Structural basis of substrate recognition by a bacterial deubiquitinase important for dynamics of phagosome ubiquitination

Identification of SdeA as a Deubiquitinating Enzyme.

To identify L. pneumophila proteins with potential DUB activity, we used the suicide inhibitor HA-Ub-VME, capable of forming a covalent complex with proteins harboring active DUB domains with a catalytic cysteine (20). When the probe was used to tag DUBs in lysates of L. pneumophila, a single band with a molecular weight larger than 150 kDa was detected (Fig. 1A). Because the CHD motif is most commonly used in DUB catalysis, we analyzed L. pneumophila effectors near the molecular weight of the identified band for this motif, and found that it is present in members of the SidE family (Fig. S1). This reactivity was then confirmed in vitro by labeling all three putative DUB domains with HA-Ub-VME as evidenced by a ∼8-kDa shift corresponding to the formation of a ubiquitin adduct (Fig. 1B). We then showed that full-length SdeA and the DUB domains of SdeA, SdeB, and SdeC could disassemble the most common chain types (Lys11, Lys48, Lys63; Fig. S2 A–D) (21).

To measure the activity of SdeA_1–193, referred to as SdeA_Dub, against polyubiquitin chains attached to a protein, we prepared a polyubiquitinated GFP substrate (GFP-Ub_n) with mostly Lys63-linked chains and small populations of chains linked via Lys11 and Lys48 (22, 23). SdeA_Dub was able to completely deubiquitinate this substrate within 10 min (Fig. 1C). We further determined the activity of SdeA_Dub against more biologically relevant substrates. To this end, we created a cell line stably expressing HA-Ubiquitin. Ubiquitinated proteins were then immunoprecipitated with an HA-specific antibody. His₆-SdeA_Dub almost completely removed the ubiquitin signals, whereas the Cys118-Ala catalytically inactive mutant could not (Fig. 1D).

To explore whether SdeA_Dub recognizes ubiquitin-like (Ubl) posttranslational modifications, we probed for its reactivity with other Ubl proteins (NEDD8, ISG15, and SUMO) conjugated to vinyl sulfone (VS). In these assays, SdeA_Dub was found to also form a covalent complex with NEDD8-VS and, to a lesser extent, ISG15-VS. There was no detectable reactivity with the SUMO-VS probe (Fig. 1E). The ability to form the complex required Cys118, further supporting the notion that members of the SidE family harbor an active DUB domain. Consistent with reactivity with NEDD8-VS, SdeA_Dub was able to remove NEDD8 signals from neddylated proteins prepared from mammalian cells transfected to express Flag-NEDD8 (Fig. 1F).

SdeA_Dub Is Involved in Polyubiquitin Regulation on the Bacterial Phagosome.

Members of the SidE family are similar proteins with sizes larger than 150 kDa (24, 25). Mutants lacking all members of the family are defective in intracellular replication in the amoebae host Dictyostelium discoideum (25). We then determined the role of DUB activity in the function of the SidE family in infections. At 72 h post infection, a mutant lacking all four members of the SidE family (∆4) displayed approximately 100-fold defects in D. discoideum, with the defect fully complemented by expressing SdeA (Fig. S3A). Interestingly, the Cys118-Ala mutant is still fully functional in the complementation of the defects (Fig. S3A), suggesting that DUB activity is not required for the role of the SidE paralogs in intracellular bacterial replication.

Phagosomes harboring virulent L. pneumophila are known to be enriched with polyubiquitinated species (10); we then determined the role of DUB activity in such enrichment. In mouse bone marrow-derived macrophages, ∼40% of the vacuoles containing WT L. pneumophila stained positively for ubiquitin (Fig. 2A). The rate of ubiquitin-positive vacuoles increased to approximately 90% in infections using the ∆4 strain; such increase can be reduced to nearly WT levels by expressing SdeA but not the SdeA_C118A mutant (Fig. 2B). Similar reduction in ubiquitin-positive vacuoles is also seen with the expression of other members of the SidE family (Fig. S4 A and B). The mutant was translocated into host cells indistinguishably to that of similarly expressed WT protein (Fig. S3 B and C), indicating that its inability to complement was caused by loss of enzymatic activity in the DUB module. Together, these results reveal that the DUB activity of SdeA is important for the dynamics of ubiquitin association with the bacterial phagosome.

SdeA_Dub Cleaves Lys11/48/63 Chains with a Preference for Lys63.

To determine the chain type preference of SdeA_Dub, we used a gel-based diubiquitin cleavage assay. These results revealed that, even though SdeA_Dub cleaves all chain types examined (Fig. 3A and Fig. S2E), there was a clear preference for Lys63-linked chains, yielding a binding constant (from the K_m) of ∼200 µM for each DUB domain and ∼100 µM for the full-length SdeA construct (Fig. 3B and Fig. S2H). We were unable to obtain kinetic parameters for Lys11 and Lys48-linked diubiquitin chains as a result of the low cleavage efficiency. Interestingly, high concentrations of Lys48-linked diubiquitin appeared to inhibit SdeA_Dub activity (Fig. S2 F and G). In line with the biochemical observations, the SdeA DUB module was able to efficiently remove Lys63-linked polyubiquitin chains from the phagosomal surface (Fig. S4 C and D).

Structure of the SdeA DUB Module.

Structures of SdeA_Dub were determined for two constructs: 1–163 SeMet (apo), 1–193 Cys118-Ala (apo), and 1–193-Ub-VME at 2.85 Å, 2.02 Å, and 2.53 Å, respectively, with each solution containing two molecules in the asymmetric unit (even though biophysical analysis suggests this protein is a monomer in solution; see Fig. S7 F and G and Table S1). The overall fold of SdeA_Dub can be described as an arrangement of periodic secondary structures forming two distinct lobes, the α-lobe and the β-lobe (Fig. 4A and Fig. S5A), which pack closely to generate a compact globular fold, connected by loops L2 and L10 (Fig. 4A and Fig. S5A). The β-lobe is composed of a twisted, mixed, six-stranded β-sheet (β1–β6; Fig. S5A), which, when viewed from one edge, appears to form a partially wound β-barrel (Fig. S5B). The concave surface of this partial β-barrel allows a curved α-helix (α3) to make tight contact. An additional, relatively short, helix (α2) seems to cap one end of the partial barrel. The outer surface of the barrel packs against two α-helices (α1 and α4) of the α-lobe (Fig. S5A). The rest of the helices in the α-lobe (α5, α6, and α7) arrange themselves by packing against at least one other helix within this lobe.

The active site is located at the interface between the two lobes, with Cys118 placed in the first turn of α4 facing the catalytic histidine, His64, with the catalytic aspartic acid, Asp80, within short hydrogen-bonding distance (Fig. 4B). This CHD catalytic triad is complemented by a suitably located Asn residue (Asn114) in loop L2 that we propose stabilizes the oxy-anion species (Fig. 4B). The catalytic residues line an open catalytic cleft (except for Glu9, which seems to function as a flap) with a predominantly negative electrostatic character complementing the distinctly electropositive C-terminal tail of ubiquitin and some Ubls. Interestingly, inspection of the structures of the three constructs reported herein reveals that the active-site residues have the propensity to assume different conformations, indicated by different orientations of the Cys and His residues, possibly reflecting plasticity in the catalytic residues, which, in addition to the open nature of the cleft, may confer some advantage in accommodating substrates with varying structure of their polyubiquitin chain.

SdeA_Dub Binds Ubiquitin Through Three Distinct Patches.

Sequence alignments and modeling with other Ulps revealed little about the nature of ubiquitin binding. To visualize the mode of ubiquitin recognition, we prepared a covalent complex of SdeA_Dub with ubiquitin by reacting the DUB module with Ub-VME. The structure of the complex reveals the formation of a covalent bond between Cys118 and the VME group of the suicide inhibitor. Inspection of the Ub-VME bound crystal structure reveals three distinct ubiquitin binding sites: the active-site cleft (site-1), a site centered around Tyr33 (site-2) referred to as the Y-site, and a site that engages the β1–β2 loop of ubiquitin (site-3; Figs. 4B and 5). The active-site cleft engages the tetrapeptide portion of the C-terminal tail of ubiquitin (Leu73-Gly-VME, with Gly-VME corresponding to Gly76) with contacts that involve mainly interbackbone hydrogen bonding. Hydrogen-bonding contacts with tail residues stabilize the tail in an extended conformation, resulting in an arrangement in which the tail residues form the central strand of a three-stranded β-sheet that includes one arm of loop L6 and residues Glu9–Gln12 of the enzyme (Fig. 4B). Engagement of the tail is facilitated by charge and shape complementarity, with the α-carbon of Gly-VME fitting snugly into a narrow groove at the end of the active-site cleft adjacent to Cys118. This matching of shape ensures proper placement of the scissile isopeptide bond adjacent to Cys118 and ensures selectivity for the terminal di-Gly motif in ubiquitin and certain Ubls. Arg72, a key determinant of ubiquitin/Ubl specificity, is unengaged, explaining the reactivity of the DUB toward ubiquitin-VS and NEDD8-VS (Fig. 1E) (26, 27).

Most DUB-ubiquitin cocrystal structures show active-site tail binding as observed here, albeit with more extensive interactions. Tyr33 in the Y-site makes three contacts with ubiquitin, the most prominent of which appears to be Gln40, as well as making van der Waals contacts with Ile36 and Leu73. The hydroxyl group is engaged in a tight hydrogen-bonding interaction with the side-chain carbonyl group of Gln40, whose side-chain amide proton serves as a hydrogen bond donor to Asp61 from the catalytic cleft (Fig. 5C). Another commonly observed interaction between DUBs and ubiquitin involves the β1–β2 loop of ubiquitin engaged at a distinct site that allows the dipeptide turn to fit in snugly. In SdeA_Dub, the β1–β2 loop of ubiquitin interacts primarily through van der Waals contacts, with only one hydrogen bond between the backbone carbonyl of Leu8 of ubiquitin and the hydroxyl group of Ser29 of SdeA (Fig. 5D). It appears that most of these interactions can occur even if the turn segment has substitutions such as those found in ISG15, possibly explaining the observed reactivity with ISG15-VS and NEDD8-VS (Fig. 1E and Fig. S6). Surprisingly, the ubiquitin Ile44 patch is not engaged in binding (Fig. 5A), quite distinct from eukaryotic DUBs, which seem to rely on the Ile44 patch for affinity, and also from the Ulp family member Den1’s recognition of NEDD8 (1, 27, 28).

Structural Similarity and Differences from Den1.

Structural similarity analysis using the DALI server (29) identified Den1 (NEDP1 or SENP8) apo and in complex with NEDD8 as high-scoring hits [Protein Data Bank (PDB) ID codes 2BKR, 2BKQ, and 1XT9] (27, 28). Den1 is a member of the Ulp family of Cys proteases, whose members share a characteristic core fold comprised of a five-stranded mixed β-sheet packed against a helix (27, 28, 30). Den1 and SdeA_Dub are structurally similar, with most of the β-sheets (excluding β6) and α4 (bearing the catalytic cysteine) adopting a similar fold (Fig. S7A). Even though the core Ulp domains were consistent, differences exist, particularly in regions used for binding to ubiquitin or Ubls (Fig. S7B). A simple S-shaped loop at the beginning of SdeA_Dub is replaced in Den1 by a helix followed by a β-hairpin and is used in NEDD8 binding. Notably, β2 of SdeA, which forms a hairpin turn with β1 to create the Tyr33 binding loop, is replaced with α3 in Den1, which mediates NEDD8 interaction via the Ile44 hydrophobic patch.

The structures of the Den1-NEDD8 and SdeA_Dub-Ubiquitin complexes also exhibit differences and similarities. The most notable similarity is the relative orientation of the C-terminal tail with respect to the adjacent arm of loop L6, which appears to be a characteristic element in the Ulp family. Loop L6 in our structures is rigid, forming a β-hairpin with a bulge (Fig. S7D). In Den1, the corresponding loop is quite dynamic, undergoing a significant conformational change upon binding to NEDD8. SdeA_Dub buries ∼900 Ų [according to the proteins, interfaces, structures and assemblies (PISA) server] of surface area upon binding ubiquitin, which is substantially less than Den1 (burying nearly 2,000 Ų when bound to NEDD8) as a result of greater recognition of NEDD8 by Den1 by structural elements absent in SdeA_Dub (31). The most prominent of these substitutions is an N-terminal β-hairpin in Den1 that nestles against the loop segment of NEDD8 (Lys48-Asn41; Fig. S7E). This hairpin is absent in SdeA_Dub and would mediate the corresponding interaction with Lys48 of ubiquitin, making it exo-specific for this chain type (that is, capable of cleaving ubiquitin chains only from the most distal end). Loss of this element in the bacterial DUB may allow it to rapidly cleave from anywhere in the chain. The other prominent difference is the absence of helix-4 (Den1 numbering) in SdeA_Dub, which plays an important role in engaging the Ile44 patch. To probe the importance of the observed contacts in solution, we generated alanine point mutants and tested their activity against Lys-63–linked diubiquitin as well as a Nedd8 fusion construct (His-MBP-NEDD8-Ub). From these experiments, we find that some variations within the catalytic cleft are tolerated, whereas mutating Y33 and D61 (Y-site) resulted in a complete loss of activity. Moreover, results observed in experiments with the NEDD8 fusion construct mirrored those of Lys63-linked diubiquitin, suggesting a similar mode of recognition (Fig. S8).

The selectivity of SdeA_Dub for Lys63-linked chains may rely on chain topology. Because Lys63-linked chains adopt an extended conformation (whereas Lys48- and Lys11-linked chains do not), they may be strung more easily through the catalytic site. Based on this observation, we hypothesize that the preference observed for Lys63-linked chains is a result of conformational differences between the chains themselves rather than structural restrictions imposed by SdeA_Dub. However, it is possible that selective binding to the amino acids that flank Lys63 contributes to the preference for this chain type; however, this remains to be determined in future studies.

Legionella pneumophila Strains Used in This Study.

Legionella strains used in the study include WT (Lp02), the DotA mutant (Lp03), Lp02 ∆SdeA-C∆SidE (referred to as ∆4), and relevant complementary strains. The SidE family in-frame deletion mutants were constructed by well-established protocols. Legionella strains used for infection were grown in ACES buffered yeast extract broth to postexponential phase (OD₆₀₀ of approximately 3.2–4.0). The expression of SdeA in Legionella pneumophila strains was detected by an SdeA-specific antibody produced in rabbit (Pocono Rabbit Farm and Laboratory).

For in vivo growth of L. pneumophila strains, Dictyostelium discoideum (strain AX4) cells were seeded onto 24-well plates at a density of 5 × 10⁵ cells per well. The seeded cells were infected with bacteria at a multiplicity of infection (MOI) of 0.05. The intracellular cfu counts after 2 h, 24 h, 48 h, and 72 h were obtained as previously published.

Cleavage of Polyubiquitinated/Polyneddylated Proteins by SdeA_Dub.

HEK293 cells were transfected with pCDNA3 HA-Ubiquitin or pCMV 4XFlag-NEDD8 by Lipofectamine 2000 (Life Technologies). Cells were collected 24 h after transfection and lysed in lysis buffer [0.2% Nonidet P-40, 50 mM Tris, pH7.5, 150 mM NaCl, 1 mM EDTA, 15% (vol/vol) glycerol, and Protease Inhibitor Mixture Set III (EMD Millipore)]. The total cell lysates were then subjected to immunoprecipitation by HA agarose or Flag affinity gel (Sigma-Aldrich) for 4 h, and the beads were washed with lysis buffer three times and resuspended in cleavage buffer containing 50 mM Tris, pH 7.5, 100 mM NaCl, and 1 mM DTT. Samples were divided into aliquots and left untreated or treated with 100 µg of SdeA_Dub or SdeA_DubCys118Ala for 1 h at room temperature. The beads were then washed three times using the cleavage buffer and boiled for 5 min at 100 °C in 1× SDS/PAGE buffer. The proteins were then separated by SDS/PAGE, and polyubiquitinated proteins or polyneddylated proteins detected by HA and Flag antibodies, respectively.

Detection of SdeA Translocation.

For detection of translocated SdeA, 2 × 10⁷ U937 cells were seeded onto a 10-cm Petri dish 12 h before infection. Cells were infected with L. pneumophila strains for 2 h at an MOI of 5. Cells were collected by centrifugation and resuspended in 100 μL of PBS solution containing 0.2% saponin. After 30 min incubation on ice, total cell lysates were cleared by centrifugation (10,000 × g for 10 min at 4 °C). The supernatants were subjected to SDS/PAGE followed by Western blot with SdeA and tubulin antibodies.

Yeast Toxicity Assay.

SdeA or SdeA_C118A was cloned into pSB157, which contains a galactose-inducible promoter. The plasmids were linearized and transformed into Saccharomyces cerevisiae (strain W303). The yeast strain (pGal::SdeA or pGal::SdeA_C118A), which consistently exhibited galactose-dependent toxicity, was isolated. The SidJ gene was cloned into p415ADH, and transformed into W303 (pGal::SdeA or pGal::SdeA_C118A). To determine the suppressor activity, yeast strains were grown in liquid selective medium containing glucose, and the cultures were serially diluted and spotted onto glucose or galactose medium. Plates were incubated at 30 °C for 3 d before the images were acquired.

Quantification of Phagosome Ubiquitination Status.

Bone marrow progenitors from 6–10-wk-old female A/J mice were differentiated for 6 d with RPMI medium containing 20% FBS and 20% (vol/vol) in L-cell supernatant-conditioned medium as described previously (41). Macrophages were seeded into 24-well plates 12 d before infection at 2 × 10⁵ per well. The cells were infected with L. pneumophila strains for 1 h at an MOI of 1. For ubiquitin recruitment analyses, macrophages were fixed and stained by standard procedures. Anti-L. pneumophila antibodies were used at a dilution of 1:5,000. Polyubiquitinated proteins were stained by the FK2 antibody (EMD Millipore) at a dilution of 1:50 followed by staining with goat anti-mouse IgG Secondary Antibody (Life Technologies). Lys63-linked polyubiquitin was likewise detected with a Lys63 linkage-specific antibody (Millipore). The recruitment of polyubiquitinated proteins was analyzed by Olympus IX-81 fluorescence microscope. Images retrieved from an Orca camera were processed with the IPlab software package (Scanalytic).

Protein Purification and Generation of Reagents.

The SidE DUB domains were cloned into pGEX6P-1 and transformed into Escherichia coli XL1-Blue or BL21 codon plus cells. Cultures were grown to an OD₆₀₀ of 0.4–0.6 at 37 °C and induced with 200 µM IPTG. Protein was expressed at 18 °C for 16–18 h and the cells were lysed via French press. The lysate was then cleared by centrifugation at 100,000 × g for 1 h. The GST-fused constructs were then purified using standard GST fusion tag purification protocol. Protein samples were further purified by size exclusion chromatography (Superdex 75). Full-length SdeA was cloned into pQE30 and transformed into E. coli XL1-Blue. Cells harboring the full-length protein construct were grown to an OD₆₀₀ of 0.5 at 37 °C and induced with 200 µM IPTG. After induction, the temperature was lowered to 18 °C for 18 h. The cells were then resuspended and lysed via French press. The lysate was then cleared by centrifugation at 100,000 × g for 1 h. The His-tagged protein was then purified following standard purification protocols. Selenomethionine-labeled protein was produced by growing E. coli in minimal media to an OD₆₀₀ of 0.4–0.6. Cultures were then induced with 200 µM IPTG, and the media was supplemented with selenomethionine. The resultant protein was then purified following standard protocol as discussed earlier. DUB mutants were generated by using site-directed mutagenesis, and the sequences were confirmed by DNA sequencing. The resulting constructs were used following virtually identical protocols as described earlier. His-MBP-Nedd8-Ub (in pLou3; gift from Ronald Hay, University of Dundee, Dundee, United Kingdom) was purified following standard His-tagged protein purification protocol as outlined earlier (27).

Ubiquitin (1-76; WT and mutants) in pET26b (a gift from Yasuke Sato, University of Tokyo, Tokyo) was transformed into E. coli Rosetta cells. The cells were then grown to an OD₆₀₀ of 0.4–0.6, induced with 200 µM IPTG, and grown for 16–18 h at 18 °C. Cultures were then harvested by centrifugation and the resultant pellet was lysed by French press. Ubiquitin was then purified on SP Sepharose resin. Ubiquitin-MESNa (in pTXB1) was transformed into BL21 cells, and cultures were grown to an OD₆₀₀ of 0.4–0.6, at which point protein expression was induced with 200 µM IPTG at 18 °C for 16–18 h. The cells were harvested by centrifugation and lysed via French press. Ub-MESNa was then purified essentially as has been previously described (20). Glycine vinyl methyl ester was synthesized following a previously published procedure and coupled to Ub-MESNa by using semisynthetic methods as has been previously described (20). Resultant Ub-VME was purified using cation exchange chromatography using a MonoS column (GE Biosciences).

Diubiquitin substrates were synthesized in-house using enzymes specific to each chain type, as has been previously described (42). For Lys-11–linked diubiquitin, ubiquitin K11R K63R was incubated at 37 °C with GST-E1 and Ube2S until an appreciable amount of Lys-11–linked diubiquitin had formed (42). Lys-48–linked diubiquitin was generated by incubating ubiquitin with GST-E1 and CDC34 at 37 °C, and Lys-63 linkages were synthesized with GST-E1, Ubc13, and Uev1a at 37 °C (42). All reactions were quenched with 50 mM sodium acetate, pH 4.5, and further purified by using cation exchange chromatography (MonoS; GE Biosciences).

Diubiquitin Cleavage Assays.

Chain cleavage assays were conducted by using SdeA as well as the DUB domains from SdeA, SdeB, and SdeC. For Lys-63 diubiquitin cleavage assays, 50 nM of each enzyme was incubated with various concentrations of diubiquitin (10–180 µM) in 50 mM Tris, pH 7.4, 50 mM NaCl, 1 mM DTT for 5 min at room temperature. The assay was quenched with 5× SDS/PAGE buffer and analyzed by SDS/PAGE. The concentration of ubiquitin was quantified by using densitometric analysis in ImageJ and determined from a standard curve prepared on the same gel (43). Initial velocities were then plotted against substrate concentration and the kinetic parameters determined by fitting the data to a model describing Michaelis–Menten kinetics. For Lys-11– and Lys-48–linked diubiquitin chains, DUB assays were carried out essentially as described earlier, but the data could not be fit into a Michaelis–Menten model, presumably because of high K_m values outside the concentration range studied.

Deubiquitination of polyubiquitinated GFP was accomplished by incubating 2 µM GFP-Ub_n with 50 nM of each protein (full-length and DUB domains) for 10 min at room temperature. The reaction was quenched by the addition of 5× SDS/PAGE loading buffer and subsequently analyzed by SDS/PAGE. Deubiquitination was evidenced by the disappearance of the high molecular weight band and the appearance of unmodified GFP.

Generation and Purification of the SdeA_Dub–Ub-VME Complex.

To generate the SdeA_Dub–Ub-VME complex, purified SdeA_Dub and Ub-VME were mixed in equimolar ratios and the pH was adjusted to 8.0. The reaction was then incubated at room temperature for 4 h and quenched with 50 mM Tris⋅HCl, pH 7.4. The covalent SdeA_Dub–Ub-VME complex was then purified from the remaining unreacted reagents by passing the mixture over a MonoQ anion exchange column (GE Biosciences).

GFP-PY was purified from E. coli Rosetta cells and polyubiquitinated following previously published protocols (23). Briefly, GFP-PY was incubated at 37 °C with ubiquitin, GST-E1, Ubc4, and RSP5 for 6 h until a molecular weight of greater than 200 kDa was reached. The reaction mixture was then flash-frozen for further use.

Protein Crystallization and Structure Determination.

Protein crystallization trials were set up using the vapor diffusion technique with a sparse matrix screening protocol. Crystals of SdeA_Dub Cys118Ala grew from 0.2 M potassium chloride, 0.05 M Hepes:NaOH, pH 7.5, 35% (vol/vol) pentaerythritol/propoxylate; crystals of the SdeA_Dub–Ub-VME complex grew in 0.1 M Tris:HCl, 3.0 M sodium chloride; and SdeA_1–163 from 0.1 M Tris:HCl, 3.0 M sodium chloride, 3% methyl-2,4-pentanediol. Diffraction data were collected at the Advanced Photon Source at Argonne National Laboratory and processed by using HKL2000 (44). The SdeA_1–163 SeMet data set was phased de novo from anomalous signal collected at the selenium edge using Phenix Autosol, and a preliminary polyalanine model was built by using Phenix Autobuild. The structure was then refined by using Phenix Refine, correcting for pseudomerohedral twinning. Stereochemistry weighting and ADP refinement was applied through various rounds of refinement (Table S1). This model then served as a search model for molecular replacement (MR) of the SdeA_Dub Cys118Ala crystals by using Phaser MR in Phenix. This search identified two subunits in the asymmetric unit. A model was built by using Phenix Autobuild and refined in Phenix refine (45). The crystallographic data collection and refinement parameters are listed in Table S1.

SdeA_Dub C118A then served as a search model for the MR of SdeA_Dub–Ub-VME crystals. The MR search was initially conducted in the P1 space group by using MolRep in the CCP4 suite (46, 47). Inspection of the map generated after rigid body refinement followed by restrained refinement revealed the presence of extra electron density corresponding to the unaccounted-for ubiquitin molecule. The C-terminal tail and β1–β2 hairpin turn of ubiquitin were first built into the density and subjected to refinement in RefMac (48, 49). When an appreciable portion of the hairpin turn and the C-terminal tail had been built, ubiquitin (PDB ID code 1UBQ) was aligned with these segments, which permitted the placement of ubiquitin near the binding interface (50). This model was used as a probe for a second round of MR by using MolRep in the P32 space group (48, 49). The structure was then refined in RefMac (48, 49). The crystallographic data collection and refinement parameters are listed in Table S1.

Diubiquitin and Nedd8 Cleavage Assays Using SdeA_Dub Mutants.

SdeA_Dub Lys63 diubiquitin cleavage assays were carried out in triplicate with 50 nM of each SdeA_Dub mutant and 10 µM diubiquitin. To analyze the efficiency of diubiquitin cleavage, ImageJ was used to quantify the amount of ubiquitin that had built up after incubating for 5 min at room temperature. The amount of ubiquitin generated by each mutant was then normalized to the amount generated by the WT protein to deduce the relative catalytic efficiency. Similar methods were used to probe the effect of each mutant on the ability to cleave His-MBP-Nedd8-Ub, but the concentration of each enzyme construct was adjusted to 1 µM and the incubation time was extended to 30 min. Analysis was carried out as described for Lys63 diubiquitin cleavage.

Analytical Ultracentrifugation.

The sedimentation velocity experiments were performed on a Beckman Coulter XLA Analytical Ultracentrifuge. The samples were dialyzed overnight into 50 mM Tris⋅HCl, 50 mM NaCl, 1 mM DTT, pH 7.6. The samples were spun at 50,000 rpm at 20 °C in a two-sector charcoal Epon centerpiece (Beckman Coulter). The moving boundary was monitored at 280 nm. The c(s) distribution and hydrodynamic properties were determined by using SEDFIT (v. 14.7g) (51). The viscosity, density, and partial specific volume was calculated using SEDNTERP (sednterp.unh.edu/). The sedimentation coefficient distribution, c (s), were created using GUSSI (v. 1.05k).