061 The roles of water in binding antifreeze proteins to ice

Abstract

The functions of ‘antifreeze proteins’ (AFPs) have expanded in recent years beyond the originally observed inhibition of ice growth in freeze-resistant fish and insects. AFPs are also involved in: freeze tolerance through their ability to inhibit ice recrystallization; the structuring of ice channels in the immediate vicinity of psychrophilic microorganisms; and in the adhesion of bacteria to ice. These diverse roles go beyond the prevention of freezing (antifreeze) and each requires that the AFPs bind to ice, hence the all-inclusive description – ice-binding proteins (IBPs). The mechanism by which IBPs can be soluble at mM concentrations in liquid water but adsorb irreversibly to water in its solid state has intrigued the field for the past four decades. To date, eleven different IBP structures have been solved and several others have been modeled. Site-directed mutagenesis identified the ice-binding sites on many of these diverse IBP structures as being relatively flat, hydrophobic and regular. Theoretical studies predicted that ice-binding sites can organize adjacent waters into an ice-like pattern to merge with, and freeze to, the quasi-liquid water layer at the ice surface. There is now support for this model from crystal structures of several IBPs. Waters form cages around hydrophobic moieties and these cages are anchored by hydrogen bonding to nearby hydrophilic groups. Thus, competing hypotheses of hydrogen-bonding to ice vs the hydrophobic effect as driving forces for adsorption have surprisingly converged. Once the IBP freezes to ice the protein is in effect hydrogen bonded to ice. Linear arrays of threonines on a beta-sheet are particularly suitable for forming these ice-like anchored clathrate waters. However, in order for the IBP to bind to ice there must be a quorum of these ice-like waters present when the IBP contacts the quasi-liquid layer. One way to increase the likelihood of a quorum is to increase the length or breadth of the ice-binding site through natural selection or protein engineering. This might explain why larger IBP isoforms have higher ‘antifreeze’ activity. Supported by the Canadian Institutes of Health Research. Source of funding: None declared. Conflict of interest: None declared. daviesp@post.queensu.ca