Search

LC8 is a ubiquitous hub protein that binds intrinsically disordered proteins and promotes their assembly into higher-order complexes. A common feature among the more than 100 essential LC8 binding proteins is that in the 10-12-amino acid recognition sequence there is a conserved QT motif but variable amino acids N- and C-terminal to the QT pair. The sequence diversity among LC8 binding partners implies that structural factors also contribute to specificity. To investigate whether one such factor is the transient secondary structure favored by an LC8 binding sequence, we report here a molecular ensemble description of IC TL , a domain of the dynein intermediate chain that includes binding sites for light chains LC8 and Tctex1. Nuclear magnetic resonance secondary chemical shifts and residual dipolar coupling values combined with ensemble generation and selection algorithms indicate a deviation from statistical (random) coil behavior with an elevated population of polyproline II (PPII) conformations for the IC TL regions that bind LC8 and Tctex1. Independent measurements of one- and three-bond scalar couplings confirm the PPII transient secondary structure propensity. Given that in the IC/Tctex1/LC8 ternary complex IC TL forms a β-strand at the interface of Tctex1 and LC8, we hypothesize that a PPII conformation may facilitate its initial docking and insertion into the binding cleft of the β-sheet LC8 dimer interface. Molecular ensemble calculations for intrinsically disordered LC8 binding partners also reveal PPII conformational sampling within and proximate to the LC8 recognition motifs, suggesting that a preference for a PPII conformation is general for LC8 binding partners.

RimO, a radical-S-adenosylmethionine (SAM) enzyme, catalyzes the specific C 3 methylthiolation of the D89 residue in the ribosomal S 12 protein. Two intact iron-sulfur clusters and two SAM cofactors both are required for catalysis. By using electron paramagnetic resonance, Mössbauer spectroscopies, and site-directed mutagenesis, we show how two SAM molecules sequentially bind to the unique iron site of the radical-SAM cluster for two distinct chemical reactions in RimO. Our data establish that the two SAM molecules bind the radical-SAM cluster to the unique iron site, and spectroscopic evidence obtained under strongly reducing conditions supports a mechanism in which the first molecule of SAM causes the reoxidation of the reduced radical-SAM cluster, impeding reductive cleavage of SAM to occur and allowing SAM to methylate a HS - ligand bound to the additional cluster. Furthermore, by using density functional theory-based methods, we provide a description of the reaction mechanism that predicts the attack of the carbon radical substrate on the methylthio group attached to the additional [4Fe-4S] cluster.

Fur is a bacterial regulator using iron as a cofactor to bind to specific DNA sequences. This protein exists in solution as several oligomeric states, of which the dimer is generally assumed to be the biologically relevant one. We describe the equilibria that exist between dimeric Escherichia coli Fur and higher oligomers. The dissociation constant for the dimer-tetramer equilibrium is estimated to be in the millimolar range. Oligomerization is enhanced at low ionic strength and pH. The as-isolated monomeric form of Fur is not in equilibrium with the dimer and contains two disulfide bridges (C92-C95 and C132-C137). Binding of the monomer to DNA is metal-dependent and sequence specific with an apparent affinity 5.5 times lower than that of the dimer. Size exclusion chromatography, EDC cross-linking, and CD spectroscopy show that reconstitution of the dimer from the monomer requires reduction of the disulfide bridges and coordination of Zn2+. Reduction of the disulfide bridges or Zn2+ alone does not promote dimerization. EDC and DMA cross-links reveal that the N-terminal NH2 group of one subunit is in an ionic interaction with acidic residues of the C-terminal tail and close to Lys76 and Lys97 of the other. Furthermore, the yields of cross-link drastically decrease upon binding of metal in the activation site, suggesting that the N-terminus is involved in the conformational change. Conversely, oxidizing reagents, H2O2 or diamide, disrupt the dimeric structure leading to monomer formation. These results establish that coordination of the zinc ion and the redox state of the cysteines are essential for holding E. coli Fur in a dimeric state.

Structure-function relationships of the membrane-embedded Saccharomyces cerevisiae mitochondrial ADP/ATP carrier were investigated through two independent approaches, namely, limited proteolysis and cysteine labeling. Experiments were carried out in the presence of either carboxyatractyloside (CATR) or bongkrekic acid (BA), two specific inhibitors of the ADP/ATP transport that bind to two distinct conformers involved in the translocation process. The proteolysis approach allowed us to demonstrate (i) that N- and C-terminal extremities of ADP/ATP carrier are facing the intermembrane space and (ii) that the central region of the carrier corresponding to the matrix loop m2 is accessible to externally added trypsin in a conformation-sensitive manner, being cleaved at the Lys163-Gly164 and Lys178-Thr179 bonds in the carrier-CATR and the carrier-BA complexes, respectively. The cysteine labeling approach was carried out on the S161C mutant of the ADP/ATP carrier. This variant of the carrier is fully active, displaying nucleotide transport kinetic parameters and inhibitor binding properties similar to that of wild-type carrier. Alkylation experiments, carried out on mitochondria with the nonpermeable reagents eosin-5-maleimide and iodoacetamidyl-3,6-dioxaoctanediamine-biotin, showed that Cys 161 is accessible from the outside in the carrier-CATR complex, whereas it is masked in the carrier-BA complex. Taken together, our results indicate that the matrix loop m2 connecting the transmembrane helices H3 to H4 intrudes to some extent into the inner mitochondrial membrane. Its participation in the translocation of ADP/ATP is strongly suggested, based on the finding that its accessibility to reagents added outside mitochondria is modified according to the conformational state of the carrier.

Microtubules assembled from purified tubulin in vitro are labile, rapidly disassembling when exposed to a variety of depolymerizing conditions such as cold temperature. In contrast, in many cell types, microtubules seem to be unaffected when the cell is exposed to the cold. This resistance of microtubules to the cold has been intriguing because the earliest and by far most studied microtubule-associated proteins such as MAP2 and tau are devoid of microtubule cold stabilizing activity. Over the past several years, it has been shown that resistance of microtubules to the cold is largely due to polymer association with a class of microtubule-associated proteins called STOPs. STOPs are calmodulin-binding and calmodulin-regulated proteins which, in mammals, are encoded by a single gene but exhibit substantial cell specific variability due to mRNA splicing and alternative promoter use. STOP microtubule stabilizing activity has been ascribed to two classes of new bifunctional calmodulin- and microtubule-binding motifs, with distinct microtubule binding properties in vivo. STOPs seem to be restricted to vertebrates and are composed of a conserved domain split by the apparent insertion of variable sequences that are completely unrelated among species. Recently, STOP suppression in mice has been found to induce synaptic defects associated with neuroleptic-sensitive behavioral disorders. Thus, STOPs are important for synaptic plasticity. Additionally, STOP-deficient mice may yield a pertinent model for the study of neuroleptics in illnesses such as schizophrenia, currently thought to result from defects in synapse function.

Exposure to solar UV radiation gives rise to mutations that may lead to skin cancer. UVA (320-340 nm) constitutes the large majority of solar UV radiation but is less effective than UVB (290-320 nm) at damaging DNA. Although UVA has been implicated in photocarcinogenesis, its contribution to sunlight mutagenesis has not been elucidated, and DNA damage produced by UVA remains poorly characterized. We employed HPLC-MS/MS and alkaline agarose gel electrophoresis in conjunction with the use of specific DNA repair proteins to determine the distribution of the various classes and types of DNA lesions, including bipyrimidine photoproducts, in Chinese hamster ovary cells exposed to pure UVA radiation, as well as UVB and simulated sunlight (lambda > 295 nm) for comparison. At UVA doses compatible with human exposure, oxidative DNA lesions are not the major type of damage induced by UVA. Indeed, single-strand breaks, oxidized pyrimidines, oxidized purines (essentially 8-oxo-7,8-dihydroguanine), and cyclobutane pyrimidine dimers (CPDs) are formed in a 1:1:3:10 ratio. In addition, we demonstrate that, in contrast to UVB and sunlight, UVA generates CPDs with a large predominance of TT CPDs, which strongly suggests that they are formed via a photosensitized triplet energy transfer. Moreover, UVA induces neither (6-4) photoproducts nor their Dewar isomers via direct absorption. We also show that UVA photons contained in sunlight, rather than UVB, are implicated in the photoisomerization of (6-4) photoproducts, a quickly repaired damage, into poorly repaired and highly mutagenic Dewar photoproducts. Altogether, our data shed new light on the deleterious effect of UVA.

Analysis of the genome of the hyperthermophilic bacterium Aquifex aeolicus has revealed the presence of a previously undetected gene potentially encoding a plant- and mammalian-type [2Fe-2S] ferredoxin. Expression of that gene in Escherichia coli has yielded a novel thermostable [2Fe-2S] ferredoxin (designated ferredoxin 5) whose sequence is most similar to those of ferredoxins involved in the assembly of iron-sulfur clusters (Isc-Fd). It nevertheless differs from the latter proteins by having deletions near its N- and C-termini, and no cysteine residues other than those involved in [2Fe-2S] cluster coordination. Resonance Raman, low-temperature MCD and EPR studies show close spectral similarities between ferredoxin 5 and the Isc-Fd from Azotobacter vinelandii. Mössbauer spectra of the reduced protein were analyzed with an S = 1/2 spin Hamiltonian and interpreted in the framework of the ligand field model proposed by Bertrand and Gayda. The redox potential of A. aeolicus ferredoxin 5 (-390 mV) is in keeping with its relatedness to Isc-Fd. Unfolding experiments showed that A. aeolicus ferredoxin 5 is highly thermostable (T(m) = 106 degrees C at pH 7), despite being devoid of features (e.g., high content of charged residues) usually associated with extreme thermal stability. Searches for genes potentially encoding plant-type [2Fe-2S] ferredoxins have been performed on the sequenced genomes of hyperthermophilic organisms. None other than the two proteins from A. aeolicus were retrieved, indicating that this otherwise widely distributed group of proteins is barely represented among hyperthermophiles.

The flavodoxin-like domain, missing in the three-dimensional structure of the monomeric, simplified model of the Escherichia coli sulfite reductase flavoprotein component (SiR-FP), has now been expressed independently. This 168 amino acid protein was named SiR-FP18 with respect to its native molecular weight and represents the FMN-binding domain of SiR-FP. This simplified biological object has kept the main characteristics of its counterpart in the native protein. It could incorporate FMN exclusively and stabilize a neutral air-stable semiquinone radical. Both the radical and the fully reduced forms of SiR-FP18 were able to transfer their electrons to DCPIP or cytochrome c quantitatively. SiR-FP18 was able to form a highly stable complex with SiR-HP, the hemoprotein component of the sulfite reductase containing an iron-sulfur cluster coupled to a siroheme. In agreement with the postulated catalytic cycle of SiR-FP, only the fully reduced form of SiR-FP18 could transfer one electron to SiR-HP, the transferred electron being localized exclusively on the heme. As isolated SiR-FP18 has kept the main characteristics of the FMN-binding domain of the native protein, a structural analysis by NMR was performed in order to complete the partial structure obtained previously. Structural modeling was performed using sequence homologues, cytochrome P450 reductase (CPR; 29% identity) and bacterial cytochrome P450 (P450-BM3; 26% identity), as conformational templates. These sequences were anchored using common secondary structural elements identified from heteronuclear NMR data measured on the protein backbone. The resulting structural model was validated, and subsequently refined using residual (C(alpha)-C', N-H(N), and C'-H(N)) dipolar couplings measured in an anisotropic medium. The overall fold of SiR-FP18 is very similar to that of bacterial flavodoxins and of the flavodoxin-like domain in CPR or P450-BM3.

A [2Fe-2S] ferredoxin (Fd1) from the hyperthermophilic bacterium Aquifex aeolicus has been obtained by heterologous expression of the encoding gene in Escherichia coli. Sequence comparisons show that this protein belongs to the extended family of plant- and mammalian-type [2Fe-2S] ferredoxins but also indicate that it is not closely similar to either the plant-type or mammalian-type subfamilies. Instead, it appears to bear some similarity to novel members of this family, in particular the Isc-type ferredoxins involved in the assembly of iron-sulfur clusters in vivo. The two redox levels of the [2Fe-2S](2+/+) metal site of A. aeolicus ferredoxin have been studied by UV-visible, resonance Raman, EPR, variable temperature magnetic circular dichroism, and Mössbauer spectroscopies. A full-spin Hamiltonian analysis is given for the Mössbauer spectra. In aggregate, the spectroscopic data reveal differences with both the plant-type and mammalian-type ferredoxins, in keeping with the sequence comparisons. The midpoint potential of the [2Fe-2S](2+/+) couple, at -375 mV versus the normal hydrogen electrode, is more negative than those of mammalian-type ferredoxins and at the upper end of the range covered by plant-type ferredoxins. A. aeolicus ferredoxin contains two cysteines in addition to the four that are committed as ligands of the [2Fe-2S] cluster. These two residues have been shown by chemical modification and site-directed mutagenesis to form a disulfide bridge in the native protein. While that cystine unit plays a significant role in the exceptional thermostability of A. aeolicus ferredoxin (T(m) = 121 degrees C at pH 7 versus T(m) = 113 degrees C in a molecular variant where the disulfide bridge has been removed), it does not bear on the properties of the [2Fe-2S](2+/+) chromophore. This observation is consistent with the large distance (ca. 20 A) that is predicted to separate the iron-sulfur chromophore from the disulfide bridge.

The low molecular weight GTP binding protein Rac is essential to the activation of the NADPH oxidase complex, involved in pathogen killing during phagocytosis. In resting cells, Rac exists as a heterodimeric complex with Rho GDP dissociation inhibitor (Rho-GDI). Two types of interactions exist between Rac and Rho-GDI: a protein-lipid interaction, implicating the polyisoprene of the GTPase, as well as protein-protein interactions. Using the two-hybrid system, we show that nonprenylated Rac1 interacts very weakly with Rho-GDI, pointing to the predominant role of protein-isoprene interaction in complex formation. In the absence of this strong interaction, we demonstrate that three sites of protein-protein interaction, Arg66(Rac)-Leu67(Rac), His103(Rac), and the C-terminal polybasic region Arg183(Rac)-Lys188(Rac), are involved and cooperate in complex formation. When Rac1 mutants are prenylated by expression in insect cells, they all interact with Rho-GDI. Rho-GDI is able to exert an inhibitory effect on the GDP/GTP exchange reaction except in the complex in which Rac1 has a deletion of the polybasic region (Arg183(Rac)-Lys188(Rac)). This complex is, most likely, held together through protein-lipid interaction only. Although able to function as GTPases, the mutants of Rac1 that failed to interact with Rho-GDI also failed to activate the NADPH oxidase in a cell-free assay after loading with GTP. Mutant Leu119(Rac)Gln could both interact with Rho-GDI and activate the NADPH oxidase. The Rac1/Rho-GDI and Rac1(Leu119Gln)/Rho-GDI complexes, in which the GTPases were bound to GDP, were found to activate the oxidase efficiently. These data suggest that Rho-GDI stabilizes Rac in an active conformation, even in the GDP-bound state, and presents it to its effector, the p67phox component of the NADPH oxidase.

Bipyrimidine photoproducts induced in DNA by UVB radiation include cyclobutane dimers, (6-4) photoproducts, and their related Dewar valence isomers. Even though these lesions have been extensively studied, their rate of formation within DNA is still not known for each possible bipyrimidine site (TT, TC, CT, and CC). Using a method based on the coupling of liquid chromatography to mass spectrometry, we determined the distribution of the 12 possible bipyrimidine photoproducts within isolated and cellular DNA. TT and TC were found to be the most photoreactive sequences, whereas lower amounts of damage were produced at CT and CC sites. In addition to this quantitative aspect, sequence effects were observed on the relative yield of (6-4) adducts with respect to cyclobutane pyrimidine dimers. Another interesting result is the lack of formation of Dewar valence isomers in detectable amounts within the DNA of cells exposed to low doses of UVB radiation. The photoproduct distribution obtained does not fully correlate with the UV mutation spectrum. A major striking observation deals with the low yield of cytosine-cytosine photoproducts which are likely to be associated with the UV-specific CC to TT tandem mutation.

A novel photoactivatable radioactive ADP derivative, namely, 2-azido-3'-O-naphthoyl-[beta-(32)P]ADP (2-azido-N-[(32)P]ADP), was synthesized with the aim at mapping the substrate binding site(s) of the yeast mitochondrial ADP/ATP carrier. It was used with mitochondria isolated from genetically modified strains of Saccharomyces cerevisiae, producing the native or the His-tagged Anc2p isoform of the carrier. In darkness, 2-azido-N-[(32)P]ADP was reversibly bound to the carrier in mitochondria, without being transported. Upon photoirradiation, only the ADP/ATP carrier was covalently radiolabeled among all mitochondrial proteins. Specificity of labeling was demonstrated since carboxyatractyloside (CATR), a potent inhibitor of ADP/ATP transport, totally prevented the incorporation of the photoprobe. To localize the radioactive region(s), the purified photolabeled carrier was submitted to CNBr or hydroxylamine cleavage. The resulting fragments were characterized and identified by SDS-PAGE, Western blotting, amino acid sequencing, and MALDI-MS and ESI-MS analyses. Two short photolabeled distinct segments, eight and nine residues long, were identified: S183-R191, located in the central part of the ADP/ATP carrier; and I311-K318, belonging to its C-terminal end. Plausible models of organization of the nucleotide binding site(s) of the carrier involving the two regions specifically labeled by 2-azido-N-[(32)P]ADP are proposed.

The Fur apoprotein has been purified and reconstituted with Co2+ and Mn2+ ions. These samples have been analyzed by UV-visible, EPR, and 1H NMR spectroscopies, by XAS, and by magnetization measurements. The apo-Fur protein is able to bind one metal dication (Co2+ or Mn2+) per monomer. A saturation magnetization study confirms the presence of a high-spin metal dication [Mn(II) S = 5/2 and Co(II) S = 3/2]. The two metal ions per Fur dimer are not in magnetic interaction (|J| < 0.1 cm-1 ). The UV-visible spectrum of the cobalt-substituted form (Co-Fur) presents two main bands at 660 nm and 540(br) nm with epsilon540 nm = 65 M-1 cm-1. The EPR spectrum gives the following g values: gx = 5.0(5), gy = 4.0(2), and gz = 2. 3(1), which are in accordance with a nearly axial (E/D < 0.11) site. The value of 55 cm-1 for the splitting (Delta) between the ground and the first excited state has been derived from an EPR saturation study and is in agreement with magnetization data. The EXAFS data of Co-Fur indicate a metal environment comprising five nitrogen/oxygen atoms at 2.11 A, the absence of sulfur, and the presence of histidines as ligands. 1H NMR of Co-Fur in H2O and D2O shows at least two exchangeable signals coming from histidine NH protons and shows the signature of carboxylate group(s). The combined spectroscopic data allow us to propose that the main metal site of Fur in Co-Fur contains at least two histidines, at least one aspartate or glutamate, and no cysteine as ligands and is in an axially distorted octahedral environment.

The redox core of the neutrophil NADPH oxidase complex is a membrane-bound flavocytochrome b in which FAD and heme b are the two prosthetic redox groups. Both FAD and heme b are able to react with diphenylene iodonium (DPI) and iodonium biphenyl (IBP), two inhibitors of NADPH oxidase activity. In this study, we show that the iodonium modification of heme b contributes predominantly to the inhibition of NADPH oxidase. This conclusion is based on the finding that both iodonium compounds decreased the absorbance of the Soret peak of flavocytochrome b in neutrophil membranes incubated with NADPH, and that this decrease was strictly correlated with the loss of oxidase activity. Furthermore, the heme component of purified flavocytochrome b reduced to no more than 95% by a limited amount of sodium dithionite could be oxidized by DPI or IBP. Butylisocyanide which binds to heme iron precludes heme b oxidation. In activated neutrophil membranes, competitive inhibition of O2 uptake by DPI or IBP occurred transiently and was followed by a noncompetitive inhibition. These results, together with those of EPR spectroscopy experiments, lead us to postulate that DPI or IBP first captures an electron from the reduced heme iron of flavocytochrome b to generate a free radical. Then, the binding of this radical to the proximate environment of the heme iron, most probably on the porphyrin ring, results in inhibition of oxidase activity. In the presence of an excess of sodium dithionite, DPI and IBP produced a biphasic decrease of the Soret band of flavocytochrome b, with a break in the dose effect curve occurring at 50% of the absorbance loss. This was consistent with the presence of two hemes in flavocytochrome b that differ by their sensitivity to DPI or IBP.

Oligonucleotides that contain a single modified pyrimidine, i.e., thymine glycol (Tg), 5,6-dihydrothymine (DHT), and 5-hydroxycytosine (5-OHC) were synthesized in order to investigate the substrate specificity and the excision mechanism of two Escherichia coli repair enzymes: endonuclease III and formamidopyrimidine DNA glycosylase (Fpg). Three techniques of analysis were employed. A gas chromatography-mass spectrometry (GC-MS) assay with HPLC prepurification was used to quantify the release of the modified bases, while polyacrylamide gel electrophoresis and matrix-assisted laser-desorption ionization-mass spectrometry (MALDI-MS) provided insights into the mechanism of oligonucleotide cleavage. Values of Vm/Km constants lead to the conclusion that the substrates are processed by endonuclease III with the following preference: Tg >> 5-OHC > DHT. This confirms that Tg is an excellent substrate for endonuclease III. Fpg-mediated cleavage of the 5-OHC-containing oligonucleotide is processed at the same rate than endonuclease III. Furthermore, Fpg was found to have a little but relevant activity on DHT-containing oligonucleotide, thus broadening the substrate specificity of this enzyme to a new modified pyrimidine. While 5-OHC-containing oligonucleotides are cleaved by the two enzymes, no or a small amount of the modified base was found to be released, as determined by GC-MS. From these data it may be suggested that 5-OHC could be modified during its enzymatic excision. Finally, MALDI-MS analyses shed new light on the mechanism of action of endonuclease III: the molecular masses of the repaired fragments of 5-OHC- and DHT-containing oligonucleotides showed that endonuclease III cleaves the DNA backbone mainly through a hydrolytic process and that no beta-elimination product was detected.

An N-terminal domain of Clostridium pasteurianum hydrogenase I, encompassing 76 residues out of the 574 composing the full-size enzyme, had previously been overproduced in Escherichia coli and shown to form a stable fold around a [2Fe-2S] cluster. This domain displays only marginal sequence similarity with [2Fe-2S] proteins of known structure, and therefore, two-dimensional 1H NMR has been implemented to elucidate features of the polypeptide fold. Despite the perturbing presence of the paramagnetic [2Fe-2S] cluster, 57 spin systems were detected in the TOCSY spectra, 52 of which were sequentially assigned through NOE connectivities. Several secondary structure elements were identified. The N terminus of the protein consists of two antiparallel beta strands followed by an alpha helix contacting both strands. Two additional antiparallel beta strands, one of them at the C terminus of the sequence, form a four-stranded beta sheet together with the two N-terminal strands. The proton resonances that can be attributed to this beta2alphabeta2 structural motif undergo no paramagnetic perturbations, suggesting that it is distant from the [2Fe-2S] cluster. In plant- and mammalian-type ferredoxins, a very similar structural pattern is found in the part of the protein farthest from the [2Fe-2S] cluster. This indicates that the N-terminal domain of C. pasteurianum hydrogenase folds in a manner very similar to those of plant- and mammalian-type ferredoxins over a significant part (ca. 50%) of its structure. Even in the vicinity of the metal site, where 1H NMR data are blurred by paramagnetic interactions, the N-terminal domains of hydrogenase and mammalian- and plant-type ferredoxins most likely display significant structural similarity, as inferred from local sequence alignments and from previously reported circular dichroism and resonance Raman spectra. These data afford structural information on a kind of [2Fe-2S] cluster-containing domain that occurs in a number of redox enzymes and complexes. In addition, together with previously published sequence alignments, they highlight the widespread distribution of the plant-type ferredoxin fold in bioenergetic systems encompassing anaerobic metabolism, photosynthesis, and aerobic respiratory chains.

The primary structure of Clostridium pasteurianum hydrogenase I appears to be composed of modules suggesting that the various iron-sulfur clusters present in this enzyme might be segregated in structurally distinct domains. On the basis of this observation, a gene fragment encoding the 76 N-terminal residues of this enzyme has been expressed in Escherichia coli. The polypeptide thus produced contains a [2Fe-2S]n+ cluster of which the oxidized level (n = 2) has been monitored by UV-visible absorption, circular dichroism, and resonance Raman spectroscopy. This cluster can be reduced by dithionite or electrochemically to the n = 1 level which has been investigated by EPR and by low-temperature magnetic circular dichroism. The redox potential of the +2 to +1 transition is -400 mV (vs the normal hydrogen electrode). The spectroscopic and redox results indicate a [2Fe-2S]2+/+ chromophore coordinated by four cysteine ligands in a protein fold similar to that found in plant- and mammalian-type ferredoxins. Among the five cysteines present in the N-terminal hydrogenase fragment, four (in positions 34, 46, 49, and 62) are conserved in other sequences and are therefore the most likely ligands of the [2Fe-2S] site. The fifth cysteine, in position 39, can be dismissed on the grounds that the Cys39Ala mutation does not alter any of the properties of the iron-sulfur cluster. The spectroscopic signatures of this chromophore are practically identical with some of those reported for full-size hydrogenase. This confirms that C. pasteurianum hydrogenase I contains a [2Fe-2S] cluster and indicates that the polypeptide fold around the metal site of the N-terminal fragment is very similar, if not identical, to that occurring in the full-size protein. The N-terminal sequence of this hydrogenase is homologous to sequences of a number of proteins or protein domains, including a subunit of NADH-ubiquinone oxidoreductase of respiratory chains. From that, it can be anticipated that the structural domain isolated and described here is a building block of electron transfer complexes involved in various bioenergetic processes.

The [2Fe-2S] cluster of the ferredoxin from Clostridium pasteurianum is coordinated by cysteines 11, 56, and 60 and by a fourth cysteine, residue 24 in the wild-type protein, located on a flexible and deletable loop around residues 14-30. New mutated forms of this ferredoxin show that the fourth cysteine ligand can be located in any one of positions 14, 16, 21, 24, or 26. Another set of molecular variants has unveiled a new case of ligand swapping on the cysteine 60 ligand site. Replacement of cysteine 60 by alanine and introduction of a cysteine in position 21 yielded a ferredoxin that assembles a [2Fe-2S] cluster of which the ligands are cysteines 11, 21, 24, and 56. This cysteine ligand pattern is similar to that occurring in plant-type or mammalian-type ferredoxins, although the overall sequence similarities are below detection. Moreover, the vibrational and electronic properties of the resulting [2Fe-2S]2+/+ center, as revealed by resonance Raman and EPR studies, are strikingly similar to those of mammalian-type ferredoxins. The extensive set of mutated forms of the C. pasteurianum ferredoxin now available indicates that cysteine ligand exchange may occur on residues 24 and 60, but not on residues 11 and 56. It is thus suggested that cysteines 24 and 60 are part of a solvent accessible aspect of the Fe-S cluster, whereas cysteines 11 and 56 are buried and form the more rigid part of the polypeptide ligand framework. In view of the unprecedented versatility of this [2Fe-2S] cluster and of its polypeptidic environment, the introduction of ligands other than cysteine in various positions has been attempted. These experiments have remained unsuccessful, and even including previous studies, noncysteinyl ligation has been obtained with this protein in only very few cases. The data provide an extensive confirmation that Fe-S clusters have a strong preference for thiolate ligation and rationalize the relatively rare occurrence of noncysteinyl ligation in native Fe-S proteins.

The flavoprotein component (SiR-FP) of the sulfite reductase from Escherichia coli is an octamer containing one FAD and one FMN per polypeptide chain. SiR-FP60, a SiR-FP fragment starting with alanine-52, was overexpressed in E. coli and purified as a monomer. The N-terminal part of the native protein contains thus all the determinants required for the polymerization. SiR-FP60 retains both FAD and FMN with comparable contributions of the two flavins and the catalytic properties of SiR-FP. Thus, SiR-FP60 can be considered as a reliable simplified model of the sulfite reductase flavoprotein component. The formation and the stabilization of the neutral FMN semiquinone is thermodynamically favorable in SiR-FP60 upon reduction with photoreduced deazaflavin, dithionite, or NADPH. Generation of FMNH* is explained from a disproportionation of electrons between the reduced and oxidized FMN moieties during an intermolecular reaction, as shown with SiR-FP23, the FMN-binding domain of SiR-FP. The neutral FAD semiquinone can be observed only within SiR-FP43, the isolated FAD-binding domain. NADPH was used as a titrant or in excess to demonstrate that electron transfer is possible only because the FMN cofactor is coupled to FAD as an electron acceptor in the protein. The electron distribution within the various reduced forms of SiR-FP60 has been compared with that of the reduced forms of cytochrome P450 reductase, bacterial cytochrome P450, and nitric-oxide synthase. Despite the conservation of the bi-flavin-domain structure between these proteins over evolutionary time, each of them provides significantly different flavin reactivities.

The zinc K-edge X-ray absorption spectra of the Fur (ferric uptake regulation) protein isolated from Escherichia coli have been analyzed in frozen solution to determine details of the zinc coordination. The spectra of apoFur and of the cobalt-substituted protein have been analyzed and compared in order to see the influence of the cobalt incorporation on the geometry of the zinc site. EXAFS analysis gave for both samples (apoFur and CoFur) a tetrahedral environment for the zinc atom with two sulfur donor ligands at a distance of 2.3 A from the zinc and two N/O donor ligands at 2.0 A. The two sulfur donor ligands are probably two of the four cysteines present in each Fur monomer and could be Cys92 and Cys95, which are known from mutagenesis studies to be essential for Fur activity [Coy, M., Doyle, C., Besser, J., and Neilands, J. B. (1994) BioMetals 7, 292-298]. The distances obtained from our fits were always too short to be compatible with penta or hexa coordination. The typical pattern observed for the Fourier transform of the EXAFS oscillations suggests the presence of at least one imidazole ligand. The XANES of these two forms of the protein are similar but significantly different. This suggests a change of the conformation of the zinc site upon cobalt incorporation. The present study provides the first unambiguous evidence for the presence of a structural zinc site in the Fur protein from Escherichia coli.

The calcium-dependent interaction between C1r and C1s, the two homologous serine proteases of the first component of human complement C1, is mediated by their N-terminal regions. The latter comprise an epidermal growth factor (EGF)-like module exhibiting the consensus sequence characteristic of Ca(2+)-binding EGF modules, surrounded by two CUB modules. Due to its Ca2+ binding ability, the C1r EGF-like module (C1r-EGF) is supposed to participate in the C1r-C1s interaction. An additional interesting feature of C1r-EGF is the unusually large loop connecting the first two conserved cysteine residues. The solution structure of synthetic C1r-EGF (residues 123-175) has been determined using nuclear magnetic resonance and combined simulated annealing-restrained molecular dynamics calculations. The resulting family of 19 structures is characterized by a well-ordered C-terminal part (residues Cys 144-Ala174) with a backbone rmsd of 0.7 A and a disordered N-terminal, including the large loop between the first two cysteines (Cys129 and Cys144). This loop is known to be surface exposed and may be expected to participate in domain-domain or protein-protein interactions. In its C-terminal part, C1r-EGF possesses the characteristic EGF fold with a major and a minor beta-sheet. The latter comprises a beta-bulge, and comparison with other EGF-like modules reveals the existence of two distinct structural and sequential motifs in the bulged part. Additional experiments in the presence of 80 mM Ca2+ did not show significant structural variation of C1r-EGF, in keeping with previous observations on blood-clotting factors IX and X.

The rubredoxin from Clostridium pasteurianum contains a single iron atom bound to the polypeptide chain by cysteines 6, 9, 39, and 42. The C42A variant of this protein has been prepared by site-directed mutagenesis and heterologous expression of the gene in Escherichia coli. The mutated protein was found to contain an unexpected chromophore that has been characterized by a variety of techniques. UV-visible absorption and resonance Raman spectra were strongly reminiscent of those of [2Fe-2S] proteins. Mössbauer spectra of the oxidized chromophore isolated in oxygen-free conditions indicated low-temperature diamagnetism resulting from antiferromagnetically coupled high-spin ferric ions. Analysis of X-ray absorption fine structure spectra yielded an Fe-Fe distance of 2.68 A. Colorimetric assays of iron and inorganic sulfide showed that the two elements are present in a 1:1 ratio. Electrospray-ionization mass spectra displayed a major component at M = 6190 Da, i.e. the molecular mass of the C42A apoprotein plus two atomic masses of iron and two atomic masses of sulfur. Taken together, these data show that a mere point mutation allows the stabilization of a binuclear [2Fe-2S] cluster in a protein that normally accommodates a mononuclear Fe(Scys)4 site. Assembly of a [2Fe-2S] cluster may occur because rubredoxin assumes a similar fold around its metal center as the [2Fe-2S] Rieske protein. Alternatively, a more extensive structural rearrangement of the polypeptide chain of the C42A rubredoxin variant may be considered as well.

Putative physiological partners of the [2Fe-2S] ferredoxin from Clostridium pasteurianum have been searched by running crude soluble extracts of this bacterium through an affinity column to which the [2Fe-2S] ferredoxin had been covalently bound. Subsequent washing of the column with buffers of increasing ionic strength revealed a strong and specific interaction of the ferredoxin with the MoFe protein of nitrogenase. This interaction was further investigated by performing cross-linking experiments with mixtures of the two purified proteins in solution. Analysis of the reactions by SDS-polyacrylamide gel electrophoresis evidenced only two covalently linked products. These were identified by N-terminal sequencing as the alpha and beta subunits of the MoFe protein, each cross-linked to a single polypeptide chain of the ferredoxin. This result, taking into account the dimeric structure of the ferredoxin in solution, strongly suggests an interaction of the ferredoxin with the MoFe protein at a site contributed to by both subunits of the MoFe protein. The ionic strength dependence of the interaction evidenced by affinity chromatography was confirmed in the cross-linking reactions, and its specificity was assessed by showing that no cross-linking occurred when the [2Fe-2S] C. pasteurianum ferredoxin was denatured or replaced by spinach ferredoxin or by clostridial rubredoxin, or when the MoFe protein from C. pasteurianum was either inactivated or replaced by its counterpart from Azotobacter vinelandii. It has also been observed that the ferredoxin inhibits cross-linking between the nitrogenase Fe protein and the MoFe protein, which suggests overlapping binding sites of the ferredoxin and of the Fe protein on the MoFe protein. Cross-linking experiments implementing a number of molecular variants of the [2Fe-2S] C. pasteurianum ferredoxin demonstrated that glutamate residues 31, 34, and 38 are important contributors to the interaction with the MoFe protein.

The rate constants for the intramolecular electron transfer between the two [4Fe-4S] clusters of a series of native and genetically engineered ferredoxins have been determined by proton magnetic resonance (1H NMR) spectroscopy. The measurement relies on the properties of the signals assigned to beta-protons of the coordinating cysteines when the protein is substoichiometrically reduced: these signals include coalesced peaks arising from the fast hopping of an extra electron between the two oxidized clusters of the protein. An upper limit of significantly less than 10(5) M(-1) s(-1) for the intermolecular and an average of the order of 5 x 10(6) s(-1) for the intramolecular electron transfer rate constants of several ferredoxins have been obtained. Owing to the edge-to-edge intercluster distance of approximately 10 A derived from the crystallographic structure of Clostridium acidurici ferredoxin, the rate constant associated with the intramolecular process is as expected for a nonadiabatic redox process, assuming a reasonable value of less than 1 eV for the reorganization energy. The latter could not be determined from the temperature dependence of the rate constant since no variation was observed over the temperature range accessible in these experiments. Structural changes introduced around and between the two [4Fe-4S] clusters in Clostridium pasteurianum ferredoxin by site-directed mutagenesis have been used to probe the potential involvement of dominant electron transfer pathways between the clusters. These changes have no major effect on the value of the intramolecular electron transfer rate constant. From this analysis, no specific amino acid side chain seems to play a central role in the process. The rate constants derived in the present work may serve as a basis for the study of enzymes containing two closely spaced [4Fe-4S] clusters such as found in these ferredoxins.

Protein kinase CK2 is a ubiquitous eukaryotic Ser/Thr kinase whose catalytic activity is enhanced several times by polyamines. We have shown previously that the regulatory beta-subunit of CK2 bears a polyamine binding site located in the region Asp51-Tyr110. In the present study, we have used spermine analogs to investigate the structural requirements of the CK2 polyamine binding site. We have observed a strong correlation between the stimulations of CK2 activity by all tested polyamines and their binding efficiencies to the enzyme. As a result, spermine was found to be the most efficient stimulator of the kinase activity and the best CK2 ligand. The effect of the pH on the stimulation of CK2 activity by spermine strongly suggests the involvement of ionic interactions between the positive charges of spermine and the negative charges of acidic amino acids of the beta-subunit. Using a fusion protein made of MBP and the beta-subunit region encompassing amino acid residues Asp51-Pro110, we have studied the binding of spermine as a function of the ionic strength. We show that this region delineates a functional and autonomous domain containing a binding site involved in the interaction with the four positive charges of spermine. Altogether, these results led to the elaboration of the first model defining the crucial structural parameters of a polyamine-protein interaction at the molecular level.

Tryptophanyl substitution of the Saccharomyces cerevisiae adenine nucleotide carrier (Anc2p isoform) was not deleterious for the transport activity or the folding of the carrier [preceding paper by Le Saux et al. (1996) Biochemistry 35, 16116-16124]. Conformational changes of the isolated wild-type and Trp-substituted Anc2p variants, induced upon binding of specific substrates [adenosine triphosphate (ATP) or diphosphate (ADP)] or inhibitors [carboxyatractyloside (CATR) or bongkrekic acid (BA)], were studied by measurement of intrinsic fluorescence. Titration of CATR and BA binding sites ended in the same number of sites, namely, 6-7 nmol/mg of wild-type and variant Anc2p. Isolated Anc2p in detergent presented similar emission spectra, suggesting that all tryptophanyl residues were in environments of similar hydrophobicity. Trp87 and Trp126 contributed largely and to a similar extent to the fluorescence enhancement observed in response to ATP binding, while Trp235 contributed negatively and to a small extent to the fluorescence change. Both Trp126 and Trp235, and to a lower extent Trp87, participate in the CATR-induced fluorescence decrease of Anc2p. Responses to BA binding were observed only in the presence of ATP; they consisted of a further fluorescence increase of the Anc2p.ATP complex, which was mainly due to Trp87 and Trp126, Trp235 being much less responsive. The different fluorescence responses of the three Trp residues of Anc2 variants to ATP, CATR, and BA are in agreement with distinct binding sites for these ligands and distinct conformations of the carrier protein recognizing specifically CATR or BA. A mechanistic model is proposed to interpret the transitions between the different conformational states of Anc2p.

The [2Fe-2S] ferredoxin from Clostridium pasteurianum is unique among ferredoxins, both by its sequence and by the distribution of its cysteine residues (in positions 11, 14, 24, 56, and 60). In previous investigations, a combination of site-directed mutagenesis and of spectroscopic techniques showed that cysteines 11, 56, and 60 are ligands of the [2Fe-2S] cluster in the wild type protein and that cysteine 14 is not, but the status of cysteine 24 remained unclear. New mutated forms of this ferredoxin have been obtained and characterized. The data show that cysteine 24 is a ligand of the cluster in the wild type protein. When cysteine 24 is mutated into alanine, it is replaced as a cluster ligand by cysteine 14. The fourth ligand of the cluster can also be a cysteine residue newly introduced in position 16 when both cysteines 14 and 24 are replaced by alanine. These results suggest that the region encompassing cysteines 14 and 24 is a solvent-exposed flexible loop, in agreement with structure predictions. A number of nondeleterious deletions of variable length (3-14 residues) have been performed in the region of residues 17-32. The deletions were found to modify only marginally the spectroscopic properties of the [2Fe-2S] cluster but resulted in variations of its redox potential over a range of nearly 100 mV. This is the first instance of ligand swapping in a [2Fe-2S] protein, and the first time in any ferredoxin that a large loop has been excised from the structure without preventing the assembly of the iron-sulfur chromophore. Some of the molecular variants described here also highlight the similarities between the C. pasteurianum [2Fe-2S] ferredoxin and the 25 kDa subunit of the proton-translocating NADH: ubiquinone oxidoreductase of Paracoccus denitrificans.

The cytochromes c' are paramagnetic heme proteins generally consisting of two identical 14 kDa subunits. The 1H and 15N resonances of the ferricytochrome c' from the purple phototrophic bacterium Rhodobacter capsulatus have been extensively assigned by the TOCSY-HSQC, NOESY-HSQC, HSQC-NOESY-HSQC, and HNHA 3D heteronuclear experiments performed on an 8 mM sample labeled with 15N. In addition, the 13C alpha and 13CO resonances were assigned by the HNCA and multiple-quantum HNCOCA 3D experiments performed on a 0.5 mM sample labeled with 13C and 15N. The assignment of the backbone 13C resonances was used to confirm the 1H and 15N assignments and to better define secondary structure. On the basis of medium-range NOEs, 3JHN alpha coupling constants, and backbone 13C chemical shifts, the secondary structure consists of four helices: helix-1 (3-29), helix-2 (33-49), helix-3 (78-97), and helix-4 (103-117). On the basis of long-range NOE contacts, the Rb. capsulatus ferricytochrome c' is a four-helix bundle protein in which consecutive helices are antiparallel with respect to one another.

In the presence of ADP and fluorometals, the ATPase activity of the catalytic sector, F1, of beef heart mitochondrial ATPase is strongly inhibited; this inhibition is dependent on the entrapment of ADP-fluoroaluminate complexes into the nucleotide binding sites of F1 [Lunardi, J., Dupuis, A., Garin, J., Issartel, J. P., Michel, L., Chabre, M., & Vignais, P. V. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 8958-8962]. We described here the effect of fluoroaluminate on the binding of 2-azido[3H]ADP and 8-azido[3H]ADP to beef heart mitochondrial F1 in the absence and presence of light. When the incubation medium was supplemented with NaF and AlCl3, and maintained in the dark, both 2-azido[3H]ADP and 8-azido[3H]ADP were able to elicit inhibition of F1-ATPase activity, exactly like ADP did. Upon photoirradiation, 2-azido[3H]ADP and 8-azido[3H]ADP bound covalently to F1. Labeling was restricted to the beta subunit of F1, and the same tyrosine residue, beta-Tyr-345, was labeled by either of the photoprobes. This is in contrast with the previous findings that in the absence of fluoroaluminate both the alpha and beta subunits of F1 were photolabeled by 8-azido[3H]ADP, and that two different regions of the beta subunits were labeled, centered on beta-Tyr-345 in the case of 2-azido[3H]ADP [Garin, J., Boulay, F., Issartel, J.P., Lunardi, J., & Vignais, P. V. (1986) Biochemistry 25, 4431-4437] and beta-Tyr-311 that of 8-azido[3H]ADP [Hollemans, M., Runswick, M., Fearnley, I.H., & Walker, J.E. (1983) J. Biol. Chem. 258, 9307-9313].(ABSTRACT TRUNCATED AT 250 WORDS)

Two-dimensional nuclear magnetic resonance spectroscopy was used to assign the proton resonances of ferrocytochrome c553 from Desulfovibrio vulgaris Hildenbourough at 37 degrees C and pH = 5.9. Only a few side-chain protons were not identified because of degeneracy or overlap. The spin systems of the 79 amino acids were identified by DQF-COSY and HOHAHA spectra in H2O and D2O. Sequential assignments were obtained from NOESY connectivities between adjacent amide, C alpha H, and C beta H protons. From sequential NH(i)----NH(i + 1) and long-range C alpha H(i)----NH(i + 3) connectivities, four stretches of helices were identified (2----8, 34----46, 53----59, 67----77). Long-range NOE between residues in three different helices provide qualitative information on the tertiary structure, in agreement with the general folding pattern of cytochrome c. The heme protons, including the propionate groups, were assigned, and the identification of Met 57 as sixth heme ligand was established. The dynamical behavior of the ring protons of the six tyrosines was analyzed in detail in terms of steric hindrance. The NMR data for ferrocytochrome c553 are consistent with the X-ray structure for the homologous cytochrome from D. vulgaris Miyazaki. On the basis of the secondary structure element and of observed chemical shift due to the heme ring current, a structural alignment of eukaryotic and prokaryotic cytochromes c is proposed.