Introduction

Methanogens are a polyphyletic group of Archaea that can reduce CO2 or simple organic compounds to methane and couple this metabolic transformation to energy conservation1,2. Based on how a chemiosmotic gradient is established, methanogens can be divided into two groups: (a) methanogens without cytochromes and (b) methanogens with cytochromes3,4. Methanogens without cytochromes rely on the penultimate step in methanogenesis, mediated by a membrane-bound enzyme called N5-methyl H4MPT: CoM Methyltransferase (Mtr), to generate a Na+ gradient. Mtr catalyzes the exergonic transfer of a methyl group from the C1 carrier tetrahydromethanopterin (H4MPT) to coenzyme M (CoM) and concomitantly pumps out two Na+ ions that can be coupled to ATP generation using a membrane-bound ATP synthase (Supplementary Fig. 1)3,4. In contrast, methanogens with cytochromes use an electron transport chain (ETC) to generate an ion gradient for energy conservation. The terminal reductase complex in the ETC of these methanogens is a membrane-bound heterodisulfide reductase that uses the heterodisulfide of coenzyme M (CoM-SH) and coenzyme B (CoB-SH) i.e., CoM-S-S-CoB, generated during the last step of methanogenesis, as the electron acceptor (Supplementary Fig. 2)3,4. Taken together, regardless of the underlying mechanism, all methanogens are obligately dependent on methanogenesis to generate an ion motive force for energy conservation and lack any alternate strategies for ATP generation.

Over the years, several studies allude to the possibility that methanogens might be able to decouple ATP generation from methanogenesis and conserve energy by iron respiration using either soluble Fe3+ or Fe(III) containing minerals like ferrihydrite as electron acceptors5,6,7,8,9,10,11,12,13,14,15,16,17. If certain methanogens are indeed capable of switching between methanogenic and non-methanogenic modes of energy conservation, it would dramatically alter our perceived view of their role in the global biogeochemical cycles of macro- and micronutrients. Hence, it is critical to assess iron respiration in methanogens rigorously by identifying specific proteins involved in the process and investigating their function in vivo and in vitro.

There is substantial disagreement in the literature about the taxonomic breadth of methanogens that can perform iron respiration and the resulting physiological and ecological consequences. While initial reports demonstrated iron respiration by methanogens with and without cytochromes5,18,19, many recent studies show that this trait is limited to methanogens with cytochromes i.e. members of the Order Methanosarcinales11,12. That said, some strains within the Methanosarcinales, like Methanolobus vulcani, cannot reduce soluble Fe3+ 5, even in the presence of a humic acid analog - anthraquinone-2,6-disulfonate or AQDS- that typically stimulates iron reduction. Altogether, the distribution of iron respiration across methanogens is strongly but not perfectly correlated with the presence of an ETC. The impact of iron respiration on methanogenesis and cell growth also warrants careful consideration. Typically, the addition of Fe3+ and/or AQDS to actively growing cultures of methanogens inhibits methanogenesis and then Fe2+ and/or AQDSH2 start to build up, which suggests that these two processes cannot occur simultaneously likely due to some molecular incompatibility8,11,18,20. In contrast, some studies12,17,21 report that Fe3+ supplementation enhances methanogenesis, implying that the two processes occur via mutually exclusive pathways. Altogether, despite conflicting evidence, it is clear that certain methanogens are metabolically active under iron-reducing conditions, however the underlying mechanisms warrant further investigation.

In recent studies, the genetically tractable methanogenic archaeon, Methanosarcina acetivorans, has been developed as a model system to probe the molecular details of energy conservation coupled to iron respiration. Biochemical assays with purified membranes or everted membrane vesicles show that a membrane-associated multiheme c-type cytochrome (MHC), likely MmcA, is involved in the electron relay from reduced ferredoxin to a membrane-bound electron carrier methanophenazine (MP) during methanogenesis and to Fe3+ during iron reduction12,22,23. Since membrane preps contains many other proteins as well as other c-type cytochromes, a direct interaction between MmcA and MP or Fe3+ cannot be concluded from these studies12,22,23. Other studies show that a mutant lacking mmcA is incapable of transferring electrons to AQDS, which further corroborates its involvement in iron reduction11. However, the phenotype of this mutant during methanogenic growth on acetate has been shown to vary. The ∆mmcA mutant generated by our group24,25 cannot grow on acetate even after a whole year of incubation whereas another group11 has reported that this mutant has no growth phenotype during methanogenesis. While multiple lines of investigation have converged on the importance of MmcA in the ETC during methanogenesis as well as iron respiration, the exact role that this protein plays under either condition remains unclear.

In this study, we use a combination of in vitro and in vivo techniques to delineate two distinct biochemical roles for MmcA that depend on the environmental context. First, we devise a strategy to enrich MmcA from M. acetivorans and provide strong evidence that it is, as predicted, a membrane-associated heptaheme MHC. Next, we demonstrate that this protein can interact with and transfer electrons to MP as well as AQDS and soluble Fe3+ species. The redox properties and evolutionary origins of MmcA further corroborate our proposed functions for this protein. Overall, MmcA is a versatile electron carrier in the ETC of methanogens and can facilitate energy conservation by methanogenesis or iron respiration depending on the availability of oxidized iron species in the environment.

Results

MmcA from M. acetivorans is a heptaheme c-type cytochrome

MmcA (locus tag: MA0658 or MA_RS03460) is the first gene of the Rnf (Rhodobacter nitrogen fixation) operon (MA0658-MA0665 or MA_RS03460-MA_RS03495) and contains five canonical (CXXCH) and two non-canonical (CXXXCH and CXXXXCH) heme-binding motifs. The holo-protein is predicted to be a 476 aa-long heptaheme c-type cytochrome (cyt c) after the signal sequence is processed (Supplementary Fig. 3). MmcA is an essential component of the Rnf complex in M. acetivorans22,24,25,26,27 and is primarily found in the membrane fraction (Supplementary Fig. 4). Since MmcA lacks an identifiable transmembrane domain, we anticipate that it is tethered in the membrane through interactions with membrane integral proteins of the Rnf complex27,28. To purify MmcA, we generated an M. acetivorans strain with an expression vector that encodes a C-terminally tagged (3 × FLAG tag and a twin-Strep tag) mmcA placed under the control of tetracycline-inducible promoter24. Our previous work shows that the tag does not interfere with maturation and can also rescue the growth defect of the ΔmmcA mutant on methylated compounds24,25. Since the protein contains a twin-Strep tag, we attempted to purify the protein using a Streptactin resin but were unsuccessful (Supplementary Fig. 5)29,30. We then attempted to purify the protein using an anti-Flag resin and were able to obtain ~250 µg of protein per liter of culture (Supplementary Fig. 6). We performed a peroxidase-based heme stain and an immunoblot with anti-Flag antibody to confirm that the protein fraction is primarily MmcA and contains covalently bound heme groups i.e. is a holo-form of MmcA (Fig. 1A). The MmcA enriched fraction has a distinct red color which is a hallmark of cyt c (Fig. 1A) and also has spectral features typical of cyt c with an absorption maximum at 408 nm (γ) and 530 nm in the oxidized state (in black) and 419.5 nm (γ), 523 nm (β), and 552 nm (α) in the reduced state (in red) (Fig. 1B). We also confirmed the presence of covalently attached heme groups in MmcA by performing a pyridine hemochrome assay and observed a characteristic 550-nm α peak (Fig. 1B, inset and Supplementary Fig. 7). Taken together, these data show that MmcA from M. acetivorans is a heme-attached cyt c protein.

Fig. 1: Purification and UV-vis characterization of MmcA.
figure 1

A Coomassie, heme staining, and Western blot (WB) with anti (α)-Flag antibody of C-terminal 3 × FLAG tagged MmcA enriched from Methanosarcina acetivorans. MmcA under aerobic conditions and concentrated to ca. 100 µM has a reddish-brown color. B UV-vis spectral analysis of MmcA in the oxidized state (black) and reduced with sodium dithionite (red). The inset shows the pyridine hemochrome assay of reduced MmcA with a characteristic alpha peak for c-type cytochromes at 550 nm. (Complete UV-vis spectra of the hemochrome assay is shown in Supplementary Fig. 7). Data shown in (A) and (B) are representative of three experiments (n = 3). Source data are provided as a source data file.

Full size image

Although MmcA is predicted to be a heptaheme cyt c, the heme occupancy of this protein has not been experimentally validated. To this end, we performed LC-MS/MS analysis of chymotrypsin-digested fragments of MmcA and observed that the seven peptides with the putative heme-binding motifs had an increase in mass of 615.17 Da corresponding to heme attachment (Table 1)31. This observation confirms that MmcA is a heptaheme cyt c, and also demonstrates that the cyt c maturation (CCM) machinery of M. acetivorans can covalently attach heme to non-canonical heme-binding motifs in MmcA (Supplementary Fig. 3)24. We did not detect the predicted signal peptide in either trypsin or chymotrypsin-digested fragments of MmcA, which suggests that it gets cleaved after membrane translocation (Supplementary Fig. 8).

Table 1 Chymotrypsin-digested MmcA peptides with heme-binding motifs
Full size table

MmcA is a methanophenazine reductase

Using purified membranes from M. acetivorans, it has been shown that MmcA is involved in transferring electrons from ferredoxin to the membrane-bound electron carrier MP via the Rnf complex22. However, whether MmcA interacts with MP directly or indirectly via another protein remains unknown. We conducted spectroscopic analyses with MmcA to measure its ability to donate electrons to MP (Fig. 2A). We added an excess of 2-hydroxyphenazine (2HP) (200 µM), a well-established soluble analog of MP22,32,33,34,35, to MmcA reduced with sodium dithionite under anoxic conditions (Fig. 2B). The addition of 2HP instantaneously oxidized MmcA as observed by a shift in its Soret peak from 419.5 nm to 408 nm and the disappearance of the α and β peaks of reduced MmcA (Fig. 2B; Supplementary Fig. 9). Since the reaction was instantaneous, we were unable to calculate kinetic parameters. Regardless, these data provide strong evidence that MmcA can interact with and donate electrons to MP and establishes a clear role for this protein in the ETC during methanogenesis.

Fig. 2: MmcA can function as a methanophenazine (MP) reductase.
figure 2

A Schematic of the MmcA-Rnf complex (adapted from Gupta et al. 24) where the multiheme cytochrome MmcA mediates the final step in the transfer of electrons from the complex to the membrane-bound electron carrier methanophenazine (MP). B Spectral analysis of MmcA-mediated reduction of the soluble MP analog, 2-hydroxyphenazine (2HP) under anaerobic conditions. Addition of 2HP to MmcA reduced with sodium dithionite (SD; red) leads to the oxidation of MmcA (olive green) as observed by the appearance of a characteristic Soret peak at 408 nm indicative of the oxidized protein (black). Data shown are representative of two experiments (n = 2). Source data are provided as a source data file.

Full size image

MmcA can donate electrons to extracellular electron acceptors like ferric iron and humic acid analogs

Previous studies have shown that a mutant of M. acetivorans lacking mmcA fails to reduce AQDS11, but these genetic data do not imply that MmcA is directly involved in the reduction of AQDS and if it can also reduce other redox-active molecules (Fig. 3A). We tested if MmcA can reduce AQDS or Fe3+ [either ferric chloride (FeCl3) or ferricyanide (K3[Fe(CN)6])] in vitro by monitoring the redox state of MmcA after adding an excess of each of these electron acceptors (Fig. 3). All three electron acceptors oxidized MmcA rapidly as confirmed by the change in the spectral profile of the protein (Fig. 3; Supplementary Figs. 10 and 11). In contrast, a control experiment with Zn2+ as the sole electron acceptor did not change the redox state of dithionite-reduced MmcA (Supplementary Figs. 12 and 13). While we were unable to measure the reaction kinetics using our experimental setup, these data confirm the interaction between MmcA and AQDS or Fe3+ and validate its important role during anaerobic respiration in M. acetivorans.

Fig. 3: MmcA can donate electrons to extracellular electron acceptors.
figure 3

A Schematic of the MmcA-Rnf complex (adapted from Gupta et al. 24) where the multiheme cytochrome MmcA mediates the final step in the transfer of electrons from the complex to extracellular electron acceptors. B–D Spectral analysis of MmcA-mediated reduction of anthraquinone-2,6-disulfonate (AQDS) (B), ferric chloride (C), and ferricyanide (D) under anaerobic conditions. Addition of AQDS (B), ferric chloride (C), and ferricyanide (D) to MmcA reduced with sodium dithionite (SD; red) leads to the oxidation of MmcA (yellow in B; green in C; blue in D) as observed by the appearance of a characteristic Soret peak at 408 nm indicative of the oxidized protein (black). Data shown are representative of two experiments (n = 2). Source data are provided as a source data file.

Full size image

MmcA enhances fitness under iron-reduction conditions in M. acetivorans

To test the role of MmcA in vivo, we measured the rate of Fe3+ reduction with methanol as the electron donor in live cell suspensions of M. acetivorans (Fig. 4A–C). Consistent with our expectations, the absence of mmcA slows down Fe3+ reduction rates by ~30% in vivo (Fig. 4B, C). To test for polar effects of deleting mmcA on the chromosome, we measured the rate of iron reduction in the ∆mmcA mutant complemented with mmcA on a plasmid. Complementing mmcA in trans restored iron reduction rates to wild-type levels (Fig. 4B, C). These data suggest that MmcA improves fitness of M. acetivorans in Fe3+-containing environments.

Fig. 4: MmcA facilitates iron reduction in M. acetivorans.
figure 4

A Coomassie, heme stain, and Western blot (WB) with anti (α)-FLAG antibody of cell lysates obtained from a ΔmmcA mutant complemented with a plasmid encoding a C-terminal 3 × FLAG tagged mmcA placed under the control of a tetracycline-inducible promoter (ΔmmcA::pmmcA). The controls are either the ΔmmcA mutant or the parent strain (WWM60; wild type or WT) with an empty vector (pDPG010; labeled here as pEmpty) as described before24. Wild-type MmcA and C-terminal 3 × FLAG tagged MmcA are depicted by open and filled arrows respectively. B The rate of Fe3+ reduction in 1 mL anaerobic cell suspensions with similar optical density (see inset, the bar diagram represents the mean with standard deviation of three technical replicates, n = 3) was determined by an increase in Fe2+ using the ferrozine assay. 100 µg/mL of tetracycline was used to induce expression. Error bars are means ± standard deviation of three technical replicates (n = 3). Asterisks represent p-values of ΔmmcA compared to the complemented strain and are only shown when it is also statistically significant for ΔmmcA compared to WT (p-values for ΔmmcA compared to WT & ΔmmcA compared to complemented strain at time points 30, 60, 90, and 120 min are 0.0489 & 0.0176, 0.0294 & 0.0046, 0.0019 & 0.0099, and 0.0225 & 0.0088, respectively). C Rate of Fe3+ reduction by the ΔmmcA and ∆mmcA::pmmcA complementation strain were calculated as the percentage activity of the parent strain (WWM60) denoted as wild type (WT). The data shown are average ± standard deviation from three independent experiments (n = 3) done with three technical replicates (n = 3) each. *p-value ≤ 0.05, **p-value ≤ 0.01, ***p-value ≤ 0.001 and ns is p-value ≥ 0.05 respectively using a two-sided Student’s t-test. The p-values are 0.00079, 0.019, and 0.124 for iron reduction rates compared to ΔmmcA vs WT, ΔmmcA vs complemented strain, and complemented strain vs WT, respectively. Source data are provided as a source data file.

Full size image

MmcA is reversibly redox active between −100 and −450 mV versus SHE

We explored the redox behavior of MmcA using protein film voltammetry (PFV) on the meso-porous indium tin oxide electrode (ITO). MmcA formed a stable film on the ITO surface and was capable of exchanging electrons directly with the electrode, giving rise to reversible redox signals spanning from −100 to −450 mV (Fig. 5A; Supplementary Figs. 14 and 15). These signals, once deconvoluted and fit to Nernstian one-electron peaks, could be separated into seven reversible signals corresponding to the reduction and oxidation of the seven heme cofactors within MmcA, with uncertainties of approximately 10 mV. We determined the midpoint potentials of each heme by taking the average of the reduction and oxidation peak potentials for each redox pair (Fig. 5B). The low potential redox range of MmcA is similar to functionally analogous MHCs like MtrA and the MtrCAB complex (0 to −400 mV and 0 to −450 mV, respectively) in Shewanella oneidensis MR-136,37 and OmcZ or OmcS (−60 to −420 mV and −40 to −360 mV, respectively) in Geobacter sulfurreducens38,39. The redox-active range of MmcA suggests that it is capable of transferring electrons to Fe3+ (+300 to +400 mV), AQDS (−185 mV), and MP (−165 mV) (Fig. 5B).

Fig. 5: MmcA is redox active at a potential between −100 and −450 mV versus SHE.
figure 5

A Background-subtracted non-turnover MmcA voltammogram (black solid) with fitting of seven reversible redox couples (red dotted). Cyclic voltammetry (CV) was performed at pH 7.4, 10 °C, and a scan rate of 20 mV/s. B Redox range of MmcA from Methanosarcina acetivorans and octaheme tetrathionate reductase (OTR) from Shewanella oneidensis (See Supplementary Fig. 17). The average midpoint potentials (Em) of the seven heme centers in MmcA (blue squares) and eight heme centers in OTR (black triangles) are calculated from three experimental replicates (n = 3) under identical conditions. The redox range of other multiheme c-type cytochromes like the metal reductase (Mtr) from Shewanella oneidensis36,37 and the outer membrane cytochromes (Omc) from Geobacter sulfurreducens38,39 as well as the midpoint potential of methanophenazine (MP) and anthraquinone-2,6-disulfonate (AQDS) are shown for reference.

Full size image

MmcA is related to the tetrathionate reductase (OTR) family of multiheme c-type cytochromes

The mmcA locus is only present in methanogens and anaerobic methane-oxidizing archaea (ANME) within the Order Methanosarcinales and has ca. 25% amino acid sequence similarity to putative octaheme tetrathionate reductases (OTR) in other Archaea within the Orders Methanosarcinales, Desulfurococcales, and Archaeoglobales. A rooted tree of MmcA and OTR sequences shows that the MmcA clade is distinct but closely related to OTRs from Archaea as well as Bacteria (Fig. 6A). These observations support an independent origin for MmcA from an ancestor of MmcA and OTR rather than MmcA being derived from OTRs within Archaea. This hypothesis is further supported by synteny analyses of mmcA and otr in methanogens that encode both loci. While mmcA is always found in an operon with other genes of the Rnf complex, otr is present at a distant locus, often near a thermosome subunit and a biotin transporter (Fig. 6B).

Fig. 6: MmcA comprises a distinct clade of multiheme cytochromes related to the octaheme tetrathionate reductases (OTR).
figure 6

A Maximum-likelihood phylogenetic tree of MmcA and representatives of the OTR family. MmcA clades are colored by members of the Genus Methanosarcina (green), other members of the Family Methanosarcinaceae (purple), unspecified MAGs from the Order Methanosarcinales (pink), and anaerobic methane-oxidizing archaea (ANME) (red). OTR clades are colored by members of the Methanosarcinaceae (orange), members of the Desulfurococcales and Archaeoglobales (yellow), and representatives from Bacteria (blue). A pentaheme nitrite reductase (NrfA), octaheme nitrite reductase (ONR and IhOCC), octaheme hydrazine dehydrogenase (HDH) and octaheme hydroxylamine oxidoreductase (HAO) derived from bacteria were used as the outgroup to root the tree. Bootstrap values of 80 and above are shown. B Chromosomal organization of genes surrounding the mmcA locus (red) and the otr locus (orange) in a few representative strains within the Family Methanosarcinaceae. Genes of the same color represent members of the same orthologous group. C Structural alignment of AlphaFold predicted model of signal-less MmcA (MA0658; 25–500 aa, cyan) to crystal structure of OTR from Shewenella oneidensis (PDB: 1SP3, pink). Heme-binding motifs of OTR (orange) and MmcA (red) are shown. The histidine (His) ligand of the seven bis-His coordinated heme groups (1, 3–8) of OTR (light green) and seven heme groups (1–7) of MmcA (dark green) are shown. The second heme-binding motif of OTR that is absent in MmcA is highlighted (pink circle). Arrangement of heme groups in the OTR crystal structure (D) and in the MmcA model (E).

Full size image

Unlike MmcA, OTR is a respiratory enzyme of cryptic function, best characterized as a reductase of tetrathionate, nitrite, and hydroxylamine in bacteria like Shewanella oneidensis40,41. All OTRs, including those derived from members of the Methanosarcinales, contain eight heme-binding motifs and structural studies show that the second heme group, involved in catalysis, is ligated by a highly conserved Lys instead of His40. Neither this heme-binding motif nor the Lys are conserved in the MmcA sequence or predicted structure (Fig. 6C, D and Supplementary Fig. 16). All seven heme-binding motifs in MmcA are within electron transfer range for the heme groups with a total head-to-head distance of approximately ~55 Å between the first to seventh heme groups (Fig. 6E). Finally, the range of the redox potentials of the heme groups in MmcA (Fig. 5) is similar to that of OTR derived from S. oneidensis (Fig. 5B, Supplementary Fig. 17). Altogether, these data suggest that the MmcA clade of MHCs is distinct in form and function but closely related to the well-characterized OTR family of MHCs.

Discussion

In principle, if methanogens with cytochromes could reconfigure their ETC to also conserve energy using extracellular electron acceptors like iron, they would survive (if not proliferate) in far more diverse ecological niches (Supplementary Table 1). In practice, this phenomenon has been demonstrated repeatedly using environmental samples as well as axenic cultures, but biochemical details of the molecular mechanism(s) involved are yet to emerge. In the absence of any extracellular electron acceptors, M. acetivorans performs methanogenesis to generate a heterodisulfide, CoM-S-S-CoB, which serves as the terminal electron acceptor for the ETC3. Under these circumstances, MmcA transfers electrons from the Rnf complex to the membrane-bound electron carrier MP (Supplementary Fig. 2; Fig. 2), which can ultimately be used to reduce the terminal electron acceptor CoM-S-S-CoB. MmcA is universally conserved in methanogens with an Rnf complex and absent in Bacteria that use the Rnf complex to transfer electron between the ferredoxin and NAD pools, further corroborating its unique role as an electron conduit between the Rnf complex and MP28. When M. acetivorans encounters soluble Fe3+ or AQDS, our data suggest that MmcA can directly interact with and reduce these electron acceptors in vivo and in vitro (Figs. 3 and 4). Here, it is worth clarifying that even though we and others12,23 have shown that methanogens like M. acetivorans are metabolically active and can conserve energy by iron respiration (Fig. 4), robust growth that spans multiple generations is yet to be demonstrated i.e. it is still not known whether methanogens can couple iron reduction to growth in addition to energy conservation. Regardless, redox transformation of iron species by methanogens has substantial biogeochemical ramifications in and of itself to merit further investigation.

Methanogenic archaea like M. acetivorans typically lack a cell wall and their cell envelope comprises of an inner membrane and a crystalline proteinaceous surface layer (S-layer)42. While MmcA is membrane-associated (Supplementary Fig. 4), it lacks any S-layer domains, suggesting that it is present in the pseudo-periplasm rather than the outer-surface of cells28. The hexagonal crystal packing of the S-layer forms three different pores of which one, the primary pore, is large enough to allow the passage of chelated metal ions and AQDS into the pseudo-periplasm43. As a result, it is entirely feasible that MmcA can interact with many soluble extracellular electron acceptors despite its pseudo-periplasmic localization. While some recent studies suggest that MmcA contributes substantially to electron uptake from metallic iron44, more work needs to be done in the future to determine the involvement of MmcA in direct electron transfer to and from minerals and electrodes.

Even though MmcA plays an important role in Fe3+ reduction, our in vivo data clearly suggest that M. acetivorans encodes additional routes for extracellular electron transport (EET) (Fig. 4). Redundancy in EET pathways seems to be a common phenomenon in other well-studied microorganisms too. Deleting any one of five MHCs (OmcS, OmcZ, OmcB, OmcT, and OmcE) has little to no effect on AQDS reduction in Geobacter sulfurrreduces, and only a quintuple mutant lacking all five genes is unable to reduce AQDS and humic acids45. M. acetivorans encodes at least four other cyt c and the expression of one of these genes (MA3739) increases by 80% in the ∆mmcA mutant (Supplementary Table 2), suggesting that other cyt c might be able to functionally complement MmcA in its absence. Alternately, it is entirely plausible that other pathway(s) for iron respiration are MHC-independent as members of the Methanosarcinales lacking cyt c have also been shown to reduce iron or participate in direct interspecies electron transfer (DIET)5,8,9,10,46. Even though it is not technically feasible at present, high-throughput Fe3+ reduction assays with an unbiased transposon mutant library of the ∆mmcA strain might help identify additional pathways for EET in the future.

In conclusion, we show that MmcA is an important constituent of the respiratory chain of methanogens like M. acetivorans. Notably, based on the availability of electron acceptors like Fe3+ or AQDS, MmcA might also facilitate energy conservation by anaerobic respiration in addition to methanogenesis, thus broadening the ecological niche of these pivotal organisms.

Methods

Growth medium

M. acetivorans strains (Supplementary Table 3) were grown in single-cell morphology47 at 37 °C without shaking in bicarbonate-buffered high-salt (HS) liquid medium containing either methanol or trimethylamine (TMA) as the carbon and energy substrate with N2/CO2 (80/20) in the headspace. Puromycin (RPI, Mount Prospect, IL) was added to a final concentration of 2 µg/mL from a sterile, anaerobic stock solution to select for M. acetivorans strains with the mmcA-expression plasmid encoding a puromycin-resistance gene (pac). Anaerobic, sterile stocks of tetracycline hydrochloride in deionized water were prepared fresh before use and added to a final concentration of 100 µg/mL to induce the expression of MmcA from the tetracycline-inducible promoter as described previously in ref. 48. Cell cultures with a volume of up to 10 mL were grown in Balch tubes and larger volume cultures were grown in anaerobic bottles.

MmcA purification

MmcA was enriched by affinity purification from 4 L of late-exponential phase culture of DDN039 grown in HS media with 100 mM TMA at 37 °C. Cells were harvested by centrifugation (6000 × g) for 20 min at 4 °C, the supernatant was discarded, and the cell pellets were stored at −80 °C. All steps of protein purification were performed under aerobic conditions. 2 U/mL DNase-I (to reduce the viscosity of the suspension) and 1 mM Phenylmethylsulfonyl fluoride (PMSF) (to inhibit protease activity) was added to 20 mL of hypotonic lysis buffer (50 mM Tris-HCl, pH = 7.4) used to resuspend the cell pellet. The cell suspension was kept on ice for 45 min with intermittent mixing using a pipette to lyse the cells. Upon complete lysis, sodium chloride was added from a 5 M stock solution to a final concentration of 150 mM to the cell lysate. The lysate was clarified by centrifugation at 10,000 × g for 20 min at 4 °C and the supernatant was separated into the soluble and membrane fractions by high-speed ultracentrifugation at 100,000 × g for 1 h at 4 °C. The membrane pellets were solubilized in 4 mL TBS buffer (50 mM Tris-HCl, 150 mM NaCl, pH = 7.4) with 2% Triton X-100 (Sigma–Aldrich, St Louis, MO, USA). The solubilized membrane fraction was loaded on a column containing 1 mL anti-DYKDDDDK (Flag) G1 affinity resin (50% suspension; GenScript, Piscataway, NJ, USA) pre-equilibrated with 3 bed volumes of TBS buffer. Five washes with 2 mL of TBS buffer were performed before the protein was eluted using competitive elution buffer (300 μg/mL Flag peptide in TBS buffer). To elute the protein, three times the bed volume (i.e., 1.5 mL) of elution buffer was added to the column and one volume (500 µL) of elute was collected right away. The column was capped and incubated at room temperature for 30 min before collecting the rest of the eluate. The elutes were quantified using Bradford reagent (Sigma–Aldrich, St Louis, MO, USA) with BSA (bovine serum albumin) as the standard following the manufacturer’s instructions and saved at −80 °C.

Heme staining and Western blot

Peroxidase-based assays for heme staining were performed as described previously24,49. Briefly, MmcA containing samples or total cell lysate was mixed with loading dye (without β-mercaptoethanol), incubated at 65 °C for 4 min, and resolved by running 12% Mini-Protean TGX SDS-PAGE gels (Bio-Rad, Hercules, CA, USA). Gels were transblotted to 0.2 µm PVDF membrane (Bio-Rad, Hercules, CA, USA) using Trans-Blot Turbo transfer system (Bio-Rad, Hercules, CA, USA) and developed with SuperSignal West Femto kit (Thermo Scientific, Waltham, MA, USA) to detect the heme signal. For Western blot, the heme stain blots were treated with 50 mL stripping buffer (60 mM Tris pH = 7 containing 2% SDS and 7 µL/mL β-mercaptoethanol) shaking at 50 rpm for 1 h at 50 °C. After confirming for the absence of any peroxidase-based signal from heme, the presence of MmcA with FLAG tag was probed with immunoblotting using monoclonal anti-Flag M2-Peroxidase (HRP) antibody (Cat # A8592, Lot # SLCF0816, Sigma–Aldrich, Saint Louis, MO, USA) (1/60,000X dilution) and developed with Immobilon Western Chemiluminescent HRP Substrate (Millipore, Burlington, MA, USA) for signal detection. The ChemiDoc MP Imaging System (Bio-Rad, Hercules, CA, USA) was used for imaging. Protein concentrations were estimated using the Bradford reagent (Sigma–Aldrich, Saint Louis, MO, USA) with BSA (bovine serum albumin) as the standard per the manufacturer’s instructions.

Proteolytic digestion of MmcA and LC-MS/MS analysis

MmcA was digested with trypsin or chymotrypsin in solution for LC-MS/MS analysis as previously described50. Briefly, 20 µL of MmcA at a concentration of 2 mg/mL (total 40 µg) was mixed with 9.6 mg of urea (8 M) in a sterile microfuge tube and incubated for 1 h at room temperature for protein denaturation. The MmcA-Urea mix was diluted 10-fold by adding 180 µL freshly prepared 50 mM ammonium bicarbonate solution. Two aliquots of 100 µL were transferred into new tubes and digested either with a sequencing-grade trypsin (Promega, Madison, WI, USA) or chymotrypsin (Promega, Madison, WI, USA) per manufacturer’s instructions. For trypsin digestion, 2.5 µL of 0.4 µg/µL trypsin (1 μg) was added to 100 µL protein solution (1:20, enzyme to protein ratio) and incubated at 37 °C overnight (ca. 16 h). Similarly, 2 µL of 0.5 µg/µL chymotrypsin (1 µg) was added to another 100 µL protein solution (1:20, enzyme to protein ratio) and incubated at 25 °C overnight (18 h). 75 µL of the overnight-digest were transferred to a clean microfuge tube and submitted for MS analysis (QB3/Chemistry Mass Spectrometry Facility, UC Berkeley). Two independently purified MmcA samples (n = 2) were digested either with trypsin or chymotrypsin and analyzed by Mass Spectrometry. Protein digests were also confirmed by running the remaining 25 µL sample on SDS-PAGE gels followed by Coomassie staining.

UV-visible (vis) absorption spectroscopy with MmcA

All UV-vis spectroscopy was performed at room temperature with a Shimadzu 1900i (Shimadzu, Torrance, CA, USA) kept inside an anaerobic chamber (97% N2, and 3% H2; Coy Laboratory, Grass Lake, USA). Unless specified, all assays were conducted with 2.5–3.5 µM MmcA in 50 µL of assay buffer (50 mM Tris-HCl, 150 mM NaCl, 2% glycerol, pH = 7.4). Stock solutions of 1 mM and 10 mM sodium dithionite were prepared in deionized water. 2-hydroxyphenazine was custom synthesized (AstaTech Inc., Bristol, PA), and a 20 mM stock solution was prepared in 100% ethanol. A 10 mM stock of Anthraquinone-2,6-disulfonate (AQDS) was prepared in deionized water as described before51. The solution was heated at 60 °C until AQDS was completely dissolved (~10 min), cooled down to room temperature, and the pH was adjusted to 7.0. Stocks of 10 mM ferric chloride (FeCl3) and potassium ferricyanide (K3[Fe(CN)6]) were prepared with deionized water. All assay components (buffer, protein, chemicals) were kept in the anaerobic chamber in small volumes (30–50 µL) at least 2 h prior to the assay and were confirmed to be anaerobic by using the redox dye resazurin (0.0001% w/v) and testing for a color change from colorless to pink after 10 min.

Pyridine hemochrome assay

This assay was performed as described before52,53. Briefly, a 0.2 M NaOH with 40% pyridine solution was made fresh using a 1 M NaOH stock and 100% pyridine solution (Sigma–Aldrich, St. Louis, MO). 5 µL (i.e., 1/200) of 0.1 M potassium ferricyanide stock solution was added to 495 µL of the aforementioned NaOH + pyridine mix to generate the pyridine hemochrome assay solution. 50 µL of the assay solution was mixed with 50 µL of TBS buffer (50 mM Tris-HCl, 150 mM NaCl, pH = 7.4) and used as a blank. Next, 50 µL of the assay solution was mixed with 50 µL of MmcA in TBS buffer, and UV-vis scans were immediately performed using a Shimadzu 1900i (Shimadzu, Torrance, CA, USA) to record the oxidized spectra. A 10 mM stock solution of sodium dithionate was added to the protein assay mixture and UV-vis scans were performed using a Shimadzu 1900i (Shimadzu, Torrance, CA, USA) to record the fully reduced pyridine hemochrome spectra.

Cell suspension assays

Cell suspension assays were performed in an anaerobic chamber (97% N2, and 3% H2; Coy Laboratory, Grass Lake, USA) at room temperature as previously described for bacterial cells54 with some modifications. Assays were performed with an M. acetivorans mutant lacking the chromosomal copy of the mmcA locus and expressing mmcA from an inducible promoter on a plasmid and a control strain containing the plasmid pDPG010 described previously24. All strains were grown in HS medium with 125 mM methanol, 2 µg/mL puromycin, and mmcA expression was induced by adding tetracycline to a final concentration of 100 µg/mL. Cells were harvested in mid-exponential phase (OD600 = 0.4–0.6) by centrifugation in the anaerobic chamber. The cell pellet was resuspended in anaerobic high-salt PIPES buffer (50 mM PIPES, 400 mM NaCl, 13 mM KCl, 54 mM MgCl2, and 2 mM CaCl2, pH 6.8) containing 5 mM methanol and washed three times. At the end of the third wash, cells were resuspended in anaerobic high-salt PIPES buffer containing 5 mM methanol and supplemented with a freshly prepared 1:1 Fe3+- nitrilotriacetic acid (NTA) mix using an anaerobic 0.4 M ferric chloride and an anaerobic 0.8 M NTA stock solutions to final concentration of 1 mM ferric chloride and 2 mM of NTA. Fe3+ reduction was monitored by sampling the suspension at different time points and measuring the Fe2+ concentration by using the ferrozine assay55.

Electrochemistry

Protein film voltammetry (PFV) experiments were carried out using a three-electrode cell configuration with the cell thermostated at 10 °C and housed inside a nitrogen-filled MBraun Labmaster glovebox (residual O2 < 1 ppm). The reference electrode was a saturated calomel electrode (SCE), and the counter electrode a platinum wire. The working electrode was a meso-porous indium tin oxide (ITO) electrode, prepared according to reported procedures56. Briefly, a pyrolytic graphite edge (PGE) electrode (3 mm diameter) was polished and sonicated in water. ITO nanoparticles (Sigma, <50 nm) were then deposited on the PGE surface by electrophoretic deposition. The PGE electrode was submerged in an acetone solution (20 mL) of I2 (0.01 g) and ITO (0.02 g), and a potential of 10 V was applied for 6 min using a graphite rod as the auxiliary electrode that was held approximately 1 cm away. The ITO electrode was then thoroughly rinsed with water and dried prior to use. To deposit the protein, a 3 μL aliquot of the protein solution (100 μM MmcA or 55 μM OTR), the latter prepared as previously described57 was placed on the electrode for 3 min. Excess protein solution was then removed by rinsing with cold buffer, and the electrode was immediately placed into the electrolyte buffer, which contained 10 mM MES, 10 mM MOPS, 10 mM TAPS, 10 mM CHES, 10 mM HEPES, 10 mM CAPS, and 200 mM NaCl (pH 7.4). Cyclic voltammograms (CVs) were collected using the GPES software package (Ecochemie) that was connected to a PGSTAT30 AutoLab potentiostat (Ecochemie). All PFV data were analyzed using the qSOAS package58, through which background electrode capacitance was subtracted, and data were filtered to remove electrical noise. Deconvolution of the redox feature was achieved within qSOAS using procedures reported previously, where the redox stoichiometry of all 7 heme cofactors was set to 1.0 (n = 1).

Bioinformatics analyses

MmcA homologs were extracted from the NCBI non-redundant protein database using the MmcA (MA0658) protein sequence from M. acetivorans as the query and Archaea as the search database. Alignments and tree building were conducted in Geneious Prime 2023.0.3 (https://www.geneious.com). Any partial sequences (<375 aa) were discarded. The sequences were aligned using MUSCLE with default parameters. Maximum-likelihood tree of MmcA was generated using RAxML (Protein Model - GAMMA BLOSUM62; Algorithm - Rapid Bootstrapping and search for best-scoring ML tree; Number of starting trees or bootstrap replicates - 100; Parsimony random seed - 1). Gene ortholog neighborhood analysis was performed on Integrated Microbial Genomes and Microbiomes platform59. For structural alignment and model building, an AlphaFold2 prediction of MmcA60 and its closest available crystal structure i.e., 1SP3 for OTR from Shewanella oneidensis (SO4144)40, were docked using matchmaker tools and 1SP3 as a reference structure in ChimeraX61.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.