|Home | About | Journals | Submit | Contact Us | Français|
Methanobactin (Mb), a 1217-Da copper chelator produced by the methanotroph Methylosinus trichosporium OB3b, is hypothesized to mediate copper acquisition from the environment, particularly from insoluble copper mineral sources. Although indirect evidence suggests that Mb provides copper for the regulation and activity of methane monooxygenase enzymes, experimental data for direct uptake of copper loaded Mb (Cu-Mb) are lacking. Uptake of intact Cu-Mb by M. trichosporium OB3b was demonstrated by isotopic and fluorescent labeling experiments. Confocal microscopy data indicate that Cu-Mb is localized in the cytoplasm. Both Cu-Mb and unchelated Cu are taken up by M. trichosporium OB3b, but by different mechanisms. Uptake of unchelated Cu is inhibited by spermine, suggesting a porin-dependent passive transport process. By contrast, uptake of Cu-Mb is inhibited by the uncoupling agents carbonyl cyanide m-chlorophenylhydrazone and methylamine, but not by spermine, consistent with an active transport process. Cu-Mb from M. trichosporium OB3b can also be internalized by other strains of methanotroph, but not by Escherichia coli, suggesting that Cu-Mb uptake is specific to methanotrophic bacteria. These findings are consistent with a key role for Cu-Mb in copper acquisition by methanotrophs and have important implications for further investigation of the copper uptake machinery.
Copper is central to the physiology and metabolism of methanotrophic bacteria, methane-oxidizing organisms with potential applications in bioremediation and greenhouse gas removal (1). Methanotrophs have a particularly high requirement for copper in part because a primary metabolic enzyme, particulate methane monooxygenase (pMMO),3 is copper-dependent (2). pMMO comprises ~20% of the total cellular protein (3) and catalyzes the first step in methanotroph metabolism, the oxidation of methane to methanol. Under conditions of copper starvation, some methanotrophs produce an alternative, soluble methane monooxygenase (sMMO) that utilizes a diiron active site (4). In these methanotroph strains, copper mediates a switch between MMOs, repressing transcription of the sMMO genes and enhancing pMMO expression and the formation of intracytoplasmic membranes (5, 6). Copper also regulates the expression of a number of other proteins (1). The molecular mechanisms of this copper switch have not been elucidated.
Methanotrophs are believed to meet their high requirement for copper by secretion of methanobactin (Mb), a 1217-Da molecule that chelates Cu(I) with high affinity (7–10) and has been investigated recently as a possible treatment for Wilson disease (11). The structure of Mb from Methylosinus trichosporium OB3b includes seven amino acid residues, two oxazolone rings, two neighboring thioamide groups, a 3-methylbutanoyl group, and a pyrollidinyl group (Fig. 1) (10, 12, 13). The peptidic nature of Mb is analogous to the structures of many iron siderophores (14–16), suggesting that it is a chalkophore (“sidero” is Greek for iron; “chalko” is Greek for copper) (13). Other methanotrophs, including Methylococcus capsulatus (Bath), Methylomicrobium album BG8, Methylocystis species strain M, and Methylocystis strain SB2, have been reported to produce Mb (17–20), but it remains unclear whether there are different structural classes of Mb. It is not known how Mb is biosynthesized, and both nonribosomal and ribosomal peptide synthesis pathways have been suggested (1, 12, 18, 21).
It has been proposed that Mb is secreted in its apo form (apo-Mb) to retrieve copper from the environment and then recognized and reinternalized in its copper-loaded form (Cu-Mb) (1, 8). In support of this functional model, addition of Cu-Mb can initiate the switch between sMMO and pMMO expression (10, 22). Notably, although both CuCl2 and Cu-Mb have this effect, addition of Cu-Mb can also promote the switch to pMMO when cells only have access to insoluble copper mineral sources (22). Mb-mediated release of copper from glass has been correlated with methane oxidation activity as well (23, 24). However, experimental evidence for the underlying tenet of this model, the uptake of Mb by methanotrophs, is lacking. To address this fundamental question, we have tracked copper and Mb uptake by isotopic and fluorescent labeling studies. These data provide direct evidence for Mb uptake by methanotrophs. Using inhibitors, we have also demonstrated that Mb transport is not mediated by porins, but by an active transport process. These findings provide a solid framework for further investigation of the Mb handling machinery.
Cultures of M. trichosporium OB3b were grown either in 250-ml Erlenmeyer flasks fitted with gas-tight rubber septa or in a 1.25-liter BioFlo 3000 bioreactor. Cells were grown at 30 °C with a constant agitation rate of 200–300 rpm. For the flasks, 50 ml of air was withdrawn and replaced with 50 ml of methane each day. For cells growing in the bioreactor, a 3:1 methane:air mixture was supplied constantly at a flow rate of ~2 liters/min. Cells were harvested in mid-exponential phase at an absorbance at 600 nm (A600) of 0.5–0.9 for flask growths and 2.0–4.0 for bioreactor growths. The components of the growth media for normal and copper-starved conditions were the same as described previously (25, 26).
M. trichosporium OB3b was grown under limiting copper concentrations as described previously (26). The spent medium was harvested by centrifugation at 9000 × g for 30 min to remove the cells. Purification of apo-Mb was performed as described previously (26). Cu-Mb was prepared by an overnight incubation with Cu(II), provided as 3 g of CuSO4 · 5H2O/liter of spent medium. Purification was performed using a DSC-18 HPLC column with methanol or acetonitrile as the elution solvent. Fractions containing Cu-Mb were lyophilized and stored at −80 °C until further use. Further purification of Cu-Mb was performed using a Bio-Sil SEC 125 size exclusion HPLC column. For isotopic labeling, 24 mg of 65CuO (Cambridge Isotopes) was dissolved in 1.2 ml of 0.5 n H2SO4 and heated at 80 °C for 1 h to yield 65CuSO4 at a final concentration of 0.25 m. The resulting 65CuSO4 solution was added to apo-Mb, incubated for 1 h at room temperature in the dark, and the 65Cu-Mb was then purified using a Sep-Pak Plus tC18 column (Waters).
MALDI-TOF MS was performed on Cu-Mb and 65Cu-Mb using a Bruker Autoflex III smart beam MALDI-TOF mass spectrometer in the negative ion mode (supplemental Fig. S1). One μl of 20 mm p-nitroaniline was added to 1 μl of Mb (~1 μm, concentration measured by the A280 using an extinction coefficient of 1.65 × 104 m−1 cm−1) and spotted on a MALDI target plate. A laser power of 30–50% was used for excitation of the sample with detection in the reflector mode.
Cu-Mb was incubated with a 10-fold molar excess of TCEP at room temperature for 1 h. Upon TCEP addition, the Cu-Mb color changed from brown to yellow. Reduced Cu-Mb was purified on a Sep-Pak Plus tC18 column using 60% methanol as the elution solvent. The eluted samples were incubated with a 5-fold molar excess of either monobromobimane (mBBr) or bodipy FL l-cysteine (Invitrogen) in 50 mm Tris, pH 8.0, overnight at room temperature. Excess label was removed by purification using a Sep-Pak Plus tC18 column, and thiol-specific labeling was confirmed by MALDI-TOF MS.
Uptake experiments were performed on cells grown to mid-exponential growth. In a typical experiment, 1 ml of cells was harvested at an A600 of 1.0 by centrifugation at 18,000 × g for 5 min, washed twice in growth medium containing only salts and phosphate buffer (salt solution contains 0.85 g/liter NaNO3, 0.17 g/liter K2SO4, 0.037 g/liter MgSO4 · 7H2O, and 0.01 g/liter CaCl2 · 2H2O in 1 liter of water, and phosphate buffer contains 48.06 g/liter Na2HPO4 · 7H2O and 23.4 g/liter KH2PO4 in 1 liter of water). No trace metals or additional copper and iron were added to this wash buffer. The twice-washed, pelleted cells were resuspended in 1 ml of the wash buffer. Cu-Mb was then added to a final concentration of 5 μm, and the mixture was incubated in the dark either in a culture tube with a rubber stopper or in a 1.5-ml Eppendorf tube. Incubation times depended on the experiment and ranged from 40 min to 16 h.
Inhibition of Cu and Cu-Mb uptake was performed by the addition of 3% methylamine (Sigma-Aldrich) or 0.2–0.5 mm carbonyl cyanide m-chlorophenylhydrazone (CCCP; Enzo Life Sciences) or 0.25–1 mm spermine (Alfa Aesar) to the cells. Following a 10–15 min incubation, cells were either analyzed directly or washed at least three times in wash buffer following the addition of 5 μm CuSO4 or 5 μm Cu-Mb after incubation for 45 min prior to analysis. The effective concentrations of methylamine and CCCP were determined using an ethidium bromide accumulation assay. Cells were harvested during mid-exponential growth, washed twice in 20 mm phosphate, pH 7.2, and resuspended in the same buffer to an A600 of ~5.0. Ethidium bromide was added to a final concentration of 100 μg/ml, and the cells were incubated for 60 min at room temperature. Published protocols (27–29) were adapted for use in a 96-well plate reader. To each well, 150 μl of 100 mm phosphate buffer, pH 7.2, and 500–1 mm inhibitors were added. The ethidium bromide-treated cells (20 ml) were then added. Fluorescence of this mixture was either monitored continuously (emission at 600 nm after excitation at 388 nm) for 8 h, or the starting and final spectra in the 420–700 nm range (excitation at 388 nm) were collected for 10-min incubations with inhibitors. Normalization was performed against fluorescence of ethidium bromide-treated cells with no added inhibitors.
After incubation with Cu-Mb or CuSO4, M. trichosporium OB3b cells were washed three times with wash buffer, and 100 μl of 60% trace metal grade nitric acid was added. The cells were then incubated in a heating water bath for 1 h at 80 °C. After 1 h, the Eppendorf (microcentrifuge) tubes were centrifuged, and metal-free water was used to dilute the sample such that the final concentration of the nitric acid was 5%. Metal content was measured using a Varian Vista-MPX CCD Simultaneous ICP-OES instrument and copper and iron atomic absorption standards (Sigma-Aldrich) prepared in 5% nitric acid. To determine the isotopic ratios of 63Cu to 65Cu, ICP-MS was performed on acid hydrolyzed samples using an X Series 2 ICP-MS (Thermo Fisher Scientific). Indium was added to a final concentration of 20 ppb as an internal mass standard. Because of the variability associated with the estimation of the metal content and cell loss during the wash procedures, the determined values for Cu content were normalized against intracellular Fe content.
Competition experiments in M. trichosporium OB3b cells were performed by incubating copper-starved cells at an A600 of 1.0 with 5 μm Cu-Mb and varying concentrations (0–10 μm) of apo-Mb. The mixtures were incubated for 45 min at room temperature, washed three times in wash buffer, and the copper and iron contents were measured as described above.
Copper-starved cells were washed three times in wash buffer and resuspended to a final A600 of 1.0. After incubation with 28.1–140.6 μm (final) mBBr-Mb or mBBr-Cu-Mb for 1 h at room temperature, cells were mounted on a poly-l-lysine-coated glass-bottom Petri dish with FM 4-64 at 5–15 μg/ml (Invitrogen) and 10% DABCO anti-fading solution (Sigma-Aldrich). Live cell imaging was performed using a Zeiss LSM510 META system equipped with a 100×/1.46 Plan-Apochromat oil-immersion TIRF objective. A 25-mW laser diode was used to excite mBBr-Mb at 405 nm, and a 1-mW HeNe laser was used to excite FM 4-64 at 543 nm. For experiments involving Mb labeled with bodipy FL l-cysteine, a 488-nm argon laser was used. Images were collected using Zeiss LSM 5 software, and further analysis was performed using Volocity (PerkinElmer Life Sciences). Images produced using the 405-nm laser were slightly offset along the x, y, and z axes compared with images produced using the 543-nm laser. For image merging and further data analysis, registration correction along the x and y axes was performed using Volocity. Conventional fluorescence microscopy was performed using a Zeiss fluorescence microscope equipped with a halogen lamp and band pass filters. A sample preparation procedure similar to that described above, with the exception of poly-l-lysine glass-bottom Petri dishes, was used for all experiments. A 100× oil immersion objective was used for visualization. Images were recorded and processed using the AxioVision software.
When M. trichosporium OB3b cells are presented with 5 μm copper as unchelated copper (Cu, supplied as CuSO4) or Cu-Mb and incubated for 45 min at room temperature, the intracellular copper content, compared with the normalized copper content of cells with no added Cu or Cu-Mb, increases by ~2-fold (Fig. 2A and supplemental Table S1). These data show that both unchelated Cu and Cu associated with Mb can be taken up by the cells. To track the added Cu, we performed uptake studies using 65Cu and 65Cu-Mb. Natural copper contains 69% 63Cu and 31% 65Cu (63Cu/65Cu ratio of 2.2), although fractionation of copper isotopes in biological systems is not well understood (30). The observed 63Cu/65Cu ratio of cells with no added Cu or Cu-Mb (control in Fig. 2B) is ~2.0. This deviation is likely due to biological and instrumental fractionation as well as the relatively poor analytical precision of ICP-MS (31). Replacing the natural isotopic copper mixture with 65Cu does not affect the ability of cells to take up Cu or Cu-Mb (Fig. 2A and supplemental Table S1), but the intracellular 63Cu/65Cu ratio decreases significantly to ~0.5 and ~0.7 for 65Cu and 65Cu-Mb (Fig. 2B). Given that the affinity of Mb for Cu(I) is (6–7) × 1020 m−1 (10), it is likely that when 65Cu is presented as 65Cu-Mb, the intact 65Cu-Mb molecule is internalized rather than 65Cu-dissociating and then entering the cell. Addition of either 65Cu and Cu-Mb or Cu and 65Cu-Mb in combination leads to an increase in intracellular copper content (Fig. 2A and supplemental Table S1) and a reduction in the intracellular 63Cu/65Cu ratio to ~1.0 (Fig. 2B). This value is in between those obtained for natural Cu and fully labeled 65Cu, indicating that natural Cu and 65Cu are taken up simultaneously, consistent with uptake of both Cu and Cu-Mb.
To determine whether Cu-Mb enters the cell as an intact complex or whether Cu dissociates from Cu-Mb before entering the cell, Cu-Mb was fluorescently labeled with the thiol-specific probe mBBr. This probe is only fluorescent when conjugated with thiols. Reduction of the Mb disulfide bond followed by overnight incubation with mBBr yielded a predominantly doubly labeled species (mBBr-Cu-Mb) with a molecular mass of 1599.23 Da (predicted mass, 1599.27 Da) (supplemental Fig. S2). Incubation of cells with mBBr-Cu-Mb leads to an increase in intracellular copper content comparable with what is observed with Cu-Mb, indicating that labeling does not interfere with Cu-Mb uptake (supplemental Fig. S3). Cells incubated with mBBr-Cu-Mb were examined using both confocal and conventional fluorescence microscopy. Confocal images clearly show that mBBr-Cu-Mb is inside the cell (Fig. 3A). To probe the localization of mBBr-Cu-Mb, the membrane-specific fluorescent dye FM 4-64 (32–34) was used. The merged images suggest that mBBr-Cu-Mb is localized primarily in the cytoplasm. The intrinsic fluorescence of Cu-Mb (35, 36) was also detectable by confocal microscopy (supplemental Fig. S4), confirming that copper uptake mediated by Mb occurs via import of the intact Cu-Mb complex. Notably, Cu-Mb labeled with bodipy FL l-cysteine appears to localize at the membrane (supplemental Fig. S4), suggesting that the larger size or chemical properties of this label may preclude uptake. Uptake of apo-Mb labeled with mBBr was also investigated, and fluorescence is detected in the cytoplasm (Fig. 3B). Although the copper-starved cells used for these experiments produce their own apo-Mb, the increase in signal intensity for cells incubated with apo mBBr-Mb suggests that the added Mb is taken up into the cell.
To assess the specificity of Mb uptake, uptake experiments were also performed using Escherichia coli and several other strains of methanotroph, including Methylocystis sp. strain M, M. capsulatus (Bath), and Methylomicrobium alcaliphilum 20Z (supplemental Fig. S5). For all three methanotroph strains, uptake was detected by an increase in intracellular copper content (normalized to the copper content of cells with no additions). It would not be surprising if Methylocystis sp. strain M could internalize Cu-Mb purified from M. trichosporium OB3b because it appears to produce a Mb of identical mass (20). The chemical composition of M. capsulatus (Bath) Mb is not known (17), and it is not clear whether M. alcaliphilum 20Z produces Mb. No uptake of Mb-associated copper was observed for E. coli, suggesting that uptake of Mb may be specific to methanotrophic bacteria.
We next investigated possible mechanisms of unchelated Cu and Cu-Mb uptake. In Gram-negative bacteria, transport of nutrients across the outer membrane can occur via passive diffusion or by active transport (37). Porins typically allow diffusion of smaller solutes, including cations and anions, and are a probable route for unchelated Cu uptake into M. trichosporium OB3b cells. However, the 1217-Da Cu-Mb molecule is likely too large for classical porins. By analogy to iron siderophores, cobalamin, and putative nickel complexes, we suggested previously (21) that Cu-Mb uptake might be mediated by outer membrane TonB-dependent transporters (TBDTs) (38–41). In TonB-dependent uptake systems, energy from the inner membrane proton motive force is transduced by interaction of the TBDTs with the inner membrane TonB-ExbB-ExbD protein complex.
To test the hypothesis that uptake of Mb is via specific outer membrane receptors, the possibility of apo-Mb competing with Cu-Mb uptake was investigated (Fig. 4 and supplemental Table S2). A 3-fold increase in intracellular copper upon addition of 5 μm Cu-Mb to copper-starved cells decreases in the presence of apo-Mb and is reduced to ~0.6 upon addition of 10 μm apo-Mb. Thus, apo-Mb and Cu-Mb may compete for uptake, which likely involves the same surface receptors. Similarly, competition for binding to the FpvA TBDT is observed for apo- and holo-pyoverdin (42–44).
To probe further the involvement of porins or TBDTs in copper uptake by M. trichosporium OB3b, we employed transport inhibitors. Polyamines such as spermine and putrescine are known to inhibit passive transport by porins (45, 46). Treatment of cells with 0.5 mm spermine inhibits Cu uptake but has no effect on Cu-Mb uptake (Fig. 5A and supplemental Table S3). Therefore, unchelated Cu likely diffuses into the cell via porins whereas Cu-Mb is transported by a different mechanism, consistent with uptake of the intact Cu-Mb molecule. Uncouplers such as CCCP, FCCP, dinitrophenol, and methylamine dissipate the inner membrane proton motive force, interfering with active transport. In particular, CCCP and FCCP have been used to block TBDT-mediated uptake of iron siderophores (42, 47, 48). Of seven uncoupling agents tested by ethidium bromide accumulation assays (27–29), CCCP and methylamine were found to inhibit active transport in M. trichosporium OB3b (supplemental Fig. S6). Treatment of cells with 3% methylamine (Fig. 6A and supplemental Table S4) or 0.5 mm CCCP (Fig. 7A and supplemental Table S5) inhibits Cu-Mb uptake, but not Cu uptake, indicating that Cu-Mb transport is an energy-dependent process.
Isotopic labeling experiments are also consistent with independent pathways for Cu and Cu-Mb uptake. The intracellular 63Cu/65Cu ratio is ~0.5 (Fig. 2B) upon incubation of cells with 65Cu and is ~2.0 in the presence of 65Cu and spermine, indicating that very little 65Cu uptake is occurring (Fig. 5B). By contrast, spermine does not affect the uptake of 65Cu-Mb (Fig. 5B). The converse is observed with methylamine (Fig. 6B) and CCCP (Fig. 7B). Addition of methylamine or CCCP prevents a decrease in the intracellular 63Cu/65Cu ratio upon addition of 65Cu-Mb but has no effect on 65Cu uptake. Isotopic ratios of less than ~1.0 are observed when mixtures of 65Cu and Cu-Mb or Cu and 65Cu-Mb are provided in the presence of each inhibitor (Figs. 5B, ,66B, and and77B), consistent with selective targeting of two different pathways.
The observations that Cu-Mb uptake is energy-dependent and that apo-Mb competes for uptake with Cu-Mb suggest that TBDTs could facilitate transport of Cu-Mb across the outer membrane, similar to uptake of iron siderophores (21). There is some precedent in the literature for TBDTs participating in copper uptake (41). The NosA protein from Pseudomonas stutzeri is an outer membrane protein necessary for formation of copper-loaded nitrous-oxide reductase (50, 51) which contains both a CuA site and an unusual tetranuclear copper cluster (52). NosA exhibits some homology to the siderophore TBDTs IutA, FepA, and FhuA (53). Copper binding by NosA has not been characterized and could involve a copper-chelate complex (53). Another relevant example is OprC from Pseudomonas aeruginosa. This protein is 65% identical to NosA and apparently binds Cu(II) (54). In the M. trichosporium OB3b draft genome (49), there are at least 45 unique TBDTs, of which some may be involved in the transport of siderophores, flavins, and other small molecules. A combination of bioinformatic and experimental approaches will be needed to determine whether any of these candidate TBDTs is involved in Mb uptake.
In summary, these data provide direct evidence for uptake of Cu-Mb by methanotrophic bacteria. Intact Cu-Mb complexes are imported into the cell rather than Cu dissociating from Mb prior to uptake. Both Cu and Cu-Mb are taken up, but by different mechanisms. Unchelated Cu uptake likely occurs through porins whereas Cu-Mb uptake is an active transport process that may involve TBDTs. The Cu-Mb from M. trichosporium OB3b can be internalized by other methanotrophs, but not E. coli, indicating that acquisition of copper via Mb may be specific to methanotrophs. These findings are consistent with a model in which Mb facilitates copper acquisition from environments where soluble copper sources are scarce. The identity and mechanism of the transport machinery represent a key direction for future research.
We thank the Keck Biophysics, Integrated Molecular Structure Education and Research Center (IMSERC), Quantitative Bioelemental Imaging Center (QBIC), and Cell Imaging Facility (CIF) facilities at Northwestern University for use of the instruments, and we thank the reviewers for experimental suggestions and the CIF staff for help with confocal microscopy.
*This work was supported by National Science Foundation Grant MCB0842366. Imaging work was performed at the Northwestern University Cell Imaging Facility supported by National Institutes of Health Grant CCSG P30 CA060553 through the NCI, awarded to the Robert H. Lurie Comprehensive Cancer Center.
3The abbreviations used are: