Please use this identifier to cite or link to this item:
Title: Study on muscle remodeling using in vivo microscopy of drosophila metamorphosis and quantitative image analysis
Authors: Kuleesha
Keywords: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
DRNTU::Engineering::Computer science and engineering::Computer applications::Life and medical sciences
Issue Date: 2018
Abstract: Skeletal muscle cells control most of the physical movements in our body, from simple actions such as smiling or frowning to critical movement of the diaphragm for breathing. Maintenance of muscle mass and strength is essential for physical functions and mobility. The human body suffers from various muscular ailments ranging from the most common age related muscle loss to fatal muscular dystrophy. Advances in the understanding of signaling pathways involved in muscle mass control would facilitate the identification of pharmacological cure for muscle diseases. The association between skeletal muscle mass and muscular disorder has been studied extensively in various organisms. However, only a few of them have used in vivo imaging to understand the development of the muscle particularly in insects such as Drosophila melanogaster and moth where the muscles undergo extreme remodeling during metamorphosis. In Drosophila melanogaster, larval muscles undergo two different developmental fates during metamorphosis; one population is removed by cell death, while the other persistent subset undergoes morphological remodeling and survives to adulthood. Thanks to the ability to perform live imaging of muscle development in transparent pupae and the power of genetics, metamorphosis in Drosophila can be used as a model to study the regulation of skeletal muscle mass. In this thesis, my objective is to explore in vivo imaging of Drosophila metamorphosis as a model to study muscle mass regulation at large scale using quantitative image analysis and discover new insights into gene functions in muscles which could not be found using traditional endpoint assays. We found new phenotypes for many genes by quantifying the morphological changes in persistent muscles and myonuclear distribution using my custom tool FMAj. A detailed analysis of muscle phenotypes in four autophagy genes, that is Atg5, Atg9, Atg12 and Atg18 revealed that autophagy acts transiently to promote autophagy. Additionally, we found that the loss of autophagy due to silencing of Atg9 and Atg18 enhanced the anti-polar migration of nuclei during mid-pupal development and decreased polar migration toward the late-pupal development. Phenotypic analysis revealed that SNF1A and Grunge are involved in regulating the cell death of Drosophila abdominal muscles during metamorphosis. In this study, our team characterized the phenotypic changes in Drosophila muscle cells caused by genetic perturbation during metamorphosis using time lapses in vivo imaging. A combination of RNAi and UAS-GAL4 system was used for muscle specific gene silencing and fluorophore expression. With the availability of advanced fluorescence microscopes and a reporter line with two markers, we acquired time lapse images of muscles and myonuclei. However, 3D time lapse imaging of Drosophila metamorphosis for 4-5 days generated large number of images (10-15 GB per sample) which is a bottleneck for the quantitative phenotypic analysis. In order to quantify the phenotypic changes in large number of images, I designed a custom tool, Fly Muscle Analysis in Java (FMAj), integrated with a relational database to perform various tasks such as annotation, segmentation, feature generation and statistical analysis. The integration of multiple tasks enhances productivity as the alternative export of data and the manual processing in a spreadsheet program would be much more time-consuming. My first objective was to study the morphological changes in muscle cells during metamorphosis for which muscle cell boundaries were required. However unlike other cells like hela, muscle cells in Drosophila pupae are surrounded by dead sequestered muscles (debris), which makes it difficult to extract muscle boundary with high accuracy. Therefore, I designed a muscle segmentation technique which is capable of differentiating muscle and debris regions. I combined watershed based region classification with chamfer distance based shape feature and an edge confidence feature to obtain muscle boundary in time series stack. The shape information extracted from muscle boundary at previous time points restricts the muscle boundary in next time point from moving to debris. Comparative studies show that the accuracy of my method is better than other segmentation techniques. I also performed nuclear spatial pattern analysis to understand the relationship between muscle mass change and the distribution of nuclei inside muscles. After nuclear segmentation, the regions were classified into two types of nuclei i.e. the nuclei inside muscle cells (internal nuclei) and the nuclei present inside debris produced during histolysis of muscle cell (external nuclei). I designed a tracking based nuclei classification algorithm which exploits the difference in the motion of internal and external nuclei. Comparative analysis show that my nuclear classification technique performs better than other techniques. I designed new nuclear spatial pattern features, each catering to a specific type of nuclei migration pattern. In conclusion, the work in this thesis has been able to justify Drosophila melanogaster as a good model to study metamorphosis and to find new insights into the genes associated with muscle mass change. My tool provides additional support by providing an image processing platform to confirm the findings quantitatively.
DOI: 10.32657/10356/73192
Fulltext Permission: open
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Theses

Files in This Item:
File Description SizeFormat 
TsceG1102023H.pdfMAIN THESIS6.93 MBAdobe PDFThumbnail

Page view(s)

Updated on May 12, 2021

Download(s) 50

Updated on May 12, 2021

Google ScholarTM




Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.