Cofilin’s activity to nucleate actin filament assembly, is regulated by phosphorylation at a single site on the amino terminus, Serine 3. Phosphorylation at this site abolishes the ability of ADF/cofilin to bind to F-actin and inhibits its severing function. This work characterizes the ability of dephosphorylated cofilin (mutation at Serine 3 site) to navigate prostate cancer actin cytoskeleton and metastatic properties in response to TGF-β. TGF-β increased Lim Domain Kinase 2 (LIMK-2) activity leading to cofilin phosphorylation and decrease actin filament severing in wild type cofilin (WTCFL) PC-3 cells. Constitutively active cofilin in Serine 3 cofilin mutants (S3ACFL) promoted prostate cancer cell filopodia formation, actin severing and directed TGF-β mediated migration and invasion. Co-culture of prostate cancer cells with prostate cancer associated fibroblasts induced cell invasion in WTCFL and S3ACFL cells. Active cofilin further enhanced the invasive response, even in the presence of a TGF-β-neutralizing antibody, implicating the contribution of the microenvironment. Active cofilin led to a significant increase in prostate cancer cell metastatic potential in vivo and cofilin correlated with metastasis in a mouse model of prostate tumor progression. In human prostate cancer, cofilin expression was significantly higher in metastasis compared to the primary tumors. Cofilin thus emerges as a regulator of the actin cytoskeleton remodeling capable of coordinating the cellular responses to TGF- β towards prostate cancer metastasis. Understanding how cancer cells interprete TGF-β signals from the microenvironment, is critical for defining the mechanism via which TGF- β function is switched from a growth suppressor to a metastasis promoter. Here we show that in prostate cancer, TGF-β action is directed by active cofilin enabling actin cytoskeleton changes and metastatic behavior. The significant association of cofilin with prostate cancer metastatic progression supports its predictive and targeting value in metastasis.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第21414号 / 医博第4404号 / 京都大学大学院医学研究科医学専攻 / (主査)教授 万代 昌紀, 教授 小川 誠司, 教授 武藤 学 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

No description available.

Wnt proteins have key roles in patterning of multicellular animals, acting at a distance from their sites of production. However, it is not well understood how these molecules propagate. This question has become even more puzzling by the discovery that Wnts harbour post-translational lipid-modifications, which enhance association with membranes and may therefore limit propagation by simple diffusion in an aqueous environment. The cellular mechanisms involved in Wnt propagation are largely unknown for vertebrate organisms. Here, I discuss my findings on the cellular localization of zebrafish Wnt8, as an example of a vertebrate Wnt. Wnt8 is a key signal for positioning the midbrain-hindbrain brain boundary (MHB) organizer along the anterior-posterior axis of the developing brain in vertebrates. However, it is not clear how this protein propagates from its source, the blastoderm margin, to the target cells, in the prospective neural plate. For this purpose, I have analysed a biologically active, fluorescently tagged Wnt8 in live zebrafish embryos. Wnt8 was present in live tissue in membrane associated punctate structures. In Wnt8 expressing cells these puncta localise to filopodial cellular processes, from which the protein is released to neighbouring cells. This filopodial release requires posttranslational palmitoylation. Although palmitoylation-defective Wnt8 retains auto- and juxtacrine signaling activity, it fails to signal over a long-range. Additionally, this Wnt8 palmitoylation is necessary for regulation of its neural plate target genes. These results suggest that vertebrate Wnt proteins use cell-to-cell contact through filopodia as a shortrange propagation mechanism while released palmitoylated Wnt is required for longrange signaling activity. Furthermore, I show that a Wnt8 receptor, Frizzled9 can negatively influence Wnt8 propagation and signaling range. Finally, I was able to determine the presence of an endogenous Wnt8 gradient in the neurectoderm. I discuss these findings in the context of Wnt8 signaling function in mediating anterior-posterior patterning during early brain development.

Epithelial cells are unique for their ability to strongly adhere to one another and coordinate communication across an asymmetrical, polar plasma membrane. These properties are necessary for carrying out normal epithelial function, such as absorbing/secreting molecules, repairing wounds, lining organs, etc. Cadherins, claudins, and occludins are major players of epithelial cell adhesion and polarity. Previously, transmembrane immunoglobulin domain containing-1, TMIGD1, was identified as a novel cell adhesion molecule, whose expression is downregulated in human renal carcinomas. Re-expression of TMIGD1 in renal tumor cells resulted in altered cell morphology and inhibition of tumor growth. In this study, we examined the hypothesis that TMIGD1 activity is associated with epithelial cell polarity. We demonstrated that TMIGD1 regulates actin stress fibril formation. A 3-dimensional (3D) cell culture assay was developed to examine the role of TMIGD1 in cell morphology and polarity. Our results demonstrate that TMIGD1 regulates actin fibril formation in Madin-Darby Canine Kidney (MDCK) cells, as blocking TMIGD1 activity by blocking antibody inhibited actin fibril formation in 3D cell culture system. Moreover, ectopic expression of TMIGD1 in rectal carcinoma cells, (RKO) , significantly inhibited filopodia formation. Taken together, our data identifies TMIGD1 as a possible regulator of epithelial cell morphology and polarity. / 2020-07-03T00:00:00Z

The lipid kinase phosphatidylinositol 4-kinase III β (PI4KIIIβ) phosphorylates phosphatidylinositol (PtdIns) to generate PI(4)P in the Golgi. PI4KIIIβ is likely involved in the development of breast cancer as it has been reported genetically amplified in a subset of human breast tumours and is a downstream effector of the eukaryotic elongation factor 1 alpha 2 (eEF1A2), a transforming gene that is amplified and highly expressed in approximately 60% of human breast tumours. The goal of my thesis is to investigate a role for PI4KIIIβ in breast oncogenesis. We show that PI4KIIIβ is highly expressed in approximately 20% of primary human breast tumours. Overexpression of PI4KIIIβ in an invasive breast ductal carcinomas cell line, BT549, increased the production of filopodial actin filament protrusions and enhanced in vitro proliferative capacity. Enhanced PI4KIIIβ expression did not impact the migratory rate of these breast cancer cells. We found that PI4KIIIβ expression activates Akt kinase in the BT549 breast cancer cell line. PI4KIIIβ overexpression led to an increase in the plasma membrane abundance of the PI3K derived PI(3,4,5)P3/PI(3,4)P2 lipids, upstream activators of Akt signalling. PI(4)P and PI(4,5)P2 are precursors to PI(3,4,5)P3 and PI(3,4)P2 generation, however, no changes in the overall cellular abundance or localization of PI(4)P or PI(4,5)P2 were detected in PI4KIIIβ-overexpressing cells. Inhibition of PI4KIIIβ kinase activity, using the drug Pik93, had no effect on PI4KIIIβ-mediated Akt activation. Additionally, ectopic expression of a catalytically inactive PI4KIIIβ also led to increased Akt activity and PI(3,4,5)P3/PI(3,4)P2 plasma membrane abundance. Together, this implies that PI4KIIIβ regulates Akt independently of PI(4)P generation. The PI4KIIIβ interacting protein, Rab11, is likely involved in PI4KIIIβ mediated Akt activation, as RNAi-mediated depletion of Rab11 suppressed the effect of PI4KIIIβ overexpression on Akt activation. Furthermore, PI4KIIIβ overexpression altered cellular Rab11 distribution and led to enhanced recruitment of PI4KIIIβ and Rab11 to recycling endosomes. Therefore, PI4KIIIβ is highly expressed in a subset of breast tumours and upregulated PI4KIIIβ expression enhances filopodia production and cell growth in vitro. Enhanced PI4KIIIβ expression increases PI(3,4,5)P3/PI(3,4)P2 plasma membrane abundance and Akt activation independently of its kinase function, through a mechanism that likely involves Rab11. This work suggests that PI4KIIIβ impacts breast oncogenesis by regulating PI3K/Akt signalling through Rab11 and endosomal trafficking.

Wnt proteins have key roles in patterning of multicellular animals, acting at a distance from their sites of production. However, it is not well understood how these molecules propagate. This question has become even more puzzling by the discovery that Wnts harbour post-translational lipid-modifications, which enhance association with membranes and may therefore limit propagation by simple diffusion in an aqueous environment. The cellular mechanisms involved in Wnt propagation are largely unknown for vertebrate organisms. Here, I discuss my findings on the cellular localization of zebrafish Wnt8, as an example of a vertebrate Wnt. Wnt8 is a key signal for positioning the midbrain-hindbrain brain boundary (MHB) organizer along the anterior-posterior axis of the developing brain in vertebrates. However, it is not clear how this protein propagates from its source, the blastoderm margin, to the target cells, in the prospective neural plate. For this purpose, I have analysed a biologically active, fluorescently tagged Wnt8 in live zebrafish embryos. Wnt8 was present in live tissue in membrane associated punctate structures. In Wnt8 expressing cells these puncta localise to filopodial cellular processes, from which the protein is released to neighbouring cells. This filopodial release requires posttranslational palmitoylation. Although palmitoylation-defective Wnt8 retains auto- and juxtacrine signaling activity, it fails to signal over a long-range. Additionally, this Wnt8 palmitoylation is necessary for regulation of its neural plate target genes. These results suggest that vertebrate Wnt proteins use cell-to-cell contact through filopodia as a shortrange propagation mechanism while released palmitoylated Wnt is required for longrange signaling activity. Furthermore, I show that a Wnt8 receptor, Frizzled9 can negatively influence Wnt8 propagation and signaling range. Finally, I was able to determine the presence of an endogenous Wnt8 gradient in the neurectoderm. I discuss these findings in the context of Wnt8 signaling function in mediating anterior-posterior patterning during early brain development.

No description available.

No description available.

No description available.