A limitation of two-state analysis for transitions between disordered and weakly ordered states

The stability of the secondary structure of particular peptide regions is often used to investigate the involvement of the region in protein folding. When analysing the relatively small populations of associated states that are formed by weak interactions (i.e. those interactions that are comparable to thermal energies), it is common practice to characterise the associated state by a parameter that is measured when this state is highly occupied. The accuracy of this method, however, has not yet been determined.

Using as a model the vancomycin group of antibiotics, either forming dimers or binding to cell wall precursors, we have investigated the dependence of the limiting (i.e. fully associated) chemical shifts of two protons on the equilibrium constants for the formation of the fully associated states. The chemical shift shows a large variation with the equilibrium constant for the formation of the fully associated state.

The results demonstrate, in two systems, that a parameter representing a fully associated state (chemical shift) varies greatly with the equilibrium constant for the formation of that associated state. The results have implications for two-state analyses of populations of protein fragments in which a parameter representing the fully associated state is taken to be independent of the equilibrium constant for its formation. Using two-state analysis to determine the population of associated states of protein fragments could result in an underestimation of the population of these associated states.