I am (and have been, since my days as a graduate student) an out-and-out believer in the kinetic control theory of protein folding championed by Levinthal in the late nineteen sixties and early seventies, which was cogently supported by arguments provided by Wetlaufer and Ristow at around the same time. I believe that all talk of two-state folding equilibria for proteins (yes, even for small, single-domain, disulphide-less, proline-less proteins), beginning with Anfinsen’s description of the behaviour of RNaseA, constitutes a smoke-screen; a comforting scenario of much satisfaction to those who hold that conformational equilibria are facile, and can be rapidly formed, like the much more simple equibria applying to small molecules, without any consideration of the kinetic barriers and time-based arguments that preclude any random search of conformational space for the conformation of the lowest energy. Yes, of course, provided infinite time, proteins can be expected to find their lowest-energy conformation; meanwhile, there is neither enough time, nor are proteins designed to fold over infinite time. The native structure, I believe, is therefore a structure which is a local energy minimum in the energy-conformation landscape of any protein, with the chain having many different routes to folding to native structure, some of which pass through ‘wayside stations’ which are kinetic traps that can be overcome by chains, i.e., traps at which molecules halt transiently, allowing their detection. Thus, I hold that there are an infinitum of folding intermediates, and that folding only ‘appears’ to be a two-state process because (i) mostly no particular intermediate is sufficiently ‘populated’ to accumulate and become detectable, and (ii) folding is so efficient that most chains avoid all the kinetic traps. Thus, intermediates which are demonstrated to be long-lived and existing on the folding pathway, in kinetic terms, are probably of the least consequence to the success in folding of the bulk of any protein population, because they are the kinetic traps that some molecules could not avoid. Their absence, signifying evolutionary avoidance of kinetic traps by chains, masquerades as two-state folding. Now, what I did for this paper was to argue that it is better to look for intermediates in an equilibrium unfolding reaction, rather than in a kinetic unfolding reaction. Using a single-tryptophan protein, azurin, and time-resolved emission studies (including the construction of time-resolved emission spectra from such data), I show that azurin subjected to equilibrium unfolding contains sub-populations of different equilibrium unfolding intermediates, discernible in time-resolved emission spectra as separate sub-populations with different degrees of exposure of azurin’s single tryptophan. Protein folding enthusiasts will love reading this paper.

Click to Read in Detail