Search tips
Search criteria 


Logo of actafjournal home pagethis articleInternational Union of Crystallographysearchsubscribearticle submission
Acta Crystallogr Sect F Struct Biol Cryst Commun. 2008 May 1; 64(Pt 5): 343–350.
Published online 2008 April 30. doi:  10.1107/S1744309108009275
PMCID: PMC2376413

Structure of Staphylococcus aureus 5′-methylthioadenosine/S-adenosylhomocysteine nucleosidase


5′-Methylthioadenosine/S-adenosylhomocysteine nucleosidase (MTAN) catalyzes the irreversible cleavage of the glycosidic bond in 5′-methylthioadenosine (MTA) and S-adenosylhomocysteine (SAH) and plays a key role in four metabolic processes: biological methylation, polyamine biosynthesis, methionine recycling and bacterial quorum sensing. The absence of the nucleosidase in mammalian species has implicated this enzyme as a target for antimicrobial drug design. MTAN from the pathogenic bacterium Staphylococcus aureus (SaMTAN) has been kinetically characterized and its structure has been determined in complex with the transition-state analogue formycin A (FMA) at 1.7 Å resolution. A comparison of the SaMTAN–FMA complex with available Escherichia coli MTAN structures shows strong conservation of the overall structure and in particular of the active site. The presence of an extra water molecule, which forms a hydrogen bond to the O4′ atom of formycin A in the active site of SaMTAN, produces electron withdrawal from the ribosyl group and may explain the lower catalytic efficiency that SaMTAN exhibits when metabolizing MTA and SAH relative to the E. coli enzyme. The implications of this structure for broad-based antibiotic design are discussed.

Keywords: 5′-methylthioadenosine/S-adenosylhomocysteine nucleosidase, structure-based drug design, Staphylococcus aureus, quorum sensing, biological methylation, methionine recycling, polyamine biosynthesis

1. Introduction

5′-Methylthioadenosine/S-adenosylhomocysteine nucleosidase (MTAN) is a dual substrate-specific enzyme that irreversibly hydrolyzes the glycosidic bond of 5′-methylthioadenosine (MTA) or S-adenosyl­homocysteine (SAH) to form adenine and 5′-methylthioribose or S-­ribosylhomocysteine, respectively (Duerre, 1962 [triangle]). The nucleosidase has been identified as an exploitable antibiotic target because it is essential for the metabolism of MTA and SAH in many microbes such as Staphylococcus aureus, Streptococcus pneumoniae, Haemophilus influenzaeMycobacterium tuberculosis and Helicobacter pylori but is not found in mammals. In mammalian systems, the breakdown of MTA and SAH requires two separate enzymes, MTA phosphorylase (Pegg & Williams-Ashman, 1969 [triangle]) and SAH hydrolase (de la Haba & Cantoni, 1959 [triangle]), respectively. Accumulation of MTA and SAH leads to negative-feedback inhibition of four critical metabolic processes: biological methylation (Miller et al., 1994 [triangle]), polyamine biosynthesis (Raina et al., 1982 [triangle]; Pajula & Raina, 1979 [triangle]), methionine recycling (Sufrin et al., 1995 [triangle]; Riscoe et al., 1989 [triangle]) and bacterial quorum sensing (Chen et al., 2002 [triangle]; Schauder et al., 2001 [triangle]).

Eight Escherichia coli MTAN structures have been determined (Lee et al., 2001 [triangle], 2003 [triangle]; Lee, Singh et al., 2005 [triangle]; Lee, Smith et al., 2005 [triangle]) and are available in the Protein Data Bank. These include complexes with substrate and transition-state analogues as well as products and provide a strong foundation for structure-based inhibitor design. The structure of MTAN from the pathogenic bacterium Streptococcus pneumoniae has also been determined in complex with a transition-state analogue. To date, two generations of transition-state analogues have been developed and the tightest binding inhibitors exhibit picomolar to femtomolar binding affinity for E. coli MTAN (Singh, Evans et al., 2005 [triangle]; Singh, Lee et al., 2005 [triangle]).

To determine whether the structure of the E. coli enzyme could be used for broad-based antibiotic design, the structure of the nucleosidase from the pathogenic bacterium Staphylococcus aureus (SaMTAN) in complex with the transition-state analogue formycin A (FMA; Fig. 1 [triangle]) has been determined. This structure provides insight into the inhibitor–enzyme interactions and the mode of nucleoside binding. A comparison of the SaMTAN–FMA and available E. coli MTAN structures reveals strong conservation of active-site residues and an identical mode of inhibitor binding. Primary sequence analysis of the enzyme from a number of bacterial pathogens further supports the proposal that any bacterial MTAN structure could serve as a good template for the design of a broad-spectrum antibiotic. However, kinetic characterization demonstrates that knowledge of the structure of the enzyme and the protein–ligand interactions are insufficient to explain the wide range of binding affinities observed for various pathogenic MTANs for a series of transition-state analogues (Gutierrez et al., 2007 [triangle]).

Figure 1
Structures of SAH, MTA and FMA. Note that for ease of comparison, the numbering of atoms in the purine base of the inhibitor FMA is based on the numbering convention of MTA and not the IUPAC standard.

2. Materials and methods

2.1. Protein expression and purification

The S. aureus pfs gene was amplified from genomic DNA by the polymerase chain reaction. The gene fragment was subsequently cloned into the NdeI and XhoI restriction sites of the pET28a vector to allow expression with an N-terminal six-His tag. The plasmid DNA was transformed into BL21 (DE3) cells. Single colonies were picked and grown overnight at 310 K in 50 ml Luria–Bertani broth (LB) containing 100 µM kanamycin. The overnight culture was used to inoculate 1.6 l LB containing a final concentration of 100 µM kanamycin and grown to an OD600 of 0.6 at 310 K. Cells were induced by the addition of isopropyl β-d-1-thiogalactopyranoside (IPTG) to 400 µM and cultured for an additional 3 h at 300 K. The induced culture was harvested by centrifugation at 6000 rev min−1 in a JA-10 rotor for 10 min. Cells were resuspended in BugBuster Protein Extraction Reagent (Novagen) and lysed by gentle vortexing for 25 min. The crude lysate was fractionated by centrifugation at 18 000 rev min−1 in a JA-20 rotor at 277 K for 20 min. The soluble fraction was applied onto a 10 ml FPLC column of Ni–NTA Superflow resin (Qiagen) pre-equilibrated with 25 mM potassium phosphate pH 7.5, 250 mM NaCl and 20 mM imidazole (buffer A). The column was washed with three column volumes of buffer A and two column volumes of buffer A with 150 mM imidazole. The recombinant nucleosidase was eluted from the Ni–NTA column with a gradient of buffer A plus 150–500 mM imidazole. Fractions containing the nucleosidase were dialyzed against 50 mM potassium phosphate pH 7.5 overnight at 277 K. Trypsin digestion at 1:1000 dilution for 30 min at room temperature removed the polyhistidine tag, leaving a four-residue (GSHM) linker sequence. The proteolytic reaction was stopped by the addition of polymethylsulfonyl fluoride (PMSF) to a final concentration of 1 mM. The nucleosidase was further purified on a Superdex-200 HR 10/30 column pre-equilibrated with 25 mM sodium HEPES pH 7.5. Isocratically eluted fractions containing the nucleosidase were pooled and concentrated to 15 mg ml−1 using a Millipore BioMax 10K Ultrafree 0.5 ml centrifugal concentrator. The purity and integrity of the sample were checked by SDS–PAGE and electrospray ionization mass spectrometry (Advanced Protein Technology Centre, The Hospital for Sick Children) prior to crystallization studies.

2.2. Kinetic characterization

The catalytic activity of SaMTAN was examined using a xanthine oxidase-coupled spectrophotometric assay (Dunn & Bryant, 1994 [triangle]). For each molecule of adenine oxidized by xanthine oxidase, two molecules of 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyltetrazo­lium chloride (INT) are reduced to formazan, which absorbs at 470 nm (de Groot et al., 1985 [triangle]). Changes in absorption at 470 nm were converted to the amount of adenine released using a molar absorption coefficient of 15 400 M −1 cm−1 at pH 7.0. All reactions were performed in 50 mM potassium phosphate pH 7.0 using 0.72–4.9 µg EcMTAN or SaMTAN, 0.28 units of xanthine oxidase (Sigma–Aldrich Chemicals) and 1 mM INT (Sigma–Aldrich Chemicals) in a final volume of 800 µl. The concentrations of MTA and SAH were varied from 0.31 to 25 µM. The reactions were monitored on a Biochrome Ultrospec 2100 UV–Vis spectrophotometer fitted with the SWIFT II enzyme-kinetics software (Amersham Pharmacia Biotech). K m and V max values were determined by fitting the initial velocity data to the Michaelis–Menten equation using the SigmaPlot Enzyme Kinetics module.

2.3. Crystallization

Although a variety of different commercial crystallization screens were tested, the best initial condition for apo-SaMTAN was obtained by the hanging-drop vapour-diffusion technique using Hampton Crystal Screen and contained 100 mM sodium HEPES pH 7.5, 0.8 M sodium phosphate and 0.8 M potassium phosphate. Grid screening was subsequently performed to optimize the buffer and precipitant concentrations. The best quality crystals were obtained using 50 mM potassium phosphate with 20%(w/v) PEG 8000. The crystals of SaMTAN–FMA were prepared by first pre-incubating the protein with 1 mM formycin A for 2 h on ice. Crystals with approximate dimensions of 0.2 × 0.2 × 0.2 mm grew within one week.

2.4. Data collection and processing

In preparation for data collection at liquid-nitrogen temperatures, the SaMTAN–FMA crystal was soaked for 2 min in a cryoprotectant [20%(w/v) PEG 8000, 50 mM potassium phosphate, 1.5 mM formycin A and 15%(v/v) glycerol] prior to flash-freezing in the cryostream. Data were measured on an R-AXIS IV++ image plate using an RUH3R rotating copper-anode X-ray generator with Osmic optics at The Hospital for Sick Children. The data were indexed, integrated, scaled and merged using d*TREK (Pflugrath, 1999 [triangle]).

2.5. Structural determination and refinement

The structure of SaMTAN–FMA was determined by molecular replacement using only the protein coordinates of the structure of E. coli MTAN complexed with FMA (EcMTAN–FMA) as the search model (PDB code 1nc3). CNS-SOLVE was used for all structure-determination and refinement steps (Brünger et al., 1998 [triangle]). Data between 15 and 4 Å resolution were included in the cross-rotation and translation functions and produced an R cryst of 46.6%. The resulting model was subjected to an initial round of rigid-body refinement and maximum-likelihood torsion-angle refinement (Brünger et al., 1998 [triangle]) prior to replacing the E. coli sequence with the S. aureus sequence. Rounds of manual rebuilding using Xfit (McRee, 1999 [triangle]) were subsequently alternated with maximum-likelihood torsion-angle refinement starting at 5000 K using all data, a bulk-solvent correction and an anisotropic B-factor correction. The progress of the refinement was monitored by reductions in R cryst and R free and σA-weighted 2F oF c and F oF c maps as implemented in CNS (Read, 1986 [triangle]) were carefully examined at the end of each round of refinement. A σA-weighted F oF c map calculated after the initial round of torsion-angle refinement revealed strong positive electron density corresponding to formycin A in the SaMTAN active site. The FMA coordinates were obtained from the EcMTAN–FMA structure and topology and parameter files were generated using the XPLO2D server (Kleywegt, 1995 [triangle]). Water molecules were added during the later rounds of refinement in accord with good electron density (σA-weighted F oF c peaks greater than 3σ) and the formation of hydrogen bonds (less than 3.2 Å) or van der Waals contacts less than 4.0 Å to other atoms. After the addition of the inhibitor and water molecules, alternating rounds of conjugate-gradient refinement and model rebuilding using Xfit were performed followed by individual isotropic B-factor refinement. In the final refined structure of SaMTAN–FMA, all 228 protein residues and two of the four residues from the N-terminal tag were modelled. Analysis of the structure using MOLPROBITY (Lovell et al., 2003 [triangle]) showed that more than 97% of residues are in the favoured regions and 100% of residues are in the allowed regions of the Ramachandran plot. Data-collection and refinement statistics are reported in Table 1 [triangle]. In the following subsequent analyses, the structure of SaMTAN–FMA was aligned with EcMTAN–FMA (PDB code 1nc3) and with the E. coli enzyme complexed with adenine (ADE; EcMTAN–ADE; PDB code 1jys) by a nonlinear least-squares fit of all main-chain (N—Cα—C) atoms using the program PROFIT (Smith, 1993 [triangle]).

Table 1
Data-collection and refinement statistics for SaMTAN–FMA

3. Results and discussion

3.1. Kinetic characterization

SaMTAN has comparable binding affinities but reduced maximum velocities for the substrates MTA and SAH compared with EcMTAN (Table 2 [triangle]). The maximum velocities for SaMTAN for MTA and SAH are reduced 52-fold and 53-fold, respectively. The catalytic efficiency of SaMTAN towards MTA and SAH is approximately ninefold and 25-fold lower, respectively, than that of EcMTAN. The reduced catalytic efficiency of SaMTAN may be rationalized by a structural difference in the active site, as described below.

Table 2
Kinetic parameters for SaMTAN and EcMTAN

3.2. Structural determination and comparison of MTANs

3.2.1. Overall structure

SaMTAN crystallizes with a monomer in the asymmetric unit. Each monomer consists of a mixed α/β domain, with a central mixed ten-stranded β-sheet flanked by six α-helices and a small 310-helix (Fig. 2 [triangle] a). The biologically active dimer is generated by the action of a crystallographic twofold axis (residues from the α2, α5, α6 and 310 helices and th,e loops between β2 and β3 and between β4 and α2). The interface area between the two monomers is 1506.3 Å2 and comprises 15% of the surface of each monomer (Krissinel & Henrick, 2007 [triangle]).

Figure 2
Structure of SaMTAN and a comparison to EcMTAN. (a) Stereo diagram of the quaternary structure of SaMTAN. Superpositions of (b) the monomeric structures, (c) the adenine and ribose subsites and (d) the 5-alkythio-binding subsites of SaMTAN–FMA ...

EcMTAN has previously been shown to exist in two different conformations: ‘open’ and ‘closed’ (Lee et al., 2001 [triangle], 2003 [triangle]). The ‘open’ state is defined by a larger and more solvent-accessible purine- and ribose-binding site, as seen in the adenine (EcMTAN–ADE; Lee et al., 2001 [triangle]) and glycerol complexes (EcMTAN–GOL; Lee, Smith et al., 2005 [triangle]). Structures of E. coli enzyme–inhibitor complexes suggest that substrate binding induces a series of conformational changes to a ‘closed’ state (Lee et al., 2003 [triangle]). In the ‘closed’ state, all water molecules except for the catalytic water are excluded from the active site. For the purposes of the following discussion, we have selected the EcMTAN–FMA (Lee et al., 2003 [triangle]) and EcMTAN–ADE (Lee et al., 2001 [triangle]) structures as the representative E. coli structures for the ‘closed’ and ‘open’ conformations, respectively. A comparison of SaMTAN–FMA with the EcMTAN–FMA and EcMTAN–ADE structures reveals Cα root-mean-square displacements of 0.55 and 1.37 Å, respectively, showing that the overall structure of SaMTAN–FMA is more similar to the ligand-induced ‘closed’ conformation of EcMTAN than the ‘open’ conformation.

3.2.2. Active site

The nucleosidase active site is typically described in terms of three subsites that accommodate the adenine, ribose and 5′-alkylthio moieties of the substrate. The initial σA-weighted F oF c map revealed strong electron density corresponding to the inhibitor formycin A in the active site (Figs. 2 [triangle] c and 2 [triangle] d). Note that the amino-acid sequence in SaMTAN is shifted by −1 relative to the numbering of the E. coli homologue. In the purine-binding subsite (Fig. 3 [triangle] a), Oδ1 and Oδ2 of Asp196 make hydrogen bonds to the N6 and N7 atoms of the purine base of formycin A, respectively. Asp196 is stabilized by two hydrogen bonds from Ser195 Oγ and the amide N atom of Ala198. The purine N6 and N1 atoms also form hydrogen bonds to the main-chain carbonyl O atom and amide N atom of Ile152, respectively. Phe150 makes a base-stacking interaction with the purine ring. A hydrogen bond (3.0 Å) exists between Ser75 Oγ and the purine N8 atom. Residues Glu11, Ser75, Met172 and Glu173 lie on the surface of the ribose-binding subsite, while two water molecules are found within this subsite. The ribose is anchored by van der Waals interactions between the side chain of Met172 and the hydrophobic face of the ribose and by a hydrogen bond between the amide N atom of Met172 and the 2′-hydroxyl. In addition, the ribosyl 2′- and 3′-­hydroxyl O atoms donate protons to the O[sm epsilon]1 and O[sm epsilon]2 carboxylate groups of Glu173. Two water molecules (WAT3 and WAT83) interact with the ribose moiety. As in the E. coli MTAN structures, WAT3 is the putative nucleophilic water molecule. This water molecule is approximately 3.6 Å from the anomeric carbon and forms hydrogen bonds to the ribosyl 3′-hydroxyl O atom (2.7 Å), Glu11 O[sm epsilon]2 (2.7 Å) and Arg192 Nη1 (2.9 Å). WAT83 forms a 2.8 Å hydrogen bond to the ribosyl O4′ ring O atom and 2.6 and 3.0 Å hydrogen bonds to Glu11 O[sm epsilon]2 and Ser75 Oγ, respectively. The 5′-substituent of formycin A makes intrasubunit van der Waals interactions with residues Met8, Ile49, Phe150, Met172 and Phe206 and intersubunit van der Waals interactions with Phe104 and Pro112. Residues Ala101 and Tyr106 also form the 5′-alkylthio-binding site but do not make any van der Waals interactions to the ligand. The 5′ hydroxyl of formycin A makes no hydrogen bonds to the protein. The longer 5′-­methylthio or homocysteinyl groups in the natural substrates, MTA and SAH, would be likely to interact with Ala101 and Tyr106.

Figure 3
Inhibitor conformation and interactions. Schematics of the interactions between the ligand and the protein in the active sites of (a) SaMTAN–FMA, (b) monomer A of EcMTAN–FMA and (c) monomer B of EcMTAN–FMA. Residues donated by ...

A comparison of the active-site architectures of S. aureus MTAN and E. coli MTAN reveals a striking similarity (Figs. 2 [triangle] c, 2 [triangle] d and 3 [triangle]). The positions of the putative catalytic acid (Asp197/Asp196), catalytic base (Glu12/Glu11) and nucleophilic water (WAT3) are all conserved. In addition, all of the residues involved in substrate binding are conserved, with the exception of Val102, which is replaced by Ala101 in SaMTAN (Fig. 2 [triangle] d). This conservative substitution does not appear to perturb the architecture of the binding site or the specificity of the enzyme. The average displacement of the Cα positions of residues in the active site is 0.3 Å and the side-chain conformations and hydrogen-bond distances are nearly identical. The only significant difference is the presence of an extra water molecule (WAT83) in the S. aureus enzyme, which forms hydrogen bonds to Ser75, Glu11 and the O4′ atom of the ligand. This additional hydrogen bond to O4′ of the ligand would cause electron withdrawal from the ribosyl group and from the σ*(C1′—N9) antibonding orbital, thereby strengthening the C1′—N9 covalent bond. This is likely to contribute to the reduction in catalytic efficiency in SaMTAN compared with EcMTAN. While WAT83 is absent from all available inhibitor-complexed structures of EcMTAN, this second water molecule is also present in the 1.6 Å ligand-bound structure of Streptococcus pneumoniae MTAN (PDB code 1zos). Like SaMTAN, S. pneumoniae MTAN is also less catalytically efficient compared with E. coli MTAN (Singh & Schramm, 2007 [triangle]). In the structures of SaMTAN and S. pneumoniae MTAN, the side chains of Met8/9 are directed away from WAT83. In EcMTAN Met9 Cξ occupies the place where WAT83 would be, hence explaining the displacement of WAT83 from the EcMTAN active site.

3.2.3. Nucleoside conformations and implications for the catalytic mechanism

In solution, MTA and SAH are found predominantly in the anti conformation with a ribosidic torsion angle (C8—N9—C1′—O4′) of 45° (Markham et al., 2002 [triangle]). The binding of nucleosides to MTAN has previously been shown to significantly alter the conformation of the nucleoside. In the SaMTAN–FMA structure, formycin A adopts a highly anti N9—C1′ ribosidic bond conformation and a high-energy C4′-endo sugar pucker. These energetically unfavourable conformations are likely to place strain on the ribosidic N9—C1′ bond and facilitate cleavage. The exocyclic C4′—C5′ bond can be described by two torsion angles, ϕoo (O5—C5′—C4′—O4′) and ϕoc (O5—C5′—C4′—C3′). In the SaMTAN–FMA structure, ϕoo = −179.6° and ϕo = −68.3°, corresponding to a transgauche conformation. Superposition of formycin A from S. aureus and E. coli reveal identical nucleoside conformations (Fig. 3 [triangle] d). Given that the overall structure, active-site residues and ligand-binding interactions are conserved between the E. coli and S. aureus enzymes, we anticipate that the catalytic mechanism will be the same. This conclusion is supported by the recent studies of the transition-state structure of various MTANs (Gutierrez et al., 2007 [triangle]). Both the S. aureus and E. coli enzymes were found to exhibit a late dissociative transition state.

3.3. Implications for structure-based drug design

The structure of S. aureus MTAN establishes that the architecture of its active site and the nucleoside–enzyme interactions are identical to those found in the E. coli enzyme. The recently determined structure of the S. pneumoniae enzyme also shows structural similarity to the E. coli enzyme (Singh et al., 2006 [triangle]). The primary sequence alignment of MTANs from multiple bacterial pathogens demonstrates sequence identity of between 30% and 100% (Fig. 4 [triangle]), with almost all active-site residues being conserved. The observed structural similarity suggests that any of the available structures could be used for modelling or docking studies for the design of broad-spectrum antibiotics, where the dynamics of the protein are ignored. As the ninefold to 25-fold decrease in catalytic efficiency observed for the S. aureus enzyme relative to the E. coli enzyme (Table 2 [triangle]) does not fully explain the 700-fold to 1000-fold differences observed in the binding affinities of the methylthio-ImmA and methylthio-DADMe-ImmA transition-state analogues (7.7 × 104 versus 77 pM for ImmA and 1.4 × 103 versus 2 pM for DADMe-ImmA, respectively; Gutierrez et al., 2007 [triangle]), inhibitor design for MTAN and the potency of an inhibitor for a particular bacterial species needs to consider not only the catalytic efficiency of the individual enzyme but also its dynamics. The structure of the SaMTAN–FMA complex presented here clearly shows that structural data alone are insufficient to explain the wide range of binding affinities found for various transition-state analogues (Gutierrez et al., 2007 [triangle]) as the protein–ligand interactions observed in the crystal structures are comparable in almost all cases.

Figure 4
Sequence alignment of E. coli and selected pathogenic MTA/SAH nucleosidases. The primary sequences of MTAN from the following bacterial pathogens are aligned: Staphylococcus aureus, Escherichia coli, Staphylococcus epidermidis, Staphylococcus saprophyticus ...

Supplementary Material

PDB reference: 5′-methylthioadenosine/S-­adenosylhomocysteine nucleosidase, 3bl6, r3bl6sf


The authors thank Dr Feher and Dr Appelt for providing the expression vector for the SaMTAN, Dr Yuri Lobsanov for his help with data collection and Dr Yi-Min Shi and Li Zhang at the Advanced Protein Technology Centre (APTC) at The Hospital for Sick Children for their assistance with the mass spectrometry. This work was supported by an operating grant from the Canadian Institutes of Health Research (#43998) and a sponsored research agreement from Quorex Pharmaceuticals. PLH is the recipient of a Canada Research Chair; JEL was supported, fully or in part, by a Graduate Doctoral training award from CIHR and a studentship from the Ontario Student Opportunities Trust Fund and Hospital for Sick Children Foundation Student Scholarship Program; KKWS was supported, fully or in part, by a Master’s research award from the Natural Science and Engineering Research Council of Canada, a studentship from the Ontario Student Opportunities Trust Fund and Hospital for Sick Children Foundation Student Scholarship Program and a Canada Graduate scholarship from CIHR.


  • Brünger, A. T., Adams, P. D., Clore, G. M., DeLano, W. L., Gros, P., Grosse-Kunstleve, R. W., Jiang, J.-S., Kuszewski, J., Nilges, M., Pannu, N. S., Read, R. J., Rice, L. M., Simonson, T. & Warren, G. L. (1998). Acta Cryst. D54, 905–921. [PubMed]
  • Chen, X., Schauder, S., Potier, N., Van Dorsselaer, A., Pelczer, I., Bassler, B. L. & Hughson, F. M. (2002). Nature (London), 415, 545–549. [PubMed]
  • Combet, C., Blanchet, C., Geourjon, C. & Deléage, G. (2000). Trends Biochem. Sci.25, 147–150. [PubMed]
  • DeLano, W. L. (2002). The PyMOL Molecular Graphics System.
  • Duerre, J. A. (1962). J. Biol. Chem.237, 3737–3741.
  • Dunn, S. & Bryant, J. (1994). Phytochem. Anal.5, 286–293.
  • Gouet, P., Courcelle, E., Stuart, D. I. & Metoz, F. (1999). Bioinformatics, 15, 305–308. [PubMed]
  • Groot, H. de, de Groot, H. & Noll, T. (1985). Biochem. J.230, 255–260. [PubMed]
  • Gutierrez, J. A., Luo, M., Singh, V., Li, L., Brown, R. L., Norris, G. E., Evans, G. B., Furneaux, R. H., Tyler, P. C., Painter, G. F., Lenz, D. H. & Schramm, V. L. (2007). ACS Chem. Biol.2, 725–734. [PubMed]
  • Haba, G. de la & Cantoni, G. (1959). J. Biol. Chem.234, 603–608. [PubMed]
  • Humphrey, W., Dalke, A. & Schulten, K. (1996). J. Mol. Graph.14, 33–38. [PubMed]
  • Kleywegt, G. (1995). Jnt CCP4/ESF–EACBM Newsl. Protein Crystallogr.31, 45–50.
  • Krissinel, E. & Henrick, K. (2007). J. Mol. Biol.372, 774–797. [PubMed]
  • Lee, J. E., Cornell, K. A., Riscoe, M. K. & Howell, P. L. (2001). Structure, 9, 941–953. [PubMed]
  • Lee, J. E., Cornell, K. A., Riscoe, M. K. & Howell, P. L. (2003). J. Biol. Chem.278, 8761–8770. [PubMed]
  • Lee, J. E., Singh, V., Evans, G. B., Tyler, P. C., Furneaux, R. H., Cornell, K. A., Riscoe, M. K., Schramm, V. L. & Howell, P. L. (2005). J. Biol. Chem.280, 18274–18282. [PubMed]
  • Lee, J. E., Smith, G. D., Horvatin, C., Huang, D. J., Cornell, K. A., Riscoe, M. K. & Howell, P. L. (2005). J. Mol. Biol.352, 559–574. [PubMed]
  • Lovell, S. C., Davis, I. W., Arendall, W. B. III, de Bakker, P. I., Word, J. M., Prisant, M. G., Richardson, J. S. & Richardson, D. C. (2003). Proteins, 50, 437–450. [PubMed]
  • McRee, D. E. (1999). J. Struct. Biol.125, 156–165. [PubMed]
  • Markham, G. D., Norrby, P. O. & Bock, C. W. (2002). Biochemistry, 41, 7636–7646. [PubMed]
  • Miller, J. W., Nadeau, M. R., Smith, J., Smith, D. & Selhub, J. (1994). Biochem. J.298, 415–419. [PubMed]
  • Pajula, R. L. & Raina, A. (1979). FEBS Lett.99, 343–345. [PubMed]
  • Pegg, A. E. & Williams-Ashman, H. G. (1969). J. Biol. Chem.244, 682–693. [PubMed]
  • Pflugrath, J. W. (1999). Acta Cryst. D55, 1718–1725. [PubMed]
  • Raina, A., Tuomi, K. & Pajula, R. L. (1982). Biochem. J.204, 697–703. [PubMed]
  • Read, R. J. (1986). Acta Cryst. A42, 140–149.
  • Riscoe, M. K., Ferro, A. J. & Fitchen, J. H. (1989). Parasitol. Today, 5, 330–333. [PubMed]
  • Schauder, S., Shokat, K., Surette, M. G. & Bassler, B. L. (2001). Mol. Microbiol.41, 463–476. [PubMed]
  • Singh, V., Evans, G. B., Lenz, D. H., Mason, J. M., Clinch, K., Mee, S., Painter, G. F., Tyler, P. C., Furneaux, R. H., Lee, J. E., Howell, P. L. & Schramm, V. L. (2005). J. Biol. Chem.280, 18265–18273. [PubMed]
  • Singh, V., Lee, J. E., Nunez, S., Howell, P. L. & Schramm, V. L. (2005). Biochemistry, 44, 11647–11659. [PubMed]
  • Singh, V. & Schramm, V. L. (2007). J. Am. Chem. Soc.129, 2783–2795. [PMC free article] [PubMed]
  • Singh, V., Shi, W., Almo, S. C., Evans, G. B., Furneaux, R. H., Tyler, P. C., Painter, G. F., Lenz, D. H., Mee, S., Zheng, R. & Schramm, V. L. (2006). Biochemistry, 45, 12929–12941. [PMC free article] [PubMed]
  • Smith, G. (1993). PROFIT: A Program for Orienting One Protein Molecule Onto Another by a Least Squares Procedure. Hauptman–Woodward Medical Research Institute, Buffalo, USA.
  • Sufrin, J. R., Meshnick, S. R., Spiess, A. J., Garofalo-Hannan, J., Pan, X. Q. & Bacchi, C. J. (1995). Antimicrob. Agents Chemother.39, 2511–2515. [PMC free article] [PubMed]

Articles from Acta Crystallographica Section F: Structural Biology and Crystallization Communications are provided here courtesy of International Union of Crystallography