Antifreeze activity

Destruction of the galactosyl residues by oxidation with periodate, acetylation of the free hydroxyl groups of the oligosaccharides, or their removal by P elimination all lead to loss of antifreeze activity. 

Franks, F., Darlington, J., Schenz, T., Mathias, S.F., Slade, L. LeVine, H. (1987). Antifreeze activity of antarctic fish glycoprotein and a synthetic polymer. Nature 325, 146-7. 

Y. Tachibana, G. L. Fletcher, N. Fujitani, S. Tsuda, K. Monde, and S.-I. Nishimura, Antifreeze glycoproteins Elucidation of the structural motifs that are essential for antifreeze activity, Angew. Chem. Int. Ed., 43 (2004) 856-862. 

VII. Other Proteins Reported To Have Antifreeze Activity. 246... 

Antifreeze activity Strong Strong Strong Strong Strong None None None... 

The active AFGP 1-5 were first reported by Chuba and co-workers (1973) to inhibit hemagglutination by a lectin prepared from the seeds of the Osage orange (Maclura pomifera). Although the structural requirements for the antilectin activity closely parallel those for antifreeze activity, these two activities appear to be unrelated, with the probable exception that they both require considerable integrity of the carbohydrate side chain and polymeric size. 

A single attempt to estimate the minimum size of the polypeptide chain needed for antifreeze activity indicated that chains shorter than... 

Fig. 24. Loss of antifreeze activity as a function of the appearance of free amino groups when a mixture of antifreeze glycoproteins (AFGP) 3 and 4 was hydrolyzed with subtilisin (A) or Pronase (B) at37°C. From Komatsuet al. (1970a), reproduced with permission.

Fig. 26. Inactivation of antifreeze and antilectin activities of antifreeze glycoprotein by /3-elimination of carbohydrate. Antilectin activity was determined by inhibition of Osage-orange lectin. Antifreeze activities were determined by measurements of freezing temperatures. /3-Elimination was done with glycoprotein concentrations of 0.1 mg/ml in 0.1 N NaOH at 20°C. Extent of/3-elimination was determined according to data of Table VIII. From Ahmed et al. (1973), reproduced with permission.

All the antifreeze activity was lost when approximately 35% of the hydroxyl groups (and the one NH2-terminal group) were acetylated (Komatsu et al., 1970a). Recovery of >95% of the activity was obtained on O-deacetylation with hydroxylamine. When 32% of the hydroxyl groups were acetylated, 18% of the antifreeze activity and 40% of the antilectin activity were retained (Ahmed et al., 1973). 

A possible explanation for these losses of activity could be that the negative charges cause undesirable ionic effects, possibly repulsions. This hypothesis appears to be at least partially substantiated by measuring the antifreeze activities under conditions where the negative charges would be decreased. When the antifreeze activities were... 

The addition of borate (Vandenheede et al., 1972 Ahmed et al., 1975) to solutions of AFGP causes reversible losses of antifreeze activity. 

Both antifreeze and antilectin activities are lost when 2 mol of borate are bound per disaccharide side chain, and these losses and bindings are pH dependent (Ahmed et al., 1976). When borate was added to AFGP components 1-4 there was a gradual decrease in antifreeze activity as the pH was raised from 7.0 to approximately 8.4, at which point about one-half of the activity had been lost. This was followed by a sharp decrease from pH 8.4 to pH 8.8 (Fig. 28). Similar effects were obtained when AFGP polyaldehyde (i.e., the product from oxidation by galactose oxidase) was used. 

Fig. 28. Effect of pH on antifreeze activity of antifreeze glycoprotein in the presence of borate. Antifreeze glycoproteins 1-4 (5 mg/ml) were measured in 0.1 M phosphate and 0.1 M borate and adjusted to the pH values indicated. The freezing temperatures of the solutions were measured by the sensing of the heat of fusion. The values for the freezing temperatures of control solutions were subtracted from the values for samples containing the antifreeze glycoprotein. From Ahmed et al. (1976), reproduced with permission.

The amount of borate bound to the antifreeze glycoprotein was strongly influenced by the pH of the solution between pH 7.0 and 9.0. The number of moles of borate bound per disaccharide unit was only about 0.5 at pH 8.0 but approached 2 at pH 9.0 (Table XIX). Comparison of these data with those of Fig. 28 shows that approximately 20% of the antifreeze activity would be lost when 0.5 mol of borate were bound per mole of disaccharide, and almost 100% of the activity would be lost when 2.0 mol of borate were bound. 

Since both the antifreeze activity in the presence of borate and the amount of borate bound to the glycoprotein were shown to be func-... 

Initial studies on the polar cod blood serum have shown that its antifreeze protein is a glycoprotein with most properties similar to those from the Antarctic species T. borchgrevinki. Both active and inactive components were obtained. The active components had the same ratios of alanine to threonine (approximately 2 1) as found in the Antarctic glycoproteins and the same contents of galactosamine and galactose. No other amino acids were found. In contrast to the comparatively low antifreeze activity of the saffron cod reported by Raymond and co-workers, the AFGP from the polar cod had an antifreeze activity similar to that from the Antarctic species. 

Fig. 31. The identification of flounder antifreeze glycoproteins on Sephadex G-75 chromatography. Freshly prepared concentrated sera (2.5 ml) were applied on the column (1.5 X 86 cm), and 70 fractions (2.5 ml each) were collected. The antifreeze activities of each fraction were monitored with an advanced osmometer. From Hew and Yip (1976), reproduced with permission.

Here, however, instead of disaccharides serving as hydrophilic links to the embryonic ice-lattice sites, one can postulate that certain hydrophilic groups of the flounder, e.g., Glu and Asp residues (Duman and DeVries, 1976), can attach to the ice surface. Then the surface disruption mechanism and temperature dependence of this disruption as previously discussed for the disaccharide group may be carried over to the Asp and Glu residues. As the temperature is lowered, gradual incompatibility of the bonding sites for the helical protein would lead to the saturation of the antifreeze activity. 

Finally, it is necessary to emphasize that, owing to the scarcity of data on the physical characterization of the flounder and the synthetic AFP systems, all these ideas are still awaiting more conclusive evidence. In particular, the degree of antifreeze activity in the AFGP, the AFP, and the synthetic AFP are different. Is there a variation of degrees of one antifreeze mechanism, or is there more than one basic mechanism of antifreeze and saturation of the antifreeze function ... 

---