Conformational rearrangements in E1 enzymes. Cartoon representations of distinct states in the catalytic cycle of canonical E1 enzymes. (a) The adenylation state based on the crystal structure of NAE1-UBA3 in complex with NEDD8 and ATP/Mg2+ [PDB: 1R4N] . The carboxy-terminal tail of the Ubl is in the adenylation site of the active Rossmann-type subunit of the E1, ready to nucleophilically attack the α-phosphate of the ATP to form the Ubl-AMP intermediate. The catalytic cysteine residue in the E1 cysteine domain is part of an α-helix and is removed from the adenylation site, giving rise to an open conformation of the cysteine domain. (b) The thioesterification state as seen in a crystal structure of SAE1-UBA2 and SUMO covalently coupled to an AMP analogue that mimics the tetrahedral intermediate generated during thioesterification [PDB: 3KYD] . Mediated by large conformational changes in the crossover and re-entry loops, the cysteine domain is rotated with respect to the Rossmann-type subunits. The helix containing the active site cysteine seen in (a) has melted. In this closed conformation of the cysteine domain, the catalytic cysteine nucleophile is in position to attack the adenylated carboxyl terminus of the Ubl. The positive dipole of helix H2 in the active Rossmann-type subunit (colored purple) is thought to favor this reaction . (c) The trans-thioesterification state as represented by a crystal structure of NAE1-UBA3 thioester-linked to NEDD8 and in complex with an additional NEDD8 molecule, an E2 enzyme (Ubc12) and ATP/Mg2+. The cysteine domain of the E1 adopts an open orientation similar to the adenylation state (a), but now holds the carboxyl terminus of the thioester-linked Ubl close to the E2 active site (a Cys-to-Ala mutant of the E2 was used in this study (see text)). The ubiquitin-fold domain has swung away from its position in the previous states (a,b) to accommodate the E2 and contributes to E2 binding. In (a,c) domains found in NAE1-UBA3 but not in SAE1-UBA2 were omitted for clarity. To see a rendition of a dynamic transition between the structures shown in the lower panels of (a-c), see Additional file 1. As noted in the movie legend, the details of the trajectory linking individual structures is not realistic and is simply meant to illustrate the nature of the conformational changes rather than identify the nature of the transition pathway.