Multinuclear magnetic resonance studies of the 2Fe.2S* ferredoxin from Anabaena species strain PCC 7120. 1. Sequence-specific hydrogen-1 resonance assignments and secondary structure in solution of the oxidized form

Complete sequence-specific assignments were determined for the diamagnetic 1H resonances from Anabaena 7120 ferredoxin (Mr = 11,000). A novel assignment procedure was followed whose first step was the identification of the 13C spin systems of the amino acids by a 13C(13C) double quantum correlation experiment [Oh, B.-H., Westler, M. W., Darba, P., & Markley, J. L. (1988) Science 240, 908-911]. Then, the 1H spin systems of the amino acids were identified from the 13C spin systems by means of direct and relayed 1H(13C) single-bond correlations [Oh, B.-H., Westler, W. M., & Markley, J. L. (1989) J. Am. Chem. Soc. 111, 3083-3085]. The sequential resonance assignments were based mainly on conventional interresidue 1H alpha i-1HNi + 1 NOE connectivities. Resonances from 18 residues were not resolved in two-dimensional 1H NMR spectra. When these residues were mapped onto the X-ray crystal structure of the homologous ferredoxin from Spirulina platensis [Fukuyama, K., Hase, T., Matsumoto, S., Tsukihara, T., Katsube, Y., Tanaka, N., Kakudo, M., Wada, K., & Matsubara, H. (1980) Nature 286, 522-524], it was found that they correspond to amino acids close to the paramagnetic 2Fe.2S* cluster. Cross peaks in two-dimensional homonuclear 1H NMR spectra were not observed for any protons closer than about 7.8 A to both iron atoms. Secondary structural features identified in solution include two antiparallel beta-sheets, one parallel beta-sheet, and one alpha-helix.     

Multinuclear magnetic resonance studies of the 2Fe.2S* ferredoxin from Anabaena species strain PCC 7120. 3. Detection and characterization of hyperfine-shifted nitrogen-15 and hydrogen-1 resonances of the oxidized form

All the nitrogen signals from the amino acid side chains and 80 of the total of 98 backbone nitrogen signals of the oxidized form of the 2Fe.2S* ferredoxin from Anabaena sp. strain PCC 7120 were assigned by means of a series of heteronuclear two-dimensional experiments [Oh, B.-H. Mooberry, E. S., & Markley, J. L. (1990) Biochemistry (second paper of three in this issue )]. Two additional nitrogen signals were observed in the one-dimensional 15N NMR spectrum and classified as backbone amide resonances from residues whose proton resonances experience paramagnetic broadening. The one-dimensional 15N NMR spectrum shows nine resonances that are hyperfine shifted and broadened. From this inventory of diamagnetic nitrogen signals and the available X-ray coordinates of a related ferredoxin [Tsukihara, T., Fukuyama, K., Nakamura, M., Katsube, Y., Tanaka, N., Kakudo, M., Wada, K., Hase, T., & Matsubara, H. (1981) J. Biochem. 90, 1763-1773], the resolved hyperfine-shifted 15N peaks were attributed to backbone amide nitrogens of the nine amino acids that share electrons with the 2Fe.2S* center or to backbone amide nitrogens of two other amino acids that are close to the 2Fe.2S* center. The seven 15N signals that are missing and unaccounted for probably are buried under the envelope of amide signals. 1H NMR signals from all the amide protons directly bonded to the seven missing and nine hyperfine-shifted nitrogens were too broad to be resolved in conventional 2D NMR spectra.(ABSTRACT TRUNCATED AT 250 WORDS)     

Nuclear magnetic resonance studies of recombinant Escherichia coli glutaredoxin. Sequence-specific assignments and secondary structure determination of the oxidized form

Escherichia coli glutaredoxin (85 amino acid residues, Mr = 9100), the glutathione-dependent hydrogen donor for ribonucleotide reductase, was purified from an inducible lambda PL, expression system both with a natural isotope content and with uniform 15N labelling. This material was used for obtaining sequence-specific 1H magnetic resonance assignments and the identification of regular secondary structures in the oxidized form of the protein, which contains the redox-active disulfide Cys11-Pro-Tyr-Cys14. Oxidized glutaredoxin contains a four-stranded beta-sheet, with the peripheral strand 32-37 arranged parallel to the strand 2-7, which further combines with the two additional strands 61-64 and 67-69 in an antiparallel fashion. The protein further contains three helices extending approximately from residues 13-28, 45-54 and 72-84.     

Isolation and sequence of the gene for ferredoxin I from the cyanobacterium Anabaena sp. strain PCC 7120.

The structural gene for ferredoxin I, petF, from the cyanobacterium Anabaena sp. strain PCC 7120 has been isolated from a recombinant lambda library. Mixtures of tetradecanucleotides and heptadecanucleotides, each containing all possible DNA sequences corresponding to two separate regions of the ferredoxin amino acid sequence, were synthesized and used as hybridization probes to identify a genomic clone containing the coding sequence for the petF gene. The sequence of the entire petF coding region and portions of the 3'- and 5'-flanking regions was determined. The DNA sequence of petF suggests that, in contrast to the nucleus-encoded plant protein, cyanobacterial apoferredoxin is not synthesized as a higher-molecular-weight precursor. The Anabaena petF gene is a single-copy gene. During growth on complete medium it was transcribed into a monocistronic mRNA species of approximately 500 bases that initiated 100 base pairs upstream from the petF coding region.     

Two-dimensional NMR studies of staphylococcal nuclease. 2. Sequence-specific assignments of carbon-13 and nitrogen-15 signals from the nuclease H124L-thymidine 3',5'-bisphosphate-Ca2+ ternary complex

Samples of staphylococcal nuclease H124L (cloned protein overproduced in Escherichia coli whose sequence is identical with that of the nuclease isolated from the V8 strain of Staphylococcus aureus) were labeled uniformly with carbon-13 (26% ul 13C), uniformly with nitrogen-15 (95% ul 15N), and specifically by incorporating nitrogen-15-labeled leucine ([98% 15N]Leu) or carbon-13-labeled lysine ([26% ul 13C]Lys), arginine ([26% ul 13C]Arg), or methionine ([26% ul 13C]Met). Solutions of the ternary complexes of these analogues (nuclease H124L-pdTp-Ca2+) at pH 5.1 (H2O) or pH* 5.5 (2H2O) at 45 degrees C were analyzed as appropriate to the labeling pattern by multinuclear two-dimensional (2D) NMR experiments at spectrometer fields of 14.09 and 11.74 T: 1H-13C single-bond correlation (1H[13C]SBC); 1H-13C single-bond correlation with NOE relay (1H[13C]SBC-NOE); 1H-13C single-bond correlation with Hartmann-Hahn relay (1H-[13C]SBC-HH); 1H-13C multiple-bond correlation (1H[13C]MBC); 1H-15N single-bond correlation (1H-[15N]SBC); 1H-15N single-bond correlation with NOE relay (1H[15N]SBC-NOE). The results have assisted in spin system assignments and in identification of secondary structural elements. Nuclear Overhauser enhancements (NOE's) characteristic of antiparallel beta-sheet (d alpha alpha NOE's) were observed in the 1H [13C]-SBC-NOE spectrum of the nuclease ternary complex labeled uniformly with 13C. NOE's characteristic of alpha-helix (dNN NOE's) were observed in the 1H[15N]SBC-NOE spectrum of the complex prepared from protein labeled uniformly with 15N. The assignments obtained from these multinuclear NMR studies have confirmed and extended assignments based on 1H[1H] 2D NMR experiments [Wang, J., LeMaster, D. M., & Markley, J. L. (1990) Biochemistry (preceding paper in this issue)].     

Crystallization and preliminary analysis of oxidized, recombinant, heterocyst [2Fe-2S] ferredoxin from Anabaena 7120.

The [2Fe-2S] ferredoxin produced in the heterocyst cells of Anabaena 7120 plays a key role in nitrogen fixation, where it serves as an electron acceptor from various sources and an electron donor to nitrogenase. Crystals of recombinant heterocyst ferredoxin, coded for by the fdx H gene from Anabaena 7120 and overproduced in Escherichia coli, have been grown from ammonium sulfate solutions and are suitable for high resolution X-ray crystallographic analysis. They belong to the hexagonal space group P6(1) or P6(5) with unit cell dimensions of a = b = 44.2 A and c = 80.6 A. The crystals contain one molecule per asymmetric unit and diffract to a nominal resolution of 1.6 A. The molecular structure of this heterocyst ferredoxin is of special interest in that 4 of the 22 amino acid positions thought to be absolutely conserved in nonhalophilic ferredoxins are different and, based on amino acid sequence alignments, three of these positions are located in the metal-cluster binding loop. Consequently, a high-resolution X-ray analysis of this [2Fe-2S] ferredoxin, and subsequent three-dimensional comparisons with other known ferredoxin models, will provide new insight into structure/function relationships for this class of redox proteins.     

Crystallization and structure determination to 2.5-A resolution of the oxidized [2Fe-2S] ferredoxin isolated from Anabaena 7120

The molecular structure of the oxidized form of the [2Fe-2S] ferredoxin isolated from the cyanobacterium Anabaena species strain PCC 7120 has been determined by X-ray diffraction analysis to a nominal resolution of 2.5 A and refined to a crystallographic R factor of 18.7%. Crystals used in this investigation belong to the space group P2(1)2(1)2(1) with unit cell dimensions of a = 37.42 A, b = 38.12 A, and c = 147.12 A and two molecules in the asymmetric unit. The three-dimensional structure of this ferredoxin was solved by a method that combined X-ray data from one isomorphous heavy-atom derivative with noncrystallographic symmetry averaging and solvent flattening. As in other plant-type [2Fe-2S] ferredoxins, the iron-sulfur cluster is located toward the outer edge of the molecule, and the irons are tetrahedrally coordinated by both inorganic sulfurs and sulfurs provided by protein cysteine residues. The main secondary structural elements include four strands of beta-pleated sheet and three alpha-helical regions.     

Activities of two dissimilar thioredoxins from the cyanobacterium Anabaena sp. strain PCC 7120.

Thioredoxin is a small redox protein that functions as a reducing agent and modulator of enzyme activity. A gene for an unusual thioredoxin was previously isolated from the cyanobacterium Anabaena sp. strain PCC 7120 and cloned and expressed in Escherichia coli. However, the protein could not be detected in Anabaena cells (J. Alam, S. Curtis, F. K. Gleason, M. Gerami-Nejad, and J. A. Fuchs, J. Bacteriol. 171:162-171, 1989). Polyclonal antibodies to the atypical thioredoxin were prepared, and the protein was detected by Western immunoblotting. It occurs at very low levels in extracts of Anabaena sp. and other cyanobacteria. No antibody cross-reaction was observed in extracts of eukaryotic algae, plants, or eubacteria. The anti-Anabaena thioredoxin antibodies did react with another unusual thioredoxin-glutaredoxin produced by bacteriophage T4. Like the T4 protein and other glutaredoxins, the unusual cyanobacterial thioredoxin can be reduced by glutathione. The Anabaena protein can also activate enzymes of carbon metabolism and has some functional similarity to spinach chloroplast thioredoxin f. However, it shows only 23% amino acid sequence identity to the spinach chloroplast protein and appears to be distantly related to other thioredoxins. The data indicate that cyanobacteria, like plant chloroplasts, have two dissimilar thioredoxins. One is related to the more common protein found in other prokaryotes, and the other is an unusual thioredoxin that can be reduced by glutathione and may function in glucose catabolism.     

Bacterial-type ferredoxin genes in the nitrogen fixation regions of the cyanobacterium Anabaena sp. strain PCC 7120 and Rhizobium meliloti.

The nucleotide sequence of a region located downstream of the nifB gene, both in the cyanobacterium Anabaena sp. strain PCC 7120 and in Rhizobium meliloti, has been determined. This region contains a gene (fdxN) whose predicted polypeptide product strongly resembles typical bacterial ferredoxins. Cyanobacteria have not previously been shown to contain bacterial-type ferredoxins. The presence of this gene suggests that nitrogen-fixing cyanobacteria have at least four distinct ferredoxins.     

Nitrate assimilation gene cluster from the heterocyst-forming cyanobacterium Anabaena sp. strain PCC 7120.

A region of the genome of the filamentous, nitrogen-fixing, heterocyst-forming cyanobacterium Anabaena sp. strain PCC 7120 that contains a cluster of genes involved in nitrate assimilation has been identified. The genes nir, encoding nitrite reductase, and nrtABC, encoding elements of a nitrate permease, have been cloned. Insertion of a gene cassette into the nir-nrtA region impaired expression of narB, the nitrate reductase structural gene which together with nrtD is found downstream from nrtC in the gene cluster. This indicates that the nir-nrtABCD-narB genes are cotranscribed, thus constituting an operon. Expression of the nir operon in strain PCC 7120 is subjected to ammonium-promoted repression and takes place from an NtcA-activated promoter located 460 bp upstream from the start of the nir gene. In the absence of ammonium, cellular levels of the products of the nir operon are higher in the presence of nitrate than in the absence of combined nitrogen.     

1H, 13C and 15N resonance assignments of the arsenate reductase from Synechocystis sp. strain PCC 6803

Arsenate reductases (ArsC) are a group of enzymes that play essential roles in biological arsenic detoxification pathways by catalyzing the intracellular reduction of arsenate to arsenite, which is subsequently extruded from the cells by specific transport systems. The ArsC protein from cyanobacterium Synechocystis sp. strain PCC 6803 (SynArsC) is related to the thioredoxin-dependent ArsC family, but uses the glutathione/glutaredoxin system for arsenate reduction. Therefore, it is classified to a novel thioredoxin/glutaredoxin hybrid arsenate reductase family. Herein we report the chemical shift assignments of ¹H, ¹³C and ¹⁵N atoms for the reduced form of SynArsC, which provides a starting point for further structural analysis and elucidation of its enzymatic mechanism.     

Growth and phage resistance of Anabaena sp. strain PCC 7120 in the presence of cyanophage AN-15

The cyanophage AN-15 was found to have a requirement for either 1 mM calcium or 1 mM magnesium ions to maintain viral stability, whereas 1 mM calcium ions alone were essential for the infection process to proceed in Anabaena sp. strain PCC 7120. Following prolonged incubation, phage-resistant cells were detected at a high frequency (approximately 10-⁵) in lysates, as either renewed growth in liquid cultures, or as colonies in confluently lysed lawns. Southern hybridisation failed to detect AN-15 DNA in any of the resistant strains, implying that resistance is unlikely to be due to the presence of temperate phages. A high rate of spontaneous mutation is therefore likely to be the cause of resistance. Two classes of resistant cells were identified; those in which AN-15 failed to attach to host cells, and those in which attachment occurred, but subsequent replication was defective. However, it was possible to overcome phage resistance by the isolation of spontaneous mutants of AN-15, capable of infecting phage-resistant cells. These observations imply that if cyanophages are to be assessed as a means of controlling cyanobacterial blooms in freshwater bodies, the ionic (notably calcium) concentration of the water must be considered, together with the possible need to employ alternative cyanophage strains if resistance to the original one arises. This revised version was published online in August 2006 with corrections to the Cover Date.     

Sequence-specific assignments of the 1H nuclear magnetic resonance spectra of reduced high-potential ferredoxin (HiPIP) from Chromatium vinosum.

The 1H resonances of the high-potential [4Fe-4S]2+ ferredoxin from Chromatium vinosum have been assigned through conventional sequential methodology applied to 2D NMR spectra. Almost 80% of the residues were identified using standard 2D COSY, HOHAHA, and NOESY pulse sequences. These residues correspond to four segments of the primary structure that do not interact strongly with the iron-sulfur cluster. A minor correction to the amino acid sequence is strongly suggested by these NMR data. Additional protons more sensitive to the proximity of the cluster were assigned by a combination of NOESY experiments with fast repetition rates and short mixing times and of HOHAHA spectra recorded with reduced spin-lock duration aimed at compensating for the short relaxation rates. Hence, the contributions of 79 residues out of 85 were identified in NMR spectra, among which the assignments of 64 residues were completed. Even the fastest relaxing protons, like those of the cysteine ligands, could be correlated, partly because the strong hyperfine shifts isolate them from the crowded diamagnetic region. However, other protons, in particular those involved in NH-S hydrogen bonds with the iron-sulfur cluster, were more difficult to identify, most probably because their relatively broad signals overlapped with those of protons not or less perturbed by the active site. The availability of the major part of the 1H NMR assignments has enabled the detection and identification of many interresidue NOESY cross peaks. These data are in full agreement with the elements of secondary structure previously revealed by X-ray crystallographic analysis of the protein.(ABSTRACT TRUNCATED AT 250 WORDS)     

Nitrogen fixation (nif) genes of the cyanobacterium Anabaena species strain PCC 7120. The nifB-fdxN-nifS-nifU operon

A second nitrogen fixation (nif) operon in the cyanobacterium (blue-green alga) Anabaena (Nostoc) sp. strain PCC 7120 has been identified and sequenced. It is located just upstream of the nifHDK operon and consists of four genes in the order nifB, fdxN, nifS, and nifU. The three nif genes were identified on the basis of their similarity with the corresponding genes from other diazotrophs. The fourth gene, fdxN, codes for a bacterial type ferredoxin (Mulligan, M. E., Buikema, W. J., and Haselkorn, R. (1988) J. Bacteriol. 167, 4406-4410). The four genes are probably transcribed as a single operon, but are expressed at a lower level than the nifHDK operon, and only after a developmentally induced DNA rearrangement occurs that excises a 55-kilobase pair element from within the fdxN gene (Golden, J. W., Mulligan, M. E., and Haselkorn, R. (1987) Nature 327, 526-529; Golden, J. W., Carrasco, C. D., Mulligan, M. E., Schneider, G. J., and Haselkorn, R. (1988) J. Bacteriol. 170, 5034-5041). The promoter for the nifB operon was located by primer extension. Comparison of the nifB 5'-flanking sequence with the nifH 5'-flanking sequence did not reveal any consensus base pairs that would define a nif promoter for Anabaena. The operon contains two instances of 7-base pair directly repeated sequences: seven copies of the repeated sequence are found between the nifB and fdxN genes and six copies are found between the nifS and nifU genes. The function of these repeats is unknown.     

Structure of the oxidized long-chain flavodoxin from Anabaena 7120 at 2 A resolution.

The structure of the long-chain flavodoxin from the photosynthetic cyanobacterium Anabaena 7120 has been determined at 2 A resolution by the molecular replacement method using the atomic coordinates of the long-chain flavodoxin from Anacystis nidulans. The structure of a third long-chain flavodoxin from Chondrus crispus has recently been reported. Crystals of oxidized A. 7120 flavodoxin belong to the monoclinic space group P2(1) with a = 48.0, b = 32.0, c = 51.6 A, and beta = 92 degrees, and one molecule in the asymmetric unit. The 2 A intensity data were collected with oscillation films at the CHESS synchrotron source and processed to yield 9,795 independent intensities with Rmerg of 0.07. Of these, 8,493 reflections had I > 2 sigma and were used in the analysis. The model obtained by molecular replacement was initially refined by simulated annealing using the XPLOR program. Repeated refitting into omit maps and several rounds of conjugate gradient refinement led to an R-value of 0.185 for a model containing atoms for protein residues 2-169, flavin mononucleotide (FMN), and 104 solvent molecules. The FMN shows many interactions with the protein with the isoalloxazine ring, ribityl sugar, and the 5'-phosphate. The flavin ring has its pyrimidine end buried into the protein, and the functional dimethyl benzene edge is accessible to solvent. The FMN interactions in all three long-chain structures are similar except for the O4' of the ribityl chain, which interacts with the hydroxyl group of Thr 88 side chain in A. 7120, while with a water molecule in the other two. The phosphate group interacts with the atoms of the 9-15 loop as well as with NE1 of Trp 57. The N5 atom of flavin interacts with the amide NH of Ile 59 in A. 7120, whereas in A. nidulans it interacts with the amide NH of Val 59 in a similar manner. In C. crispus flavodoxin, N5 forms a hydrogen bond with the side chain hydroxyl group of the equivalent Thr 58. The hydrogen bond distances to the backbone NH groups in the first two flavodoxins are 3.6 A and 3.5 A, respectively, whereas in the third flavodoxin the distance is 3.1 A, close to the normal value. Even though the hydrogen bond distances are long in the first two cases, still they might have significant energy because their microenvironment in the protein is not accessible to solvent.(ABSTRACT TRUNCATED AT 400 WORDS)