronic region of an endogenous gene (blue boxes) allows generation of chimeric transcripts. The reporter gene can be fused to the upstream sequence of an endogenous gene by exploiting the endogenous splice donor (SD) site. Since reporter expression of GT constructs depends on endogenous enhancer (en) and promoter (P) sequences, its pattern mainly reflects the expression of the trapped gene

downstream genes. Similar to the Gal4/UAS system, spatial control of gene expression is attained through the enhancer– promoter elements driving LexPR, whilst temporal control is achieved through the administration of mifepristone. LexPRdependent transcription of reporter genes is induced in zebrafish larvae within 1 h, but is switched off very slowly. Even 5 days after mifepristone withdrawal, reporter transcripts are detectable [115]. In contrast, temporal regulation of GV-EcR appears to be tighter; inactivation of GV-EcR occurs within 1 h after tebufenozide removal [113]. However, it should be noted that the LexPR and GV-EcR activities were assayed differently, making the results of these studies not directly comparable. In the future, the LexPR/LexA system can substitute for the Gal4/UAS system in enhancer trap screens to generate zebrafish lines that drive LexPR-dependent transgene expression in components of the visual circuitry. Even more flexibility in regulating transgene expression can be obtained when the Gal4/UAS or LexPR/LexA system is combined with the Cre/LoxP system. The Cre enzyme was originally isolated from bacteriophage P1; it promotes specific recombination between two loxP sites [116], each of a length of 34 bp. Depending on the loxP sites, Cre-dependent recombination can excise, invert or insert DNA sequences (Fig. 5). Sato et al. [114] exploited the Cre/loxP system to excise a loxP flanked DsRed sequence and stochastically induce GFP expression in order to label single neurons in the zebrafish tectum and study their projections. Inducibility and hence temporal control of such recombination events can be achieved through the expression of Cre recombinase under a heat shock promoter [117, 118].

Alternatively, inducible Cre protein variants have successfully been tested in zebrafish. Cre was fused to one or two mutated ligand-binding domains of the human oestrogen receptor (CreERT2 or ERT2CreERT2). Activity of this modified Cre enzyme can be induced by the administration of tamoxifen [119, 120]. Depending on the tamoxifen concentration and the level of CreERT2 protein, the recombination rates can be variable [121]. Recently, the flip-excision (FlEx) system has been incorporated to modify the irreversible Cre-mediated recombination events and allow re-inversion of DNA sequences [122].

Other techniques to study eye development and visual function in zebrafish Cell labelling Transgenic techniques greatly complement conventional methods such as tracer injection or antibody staining (Fig. 6) for the visualisation of complex neuronal networks. A major advantage of transgenic techniques is that fluorescent marker expression not only facilitates labelling of single cells within the complex visual circuitry but also allows studying neuronal connectivity or plasticity in vivo. By combining transient and stable expression of membrane-tagged fluorophores, Mumm et al. [123] revealed diverse dendritic growth and arborisation patterns for zebrafish retinal ganglion cells (RGCs) and precise targeting of synaptic strata in the inner plexiform layer. Another study described horizontal precursor cells undergoing final non-apical symmetric cell division and pro-

Neurogenetics

Fig. 5 Cre/loxP system. Cre recombinase-regulated gene expression necessitates transgenic constructs harbouring a DNA sequence of interest (seq) flanked by loxP sites (black triangle) on either side. a In cells expressing Cre recombinase, such floxed DNA elements are excised if loxP sites are oriented in parallel (head-to-tail). b If loxP sites are anti-parallel-oriented (head-to-head), Cre-mediated recombination results in the inversion of the floxed DNA element. c Sitespecific gene integration exploits target zebrafish lines harbouring a

single genomic lox site of modified sequence (striped triangle). Knock-in constructs, encoding the DNA sequence of interest (red cassette), and a second lox site, are integrated into the target site in presence of Cre recombinase (for details, see [174]). Cre recombinase can be delivered by microinjection of Cre mRNA into zebrafish eggs. Alternatively, Cre enzyme can be expressed as transgene which allows spatiotemporal restriction of Cre-mediated recombination events

posed a mechanism of neuronal layer formation through a lamina-specific mitosis facilitating rapid contact between synaptic partners [124]. Furthermore, visualisation of dynamic developmental processes during wiring of a neuronal circuit became recently accessible through the development of fluorescent pre- and postsynaptic markers [125, 126]. Synaptophysin:GFP expression under the Brn3C promoter was used to visualise synapse formation of RGC in the zebrafish tectum [125]. Similarly, fusion protein PSD-95:GFP expressed in tectal

neurons localises postsynaptically in dendritic arbours. Long-time in vivo imaging revealed that synapse formation stabilises filopodia and directs dendrite arborisation [126]. These studies nicely outline the potential of specific transgenic markers to visualise highly dynamic events during neural circuit formation in the zebrafish visual system. Stochastic expression of fluorophores by DNA microinjection is a valuable tool to visualise single cells and their connections within a dense neuronal network. Within the zebrafish visual circuit, individual neurons can also be

Fig. 6 Visualisation of neuronal structures in 3-day-old zebrafish larvae. a Antibody staining against acetylated Tubulin (green) visualises neuronal projections. Additionally, quantum dots QD605 (red) were intracardially injected to label the vasculature of the larva (picture adapted from [175]). b Projections of tectal neurons to the neuropil were stochastically labelled by transient transgenic expression of the unc76:EGFP-fusion protein (green). The egfp was fused to unc76 for membrane localisation of the fusion product and expressed under the α-tubulin promoter. Transcription was mediated using the

Gal4/UAS system. Transgenic constructs were introduced by microinjection of circular plasmid DNA into zebrafish eggs. To visualise the overall morphology, the larva was counterstained with Bodipy Ceramide (red), which labels extracellular matrix and cell membranes (for details, see [176]). cb cerebellum, np neuropil, ot optic tectum, rh rhombencephalon. Images courtesy of Dr. Reinhard Köster, Neuroimaging Group, Institute of Developmental Genetics, German Research Center for Environmental Health, Munich

Neurogenetics

visualised by exploiting the variegated expression of GFP in the Brn3c:Gal4, UAS:GFP (BGUG) transgenic line [104, 127]. Recently, this transgenic line has successfully been applied to describe the cellular architecture of the larval zebrafish optic tectum [128]. An alternative approach to label single cells or whole groups of neurons exploits photoconvertible fluorophores [104, 129]. The Kaede protein and the Dendra protein are two examples of such fluorophores that can be transgenically expressed in zebrafish neurons. Their fluorescence can be stably converted from green to red upon illumination with UV and blue light, respectively [130, 131]. Since these proteins diffuse easily throughout the cell, they allow tracing of neuronal processes as axons and dendrites. Photoconvertion of Kaede within groups of neurons is already feasible using an epifluorescent microscope with a standard DAPI filter set. For the conversion of Kaede within individual neurons, confocal microscopes are instead needed to spatially restrict the UV laser light. However, the photoconvertion of fluorophores within individual neurons of deep and densely packed cell layers is challenging [128] and demands for more sophisticated setups such as the two-photon microscope. Recent advances in multicolour labelling of zebrafish neuronal circuits have been achieved through the use of stochastic Cre/LoxP recombination [114] and the combination of the Cre/LoxP and FlEx system [122]. Both studies opened the door for a more advanced technique that has recently been described in mouse [9]. The Brainbow technology exploits Cre-mediated stochastic reshuffling of tandem-arranged fluorescent marker genes to colour-tag individual cells within a neuronal network. Current Brainbow constructs encode up to four different fluorescent markers. Multiple pairs of lox sites in these constructs allow Cremediated excision and/or inversion to position individual marker genes for transcription. In the presence of multiple transgene copies, different fluorescent markers are coexpressed within one cell. Thereby, a variety of different colours are generated and each cell is labelled individually according to its unique set of fluorescent markers. Applying the Brainbow technology in zebrafish will in future ease the visualisation of individual neurons and in the long run might allow the reconstruction of visual circuits. Cell ablation Besides the anatomical description of the visual circuitry, transgenic technologies also contribute substantially to the functional characterization of this complex neuronal network. Transgenic zebrafish lines have been used to visualise cell structures and guide laser ablation of specific cells. For instance, ablation of Müller glia cells in the larval retina revealed that glial contact is not required for

photoreceptor circuit formation and immediate stabilisation [132]. Ablation of retinotectal projections in transgenic zebrafish larvae destroyed tectum function and revealed that OKR and OMR behaviour does not depend on the tectum [69]. Orientating movements during prey capture, in contrast, strongly rely on tectum function [36]. Recently, cell ablation for functional analysis has become easily feasible through the use of bacterial nitroreductase (NTR). Exploiting the transgenic tools described above, NTR can be expressed cell type specifically, e.g. in rod photoreceptors [133] or in a subset of bipolar cells [134]. After administration of metronidazole (Met), the prodrug Met is converted in NTR-expressing cells into a cytotoxic agent causing DNA damage and apoptosis [135]. Using NRT-mediated cell ablation, Montgomery et al. [133] confirmed that rod photoreceptor cell death can induce Müller glial proliferation, though rod cell death must be sufficiently high and acute. This study nicely complements previous regeneration studies using light damage [40, 47], ouabain injection [41, 136] or surgical injury [43] to induce cell death. Cell ablation through NTR-induced apoptosis is straightforward and highly reproducible also in a large-scale approach. Cells can be targeted specifically through the restricted expression of NTR and the time point of ablation can be controlled through the administration of Met. This technique will allow studying cell death and regeneration dynamics in more detail. To visualise cellular behaviour during apoptosis, a fusion protein composed of the NRT and a fluorescent reporter can be expressed. Such NTR fusion proteins get homogeneously distributed throughout the cytoplasm, thereby labelling the entire cell. Alternatively, the NTR-encoding nfsB gene and a reporter gene can be linked via the viral 2A peptide sequence. An inefficient peptide bond formation within the 2A peptide allows stochastic translation of NTR and the reporter without fusion [137]. Another approach involves the Gal4/UAS system. If NTR is expressed in a Gal4-dependent manner, additional UAS-linked transgenes can be co-expressed. By exploiting both trans activation through Gal4 or inefficient peptide bond formation, reporters with distinct cellular localisations can be expressed to visualise defined cell structures as the cell nucleus or the neural synapse. Having a tool in hand that allows highly specific cell ablation, the visual circuitry can now be dissected and distinct cellular functions within the network can be revealed. Such cell ablation studies can be supplemented through an alternative approach in zebrafish. Instead of killing a cell, its function can be manipulated (reviewed in [138, 139]). For instance, the neurosecretory function of a neuron can be blocked through the transgenic expression of tetanus toxin light chain (TeTxLC) or dominant negative vesicle-associated protein (dnVAMP) [107, 140, 141]. Such

Neurogenetics

suppression of neurotransmitter release in RGCs during circuit formation has already revealed the importance of synaptic activity for filopodia stabilisation and axon arborisation of RGCs in the zebrafish tectum [140, 142]. Blocking neuronal activity through TeTxLC or dnVAMP is convenient to study the wiring of a developing neuronal circuit. To apply this method on an already existing neuronal network, TeTxLC or dnVAMP expression has to be tightly controlled in a spatiotemporal manner to avoid unspecific effects. Potential side effects through leaky expression can be circumvented using optogenetic tools instead to modulate neuronal activity. A transgenically expressed, modified ionotropic glutamate receptor, LiGluR, depolarizes neurons only upon interaction with synthetic maleimide azobenzene glutamate and activation through UV light; blue light can switch off LiGluR function [143]. Since LiGluR retained the ability to be activated by free glutamate, its application in glutamatergic synapses of the visual circuitry is of great interest. In zebrafish, this optogenetic control system has already been successfully used to manipulate zebrafish escape response to a touch stimulus [144]. Alternatively to LiGluR, the light-activated chloride pump halorhodopsin (NpHR) can be used to silence neuronal activity [145, 146]. NpHR has recently been used in combination with light-gated cation channel channelrhodopsin (ChR2) to manipulate the activity of hindbrain neurons [147, 148]. One drawback of NpHR and ChR2 is that their activation relies on strong pulses of yellow and blue light, respectively. Efficient NpHR or ChR2 stimulation in neurons of deep cell layers might therefore be challenging. Moreover, ChR2-evoked currents are not as large as LiGluR-generated currents and less stable. Thus, LiGluR is currently the most effective means to manipulate neuronal activity and study visual function.

Recently, neuronal activity was imaged even in freeswimming zebrafish larvae [153]. Fluorescent indicators of neuronal activity can be applied as synthetic dyes (through superfusion or microinjection) or they can be encoded genetically. Most commonly used are synthetic, voltage- or calcium-sensitive dyes. The advantage of Ca2+ indicators over voltage-sensitive dyes is their brighter fluorescence and lower phototoxicity. Using such a synthetic Ca2+ indicator, population imaging of the larval zebrafish tectum revealed that direction and size selectivity of tectal neurons is established right after RGCs start to arborise in the tectum [154]. Although synthetic dyes have a favourable signal-to-noise ratio, they cannot easily be targeted to specific neuronal cell types. This drawback can be overcome by the use of genetically encoded indicators [150]. Calmodulin-based (e.g. cameleon or GCaMP3) or troponin-based (TN-XXL) indicators are sensitive to changes in cellular Ca2+ concentrations [155–158]. Colmeleon, in contrast, may serve as a chloride indicator to monitor synaptic inhibition [159]. The important advantage of genetically encoded indicators is that their expression can be targeted to specific cell types by using appropriate regulatory elements. When combined with the Gal4/UAS or LexPR/LexA system, their expression can be controlled in a spatiotemporal manner. In addition, Ca2+ or Cl− indicators can be co-expressed with additional fluorescent proteins to facilitate anatomical description of the imaged neurons and reconstruction of their connectivity. The Ca2+ indicator cameleon has already been used to monitor motoneuron and spinal interneuron activity in behaving larval zebrafish [160], but none of the genetically encoded indicators have yet been used to study the visual circuitry of zebrafish.

Activity imaging

Zebrafish models of human ocular diseases and visual impairment

Characterising the properties of a cell within the visual circuitry is essential to understand how visual information is integrated and processed. Electrophysiological methods, such as ERG or single-cell recording, have recently been supplemented by fluorescent techniques. Fluorescent indicators are used to monitor sudden changes in cellular ion concentrations, membrane potentials or neurotransmitter release by changing their fluorescent emission (reviewed in [149–151]). Whilst conventional microelectrode recordings have the advantage that fast neuronal kinetics can be revealed, fluorescent indicators allow concurrent monitoring of groups of neurons within a neural ensemble. Their use also provides the opportunity to record neuronal activity in animals that are awake and to link neuronal activity with behavioural functions (reviewed in [152]).

With the innovations described above, the zebrafish has become a valuable model organism to study human eye diseases. Its retina has comparable morphological and physiological properties to the human retina. In particular, its cone dominance closely mimics the functional cone dominance of the human visual system. This is of importance when studying, for instance, macular degeneration, a disease which affects the cone photoreceptors and is hence difficult to model in mice. Human genetic diseases can be mimicked in zebrafish relative cheaply and rapidly through morpholino-mediated gene knockdown or overexpression of dominant negative gene variants. Moreover, transplantation studies can help circumvent early embryonic lethal phenotypes or to reveal cell autonomous gene functions. In addition to morphant

Neurogenetics

analysis, several mutant lines with visual defects were isolated in mutagenesis screens that might serve as models for human diseases. Retinal mutants Inherited blindness is most commonly caused by defects in photoreceptor function resulting in photoreceptor degeneration. Several zebrafish mutant lines with a photoreceptor degeneration phenotype have been isolated (reviewed in [161]). Analysis of such mutants has already contributed significantly to our understanding of degeneration mechanisms. For instance, pde6cw59 mutants, which harbour a mutation in the cone phosphodiesterase resulting in a disrupted cone phototransduction, revealed a mechanism for secondary photoreceptor degeneration triggered by a reduced cell density in consequence of primary cone degeneration [162, 163]. pde6cw59 mutants might be used to screen for potential drugs preventing photoreceptor degeneration. Another example constitutes the analysis of different intraflagellar transport (IFT) mutants [164, 165]. Mutant analysis implicated the importance of the IFT complex B in the maintenance of photoreceptor cell outer segments and gave functional evidence for distinct ift genes in causation of retinal diseases such as retinitis pigmentosa. Besides photoreceptor-specific functions, analyses of zebrafish mutants helped associate deficiencies in RPE function with photoreceptor degeneration. The zebrafish rep1 mutant, for example, linked photoreceptor degeneration observed in human choroideremia to defects in the RPE [166]. Hence, rep1 mutant analysis not only provided new insights into the pathology of this disease but also pointed out new strategies for future therapies. Similar to rep1, mutations found in vamp6 [167], silva [83] or in different components of the vacuolar ATPase complex [168] cause defects in RPE functions that result in an altered photoreceptor morphology and impaired vision. None of these genes have yet been found to co-segregate with known disease genes. The phenotypes of these mutants, however, mimic characteristics of human diseases. Hence, these genes might be candidates for unknown genes underlying human diseases. Besides congenital blindness, zebrafish was shown to model various retinal diseases. Recently, vhl mutants, carrying a mutation in the von Hippel–Lindau tumour suppressor gene, were for instance described as a model for vascular retinopathies [169]. Additionally, retinopathy can be induced in adult animals exposed to hypoxic conditions [170]. By combining mutant analysis with transgenic methods, these studies outline the potential of zebrafish to study the pathogenesis of a disease and to screen for novel pharmacological treatments.

Mutants with defects downstream of the retina Zebrafish mutants with retinotopic projection defects provide insight into axon guidance and targeting mechanisms as well as synaptic specificity. Moreover, such mutants help assign functions to distinct circuits within the visual system [7, 62]. Interestingly, recent analyses associated zebrafish projection mutants with human diseases. The zebrafish dragnet/ col4a5 mutant has been isolated due to patterning defects of tectal layers [70, 127]. In humans, Col4a5 mutations cause Alport syndrome, which includes defects in the kidney, ear and lens. Defects in the central nervous system are rarely described in humans. Hence, the analysis of dragnet mutants helped associate possible neurological abnormalities with human collagen IV disorders and nicely outlined the potential of zebrafish projection mutants in revealing patterning mechanisms of the vertebrate nervous system. Similarly, the zebrafish belladonna (bel) mutant was first described as achiasmatic exhibiting a sign-reversed OKR [30, 171]. bel mutants also show strong spontaneous eye oscillations (SOs) in the absence of motion stimuli [172]. These SOs closely resemble infantile nystagmus (IN) in human patients, indicating that IN might be caused by projection defects. These findings establish bel as model for human IN.

Future perspectives The embryological advantages and the genetic tractability of the zebrafish, combined with a visual system that closely resembles the human eye, have led to this species becoming one of the main model organisms to study vertebrate eye development and disease. Moreover, the possibility to easily measure several visually evoked behaviours allows sophisticated studies of visual processing and function. With the recent development of new techniques that allow the targeting of specific genes and controlling gene expression in a spatiotemporal manner, it is now possible to dissect the genetic control of eye development and function with unprecedented detail in this species, for example by facilitating the observation of specific cell types to elucidate their physiological characteristics or contribution to the processing of visual information. The zebrafish is therefore expected to gain in importance as a model for eye research, including in our understanding of human ocular disorders and diseases. Acknowledgements The authors are very grateful to Dr. Reinhard Köster for contributing unpublished images to this article. SLR and SCFN are supported by the Swiss National Science Foundation (3100A0-117782).

Neurogenetics

References 1. Streisinger G, Walker C, Dower N, Knauber D, Singer F (1981) Production of clones of homozygous diploid zebra fish (Brachydanio rerio). Nature 291(5813):293–296 2. Clark DT (1981) Visual responses in the developing zebrafish (Brachydanio rerio). Universtiy of Oregon Press, Eugene 3. Dahm R, Geisler R (2006) Learning from small fry: the zebrafish as a genetic model organism for aquaculture fish species. Mar Biotechnol (NY) 8(4):329–345 4. Schulte-Merker S (2002) Looking at embryos. In: Nüsslein-Volhard C, Dahm R (eds) Zebrafish—a practical approach. Oxford University Press, Oxford, pp 39–58 5. Frohnhoeffer HG (2002) Table of zebrafish mutations. In: Nüsslein-Volhard C, Dahm R (eds) Zebrafish—a practical approach. Oxford University Press, Oxford, pp 237–292 6. Keller PJ, Schmidt AD, Wittbrodt J, Stelzer EH (2008) Reconstruction of zebrafish early embryonic development by scanned light sheet microscopy. Science 322(5904):1065–1069 7. Baier H, Klostermann S, Trowe T, Karlstrom RO, Nusslein-Volhard C, Bonhoeffer F (1996) Genetic dissection of the retinotectal projection. Development 123:415–425 8. Karlstrom RO, Trowe T, Klostermann S, Baier H, Brand M, Crawford AD, Grunewald B, Haffter P, Hoffmann H, Meyer SU, Muller BK, Richter S, van Eeden FJ, Nusslein-Volhard C, Bonhoeffer F (1996) Zebrafish mutations affecting retinotectal axon pathfinding. Development 123:427–438 9. Livet J, Weissman TA, Kang H, Draft RW, Lu J, Bennis RA, Sanes JR, Lichtman JW (2007) Transgenic strategies for combinatorial expression of fluorescent proteins in the nervous system. Nature 450(7166):56–62 10. Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF (1995) Stages of embryonic development of the zebrafish. Dev Dyn 203(3):253–310 11. Easter SS Jr, Nicola GN (1996) The development of vision in the zebrafish (Danio rerio). Dev Biol 180(2):646–663 12. Easter SS Jr, Nicola GN (1997) The development of eye movements in the zebrafish (Danio rerio). Dev Psychobiol 31 (4):267–276 13. Meyer A, Van de Peer Y (2005) From 2r to 3r: evidence for a fish-specific genome duplication (FSGD). Bioessays 27(9):937– 945 14. Braasch I, Brunet F, Volff JN, Schartl M (2009) Pigmentation pathway evolution after whole-genome duplication in fish. Genome Biol Evol 1:479–493 15. Kasahara M, Naruse K, Sasaki S, Nakatani Y, Qu W, Ahsan B, Yamada T, Nagayasu Y, Doi K, Kasai Y, Jindo T, Kobayashi D, Shimada A, Toyoda A, Kuroki Y, Fujiyama A, Sasaki T, Shimizu A, Asakawa S, Shimizu N, Hashimoto S, Yang J, Lee Y, Matsushima K, Sugano S, Sakaizumi M, Narita T, Ohishi K, Haga S, Ohta F, Nomoto H, Nogata K, Morishita T, Endo T, Shin IT, Takeda H, Morishita S, Kohara Y (2007) The medaka draft genome and insights into vertebrate genome evolution. Nature 447(7145):714–719 16. Sato Y, Hashiguchi Y, Nishida M (2009) Temporal pattern of loss/persistence of duplicate genes involved in signal transduction and metabolic pathways after teleost-specific genome duplication. BMC Evol Biol 9:127 17. Postlethwait JH, Woods IG, Ngo-Hazelett P, Yan YL, Kelly PD, Chu F, Huang H, Hill-Force A, Talbot WS (2000) Zebrafish comparative genomics and the origins of vertebrate chromosomes. Genome Res 10(12):1890–1902 18. Force A, Lynch M, Pickett FB, Amores A, Yan YL, Postlethwait J (1999) Preservation of duplicate genes by complementary, degenerative mutations. Genetics 151(4):1531–1545

19. Fleisch VC, Schonthaler HB, von Lintig J, Neuhauss SC (2008) Subfunctionalization of a retinoid-binding protein provides evidence for two parallel visual cycles in the cone-dominant zebrafish retina. J Neurosci 28(33):8208–8216 20. Collery R, McLoughlin S, Vendrell V, Finnegan J, Crabb JW, Saari JC, Kennedy BN (2008) Duplication and divergence of zebrafish CRALBP genes uncovers novel role for RPE- and Müller-CRALBP in cone vision. Invest Ophthalmol Vis Sci 49 (9):3812–3820 21. Schmitt EA, Dowling JE (1994) Early eye morphogenesis in the zebrafish, Brachydanio rerio. J Comp Neurol 344(4):532–542 22. Soules KA, Link BA (2005) Morphogenesis of the anterior segment in the zebrafish eye. BMC Dev Biol 5:12 23. Dahm R, Schonthaler HB, Soehn AS, van Marle J, Vrensen GF (2007) Development and adult morphology of the eye lens in the zebrafish. Exp Eye Res 85(1):74–89 24. Schmitt EA, Dowling JE (1996) Comparison of topographical patterns of ganglion and photoreceptor cell differentiation in the retina of the zebrafish, Danio rerio. J Comp Neurol 371(2):222–234 25. Neumann CJ, Nuesslein-Volhard C (2000) Patterning of the zebrafish retina by a wave of sonic hedgehog activity. Science 289(5487):2137–2139 26. Schmitt EA, Dowling JE (1999) Early retinal development in the zebrafish, Danio rerio: light and electron microscopic analyses. J Comp Neurol 404(4):515–536 27. Hu M, Easter SS (1999) Retinal neurogenesis: the formation of the initial central patch of postmitotic cells. Dev Biol 207 (2):309–321 28. Biehlmaier O, Neuhauss SC, Kohler K (2003) Synaptic plasticity and functionality at the cone terminal of the developing zebrafish retina. J Neurobiol 56(3):222–236 29. Kimmel CB, Patterson J, Kimmel RO (1974) The development and behavioral characteristics of the startle response in the zebra fish. Dev Psychobiol 7(1):47–60 30. Neuhauss SC, Biehlmaier O, Seeliger MW, Das T, Kohler K, Harris WA, Baier H (1999) Genetic disorders of vision revealed by a behavioral screen of 400 essential loci in zebrafish. J Neurosci 19(19):8603–8615 31. Orger MB, Smear MC, Anstis SM, Baier H (2000) Perception of Fourier and non-Fourier motion by larval zebrafish. Nat Neurosci 3(11):1128–1133 32. Branchek T (1984) The development of photoreceptors in the zebrafish, Brachydanio rerio: II. Function. J Comp Neurol 224 (1):116–122 33. Bilotta J, Saszik S, Sutherland SE (2001) Rod contributions to the electroretinogram of the dark-adapted developing zebrafish. Dev Dyn 222(4):564–570 34. Saszik S, Bilotta J, Givin CM (1999) ERG assessment of zebrafish retinal development. Vis Neurosci 16(5):881–888 35. Burrill JD, Easter SS Jr (1994) Development of the retinofugal projections in the embryonic and larval zebrafish (Brachydanio rerio). J Comp Neurol 346(4):583–600 36. Gahtan E, Tanger P, Baier H (2005) Visual prey capture in larval zebrafish is controlled by identified reticulospinal neurons downstream of the tectum. J Neurosci 25(40):9294–9303 37. Marcus RC, Delaney CL, Easter SS Jr (1999) Neurogenesis in the visual system of embryonic and adult zebrafish (Danio rerio). Off Vis Neurosci 16(3):417–424 38. Hitchcock PF, Raymond PA (2004) The teleost retina as a model for developmental and regeneration biology. Zebrafish 1(3):257–271 39. Raymond PA, Barthel LK, Bernardos RL, Perkowski JJ (2006) Molecular characterization of retinal stem cells and their niches in adult zebrafish. BMC Dev Biol 6:36 40. Vihtelic TS, Soverly JE, Kassen SC, Hyde DR (2006) Retinal regional differences in photoreceptor cell death and regeneration in light-lesioned albino zebrafish. Exp Eye Res 82(4):558–575

Neurogenetics 41. Fimbel SM, Montgomery JE, Burket CT, Hyde DR (2007) Regeneration of inner retinal neurons after intravitreal injection of ouabain in zebrafish. J Neurosci 27(7):1712–1724 42. Fausett BV, Goldman D (2006) A role for alpha1 tubulinexpressing Müller glia in regeneration of the injured zebrafish retina. J Neurosci 26(23):6303–6313 43. Yurco P, Cameron DA (2005) Responses of Müller glia to retinal injury in adult zebrafish. Vis Res 45(8):991–1002 44. Bernardos RL, Barthel LK, Meyers JR, Raymond PA (2007) Late-stage neuronal progenitors in the retina are radial Müller glia that function as retinal stem cells. J Neurosci 27(26):7028– 7040 45. Cameron DA, Gentile KL, Middleton FA, Yurco P (2005) Gene expression profiles of intact and regenerating zebrafish retina. Mol Vis 11:775–791 46. Fausett BV, Gumerson JD, Goldman D (2008) The proneural basic helix-loop-helix gene ascl1a is required for retina regeneration. J Neurosci 28(5):1109–1117 47. Kassen SC, Ramanan V, Montgomery JE, TB C, Liu CG, Vihtelic TS, Hyde DR (2007) Time course analysis of gene expression during light-induced photoreceptor cell death and regeneration in albino zebrafish. Dev Neurobiol 67(8):1009– 1031 48. Qin Z, Barthel LK, Raymond PA (2009) Genetic evidence for shared mechanisms of epimorphic regeneration in zebrafish. Proc Natl Acad Sci USA 106(23):9310–9315 49. Wehman AM, Staub W, Meyers JR, Raymond PA, Baier H (2005) Genetic dissection of the zebrafish retinal stem-cell compartment. Dev Biol 281(1):53–65 50. Greiling TM, Houck SA, Clark JI (2009) The zebrafish lens proteome during development and aging. Mol Vis 15:2313–2325 51. Greiling TM, Aose M, Clark JI (2010) Cell fate and differentiation of the developing ocular lens. Invest Ophthalmol Vis Sci 51(3):1540–1546 52. Greiling TM, Clark JI (2009) Early lens development in the zebrafish: a three-dimensional time-lapse analysis. Dev Dyn 238 (9):2254–2265 53. Fan L, Moon J, Crodian J, Collodi P (2006) Homologous recombination in zebrafish ES cells. Transgenic Res 15(1):21–30 54. Driever W, Solnica-Krezel L, Schier AF, Neuhauss SC, Malicki J, Stemple DL, Stainier DY, Zwartkruis F, Abdelilah S, Rangini Z, Belak J, Boggs C (1996) A genetic screen for mutations affecting embryogenesis in zebrafish. Development 123:37–46 55. Haffter P, Granato M, Brand M, Mullins MC, Hammerschmidt M, Kane DA, Odenthal J, van Eeden FJ, Jiang YJ, Heisenberg CP, Kelsh RN, Furutani-Seiki M, Vogelsang E, Beuchle D, Schach U, Fabian C, Nusslein-Volhard C (1996) The identification of genes with unique and essential functions in the development of the zebrafish, Danio rerio. Development 123:1–36 56. Amsterdam A, Burgess S, Golling G, Chen W, Sun Z, Townsend K, Farrington S, Haldi M, Hopkins N (1999) A large-scale insertional mutagenesis screen in zebrafish. Genes Dev 13 (20):2713–2724 57. Gaiano N, Allende M, Amsterdam A, Kawakami K, Hopkins N (1996) Highly efficient germ-line transmission of proviral insertions in zebrafish. Proc Natl Acad Sci USA 93(15):7777– 7782 58. Gaiano N, Amsterdam A, Kawakami K, Allende M, Becker T, Hopkins N (1996) Insertional mutagenesis and rapid cloning of essential genes in zebrafish. Nature 383(6603):829–832 59. Amsterdam A, Hopkins N (2006) Mutagenesis strategies in zebrafish for identifying genes involved in development and disease. Trends Genet 22(9):473–478 60. Brockerhoff SE, Dowling JE, Hurley JB (1998) Zebrafish retinal mutants. Vis Res 38(10):1335–1339

61. Brockerhoff SE, Hurley JB, Janssen-Bienhold U, Neuhauss SC, Driever W, Dowling JE (1995) A behavioral screen for isolating zebrafish mutants with visual system defects. Proc Natl Acad Sci USA 92(23):10545–10549 62. Muto A, Orger MB, Wehman AM, Smear MC, Kay JN, Page-McCaw PS, Gahtan E, Xiao T, Nevin LM, Gosse NJ, Staub W, Finger-Baier K, Baier H (2005) Forward genetic analysis of visual behavior in zebrafish. PLoS Genet 1(5):e66 63. Orger MB, Baier H (2005) Channeling of red and green cone inputs to the zebrafish optomotor response. Vis Neurosci 22 (3):275–281 64. Fleisch VC, Neuhauss SC (2006) Visual behavior in zebrafish. Zebrafish 3(2):191–201 65. Orger MB, Gahtan E, Muto A, Page-McCaw P, Smear MC, Baier H (2004) Behavioral screening assays in zebrafish. Meth Cell Biol 77:53–68 66. Emran F, Rihel J, Dowling JE (2008) A behavioral assay to measure responsiveness of zebrafish to changes in light intensities. J Vis Exp (20) pii:293 67. Fadool JM, Brockerhoff SE, Hyatt GA, Dowling JE (1997) Mutations affecting eye morphology in the developing zebrafish (Danio rerio). Dev Genet 20(3):288–295 68. Gross JM, Perkins BD, Amsterdam A, Egana A, Darland T, Matsui JI, Sciascia S, Hopkins N, Dowling JE (2005) Identification of zebrafish insertional mutants with defects in visual system development and function. Genetics 170(1):245–261 69. Roeser T, Baier H (2003) Visuomotor behaviors in larval zebrafish after GFP-guided laser ablation of the optic tectum. J Neurosci 23(9):3726–3734 70. Xiao T, Roeser T, Staub W, Baier H (2005) A GFP-based genetic screen reveals mutations that disrupt the architecture of the zebrafish retinotectal projection. Development 132(13):2955– 2967 71. Amsterdam A, Nissen RM, Sun Z, Swindell EC, Farrington S, Hopkins N (2004) Identification of 315 genes essential for early zebrafish development. Proc Natl Acad Sci USA 101 (35):12792–12797 72. Golling G, Amsterdam A, Sun Z, Antonelli M, Maldonado E, Chen W, Burgess S, Haldi M, Artzt K, Farrington S, Lin SY, Nissen RM, Hopkins N (2002) Insertional mutagenesis in zebrafish rapidly identifies genes essential for early vertebrate development. Nat Genet 31(2):135–140 73. Wang D, Jao LE, Zheng N, Dolan K, Ivey J, Zonies S, Wu X, Wu K, Yang H, Meng Q, Zhu Z, Zhang B, Lin S, Burgess SM (2007) Efficient genome-wide mutagenesis of zebrafish genes by retroviral insertions. Proc Natl Acad Sci USA 104(30):12428– 12433 74. Maddison LA, Lu J, Victoroff T, Scott E, Baier H, Chen W (2009) A gain-of-function screen in zebrafish identifies a guanylate cyclase with a role in neuronal degeneration. Mol Genet Genomics 281(5):551–563 75. Kawakami K, Takeda H, Kawakami N, Kobayashi M, Matsuda N, Mishina M (2004) A transposon-mediated gene trap approach identifies developmentally regulated genes in zebrafish. Dev Cell 7(1):133–144 76. Nagayoshi S, Hayashi E, Abe G, Osato N, Asakawa K, Urasaki A, Horikawa K, Ikeo K, Takeda H, Kawakami K (2008) Insertional mutagenesis by the Tol2 transposonmediated enhancer trap approach generated mutations in two developmental genes: Tcf7 and synembryn-like. Development 135(1):159–169 77. Sivasubbu S, Balciunas D, Davidson AE, Pickart MA, Hermanson SB, Wangensteen KJ, Wolbrink DC, Ekker SC (2006) Gene-breaking transposon mutagenesis reveals an essential role for histone H2afza in zebrafish larval development. Mech Dev 123(7):513–529

Neurogenetics 78. Balciunas D, Davidson AE, Sivasubbu S, Hermanson SB, Welle Z, Ekker SC (2004) Enhancer trapping in zebrafish using the sleeping beauty transposon. BMC Genomics 5(1):62 79. Ellingsen S, Laplante MA, Konig M, Kikuta H, Furmanek T, Hoivik EA, Becker TS (2005) Large-scale enhancer detection in the zebrafish genome. Development 132(17):3799– 3811 80. Parinov S, Kondrichin I, Korzh V, Emelyanov A (2004) Tol2 transposon-mediated enhancer trap to identify developmentally regulated zebrafish genes in vivo. Dev Dyn 231(2):449– 459 81. Jao LE, Maddison L, Chen W, Burgess SM (2008) Using retroviruses as a mutagenesis tool to explore the zebrafish genome. Brief Funct Genomic Proteomic 7(6):427–443 82. Bill BR, Petzold AM, Clark KJ, Schimmenti LA, Ekker SC (2009) A primer for morpholino use in zebrafish. Zebrafish 6 (1):69–77 83. Schonthaler HB, Lampert JM, von Lintig J, Schwarz H, Geisler R, Neuhauss SC (2005) A mutation in the silver gene leads to defects in melanosome biogenesis and alterations in the visual system in the zebrafish mutant fading vision. Dev Biol 284 (2):421–436 84. Rinner O, Makhankov YV, Biehlmaier O, Neuhauss SC (2005) Knockdown of cone-specific kinase GRK7 in larval zebrafish leads to impaired cone response recovery and delayed dark adaptation. Neuron 47(2):231–242 85. McNulty CL, Peres JN, Bardine N, van den Akker WM, Durston AJ (2005) Knockdown of the complete Hox paralogous group 1 leads to dramatic hindbrain and neural crest defects. Development 132(12):2861–2871 86. Russek-Blum N, Nabel-Rosen H, Levkowitz G (2009) High resolution fate map of the zebrafish diencephalon. Dev Dyn 238 (7):1827–1835 87. England SJ, Blanchard GB, Mahadevan L, Adams RJ (2006) A dynamic fate map of the forebrain shows how vertebrate eyes form and explains two causes of cyclopia. Development 133 (23):4613–4617 88. Kozlowski DJ, Murakami T, Ho RK, Weinberg ES (1997) Regional cell movement and tissue patterning in the zebrafish embryo revealed by fate mapping with caged fluorescein. Biochem Cell Biol 75(5):551–562 89. Porteus MH, Carroll D (2005) Gene targeting using zinc finger nucleases. Nat Biotechnol 23(8):967–973 90. Doyon Y, McCammon JM, Miller JC, Faraji F, Ngo C, Katibah GE, Amora R, Hocking TD, Zhang L, Rebar EJ, Gregory PD, Urnov FD, Amacher SL (2008) Heritable targeted gene disruption in zebrafish using designed zinc-finger nucleases. Nat Biotechnol 26(6):702–708 91. Meng X, Noyes MB, Zhu LJ, Lawson ND, Wolfe SA (2008) Targeted gene inactivation in zebrafish using engineered zincfinger nucleases. Nat Biotechnol 26(6):695–701 92. Ekker SC (2008) Zinc finger-based knockout punches for zebrafish genes. Zebrafish 5(2):121–123 93. Foley JE, Maeder ML, Pearlberg J, Joung JK, Peterson RT, Yeh JR (2009) Targeted mutagenesis in zebrafish using customized zinc-finger nucleases. Nat Protoc 4(12):1855–1867 94. Foley JE, Yeh JR, Maeder ML, Reyon D, Sander JD, Peterson RT, Joung JK (2009) Rapid mutation of endogenous zebrafish genes using zinc finger nucleases made by Oligomerized Pool ENgineering (OPEN). PLoS ONE 4(2):e4348 95. Grabher C, Wittbrodt J (2008) Recent advances in meganuclease- and transposon-mediated transgenesis of medaka and zebrafish. Meth Mol Biol 461:521–539 96. Hendricks M, Jesuthasan S (2007) Electroporation-based methods for in vivo, whole mount and primary culture analysis of zebrafish brain development. Neural Dev 2:6

97. Cerda GA, Thomas JE, Allende ML, Karlstrom RO, Palma V (2006) Electroporation of DNA, RNA, and morpholinos into zebrafish embryos. Methods 39(3):207–211 98. Amsterdam A, Becker TS (2005) Transgenes as screening tools to probe and manipulate the zebrafish genome. Dev Dyn 234 (2):255–268 99. Ivics Z, Kaufman CD, Zayed H, Miskey C, Walisko O, Izsvak Z (2004) The sleeping beauty transposable element: evolution, regulation and genetic applications. Curr Issues Mol Biol 6 (1):43–55 100. Kawakami K (2005) Transposon tools and methods in zebrafish. Dev Dyn 234(2):244–254 101. Asakawa K, Kawakami K (2008) Targeted gene expression by the Gal4-UAS system in zebrafish. Dev Growth Differ 50 (6):391–399 102. Baier H, Scott EK (2009) Genetic and optical targeting of neural circuits and behavior—zebrafish in the spotlight. Curr Opin Neurobiol 19(5):553–560 103. Halpern ME, Rhee J, Goll MG, Akitake CM, Parsons M, Leach SD (2008) Gal4/UAS transgenic tools and their application to zebrafish. Zebrafish 5(2):97–110 104. Scott EK, Mason L, Arrenberg AB, Ziv L, Gosse NJ, Xiao T, Chi NC, Asakawa K, Kawakami K, Baier H (2007) Targeting neural circuitry in zebrafish using Gal4 enhancer trapping. Nat Meth 4(4):323–326 105. Gehrig J, Reischl M, Kalmar E, Ferg M, Hadzhiev Y, Zaucker A, Song C, Schindler S, Liebel U, Muller F (2009) Automated highthroughput mapping of promoter–enhancer interactions in zebrafish embryos. Nat Meth 6(12):911–916 106. Picker A, Cavodeassi F, Machate A, Bernauer S, Hans S, Abe G, Kawakami K, Wilson SW, Brand M (2009) Dynamic coupling of pattern formation and morphogenesis in the developing vertebrate retina. PLoS Biol 7(10):e1000214 107. Asakawa K, Suster ML, Mizusawa K, Nagayoshi S, Kotani T, Urasaki A, Kishimoto Y, Hibi M, Kawakami K (2008) Genetic dissection of neural circuits by Tol2 transposon-mediated Gal4 gene and enhancer trapping in zebrafish. Proc Natl Acad Sci USA 105(4):1255–1260 108. Ogura E, Okuda Y, Kondoh H, Kamachi Y (2009) Adaptation of Gal4 activators for Gal4 enhancer trapping in zebrafish. Dev Dyn 238(3):641–655 109. Davison JM, Akitake CM, Goll MG, Rhee JM, Gosse N, Baier H, Halpern ME, Leach SD, Parsons MJ (2007) Transactivation from Gal4-VP16 transgenic insertions for tissue-specific cell labeling and ablation in zebrafish. Dev Biol 304(2):811–824 110. Kikuta H, Fredman D, Rinkwitz S, Lenhard B, Becker TS (2007) Retroviral enhancer detection insertions in zebrafish combined with comparative genomics reveal genomic regulatory blocks—a fundamental feature of vertebrate genomes. Genome Biol 8 (Suppl 1):S4 111. Kondrychyn I, Garcia-Lecea M, Emelyanov A, Parinov S, Korzh V (2009) Genome-wide analysis of Tol2 transposon reintegration in zebrafish. BMC Genomics 10:418 112. Distel M, Wullimann MF, Koster RW (2009) Optimized Gal4 genetics for permanent gene expression mapping in zebrafish. Proc Natl Acad Sci USA 106(32):13365–13370 113. Esengil H, Chang V, Mich JK, Chen JK (2007) Small-molecule regulation of zebrafish gene expression. Nat Chem Biol 3 (3):154–155 114. Sato T, Hamaoka T, Aizawa H, Hosoya T, Okamoto H (2007) Genetic single-cell mosaic analysis implicates ephrinB2 reverse signaling in projections from the posterior tectum to the hindbrain in zebrafish. J Neurosci 27(20):5271–5279 115. Emelyanov A, Parinov S (2008) Mifepristone-inducible LexPR system to drive and control gene expression in transgenic zebrafish. Dev Biol 320(1):113–121

Neurogenetics 116. Hamilton DL, Abremski K (1984) Site-specific recombination by the bacteriophage P1 lox-Cre system Cre-mediated synapsis of two lox sites. J Mol Biol 178(2):481–486 117. Thummel R, Burket CT, Brewer JL, Sarras MP Jr, Li L, Perry M, McDermott JP, Sauer B, Hyde DR, Godwin AR (2005) Cremediated site-specific recombination in zebrafish embryos. Dev Dyn 233(4):1366–1377 118. Hans S, Freudenreich D, Geffarth M, Kaslin J, Machate A, Brand M (2011) Generation of a non-leaky heat shock-inducible Cre line for conditional Cre/lox strategies in zebrafish. Dev Dyn 240:108–115 119. Indra AK, Warot X, Brocard J, Bornert JM, Xiao JH, Chambon P, Metzger D (1999) Temporally-controlled sitespecific mutagenesis in the basal layer of the epidermis: comparison of the recombinase activity of the tamoxifeninducible Cre-Er(T) and Cre-Er(T2) recombinases. Nucleic Acids Res 27(22):4324–4327 120. Feil R, Brocard J, Mascrez B, LeMeur M, Metzger D, Chambon P (1996) Ligand-activated site-specific recombination in mice. Proc Natl Acad Sci USA 93(20):10887–10890 121. Hans S, Kaslin J, Freudenreich D, Brand M (2009) Temporallycontrolled site-specific recombination in zebrafish. PLoS ONE 4 (2):e4640 122. Boniface EJ, Lu J, Victoroff T, Zhu M, Chen W (2009) FlEXbased transgenic reporter lines for visualization of Cre and Flp activity in live zebrafish. Genesis 47(7):484–491 123. Mumm JS, Williams PR, Godinho L, Koerber A, Pittman AJ, Roeser T, Chien CB, Baier H, Wong RO (2006) In vivo imaging reveals dendritic targeting of laminated afferents by zebrafish retinal ganglion cells. Neuron 52(4):609–621 124. Godinho L, Williams PR, Claassen Y, Provost E, Leach SD, Kamermans M, Wong RO (2007) Nonapical symmetric divisions underlie horizontal cell layer formation in the developing retina in vivo. Neuron 56(4):597–603 125. Meyer MP, Smith SJ (2006) Evidence from in vivo imaging that synaptogenesis guides the growth and branching of axonal arbors by two distinct mechanisms. J Neurosci 26(13):3604–3614 126. Niell CM, Meyer MP, Smith SJ (2004) In vivo imaging of synapse formation on a growing dendritic arbor. Nat Neurosci 7 (3):254–260 127. Xiao T, Baier H (2007) Lamina-specific axonal projections in the zebrafish tectum require the type IV collagen dragnet. Nat Neurosci 10(12):1529–1537 128. Scott EK, Baier H (2009) The cellular architecture of the larval zebrafish tectum, as revealed by Gal4 enhancer trap lines. Front Neural Circ 3:13 129. Sato T, Takahoko M, Okamoto H (2006) Huc:Kaede, a useful tool to label neural morphologies in networks in vivo. Genesis 44(3):136–142 130. Ando R, Hama H, Yamamoto-Hino M, Mizuno H, Miyawaki A (2002) An optical marker based on the UV-induced green-to-red photoconversion of a fluorescent protein. Proc Natl Acad Sci USA 99(20):12651–12656 131. Gurskaya NG, Verkhusha VV, Shcheglov AS, Staroverov DB, Chepurnykh TV, Fradkov AF, Lukyanov S, Lukyanov KA (2006) Engineering of a monomeric green-to-red photoactivatable fluorescent protein induced by blue light. Nat Biotechnol 24 (4):461–465 132. Williams PR, Suzuki SC, Yoshimatsu T, Lawrence OT, Waldron SJ, Parsons MJ, Nonet ML, Wong RO (2010) In vivo development of outer retinal synapses in the absence of glial contact. J Neurosci 30(36):11951–11961 133. Montgomery JE, Parsons MJ, Hyde DR (2009) A novel model of retinal ablation demonstrates that the extent of rod cell death regulates the origin of the regenerated zebrafish rod photoreceptors. J Comp Neurol 518(6):800–814

134. Zhao XF, Ellingsen S, Fjose A (2009) Labelling and targeted ablation of specific bipolar cell types in the zebrafish retina. BMC Neurosci 10:107 135. Curado S, Anderson RM, Jungblut B, Mumm J, Schroeter E, Stainier DY (2007) Conditional targeted cell ablation in zebrafish: a new tool for regeneration studies. Dev Dyn 236(4):1025– 1035 136. Sherpa T, Fimbel SM, Mallory DE, Maaswinkel H, Spritzer SD, Sand JA, Li L, Hyde DR, Stenkamp DL (2008) Ganglion cell regeneration following whole-retina destruction in zebrafish. Dev Neurobiol 68(2):166–181 137. Provost E, Rhee J, Leach SD (2007) Viral 2A peptides allow expression of multiple proteins from a single orf in transgenic zebrafish embryos. Genesis 45(10):625–629 138. Luo L, Callaway EM, Svoboda K (2008) Genetic dissection of neural circuits. Neuron 57(5):634–660 139. Scott EK (2009) The Gal4/UAS toolbox in zebrafish: new approaches for defining behavioral circuits. J Neurochem 110 (2):441–456 140. Hua JY, Smear MC, Baier H, Smith SJ (2005) Regulation of axon growth in vivo by activity-based competition. Nature 434 (7036):1022–1026 141. Wyart C, Del Bene F, Warp E, Scott EK, Trauner D, Baier H, Isacoff EY (2009) Optogenetic dissection of a behavioural module in the vertebrate spinal cord. Nature 461(7262):407–410 142. Ben Fredj N, Hammond S, Otsuna H, Chien CB, Burrone J, Meyer MP (2010) Synaptic activity and activity-dependent competition regulates axon arbor maturation, growth arrest, and territory in the retinotectal projection. J Neurosci 30 (32):10939–10951 143. Volgraf M, Gorostiza P, Numano R, Kramer RH, Isacoff EY, Trauner D (2006) Allosteric control of an ionotropic glutamate receptor with an optical switch. Nat Chem Biol 2(1):47–52 144. Szobota S, Gorostiza P, Del Bene F, Wyart C, Fortin DL, Kolstad KD, Tulyathan O, Volgraf M, Numano R, Aaron HL, Scott EK, Kramer RH, Flannery J, Baier H, Trauner D, Isacoff EY (2007) Remote control of neuronal activity with a light-gated glutamate receptor. Neuron 54(4):535–545 145. Gradinaru V, Thompson KR, Deisseroth K (2008) ENpHR: a Natronomonas halorhodopsin enhanced for optogenetic applications. Brain Cell Biol 36(1–4):129–139 146. Zhao S, Cunha C, Zhang F, Liu Q, Gloss B, Deisseroth K, Augustine GJ, Feng G (2008) Improved expression of halorhodopsin for light-induced silencing of neuronal activity. Brain Cell Biol 36(1–4):141–154 147. Arrenberg AB, Del Bene F, Baier H (2009) Optical control of zebrafish behavior with halorhodopsin. Proc Natl Acad Sci USA 106(42):17968–17973 148. Schoonheim PJ, Arrenberg AB, Del Bene F, Baier H (2010) Optogenetic localization and genetic perturbation of saccadegenerating neurons in zebrafish. J Neurosci 30(20):7111– 7120 149. Baker BJ, Mutoh H, Dimitrov D, Akemann W, Perron A, Iwamoto Y, Jin L, Cohen LB, Isacoff EY, Pieribone VA, Hughes T, Knopfel T (2008) Genetically encoded fluorescent sensors of membrane potential. Brain Cell Biol 36(1–4):53–67 150. Miyawaki A (2005) Innovations in the imaging of brain functions using fluorescent proteins. Neuron 48(2):189–199 151. Yuste R, Miller RB, Holthoff K, Zhang S, Miesenbock G (2000) Synapto-phluorins: chimeras between pH-sensitive mutants of green fluorescent protein and synaptic vesicle membrane proteins as reporters of neurotransmitter release. Meth Enzymol 327:522–546 152. Fetcho JR, Bhatt DH (2004) Genes and photons: new avenues into the neuronal basis of behavior. Curr Opin Neurobiol 14 (6):707–714

Neurogenetics 153. Naumann EA, Kampff AR, Prober DA, Schier AF, Engert F (2010) Monitoring neural activity with bioluminescence during natural behavior. Nat Neurosci 13(4):513–520 154. Niell CM, Smith SJ (2005) Functional imaging reveals rapid development of visual response properties in the zebrafish tectum. Neuron 45(6):941–951 155. Mank M, Santos AF, Direnberger S, Mrsic-Flogel TD, Hofer SB, Stein V, Hendel T, Reiff DF, Levelt C, Borst A, Bonhoeffer T, Hubener M, Griesbeck O (2008) A genetically encoded calcium indicator for chronic in vivo two-photon imaging. Nat Meth 5 (9):805–811 156. Miyawaki A, Llopis J, Heim R, McCaffery JM, Adams JA, Ikura M, Tsien RY (1997) Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin. Nature 388 (6645):882–887 157. Nakai J, Ohkura M, Imoto K (2001) A high signal-to-noise Ca(2 +) probe composed of a single green fluorescent protein. Nat Biotechnol 19(2):137–141 158. Tian L, Hires SA, Mao T, Huber D, Chiappe ME, Chalasani SH, Petreanu L, Akerboom J, McKinney SA, Schreiter ER, Bargmann CI, Jayaraman V, Svoboda K, Looger LL (2009) Imaging neural activity in worms, flies and mice with improved GCaMP calcium indicators. Nat Meth 6(12):875–881 159. Kuner T, Augustine GJ (2000) A genetically encoded ratiometric indicator for chloride: capturing chloride transients in cultured hippocampal neurons. Neuron 27(3):447–459 160. Higashijima S, Masino MA, Mandel G, Fetcho JR (2003) Imaging neuronal activity during zebrafish behavior with a genetically encoded calcium indicator. J Neurophysiol 90 (6):3986–3997 161. Samardzija M, Neuhauss SCF, Joly S, Kurz-Levin M, Grimm C (2010) Zebrafish vision—structure and function of the zebrafish visual system. Animal models of retinal diseases. Humana, New York 162. Stearns G, Evangelista M, Fadool JM, Brockerhoff SE (2007) A mutation in the cone-specific pde6 gene causes rapid cone photoreceptor degeneration in zebrafish. J Neurosci 27 (50):13866–13874 163. Nishiwaki Y, Komori A, Sagara H, Suzuki E, Manabe T, Hosoya T, Nojima Y, Wada H, Tanaka H, Okamoto H, Masai I (2008) Mutation of cGMP phosphodiesterase 6alpha′-subunit gene causes progressive degeneration of cone photoreceptors in zebrafish. Mech Dev 125(11–12):932–946 164. Krock BL, Perkins BD (2008) The intraflagellar transport protein IFT57 is required for cilia maintenance and regulates IFT-

165. 166. 167.

168.

169.

170. 171. 172.

173.

174.

175. 176.

particle-kinesin-II dissociation in vertebrate photoreceptors. J Cell Sci 121(Pt 11):1907–1915 Tsujikawa M, Malicki J (2004) Intraflagellar transport genes are essential for differentiation and survival of vertebrate sensory neurons. Neuron 42(5):703–716 Krock BL, Bilotta J, Perkins BD (2007) Noncell-autonomous photoreceptor degeneration in a zebrafish model of choroideremia. Proc Natl Acad Sci USA 104(11):4600–4605 Schonthaler HB, Fleisch VC, Biehlmaier O, Makhankov Y, Rinner O, Bahadori R, Geisler R, Schwarz H, Neuhauss SC, Dahm R (2008) The zebrafish mutant lbk/vam6 resembles human multisystemic disorders caused by aberrant trafficking of endosomal vesicles. Development 135(2):387–399 Nuckels RJ, Ng A, Darland T, Gross JM (2009) The vacuolarATPase complex regulates retinoblast proliferation and survival, photoreceptor morphogenesis, and pigmentation in the zebrafish eye. Invest Ophthalmol Vis Sci 50(2):893–905 van Rooijen E, Voest EE, Logister I, Bussmann J, Korving J, van Eeden FJ, Giles RH, Schulte-Merker S (2010) Von Hippel– Lindau tumor suppressor mutants faithfully model pathological hypoxia-driven angiogenesis and vascular retinopathies in zebrafish. Dis Model Mech 3(5–6):343–353 Cao Z, Jensen LD, Rouhi P, Hosaka K, Lanne T, Steffensen JF, Wahlberg E, Cao Y (2010) Hypoxia-induced retinopathy model in adult zebrafish. Nat Protoc 5(12):1903–1910 Rick JM, Horschke I, Neuhauss SC (2000) Optokinetic behavior is reversed in achiasmatic mutant zebrafish larvae. Curr Biol 10 (10):595–598 Huang YY, Rinner O, Hedinger P, Liu SC, Neuhauss SC (2006) Oculomotor instabilities in zebrafish mutant belladonna: a behavioral model for congenital nystagmus caused by axonal misrouting. J Neurosci 26(39):9873–9880 Dahm R, Geisler R, Nüsslein-Volhard C (2005) Zebrafish (Danio rerio) genome and genetics. In: Meyers RA (ed) Encyclopedia of molecular cell biology and molecular medicine, 2nd edn. WileyVCH, Weinheim, pp 593–626 Liu WY, Wang Y, Qin Y, Wang YP, Zhu ZY (2007) Site-directed gene integration in transgenic zebrafish mediated by Cre recombinase using a combination of mutant lox sites. Mar Biotechnol NY 9(4):420–428 Rieger S, Kulkarni RP, Darcy D, Fraser SE, Koster RW (2005) Quantum dots are powerful multipurpose vital labeling agents in zebrafish embryos. Dev Dyn 234(3):670–681 Koster RW, Fraser SE (2001) Tracing transgene expression in living zebrafish embryos. Dev Biol 233(2):329–346