Figure Legends:

Figure 1. The structure of avian cytochrome bc1 complex. An overall view of the 10-subunit complex from 2 orientations. The dimer 2-fold axis is vertical in the plane of the page. The shaded band represents the approximate location of the membrane hydrophobic phase, with the mitochondrial matrix below and inter-membrane space above. Ten subunits are shown. Subunit 11, if it exists in the chicken complex, is not present in the crystals and not consistently present in the preparation of the enzyme. This figure was made from PDB entry 1BCC, first revision after release.

Figure 2. Different positions of the Rieske Iron-sulfur cluster relative to heme irons. Iron-Iron distances are taken from references 1, 2, and 3. The distance between the Rieske cluster and the low potential heme of cytochrome b is plotted on the vertical axis, and that to the heme of cytochrome c1 on the horizontal axis. Blue dots represent positions approximating the cytochrome b position, red dots approximate the cytochrome c1 position, and the green dot represents the intermediate position found in one monomer of the P65 crystal. The data labels identify the following structures: I4122- entry 1QCR (bovine tetragonal [1]). Chicken + stigmatellin- entry 3BCC. P65- entry 1BGY, chain E [3]. P212121- entry 1BCC (chicken); P6522R- rabbit hexagonal[2]; P21- beef monoclinic[2]; P6522B- bovine hexagonal[2]; P6522I- entry 1BE3 (flash-frozen, dehydrated beef hexagonal[3]).

Figure 3. Two conformations of the Rieske Iron-sulfur Protein (Stereo pair). Cytochrome b (salmon) and the Rieske protein in the cytochrome b position (blue) are from PDB entry 3BCC. Cytochrome c1 (orange) and the Rieske protein in cyt c1 position (yellow) are from our unsubmitted beef P6522 structure (2). Stigmatellin (green) marks the position of the Qo site, and the hemes of cytochrome b and cytchrome c1 are red. The ironsulfur cluster in each position is shown as large red and yellow balls, and H161 is solid blue (b position) or yellow (c1 position).

Figure 4. Cytochrome b backbone movements induced by inhibitor binding. The structures of entry 1BCC (native), 3BCC (Stig + Anti), and two unsubmitted datasets containing only stigmatellin (stig) or antimycin (anti) were superimposed so as to minimize the rms deviation of C-a atoms of residues 32-129, 180-248, and 294-380. Then the four pairwise comparisons indicated were made by subtracting the coordinates of corresponding C-a atoms in the superimposed structures. To reduce noise and emphasize changes involving more than a single residue, the resulting values were averaged in a window of 3 residues. The resulting value is plotted vs residue number to indicate where the changes occur. A cartoon representing the secondary structure and surface exposure (prepared by the program Procheck) is aligned above.

Figure 5. Movements in the surface helices of cytochrome b in the presence of stigmatellin. The brown net is the electron density calculated from a native crystal (no inhibitors) and the red model is the backbone of cytochrome b built into this density. The blue net is density from a crystal containing stigmatellin. The purple backbone and connected side chains are the Rieske protein located in this density, and the other ball-and-stick models are depict stigmatellin and parts of cytochrome b. Both electron density maps are made using experimental phases after improvement by density modification, and are contoured at 1 s . The models are unrefined and intended only to identify features of the density. It can be seen that the cd1 helix moves down, perhaps pushed by the Rieske protein in docking, and the ef helix and residue Y279 move to the right, perhaps as expansion of the Qo site to accommodate stigmatellin. Rieske H161 which H-bonds stigmatellin is deep in the picture; the electron density can be seen dimmed by the depth-cueing. The descending arm of the ef linker with P271 and E272 is not seen because it is in front of the slab of space used for the picture. Artifacts due to differences in cell parameters and different positioning of the protein in the two cells was avoided by transforming the models and skewing the maps to a reference cell in sucha a way as to optimally superimpose the rest of the protein.

Figure 6. Electron density at the Qi site in native chicken crystals. the refined model of the complex, superimposed on a 2Fo-Fc map, where Fc and Fc are calculated from the refined model with ubiquinone omitted to avoid phase bias, contoured at 1.1 s. The superimposed model is the refined structure of the complex, including ubiquinone with its headgroup in the center of the figure. Panels A and B show two different views related by approximately 90 rotation about a vertical axis.

Figure 7. Stereo view of the Qo site and the ef linker peptide. Above, the view is zoomed out to show the region in the context of cytochrome b (pink and red), cytochrome c1 (blue), and the Rieske protein (green). Below, the view is zoomed in to make details visible. The transmembrane helices E and F of cytochrome b, as well as the connecting ef linker (residues 246 to 288) are red while the rest of cytochrome b is pink. The binding position of Qo site inhibitors is shown by the yellow and red model of stigmatellin behind and to the right of the descending arm of the ef linker. Behind this inhibitor the cd1 and C helices come together at an acute angle. Purple spheres are the iron atoms in the low potential cytochrome b heme and in the Rieske ironsulfur cluster. Some residues dicussed in the text are labeled.

Figure 8. Comparison of the structures of cytochrome c1 and cytochrome c (stereo pair). Above is a ribbon diagram of mitochondrial cytochrome c in a standard orientation with the open C corner of the heme facing the viewer, and the heme propionates directed downward. Below is cytochrome c1, rotated to put the common features between the two cytochromes in the same orientation. Corresponding segments of each cytochrome are drawn with the same color. Helices labeled a1, a3, and a5 correspond to similarly labeled helices in cytochrome c. The N- and C- termini are labeled in cytochrome c1.

Figure 9. Comparison of cytochrome c1 and hinge protein from the orthorhombic crystals of chicken bc1 (1bcc) with that from the p6522 bovine enzyme (1be3). The bovine structure is on the left, the chicken on the right. A significant difference (described in the text) is indicated by the gray arrow. Because this area has poor order in both structures, the difference could be due to mis-tracing the polypeptide chain as well as to crystal packing or species differences.

Figure 10. Structure-Based Sequence Alignment of cytochromes c1, c, and c2. The sequences are vertebrate cytochrome c1 (1BCC, middle sequence), mitochondrial cytochrome c (eqine 1HRC, top sequence), and bacterial cytochrome c2 (Rb. capsulatus 1C2R, bottom sequence). The symbols between the lines indicate closeness of superposition of Ca atoms of the aligned residues: ( | ) 1 Å or less, (~) between 1 and 2 Å. The sequence numbering of the beef cytochrome c is indicated above, and that of bovine cytochrome c1 below, the alignment.

Figure 11. Sequence alignment of cytochromes c1 with other class I cytochromes. The alignment of cytochromes c1 (1BCC), c (1HRC, 1YCC), c2 (1C2R), c6 (1CYI), and c8 (351C) are based on superimposed structures from the indicated PDB entries. The other cytochromes c1 for which no structure is available were aligned by conventional sequence alignment. Boxed regions of the sequence indicate the 3 conserved helices 1, 3, and 5; helices 2 and 4 in cytochromes c and c2; helix 2' in cytochrome c1; and the heme-bracing PXL tripeptide. Highly conserved residues are indicated in bold type. The sequence numbering of horse cytochrome c is indicated at the top, and that for bovine cytochrome c1 at the bottom.