Overlapping role of the outer membrane cytochromes of Shewanella oneidensis MR-1 in the reduction of manganese(IV) oxide

Authors


Charles R. Myers, Department of Pharmacology and Toxicology, Medical College of Wisconsin, 8701 Watertown Plank Road, Milwaukee, WI 53226, USA (e-mail: cmyers@mcw.edu).

Abstract

Aim: To determine if the outer membrane (OM) cytochromes OmcA and OmcB of the metal-reducing bacterium Shewanella oneidensis MR-1 have distinct or overlapping roles in the reduction of insoluble manganese(IV) oxide.

Methods and Results: The gene replacement mutant (OMCA1) which lacks OmcA was partially deficient in Mn(IV) reduction. Complementation of OMCA1 with a vector (pVK21) that contains omcB but not omcA restored Mn(IV) reduction to levels that were even greater than those of wild-type. Examination of the OM of OMCA1/pVK21 revealed greater than wild-type levels of OmcB protein and specific haem content.

Conclusions: Overexpression of OmcB can compensate for the absence of OmcA in the reduction of insoluble Mn(IV) oxides. Therefore, there is at least a partial overlap in the roles of these OM cytochromes in the reduction of insoluble Mn(IV) oxide.

Significance: The overlapping roles of these two cytochromes has important implications for understanding the mechanism by which MR-1 reduces insoluble metal oxides. There is no obligatory sequential electron transfer from one cytochrome to the other. They could both potentially serve as terminal reductases for extracellular electron acceptors.

Introduction

Shewanella oneidensis (formerly putrefaciens) MR-1 uses a wide variety of terminal electron acceptors for anaerobic respiration including insoluble manganese (Mn) and iron (Fe) oxides (Myers and Nealson 1988, 1990b; Myers and Myers 1994). Many aquatic environments, sediments and soils are rich in Mn and Fe oxides, and metal-reducing bacteria such as MR-1 may contribute to the reductive solubilization of metal oxides in these environments (Myers and Nealson 1990a; Nealson et al. 1991). The ability of Shewanella species to reduce other metals including chromium(VI) (Myers et al. 2000), uranium(VI) (Lovley et al. 1991), and technetium(VII) (Lloyd and Macaskie 1996) could be useful in the bioremediation of metal-contaminated sites. Understanding the mechanisms by which these bacteria mediate metal reduction will contribute to a better understanding of their roles in the environment.

The respiratory versatility of MR-1 is reflected in its diverse array of electron transport components. It synthesizes a variety of quinones (Myers and Myers 1993, 2000) as well as many c-type cytochromes (Myers and Myers 1992, 1997a,b). MR-1 was the first bacterium shown to localize cytochromes to its outer membrane (OM) when grown under anaerobic conditions (Myers and Myers 1992). Two of these OM cytochromes, OmcA and OmcB, have been purified and gene replacement mutants indicate that they have a role in the reduction of insoluble Mn(IV) oxides (e.g. MnO2) (Myers and Myers 2001, 2002b). The OM localization of OmcA and OmcB places them where they could make direct contact with extracellular metal oxides at the cell surface, thereby serving as terminal Mn(IV) reductases. As electron transport components in the cytoplasmic membrane (CM) are also required for the use of insoluble electron acceptors including MnO2 (Myers and Myers 1993, 1997a, 2000), MR-1 represents an important model for anaerobic respiration because it includes components of both the CM and OM. It was the first example of a bacterium with electron transport components in direct contact with the extracellular environment.

While both OmcA and OmcB have roles in MnO2 reduction, it was not clear if their roles were overlapping or inter-dependent. The studies reported here provide evidence for overlapping roles.

Materials and methods

Shewanella oneidensis MR-1 (wild-type) (Myers and Nealson 1988), and the gene replacement mutant OMCA1 (Myers and Myers 2001) that is lacking the gene encoding the OM cytochrome OmcA have been previously described. Plasmid pVK100 (Knauf and Nester 1982), a 23 kb broad host range low copy number cosmid encoding kanamycin and tetracycline resistance, was obtained from the American Type Culture Collection (ATCC 37156). Plasmid pVK21 consists of a genomic fragment of MR-1 cloned into the HindIII site of pVK100 (Myers and Myers 2002b). The insert in pVK21 includes omcB, but not omcA.

Vernadite (δMnO2) (Myers and Nealson 1988) was prepared as described. Mn(II) was determined in filtrates by a formaldoxime method (Brewer and Spencer 1971; Armstrong et al. 1979). Protein was determined by a modified Lowry method (Collins and Hughes 1983) using bovine serum albumin as the standard.

Shewanella oneidensis was grown at room temperature (23–25°C). For the preparation of subcellular fractions, S. oneidensis was grown under anaerobic conditions as previously described (Myers and Myers 1992) in M1 defined medium (Myers and Nealson 1990b) supplemented with 15 mmol l−1 lactate, 24 mmol l−1 fumarate, vitamin-free Casamino acids (0·2 g l−1), and kanamycin (50 μg ml−1). For testing Mn(IV) reduction, the same medium was used except it was also supplemented with 15 mmol  l−1 formate plus 4·5 mmol l−1δMnO2. For Mn(IV) reduction experiments, inocula were prepared using cells grown aerobically on Luria-Bertani medium pH 7·4 (Sambrook et al. 1989) supplemented with kanamycin and tetracycline (20 μg ml−1); the cells were suspended in sterile distilled water and the inocula densities were adjusted to equalize turbidity (O.D. at 500 nm using a Beckman DU-64 spectrophotometer, Beckman Coulter, Fullerton, CA, USA). The Mn(IV) reduction experiments were conducted in medium that had been pre-equilibrated to anoxic conditions in the anaerobic chamber.

The CM, intermediate membrane (IM), OM, and soluble fractions (periplasm plus cytoplasm) were purified from cells using an EDTA-lysozyme-Brij protocol as previously described (Myers and Myers 1992) and utilized (Myers and Myers 1997b, 1998, 2001). The IM closely resembles the OM except for a buoyant density intermediate between that of the CM and OM (Myers and Myers 1997b). The separation and purity of these subcellular fractions were assessed by spectral cytochrome content (Myers and Myers 1992), membrane buoyant density (Myers and Myers 1992), and NADH oxidase activity (Osborn et al. 1972). SDS-PAGE gels (Myers and Myers 1998) were stained for protein with Pro-Blue (Owl Separation Systems, Woburn, MA, USA) or for haem (Thomas et al. 1976).

Polyclonal antibodies specific for OmcA and OmcB were those described in a previous study (Myers and Myers 2002b). The purification of IgG from rabbit sera, the removal of non-specific antibodies by absorption with Escherichia coli, and western blotting with chemiluminescent detection were carried out as previously described (Myers and Myers 1997b, 2002a).

Statistical analysis was performed using analysis of variance and Tukey's multiple comparison post-test (InStat software; GraphPad, San Diego, CA, USA).

Results

According to the genomic sequence for MR-1 (http://www/tigr/org), omcB lies in the same orientation and just downstream of omcA. However, these two genes are transcribed independently (Myers and Myers 1998, 2001, 2002b). The gene replacement mutant OMCA1, which was generated from MR-1, lacks functional omcA and does not contain omcA transcript or OmcA protein (Myers and Myers 2001). Wild-type levels of OmcB were properly localized to the OM of OMCA1 (Fig. 1). The vector pVK21 was introduced into OMCA1. The insert in pVK21 is an MR-1 genomic fragment that contains omcB, but not omcA. pVK21 has been previously used to restore OmcB to an omcB-minus mutant (Myers and Myers 2002b). When OMCA1/pVK21 was examined by western blotting, its OmcB levels were much greater than those in MR-1 or OMCA1 containing the empty vector pVK100 (Fig. 1). The increased levels of OmcB in OMCA1/pVK21 were properly localized to the OM and IM fractions with only minor amounts in the CM and soluble fractions (Fig. 1). As expected, OMCA1/pVK100 and OMCA1/pVK21 did not express OmcA (Fig. 1).

Figure 1.

Western blots using polyclonal IgGs specific for OmcA (a) or OmcB (b). Subcellular fractions were isolated from fumarate-grown cells of MR-1/pVK100 (lanes 1, 4, 7 and 10), OMCA1/pVK100 (lanes 2, 5, 8 and 11) and OMCA1/pVK21 (lanes 3, 6, 9 and 12). The lanes were loaded with 5 μg portions of the following subcellular fractions: CM (lanes 1–3), IM (lanes 4–6), OM (lanes 7–9) and soluble fraction (lanes 10–12). The positions and molecular masses of protein standards are indicated on the left

Visual examination of haem-stained SDS-PAGE gels (not shown) were consistent with Fig. 1, i.e. the band corresponding to OmcA was absent in the OMCA1 strains, and the OmcB band was more intense in OMCA1/pVK21 than in OMCA1/pVK100. Reduced-minus-oxidized cytochrome spectra were obtained to determine the relative total cytochrome contents of the subcellular fractions (Fig. 2). The cytochrome contents of the CM and soluble fractions were similar for all three strains, whereas marked differences were seen in the IM and OM (Fig. 2). The cytochrome contents of the IM and OM of OMCA1/pVK100 were 54 and 59%, respectively, of those of the IM and OM of MR-1 (Fig. 2). Because OMCA1 retains wild-type levels of OmcB (Fig. 1), this decline is due to the absence of OmcA in this mutant. The cytochrome contents of the IM and OM of OMCA1/pVK21 were significantly elevated; quantitatively, they were 1·9- and 2·2-fold of those of the corresponding wild-type fractions (Fig. 2). This increase in cytochrome content results from the overexpression of OmcB despite the continued absence of OmcA (Fig. 1). These spectral studies demonstrate that the additional OmcB in OMCA1/pVK21 has attached haem and should therefore be functional.

Figure 2.

Specific cytochrome contents of the subcellular fractions prepared from strains that were grown anaerobically with fumarate as the electron acceptor. The specific cytochrome content is the difference between the absorbance at the peak and the absorbance at the trough for the Soret region from reduced-minus-oxidized difference spectra per milligram of protein. The values are mean ± s.d. of results of independent experiments (two independent cultures of each strain)

The rate of Mn(IV) reduction by OMCA1/pVK100 cells was significantly depressed relative to that of wild-type (Fig. 3), in agreement with findings previously reported for OMCA1 (Myers and Myers 2001). In contrast, the rate of Mn(IV) reduction by OMCA1/pVK21 was significantly elevated relative to that of wild-type (Fig. 3). The relative differences between the strains were even more pronounced over a shorter timeframe (24 h) (data not shown). The excess OmcB in OMCA1/pVK21 was therefore able to more than compensate for the deficiency in OmcA.

Figure 3.

Reduction of δMnO2 by cells of various strains grown under anaerobic conditions as determined by the increases in Mn(II) over the first 48 h. All values are the mean ± s.d. for three independent cultures of each strain. †, statistically different from MR-1/pVK100 to a P < 0·001

Discussion

If OmcA and OmcB were sequential components in an electron transfer pathway with Mn(IV) as the terminal electron acceptor, then an excess of OmcB would not be able to compensate for an absence of OmcA. However, the data demonstrate that overproduction of OmcB can more than compensate for a lack of OmcA. The overproduction of OmcB in OMCA1/pVK21 did not affect its proper localization to the OM (Fig. 1) and the markedly increased haem content of the OM (Fig. 2) indicates that haem was properly incorporated into at least a significant portion of the excess OmcB.

mtrF lies upstream of omcA and encodes a third possible decahaem OM cytochrome. While definitive evidence for the presence of MtrF protein has not been provided, an mtrF knockout exhibited wild-type rates of Mn(IV) reduction (Myers and Myers 2002a). An mtrF-omcA double knockout resembled OMCA1 with respect to OM cytochrome content and Mn(IV) reduction (Myers and Myers 2002a). Therefore, a possible role for MtrF does not enter into the interpretation of the results of these studies.

Previous studies have indicated that both OmcA and OmcB are required for wild-type rates of Mn(IV) reduction (Myers and Myers 2001). Energy generation from the reduction of insoluble electron acceptors such as MnO2 will be limited by the kinetics associated with surface chemical reactions (Stone and Morgan 1990). Therefore, the optimal functioning of both OmcA and OmcB may be critical to successful competition in Mn(IV)-rich environments, especially given that MR-1 is an obligate respirer. The potential for an excess of OmcA to compensate for OmcB cannot be tested, however, because OmcB is required for the proper localization of OmcA to the OM (Myers and Myers 2001, 2002b). While the western blots (Fig. 1) indicate that MR-1 does not have a mechanism to upregulate the levels of OmcB in the absence of OmcA, our complementation studies demonstrate that an excess of OmcB can compensate for the absence of OmcA, consistent with an overlapping role for these cytochromes.

Acknowledgements

This work was supported by National Institute of Health grant R01GM50786 to C.R. Myers. We are also grateful for the support of the STRATTEC Foundation.

Ancillary