The parallel actin filament bundling protein Fascin is a critical protein in both disease and development. Overexpression of Fascin is linked to increased aggressiveness in a number of cancer types, including breast and colon carcinomas. Importantly, Fascin is not normally expressed in adult epithelial cells from which many of these cancers arise. Therefore, Fascin is increasingly cited as both a potential biomarker and therapeutic target in many types of cancer. Fascin is most commonly associated with the formation of filopodia and invadapodia (parallel actin filament bundle structures) to drive migration and invasion. However, Fascin activity and regulation remain poorly understood. In order for Fascin to be an effective target for cancer therapeutics, a better understanding of the mechanisms regulating Fascin activities in the cell is necessary.
Prostaglandins (PGs) are short-lived lipid signaling molecules that mediate a wide range of biological activities. PGs act through G protein-coupled receptors to initiate signaling cascades that affect downstream targets, including actin cytoskeletal remodeling. Importantly, the key enzymes in the synthesis of PGs, cyclooxygenase (COX) 1 and 2, are the targets of non-steroidal anti-inflammatory drugs like aspirin. Interestingly, like Fascin, PGs have been independently implicated in cancer development and metastasis and aspirin may reduce the risk of aggressive cancer. However, the exact mechanisms by which PGs mediate cancer development are unknown. The work presented in this thesis focused on novel PG-dependent regulation and activity of Fascin.
The research presented here utilized Drosophila oogenesis as a model system to analyze PG-dependent Fascin activity. Drosophila oogenesis is an ideal model in which to study the activity and regulation of actin binding proteins like Fascin. Oogenesis consists of 14 morphologically defined stages, which are observable many times over within a single isolated pair of ovaries. A developing follicle consists of 16 germline cells – 15 nurse, or support cells, and a single oocyte. The nurse cells are of particular interest because they are the sites of dynamic actin remodeling during mid-late oogenesis. During stage 10B, an array of radially-aligned actin filament bundles form at the nurse cell membranes and extend inwards towards the nucleus. A network of cortical actin is also strengthened during this stage. These actin structures are essential for the completion of oogenesis, and ultimately female fertility. Importantly, PGs and Fascin are required for this actin remodeling; genetic loss of Fascin or the Drosophila COX-like enzyme Pxt (Peroxinectin-like) leads to disruption of cytoplasmic actin remodeling, and ultimately, female sterility.
Using this model system, work presented here describes the discovery of Fascin as a downstream target of PGs to promote actin bundle formation, described in Chapter 2. Additionally, Fascin is required for strengthening of the cortical actin network downstream of PGs. This observation is one of the first to describe a role for Fascin in a branched actin network. Additionally, Fascin is regulated by a specific PG – PGF2α – during S10B to promote follicle development. Finally, Chapter 2 shows that PGs target specific actin binding proteins to promote cytoskeletal remodeling; Villin, another actin bundling protein, does not interact with PGs.
Chapter 3 describes the novel observation that Fascin localizes to the nucleus and the nuclear periphery in Drosophila nurse cells. This finding is significant, as it is the first to describe Fascin in a context other than cytoplasmic. Fascin localization in and around the nucleus is specific and dynamic, and changes throughout late stage oogenesis, suggesting regulated functions at these sites. Fascin localization is regulated by PGs, and loss of Pxt leads to reduced nuclear Fascin localization and failure to localize to the nuclear periphery. Additionally, Fascin has novel potential functions in the nucleus and at the nuclear periphery. Loss of Fascin leads to disruption of nucleolar morphology in the nurse cell nuclei. Additionally, loss of PGs, which cause reduced nuclear Fascin levels, also causes abnormal nucleolar morphology. These data suggest that PGs regulate Fascin to control nucleolar organization. At the nuclear periphery, Fascin localization requires components of the protein complex that links the nucleoplasm to the cytoplasm, termed the LINC complex. Loss of an essential LINC complex protein, Koi, leads to a loss of nuclear periphery Fascin localization. These data suggest that Fascin may be a novel component of the LINC complex.
Finally, Chapter 4 describes regulation of Fascin by phosphorylation at conserved serine residues. PGs affect Fascin phosphorylation, and loss of PGs leads to more heavily phosphorylated Fascin. Additionally, phosphorylation of Fascin alters localization to the nucleus and to the nuclear periphery. These data suggest that one mechanism by which PGs regulate Fascin is to control its phosphorylation status to affect subcellular distribution.
In summary, the work presented in this thesis has demonstrated novel regulation and function of the actin bundling protein Fascin using Drosophila oogenesis as a model. Importantly, these functions and regulation of Fascin are likely conserved in mammals, and may have implications in human health and disease. Continued study of the activity and regulation of actin binding proteins like Fascin in Drosophila will likely have great effect on our understanding of many human diseases.