Structural and functional characterisation of FOXP1 and FOXP2 near-full length variants and their protein-protein interactions
Date
2019
Authors
Thulo, Monare
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
FOXP1 and FOXP2 are members of the FOXP subfamily of transcription factors characterised by multiple domains, including a highly conserved C-terminal DNA binding domain (DBD). Additionally, the FOXP proteins are characterised by two functional dimerisation interfaces, comprising of the DBD, the forkhead domain (FHD) and an N-terminal leucine zipper domain. Sequence analysis also shows several regions of intrinsic disorder in the sequence, including the polyglutamine region, the domain linkers and a C-terminal acid rich region, conserved in FOXP1, FOXP2 and FOXP4. FOXP1 and FOXP2 are expressed in several organs including the lungs and brain and are important during embryonic development. FOXP1 is associated with cognition and retardation while FOXP2 has been implicated in language development. The similarities in structure and function between FOXP1 and FOXP2 and the partial overlap that has been reported in their expression patterns in brain regions important for language may indicate a possible interaction between them. More recently, disruptions in FOXP1 have been implicated in speech delay, intellectual disability, and autism, suggesting that just like FOXP2, FOXP1 has a significant involvement in speech and language development. Because of multiple domains, the expression and purification of the full length FOXP protein is relatively difficult to achieve in vitro. In this study, the structural and functional characteristics of FOXP1 and FOXP2 variants were determined using biophysical techniques. Four variants, two of FOXP1 and two of FOXP2 were constructed containing the FHD and the leucine zipper domain and differing in whether or not they additionally contained the C-terminal acid rich tail. The four variants were expressed in the soluble Escherichia coli cell fraction and were purified to relative homogeneity using chromatography. Although it has been suggested that FOXP1 and FOXP2 regulate transcription as dimers, we cannot rule out the possibility of both proteins performing their functions as higher oligomers. Indeed, the formation of higher order oligomers was observed in this study through SEC, DLS and native PAGE measurements. Higher order oligomers-DNA complexes were observed in EMSA, therefore suggesting that formation of oligomers may also occur during DNA binding. The significance of the disordered C-terminal tail, ART, to structural stability was studied in the presence of a simple alcohol, ethanol. Simple alcohols are enough to destabilise the tertiary structure of a protein and to stabilise the helical structure, leading to a possible quaternary conformational state change. FOXP1 and FOXP2 variants exist in solution predominantly as higher order oligomers at different concentrations. However, for FOXP1 and FOXP2 LZ-FHD, decreases in solvent dielectric result in dissociation of the higher oligomers to form a mixture of monomer and dimer. Although there is also a decrease in quaternary state in the presence of ethanol for FOXP1 and FOXP2 LZ-End, the ART seem to be enough to maintain dimeric and higher oligomeric state. In agreement with other studies, the use of ethanol was able to destabilise the tertiary structure, shown by a red shift in fluorescence emission for each variant, and to a characteristic increase in helical content. The folding of FOXP1 and FOXP2 LZ-End in mixtures of water with a simple alcohol directly relate to decrease in the dielectric constant of the solution. The hetero-association of FOXP1 with FOXP2 was defined using pull down assays on purified FOXP1 and FOXP2 truncated variants encompassing the leucine zipper, FHD and ART and fluorescence anisotropy to study the binding affinity of FOXP2 variants for FOXP1 variants in vitro. Size exclusion chromatography showed that the hetero-associated proteins exist mainly as dimers and fluorescence anisotropy revealed relatively weak association, both in the presence and absence of DNA in comparison to the isolated FHD of both proteins. The loose assembly of FOXP1 and FOXP2 near full-length variants native structure suggest that this is a mechanism that is needed for both interaction with binding partners such as the other FOXP proteins during heterodimerisation and during DNA binding. These results also suggest a possibility of regulation by a dynamic equilibrium between different states which leads to a ‘partial occupancy’ upon DNA binding. The existence of different quaternary states and conformation suggests that both FOXP1 and FOXP2 might also control transcription as components of supramolecular regulatory complexes stabilised by different subdomains, including the C-terminal acid rich region. This, therefore, suggests that local folding of the proteins must be frequently coupled to DNA binding. Consequently, hetero-oligomerisation of the full-length protein could be a transient event, that occurs for limited time.
Description
Submitted in fulfilment of the requirements for the degree
Doctor of Philosophy in
Molecular and Cell Biology in the Faculty of Science, University of the Witwatersrand, Johannesburg, South Africa June 2019
Keywords
Citation
Thulo, Monare. (2019). Structural and functional characterisation of FOXP1 and FOXP2 near-full length variants and their protein-protein interactions. University of the Witwatersrand, https://hdl.handle.net/10539/29641