To integrate the dynamics of transcription, my team has and continues to optimize current live imaging (MS2-MCP), single molecule labelling (smiFISH) and multiplex genomic DNA painting (Hi-M) techniques to adapt them to a multicellular organism. Technological developments allowed us to characterize key features of transcriptional regulation during development summarized below.
1-We have been invited to contribute to a chapter dedicated to imaging gene expression (Book ‘Imaging Gene Expression’ edited by Pr. Y.Shav-Tal Springer editions)[Fernandez and Lagha 2019].
2-We have contributed to setting up the Hi-M technique in Drosophila embryos [Gizzi et al 2019] and then employed this single cell spatial genomics approach to decipher the impact of 3D contacts on transcriptional control [Espinola et al, 2021].
3-To extract quantitative data from these imaging-based approaches, we implemented specific image analysis software, able to segment nuclei and transcriptional spots and track them even during divisions [Trullo et al., 2019].
4-With these tools, we showed, for the first time in a multicellular developing organism, that the transcriptional status of mother nuclei biases that of their daughters after division, a phenomenon referred to as ‘transcriptional memory’ [Ferraro et al., 2016].
5-We showed that the pioneer factor Zelda is not essential for this memory and developed [Dufourt, Trullo et al., 2018] a modelling framework for transcriptional memory [Bellec et al., 2018].
6-We discovered that the transcriptional activator and pioneer factor GAF acts as a mitotic bookmarker with key functions during early development [Bellec et al., 2022] . For this study, we developed a protocol to perform ChIP-Seq from a pure population of mitotic embryos; embryonic mitotic ChIP has never been performed.
7-By imaging Zelda dynamic behaviour in living embryos, we discovered that Zelda proteins accumulate in nuclear hubs [Dufourt et al., 2018]. We then questioned the impact of these hubs on 3D genome organization during zygotic genome activation. Surprisingly, we found that 3D organization of cis-regulatory hubs is similar between cell types and occurs prior to the emergence of TADs in the embryo [Espinola et al., 2021].
8-We revealed the contribution of specific core promoter motifs on promoter state dynamics and how this creates coordinated gene expression in the developing drosophila embryo. In this study, we employed fast live imaging and an innovative machine-learning approach that allowed us to replace each individual polymerase initiation event for each nucleus in vivo. This analysis framework revealed that the INR promoter motif leads to an extra checkpoint of regulation associated with stochastic pausing, imposed on only a subset of polymerases. This novel view of non-obligatory pausing and its instruction by core promoter motifs, has fascinating implications for the control of gene expression variation in both development (during ZGA) and disease [Pimmett, Dejean et al., 2021] .
9-We implemented the Suntag labelling method in Drosophila, which allowed for the first time to visualize and quantify translational dynamics in a multicellular living embryo. By focusing on the translation of the conserved major epithelial-mesenchymal transition (EMT)-inducing transcription factor Twist, we identified spatial heterogeneity in mRNA translation efficiency and revealed the existence of translation factories, where clustered mRNAs are co-translated preferentially at basal perinuclear regions [Dufourt, Bellec et al., 2021].
