The morphogenetic movements in epithelial sheets seen during organogenesis are driven in part by changes in cell shape, which in turn depend on changes in the cytoskeleton. In this thesis I used the formation of the salivary glands in the Drosophila embryo, which are formed by coordinated apical constriction and invagination of a polarised epithelial placode, as a model to understand these changes. The role of microtubules (MTs) during morphogenetic processes such as this are not well understood. In order to ascertain their role in successful tubulogenesis I looked at both stable and newly polymerized MTs within the salivary gland placode. Apical constriction is accompanied by the loss of a centrosomal apically localised MT network and the appearance of a longitudinal acentrosomal MT network. In parallel, MTs become increasingly more stable, starting from the apical surface and spreading more basally, leading to stable MT bundles that appear nucleated and anchored near the apical surface independent of centrosomes. Disrupting the MTs specifically in the salivary gland placode using the MT-severing protein Spastin results in defects in apical constriction, and at later stages tube lumen defects. Wild type cells generate an apical medial actomyosin meshwork that likely drives apical constriction. MT ends present at the apical surface colocalise with apical medial actomyosin, and cells lacking MTs fail to maintain apical medial actomyosin. This indicates crosstalk between the MT and actomyosin cytoskeleton during apical constriction and tissue bending. The cytolinker protein Short Stop (Shot) has previously been shown to be able to link the actin and MT cytoskeleton. Shot rearranges concomitant with the MT cytoskeleton and colocalises with both with the tips of MT bundles and apical medial actin. Further to Shot, the formin Diaphanous regulates both actin and microtubules in morphogenesis, and was also found to localize to MT tips at the apical surface. Overexpression of constitutively active Diaphanous results in precocious invagination of the salivary gland. Overall this suggests a role for acentrosomal MT bundles in maintaining an apical actomyosin meshwork for successful apical constriction and subsequent tubulogenesis. Cytoskeletal crosstalk may be mediated by the cytolinker Shot and the formin Diaphanous.

The conserved human protein KANK1 has been identified as a tumour suppressor and its expression is down-regulated in several tumour types. Roles for this protein in actin regulation, cell migration and cell polarity have been documented in cultured mammalian cells. In C. elegans the KANK1 orthologue, VAB-19, is required for normal development as it helps stabilise attachment structures between muscle and epidermal cells. Despite these studies, the precise cellular role of KANK remains elusive. It was found that the Drosophila KANK orthologue binds directly to EB1, a crucial regulator of microtubule plus-end dynamics. I aimed to determine the role of KANK with respect to this indirect microtubule interaction using Drosophila. I identified residues which mediate the interaction between KANK and EB1, and showed they are essential for localisation of KANK to microtubule plus-ends in Drosophila culture cells. I found that KANK expression increases during embryogenesis and peaks in the late embryonic development when KANK is shown to localise to sites of attachment between muscle and epidermal cells. This suggests a role for the protein in stabilisation of muscle attachment during embryonic development, a process previously shown to require EB1. I generated a KANK deletion mutant and found they are viable and fertile but show a mild neuronal phenotype, specifically early branching of the neurons and less organised neuron bundles. My results suggest previously unknown roles for KANK in myogenesis and neurogenesis in Drosophila embryogenesis.