Integrative Approaches for Characterising Small-Molecule Binding to Disordered Proteins

Many processes in biology (e.g. signalling, stimulation) and how we manipulate them (e.g. pesticides, drugs) depend, at the molecular level, on large biomolecules called proteins, and how they interact with smaller chemicals (or 'small molecules').

Most proteins (termed 'structured proteins') adopt well-defined 3D shapes which are associated with a specific function. Generally, small molecules insert themselves into grooves on the surfaces of structured proteins, which in turn, alter protein movement and function. Often, this is referred to as the 'lock-and-key' binding mechanism because the small molecule fits into the protein's grooves much like a key fits into a lock.

Nevertheless, many proteins (including those from humans, viruses, & plants) never adopt a single shape and instead rapidly interconvert between many shapes. These proteins, called 'disordered proteins' lack long-lived grooves ('locks') to which small molecules ('keys') can bind. For many decades it was believed that small molecules do not interact with disordered proteins, because it was unclear how these interactions could take place. Nevertheless, recent work suggests that disordered proteins can indeed bind small molecules, but the mechanisms differ from 'lock-and-key' binding. Instead, disordered proteins 'dance' with small molecules, such that each protein shape interacts with the small molecule in a unique way. I was one of the first to describe these new mechanisms (for a disordered protein involved in Alzheimer's disease), but there is still much to be understood about the molecular details of binding. The greatest bottleneck to addressing this gap is the lack of available tools to study these processes.

As a BBSRC Fellow, I will combine two approaches, one experimental and one computational, towards the development of new interdisciplinary tools. The experimental technique is called Nuclear Magnetic Resonance spectroscopy (NMR), in which strong magnetic fields are used to study the physical and chemical properties of proteins and small molecules. Most NMR experiments were developed for structured proteins. Thus, I will establish new methods specifically for the application of small-molecule binding to disordered proteins. I will also employ a computational approach called molecular dynamics (MD) simulations, in which I will model the movement of disordered proteins and small molecules using supercomputers. The MD simulations will allow me to create 'movies' to 'see' how small molecules and disordered proteins interact with one-another, which is valuable insight that I cannot get from NMR alone. Nevertheless, certain approximations must be made in my models to make the calculations affordable, and these can lead to inaccuracies. Thus, I will develop new tools that allow me to incorporate experimental NMR data into the simulations to improve accuracy. By combining both NMR and MD simulations, I can overcome the limitations of each technique alone and provide new insight into the molecular mechanisms of how disordered proteins interact with small molecules.

I will apply these tools to specific disordered proteins to discover new small-molecule binding mechanisms. For example, the Non-Structural protein 5A (NS5A) from hepatitis C virus interacts with several small molecules, including an antiviral, but the mechanisms of how these molecules bind (and how the antiviral works) remain unclear. I will also apply these tools to study a human disordered protein called FUsed in Sarcoma (FUS). FUS undergoes a phenomenon termed 'liquid-liquid phase separation' which is very similar to the formation of oil droplets in water. In the presence of high concentrations of certain small molecules, called 'nucleotides', FUS does not undergo phase separation, but it is not clear why. My new tools will allow me to 'see' and understand these binding mechanisms, discover new fundamental biology, and exploit it towards the development of novel biotechnology.

Intrinsically disordered proteins (IDPs) are highly prevalent biomolecules, estimated to make up one-third of all eukaryotic proteins. In contrast to structured proteins, which often have a single, well-defined conformation, IDPs are highly dynamic, rapidly interconverting between different states. IDPs are generally considered to be 'untargetable' by small molecules because they lack long-lived rigid binding sites. Nevertheless, recent evidence suggests that IDPs can indeed bind small molecules; however, the mechanisms of these interactions differ greatly from 'lock-and-key' binding. Instead, IDPs often remain highly dynamic in their small-molecule bound states, but much remains unknown about these interactions, including underlying biophysical mechanisms, their generalisability, and roles of specificity. The major bottleneck to fully uncovering the molecular details underpinning the interactions between small molecules and IDPs is the lack of available techniques available to detect and characterise such highly dynamic binding. As a BBSRC Discovery Fellow, I propose to combine computational approaches, namely molecular dynamics simulations, and experimental approaches, primarily Nuclear Magnetic Resonance spectroscopy, to develop new interdisciplinary tools to bridge this major technological gap. I will apply these methods to answer long-standing questions about how small molecules, including antivirals and metabolites, interact with viral and human IDPs, respectively. This insight will help elucidate the 'rules' of small-molecule interactions with IDPs, towards the establishment of novel binding paradigms. Given the high prevalence of IDPs and proteins with long flexible regions in plants, animals, and humans, I anticipate that the tools I propose to develop and the resulting knowledge will be broadly applied for both basic research and industrial applications, including drug and pesticide discovery.