The Long-wavelength Macromolecular Crystallography I23 at Diamond Light Source

Abstract

Long-wavelength macromolecular crystallographyMXhas been proposed for a long time as a tool for phasing novel macromolecular crystal structures without additional heavy atom labelling. Making use of anomalous diffraction from atoms natively present in the crystal, such as sulphur and phosphorus, has become increasingly popular over the past years. Nevertheless, the full potential of this technique, has not been fully exploited due to lack of dedicated experimental setups able to easily access wavelengths longer than 2 Å. Since the wavelengths for the absorption edges of sulphur and phosphorus are significantly longer, standard beamline setups are not suitable to provide high-quality, high-resolution data, as the experiments are limited by the increasing absorption effects and the diffraction angles for longer wavelengths. Currently only two synchrotron beamlines offer access to optimised sample environments using wavelengths longer than 2.7 ÅBL1A at Photon Factory, Japan and I23 at Diamond Light Source, UK. Here, we describe the challenges and solutions implemented at the in-vacuum long-wavelength MX beamline I23 and present first results. 1．Introduction Macromolecular crystallography is the most successful technique in structural biology, contributing to date more than 130,000 structures to the total 145,000 structures in the protein data bank. The success of this technique is owed to the significant investment in automation of all aspects of the structure determination pipelines, from cloning, protein expression and purification, crystallisation, automated sample changing and data collections, all the way through to automated processing and phasing. Recently, Grimes et al. 1 reviewed these developments at Diamond Light Source with an outlook towards the future of MX. The majority of MX experiments are nowadays performed at synchrotron beamlines, typically at wavelengths around 1 Å. The wavelength is an important parameter to select in an MX experiment and synchrotron beamlines typically cover a range from 0.7 to 2.0 Å. For crystal screening and high resolution data collections, λ 1 Å is a good comprise between diffraction and detector efficiency of Si based pixel-array detectors. Tuning the wavelength also allows exploiting anomalous dispersion, a resonance effect observed close to the characteristic absorption edges from elements present in the crystal structure. Three main applications exist to exploit the anomalous contrast in MX when tuning the wavelength close and across absorption edges: element identification, location of sulphur positions to assist model building and experimental phasing. To identify particular anomalously scattering elements in the electron density, typically two data sets are needed, one above the absorption edgehigher energy or shorter wavelength sideand a second one belowlower energy or longer-wavelength side. Atoms from the element under investigation will show peaks in anomalous difference Fourier maps calculated from the data above the absorption edge, which are not present below the edgeor significantly