Structure of type I antifreeze protein and mutants in supercooled water.

Many organisms are able to survive subzero temperatures at which bodily fluids would normally be expected to freeze. These organisms have adapted to these lower temperatures by synthesizing antifreeze proteins (AFPs), capable of binding to ice, which make further growth of ice energetically unfavorable. To date, the structures of five AFPs have been determined, and they show considerable sequence and structural diversity. The type I AFP reveals a single 37-residue alpha-helical structure. We have studied the behavior of wild-type type I AFP and two "inactive" mutants (Ala17Leu and Thr13Ser/Thr24Ser) in normal and supercooled solutions of H(2)O and deuterium oxide (D(2)O) to see if the structure at temperatures below the equilibrium freezing point is different from the structure observed at above freezing temperatures. Analysis of 1D (1)H- and (13)C-NMR spectra illustrate that all three proteins remain folded as the temperature is lowered and even seem to become more alpha-helical as evidenced by (13)C(alpha)-NMR chemical shift changes. Furthermore, (13)C-T(2) NMR relaxation measurements demonstrate that the rotational correlation times of all three proteins behave in a predictable manner under all temperatures and conditions studied. These data have important implications for the structure of the AFP bound to ice as well as the mechanisms for ice-binding and protein oligomerization.

Modeling Pseudomonas syringae ice-nucleation protein as a beta-helical protein.

Antifreeze proteins (AFPs) inhibit the growth of ice, whereas ice-nucleation proteins (INPs) promote its formation. Although the structures of several AFPs are known, the structure of INP has been modeled thus far because of the difficulty in determining membrane protein structures. Here, we present a novel model of an INP structure from Pseudomonas syringae based on comparison with two newly determined insect AFP structures. The results suggest that both this class of AFPs and INPs may have a similar beta-helical fold and that they could interact with water through the repetitive TXT motif. By theoretical arguments, we show that the distinguishing feature between an ice inhibitor and an ice nucleator lies in the size of the ice-interacting surface. For INPs, the larger surface area acts as a template that is larger than the critical ice embryo surface area required for growth. In contrast, AFPs are small enough so that they bind to ice and inhibit further growth without acting as a nucleator.

Structure-function relationships in spruce budworm antifreeze protein revealed by isoform diversity.

The spruce budworm, Choristoneura fumiferana, produces antifreeze protein (AFP) to assist in the protection of the overwintering larval stage. AFPs are thought to lower the freezing point of the hemolymph, noncolligatively, by interaction with the surface of ice crystals. Previously, we had identified a cDNA encoding a 9-kDa AFP with 10-30 times the thermal hysteresis activity, on a molar basis, than that shown by fish AFPs. To identify important residues for ice interaction and to investigate the basis for the hyperactivity of the insect AFPs, six new spruce budworm AFP cDNA isoforms were isolated and sequenced. They differ in amino-acid identity as much as 36% from the originally characterized AFP and can be divided into three classes according to the length of their 3' untranslated regions (UTRs). The new isoforms have at least five putative 'Thr-X-Thr' ice-binding motifs and three of the new isoforms encode larger, 12-kDa proteins. These appear to be a result of a 30 amino-acid insertion bearing two additional ice-binding motifs spaced 15 residues apart. Molecular modeling, based on the NMR structure of a short isoform, suggests that the insertion folds into two additional beta-helix loops with their Thr-X-Thr motifs in perfect alignment with the others. The first Thr of the motifs are often substituted by Val, Ile or Arg and a recombinantly expressed isoform with both Val and Arg substitutions, showed wild-type thermal hysteresis activity. The analysis of these AFP isoforms suggests therefore that specific substitutions at the first Thr in the ice binding motif can be tolerated, and have no discernible effect on activity, but the second Thr appears to be conserved. The second Thr is thus likely important for the dynamics of initial ice contact and interaction by these hyperactive antifreezes.

Beta-helix structure and ice-binding properties of a hyperactive antifreeze protein from an insect.

Insect antifreeze proteins (AFP) are considerably more active at inhibiting ice crystal growth than AFP from fish or plants. Several insect AFPs, also known as thermal hysteresis proteins, have been cloned and expressed. Their maximum activity is 3-4 times that of fish AFPs and they are 10-100 times more effective at micromolar concentrations. Here we report the solution structure of spruce budworm (Choristoneura fumiferana) AFP and characterize its ice-binding properties. The 9-kDa AFP is a beta-helix with a triangular cross-section and rectangular sides that form stacked parallel beta-sheets; a fold which is distinct from the three known fish AFP structures. The ice-binding side contains 9 of the 14 surface-accessible threonines organized in a regular array of TXT motifs that match the ice lattice on both prism and basal planes. In support of this model, ice crystal morphology and ice-etching experiments are consistent with AFP binding to both of these planes and thus may explain the greater activity of the spruce budworm antifreeze.

Crystallization and preliminary X-ray crystallographic analysis of spruce budworm antifreeze protein.

Antifreeze proteins have the ability to bind to ice with high affinity and inhibit further crystal growth. The insect antifreeze protein from spruce budworm exhibits very high thermal hysteresis activity and is implicated in the protection of overwintering larvae from freezing. This protein has been crystallized in 20-25% polyethylene glycol (Mr 6000), 0.4 M NaCl, 0.1 M Tris-HCl, pH 8.5, by vapor diffusion using the hanging drop method. The resulting crystals are very thin (typically <0.01 mm in the shortest dimension), and only after repeated seeding could crystals be grown large enough for data collection using synchrotron radiation. The crystals belong to the monoclinic space group C2, with cell dimensions a = 82.28 A, b = 62.29 A, c = 63.63 A, and beta = 113.7 degrees. Molecules in the asymmetric unit are related by a twofold axis of symmetry with two molecules present. Native data to a resolution of 2.6 A have been collected with 90.3% completeness and a Rsym of 6.9%.

Quantitative and qualitative analysis of type III antifreeze protein structure and function.

Some cold water marine fishes avoid cellular damage because of freezing by expressing antifreeze proteins (AFPs) that bind to ice and inhibit its growth; one such protein is the globular type III AFP from eel pout. Despite several studies, the mechanism of ice binding remains unclear because of the difficulty in modeling the AFP-ice interaction. To further explore the mechanism, we have determined the x-ray crystallographic structure of 10 type III AFP mutants and combined that information with 7 previously determined structures to mainly analyze specific AFP-ice interactions such as hydrogen bonds. Quantitative assessment of binding was performed using a neural network with properties of the structure as input and predicted antifreeze activity as output. Using the cross-validation method, a correlation coefficient of 0.60 was obtained between measured and predicted activity, indicating successful learning and good predictive power. A large loss in the predictive power of the neural network occurred after properties related to the hydrophobic surface were left out, suggesting that van der Waal's interactions make a significant contribution to ice binding. By combining the analysis of the neural network with antifreeze activity and x-ray crystallographic structures of the mutants, we extend the existing ice-binding model to a two-step process: 1) probing of the surface for the correct ice-binding plane by hydrogen-bonding side chains and 2) attractive van der Waal's interactions between the other residues of the ice-binding surface and the ice, which increases the strength of the protein-ice interaction.

Tryptophan residues in caldesmon are major determinants for calmodulin binding.

Calmodulin has been shown to interact with the COOH-terminal domain of gizzard h-caldesmon at three sites, A (residues 658-666), B (residues 687-695), and B' (residues 717-725), each of which contains a Trp residue [Zhan et al. (1991) J. Biol. Chem. 266, 21810-21814; Marston et al. (1994) J. Biol. Chem. 296, 8134-8139; Mezgueldi et al. (1994) J. Biol. Chem. 269, 12824-12832]. To determine the contribution of each of the three Trp residues in the calmodulin-caldesmon interaction, we have mutated the Trp residues to Ala in the COOH-terminal domain of fibroblast caldesmon (CaD39) and studied the effects on calmodulin binding by fluorescence measurements and using immobilized calmodulin. Wild-type CaD39 binds with a Kd of 0.13 x 10(-6) M and a stoichiometry of 1 mol of calmodulin per mol of caldesmon. Replacing Trp 659 at site A or Trp 692 at site B to Ala reduces binding by 22- and 31-fold (Kd = 2.9 x 10(-6) and 4.0 x 10(-6) M), respectively, and destabilizes the CaD39-calmodulin complex by 1.75 and 1.94 kcal mol-1, respectively. Mutation of both Trp 659 and Trp 692 to Ala further reduces binding with a Kd of 6.1 x 10(-6) M and destabilizes the complex by 2.17 kcal mol-1. On the other hand, mutation of Trp 722 at site B' to Ala causes a much smaller decrease in affinity (Kd = 0.6 x 10(-6) M) and results in a destabilization energy of 0.87 kcal mol-1. To investigate the relative importance of the amino acid residues near each Trp residue in the caldesmon-calmodulin interaction, deletion mutants were constructed lacking site A, site B, and site A + B. Although deletion of site A decreases binding of CaD39 to calmodulin by 13-fold (Kd = 1.7 x 10(-6) M), it results in tighter binding than mutation of Trp 659 to Ala at this site, suggesting that the residues neighboring Trp 659 may contribute negatively to the interaction. Deletion of site B causes a similar reduction in binding (Kd = 4.1 x 10(-6) M) as observed for replacing Trp 692 to Ala at this site, indicating that Trp 692 is the major, if not the only, binding determinant at site B. Deletion of both site A and site B drastically reduces binding by 62-fold. Taken together, these results suggest that Trp 659 and Trp 692 are the major determinants in the caldesmon-calmodulin interaction and that Trp 722 in site B' plays a minor role.