Atomic Resolution Structures and Solution Behavior of Enzyme-Substrate Complexes of Enterobacter cloacae PB2 Pentaerythritol Tetranitrate Reductase
2004; Elsevier BV; Volume: 279; Issue: 29 Linguagem: Inglês
10.1074/jbc.m403541200
ISSN1083-351X
AutoresHuma Khan, T. Barna, Richard J. Harris, Neil C. Bruce, Igor Barsukov, Andrew W. Munro, P.C.E. Moody, Nigel S. Scrutton,
Tópico(s)Metal complexes synthesis and properties
ResumoThe structure of pentaerythritol tetranitrate (PETN) reductase in complex with the nitroaromatic substrate picric acid determined previously at 1.55 Å resolution indicated additional electron density between the indole ring of residue Trp-102 and the nitro group at C-6 of picrate. The data suggested the presence of an unusual bond between substrate and the tryptophan side chain. Herein, we have extended the resolution of the PETN reductase-picric acid complex to 0.9 Å. This high-resolution analysis indicates that the active site is partially occupied with picric acid and that the anomalous density seen in the original study is attributed to the population of multiple conformational states of Trp-102 and not a formal covalent bond between the indole ring of Trp-102 and picric acid. The significance of any interaction between Trp-102 and nitroaromatic substrates was probed further in solution and crystal complexes with wild-type and mutant (W102Y and W102F) enzymes. Unlike with wild-type enzyme, in the crystalline form picric acid was bound at full occupancy in the mutant enzymes, and there was no evidence for multiple conformations of active site residues. Solution studies indicate tighter binding of picric acid in the active sites of the W102Y and W102F enzymes. Mutation of Trp-102 does not impair significantly enzyme reduction by NADPH, but the kinetics of decay of the hydride-Meisenheimer complex are accelerated in the mutant enzymes. The data reveal that decay of the hydride-Meisenheimer complex is enzyme catalyzed and that the final distribution of reaction products for the mutant enzymes is substantially different from wild-type enzyme. Implications for the mechanism of high explosive degradation by PETN reductase are discussed. The structure of pentaerythritol tetranitrate (PETN) reductase in complex with the nitroaromatic substrate picric acid determined previously at 1.55 Å resolution indicated additional electron density between the indole ring of residue Trp-102 and the nitro group at C-6 of picrate. The data suggested the presence of an unusual bond between substrate and the tryptophan side chain. Herein, we have extended the resolution of the PETN reductase-picric acid complex to 0.9 Å. This high-resolution analysis indicates that the active site is partially occupied with picric acid and that the anomalous density seen in the original study is attributed to the population of multiple conformational states of Trp-102 and not a formal covalent bond between the indole ring of Trp-102 and picric acid. The significance of any interaction between Trp-102 and nitroaromatic substrates was probed further in solution and crystal complexes with wild-type and mutant (W102Y and W102F) enzymes. Unlike with wild-type enzyme, in the crystalline form picric acid was bound at full occupancy in the mutant enzymes, and there was no evidence for multiple conformations of active site residues. Solution studies indicate tighter binding of picric acid in the active sites of the W102Y and W102F enzymes. Mutation of Trp-102 does not impair significantly enzyme reduction by NADPH, but the kinetics of decay of the hydride-Meisenheimer complex are accelerated in the mutant enzymes. The data reveal that decay of the hydride-Meisenheimer complex is enzyme catalyzed and that the final distribution of reaction products for the mutant enzymes is substantially different from wild-type enzyme. Implications for the mechanism of high explosive degradation by PETN reductase are discussed. Pentaerythritol tetranitrate (PETN) 1The abbreviations used are: PETN, pentaerythritol tetranitrate; TNT, trinitrotoluene; GTN, glycerol trinitrate; OYE, old yellow enzyme; 2,4-DNP, 2,4-dinitrophenol. reductase is a member of the old yellow enzyme (OYE) family of flavoproteins and was purified from a strain of Enterobacter cloacae (strain PB2) originally isolated on the basis of its ability to utilize nitrate ester explosives such as PETN and glycerol trinitrate (GTN) as sole nitrogen source (1Binks P.R. French C.E. Nicklin S. Bruce N.C. Appl. Environ. Microbiol. 1996; 62: 1214-1219Crossref PubMed Google Scholar). The structure of PETN reductase (2Barna T.M. Khan H. Bruce N.C. Barsukov I. Scrutton N.S. Moody P.C. J. Mol. Biol. 2001; 310: 433-447Crossref PubMed Scopus (87) Google Scholar) is similar to that of OYE (3Fox K.M. Karplus P.A. Structure. 1994; 2: 1089-1105Abstract Full Text Full Text PDF PubMed Scopus (227) Google Scholar) and morphinone reductase (4Barna T. Messiha H.L. Petosa C. Bruce N.C. Scrutton N.S. Moody P.C. J. Biol. Chem. 2002; 277: 30976-30983Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar), confirming the close evolutionary relationship with OYE and other FMN-dependent flavoprotein oxidoreductases inferred from sequence analysis of the genes encoding these enzymes (5French C.E. Bruce N.C. Biochem. J. 1995; 312: 671-678Crossref PubMed Scopus (68) Google Scholar, 6French C.E. Nicklin S. Bruce N.C. J. Bacteriol. 1996; 178: 6623-6627Crossref PubMed Google Scholar). Consistent with this close relationship is the ability of the OYE family of enzymes to reduce a variety of cyclic enones, including 2-cyclohexenone and steroids. Some steroids act as substrates, whereas others are potent inhibitors of these enzymes. PETN reductase, and the related orthologues from strains of Pseudomonas (7Blehert D.S. Fox B.G. Chambliss G.H. J. Bacteriol. 1999; 181: 6254-6263Crossref PubMed Google Scholar) and Agrobacterium (8Snape J.R. Walkley N.A. Morby A.P. Nicklin S. White G.F. J. Bacteriol. 1997; 179: 7796-7802Crossref PubMed Google Scholar), show reactivity against explosive substrates. PETN reductase degrades major classes of explosive, including nitroaromatic compounds (e.g. trinitrotoluene TNT) and nitrate esters (GTN and PETN) (9French C.E. Nicklin S. Bruce N.C. Appl. Environ. Microbiol. 1998; 64: 2864-2868Crossref PubMed Google Scholar, 10Williams R.E. Rathbone D. Bruce N.C. Scrutton N.S. Moody P.C.E. Nicklin S. Ghisla S. Kroneck P. Macheroux P. Sund H. Flavins and Flavoproteins. Rudolf Weber, Berlin1999: 663-666Google Scholar, 11Williams R. Rathbone D. Moody P. Scrutton N. Bruce N. Biochem. Soc. Symp. 2001; 8: 143-153Google Scholar). Degradation of TNT involves reductive hydride addition to the aromatic nucleus (Fig. 1). In the case of members of the old yellow enzyme family of enzymes that are closely related to PETN reductase, the products of TNT reduction have been shown to result from both reductive hydride addition at the aromatic nucleus and also nitro group reduction in two competing pathways in the oxidative half-reaction of the enzyme (9French C.E. Nicklin S. Bruce N.C. Appl. Environ. Microbiol. 1998; 64: 2864-2868Crossref PubMed Google Scholar, 12Pak J.W. Knoke K.L. Noguera D.R. Fox B.G. Chambliss G.H. Appl. Environ. Microbiol. 2000; 66: 4742-4750Crossref PubMed Scopus (127) Google Scholar, 13Williams R.E. Rathbone D.A. Scrutton N.S. Bruce N.C. Appl. Environ. Microbiol. 2004; (in press)Google Scholar). The reaction of PETN reductase comprises two half-reactions: in the reductive half-reaction, enzyme is reduced by NADPH to yield the dihydroquinone form of the enzyme-bound FMN, and in the oxidative half-reaction the flavin is oxidized by the nitro-containing explosive substrates or cyclic enone substrates. A detailed kinetic mechanism based on stopped-flow data has been proposed (14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar), and recently hydride transfer in the reductive half-reaction was shown to proceed by quantum mechanical tunneling (15Basran J. Harris R.J. Sutcliffe M.J. Scrutton N.S. J. Biol. Chem. 2003; 278: 43973-43982Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar). The structures of PETN reductase in complex with steroid substrates and inhibitors have been determined (2Barna T.M. Khan H. Bruce N.C. Barsukov I. Scrutton N.S. Moody P.C. J. Mol. Biol. 2001; 310: 433-447Crossref PubMed Scopus (87) Google Scholar), as have complexes of the enzyme with 2-cyclohexenone, the inhibitor 2,4-dinitrophenol (2,4-DNP), and the substrates TNT and picric acid (14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar). The 1.55 Å structure of the enzyme in complex with picric acid indicated additional electron density between the indole ring of residue Trp-102 and the nitro group at C-6 of picrate, which at this resolution suggested the presence of an unusual bond between substrate and the tryptophan side chain (14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar). The formation of such an unusual bond would clearly have profound implications for the mechanism of picric acid and TNT reduction in the enzyme. Thus, to address the origin of the additional electron density observed between residue Trp-102 and picric acid, in this paper we have extended the resolution of the wild-type enzyme in complex with picric acid to 0.9 Å. Our high resolution studies reveal that the additional density observed in the 1.55 Å structure of the wild-type-picric acid complex is attributed to the population of multiple conformational states of Trp-102 and not a formal covalent bond between the indole ring of Trp-102 and picric acid. To understand further the reaction mechanism of PETN reductase, and in particular the role of Trp-102, we have mutated this residue to phenylalanine (W102F) and tyrosine (W102Y) and we describe detailed kinetic studies of these mutant PETN reductases. Solution studies indicate that exchange of Trp-102 for tyrosine and phenylalanine favors the binding of picric acid through the removal of steric constraints, an aspect also confirmed by determination of the crystal structures of these enzymes bound to picric acid. Kinetic and NMR studies with the mutant and wild-type enzymes indicate that removal of the indole side chain also influences the distribution of the reaction products with the nitroaromatic substrate TNT. The role of Trp-102 in catalysis by PETN reductase and related enzymes, and implications for the partitioning of reaction pathways in nitroaromatic reduction, are discussed. Chemicals—All chemicals were of analytical grade where possible. Complex bacteriological media were from Unipath, and all media were prepared as described in Sambrook et al. (16Sambrook J. Fritsch E. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1989Google Scholar). Mimetic Orange 2 affinity chromatography resin was from Affinity Chromatography Ltd. Q-Sepharose resin was from Amersham Biosciences. NADPH, glucose 6-phosphate dehydrogenase, glucose 6-phosphate, benzyl viologen, methyl viologen, 2-hydroxy-1,4-naphthaquinone, phenazine methosulfate and 2,4-DNP were from Sigma. 2-Cyclohexenone was from Acros Organics. Dr. S. Nicklin (UK Defense and Evaluation Research Agency) supplied TNT, GTN, and picric acid. The following extinction coefficients were used to calculate the concentration of substrates and enzyme: NADPH (ϵ340 = 6.22 × 103m-1 cm-1); PETN reductase (ϵ464 = 11.3 × 103m-1 cm-1). Stock solutions of TNT (600 mm) were made up in acetone. Dilutions were then made into 50 mm potassium phosphate buffer, pH 7.0, and the acetone concentration was maintained at 1% (v/v). The presence of acetone in buffers at 1% (v/v) was shown not to affect enzyme activity. Mutagenesis and Protein Purification—Site-directed mutagenesis of Trp-102 was achieved using the QuikChange mutagenesis method (Stratagene) and the following oligonucleotides (and their complementary sequences): 5′-CGG TTC AGC TGT TTC ACA CCG GTC G-3′ (W102F forward primer), 5′-CGA CCG GTG TGA AAC AGC TGA ACC G-3′ (W102F, reverse primer), 5′-CGG TTC AGC TGT ATC ACA CCG GTC G-3′ (W102Y, forward primer), and 5′-CGA CCG GTG TGA TAC AGC TGA ACC G-3′(W102Y, reverse primer). Plasmid pONR1 (6French C.E. Nicklin S. Bruce N.C. J. Bacteriol. 1996; 178: 6623-6627Crossref PubMed Google Scholar) was used as template for mutagenesis reactions. All mutant genes were completely sequenced to ensure that spurious changes had not arisen during the mutagenesis reaction. The expression and purification of the wild-type and mutant PETN reductase enzymes was as described previously for wild-type enzyme (6French C.E. Nicklin S. Bruce N.C. J. Bacteriol. 1996; 178: 6623-6627Crossref PubMed Google Scholar). Redox Potentiometry Ligand Binding and Kinetic Analyses—Redox titrations and the determination of redox potential of the enzyme-bound FMN were performed as described previously for wild-type PETN reductase (14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar). 2In Ref. 14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar the midpoint reduction potential for the concerted two electron reduction of wild-type PETN reductase is reported as -195 ± 5 mV. We have shown subsequently that this value is incorrect because of incorrect calibration of the electrode used in potentiometric analysis. Correct values for wild-type PETN reductase (-267 ± 5 mV) and the W102Y (-236 ± 5 mV) and W102F (-241 ± 5 mV) enzymes were determined and reported herein. The spectral changes occurring during redox titration of the wild-type enzyme were identical to those reported in Ref. 14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar; similar spectral changes were also observed for the W102Y and W102F enzymes. Ligand binding studies were also as performed previously with wild-type PETN reductase (14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar), relying on the perturbation of the flavin electronic absorption spectrum on binding ligand in the active site of PETN reductase. Data were collected in the UV-visible region (250–600 nm), and the absorption at 518 nm plotted as a function of ligand concentration. Data in the plots were analyzed by fitting to Equation 1,ΔA=ΔAmax2ET[(LT+ET+Kd)−((LT+ET+Kd)2−(4LTET))0.5](Eq. 1) where ΔAmax is the maximum absorption change at 518 nm, LT is the total ligand concentration and ET the total enzyme concentration. Rapid reaction kinetic experiments using single wavelength absorption and photodiode array detection were performed using an Applied Photophysics SF.17MV stopped-flow instrument contained within an anaerobic glove box as described previously for wild-type PETN reductase (14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar). Multiple Turnover Studies and NMR Analysis of Reaction Products— Multiple turnover studies were performed under anaerobic conditions and the reaction progress monitored by absorption spectroscopy. The reaction mix (total volume, 1 ml) comprised 0.2 μm PETN reductase, 30 μm NADPH, and 100 μm TNT contained in 50 mm potassium phosphate buffer, pH 7.0, and reactions were performed at 25 °C. A NADPH-generating system comprising 10 mm glucose 6-phosphate, and 1 unit of glucose-6-phosphate dehydrogenase was also included in the reaction mix. UV-visible spectra were recorded using a Jasco V530 spectrophotometer contained within a Belle Technology anaerobic glove box. Multiple turnover studies were also performed directly in the NMR instrument. Reference one- and two-dimensional COSY proton spectra were collected for the reaction components. The reaction mixture of 1 mm TNT, 0.2 μm PETN reductase, 1.5 mm NADPH, in the D2O buffer (50 mm potassium phosphate, pH 7.0) was transferred into the NMR tube under anaerobic conditions, and a series of one-dimensional NMR spectra were collected at 5-min intervals. Upon completion of the reaction, as evidenced by the absence of changes in the NMR spectra, a two-dimensional COSY spectrum was acquired. The dead time of the reaction between the mixing of the components and the acquisition of the first one-dimensional spectrum was ∼10 min. All spectra were acquired at 600 MHz on a Bruker DRX600 spectrometer at 25 °C. The spectra were processed and analyzed using XWINNMR 3.5 software (Bruker), and were referenced to the external DSS standard at 0.000 ppm. Crystallography—Crystals of wild-type, W102F, and W102Y PETN reductase in complex with picric acid were grown in the manner described previously for wild-type PETN reductase (2Barna T.M. Khan H. Bruce N.C. Barsukov I. Scrutton N.S. Moody P.C. J. Mol. Biol. 2001; 310: 433-447Crossref PubMed Scopus (87) Google Scholar, 14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar). The crystals were isomorphous with those previously described, having space group P212121 and one molecule per asymmetric unit. Data for the wild-type and W102Y mutants were collected at the Daresbury Synchrotron Radiation Source (Daresbury, UK) and data for the W102F enzyme were collected at the European Synchrotron Radiation Facility (Grenoble, France). The data were measured and reduced with the HKL suite (17Otwinowski Z. Minor W. Methods Enzymol. 1997; 276: 307-326Crossref PubMed Scopus (38570) Google Scholar). Data collection and processing statistics are given in Table I. The mutant structures were refined using CNS (18Brunger A. Adams P. Clore G. DeLano W. Gros P. Grosse-Kunstleve R. Jiang J.-S. Kuszewski J. Nilges M. Pannu N. Read R. Rice L. Simonson T. Warren G. Acta Crystallogr. Sect. D. 1998; 54: 905-929Crossref PubMed Scopus (16967) Google Scholar) and Refmac5 (19Murshudov G. Vagin A. Dodson E. Acta Crystallogr. Sect. D. 1997; 53: 240-255Crossref PubMed Scopus (13869) Google Scholar), but in the refinement of the wild-type enzyme SHELXL (20Sheldrik G.M. Schnieder T.R. Carter C.W.J. Sweet R.M. Macromolecular Crystallography, Part B. 277. Academic Press, London1997: 319-343Google Scholar) was also used. Xtalview (21McRee D. J. Mol. Graphics. 1992; 10: 44-46Crossref Google Scholar) was used throughout for the display of electron density and model fitting. Final refinement statistics are given in Table I. The refined structures and diffraction data for the wild-type, W102F, and W012F have been deposited with the Protein Data Bank with accession codes 1vyr, 1vyp, and 1vys, respectively.Table IData collection and refinement statistics for PETN reductaseWild typeW102YW102FTotal observations998,70978,865332,712Unique reflections254,70429,34091,259Resolution (Å)aValues in parentheses for the outer bin0.90 (0.94–0.90)1.8 (1.86–1.80)1.27 (1.32–1.27)Completeness (%)aValues in parentheses for the outer bin94.8 (65.7)89.9 (92.7)93.8 (90.4)Rmerge (%)aValues in parentheses for the outer bin8.7 (44.0)5.2 (22.4)5.5 (26.0)I/sig(I)29.7 (2.4)10.3 (2.2)17.1 (2.5)Rwork13.2 (32.6)14.2 (19.3)13.3 (17.3)Rfree14.3 (30.9)19.9 (28.1)15.3 (22.3)Root mean square deviations from idealBond lengths (Å)0.0120.0220.014Bond angles (°)1.621.871.48a Values in parentheses for the outer bin Open table in a new tab Structure of PETN Reductase Bound to Picric Acid at High Resolution—Our previous structural studies of the complex formed between PETN reductase and picric acid at 1.55 Å resolution indicated the presence of additional density between the indole side chain of Trp-102 and the C-6 nitro group of picric acid (14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar). At subatomic resolution it becomes clear that this density does not represent a covalent bond between the nitroaromatic substrate and protein, but indicates that there are at least two conformations of the side chain of Trp-102 (Fig. 2). One of these conformations overlaps with bound picric acid. The occupancy of the picric acid and the non-overlapping conformer of Trp-102 is estimated at 33%. The electron density of the 1.55 Å structure is compared with that of the resolved structure in Fig. 2. Comparison with the structures of PETN reductase in complex with smaller ligands such as 2,4-dinitrophenol (14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar) suggests that, upon binding picric acid, the side chain of Trp-102 moves to accommodate the bulky nitro group at C-6. The relatively low occupancy of the picric acid leads to an apparent connection between the electron densities even at near-atomic resolution. The electron density of the minor conformation was initially interpreted as evidence for the lack of planarity of the indole side chain of Trp-102, which would be expected upon formation of a link between picric acid and the indole side chain but, as demonstrated herein, extension to high resolution indicates that this is not the case. The multiple conformations of Trp-102 have a further effect on the side chain of Thr-104, which is seen to adopt two conformations. Tyr-68, which is also involved in formation of the active site is also seen to adopt two conformations about chi-2; these multiple conformations are seen to extend as far as the main chain of Gly-67 (Fig. 3). The side chain of Ser-132, distal to the active site from Tyr-68, is also seen to have two conformations. There are also other regions of the enzyme away from the active site where multiple conformations are found, which is consistent with observations made with other structures reported at subatomic resolution (e.g. Ref. 22Lario P.I. Sampson N. Vrielink A. J. Mol. Biol. 2003; 326: 1635-1650Crossref PubMed Scopus (108) Google Scholar). In particular, methionine and proline show as mixed conformational populations. There is also evidence for the radiation-induced decarboxylation of some amino acid residue side chains and the carboxyl-terminal group as reported by Ravelli and McSweeny (23Ravelli R.B. McSweeney S.M. Structure Fold Des. 2000; 8: 315-328Abstract Full Text Full Text PDF Scopus (332) Google Scholar); deamination of some side chains is also seen at a lower level. Crystallographic Structures of the W102Y and W102F Enzyme-Ligand Complexes and Ligand Binding to W102Y and W102F Enzymes—In contrast to the multiple conformations found in the active site of the wild-type enzyme, the W102F and W102Y mutant enzymes, where a potential steric clash upon picrate binding has been removed by engineering, show no evidence for multiple conformations of the residues in the active site. Furthermore, the picric acid appears to fully occupy the active site. The active sites of the mutants are shown in Fig. 3B. The mutations at residue 102 have not introduced any other significant alteration. The structures suggest that the nitro group at C-6 of picric acid is accommodated without steric clash, which should favor binding of picric acid (and by inference TNT which is likely to bind in a similar mode, Ref. 14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar) in the active site of PETN reductase. The binding of nitroaromatic compounds (picric acid and TNT) to wild-type and mutant forms of PETN reductase was investigated by perturbation of the electronic absorption spectrum of the enzyme-bound FMN. Equilibrium binding measurements with wild-type PETN reductase have been reported (14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar), and an identical approach was used with the mutant enzymes. Addition of picric acid and 2,4-DNP to the mutant PETN reductase enzymes elicited changes in the electronic absorption spectrum with well defined isosbestic points (Fig. 4) and enzyme-ligand dissociation constants were calculated by fitting data obtained at 518 nm to Equation 1. Enzyme-ligand dissociation constants are collected in Table II. With picric acid, the dissociation constants indicate tighter binding of the ligand to the mutant enzymes compared with wild-type PETN reductase. The wild-type and mutant enzymes have similar affinities for 2,4-DNP. These observations are consistent with the crystallographic data, which indicate a steric interaction between the side chain of Trp-102 and one of the nitro groups of picric acid, but not 2,4-DNP. Titrations were also performed with progesterone, an inhibitor of PETN reductase that also elicits spectral changes on binding in the active site (14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar). In this case, the dissociation constant for the progesterone-enzyme complex is not affected by mutation, which is consistent with the lack of interaction between Trp-102 and progesterone in the crystal structure of the steroid-enzyme complex (2Barna T.M. Khan H. Bruce N.C. Barsukov I. Scrutton N.S. Moody P.C. J. Mol. Biol. 2001; 310: 433-447Crossref PubMed Scopus (87) Google Scholar). As with wild-type PETN reductase (14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar), the addition of TNT to the W102Y and W102F enzymes did not elicit spectral change, suggesting weak binding of this ligand is attributed to poor interaction of the methyl group of TNT with His-181 and His-184. These latter residues form strong interactions with the hydroxy group of picric acid, which is replaced by a methyl group in TNT.Table IICalculated dissociation constants for wild-type, W102F, and W102Y PETN reductases Wild-type PETN reductase and W102F and W102Y mutant enzymes (10 μm) were each titrated with picric acid, progesterone, and 2,4-DNP in 50 mm sodium phosphate buffer, pH 7.0, 25 °C. Data at 518 nm were plotted as a function of ligand concentration and fitted using Equation 1 (see Fig. 4, panels B, D, and F for the W102Y enzyme).LigandDissociation constant (Kd) for each ligand-PETN reductase complexWild-typeW102FW102YμmμmμmPicric acid5.4 ± 1.10.07 ± 0.030.24 ± 0.04Progesterone0.07 ± 0.030.07 ± 0.040.03 ± 0.022,4-DNP0.95 ± 0.130.21 ± 0.121.05 ± 0.09 Open table in a new tab Kinetic Analysis of the Reductive and Oxidative Half-reactions of W102Y and W102F PETN Reductases—The kinetic mechanism for the reductive half-reaction of wild-type PETN reductase has been described elsewhere (14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar) and involves the formation of an enzyme-NADPH charge-transfer complex (E-NADPH) followed by enzyme reduction (Scheme 1). The rate of formation and decay of the charge-transfer complex for the mutant enzymes was measured in stopped-flow studies with absorption detection at 560 nm (Fig. 5). Reduction of the flavin was monitored at 464 nm as a single exponential decrease in absorbance and was shown to be kinetically equivalent to the slow-down phase observed in the biphasic transients observed at 560 nm. As with wild-type PETN reductase (14Khan H. Harris R.J. Barna T. Craig D.H. Bruce N.C. Munro A.W. Moody P.C. Scrutton N.S. J. Biol. Chem. 2002; 277: 21906-21912Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar), the rate of flavin reduction for the mutant enzymes was essentially independent of NADPH concentration (range 100–1000 μm coenzyme) at 5 °C (Fig. 5A). Global analysis of photodiode array data has indicated that flavin reduction is essentially irreversible in wild-type PETN reductase (i.e. k-2 ∼0; see Ref. 15Basran J. Harris R.J. Sutcliffe M.J. Scrutton N.S. J. Biol. Chem. 2003; 278: 43973-43982Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar for more detailed discussion), and similar observations were made for the W102Y and W102F enzymes by global fitting of spectral data sets obtained in stopped-flow studies of the reductive half-reaction (data not shown). Consistent with the kinetic model proposed for wild-type enzyme (Scheme 1), charge-transfer complex (E-NADPH) formation in the mutant enzymes is dependent on NADPH concentration and is reversible (Fig. 5B). Rate constants for formation and decay of the charge-transfer complex and for flavin reduction are given in Table III. From these data it is clear that mutation of Trp-102 does not substantially affect the rate of flavin reduction by NADPH or the rate of formation and decay of the E-NADPH charge-transfer species, although a modest ∼2-fold increase in the hydride transfer rate and the rate of decay of the enzyme-NADPH complex is seen with the W102F PETN reductase.Fig. 5Plot of observed
Referência(s)