The overall goal of this procedure, is to generate tumerogenic mosaic clones in developing Drosophila imaginal discs, and then classify the clonal lesions into various stages of tumor development. This method can help answer a key question in cancer modeling. Such as how to consistently induce tumors, and how to classify the stage of progression.
The main advantages of this technique are that tumor injection in the Drosophila wing imaginal disks is accomplished using simple fly genetics, and that, classification of the tumor development stages is a simple diagnosis. Along with Kenta Morimoto, demonstrating the procedure, will be Kimiko Morimoto, a technician from my laboratory. Begin with setting up the genetic cross.
12 hours before collecting virgins for the cross, clear all the adult flies from vials containing mature pupae. The following day, set up a cross to generate genetically-mosaic progeny. Transfer 12 to 20 virgins, and 10 males, into a fresh vial, and incubate them for one day at 25 degrees Celsius.
After 24 hours, transfer the flies into a fresh vial, and discard the first vial they populated. After 12 more hours, eggs will start becoming available. On 12-hour intervals transfer the adults to new vials, and use the progeny for the experiment.
After incubating egg vials for two to four days at 25 degrees Celsius, heat-shock the larvae for ten to 45 minutes, using a 37-degree Celsius water bath, for induction of mosaic clones by flip-out GAL4. Heat-shocking at two days of development will generate clones in immature discs. Heat-shocking at three or four days of development, will generate clones in more later-stage discs.
After the heat-shock, return the vials to the 25-degree Celsius incubator, until the third-instar larvae are available. To begin, use a wooden stick, or blunt forceps, to transfer wandering third-instar larvae into a Petri dish with PBS. Then, genotype them under fluorescent light to identify the appropriate fluorescent markers for mosaic larvae.
Before proceeding, wash the larvae with PBS. Then, under a dissection microscope, use two pairs of forceps to pinch the center of the larva, and tear the body in half. Next, pinch the mouth hooks, while pushing the mouth towards the body, to turn the body inside-out.
Then, remove unnecessary materials, such as the salivary glands and fat bodies. The larvae are now prepared for fixation and antibody staining, which is detailed in the text protocol. Transfer the stained tissue preparations to a microscope slide using a two-milliliter plastic transfer pipette.
Then, using forceps, move individual larva into drops of mounting medium on a microscope slide. Now, using forceps, isolate the imaginal discs. Start with holding down the end of dissected tissue, and pulling away the brain and eye-antennal discs with the other forceps.
Keep the brains ready for use. Now locate the wing imaginal discs. The wing imaginal discs are stuck to the lateral side of dissected tissue.
Next, secure the tissue with one forcep, and gently scratch off the wing discs. Then, position the brains near the wing discs, to shield them from being crushed by the coverslip. Now, gently cover the structures with a coverslip, and seal the coverslip with nail polish.
Store the slide at four degrees Celsius. After using a confocal microscope to acquire the confocal Z-stack images, open them in ImageJ, and use the Reslice command to create virtual sections. They can be made on the X-Z axis, the Y-Z axis, or in a customized plane.
To create the customized plane, draw a straight line, or rectangle, onto the Z-stack image, And follow the prompts to make the computer generate its section. Use such sections to diagnose the tissue for tumor phenotypes. First, identify the cell masses that deviate, or shows outgrowth from the main epithelial layer.
Then, for each mass, determine whether the sub-cellular localization of the junctional proteins appears normal. Other considerations include the size of the mass, and any indications of increased cell proliferation within the mass. Thus, the type of tumor is identified.
In this example, four clones were classified, using vertical sections generated by the imaging software. Two were classified as dysplasia, because the labeled DLG protein was mis-localized. Clone B is less than four cells large, and Clone C does not deviate from the epithelium.
Therefore, neither B or C were classified as a neoplasm. The other two clones also showed mis-localization of DLG, and formed tumorous cell-masses outside of the epithelium. As the size of these displaced tumorous clones was more than four cells in diameter, these clones were classified as neoplastic.
Using the described methods, the gene LGL was knocked-down in the wing-pouch and the hinge regions, using sd-GAL4-driven RNAi. In controls lacking the RNAi transgene, there were no morphological alterations in the third-instar wing discs. In knock-downs, epithelial deformations and tumorogenic over-growths were clearly observed in certain parts of the hinge.
However, dysplastic over-growth was not observed in the wing-pouch region. This correlated inversely with MMP1 expression in those regions:a protein that provides metastatic capability. To monitor the progression of tumor growth induced by LGL knock-down, sporadic clones, expressing LGL-RNAi were generated in wing discs, as described.
Two days after heat-shock to second-instar larvae, some knock-down clones in the hinge region showed deviation from the apical side of the epithelial layer. Four days after heat-shock induction, dysplastic lesions in the hinge region became clear, suggesting that LGL knock-down clones, displaced from the apical side, proliferate in the lumen. Seven days after heat-shock induction, tumorogenic over-growths dominated the hinge regions.
When using flip-out GAL4 systems, it's important to remember to examine the effect of altered heat-shock timing on clonal phenotypes, because the tumorogenic potential of imaginal disc cells may be dependent on the developmental stage endogenous signaling events, or cellular differentiation.
