Technical Highlight - April 2015
Short description: Two recent methods increase the range of SAD phasing to more challenging protein structures.
Today many crystallographers utilize molecular replacement to obtain phasing information for unknown protein structures, but this requires that the structure of a close homolog be available; the approach can also be subject to bias. Since its introduction more than three decades ago, single-wavelength anomalous diffraction (SAD) has become a widely used method for solving protein structures. This experimental phasing technique typically relies on the incorporation of an unnatural amino acid, selenomethione, into the protein of interest; the heavy selenium atom generates anomalous scatter and thus provides the phasing information necessary for structure determination.
Selenomethione can be toxic, and can also lower protein expression yields, so selenomethione-SAD is not a universally applicable approach. The native element of sulfur can also generate anomalous scatter, albeit much weaker scatter than selenium; thus, sulfur-SAD has not been widely applied. Two recently reported methods increase the sensitivity of sulfur-SAD and will likely enable experimental phasing of more challenging protein structures without resorting to heavy-atom labeling.
One way to increase the sensitivity of sulfur-SAD is to merge data from multiple crystals, but this has obvious limitations. Wang and colleagues developed a data-collection strategy for sulfur-SAD phasing that requires a single crystal only. With the help of a specialized goniometer, diffraction data are collected from a single crystal at multiple orientations as a tolerable X-ray dose is applied. The authors applied this method to eleven proteins, including a membrane protein and two protein complexes. Even proteins with weak anomalous sulfur signals could be phased and their structures solved. Notably, the approach should be implementable on any synchrotron beamline.
Terwilliger, Read and colleagues report a different approach for structure determination from weak anomalous scattering data (including sulfur-SAD data) by focusing not on data collection but on data analysis. They developed a likelihood algorithm for determining the substructure of anomalously scattering atoms, which is used to estimate phases, and also described an enhanced phase improvement approach. They have implemented the approach in the popular Phenix software package for crystal structure determination. In total, they applied their method to 159 SAD datasets available from the Protein Data Bank and showed that 79% of structures could be determined. Without the algorithm improvements, just 50% of the structures could be solved.
With these two new SAD approaches, researchers should be able to better utilize experimental phase information for structure determination without heavy-atom labeling and with little additional effort.
T. Weinert et al. Fast native-SAD phasing for routine macromolecular structure determination.
Nat. Methods. 12, 131-133 (2015). doi:10.1038/nmeth.3211
G. Bunkóczi et al. Macromolecular X-ray structure determination using weak, single-wavelength anomalous data.
Nat. Methods. 12, 127-130 (2015). doi:10.1038/nmeth.3212