Temporal Control of Transcription by Zelda in living Drosophila embryos

Jeremy Dufourt, Antonio Trullo, Jennifer Hunter, Carola Fernandez, Jorge Lazaro, Matthieu Dejean, Lucas Morales, Katharine N Schulz, Melissa M Harrison, Ovidiu Radulescu, Cyril Favard, Mounia Lagha

Preprint posted on March 14, 2018

Measuring enhancers in motion: Quantitative live imaging coupled with modeling to study temporal control of gene expression and transcriptional memory.

Selected by Teresa Rayon

Categories: developmental biology


In this preprint, the Lagha lab studies the role of pioneer transcription factors on the temporal control gene expression. Pioneer factors are known for their ability to bind their sites when the chromatin is closed. However, how pioneer factors prime enhancers to control the time of gene activation, and how this priming affects gene expression after mitosis, remains elusive.  In the Drosophila embryo, the maternally provided Zelda pioneer factor is responsible for the awakening of the zygotic genome. In this preprint, the authors examine the enhancer expression dynamics of a Zelda responsive element, and the dynamic properties of Zelda in living Drosophila embryos, through two cycles of cell division. In particular, they measure the temporal coordination in Zelda activation through a synthetic regulatory element. The authors also test how the initial state of activity of the enhancer (memory) impacts on gene activation. By examining Zelda dynamics in living embryos during mitosis, they find that Zelda accelerates transcription and fosters synchrony but is not the basis for transcriptional mitotic memory.

Movie kindly provided by the authors showing the dynamics during various nuclear divisions of the snail shadow enhancer. MS2 transcription sites are in green (using MCP-GFP), and nuclei are labeled in red (using His2Av-RFP).


Why I chose the paper:

How gene regulation is precisely temporally controlled is one of the most interesting questions in developmental biology. One way to understand what’s going on is to characterize and study the dynamic activity of enhancers, and we need to find new ways to study how enhancers regulate gene expression.

This preprint offers a novel way to interpret enhancer dynamics through cell division and study transcriptional regulation by enhancers. The authors had previously shown how cells with memory activate transcription twice as fast as those with inactive mothers (1), but this time they focus on the role of Zelda in temporal coordination. I really like the fact that the authors are able to reproduce the activation kinetics of the endogenous shadow enhancer with their synthetic version that contains two Zelda binding sites. In addition, they implement an automated segmentation and tracking system that allows them to study hundreds of nuclei during multiple cell divisions. This system, once implemented, can save hours of tedious manual tracking, and provides a huge number of nuclei to analyse quantitatively. With this data, the authors move on to design a mathematical model to learn how memory could work. They model the time of reactivation after mitosis of cells with or without memory and find that inactive cells require at least three transitions to become active. The fitting to the model also shows that memory may be potentiated by reducing the number of steps required for post-mitotic transcriptional activation; something that couldn’t have been proposed without the model.

Unexpectedly, they find that Zelda does not play a role in setting the memory, since reducing Zelda levels leads to a reduction in activation kinetics but memory is still present. Neither does Zelda function as a mitotic bookmarking protein, since it is not retained on the chromosomes during mitosis. To pin down Zelda’s mode of action, the authors go one step further and perform fluorescent recovery after photo-bleaching and fluorescent correlation spectroscopy experiments. This thorough analysis of Zelda kinetic properties demonstrates that Zelda is highly dynamic and binds transiently to chromatin. The preprint concludes that Zelda accelerates the various transitions required prior to transcriptional activation.

How this work moves the field forward:

Regulatory sequences contain multiple transcription factor binding sites that are responsible for the spatio-temporal expression patterns of genes. We have long tried to decode the map to predict how the number, affinity and arrangement of sites determines transcription. This gets even more complicated if we consider that a single gene can respond to multiple enhancers, and these can exhibit overlapping spatiotemporal activity. This preprint advances our understanding of how enhancers coordinate gene expression, and emphasises how quantitative measurements coupled with mathematical modelling could provide precise and accurate models for transcription.

References :

  1. Transcriptional Memory in the Drosophila Embryo. Ferraro T, Esposito E, Mancini L, Ng S, Lucas T, Coppey M, Dostatni N, Walczak AM, Levine M, Lagha M. Curr Biol. 2016 Jan 25;26(2):212-8.
  2. Transcriptional precision and accuracy in development: from measurements to models and mechanisms. Bentovim L, Harden TT, DePace AH. Development. 2017 Nov 1;144(21):3855-3866.

Questions to the authors:

  1. In the model, there are two possible parameters that define the memory function, (1) the number of transitions between discrete transcriptional states and (2) the duration of each transition. Have the authors considered to describe/identify the three transitions and/or metastable states, and explore parameter ‘a’ further?
  2. I wonder if the authors could clarify further how the model supports their finding that the increase number of binding sites has an effect on the durations of the transitions and not on the number of states.

Tags: enhancer, fruit fly, gene expression, quantitative

Read preprint (No Ratings Yet)

  • Author's response

    Mounia Lagha shared

    In this preprint, we studied the dynamics of enhancer expression during early Drosophila embryogenesis. In particular, we tried to understand the role of the pioneer factor Zelda on temporal control of gene expression using quantitative live imaging. Our three main findings are:

    1. Zelda is able to critically accelerate transcriptional initiation following mitosis, regardless of the past transcriptional status. Using mathematical modelling we propose that Zelda is able to accelerate transitions between discrete transcriptional states. (A detailed analysis and discussion of the mathematical model will be provided in an independent manuscript by O.Radulescu). This result raised one question: what could these discrete transcriptional states correspond to?Although purely speculative, we think these steps could correspond to chromatin opening, transcription factor binding, and to the assembly of the pre-initiation complex. Further investigations in that direction would be very interesting.
    2. Zelda is not required for mitotic memory. As a pioneer factor, Zelda was a good candidate for this mitotic memory. But surprisingly in the absence of Zelda, mitotic memory persists.
    3. In spite of its role in fostering a rapid transcriptional activation, we reveal that Zelda is highly dynamic, with transient chromatin binding and is evicted during mitosis.  We also found that Zelda is able to form local hubs within the nucleus. Similarly to what has been recently shown for Ubx (Crocker lab) and Bcd (Darzacq lab), we propose that Zelda local aggregation can compensate for its transient binding properties. Further investigations would be required to show if Zelda local accumulation could lead to a phase transition type of compartmentalization within the fly blastoderm nucleus.

    Answers to the questions:

    1. Indeed it would very exciting to further characterize what these various discrete transcriptional states could correspond to. We are currently trying to examine the effect of varying the concentrations of one activator (dorsal) to examine its effect on the ‘a’ parameter of our model. Future quantitative analysis of pre-initiation complex mutants in trans and minimal promoter mutants in cis should help clarifying the nature of these steps.
    2. The estimated a and b parameter of the model are provided for each genotype in Table 1. We did not demonstrate nor stated that Zld binding sites increase had solely an effect on the durations of the transitions. However we do indeed see that the main striking effect of adding extra Zelda binding sites is on the reduction in b parameter, in other words in the acceleration of transitions. Given what is known about Zelda, we speculate that by locally opening chromatin, Zelda binding creates a permissive environment, where jumping from one step to another would be faster.

    Have your say

    Your email address will not be published. Required fields are marked *

    Sign up to customise the site to your preferences and to receive alerts

    Register here

    Also in the developmental biology category:

    A novel mechanism of gland formation in zebrafish involving transdifferentiation of renal epithelial cells and live cell extrusion

    Richard W Naylor, Alan J Davidson

    Selected by Giuliana Clemente


    An intrinsic cell cycle timer terminates limb bud outgrowth

    Joseph Pickering, Kavitha Chinnaiya, Constance A Rich, et al.

    Selected by Ashrifia Adomako-Ankomah


    Fbxw7 is a critical regulator of Schwann cell myelinating potential

    Breanne L Harty, Fernanda Coelho, Sarah D Ackerman, et al.

    Selected by Yen-Chung Chen

    Kinetic sculpting of the seven stripes of the Drosophila even-skipped gene

    Augusto Berrocal, Nicholas C Lammers, Hernan G Garcia, et al.

    Selected by Erik Clark

    TORC1 modulation in adipose tissue is required for organismal adaptation to hypoxia in Drosophila.

    Byoungchun Lee, Elizabeth C Barretto, Savraj S Grewal

    Selected by Sarah Bowling

    The Ly6/uPAR protein Bouncer is necessary and sufficient for species-specific fertilization

    Sarah Herberg, Krista R Gert, Alexander Schleiffer, et al.

    Selected by James Gagnon

    Functional testing of a human PBX3 variant in zebrafish reveals a potential modifier role in congenital heart defects

    Gist H. Farr III, Kimia Imani, Darren Pouv, et al.

    Selected by Hannah Brunsdon

    Mitotic chromosome alignment is required for proper nuclear envelope reassembly

    Cindy L Fonseca, Heidi LH Malaby, Leslie A Sepaniac, et al.

    Selected by Maiko Kitaoka

    Timed collinear activation of Hox genes during gastrulation controls the avian forelimb position

    Chloe Moreau, Paolo Caldarelli, Didier Rocancourt, et al.

    Selected by Wouter Masselink

    WNT signaling memory is required for ACTIVIN to function as a morphogen in human gastruloids

    Anna Yoney, Fred Etoc, Albert Ruzo, et al.

    Selected by Sundar Naganathan

    Human macrophages survive and adopt activated genotypes in living zebrafish

    Colin D. Paul, Alexus Devine, Kevin Bishop, et al.

    Selected by Giuliana Clemente


    Altering the temporal regulation of one transcription factor drives sensory trade-offs

    Ariane Ramaekers, Simon Weinberger, Annelies Claeys, et al.

    Selected by Mariana R.P. Alves

    Presence of midline cilia supersedes the expression of Lefty1 in forming the midline barrier during the establishment of left-right asymmetry

    Natalia A Shylo, Dylan A Ramrattan, Scott D Weatherbee

    Selected by Hannah Brunsdon

    Genetic compensation is triggered by mutant mRNA degradation

    Mohamed El-Brolosy, Andrea Rossi, Zacharias Kontarakis, et al.

    Selected by Andreas van Impel


    Cellular Crowding Influences Extrusion and Proliferation to Facilitate Epithelial Tissue Repair

    Jovany Jeomar Franco, Youmna Maryline Atieh, Chase Dallas Bryan, et al.

    Selected by Helen Weavers

    Photoperiod sensing of the circadian clock is controlled by ELF3 and GI

    Usman Anwer, Amanda Davis, Seth Jon Davis, et al.

    Selected by Annika Weimer

    We want to make our website, and the services we provide, useful and reliable. This sometimes involves placing small amounts of information called cookies on the device you used to access the internet. If you continue to use this website we will assume you are happy to accept our cookies.