Crystallographic and molecular dynamics simulation analysis of Escherichia coli dihydroneopterin aldolase
- Jaroslaw Blaszczyk†1, 2, 3,
- Zhenwei Lu†2, 4,
- Yue Li2,
- Honggao Yan2Email author and
- Xinhua Ji1Email author
© Blaszczyk et al.; licensee BioMed Central Ltd. 2014
Received: 2 July 2014
Accepted: 26 August 2014
Published: 2 September 2014
Dihydroneopterin aldolase (DHNA) catalyzes the conversion of 7,8-dihydroneopterin to 6-hydroxymethyl-7,8-dihydropterin and also the epimerization of DHNP to 7,8-dihydromonapterin. Previously, we determined the crystal structure of Staphylococcus aureus DHNA (SaDHNA) in complex with the substrate analogue neopterin (NP). We also showed that Escherichia coli DHNA (EcDHNA) and SaDHNA have significantly different binding and catalytic properties by biochemical analysis. On the basis of these structural and functional data, we proposed a catalytic mechanism involving two proton wires.
To understand the structural basis for the biochemical differences and further investigate the catalytic mechanism of DHNA, we have determined the structure of EcDHNA complexed with NP at 1.07-Å resolution [PDB:2O90], built an atomic model of EcDHNA complexed with the substrate DHNP, and performed molecular dynamics (MD) simulation analysis of the substrate complex. EcDHNA has the same fold as SaDHNA and also forms an octamer that consists of two tetramers, but the packing of one tetramer with the other is significantly different between the two enzymes. Furthermore, the structures reveal significant differences in the vicinity of the active site, particularly in the loop that connects strands β3 and β4, mainly due to the substitution of nearby residues. The building of an atomic model of the complex of EcDHNA and the substrate DHNP and the MD simulation of the complex show that some of the hydrogen bonds between the substrate and the enzyme are persistent, whereas others are transient. The substrate binding model and MD simulation provide the molecular basis for the biochemical behaviors of the enzyme, including noncooperative substrate binding, indiscrimination of a pair of epimers as the substrates, proton wire switching during catalysis, and formation of epimerization product.
The EcDHNA and SaDHNA structures, each in complex with NP, reveal the basis for the biochemical differences between EcDHNA and SaDHNA. The atomic substrate binding model and MD simulation offer insights into substrate binding and catalysis by DHNA. The EcDHNA structure also affords an opportunity to develop antimicrobials specific for Gram-negative bacteria, as DHNAs from Gram-negative bacteria are highly homologous and E. coli is a representative of this class of bacteria.
Folate cofactors are essential for life . Mammals obtain folates from their diet because they cannot synthesize folates de novo but have an active transport system. In contrast, most microorganisms must synthesize folates de novo because they cannot take folates from their environments due to the lack of an active transport system . Therefore, the folate biosynthetic pathway has been one of the principal targets for developing antimicrobial agents [5–9]. Among the folate pathway enzymes, the four enzymes in the mid pathway are particularly attractive because they are absent in mammals: DHNA, 6-hydroxymethyl-7,8-dihydropterin pyrophosphokinase (HPPK), dihydropteroate synthase (DHPS), and dihydrofolate synthase (DHFS). DHPS is the target of sulfa drugs, the clinical use of which marks the beginning of the modern era of antimicrobial chemotherapy . The multiple targets afforded by this pathway also provide opportunities to develop antibiotics with synergetic effects. For example, in clinical use, sulfonamides, which target DHPS, are combined with trimethoprim, an antibiotic targeting DHFS, the last enzyme in the folate pathway .
Interestingly, DHNAs from Gram-positive and Gram-negative bacteria have some unique sequence motifs . The sequence identities between enzymes from Gram-positive bacteria range from 39% to 45% and those between Gram-negative bacteria are 49-91%, but the identities between Gram-positive and Gram-negative bacterial enzymes are <30% . Many differences between the amino acid sequences of DHNAs from Gram-positive and Gram-negative bacteria are at or near the active center. In accordance with the significant differences between their sequences, biochemical studies have shown that EcDHNA and SaDHNA have significantly different ligand binding and catalytic properties [11–13].
To date, crystal structures have been reported for DHNAs from Gram-positive bacteria S. aureus[14–16], Mycobacterium tuberculosis (MtDHNA) , Streptococcus pneumoniae, and the plant Arabidopsis thaliana. The active site of the enzyme has been identified by the protein-product (HP) structures of SaDHNA [PDB:2DHN]  and MtDHNA [PDB:1NBU] . The structural information about the critical interactions between DHNA and the trihydroxypropyl moiety of the substrate, which undergoes bond cleavage and formation, has been revealed by our structures of SaDHNA in complex with neopterin (NP) [PDB:2NM2] and with monapterin (MP) [PDB:2NM3], respectively . NP and MP are excellent inhibitors for SaDHNA, because the only difference between these inhibitors and the corresponding substrates is that the single bond between C7 and N8 in the substrates is replaced by a double bond in the inhibitors (Figure 1). The crystal structures of SaDHNA in complex with NP or MP have provided important insights into the catalytic mechanism of the enzyme .
No crystal structure has been reported for DHNAs from Gram-negative bacteria. Because E. coli is a representative of Gram-negative bacteria and EcDHNA has been well characterized biochemically , we have determined the crystal structures of EcDHNA in complex with the substrate analogue NP (EcDHNA:NP) [PDB:2O90]. Based on this crystal structure, we have built an atomic model of the enzyme in complex with the substrate DHNP (EcDHNA:DHNP) and performed molecular dynamics (MD) simulation of the enzyme:substrate complex. The results provide insights into the mechanism of DHNA catalysis, the structural basis of biochemical differences between SaDHNA and EcDHNA, and valuable information for structure-based design of novel antimicrobial agents.
Overall structure of the EcDHNA:NP complex
The EcDHNA:NP structure has been determined at 1.07-Å resolution. The asymmetric unit of the structure contains one DHNA polypeptide, one NP molecule, and 279 water molecules. Thus, the octamer of EcDHNA:NP contains eight identical active sites. Seven residues at the C-terminus (Asn116-Asn122) are not observed and thus presumably disordered. Met1 exhibits three conformations of equal probabilities; 20 residues (Ile3, Gln8, Ser10, Val17, Tyr18, Asp19, Lys27, Asp31, Glu33, Arg39, Ser62, Arg68, Leu82, Arg93, Ile94, Ser97, Pro99, Gly100, Ala101, and Glu113) assume two conformations.
Interactions between EcDHNA and NP
Comparison with the SaDHNA structure
Dynamic properties of the enzyme-substrate complex EcDHNA:DHNP
To assess whether the motions of individual protomers are independent, we calculated the RMSD of each protomer (Figure 6). The RMSD plots of protomers 3, 5, and 7 are similar, rising quickly to and staying at ~1 Å. The RMSD plots of protomers 6 and 8 are similar, with the RMSD values increasing in two stages, first rising to and staying at ~1 Å and then rising to and staying at ~2 Å. It is worthy to note that the transition between these two stages is different for the two protomers, with one much earlier than the other. The RMSD plots of the other three protomers are in between. The differences in the RMSD plots and their uncorrelated nature indicate that the motions of individual protomers are largely independent, in consistence with the uncooperative binding of NP and DHNP to the enzyme observed in biochemical experiments . Consequently, the analyses of the MD data are based on the behaviors of individual protomers rather than the octomer as a whole. The MD data are effectively composed of eight 27-ns trajectories, equivalent to an aggregate simulation time of 216 ns, and the results below are the average of the eight trajectories.
Interactions between the substrate DHNP and the enzyme
Hydrogen bonds between DHNP, the catalytic water, and DHNA revealed by molecular dynamics simulation
Leu72 - NH…O = C4
2.91 ± 0.14
WAT - OH… O = C4
2.94 ± 0.20
Glu73 - OE1…H - N3
2.79 ± 0.09
Glu73 - OE2…H - NC2
2.84 ± 0.13
Leu51′ - O…H - NC2
2.93 ± 0.15
Tyr53′ - NH…N1
3.33 ± 0.12
WAT - OH…N5
3.18 ± 0.19
Ser52′-OG…H - N8
3.03 ± 0.17
Glu21 - OE1…H - OC1′
2.69 ± 0.24
Val17 - NH…OC1′
2.98 ± 0.20
Lys98 - NZ… H - OC2′
2.95 ± 0.16
Tyr53′ - OH…OC2′
3.10 ± 0.19
Ala101 - O…H - OC3′
3.00 ± 0.24
Lys98 - NZH…WAT
2.98 ± 0.17
Insights into substrate binding and catalysis
DHNA is a special aldolase because it does not require a zinc ion or the formation of a Schiff base between the enzyme and the substrate for catalysis and generates an epimerization product at a significant rate. The hallmark of its catalytic power is general acid/base catalysis. Two proton wires have been proposed for DHNA catalysis . The first proton wire consists of a conserved Lys residue, a catalytic water molecule, and 2′-OH and N5 of the substrate DHNP. This proton wire is revealed by the crystal structure of SaDHNA in complex with NP [PDB:2NM2] and is required for the cleavage of the C1′-C2′ bond, which generates the enol intermediate. The importance of the conserved Lys residue, Lys98 in EcDHNA, for catalysis has been demonstrated by site-directed mutagenesis. The second proton wire consists of a conserved Tyr residue, Tyr53 in EcDHNA, the conserved Lys residue, the catalytic water, and C1′ and N5 of the reaction intermediate. This proton wire is proposed based on a mutagenesis study of the conserved Tyr residue of DHNA . The mutagenesis study has shown that the Tyr residue is not required for the generation of the reaction intermediate but for protonation of it to generate the product HP. The first proton wire is present in the crystal structure of EcDHNA in complex with NP and the atomic model of the complex of EcDHNA and the substrate DHNP reported here. The formation of the second proton wire needs significant local conformational adjustment. In consistence with this requirement, the segment of the protein containing Tyr53 (residues 38-53) is most flexible in the MD simulation with the highest RMSF. This flexibility may also allow the rotation of the product glycoaldehyde and generation of the epimerization product MP.
The MD simulation of the EcDHNA complex also shows that the substrate DHNP is anchored in the active site by four hydrogen bonds between the substrate (the groups at positions 2, 3, and 4) and the protein (backbone amides of Leu51′ and Leu72 and carboxylate of Glu73). This is manifested by the persistence of these hydrogen bonds, found with >95% of the snapshots of the MD simulation. Particularly, the carboxylate of Glu73 forms two hydrogen bonds, one with 1-NH and the other with 2-NH2, the former with a 100% of occurrence and the latter with a 99.7% of occurrence. Indeed, mutagenesis studies have shown that Glu73 is the most important residue for binding of the substrate DHNP, the inhibitor NP, and the product HP. In contrast, the hydrogen bonds between the trihydroxyl moiety of DHNP and the protein are transient, with occurrence in the range of 7.3 - 80.6%. In particular, the hydrogen bond between 3′-OH of the substrate and the backbone amide of Ala101 observed in the crystal structure is largely absent in the MD simulation, with an occurrence of only 7.3%. The transient nature of these hydrogen bonds, particularly the rare occurrence of the hydrogen bonding of 3′-OH, may allow the epimer DHMP serve as a good substrate as demonstrated by biochemical analysis .
The asynchronous motions of individual protomers observed in the MD simulation are also biochemically important. Although DHNA is a homooctomeric protein, it does not show cooperativity in binding substrate or other ligands. This lack of cooperativity is consistent with the asynchronous motions of individual protomers in the MD simulation.
Implications for structure-based drug design
Infectious diseases are the leading causes of death and the main causes of premature death (0-44 years) . Widespread and persistent antibiotic resistance has caused a worldwide health care crisis [5, 25, 26]. The crisis has been aggravated by the decisions by many major pharmaceutical companies to abandon or curtail their antibacterial programs for business reasons [27–29] and the fact that most new antibiotics are chemical modifications of existing antimicrobial agents . These compounds act against old targets and are therefore less effective in dealing with widespread antibiotic resistance. New targets for the development of novel antimicrobial agents are thus urgently needed for combating the antibiotic crisis.
DHNA is an attractive target for developing new antibiotics, because the enzyme is in a biosynthetic pathway proven effective in developing antibiotics and is absent in human. Sanders and coworkers have reported the development of inhibitors against SaDHNA . The fact that EcDHNA and SaDHNA have significant differences in their biochemical properties and active center structures, structure-based drug design must take into account the structure of EcDHNA for developing broad-spectrum antibiotics. On the other hand, DHNAs from Gram-negative bacteria are highly homologous, and so their active centers are expected to have very similar shapes and physicochemical characteristics. The structure of EcDHNA reported here offers an opportunity to develop antibiotics specific for Gram-negative bacteria.
The EcDHNA:NP structure has been determined at 1.07 Å, the highest resolution among all of the DHNA structures reported to date. This crystal structure of the inhibitor complex and the MD simulation of the substrate complex EcDHNA:DHNP have provided important insights into substrate binding and catalysis. The substrate DHNP is anchored in the active site via four persistent hydrogen bonds. The transient nature of hydrogen bonding between the trihydroxyl moiety and the protein and the rare occurrence of hydrogen bonding with 3′-hydroxyl allow the pair of epimers DHNP and DHMP to serve as good substrates. The high flexibility of segments of the protein, particularly residues 38-53, may permit the switching from the first to the second proton wire during catalysis and the generation of the epimerization product. The asynchronous motions of individual protomers are consistent with the noncooperative binding of the substrate by DHNA. Between EcDHNA and SaDHNA, the two best characterized DHNAs, there are three outstanding structural differences. First, the packing of the two tetramers in the EcDHNA octamer is significantly different from that in the SaDHNA octamer due to a single mutation. Residue Trp20 in EcDHNA is an Ala in SaDHNA. The indole rings of the Trp20 residue in one tetramer stack on those of the corresponding Trp residues in the other tetramer, stabilizing the EcDHNA octamer. Second, the active site structures of the two enzymes, especially in the β3-β4 loop, are significantly different due to the difference in amino acid sequences. In SaDHNA, the loop is 11 residues in length, including three prolines, whereas in EcDHNA, it has nine residues, containing only one proline. Third, the entrance to the active site is significantly different in the two enzymes. The combination of Leu19 and His53 in SaDHNA result in a wide open entrance, whereas the Tyr18 and Ser52 in EcDHNA largely block the entrance. These structural differences provide the basis for the biochemical differences between the S. aureus and E. coli enzyme that represents Gram-positive and Gram-negative DHNA, respectively.
Cloning, expression, and purification of EcDHNA have been reported previously . NP was purchased from the Schircks Laboratories. The crystals of EcDHNA:NP were obtained via co-crystallization using the hanging-drop technique at well-controlled room temperature (19 ± 1°C). The protein solution was mixed and incubated with the ligand prior to crystallization experiments. The drops contained equal volumes of protein and reservoir solutions. The protein solution contained 11 mg/mL protein and 25 mM NP in 10 mM Tris-HCl (pH 8.0). The well solution contained 4.0 M sodium formate (pH 7.0). Crystals reached the size of 0.25 × 0.35 × 0.60 mm in about three months.
X-ray data and refinement statistics for the EcDHNA:NP structure [PDB:2O90]
Resolution range (Å)
R scaling a
I ≥ 2σ(I)
Reflections used for refinement
Reflections used for Rfree
Number of least-squares parameters
Crystallographic R b
R free c
Number of protein atoms/average B factor (Å2)
Number of ligand atoms/average B factor (Å2)
Number of water oxygen atoms/average B factor (Å2)
RMSD from ideal geometry:
Bond distances (Å)
Angle distances (Å)
Most favored φ/ψ angles (%)
Disallowed φ/ψ angles (%)
The structure was solved by molecular replacement (MR) using the apo-DHNA structure [PDB:1DHN] as the search model after the solvent molecules were removed. The MR solutions were subjected to rigid body refinement, energy minimization, and grouped B-factor refinement followed by a difference Fourier synthesis, which revealed the locations of the ligand molecule.
The structure was refined using CNS  at the initial stage and SHELXL  till completion. Model building was carried out using the graphics package O . The structure was refined with anisotropic temperature factors for non-hydrogen atoms, except for regions of increased mobility where the temperature factors were refined isotropically. For atoms with anisotropic temperature factors, the hydrogen atoms were built at idealized positions, with assigned isotropic temperature parameters equal to 1.2 times the equivalent isotropic temperature parameters of their parent atoms. Positional parameters of ligands have been refined with geometric restrictions for bond lengths, bond angles and planarity. The geometry of the final structure was assessed using PROCHECK  and WHAT IF . The asymmetric unit of EcDHNA:NP contains residues 1-115 of the DHNA polypeptide chain (122 residues), one NP molecule, and 263 water molecules. The details of structure refinement and the statistics of the final structure are summarized in Table 2.
Atomic coordinates and structure factors have been deposited with the Protein Data Bank under accession codes 2O90.
Molecular dynamics simulations
The PMEMD module of the Amber molecular dynamics package (version 10)  and the Amber ff03 force field  were used for the MD simulation of the EcDHNA:DHNP complex. RESP charges for DHNP were derived by using the Antechamber module of Amber 10 with DHNP optimized and its molecular electrostatic potential calculated using the Gaussian program (version 03)  at the HF/6-31G* level. The starting coordinate was the crystal structure of EcDHNA:NP with NP replaced by DHNP. All the crystallographic water molecules were removed except the one which is hydrogen bonded to N5 of NP and the amino group of Lys98. The side chain of Lys98 was in the deprotonated form as deemed to be the active form. The protonation and tautomerization states of the histidine residues were assigned taking into account their hydrogen bond interactions in the crystal structure. The default protonation states were used for all of the other amino acids. The system was solvated by a periodic box of TIP3P water molecules that extended at least 12 Å from the protein atoms and neutralized by the addition of Na+ ions. The solvent and Na+ ions were subjected to energy minimization by 1000 steps steepest descent followed by 1500 steps conjugated gradient minimization while both the protein and DHNP were restrained harmonically with a force constant 50.0 kcal/mol/Å2. The MD simulation was run at constant volume for 200 ps to heat up the system from 0 to 300 K while the solutes were restrained in the same way as in the energy minimization, and then at constant temperature and pressure, which were regulated by Langevin dynamics and isotropic position scaling, respectively. The SHAKE algorithm was used to constrain all bond lengths involving hydrogen atoms, permitting a 2-fs time step . The Particle-Mesh-Ewald method was used to evaluate the contribution of long-range electrostatic interactions . A non-bonded pair list cut off 12.0 Å was used, and the list updated every 25 steps. Coordinates were saved every 2 ps. PCA of the MD trajectory was carried out using the program GROMACS 3.2 [42, 43]. The overall translational and rotational motions for all snapshots were removed, and the covariance matrix for all Cα atom fluctuations from their trajectory-averaged values was calculated. The conformational clustering was based on the RMSD for main-chain atoms using the average-linkage algorithm as implemented in the PTRAJ module of Amber 10 as previously described .
We thank Zbigniew Dauter for assistance during data collection and processing. The X-ray diffraction data were collected at the X9B beamline of National Synchrotron Light Source at Brookhaven National Laboratory. This research was supported by the NIH grant GM51901 (to HY) and the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research (to XJ).
- Mathis JB, Brown GM: The biosynthesis of folic acid. XI. Purification and properties of dihydroneopterin aldolase. J Biol Chem. 1970, 245: 3015-3025.PubMedGoogle Scholar
- Hauβmann C, Rohdich F, Schmidt E, Bacher A, Richter G: Biosynthesis of pteridines in Escherichia coli. Structural and mechanistic similarity of dihydroneopterin-triphosphate epimerase and dihydroneopterin aldolase. J Biol Chem. 1998, 273: 17418-17424. 10.1074/jbc.273.28.17418View ArticleGoogle Scholar
- Blakley RL, Benkovic SJ: Chemistry and Biochemistry of Folates. Folates and Pterins. Volume 1. Edited by: Blakley RL, Benkovic SJ. 1984, New York: John Wiley & Sons, Inc.Google Scholar
- Hitchings GH, Burchall JJ: Inhibition of folate biosynthesis and function as a basis for chemotherapy. Adv Enzymol Relat Areas Mol Biol. 1965, 27: 417-468.PubMedGoogle Scholar
- Walsh C: Where will new antibiotics come from?. Nat Rev Microbiol. 2003, 1: 65-70. 10.1038/nrmicro727View ArticlePubMedGoogle Scholar
- Bermingham A, Derrick JP: The folic acid biosynthesis pathway in bacteria: evaluation of potential for antibacterial drug discovery. Bioessays. 2002, 24: 637-648. 10.1002/bies.10114View ArticlePubMedGoogle Scholar
- Kompis IM, Islam K, Then RL: DNA and RNA synthesis: antifolates. Chem Rev. 2005, 105: 593-620. 10.1021/cr0301144View ArticlePubMedGoogle Scholar
- Nzila A: Inhibitors of de novo folate enzymes in Plasmodium falciparum. Drug Discov Today. 2006, 11: 939-944. 10.1016/j.drudis.2006.08.003View ArticlePubMedGoogle Scholar
- Nzila A: The past, present and future of antifolates in the treatment of Plasmodium falciparum infection. J Antimicrob Chemother. 2006, 57: 1043-1054. 10.1093/jac/dkl104View ArticlePubMedGoogle Scholar
- Zinner SH, Mayer KH: Sulfonamides and Trimethoprim. Principles and Practice of Infectious Diseases. Volume 1. Edited by: Mandell GL, Bennett JE, Dolin R. 2009, 475-485. Philadelphia: Churchill Livingstone, 6.Google Scholar
- Wang Y, Li Y, Wu Y, Yan H: Mechanism of dihydroneopterin aldolase. NMR, equilibrium and transient kinetic studies of the Staphylococcus aureus and Escherichia coli enzymes. FEBS J. 2007, 274: 2240-2252. 10.1111/j.1742-4658.2007.05761.xView ArticlePubMedGoogle Scholar
- Wang Y, Li Y, Yan HG: Mechanism of dihydroneopterin aldolase: Functional roles of the conserved active site glutamate and lysine residues. Biochemistry. 2006, 45: 15232-15239. 10.1021/bi060949jPubMed CentralView ArticlePubMedGoogle Scholar
- Wang Y, Scherperel G, Roberts KD, Jones AD, Reid GE, Yan H: A point mutation converts dihydroneopterin aldolase to a cofactor-independent oxygenase. J Am Chem Soc. 2006, 128: 13216-13223. 10.1021/ja063455iView ArticlePubMedGoogle Scholar
- Hennig M, D’Arcy A, Hampele IC, Page MG, Oefner C, Dale GE: Crystal structure and reaction mechanism of 7, 8-dihydroneopterin aldolase from Staphylococcus aureus. Nat Struct Biol. 1998, 5: 357-362. 10.1038/nsb0598-357View ArticlePubMedGoogle Scholar
- Sanders WJ, Nienaber VL, Lerner CG, McCall JO, Merrick SM, Swanson SJ, Harlan JE, Stoll VS, Stamper GF, Betz SF, Condroski KR, Meadows RP, Severin JM, Walter KA, Magdalinos P, Jakob CG, Wagner R, Beutel BA: Discovery of potent inhibitors of dihydroneopterin aldolase using CrystaLEAD high-throughput X-ray crystallographic screening and structure-directed lead optimization. J Med Chem. 2004, 47: 1709-1718. 10.1021/jm030497yView ArticlePubMedGoogle Scholar
- Blaszczyk J, Li Y, Gan J, Yan H, Ji X: Structural basis for the aldolase and epimerase activities of Staphylococcus aureus dihydroneopterin aldolase. J Mol Biol. 2007, 368: 161-169. 10.1016/j.jmb.2007.02.009PubMed CentralView ArticlePubMedGoogle Scholar
- Goulding CW, Apostol MI, Sawaya MR, Phillips M, Parseghian A, Eisenberg D: Regulation by oligomerization in a mycobacterial folate biosynthetic enzyme. J Mol Biol. 2005, 349: 61-72. 10.1016/j.jmb.2005.03.023View ArticlePubMedGoogle Scholar
- Garçon A, Levy C, Derrick JP: Crystal structure of the bifunctional dihydroneopterin aldolase/6-hydroxymethyl-7, 8-dihydropterin pyrophosphokinase from Streptococcus pneumoniae. J Mol Biol. 2006, 360: 644-653. 10.1016/j.jmb.2006.05.038View ArticlePubMedGoogle Scholar
- Bauer S, Schott AK, Illarionova V, Bacher A, Huber R, Fischer M: Biosynthesis of tetrahydrofolate in plants: crystal structure of 7, 8-dihydroneopterin aldolase from Arabidopsis thaliana reveals a novel aldolase class. J Mol Biol. 2004, 339: 967-979. 10.1016/j.jmb.2004.04.034View ArticlePubMedGoogle Scholar
- Kraulis PJ: MOLSCRIPT: a program to produce both detailed and schematic plots of protein structures. J Appl Crystallogr. 1991, 24: 946-950. 10.1107/S0021889891004399. 10.1107/S0021889891004399View ArticleGoogle Scholar
- Bacon DJ, Anderson WF: A fast algorithm for rendering space-filling molecule pictures. J Mol Graph. 1988, 6: 219-220.View ArticleGoogle Scholar
- Merritt EA, Bacon DJ: Raster3D: photorealistic molecular graphics. Methods Enzymol. 1997, 277: 505-524.View ArticlePubMedGoogle Scholar
- Yao LS, Yan HG, Cukier RI: Mechanism of dihydroneopterin aldolase: a molecular dynamics study of the apo enzyme and its product complex. J Phys Chem B. 2006, 110: 1443-1456. 10.1021/jp054854nView ArticlePubMedGoogle Scholar
- , : Report on Infectious Diseases: Removing Obstacles to Healthy Development. 1999, Geneva: World Health Organization.Google Scholar
- Cohen ML: Changing patterns of infectious disease. Nature. 2000, 406: 762-767. 10.1038/35021206View ArticlePubMedGoogle Scholar
- Murray BE: Antibiotic resistance. Adv Intern Med. 1997, 42: 339-367.PubMedGoogle Scholar
- Projan SJ: Why is big Pharma getting out of antibacterial drug discovery?. Curr Opin Microbiol. 2003, 6: 427-430. 10.1016/j.mib.2003.08.003View ArticlePubMedGoogle Scholar
- Spellberg B, Powers JH, Brass EP, Miller LG, Edwards JE: Trends in antimicrobial drug development: implications for the future. Clin Infect Dis. 2004, 38: 1279-1286. 10.1086/420937View ArticlePubMedGoogle Scholar
- Wenzel RP: The antibiotic pipeline-challenges, costs, and values. N Engl J Med. 2004, 351: 523-526. 10.1056/NEJMp048093View ArticlePubMedGoogle Scholar
- Barrett CT, Barrett JF: Antibacterials: are the new entries enough to deal with the emerging resistance problems?. Curr Opin Biotechnol. 2003, 14: 621-626. 10.1016/j.copbio.2003.10.003View ArticlePubMedGoogle Scholar
- Otwinowski Z, Minor W: Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 1997, 276: 307-326.View ArticleGoogle Scholar
- Brünger AT, Adams PD, Clore GM, DeLano WL, Gros P, Grosse-Kunstleve RW, Jiang JS, Kuszewski J, Nilges M, Pannu NS, Read RJ, Rice LM, Simonson T, Warren GL: Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr D. 1998, 54: 905-921.View ArticlePubMedGoogle Scholar
- Sheldrick GM, Schneider TR: SHELXL: high-resolution refinement. Methods Enzymol. 1997, 277: 319-343.View ArticlePubMedGoogle Scholar
- Jones TA, Kjeldgaard M: Electron-density map interpretation. Methods Enzymol. 1997, 277: 173-208.View ArticlePubMedGoogle Scholar
- Laskowski RA, MacArthur MW, Moss DS, Thornton JM: PROCHECK: a program to check the stereochemical quality of protein structures. J Appl Crystallogr. 1993, 26: 283-291. 10.1107/S0021889892009944. 10.1107/S0021889892009944View ArticleGoogle Scholar
- Vriend G: WHAT IF: a molecular modeling and drug design program. J Mol Graph. 1990, 8: 52-56. 29. 10.1016/0263-7855(90)80070-VView ArticlePubMedGoogle Scholar
- Case DA, Cheatham TE, Darden T, Gohlke H, Luo R, Merz KM, Onufriev A, Simmerling C, Wang B, Woods RJ: The Amber biomolecular simulation programs. J Comput Chem. 2005, 26: 1668-1688. 10.1002/jcc.20290PubMed CentralView ArticlePubMedGoogle Scholar
- Duan Y, Wu C, Chowdhury S, Lee MC, Xiong GM, Zhang W, Yang R, Cieplak P, Luo R, Lee T, Caldwell J, Wang JM, Kollman PA: A point-charge force field for molecular mechanics simulations of proteins based on condensed-phase quantum mechanical calculations. J Comput Chem. 2003, 24: 1999-2012. 10.1002/jcc.10349View ArticlePubMedGoogle Scholar
- Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Montgomery JA, Vreven T, Kudin KN, Burant JC, Millam JM, Iyengar SS, Tomasi J, Barone V, Mennucci B, Cossi M, Scalmani G, Rega N, Petersson GA, Nakatsuji H, Hada M, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H: Gaussian 03, Revision B.05. 2003, Wallingford CT: Gaussian, Inc.Google Scholar
- Ryckaert JP, Ciccotti G, Berendsen HJC: Numerical-integration of cartesian equations of motion of a system with constraints - molecular-dynamics of N-alkanes. J Comput Phys. 1977, 23: 327-341. 10.1016/0021-9991(77)90098-5. 10.1016/0021-9991(77)90098-5View ArticleGoogle Scholar
- Essmann U, Perera L, Berkowitz ML, Darden T, Lee H, Pedersen LG: A smooth particle mesh Ewald method. J Chem Phys. 1995, 103: 8577-8593. 10.1063/1.470117. 10.1063/1.470117View ArticleGoogle Scholar
- Berendsen HJC, Vanderspoel D, Vandrunen R: GROMACS - A message-passing parallel molecular-dynamics implementation. Comput Phys Commun. 1995, 91: 43-56. 10.1016/0010-4655(95)00042-E. 10.1016/0010-4655(95)00042-EView ArticleGoogle Scholar
- Van der Spoel D, Lindahl E, Hess B, Groenhof G, Mark AE, Berendsen HJC: GROMACS: Fast, flexible, and free. J Comput Chem. 2005, 26: 1701-1718. 10.1002/jcc.20291View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.