Oligosaccharide Conformations by Diffraction Methods

Serge Perez, Catherine Gautier, and Anne Imherty 

Crystal structure have been solved for most of the pentoses and hexoses, in one or more of the isomeric forms or as a 1-O-methyl derivatives. A significant amount of crystal structures of carbohydrates have been determined from neutron diffraction experiments. From these highly accurate structural determination, standard molecular dimensions of the constituting units have been established (Table 2). These data are an update of the ones published previously by Jeffrey Taylor [14] and, in a similar way, they can serve as a basis for parameterizations of molecular mechanics force fields. 

There is an obvious reluctance of carbohydrates to crystallize in a form suitable for X-ray or neutron diffraction studies. This is particularly true for aldose and ketose containing carbohydrates as a configurational mixture of four isomers, a-and P-pyranoses and a- and P-furanose, is likely to occur. Configurational heterogeneity in solution tends to inhibit crystallization. For this reason 1-O-methyl derivatives, which cannot epimerize, are likely to crystallize more readily. In some instances, the a and P epimers can co-crystallize, and more than 10 examples of such co-crystallization can be found. The a and p ratio may be dependent on the temperature and solvent of crystallization, and may not be reproducible between independent investigators. One extreme example is provide by the crystal struc- 

Mana( 1 -3)ManP( 1 -4)GlcNAc Fragment N-glycan MPYAGL [35] 

The reluctance of carbohydrate to crystallize in a form suitable for X-ray diffraction studies, is more pronounced for compounds having molecular weights ranging from 1000 to 5000. This is true but for the exception of cyclic compounds such as cyclodextrins and cycloamyloses. One of the reasons may be the lack of sufficient amount of material available for crystal growth. The other reason is that the techniques of growing organic crystals of medium size bio-molecules has not parallel the revolution of growing protein and viruses crystals. 

---