Expression of the gene encoding cytochrome c3 from the sulfate-reducing bacterium Desulfovibrio vulgaris in the purple photosynthetic bacterium Rhodobacter sphaeroides.

The gene encoding cytochrome c3 (cyc-gene) from Desulfovibrio vulgaris (Hildenborough) was cloned by G. Voordouw and S. Brenner (1986, Eur. J. Biochem. 159, 347-351). The gene was expressed in Escherichia coli but only the apoprotein was observed (W. Pollock, P. Chemerika, M. Forrest, J. Beatty, and G. Voordouw, 1989, J. Gen. Microbiol. 135, 2319-2328). In this study, the cyc-gene was cloned into the broad host range vector pRK404 and then introduced into the purple photosynthetic bacterium Rhodobacter sphaeroides. Cells grown anaerobically produced a significant amount of recombinant cytochrome c3. The purified protein contains four hemes and the N-terminal protein sequence is identical to the published sequence of the native cytochrome c3. Thus, R. sphaeroides was able to produce the mature cytochrome c3 by combining the four steps of protein synthesis, exporting the protein across the membrane, cleaving the signal peptide, and inserting four hemes. It appears that the D. vulgaris promoter is not very efficiently used by R. sphaeroides. However, replacement of the promoter with a R. sphaeroides promoter should result in cytochrome c3 overproduction.

Soluble cytochromes and ferredoxins from the marine purple phototrophic bacterium, Rhodopseudomonas marina.

Four soluble c-type cytochromes, the high redox potential 4-Fe-S ferredoxin known as HiPIP, a large molecular weight 2-Fe-S ferredoxin and a 4-Fe-S 'bacterial' ferredoxin, were isolated from extracts of two strains of Rps. marina. Cytochrome c-550, cytochrome c' and cytochrome c-549 were previously described, and we have extended their characterization. Cytochrome c-558, which has not previously been observed in Rps. marina, appears to be a low-spin isozyme of the more commonly observed high-spin cytochrome c'. HiPIP, which was not observed in previous work, was found to be abundant in Rps. marina. The 2-Fe-S ferredoxin, which has previously been observed only in Rps. palustris, has a native size greater than 100 kDa and a subunit size of 17 kDa. The 'bacterial' ferredoxin appears to have only a single four-iron-sulfur cluster. We examined photosynthetic membranes by difference spectroscopy and found abundant c-type cytochromes. Approximately one-quarter of the heme can be reduced by ascorbate and the remainder by dithionite. There is 2 nm difference between the high-potential heme (554 nm) and the low (552 nm). These characteristics resemble those of the tetraheme reaction center cytochrome of Rps. viridis. In addition to the electron transfer components, we found small amounts of a fluorescent yellow protein which has spectral resemblance to a photoactive yellow protein from Ec. halophila.

Expression of a cytochrome c2 isozyme restores photosynthetic growth of Rhodobacter sphaeroides mutants lacking the wild-type cytochrome c2 gene.

Deletion of the cytochrome c2 gene in the purple bacterium Rhodobacter sphaeroides renders it incapable of phototrophic growth (strain cycA65). However, suppressor mutants which restore the ability to grow phototrophically are obtained at relatively high frequency (1-10 in 10(7)). We examined two such suppressors (strains cycA65R5 and cycA65R7) and found the expected complement of electron transfer proteins minus cytochrome c2: SHP, c', c551.5, and c554. Instead of cytochrome c2 which elutes from DEAE-cellulose between SHP and cytochrome c', at about 50 mM ionic strength in wild-type extracts, we found a new high redox potential cytochrome c in the mutants which elutes with cytochrome c551.5 at about 150 mM ionic strength. The new cytochrome is more acidic than cytochrome c2, but is about the same size or slightly smaller (13,500 Da). The redox potential of the new cytochrome from strain cycA65R7 (294 mV) is about 70 mV lower than that of cytochrome c2. The 280 nm absorbance of the new cytochrome is smaller than that of cytochrome c2, which suggests that there is less tryptophan (the latter has two residues). In vitro kinetics of reduction by lumiflavin and FMN semiquinones show that the reactivity of the new cytochrome is similar to that of cytochrome c2, and that there is a relatively large positive charge (+2.6) at the site of reduction, despite the overall negative charge of the protein. This behavior is characteristic of cytochromes c2 and unlike the majority of bacterial cytochromes examined. Fourteen out of twenty-four of the N-terminal amino acids of the new cytochrome are identical to the sequence of cytochrome c2. The N-termini of the cycA65R5 and cycA65R7 cytochromes were the same. The kinetics and sequence data indicate that the new protein may be a cytochrome c2 isozyme, which is not detectable in wild-type cells under photosynthetic growth conditions. We propose the name iso-2 cytochrome c2 for the new cytochrome produced in the suppressor strains.