Volume 124, Issue 1, Pages 61-73 (January 2006) The Crystal Structure of Yeast Protein Disulfide Isomerase Suggests Cooperativity between Its Active Sites  Geng Tian, Song Xiang, Robert Noiva, William J. Lennarz, Hermann Schindelin  Cell  Volume 124, Issue 1, Pages 61-73 (January 2006) DOI: 10.1016/j.cell.2005.10.044 Copyright © 2006 Elsevier Inc. Terms and Conditions

Figure 1 Primary Structure of Mammalian and Yeast PDI (A) Domain organization of bovine PDI as deduced from biochemical studies and that of yeast PDI based on the crystal structure (c represents the C-terminal tail). The C-G-H-C motifs indicate the location of the active sites, the Cs the nonactive site cysteines, and x the loop connecting the b′ and a′ domains. (B) Multiple sequence alignment of yeast and mammalian PDIs. Helices and strands are represented by cylinders and arrows, respectively. Active-site residues in the a and a′ domain are boxed. Black arrows highlight the two buried polar residues in the vicinity of the active site, white arrows the nonactive site cysteines, and blue arrows the cis-prolines near each active site. Boxed residues in the b′ domain form a hydrophobic pocket presumably involved in substrate binding. The bar above the sequence represents the domain architecture. Cell 2006 124, 61-73DOI: (10.1016/j.cell.2005.10.044) Copyright © 2006 Elsevier Inc. Terms and Conditions

Figure 2 Overall Structure of PDI (A) Ribbon diagram of PDI with the a, b, b′, and a′ domains in magenta, cyan, yellow, and red, respectively, and the C-terminal extension in green. The two orientations roughly differ by a 90° rotation around the horizontal axis. The side chains of the active site cysteines in the a and a′ domains are shown in space-filling representation with the sulfur atoms in yellow. (B) Structural comparison of the individual domains of PDI. The domains are shown in the same relative orientation, with the “long helix” side below the β sheet. The active-site cysteine residues in the a and a′ domains are shown in space-filling representation. (C) Secondary structure diagram of the canonical thioredoxin fold with α helices in green and β strands in red. The location of the active site is indicated by a red oval. Cell 2006 124, 61-73DOI: (10.1016/j.cell.2005.10.044) Copyright © 2006 Elsevier Inc. Terms and Conditions

Figure 3 Functional Characterization of PDI In vitro PDI activity assay using scrambled ribonuclease (sRNase) and reduced ribonuclease (rRNase) as substrates. The domain boundaries of the deletion constructs are defined in Figure 1A and ∗ indicate the relative location of the altered residues. The standard deviations are derived from three independent experiments. Cell 2006 124, 61-73DOI: (10.1016/j.cell.2005.10.044) Copyright © 2006 Elsevier Inc. Terms and Conditions

Figure 4 Hydrophobic Surface Features of PDI (A) Surface representation of PDI color coded ranging from hydrophobic (green) to hydrophilic (gray) according to the normalized consensus hydrophobicity scale (Eisenberg et al., 1984) of the exposed residues. The orientation is the same as in Figure 2A, lower panel. The lower left and right panels represent the a and a′ domain, respectively, which were rotated by +60° and −60° as indicated to allow visualization of the face surrounding the active site, highlighted in red. (B) Close-up view of the peptide binding pocket in the b′ domain. The orientation is the same as in (A). Residues in the center line the bottom of the hydrophobic pocket. (C) Packing interactions between the b domain of a symmetry-related molecule and PDI. PDI is shown in a surface representation in roughly the same orientation as in Figure 2A, top panel, with the domains in standard color code. The b domain of the symmetry mate (cyan) is located between the a and a′ domains. Cell 2006 124, 61-73DOI: (10.1016/j.cell.2005.10.044) Copyright © 2006 Elsevier Inc. Terms and Conditions

Figure 5 Active Site Features of PDI (A) Stereo view of a SIGMAA weighted electron density map (one rms deviation contour level) near the a domain active site. The S atoms of the active site cysteines are shown in yellow for the oxidized state and orange for the reduced state. (B) Close-up view into the active site of the a domain. β strands are shown in yellow, α helices in green, and loops in gray. Selected residues are displayed in stick representation, the buried water molecule as a red sphere, and hydrogen bonds as dashed lines. Only the oxidized state of the active site is displayed. Not all secondary structure elements are shown to increase visibility, but the surface is displayed for the entire domain. (C) Close-up view into the active site of the a′ domain. (D) Surface representation of the a domain after a 180° rotation around the horizontal axis relative to the view in (B). The N-terminal extension covering the pocket adjacent to the active site has not been included in the surface calculation and is shown as a magenta loop with residues 26–28 in stick representation. Hydrogen bonds are shown as dashed lines. (E) View into the active site of the human a domain based on structure 1 of the NMR ensemble (Kemmink et al., 1996). In this and all other NMR models, the cleft is not covered by the N-terminal extension (shown in magenta and excluded from the surface calculation). (F) View of the yeast a′ domain highlighting the putative Ero1 binding pocket. The orientation is related to that of the a′ domain shown in (C) by a 180° rotation around the horizontal axis. Cell 2006 124, 61-73DOI: (10.1016/j.cell.2005.10.044) Copyright © 2006 Elsevier Inc. Terms and Conditions

Figure 6 Relationship to DsbC (A) Ribbon diagram of the DsbC dimer with the thioredoxin domains in green, dimerization modules in orange and magenta, and connecting helices in blue. (B) Surface representation of DsbC viewed into the active site cleft. The surface is color coded according to the same hydrophobicity scale utilized in Figure 4A, and the active sites are labeled in red. Cell 2006 124, 61-73DOI: (10.1016/j.cell.2005.10.044) Copyright © 2006 Elsevier Inc. Terms and Conditions