Residue-specific contributions to the energetics of the catalytic cycle of PGK

Biological molecules interact through multiple weak bonds, which define the specificity and the affinity of the interaction. One would expect that the total strength of this interaction would equal the sum of all the contributing weak bonds in isolation. However this is not the case, and interactions are often much weaker or stronger than expected (known as cooperativity). This often corresponds with the function of the molecules. The majority of functional biological molecules are proteins, the large macromolecules that are encoded by DNA. Proteins that bind to rare nutrients (biotin or iron) or highly unstable structures (rate-defining intermediates of chemical reactions) bind more tightly than expected, whereas other proteins bind and release abundant molecules quickly (for example the reactants and products of biochemical reactions, like glucose or lactate), but must still bind specifically. This is most striking in enzymes, which speed up biochemical reactions by binding to rate-defining intermediates of chemical reactions (transition states). They must also bind to the reactants and products of the reactions, which are very similar in chemistry to the transition state, but must be bound much more weakly. The focus of this study is how enzymes combine these two modes of binding in their reaction cycles, and how they use their intrinsic flexibility to do so. We wish to test whether structural tightening provides a mechanism of achieving this discrimination. NMR allows the measurement of the properties of individual atoms within large molecules, but there is a size limit to the size of molecules that can be studied. Over time this size limit is increasing as technology improves and is now at a stage where large enzymes like phosphoglycerate kinase (PGK) can be studied. This project will use this technology to determine the contributions that different atoms within this enzyme make to the binding of the transition state of the reaction it catalyses, using stable chemicals that resemble it (called transition state analogues). The conclusions should be broadly applicable to other enzymes. An understanding of this process is vital to the design inhibitors of enzymes for use as therapeutic agents (drugs) and to technologies that use enzymes out of their biological context, for example bioremediation. It will also help the theoretical understanding of how important biological molecules work.

In systems that feature multiple intermolecular interactions, the contribution that an individual interaction makes to the free energy of an assembly can be significantly larger than expected from the properties of that interaction studied in isolation. Classically, this cooperative effect is entropic, i.e. the penalty associated with forming a complex is only paid once for the assembly, as opposed to when each of the interactions forms in isolation. Recently, a more general phenomenon was suggested that results in enthalpic cooperativity. In a complex that is held together by multiple weak non-covalent interactions, individual intermolecular bonding interactions are weakened by residual intermolecular motion. If additional interaction sites are added to generate a more strongly bound complex, this motion is damped, and all the individual interactions become more favourable. However, the implications of such a 'structural tightening' model in general, and for enzyme catalysis in particular, are not clear. Our approach is to mutate selected positions remote from the active site and to investigate their effects on structural tightening, the thermodynamics of substrate and transition state analogue binding, and catalysis. The structural context of the tightening will be determined by NMR, using amide hydrogen exchange, chemical shift changes and relaxation to determine the stability, length and mobility of intramolecular interactions. The global effects will be determined by ITC and linked spectrophotometric enzyme assay. Recent advances in NMR mean that our target enzyme, PGK, a monomeric, 43kDa 'hinge-bending' enzyme, is amenable to such a study, and we have recently assigned the backbone resonances of a ligand free and a liganded form. Two transition state analogues that stimulate domain closure are available, in addition to substrate analogues and products.