Water in protein crystals

Finney, J. L. The Organization and Function of Water in Protein Crystals, in Water — a Comprehensive Treatise (ed. Franks, F.), Vol. 6, chapter 2, New York, Plenum Press 1979... 

Nayal, M. and Di Cera, E. (1996). Valence screening of water in protein crystals reveals potential Na binding sites. J. Mol. Biol. 256, 228-34. 

Jensen, L. H, The structure of water in protein crystals. Developments in Biological Standardization, Vol. 74, p. 53-61. Acting Editors JoanC. May-F. Brown. S. Karger AG, CH-4009 Basel (Switzerland), 1992... 

Finney JL (1979) The organisation and function of water in protein crystals. In Franks F (ed) The physics and chemistry of water. A comprehensive treatise, vol. 6. Plenum Press, New York, pp 47-122... 

Structural information and tabulations of the type given for lysozyme are of the greatest importance in understanding hydration. Nevertheless, the consumer of results of antdyses of ordered water in protein crystals perhaps should keep in mind several caveats ... 

Finney, J.L. "The Organization and Function of Water in Protein Crystals" in "Water a Comprehensive Treatise" (Franks, F. ed) Plenum Press, New York and London 1979 ... 

The information obtained from X-ray measurements on the arrangement of the water molecules naturally depends very much on the resolution and state of refinement of the crystal structure investigated. For detailed information on the organization of water molecules in the protein hydration shell at the surface and on the bulk water in the crystals a 1,2 to 1,8 A resolution range is necessary 153>. 

These compounds have remarkably high melting points (see Thble 21.2), considering their water content, e.g., [(i-C5H11)4N] + [F 38H20] has a melting point of 31 °C and contains 68 wtM of water [795], which is comparable to the water content in protein crystals. 

This water structure illustrates the level of complexity of these periodic four-connected nets of water molecules that can occur when perturbed by molecular species that have both hydrophilic and hydrophobic character. If the condition of high-resolution periodicity throughout the water structure is relaxed, as in protein crystals, even greater complexity is possible. 

Blake et al. (1983) refined the structures of human lysozyme (HL) and tortoise egg white lysozyme (TEWL) to 1.5 and 1.6 A resolution, respectively. The diffraction was modeled as arising from three components the protein, ordered water, and disordered water. Most of the water in the crystals (i.e., 60—80%) is disordered. The analysis located 143 molecules of ordered water out of about 350 per HL molecule, and 122 molecules out of 650 per TEWL molecule. The ordered water covers 75% of the available surface of the the protein. One-third (TEWL) to one-half (HL) of the total surface is unavailable for analysis of the adjacent water, owing to crystal contacts or disorder in the protein region. Thus, the estimate of surface coverage is in good agreement with the 300 molecules of water estimated by heat capacity measurements as full hydration (0.38 h). The area covered per water molecule is estimated as 18.9... 

Parak and collaborators (Hartmann et al., 1987 Parak et al., 1987) carried out X-ray structure analyses for metmyoglobin at temperatures of 80—300 K. One hundred sixty water molecules, more than one-third of the water in the crystals, were included in the refinement. The disorder is higher for water that for protein atoms (Fig. 31). Reduction in temperature partially freezes out the disorder for the water, as for the protein. The residual disorder at low temperature has been understood to represent conformational substates or a distribution of conformations, frozen in at low temperature and in mobile equilibrium at high temperature. 

X-ray diffraction data from crystals are either collected at room temperature or under cryogenic conditions at liquid nitrogen temperatures [around 100°K(-170°C)]. For room temperature data collection, crystals are normally mounted in thin-walled glass capillaries, with a small amount of mother liquor about 5 mm from the crystal. The mother liquor in the capillary is critical because protein crystals are 40-80% water—dried protein crystals do not diffract. The nearby mother... 

In protein crystal structures, ordered water molecules were frequently observed at instances where a-helices bend or fold. Molecular dynamic simu-... 

A. McKenzie of Canberra and I have embarked on a review of the present state of our knowledge of water, especially in its relation to proteins. The first part of this study has been published ( ) we are still at work on the second and final part. We are not reporting on original research of our own, and I have indeed retired from the laboratory several years ago but we hope to provide some perspective on aspects of aqueous protein systems that have long preoccupied us. Here I will briefly discuss only two points (1) an aspect of hydro-phobic interactions that has only recently become apparent, and is still not widely noted, and (2) a few aspects of the location and mobility of water molecules in protein crystals, and (by inference) in solutions. 

This paper presents a very brief history of some past thought and research concerning water, a discussion of the unusual features of the partial molal compressibilities of hydrophobic substances in water, and a brief discussion of the evidence from x-ray and neutron diffraction for the locations of water nwlecules in protein crystals. 

Nonexponential NMR relaxation was reported several years ago for water protons in protein crystals (12). This report has been largely ignored apparently because of a less than adequate explanation that leaned heavily on a chemical exchange model. 

Partial occupancy of atom sites is a relatively common special case of substitutional disorder, and non-coordinating solvent molecules are frequently found to occupy only about half of the voids in the crystal lattice. The presence of half waters in protein stractures is a typical example. Unusually high displacement parameters are a sign for partially occupied solvent molecules however, one should take into account that, due to their mobility, even fully occupied non-coordinating solvent molecules tend to show relatively high displacement parameters. Therefore, the ADPs should be drastically larger to justify a reduction of the occupancy factors. The residual electron density map, which shows negative electron density at or around the nuclear positions if the true occupancy is lower than one, is a better criterion. 

Water in proteins has been determined from hydration isotherms obtained gravimetrically and the more recent quartz crystal resonator techniques. The steady state conductivity of protein samples increases rapidly with water absorbed. The weight percentage, m, relates to the conductivity according to the Spivey equation,... 

---