Crystallization problems associated with protein

In addition to their varied biological roles, non-heme iron proteins contain a magnificent assortment of iron sites having a multitude of chemical and structural properties. Indeed, the catalog of iron centers is a bit like the taxonomy of insects—a seemingly limitless variation of a few structural themes, yet each new form sufficiently different to define a new species. It is beyond the scope of any review of non-heme iron proteins to be inclusive, and there are excellent recent reviews which detail selected topics. Rather, it is our intention to provide in one chapter an overview of the major classes with an emphasis on proteins for which a crystal structure is available. This review begins with a survey of the types of protein iron structures and a discussion of some methods and problems associated with establishing the iron center type. This should provide an introduction to readers less familiar with the area. Sections II to IV include the current status and recent developments for a limited number of proteins from the major iron classes. These have been chosen in the subjective vein of a limited review the omission of a topic does not indicate its relative importance or interest, only the limitation of space. The purpose of this section is to emphasize the diversity of iron center structures and functions. 

In particular the structure determination of cucumber basic protein has been made using MAD data recorded on this instrument around the CuK edge (Guss et al 1988, section 9.7.5). This protein crystal structure had previously defied solution for many years because of problems associated with preparing heavy atom derivatives. 

Special proteins create special problems. It is difficult to incorporate membrane proteins into the crystals. In addition, it is unclear whether the membrane proteins remain associated with lipids and detergents and protein/lipid/detergent complexes jump into the gas (if anything jumps at all). It seems like the membrane protein bacteriorhodopsin was successfully analyzed with MALDI-TOF. With glycoproteins, the sugar residues sometimes shift position or are cut off by the acidic matrix or the photon current. Finally, a part of the matrix molecules disintegrates under the laser, reacts with the proteins, and thereby increases their MW. This becomes noticeable through so-called adduct peaks in the spectrum. 

A number of more general issues, associated with the selection and preparation of a receptor structure, should be noted. In particular for crystal structures, water molecules and ions present at the protein-ligand interface can pose serious problems for identifying or optimizing lead compounds. In many cases, it is difficult to predict whether a water molecule or ion can be replaced. Ions are especially difficult to treat computationally since their formal charge and the associated desolvation energy are often difficult to model. Their inclusion - or exclusion - is arguable. 

---