Previous research on JNK-mediated stress signals demonstrated that stem cells in the posterior midgut of Drosophila Melanogaster only undergo compensatory proliferation or apoptosis. However, our group discovered that the stem cells can also undergo the process of basal extrusion and dissemination, resulting in the cells being eliminated from the tissue into the hemocoel, the blood containing intertissue body cavity. The JNK signal promotes the cells to exit the epithelium of the gut, move through the muscle layer, and be released to the hemocoel, which resembles the process of metastasis in human cancer. In order to understand the mechanism of this extrusion process, we carried out a RNA Interference (RNAi) screen and sought to find genes that are necessary for the stem cell extrusion in the JNK activators, rasv12 and HepCA, expressed flies. We used the ESG-GAL4, UAS-GFP, TUB-GAL80TS(EGT) genetic system to study the knockdown effect from the RNAi of each gene. After 4 days of inducement, which is the sufficient time for the JNK signal to promote complete extrusion of the intestinal stem cells, the intestines were dissected, fixed, mounted, and examined with a fluorescent microscope for any presence of stem cells. Our previous research from last year focused on the first phase of screening of possible RNAi lines involved in cell extrusion. Our results from this year focuses on our second phase of further identifying characteristics of cell movement in rasv12 flies.

Eighty-four percent of genes associated with human disease have a zebrafish counterpart. This is one reason that zebrafish are staples for developmental and genetic studies. Recently, there has been an interest in studying genetic control of skeletal phenotype, particularly in the craniofacial skeleton. Experiments typically use micro-computed tomography (micro-CT) to capture skeletal phenotype in 3D, and rely on manually annotated anatomical landmarks for statistical shape analysis of individual bone structures. However, manual segmentation has some fundamental problems. Subtle differences in how anatomical landmarks are recorded create differences between data sets that can influence the fidelity of the work. Varying lab practices can make exchanging data between researchers and reproducing findings difficult. Additionally, the significant time it takes to manually annotate an image makes large data sets an obstacle. Here, we propose the use of a fully segmented synthetic template of the zebrafish skeleton, created from common inbred strains, to segment and isolate cranial structures in a nearly automated procedure. Instead of segmenting each image individually, manual segmentation only needs to occur once on the template. Then, large quantities of images may be segmented by matching their skeletal landmarks to those of the template. This would not only standardize these micro-CT studies but also reduce the time burden of image analysis. We demonstrate an application of nearly automated shape analysis in proof-of-concept studies that compare the template-segmented volumes of 200 zebrafish otoliths with the volumes measured with manual annotation. Further proof-of-concept studies will compare measurements of other skeletal structures, specifically cranial bone width and parasphenoid bone length. Future work will include developing a micro-CT scanning protocol for the zebrafish Weberian apparatus and applying the template segmentation method to studying the relationship between swim bladder defects and the Weberian apparatus by examining the skeletal phenotypes of a number of zebrafish mutants.