Periodic Reporting for period 1 - LivAdapt (Transcriptional adaptation during vertebrate development at the single-cell level)
Periodo di rendicontazione: 2024-04-01 al 2026-03-31
The central motivation of this project was to close this gap by developing, for the first time, a system capable of imaging translation live and in real time within an intact vertebrate embryo. The zebrafish (Danio rerio) was chosen as the model organism of choice: it is genetically tractable, produces optically transparent embryos amenable to high-resolution microscopy, and develops rapidly, making it uniquely suited to live imaging approaches. Prior to this work, live imaging of translation had been demonstrated in unicellular organisms and in cell culture systems, but never in a vertebrate embryo — leaving a fundamental question unanswered: how is protein synthesis regulated in space and time during the development of a complex animal?
The project was initially designed around the phenomenon of Transcriptional Adaptation (TA), a form of genetic compensation whereby the mutation or loss of a gene triggers the upregulation of related genes, buffering the organism against genetic perturbation. TA is of broad biological and biomedical relevance, as it underlies many cases where gene knockout animals fail to display the expected phenotypes — a common and poorly understood phenomenon in genetics. The original objectives aimed to dissect TA at the levels of transcription, mRNA localisation, and translation, using live imaging approaches in zebrafish. However, when experimental evidence indicated that the TA response could not be reliably recapitulated under the conditions tested, the project was strategically reoriented — a decision that ultimately led to a stronger and broader scientific outcome.
The reorientation focused on deploying the newly developed live translation imaging platform to study bmp2b, a gene encoding a key morphogen involved in dorsoventral patterning and organogenesis. BMP2 acts as a signalling gradient across the embryo, and while its distribution at the mRNA and protein levels had been described, nothing was known about how its translation is regulated in vivo. By applying the new imaging system to this biologically important gene, the project was able to reveal, for the first time, translational regulation of a morphogen gradient in a living vertebrate embryo — including control at the level of translation initiation and in the proportion of mRNAs actively engaged with ribosomes.
Beyond this primary finding, the project produced a significant and unexpected discovery: evidence of non-canonical translation occurring in the zebrafish embryo. This refers to translation initiation through mechanisms that deviate from the classical cap-dependent pathway, a phenomenon previously associated with stress responses, viral infection, and cancer, but never before described in the context of normal vertebrate embryogenesis. This discovery opens an entirely new research avenue and reframes fundamental questions about how protein synthesis is controlled during development.
The expected impact of this project operates on multiple levels. Technologically, it delivers a versatile live imaging platform — comprising multiple transgenic zebrafish lines with different fluorophores — that can be adopted by the broader zebrafish and vertebrate biology community to study the translation of any gene of interest in a living embryo. The spontaneous requests for tool access already received from two internationally leading research groups (the Pauli lab, Vienna, and the Raz lab, Münster) reflect the community-wide relevance of this resource. Scientifically, the results provide the first direct view of translational regulation during vertebrate embryogenesis, and uncover a previously unknown layer of translational control whose implications extend from developmental biology to disease.
The central technological achievement of this project was the establishment of the first system capable of imaging translation in real time in an intact vertebrate embryo. This required overcoming substantial technical challenges, as no existing platform was directly transferable to the zebrafish embryo context.
A first attempt was made using the SunTag system, a widely used approach for translation imaging in cell culture. However, this strategy proved incompatible with the zebrafish system: the single-chain antibody (scFv) recognising the SunTag epitope showed signs of toxicity, as evidenced by very low efficiency in generating stable transgenic lines, and formed spontaneous aggregates that precluded reliable signal detection. This approach was therefore abandoned in favour of an alternative strategy.
A new system was developed based on the ALFA-tag and its cognate Nanobody. This approach proved highly compatible with the zebrafish embryo context. Multiple stable transgenic lines were generated, each expressing the Nanobody fused to a different fluorophore, making the platform versatile and combinable with a wide range of existing tools in the zebrafish community. The small size of the ALFA-tag and its tolerance in vivo made it particularly well suited for endogenous tagging strategies. These lines are now being adopted by other members of the host laboratory and have been requested by external groups, reflecting their broad utility.
Optimisation of the imaging setup
Establishing the biological tools was only one part of the challenge — achieving the spatial and temporal resolution required to detect individual translation events in a living embryo demanded equally significant work on the imaging side. Classical confocal microscopy was initially tested but found to be insufficient in terms of resolution and acquisition speed for this application. Following an evaluation of multiple state-of-the-art imaging systems at the Zeiss Demo Center in Oberkochen, Germany, the lattice light-sheet microscope was identified as the only platform capable of meeting the requirements of this project.
A collaboration was initiated with the D. Muriaux team at the Institut de Recherche en Infectiologie de Montpellier (IRIM), which operates one of the few facilities in France equipped with this technology in proximity to a fish facility. Extended experimental visits to this facility enabled the optimisation of the full imaging pipeline, from embryo mounting and sample preparation to image acquisition and downstream quantitative analysis. This work was carried out in close collaboration with the institute's imaging platform, in particular with Kenny Mattonet, whose contribution to image analysis was instrumental. The result was a robust and reproducible protocol for live translation imaging in intact zebrafish embryos — an achievement with no prior equivalent in vertebrate systems.
Visualisation of transcription and single mRNA molecules
In parallel with the translation imaging work, significant effort was invested in developing tools for live transcription imaging. Two transgenic lines expressing the MS2 coat protein (MCP-msGFP and MCP-Scarlet) were generated, representing optimised versions of previously published constructs. Using these lines in combination with the lattice light-sheet microscope, active transcription sites were visualised in live zebrafish embryos, and individual mRNA molecules were tracked in real time within a living vertebrate embryo for the first time. These results are included in the revised manuscript currently under review at Science Advances.
Biological application: translation dynamics of BMP2b
The live translation imaging platform was applied to bmp2b, a gene encoding a morphogen essential for dorsoventral patterning and organogenesis in the zebrafish embryo. Transgenic lines expressing bmp2b tagged with the ALFA-array system were generated, and live imaging of translation was performed across the embryo. Quantitative analysis, carried out in collaboration with the physics teams of T. Stasevich and T. Morisaki at the University of Colorado, enabled measurement of translation kinetics with single-molecule resolution. The results revealed that BMP2b translation is regulated at the level of initiation, and that the fraction of bmp2b mRNAs actively engaged in translation varies across the embryo. These findings provide the first direct characterisation of morphogen translation dynamics in a living vertebrate, and shed new light on the mechanisms controlling gradient formation during embryonic patterning.
An unexpected discovery: non-canonical translation in the vertebrate embryo
In the course of these experiments, an entirely unanticipated observation was made: evidence of non-canonical translation occurring in the zebrafish embryo. This refers to translation initiation through mechanisms that deviate from the classical cap-dependent pathway — a phenomenon previously described in contexts such as viral infection, cellular stress, and cancer, but never before reported during normal vertebrate embryogenesis. This finding emerged from the quantitative analysis of translation events and was not part of the original project design. It constitutes a significant and novel contribution to the field, raising fundamental questions about the extent and role of non-canonical translation during early vertebrate development. This discovery is reported in the manuscript under revision at Science Advances and will form a central focus of future independent research.
Work on Transcriptional Adaptation
Although the original objectives centred on the Transcriptional Adaptation (TA) response could not be fully achieved, substantial experimental work was carried out in this direction. Multiple transgenic lines were generated to visualise and quantify the expression of candidate adapting genes (aldh1a3, vclb, alcamb) and their mutant counterparts, including constructs with and without introns to assess the potential role of splicing. Expression of adapting genes was assessed by single-embryo genotyping followed by RT-qPCR at multiple developmental stages, but no significant or reproducible upregulation could be detected. The absence of a reliable TA response under all tested conditions, including in non-transgenic mutant backgrounds, indicated a fundamental experimental constraint rather than a technical failure. This conclusion prompted the strategic reorientation described above, which ultimately led to a scientifically stronger outcome.
The most significant result of this project is the establishment of the first platform capable of imaging translation in real time in an intact vertebrate embryo. Prior to this work, live imaging of translation had been demonstrated exclusively in unicellular organisms such as yeast and bacteria, and in mammalian cell culture systems. The leap to a living vertebrate embryo represents a qualitative change in what is experimentally accessible, not merely an incremental improvement. The zebrafish embryo is orders of magnitude more complex than a cell in culture — it is a three-dimensional, multicellular organism undergoing active patterning, differentiation, and morphogenesis — and imaging translation within it with single-molecule resolution required overcoming challenges that had no established solution in the field.
This achievement places the project at the frontier of live imaging in developmental biology, and defines a new benchmark for what can be observed directly in a vertebrate system. The tools developed — multiple stable transgenic zebrafish lines based on the ALFA-array Nanobody system, combined with an optimised lattice light-sheet imaging pipeline — constitute a reusable and transferable resource that extends well beyond the immediate scientific questions addressed here. Any gene of interest can in principle be tagged and its translation imaged using this platform, making it applicable across a wide range of biological questions in vertebrate development.
First direct measurement of morphogen translation dynamics in vivo
The application of this platform to bmp2b yielded the first direct quantitative characterisation of morphogen translation dynamics in a living vertebrate embryo. While the BMP2 signalling gradient had previously been described at the levels of mRNA distribution and protein accumulation, the translational step — how, where, and at what rate mRNA molecules are converted into protein — had remained entirely inaccessible. The results reveal that bmp2b translation is regulated at the level of initiation, and that the proportion of mRNAs actively engaged with ribosomes varies spatially across the embryo. These findings add a previously missing layer to our understanding of how morphogen gradients are established and maintained, with implications for models of embryonic patterning more broadly.
This result advances the state of the art by demonstrating that translational regulation is not merely a downstream consequence of mRNA distribution, but an active and spatially controlled process that contributes independently to gradient formation. This conceptual advance is likely to stimulate renewed interest in the translational dimension of morphogen biology, a dimension that has been largely neglected due to the absence of tools capable of addressing it in vivo.
Discovery of non-canonical translation in the vertebrate embryo
Perhaps the most far-reaching result of the project is the discovery of non-canonical translation occurring in the zebrafish embryo during normal development. Non-canonical translation — whereby ribosomes initiate protein synthesis through mechanisms that bypass the classical cap-dependent pathway — had previously been described in the context of viral infection, integrated stress responses, and cancer biology. Its occurrence during normal vertebrate embryogenesis had not been reported, and was not anticipated at the outset of this project.
This discovery is significant for several reasons. It reveals an unexpected layer of translational control operating during early vertebrate development, at a stage when the regulation of gene expression is particularly consequential for cell fate and tissue patterning. It raises fundamental questions about which mRNAs are subject to non-canonical translation, under what conditions, and with what functional consequences for the embryo. It also opens the possibility that mechanisms previously studied exclusively in pathological contexts may play physiological roles during normal development — a conceptual reframing with broad implications.
The identification of non-canonical translation in this context was made possible only by the quantitative, single-molecule resolution of the imaging platform developed here. It exemplifies the kind of discovery that becomes accessible when genuinely new observational tools are introduced into a field — results that could not have been anticipated, and that redefine the questions the field will need to address.
Potential impacts and pathways to further uptake
The results of this project are positioned to have impact across several interconnected domains.
In the field of developmental biology, the platform enables a new class of experiments addressing translational regulation in vivo, with direct applicability to questions of morphogen gradient formation, cell fate specification, and embryonic patterning. The discovery of non-canonical translation opens a specific new research programme that will be pursued as the core focus of my new position next.
In the RNA biology community, the tools and results are immediately relevant to researchers studying translational control, mRNA regulation, and the coupling between transcription and translation. The platform is compatible with existing zebrafish transgenic resources, lowering the barrier to adoption for any group already working in this system.
For further uptake, the key needs are the dissemination of both the biological materials and the associated protocols to the community. This is already underway: two leading international groups — the Pauli lab in Vienna and the Raz lab in Münster — have independently requested access to the transgenic lines and imaging tools developed during this project. The forthcoming publication in Science Advances, combined with the existing preprint, will provide the community with the full methodological framework required for independent replication and adaptation.
No intellectual property barriers are anticipated, and no patent applications have been filed, reflecting a deliberate commitment to open science and maximal community access. The transgenic lines will be made available through standard biological material transfer mechanisms. Further dissemination will be supported through the workshop on live imaging of translation organised at Mifobio2025, and through ongoing scientific communication via conference presentations and collaborative networks.
In the longer term, the discovery of non-canonical translation in the vertebrate embryo may attract interest beyond the developmental biology community, given the established links between non-canonical translation mechanisms and disease contexts including cancer and viral pathogenesis. Understanding the physiological role of these mechanisms during normal development may ultimately inform our understanding of how they are dysregulated in pathological settings — a translational research perspective that could support future applications in biomedical research.