The nucleotide sequence of the Desulfovibrio gigas desulforedoxin gene indicates that the Desulfovibrio vulgaris rbo gene originated from a gene fusion event.

Expression of the rbo gene from Desulfovibrio vulgaris Hildenborough in Escherichia coli minicells and Western blotting (immunoblotting) of Desulfovibrio cell extracts with antibodies raised against a synthetic peptide indicated the presence of a 14-kDa polypeptide product, as expected from the gene sequence. Cloning and sequencing of the gene (dsr) for desulforedoxin, a 4-kDa redox protein from Desulfovibrio gigas, showed that it is formed by expression of an autonomous gene of 111 bp, not by processing of a 14-kDa protein. The results indicate that the rbo gene product, which has a 4-kDa desulforedoxin domain as the NH2 terminus, may have arisen by gene fusion. Shuffling and fusion of genes for redox protein domains can explain the large variety of redox proteins found in sulfate-reducing bacteria.     

Purification and characterization of desulfoferrodoxin. A novel protein from Desulfovibrio desulfuricans (ATCC 27774) and from Desulfovibrio vulgaris (strain Hildenborough) that contains a distorted rubredoxin center and a mononuclear ferrous center

A new type of non-heme iron protein was purified to homogeneity from extracts of Desulfovibrio desulfuricans (ATCC 27774) and Desulfovibrio vulgaris (strain Hildenborough). This protein is a monomer of 16-kDa containing two iron atoms per molecule. The visible spectrum has maxima at 495, 368, and 279 nm and the EPR spectrum of the native form shows resonances at g = 7.7, 5.7, 4.1 and 1.8 characteristic of a high-spin ferric ion (S = 5/2) with E/D = 0.08. M?ssbauer data indicates the presence of two types of iron: an FeS4 site very similar to that found in desulforedoxin from Desulfovibrio gigas and an octahedral coordinated high-spin ferrous site most probably with nitrogen/oxygen-containing ligands. Due to this rather unusual combination of active centers, this novel protein is named desulfoferrodoxin. Based on NH2-terminal amino acid sequence determined so far, the desulfoferrodoxin isolated from D. desulfuricans (ATCC 27774) appears to be a close analogue to a recently discovered gene product from D. vulgaris (Brumlik, M.J., and Voordouw, G. (1989) J. Bacteriol. 171, 49996-50004), which was suggested to be a rubredoxin oxidoreductase. However, reduced pyridine nucleotides failed to reduce the desulforedoxin-like center of this new protein.     

Five-gene cluster in Clostridium thermoaceticum consisting of two divergent operons encoding rubredoxin oxidoreductase- rubredoxin and rubrerythrin-type A flavoprotein- high-molecular-weight rubredoxin

A five-gene cluster encoding four nonheme iron proteins and a flavoprotein from the thermophilic anaerobic bacterium Clostridium thermoaceticum (Moorella thermoacetica) was cloned and sequenced. Based on analysis of deduced amino acid sequences, the genes were identified as rub (rubredoxin), rbo (rubredoxin oxidoreductase), rbr (rubrerythrin), fprA (type A flavoprotein), and a gene referred to as hrb (high-molecular-weight rubredoxin). Northern blot analysis demonstrated that the five-gene cluster is organized as two subclusters, consisting of two divergently transcribed operons, rbr-fprA-hrb and rbo-rub. The rbr, fprA, and rub genes were expressed in Escherichia coli, and their encoded recombinant proteins were purified. The molecular masses, UV-visible absorption spectra, and cofactor contents of the recombinant rubrerythrin, rubredoxin, and type A flavoprotein were similar to those of respective homologs from other microorganisms. Antibodies raised against Desulfovibrio vulgaris Rbr reacted with both native and recombinant Rbr from C. thermoaceticum, indicating that this protein was expressed in the native organism. Since Rbr and Rbo have been recently implicated in oxidative stress protection in several anaerobic bacteria and archaea, we suggest a similar function of these proteins in oxygen tolerance of C. thermoaceticum.     

Cloning of genes encoding redox proteins of known amino acid sequence from a library of the Desulfovibrio vulgaris (Hildenborough) genome

A library of 900 recombinant phages has been constructed for the genome of Desulfovibrio vulgaris Hildenborough (1.7 x 10(6) bp) by cloning size-fractionated Sau3A fragments (15-20 kb) into the replacement vector lambda-2001. When a hydrogenase gene probe, a 4.7-kb SalI-EcoRI fragment of known nucleotide sequence, was used to screen the plaque lifted library, 23 positive clones were found, which together span 31 kb of D. vulgaris DNA. To facilitate the cloning of genes with oligodeoxynucleotides as probes, DNA was purified for all clones in the library and spotted on a 16 x 16-cm grid of nitrocellulose. This grid was incubated sequentially to identify lambda clones containing the gene for redox proteins of known amino acid sequence: cytochrome c3 (one 18-mer----four clones), flavodoxin (one 17-mer and one 26-mer----one clone) and rubredoxin (one 44-mer----21 clones). The four cyc-positive clones are also recognized by the rubredoxin oligodeoxynucleotide probe. Restriction mapping defines a 35-kb region of the D. vulgaris chromosome in which the rub and cyc loci are separated by 17.5 kb. The nucleotide sequence of the rubredoxin gene was determined and the deduced amino acid sequence found to agree with that determined in Bruschi [Biochim. Biophys. Acta 434 (1976) 4-17] with the exception of Thr-21 which is found to be encoded by GAC, an Asp codon. A plausible ribosome-binding site precedes the N-terminal initiator methionine residue. Rubredoxin does not have an N-terminal signal sequence which is in agreement with the cytoplasmic location of this redox protein.     

Amino acid sequence and function of rubredoxin from Desulfovibrio vulgaris Miyazaki

Rubredoxin was purified from Desulfovibrio vulgaris Miyazaki. It was sequenced and some of its properties determined. Rubredoxin is composed of 52 amino acids. It is highly homologous to that from D. vulgaris Hildenborough. Its N-terminal methionyl residue is partially formylated. The millimolar absorption coefficients of the rubredoxin at 489 nm and 280 nm are 8.1 and 18.5, respectively, and the standard redox potential is +5 mV, which is slightly higher than those of other rubredoxins. Rubredoxin, as well as cytochrome c-553, was reduced with lactate by the action of lactate dehydrogenase of this organism, and the reaction was stimulated with 2-methyl-1,4-naphthoquinone. It is suggested that rubredoxin, in collaboration with membranous quinone, functions as a natural electron carrier for cytoplasmic lactate dehydrogenase of this organism, whereas cytochrome c-553 plays the same role for periplasmic lactate dehydrogenase.     

Rubrerythrin and rubredoxin oxidoreductase in Desulfovibrio vulgaris: a novel oxidative stress protection system

Evidence is presented for an alternative to the superoxide dismutase (SOD)-catalase oxidative stress defense system in Desulfovibrio vulgaris (strain Hildenborough). This alternative system consists of the nonheme iron proteins, rubrerythrin (Rbr) and rubredoxin oxidoreductase (Rbo), the product of the rbo gene (also called desulfoferrodoxin). A Deltarbo strain of D. vulgaris was found to be more sensitive to internal superoxide exposure than was the wild type. Unlike Rbo, expression of plasmid-borne Rbr failed to restore the aerobic growth of a SOD-deficient strain of Escherichia coli. Conversely, plasmid-borne expression of two different Rbrs from D. vulgaris increased the viability of a catalase-deficient strain of E. coli that had been exposed to hydrogen peroxide whereas Rbo actually decreased the viability. A previously undescribed D. vulgaris gene was found to encode a protein having 50% sequence identity to that of E. coli Fe-SOD. This gene also encoded an extended N-terminal sequence with high homologies to export signal peptides of periplasmic redox proteins. The SOD activity of D. vulgaris is not affected by the absence of Rbo and is concentrated in the periplasmic fraction of cell extracts. These results are consistent with a superoxide reductase rather than SOD activity of Rbo and with a peroxidase activity of Rbr. A joint role for Rbo and Rbr as a novel cytoplasmic oxidative stress protection system in D. vulgaris and other anaerobic microorganisms is proposed.     

Nucleotide sequence of the gene encoding the hydrogenase from Desulfovibrio vulgaris (Hildenborough)

The nucleotide sequence of the 4.7-kb SalI/EcoRI insert of plasmid pHV 15 containing the hydrogenase gene from Desulfovibrio vulgaris (Hildenborough) has been determined with the dideoxy chain-termination method. The structural gene for hydrogenase encodes a protein product of molecular mass 45820 Da. The NH2-terminal sequence of the enzyme deduced from the nucleic acid sequence corresponds exactly to the amino acid sequence determined by Edman degradation. The nucleic acid sequence indicates that a N-formylmethionine residue precedes the NH2-terminal amino acid Ser-1. There is no evidence for a leader sequence. The NH2-terminal part of the hydrogenase shows homology to the bacterial [8Fe-8S] ferredoxins. The sequence Cys-Ile-Xaa-Cys-Xaa-Xaa-Cys-Xaa-Xaa-Xaa-Cys-Pro-Xaa-Xaa-Ala-(Ile) occurs twice both in the hydrogenase and in [8Fe-8S] ferredoxins, where the Cys residues have been shown to coordinate two [4Fe-4S] clusters [Adman, E. T., Sieker, L. C. and Jensen, L. H. (1973) J. Biol. Chem. 248, 3987-3996]. These results, therefore, suggest that two electron-transferring ferredoxin-like [4Fe-4S] clusters are located in the NH2-terminal segment of the hydrogenase molecule. There are ten more Cys residues but it is not clear which four of these could participate in the formation of the third cluster, which is thought to be the hydrogen binding centre. Another gene, encoding a protein of molecular mass 13493 Da, was found immediately downstream from the gene for the 46-kDa hydrogenase. The nucleic acid sequence suggests that the hydrogenase and the 13.5-kDa protein belong to a single operon and are coordinately expressed. Since dodecylsulfate gel electrophoresis of purified hydrogenase indicates the presence of a 13.5-kDa polypeptide in addition to the 46-kDa component, it is proposed that the hydrogenase from D. vulgaris (Hildenborough) is a two-subunit enzyme.     

Organization of the genes encoding [Fe] hydrogenase in Desulfovibrio vulgaris subsp. oxamicus Monticello.

The genes encoding the periplasmic [Fe] hydrogenase from Desulfovibrio vulgaris subsp. oxamicus Monticello were cloned by exploiting their homology with the hydAB genes from D. vulgaris subsp. vulgaris Hildenborough, in which this enzyme is present as a heterologous dimer of alpha and beta subunits. Nucleotide sequencing showed that the enzyme is encoded by an operon in which the gene for the 46-kilodalton (kDa) alpha subunit precedes that of the 13.5-kDa beta subunit, exactly as in the Hildenborough strain. The pairs of hydA and hydB genes are highly homologous; both alpha subunits (420 amino acid residues) share 79% sequence identity, while the unprocessed beta subunits (124 and 123 amino acid residues, respectively) share 71% sequence identity. In contrast, there appears to be no sequence homology outside these coding regions, with the exception of a possible promoter element, which was found approximately 90 base pairs upstream from the translational start of the hydA gene. The recently discovered hydC gene, which may code for a 65.8-kDa fusion protein (gamma) of the alpha and beta subunits and is present immediately downstream from the hydAB genes in the Hildenborough strain, was found to be absent from the Monticello strain. The implication of this result for the possible function of the hydC gene product in Desulfovibrio species is discussed.     

Solution structure of the two-iron rubredoxin of Pseudomonas oleovorans determined by NMR spectroscopy and solution X-ray scattering and interactions with rubredoxin reductase

Here we provide insights into the molecular structure of the two-iron 19-kDa rubredoxin (AlkG) of Pseudomonas oleovorans using solution-state nuclear magnetic resonance (NMR) and small-angle X-ray scattering studies. Sequence alignment and biochemical studies have suggested that AlkG comprises two rubredoxin folds connected by a linker region of approximately 70 amino acid residues. The C-terminal domain (C-Rb) of this unusual rubredoxin, together with approximately 35 amino acid residues of the predicted linker region, was expressed in Escherichia coli, purified in the one-iron form and the structure of the cadmium-substituted form determined at high-resolution by NMR spectroscopy. The structure shows that the C-Rb domain is similar in fold to the conventional one-iron rubredoxins from other organisms, whereas the linker region does not have any discernible structure. This tandem "flexible-folded" structure of the polypeptide chain derived for the C-Rb protein was confirmed using solution X-ray scattering methods. X-ray scattering studies of AlkG indicated that the 70-amino acid residue linker forms a structured, yet mobile, polypeptide segment connecting the globular N- and C-terminal domains. The X-ray scattering studies also showed that the N-terminal domain (N-Rb) has a molecular conformation similar to that of C-Rb. The restored molecular shape indicates that the folded N-Rb and C-Rb domains of AlkG are noticeably separated, suggesting some domain movement on complex formation with rubredoxin reductase to allow interdomain electron transfer between the metal centers in AlkG. This study demonstrates the advantage of combining X-ray scattering and NMR methods in structural studies of dynamic, multidomain proteins that are not suited to crystallographic analysis. The study forms a structural foundation for functional studies of the interaction and electron-transfer reactions of AlkG with rubredoxin reductase, also reported herein.     

A role for rubredoxin in oxidative stress protection in Desulfovibrio vulgaris: catalytic electron transfer to rubrerythrin and two-iron superoxide reductase

Desulfovibrio vulgaris rubredoxin, which contains a single [Fe(SCys)4] site, is shown to be a catalytically competent electron donor to two enzymes from the same organism, namely, rubrerythrin and two-iron superoxide reductase (a.k.a. rubredoxin oxidoreductase or desulfoferrodoxin). These two enzymes have been implicated in catalytic reduction of hydrogen peroxide and superoxide, respectively, during periods of oxidative stress in D. vulgaris, but their proximal electron donors had not been characterized. We further demonstrate the incorrectness of a previous report that rubredoxin is not an electron donor to the superoxide reductase and describe convenient assays for demonstrating the catalytic competence of all three proteins in their respective functions. Rubrerythrin is shown to be an efficient rubredoxin peroxidase in which the rubedoxin:hydrogen peroxide redox stoichiometry is 2:1 mol:mol. Using spinach ferredoxin-NADP+ oxidoreductase (FNR) as an artificial, but proficient, NADPH:rubredoxin reductase, rubredoxin was further found to catalyze rapid and complete reduction of all Fe3+ to Fe2+ in rubrerythrin by NADPH under anaerobic conditions. The combined system, FNR/rubredoxin/rubrerythrin, was shown to function as a catalytically competent NADPH peroxidase. Another small rubredoxin-like D. vulgaris protein, Rdl, could not substitute for rubredoxin as a peroxidase substrate of rubrerythrin. Similarly, D. vulgaris rubredoxin was demonstrated to efficiently catalyze reduction of D. vulgaris two-iron superoxide reductase and, when combined with FNR, to function as an NADPH:superoxide oxidoreductase. We suggest that, during periods of oxidative stress, rubredoxin could divert electron flow from the electron transport chain of D. vulgaris to rubrerythrin and superoxide reductase, thereby simultaneously protecting autoxidizable redox enzymes and lowering intracellular hydrogen peroxide and superoxide levels.     

The primary structure of rubrerythrin, a protein with inorganic pyrophosphatase activity from Desulfovibrio vulgaris. Comparison with hemerythrin and rubredoxin.

The complete polypeptide chain of rubrerythrin from the sulfate reducing bacterium Desulfovibrio vulgaris, strain Hildenborough NCIB 8303, was found by protein chemical techniques to consist of 191 residues and to have the amino acid sequence [sequence: see text] The C-terminal part of the protein (position 153----191) shows the typical sequence features of rubredoxin, a protein with a nonheme iron center also present in the same and other Desulfovibrio species. Based on the known three-dimensional structure of D. desulfuricans rubredoxin, we propose that the C-terminal part of rubrerythrin is folded in a similar way and suggest that the deletion of the extra 10 residues is compatible with the same basic rubredoxin-fold. After characterization of the C-terminal region, and in contrast to what could be expected from previously published spectroscopic analyses, the N-terminal region 1-152 of rubrerythrin appears to have no sequence similarity with the eukaryotic protein hemerythrin which is known to contain a binuclear iron center bound by 5 histidine ligands. However, the N-terminal region of rubrerythrin does contain 5 histidine residues but they are differently spaced along the peptide chain. We suggest that at least one of the 3 histidine residues located in the rubredoxin-like center of rubrerythrin may be liganded to one iron atom of the hemerythrin-like center. This paper is the first sequence report of a protein with pyrophosphatase activity although the physiological substrate for the rubrerythrin may be not inorganic pyrophosphate.     

Deletion of the rbo Gene Increases the Oxygen Sensitivity of the Sulfate-Reducing Bacterium Desulfovibrio vulgaris Hildenborough

The rbo gene of Desulfovibrio vulgaris Hildenborough encodes rubredoxin oxidoreductase (Rbo), a 14-kDa iron sulfur protein; forms an operon with the gene for rubredoxin; and is preceded by the gene for the oxygen-sensing protein DcrA. We have deleted the rbo gene from D. vulgaris with the sacB mutagenesis procedure developed previously (R. Fu and G. Voordouw, Microbiology 143:1815–1826, 1997). The absence of the rbo-gene in the resulting mutant, D. vulgaris L2, was confirmed by PCR and protein blotting with Rbo-specific polyclonal antibodies. D. vulgaris L2 grows like the wild type under anaerobic conditions. Exposure to air for 24 h caused a 100-fold drop in CFU of L2 relative to the wild type. The lag times of liquid cultures of inocula exposed to air were on average also greater for L2 than for the wild type. These results demonstrate that Rbo, which is not homologous with superoxide dismutase or catalase, acts as an oxygen defense protein in the anaerobic, sulfate-reducing bacterium D. vulgaris Hildenborough and likely also in other sulfate-reducing bacteria and anaerobic archaea in which it has been found.     

Complementation of an Escherichia coli pyrF mutant with DNA from Desulfovibrio vulgaris.

A PyrF- mutant of Escherichia coli (SK1108, pyrF::Tn5 Kanr) was complemented with the Desulfovibrio vulgaris (Hildenborough) structural gene for orotidine-5'-phosphate decarboxylase (EC 4.1.1.23). Either orientation of a 1.6-kilobase-pair D. vulgaris DNA fragment (pLP3B or pLP3A) complemented the PyrF- strain suggesting that the D. vulgaris pyrF promoter was functional. The apparent product of the D. vulgaris pyrF gene was a single 26-kilodalton polypeptide. These results demonstrate the utility of E. coli cloning systems in studying metabolic and energetic pathways in sulfate-reducing bacteria.     

Cloning, sequencing, and expression of the gene encoding the high-molecular-weight cytochrome c from Desulfovibrio vulgaris Hildenborough.

By using a synthetic deoxyoligonucleotide probe designed to recognize the structural gene for cytochrome cc3 from Desulfovibrio vulgaris Hildenborough, a 3.7-kb XhoI genomic DNA fragment containing the cc3 gene was isolated. The gene encodes a precursor polypeptide of 58.9 kDa, with an NH2-terminal signal sequence of 31 residues. The mature polypeptide (55.7 kDa) has 16 heme binding sites of the form C-X-X-C-H. Covalent binding of heme to these 16 sites gives a holoprotein of 65.5 kDa with properties similar to those of the high-molecular-weight cytochrome c (Hmc) isolated from the same strain by Higuchi et al. (Y. Higuchi, K. Inaka, N. Yasuoka, and T. Yagi, Biochim. Biophys. Acta 911:341-348, 1987). Since the data indicate that cytochrome cc3 and Hmc are the same protein, the gene has been named hmc. The Hmc polypeptide contains 31 histidinyl residues, 16 of which are integral to heme binding sites. Thus, only 15 of the 16 hemes can have bis-histidinyl coordination. A comparison of the arrangement of heme binding sites and coordinated histidines in the amino acid sequences of cytochrome c3 and Hmc from D. vulgaris Hildenborough suggests that the latter contains three cytochrome c3-like domains. Cloning of the D. vulgaris Hildenborough hmc gene into the broad-host-range vector pJRD215 and subsequent conjugational transfer of the recombinant plasmid into D. desulfuricans G200 led to expression of a periplasmic Hmc gene product with covalently bound hemes.     

The biosynthesis of the ubiquinol-cytochrome c reductase complex in yeast. DNA sequence analysis of the nuclear gene coding for the 14-kDa subunit

The nuclear gene coding for the imported 14-kDa subunit of the ubiquinol-cytochrome c reductase of yeast mitochondria has been sequenced in an attempt to define regulatory and protein topogenic elements. The gene has a length of 381 base pairs and is potentially capable of encoding a polypeptide of 14561 Da. It is transcribed into a single low-abundance RNA of 680 nucleotides whose 5' and 3' termini map, respectively, 30-35 nucleotides upstream and 180-190 nucleotides downstream of the initiator and termination codons. Consistent with the estimated low level of the mRNA, codon usage in the gene is not strongly biased and other features, characteristic of highly expressed genes in yeast, are absent. The 14-kDa protein is predicted to be a predominantly hydrophilic protein, with only a single, short hydrophobic stretch located between positions 19-38. Comparison with other imported mitochondrial proteins so far sequenced has failed to reveal unifying features that might serve as targeting elements. Steady-state levels of the 14-kDa and 11-kDa subunits are reduced in mit- mutants which synthesize truncated forms of apocytochrome b and in these, newly synthesized subunits exhibit a specifically increased turnover rate. We suggest that association of these two subunits with the complex may be mediated or enhanced by interaction with other subunits, in particular cytochrome b.     

Cloning and nucleotide sequence of a gene for Actinomyces naeslundii WVU45 type 2 fimbriae.

A genomic library of Actinomyces naeslundii WVU45 DNA in Escherichia coli was screened for antigen expression with rabbit antibody against A. naeslundii fimbriae. Western blotting (immunoblotting) of one recombinant clone carrying a 13.8-kilobase-pair insert revealed a 59-kilodalton (kDa) immunoreactive protein. A protein of similar electrophoretic mobility was detected from the isolated fimbrial antigen. Expression of the 59-kDa cloned protein in E. coli was directed by a promoter from the insert. The DNA sequence of the subunit gene was determined, and an open reading frame of 1,605 nucleotides was identified which was preceded by a putative ribosome-binding site and followed by two inverted repeats of 14 and 17 nucleotides, respectively. The reading frame encoded a protein of 534 amino acids (calculated molecular weight, 57,074), and the N-terminal sequence resembled that of a signal peptide. The presence of a 32-amino-acid signal peptide was indicated by amino-terminal sequencing of the fimbriae from A. naeslundii. The sequence, as determined by Edman degradation, was identical to that deduced from the DNA sequence beginning at predicted residue 33 of the latter sequence. Moreover, the amino acid composition of the predicted mature protein was similar to that of the isolated fimbriae from A. naeslundii. Thus, the cloned gene encodes a subunit of A. naeslundii fimbriae.     

Expression of a synthetic gene coding for the amino acid sequence of Clostridium pasteurianum rubredoxin.

A synthetic gene based on the published amino acid sequence for Clostridium pasteurianum rubredoxin was constructed, cloned in Escherichia coli 71/18 and expressed using the T7 RNA polymerase/promoter system in E. coli HMS273. UV/visible spectroscopy and metal analyses indicated that the as-isolated synthetic gene product is a mixture of holo-(i.e. iron-containing) rubredoxin and zinc-substituted rubredoxin, with the latter amounting to approximately 70% of the total rubredoxin. The UV/visible absorption and resonance Raman spectra of the cloned holorubredoxin are characteristic of the native rubredoxin-type iron site. N-terminal amino acid sequencing suggests that the gene product consists of at least three polypeptide species with the initial sequences (approximate relative abundances): Met-Met-Lys-... (63%), blocked (30%) and Met-Lys-... (7%). The blocked portion presumably consists of a mixture of nMet-Met-Lys-... and nMet-Lys-..., where nMet represents an amino-blocked methionine residue.     

Expression of the gene encoding cytochrome c3 from Desulfovibrio vulgaris (Hildenborough) in Escherichia coli: export and processing of the apoprotein

The expression of cytochrome c3 from Desulfovibrio vulgaris (Hildenborough) was examined in Escherichia coli transformed with either of two plasmids, pJ8 and pJ81. The former has an 840 bp insert of D. vulgaris DNA, containing the structural gene for cytochrome c3 (387 bp) and its promoter region. Plasmid pJ81 was generated from pJ8 by deoxyoligonucleotide-directed mutagenesis to direct the synthesis of a protein with an altered signal peptidase cleavage site [Ala(-1)----Asp(-1)]. Synthesis of the 14 kDa precursor, which was partly processed to the 12 kDa mature protein, was observed in cells of E. coli TG2(pJ8) by SDS gel electrophoresis and Western blotting. Analysis of spheroplasts revealed that the processed polypeptide was present in the periplasm while the precursor was found only in the membrane/cytoplasmic fraction. No processing was observed in E. coli TG2(pJ81) cells, due to the mutation of the signal peptide cleavage site. No insertion of haem into the E. coli product could be detected in E. coli TG2(pJ8) cells by post-electrophoretic protohaem fluorescence analysis. The sensitivity of the cytochrome c3 synthesized in E. coli TG2(pJ8) to digestion by chymotrypsin also indicated that the apoprotein was formed. The results indicate that E. coli is capable of synthesizing and exporting the cytochrome c3 polypeptide, but fails to insert the haems.