Superoxide reactivity of KatG: insights into isoniazid resistance pathways in TB

To gain insight into the mechanism of INH activation by KatG and to understand how resistance is conferred by the single active-site point mutation of KatG(S315T), we have employed pulse radiolysis as the means to initiate a catalytic pathway capable of mimicking the in vivo oxidation of isoniazid (INH). Radiolysis of a solution containing WT KatG revealed two intermediates: compound III (oxyferrous KatG) [415 (Soret), 545, 580 nm] formed [k1 = (4.47 +/- 0.91) x 105 M-1 s-1] in the absence of INH and compound II (410 (Soret), 540, 575 nm) formed [k1 = (4.43 +/- 0.69) x 105 M-1 s-1] in the presence of INH, with a comparison of the rates suggesting that compound III (rate-limiting) precedes compound II formation. By contrast, radiolysis of KatG(S315T) only led to compound III formation, whether INH was present [k1 = (4.72 +/- 0.99) x 105 M-1 s-1] or not [k1 = (4.51 +/- 1.38) x 105 M-1 s-1]. HPLC studies to determine the rates of INH-NADH adduct formation (an inhibitor of InhA) as catalyzed by KatG were also performed employing various oxidants: air [WT: (7.18 +/- 1.25) x 10-4, S315T: (0.74 +/- 0.39) x 10-4], superoxide (SOTS-1) [WT: (9.22 +/- 1.10) x 10-4, S315T: not detected], and tert-butylhydroperoxide [WT: (20.5 +/- 1.13) x 10-4, S315T: (10.15 +/- 0.19) x 10-4]. Taken together, the results from the pulse radiolysis work as well as the InhA inhibitor studies allow us to propose a mechanism capable of correlating the inability for the oxyferrous intermediate of KatG(S315T) to oxidize ("activate") INH to the suppressed formation of the INH-NADH adduct, thereby leading to INH resistance in Mycobacterium tuberculosis.     

Two [Fe(IV)=O Trp*] intermediates in M. tuberculosis catalase-peroxidase discriminated by multifrequency (9-285 GHz) EPR spectroscopy: reactivity toward isoniazid

We have characterized the intermediates formed in the peroxidase cycle of the multifunctional heme-containing enzyme KatG of M. tuberculosis. Selected Trp variants from the heme proximal (W321F) and distal (W107F and W91F) sides were analyzed together with the wild-type enzyme with regard to the reaction with peroxyacetic acid and hydrogen peroxide (in the catalase-inactive W107F). The 9 GHz EPR spectrum of the enzyme upon reaction with peroxyacetic acid showed the contribution of three protein-based radical species, two Trp* and a Tyr*, which could be discerned using a combined approach of multifrequency Electron Paramagnetic Resonance (EPR) spectroscopy with selective deuterium labeling of tryptophan and tyrosine residues and site-directed mutagenesis. Trp321, a residue in H-bonding interactions with the iron through Asp381 and the heme axial ligand His270, was identified as one of the radical sites. The 9 GHz EPR signal of the Trp321 radical species was consistent with an exchange-coupled species similar to the oxoferryl-Trp radical intermediate in cytochrome c peroxidase. On the basis of the possibility of distinguishing among the different radical intermediates of the peroxidase cycle in M. tuberculosis KatG (MtKatG), we used EPR spectroscopy to monitor the reactivity of the enzyme and its W321F variant with isoniazid, the front-line drug used in the treatment of tuberculosis. The EPR experiments on the W321F variant preincubated with isoniazid allowed us to detect the short-lived [Fe(IV)=O Por*+] intermediate. Our results showed that neither the [Fe(IV)=O Por*+] nor the [Fe(IV)=O Trp321*+] intermediates were the reactive species with isoniazid. Accordingly, the subsequent intermediate (most probably the other Trp*) is proposed to be the oxidizing species. Our findings demonstrate that the protein-based radicals formed as alternative intermediates to the [Fe(IV)=O Por*+] can play the role of cofactors for substrate oxidation in the peroxidase cyle of KatGs.     

Versatility of the electronic structure of compound I in catalase-peroxidases

Catalase-peroxidases (KatGs) are bifunctional heme proteins, belonging to the family of class I peroxidases, that are able to catalyze both catalatic and peroxidatic reactions within a peroxidase-like structure. We investigated the electronic structure of reaction intermediates of the catalytic cycle of KatGs by means of density functional theory (DFT) QM/MM calculations. The outcome was that the ionization state of the KatG-specific covalent adduct (Met264-Tyr238-Trp111) affects the radical character of compound I (Cpd I). Specifically, in the optimized structures, substantial radical character is observed on the proximal Trp330 when Tyr238 is protonated, whereas when Tyr238 is deprotonated the radical localizes on the Met+-Tyr(O-)-Trp adduct. These findings are not affected by protein thermal fluctuations, although details of the spin density distribution are affected by the geometry of the active site. Calculations provide structures in good agreement with the crystal structure of BpKatG Cpd I. They also provide an explanation for the experimental findings of the mobile and catalatic-specific residue Arg426 being 100% in conformation R in the X-ray structure of BpKatG treated with organic peroxides. The role of different Cpd I forms in the catalase and peroxidase reaction pathways is discussed.     

Protein engineering of nitrile hydratase activity of papain: molecular dynamics study of a mutant and wild-type enzyme

The mechanism of hydrolysis of the nitrile (N-acetyl-phenylalanyl-2-amino-propionitrile, I) catalyzed by Gln19Glu mutant of papain has been studied by nanosecond molecular dynamics (MD) simulations. MD simulations of the complex of mutant enzyme with I and of mutant enzyme covalently attached to both neutral (II) and protonated (III) thioimidate intermediates were performed. An MD simulation with the wild-type enzyme.I complex was undertaken as a reference. The ion pair between protonated His159 and thiolate of Cys25 is coplanar, and the hydrogen bonding interaction S(-)(25).HD1-ND1(159) is observed throughout MD simulation of the mutant enzyme.I complex. Such a sustained hydrogen bond is absent in nitrile-bound wild-type papain due to the flexibility of the imidazole ring of His159. The nature of the residue at position 19 plays a critical role in the hydrolysis of the covalent thioimidate intermediate. When position 19 represents Glu, the imidazolium ion of His159-ND1(+).Cys25-S(-) ion pair is distant, on average, from the nitrile nitrogen of substrate I. Near attack conformers (NACs) have been identified in which His159-ImH(+) is positioned to initiate a general acid-catalyzed addition of Cys-S(-) to nitrile. Though Glu19-CO(2)H is distant from nitrile nitrogen in the mutant.I structure, MD simulations of the mutant.II covalent adduct finds Glu19-CO(2)H hydrogen bonded to the thioimide nitrogen of II. This hydrogen bonded species is much less stable than the hydrogen bonded Glu19-CO(2)(-) with mutant-bound protonated thioimidate (III). This observation supports Glu19-CO(2)H general acid catalysis of the formation of mutant.III. This is the commitment step in the Gln19Glu mutant catalysis of nitrile hydrolysis.     

Investigations on the synthesis, structures, and properties of new copper(I) 2,3-dimethylpyrazine coordination compounds

Five new coordination compounds were prepared, structurally characterized, and investigated for their thermal properties. In the structure of the ligand-rich 4:9 compound, tetra(mu2-chloro)bis(mu2-2,3-dimethylpyrazine-N,N')tetrakis(2,3-dimethylpyrazine-N)tetracopper(I) tris(2,3-dimethylpyrazine)solvate (I), discrete complexes are formed by build up of two [(CuCl-(2,3-dimethylpyrazine)2]2 dimers, which are connected by two 2,3-dimethylpyrazine ligands via mu-N,N' coordination. In the 1:1 compound poly[mu2-chloro-mu2-2,3-dimethylpyrazine-N,N'-copper(I)] (II), (CuCl)2 dimers are found, which are connected by the 2,3-dimethylpyrazine ligands into layers. For this composition, a second polymorphic modification was found (III), which exhibits a different topology of the coordination network and a different packing of the layers. In the most stable 3:2 compound catena[tri(mu2-chloro)bis(mu2-2,3-dimethylpyrazine-N,N')tricopper(I)] (IV), six-membered rings of (CuCl)3 are found, which are connected by the 2,3-dimethylpyrazine ligands into chains. In the ligand-deficient 2:1 compound, poly[di(mu3-chloro)(mu2-2,3-dimethylpyrazine-N,N')dicopper(I)] (V), CuCl double chains are found, which are connected by the 2,3-dimethylpyrazine ligands into layers. On heating, compound I transforms quantitatively into the 3:2 compound IV without the formation of II or III as intermediates. Compound IV is also obtained by heating either the 1:1 compound II or III. On further heating, the 3:2 compound IV loses additional ligands, forming the ligand-deficient 2:1 compound V, which then decomposes into CuCl. The stability, thermal reactivity, and the transition behavior of all compounds were investigated using different thermoanalytical methods. These results are compared with those previously reported for the structurally similar CuCl(2-ethylpyrazine) coordination compounds. The formation and the stability of the different compounds in solution were also investigated.     

Synthesis of 1,2,3,4-tetrahydroquinazoline derivatives as potential alpha-adrenergic blocking agents

The synthesis of N,N′-bis[6-(1,2,3,4-tetrahydro-3-quinazolidyl)hexyl]cystamine (I) and 3-(6-aminohexyl)-1,2,3,4-tetrahydroquinazoline (II) are described. Compound I is obtained by condensation of o-nitrobenzoyl chloride with 3-(6-aminohexyl)-1,3-thiazolidine (III) followed by dimerization, reduction and formation of tetrahydroquinazoline ring. A similar method was used for preparation of compound II. These compounds and some synthesis intermediates are potential alpha-adrenergic blockers.     

The nonenzymatic activation of isoniazid by MnIII-pyrophosphate in the presence of NADH produces the inhibition of the enoyl-ACP reductase InhA from Mycobacterium tuberculosis

The antituberculosis drug isoniazid (INH) is quickly oxidized by stoichiometric amounts of manganese(III)-pyrophosphate. In the presence of the nicotinamide coenzyme, the INH oxidation produced the formation of INH-NAD(H) adducts and allowed the in vitro inhibition of the enoyl-acyl carrier protein reductase InhA, an INH target in the biosynthetic pathway for mycolic acids. Manganese(III)-pyrophosphate is an efficient alternative oxidant to mimic the activity of the Mycobacterium tuberculosis KatG catalase-peroxidase and will be useful for further mechanistic studies of INH activation and for structural investigations on reactive INH species and resulting InhA inhibitors.     

Design of an optical sensor for indirect determination of isoniazid

The characterization of an optical sensor membrane is described for indirect determination of isoniazid. The sensing membrane was consisted of immobilized 3-(2-pyridyl)-5,6-diphenyl-1,2,4-triazine (PDT) on a triacetylcellulose membrane. The procedure is based on the reaction of Fe(III) with isoniazid in the presence of PDT. Fe(III) is reduced by isoniazid to Fe(II) which forms a complex with PDT. The complex shows an absorption maximum at 558nm. By measuring the absorbance of the complex at this wavelength, isoniazid can be determined in the range of 0.62-6.15mugmL(-1). This method was applied to the determination of isoniazid in pharmaceutical formulation and enabled the determination of isoniazid in microgram quantities.     

Formation of compound I by photo-oxidation of compound II

[reaction: see text] Compound I is the heme-iron(IV)-oxo porphyrin radical cation formed in peroxidase and catalase enzymes by reaction with hydrogen peroxide. As an alternative to chemical oxidations of porphyrin-iron(III) species, various compound I species were produced by 355 nm laser flash photolysis photooxidation of the corresponding compound II species, porphyrin-iron(IV)-oxo derivatives. The method is demonstrated by production and kinetic studies of the compound I species from 5,10,15,20-tetrakis(pentafluorophenyl)porphyrin-iron, from horseradish peroxidase, and from wild-type horse skeletal myoglobin.     

Photolysis of Isoniazid

The UV light irradiation of isoniazid (I) in methanol four products, isonicotinic acid (II), isonicotinamide (III), N, N′-bis(isonicotinic acid)hydrazide (IV) and isonicotinaldehyde isonicotinyl hydrazone (V), in ethanol three products, (III), (IV) and acetaldehyde isonicotinyl hydrazone (VI) were isolated and identified. Also, the photoreaction mechanism of isoniazid in methanol and ethanol were discussed.     

Voltammetric determination of mercury(II) at a carbon paste electrode in aqueous solutions containing tetraphenylborate ion

The determination of mercury(II) ions can be achieved by monitoring the decrease in the oxidation peak of the tetraphenylborate ion in the presence of this metal ion at a carbon paste electrode. The reaction between mercury(II) and the tetraphenylborate ion results in the formation of diphenylmercury, thus providing the method with good selectivity over other metal ions. Using anodic stripping voltammetry in a neutral electrolyte, a linear dependence of the decrease of peak height was observed on increasing the mercury(II) concentration in the range 1 x 10(-6)-8 x 10(-9)M mercury(II). Zinc(II), cadmium(II), lead(II), nickel(II), cobalt(II), tin(II), potassium(I) and ammonium(I) ions did not interfere at a 1000-fold concentration excess. Iron(III) and chromium(III) did not interfere at a 250-fold and 50-fold concentration excess, respectively. Following masking procedures, copper(II), bismuth(III) and silver(I) did not interfere at a 100-fold concentration excess. The method can be used to determine the concentration of mercury(II) in natural waters contaminated by this metal.     

Identification of intermediates in the catalytic cycle of chloroperoxidase

Background: Chloroperoxidase (CPO) is the most versatile of the hemethiolate proteins, catalyzing the chlorination of activated CH bonds and reactions reminiscent of peroxidase, catalase, and cytochrome P450. Despite 30 years of continuous efforts, no intermediates of the enzyme's catalytic cycle have been identified except for compound I. Thus, in the absence of conclusive evidence it is generally believed that the halogenation of substrates proceeds by means of ‘free HOCl’ in solution.Results: The pH profile of chloroperoxidase from Caldariomyces fumago revealed a new active-site complex that can be detected only at pH 4.4. According to ultra-violet (UV) spectroscopy, and by comparison with suitable enzyme models, this intermediate is the HOCl adduct of the iron(III) protoporphyrin(IX). Inactivation of chloroperoxidase by diethyl pyrocarbonate, which interrupts the proton shuttle by modification of the distal histidine, led to the formation of the ⁻OCl adduct of the iron complex, which was identified by comparison with a corresponding active site analogue.Conclusions: The availability of enzyme models of heme-thiolate proteins allowed the identification by UV spectroscopy of both the ⁻OCl adduct and the HOCI adduct of the iron(III) protoporphyrin(IX) of chloroperoxidase. The existence of these previously elusive intermediates suggests that the chlorination catalyzed by CPO, and its corresponding active site analogue, proceeds by Cl⁺ transfer from the HOCl adduct to the substrate bound in the distal pocket of the enzyme.     

DFT calculations, structural and spectroscopic studies on the products formed between IBr and N,N'-dimethylbenzoimidazole-2(3H)-thione and -2(3H)-selone

The NBO charge distribution calculated at DFT level on the [LEX](+) species [LE=N,N'-dimethylbenzoimidazole-2(3H)-thione (3) and -2(3H)-selone (4)(Scheme 1); X=I, Br] suggests that the most likely products from the reaction 3 of 4 and with IBr are the 10-X-2 charge-transfer (CT) adduct and the 10-Se-3 "T-shaped" hypervalent adduct featuring a linear Br--Se--I system, respectively. This prediction is confirmed by the synthesis, and X-ray diffraction analysis of 3.IBr (I) and 4.I(0.72)Br(1.28)(II). In particular II, is a 10-Se-3 "T-shaped" hypervalent adduct containing an almost linear X--Se--X system [X--Se--X 179.07(3) degrees, X=I(0.36)/Br(0.64)], which is roughly perpendicular to the average plane of the benzoimidazole moiety. The FT-Raman spectra of I and II agree very well with their structural features. In particular, the complexity of the FT-Raman spectrum of II reflects the disorder in the X-ray crystal structure of this compound.     

Mechanism of reaction of alkyl radicals with (Ni(II)L)2+ complexes in aqueous solutions

The reactions of methyl radicals, CH(3), with the macrocyclic complexes Ni(II)L(1-5) (L(1-5) = cyclam derivatives, vide infra) and Ni(II)edta in aqueous solutions were studied. Methyl radicals react with all these nickel complexes, forming intermediates with Ni(III)-C sigma-bonds. The L(m)Ni(III)-CH(3) complexes are formed in equilibria processes with relatively fast forward rate constants of k(f) > 1 x 10(8) M(-1) s(-1) (except in the case of NiL(2)-trans I cyclam, where the reaction is slower). In all cases the decomposition of the transient complexes occurs via the homolytic cleavage of the metal-carbon sigma-bond. When the homolysis is relatively slow, an isomerisation process of the transient is also observed with the exception of NiL(2), where no isomerisation was observed. The results suggest that the strength of the Ni(III)-CH(3)sigma-bond is mainly affected by steric hindrance.     

Photoaquation of methylated cis-dichlorobis(1,10-phenanthroline)rhodium(III)chloride compounds by direct population of a photoactive triplet excited state

Photoaquation in compounds II and III by direct excitation into a photoactive triplet excited state is reported. The location of the singlet to triplet transition in compound III is estimated by a combination of the action spectrum for photoaquation in the region between 520 and 600 nm and the phosphorescence spectrum at 77 K. The significant increment of the reactivity (10-fold) of the triplet states in II and III as compared to that in I is explained in terms of increasing sigma-donation from the phen ligands stabilizing the pentacoordinated rhodium intermediate formed by chloride expulsion.     

Kinetics and mechanism of the reversible binding of nitric oxide to reduced cobalamin B(12r) (Cob(II)alamin)

The reduced form of aquacobalamin binds nitric oxide very effectively to yield a nitrosyl adduct, Cbl(II)-NO. UV-vis, (1)H-, (31)P-, and (15)N NMR data suggest that the reaction product under physiological conditions is a six-coordinate, "base-on" form of the vitamin with a weakly bound alpha-dimethylbenzimidazole base and a bent nitrosyl coordinated to cobalt at the beta-site of the corrin ring. The nitrosyl adduct can formally be described as Cbl(III)-NO-. The kinetics of the binding and dissociation reactions was investigated by laser flash photolysis and stopped-flow techniques, respectively. The activation parameters, DeltaH, DeltaS, and DeltaV, for the forward and reverse reactions were estimated from the effect of temperature and pressure on the kinetics of these reactions. For the "on" reaction of Cbl(II) with NO, the small positive DeltaS and DeltaV values suggest the operation of a dissociative interchange (I(d)) substitution mechanism at the Co(II) center. Detailed laser flash photolysis and (17)O NMR studies provide evidence for the formation of water-bound intermediates in the laser flash experiments and strongly support the proposed I(d) mechanism. The kinetics of the "off" reaction was studied using an NO-trapping technique. The respective activation parameters are also consistent with a dissociative interchange mechanism.     

An inorganic iron complex that inhibits wild-type and an isoniazid-resistant mutant 2-trans-enoyl-ACP (CoA) reductase from Mycobacterium tuberculosis

The in vitro kinetics of inactivation of both wild-type and I21V InhA enzymes by [FeII(CN)5(INH)]3- indicate that this process requires no activation by KatG, and no need for the presence of NADH. This inorganic complex may represent a new class of lead compounds to the development of anti-tubercular agents aiming at inhibition of a validated target.     

Switching off electron transfer reactions in confined media: reduction of [Co(dipic)2]- and [Co(edta)]- by hexacyanoferrate(II)

The kinetics of reduction of two cobalt(III) complexes with similar redox potentials by hexacyanoferrate(II) were investigated in water and in reverse micelle (RM) microemulsions. The RMs were composed of water, surfactant [(sodium(bis(2-ethylhexylsulfosuccinate)), NaAOT], and isooctane. Compared to the reaction in water, the reduction rates of (ethylenediaminetetraacetato)cobaltate(III) by hexacyanoferrate(II) were dramatically suppressed in RM microemulsions whereas a slight rate increase was observed for reduction of bis-(2,6-dipicolinato)cobaltate(III). For example, the ferrocyanide reduction of [Co(dipic)(2)](-) increased from 55 M(-1) s(-1)in aqueous media to 85 M(-1) s(-1) in a w(o) = 20 RM. The one-dimensional (1-D) and two-dimensional (2-D) (1)H NMR and FT-IR studies are consistent with the reduction rate constants of these two complexes being affected by their location within the RM. Since reduction of [Co(edta)](-) is switched off, in contrast to [Co(dipic)(2)](-), these observations are attributed to the penetration of the [Co(edta)](-) into the interfacial region of the RM whereas [Co(dipic)(2)](-) is in a region highly accessible to the water pool and thus hexacyanoferrate(II). These results demonstrated that compartmentalization completely turns off a redox reaction in a dynamic microemulsion system by either reactant separation or alteration of the redox potentials of the reactants.     

Electron Transfer. 129. Copper Catalysis in the Thiol Reduction of Oxime-Bound Nickel(IV)(1)

Solutions of the Ni(IV) complex of the dianion of 2,6-diacetylpyridine dioxime (chelate II in text) are reduced very slowly by 2-aminoethanethiol at pH 2.3-3.0, but this reaction is catalyzed dramatically and specifically by dissolved copper, with Cu(I) the active reductant. When the [thiol]/[Ni(IV)] ratio exceeds 1.6, each Ni(IV) oxidizes two molecules of thiol, forming Ni(II) and R(2)S(2). At low concentrations of catalyst and reductant, reaction profiles are almost exponential, but at higher concentrations of either, curves become progressively more nearly linear. Reactions are sharply retarded by increases in acidity. Profiles for 14 runs, carried out with [H(+)] = 0.001-0.0040 M, [Ni(IV)] = (0.94-1.2) x 10(-)(5) M, [thiol] = (2.0-32) x 10(-)(4) M, and [Cu(2+)] = (2.5-80) x 10(-)(6)M, are consistent with a reaction sequence (eqs 2-10 in text) in which Cu(I) is generated in competing homolyses of the complexes Cu(II)(SRH) and Cu(II)(SRH)(2). Reduction of Ni(IV) appears then to proceed through a Ni(IV)Cu(I) adduct, which can undergo electron transfer (yielding Ni(III) and Cu(II)), either in a unimolecular fashion or, alternatively, as a result of attack by a second Cu(I) species. The Ni(IV)Cu(I) + Cu(I) process is reflected in approach to second-order dependences on [Cu(II)] and [thiol] (which generate Cu(I)) at high concentrations of these reagents. Reductions of the Ni(III) intermediate are taken to be much more rapid than those of Ni(IV). Kinetic trends in the present system stand in contrast to the more familiar catalytic patterns such as those seen when the same combination of thiol and catalyst is used to reduce superoxo complexes of cobalt(III). With the latter reactions, decay profiles for the oxidant tend to be exponential at high reagent concentrations but approach linearity at low.