The bane of protein crystallographers is the common problem of proteins that simply will not crystallize. This is especially poignant when it comes to some of the more biomedically interesting of their number, such as the numerous membrane proteins, many of which do not succumb to even the most sophisticated crystallization techniques. Now, researchers at Imperial College London and the University of Surrey, both in the UK, have developed a new technique for crystallizing proteins, which could open up a whole range of materials to this powerful analytical technique.

Pioneering three-dimensional crystal structures of proteins, such as hemoglobin and insulin and other biological macromolecules, including DNA, using X-ray diffraction techniques plays a key role in laying the foundations of molecular biology. However, protein crystallographers hit a brick wall repeatedly when it comes to certain types of protein. These intractable compounds are very important to understanding the machinery of living things and as targets for drug therapy in a range of diseases that crystallographers must persist in finding a way to crystallize them.

Usually, to direct a protein to crystallize, a nucleant is used. This "seeds" the formation of a crystal lattice from a solution of the protein. Now, a new theory concerning the design of porous materials for protein crystallization has been put into practice. The theory is based on the rationale that the porous structure of a material traps the protein molecules, and encourages them to crystallize. Imperials' Naomi Chayen and Emmanuel Saridakis and Surrey colleague Richard Sear have now tested the theory on the stubborn lobster shell protein, alpha-crustacyanin, using BioGlass, a substance developed at Imperial as a scaffold, to trap and encourage the growth of protein crystals. BioGlass is a porous material, with a variety of different sized pores able to trap different sized proteins. The researchers found that BioGlass induced the crystallization of the largest number of proteins ever crystallized using a single nucleant.

	A porous structure traps a lobster protein crystal at last

"The first step in obtaining a good crystal is to get it to nucleate in an ordered way," explains Chayen. "The 'holy grail' is to find a 'universal nucleant' which would induce crystallization of any protein. Although there has been considerable research in search of a universal nucleant, this is the first time we have designed one which works on a large number of materials."

Proc Natl Acad Sci, 2006; http://dx.doi.org/10.1073/pnas.0504860102

http://www.imperial.ac.uk/people/n.chayen