Technical Highlight - August 2012
Short description: Anomalous diffraction from multiple crystals allows phasing of data from native proteins.
Solving the crystal structure of a new protein without prior structural information on closely related proteins requires a solution to the phase problem. Since the advent of multi- and single-wavelength anomalous diffraction (MAD and SAD) experiments many new structures have been solved once the necessary selenomethionyl or heavy atom sites were incorporated. While MAD and SAD experiments are now very widely used to acquire phases there remain problematic cases where appropriate heavy atoms cannot be introduced readily and are not intrinsically present.
Hendrickson and colleagues (PSI NYCOMPS) have investigated the possibility of routinely obtaining phases using anomalous diffraction from lighter atoms (notably sulfur and phosphorous) intrinsically present in native macromolecules. The underlying technique is not new, but the success of experiments on selenomethionyl proteins combined with the weakness of anomalous scattering signals from intrinsic sites have limited application mainly to test cases. The key solution is to achieve data multiplicity while avoiding radiation damage and other sources of error.
The authors extended their earlier work to develop a robust native SAD method for enhancing weak signals by combining data from multiple crystals. They optimized the x-ray energy and beam path parameters by minimizing background absorption and matching the beam size to crystal size. To address concerns about crystal variation, for example as introduced by freezing, the authors used cluster analyses to check the statistical equivalence and integrity of the crystals used. The method was applied to four proteins, three of which (HK9S, CysZ and netrin G2) yielded new structures while the fourth (TorT/TorSS) is a complex with a known structure, but was challenging due to its size and complexity. The resulting structures used 5 to 13 crystals each and were solved at resolutions of 2.3 to 2.8 Å.
Although the method was successful as implemented, the authors highlight ways to improve both data collection and analysis. They suggest that an advanced undulator beamline optimized for low-energy x-ray experiments and a high-angle detector could be combined with improved data handling. While the method requires multiple crystals, it is possible that benefits will accrue from further experiments on native phasing sites in both nucleic acids and proteins.
Q. Liu et al. Structures from anomalous diffraction of native biological macromolecules.
Science 336, 1033-1037 (2012). doi:10.1126/science.1218753