|Kelterer, Anne-Marie - TECHNISCHE UNIVERSITAT|
|Cramer, Christopher - UNIV OF MINNESOTA|
Submitted to: Sugar Processing Research Conference Proceedings
Publication Type: Proceedings
Publication Acceptance Date: April 5, 2004
Publication Date: September 1, 2004
Citation: French, A.D., Johnson, G.P., Kelterer, A., Cramer, C.J. The shape of sucrose molecules. Proceedings of Sugar Processing Research Conference. 2004. p. 417-427. Interpretive Summary: The purpose of this paper was to convey progress in understanding the shapes of sucrose (table sugar) molecules. These shapes are important keys to understanding its interactions with surrounding molecules, including those in the mouth that signal a sweet taste. Sucrose is one of a number of carbohydrates of agricultural interest, and it has proven especially difficult to learn its shape properties by computerized molecular modeling. This paper finds that the shapes that have been determined by experiment of sucrose-like molecules in their crystals are similar to each other, but there is substantial flexibility. Also, sophisticated molecular modeling methods can predict these shapes well only under certain conditions. Those conditions should be applicable to the modeling of other agricultural carbohydrates such as cellulose, the main molecule found in cotton fiber, and starch. This work is primarily of interest to those who study the chemistry and biochemistry of carbohydrate molecules, as well as computerized molecular modeling.
Technical Abstract: The shape properties of sucrose have many important ramifications. They are responsible for its sweet taste, for its crystallization behavior and for many of the intermolecular interactions that are unique to sucrose. The shapes of sucrose are conveniently described in terms of the extent of rotation of the glucose and fructose monomers about their bonds to the mutually held oxygen atom. The most definitive method for studying shape is diffraction crystallography. It precisely locates each of the atoms but it requires a crystal and in most cases gives only a single structure when the desired result is probabilities for the range of plausible structures. The problem of a limited number of structures has been overcome by finding crystals of very similar molecules and complexes, giving a wide range of observed structures. Another precise method is to calculate the energy of the different shapes to learn the shape with lowest energy. In principle, that structure is the most likely one, and structures with progressively higher energy are progressively less probable. Although this method can give a very precise answer, it may not be accurate. There are many ways to calculate the energy and they give different answers. Our poster presents a method that shows all of the crystallographically observed structures to have fairly low energy. This suggests that other shapes are unlikely to be observed in future experiments. The lowest energy occurs when hydrogen bonds connect the glucose and fructose rings, but many of the observed structures do not have such hydrogen bonds. This suggests that hydrogen bonds do not determine the structure but that they can form if the molecule otherwise has the correct shape. Other workers have proposed that water plays a special role in determining the shape of the molecule in solution, but that is not indicated by our modeling method.