Theory of the Hydrophobic Effect.

Lawrence Pratt (Los Alamos Natl. Laboratory) and Andrew Pohorille

Hydrophobic effects are the driving force for the formation of membranes and micelles from aqueous solution. Hydrophobic effects are commonly believed to play a similar role in the folding of globular proteins, although proteins are very much more complex systems. However, because hydrophobic effects are an entropic manifestation of solvation by a remarkably complex liquid (water!), construction of a valid molecular scale description of hydrophobic interactions has proved to be very difficult. No such molecular-scale description is currently available.

Recently, a collaborative effort between scientists and Los Alamos Natl. Laboratory, NASA-Ames and Univ. of Delaware has led to the development of computational tools which provide new molecular-scale information on hydrophobic effects. Applications of these methods have provided answers to basic questions about the molecular origin of hydrophobic effects, questions that had been asked but not answered for some decades. In particular, venerable and very different theoretical models suggestive of very different physical conclusions were successfully distinguished for the first time on the basis of the results obtained by these new methods. Moreover, these new methods couple naturally to the preferred theoretical tools for study of the molecular structure of biological macromolecules, namely computer simulation by molecular dynamics or Monte Carlo techniques. In fact, these developments have lead to a proposal for the construction of ``hydrophobic images'' of proteins on the basis of established molecular principles and computer simulation data. Much as a suitable solution of the Poisson equation can produce a ``map'' of the electrostatic potential, such a hydrophobic map would distinguish hydrophobic from hydrophilic regions of a protein by answering a question: ``where would inert gas molecules stick?'' Answers to such a question would provide entirely new information about protein structures, and we can reasonably expect that it would lead to new insights into the factors which determine these structures. So far, classifications of protein surfaces based on hydrophobicity were of a taxonomic sort, i.e., surfaces of a protein were labeled as hydrophobic merely on the basis of the a classification scheme for amino acids rather than physical properties characteristic of these regions.

References:

``An information theory model of hydrophobic interactions'', G. Hummer, S. Garde, A. E. Garcia, A. Pohorille and L. R. Pratt, Proc. Natl. Acad. Sci. USA, 93, 8951-8955 (1996).

``Transient cavities in liquids and the nature of the hydrophobic effect'', A. Pohorille, Polish J. Chem., 72 1680-1690 (1998).

``Theory of hydrophobicity: transient cavities in molecular liquids'', L. R. Pratt and A. Pohorille Proc. Natl. Acad. Sci., 89 2995-2999 (1992).

"Cavities in Molecular Liquids and the Theory of Hydrophobic Solubilities." A. Pohorille and L. R. Pratt, J. Am. Chem. Soc. 112, 5066 (1990).