Duox, Flotillin-2, and Src42A are required to activate or delimit the spread of the transcriptional response to epidermal wounds in Drosophila

Abstract

The epidermis is the largest organ of the body for most animals, and the first line of defense against invading pathogens. A breach in the epidermal cell layer triggers a variety of localized responses that in favorable circumstances result in the repair of the wound. Many cellular and genetic responses must be limited to epidermal cells that are close to wounds, but how this is regulated is still poorly understood. The order and hierarchy of epidermal wound signaling factors are also still obscure. The Drosophila embryonic epidermis provides an excellent system to study genes that regulate wound healing processes. We have developed a variety of fluorescent reporters that provide a visible readout of wound-dependent transcriptional activation near epidermal wound sites. A large screen for mutants that alter the activity of these wound reporters has identified seven new genes required to activate or delimit wound-induced transcriptional responses to a narrow zone of cells surrounding wound sites. Among the genes required to delimit the spread of wound responses are Drosophila Flotillin-2 and Src42A, both of which are transcriptionally activated around wound sites. Flotillin-2 and constitutively active Src42A are also sufficient, when overexpressed at high levels, to inhibit wound-induced transcription in epidermal cells. One gene required to activate epidermal wound reporters encodes Dual oxidase, an enzyme that produces hydrogen peroxide. We also find that four biochemical treatments (a serine protease, a Src kinase inhibitor, methyl-ß-cyclodextrin, and hydrogen peroxide) are sufficient to globally activate epidermal wound response genes in Drosophila embryos. We explore the epistatic relationships among the factors that induce or delimit the spread of epidermal wound signals. Our results define new genetic functions that interact to instruct only a limited number of cells around puncture wounds to mount a transcriptional response, mediating local repair and regeneration.

Figure 1. Summary of epidermal wound reporter mutants and their phenotypes.

(A) Maps of X and 2nd chromosomes. Black bars represent deleted regions that have a normal epidermal wound reporter phenotype. Red bars represent deleted regions that have an altered epidermal wound reporter express pattern. (B) A list of new genes identified that affect the localization of the epidermal wound reporter. X:1.2M, for example, denotes 1.2 Megabases on the Drosophila genome map of the X chromosome .

Figure 2. Flo-2 functions to inhibit widespread activation of epidermal wound reporters.

(A) Confocal images of Ddc .47-GFP epidermal wound reporter activity (4–6 hours after puncture wounding). Wild type (wt) embryos show the reporter in cells around the wound site. Flo-2 mutant embryos show expansion of Ddc .47 reporter expression to most epidermal cells. (B) ple-WE1-dsRed epidermal wound reporter expression expands to most epidermal cells in Flo-2 mutant embryos 6 hours after wounding. Arrows mark the site of the wound. Dashed lines in the data panels mark the outlines of embryos. Scale bars = 50 µM.

Figure 3. Flo-2 transcriptional activation at wound site depends on grainy head function.

Confocal images of in situ hybridizations with fluorescently labeled probes made from Flo-2 and Ddc cDNA clones. (A) Flo-2 transcripts accumulate in all epidermal cells of late stage embryos. (B) 30-minutes after puncture wounding, Flo-2 transcripts show increased accumulation in cells around wound sites, in a similar pattern as wound-induced Ddc transcript activation in the epidermis. (C) 30-minutes after puncture wounding, grainy head (grh) mutant embryos fail to show increased accumulation of Flo-2 transcripts around epidermal wound sites. Arrows mark the wound site. Dashed lines in the data panels mark the outlines of embryos. Scale bar = 50 µM.

Figure 4. Overexpression of Flo-2 inhibits activation of an epidermal wound reporter.

(A) Brightfield and fluorescent confocal images of late stage embryos with Ddc .47-GFP epidermal wound reporter genes. Wild type (wt) embryos, or embryos with overexpressed Flo-2 (arm>Flo-2) have similar melanized clots in the cuticle at puncture wound sites. Ddc .47 reporter expression accumulates around punctures to high levels at 4–6 hours post-wounding. In arm>Flo-2 embryos, Ddc .47 wound reporter expression is not detected around punctures at any time post-wounding. (B) After wounding with needles loaded with trypsin, wild type (wt) late stage embryos dramatically activate Ddc .47-GFP reporter expression throughout most or all epidermal cells at 6 hours post-wounding. In late stage embryos overexpressing Flo-2 (arm>Flo-2), the trypsin-induced Ddc .47 wound reporter activation is completely repressed. (C,D) Embryos with the Ddc .47 epidermal wound reporter that overexpress Flo-2 in stripes using the en-GAL4 driver (en>Flo-2) show inhibition of both wound-induced (C) and trypsin-induced (D) activation of the Ddc .47 wound reporter in all epidermal cells, not just the epidermal cells that overexpress Flo-2 (see Figure S2B). Arrows show wound sites. Dashed lines in the data panels mark the outlines of embryos. Scale bar = 50 µM.

Figure 5. Src42A functions to inhibit widespread epidermal wound reporter activity after wounding.

(A) Fluorescent confocal image of ple-WE1-dsRed epidermal wound reporter activity in a late stage Src42A mutant embryo. At 6 hours after wounding, the Src42A mutants show activation of the ple-WE1 wound reporter in a very broad area of the embryonic epidermis surrounding the wound site (compare with Figure 1B). (B,C) Confocal images of fluorescent in situ hybridization experiments. (B) 30-minutes after wounding, Src42A mutant embryos show accumulation of Ddc transcripts in a widespread zone of epidermal cells around the wound site (compare with Ddc expression in wounded wild type (wt) in panel C- right side). (C) Using double in situ hybridization with probes labeled with different fluorophores (both images in C are taken from the same wild type (wt) embryo 30 minutes after wounding), we observed that Src42A transcripts are activated to high levels around wound sites (left panel), and in a slightly smaller zone than Ddc transcripts at the same stage. (D,E) Fluorescent confocal images of ple-WE1-dsRed epidermal wound reporter activity in late stage embryos (6 hours post-wounding) after body cavity injection of trypsin. (D) Ubiquitous expression with Src42A.CA inhibits the trypsin-induced ple-WE1 reporter in late stage embryos (compare with Figure S3B). (E) Overexpression of Src42A.CA in stripes with the en-GAL4 driver only inhibits ple-WE1 reporter expression in the cells where en-GAL4 is activating high levels of Src42A.CA protein expression. The insert provides higher magnification, and the bracket shows the stripe of cells where Src42A.CA is overexpressed by the en-GAL4 driver. (F,G) Confocal images of Ddc .47 epidermal wound reporter activity in late stage embryos (6 hours post-wounding) after body cavity injection of trypsin. (F) Ubiquitous expression with Src42A.CA inhibits the trypsin-induced Ddc .47 wound reporter in late stage embryos. (G) Overexpression of Src42A.CA in stripes with the en-GAL4 driver only inhibits Ddc .47 reporter expression in the cellular stripes where en-GAL4 is activating high levels of Src42A.CA protein expression. Arrows mark sites of wounds. Dashed lines in the data panels mark the outlines of embryos. Scale bar = 50 µM.

Figure 6. Chemical activation of the epidermal wound reporters by H₂O₂ and methyl-ß-cyclodextrin (MßCD) is consistent with the affects of Duox and Flo-2 mutations on wound reporter expression.

(A–F) Fluorescent confocal images of Ddc .47 or ple-WE1 epidermal wound reporters (all embryos shown are 6 hours post-wounding). Body cavity injection of MßCD (A) and H₂O₂ (B) into wild type (wt) embryos activates the Ddc .47 wound reporter throughout the entire epidermis. Overexpression of Flo-2 with an arm-GAL4 driver (arm>Flo-2) can inhibit both MßCD (C) and H₂O₂-induced (D) activation of the Ddc .47 wound reporter (or the ple-WE1 wound reporter, not shown). (E) In Duox mutant embryos, wound-induced ple-WE1 reporter expression is not detected (F) However, in Duox mutant embryos treated with trypsin ple-WE1 reporter expression is detected in all epidermal cells. Scale bar = 50 µM.