Technical Highlight - April 2012
Short description: Serial femtosecond crystallography coupled to an X-ray free electron laser may be the next big step for protein crystallography.
Macromolecular X-ray crystallography makes use of electron scattering by X-ray beams to obtain structural information about biological molecules. However, electrons do not scatter hard X-rays very efficiently, so larger crystals (1-50 μm) are needed to scatter enough X-rays to generate high-quality diffraction images. The development of a fourth-generation light source based upon the free-electron laser (FEL) provides a brighter hard X-ray source that allows smaller samples to be examined.
Two research groups, led by Duszenko and Neutze, now describe the application of FEL-based serial femtosecond crystallography to examine in vivo-grown crystals and crystals grown in the lipidic-sponge phase (LSP), respectively. Both groups make use of the Linac Coherent Light Source (LCLS), an X-ray FEL (XFEL) based at the SLAC Linear Accelerator Laboratory in Stanford. Using a liquid jet, protein nanocrystals were injected across the path of the XFEL, which was firing in femtosecond-length pulses. The ultrafast pulse allowed data collection before structural damage to the protein occurred, known as the “diffraction before destruction” principle.
Duszenko's group examined in vivo crystals formed by the overexpression of Trypanosomal cathepsin in Sf9 insect cells using a baculovirus plasmid transfection system. The crystal needles were resuspended in water and crushed before being injected across the XFEL beam. Over 83,000 images were collected, of which 988 were determined to contain protein diffraction patterns with a resolution of up to 7.5 Å. While the collection time was not enough to collect a full dataset, unit cell parameters could be assigned.
Neutze and colleagues used XFEL to examine the photosynthetic reaction center from Blastochloris viridis. The authors had originally grown the membrane protein crystals in the lipidic cubic phase, a semisolid state not suitable for use in the liquid jet injector. Their challenge was to adapt crystallographic conditions to the LSP. Once this was achieved, the microcrystal-LSP suspension could be passed across the XFEL beam as a liquid micro-jet. They were able to collect sufficient data to derive a molecular replacement-guided solution for the reaction center at 8.2- Å resolution.
Both studies reveal the promise of XFEL serial femtosecond crystallography, particularly for membrane protein structure solution. Improvements in the technical aspects of the XFEL at LCLS and other experimental stations and greater user accessibility will provide more opportunities to gain interesting biological insights.