Evolution and Binding Mechanisms of Intrinsically Disordered Proteins

Sammanfattning: Intrinsically disordered proteins (IDPs) make up a considerable fraction of the proteome in eukaryotic organisms. These proteins often act as hubs in interaction networks, harbouring multiple interaction with other proteins, and thus evolution has to walk a tightrope to accommodate new interactions while maintaining the previously established interactions. The ability to accommodate multiple ligands with high specificity is one of the fascinating properties of IDPs, and the molecular details of how this is achieved throughout evolution are largely unknown. The nuclear co-activator binding domain (NCBD) and CREBBP-interacting domain (CID) are intrinsically disordered domains of two transcriptional co-activator proteins. This interaction constitutes one of the earliest examples of a binding reaction where two binding partners fold synergistically upon binding. Previous phylogenetic analysis showed that NCBD is evolutionarily older than CID, which likely emerged after the divergence of the deuterostome and protostome clades of the animal kingdom. When NCBD adapted to the new ligand CID, the affinity of this interaction increased 10-20-fold, while the affinity for some other NCBD ligands were maintained. My thesis work has largely focused around understanding the evolutionary adaptation of the NCBD-CID complex. I have characterised a reconstructed ancestral NCBD-CID complex with respect to structure, folding and binding mechanism and compared these properties to those of the present-day human complex. The results show that the structure and disordered properties of NCBD and CID, as well as their overall binding mechanism, have been moderately conserved throughout evolution. Small differences in the binding mechanism and compactness of the complexes were observed between the most ancestral and present-day human complexes, indicating a somewhat malleable protein-protein interaction that allows for fine-tuning of biophysical properties when new ligands are adopted. Furthermore, I have investigated the impact of disordered regions flanking the binding interface of present-day human CID, using stopped-flow fluorimetry. The disordered regions contributed to an increased affinity to NCBD, although no additional structure was formed upon binding. Ionic strength-dependence curves of the obtained kinetic parameters showed that electrostatic interactions likely do not contribute to the increase in affinity mediated by the disordered flanking regions. These results demonstrate how disordered regions flanking the binding interface regions can contribute to affinity, and highlights the importance of including larger parts of proteins when conducting studies of proteins in vitro.