Journal of Biological Chemistry · 1999 · 53 citations · 21 references
Life SciencesBiochemistryMicrobiologyMicrobial MetabolismMicrobial PhysiologyElectron Transfer ReactionsRedox CofactorsRedox EnzymesRedox ReactionsCytochrome C Oxidation StateCytochrome Electron TransferCytochrome EnzymesCytochrome SystemsElectron TransferHeme Electron TransferMicrobial Electron TransferMicrobial Redox Processes
The dorC gene of the dimethyl sulfoxide respiratory (dor) operon of Rhodobacter capsulatus encodes a pentaheme c-type cytochrome that is involved in electron transfer from ubiquinol to periplasmic dimethyl sulfoxide reductase. DorC was expressed as a C-terminal fusion to an 8-amino acid FLAG epitope and was purified from detergent-solubilized membranes by ion exchange chromatography and immunoaffinity chromatography. The DorC protein had a subunitM r = 46,000, and pyridine hemochrome analysis indicated that it contained 5 mol heme c/mol DorC polypeptide, as predicted from the derived amino acid sequence of the dorCgene. The reduced form of DorC exhibited visible absorption maxima at 551.5 nm (α-band), 522 nm (β-band), and 419 nm (Soret band). Redox potentiometry of the heme centers of DorC identified five components (n = 1) with midpoint potentials of −34, −128, −184, −185, and −276 mV. Despite the low redox potentials of the heme centers, DorC was reduced by duroquinol and was oxidized by dimethyl sulfoxide reductase. The dorC gene of the dimethyl sulfoxide respiratory (dor) operon of Rhodobacter capsulatus encodes a pentaheme c-type cytochrome that is involved in electron transfer from ubiquinol to periplasmic dimethyl sulfoxide reductase. DorC was expressed as a C-terminal fusion to an 8-amino acid FLAG epitope and was purified from detergent-solubilized membranes by ion exchange chromatography and immunoaffinity chromatography. The DorC protein had a subunitM r = 46,000, and pyridine hemochrome analysis indicated that it contained 5 mol heme c/mol DorC polypeptide, as predicted from the derived amino acid sequence of the dorCgene. The reduced form of DorC exhibited visible absorption maxima at 551.5 nm (α-band), 522 nm (β-band), and 419 nm (Soret band). Redox potentiometry of the heme centers of DorC identified five components (n = 1) with midpoint potentials of −34, −128, −184, −185, and −276 mV. Despite the low redox potentials of the heme centers, DorC was reduced by duroquinol and was oxidized by dimethyl sulfoxide reductase. trimethylamine-N-oxide polyacrylamide gel electrophoresis The ability to use Me2SO and trimethylamine-N-oxide (TMAO)1 as electron acceptors is widespread among facultative aerobic bacteria and the organization of the Me2SO and TMAO respiratory chains is now well defined (1McEwan A.G. Hanson G.R. Bailey S. Biochem. Soc. Trans. 1998; 26: 390-394Crossref PubMed Scopus (6) Google Scholar). In the Me2SO respiratory system of purple photosynthetic bacteria such as Rhodobacter capsulatus, electrons are transferred from primary dehydrogenases via the ubiquinone pool to a periplasmic Me2SO reductase (2McEwan A.G. Richardson D.J. Hudig H. Ferguson S.J. Jackson J.B. Biochim. Biophys. Acta. 1989; 973: 308-314Crossref Scopus (28) Google Scholar). Recent sequence and mutational analysis of the Me2SO respiratory gene cluster from photosynthetic bacteria has identified a pentaheme c-type cytochrome (DorC) as the likely mediator of electron transfer from ubiquinol to Me2SO reductase (3Mouncey N.J. Chaudhary M. Kaplan S. J. Bacteriol. 1997; 179: 7617-7624Crossref PubMed Google Scholar). 2Shaw, A. L., Knaeblein, J., Leimkuhler, S., Hanson, G. R., Klipp, W., and McEwan, A. G. (1999)Microbiology, in press. 2Shaw, A. L., Knaeblein, J., Leimkuhler, S., Hanson, G. R., Klipp, W., and McEwan, A. G. (1999)Microbiology, in press. The TMAO respiratory (Tor) system of Escherichia coli (5Mejean V. Iobbi-Nivol C. Lepetellier M. Giordano G. Chippaux M. Pascal M.-C. Mol. Microbiol. 1994; 11: 1169-1179Crossref PubMed Scopus (214) Google Scholar) is very similar to the Dor system of R. capsulatus, and they differ from the Me2SO respiratory (Dms) system of E. coli. The Me2SO reductase (DmsABC) of E. coli can be purified as a menaquinol-oxidizing Me2SO reductase complex that lacks c-type cytochromes (6Weiner J.H. Rothery R.A. Sambasivarao D. Trieber C.A. Biochim. Biophys. Acta. 1992; 1102: 1-18Crossref PubMed Scopus (118) Google Scholar). Me2SO reductase from R. capsulatus has been purified as a monomeric protein containing a pterin molybdenum cofactor (Moco) as its only prosthetic group (7McEwan A.G. Ferguson S.J. Jackson J.B. Biochem. J. 1991; 274: 305-308Crossref PubMed Scopus (91) Google Scholar). This property of Me2SO reductase has many advantages for spectroscopic characterization of Moco, but a major deficiency has been the lack of a physiologically relevant electron donor. To overcome this problem we describe in this paper the purification and characterization of DorC. The derived amino acid sequence of Rhodobacter DorC indicated that it was related to members of the NirT class of tetrahemec-type cytochromes (3Mouncey N.J. Chaudhary M. Kaplan S. J. Bacteriol. 1997; 179: 7617-7624Crossref PubMed Google Scholar).2 However, DorC is predicted to be a pentaheme cytochrome with a fifth c-type heme binding motif in the C-terminal polypeptide, which is absent from the tetraheme members of the NirT class. Almost all of the members of the NirT class are involved in an anaerobic respiratory pathway, and their role appears to be to catalyze electron transfer from the Q-pool to a periplasmic terminal reductase (8Berks B.C. Ferguson S.J. Moir J.W.B. Richardson D.J. Biochim. Biophys. Acta. 1995; 1232: 97-173Crossref PubMed Scopus (487) Google Scholar). Although molecular genetic studies have provided an insight into the function and likely properties of cytochromes of the NirT class, very little is known about their biochemical and spectroscopic properties. A 1.4-kilobase DNA fragment containing the complete dorC nucleotide sequence except for the stop codon and including the upstream regulatory sequences of the dor operon was amplified by polymerase chain reaction using plasmid pALS3 as a template.2 The forward primer was designed so that a BamHI site would be synthesized while the reverse primer directed the synthesis of anEcoRI site plus a sequence to include a C-terminal 8-amino acid FLAG epitope (9Hopp T.P. Prockett K.S. Price V. Libby R.T. March C.J. Ceretti P. Urdal D.L. Conlon P.J. BioTechnology. 1988; 6: 1205-1210Crossref Scopus (743) Google Scholar). The 1.4-kilobase polymerase chain reaction product was cloned into pDSK519 (10Keen S. Tamaki D. Kobayashi D. Trollinger D. Gene (Amst.). 1988; 70: 191-197Crossref PubMed Scopus (1263) Google Scholar) to create plasmid pALS300. pALS300 was transferred from E. coli S17–1 to R. capsulatus 37b4 by conjugation (11Simon R. Priefer U. Puhler A. BioTechnology. 1983; 1: 784-791Crossref Scopus (5562) Google Scholar). DorC was expressed in 2l R. capsulatus cells grown for 24 h under phototrophic conditions in RCV medium (12Weaver P.F. Wall J.D. Gest H. Arch. Microbiol. 1975; 105: 207-216Crossref PubMed Scopus (543) Google Scholar) in the presence of 60 mm Me2SO. Membranes were prepared from the cells using a French Press as in Ref. 2McEwan A.G. Richardson D.J. Hudig H. Ferguson S.J. Jackson J.B. Biochim. Biophys. Acta. 1989; 973: 308-314Crossref Scopus (28) Google Scholar and were resuspended in Tris-buffered saline. Proteins were solubilized by treatment of membranes with 4% Triton X-100 for 90 min at room temperature. The mixture was then centrifuged (145,000 ×g, 90 min, 4 °C), and the soluble fraction was collected. The soluble fraction containing DorC (containing approximately 100 mg of protein) was dialyzed against 50 mm Tris-HCl, pH 8.0, 0.2% Triton X-100 and then charged onto a DEAE-Trisacryl column (bed volume, 15 ml), equilibrated with the same buffer. DorC was eluted using a 100-ml linear gradient of NaCl from 0 to 1 m. The DorC-enriched fraction from the anion exchange chromatography was dialyzed against Tris-buffered saline plus 0.2% Triton X-100 and then charged onto a 2-ml column containing FLAG affinity resin (Eastman Kodak). Bound DorC was eluted with 0.1 m glycine, pH 3.5, 0.2% Triton X-100. 2-ml fractions were collected, and the pH was immediately raised by addition of 50 μl 1 m Tris-HCl, pH 8.0. Optical spectra of DorC were collected using a Hitachi U-3000 spectrophotometer. Pyridine hemochrome analysis of DorC was carried out according to the method of Berry and Trumpower (13Berry E.A. Trumpower B.L. Anal. Biochem. 1987; 161: 1-15Crossref PubMed Scopus (728) Google Scholar). For reduction of DorC with duroquinol, the cytochrome was incubated with 2 units ml−1 Clostridial diaphorase, 25 μm duroquinone and 200 μmNADH. Redox potentiometry of DorC was performed by the method of Dutton (14Dutton P.L. Methods Enzymol. 1978; 54: 411-435Crossref PubMed Scopus (724) Google Scholar). Absorption spectra were recorded between 500 and 600 nm using a Jasco 7850 UV-visible spectrophotometer. Redox titrations were performed with dithionite as reductant and ferricyanide as oxidant in the presence of the following mediators: 2.5 μmbenzoquinone (Eo′ = +280 mV), 2.5 μm 1,2-naphthoquinone (Eo′ = +145 mV), 2.5 μm 1,4-naphthoquinone (Eo′ = +60 mV), 20 μm duroquinone (Eo′ = +5 mV), 20 μm1,4-dihydroxynaphthoquinone (Eo′ = −145 mV), 5 μmanthroquinone-2-sulfonate (Eo′ = −225 mV), and 5 μmbenzylviologen (Eo′ = −350 mV). The absorbance change at 551.5–540 nm was plotted against redox potential, and theoretical Nernstian curves were fitted to the data using Sigma Plot. SDS-PAGE analysis of proteins and Western blotting using an anti-FLAG antibody were performed as in Refs. 15Laemmli U. Nature. 1971; 227: 680-685Crossref Scopus (205531) Google Scholar and16Hatton A.D. Malin G. McEwan A.G. Arch. Microbiol. 1994; 162: 148-150Crossref Scopus (14) Google Scholar. Protein concentration was determined as in Ref. 17Smith P.K. Krohn R.I. Hermanson G.T. Maliia A.K. Gartner F.H. Provenzano M.D. Fujimoto E.K. Goeke N.M. Olson B.J. Klenk D.C. Anal. Biochem. 1985; 150: 76-85Crossref PubMed Scopus (18349) Google Scholar. A plasmid for the expression of DorC with an 8-amino acid FLAG epitope attached to the C terminus was constructed as described under “Experimental Procedures.” This plasmid was expressed in R. capsulatuscells grown phototrophically in the presence of Me2SO. Membrane and soluble fractions were prepared, and production of DorC was monitored by Western blotting. A single immunoreactive polypeptide with a molecular mass of 46 kDa was observed; this was the expected size of the DorC-FLAG polypeptide (data not shown). The putative DorC polypeptide was located exclusively in the membrane fraction. Several detergents were tested for their ability to solubilize DorC. Triton X-100 proved to be the most efficient, but even this detergent failed to completely solubilize DorC because an immunoreactive polypeptide could still be detected in the detergent-insoluble membrane fraction. The detergent-solubilized fraction was applied to a DEAE-Trisacryl anion exchange column. DorC was eluted from the column at about 450 mm NaCl. The fractions containing DorC were then applied to a FLAG-affinity resin and eluted as described under “Experimental Procedures.” The pooled fractions containing the colored cytochrome were analyzed by SDS-PAGE. Fig. 1 shows the presence of a single polypeptide M r = 46,000 corresponding to DorC. This protein cross-reacted with the anti-FLAG antibody in Western blots and stained for heme-dependent peroxidase activity (data not shown), confirming that the DorC-FLAG polypeptide had been purified. Fig.2 shows a dithionite-reduced minus ferricyanide-oxidized spectrum of DorC. Absorption peaks at 551.5 nm (α-band), 522 nm (β-band), and 419 nm (Soret band) were typical of a reduced cytochrome. The absorption minimum at 408 nm represents the Soret absorption band of the oxidized form of DorC. A spectrum of the resting form of DorC exhibited this 408 nm peak, and the absence of the α-band and β-band of the reduced cytochrome indicates that the DorC, as prepared, is in a fully oxidized state. The number of heme groups per DorC polypeptide was determined by pyridine hemochrome analysis. Fig. 3shows hemichrome and hemochrome spectra of DorC in alkaline pyridine. The presence of an absorption maximum at 550 nm in the pyridine hemochrome spectrum (Fig. 3) confirmed that DorC contained onlyc-type hemes. Using the method of Berry and Trumpower (13Berry E.A. Trumpower B.L. Anal. Biochem. 1987; 161: 1-15Crossref PubMed Scopus (728) Google Scholar), it was calculated that there were 4.9 mol c-type heme/mol DorC. This confirms that DorC is a pentaheme, as indicated from the primary structure analysis.2 The thermodynamic properties of the heme centers in DorC were analyzed by redox potentiometry. It was observed that during reductive titration no reduction of the cytochrome was seen until the ambient redox potential (at pH 7.0) was below 0 mV. This is consistent with the observation that sodium ascorbate (Em = +65 mV) did not reduce DorC as seen in reduced minus oxidized difference spectra (data not shown). Fig.4 shows the change in absorbance of the a band (551.5 nm minus 540 nm as the reference wavelength) of DorC as a function of the ambient redox potential. The data obtained could be fitted to a five-component Nernstian curve (n = 1) with midpoint potentials (Em7.0) of −34 ± 2, −128 ± 2, −184 ± 3, −185 ± 3, and −276 ± 3 mV. DorC is the only protein encoded by the dor operon that could act as a mediator of electron transfer between ubiquinol and Me2SO reductase.2 Before reconstitution of this electron transfer pathway was attempted, the ability of duroquinol to reduce DorC was investigated. A duroquinol-reduced spectrum of DorC indicated that at least some of the heme centers of the protein had been reduced (data not shown). For the reconstitution of electron transfer from quinol to Me2SO reductase DorC was titrated with an approximately stoichiometric amount of duroquinol under anaerobic conditions. This led to an increase in absorbance at 551.5 nm (Fig. 5). Upon addition of Me2SO reductase there was a decrease in the absorption of the α-band of DorC (Fig. 5), indicating that electron transfer from DorC to Me2SO reductase had occurred. Control experiments showed that the oxidation of DorC was specifically dependent upon Me2SO reductase. DorC from R. capsulatus is the first pentaheme cytochrome in the NirT class to be purified and characterized. The presence of photosynthetic pigments in the membrane of R. capsulatus made the detection of the native cytochrome very difficult and necessitated the strategy of attaching a FLAG peptide as a tag for detection and affinity purification. The protein was purified from R. capsulatus because all of our attempts to express DorC in E. coli were unsuccessful. 3A. L. Shaw and A. G. McEwan, unpublished observations. DorC was located exclusively in the membrane of R. capsulatus, and this would be consistent with a role in ubiquinol oxidation. In vivo, it is expected that DorC would be a peripheral membrane protein anchored to the membrane by its hydrophobic N terminus, and this explains why detergents were required to solubilize the protein. The optical spectroscopy of DorC identified an absorption maximum at 551.5 nm for the α-band. This is a symmetrical peak with very little complexity, and it suggests that although the midpoint redox potentials of the heme centers differ, it is likely that each possesses the same fifth and sixth ligands. There was no evidence for high spin five coordinate heme with an optical band beyond 600 nm as seen in cytochrome c′ for example (18Moore G.R. Pettigrew G.W. Cytochromes c: Biological Aspects. Springer-Verlag, Berlin1987Google Scholar). Redox potentiometry was consistent with the presence five heme centers, all of which have midpoint redox potentials below 0 mV. These are surprisingly low values for a protein that is thought to act as an electron acceptor from ubiquinol because the Eo′ for the ubiquinol/ubiquinone redox couple is +40 mV (19Matthews C.K. van Holde K.E. Biochemistry. 2nd Ed. Benjamin Cummins, Menlo Park, CA1995: 524Google Scholar). However, it was observed that water-soluble duroquinol could reduce DorC. The thermodynamic quantity described in this paper is the “macroscopic” potential of the heme centers of DorC (20Santos H. Moura J.J.G. Mours I. LeGall J. Xavier A.V. Eur. J. Biochem. 1984; 141: 283-296Crossref PubMed Scopus (150) Google Scholar,21Bertrand P. Asso M. Mbarki O. Camensuli P. More C. Guigliarelli B. Biochim. Biophys. Acta. 1994; 76: 524-536Google Scholar). These parameters are not influenced by electron/electron cooperativity between heme centers. However, it has been shown that in the tetraheme cytochrome c 3 fromDesulfovibrio vulgaris each heme redox potential is dependent on the oxidation state of the other three hemes (22Gayda J.-P. Benosman H. Bertrand P. More C. Asso M. Eur. J. Biochem. 1988; 177: 199-206Crossref PubMed Google Scholar), and this can affect the thermodynamic properties of a multi-heme protein (21Bertrand P. Asso M. Mbarki O. Camensuli P. More C. Guigliarelli B. Biochim. Biophys. Acta. 1994; 76: 524-536Google Scholar). This can lead to a variety of potential microscopic heme-heme interacting potentials that may be higher than the macroscopic potentials. The ability of DorC to be reduced by duroquinol can be rationalized if the heme with the highest potential (−34 mV) is reducible by duroquinol. Under physiological conditions this would suggest that the DorC protein would be reduced when the Ehof the ubiquinol/ubiquinone pool was lower than the Eo′ for the ubiquinol/ubiquinone redox couple. Such a situation could occur under phototrophic (anaerobic) conditions in the presence of reduced carbon sources (23Richardson D.J. King G.F. Kelly D.J. McEwan A.G. Jackson J.B. Ferguson S.J. Arch Microbiol. 1988; 150: 130-137Google Scholar). Further work to demonstrate and analyze heme-heme interactions in DorC using EPR and/or NMR spectroscopy is now required. There has been intense interest in the structure and mechanism of Me2SO reductase (4McAlpine A.S. Shaw A.L. McEwan A.G. Bailey S. J. Biol. Inorg. Chem. 1997; 2: 690-701Crossref Scopus (151) Google Scholar, 24Schindelin H. Kisker C. Hilton J.C. Rajagopalan K.V. Rees D.C. Science. 1996; 272: 1615-1621Crossref PubMed Scopus (435) Google Scholar, 25Schneider F. Lowe J. Huber R. Schindelin F. Kisker C. Knaeblein J. J. Mol. Biol. 1996; 263: 53-69Crossref PubMed Scopus (252) Google Scholar), and many questions remain regarding the route of electron transfer to the molybdenum atom. It seems almost certain that one of the molybdopterin guanine dinucleotide moieties of Moco will act as a conduit for electrons from DorC. The purification of DorC that we have described opens the way for experiments to look at electron transfer to Me2SO reductase using the physiological electron donor. This may help resolve some current controversies relating to the mechanism of Me2SO reductase by avoiding the need to use dithionite and viologens as electron donors. Although we showed that DorC was reduced by duroquinol and that reduced DorC was oxidized by Me2SO reductase, it was not possible to use DorC to catalyze steady state electron transfer between duroquinol and Me2SO reductase. The reason is that Me2SO reductase itself exhibited duroquinol-Me2SO oxidoreductase activity.3However, under physiological conditions DorC would be required for the oxidation of ubiquinol-10, which, unlike water-soluble duroquinol, is buried within the hydrophobic domain of the cytoplasmic membrane and thus is inaccessible to periplasmic Me2SO reductase. We thank Bruno Guigliarelli and David Richardson for helpful comments and discussion.
21
Cleavage of Structural Proteins during the Assembly of the Head of Bacteriophage T4
Ulrich K. Laemmli · Nature · 1970 · 251.6K citations
Measurement of protein using bicinchoninic acid
Pam Smith, Randall I. Krohn, Greg T. Hermanson et al. · Analytical Biochemistry · 1985 · 18.4K citations
Improved broad-host-range plasmids for DNA cloning in Gram-negative bacteria
N. T. Keen, Stanley Tamaki, Daisuke Kobayashi et al. · Gene · 1988 · 1.5K citations