JBIC Journal of Biological Inorganic Chemistry (v.17, #5)

Role of an isoform-specific serine residue in FMN–heme electron transfer in inducible nitric oxide synthase by Wenbing Li; Weihong Fan; Li Chen; Bradley O. Elmore; Mike Piazza; J. Guy Guillemette; Changjian Feng (675-685).
In the crystal structure of a calmodulin (CaM)-bound FMN domain of human inducible nitric oxide synthase (NOS), the CaM-binding region together with CaM forms a hinge, and pivots on an R536(NOS)/E47(CaM) pair (Xia et al. J Biol Chem 284:30708–30717, 2009). Notably, isoform-specific human inducible NOS S562 and C563 residues form hydrogen bonds with the R536 residue through their backbone oxygens. In this study, we investigated the roles of the S562 and C563 residues in the NOS FMN–heme interdomain electron transfer (IET), the rates of which can be used to probe the interdomain FMN/heme alignment. Human inducible NOS S562K and C563R mutants of an oxygenase/FMN (oxyFMN) construct were made by introducing charged residues at these sites as found in human neuronal NOS and endothelial NOS isoforms, respectively. The IET rate constant of the S562K mutant is notably decreased by one third, and its flavin fluorescence intensity per micromole per liter is diminished by approximately 24 %. These results suggest that a positive charge at position 562 destabilizes the hydrogen-bond-mediated NOS/CaM alignment, resulting in slower FMN–heme IET in the mutant. On the other hand, the IET rate constant of the C563R mutant is similar to that of the wild-type, indicating that the mutational effect is site-specific. Moreover, the human inducible NOS oxyFMN R536E mutant was constructed to disrupt the bridging CaM/NOS interaction, and its FMN–heme IET rate was decreased by 96 %. These results demonstrated a new role of the isoform-specific serine residue of the key CaM/FMN(NOS) bridging site in regulating the FMN–heme IET (possibly by tuning the alignment of the FMN and heme domains).
Keywords: Electron transfer; Nitric oxide synthase; Calmodulin; Laser flash photolysis; Heme

N–O bond cleavage mechanism(s) in nitrous oxide reductase by Mehmed Z. Ertem; Christopher J. Cramer; Fahmi Himo; Per E. M. Siegbahn (687-698).
Quantum chemical calculations of active-site models of nitrous oxide reductase (N2OR) have been undertaken to elucidate the mechanism of N–O bond cleavage mediated by the supported tetranuclear Cu4S core (CuZ) found in the enzymatic active site. Using either a minimal model previously employed by Gorelsky et al. (J. Am. Chem. Soc. 128:278–290, 2006) or a more extended model including key residue side chains in the active-site second shell, we found two distinct mechanisms. In the first model, N2O binds to the fully reduced CuZ in a bent μ-(1,3)-O,N bridging fashion between the CuI and CuIV centers and subsequently extrudes N2 while generating the corresponding bridged μ-oxo species. In the second model, substrate N2O binds loosely to one of the coppers of CuZ in a terminal fashion, i.e., using only the oxygen atom; loss of N2 generates the same μ-oxo copper core. The free energies of activation predicted for these two alternative pathways are sufficiently close to one another that theory does not provide decisive support for one over the other, posing an interesting problem with respect to experiments that might be designed to distinguish between the two. Effects of nearby residues and active-site water molecules are also explored.
Keywords: Density functional theory; Molecular modeling; Electronic structure; Transition state

Effect of reactivity on cellular accumulation and cytotoxicity of oxaliplatin analogues by Irina Buß; Ganna V. Kalayda; Andreas Lindauer; Michael R. Reithofer; Markus Galanski; Bernhard K. Keppler; Ulrich Jaehde (699-708).
The purpose of this study was to systematically investigate the relationships between reactivity, cellular accumulation, and cytotoxicity of a panel of oxaliplatin analogues with different leaving groups in human carcinoma cells. The reactivity of the complexes towards the nucleotides 2′-deoxyguanosine 5′-monophosphate and 2′-deoxyadenosine 5′-monophosphate was studied using capillary electrophoresis. Cellular accumulation and cytotoxicity were measured in an oxaliplatin-sensitive and oxaliplatin-resistant ileocecal colorectal adenocarcinoma cell line pair (HCT-8/HCT-8ox). Platinum concentrations were determined by flameless atomic absorption spectrometry. The 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay was used to assess cytotoxicity. Early cellular platinum accumulation was predominantly affected by lipophilicity. A relationship between reactivity and cellular accumulation was observed for three of four platinum complexes investigated, whereas the most lipophilic oxaliplatin analogue was an exception. Increased reactivity and reduced lipophilicity were associated with high cytotoxic activity. Resistance was influenced by lipophilicity but not by reactivity. The observed relationships may help in the design of analogues with high antitumoral activity in oxaliplatin-sensitive as well as oxaliplatin-resistant cells.
Keywords: Cytotoxicity; Influx; Oxaliplatin; Reactivity; Resistance

177Lu–DO3A–HSA–ZEGFR:1907: characterization as a potential radiopharmaceutical for radionuclide therapy of EGFR-expressing head and neck carcinomas by Susan Hoppmann; Shibo Qi; Zheng Miao; Hongguang Liu; Han Jiang; Cathy S. Cutler; Ande Bao; Zhen Cheng (709-718).
Epidermal growth factor receptor 1 (EGFR) is an attractive target for radionuclide therapy of head and neck carcinomas. Affibody molecules against EGFR (ZEGFR) show excellent tumor localizations in imaging studies. However, one major drawback is that radiometal-labeled Affibody molecules display extremely high uptakes in the radiosensitive kidneys which may impact their use as radiotherapeutic agents. The purpose of this study is to further explore whether radiometal-labeled human serum albumin (HSA)–ZEFGR bioconjugates display desirable profiles for the use in radionuclide therapy of EGFR-positive head and neck carcinomas. The ZEFGR analog, Ac–Cys–ZEGFR:1907, was site-specifically conjugated with HSA. The resulting bioconjugate 1,4,7,10-tetraazacyclododecane-1,4,7-triacetic acid (DO3A)–HSA–ZEGFR:1907 was then radiolabeled with either 64Cu or 177Lu and subjected to in vitro cell uptake and internalization studies using the human oral squamous carcinoma cell line SAS. Positron emission tomography (PET), single photon emission computed tomography (SPECT), and biodistribution studies were conducted using SAS-tumor-bearing mice. Cell studies revealed a high (8.43 ± 0.55 % at 4 h) and specific (0.95 ± 0.09 % at 4 h) uptake of 177Lu–DO3A–HSA–ZEGFR:1907 as determined by blocking with nonradioactive ZEGFR:1907. The internalization of 177Lu–DO3A–HSA–ZEGFR:1907 was verified in vitro and found to be significantly higher than that of 177Lu-labeled ZEFGR at 2–24 h of incubation. PET and SPECT studies showed good tumor imaging contrasts. The biodistribution of 177Lu–DO3A–HSA–ZEGFR:1907 in SAS-tumor-bearing mice displayed high tumor uptake (5.1 ± 0.44 % ID/g) and liver uptake (31.5 ± 7.66 % ID/g) and moderate kidney uptake (8.5 ± 1.08 % ID/g) at 72 h after injection. 177Lu–DO3A–HSA–ZEGFR:1907 shows promising in vivo profiles and may be a potential radiopharmaceutical for radionuclide therapy of EGFR-expressing head and neck carcinomas.
Keywords: 177Lu; Affibody; Epidermal growth factor receptor; Human serum albumin; Radionuclide therapy

Structural and functional insights into the heme-binding domain of the human soluble guanylate cyclase α2 subunit and heterodimeric α2β1 by Hongyan Wang; Fangfang Zhong; Jie Pan; Wei Li; Jihu Su; Zhong-Xian Huang; Xiangshi Tan (719-730).
Soluble guanylate cyclase (sGC) mediates NO signaling for a wide range of physiological effects in the cardiovascular system and the central nervous system. The α1β1 isoform is ubiquitously distributed in cytosolic fractions of tissues, whereas α2β1 is mainly found in the brain. The major occurrence and the unique characteristic of human sGC α2β1 indicate a special role in the mediation of neuronal communication. We have efficiently purified and characterized the recombinant heme-binding domain of the human sGC α2 subunit (hsGC α2H) and heterodimeric α2β1 (hsGC β1H–α2H) by UV–vis spectroscopy, circular dichrosim spectroscopy, EPR spectroscopy, and homology modeling. The heme dissociation and related NO/CO binding/dissociation of both hsGC α2H and hsGC β1H–α2H were investigated. The two truncated proteins interact with heme noncovalently. The CO binding affinity of hsGC α2H is threefold greater than that of human sGC α1H, whereas the dissociation constant k 1 for dissociation of NO from hsGC α2H is sevenfold larger than that for dissociation of NO from hsGC α1H, although k 2 is almost identical. The results indicate that in comparison with the α1β1 isoform, the brain α2β1 isoform exhibits a distinctly different CO/NO affinity and binding rate in favor of NO signaling, and this is consistent with its physiological role in the activation and desensitization. Molecular modeling and sequence alignments are consistent with the hypothesis that His105 contributes to the different CO/NO binding properties of different isoforms. This valuable information is helpful to understand the molecular mechanism by which human sGC α2β1 mediates NO/CO signaling.
Keywords: Human soluble guanylate cyclase; Human soluble guanylate cyclase α2β1; NO/CO singling; NO/heme dissociation

There is growing experimental evidence that tracing the elements involved in brain hyperexcitability, excitotoxicity, and/or subsequent neurodegeneration could be a valuable source of data on the molecular mechanisms triggering or promoting further development of epilepsy. The most frequently used experimental model of the temporal lobe epilepsy observed in clinical practice is the one based on pilocarpine-induced seizures. In the frame of this study, the elemental anomalies occurring for the rat hippocampal tissue in acute and silent periods after injection of pilocarpine in rats were compared. X-ray fluorescence microscopy was applied for the topographic and quantitative elemental analysis. The differences in the levels of elements such as P, S, K, Ca, Fe, Cu, and Zn between the rats 3 days (SE72) and 6 h (SE6) after pilocarpine injection as well as naive controls were examined. Comparison of SE72 and control groups showed, for specific areas of the hippocampal formation, lower levels of P, K, Cu, and Zn, and an increase in Ca accumulation. These results as well as further analysis of the differences between the SE72 and SE6 groups confirmed that seizure-induced excitotoxicity as well as mossy fiber sprouting are the mechanisms involved in the neurodegenerative processes which may finally lead to spontaneous seizures in the chronic period of the pilocarpine model. Moreover, in the light of the results obtained, Cu seems to play a very important role in the pathogenesis of epilepsy in this animal model. For all areas analyzed, the levels of this element recorded in the latent period were not only lower than those for controls but were even lower than the levels found in the acute period. The decreased hippocampal accumulation of Cu in the phase of behavior and EEG stabilization, a possible inhibitory effect of this element on excitatory amino acid receptors, and enhanced seizure susceptibility in Menkes disease (an inherited Cu transport disorder leading to Cu deficiency in the brain) suggest a neuroprotective role rather than neurodegenerative and proconvulsive roles of Cu in pilocarpine-induced epilepsy.
Keywords: Metal determination; X-ray microprobe; Neurochemistry

Cytochrome c nitrite reductase catalyzes the six-electron, seven-proton reduction of nitrite to ammonia without release of any detectable reaction intermediate. This implies a unique flexibility of the active site combined with a finely tuned proton and electron delivery system. In the present work, we employed density functional theory to study the recharging of the active site with protons and electrons through the series of reaction intermediates based on nitrogen monoxide [Fe(II)–NO+, Fe(II)–NO·, Fe(II)–NO, and Fe(II)–HNO]. The activation barriers for the various proton and electron transfer steps were estimated in the framework of Marcus theory. Using the barriers obtained, we simulated the kinetics of the reduction process. We found that the complex recharging process can be accomplished in two possible ways: either through two consecutive proton-coupled electron transfers (PCETs) or in the form of three consecutive elementary steps involving reduction, PCET, and protonation. Kinetic simulations revealed the recharging through two PCETs to be a means of overcoming the predicted deep energetic minimum that is calculated to occur at the stage of the Fe(II)–NO· intermediate. The radical transfer role for the active-site Tyr218, as proposed in the literature, cannot be confirmed on the basis of our calculations. The role of the highly conserved calcium located in the direct proximity of the active site in proton delivery has also been studied. It was found to play an important role in the substrate conversion through the facilitation of the proton transfer steps.
Keywords: Cytochrome c nitrite reductase; Proton-coupled electron transfer; Density functional theory; Heme enzymes

The coordination cage of the metal center in Fe(II)–bleomycin has been proposed to consist of the secondary amines in β-aminoalanine, the pyrimidinylpropionamide and imidazole rings, and the amide nitrogen in β-hydroxyhistidine as equatorial ligands, and the primary amine in β-aminoalanine and either the carbamoyl group in mannose or a solvent molecule occupying the axial sites. With the aim of supporting or not supporting coordination of a water molecule to the metal center in Fe(II)–bleomycin, the solution structure of Fe(II)–azide–bleomycin has been derived from NMR data. The structural changes that occur in Fe(II)–bleomycin upon azide binding have been monitored by comparing the experimental results with those obtained from the calculated structures for both bleomycin adducts. The results of this investigation strongly support a model of Fe(II)–bleomycin with six endogenous ligands as the most likely structure held in solution by this metallobleomycin in the absence of DNA.
Keywords: Bleomycin; Coordination chemistry; Glycopeptide antibiotic; Molecular dynamics

Application of DFT methods to the study of the coordination environment of the VO2+ ion in V proteins by Daniele Sanna; Vincent L. Pecoraro; Giovanni Micera; Eugenio Garribba (773-790).
Density functional theory (DFT) methods were used to simulate the environment of vanadium in several V proteins, such as vanadyl-substituted carboxypeptidase (sites A and B), vanadyl-substituted chloroplast F1-ATPase (CF1; site 3), the reduced inactive form of vanadium bromoperoxidase (VBrPO; low- and high-pH sites), and vanadyl-substituted imidazole glycerol phosphate dehydratase (IGPD; sites α, β, and γ). Structural, electron paramagnetic resonance, and electron spin echo envelope modulation parameters were calculated and compared with the experimental values. All the simulations were performed in water within the framework of the polarizable continuum model. The angular dependence of $$ left| {A_{ m{iso}}^{ m{N}} } ight| $$ and $$ left| {A_{z}^{ m{N}} } ight| $$ on the dihedral angle θ between the V=O and N–C bonds and on the angle φ between the V=O and V–N bonds, where N is the coordinated aromatic nitrogen atom, was also found. From the results it emerges that it is possible to model the active site of a vanadium protein through DFT methods and determine its structure through the comparison between the calculated and experimental spectroscopic parameters. The calculations confirm that the donor sets of sites B and A of vanadyl-substituted carboxypeptidase are [ $$ { ext{COO}}_{ ext{Glu}}^{ - } $$ , H2O, H2O, H2O] and [NHis(||), NHis(⊥), $$ { ext{COO}}_{ ext{Glu}}^{ - } $$ , H2O], and that the donor set of site 3 of CF1-ATPase is [ $$ { ext{COO}}_{ ext{Asp}}^{ - } $$ , OHThr, H2O, H2O, $$ { ext{NH}}_{{ 2 { ext{Lys}}}}^{ ext{ax}} $$ ]. For VBrPO, the coordination modes [NHis(||), NHis(∠), OHSer, H2O, H2Oax] for the low-pH site and [NHis(||), NHis(∠), OHSer, OH, H2Oax] or [NHis(||), NHis(∠), $$ { ext{O}}_{ ext{Ser}}^{ - } $$ , H2O] for the high-pH site, with an imidazole ring of histidine strongly displaced from the equatorial plane, can be proposed. Finally, for sites α, β, and γ of IGPD, the subsequent deprotonation of one, two, and three imidazole rings of histidine and the participation of a carboxylate group of a glutamate residue ([NHis(||), $$ { ext{COO}}_{ ext{Glu}}^{ - } $$ , H2O, H2O], [NHis(||), NHis(||), $$ { ext{COO}}_{ ext{Glu}}^{ - } $$ , H2O], and [NHis(||), NHis(||), $$ { ext{COO}}_{ ext{Glu}}^{ - } $$ , OH, $$ { ext{N}}_{ ext{His}}^{ ext{ax}} $$ ], respectively) seems to be the most plausible hypothesis.
Keywords: Vanadium; Proteins; Density functional theory methods; Electron paramagnetic resonance spectroscopy; Electron spin echo envelope modulation spectroscopy

Control of enzyme reaction by a designed metal-ion-dependent α-helical coiled-coil protein by Shigeo Murase; Sonoko Ishino; Yoshizumi Ishino; Toshiki Tanaka (791-799).
Regulation of protein function by external stimuli is a fascinating target for de novo design. We have constructed a peptide that assembles into a homotrimer in the presence of metal ions, such as Ni2+, Cu2+, and Zn2+. We fused the peptide construct to the DNA-binding domain (DBD) of the heat shock factor from Saccharomyces cerevisiae, which binds tandem repeats of the heat shock element (HSE). However, the fusion protein bound to the natural three tandem HSEs even in the absence of metal ions, although mainly as the dimerized protein. Using “skipped” HSEs containing six additional nucleotides inserted between two adjacent HSEs, to prevent interactions between the DBDs, we found the fusion protein bound to the new DNA target in a metal-ion-dependent manner, as monitored by a HindIII protection assay. The fusion protein containing two metal binding sites in the metal-ion-controlled domain inhibited RNA transcription by T7 RNA polymerase in the presence of metal ions, in a template containing skipped HSEs downstream of the T7 promoter. The designed protein therefore regulates the functions of the enzyme in a metal-ion-dependent manner.
Keywords: Coiled-coil; Protein engineering; Metalloregulation; DNA binding

This study elucidates the role of the protein structure in the catalysis of β-diketone cleavage at the three-histidine metal center of diketone cleaving enzyme (Dke1) by computational methods in correlation with kinetic and mutational analyses. Molecular dynamics simulations, using quantum mechanically deduced parameters for the nonheme Fe(II) cofactor, were performed and showed a distinct organization of the hydrophilic triad in the free and substrate-ligated wild-type enzyme. It is shown that in the free species, the Fe(II) center is coordinated to three histidines and one glutamate, whereas the substrate-ligated, catalytically competent enzyme–substrate complex has an Fe(II) center with three-histidine coordination, with a small fraction of three-histidine, one-glutamate coordination. The substrate binding modes and channels for the traffic of water and ligands (2,4-pentandionyl anion, methylglyoxal, and acetate) were identified. To characterize the impact of the hydrophobic protein environment around the metal center on catalysis, a set of hydrophobic residues close to the active site were targeted. The variations resulted in an up to tenfold decrease of the O2 reduction rates for the mutants. Molecular dynamics studies revealed an impact of the hydrophobic residues on the substrate stabilization in the active site as well as on the orientations of Glu98 and Arg80, which have previously been shown to be crucial for catalysis. Consequently, the Glu98–His104 interaction in the variants is weaker than in the wild-type complex. The role of protein structure in stabilizing the primary O2 reduction step in Dke1 is discussed on the basis of our results.
Keywords: Dke1; Fe2+ parameters in nonheme enzyme; Molecular dynamics; Mutants; Random acceleration molecular dynamics

The structure of the periplasmic nickel-binding protein NikA provides insights for artificial metalloenzyme design by Mickaël V. Cherrier; Elodie Girgenti; Patricia Amara; Marina Iannello; Caroline Marchi-Delapierre; Juan C. Fontecilla-Camps; Stéphane Ménage; Christine Cavazza (817-829).
Understanding the interaction of a protein with a relevant ligand is crucial for the design of an artificial metalloenzyme. Our own interest is focused on the synthesis of artificial monooxygenases. In an initial effort, we have used the periplasmic nickel-binding protein NikA from Escherichia coli and iron complexes in which N2Py2 ligands (where Py is pyridine) have been varied in terms of charge, aromaticity, and size. Six “NikA/iron complex” hybrids have been characterized by X-ray crystallography, and their interactions and solution properties have been studied. The hybrids are stable as indicated by their K d values, which are all in the micromolar range. The X-ray structures show that the ligands interact with NikA through salt bridges with arginine residues and π-stacking with a tryptophan residue. We have further characterized these interactions using quantum mechanical calculations and determined that weak CH/π hydrogen bonds finely modulate the stability differences between hybrids. We emphasize the important role of the tryptophan residues. Thus, our study aims at the complete characterization of the factors that condition the interaction of an artificial ligand and a protein and their implications for catalysis. Besides its potential usefulness in the synthesis of artificial monooxygenases, our approach should be generally applicable in the field of artificial metalloenzymes.
Keywords: Artificial metalloenzyme; Ligand binding; X-ray crystallography; Iron chemistry; CH/π bonds

Electron transfer between periplasmic formate dehydrogenase and cytochromes c in Desulfovibrio desulfuricans ATCC 27774 by Sofia Marques da Silva; Isabel Pacheco; Inês A. Cardoso Pereira (831-838).
Desulfovibrio spp. are sulfate-reducing organisms characterized by having multiple periplasmic hydrogenases and formate dehydrogenases (FDHs). In contrast to enzymes in most bacteria, these enzymes do not reduce directly the quinone pool, but transfer electrons to soluble cytochromes c. Several studies have investigated electron transfer with hydrogenases, but comparatively less is known about FDHs. In this work we conducted experiments to assess potential electron transfer pathways resulting from formate oxidation in Desulfovibrio desulfuricans ATCC 27774. This organism can grow on sulfate and on nitrate, and contains a single soluble periplasmic FDH that includes a cytochrome c 3 like subunit (FdhABC3). It has also a unique cytochrome c composition, including two cytochromes c not yet isolated from other species, the split-Soret and nine-heme cytochromes, besides a tetraheme type I cytochrome c 3 (TpIc 3). The FDH activity and cytochrome composition of cells grown with lactate or formate and nitrate or sulfate were determined, and the electron transfer between FDH and these cytochromes was investigated. We studied also the reduction of the Dsr complex and of the monoheme cytochrome c-553, previously proposed to be the physiological partner of FDH. FdhABC3 was able to reduce the c-553, TpIc 3, and split-Soret cytochromes with a high rate. For comparison, the same experiments were performed with the [NiFe] hydrogenase from the same organism. This study shows that FdhABC3 can directly reduce the periplasmic cytochrome c network, feeding electrons into several alternative metabolic pathways, which explains the advantage of not having an associated membrane subunit.
Keywords: Formate dehydrogenase; Cytochrome; Sulfate-reducing bacteria; Desulfovibrio ; Hydrogenase

A note from the President by José J. G. Moura (839-839).