Producing diffraction-quality protein crystals can be a huge problem and is a major bottleneck in X-ray crystallography. Several methods have been developed in an attempt to overcome this obstacle. One promising technique is chemical modification of proteins using reductive methylation.
Dimethyl lysine (dmLys) promotes intermolecular interactions. dmLys A62 in the structure of VPA0580 from Vibrio parahaemolyticus (PDB 2QHQ) makes contact with Glu A35 and the main-chain carbonyl of Leu A91.
This approach results in methylation of the solvent-exposed ε-amino group of lysines, which changes the protein's surface and biophysical properties (such as pI, solubility and hydropathy). Previous studies using this method indicated that methylation generally does not affect the protein's function and structure.
Now, a team from the PSI MCSG has carried out a large-scale test of the general applicability of the method. The team analysed 370 proteins from diverse protein families, with no sequence homologs within the set. Of these proteins, 269 had produced poor-quality crystals unsuitable for diffraction studies, 85 had failed to crystallize, and 16 were a reference set.
Kim et al. used a simple method involving the addition of 1 M formaldehyde and 1 M dimethylamine–borane complex (ABC) as a reducing agent. After reductive methylation, they obtained diffraction-quality crystals for 40 out of the 370 proteins, and so far have solved 26 crystal structures. The crystallization success rate did not correlate with the number of lysines, pI, hydropathy or molecular weight.
The MCSG team solved structures of four proteins that had not been methylated and compared them with the same proteins after methylation. The structures were virtually identical and the resolution of the methylated structures was better than that of their native counterparts. The results suggest that methylation produces a similar effect to the replacement of lysine residues with arginine, another approach to improving protein crystallization.
This reductive methylation technique produced an overall success rate, from purified protein to a structure, of 7.0% for proteins that did not produce useable structures during the initial crystallization trials, and allowed structure determination of some proteins that failed to crystallize in their native state.