Access: Stay Flexible

Variation of the ¹³C chemical shift for the anomeric
carbon (shown in red) with dihedral angles <

> and
<

>of the model glycopeptide GlcNac-

-NAcSerNH₂.

Molecular model for a model glycopeptide GlcNac-

-
NAcSerNH₂.

The nucleus of the Moyna team's method is the nuclei of the polysaccharide's carbon atoms plus a phenomenon known as chemical shift. When an atom is exposed to a magnetic field, as in NMR spectroscopy, the direction in which the atom's electrons orbit its nucleus is affected by that magnetic field. This circulation in turn causes a second, tiny magnetic field at the nucleus that opposes the first field. The difference between these fields allows scientists to distinguish nuclei from one another.

Nuclei are never alone in molecules, however. They are surrounded by electrons that create their own magnetic dipole and are constantly in motion. The electron clouds around the nuclei shield the nuclei and alter the tiny magnetic field around them. The difference between the external magnetic field and the effective field created in a given nuclei is known as chemical shift. The amount of shielding, and thus the chemical shift, is related to the number and arrangement of the electrons surrounding the nuclei. In other words, the chemical shift is determined by the company that the nuclei keep.

"Different nuclei have different shifts because of the chemical environment around them," Moyna says. "We know that chemical shift depends, in part, on the conformation of the molecules, which is an aspect of that chemical environment. The exact nature of the relationship is hard to quantify. But even without that knowledge, we've come up with a nice way of predicting conformation based on the chemical shift of certain carbon atoms within the molecules."

To develop this method, Moyna's team started with molecular models, or computational representations, of eight different disaccharides. Disaccharides are two sugars bonded to one another, polysaccharides pared down to the absolute basics. The team fixed the angle of the two bonds that connect the sugars within the model dissacharide and calculated the optimal structure for the rest of the disaccharide. The angle of one of the bonds was then shifted by 20 slim degrees and the structure was reoptimized. Once a complete 360-degree turn was achieved, the process was repeated on the second bond. A total of 324 distinct conformations was created for each of the eight disaccharides.

Once these models were complete, the team used the University of Kentucky's HP-Convex Exemplar array to calculate the carbon chemical shift that each of the conformations would create, taking advantage of what is known as the Gauge Independent Atomic Orbital method. Using this data, the team created a series of empirical functions that established the relationship between the chemical shift of carbons within polysaccharides and their conformations as described by the molecules' bond torsions. In effect, they turned their year's worth of supercomputing calculations and hard work into a mathematical equation.