Water structure theory and some implications for drug design

The development of theories of water structure has been hindered in the past by the difficulty of experimental measurement. Both measurement and computer modelling studies have now reached the stage where theoretical treatments of water structure are converging to a broadly acceptable model. In current understanding, water is a mixture of randomly hydrogen-bonded molecules and larger structures comprised of tetrahedral oxygen centres which, when hydrogen-bonded to each other, lead to five-membered and other rings which can aggregate to form three-dimensional structures. Evidence is taken from studies of the ices, from clathrates and other solid solutions, as well as from liquid solutions, that certain motifs occur very frequently and have relatively high stability, such as the (H2O)20 cavity-forming structure known from studies on clathrates. The implications of recent models of water structure for an understanding of biological events, including the interactions of drugs with receptors, are profound. It is becoming clear that modelling of aqueous solutions of any molecule must consider the explicit interactions with water molecules, which should not be regarded as a continuum: water itself is not a continuum. Solute molecules which possess hydrogen-bonding groups will provoke the formation of further hydrogen-bonding chains of water molecules: if these can form rings, such ringswilltend to persist longerthan chains, givingthesolute a secondary identity of associated water which may play a role in molecular recognition. Solutes that do not have hydrogen-bonding capability, or regions of solutes which are non-polar, may also produce partial cage-like water structures that are characteristic of the solute. The classification of many solutes as structure makers or structure breakers has relevance to the interactions between ligands and large biomolecules such as proteins. While it is generally accepted that sulfate and urea, respectively structure maker and breaker, may alter protein conformation through effects on water, it has not been recognised that bioactive ligands, which also change the conformation of proteins, may do so by a related, but more selective, mechanism. Very early studies of cell contents suggested that the associated water might be different from bulk water, a concept that lost support in the mid-20th century. Current theories of water structure may invite a reappraisal of this position, given the observation that structuring may extend for many molecular diameters from an ordered surface.