Share this article.

Small but perfectly formed: how zebrafish model the human brain

ArticleQ&ADetailDownload pdf
For novel medical advances, such as in regenerative medicine, animal models are a key first stage of research. However, for subtle conditions such as visual impairment, the effectiveness of treatments is often difficult to measure using animal models. Robbert J. Creton, Professor at Brown University has developed a powerful system that uses the behaviour of zebrafish as a highly accurate proxy of their visual acuity. His high-throughput techniques could speed up advances in the treatment of many developmental disorders.
Almost 40 million people in the world are blind, and another 250 million visually impaired. Visual impairment has many potential causes, but at least some may conceivably become curable in the recent future through ground-breaking developments in regenerative medicine, enabling doctors to rebuild neural connections in both the eye and brain.
Few regenerative approaches have yet reached the stage of human clinical trials, so the scientists studying them must use model organisms, sharing some features of the human eye and brain, to conduct their research. A key question facing these researchers is how to monitor the success of new treatments in living animals, i.e., how to assess vision before and after treatment.

Animal behaviour is a good proxy for vision: behavioural changes can reveal subtle differences even when the eye appears structurally normalQuote_brain

particle analysis using ImageJ
The unique method developed by the Creton lab automatically analyses both the location and orientation of individual zebrafish larvae in multiwell plates. Larvae are detected with a threshold for dark objects (red) and a particle analysis using ImageJ, a freely available open-source software package. Visual stimuli can be shown to the larvae on a computer screen.
Professor Robbert Creton is pioneering the use of a model system using the behaviour of zebrafish larvae to evaluate treatments for visual impairment. He believes animal behaviour is a good proxy for vision, since behavioural changes can reveal subtle functional or physiological differences even when the eye appears structurally normal. This led him to the hypothesis, explored in his current National Institutes of Health project, that automated analysis of zebrafish behaviour is an effective, sensitive and non-invasive way to identify visual defects and monitor recovery from them.
From the Himalayas to the lab
A small, stripy tropical fish native to the streams of the southern Himalayas, the zebrafish has been at the forefront of vertebrate research since the 1970s and is now a well-established developmental model, although its use in behavioural studies is relatively new. It is ideally suited to studying vision, since the development, genetics and physiology of its visual system are very similar to those of humans: like us, zebrafish are diurnal and see in colour. Unlike us, they show a remarkable capacity to regenerate damaged parts of their visual system, and are cheap and easy to keep in the lab.
Calcium imaging in a zebrafish embryo
Calcium imaging in a zebrafish embryo. Far left: Brightfield image; Fluorescence imaging using a green fluorescent calcium indicator and a red fluorescent control; Overlay of brightfield and fluorescence; Ratiometric analysis showing low calcium in blue and high calcium in red (far right). These calcium signals are important for early embryonic development, neural function and vision.
Zebrafish exhibit defined behavioural responses to visual cues, which evolved to help them seek food and avoid predators. For example, the ‘optokinetic’ response causes the eyes of immobilised larvae to track a rotating pattern. The ‘optomotor’ response causes them to swim in the same direction as a moving, coloured stripe. Both responses can be abolished by damage to the eye, and restored as the visual system regenerates. They have already been used to identify a wide range of genes implicated in vision.
Automating the process
Another advantage of zebrafish is the potential to scale up and automate analyses of their behaviour. This perfectly taps into the quest of modern biological research to achieve ‘big data,’ by computerised, high-throughput analyses of vast numbers of organisms to rapidly and objectively draw out patterns and correlations.
Kaede fluorescence in eye and optic tectum of a 7 day-old zebrafish larva
Kaede fluorescence in eye and optic tectum of a 7 day-old zebrafish larva. Kaede fluorescence was converted from green to red at 5 days post-fertilisation by UV illumination. This procedure allows one to image older neurons in red and newly-formed neurons in green.
Neural networks in the developing zebrafish brain
Neural networks in the developing zebrafish brain. Neurons were labeled with an acetylated tubulin antibody and imaged by confocal microscopy. The acquired images were used for 3D reconstruction and visualisation of neural patterns in the brain.
A single tank of zebrafish can produce a cohort of hundreds of synchronously developing larvae, every single day. Crucially, these larvae are small enough to analyse in a multiwell plate – the standard tool of analytical and diagnostic laboratories worldwide (alternatively, for evaluation of the optomotor response, the system of choice is a set of multiple lanes resembling a tiny racetrack). Larvae just four or five days old can respond to moving, coloured lights on a screen, and their behaviour can be analysed as their vision is destroyed and then regained over a matter of days.
Prof Creton’s lab is working to scale up and optimise these imaging protocols, finding creative ways to solve problems such as the difficulty of imaging the tiny larvae when they get into the corners of the wells. The system they have developed over the past decade started life as a means to study the development of asymmetrical body parts and behaviours, but now fills a crucial gap in many studies, between high-throughput screening technologies and detailed analyses of behaviour. It can subject the fish to increasingly complex, coloured and mobile visual stimuli, and is unique in its ability to analyse automatically both the location and orientation of multiple, individual zebrafish larvae using multiwell plates.
Shedding light on visual impairment
Prof Creton’s current NIH project aims to further advance the automated analysis system, and to use it to identify changes in zebrafish behaviour due to specific visual defects, by comparing healthy zebrafish with strains known to have genetic mutations causing defects in different parts of the eye. Then, he will characterise the behavioural changes that accompany recovery from these defects. Ultimately, the tools developed will enable rapid screening for both genetic and environmental factors that may cause visual impairment, and comparison of the effectiveness of potential treatments.
Zebrafish larva at 7 days post-fertilisation
Zebrafish larva at 7 days post-fertilisation. The larva is only 5 mm long, but has well-developed organs, including eyes, a heart and a brain.
The zebrafish system has already proved useful for a huge range of neurological, developmental, behavioural and environmental studies in collaboration with researchers from a wide range of disciplines. Among the growing ‘Zebrafish at Brown’ research group, teams are using the system to shed light on the biological mechanisms underlying brain development, and contributing to the prevention and treatment of conditions including not just blindness, but developmental brain disorders and Alzheimer’s disease.

Ultimately, the tools developed will enable rapid screening for both genetic and environmental factors that may cause visual impairmentQuote_brain

Prof Creton’s most recent proposal plans to optimise the system to study signalling in the vertebrate brain by calcineurin, a molecule implicated in the behavioural features of Down Syndrome and Alzheimer’s Disease. Other future projects will expand to studies of the neural, digestive and muscular systems.
Of course, zebrafish are not people, and Prof Creton is quick to point out that complementary data from multiple model animals as well as human studies must be integrated to provide a full understanding of development. His overall goal is to identify clinically-relevant treatments in the zebrafish system, explore these further in mammalian models such as mice, and finally bring them to clinical trials before rolling out new treatments. However, the statistical power provided by the sheer numbers of zebrafish larvae that can be analysed simultaneously through automated analyses makes this an extremely promising approach.

What first drew you to zebrafish as a potential model organism to use in your research?
When I first looked at a zebrafish embryo on a light microscope, I was amazed by its transparency and rapid development. The embryo was only two days old and it was possible to see two large eyes, distinct brain regions, a beating heart and blood cells flowing through the cardiovascular system. In addition, it was easy to collect fertilised eggs from the bottom of the tank and to grow the embryos in a culture dish. The embryo’s transparency and rapid external development are still used in my current research to examine how specific neurons in the brain control behaviour.
What forms of zebrafish behaviour have proven most useful for testing visual defects?
Zebrafish larvae quickly swim away from large moving objects, including moving objects that are displayed on a computer screen. The larvae have full colour vision and show a particularly strong response to red objects, although responses to green and blue objects can be measured as well. These avoidance behaviours are easily quantified and can be used to measure an overall loss of vision, as well as specific defects in colour vision.
Visual impairment has so many different causes; which ones do you think we may be able to treat?
While prevention is often more successful than treatment, the zebrafish model may be particularly useful in finding novel treatments for the most challenging cases of visual impairment, in which photoreceptors or neural connections are lost. We are currently working on photoreceptor damage and were able to measure a loss and recovery of vision in zebrafish larvae. A better understanding of the underlying mechanisms will help in designing novel treatments for photoreceptor damage in humans.
How can you use zebrafish to model Alzheimer’s disease?
Various studies suggest that calcineurin inhibitors have therapeutic potential in the prevention of Alzheimer’s disease. For example, transplant patients who use calcineurin inhibitors to prevent organ rejection rarely develop Alzheimer’s disease, even when patients get older. The automated analysis of zebrafish behaviour may be used to optimise such calcineurin-based therapies. It is possible to examine effects of calcineurin activation and inhibition in a high-throughput format and screen for novel drugs that inhibit calcineurin signalling in the brain, while avoiding side effects in other organ systems.
Where do you see your research going over the next decade?
I believe that high-throughput analyses will become increasingly important, even when studying complex systems such as the eye and brain. Rather than studying one gene, one drug, or one environmental toxicant, it may become possible to screen thousands of factors in integrated signalling networks. The development and use of high-throughput technologies will provide a more holistic understanding of biological processes and will facilitate the discovery of novel treatments of visual impairment as well as other disorders.
Research Objectives
Professor Robbert Creton’s research group is focused on brain development and behaviour, using zebrafish as a model system. The lab seeks to provide a better understanding of basic biological mechanisms, with the aim of preventing and treating disorders including visual impairment, developmental brain disorders and Alzheimer’s disease.

  • National Institutes of Health (NIH)


  • Dr Ruth Colwill, Professor of Cognitive, Linguistic and Psychological Sciences, Brown University
  • Dr Jason Sello, Associate Professor of Chemistry, Brown University

Dr Robbert CretonRobbert Creton received his B.S. in Biology (1990) and Ph.D. in Developmental Biology (1994) from the University of Utrecht in the Netherlands. He currently studies brain development and visually-guided behaviours at Brown University. He is the director of the Leduc Bioimaging Facility and the director of the Molecular Pathology Core.
Professor Robbert J Creton
Professor of Medical Science (Research)
Brown University
185 Meeting Street
Providence, Rhode Island 02912
E: [email protected]
T: +1 401 863 9646

Creative Commons Licence

(CC BY-NC-ND 4.0) This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. Creative Commons License

What does this mean?
Share: You can copy and redistribute the material in any medium or format
Related posts.