Investigating the Non-globular Proteins of the Canonical Wnt Signalling Pathway

The canonical Wnt pathway is a vitally important signalling pathway that plays an important role in cell proliferation, differentiation and fate decisions in embryonic development and in the maintenance of adult tissues. The twelve Armadillo (ARM) repeat-containing protein beta-catenin acts as the signal transducer in this pathway and is continuously degraded in the cytosol by the beta-catenin destruction complex (BDC). Upon receiving the Wnt signal the BDC is inactivated, allowing beta-catenin to accumulate in the cytosol and be transported to the nucleus where it binds to the TCF/LEF family of transcription factors, inducing the expression of cell cycle promotor genes. In this Thesis I describe investigations into the roles of leucine-rich repeat kinase 2 (LRRK2) and the transcription factor TCF7L2 within this signalling pathway. LRRK2 is a large multi-domain protein with strong links to Parkinson’s disease and suggested to play a role in inactivating the BDC in response to the Wnt signal. A recent paper proposed that the previously uncharacterised regions of LRRK2 contain a series of tandem repeat sub-domains. I began an investigation into these sub-domains but I was unable to produce soluble protein constructs despite the use of a range of common techniques, and so I was forced to conclude this project early. The main body of this thesis focuses on the interaction between the intrinsically disordered TCF7L2 and the repeat protein beta-catenin, a very long interface of approximately 4800 Å2 that spans from the third to the eleventh ARM repeat of beta-catenin and residues 12 to 50 of TCF7L2, as determined by X-ray crystal structures. First, a fluorescence reporter system for the binding interaction was developed and used to determine the kinetic rate constants for the association and dissociation of the wild-type construct using stopped-flow fluorescence spectroscopy and time-dependent fluorescence spectroscopy. It was found that association of TCF7L2 and beta-catenin was rapid (7.3 ± 0.1 x107 M-1s-1) with only a single phase was observed, whereas dissociation was biphasic and slow (5.7 ± 0.4 x10-4 s-1, 15.2 ± 2.8 x10-4 s-1). Using either of these two dissociation rate constants the calculated Kd value obtained is much lower than the values previously reported in the literature (8 ± 1 / 20 ± 2 pM compared with 16 nM). This reporter system was then used to investigate the striking variability between three crystal structures previously obtained for the TCF7L2-beta-catenin complex, in which different regions of TCF7L2 show different elements of secondary structure. Mutational analysis revealed that the interface residues on TCF7L2 identified in these structures make little or no contribution to the overall binding affinity, pointing to a transient nature of these contact in solution and suggesting that the observed differences between the structures are due to differences in crystal packing. Further experiments into the effect of osmolarity on the binding equilibrium and kinetics supported this conclusion and suggest a change in the association/dissociation mechanism as a function of ionic strength. Lastly, further mutational analysis of TCF7L2 revealed two regions that contribute particularly strongly to the binding kinetics, suggesting that TCF7L2-beta-catenin assembly proceeds via a two-site avidity mechanism. Some of the most destabilising variants display two additional dissociation phases, indicating the presence of an alternative dissociation pathway that is inaccessible to the wild-type. In summary, the results presented here provide insights into the kinetics of molecular recognition of a long intrinsically disordered region with an extended repeat protein surface, a process shown to involve multiple routes with multiple steps in each