Previous Article | Next Article ![]()
Journal of Bacteriology, February 2007, p. 1036-1043, Vol. 189, No. 3
0021-9193/07/$08.00+0 doi:10.1128/JB.01249-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.
,
School of Biology, Georgia Institute of Technology, Atlanta, Georgia 30332
Received 8 August 2006/ Accepted 19 November 2006
|
|
|---|
-FeOOH, SO32, and S2O32]. Genetic complementation and nucleotide sequence analyses indicated that the CCMB1 respiratory mutant phenotype was due to mutation of a conserved histidine residue (H108Y) in a protein that displayed high homology to Escherichia coli CcmB, the permease subunit of an ABC transporter involved in cytochrome c maturation. Although CCMB1 retained the ability to grow on electron acceptors with high E'0, the cytochrome content of CCMB1 was <10% of that of the wild-type strain. Periplasmic extracts of CCMB1 contained slightly greater concentrations of the thiol functional group (-SH) than did the wild-type strain, an indication that the Eh of the CCMB1 periplasm was abnormally low. A ccmB deletion mutant was unable to respire anaerobically on any electron acceptor, yet retained aerobic respiratory capability. These results suggest that the mutation of a conserved histidine residue (H108) in CCMB1 alters the redox homeostasis of the periplasm during anaerobic growth on electron acceptors with low (but not high) E'0. This is the first report of the effects of Ccm deficiencies on bacterial respiration of electron acceptors whose E'0 nearly span the entire redox continuum. |
|
|---|
In several metal-respiring bacteria, c-type cytochromes are involved in electron transport to oxidized forms of metals and radionuclides. In Geobacter sulfurreducens, outer membrane c-type cytochromes OmcB and OmcF are required for respiration on solid Fe(III) oxides, and diheme c-type cytochrome MacA is an intermediate electron carrier (7, 29, 35). Metal-respiring members of the genus Shewanella also require outer membrane c-type cytochromes OmcA and OmcB (MtrC) for respiration on U(VI) and solid Fe(III) or Mn(IV) oxides (3, 39, 44). c-type cytochromes involved in periplasmic electron transport in S. oneidensis include MtrA and CymA (41, 50). CymA oxidizes quinol and transfers electrons to downstream components of the electron transport pathway terminating with the reduction of Fe(III), NO3, NO2, fumarate, and dimethyl sulfoxide (DMSO) (41, 53, 54). A purified cytochrome c3-hydrogenase complex isolated from U(VI)-reducing Desulfovibrio vulgaris Hildenborough displays H2-U(VI) oxidoreductase activity in vitro (38). In addition, cytochrome c3 mutants of Desulfovibrio desulfuricans G20 are deficient in U(VI) reduction activity with H2 as electron donor (47, 48).
Bacteria, archaea, and the mitochondria and chloroplasts of eukarya require maturation systems to complete c-type cytochrome synthesis. Cytochrome c maturation (Ccm) systems attach heme groups to the CXXCH motifs of apocytochrome c via stereospecific thioether covalent bonds. Three Ccm systems are currently known (for recent reviews, see references 57 and 59). System I, found predominantly in alpha- and gammaproteobacteria, land plant and protozoan mitochondria, and some archaea, consists of eight dedicated components (CcmA to -H). System II, commonly found in gram-positive bacteria, cyanobacteria, beta- and deltaproteobacteria, plant and algal chloroplasts, and some archaea, includes the major components ResA to -C and CcdA. The third cytochrome c maturation system, system III, is restricted to fungal and animal mitochondria and consists of a single heme lyase component (CCHL).
Cytochrome c maturation system I is the most extensively studied cytochrome c maturation pathway. System I is organized into two branches that converge at heme lyase (CcmF): a heme delivery branch comprised of CcmA to -E and a thioredoxin branch that includes DsbD and CcmGH (59). The heme delivery branch transfers heme to apocytochrome c and includes ABC transporter subunits CcmABC, membrane protein CcmD, and heme chaperone CcmE. The thioredoxin branch transfers reducing equivalents to the periplasm and ensures that the thiol groups of apocytochrome c remain reduced for heme attachment. Maintenance of redox homeostasis in the periplasm is crucial for optimal Ccm activity (2, 8).
A random point mutant of S. putrefaciens strain 200 (originally designated Urr14 and here renamed CCMB1) was isolated by combining chemical mutagenesis (ethyl methanesulfonate) procedures with a rapid mutant screen designed to detect respiratory mutants deficient in U(VI) reduction activity (63). Subsequent anaerobic liquid growth experiments demonstrated that CCMB1 was unable to respire U(VI) or NO2 as an anaerobic electron acceptor, yet retained the ability to respire on O2, NO3, MnO2, and Fe(III) citrate. In the present study, CCMB1 was found to contain a mutation in a gene displaying high similarity to ccmB, encoding a permease involved in system I cytochrome c maturation in Escherichia coli (60, 61). A comparison of wild-type and CCMB1 growth rates, cytochrome content, and periplasmic free thiol content indicated that wild-type ccmB was required for growth on electron acceptors with low (but not high) E'0 and for maintaining periplasmic redox homeostasis.
|
|
|---|
-FeOOH (55), 40 mM; MnO2 (as colloidal MnO2), 10 mM (49); Mn(III) pyrophosphate, 10 mM (31); TMAO (trimethylamine N-oxide), 30 mM; SO32, 10 mM; S2O32, 10 mM; DMSO, 50 mM; fumarate, 35 mM. Aerobically grown cells were inoculated in batch cultures vigorously aerated with atmospheric gas. U(VI) growth experiments were carried out in SM with the headspace gas consisting of 10% H2, 5% CO2, and the balance N2. Control experiments consisted of incubating wild-type S. putrefaciens in SM liquid growth medium with electron acceptor omitted or with heat-killed cells as inoculum. Antibiotics were added from filter-sterilized stocks at the indicated concentrations: chloramphenicol, 25 µg ml1, and tetracycline, 15 µg ml1. Genetic complementation and nucleotide sequence analyses. Standard genetic procedures were used in genetic complementation and subcloning experiments (52). A previously constructed clone library of partially digested HindIII chromosomal DNA fragments of the S. putrefaciens strain (harbored in broad-host-range cosmid pVK100 and maintained in Escherichia coli strain HB101) was mobilized into respiratory mutant CCMB1 following previously described triparental mating procedures that included helper strain E. coli HB101(pRK2013) (14). To identify CCMB1 transconjugates with restored U(VI) reduction activity, Rifr and Tetr transconjugate colonies were screened for a positive U(VI) reduction phenotype on SM agar growth medium supplemented with U(VI) carbonate (1 mM) (63). After screening of approximately 1,000 transconjugates, a CCMB1 transconjugate (designated CCMB1-D14) with restored U(VI) reduction ability was identified and subsequently confirmed for wild-type U(VI) reduction activity in liquid culture.
Transconjugate CCMB1-D14 was found to contain a 34-kb complementing cosmid (designated pD14; see Fig. S1 in the supplemental material for subcloning strategy). D14 contained two internal HindIII restriction sites resulting in three HindIII fragments (D14-1, D14-2, and D14-3), each of which was ligated into cosmid pVK100. The three resulting recombinant cosmids were electroporated into E. coli HB101, and the three corresponding transformants were mated triparentally with HB101(pRK2013) and CCMB1. Each CCMB1 transconjugate was tested for anaerobic growth in liquid culture with U(VI) as electron acceptor. CCMB1 transconjugates containing complementing subclone D14-2 were restored for U(VI) reduction activity. Complementing subclone D14-2 was sequenced (sixfold coverage) with an ABI 3700 automated sequencer. Initial sequencing primers complementary to the unique cloning site of cosmid pVK100 included pvkF (GAT CCT GGT ATC GGT CTG CGA TTC CGA CTC GTC) and pvkR (GTA CTC CTG ATG ATG CAT GGT TAC TCA CCA CTG CGA TCC). Overlapping internal regions of D14-2 (designated DC1 to -10) were PCR amplified and cloned into broad-host-range plasmid pBBR1MCS (32) for further genetic complementation analyses (Table 1 shows strains and plasmids; see Table S1 in the supplemental material for primers and restriction sites used for subcloning). The altered nucleotide in mutant CCMB1 was identified via PCR amplification of the CCMB1 chromosomal region that corresponded to complementing fragment DC9 of wild-type S. putrefaciens (primers were identical to those used to PCR amplify the DC9 region of wild-type S. putrefaciens). Nucleotide sequencing of the DC9 region of CCMB1 was carried out as described above.
|
View this table: [in a new window] |
TABLE 1. Bacterial strains and plasmids used in the present study
|
ccmB) was verified by PCR amplification with primers flanking the deletion (forward, CGCTTGTTATGATGAGTACCG, and reverse, CCTTGGTGGAGGCAGACTCAT) and DNA sequencing. Analytical techniques. Cell growth was monitored by simultaneously measuring cell number and electron acceptor depletion or end product production. Acridine orange-stained cells were counted directly via epifluorescence microscopy (Nikon Diaphot 300 microscope). Fe(III) reduction was monitored by measuring HCl-extractable Fe(II) with the ferrozine technique (58). MnO2 reduction was monitored by measuring MnO2 depletion with the benzidine blue colorimetric assay (6). Mn(III) pyrophosphate depletion was monitored spectrophotometrically at 480 nm (31). U(VI) was measured colorimetrically with Arsenazo III reagent (33). NO2 was measured spectrophotometrically with sulfanilic acid-N-1-naphthylethylene-diamine dihydrochloride reagent (40). Growth on O2, TMAO, DMSO, fumarate, S2O32, and SO32 was monitored by cell growth only. Free thiol equivalents (exposed thiol functional groups [-SH] free to react with Ellman's reagent) were determined with 5,5'-dithio-bis-2-nitrobenzoic acid (17) standardized with reduced glutathione per µg protein. Protein concentration was determined with the Coomassie Plus Bradford assay (Pierce Biotechnology) standardized with bovine serum albumin. Homologous and orthologous protein sequences were identified by BLAST analysis and aligned with ClustalW (1, 27). Membrane topology was predicted in silico with World Wide Web-interfaced software HMMTOP and TopPred2 (9).
Cytochrome detection. Wild-type and CCMB1 mutant cell cultures were grown on the array of electron acceptors in SM liquid medium supplemented with lactate (30 mM) as a carbon and energy source. Cells were harvested during late logarithmic growth phase and washed three times in anaerobic phosphate-buffered saline (130 mM NaCl, 50 mM sodium phosphate, pH 7.2, 4°C). Cell extracts were prepared as in previously described procedures (18). Briefly, cells were resuspended in cold extraction buffer (50 mM sodium phosphate, 2 mg/ml polymyxin B sulfate, 300 mM NaCl, 5 mM EDTA, pH 7.2, 4°C). The suspension was stirred gently for 1 h at 4°C and centrifuged (Beckman Coulter Optima L-100 XP Ultracentrifuge, 40,000 x g for 20 min at 4°C). Extracts from each sample were collected, and specific cytochrome content was determined by measuring dithionite-reduced-minus-ferricyanide-oxidized difference spectra on a Shimadzu UV-1601 spectrophotometer operated in the split-beam mode. Estimates of total cytochrome content were based on the difference between absorbance values at the peak and trough of the Soret peak standardized per mg total protein (10, 43). Protein fractions were separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis on a linear gradient gel (4% to 20% resolving) with 200 µg/ml total protein loaded per well (34). Gels were stained for heme peroxidase activity by being incubated in a solution containing 3,3-dimethoxybenzidine-peroxide (1 µg ml1) and H2O2 (30% stock solution, 0.0025 µl ml1) (24).
Nucleotide sequence accession number. The nucleotide sequence of D14-2 has been submitted to the GenBank database under accession number DQ682922.
|
|
|---|
![]() View larger version (121K): [in a new window] |
FIG. 1. U(VI) reduction screening plate from which complementing transconjugate CCMB1-pD14 (second colony from left in fourth row) was identified. Strains 200-pVK100 (colony in top row) and CCMB1-pVK100 (first colony from left in second row) were included as U(VI) reduction-positive and U(VI) reduction-negative control strains, respectively.
|
![]() View larger version (87K): [in a new window] |
FIG. 2. CcmB amino acid sequence of S. putrefaciens (Sputr) aligned with E. coli (Ecoli), orthologous Ccb206 of Arabidopsis thaliana (Athal), Orf206 of Triticum aestivum (Taest), Orf277 of Marchantia polymorpha (Mpoly), and YejV of Cyanidioschyzon merolae (Cmero). Identical residues are highlighted. H108 of S. putrefaciens and corresponding identical residues are boxed. Predicted transmembrane domains in S. putrefaciens are indicated by bars above the sequence.
|
|
View this table: [in a new window] |
TABLE 2. Sequence analysis of S. putrefaciens CcmABCDE gene cluster
|
Respiratory mutant CCMB1 retains the ability to respire on electron acceptors with high (but not low) reduction potentials (E'0). To determine the overall respiratory capability of CCMB1, anaerobic growth experiments were carried out on a set of 13 electron acceptors. CCMB1 retained the ability to grow at wild-type rates with O2, MnO2, Mn(III) pyrophosphate, NO3, and Fe(III) citrate, yet was severely impaired for growth on NO2, U(VI), DMSO, TMAO, fumarate, Fe(III) oxide, SO32, or S2O32 as an electron acceptor (Fig. 3; see also Fig. S2 in the supplemental material). CCMB1 transconjugates containing plasmid pDC9 in trans displayed wild-type rates of growth and reduction of each electron acceptor (see Fig. S2 in the supplemental material). CCMB1 was unable to grow at wild-type rates on electron acceptors with E'0 values below a threshold value of approximately 0.36 V (NO2/NH4+ couple) (Fig. 4; see also Fig. S2 in the supplemental material). CCMB1 grew at wild-type rates on NO3 yet was unable to grow on NO2. After stoichiometric reduction of NO3 to NO2, growth of the wild-type strain continued on NO2 until all NO2 was depleted (see Fig. S2D in the supplemental material). CCMB1, on the other hand, was able to grow at wild-type rates on NO3 and to stoichiometrically convert NO3 to NO2 at a rate similar to that of the wild-type strain, yet was unable to sustain growth or deplete NO2 after all NO3 was reduced. CCMB1 cell density and NO2 concentrations remained unchanged after NO3 was completely reduced to NO2.
![]() View larger version (14K): [in a new window] |
FIG. 3. Growth rate of CCMB1 on a set of 13 electron acceptors (rates normalized to the wild-type strain of S. putrefaciens). Wild-type growth rates are as indicated (h1): O2, 1.39; MnO2, 0.71; Mn(III) pyrophosphate, 0.32; NO3, 0.38; Fe(III) citrate, 0.21; NO2, 0.23; DMSO, 0.87; TMAO, 0.36; fumarate, 0.91; U(VI), 0.04; SO32, 0.18; -FeOOH, 0.25; S2O32, 0.14. See Fig. S1 in the supplemental material for individual growth curves. Error bars represent the standard deviations of three parallel yet independent incubations.
|
![]() View larger version (10K): [in a new window] |
FIG. 4. Reduction potential (E'0) of electron acceptors included in the present study. E'0 values were obtained from published values for the indicated couples at neutral pH. CCMB1 displayed wild-type growth rates on the set of electron acceptors that are boxed. DMS, dimethyl sulfide.
|
![]() View larger version (26K): [in a new window] |
FIG. 5. Analysis of cytochrome content of wild-type S. putrefaciens and respiratory mutant CCMB1. (A and B) SDS-polyacrylamide gel electrophoresis heme stains of periplasmic fractions of S. putrefaciens (A) and CCMB1 (B) cells grown on the indicated electron acceptors. Horse heart cytochrome c was included as a positive control. Molecular mass markers (daltons) are indicated with arrows. (C) Reduced-minus-oxidized difference spectra of cell extracts from S. putrefaciens (spectra 1 and 2) and CCMB1 (3 and 4) grown on Fe(III) citrate (1 and 3) and fumarate (2 and 4). The scale bar corresponds to absorbance units.
|
-FeOOH, or S2O32 as electron acceptor due to low growth rates (and correspondingly low cell yields) under these growth conditions (data not shown). Periplasmic extracts from the wild-type strain contained the highest thiol content after growth on
-FeOOH, SO32, or S2O32 as electron acceptor, ranging from 682 to 1,130 pmol µg protein1 (approximately 5 to 10 times greater than those of the other electron acceptors). Overall, the periplasmic extracts recovered from CCMB1 contained a slightly greater concentration of thiol groups than the wild-type strain did (Fig. 6).
![]() View larger version (17K): [in a new window] |
FIG. 6. Free thiol content of periplasmic fractions of S. putrefaciens and CCMB1 cells grown on the indicated electron acceptors. The box below the bar graph indicates CCMB1 growth ability on the electron acceptor (a). +, growth rate >80% of wild-type rate; , growth rate <25% of wild-type rate.
|
ccmB) was constructed and subsequently tested for growth on the entire suite of 13 electron acceptors respired by the wild-type strain. The
ccmB strain was unable to grow anaerobically on any electron acceptor, yet retained the ability to grow aerobically (see Fig. S3 in the supplemental material). Transconjugates of the
ccmB strain, containing a wild-type copy of ccmB in trans, were restored for the ability to respire at wild-type rates on all electron acceptors (see Fig. S3 in the supplemental material). |
|
|---|
S. putrefaciens respiratory mutant CCMB1 grows on electron acceptors with E'0 values greater than a threshold level of approximately 0.36 V [O2, Fe(III) citrate, MnO2, Mn(III) pyrophosphate, and NO3]. This finding is unexpected for several reasons. First, anaerobic respiration by S. oneidensis on Fe(III) citrate, MnO2, and NO3 as electron acceptor involves several c-type cytochromes including CymA, MtrA, MtrC, and OmcA (3, 41, 44, 50). Second, ccmC mutants of S. oneidensis are unable to produce mature cytochrome c or grow anaerobically on U(VI), fumarate, TMAO, DMSO, Fe(III) citrate, MnO2, NO3, or NO2 as electron acceptor (4, 39). Several possibilities may explain these differences in the anaerobic growth capability of the Shewanella ccm mutants: (i) S. putrefaciens ccmB and S. oneidensis ccmC mutant genotypes display different respiratory deficiencies, (ii) S. putrefaciens contains cytochrome c-independent respiratory pathways, or (iii) replacement of the histidine residue at position 108 with tyrosine permits limited, but sufficient, CcmB activity to sustain growth of CCMB1 on electron acceptors with high E'0. Results from the present study favor the third possibility. The
ccmB deletion mutant does not respire anaerobically on any electron acceptor, indicating that anaerobic electron transport pathways in S. putrefaciens require mature c-type cytochromes. Point mutant CCMB1, on the other hand, retains growth on anaerobic electron acceptors with an E'0 of >0.36 V, despite lacking the ability to produce c-type cytochromes at detectable levels. These findings suggest that the H108Y CcmB mutant may retain partial Ccm activity and produce low levels of mature cytochrome c that are adequate to sustain wild-type growth rates on electron acceptors above a threshold E'0.
Mutations in ccmA, ccmB, or ccmC in other bacteria often result in mutant phenotypes not associated with cytochrome c activity. Legionella pneumophila CcmB and CcmC mutants, for example, display defective iron acquisition capability and aberrant virulence phenotypes (8, 45, 62). Pseudomonas fluorescens CcmC mutants are unable to complete synthesis of the siderophore pyoverdine (2). CcmC mutants of Pantoea citrea, Gluconacetobacter diazotrophicus, and Pseudomonas putida are unable to oxidize gluconate and 2-ketogluconate, synthesize indole-3-acetic acid, or isomerize (cis-trans) unsaturated fatty acids, respectively (28, 36, 51). The seemingly unrelated Ccm mutant phenotypes are postulated to reflect improper redox homeostasis in the periplasm (2, 8).
To determine if CcmB plays a similar role in maintaining periplasmic redox homeostasis in S. putrefaciens, CCMB1 and wild-type S. putrefaciens were grown on the suite of 13 electron acceptors and free thiol content was determined in the corresponding periplasmic fractions. The wild-type strain contained approximately 100 pmol µg protein1 free thiol after growth on O2, NO3, Mn(III) pyrophosphate, and Mn(IV). Free thiol content was approximately two to three times greater after growth on NO2, TMAO, DMSO, Fe(III) citrate, and fumarate and approximately 10 times greater after growth on
-FeOOH, SO32, and S2O32. The pattern of free thiol content in the wild-type strain reflects the ability of CCMB1 to grow on electron acceptors with an E'0 of >0.36 V (NO3/NO2 threshold). Periplasmic redox conditions corresponding to a free thiol content of >300 pmol µg protein1 require wild-type CcmB activity to sustain growth. These results parallel those obtained with P. fluorescens ccmC mutants that are unable to complete pyoverdine biosynthesis (2). The periplasm of the P. fluorescens mutant strain was abnormally reducing and contained high free thiol content.
Increased free thiol in the periplasm of CCMB1 after growth on electron acceptors with low E'0 may result from several factors. First, the low redox poise associated with growth of CCMB1 on electron acceptors with low E'0 may result in inadvertent reduction of disulfide bonds that otherwise remain unaltered in the wild-type strain. In contrast, the redox poise associated with growth of CCMB1 on electron acceptors with high E'0 does not dramatically alter disulfide bond formation. Second, increased periplasmic free thiol content in CCMB1 may reflect an overabundance of periplasmic apocytochrome c (i.e., immature cytochrome c with CXXCH heme binding motifs not bound by heme). Finally, CcmABC may mediate periplasmic redox homeostasis via an as-yet-unknown mechanism (2, 8).
The present study is the first to identify an amino acid residue critical for CcmB function. Residue H108 is conserved in >88% of the 185 bacterial and land plant mitochondrial CcmB orthologs currently available in the nonredundant database. Residue H108 may therefore play a pivotal role in CcmB activity. Membrane protein topology prediction in S. putrefaciens indicates that residue H108 is found at the interface of a hydrophobic transmembrane helix and a hydrophilic cytoplasmic loop, well situated for controlling CcmB complex formation with other Ccm subunits or for allocrite recognition and subsequent transport across the cytoplasmic membrane. The identity of the CcmB allocrite in E. coli, however, remains controversial. Initial studies of E. coli indicate that heme is the CcmB allocrite and that CcmABC functions as a high-affinity heme transporter (21, 22). More recent results suggest that E. coli CcmABC drives the release of holo-CcmE (CcmE with covalently attached heme) from CcmC by coupling the release with ATP hydrolysis (22). Current work is aimed at identifying the CcmABC allocrite and determining the role that H108 plays in cytochrome c maturation in S. putrefaciens.
We thank Alana Reed, Amanda Payne, David Bates, and Justin Burns for laboratory assistance.
Published ahead of print on 1 December 2006. ![]()
Supplemental material for this article may be found at http://jb.asm.org/. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»