Zebrafish in Translation
Written by Yevgenya Grinblat
We continue our “Zebrafish in Translation” feature with a spotlight on the work of Jean Giacomotto, a tenured Group Leader (Senior Lecturer and NHMRC Research Fellow) at Griffith University in Brisbane, Australia. A study from his lab, just featured on the front cover of EMBO Molecular Medicine, is a beautiful illustration of the power of zebrafish to model disease and inform clinical practice (Stringer et al., 2026)
To start with, could you briefly explain the focus of this study and what motivated you to undertake it?
My lab works on an early childhood disease, spinal muscular atrophy (SMA) that causes muscle atrophy, paralysis and death. SMA is known to be caused by mutations in SMN1, and current newborn screening programs routinely identify infants carrying SMN1 variants. For a subset of these variants, their pathogenicity remains uncertain.
Disease-modifying therapies for SMA are now available and can be highly effective, particularly when administered early, before irreversible motor neuron loss has occurred. As a result, clinical decisions regarding treatment—often involving extremely costly therapies with potential risks—must sometimes be made in the absence of clear functional evidence indicating whether the specific SMN1 variants are truly disease-causing or not.
In my lab, we’ve developed a variety of zebrafish models of smn1-linked SMA, including a model originally based on a smn1 mutation generated by Christine Beattie (Boon et al., 2009). We used these models to test SMN1 variants of uncertain significance (VUS) newly identified in two infants. We overexpressed each variant in smn1-deficient zebrafish and evaluated their ability to rescue the disease, which normally progresses rapidly and results in larval death by 5 dpf. Using this assay, we found that both variants were functional. Importantly, the test took only a few weeks to complete, with several replicates, so our clinical colleagues and the parents had this information in plenty of time for the big decision.
Based on our findings in zebrafish, the team decided not to initiate drug treatments. The babies were 4 and 6 months old at the time, and we wouldn’t know for another 6 months if we’d made the correct decision. The wait was really stressful! And now, at 24 months, both children are perfectly healthy and happy, with neuromuscular functions at the high end of the normal range.
That’s really inspirational. How are you planning to build on these findings?
We’ve continued to dive deeper into characterizing the two novel variants using zebrafish, as there remains a possible risk of a late-onset form of the disease. We also used this opportunity to optimize our approach for detecting hypomorphic variants. Part of these results are presented in the final version of our EMBO article, with additional data to be reported later this year. This includes a follow-up study -incorporating our clinical data and epidemiological analyses- that is nearly accepted in the American Journal of Human Genetics and should be coming out in March.
What are your long-term plans for disease modeling in zebrafish? I saw that your lab has begun modeling schizophrenia— quite an ambitious undertaking. Could you tell us a bit about that approach?
Yes, while we are working hard on a variety of other neurological and pediatric diseases, we’ve also recently been working on modelling late-onset diseases. We started with the zebrafish neurexin gene family because they’ve been linked to schizophrenia, autism and other neurological conditions. We’ve generated mutant lines, initially expecting to observe developmental defects in the CNS. Despite careful examination of mutant brains for obvious deficits, we found that neurexin mutants instead develop into healthy adults. However, despite this absence of overt early circuitry changes, they later develop pronounced adult behavioral defects, including marked social withdrawal, depression-like behaviors, and, in some cases, seizure-like or epileptic-like episodes. The significant social withdrawal observed in the neurexin-1 lines is particularly interesting, as it is reminiscent of the negative symptoms of schizophrenia. You can read about it here.
The best known symptom of schizophrenia is auditory hallucinations. My first reaction when clinicians suggested a model of schizophrenia in zebrafish was disbelief. How could we possibly know if a fish is “hearing voices”? However, thanks to recent advances in imaging, we can now use calcium imaging to look directly into the brains of these animals. And despite a normally developing brain at early age, imagine if we could observe their tiny brain to activate sound-associated circuitries in the absence of acoustic signals. This is a work in progress.
Talking about this brings me back to why I got interested in drug discovery in the first place. Most antipsychotics used to treat schizophrenia today were discovered in the 1950s and 1960s, and in the 70 years since then, we haven’t seen truly new drugs that work in fundamentally different ways. Randy Peterson and David Kokel have made this point very clearly: traditional drug discovery pipelines are too target focused. If you don’t work in an intact organism—or at least in a system that captures multiple interacting targets—you’refocused. If you don’t work in an intact organism—or at least in a system that captures multiple interacting targets—you’re unlikely to discover something genuinely new in that field. And that’s exactly where we are now. After seven decades, we’re still relying on variations of the same drug scaffolds that were identified all those years ago to look for better treatments for schizophrenia. I am hoping zebrafish models like ours will open a way to change that.
One of the motivations for this series of interviews is to offer a roadmap for students and postdocs interested in translational science. Could you walk us through your career path from the start to the present?
I started my career at Sanofi, a large pharma company, where drug discovery was primarily based on screening compounds that can bind to, or modulate known molecular targets. For my Ph. D., which I completed in association with the CNRS in France, I decided to shift from target-centered to phenotype-based drug discovery using C. elegans. We developed several disease models, including models of Duchenne muscular dystrophy and spinal muscular atrophy, and genetically engineered them to be compatible with the same high-throughput automated pipelines used at Sanofi. These models exhibited robust and quantifiable phenotypes suitable for high-throughput screening –the progress of these diseases in the worm simply translated into progressive loss of motor function– and that led to the discovery of compounds that ended up being translated into preclinical and clinical studies.
About the same time, the pioneering work of Randy Peterson on chemical screening using zebrafish, together with Peter Currie’s work on muscular dystrophy, strongly influenced my thinking. I became increasingly convinced that phenotypic screening was underused in the field—not only as a way to identify compounds for translation to patients, but also as a powerful approach to interrogate disease mechanisms using existing drug libraries with well-characterized mechanisms of action. This realization motivated my move fully into academia and the start of my postdoctoral work.
Since then, I have continued to build and engineer robust genetic disease models and to combine pharmacology and genetics to both treat disease phenotypes and gain fundamental insight into human biology and disease mechanisms. Building and engineering physiologically relevant disease models for phenotypic screening—including zebrafish models of spinal muscular atrophy—has been the central driving force of my research ever since.
Coming from a more industrial background, I did not initially appreciate how challenging it can be to secure funding in academia, nor did I begin my career with a strong publication-driven mindset—both of which I later learned are critical for long-term academic success. Fortunately, I had the opportunity to interact closely with clinicians early on, and this turned out to be really valuable.
With the recent emergence of precision medicine and the rapid expansion in the identification of disease-associated gene variants, I strongly believe that zebrafish will play an increasingly important and supportive role in clinical research settings. In the mid- to long term, this should create new career paths and funding opportunities for young scientists. While there remains, at least in my experience, some hesitation in the field regarding the utility and value of zebrafish as a preclinical model, I hope that studies such as our recent work published in EMBO Molecular Medicine help demonstrate what this model and community can offer to translational research.
As for early career scientists with an interest in translational research, my advice is to join labs that are embedded with clinicians, or labs that are headed by clinicians with access to patients. In such environments, meaningful translational opportunities are far more likely to emerge.
What do you see as the next goals for the field and what are the limitations?
Expanding close collaborations with clinicians is one of the most important next steps for the field. The logistics are challenging and barriers exist but are definitely worth overcoming. Shared databases or platforms that actively facilitate these connections would help and we should invest in building this infrastructure. We should also publish these collaborations, so their impact is visible. And speaking that, the costs of publishing have become unsustainably high, and that needs to change if we want translational, collaborative science to thrive.
References:
Stringer, B. W. et al. (2026) Clinical relevance of zebrafish for gene variants testing. Proof-of-principle with SMN1/SMA. EMBO molecular medicine 18, 41-54.
Boon, K. L. et al. (2009) Zebrafish survival motor neuron mutants exhibit presynaptic neuromuscular junction defects. Hum Mol Genet 18, 3615-3625.