Novel insights into dishevelled and frizzled function

Abstract

The formation of a multicellular body during embryonic development is one of the most fascinating processes in biology. The cells are in constant communication and must be precisely coordinated. Several essential signalling pathways have been identified in cell-to-cell signal transmission. One such pathway is the WNT signalling network, which is highly conserved and plays an important role not only in embryogenesis but also in the maintenance of adult tissue homeostasis. Disturbances of the WNT signalling system are connected to many pathophysiological processes including cancer, rendering this pathway a great target for pharmacological treatment. The aim of this thesis is to broaden our knowledge regarding the function of the two main components of the WNT signalling cascade: the intracellular scaffold protein Dishevelled (DVL) and the membrane receptors from the Class Frizzled (Class F).

First, we have shown that DVL is located at the centrosome. We have explored the function of DVL in the regulation of the centrosomal cycle – a chain of events that centrosome undergoes during the cell cycle progression. DVL constitutes a part of the linker connecting two centrosomes that coordinate centrosomal separation. As a target of several kinases, DVL possesses many phosphorylation sites, which regulate its function and selectivity for its binding partners. We added another kinase to DVL’s portfolio of kinases – the centrosomal kinase Nima related kinase 2 (NEK2). Interaction between NEK2 and DVL is important for the regulation of the centrosomal cycle, where phosphorylation of DVL by NEK2 displaces DVL from the centrosome together with other linker proteins subsequently enabling centrosomal separation. Therefore, DVL represents a newly identified key player in the centrosomal cycle.

Next, we investigated the ability of receptors from the Class F to signal through heterotrimeric G proteins. We have shown that Frizzled 10 (FZD10) forms an inactive state complex with one specific isoform of heterotrimeric G proteins, G13, and this complex dissociates after addition of ligands for FZD10. Additionally, FZD10 overexpression induced Yes-associated protein/Tafazzin (YAP/TAZ) transcriptional activity. We have also found that FZD10 mRNA is expressed in endothelial cells of the developing brain. Since G13 is a crucial molecule in vascular development, particularly for the endothelial cells, we propose that FZD10 signalling through G13 and YAP/TAZ might represent a novel signalling axis in the development of the central nervous system (CNS) vasculature. Further, we explored the structure and function of another FZD homologue, FZD6. We identified a triad of conserved cysteine residues in the linker region, which is connecting the N-terminal cysteine rich domain (CRD) with the core of the receptor. These cysteines are crucial for the localization of FZD6 in the plasma membrane as well as for its interaction with the cytoplasmic scaffold protein DVL. Additionally, we utilized a method involving receptor purification and insertion into artificial phospholipid bilayer particles to investigate the ability of FZD6 to directly activate heterotrimeric G proteins. We showed that FZD6 activates Gi as a monomeric unit by its constitutive activity. The method itself harbours great potential for future biochemical and pharmacological studies.

Taken together, this thesis provides new evidence for the involvement of the WNT pathway components in the less investigated WNT signalling events, connects DVL biology with the centrosomal cycle regulation, expands the portfolio of FZD homologues forming an inactive preassembly complex with heterotrimeric G proteins, and presents FZDs as direct activators of heterotrimeric G proteins.