Investigation of alternative states of barnase

Enzymes are the molecules that catalyse biological reactions. Almost all enzymes are proteins, and they are large and complicated molecules. This is inevitable, because of the job they have to do. All enzymes work by being able to stabilise the 'transition state', which is the highest energy part of the reaction pathway. They therefore have to be able to recognise, and bind to, the starting materials of the reaction ('substrates'), the transition state, and the products of the reaction. This means that an enzyme has to be flexible, to accommodate these three stages of the reaction, which always have different shapes and (usually) a different distribution of electric charges. This much is known, but much else is surprisingly poorly understood. It was for example suggested a long time ago that enzymes undergo 'induced fit', in which binding of the substrate causes the structure of the enzyme to change, in order to better match the transition state. However, more recently it has become clear that enzymes may undergo induced fit motions even in the absence of substrates. In which case the question arises; what exactly does the substrate do? For example, does it change the motions of the enzyme, or does it stop them, by freezing the enzyme in an active state? Or does it redirect the motion so that the enzyme 'pushes' the substrate in the appropriate way to help the reaction to happen? These are fundamental questions, which are important to answer because they will enable us to engineer better enzymes in future. The enzyme being studied here functions to digest RNA, and is called barnase. Rather than study binding of substrate, we will study inhibitors, which are more stable. We will study two different types of inhibitor: a mimic of the substrate, plus a naturally occurring protein inhibitor. The technique we will use is nuclear magnetic resonance (NMR), which provides detailed information about motional states of individual atoms in the enzyme. We have been developing a novel technique to characterise alternative states of proteins, that are only populated a few percent: these are the types of states involved in these induced fit motions. The purpose of this research is to make detailed comparisons of our technique to other NMR techniques for probing alternative states, in particular a technique called relaxation dispersion. Relaxation dispersion is an exciting measurement, because it provides timescales and populations of alternative states. However, it is only sensitive to a relatively limited range of timescales, between about 10-3 s and 10-6 s. There are other NMR techniques that can look at the range from 10-9 s and faster, but so far nothing that can look in the intermediate range, between 10-6 and 10-9 s. This is a big gap, and one that includes many of the motions that are suspected to be important for enzyme function. Our research will provide a complete picture of what motion is happening, where in the protein, and how fast. We will also measure whether motions of different atoms are correlated, that is, whether they are part of the same movements or are independent. These are detailed measurements, but they will for the first time enable us to say with confidence how the protein moves, and therefore how the motions relate to its function.

We are still very far from understanding how enzymes move, and how this is related to their catalytic activity. We have shown using a novel temperature-based technique that barnase populates catalytically relevant non-native states (ie, different from the ground state) in both free and bound forms. The questions we are looking to answer, using barnase as an example, are (1) do the same slow-timescale motions occur in the free enzyme and in complexes with inhibitors? And can we characterise these motions, measuring for example the amplitude and timescale of motions at each residue throughout the protein? (2) are these motions correlated, ie are there concerted motions for example to move whole loops or domains in 'induced fit' type motions? (3) how does the binding of an inhibitor change these motions? / does it freeze enzyme motion partially or completely, or redirect motions to more 'catalytically relevant' ones? (4) do different inhibitors (ie substrate-like and protein inhibitors) have the same effect on the alternative states present? (5) can we probe concerted motions using selected mutants? And finally (6) are our temperature-dependent experiments a good alternative to the much more difficult and time-consuming relaxation experiments, for measuring motional details? We will therefore compare our recent results (and novel results, to be obtained from a barnase/barstar complex) to relaxation dispersion measurements. These measurements will be made at several different field strengths and several different temperatures, to provide as complete a picture as possible of timescales, populations and energies, and will be made with the support of Dr Akke. We shall also use a novel relaxation experiment to quantify correlated motions. These measurements will be combined with exisinting relaxation and molecular dynamics results to build up a detailed picture of how this enzyme functions.