Proton NMR study of the heme environment in bacterial quinol oxidases

The heme environment and ligand binding properties of two relatively large membrane proteins containing multiple paramagnetic metal centers, cytochrome bo3 and bd quinol oxidases, have been studied by high field proton nuclear magnetic resonance (NMR) spectroscopy. The oxidized bo3 enzyme displays well-resolved hyperfine-shifted 1H NMR resonance assignable to the low-spin heme b center. The observed spectral changes induced by addition of cyanide to the protein were attributed to the structural perturbations on the low-spin heme (heme b) center by cyanide ligation to the nearby high-spin heme (heme o) of the protein. The oxidized hd oxidase shows extremely broad signals in the spectral region where protons near high-spin heme centers resonate. Addition of cyanide to the oxidized bd enzyme induced no detectable perturbations on the observed hyperfine signals, indicating the insensitive nature of this heme center toward cyanide. The proton signals near the low-spin heme b558 center are only observed in the presence of 20% formamide, consistent with a critical role of viscosity in detecting NMR signals of large membrane proteins. The reduced bd protein also displays hyperfine-shifted 1H NMR signals, indicating that the high-spin heme centers (hemes b595 and d) remain high-spin upon chemical reduction. The results presented here demonstrate that structural changes of one metal center can significantly influence the structural properties of other nearby metal center(s) in large membrane paramagnetic metalloproteins.     

Comparison of redox and ligand binding behaviour of yeast and bovine cytochrome c oxidases using FTIR spectroscopy

Redox and CO photolysis FTIR spectra of yeast cytochrome c oxidase WT and mutants are compared to those from bovine and P. denitrificans CcOs in order to establish common functional features. All display changes that can be assigned to their E242 (bovine numbering) equivalent and to weakly H-bonded water molecules. The additional redox-sensitive band reported at 1736?cm^?1 in bovine CcO and previously assigned to D51 is absent from yeast CcO and couldn't be restored by introduction of a D residue at the equivalent position of the yeast protein. Redox spectra of yeast CcO also show much smaller changes in the amide I region, which may relate to structural differences in the region around D51 and the subunit I/II interface.     

Long lived intermediate metal site structure upon binding of cadmium to plastocyanin

Perturbed angular correlation of γ-rays (PAC) spectroscopy of cadmium substituted plastocyanin shows one dominant metal site configuration at pH 7.5. Lowering the pH to 4.8 a fraction of the molecules undergoes structural change and loses the bound cadmium ion. At pH 4.4 all plastocyanin is in the apo-form. Increasing the pH back to neutral pH values two distinct metal site coordination geometries were observed. One of the two signals is the same as that found initially at pH 7.5; the other form is stable for hours at 1 °C, indicating the existence of a long lived intermediate metal site structure. The cadmium ion is surrounded by the same ligands (His37, Cys84, His87 and Met92) in both forms, however the metal center in the long lived intermediate metal site structure can be best described by a larger His–metal–His angle.     

Analysis of the binding site architecture of monoclonal antibodies to morphine by using competitive ligand binding and molecular modeling

The structural features of mAb directed against the opiate morphine were analyzed by using competitive ligand analog-binding studies, examination of the V region amino acid sequence, and computer-aided molecular modeling of the fragment V region. The antibody response in BALB/c mice to morphine is relatively restricted, in that all of the mAb examined in this study contained the same lambda L chain and very similar H chain V regions. A three-dimensional model of the antimorphine-binding site was constructed by using computational and graphic display techniques. Each of the six complementary-determining regions was constructed by using fragment replacement methods employing canonical loop conformations of known "parent" structures. Experimental competitive ligand-binding data and theoretical modeling suggest that a charged glutamate residue at position H:50 and aromatic side chains of residues H:33W, H:47W, H:58F, H:95W, H:101iY, and L:91W are key features in ionic and hydrophobic interactions with the ligand. This study represents the first use of theoretical and experimental modeling techniques to describe the Ag-binding site of a mouse fragment V region containing a lambda L chain.     

Modeling binding kinetics at the Q(A) site in bacterial reaction centers

Bacterial reaction centers (RCs) catalyze a series of electron-transfer reactions reducing a neutral quinone to a bound, anionic semiquinone. The dissociation constants and association rates of 13 tailless neutral and anionic benzo- and naphthoquinones for the Q(A) site were measured and compared. The K(d) values for these quinones range from 0.08 to 90 microM. For the eight neutral quinones, including duroquinone (DQ) and 2,3-dimethoxy-5-methyl-1,4-benzoquinone (UQ(0)), the quinone concentration and solvent viscosity dependence of the association rate indicate a second-order rate-determining step. The association rate constants (k(on)) range from 10(5) to 10(7) M(-)(1) s(-)(1). Association and dissociation rate constants were determined at pH values above the hydroxyl pK(a) for five hydroxyl naphthoquinones. These negatively charged compounds are competitive inhibitors for the Q(A) site. While the neutral quinones reach equilibrium in milliseconds, anionic hydroxyl quinones with similar K(d) values take minutes to bind or dissociate. These slow rates are independent of ionic strength, solvent viscosity, and quinone concentration, indicating a first-order rate-limiting step. The anionic semiquinone, formed by forward electron transfer at the Q(A) site, also dissociates slowly. It is not possible to measure the association rate of the unstable semiquinone. However, as the protein creates kinetic barriers for binding and releasing anionic hydroxyl quinones without greatly increasing the affinity relative to neutral quinones, it is suggested that the Q(A) site may do the same for anionic semiquinone. Thus, the slow semiquinone dissociation may not indicate significant thermodynamic stabilization of the reduced species in the Q(A) site.     

Ultrafast ligand binding dynamics in the active site of native bacterial nitric oxide reductase

The active site of nitric oxide reductase from Paracoccus denitrificans contains heme and non-heme iron and is evolutionarily related to heme-copper oxidases. The CO and NO dynamics in the active site were investigated using ultrafast transient absorption spectroscopy. We find that, upon photodissociation from the active site heme, 20% of the CO rebinds in 170 ps, suggesting that not all the CO transiently binds to the non-heme iron. The remaining 80% does not rebind within 4 ns and likely migrates out of the active site without transient binding to the non-heme iron. Rebinding of NO to ferrous heme takes place in approximately 13 ps. Our results reveal that heme-ligand recombination in this enzyme is considerably faster than in heme-copper oxidases and are consistent with a more confined configuration of the active site.     

Probing site specificity of DNA binding metallointercalators by NMR spectroscopy and molecular modeling

The molecular recognition of oligonucleotides by chiral ruthenium complexes has been probed by NMR spectroscopy using the template Delta-cis-alpha- and Delta-cis-beta-[Ru(RR-picchxnMe(2)) (bidentate)](2+), where the bidentate ligand is one of phen (1,10-phenanthroline), dpq (dipyrido[3,2-f:2',3'-h]quinoxaline), or phi (9,10-phenanthrenequinone diimine) and picchxnMe(2)() is N,N'-dimethyl-N,N'-di(2-picolyl)-1,2-diaminocyclohexane. By varying only the bidentate ligand in a series of complexes, it was shown that the bidentate alone can alter binding modes. DNA binding studies of the Delta-cis-alpha-[Ru(RR-picchxnMe(2))(phen)](2+) complex indicate fast exchange kinetics on the chemical shift time scale and a "partial intercalation" mode of binding. This complex binds to [d(CGCGATCGCG)](2) and [d(ATATCGATAT)](2) at AT, TA, and GA sites from the minor groove, as well as to the ends of the oligonucleotide at low temperature. Studies of the Delta-cis-beta-[Ru(RR-picchxnMe(2))(phen)](2+) complex with [d(CGCGATCGCG)](2) showed that the complex binds only weakly to the ends of the oligonucleotide. The interaction of Delta-cis-alpha-[Ru(RR-picchxnMe(2))(dpq)](2+) with [d(CGCGATCGCG)](2) showed intermediate exchange kinetics and evidence of minor groove intercalation at the GA base step. In contrast to the phen and dpq complexes, Delta-cis-alpha- and Delta-cis-beta-[Ru(RR-picchxnMe(2))(phi)](2+) showed evidence of major groove binding independent of the metal ion configuration. DNA stabilization induced by complex binding to [d(CGCGATCGCG)](2) (measured as DeltaT(m)) increases in the order phen < dpq and DNA affinity in the order phen < dpq < phi. The groove binding preferences exhibited by the different bidentate ligands is explained with the aid of molecular modeling experiments.     

Tight binding of inhibitors to bovine bc1 complex is independent of the Rieske protein redox state. Consequences for semiquinone stabilization in the quinol oxidation site

To determine the effect of the redox state of the Rieske protein on ligand binding to the quinol oxidation site of the bc(1) complex, we measured the binding rate constants (k(1)) for stigmatellin and myxothiazol, at different concentrations of decylbenzoquinone or decylbenzoquinol, in the bovine bc(1) complex with the Rieske protein in the oxidized or reduced state. Stigmatellin and myxothiazol bound tightly and competitively with respect to quinone or quinol, independently of the redox state of the Rieske protein. In the oxidized bc(1) complex, the k(1) values for stigmatellin ( approximately 2.6 x 10(6) m(-1)s(-1)) and myxothiazol ( approximately 8 x 10(5) m(-1)s(-1)), and the dissociation constant (K(d)) for quinone, were similar between pH 6.5 and 9, indicating that ligand binding is independent of the protonation state of histidine 161 of the Rieske protein (pK(a) approximately 7.6). Reduction of the Rieske protein increased the k(1) value for stigmatellin and decreased the K(d) value for quinone by 50%, without modifying the k(1) for myxothiazol. These results indicate that reduction of the Rieske protein and protonation of histidine 161 do not induce a strong stabilization of ligand binding to the quinol oxidation site, as assumed in models that propose the existence of a highly stabilized semiquinone as a reaction intermediate during quinol oxidation.     

Design of a redox-linked active metal site: manganese bound to bacterial reaction centers at a site resembling that of photosystem II

Metals bound to proteins perform a number of crucial biological reactions, including the oxidation of water by a manganese cluster in photosystem II. Although evolutionarily related to photosystem II, bacterial reaction centers lack both a strong oxidant and a manganese cluster for mediating the multielectron and proton transfer needed for water oxidation. In this study, carboxylate residues were introduced by mutagenesis into highly oxidizing reaction centers at a site homologous to the manganese-binding site of photosystem II. In the presence of manganese, light-minus-dark difference optical spectra of reaction centers from the mutants showed a lack of the oxidized bacteriochlorophyll dimer, while the reduced primary quinone was still present, demonstrating that manganese was serving as a secondary electron donor. On the basis of these steady-state optical measurements, the mutant with the highest-affinity site had a dissociation constant of approximately 1 microM. For the highest-affinity mutant, a first-order rate with a lifetime of 12 ms was observed for the reduction of the oxidized bacteriochlorophyll dimer by the bound manganese upon exposure to light. The dependence of the amplitude of this component on manganese concentration yielded a dissociation constant of approximately 1 muM, similar to that observed in the steady-state measurements. The three-dimensional structure determined by X-ray diffraction of the mutant with the high-affinity site showed that the binding site contains a single bound manganese ion, three carboxylate groups (including two groups introduced by mutagenesis), a histidine residue, and a bound water molecule. These reaction centers illustrate the successful design of a redox active metal center in a protein complex.     

Refined structure and metal binding site of the kalata B1 peptide

The cyclic polypeptide kalata B1 from the African plant Oldenlandia affinis DC consists of 29 amino acid residues with three disulfide linkages. In this study we used two-dimensional NMR spectroscopy to investigate the three-dimensional structure of the peptide and to determine the disulfide connectivities. Nuclear Overhauser effects (NOEs) between neighboring beta-protons of the cysteines detected at 750 MHz provided evidence for the disulfide connectivity pattern 5-13, 17-29, and 22-27. These disulfide linkages were confirmed by three-dimensional structures calculated from input constraints derived solely from NOEs without explicit disulfide connectivities. Kalata B1 is insoluble in aqueous solution above pH 3.5, but in a 50-50 water-methanol mixture, it was possible to use natural abundance two-dimensional (15)N-(1)H heteronuclear single quantum coherence spectroscopy to study the hydrophobic peptide from pH 2 to 10. The addition of methanol resulted in no significant structural changes. Although the peptide contains three prolyl residues, no evidence of multiple conformers was detected at any pH. The addition of Mn(2+) to kalata B1 resulted in selective broadening of resonances from Asn 23, Thr 24, and Glu 15; these results suggest that these three residues are involved in a specific metal binding site.     

Active site structure of the aa3 quinol oxidase of Acidianus ambivalens

The membrane bound aa(3)-type quinol:oxygen oxidoreductase from the hyperthermophilic archaeon, Acidianus ambivalens, which thrives at a pH of 2.5 and a temperature of 80 degrees C, has several unique structural and functional features as compared to the other members of the heme-copper oxygen reductase superfamily, but shares the common redox-coupled, proton-pumping function. To better understand the properties of the heme a(3)-Cu(B) catalytic site, a resonance Raman spectroscopic study of the enzyme under a variety of conditions and in the presence of various ligands was carried out. Assignments of several heme vibrational modes as well as iron-ligand stretching modes are made to serve as a basis for comparing the structure of the enzyme to that of other oxygen reductases. The CO-bound oxidase has conformations that are similar to those of other oxygen reductases. However, the addition of CO to the resting enzyme does not generate a mixed valence species as in the bovine aa(3) enzyme. The cyanide complex of the oxidized enzyme of A. ambivalens does not display the high stability of its bovine counterpart, and a redox titration demonstrates that there is an extensive heme-heme interaction reflected in the midpoint potentials of the cyanide adduct. The A. ambivalens oxygen reductase is very stable under acidic conditions, but it undergoes an earlier alkaline transition than the bovine enzyme. The A. ambivalens enzyme exhibits a redox-linked reversible conformational transition in the heme a(3)-Cu(B) center. The pH dependence and H/D exchange demonstrate that the conformational transition is associated with proton movements involving a group or groups with a pK(a) of approximately 3.8. The observed reversibility and involvement of protons in the redox-coupled conformational transition support the proton translocation model presented earlier. The implications of such conformational changes are discussed in relation to general redox-coupled proton pumping mechanisms in the heme-copper oxygen reductases.     

Guanidine-II aptamer conformations and ligand binding modes through the lens of molecular simulation

Regulation of gene expression via riboswitches is a widespread mechanism in bacteria. Here, we investigate ligand binding of a member of the guanidine sensing riboswitch family, the guanidine-II riboswitch (Gd-II). It consists of two stem–loops forming a dimer upon ligand binding. Using extensive molecular dynamics simulations we have identified conformational states corresponding to ligand-bound and unbound states in a monomeric stem–loop of Gd-II and studied the selectivity of this binding. To characterize these states and ligand-dependent conformational changes we applied a combination of dimensionality reduction, clustering, and feature selection methods. In absence of a ligand, the shape of the binding pocket alternates between the conformation observed in presence of guanidinium and a collapsed conformation, which is associated with a deformation of the dimerization interface. Furthermore, the structural features responsible for the ability to discriminate against closely related analogs of guanidine are resolved. Based on these insights, we propose a mechanism that couples ligand binding to aptamer dimerization in the Gd-II system, demonstrating the value of computational methods in the field of nucleic acids research.     

On the influence of local molecular environment on the redox potential of electron transfer cofactors in bacterial photosynthetic reaction centers

The addition of cryosolvents (glycerol, dimethylsulfoxide) to a water solution containing bacterial photosynthetic reaction centers changes the redox potential of the bacteriochlorophyll dimer, but does not affect the redox potential of the quinone primary acceptor. It has been shown that the change in redox potential can be produced by changes of the electrostatic interactions between cofactors and the local molecular environment modified by additives entered into the solution. The degree of influence of a solvent on the redox potential of various cofactors is determined by degree of availability of these cofactors for molecules of solvent, which depends on the arrangement of cofactors in the structure of reaction centers.     

On the influence of local molecular environment on the redox potential of electron transfer cofactors in bacterial photosynthetic reaction centers

The addition of cryosolvents (glycerol, dimethylsulfoxide) to a water solution containing bacterial photosynthetic reaction centers changes the redox potential of the bacteriochlorophyll dimer but does not affect the redox potential of the quinone primary acceptor. It is shown that the change in redox potential can be due to variation of the electrostatic interactions between cofactors and the local molecular environment, which is modified by solvent additives. The degree of influence of a solvent on the redox potential of various cofactors is determined by their solvent accessibility, i.e. on their arrangement in the structure of reaction centers.     

Tuning the redox potential of the primary electron donor in bacterial reaction centers by manganese binding and light-induced structural changes

The influence of transition metal binding on the charge storage ability of native bacterial reaction centers (BRCs) was investigated. Binding of manganous ions uniquely prevented the light-induced conformational changes that would yield to long lifetimes of the charge separated state and the drop of the redox potential of the primary electron donor (P). The lifetimes of the stable charge pair in the terminal conformations were shortened by 50-fold and 7-fold upon manganous and cupric ion binding, respectively. Nickel and zinc binding had only marginal effects. Binding of manganese not only prevented the drop of the potential of P/P⁺ but also elevated it by up to 117 mV depending on where the metal was binding. With variable conditions, facilitating either manganese binding or light-induced structural changes a controlled tuning of the potential of P/P⁺ in multiple steps was demonstrated in a range of ~200 mV without the need of a mutation or synthesis. Under the selected conditions, manganese binding was achieved without its photochemical oxidation thus, the energized but still native BRCs can be utilized in photochemistry that is not reachable with regular BRCs. A 42 Å long hydrophobic tunnel was identified that became obstructed upon manganese binding and its likely role is to deliver protons from the hydrophobic core to the surface during conformational changes.     

Probing the ligand preferences of the three types of bacterial pantothenate kinase

Pantothenate kinase (PanK) catalyzes the transformation of pantothenate to 4′-phosphopantothenate, the first committed step in coenzyme A biosynthesis. While numerous pantothenate antimetabolites and PanK inhibitors have been reported for bacterial type I and type II PanKs, only a few weak inhibitors are known for bacterial type III PanK enzymes. Here, a series of pantothenate analogues were synthesized using convenient synthetic methodology. The compounds were exploited as small organic probes to compare the ligand preferences of the three different types of bacterial PanK. Overall, several new inhibitors and substrates were identified for each type of PanK.     

Metal binding to angiotensin converting enzyme: implications for the metal binding site

Angiotensin converting enzyme interacts with the chelator, 1,10-phenanthroline (OP) to form an OP-Zn-ACE ternary complex, which subsequently dissociates to OP-Zn and apoenzyme. The association and dissociation rate constants for the reaction OP + Zn-ACE in equilibrium OP-Zn-ACE have been determined and compared with those of known OP-metal complexes. Such constants were also used to calculate the rate constant for formation of the OP-Zn complex from OP-Zn-ACE. The rate of dissociation of zinc from ACE has been measured in the presence of EDTA (which acts only as a metal scavenger) as a function of chelator concentration, at different pH values, and with different buffers. The stability constant for the binding of zinc to apoACE log Kc = 8.2, determined by equilibrium dialysis using atomic absorption spectroscopy to assess metal concentration, is much smaller than that for Zn-carboxypeptidase A. Zn-thermolysin, or Zn-carbonic anhydrase. This weak binding is attributable to the zinc dissociation rate constant of ACE, 7.5 X 10(-3) sec-1 at pH 7.0, which is much greater than that of the other zinc metalloenzymes. These results lead to inferences regarding the metal binding site of ACE.     

Mapping the ligand binding site at protein side-chains in protein-ligand complexes through NOE difference spectroscopy

This report describes a novel NMR approach for mapping the interaction surface between an unlabeled ligand and a ¹³C,¹⁵N-labeled protein. The method relies on the spin inversion properties of the dipolar relaxation pathways and records the differential relaxation of two spin modes, where ligand and protein ¹H magnetizations are aligned either in a parallel or anti-parallel manner. Selective inversion of protein protons is achieved in a straightforward manner by exploiting the one-bond heteronuclear scalar couplings (¹J_CH, ¹J_NH). Suppression of indirect relaxation pathways mediated by bulk water or rapidly exchanging protons is achieved by selective inversion of the water signal in the middle of the NOESY mixing period. The method does not require deuteration of the protein or well separated spectral regions for the protein and the ligand, respectively. Additionally, in contrast to previous methods, the new experiment identifies side-chain enzyme ligand interactions along the intermolecular binding interface. The method is demonstrated with an application to the B₁₂-binding subunit of glutamate mutase from Clostridium tetanomorphum for which NMR chemical shift changes upon B₁₂-nucleotide loop binding and a high-resolution solution structure are available.