Microfluidics and imaging techniques for high-throughput studies of early embryonic development

Abstract

Understanding how developmental systems achieve robustness is a key goal of developmental biology. The fruit fly Drosophila melanogaster is a model of development and developmental genetics owing to high genetic conservation that can provide insight into human development. Drosophila is compatible with in vivo live imaging: a powerful technique that allows researchers to visualize dynamic processes in real time within developing organisms, but is technically challenging to perform. As a result, large-scale data collection is virtually impossible preventing researchers from obtaining highly quantitative information regarding live embryo development. To address this issue, this thesis advances the quantitative imaging toolsets available to biologists by developing microfluidic technologies for high-throughput time-lapse microscopy of live Drosophila embryos as well as image processing and analysis software for automated quantitative phenotyping of dynamic processes. Significant engineering feats allowed for the expansion of microsystem functionality and integration with computer vision algorithms facilitate rapid microscopy and quantitative analysis of a wide range of biological applications. As a result of these technological advances, insight regarding anoxia-induced developmental arrest and recovery, mitotic wave-front propagation dynamics, and the effects of RTK-ERK pathway mutations on downstream signaling kinetics were uncovered and quantitatively characterized. The technologies developed in this dissertation are generalizable, and should facilitate rapid microscopy and quantitative phenotyping throughout developmental biology.