The maturation of the epithelial barrier in Drosophila wing imaginal discs is ecdysone dependent and is regulated to communicate the completion of regeneration

Regeneration is a highly coordinated process that results in the complete and scar-less restoration of damaged tissues. Over the last ten years we have improved our understanding of how tissues initiate and progress through regeneration, as well as how they coordinate regrowth with the growth of undamaged tissues. Still, little is known about how fully regenerated tissues communicate the completion of regeneration. Using the Drosophila melanogaster wing imaginal discs (adult wing precursors) as a regenerative model, I have found that one mechanism signaling the completion of regeneration is the re-establishment of epithelial barrier, which is disrupted during damage. The epithelial barrier is a semi-permeable diffusion barrier and a critical component of an epithelium. In the larval wing disc a functional barrier should separate the disc lumen from the larval hemolymph. However, one of the damage-response peptides, Dilp8, accumulates in the imaginal disc lumen following damage or exogenous overexpression even though it is only known to function in the brain and prothoracic gland. This indicates that some aspect of the imaginal disc epithelium contains Dilp8 in the lumen. My data indicate that the aspect of the disc providing this containment of Dilp8 is the epithelial barrier.
To investigate this, I used a phenotypic effect of Dilp8: Dilp8 causes a developmental delay, allowing the larvae to regenerate. This delay results from the inhibition of the production of the steroid hormone ecdysone in the brain and prothoracic gland. In Chapter 2, I show that disrupting the epithelial barrier by RNAi against the septate junction components Kune and Nrx produces an extended Dilp8-mediated delay indicating that the barrier limits Dilp8 signaling. I observed the same result in damaged larvae, indicating that the barrier regulates the length of the regenerative response. Thus, the barrier is a mechanism to signal the completion of the regeneration.
To better understand this process, I characterized the functionality of the epithelial barrier in wing imaginal discs. I found that the epithelial barrier grows increasingly restrictive during the third instar between 92 and 116 hours after egg deposition (h AED). Over time, the barrier becomes less permeable to the 10 kDa dextran used to assay barrier function. Interestingly, although the barrier becomes more exclusive, it is still functional at 92h AED, excluding significantly more dextran than a disrupted barrier. This change in function results from a change in the localization of the protein Coracle, a component of the septate junctions that form the epithelial barrier. Between 92h and 116h AED, Coracle shifts from being diffusely localized along the membrane to being localized only at the septate junctions. Diffusely localized Coracle is not necessary for the less restrictive barrier, as disruption by RNAi at 92h AED produces a barrier that excludes the dextran similarly to wild type controls.
In contrast, at 116h AED, Coracle localized only at the septate junctions is necessary for the more restrictive barrier, as Coracle disruption by RNAi also disrupts the barrier. The localization of Coracle is dependent on ecdysone, the same hormone that Dilp8 limits to produce developmental delay. Coracle localization at the septate junctions is significantly reduced in discs expressing a dominant-negative allele for the ecdysone receptor, and the barrier does not mature in these discs either. Inducing ecdysone signaling early causes the barrier to prematurely become more restrictive. These data indicate a model where a damaged tissue produces Dilp8, Dilp8 limits ecdysone in the brain and prothoracic gland, and then, once the tissue regenerates, the epithelial barrier begins to mature and limit Dilp8. This could act in a feedback loop where trapped Dilp8 results in higher levels of ecdysone which causes the barrier to mature more, which in turn traps more Dilp8.
I began to investigate this hypothesis in the Appendix. I first demonstrated that damage disrupts the epithelial barrier. Preliminary experiments indicate that the barrier is likely to recover after damage, but the timeframe of that recovery depends on when the larvae are damaged. Drosophila larvae lose the ability to regenerate during late larval development in response to rising ecdysone levels. When larvae are damaged with X-irradiation before regeneration is restricted, the barrier is disrupted and remains disrupted for an extended amount of time, beginning to recover 48 hours after damage (the last timepoint collected). During this time the septate junction components are also depleted from the septate junctions. In contrast, when larvae are damaged after regeneration is restricted, the barrier is disrupted but recovers within 24 hours and the septate junctions are never depleted. These data preliminarily indicate that the barrier is downregulated during regeneration.
Overall, this work furthers our understanding of the hormonal regulation of the epithelial barrier during development and demonstrates how the restoration of a tissue’s function can signal the completion of regeneration.