A direct and widespread role for the nuclear receptor EcR in mediating the response to ecdysone in Drosophila

Christopher M Uyehara, Daniel J McKay

Preprint posted on January 10, 2019

Article now published in Proceedings of the National Academy of Sciences at

Here, there, everywhere: extensive and dynamic genome binding by a steroid hormone receptor highlights the interconnection between systemic and local cues for organ development

Selected by Natalie Dye

Hormones regulate so much of our physiology and development. Just think back to those gloriously awkward times at puberty or consider what happens during pregnancy– so many changes in the body happening at once, all in response to changing hormones. But of course, every tissue responds differently – some tissues grow while others shrink or change shape. How does that work?

The authors of this preprint start addressing this question by profiling gene expression changes and genome-wide binding of the steroid hormone receptor in the Drosophila wing during the transition from larval to pupal life stages – the fly’s puberty, if you will. The steroid in flies is called ecdysone and it is received by a nuclear hormone receptor called EcR.

Drosophila has long been a favorite model for studying hormone signaling, and (for once) it wasn’t because of the fly’s genetic tractability (at least not at first). Researchers in the 1950s-60s figured out how to culture salivary glands ex vivo to directly study the response to exogenously added hormone. The glands have huge chromosomes that make visible bulges when gene expression is activated, so researchers didn’t even need fancy microscopy to track the response to hormone. From these experiments, Michael Ashburner proposed that ecdysone binds to a (then unknown) receptor that activates relatively few targets, who then go on to activate many additional downstream targets. Thus, Ashburner proposed that much of ecdysone’s effects on the tissue are only indirectly mediated by its receptor.

While this model provided an influential framework for hormone-induced gene expression, it remained unclear how tissue-specific responses are elicited. After all, the salivary glands die after larval stages, while “imaginal” tissues like the wing undergo dramatic morphogenesis to make adult structures. Previous works have profiled gene expression changes occurring in the wing at pupariation, but the mechanisms and direct role of EcR were not addressed. Thus, we are overdue for a systems-level analysis of EcR’s role in eliciting steroid dependent expression in the wing.


Summary of the preprint:

  • Authors perform RNA-seq on wild type and EcR-depleted wing discs at two timepoints: 6hr before and 6hr after the larval-to-pupal transition (Fig 1C). Their results suggest that:
  • Before the transition, EcR mostly prevents the precocious activation of ~400 genes whose expression will increase at pupariation (Fig 1D).
  • After the transition, EcR affects both the activation and repression of >1000 genes. Most of these genes either increase or decrease during the larval-to-pupal transition (Fig 1F).
Fig 1: Reproduced from Fig 1 of the preprint with author permission. Part C shows the genes whose expression changes over time in wild type wing discs during the larval-to-pupal transition. Part D and F indicate how the expression pattern changes between wild type and EcR-depleted wing discs either before (D) or after (F) the larval-to-pupal transition.


  • Using CUT&RUN (a recently developed alternative to ChIP-seq) to identify the sites on the genome where EcR binds, authors find that:
  • Before the larval-to-pupariation transition, EcR binds extensively to the genome at numerous sites, both canonical and wing-specific genes.
  • After, EcR is lost from most of its previous targets but gained at hundreds of new sites.
Fig 3b: Reproduced from the preprint, with author permission. The graph visualizes how many genes bind EcR only before the transition (magenta), only after the transition (green), or stably through both timepoints (yellow).


  • The sites that stably bind EcR through the transition have the highest density of the EcR-motif and the highest overlap with sites bound by EcR in the S2 culture cell line. These sites tend to be genes involved in steroid-mediated signaling.
  • Most of the sites bound by EcR (except for those relatively few that also bind in S2 cells) are functionally classified as imaginal disc-derived wing morphogenesis genes.
  • Not all of the genes found with RNA-seq to vary during the larval-to-pupal transition are bound by EcR – suggesting that there are both direct and indirect effects of ecdysone.
  • By examining two EcR target genes, authors conclude that EcR can affect both the temporal and spatial patterns of gene expression in the developing wing:
    • The canonical target gene, broad, does not require EcR for its expression and in fact is ectopically high in the absence of EcR, suggesting that EcR is a repressor of broad, preventing its concentration from rising too fast. Interestingly, broad’s expression still increases over time without EcR, suggesting that other factors also regulate this key steroid target.
    • The non-canonical target gene, delta, was found to have two EcR-enhancers with spatially restricted expression patterns. Mutant analysis suggests that EcR, together with its binding partner Usp, repress delta. Unlike broad, delta expression does not change through the larval-to-pupal transition. Thus, EcR/Usp is most likely to influence the spatial but not temporal activation of delta – a somewhat surprising role for a hormone receptor.


My 2-cents:

While the authors do identify a relatively small set of core target genes bound by EcR in the wing and in other cell types, what is more striking is that they also find EcR directly, dynamically, and tissue-specifically binding the genome upstream of numerous genes that are required specifically for wing development. Thus, the role of EcR seems much broader than the Ashburner model would have predicted.

The connection between ecdysone and wing development genes is actually not surprising to me. Using culture experiments analogous to those early ones in salivary glands, I previously showed that the expression of numerous genes involved in wing patterning, including delta, are sensitive to steroid during larval stages. But I could not conclude much about the mechanism at the time. This preprint now makes me believe that what I observed is directly attributable to EcR acting as a repressor for many wing patterning genes, and that the low levels of steroid circulating during larval stages partially relieves this repression to control expression in time and/or space.

Importantly, wing patterning was heretofore considered to be largely tissue-autonomous: activity gradients in developmental signaling pathways such as Hedgehog and Wnt collaborate to regionally subdivide the wing tissue, specifying cells that will make, for example, the wing veins. While extensively studied, what controls the timing has remained mysterious. For example, the Hedgehog gradient is present from the very beginning of wing development, so why are veins only specified in late larval stages? Furthermore, morphogen gradients persist through the larval-to-pupal transition, yet the cells do something different with the information, undergoing morphogenesis instead of growth. How?

The intricate connection between ecdysone and wing patterning revealed in this preprint and other recent work suggests that systemic cues like ecdysone provide key tissue-non-autonomous inputs. But importantly, this work also indicates that the information flow is not unidirectional: tissue-specific genes clearly affect the site-specific binding of EcR and thereby the response to steroid. Thus, the picture now emerging is that bidirectional feedback exists between tissue autonomous and non-autonomous signals for wing patterning and growth. The exact nature of these interactions – e.g., which tissue-specific factors interact with EcR – will be interesting to discover in future work.

One last thing that I’m curious about is the potential role of insulin in regulating the dynamics or specificity of EcR genome binding. Insulin is a major signal coordinating systemic development with nutrient availability, and at the larval-to-pupal transition, the animals stop feeding and insulin signaling changes. Crosstalk exists between insulin and steroid signaling pathways in many animals: in humans, consider the link between childhood obesity and the onset of puberty; and in insects, direct interactions have been observed between EcR and the insulin-responsive transcription factor FoxO. Thus, I would be curious to know how EcR binding changes during starvation or mutants of insulin receptor or FoxO.


For more information…

Review of EcR and a little history of the Ashburner model:

Transcriptional profiling of wing discs grown ex vivo +/- ecdysone (NOTE: shameless self-promotion):

Review of inter-organ signaling coordinating growth in Drosophila:

Ecdysone-induced expression in cell lines:  &


Tags: drosophila, ecdysone, nuclear hormone receptor, transcriptomics, wing disc

Posted on: 4th February 2019

Read preprint (No Ratings Yet)

  • Have your say

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

    This site uses Akismet to reduce spam. Learn how your comment data is processed.

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

    Register here