New perspectives on bacterial ferredoxin evolution

Summary Recent evidence indicates that a gene transposition event occurred during the evolution of the bacterial ferredoxins subsequent to the ancestral intrasequence gene duplication. In light of this new information, the relationships among the bacterial ferredoxins were reexamined and an evolutionary tree consistent with this new understanding was derived. The bacterial ferredoxins can be divided into several groups based on their sequence properties; these include the clostridial-type ferredoxins, theAzotobacter-type ferredoxins, and a group containing the ferredoxins from the anaerobic, green, and purple sulfur bacteria. Based on sequence comparison, it was concluded that the amino-terminal domain of theAzotobacter-type ferredoxins, which contains the novel 3Fe3S cluster binding site, is homologous with the carboxyl-terminal domain of the ferredoxins from the anaerobic photosynthetic bacteria.A number of ferredoxin sequences do not fit into any of the groups described above. Based on sequence properties, these sequences can be separated into three groups: a group containingMethanosarcina barkeri ferredoxin andDesulfovibrio desulfuricans ferredoxin II, a group containingDesulfovibrio gigas ferredoxin andClostridium thermoaceticum ferredoxin, and a group containingDesulfovibrio africanus ferredoxin I andBacillus stearothermophilus ferredoxin. The last two groups differ from all of the other bacterial ferredoxins in that they bind only one FeS cluster per polypeptide, whereas the others bind two. Sequence examination indicates that the second binding site has been either partially or completely lost from these ferredoxins.Methanosarcina barkeri ferredoxin andDesulfovibrio desulfuricans ferredoxin II are of interest because, of all the ferredoxins whose sequences are presently known, they show the strongest evidence of internal gene duplication. However, the derived evolutionary tree indicates that they diverged from theAzotobacter-type ferredoxins well after the ancestral internal gene duplication. This apparent discrepancy is explained by postulating a duplication of one halfchain sequence and a deletion of the other halfchain. TheClostridium thermoaceticum andBacillus stearothermophilus groups diverged from this line and subsequently lost one of the FeS binding sites.It has recently become apparent that gene duplication is ubiquitous among the ferredoxins. Several organisms are now known to have a variety of ferredoxins with widely divergent properties. Unfortunately, in only one case are the sequences of more than one ferredoxin from the same organism known. Thus, although the major features of the bacterial ferredoxin tree are now understood, a complete bacterial phylogeny cannot be inferred until more sequence information is available.     

Examination of protein sequence homologies: V. New perspectives on evolution between bacterial and chloroplast-type ferredoxins inferred from sequence evidence

Summary Sequence homologies among 34 chloroplast-type ferredoxins were examined using a computer program that quantitatively evaluates the extent of sequence similarity as a correlation coefficient. The resultant alignment contains six gaps representing insertions or deletions of some residues, all of which are located such that they precisely preserve the domains of structural fragments as determined by crystallographic data onSpirulina platensis ferredoxin.In the search for any total correlation between the chloroplast-type and 27 bacterial ferredoxins, 1891 comparison matrices prepared for possible combinations indicated that the bacterial basal sequence of 55 residues has been conserved evolutionarily in the chloroplast-type sequences corresponding to residue positions 36–90 ofSpirulina platensis ferredoxin. In addition, the bacterial connector sequence region was found to be conserved. These findings strongly suggest that the bacterial and chloroplast-type ferredoxins descended from a common ancestor, and branched off after the bacterial gene duplication, whereas the chloroplast-type ferredoxins originally were generated by duplicating the already duplicated bacterial gene, i.e., by double-duplication.     

Complete amino acid sequence of ferredoxin from Peridinium bipes (Dinophyceae)

The amino acid sequence of the major ferredoxin component isolated from a dinoflagellate, Peridinium bipes, was completely determined. Staphylococcus aureus V8 proteolytic, tryptic and chymotryptic peptides of Cm-ferredoxin were prepared and sequenced. The sequence was Phe-Lys-Val-Thr-Leu-Asp-Thr-Pro-Asp-Gly-Lys-Lys-Ser-Phe-Glu-Cys- Pro-Gly-Asp-Ser-Tyr-Ile-Leu-Asp-Lys-Ala-Glu-Glu-Glu-Gly-Leu-Glu-Leu-Pro- Tyr-Ser - Cys-Arg-Ala-Gly-Ser-Cys-Ser-Ser-Cys-Ala-Gly-Lys-Val-Leu-Thr-Gly-Ser-Ile- Asp-Gln - Ser-Asp-Gln-Ala-Phe-Leu-Asp-Asp-Asp-Gln-Gly-Gly-Asp-Gly-Tyr-Cys-Leu-Thr- Cys-Val - Thr-Tyr-Pro-Thr-Ser-Asp-Val-Thr-Ile-Lys-Thr-His-Cys-Glu-Ser-Glu-Leu. It was composed of 93 amino acid residues with 7 cysteine residues, the highest number found among the chloroplast-type ferredoxins so far sequenced. A cysteine residue was found for the first time at the 89th position in a chloroplast-type ferredoxin. Calculation of the numbers of amino acid differences among chloroplast-type ferredoxins indicates that the Peridinium ferredoxin is far divergent not only from higher plant ferredoxins but also from blue-green algal ferredoxins.     

Structure of the extracellular ferredoxin from Rhodospirillum rubrum: close similarity to clostridial ferredoxins

The amino acid sequence of an [8Fe-8S] ferredoxin isolated from the culture medium of Rhodospirillum rubrum, a photosynthetic purple non-sulfur bacterium, was determined by a combination of various conventional procedures. The sequence was A-Y-K-I-E-E-T-C-I-S-C-G-A-C-A-A-E-C-P-V-N-A-I-E-Q-G-D-T-I-F-V-V-N-A-D-T-C-I-D-C - G-N-C-A-N-V-C-P-V-G-A-P-V-A-E (55 amino acid residues). It lacked methionine, leucine, histidine, arginine, and tryptophan. The molecular weight was calculated to be 5,568 excluding iron and sulfur atoms. The distribution of 8 cysteine residues was exactly the same as that of clostridial-type ferredoxin, suggesting retention of the duplication of the bacterial ancestral ferredoxin gene. The extracellular ferredoxin of R. rubrum was compared with other ferredoxins observed in closely related photosynthetic bacteria and the evolutionary significance of this ferredoxin is discussed.     

Two distinct ferredoxins from Rhodobacter capsulatus: complete amino acid sequences and molecular evolution

Two distinct ferredoxins were purified from Rhodobacter capsulatus SB1003. Their complete amino acid sequences were determined by a combination of protease digestion, BrCN cleavage and Edman degradation. Ferredoxins I and II were composed of 64 and 111 amino acids, respectively, with molecular weights of 6,728 and 12,549 excluding iron and sulfur atoms. Both contained two Cys clusters in their amino acid sequences. The first cluster of ferredoxin I and the second cluster of ferredoxin II had a sequence, CxxCxxCxxxCP, in common with the ferredoxins found in Clostridia. The second cluster of ferredoxin I had a sequence, CxxCxxxxxxxxCxxxCM, with extra amino acids between the second and third Cys, which has been reported for other photosynthetic bacterial ferredoxins and putative ferredoxins (nif-gene products) from nitrogen-fixing bacteria, and with a unique occurrence of Met. The first cluster of ferredoxin II had a CxxCxxxxCxxxCP sequence, with two additional amino acids between the second and third Cys, a characteristics feature of Azotobacter-[3Fe-4S] [4Fe-4S]-ferredoxin. Ferredoxin II was also similar to Azotobacter-type ferredoxins with an extended carboxyl (C-) terminal sequence compared to the common Clostridium-type. The evolutionary relationship of the two together with a putative one recently found to be encoded in nifENXQ region in this bacterium [Moreno-Vivian et al. (1989) J. Bacteriol. 171, 2591-2598] is discussed.     

The amino acid sequence of ferredoxin from Sambucus nigra

The amino acid sequence of the ferredoxin from Sambucus nigra consists of a single polypeptide chain of 97 amino acid residues, 5 of which are cysteine. The positions of the 4 cysteine residues which bind the iron atoms of the active centre are identical to those of other ferredoxins. Due to difficulties of obtaining pure protein, residues 87–90 have only been identified from the amino acid analysis of peptide C 10 and by homology with other higher plant ferredoxins.     

Amino acid sequence of a four-iron-four-sulphur ferredoxin isolated from Bacillus stearothermophilus.

1. The primary structure of a 4Fe-4S ferredoxin from Bacillus stearothermophilus was determined and shown to consist of a single polypeptide chain of 81 amino acid residues. The molecular weight of the holoprotein is about 9120. 2. There are only four cysteine residues in the molecule; three of these are located near the N-terminus as a Cys-X-X-Cys-X-X-Cys segment, and the fourth cysteine residue is followed by a proline and located in the C-terminal half. 3. The Fe-S chromophore in B. stearothermophilus ferredoxin was previously well characterized and was shown to consist of a single 4Fe-4S cluster. This ferredoxin sequence establishes for the first time the relative location of the four cysteine residues necessary to bind the 4Fe-4S cluster of a 4Fe ferredoxin, and is in agreement with the criteria for the relative positions of the cysteines proposed from X-ray-crystallographic studies on an 8Fe (two 4Fe-4S clusters) ferredoxin. 4. The sequence of B. stearothermophilus ferredoxin is homologous in many segments to that of other bacterial ferredoxins, the degree of homology being greater towards ferredoxins from Desulfovibrio gigas and photosynthetic bacteria than to Clostridial ferredoxins. 5. The presence of a relatively higher number of glutamic acid and lower number of cysteine residues in the molecule may explain the greater thermal stability and oxygen-insenstivity of this ferredoxin.     

Functional analysis of the cysteine motifs in the ferredoxin-like protein FdxN of Rhizobium meliloti involved in symbiotic nitrogen fixation.

Summary TheRhizobium meliloti fdxN gene, which is part of thenifA-nifB fdxN operon, is absolutely required for symbiotic nitrogen fixation. The deduced sequence of the FdxN protein is characterized by two cysteine motifs typical of bacterial-type ferredoxins. The Fix^– phenotype of anR. meliloti fdxN: :[Tc] mutant could be rescued by theR. leguminosarum fdxN gene, whereas no complementation was observed withnif-associated genes encoding ferredoxins fromBradyrhizobium japonicum, Azotobacter vinelandii, A. chroococcum andRhodobacter capsulatus. In addition to these heterologous genes, severalR. meliloti fdxN mutant genes constructed by site-directed mutagenesis were analyzed. Not only a cysteine residue within the second cysteine motif (position 42), which is known to coordinate the Fe-S cluster in homologous proteins, but also a cysteine located down-stream of this motif (position 61), was found to be essential for the activity of theR. meliloti FdxN protein. Changing the amino acid residue proline in position 56 into methionine resulted in a FdxN mutant protein with decreased activity, whereas changes in positions 35 (Asp35Glu) and 45 (Gly45Glu) had no significant effect on the function of the FdxN mutant proteins. In contrast to bacterial-type ferredoxins, which contain two identical cysteine motifs of the form C-X₂-C-X₂-C-X₃-C,nif-associated ferredoxins, includingR. meliloti FdxN, are characterized by two different cysteine motifs. Six additional amino acids separate the second (Cys₄₂) and the third cysteine (Cys₅₁) in the C-terminal motif (C-X₂-C-X₈-C-X₃-C). By molecular modelling, it was predicted that these amino acid residues form a loop, which does not alter the relative positions of the neighbouring cysteines. Deletion of this loop resulted in anR. meliloti FdxN mutant protein, which exhibited almost 70% wild-type activity, indicating that the predicted loop does not affect Fe-S cluster binding and plays no crucial role in activity of the FdxN protein.     

Refined crystal structure of ferredoxin II from Desulfovibrio gigas at 1.7 A

The crystal structure of ferredoxin II from Desulfovibrio gigas has been determined using phasing from anomalous scattering data at a resolution of 1.7 A and refined to an R-factor of 0.157. The molecule has an overall chain fold similar to that of the other bacterial ferredoxins of known structure. The molecule contains a single 3Fe-4S cluster with geometry indistinguishable from the 4Fe-4S clusters, and a disulfide bond near the site corresponding to the position of the second cluster of two-cluster ferredoxins. The cluster is bound by cysteine residues 8, 14 and 50. The side-chain of cysteine 11 extends away from the cluster, but could rotate to become the fourth cysteine ligand in the four-iron form of the molecule given a local adjustment of the polypeptide chain. This residue is modified, however, by what appears to be a methanethiol group. There are a total of eight NH . . . S bonds to the inorganic and cysteine sulfur atoms of the Fe-S cluster. There is an additional residue found that is not reported for the chemical sequence: according to the electron density a valine residue should be inserted after residue 55.     

Amino acid sequence of the [4Fe-4S] ferredoxin isolated from Desulfovibrio desulfuricans Norway

The complete amino acid sequence of the [4Fe-4S] ferredoxin from Desulfovibrio desulfuricans Norway was determined by repetitive Edman degradation of the whole protein and peptides derived from tryptic digestion. The protein has 59 residues. Four of the six cysteine residues are involved in the binding of the [4Fe-4S] cluster in the same arrangement as in clostridial ferredoxins. This sequence is compared to various Desulfovibrio ferredoxin sequences and to the sequence and three-dimensional structure of Peptococcus aerogenes ferredoxin. Evidence of gene duplication is indicated. The requirement of some sequence features in the ferredoxin for an interaction process with its electron transfer partner, cytochrome c3, is postulated in the discussion.     

Isolation and characterization of a rubredoxin and two ferredoxins from Desulfovibrio africanus

Rubredoxin and two distinct ferredoxins have been purified from Desulfovibrio africanus. The rubredoxin has a molecular weight of 6000 while the ferredoxins appear to be dimers of identical subunits of approximately 6000 to 7000 molecular weight. Rubredoxin contains one iron atom, no acid-labile sulfide and four cysteine residues per molecule. Its absorbance ratio A₂₇₈/A₄₉₀ is 2.23 and its amino acid composition is characterized by the absence of leucine and a preponderance of acidic amino acids.

The two ferredoxins, designated I and II, are readily separated on DEAE-cellulose. The amino acid compositions of ferredoxins I and II show them to be different protein species; the greater number of acidic amino acid residues in ferredoxin I than in ferredoxin II appears to account for separation based on electronic charge. Both ferredoxins contain four iron atoms, four acid-labile sulfur groups and either four (ferredoxin II) or six (ferredoxin I) cysteine residues per molecule. Spectra of the two ferredoxins differ from those of ferredoxins of other Desulfovibrio species by exhibiting a pronounced absorption peak at 283 nm consistent with an unusual high content of aromatic residues. The A₃₈₅/A₂₈₃ absorbance ratio of ferredoxins I and II are 0.56 and 0.62, respectively.

The N-terminal sequencing data of the two ferredoxins clearly indicate that ferredoxins I and II are different protein species. However, the two proteins exhibit a high degree of homology.

The physiological activity of ferredoxins I and II appears to be similar as far as the electron transfer in the phosphoroclastic reaction is concerned.     

Two EPR-detectable [4Fe-4S] clusters,N2a and N2b,are bound to the NuoI (TYKY) subunit of NADH:ubiquinone oxidoreductase (Complex I) from Rhodobacter capsulatus

NADH:ubiquinone oxidoreductases (Complex I) contain a subunit, TYKY in the bovine enzyme and NuoI in the enzyme from Rhodobacter capsulatus, which is assumed to bind two [4Fe-4S] clusters because it contains two sets of conserved cysteine motifs similar to those found in the 2[4Fe-4S] ferredoxins. It was recently shown that the TYKY subunit is not an ordinary 2[4Fe-4S] ferredoxin, but has a unique amino acid sequence, which is only found in NAD(P)H:quinone oxidoreductases and certain membrane-bound [NiFe]-hydrogenases expected to be involved in redox-linked proton translocation [FEBS Lett. 485 (2000) 1]. We have generated a set of R. capsulatus mutants in which five out of the eight conserved cysteine residues in NuoI were replaced by other amino acids. The resulting mutants fell into three categories with virtually no, intermediate or quite normal Complex I activities. EPR-spectroscopic analysis of the membranes of the C67S and C106S mutants, two mutants belonging to the second and third group, respectively, showed a specific 50% decrease of the EPR signal attributed to cluster N2. It is concluded that the NuoI (TYKY) subunit binds two clusters N2, called N2a and N2b, which exhibit very similar spectral features when analyzed by X-band EPR spectroscopy.     

Isolation and characterization of a rubredoxin and two ferredoxins from Desulfovibrio africanus

Rubredoxin and two distinct ferredoxins have been purified from Desulfovibrio africanus. The rubredoxin has a molecular weight of 6000 while the ferredoxins appear to be dimers of identical subunits of approximately 6000 to 7000 molecular weight. Rubredoxin contains one iron atom, no acid-labile sulfide and four cysteine residues per molecule. Its absorbance ratio A278/A490 is 2.23 and its amino acid composition is characterized by the absence of leucine and a preponderance of acidic amino acids. The two ferredoxins, designated I and II, are readily separated on DEAE-cellulose. The amino acid compositions of ferredoxins I and II show them to be different protein species; the greater number of acidic amino acid residues in ferredoxin I than in ferredoxin II appears to account for separation based on electronic charge. Both ferredoxins contain four iron atoms, four acid-labile residues per molecule. Spectra of the two ferredoxins differ from those of ferredoxins of other Desulfovibrio species by exhibiting a pronounced absorption peak at 283 nm consistent with an unusual high content of aromatic residues. The A385/A283 absorbance ratio of ferredoxins I and II are 0.56 and 0.62, respectively. The N-terminal sequencing data of the two ferredoxins clearly indicate that ferredoxins I and II are different protein species. However, the two proteins exhibit a high degree of homology.     

Iron-sulfur clusters and cysteine distribution in a ferredoxin from Azotobactervinelandii

In 80% dimethyl sulfoxide/H₂O, $\text{Azotobacter}$ ferredoxin FeS clusters can be extruded with benzene thiol. The extruded clusters have an absorption spectra maximum at 458 nm which is characteristic of 4Fe4S centers. The amino terminal sequence of the $\text{Azotobacter}$ ferredoxin has 7 of the 8 Cys residues at residue numbers 8, 11, 16, 20, 24, 39 and 42. Except for Cys 24, all of these residues can be correlated to homologous Cys residues in other bacterial ferredoxins. Although two thirds of the first 45 residues are identical to or conservative replacements for the first 43 residues of other bacterial ferredoxins, the insertion of Cys-24 indicates a major change in the environment of one of the two 4Fe4S clusters.     

Gene Triplication deduced from the Tertiary Structure of a Muscle Calcium Binding Protein

GENE duplication was first observed cytologically in the bar gene of drosophila¹. The idea, extended by Dixon², has been used to explain the obvious repeats in amino-acid sequence in the immunoglobulins, in bacterial ferredoxins, in haptoglobin α-2 (summarized in ref. 3) and in neurophysin⁴.     

Amino acid sequences of two ferredoxins from pokeweed, Phytolacca americana

The amino acid sequences of two ferredoxins isolated from pokeweed, Phytolacca americana, were determined. Tryptic peptides of maleyl-carboxymethyl-ferredoxin I and carboxymethyl-ferredoxin II were prepared and analyzed. The large peptides were further digested with staphylococcal protease and chymotrypsin. Ferredoxins I and II were composed of 96 and 98 amino acid residues, respectively. Though ferredoxin I lacks tryptophan and methionine, ferredoxin II contains both of them. In a comparison of the amino acid sequences with those of other higher plant ferredoxins, ferredoxin I is one residue shorter than others at the carboxyl-terminus and ferredoxin II one longer than others at the amino-terminus. Ferredoxins I and II differ in 23 sites from each other and in 27 to 37 sites from other higher plant ferredoxins. This suggests that duplication of the ferredoxin gene occurred after the divergence of pokeweed from other higher plants. A phylogenetic tree including all other ferredoxins was constructed.     

Amino acid sequence of ferredoxin from Arctium lappa

Ferredoxin from Arctium lappa consists of a single polypeptide chain of 97 residues, four of which are cysteine. These residues, which are in the active centre, are in identical positions to those of other ferredoxins. Overlap between residue positions 50/51 was not obtained, but amino acid composition of the two cyanogen bromide fragments which were overlapped corresponded with the amino acid composition of the total protein.     

Complete determination of disulfide bonds localized within the short consensus repeat units of decay accelerating factor (CD55 antigen).

Decay accelerating factor (DAF) has 4 SCR (short consensus repeat) units. Each SCR unit consists of approx. 60 amino acids characterized by having four conserved cysteine residues and several other highly conserved residues which include proline, tryptophan, tyrosine/phenylalanine and glycine. To determine the disulfide-bonding pattern, we used the urine form of DAF. After thermolysin and trypsin digestion, we isolated seven disulfide-linked peptides by HPLC purification. Because all of the cysteine residues are disulfide-bonded, DAF should contain eight disulfide bonds. After subtilisin and trypsin digestion, we isolated the eighth disulfide-bonded peptides by HPLC purification. From sequence analyses of these peptides, we could identify all disulfide bonds in the 4 SCR units of DAF as being between the first and the third and between the second and the fourth half-cystines within each SCR unit.