A pathway for biological methane production using bacterial iron-only nitrogenase


Methane (CH4) is a potent greenhouse gas that is released from fossil fuels and is also produced by microbial activity, with at least one billion tonnes of CH4 being formed and consumed by microorganisms in a single year1. Complex methanogenesis pathways used by archaea are the main route for bioconversion of carbon dioxide (CO2) to CH4 in nature2,3,4. Here, we report that wild-type iron-iron (Fe-only) nitrogenase from the bacterium Rhodopseudomonas palustris reduces CO2 simultaneously with nitrogen gas (N2) and protons to yield CH4, ammonia (NH3) and hydrogen gas (H2) in a single enzymatic step. The amount of CH4 produced by purified Fe-only nitrogenase was low compared to its other products, but CH4 production by this enzyme in R. palustris was sufficient to support the growth of an obligate CH4-utilizing Methylomonas strain when the two microorganisms were grown in co-culture, with oxygen (O2) added at intervals. Other nitrogen-fixing bacteria that we tested also formed CH4 when expressing Fe-only nitrogenase, suggesting that this is a general property of this enzyme. The genomes of 9% of diverse nitrogen-fixing microorganisms from a range of environments encode Fe-only nitrogenase. Our data suggest that active Fe-only nitrogenase, present in diverse microorganisms, contributes CH4 that could shape microbial community interactions.


The microbial degradation of complex organic material in anoxic freshwater environments involves a microbial food chain that culminates in the production of CH4 by methanogenic archaea5. In aerobic oceans, marine microorganisms can produce CH4 as a by-product of using the phosphate source methylphosphonate, which is produced by Nitrosopumilus maritimus6. Finally, small amounts of CH4 are generated by bacteria and archaea that use the Wood–Ljungdahl pathway to either fix CO2 to acetyl-CoA or to oxidize acetyl-CoA to CO27,8.

In common with CH4 production, biological nitrogen fixation is solely a microbial process. Carried out by members of the bacteria and archaea, biological nitrogen fixation accounts for 50% of the conversion of nitrogen gas (N2) to NH3 on Earth today, with the balance contributed by the industrial Haber–Bosch process9. Three homologous nitrogenase enzymes exist and are known as the molybdenum-iron or Mo nitrogenase, the vanadium-iron or V nitrogenase and the iron-iron or Fe-only nitrogenase10. Mo nitrogenase (comprised of NifHDK proteins) is found in all microorganisms that fix N2 and some microorganisms have, in addition, anfHDGK genes encoding Fe-only nitrogenase and/or vnfHDGK genes encoding V nitrogenase (Supplementary Table 1). Mo nitrogenase is more efficient at fixing N2 than the two alternative nitrogenases10 and, in the laboratory, it is preferentially expressed by microorganisms when Mo is present. From this it has been concluded that, in nature, the alternative nitrogenases are back-up enzymes that are deployed in situations where Mo is limiting. A number of types of environments with limited Mo-availability are known in nature11,12,13 and alternative nitrogenases would be expected to be expressed in nitrogen-fixing bacteria at such sites or in microenvironments that transiently become Mo-depleted. Recent reports using an isotope fractionation technique, that can distinguish between the forms of nitrogenase by measuring 13C isotopes in the acetylene reduction assay, indicate that V- and Fe-only nitrogenases are active in two coastal marine environments, both of which contain measurable Mo at the sampling sites14,15. This suggests that the relationship between the presence of Mo and the expression of active alternative nitrogenases in natural environments may be complex.

Nitrogenases are known to be promiscuous in the reduction reactions that they can catalyse. For example, V and Mo nitrogenases have been shown to reduce and concatenate carbon monoxide (CO) to a variety of hydrocarbon products in vitro16,17,18,19,20. In an exploration of additional reduction reactions, it was reported that a variant of Mo nitrogenase with two amino acid substitutions near its active site catalyses the reduction of CO2 to CH4 in vitro21. This extremely difficult eight-electron reduction reaction has an activation energy barrier similar to that of N2 reduction. The CO2 reduction reaction was accompanied by the production of H2, a known feature of nitrogenases22, but the variant enzyme did not retain its ability to convert N2 to NH3. The variant form of Mo nitrogenase also functioned in vivo to catalyse the production of CH4 by the anoxygenic phototrophic bacterium R. palustris23. As expected from in vitro experiments, the R. palustris cells did not retain their ability to fix nitrogen when they expressed the variant Mo nitrogenase.

The two amino acids NifDV75 and NifDH201 in the R. palustris enzyme that, when changed to NifDA75 and NifDQ201, conferred the ability to reduce CO2, are conserved in all known sequences of Mo, V and Fe-only nitrogenases (Supplementary Figs 1, 2). To extend these studies we made the corresponding amino acid changes in the R. palustris V and Fe-only nitrogenases, expressed these enzymes in mutants that can synthesize only the V enzyme or only the Fe-only enzyme24 and measured CH4 production. R. palustris cells expressing the wild-type V nitrogenase produced a very small, but detectable, amount of CH4, whereas the cells expressing the variant V nitrogenase with VnfDV57AH180Q amino acid substitutions produced a much higher amount of CH4 (Fig. 1a). In view of this expected result, we were surprised to find that R. palustris cells expressing wild-type Fe-only nitrogenase produced significant amounts of CH4, whereas a strain expressing a variant Fe-only nitrogenase with AnfDV57AH180Q amino acid substitutions failed to produce detectable CH4 (Fig. 1a). These results suggest that the Fe-only nitrogenase enzyme is naturally capable of reducing CO2 to CH4. CH4 production by Fe-only nitrogenase in R. palustris was light-dependent, with the rate of CH4 production increasing as light intensity was increased to 30 μmol photons m-2 s-1 (Fig. 1b).

Fig. 1: Cells expressing wild-type Fe-only nitrogenase produce CH4

a, CH4 production by non-growing cell suspensions of R. palustris strains expressing either wild-type V (strain CGA766) or Fe-only (strain CGA755) nitrogenases (left) or their variants (strains CGA7661 and CGA7551, respectively) (right). Cell suspensions were incubated in light for 10 days until maximal CH4 production was achieved. These data are the average of three independent experiments, and the error bars represent the s.d. b, CH4 production by non-growing cell suspensions of an R. palustris mutant (strain CGA7553) that encodes only the Fe-only nitrogenase as a function of light intensity. Data are the average of three biological replicates and the error bars represent the s.d. c, GC–MS traces showing conversion of 13C-labeled bicarbonate to CH4 by cell suspensions of R. palustris CGA7553. The black traces are the GC–MS data for monitoring 12CH4 (m/z = 16) and the red traces are the data for monitoring 13CH4 (m/z = 17). This figure is representative of two biological repetitions. d,e, Metabolic route of CH4 production by R. palustris expressing wild-type Fe-only nitrogenase. ATP is produced by cyclic photophosphorylation, in which electrons energized by light are cycled through a proton-pumping electron transport chain. Electrons derived from oxidation of the inorganic compound thiosulfate (d) or the organic compound acetate (e) are used in biosynthesis and in CO2 fixation and are also diverted to nitrogenase. CO2 generated from bicarbonate or from the oxidation of acetate and N2 from the atmosphere are converted by Fe-only nitrogenase to CH4 and NH3, respectively. Not shown here is H2, which is also a product of nitrogenase. CBB, Calvin-Benson-Bassham; ND, not detected; WT, wild type; hv, light photons.

To verify that the CH4 produced in vivo by the Fe-only nitrogenase-expressing strain of R. palustris came from CO2 reduction, we incubated cell suspensions of this strain with thiosulfate as a source of electrons and 13C-labeled sodium bicarbonate (NaH13CO3) as a source of 13CO2. R. palustris has a carbonic anhydrase that converts HCO3- to CO2. Such cells produced a compound with an m/z ratio of 17 and a gas chromatography retention time corresponding to 13CH4. When NaH12CO3 was used as substrate, a product with the same retention time as 13CH4, but an m/z of 16, was detected (Fig. 1c). A diagram of the probable metabolic route to CH4 production taken by cells growing with thiosulfate as a source of electrons, HCO3- as a source of carbon and light as a source of ATP is shown in Fig. 1d. In other experiments, we found that Fe-only nitrogenase-expressing R. palustris cells grown with acetate as a carbon and electron source also produced CH4. Most likely the CH4 was derived from CO2 that is released when acetate is metabolized by cells25 (Fig. 1e).

To confirm that the Fe-only nitrogenase was responsible for in vivo CO2 reduction to CH4, a His-tagged version of the FeFe (AnfDGK) protein was purified from R. palustris (Fig. 2a, Supplementary Fig. 3) and its peptide identity was verified by mass spectrometry. The purified enzyme converted CO2 to CH4 at a low yield (1 nmol CH4 per nmol FeFe protein) compared to the yields with H+ (1793 nmol H2 per nmol FeFe protein) and N2 (296 nmol NH3 per nmol FeFe protein) (Fig. 2b). It can carry out these three reactions concurrently. This confirms that CH4 production along with NH3 and H2 production are intrinsic properties of R. palustris Fe-only nitrogenase.

Fig. 2: Fe-only nitrogenase purified from R. palustris reduces CO2 to CH4 in vitro

a, SDS–PAGE analysis of the AnfDGK protein purified from R. palustris strain CGA7554. A single gel was run and the identities of the Anf peptides were confirmed by mass spectrometry. AnfH was purified from A. vinelandii strain DJ1255 and used as the Fe protein in nitrogenase assays. A. vinelandii and R. palustris AnfH proteins share 87% amino acid identity. b, Products formed by purified Fe-only nitrogenase. These data are the average of two independent experiments, with the error bars representing the range of values.

To determine if CH4 production is a general property of Fe-only nitrogenase and not an exclusive feature of the R. palustris enzyme, we tested the abilities of three additional bacterial species that have Fe-only nitrogenase to produce CH4 when growing under nitrogen-fixing conditions. As shown in Fig. 3, R. palustris, as well as two other phototrophic bacteria, Rhodospirillum rubrum and Rhodobacter capsulatus, and the obligate aerobic bacterium Azotobacter vinelandii, each produced CH4 when grown with N2 as a sole nitrogen source in Mo-depleted medium, a condition that promotes the expression of active Fe-only nitrogenase in preference of Mo nitrogenase in each of these species26,27,28. None of the species produced significant amounts of CH4 when grown in medium supplemented with Mo, which induces expression of the Mo nitrogenase in these species. A. vinelandii and R. palustris also encode V nitrogenase, but V was not added to the growth media used for these experiments. These results suggest that reduction of CO2 to CH4 is a general property of Fe-only nitrogenase. We identified 358 diazotrophs (those encoding for at least Mo nitrogenase) among available bacterial and archaeal genomes (7.8% of total genomes) and, of these, 31 (9%) also encoded for Fe-only nitrogenase (Supplementary Table 1). Fe-only nitrogenase genes were detected in the genomes of physiologically diverse organisms, including those characterized as aerobes, anaerobes and anoxygenic phototrophs, as well as chemoautotrophs and chemoheterotrophs. Moreover, these genes were identified in taxonomically diverse strains, including those that belong to the bacterial phyla Bacteroidetes, Chlorobi, Firmicutes, Proteobacteria and Verrucomicrobia and the archaeal phylum Euryarchaeota. These data, along with other recently published data15, show that Fe-only nitrogenase is widely distributed in nature.

Fig. 3: Effect of molybdenum on CH4 production by nitrogen-fixing bacteria.

a, CH4 production by R. palustris growing under nitrogen-fixing conditions in medium with (+Mo) and without (–Mo) molybdenum. b, Growth of R. palustris with N2 gas as the sole nitrogen source in the presence (+Mo) or absence (-Mo) of molybdenum. c, CH4 production by wild-type photosynthetic bacteria grown anaerobically in light with N2 as a sole nitrogen source in the presence (+Mo) or absence (-Mo) of molybdenum. d, CH4 production by A. vinelandii grown aerobically in nitrogen-free medium in the presence (+Mo) or absence (–Mo) of molybdenum. Data are the average of three (ac) or four (d) biological replicates and the error bars represent the s.d.

A consequence of the release of CH4 by microorganisms is that it provides anaerobic CH4-utilizing archaea and aerobic methanotropic bacteria with an energy and carbon source and thereby contributes not only to the global carbon cycle29, but also to microbial community interactions. To demonstrate a community interaction, we set up a co-culture of R. palustris with Methylomonas sp. LW13, an obligate CH4-utilizing bacterium isolated from a freshwater lake30. When the co-culture was incubated in Mo-depleted medium (-Mo) and spiked with O2 to enable Methylomonas cells to convert CH4 to CH3OH, the number of Methylomonas sp. LW13 cells increased by over threefold in 15 days, suggesting that cells were growing on CH4 produced by R. palustris. Methylomonas sp. LW13 did not increase in cell number when incubated with R. palustris in medium supplemented with Mo (+Mo) (Fig. 4), but it grew under this condition when CH4 was added. In complementary experiments, we incubated Methylomonas sp. LW13 with headspace gases collected from cultures of the R. palustris Fe-only nitrogenase strain that had been grown phototrophically with acetate and 13C-labeled bicarbonate. After a period of incubation, mass spectrometry revealed compounds with fragmentation patterns consistent with 13C labelled glucose-6-phosphate and 13C fructose-6-phosphate in Methylomonas sp. LW13 cells (Supplementary Fig. 4), supporting the conclusion that 13CH4 produced by the Fe-only nitrogenase was incorporated into Methylomonas cellular components. It is likely that microbial interactions like these occur in nature because CH4 is produced concomitantly with NH3 by Fe-only nitrogenase, which is known to be active in nature14,15, and Fe-only nitrogenase in bacteria like R. palustris and A. vinlandeii generates CH4 when O2 is present and available to support the activities of CH4 monoxygenases in CH4-oxidizing bacteria.

Fig. 4: Co-culture of Methylomonas sp. LW13 and R. palustris

Methylomonas sp. LW13 increased in cell numbers when the co-culture was incubated in Mo-depleted medium, a condition that derepresses expression of the R. palustris Fe-only nitrogenase. The relative cell numbers were determined by quantitative PCR. These data are the average of two independent experiments, and the error bars represent the range of values.

Fe-only nitrogenase would be expected to make a minor contribution to net CH4 production in freshwater anaerobic ecosystems dominated by methanogenic archaea1,3. However, it may play a role in shaping microbial community interactions in marine sediments where sulfidogenic microorganisms tend to outcompete methanogens, in host-associated microbiomes where fermentative microorganisms often dominate and in aerobic soil environments. Microorganisms that encode Fe-only nitrogenase and CH4-oxidizing bacteria are known to coexist in each of these types of habitats.


Bacterial strains and growth conditions

For genetic manipulations, R. palustris strains were grown aerobically on defined mineral medium (PM)31 agar supplemented with 10 mM succinate at 30 °C. Escherichia coli S17-1 was grown in LB medium at 37 °C. When appropriate, R. palustris was grown with gentamicin at 100 μg ml-1. E. coli cultures were supplemented with gentamicin at 20 μg ml-1. R. palustris strains, R. rubrum UR2 and R. capsulatus SB1003 were grown anaerobically in nitrogen-fixing medium (NFM)32. This defined mineral medium is the same as PM medium but lacks ammonium sulfate. N2 gas was provided in the headspace of sealed culture tubes. For the studies described here, we also omitted the Mo salt from the trace element solution used in NFM and added Wolfe’s vitamins (0.05 mg l-1 para-aminobenzoic acid, 0.02 mg l-1 folic acid, 0.05 mg l-1 lipoic acid, 0.05 mg l-1 riboflavin, 0.05 mg l-1 thiamine, 0.05 mg l-1 nicotinic acid amide, 0.1 mg l-1 pyridoxamine, 0.05 mg l-1 pantothenic acid, 0.001 mg l-1 cobalamin and 0.02 mg l-1 biotin, pH = 7.0), 10 μM nickel chloride and 10 mM sodium bicarbonate, unless otherwise indicated. Acetate (20 mM or as indicated) was included as the carbon source and the medium was supplemented with 10 µM sodium molybdate (Na2MoO4) where indicated. Non-growing R. palustris cell suspensions were prepared as follows. Cells were first grown in NFM supplemented with 20 mM acetate and 0.1% yeast extract to obtain sufficient biomass and derepress the expression of V or Fe-only nitrogenases. N2 gas was provided in the headspace of sealed culture tubes. Cultures in late log phase were harvested by centrifuging at 4,000 rpm for 6 minutes. The cell pellets were washed twice with Mo-free NFM medium, and then transferred to 27-ml sealed culture tubes with 10 ml NFM medium supplemented with 20 mM acetate and 10 mM sodium bicarbonate. Ar gas was provided in the headspace of the sealed incubation tubes. In the media used to derepress the expression of wild-type or variant V nitrogenases, 10 µM vanadium chloride (VCl3) was added. All cultures were incubated anaerobically with 30 μmol photons m-2 s-1 illumination from a 60 W incandescent light bulb (General Electric), unless otherwise indicated. A. vinelandii (DJ1255) was grown in Burke’s medium33 with the Na2MoO4 omitted under nitrogen-fixing conditions in a 100 l custom made fermenter with stirring and aeration to an optical density (OD)600 of 1.8–2.0 and then harvested.

Genetic manipulation of R. palustris

All strains and plasmids used are listed in Supplementary Table 2. R. palustris mutants CGA7661 and CGA7551 were prepared as follows. Polymerase chain reaction (PCR) products containing vnfDV57AH180Q and anfDV57AH180Q were inserted into PstI-digested pJQ200SK using the In-Fusion PCR cloning system (Clontech). The obtained plasmids pJQ-vnfDV57AH180Q and pJQ-anfDV57AH180Q were mobilized into R. palustris by conjugation with E. coli S17-1 and double-crossover events for allelic exchange were achieved using a selection and screening strategy as described previously34. Allelic exchange was verified using PCR amplification and sequencing of the resulting PCR product. In-frame deletions of nifHDK and vnfDGK were created by PCR using the Q5 high-fidelity DNA polymerase to amplify 1,014 bp (nifHDK) or 1,003 bp (vnfHDGK) of DNA upstream of the start codon for nifH or vnfD and 1,032 bp (nifHDK) or 1,013 bp (vnfDGK) of DNA downstream of the stop codon for nifK or vnfK. These fragments were then incorporated into PstI-digested pJQ200SK suicide vector using the In-Fusion PCR cloning system. Plasmid pJQ-ΔnifHDK was mobilized into R. palustris CGA755 by conjugation with E. coli S17-1 and double-crossover events for allelic exchange were achieved using a selection and screening strategy as described previously34. The obtained R. palustris CGA7551 strain was then used as the parent strain and pJQ-ΔvnfDGK was used as the suicide plasmid for the next round of in-frame deletion of vnfDGK, using the same protocol as mentioned above. All deletions were verified by PCR. For the preparation of R. palustris CGA7554, Q5 High-Fidelity DNA polymerase and primers were used to incorporate eight histidines (CAC) before the stop codon of anfD. The product was incorporated into PstI-digested pJQ200SK using the In-Fusion PCR cloning system and inserted into the R. palustris CGA009 chromosome by allelic exchange. A. vinelandii DJ1255 was constructed in the following way. First, an in-frame deletion within nifD and nifK35 was placed within strain CA636 to yield DJ1254. CA6 is tungsten tolerant and, therefore, deficient in the accumulation of Mo37. A kanamycin resistance cartridge was then placed within the MfeI restriction enzyme site of the vnfD-coding region of DJ1254 to yield DJ1255. Strain DJ1255 is therefore deficient in the capacity to fix nitrogen using either the Mo-dependent system or the V-dependent system. However, it retains the capacity to form the anf-encoded nitrogen fixation system and can grow diazotrophically in medium that contains neither Mo nor V. Details of strain constructions were performed as described previously35.

CH4 measurements from whole cells

When cultures of R. palustris CGA010, R. rubrum UR2 and R. capsulatus SB 1003 reached their maximal OD660, gas-phase samples were withdrawn with a Hamilton sample lock syringe from the culture vial headspace. CH4 was measured with a Shimadzu GC-2014 gas chromatograph as described previously23. A. vinelandii (wild type) was grown in Burke’s medium33 under nitrogen-fixing conditions. When cells reached an OD600 of 0.8–1.0, 15 ml aliquots of the cell suspension were transferred to 25 ml degassed vials containing sodium bicarbonate (10 mM final concentration). Vials were equilibrated and then 0.2 atm O2 was added. O2 was subsequently injected at 0.15 atm every 2 hours over the course of the first 8 hours to maintain cell viability. Following a 24 hour incubation period at 30 °C, the headspace was charged with 1 ml of phosphate buffer. Gas-phase samples were taken and measured with a Shimadzu GC-8A as described previously21. Total protein concentrations were determined using the Bio-Rad protein assay kit.

Capillary gas chromatography mass spectrometry analysis for in vivo CO2 reduction to CH4 by R. palustris

Gas chromatography–mass spectrometry (GC–MS) analysis was conducted to confirm the production of CH4 from CO2 reduction, using a Shimadzu GC-2010 gas chromatograph equipped with a programmed temperature vaporizing injector and a Shimadzu GCMS-QP2010S mass spectrometer by using 12/13C-enriched NaHCO3 as CO2 source for an R. palustris ΔnifHΔvnfH mutant. Cells were grown in several 10 ml tubes of NFM medium with 20 mM acetate and 0.1% yeast extract. Then, cells were harvested by centrifugation, transferred to 100 ml fresh NFM with 10 mM thiosulfate but without bicarbonate and finally transferred to NFM with 10 mM thiosulfate without bicarbonate or with 10 mM 12C- or 13C-bicarbonate (Cambridge Isotope Lab). Identification of 12CH4 and 13CH4 was done according to a standard protocol previously published23.

Nitrogenase purification

For protein purification, a 1 l culture was used to inoculate 10 l of NFM medium containing 20 mM acetate and 0.1% yeast extract and incubated at 30 °C in light until it reached an OD660 of ~0.7. Cell extracts from R. palustris cells were prepared by an osmotic shock method using a French pressure cell operated at 1,500 lb in-2 in a degassed 50 mM Tris-HCl buffer (pH 8.0) with 2 mM sodium dithionite under Ar. His-tagged Fe-only nitrogenase protein was purified by an immobilized metal-affinity chelation chromatography protocol38, with minor modifications. Fe protein (AnfH) was purified from A. vinelandii cells using a previously described protocol39, with minor modifications. Protein concentrations were determined by the Biuret assay using BSA as standard. The purities of these proteins were checked based on SDS–PAGE analysis with Coomassie staining.

CO2 reduction assays

CO2 reduction assays were conducted in 9.4 ml serum vials containing an assay buffer consisting of an MgATP regeneration system (15 mM MgCl2, 90 mM phosphocreatine, 15 mM ATP, 0.6 mg ml-1 creatine phosphokinase and 1.2 mg ml-1 BSA) and 12 mM sodium dithionite in 100 mM MOPS buffer at pH 7.0. After solutions were made anoxic, 0.45 atm CO2 was added and the gas and liquid phases were allowed to equilibrate for approximately 20 minutes. AnfDGK protein was then added, the vials ventilated to atmospheric pressure and the reaction initiated by the addition of A. vinelandii AnfH protein. Reactions were conducted at 30 °C for 6 hours and then quenched by the addition of 700 µl of 400 mM EDTA pH 8.0. Quantification of CH4 was done according to a published protocol21.

N2 and H+ reduction assays

Reduction assays were conducted in 9.4 ml serum vials containing an assay buffer consisting of a MgATP regeneration system (6.7 mM MgCl2, 30 mM phosphocreatine, 5 MM ATP, 0.2 mg ml-1 creatine phosphokinase, 1.2 mg ml-1 BSA) and 12 mM sodium dithionite in 100 mM Mops buffer at pH 7.0. After making solutions anoxic, headspace gases in reaction vials were adjusted to appropriate partial pressures per substrate (1 atm N2 for N2 and 1 atm Ar for proton). AnfDGK protein was then added, vials equilibrated to atmospheric pressure and reactions initiated by addition of A. vinelandii AnfH protein. Reactions were conducted at 30 °C for 15 minutes and then quenched by addition of 300 µl of 400 mM EDTA (pH 8.0). NH3 and H2 were quantified according to published methods with minor modifications40,41.

Identification and compilation of nitrogenases homologues

AnfD (AGK17546), VnfD (AGK18954) and NifD (AGK18378) amino acid sequences from A. vinelandii were used in separate  protein Basic Local Alignment Search Tool queries to identify homologues in complete genome sequences (n = 4,586 genomes), specifying 30% per cent sequence identities and 60% sequence coverage.

Co-culture of R. palustris CGA010 and Methylomonas sp. LW13

The co-culture was incubated in a modified NFM medium supplemented with 20 mM acetate, 10 mM NaHCO3 and 20% nitrate mineral salts medium42 at room temperature. Light was provided from a 60-W incandescent light bulb. The initial inoculation ratio of R. palustris CGA010 to Methylomonas sp. LW13 was ~50:1. O2 was provided for the growth of Methylomonas sp. LW13 by adding 10% filtered air every 48 hours into a 250 ml serum bottle containing 50 ml co-culture. Samples were taken from the co-culture at day 1, 5, 10 and 15, and quantitative PCR was employed to determine the cell number of Methylomonas sp. LW13, as previously described43.

Liquid chromatography–tandem mass spectrometry analysis for CH4 uptake by Methylomonas sp. LW13

R. palustris CGA010 cells, which were pelleted from a 1.2 l growing culture, resuspended in 100 ml Mo-free NFM medium supplemented with 5 mM acetate, 10 mM thiosulfate and 20 mM 13C-enriched NaHCO3 and transferred to a 160 ml serum bottle, were used to make 13CH4. The headspace of a 50 ml culture of Methylomonas sp. LW13 at OD600 = 0.05 growing in nitrate mineral salts medium42 was refreshed using the following procedure daily for 5 days. The headspace of the Methylomonas sp. LW13 culture was flushed with filtered air for 1 min, and then 150 ml of headspace was extracted. The 150 ml collected headspace from R. palustris CGA010 cultures was injected into Methylomonas sp. LW13 culture vials using an air-tight syringe with an inline 0.22 μm filter. The vials were further over-pressured with 50 ml of filtered air to maintain similar pressure as control groups. The negative control group was grown on 12CH4, and the positive control groups were grown on 13CH4 (99% atom abundance, Sigma-Aldrich). The Methylomonas sp. LW13 cell cultures were quenched and harvested with the fast filtration procedure described previously44. The intermediate metabolites were extracted and identified by liquid chromatography–tandem mass spectrometry (Xevo, Waters, Milford, MA) equipped with a Zic-pHilic column (SeQuant, PEEK 150 × 2.1 mm × 5 μm) according to a standard previously published protocol44.

Life Sciences Reporting Summary

Further information on experimental design is available in the Life Sciences Reporting Summary.

Data availability

The datasets generated and analysed during the current study are available from the corresponding author on reasonable request.


  1. 1.

    Thauer, R. K. & Shima, S. Biogeochemistry: methane and microbes. Nature 440, 878–879 (2006).

    CAS  Article  Google Scholar 

  2. 2.

    Yvon-Durocher, G. et al. Methane fluxes show consistent temperature dependence across microbial to ecosystem scales. Nature 507, 488–491 (2014).

    CAS  Article  Google Scholar 

  3. 3.

    Thauer, R. K., Kaster, A. K., Seedorf, H., Buckel, W. & Hedderich, R. Methanogenic archaea: ecologically relevant differences in energy conservation. Nat. Rev. Microbiol. 6, 579–591 (2008).

    CAS  Article  Google Scholar 

  4. 4.

    Brauer, S. L., Cadillo-Quiroz, H., Yashiro, E., Yavitt, J. B. & Zinder, S. H. Isolation of a novel acidiphilic methanogen from an acidic peat bog. Nature 442, 192–194 (2006).

    Article  Google Scholar 

  5. 5.

    Ferry, J. G. Fundamentals of methanogenic pathways that are key to the biomethanation of complex biomass. Curr. Opin. Biotech. 22, 351–357 (2011).

    CAS  Article  Google Scholar 

  6. 6.

    Metcalf, W. W. et al. Synthesis of methylphosphonic acid by marine microbes: a source for methane in the aerobic ocean. Science 337, 1104–1107 (2012).

    CAS  Article  Google Scholar 

  7. 7.

    Schauder, R., Eikmanns, B., Thauer, R. K., Widdel, F. & Fuchs, G. Acetate oxidation to CO2 in anaerobic bacteria via a novel pathway not involving reactions of the citric acid cycle. Arch. Microbiol. 145, 162–172 (1986).

    CAS  Article  Google Scholar 

  8. 8.

    Vorholt, J., Kunow, J., Stetter, K. O. & Thauer, R. K. Enzymes and coenzymes of the carbon monoxide dehydrogenase pathway for autotrophic CO2 fixation in Archaeoglobus lithotrophicus and the lack of carbon monoxide dehydrogenase in the heterotrophic A. profundus. Arch. Microbiol. 163, 112–118 (1995).

    CAS  Article  Google Scholar 

  9. 9.

    Spatzal, T. The center of biological nitrogen fixation: FeMo-cofactor. Z. Anorg. Allg. Chem. 641, 10–17 (2015).

    CAS  Article  Google Scholar 

  10. 10.

    Eady, R. R. Structure-function relationships of alternative nitrogenases. Chem. Rev. 96, 3013–3030 (1996).

    CAS  Article  Google Scholar 

  11. 11.

    Bishop, P. E. & Joerger, R. D. Genetics and molecular biology of alternative nitrogen fixation systems. Annu. Rev. Plant Phys. 41, 109–125 (1990).

    CAS  Article  Google Scholar 

  12. 12.

    Helz, G. R. et al. Mechanism of molybdenum removal from the sea and its concentration in black shales: EXAFS evidence. Geochim. Cosmochim. Acta 60, 3631–3642 (1996).

    CAS  Article  Google Scholar 

  13. 13.

    Barron, A. R. et al. Molybdenum limitation of asymbiotic nitrogen fixation in tropical forest soils. Nat. Geosci. 2, 42–45 (2009).

    CAS  Article  Google Scholar 

  14. 14.

    Zhang, X. et al. Alternative nitrogenase activity in the environment and nitrogen cycle implications. Biogeochemistry 127, 189–198 (2016).

    CAS  Article  Google Scholar 

  15. 15.

    McRose, D. L., Zhang, X., Kraepiel, A. M. & Morel, F. M. Diversity and activity of alternative nitrogenases in sequenced genomes and coastal environments. Front. Microbiol. 8, 267 (2017).

    Article  Google Scholar 

  16. 16.

    Seefeldt, L. C., Yang, Z.-Y., Duval, S. & Dean, D. R. Nitrogenase reduction of carbon-containing compounds. Biochim. Biophys. Acta 1827, 1102–1111 (2013).

    CAS  Article  Google Scholar 

  17. 17.

    Lee, C. C. et al. Uncoupling binding of substrate CO from turnover by vanadium nitrogenase. Proc. Natl Acad. Sci. USA 112, 13845–13849 (2015).

    CAS  Article  Google Scholar 

  18. 18.

    Lee, C. C., Hu, Y. & Ribbe, M. W. Vanadium nitrogenase reduces CO. Science 329, 642 (2010).

    CAS  Article  Google Scholar 

  19. 19.

    Hu, Y., Lee, C. C. & Ribbe, M. W. Extending the carbon chain: hydrocarbon formation catalyzed by vanadium/molybdenum nitrogenases. Science 333, 753–755 (2011).

    CAS  Article  Google Scholar 

  20. 20.

    Yang, Z.-Y., Dean, D. R. & Seefeldt, L. C. Molybdenum nitrogenase catalyzes the reduction and coupling of CO to form hydrocarbons. J. Biol. Chem. 286, 19417–19421 (2011).

    CAS  Article  Google Scholar 

  21. 21.

    Yang, Z.-Y., Moure, V. R., Dean, D. R. & Seefeldt, L. C. Carbon dioxide reduction to methane and coupling with acetylene to form propylene catalyzed by remodeled nitrogenase. Proc. Natl Acad. Sci. USA 109, 19644–19648 (2012).

    CAS  Article  Google Scholar 

  22. 22.

    Hoffman, B. M., Lukoyanov, D., Yang, Z.-Y., Dean, D. R. & Seefeldt, L. C. Mechanism of nitrogen fixation by nitrogenase: the next stage. Chem. Rev. 114, 4041–4062 (2014).

    CAS  Article  Google Scholar 

  23. 23.

    Fixen, K. R. et al. Light-driven carbon dioxide reduction to methane by nitrogenase in a photosynthetic bacterium. Proc. Natl Acad. Sci. USA 113, 10163–10167 (2016).

    CAS  Article  Google Scholar 

  24. 24.

    Oda, Y. et al. Functional genomic analysis of three nitrogenase isozymes in the photosynthetic bacterium Rhodopseudomonas palustris. J. Bacteriol. 187, 7784–7794 (2005).

    CAS  Article  Google Scholar 

  25. 25.

    McKinlay, J. B. & Harwood, C. S. Carbon dioxide fixation as a central redox cofactor recycling mechanism in bacteria. Proc. Natl Acad. Sci. USA 107, 11669–11675 (2010).

    CAS  Article  Google Scholar 

  26. 26.

    Masepohl, B. & Hallenbeck, P. C. Nitrogen and molybdenum control of nitrogen fixation in the phototrophic bacterium Rhodobacter capsulatus. Adv. Exp. Med. Biol. 675, 49–70 (2010).

    CAS  Article  Google Scholar 

  27. 27.

    Waugh, S. I. et al. The genes encoding the delta subunits of dinitrogenases 2 and 3 are required for Mo-independent diazotrophic growth by Azotobacter vinelandii. J. Bacteriol. 177, 1505–1510 (1995).

    CAS  Article  Google Scholar 

  28. 28.

    Lehman, L. J. & Roberts, G. P. Identification of an alternative nitrogenase system in Rhodospirillum rubrum. J. Bacteriol. 173, 5705–5711 (1991).

    CAS  Article  Google Scholar 

  29. 29.

    Stanley, E. H. et al. The ecology of methane in streams and rivers: patterns, controls, and global significance. Ecol. Monogr. 86, 146–171 (2016).

    Article  Google Scholar 

  30. 30.

    Auman, A. J., Stolyar, S., Costello, A. M. & Lidstrom, M. E. Molecular characterization of methanotrophic isolates from freshwater lake sediment. Appl. Environ. Microbiol. 66, 5259–5266 (2000).

    CAS  Article  Google Scholar 

  31. 31.

    Kim, M. K. & Harwood, C. S. Regulation of benzoate-CoA ligase in Rhodopseudomonas palustris. FEMS Microbiol. Lett. 83, 199–203 (1991).

    CAS  Google Scholar 

  32. 32.

    Huang, J. J., Heiniger, E. K., McKinlay, J. B. & Harwood, C. S. Production of hydrogen gas from light and the inorganic electron donor thiosulfate by Rhodopseudomonas palustris. Appl. Environ. Microbiol. 76, 7717–7722 (2010).

    CAS  Article  Google Scholar 

  33. 33.

    Toukdarian, A. & Kennedy, C. Regulation of nitrogen metabolism in Azotobacter vinelandii: isolation of ntr and glnA genes and construction of ntr mutants. EMBO J. 5, 399–407 (1986).

    CAS  Article  Google Scholar 

  34. 34.

    Rey, F. E., Heiniger, E. K. & Harwood, C. S. Redirection of metabolism for biological hydrogen production. Appl. Environ. Microbiol. 73, 1665–1671 (2007).

    CAS  Article  Google Scholar 

  35. 35.

    Robinson, A. C., Burgess, B. K. & Dean, D. R. Activity, reconstitution, and accumulation of nitrogenase components in Azotobacter vinelandii mutant strains containing defined deletions within the nitrogenase structural gene cluster. J. Bacteriol. 166, 180–186 (1986).

    CAS  Article  Google Scholar 

  36. 36.

    Bishop, P. E., Jarlenski, D. M. L. & Hetherington, D. R. Evidence for an alternative nitrogen fixation system in Azotobacter vinelandii. Proc. Natl Acad. Sci. USA 77, 7342–7346 (1980).

    CAS  Article  Google Scholar 

  37. 37.

    Premakumar, R., Jacobitz, S., Ricke, S. C. & Bishop, P. E. Phenotypic characterization of a tungsten-tolerant mutant of Azotobacter vinelandii. J. Bacteriol. 178, 691–696 (1996).

    CAS  Article  Google Scholar 

  38. 38.

    Christiansen, J., Goodwin, P. J., Lanzilotta, W. N., Seefeldt, L. C. & Dean, D. R. Catalytic and biophysical properties of a nitrogenase apo-MoFe protein produced by a nifB-deletion mutant of Azotobacter vinelandii. Biochemistry 37, 12611–12623 (1998).

    CAS  Article  Google Scholar 

  39. 39.

    Peters, J. W., Fisher, K. & Dean, D. R. Identification of a nitrogenase protein-protein interaction site defined by residues 59 through 67 within the Azotobacter vinelandii Fe protein. J. Biol. Chem. 269, 28076–28083 (1994).

    CAS  PubMed  Google Scholar 

  40. 40.

    Corbin, J. L. Liquid chromatographic-fluorescence determination of ammonia from nitrogenase reactions: a 2-min assay. Appl. Environ. Microbiol. 47, 1027–1030 (1984).

    CAS  PubMed  PubMed Central  Google Scholar 

  41. 41.

    Barney, B. M., Igarashi, R. Y., Dos Santos, P. C., Dean, D. R. & Seefeldt, L. C. Substrate interaction at an iron-sulfur face of the FeMo-cofactor during nitrogenase catalysis. J. Biol. Chem. 279, 53621–53624 (2004).

    CAS  Article  Google Scholar 

  42. 42.

    Whittenbury, R., Phillips, K. C. & Wilkinson, J. F. Enrichment, isolation and some properties of methane-utilizing bacteria. J. Gen. Microbiol. 61, 205–218 (1970).

    CAS  Article  Google Scholar 

  43. 43.

    Yu, Z., Krause, S. M. B., Beck, D. A. C. & Chistoserdova, L. A synthetic ecology perspective: how well does behavior of model organisms in the laboratory predict microbial activities in natural habitats? Front. Microbiol. 7, 946 (2016).

    PubMed  PubMed Central  Google Scholar 

  44. 44.

    Fu, Y., Beck, D. A. C. & Lidstrom, M. E. Difference in C3-C4 metabolism underlies tradeoff between growth rate and biomass yield in Methylobacterium extorquens AM1. BMC Microbiol. 16, 156 (2016).

    Article  Google Scholar 

Download references


We thank the entire Biological Electron Transfer and Catalysis (BETCy) team for informative discussions. We also thank G. Roberts, Y. Zhang, J. McKinlay and F. Daldal for the generous gifts of R. rubrum and R. capsulatus strains, S. Shaw for assistance with activity assays and M. Tokmina-Lukaszewska for the verification of purified Fe-only nitrogenase by mass spectrometry. This work was supported as part of the BETCy Energy Frontier Research Center (EFRC), an EFRC funded by the US Department of Energy, Office of Science Grant DE-SC0012518.

Author information




Y.Z., D.F.H., Z.Y, K.R.F., E.S.B., M.E.L., L.C.S. and C.S.H. designed the research. Y.Z. performed the in vivo studies of photosynthetic bacteria. Y.Z. and K.R.F. made the R. palustris mutants. Y.Z. grew the R. palustris cells. D.F.H. purified the Fe-only nitrogenase and did the enzyme assays. Z.Y. and Y.F. performed the co-culture experiments. Y.F. carried out the analysis of 13C-labeled metabolites. S.P. completed the taxonomic distribution of nitrogenases; R.N.L. performed the CH4 measurement of A. vinelandii. Z.-Y.Y. performed the GC–MS analysis of CH4. Y.Z., D.F.H., K.R.F., E.S.B., M.E.L., L.C.S. and C.S.H. analysed the data. Y.Z., E.S.B., M.E.L., L.C.S. and C.S.H. wrote the paper. All authors contributed to the revision of the manuscript.

Corresponding author

Correspondence to Caroline S. Harwood.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Supplementary Figures 1–4, Supplementary Tables 1 and 2.

Life Sciences Reporting Summary

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Zheng, Y., Harris, D.F., Yu, Z. et al. A pathway for biological methane production using bacterial iron-only nitrogenase. Nat Microbiol 3, 281–286 (2018). https://doi.org/10.1038/s41564-017-0091-5

Download citation

Further reading


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing