Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 525 (7567), 140-3

A Four-Helix Bundle Stores Copper for Methane Oxidation

Affiliations

A Four-Helix Bundle Stores Copper for Methane Oxidation

Nicolas Vita et al. Nature.

Abstract

Methane-oxidizing bacteria (methanotrophs) require large quantities of copper for the membrane-bound (particulate) methane monooxygenase. Certain methanotrophs are also able to switch to using the iron-containing soluble methane monooxygenase to catalyse methane oxidation, with this switchover regulated by copper. Methane monooxygenases are nature's primary biological mechanism for suppressing atmospheric levels of methane, a potent greenhouse gas. Furthermore, methanotrophs and methane monooxygenases have enormous potential in bioremediation and for biotransformations producing bulk and fine chemicals, and in bioenergy, particularly considering increased methane availability from renewable sources and hydraulic fracturing of shale rock. Here we discover and characterize a novel copper storage protein (Csp1) from the methanotroph Methylosinus trichosporium OB3b that is exported from the cytosol, and stores copper for particulate methane monooxygenase. Csp1 is a tetramer of four-helix bundles with each monomer binding up to 13 Cu(I) ions in a previously unseen manner via mainly Cys residues that point into the core of the bundle. Csp1 is the first example of a protein that stores a metal within an established protein-folding motif. This work provides a detailed insight into how methanotrophs accumulate copper for the oxidation of methane. Understanding this process is essential if the wide-ranging biotechnological applications of methanotrophs are to be realized. Cytosolic homologues of Csp1 are present in diverse bacteria, thus challenging the dogma that such organisms do not use copper in this location.

Figures

<b>Extended Data Figure 1</b>
Extended Data Figure 1. Purification of proteins from M. trichosporium OB3b and the amino-acid sequence of Csp1
a, The copper content of anion-exchange fractions (NaCl gradient shown as a dashed line) and the SDS-PAGE analysis of selected fractions (1 mL) from the purification of soluble extract from M. trichosporium OB3b cells. The band just below the 14.4 kDa marker, indicated with an arrow, is present. Fraction 32 was judged to have the lowest level of contaminating proteins and was further purified by gel-filtration chromatography on a Superdex 75 column (b). Csp1 is present in the peak that elutes at ~ 11 mL and contains considerable copper (see Fig. 1c). c, The amino-acid sequence of Csp1 showing the predicted Tat leader peptide (the first 24 residues of the pre-protein) in italics. The 13 Cys residues are highlighted in yellow and His36 (cyan), Met40, Met43 and Met48 (magenta) are also indicated (the numbering of these residues refers to the mature protein). The CXXXC and CXXC motifs are underlined. The region in bold corresponds to the single tryptic fragment identified on two separate occasions in MS analysis, representing 11% sequence coverage of the mature protein (Mascot search of peptide mass fingerprint, expect value = 1.9 × 10−5). The sequence of this fragment was confirmed by liquid chromatography/MS/MS (data not shown). This is the only tryptic peptide from the mature protein that would be anticipated to be readily detected by MS (due to either small mass or presence of Cys residues in all other theoretical tryptic fragments) and is unique to this protein amongst all proteobacterial protein sequences in the NCBInr database.
<b>Extended Data Figure 2</b>
Extended Data Figure 2. Cu(I) binding to Csp1
a, UV-Vis difference spectra upon the addition of Cu(I) to apo-Csp1 (5.32 μM) showing the appearance of S(Cys)→Cu(I) LMCT bands,,. b, Plots of absorbance at 250 (filled squares), 275 (filled circles), and 310 (open circles) nm against [Cu(I)]/[Csp1] ratio taken from the spectra in (a). The absorbance rises steeply until ~ 11-15 Cu(I) equivalents but continues to rise, particularly at lower wavelengths, making binding stoichiometry difficult to determine precisely with this approach. Systems that bind multiple Cu(I) ions in clusters such as those found in metallothioneins, typically give rise to luminescence at around 600 nm,. However, limited luminescence is observed at 600 nm during the titration of Cu(I) into Csp1 (data not shown). c, X-ray absorption near edge spectrum of a fresh crystal of Cu(I)-Csp1 at 100 K. d, Plots of [Cu(BCS)2]3− formation against time after the addition of Cu(I)-Csp1 (0.93 μM) loaded with 11.8 equivalents of Cu(I) to 2510 μM BCS either in the absence (dashed line) or presence (solid line) of 7.9 M Urea. Cu(I) is removed faster in urea and is limited by the rate of Cu(I)-Csp1 unfolding (Extended Data Fig. 5i). The presence of urea has little effect on the end point for this reaction. a, b and d were all performed in 20 mM Hepes pH 7.5 containing 200 mM NaCl.
<b>Extended Data Figure 3</b>
Extended Data Figure 3. Sequence comparison of Csp1 homologues in M. trichosporium OB3b
The M. trichosporium OB3b genome possesses two genes that code for Csp1 homologues; Csp2 and Csp3 having 58 and 19% sequence identity to Csp1, respectively. The predicted Tat leader peptides of Csp1 (MERRDFVTAFGALAAAAAASSAFA) and Csp2 (MERRQFVAAIGAAAAAASASRAFA) are omitted. The Cys residues (13 in Csp1 and Csp2 and 18 in Csp3) are highlighted in yellow with CXXXC and CXXC motifs underlined. A CXXXC motif in an α-helix allows both of the Cys residues to coordinate the same Cu(I) ion (Fig. 3d, e), which is not the case for a CXXC motif. This is consistent with the observation that a synthetic peptide containing a CXXC motif binds a Cu4S4 cluster via a 4-helix bundle made from four peptides, with coordination involving only one Cys per peptide,. The alignment was produced using the T-coffee alignment tool. The * symbol indicates fully conserved sequence positions, whilst the: and. symbols indicate strongly and weakly similar sequence positions respectively.
<b>Extended Data Figure 4</b>
Extended Data Figure 4. sMMO activity of wild type M. trichosporium OB3b and the Δcsp1/csp2 strain
Purple colour, indicating sMMO activity, is evident at 19.25 h in the Δcsp1/csp2 strain (tubes 4 – 6), but not until 24.5 h in the wild type (WT, tubes 1 – 3) when using a qualitative assay. When quantified spectrophotometrically at 27.75 h, the average sMMO activity in the Δcsp1/csp2 strain (grey) is 1.8 fold greater (p = 0.04, one-tailed t-test) than that of the WT (white), as shown in the bar chart (mean ± s.d. of three replicates).
<b>Extended Data Figure 5</b>
Extended Data Figure 5. The dependence on pH of competition between Csp1 and BCA for Cu(I), and far-UV CD spectra showing pH stability and unfolding of Csp1 in urea
a, Plots of [Cu(BCA)2]3− concentration against [Cu(I)]/[Csp1] ratio for the addition of Cu(I) to apo-Csp1 (2.38-2.56 μM) in the presence of 103 μM BCA in 20 mM buffer (see Methods) at pH 5.5 (filled squares), 6.5 (filled circles), 7.5 (filled triangles), 8.5 (open circles) and 9.5 (open squares) plus 200 mM NaCl. Equilibration is fast (< 20 min) at pH 6.5 and higher and the data shown are from titrations of Cu(I) into apo-Csp1. At pH 5.5 equilibration is slower and the data are for mixtures incubated for 21 h. Also shown are results for mixtures of Cu(I) with apo-Csp1 (3.31-3.67 μM) at pH 6.5 (b) and 9.5 (c) in the presence of 120 (filled squares), 300 (open circles), 450 (stars), 600 (filled triangles), and 900 (open squares) μM BCA, all after 15 h incubation. At lower BCA concentrations Csp1 is able to effectively compete for Cu(I) in the pH 6.5 to 9.5 range giving Cu(I) binding stoichiometries of 12-14 (see also Fig. 3a and Fig. 4a). At pH 5.5 Csp1 competes less effectively with BCA for Cu(I) most likely due to the protonation of Cys ligands. This is consistent with greater competition by 600 μM BCA at pH 6.5 (b) compared to pH 7.5 (Fig. 4a). The stability of apo-Csp1 over the pH and time range used for experiments with BCA (and BCS) was determined using far-UV CD spectroscopy. The spectra of apo-Csp1 (solid lines) at pH (d) 5.5 (34.1 μM, 0.43 mg/mL), (e) 7.5 (36.5 μM, 0.46 mg/mL) and (f) 9.5 (32.6 μM, 0.41 mg/mL) are compared with those for samples incubated for 43 h (dashed lines), and also for 3 (dotted line) and 17 (dashed/dotted line) h at pH 9.5. At pH 9.5 and in the presence of higher BCA concentrations (c), Csp1 binds approximately one less equivalent of Cu(I) that must be due to changes in structure that are observed at this pH value (no change after 3 h but there is a decrease of ~ 15-20% α-helical content at longer incubation times, see f). However, the remaining sites bind Cu(I) more tightly (c) than at pH 7.5 (Fig. 4a) due to deprotonation of the Cys ligands. g, Far-UV CD spectra of apo-Csp1 (19.9 μM, 0.25 mg/mL) in 20 mM Hepes pH 7.5 containing 200 mM NaCl at 0, 30, 60, 120, and 240 min (solid lines) after the addition of urea (7 M) compared to the spectrum for apo-Csp1 in the same buffer but with no urea (dashed line). h, Far-UV CD spectra of apo-Csp1 (7.94 μM, 0.10 mg/mL) as in (g) except that spectra were acquired at 0, 15, 30, 45 and 60 min (solid lines) after addition of urea (7 M); unfolding is significantly faster at lower protein concentrations and is consistent with the reaction with DTNB in urea being complete in 20 min at Csp1 concentrations < 4 μM. i, Far-UV CD spectra of Csp1 incubated with 14.0 equivalents of Cu(I) (19.9 μM, 0.25 mg/mL) as in (g) but at 0, 60, 240, 360, and 480 min and 24 h (solid lines) after addition of urea (7 M) compared to the spectrum for Cu(I)-Csp1 in buffer with no urea (dashed line). The arrow in g to i indicates how the spectrum changes with time.
<b>Extended Data Figure 6</b>
Extended Data Figure 6. Competition for Cu(I) between Csp1 and chromophoric ligands and the determination of the apparent average Cu(I) dissociation constant for Csp1 using BCS
a, Plots of [Cu(L)2]3− concentration against [L] (BCA or BCS) after the incubation of Cu(I)-Csp1 (2.59 μM) loaded with 10.4 equivalents of Cu(I) with different concentrations of BCA (filled circles) and BCS (filled squares) for 17 h. b, Plots of [Cu(BCS)2]3− concentration against [Cu(I)] for apo-Csp1 (2.71 μM) in the presence of 99.2 (open squares) and 248 (filled squares) μM BCS incubated with increasing concentrations of Cu(I) (0, 4.38, 11.0, 15.3 and 21.9 equivalents; data shown after 17 h incubation). BCS competes much more effectively with Csp1 for Cu(I) than BCA and [Cu(BCS)2]3− is stoichiometrically formed at 248 μM BCS. c, Plot of [Cu(BCS)2]3− concentration against the [Cu(I)]/[Csp1] ratio for mixtures of Cu(I) plus apo-Csp1 (2.70 μM) in the presence of 101 μM BCS (open squares) for 19 h. For comparison the data from (b) (2.71 μM Csp1 in the presence of 99.2 μM BCS for 17 h) are also shown (filled squares). The data in ac were all acquired in 20 mM Hepes plus 200 mM NaCl at pH 7.5. d, Fractional occupancy of Cu(I)-binding sites in Csp1 (maximum value is 11.3 equivalents in this experiment) at different concentrations of free Cu(I) for the experiment shown in (c). The solid line shows the fit of the data to the non-linear Hill equation giving KCu= (1.3 ± 0.1) × 10−17 M (n = 2.7 ± 0.2). Hill coefficients larger than 1 indicate positive cooperativity for Cu(I) binding by Csp1. Confirmation, and the cause, of this effect will be the subject of further studies.
<b>Extended Data Figure 7</b>
Extended Data Figure 7. Cu(I) exchange between Csp1 and mbtin
UV-Vis spectra of apo-mbtin (dashed lines) and at various times up to 360 min (thick lines) after the addition of either Cu(I)-Csp1 or Cu(I). Cu(I)-Csp1 (1.02 μM) loaded with 13.0 equivalents of Cu(I) was added to either 13.4 (a) or 27.4 (c) μM apo-mbtin. Cu(I) alone (13.3 μM) was added to 13.4 (b) or 27.1 (d) μM apo-mbtin. Plots of absorbance at 394 nm against time for (a) to (d) are shown in Fig. 4d. Mbtin from M. trichosporium OB3b has a Cu(I) affinity of (6-7) × 1020 M−1 at pH 7.5 (determined using a logβ2 value of 19.8 for [Cu(BCS)2]3−, but is an order of magnitude tighter if the more recent logβ2 value of 20.8 (25) is used) and stoichiometrically removes Cu(I) from Csp1 within 1 h. e, UV-Vis spectra of Cu(I)-mbtin (2.71 μM, black line) immediately after mixing with apo-Csp1 (234 μM, green line) and after incubation under anaerobic conditions for 1 (blue line) and 20 (red line) h. Small increases in absorbance are observed due to the absorbance of apo-Csp1 at these wavelengths and precipitation. The latter was more of a problem at longer incubation times and the sample at 20 h required filtering prior to running the spectrum shown. The small changes observed are not consistent with the formation of apo-mbtin. All experiments were performed in 20 mM Hepes pH 7.5 plus 200 mM NaCl.
<b>Extended Data Figure 8</b>
Extended Data Figure 8. Sequence comparison of Csp homologues from diverse bacteria
Homology searches show that Csp homologues are encoded in the genomes of diverse bacteria. Multiple sequence alignment of the three M. trichosporium OB3b proteins (OB3b Csp1, OB3b Csp2 and OB3b Csp3) with a selection of these proteins, including one member (from Neisseria gonorrhoeae) that also possesses a putative Tat signal sequence (underlined), shows that the Cys residues (highlighted in yellow) are highly conserved. The alignment was produced using the T-coffee alignment tool. The * symbol indicates fully conserved sequence positions, whilst the : and . symbols indicate strongly and weakly similar sequence positions respectively. N. gonorrhoeae sequence: ORF NGAG_01502, UniProt accession C1I025; Pseudomonas aeruginosa sequence: ORF PA96_2930, UniProt accession X5E748 (PDB ID 3KAW); Streptomyces coelicolor sequence: ORF SCO3281, UniProt accession Q9X8F4; Nitrosospira multiformis sequence: ORF NmuI_A1745, UniProt accession Q2Y879 (PDB ID 3LMF); Rhizobium leguminosarum sequence: ORF RLEG_20420, UniProt accession W0IHZ3; Ralstonia metallidurans sequence: ORF Rmet_5753, UniProt accession Q1LB64; Salmonella enterica sv. Typhimurium sequence: ORF STM14_1521, UniProt accession D0ZVJ6; Bacillus subtilis sequence: ORF BSU10600, UniProt accession O07571; Legionella pneumophila sequence: ORF LPE509_p00081, UniProt accession M4SK87.
Figure 1
Figure 1. Identification and purification of Csp1 from M. trichosporium OB3b
a, Copper content of anion-exchange fractions (NaCl gradient shown as a dashed line) of extract from M. trichosporium OB3b cells and the sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) analysis of fractions 20 to 33. b, Copper content and SDS-PAGE analysis of the purification of the fraction containing the highest copper concentration (fraction 28) from (a) on a G100 gel-filtration column. A similar anion-exchange fraction (Extended Data Fig. 1a) was purified on a Superdex 75 column (Extended Data Fig. 1b), with the copper content and SDS-PAGE analyses of eluted fractions shown in (c). The band of interest that migrates below the 14.4 kDa marker is indicated in each panel with an arrow, and protein identification was performed on the bands from the 7.0 (b) and 12.0 (c) mL fractions.
Figure 2
Figure 2. The structure of apo-Csp1
a, Analytical gel-filtration chromatograms of apo-Csp1 (red line) and protein to which 14.0 molar equivalents of Cu(I) were added (blue line) for samples (100 μM when injected) in 20 mM 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid (Hepes) pH 7.5 containing 200 mM NaCl. The absorbance was monitored at 280 nm with the values for Cu(I)-Csp1 divided by 10 (see Extended Data Fig. 2a, b). The inset shows SDS-PAGE analysis of the purified protein. b, Far-UV CD spectra of apo-Csp1 (red line) and Csp1 plus 14.0 equivalents of Cu(I) (blue line) at 39.6 and 35.7 μM respectively in 100 mM phosphate pH 8.0. c, The tetrameric arrangement in the asymmetric unit of the crystal structure of apo-Csp1, with the side-chains of the Cys residues that point into the core of the 4-helix bundle shown as sticks for one monomer in (d). The opening into the core of the 4-helix bundle is facing out in (d), and involves His36, Met40, Met43 (on the extended α1) and Met48.
Figure 3
Figure 3. Cu(I)-binding by Csp1
a, Plot of [Cu(BCA)2]3− concentration against the [Cu(I)]/[Csp1] ratio upon titrating Cu(I) into apo-Csp1 (2.43 μM) in the presence of 103 μM BCA (same buffer as for Fig. 2a). [Cu(BCA)2]3− starts forming after ~ 12 equivalents of Cu(I) are added. b, Analytical gel-filtration chromatogram of Csp1 (116 μM) mixed with ~ 25 equivalents of Cu(I) in the same buffer. Csp1 (Bradford, red squares), copper (atomic absorption spectroscopy, blue triangles) and Cu(I) (bathocuproine disulfonate (BCS) in the presence of 7.6 M urea, open cyan circles) concentrations are shown. The main Csp1-containing fractions bind 11.8 to 12.9 equivalents of Cu(I). c, The structure of Cu(I)-Csp1 (chain A) including the anomalous difference density for copper contoured at 3.5 σ (orange mesh). The copper ions (Cu1 to Cu13 correspond to A1123 to A1135 in the PDB file 5AJF) are represented as dark grey spheres and the side chains of Cys, and other key residues as sticks. The coordination of Cu(I) ions at the two ends of the 4-helix bundle are shown in (d) and (e) with bond distances (Å) in red.
Figure 4
Figure 4. Cu(I) affinity of Csp1 and Cu(I) release
a, Plots of [Cu(BCA)2]3− concentration against the [Cu(I)]/[Csp1] ratio for mixtures of apo-Csp1 (3.57 μM) and Cu(I) in the presence of 120 (filled squares), 300 (open circles), 600 (open triangles), 900 (cross) and 1200 (plus sign) μM BCA (all data acquired after 41 h incubation). b, Plot of [Cu(BCA)2]3− concentration against the [Cu(I)]/[Csp1] ratio for mixtures of apo-Csp1 (3.61 μM) and Cu(I) in the presence of 1210 μM BCA (open squares) for 20 h, along with the data from (a) at 1200 μM BCA (filled squares). c, Fractional occupancy of Cu(I)-binding sites in Csp1 (maximum value 11.7 equivalents) at different concentrations of free Cu(I) from the data shown in (b) at 1210 μM BCA up to a [Cu(I)]/[Csp1] ratio of 19.2. The solid line shows the fit of the data to the non-linear Hill equation giving an average dissociation constant for Cu(I), KCu, of (7.5 ± 0.1) × 10−18 M (n = 3.1 ± 0.2, see Extended Data Fig. 6c, d). d, Plots of the absorbance at 394 nm (spectra in Extended Data Fig. 7a-d) against time after the addition of Cu(I)-Csp1 (1.02 μM) loaded with 13.0 equivalents of Cu(I)) to 13.4 (filled squares) and 27.4 (filled circles) μM apo-mbtin and Cu(I) (13.3 μM) to 13.4 (open squares) and 27.1 (open circles) μM apo-mbtin. All experiments were performed in the same buffer as for Fig. 2a.

Comment in

Similar articles

See all similar articles

Cited by 24 PubMed Central articles

See all "Cited by" articles

References

    1. Hanson RS, Hanson TE. Methanotrophic bacteria. Microbiol. Rev. 1996;60:439–471. - PMC - PubMed
    1. Balasubramanian R, et al. Oxidation of methane by a biological dicopper centre. Nature. 2010;465:115–119. - PMC - PubMed
    1. Murrell JC, McDonald IR, Gilbert B. Regulation of expression of methane monooxygenases by copper ions. Trends Microbiol. 2000;8:221–225. - PubMed
    1. Hakemian AS, Rosenzweig AC. The biochemistry of methane oxidation. Annu. Rev. Biochem. 2007;76:223–241. - PubMed
    1. Jiang H, et al. Methanotrophs: multifunctional bacteria with promising applications in environmental bioengineering. Biochem. Eng. J. 2010;49:277–288.

Publication types

LinkOut - more resources

Feedback