On the origin of diffuse X-ray scattering from protein crystals

Abstract

X-ray crystallography is the primary source of information on protein structure at high resolution. As we seek insight into dynamics that are critical for function, we must go beyond average structure models toward an ensemble description. Traditional X-ray crystallography, however, relies on analysis of intense Bragg reflections and is sensitive only to the unit cell’s average electron density. Although ensembles are often used in crystal structure refinement, the Bragg data contain no information about displacement correlations which would be needed to provide a unique result. One solution, first proposed several decades ago, is to analyze the continuous diffuse scattering between the Bragg peaks. The diffuse scattering pattern is highly sensitive to displacement correlations and it is often observed in protein crystal diffraction. Newly-available X-ray sources and detectors are ideal for these measurements and have inspired new interest in the technique. However, the microscopic origin of diffuse scattering from proteins has not been firmly established, and this has led to considerable controversy over which models must be applied, and how the data should be treated. To resolve this issue, we measured highly detailed and precise maps of total diffraction from lysozyme crystals in three space groups at room temperature. To aid in interpretation, all-atom MD simulations were performed by David A.Case (Rutgers) on one or more unit cells. These measurements and simulations highlight the variety of diffuse features that can be observedeven in very well-diffracting systems and provide insight into the types of disorder that are likely common to all protein crystals.