Abstract
The last decade has seen a dramatic rise in the number of genes linked to neurological disorders, necessitating new models to explore underlying mechanisms and to test potential therapies. Over a similar period, many laboratories adopted zebrafish as a tractable model for studying brain development, defining neural circuits and performing chemical screens. Here we discuss strengths and limitations of using the zebrafish system to model neurological disorders. The underlying premise for many disease models is the high degree of homology between human and zebrafish genes, coupled with the conserved vertebrate Bauplan and repertoire of neurochemical signaling molecules. Yet, we caution that important evolutionary divergences often limit the extent to which human symptoms can be modeled meaningfully in zebrafish. We outline advances in genetic technologies that allow human mutations to be reproduced faithfully in zebrafish. Together with methods that visualize the development and function of neuronal pathways at the single cell level, there is now an unprecedented opportunity to understand how disease-associated genetic changes disrupt neural circuits, a level of analysis that is ideally suited to uncovering pathogenic changes in human brain disorders.
Keywords: zebrafish, disease models, neuroanatomy, neural development, CRISPR, transgenics
INTRODUCTION
Over the last two decades, zebrafish models have proliferated in laboratories internationally, used to study everything from infectious disease to social behavior. Breeding zebrafish embryos to ‘demonstrate development of a living egg and the circulation of the blood’ in student laboratories was suggested by Creaser as early as 1934 [1]. However, George Streisinger pioneered use of zebrafish as a research model in the late 1960s. His foundational studies, motivated primarily by the need for a species that would enable him to probe the genetic mechanisms underlying neural development experimentally, yielded the first laboratory zebrafish strains [2]. His laboratory developed methods for inducing genetic lesions in zebrafish and isolated the earliest mutants affecting brain development [3, 4]. During the 1990s and 2000s, many other mutants affecting neural development were isolated through mutagenesis screens, leading to rich insights into genetic mechanisms that control vertebrate brain development [5–7].
At around the same time, interest by the aquaculture industry in developing genetically modified fish spurred development of protocols for generating transgenic fish, initially by simple injection of linear DNA [8, 9]. Early attempts to study human disorders arising from gain of function mutations were able to leverage these techniques, creating transgenic zebrafish that modeled human cancers and neurodegenerative disorders [10–13]. However, until recently, two main obstacles limited the utility of zebrafish for studying many brain disorders associated with loss of function mutations. First, relatively few genes had been linked reliably to common disorders, such as autism and schizophrenia. Indeed, there remains substantial debate about the genetic architecture of many such diseases, including the extent to which rare single gene variants contribute to risk. Second, even where a gene mutation had been linked decisively to a brain disorder, it was technically very difficult to disrupt the homologous gene in zebrafish. Knockout mice were developed in the 1980s based on techniques for inducing and selecting rare recombination events in embryonic stem cells. In contrast, the first broadly applicable method for creating targeted genetic lesions in zebrafish appeared only in 2008, with the availability of zinc finger nucleases [14, 15].
Over the last decade, each of these obstacles has been surmounted. GWAS and exome-sequencing projects in large carefully phenotyped populations have revealed hundreds of single gene mutations linked causally to common neurodevelopmental disorders, such as autism, schizophrenia and intellectual disability [16–18], and genetic variants associated with sporadic neurodegenerative diseases, including Parkinson's disease (PD) and Alzheimer's disease [19, 20]. Many genes newly implicated in neurodevelopmental disorders were already well known to play essential roles in brain development, regulating neuronal proliferation, maturation or connectivity within the brain. Similarly, several genes associated with neurodegenerative diseases had well-established roles in neuronal proteostasis, mitochondrial function and CNS innate immunity. Nevertheless, for the majority of neurological disorders, the specific cellular processes that are disrupted by disease-associated gene variants still remain unclear. Thanks to a second advance, zebrafish are now ideally placed to help to resolve these questions. The adaptation of the bacterial CRISPR/Cas9 system for genome editing in eukaryotic cells allows loss of function alleles in almost any gene to be generated efficiently at minimal cost. Together, the wealth of new information about the genetic underpinnings of major human brain diseases and the ease with which homologous genes can be mutated in zebrafish, provide fresh impetus for using zebrafish to study neurological disorders.
In this review, we outline aspects of the zebrafish model that provide exceptional value for studying human brain disorders along with practical and theoretical limitations. The present paper considers the premise for zebrafish studies: the extent of neural and genetic homology between fish and humans, the feasibility of genetic manipulations and how these considerations may limit the types of human symptoms may be studied in this system. In a companion piece, we consider phenotyping strategies, focusing on behavioral analysis and brain imaging and describe how zebrafish may be successfully used to perform discovery screens for chemical modifiers [21].
Before embarking on a survey of this burgeoning field, we offer a few notes concerning terminology. First, although we refer to genetically manipulated zebrafish as ‘models’ for studying brain disorders, we stress that no species can faithfully reproduce the manifold changes that lead to the entire range of pathological and functional abnormalities seen in humans with CNS diseases. An important component of the disabling nature of brain disorders arises from the interaction between behavioral changes intrinsic to each disorder, and the variety of social environments through which people must navigate. Thus, as has been emphasized elsewhere [22], we do not see zebrafish as phenotypic models for the clinical manifestations of human brain diseases, but rather as providing an experimental system to answer questions related to the neurobiological basis for disorders and their treatment. As George Box wrote ‘All models are wrong, but some are useful’ [23].
Second, many studies on zebrafish are performed during the first week after fertilization. An occasional source of confusion is the term ‘larva’ to refer to young zebrafish. Zebrafish researchers customarily refer to animals as larvae during the period from around 72 hours post-fertilization (the ‘protruding mouth stage’) until reaching the juvenile stage several weeks later [24, 25]. In many insects, there is extensive remodeling of the brain between larval and adult stages; in Drosophila for instance, most larval neurons either die or acquire new morphology and function in adults [26]. In contrast, however, changes are far less drastic in zebrafish: for the most part, nuclei present in zebrafish larvae are readily identified in adults, gradually expanding as fish mature and developing characteristic internal organization and becoming more clearly segregated from their neighbors.
Why make zebrafish models?
Mammalian laboratory organisms are phylogenetically closer to humans than fish, although there are some exceptions in the degree to which they replicate human biology. For example, fish and humans—but not most rodents—are diurnal species whose visual behaviors are primarily driven by a cone-rich retina. Nevertheless, the zebrafish system has important strengths for understanding human brain diseases. Several excellent reviews have covered the general advantages of the zebrafish system [27]; however, we draw attention to five areas where we see a special niche for the zebrafish system in studying brain disorders.
Circuit-level analysis.
There is increasing recognition that many disorders affect the function of the brain at the circuit level. To understand these disorders, we need to decode how information is processed and transmitted as it passes between neurons, and then understand how this is modified by disease. Such studies are facilitated by simple organisms, where precise connections between neurons can be observed and manipulated readily. Rapid embryonic development enables zebrafish to become free-swimming just 5 days after fertilization. By this stage, the major burst of neurogenesis is complete, neurons have found their synaptic partners, and the brain is able to control a rich repertoire of behavior: 5-day-old zebrafish hunt, avoid predators, sleep and process a wide range of sensory cues to navigate in their environment. Yet remarkably, all this is controlled by fewer than 100 000 neurons (compared to around 35 million in the neonatal mouse) [28, 29]. A relatively simple brain greatly reduces the search space for pinpointing disease-associated changes. But just as importantly, most of these 100 000 neurons are identifiable in the sense that the same neuron, or cluster of neurons, can be identified readily in different fish thanks to extensive libraries of transgenic lines that express reporter genes in particular subsets of neurons [30–32]. Using chemogenetic and optogenetic methods, neuronal activity in the same neurons across fish can be monitored and manipulated to resolve how specific connections between neurons regulate physiology and behavior in wildtype and mutant animals.
The optical clarity of larval stage zebrafish further offers opportunities that are not readily available in rodent systems, to define and manipulate cell biological events occurring in intact CNS neurons in situ, using fluorescent reporters and intravital microscopy (Figure 1). Such studies have provided insight into microglial dynamics, mitochondrial structure, morphology and transport and cytosolic RedOx fluxes in mature neurons or used chemoptogenetics to induce oxidative damage to the respiratory chain of mitochondria within CNS neurons in vivo (Figure 2) [33–36]. Zebrafish thus offer the opportunity to study perturbations in neural circuits, a level of analysis most relevant to many brain disorders.
Figure 1.

Fluorescence microscopy and photoconversion in live transgenic zebrafish. Confocal Z-plane projections of zebrafish larva expressing the photoconvertible fluorophore Kaede in a pan-neuronal pattern throughout the CNS at 3 dpf. At baseline, fluorescence is seen in the green, but not red, channel (above). Immediately following exposure to UV light, fluorescence is seen only in the red channel (below)
Figure 2.

Imaging and manipulating RedOx biochemistry in vivo. (A) Ratiometric intravital confocal images of diencephalic neurons expressing the glutathione oxidation reporter GRX1-roGFP2. Dramatic changes in the fluorescence excitation properties of the reporter following application of hydrogen peroxide to the bath reveal a significant change in the intracellular glutathione RedOx potential [35]. (B) Intravital confocal images of mCerulean3-labeled mitochondria in the lateral line nerve of transgenic zebrafish also expressing a mitochondrially targeted form of the fluorogen activating protein dL5**. Exposure to far-red light in the presence (below), but not absence (above), of the iodine substituted fluorogen MG2I results in mitochondrial fragmentation as a consequence of singlet oxygen formation within the mitochondrial matrix [36].
Testing candidate genes.
Zebrafish offer a relatively rapid and inexpensive test system for determining whether candidate human gene mutations are in fact causally responsible for a disease phenotype. So few patients may be ascertained with rare mutations that it can be difficult to ascribe the disorder confidently to genetic variants that are present. Not only can mutant zebrafish lines be generated easily using CRISPR/Cas9, but large clutch sizes (often 200 or more), allow mutants to be compared to genetically well-controlled siblings. A close match between human and zebrafish phenotypes may provide strong support for a causal role of a candidate gene. Moreover, several gene mutations that show prenatal lethality in mice do not prevent embryonic development in zebrafish, making it possible to observe developmental effects of gene mutations up to mid-larval stages where major phases of brain development are mostly complete.
Phenotypically classifying disease mutations.
A special niche for the zebrafish system is in stratifying effects of gene mutations. So many causal genes have been linked to major disorders including autism, intellectual disability, schizophrenia and epilepsy, that there is an impetus to stratify mutations based on similar phenotypic effects. For these disorders, substantial variability in symptoms and severity has impaired diagnosis and confounded testing of novel treatments in homogeneous patient groups. Defining subgroups of patients based on genotypes that provoke shared disease phenotypes through common underlying neurobiological mechanisms may improve treatment and facilitate the search for new therapies [37]. Yet despite the wealth of new genetic information, in most cases, specific gene variants have each been described in only a handful of patients, thwarting efforts to use patient phenotypes to stratify disease groups. Detailed phenotyping of animals with relevant gene mutations can overcome this difficulty. Zebrafish are an attractive system for these studies as targeted gene mutations and phenotyping can both be performed with medium throughput compared to rodent models. Thus, comparison of mutant phenotypes in zebrafish with a range of targeted mutations has enabled patient mutations to be grouped based on phenotypic effect [38].
Linking genetic risk to underlying neurobiology.
In the majority of cases, single gene mutations do not account for common, sporadic, CNS diseases, such as PD, Alzheimer's disease, autism spectrum disorder, schizophrenia, most types of epilepsy or amyotrophic lateral sclerosis. Recent advances in these disorders have included the identification of SNPs that are associated with disease risk, with the implication that overall genetic risk may be an aggregate of quantitative changes in multiple different cellular processes, and overall disease risk may be a composite of genetic and environmental risk factors. However, moving from a catalog of risk alleles to understanding the relevant biology is a major challenge. Many disease-associated variants are in non-coding segments of the genome and are presumed to be involved in gene regulation, but the disease-relevant functional consequences of a variant might not even relate to the nearest gene. In addition, disease risk variants may work through cell non-autonomous mechanisms or alter aspects of synaptic transmission or information processing in neurons relevant to the phenotypic expression of the diseases, which would be difficult to analyse in cell culture. Consequently, an intact vertebrate brain may be a prerequisite for understanding the functions of genes whose expression is altered by disease risk variants. Together with the complexity resulting from changes in the expression of combinations of multiple genes (whose identities might not even be clear), this presents a significant practical challenge. The possibility of high-throughput reverse genetics in zebrafish provides a means to analyse the functions and interactions of multiple genes implicated in GWAS in an intact vertebrate brain.
Chemical screens.
Finally, an area where zebrafish models are well placed to make key contributions is in their utility for chemical modifier screens, with translational applications in the discovery and development of drugs to treat CNS disorders.
STRUCTURAL BRAIN CONSERVATION IN VERTEBRATES
All vertebrate central nervous systems share a common organizational scheme so that someone who is familiar with the brain of any vertebrate will also be on familiar ground when first encountering the brain of any other species [39].
Vertebrate brains come in an impressive range of shapes and sizes, yet all share the same basic architecture [40]. The same major subdivisions of the brain, from the visibly segmented rhombencephalon to the labyrinthine telencephalon, are established by genetic patterning mechanisms that are highly conserved in the vertebrate lineage (Figure 3) [41]. Neuroanatomical novelties (brain regions or connections that can not be recognized in distant species) undeniably exist. For instance, in teleost fish, the cerebellum includes the valvula cerebellum, a division that is simply not present outside this clade [40]. Even more unusual is the torus longitudinalis, a structure that occupies the dorsal midline of the mesencephalon only in ray-finned fish. However, most of the salient differences between brains simply reflect the degree to which a given structure is emphasized or elaborated. For example, in many mammals, extensive growth of the cortical plate results in the relatively thin cerebral cortex wrapping over the dorsal and lateral aspects of the brain, obscuring much of the diencephalon and midbrain. For similar reasons, the optic tectum—the direct homolog of the mammalian superior colliculus—is conspicuous in many fish species. Of course, divergences in the size or organization of brain regions often reflect differences in how functionally important they are to the biological success of a species. Such novelties are of intense interest for studying the biological basis of behavior and evolutionary mechanisms that adaptively modify it. Nevertheless, most brain regions that are affected by diseases in human patients can be identified readily in zebrafish, making it possible to study disease-related changes in brain architecture and to complete studies in cellular populations with strong functional homology.
Figure 3.

Comparative neuroanatomy of human and larval zebrafish. Schematic representations of sagittal sections of adult human (above) and 5-day post-fertilization zebrafish (below) brains. Homologous areas are shaded the same colors to illustrate the conservation of basic vertebrate brain organization from human to larval zebrafish, despite the dramatic difference in scale and complexity.
There are, however, challenges in translating neuroanatomy between zebrafish and mammals. Brain atlases for zebrafish are still relatively incomplete, especially in young fish that are most easily studied. As a result, brain regions of interest to a particular disease are sometimes poorly annotated. In part, this is because the zebrafish model is still relatively new; however, it is also true that neurons lack an evident nuclear organization in most brain regions of young zebrafish, greatly complicating the task of annotation. Only a minority of structures can be identified based on position or shape alone: the zebrafish cerebellum is quite obviously related to its corresponding structure in mammals, but it is less easy to define how its internal architecture relates to subdivisions of the mammalian cerebellum. To assign homology between vertebrate species, neuroanatomists rely not only on the topological position of a given structure, but also consider its connectivity with other parts of the brain, developmental origin, patterns of gene expression and function. Thus, for example, in zebrafish, a cluster of neurons below the cerebellum initiates escape behavior in response to vibration: based on their origin in rhombomere 1, commissural and cerebellar projections and expression of marker genes, this cluster likely represents the zebrafish homolog of the superior vestibular nucleus, a region that triggers vestibular startle behavior in mammals [42]. On the positive side, several three-dimensional digital atlases have already been developed by the zebrafish field and these are invaluable resources that enable results from experiments to be compared accurately with cutting edge brain annotations [43–47].
Nomenclature is another impediment to zebrafish studies. The accepted names for many brain regions in fish do not reflect established homology with mammalian counterparts. For instance, one may not immediately suspect that the counterpart of the nucleus isthmi in fish is the mammalian parabigeminal nucleus [48]. More subtly the medial habenula in larval zebrafish is (gratifyingly) homologous to the medial habenula in mammals—but then sadly transposes into the ‘dorsal’ habenula in adult fish [49]. These labeling issues are not unique to zebrafish. In mammals, brain nomenclature has been described charitably as ‘chaotic and exceptionally confusing’, reflecting conflicting parcellation schemes and labels employed by different anatomical traditions [50, 51]. As a result, it is frequently essential to dive deeply into decades old literature to determine if a brain region in mammals has a homolog in fish.
Finally, it is critical to acknowledge that despite the high degree of conservation of the vertebrate brain Bauplan, there are substantive differences between the precise pathways by which signals are relayed in fish and human brains. Major pathways in mammals may be greatly reduced, or even absent in fish. For instance, projections from the lateral geniculate of the thalamus to the visual cortex that transmit crucial visual information in humans do not have a homologous pathway in fish [52]. Rather, in fish, a direct projection from the retina to the optic tectum is essential for processing many visual cues. And although the homologous retinocollicular pathway in mammals is also used to guide visuo-motor behavior, it is typically involved only in relatively simple orienting behaviors [53]. Moreover, even neurons with well conserved projections and firing patterns may generate spontaneous action potentials using distinct molecular mechanisms, as in the case of pacemaker activity in zebrafish diencephalic dopamine neurons [54].
The most overt dissimilarity between fish and mammals is in the forebrain. Already in 1908, Ludwig Edinger recognized that the brainstem (or palaeencephalon as he called it) was highly conserved in vertebrates, but that the size and complexity of the forebrain varies greatly [55]. Indeed, Edinger described fish as ‘as strictly palaeencephalic animals’. We now recognize that the human cerebral cortex is a highly elaborated derivative of the dorsal pallium, a region that is present in all vertebrate species [41]. However, differences between cortical areas in humans, and pallial derivatives in fish are profound. Whereas molecular markers have revealed clear associations between subpallial derivatives (including the striatum and pallidum) in fish and mammals [56], it has proven more difficult to relate pallial territories in zebrafish unambiguously to cortical areas in mammals. In part this is because of unusual morphogenetic movements of the developing forebrain in zebrafish and other ray-finned fish. In most vertebrates, the side walls of the anterior neural tube balloon outward around a large internal ventricle. In contrast, in ray-finned fish, the homologous region expands along the anterior–posterior axis, eventually forming solid everted telencephalic lobes [57, 58]. No part of the resulting territory comprises a six-layered structure reminiscent of the mammalian neocortex. However, molecular markers have disclosed domains that are homologous to hippocampus, entorhinal cortex and amygdala [58, 59]. In addition, just as cortical interneurons are generated in the subpallial ganglionic eminences in mammals, before migrating long distances into the developing cortical plate, inhibitory neurons also migrate from the subpallium into the dorsal telencephalon in zebrafish [60]. It is important to recognize that features associated with dysfunction of the human cerebral cortex may be difficult to study using zebrafish.
NEUROGENETIC HOMOLOGY BETWEEN FISH AND MAMMALS
The high degree of structural similarity between fish and human brains is also reflected in conservation of the underlying genetic architecture. Of the 20 265 protein-coding genes in the human genome, 72% have an ortholog in zebrafish (compared to 90% in mouse). This number rises to 86% (97% in mouse) when considering only the subset of 4435 disease-associated protein-coding genes, and to 90% for disease genes that are expressed during brain development (Figure 4-5). Thus, a zebrafish homolog can be identified for most human neurodevelopmental disease genes. Indeed, occasionally, disease gene homologs that were lost in mouse are retained in zebrafish. For example, mutations in ANOS1 cause Kallmann syndrome, characterized by hypogonadotropic hypogonadism with anosmia. Mice lack an ANOS1 homolog; experiments in zebrafish demonstrated its molecular role in regulating sensory system development [61, 62].
Figure 4.

Conservation of protein coding genes in zebrafish and mouse models. Human protein-coding genes with orthologs in zebrafish (left) and mouse (right) shaded in purple, while those without orthologs are shaded red. Blue shading for zebrafish and mouse specific genes. The subset of human genes known to cause disease is also indicated. Data downloaded from ensembl biomart, filtering for protein-coding genes (GRCh38.p13). Genes on non-main contigs (i.e. patches and alternative assemblies) are excluded. Disease-associated genes were downloaded using the MIM morbid description annotation.
Figure 5.

A high fraction of neurodevelopmental disease genes have zebrafish orthologs. Chart indicates the percent of human genes with zebrafish paralogs, for all 19 988 human protein coding genes (left), and the 1052 human disease genes that are highly expressed during brain development (right). Outer rings show the percentage of human genes with either no (grey) or at least one zebrafish paralog (blue), while inner rings break down zebrafish genes according to whether zebrafish have one, two or more than three paralogs. Human protein-coding genes and their zebrafish homologs were downloaded from ensembl biomart, using the MIM morbid description to designate disease genes. For genes highly expressed during human neural development we downloaded the BrainSpan developmental transcriptome dataset (RNA-Seq Gencode v10 archive) [164] and identifed those genes whose maximal expression across developmental stages and timepoints were in the top 10% quantile.
However, complicating the picture, in many cases, there is no one-to-one correspondence between human and zebrafish genes. Around 350 million years ago, a genome duplication occurred in the ray-finned fish lineage, leading to teleosts. Many of the duplicated genes were subsequently lost (more than half over the next 75 million years); however, around 20% of human protein-coding genes are orthologous to multiple (generally two) equivalently related paralogs rather than to a single ortholog in zebrafish [63, 64]. For neurodevelopmental disease genes, the proportion is slightly higher, around 26% (Figure 5). To study the effects of a human loss of function genetic disorder, it may therefore be necessary to inactivate both zebrafish paralogs, or risk a phenotype being masked by the presence of an intact paralog [65]. For instance, in humans, mutations in TDP-43 cause amyotrophic lateral sclerosis; only zebrafish with mutations disrupting both paralogs of TDP-43 showed a muscle degeneration phenotype similar to patients [66]. Although in theory, the presence of two equivalent genes may free one copy to acquire new molecular functions (potentially confounding the interpretation of genetic analyses in fish), there is little evidence for such ‘neofunctionalization’ in zebrafish [67, 68]. However, there are several examples where zebrafish paralogs have diverged through the accumulation of genetic changes in promoter or enhancer regions. As a result, the ancestral expression pattern may become effectively partitioned between paralogs [69]. In some cases, this is actually advantageous for experimentation, because separately inactivating paralogs may enable a more detailed dissection of molecular function in different tissues and stages [70].
Signaling between neurons in zebrafish is based on the same main amino acid neurotransmitters (glutamate, GABA and glycine) as in mammals, and cognate receptors and transporters are also highly conserved [71–75]. Even key developmental changes in neurotransmitter signaling that occur in mammals are also present in zebrafish. During prenatal development, signaling by the classically inhibitory neurotransmitters GABA and glycine often paradoxically depolarizes target neurons due to high intracellular chloride levels [76, 77]. Depolarization of nascent neurons plays a critical role in the normal acquisition of neuronal identity and circuit formation in both fish and mammals [78, 79].
In mammals, monoaminergic neurotransmitters are synthesized and released from neurons located in specific nuclei that each project to a wide range of targets within the brain. The spatial locations of nuclei that synthesize monoaminergic neurotransmitters are generally conserved in zebrafish, with some notable exceptions. For example, zebrafish lack homologs of the mammalian ventral midbrain dopamine neurons that give rise to the important mesostriatal dopaminergic pathway, and it is thought that their function is instead performed by dopaminergic neurons in the striatum itself, or by diencephalic dopaminergic neurons that project to forebrain [54, 80, 81]. As in mammals, serotonin synthesis is concentrated in the raphe nuclei, located in the midline of the brainstem and midbrain; however, there is evidence for additional serotonergic nuclei in zebrafish in diencephalic areas [82, 83]. In both fish and mammals, major clusters of noradrenergic neurons are located in the locus coeruleus, and in the caudal medulla [84]. In the mammalian brain, histamine is exclusively produced by tuberomamillary neurons of the hypothalamus; in zebrafish, histaminergic neurons are likewise confined to a putatively homologous region of the posterior hypothalamus [85]. Projections have only been systematically studied for catecholaminergic neurons, thus limited information currently supports the functional homology of these systems [81]. Cholinergic neurons are distributed in several clusters in zebrafish that show varying degrees of conservation with mammals [86, 87]. Some clusters, such as the cranial motor nuclei and cholinergic nuclei of rhombomere 1, are well conserved across the entire vertebrate lineage; however, other regions are unique to teleosts. In the cholinergic system, synaptic transmission is terminated by neurotransmitter hydrolysis rather than uptake. One of the two enzymes responsible for hydrolysis, butyrylcholinesterase, is missing in zebrafish, although intriguingly, zebrafish have evolved a novel cholinesterase with similar activity [88].
Neuropeptides are another important class of signaling molecule in the mammalian brain [89]. All major classes of neuropeptides in mammals are also present in zebrafish, and indeed, at least one ortholog can be identified in zebrafish for 85 of 97 mammalian genes that produce neuropeptide precursors (Figure 6). Many biological differences have arisen during the millennia separating fish and humans. For instance, a duplication of the calcitonin-related polypeptide gene in mammals lead to the origin of CALCA and CALCB genes, where only a single orthologous gene is present in zebrafish [90]. Gonadotropin releasing hormone (GnRH) genes have an even more tangled evolutionary history with three ancestral genes being differentially lost, resulting in both humans and zebrafish each having two GnRH genes, only one of which is a true ortholog [91]. In humans, GnRH1 expressing neurons are located in the preoptic region. Zebrafish lack GnRH1, but instead retain GnRH3 (which humans lack), and a population of GnRH3 expressing neurons in zebrafish localize to the preoptic region. Thus, the high degree of conservation of neuropeptides between zebrafish and humans is often only the beginning of a more complex story where particular genes in each lineage acquire new roles or are co-opted to compensate for the loss of an ancestral gene function.
Figure 6.

Neuropeptide conservation in zebrafish. Number of paralogs of each major human neuropeptide in zebrafish. Green bar: Human genes that lack a zebrafish paralog. Successive bars: Human genes with 1 to 4 zebrafish paralogs each. The list of neuropeptides was taken from the neuropeptide database [89], and zebrafish homologs identified for each using ZFIN, PANTHER phylogenetic trees and reciprocal blast [165, 166]
DEVELOPMENTAL STAGING FOR DISEASE MODELS
Embryonic development in zebrafish is extremely rapid. After just 5 days of development, larvae are free swimming and possess a sophisticated behavioral repertoire [92, 93]. How can we compare developmental stages in zebrafish, with the much more slowly maturing human embryo? Naively, one might associate hatching in fish with birth in mammals, or with hatching in other classes of the vertebrate subphylum, to guide comparisons. However, fish do not hatch at a stage which is comparable to either birth in mammals or hatching in birds [94]. Postnatally, birds and mammals require sustained parental care, whereas zebrafish embryos have a yolk sac that is capable of sustaining life for at least the first week of development after fertilization. In fact, larvae that begin exogenous feeding starting at the 8th day of development show normal growth and survival [95]. Moreover, brain development does not correlate strongly with non-neural development across species, precluding the use of somatic characteristics to link brain development in fish and mammals [96].
In broad terms, zebrafish brain development is characterized by two waves of neurogenesis. Simple sensorimotor reflexes that appear within 24 hours of fertilization are mediated by a small cohort of early-born neurons. However, neurons that participate in more complex behaviors, such as navigation, hunting and visually mediated predator avoidance, are born during secondary neurogenesis that starts around 2 days post-fertilization (dpf). Post-mitotic neurons from this wave start to mature by 3 dpf, and by around 4 dpf, synaptic connections supporting a rich repertoire of behaviors are present [97].
However, there is an important exception to this timetable. Maturation of the zebrafish forebrain is delayed compared with the rest of the brain. Whereas most brainstem neurons have been generated and are synaptically connected by 5 dpf, neurogenesis for the pallium and other forebrain regions is in its inception, with the pallium fully maturing by 1 to 1.5 months post-fertilization [58–60, 98–100]. The delayed maturation of the forebrain explains why zebrafish in the first week do not perform complex behaviors that are known to require telencephalic neurons. For example, social behavior, which begins in the second week of development, is dependent on a population of ventral telencephalic cholinergic neurons [101–103]. Similarly, various forms of associative learning and memory, which are not robust in the first week of development, are linked to telencephalic function [104, 105]. Avoidance conditioning, for instance, requires neurons in the adult dorsomedial telencephalon that are postulated to be analogous to the mammalian amygdala [106], whereas adult maze learning correlates with neurogenesis in caudal and medial regions of the pallium [107].
The delayed development of the pallium has implications for the regulation of zebrafish research. Critical systems necessary for the emotional experience of pain in humans include the cortex and amygdala. As pallial derivatives, these are not yet mature in the first week of zebrafish development. As noted above, zebrafish do not have a telencephalic region that is organized in the same way as the human cerebral cortex. As most contemporary views ascribe a critical and necessary role to the cortex in mediating consciousness in humans, there is vigorous debate over whether fish sustain any conscious experience, including pain. In this respect, it is critical to distinguish between pain (defined by the International Association for the Study of Pain as a subjective unpleasant emotional experience), and nociception (which includes detecting and signaling aversive or dangerous cues) [108]. Unlike pain, nociception does not require processing by forebrain structures. Chronic decerebrate rats perform largely normal withdrawal responses to noxious stimuli, yet do not exhibit signs of fear or distress that would indicate the emotional experience of pain [109]. Conversely, nociceptive reflexes mediated by brainstem and spinal cord circuits allow animals to avoid injury but do not in themselves provoke the internal experience of pain [110]. Thus, pain requires a mature forebrain, and cannot be inferred from the presence of nociceptive sensorimotor reflexes [110]. In zebrafish, withdrawal behaviors are present by 5 dpf, enabling studies of nociception. However, in the absence of mature forebrain structures, these reflexes are unlikely to be accompanied by an emotional experience of pain. Indeed, the American Veterinary Medical Association Guidelines on Euthanasia note that experience of distress and pain might be present in zebrafish only after the first 7 days [111].
Following the embryonic period, fish differ from mammals in the presence of constitutive neurogenesis throughout life, supporting the continuous growth of the brain over the lifespan. In mammals, neural stem cells are localized to a few niches in the adult telencephalon (reviewed in [112]). In contrast, stem cell niches in the adult zebrafish brain are abundant and broadly distributed, with individual niches dedicated to producing specific neuronal subtypes [113, 114]. There are tantalizing indications that adult neurogenesis does not merely serve to slowly expand brain volume, but is also stimulated by sensory cues, perhaps helping to match cognitive power to environmental demands [115, 116].
Consistent with ongoing neurogenesis and growth, the zebrafish CNS can regenerate after injury (Figure 7). This has been demonstrated in multiple experimental settings, including compressive injuries or transection of the optic nerves and spinal cord, or laceration of the brain. For example, optic nerve compression in the zebrafish, an injury that would lead to permanent blindness in humans, triggers regrowth of retinal ganglion cell axons, pathfinding to the optic tectum, reestablishment of synaptic contacts, remyelination of regenerated axons, and restoration of visual function. It was recognized several decades ago that a full understanding of the molecular events underlying this extraordinary display of virtuoso CNS repair might lead to better treatments for human patients with a range of CNS injuries resulting in permanent neurological deficits. Zebrafish CNS axonal regeneration is broadly dependent on both cell-intrinsic properties of neurons that can survive axonal damage and upregulate axonal growth genes afterwards, and a permissive tissue environment for axonal growth in the zebrafish CNS (reviewed in [117]). For example, the zebrafish homolog of Nogo-A, a prominent myelin-derived inhibitory signal in mammals, promotes axonal growth [118]. Little is known about the mechanisms presiding over reestablishment of synapses and restoration of function. Zebrafish CNS regeneration is a very active area of research that holds great translational promise.
Figure 7.

Spinal cord regeneration in a larval zebrafish. Serial confocal Z-plane projections show the spinal cord of a zebrafish (A) at 10 dpf before any experimental manipulation, (B) at the same age immediately after spinal cord transection, and (C) 2 days later. Axons of projection neurons are labeled with membrane bound EGFP. Disrupted spinal tracts in (B) have regenerated in (C), such that abundant axons are seen crossing the transection site at the later time point. Lesion and imaging performed as previously described [167].
The remarkable capacity to effect repairs in response to damage raises important questions about using zebrafish models to study neurological diseases. It might be concluded from the previous paragraph that genetic and environmental manipulations designed to provoke disruption of the nervous system in a way that mimics neurological diseases might trigger a compensatory regenerative response that would prevent emergence of any informative phenotypes. It is certainly true that acute ablation of cellular groups can give rise to neurogenesis, differentiation and replacement of the relevant neuronal population. However, for genetic mutations and transgene products that impair the health or function of discrete cellular populations, it is currently unclear whether these compensatory responses can outmatch pathophysiology. Certainly, there are plenty of models with useful phenotypes that suggest this is not the case, and the added value of in vivo models relates to understanding altered cellular function in the context of intact circuits mediating behavior, including events upstream of cell death that might be less susceptible to mitigation by repair.
GENETIC EXPERIMENTAL APPROACHES
For many years, reverse genetics in zebrafish primarily relied on antisense knockdown of genes of interest, technically limiting analyses to the first few days of development. However, the zebrafish genetics toolbox now includes facile methods to introduce mutations into genes of interest, causing loss of function alleles that may model the pathophysiology of recessive disorders, or those characterized by haploinsufficiency or dominant negative mechanisms. Complementing this approach, efficient methods to introduce transgenes into the genome allow targeted expression of exogenous sequences in specific cells of interest enabling genetic rescue experiments, models of dominant disorders, and the expression of labels, biosensors or effectors. Together with more recent genome editing protocols, these powerful approaches to manipulate the zebrafish genome form the bedrock of its usefulness as a model organism and experimental system.
Loss of function models
Many loss of function zebrafish models have been generated for a wide range of human diseases, providing new insights into disease mechanisms. For example, studies on zebrafish with mutations in dyrk1aa, one of two paralogs of a gene associated with both autism and Down syndome, revealed dysregulated calcium homeostasis, leading to a high incidence of cerebrovascular hemorrhage, a clinical manifestation of Down syndrome [119]. Simple loss of function mutations in human genes are modeled readily by using the CRISPR/Cas9 system to introduce a small insertion or deletion mutation (indel) in an early exon of the homologous zebrafish gene, leading to a frameshift in the coding sequence and appearance of a premature stop codon. CRISPR/Cas variants are being developed constantly and will not be described here (see [120] for a comprehensive review). Typically, Cas9 protein is pre-incubated with synthetic RNA duplexes that target a gene of interest, then injected into fertilized zebrafish embryos, efficiently generating double-strand breaks that are repaired by non-homologous end joining, resulting in indel mutations, and allowing stable mutant lines to be isolated in around 6 months [121]. Unfortunately, many indel mutants fail to manifest a phenotype due to genetic compensation [122]. This was a revelation to many in the field, as hundreds of zebrafish mutants with nonsense mutations causing stable and robust phenotypes had previously been recovered through mutagenesis screens. The molecular mechanisms that contribute to compensation remain incompletely understood, but in some cases, compensation is triggered via the nonsense-mediated decay pathway. In nonsense-mediated decay, a stop codon is detected in any but the last exon of a gene, leading to partial degradation of the mRNA transcript [123]. mRNA degradation triggers increased expression of hundreds of other genes, including some with sequence similarity to the mutated gene that may compensate for its deficiency [124, 125]. It is difficult to ascertain whether the lack of an anticipated mutant phenotype is due to genetic compensation: one can check for nonsense-mediated decay, but compensation can also occur in its absence [126]. Fortunately, the CRISPR/Cas system can also generate small deletions efficiently. Thus, to avoid genetic compensation in loss of function mutants, CRISPR/Cas9 can be used to excise the promoter of a gene of interest [121, 124].
Effects of human loss of function mutations can be assessed rapidly in zebrafish using ‘morphants’ or ‘crispants’. In the former case, zygotes are injected with stabilized antisense morpholino oligonucleotides that suppress normal translation or splicing of the target transcript [127]. Morpholinos have been used for decades in zebrafish and other organisms, and well-controlled morpholino studies enable mutant phenotypes to be quickly assessed. Unfortunately, morpholinos often have off-target effects, producing phenotypes through interactions with mRNAs other than those of interest. These artifacts can confound even well controlled studies: zebrafish with morpholino knockdown of the Fragile X gene, fmr1 showed a strong increase in peripheral sensory neurite branching, a phenotype not subsequently observed in fmr1 mutants [128, 129]. Up to 20% of morpholinos also activate the p53 pathway, leading to non-specific neuronal death and craniofacial defects, including microcephaly [130]. Moreover, serial dilution that occurs as cells divide usually limits morpholino studies to processes that occur in the first few days of development. However, later phenotypes may be observed if morpholinos disrupt early developmental processes. Similar to the association between chd8 mutations and macrocephaly in humans, morpholino knockdown of chd8 in zebrafish causes an enlarged inter-ocular distance at larval stages, which was due to increased proliferation by neuronal progenitors in early development [131, 132].
As an alternative to morpholino studies, the CRISPR/Cas9 system is so efficient, that bi-allelic loss of function ‘crispants’ can frequently be generated simply by injecting reagents into fertilized embryos. To maximize success, a cocktail targeting 2–5 sites within a gene of interest is injected into embryos, which are then examined for phenotypes, and genotyped post-hoc to ascertain bi-allelic mutants [133]. Crispants are not subject to the limited window of efficacy of morpholinos. However, as for morpholinos, off-target effects in crispants may produce phenotypes unrelated to the gene of interest. Each crispant is a mosaic of cells, each potentially comprising a unique complement of mutations. This mosaicism may explain why, in some cases, crispants have demonstrated a phenotype that failed to manifest in stable mutants, possibly because genetic compensation mechanisms were only activated in a subset of cells [121, 126].
Disease-associated mutations in humans are often attributable to mechanisms other than simple loss of function, such as changes in amino acid sequence that have highly specific effects on protein function. Such changes may give rise to gain of function or dominant negative effects, or hypomorphic alleles caused by protein instability, aberrant trafficking, or loss of binding interactions. To reproduce these effects, it may be necessary to introduce the same change into the protein sequence of the homologous zebrafish gene. Precise genome editing can be achieved by co-injecting CRISPR/Cas reagents against a target genomic sequence, with DNA fragments containing the same sequence that has been modified to include the mutation of interest. Cas9 introduces a double-stranded break, triggering cellular homology-directed repair mechanisms that re-join the lesion using the DNA template molecule. Many efforts have been made to improve the relatively low success rate of this technique, which typically requires screening large numbers of founder fish, but the most influential factors are not yet clear [134]. Successful genome edits must be carefully validated, as standard PCR reactions used to verify template-directed repair events are subject to a well-known artifact that can generate misleading bands of the expected size, even when the template has integrated randomly [135].
Recent studies have revealed an important role for dominant or haploinsufficient mutations in several prevalent neurological disorders, often occurring when de novo mutations disrupt only one copy of an autosomal gene. Exome-sequencing studies in autism have revealed more than 100 genes with de novo mutations affecting one copy of a gene, many of which are predicted to be damaging loss of function mutations [17]. Haploinsufficiency is widespread, affecting around three thousand genes in humans [136, 137]. Gene dosage effects may vary between species, making it unclear how to recapitulate human haploinsufficient diseases most authentically. In mice there have been several instances where severe human haploinsufficient disorders are reproduced only in homozygous knockout mice but not in heterozygotes. For example, single gene copy mutations in shank2 cause autism and intellectual disability. Homozygous shank2 knockout mice have impairments in learning and social behavior; however these phenotypes are absent in shank2 heterozygotes [138]. Likewise, heterozygous zebrafish mutants have in some cases failed to manifest a phenotype that was anticipated from haploinsufficient disease genes. Thus, heterozygous mutations in TBX5 cause hand abnormalities as part of Holt-Oram syndrome in humans, whereas pectoral fin development in zebrafish is only impaired in tbx5 homozygous mutants [139]. Similarly, the GATA3 gene produces haploinsufficient birth defects in humans; heterozygous gata3 mutant zebrafish develop normally but are sensitized to the effects of Shh signaling inhibitors [140]. Diverse mechanisms underlie haploinsufficiency, including non-linear binding of transcription factors to promoter elements, and mono-allelic expression of some autosomal genes, such as pax5 [141, 142]. However, a major class of haploinsufficient genes encode proteins that are members of macromolecular protein complexes. This observation led to the proposal that subunits of protein complexes may be tightly regulated to ensure stoichiometric expression, because abnormal expression of one component may disrupt complex formation [143]. Consistent with this idea, in yeast, haploinsufficient genes also impair growth when expressed at increased levels [144]. Moreover, dosage-sensitive human genes rarely have paralogs, or are lost, in other mammals [145]. In this respect, it may be worth noting that around 15% of the zebrafish genome (covering around 20% of genes) is subject to copy-number variation - more than 4 times the rate in human [146]. A substantial portion of zebrafish genes may be somehow buffered against gene dosage effects, and it would therefore not be surprising if many haploinsufficient disease genes fail to manifest phenotypes in heterozygous zebrafish.
Transgenic alleles
Many human disorders are modeled using transgenic zebrafish. Transgenic overexpression is a powerful tool to determine whether disease alleles operate through loss or gain of function mechanisms. Thus, multiple prpf31 mutations are associated with autosomal dominant forms of retinitis pigmentosa in humans. Whereas zebrafish transgenically expressing one such mutant allele were overtly normal—suggesting haploinsufficiency as the cause of retinitis pigmentosa—transgenic expression of a distinct allele caused rod photoreceptor apoptosis, and the likelihood of a dominant negative mechanism occurring in patients [147]. Transgenic overexpression is useful not only to mimic the abnormalities caused by dominant human mutations, but also to induce pathology resembling that seen in sporadic diseases [10], mark groups of cells for ablation in functional circuitry studies [148], and label cells and organelles for in vivo cell biology studies using intravital imaging [33].
Although zebrafish transgenesis has a relatively long history going back to the 1990s, several technical innovations have improved both its efficiency and utility. Integration of microinjected plasmid DNA into the genome is extremely inefficient; this was improved significantly by the introduction of meganuclease-mediated transgenesis [149]. However, Koichi Kawakami’s isolation of an active transposase from the Medaka Tol2 transposon and subsequent exploitation for efficient integration of transgene constructs into the zebrafish genome was transformational for the widespread and practicable application of transgenic zebrafish in research [150]. In the most commonly employed scheme, plasmid DNA, in which the transgene cassette is flanked by 150 – 200 bp inverted terminal repeats from Tol2, is co-microinjected with mRNA encoding the 649 amino acid tol2 transposase into single cell zebrafish embryos. This results in efficient, single-copy, integration of the transgene cassette at one or more genomic sites. Although Tol2 is natively 4.7 kb in length, the system is effective with much larger transgenes and has even been exploited to introduce bacterial artificial chromosomes containing >100 kb of foreign DNA into the zebrafish genome [151]. The high efficiency with smaller constructs allows a relatively small number of embryos to be microinjected and raised, with a high level of confidence of finding F1 offspring that harbor and stably transmit the transgene. Importantly, the high efficiency also allowed this system to be exploited to find random insertions in genes with expression patterns of interest using gene trap and enhancer trap approaches [152].
A second major technical innovation that further increased the utility of zebrafish transgenesis was the introduction of the bipartite Gal4-UAS gene expression system [153]. In this approach, a transcriptional activator, comprising the DNA binding domain from yeast Gal4 fused to a potent transactivation domain, is expressed in a transgenic zebrafish ‘driver’ line. The cognate DNA site for Gal4 binding, the upstream activating sequence (UAS), is not natively present in the zebrafish genome. Driver zebrafish can be crossed with a second line, termed a ‘responder’ zebrafish, which expresses a transgene of interest under regulatory elements that contain multiple copies of the UAS. In the resulting double transgenic progeny, Gal4 from the driver allele transactivates the responder allele leading to transgene expression. The use of a bipartite expression system has distinct advantages. Each responder transgene only needs to be incorporated into a single UAS line, as its expression pattern and level can be controlled by crossing with different Gal4 drivers. In addition, the driver and responder lines can be maintained separately, which is important when transgene expression causes a lethal phenotype. Refinements of this system in zebrafish have yielded tools that allow conditional expression of a transgene under control of a UAS construct, in a cell or tissue-specific pattern dictated by Gal4 driver lines [30, 154]. This technique was leveraged in a recent study that used zebrafish to analyse the pathophysiological functions of α-synuclein [35]. In PD in humans, degeneration of substantia nigra dopaminergic neurons is accompanied by impaired mitochondrial respiratory chain function and oxidative stress, and by accumulation of α-synuclein [155]. Clinical phenocopies caused by over-expression of α-synuclein, or by exposure to mitochondrial inhibitors support a pathogenic role for both α-synuclein and mitochondrial dysfunction in PD, but how is unclear how these factors interact. Transgenic zebrafish expressing both human α-synuclein and a RedOx reporter protein in dopaminergic neurons using a dopaminergic Gal4 driver showed dramatically elevated dynamic cytoplasmic hydrogen peroxide flux after pharmacological inhibition of mitochondrial complex I, indicating that α-Synuclein amplifies the effects of mitochondrial dysfunction to promote oxidative damage in dopamine neurons [35].
The Gal4-UAS system has been combined with the high-efficiency generation of random transgene insertions in studies to generate libraries of Gal4 drivers using enhancer trap methods [30, 32]. Gaining genetic control over individual groups of CNS neurons is both powerful and useful, since analysis of cellular mechanisms that selectively involve specific neuronal types depends on the ability to express reporters or biosensors in these populations. Precisely targeted transgene expression is also critical for chemogenetic and optogenetic technologies that are used to probe the specific roles of circuit components in behavior. A major strength of the enhancer trap method is that lines can be selected based on their expression patterns in populations of interest [156], in contrast to the more traditional—and frequently unreliable—route of finding a gene that is expressed in target cells, then attempting to clone its regulatory elements. The availability of enhancer trap Gal4 driver libraries has also allowed identification of cell groups mediating specific behaviors by targeted ablation, a powerful use of genetics for ‘reverse experimental neurology’ [148].
More recently, CRISPR/Cas9 has been used to insert Gal4 into a gene of interest, so that its expression is controlled by endogenous regulatory elements [157, 158]. Although the number of reported knock-in Gal4 CNS driver lines is currently comparatively small, these may provide an important resource for future transgenic work. A potential concern is whether interfering with an endogenous gene locus by transgene insertion itself causes or modulates phenotypes of interest. Continued vigilance is necessary, especially when measuring sensitive phenotypes, such as behavioral endpoints.
A major limitation of the Gal4-UAS system is that it is not possible to express two transgenes in different patterns, for example, to analyse neuron–glia interactions using two-channel fluorescence, or to determine how an experimental perturbation in one cell type alters the readout from a sensor in another. This limitation has prompted interest in orthogonal expression systems. Although a Gal4–UAS pair can be combined with a traditional promoter-cDNA transgenic line, this loses the advantages of a bipartite system discussed above. The QF/QUAS system from Neurospora has been employed in zebrafish as a complementary transgene expression system to Gal4/UAS [159]. Like Gal4/UAS, the native components of QF/QUAS have some limitations, including a high degree of toxicity of the transcriptional activator QF (and its variants) and the presence of CG dinucleotides in the cognate QUAS sequence that may be susceptible to epigenetic silencing. Recent work may have addressed these limitations potentially providing an orthogonal gene expression system to Gal4/UAS [160].
Further complementing the Gal4-UAS system, transgenic recombinase methods using Cre/lox, Flp/FRT and ΦC31 have been applied in zebrafish to modify, excise or rearrange transgene cassettes (reviewed in [161]). Recombinases are particularly useful for lineage tracing experiments and conditional mutagenesis, and they enable powerful intersectional approaches to restrict transgene expression to cells that express two different complementary genes. For example, a gene-switch cassette might contain LoxP sites (the cognate sequences for Cre recombinase) arranged to flank one transgene whose expression precludes another, for example, LoxP-GFP-polyA-LoxP-RFP-polyA. Expressing this cassette under UAS regulatory elements, and crossing with a cell type specific Gal4 driver, leads to GFP expression in cells expressing Gal4. When further crossed with a complementary Cre driver, the subset of cells expressing both Cre and Gal4 switch permanently to expressing RFP. This is a powerful approach to selectively label subsets of cells for which specific Gal4 drivers do not exist, since minimal intersectional overlap between Gal4 and Cre can give rise to highly selective expression of the second transgene in the cassette [162]. Cre enhancer trap libraries have been generated using tol2 transgenesis as described above, providing opportunities for recombination-based transgenesis in a variety of neuronal patterns [28].
Cre recombination has been used extensively in mouse transgenic experiments to inactivate a gene in specific cell types, without the potential confounds of systemic deficits, organism lethality or cell nonautonomous phenotypes, which could limit analysis of constitutive knockouts [163]. This approach is dependent on targeted insertion of LoxP into key regions of specific genes. Advances in CRISPR-based methods for targeted transgene insertions into the zebrafish genome hold great potential for making this approach more widespread in zebrafish, which is an exciting prospect for dissecting cell-specific functions of genes in development and diseases [157, 158].
CONCLUSION
The substantial genetic homology between zebrafish and humans, in particular for disease-associated genes, and conservation of neurochemical signaling pathways provides a strong foundation for models of neurological disorders. It is more difficult to quantify homology of brain structure and composition or to generalize about the degree to which zebrafish may be able to reproduce changes associated with human lesions. However, with the explosion of basic research into neural pathways in zebrafish, it is becoming possible to determine whether there is a valid premise for generating a zebrafish model for a disease of interest. Genetic tools increasingly enable precise genome engineering in zebrafish, faithfully reproducing pathogenic mutations found in human patients. In the accompanying article, we turn our attention to strategies for pinpointing phenotypic changes in zebrafish neurological disease models that may provide insight into underlying pathology in patients [21].
Acknowledgment
This article does not represent the views of the United States government.
Author Contributions
H.A.B. and E.A.B. co-wrote and edited the manuscript.
Conflict of Interest
None declared.
Data availability
The data underlying this article will be shared on reasonable request to the corresponding author.
Supplementary Material
Funding
Dr. Burgess was supported by the Intramural Research Program of the Eunice Kennedy Shriver National Institute for Child Health and Human Development (NICHD). Dr. Burton was supported by research grants from NINDS (NS123211, NS125280, NS080881), NIEHS (ES025606, ES022644), The United States Department of Veterans Affairs (BX003168), CurePSP (655-2018-06, 468-08), Pittsburgh Foundation (AD2015-77642), and the University of Pittsburgh (Endowed Chair).
Supplementary Material
Supplementary data is available at Oxford Open Neuroscience online.
Contributor Information
Harold A Burgess, Division of Developmental Biology, Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, MD, 20892, USA.
Edward A Burton, Pittsburgh Institute for Neurodegenerative Diseases, University of Pittsburgh, Pittsburgh, PA, 15260, USA; Department of Neurology, University of Pittsburgh, Pittsburgh, PA,15260, USA; Geriatric Research, Education, and Clinical Center, Pittsburgh VA Healthcare System, Pittsburgh, PA, 15240, USA.
References
- 1.Creaser CW. The technic of handling the zebra fish (Brachydanio rerio) for the production of eggs which are favorable for embryological research and are available at any specified time throughout the year. Copeia. 1934;1934:159–61 [Google Scholar]
- 2.Grunwald DJ, Eisen JS. Headwaters of the zebrafish -- emergence of a new model vertebrate. Nat Rev Genet. 2002;3:717–24 [DOI] [PubMed] [Google Scholar]
- 3.Grunwald D, Kimmel C, Westerfield Met al. A neural degeneration mutation that spares primary neurons in the zebrafish. Dev Biol. 1988;126:115–28 [DOI] [PubMed] [Google Scholar]
- 4.Walker C, Streisinger G. Induction of mutations by gamma-rays in Pregonial germ cells of zebrafish embryos. Genetics. 1983;103:125–36 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Brockerhoff SE, Hurley JB, Janssen-Bienhold Uet al. A behavioral screen for isolating zebrafish mutants with visual system defects. Proc Natl Acad Sci U S A. 1995;92:10545–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Granato M, van Eeden FJ, Schach Uet al. Genes controlling and mediating locomotion behavior of the zebrafish embryo and larva. Development. 1996;123:399–413 [DOI] [PubMed] [Google Scholar]
- 7.Schier AF, Neuhauss SC, Harvey Met al. Mutations affecting the development of the embryonic zebrafish brain. Development. 1996;123:165–78 [DOI] [PubMed] [Google Scholar]
- 8.Stuart GW, McMurray JV, Westerfield M. Replication, integration and stable germ-line transmission of foreign sequences injected into early zebrafish embryos. Development. 1988;103:403–12 [DOI] [PubMed] [Google Scholar]
- 9.Zhu Z, He L, Chen S. Novel gene transfer into the fertilized eggs of gold fish (Carassius auratus L. 1758). J Appl Ichthyol. 1985;1:31–4 [Google Scholar]
- 10.Bai Q, Garver JA, Hukriede NA, Burton EA. Generation of a transgenic zebrafish model of Tauopathy using a novel promoter element derived from the zebrafish eno2 gene. Nucleic Acids Res. 2007;35:6501–16 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Le X, Langenau DM, Keefe MDet al. Heat shock-inducible Cre/lox approaches to induce diverse types of tumors and hyperplasia in transgenic zebrafish. Proc Natl Acad Sci U S A. 2007;104:9410–5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Sabaawy HE, Azuma M, Embree LJet al. TEL-AML1 transgenic zebrafish model of precursor B cell acute lymphoblastic leukemia. Proc Natl Acad Sci U S A. 2006;103:15166–71 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Tomasiewicz HG, Flaherty DB, Soria JP, Wood JG. Transgenic zebrafish model of neurodegeneration. J Neurosci Res. 2002;70:734–45 [DOI] [PubMed] [Google Scholar]
- 14.Doyon Y, McCammon JM, Miller JCet al. Heritable targeted gene disruption in zebrafish using designed zinc-finger nucleases. Nat Biotechnol. 2008;26:702–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Meng X, Noyes MB, Zhu LJet al. Targeted gene inactivation in zebrafish using engineered zinc-finger nucleases. Nat Biotechnol. 2008;26:695–701 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Fitzgerald TW, Gerety SS, Jones WDet al. Large-scale discovery of novel genetic causes of developmental disorders. Nature. 2015;519:223–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Satterstrom FK, Kosmicki JA, Wang Jet al. Large-scale exome sequencing study implicates both developmental and functional changes in the neurobiology of autism. Cell. 2020;180:568–584.e23 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Singh T, Poterba T, Curtis D. et al. Rare coding variants inten genes confer substantial risk for schizophrenia. Nature. 2022; 604:509–16 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Nalls MA, Blauwendraat C, Vallerga CLet al. Identification of novel risk loci, causal insights, and heritable risk for Parkinson’s disease: a meta-analysis of genome-wide association studies. Lancet Neurol. 2019;18:1091–102 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wightman DP, Jansen IE, Savage JEet al. A genome-wide association study with 1,126,563 individuals identifies new risk loci for Alzheimer’s disease. Nat Genet. 2021;53:1276–82 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Burton EA, Burgess HA.. A critical review of zebrafish neurological disease models - 2, Application: functional and neuroanatomical phenotyping strategies and chemical screens. Oxford Open Neuroscience. 2023. 10.1093/oons/kvac019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.NOT-MH-19-053: Notice of NIMHs Considerations Regarding the Use of Animal Neurobehavioral Approaches in Basic and Pre-clinical Studies. 2019.
- 23.Box GEP Robustness in the Strategy of Scientific Model Building. In: Launer RL, Wilkinson GN, (eds.), Robustness in Statistics. Academic Press, 1979, 201–36 [Google Scholar]
- 24.Kimmel CB, Ballard WW, Kimmel SRet al. Stages of embryonic development of the zebrafish. American Journal of Anatomy. 1995;203:253–310 [DOI] [PubMed] [Google Scholar]
- 25.Parichy DM, Elizondo MR, Mills MGet al. Normal table of postembryonic zebrafish development: staging by externally visible anatomy of the living fish. Dev Dyn. 2009;238:2975–3015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Tissot M, Stocker RF. Metamorphosis in drosophila and other insects: the fate of neurons throughout the stages. Prog Neurobiol. 2000;62:89–111 [DOI] [PubMed] [Google Scholar]
- 27.Veldman MB, Lin S. Zebrafish as a developmental model organism for pediatric research. Pediatr Res. 2008;64:470–6 [DOI] [PubMed] [Google Scholar]
- 28.Tabor KM, Marquart GD, Hurt Cet al. Brain-wide cellular resolution imaging of Cre transgenic zebrafish lines for functional circuit-mapping. elife. 2019;8. 10.7554/eLife.42687 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Young DM, Fazel Darbandi S, Schwartz Get al. Constructing and optimizing 3D atlases from 2D data with application to the developing mouse brain. elife. 2021;10:e61408 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Asakawa K, Suster ML, Mizusawa Ket al. Genetic dissection of neural circuits by Tol2 transposon-mediated Gal4 gene and enhancer trapping in zebrafish. Proc Natl Acad Sci U S A. 2008;105:1255–60 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Bergeron SA, Hannan MC, Codore Het al. Brain selective transgene expression in zebrafish using an NRSE derived motif. Front Neural Circuits. 2012;6:110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Scott EK, Mason L, Arrenberg ABet al. Targeting neural circuitry in zebrafish using GAL4 enhancer trapping. Nat Methods. 2007;4:323–6 [DOI] [PubMed] [Google Scholar]
- 33.Dukes AA, Bai Q, Van Laar VSet al. Live imaging of mitochondrial dynamics in CNS dopaminergic neurons in vivo demonstrates early reversal of mitochondrial transport following MPP+ exposure. Neurobiol Dis. 2016;95:238–49 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hassan-Abdi R, Brenet A, Bennis Met al. Neurons expressing pathological tau protein trigger dramatic changes in microglial morphology and dynamics. Front Neurosci. 2019;13:1199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Van Laar VS, Chen J, Zharikov ADet al. α-Synuclein amplifies cytoplasmic peroxide flux and oxidative stress provoked by mitochondrial inhibitors in CNS dopaminergic neurons in vivo. Redox Biol. 2020;37:101695 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Xie W, Jiao B, Bai Qet al. Chemoptogenetic ablation of neuronal mitochondria in vivo with spatiotemporal precision and controllable severity. elife. 2020;9. 10.7554/eLife.51845 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Stessman HA, Bernier R, Eichler EE. A genotype-first approach to defining the subtypes of a complex disease. Cell. 2014;156:872–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Thyme SB, Pieper LM, Li EHet al. Phenotypic landscape of schizophrenia-associated genes defines candidates and their shared functions. Cell. 2019;177:478–491.e20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Butler AB, Hodos W. Comparative Vertebrate Neuroanatomy: Evolution and Adaptation. John Wiley & Sons, 2005 [Google Scholar]
- 40.Striedter GF Principles of Brain Evolution. Sunderland, Mass: Sinauer Associates, 2005 [Google Scholar]
- 41.Puelles L, Harrison M, Paxinos G, Watson C. A developmental ontology for the mammalian brain based on the prosomeric model. Trends Neurosci. 2013;36:570–8 [DOI] [PubMed] [Google Scholar]
- 42.Marquart GD, Tabor KM, Bergeron SAet al. Prepontine non-giant neurons drive flexible escape behavior in zebrafish. PLoS Biol. 2019;17:e3000480 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Kenney JW, Steadman PE, Young Oet al. A 3D adult zebrafish brain atlas (AZBA) for the digital age. elife. 2021;10:e69988 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Kunst M, Laurell E, Mokayes Net al. A cellular-resolution atlas of the larval zebrafish brain. Neuron. 2019;103:21–38.e5 [DOI] [PubMed] [Google Scholar]
- 45.Marquart GD, Tabor KM, Brown Met al. A 3D searchable database of transgenic zebrafish Gal4 and Cre lines for functional neuroanatomy studies. Front Neural Circuits. 2015;9:78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Randlett O, Wee CL, Naumann EAet al. Whole-brain activity mapping onto a zebrafish brain atlas. Nat Methods. 2015;12:1039–46 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Ronneberger O, Liu K, Rath Met al. ViBE-Z: a framework for 3D virtual colocalization analysis in zebrafish larval brains. Nat Methods. 2012;9:735–42 [DOI] [PubMed] [Google Scholar]
- 48.Volkmann K, Chen YY, Harris MPet al. The zebrafish cerebellar upper rhombic lip generates tegmental hindbrain nuclei by long-distance migration in an evolutionary conserved manner. J Comp Neurol. 2010;518:2794–817 [DOI] [PubMed] [Google Scholar]
- 49.Amo R, Aizawa H, Takahoko Met al. Identification of the zebrafish ventral habenula as a homolog of the mammalian lateral habenula. J Neurosci. 2010;30:1566–74 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Mai JK, Majtanik M. Toward a common terminology for the thalamus. Front Neuroanat. 2018;12:114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Swanson LW. Brain maps online: toward open access atlases and a pan-mammalian nomenclature. J Comp Neurol. 2015;523:2272–6 [DOI] [PubMed] [Google Scholar]
- 52.Mueller T. What is the thalamus in zebrafish? Front Neurosci. 2012;6:64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Ingle D. Evolutionary perspectives on the function of the optic tectum. Brain Behav Evol. 1973;8:211–37 [DOI] [PubMed] [Google Scholar]
- 54.Ilin VA, Bai Q, Watson AMet al. Mechanism of pacemaker activity in zebrafish DC2/4 dopaminergic neurons. J Neurosci. 2021;41:4141–57 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Edinger L. The Relations of Comparative Anatomy to Comparative Psychology J Comp Neurol and Psych. 1908;18(5) 437–57 [Google Scholar]
- 56.Mueller T, Wullimann MF, Guo S. Early teleostean basal ganglia development visualized by zebrafish Dlx2a, Lhx6, Lhx7, Tbr2 (eomesa), and GAD67 gene expression. J Comp Neurol. 2008;507:1245–57 [DOI] [PubMed] [Google Scholar]
- 57.Folgueira M, Bayley P, Navratilova Pet al. Morphogenesis underlying the development of the everted teleost telencephalon. Neural Dev. 2012;7:32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Porter BA, Mueller T. The zebrafish Amygdaloid complex - functional ground plan, molecular delineation, and everted topology. Front Neurosci. 2020;14:608. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Dirian L, Galant S, Coolen Met al. Spatial regionalization and heterochrony in the formation of adult pallial neural stem cells. Dev Cell. 2014;30:123–36 [DOI] [PubMed] [Google Scholar]
- 60.Mione M, Baldessari D, Deflorian Get al. How neuronal migration contributes to the morphogenesis of the CNS: insights from the zebrafish. Dev Neurosci. 2008;30:65–81 [DOI] [PubMed] [Google Scholar]
- 61.Wang J, Yin Y, Lau Set al. Anosmin1 shuttles Fgf to facilitate its diffusion, increase its local concentration, and induce sensory organs. Dev Cell. 2018;46:751–766.e12 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Yanicostas C, Herbomel E, Dipietromaria A, Soussi-Yanicostas N. Anosmin-1a is required for fasciculation and terminal targeting of olfactory sensory neuron axons in the zebrafish olfactory system. Mol Cell Endocrinol. 2009;312:53–60 [DOI] [PubMed] [Google Scholar]
- 63.Postlethwait J, Amores A, Cresko Wet al. Subfunction partitioning, the teleost radiation and the annotation of the human genome. Trends Genet. 2004;20:481–90 [DOI] [PubMed] [Google Scholar]
- 64.Sato Y, Hashiguchi Y, Nishida M. Temporal pattern of loss/persistence of duplicate genes involved in signal transduction and metabolic pathways after teleost-specific genome duplication. BMC Evol Biol. 2009;9:127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 65.Hoffman EJ, Turner KJ, Fernandez JMet al. Estrogens suppress a behavioral phenotype in zebrafish mutants of the autism risk gene, CNTNAP2. Neuron. 2016;89:725–33 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66.Schmid B, Hruscha A, Hogl Set al. Loss of ALS-associated TDP-43 in zebrafish causes muscle degeneration, vascular dysfunction, and reduced motor neuron axon outgrowth. Proc Natl Acad Sci U S A. 2013;110:4986–91 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Braasch I, Gehrke AR, Smith JJet al. The spotted gar genome illuminates vertebrate evolution and facilitates human-teleost comparisons. Nat Genet. 2016;48:427–37 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 68.Ohno S Evolution by Gene Duplication. Springer-Verlag, 1970 [Google Scholar]
- 69.Force A, Lynch M, Pickett FBet al. Preservation of duplicate genes by complementary. Degenerative Mutations. 1999;151:1531–45 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 70.Thummel R, Enright JM, Kassen SCet al. Pax6a and Pax6b are required at different points in neuronal progenitor cell proliferation during zebrafish photoreceptor regeneration. Exp Eye Res. 2010;90:572–82 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 71.Cox JA, Kucenas S, Voigt MM. Molecular characterization and embryonic expression of the family of N-methyl-D-aspartate receptor subunit genes in the zebrafish. Dev Dyn. 2005;234:756–66 [DOI] [PubMed] [Google Scholar]
- 72.Haug MF, Gesemann M, Mueller T, Neuhauss SCF. Phylogeny and expression divergence of metabotropic glutamate receptor genes in the brain of zebrafish (Danio rerio). J Comp Neurol. 2013;521:1533–60 [DOI] [PubMed] [Google Scholar]
- 73.Hirata H, Carta E, Yamanaka Iet al. Defective glycinergic synaptic transmission in zebrafish motility mutants. Front Mol Neurosci. 2009;2:26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 74.Lin W-H, Wu C-H, Chen Y-C, Chow WY. Embryonic expression of zebrafish AMPA receptor genes: zygotic gria2alpha expression initiates at the midblastula transition. Brain Res. 2006;1110:46–54 [DOI] [PubMed] [Google Scholar]
- 75.Sadamitsu K, Shigemitsu L, Suzuki Met al. Characterization of zebrafish GABA(a) receptor subunits. Sci Rep. 2021;11:6242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 76.Rivera C, Voipio J, Payne JAet al. The K+/cl- co-transporter KCC2 renders GABA hyperpolarizing during neuronal maturation. Nature. 1999;397:251–5 [DOI] [PubMed] [Google Scholar]
- 77.Yamada J, Okabe A, Toyoda Het al. Cl- uptake promoting depolarizing GABA actions in immature rat neocortical neurones is mediated by NKCC1. J Physiol. 2004;557:829–41 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 78.Ben-Ari Y. Excitatory actions of gaba during development: the nature of the nurture. Nat Rev Neurosci. 2002;3:728–39 [DOI] [PubMed] [Google Scholar]
- 79.Reynolds A, Brustein E, Liao Met al. Neurogenic role of the depolarizing chloride gradient revealed by global overexpression of KCC2 from the onset of development. J Neurosci. 2008;28:1588–97 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 80.Rink E, Wullimann M. The teleostean (zebrafish) dopaminergic system ascending to the subpallium (striatum) is located in the basal diencephalon (posterior tuberculum). Brain Res. 2001;889:316–30 [DOI] [PubMed] [Google Scholar]
- 81.Tay TL, Ronneberger O, Ryu Set al. Comprehensive catecholaminergic projectome analysis reveals single-neuron integration of zebrafish ascending and descending dopaminergic systems. Nat Commun. 2011;2:171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 82.Bellipanni G, Rink E, Bally-Cuif L. Cloning of two tryptophan hydroxylase genes expressed in the diencephalon of the developing zebrafish brain. Mech Dev. 2002;119:S215–20 [DOI] [PubMed] [Google Scholar]
- 83.Lillesaar C, Tannhauser B, Stigloher Cet al. The serotonergic phenotype is acquired by converging genetic mechanisms within the zebrafish central nervous system. Dev Dyn. 2007;236:1072–84 [DOI] [PubMed] [Google Scholar]
- 84.McLean DL, Fetcho JR. Ontogeny and innervation patterns of dopaminergic, noradrenergic, and serotonergic neurons in larval zebrafish. J Comp Neurol. 2004;480:38–56 [DOI] [PubMed] [Google Scholar]
- 85.Kaslin J, Panula P. Comparative anatomy of the histaminergic and other aminergic systems in zebrafish (Danio rerio). J Comp Neurol. 2001;440:342–77 [DOI] [PubMed] [Google Scholar]
- 86.Arenzana FJ, Clemente D, Sánchez-González Ret al. Development of the cholinergic system in the brain and retina of the zebrafish. Brain Res Bull. 2005;66:421–5 [DOI] [PubMed] [Google Scholar]
- 87.Clemente D, Porteros A, Weruaga Eet al. Cholinergic elements in the zebrafish central nervous system: histochemical and immunohistochemical analysis. J Comp Neurol. 2004;474:75–107 [DOI] [PubMed] [Google Scholar]
- 88.Tan KS, Zhang Y, Liu Let al. Molecular cloning and characterization of an atypical butyrylcholinesterase-like protein in zebrafish. Comp Biochem Physiol B Biochem Mol Biol. 2021;255:110590 [DOI] [PubMed] [Google Scholar]
- 89.Burbach JPH. Neuropeptides from concept to online database www.neuropeptides.nl. Eur J Pharmacol. 2010;626:27–48 [DOI] [PubMed] [Google Scholar]
- 90.Katafuchi T, Yasue H, Osaki T, Minamino N. Calcitonin receptor-stimulating peptide: its evolutionary and functional relationship with calcitonin/calcitonin gene-related peptide based on gene structure. Peptides. 2009;30:1753–62 [DOI] [PubMed] [Google Scholar]
- 91.Ogawa S, Yamamoto N, Hagio Het al. Multiple gonadotropin-releasing hormone systems in non-mammalian vertebrates: ontogeny, anatomy, and physiology. J Neuroendocrinol. 2022;34:e13068 [DOI] [PubMed] [Google Scholar]
- 92.Fero K, Yokogawa T, Burgess HA. The behavioral repertoire of larval zebrafish. In: Kalueff A, Cachat J, (eds.), Zebrafish Models in Neurobehavioral Research. New York: Springer, 2011, 249–91 [Google Scholar]
- 93.Marques JC, Lackner S, Félix R, Orger MB. Structure of the zebrafish locomotor repertoire revealed with unsupervised behavioral clustering. Curr Biol. 2018;28:181–195.e5 [DOI] [PubMed] [Google Scholar]
- 94.Balon EK. Saltatory processes and Altricial to Precocial forms in the ontogeny of fishes. Amer Zool. 1981;21:573–96 [Google Scholar]
- 95.Hernandez RE, Galitan L, Cameron Jet al. Delay of initial feeding of zebrafish larvae until 8 days Postfertilization has no impact on survival or growth through the juvenile stage. Zebrafish. 2018;15:515–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 96.Clancy B, Finlay BL, Darlington RB, Anand KJS. Extrapolating brain development from experimental species to humans. Neurotoxicology. 2007;28:931–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 97.Mueller T, Wullimann MF. Atlas of Early Zebrafish Brain Development. A Tool for Molecular Neurogenetics, 1st edn. Amsterdam: Elsevier B.V, 2005 [Google Scholar]
- 98.Filippi A, Mueller T, Driever W. vglut2 and gad expression reveal distinct patterns of dual GABAergic versus glutamatergic cotransmitter phenotypes of dopaminergic and noradrenergic neurons in the zebrafish brain. J Comp Neurol. 2014;522:2019–37 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 99.Furlan G, Cuccioli V, Vuillemin Net al. Life-long neurogenic activity of individual neural stem cells and continuous growth establish an outside-in architecture in the teleost pallium. Curr Biol. 2017;27:3288–3301.e3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 100.Mueller T, Dong Z, Berberoglu MA, Guo S. The dorsal pallium in zebrafish, Danio rerio (Cyprinidae, Teleostei). Brain Res. 2011;1381:95–105 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 101.Dreosti E, Lopes G, Kampff ARet al. Development of social behavior in young zebrafish. Front Neural Circuits. 2015;39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 102.Engeszer RE, Da Barbiano LA, Ryan MJet al. Timing and plasticity of shoaling behaviour in the zebrafish, Danio rerio. Anim Behav. 2007;74:1269–75 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 103.Stednitz SJ, McDermott EM, Ncube Det al. Forebrain control of behaviorally driven social orienting in zebrafish. Curr Biol. 2018;28:2445–2451.e3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 104.Bretaud S, Li Q, Lockwood BLet al. A choice behavior for morphine reveals experience-dependent drug preference and underlying neural substrates in developing larval zebrafish. Neuroscience. 2007;146:1109–16 [DOI] [PubMed] [Google Scholar]
- 105.Valente A, Huang KH, Portugues R, Engert F. Ontogeny of classical and operant learning behaviors in zebrafish. Learn Mem. 2012;19:170–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 106.Lal P, Tanabe H, Suster MLet al. Identification of a neuronal population in the telencephalon essential for fear conditioning in zebrafish. BMC Biol. 2018;16:45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 107.Mazzitelli-Fuentes LS, Román FR, Castillo Elías JRet al. Spatial learning promotes adult neurogenesis in specific regions of the zebrafish pallium. Front Cell Dev Biol. 2022;10:840964 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 108.Raja SN, Carr DB, Cohen Met al. The revised International Association for the Study of Pain definition of pain: concepts, challenges, and compromises. Pain. 2020;161:1976–82 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 109.Woolf CJ. Long term alterations in the excitability of the flexion reflex produced by peripheral tissue injury in the chronic decerebrate rat. Pain. 1984;18:325–43 [DOI] [PubMed] [Google Scholar]
- 110.Barik A, Thompson JH, Seltzer Met al. A brainstem-spinal circuit controlling Nocifensive behavior. Neuron. 2018;100:1491–1503.e3 [DOI] [PubMed] [Google Scholar]
- 111.Underwood W, Anthony R. AVMA guidelines for the euthanasia of animals: 2020 edition Retrieved on March. 2020;2013:2020–1 [Google Scholar]
- 112.Matsubara S, Matsuda T, Nakashima K. Regulation of adult mammalian neural stem cells and neurogenesis by cell extrinsic and intrinsic factors. Cell. 2021;10:1145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 113.Grandel H, Kaslin J, Ganz Jet al. Neural stem cells and neurogenesis in the adult zebrafish brain: origin, proliferation dynamics, migration and cell fate. Dev Biol. 2006;295:263–77 [DOI] [PubMed] [Google Scholar]
- 114.Kaslin J, Ganz J, Geffarth Met al. Stem cells in the adult zebrafish cerebellum: initiation and maintenance of a novel stem cell niche. J Neurosci. 2009;29:6142–53 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 115.Hall ZJ, Tropepe V. Movement maintains forebrain neurogenesis via peripheral neural feedback in larval zebrafish. elife. 2018;7:e31045 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 116.Lindsey BW, Di Donato S, Kaslin Jet al. Sensory-specific modulation of adult neurogenesis in sensory structures is associated with the type of stem cell present in the neurogenic niche of the zebrafish brain. Eur J Neurosci. 2014;40:3591–607 [DOI] [PubMed] [Google Scholar]
- 117.Rasmussen JP, Sagasti A. Learning to swim, again: axon regeneration in fish. Exp Neurol. 2017;287:318–30 [DOI] [PubMed] [Google Scholar]
- 118.Welte C, Engel S, Stuermer CAO. Upregulation of the zebrafish Nogo-a homologue, Rtn4b, in retinal ganglion cells is functionally involved in axon regeneration. Neural Dev. 2015;10:6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 119.Cho H-J, Lee J-G, Kim J-Het al. Vascular defects of DYRK1A knockouts are ameliorated by modulating calcium signaling in zebrafish. Dis Model Mech. 2019;12:dmm037044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 120.Liu K, Petree C, Requena Tet al. Expanding the CRISPR toolbox in zebrafish for studying development and disease. Front Cell Dev Biol. 2019;7:13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 121.Hoshijima K, Jurynec MJ, Klatt Shaw Det al. Highly efficient CRISPR-Cas9-based methods for generating deletion mutations and F0 embryos that lack gene function in zebrafish. Dev Cell. 2019;51:645–657.e4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 122.Rossi A, Kontarakis Z, Gerri Cet al. Genetic compensation induced by deleterious mutations but not gene knockdowns. Nature. 2015;524:230–3 [DOI] [PubMed] [Google Scholar]
- 123.Celik A, Kervestin S, Jacobson A. NMD: at the crossroads between translation termination and ribosome recycling. Biochimie. 2015;114:2–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 124.El-Brolosy MA, Kontarakis Z, Rossi Aet al. Genetic compensation triggered by mutant mRNA degradation. Nature. 2019;568:193–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 125.Ma Z, Zhu P, Shi Het al. PTC-bearing mRNA elicits a genetic compensation response via Upf3a and COMPASS components. Nature. 2019;568:259–63 [DOI] [PubMed] [Google Scholar]
- 126.Buglo E, Sarmiento E, Martuscelli NBet al. Genetic compensation in a stable slc25a46 mutant zebrafish: a case for using F0 CRISPR mutagenesis to study phenotypes caused by inherited disease. PLoS One. 2020;15:e0230566 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 127.Eisen JS, Smith JC. Controlling morpholino experiments: don’t stop making antisense. Development. 2008;135:1735–43 [DOI] [PubMed] [Google Scholar]
- 128.Tucker B, Richards RI, Lardelli M. Contribution of mGluR and Fmr1 functional pathways to neurite morphogenesis, craniofacial development and fragile X syndrome. Hum Mol Genet. 2006;15:3446–58 [DOI] [PubMed] [Google Scholar]
- 129.den Broeder MJ, van der Linde H, Brouwer JRet al. Generation and characterization of FMR1 knockout zebrafish. PLoS One. 2009;4:e7910 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 130.Bedell VM, Westcot SE, Ekker SC. Lessons from morpholino-based screening in zebrafish. Brief Funct Genomics. 2011;10:181–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 131.Bernier R, Golzio C, Xiong Bet al. Disruptive CHD8 mutations define a subtype of autism early in development. Cell. 2014;158:263–76 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 132.Sugathan A, Biagioli M, Golzio Cet al. CHD8 regulates neurodevelopmental pathways associated with autism spectrum disorder in neural progenitors. Proc Natl Acad Sci U S A. 2014;111:E4468–77 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 133.Wu RS, Lam II, Clay Het al. A rapid method for directed gene knockout for screening in G0 zebrafish. Dev Cell. 2018;46:112–125.e4 [DOI] [PubMed] [Google Scholar]
- 134.Prill K, Dawson JF. Homology-directed repair in zebrafish: witchcraft and wizardry? Front Mol Biosci. 2020;7:595474 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 135.Won M, Dawid IB. PCR artifact in testing for homologous recombination in genomic editing in zebrafish. PLoS One. 2017;12:e0172802 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 136.Collins RL, Glessner JT, Porcu Eet al. A cross-disorder dosage sensitivity map of the human genome. Cell. 2022;185:3041–3055.e25 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 137.Lek M, Karczewski KJ, Minikel EVet al. Analysis of protein-coding genetic variation in 60,706 humans. Nature. 2016;536:285–91 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 138.Won H, Lee H-R, Gee HYet al. Autistic-like social behaviour in Shank2-mutant mice improved by restoring NMDA receptor function. Nature. 2012;486:261–5 [DOI] [PubMed] [Google Scholar]
- 139.Garrity DM, Childs S, Fishman MC. The heartstrings mutation in zebrafish causes heart/fin Tbx5 deficiency syndrome. Development. 2002;129:4635–45 [DOI] [PubMed] [Google Scholar]
- 140.Swartz ME, Lovely CB, Eberhart JK. Variation in phenotypes from a bmp-Gata3 genetic pathway is modulated by Shh signaling. PLoS Genet. 2021;17:e1009579 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 141.Nutt SL, Busslinger M. Monoallelic expression of Pax5: a paradigm for the haploinsufficiency of mammalian Pax genes? Biol Chem. 1999;380:601–11 [DOI] [PubMed] [Google Scholar]
- 142.Seidman JG, Seidman C. Transcription factor haploinsufficiency: when half a loaf is not enough. J Clin Invest. 2002;109:451–5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 143.Veitia RA. Exploring the etiology of haploinsufficiency. BioEssays. 2002;24:175–84 [DOI] [PubMed] [Google Scholar]
- 144.Morrill SA, Amon A. Why haploinsufficiency persists. Proc Natl Acad Sci U S A. 2019;116:11866–71 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 145.Rice AM, McLysaght A. Dosage sensitivity is a major determinant of human copy number variant pathogenicity. Nat Commun. 2017;8:14366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 146.Brown KH, Dobrinski KP, Lee ASet al. Extensive genetic diversity and substructuring among zebrafish strains revealed through copy number variant analysis. Proc Natl Acad Sci U S A. 2012;109:529–34 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 147.Yin J, Brocher J, Fischer U, Winkler C. Mutant Prpf31 causes pre-mRNA splicing defects and rod photoreceptor cell degeneration in a zebrafish model for retinitis pigmentosa. Mol Neurodegener. 2011;6:56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 148.Bergeron SA, Carrier N, Li GHet al. Gsx1 expression defines neurons required for Prepulse inhibition. Mol Psychiatry. 2014;20:974–85. 10.1038/mp.2014.106 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 149.Thermes V, Grabher C, Ristoratore Fet al. I-SceI meganuclease mediates highly efficient transgenesis in fish. Mech Dev. 2002;118:91–8 [DOI] [PubMed] [Google Scholar]
- 150.Kawakami K, Shima A, Kawakami N. Identification of a functional transposase of the Tol2 element, an ac-like element from the Japanese medaka fish, and its transposition in the zebrafish germ lineage. Proc Natl Acad Sci U S A. 2000;97:11403–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 151.Suster ML, Sumiyama K, Kawakami K. Transposon-mediated BAC transgenesis in zebrafish and mice. BMC Genomics. 2009;10:477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 152.Kawakami K, Takeda H, Kawakami Net al. A transposon-mediated gene trap approach identifies developmentally regulated genes in zebrafish. Dev Cell. 2004;7:133–44 [DOI] [PubMed] [Google Scholar]
- 153.Scheer N, Campos-Ortega JA. Use of the Gal4-UAS technique for targeted gene expression in the zebrafish. Mech Dev. 1999;80:153–8 [DOI] [PubMed] [Google Scholar]
- 154.Distel M, Wullimann MF, Koster RW. Optimized Gal4 genetics for permanent gene expression mapping in zebrafish. Proceedings of the National Academy of Sciences, USA. 2009;106:13365–70 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 155.Bartel WP, Van Laar VS, Burton EA. Parkinson’s Disease. In: Gerlai R, (ed.), Behavioral and Neural Genetics of Zebrafish. Academic Press, 2020, 377–412 [Google Scholar]
- 156.Wen L, Wei W, Gu Wet al. Visualization of monoaminergic neurons and neurotoxicity of MPTP in live transgenic zebrafish. Dev Biol. 2008;314:84–92 [DOI] [PubMed] [Google Scholar]
- 157.Auer TO, Duroure K, De Cian Aet al. Highly efficient CRISPR/Cas9-mediated knock-in in zebrafish by homology-independent DNA repair. Genome Res. 2014;24:142–53 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 158.Kimura Y, Hisano Y, Kawahara A, Higashijima SI. Efficient generation of knock-in transgenic zebrafish carrying reporter/driver genes by CRISPR/Cas9-mediated genome engineering. Sci Rep. 2014;4:6545. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 159.Subedi A, Macurak M, Gee STet al. Adoption of the Q transcriptional regulatory system for zebrafish transgenesis. Methods. 2014;66:433–40 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 160.Burgess J, Burrows JT, Sadhak Ret al. An optimized QF-binary expression system for use in zebrafish. Dev Biol. 2020;465:144–56 [DOI] [PubMed] [Google Scholar]
- 161.Carney TJ, Mosimann C. Switch and trace: recombinase genetics in zebrafish. Trends Genet. 2018;34:362–78 [DOI] [PubMed] [Google Scholar]
- 162.Satou C, Kimura Y, Hirata Het al. Transgenic tools to characterize neuronal properties of discrete populations of zebrafish neurons. Development. 2013;140:3927–31 [DOI] [PubMed] [Google Scholar]
- 163.Gu H, Marth JD, Orban PCet al. Deletion of a DNA polymerase beta gene segment in T cells using cell type-specific gene targeting. Science. 1994;265:103–6 [DOI] [PubMed] [Google Scholar]
- 164.Kang HJ, Kawasawa YI, Cheng Fet al. Spatio-temporal transcriptome of the human brain. Nature. 2011;478:483–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 165.Bradford YM, Van Slyke CE, Ruzicka Let al. Zebrafish information network, the knowledgebase for Danio rerio research. Genetics. 2022;220:iyac016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 166.Mi H, Muruganujan A, Thomas PD. PANTHER in 2013: modeling the evolution of gene function, and other gene attributes, in the context of phylogenetic trees. Nucleic Acids Res. 2013;41:D377–86 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 167.Hossainian D, Shao E, Jiao Bet al. Quantification of functional recovery in a larval zebrafish model of spinal cord injury. J Neurosci Res. 2022;100:2044–54. 10.1002/jnr.25118 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article will be shared on reasonable request to the corresponding author.
