Structural snapshots of conformational changes in a seven-helix membrane protein: lessons from bacteriorhodopsin

Abstract

Recent advances in crystallizing integral membrane proteins have led to atomic models for the structures of several seven-helix membrane proteins, including those in the G-protein-coupled receptor family. Further steps toward exploring structure-function relationships will undoubtedly involve determination of the structural changes that occur during the various stages of receptor activation and deactivation. We expect that these efforts will bear many parallels to the studies of conformational changes in bacteriorhodopsin, which still remains the best-studied seven-helix membrane protein. Here, we provide a brief review of some of the lessons learned, the challenges faced, and the controversies over the last decade with determining conformational changes in bacteriorhodopsin. Our hope is that this analysis will be instructive for similar structural studies, especially of other seven-helix membrane proteins, in the coming decade.

Figure 1

Structural similarity among proteins in the seven-helix family. (a) Structural diversity of bacterial rhodopsins. Bacteriorhodopsin (PDB code; 1M0L) [9], halorhodopsin (1E12) [10], sensory rhodopsin II (1JGJ) [11], and xanthorhodopsin (3DDL) [12] are aligned using residues 11-29, 43-59, 82-98, 110-124, 137-153, 170-189, and 205-223 of bacteriorhodopsin and equivalent residues of others and shown colored in blue, light green, yellow, and coral respectively. (b) Structural diversity of the G-protein coupled receptor family. Bovine rhodopsin (PDB code; 1U19) [13,14], squid rhodopsin (2Z73) [15], β₂-adrenergic receptor (2RH1) [16], β₁-adrenergic receptor (2VT4) [17], and A_2A adenosine receptor (3EML) [18] are aligned using residues 37-62, 72-96, 109-137, 151-171, 201-223, 249-275, and 289-319 of bovine rhodopsin and equivalent residues of others and shown colored in green, light blue, yellow, coral, and cyan respectively. (c) Direct structural comparisons of transmembrane regions of bacteriorhodopsin and bovine rhodopsin, as viewed from the cytoplasmic side. Residues 13-29, 43-59, 82-98, 110-124, 142-152, 170-189, and 205-222 of bacteriorhodopsin and residues of 44-60, 73-89, 117-133, 156-170, 213-223, 249-268, and 292-307 of bovine rhodopsin are aligned and shown colored in magenta and green respectively in stereo.

Figure 2

Structural differences between different models reported for the M intermediate of the bacteriorhodopsin photocycle. (a) Plot of the root mean square deviation of Cα atoms from the unilluminated state of bacteriorhodopsin in models of the M intermediate obtained using electron crystallography (blue; [19]), and with a model derived from X-ray crystallography that shows the largest changes (red; [27]). The inset shows an expanded view of the deviations at the cytoplasmic end of helix F for two M intermediate states shown in the main panel (blue and red, respectively) and from other reports of M intermediate structures (yellow; [39], purple; [26], green; [30], cyan; [36]). (b) Experimentally derived difference map (calculated using both amplitude and phase data from diffraction and image of triple mutant and wild type [38]) showing the nature and extent of the large conformational changes in the bacteriorhodopsin photocycle. (c and d) Difference maps computed from deposited coordinates for the M intermediate by electron crystallography [19] showing a close match (panel c) or by X-ray crystallography [27], showing a poor match (panel d) with the experimentally derived map in panel b. See [38] for a detailed discussion of the use of experimentally derived difference maps as a strategy for validating structures derived by crystallography for the conformational change.