Overview and new developments in softer X-ray (2A < lambda < 5A) protein crystallography.

New methodologies with synchrotron radiation and X-ray free electron lasers (XFELs) in structural biology are being developed. Recent trends in harnessing softer X-rays in protein crystallography for phase determination are described. These include reference to a data-collection test at 2.6 A wavelength with a lysozyme crystal on SRS station 7.2 (Helliwell, 1983) and also use of softer X-rays (2 A wavelength) to optimise f " at the xenon L1 absorption edge in the Single Isomorphous Replacement Optimised Anomalous Scattering ('SIROAS') structure determination of apocrustacyanin A1 with four, partially occupied, xenon atoms (Cianci et al., 2001; Chayen et al., 2000). The hand of the protein was determined using the f " enhanced sulphur anomalous signal from six disulphides in the protein dimer of 40 kDa. In a follow-up study the single wavelength xenon L1-edge f " optimised data set alone was used for phase determination and phase improvement by solvent flattening etc. (CCP4 DM) (Olczak et al., 2003). Auto-tracing of the protein was feasible but required additional diffraction data at higher resolution. This latter could be avoided in future by using improved tilted detector settings during use of softer X-rays, i.e. towards back-scattering recording (Helliwell, 2002). The Olczak et al. study has already led to optimisation of the new SRS beamline MPW MAD 10 (see www.nwsgc.ac.uk) firstly involving the thinning of the beryllium windows as much as possible and planning for a MAR Research tilted detector 'desk top beamline' geometry. Thus the use of softer, i.e. 2 to 3 A wavelength range, X-rays will allow optimisation of xenon and iodine L-edge f " and enhancing of sulphur f " signals for higher throughput protein crystallography. Softer X-rays utilisation in protein crystallography includes work done on SRS bending-magnet station 7.2 in the early 1980s by the author as station scientist (Helliwell, 1984). In the future development of XFELs these softer X-ray wavelengths could also be harnessed and relax the demands to some extent on the complexity and cost of an XFEL. Thus, by use of say 4 A XFEL radiation and use of a back-scattering geometry area detector the single molecule molecular transform could be sampled to a spatial resolution of 2 A, sufficient, in principle, for protein model refinement (Miao et al., 1999). Meanwhile, Miao et al. (2003) report the first experimental recording of the diffraction pattern from intact Escherichia coli bacteria using coherent X-rays, with a wavelength of 2 A, at a resolution of 30 nm and a real-space image constructed. The new single-particle X-ray diffraction-imaging era has commenced.