research papers 776 doi:10.1107/S0907444906017318 Acta Cryst. (2006). D62, 776–783 Acta Crystallographica Section D Biological Crystallography ISSN 0907-4449 Time-dependent atomic coordinates for the dissociation of carbon monoxide from myoglobin Roman Aranda IV, a ‡ Elena J. Levin, b ‡ Friedrich Schotte, c Philip A. Anfinrud c and George N. Phillips Jr b * a Department of Biomolecular Chemistry, University of Wisconsin, Madison, USA, b Department of Biochemistry, University of Wisconsin, Madison, USA, and c Laboratory of Chemical Physics, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD, USA ‡ These authors contributed equally to this work. Correspondence e-mail: phillips@biochem.wisc.edu # 2006 International Union of Crystallography Printed in Denmark – all rights reserved Picosecond time-resolved crystallography was used to follow the dissociation of carbon monoxide from the heme pocket of a mutant sperm whale myoglobin and the resultant conforma- tional changes. Electron-density maps have previously been created at various time points and used to describe amino-acid side-chain and carbon monoxide movements. In this work, difference refinement was employed to generate atomic coordinates at each time point in order to create a more explicit quantitative representation of the photo-dissociation process. After photolysis the carbon monoxide moves to a docking site, causing rearrangements in the heme-pocket residues, the coordinate changes of which can be plotted as a function of time. These include rotations of the heme-pocket phenylalanine concomitant with movement of the distal histidine toward the solvent, potentially allowing carbon monoxide movement in and out of the protein and proximal displacement of the heme iron. The degree of relaxation toward the intermediate and deoxy states was probed by analysis of the coordinate movements in the time-resolved models, revealing a non-linear progression toward the unbound state with coordinate movements that begin in the heme-pocket area and then propagate throughout the rest of the protein. Received 27 February 2006 Accepted 10 May 2006 PDB References: myoglobin, crystal 1, ‘laser off’, 2g0r, r2g0rsf; crystal 2, ‘laser off’, 2g0s, r2g0ssf; 100 ps coordi- nate model, 2g0v, r2g0vsf; 316 ps coordinate model, 2g0x, r2g0xsf; 1 ns coordinate model, 2g0z, r2g0zsf; 3.16 ns coordinate model, 2g10, r2g10sf; 31.6 ns coordinate model, 2g11, r2g11sf; 316 ns coordinate model, 2g12, r2g12sf; 3.16 ms coordinate model, 2g14, r2g14sf. 1. Introduction Since the first determination of the three-dimensional struc- tures of hemoglobin and myoglobin in the 1950s (Kendrew et al., 1958; Perutz, 1960), X-ray crystallography has become the most widely applied technique for obtaining structural infor- mation from biological macromolecules. However, knowledge of a static structure alone is not adequate to obtain a complete understanding of how a protein performs its biological func- tion. A complete mechanistic explanation must also include a description of its dynamic behavior. The desire to understand the dynamic properties of proteins has led to the recent development of time-resolved X-ray crystallography, which allows the researcher to supplement the static time-averaged image obtained from conventional crystallography with a series of images representing proteins in motion (Moffat, 1998a,b, 2001, 2003; Schotte et al. , 2003, 2004; Helliwell et al., 1998, 2003). Myoglobin (Mb) is a heme protein that reversibly binds diatomic oxygen and other small gaseous ligands such as carbon monoxide (CO) and nitric oxide and serves as a mobile oxygen buffer in muscle (Radding & Phillips, 2004; Springer et al., 1994). Owing to its simple well characterized structure, the presence of a photolabile bond between the ligand and the heme iron (Gibson & Ainsworth, 1957) and the availability of a large number of mutants with altered functional and kinetic