Combining proteomics and functional genetics to find new regulators of tissue growth and development in Drosophila melanogaster

© 2015 Dr. Lucas Graham DentBackground: The Hippo signalling pathway was discovered is Drosophila melanogaster and is a conserved regulator of tissue growth in animals. This pathway is also deregulated in human cancers. The Hippo pathway consists of upstream growth-suppressing proteins that work together to restrict the activity of the growth promoting protein, Yorkie (Yki). Despite its importance in disease and development, the complete set of signals that regulate the Hippo pathway remain to be determined. Approach and aims: I wanted to improve our understanding of the mechanisms controlling the Hippo pathway in Drosophila. To do this, I considered the results of a proteomic screen to identify proteins that bind to the Hippo (Hpo) kinase. Prominent amongst these proteins were Pak-interacting exchange factor (Pix), and G-protein coupled receptor kinase-interacting protein (Git). Pix, is a conserved rho-type guanine nucleotide exchange factor (RhoGEF), and is homologous to Beta-PIX and Alpha-PIX in mammals. Git, is the Drosophila homolog of the mammalian GIT1 and GIT2 proteins, and acts as an Arf-GTPase activating protein (ArfGAP). Pix and Git also heterodimerise with each other to form an oligomeric scaffold and facilitate sterile 20-like kinase activation. In this thesis, I used genetics and biochemistry approaches to investigate the following questions: 1) Do Pix and Git regulate the activity of the Hippo pathway? 2) Do Pix and Git regulate the activation of the Hpo kinase? 3) Does the Pix-Git complex have non-growth roles in epithelial development and are these related to Hippo signalling? Results: 1) I found that Pix and Git co-operate to modulate tissue growth through the Hippo pathway. Mutation of pix and git did not produce growth phenotypes, however mutating either of these genes in conjunction with other Hippo pathway regulators resulted in striking tissue overgrowth. Consistent with this, simultaneous expression of pix, git and hpo, severely limits tissue growth. These effects occur through the Hippo signalling pathway, as silencing pix increased Yki activity. Moreover, Pix and Git form a physical complex with Hpo, and epistasis experiments show that pix and git function upstream of hpo in growth control. 2) I showed that the Pix-Git complex controls growth by directly regulating the activity of the Hpo kinase. Co-expression of pix and git promoted Hpo homodimerisation, and increased Hippo autophosphorylation at its activation loop. Pix and Git appear to serve as scaffolds to recruit and activate Hpo, as their enzymatic domains are dispensable for Hpo activation. 3) Loss of pix and git resulted in a range of epithelial defects that were not directly growth related. These include defects in the coordinated migration of epithelial cells, the maintenance of basal filamentous actin, and disruption of epithelial polarity in developing egg chambers. Genetic interaction experiments showed that the cell migration phenotypes of pix and git are related to signalling via the sterile 20-kinase Pak, and not Hippo signalling. Significance: These discoveries provide new insight into the regulation of an ancient signalling network that controls the growth of animals, and describe new functions of the conserved Pix-Git complex that are critical for epithelial development