Influence of sulfide on diazotrophic growth of the methanogen Methanococcus maripaludis and its implications for the origin of nitrogenase

Communications Biology volume 6, Article number: 799 (2023) Cite this article

634 Accesses

8 Altmetric

Metrics details

Methanogens inhabit euxinic (sulfide-rich) or ferruginous (iron-rich) environments that promote the precipitation of transition metals as metal sulfides, such as pyrite, reducing metal or sulfur availability. Such environments have been common throughout Earth’s history raising the question as to how anaerobes obtain(ed) these elements for the synthesis of enzyme cofactors. Here, we show a methanogen can synthesize molybdenum nitrogenase metallocofactors from pyrite as the source of iron and sulfur, enabling nitrogen fixation. Pyrite-grown, nitrogen-fixing cells grow faster and require 25-fold less molybdenum than cells grown under euxinic conditions. Growth yields are 3 to 8 times higher in cultures grown under ferruginous relative to euxinic conditions. Physiological, transcriptomic, and geochemical data indicate these observations are due to sulfide-promoted metal limitation, in particular molybdenum. These findings suggest that molybdenum nitrogenase may have originated in a ferruginous environment that titrated sulfide to form pyrite, facilitating the availability of sufficient iron, sulfur, and molybdenum for cofactor biosynthesis.

Nitrogen (N) is essential for the synthesis of nucleic and amino acids and other key biomolecules in all forms of life. Earth’s largest reservoir of N is dinitrogen (N2) gas in the atmosphere; however, it is not bioavailable and must be fixed to nitrate (NO3-) or ammonia (NH3) prior to its assimilation. As such, the availability of fixed N often limits the productivity of ecosystems1. On early Earth, fixed N is thought to have been supplied by abiotic processes such as lightning-based oxidation of atmospheric N2 or mineral reduction of N22,3. However, fixed N from these sources would have been minimal and finite, and together these features are thought to have limited ecosystem productivity during this time1. Today, approximately 50% of all fixed N is generated through the biological process of N2 fixation1,4, whereby N2 is reduced to NH3 by the enzyme nitrogenase (diazotrophy) with the remaining fixed N largely generated through the industrial Haber-Bosch process.

Three different forms of nitrogenase have been described to date and these are differentiated by the (hetero)metals comprising the active site of each enzyme complex5. This includes molybdenum (Mo), vanadium (V), and iron (Fe)-only forms of nitrogenase6,7. Mo-nitrogenase (Nif) is taxonomically the most widely distributed and oldest form of nitrogenase8,9 and, at a minimum, consists of the structural proteins NifHDK and maturase proteins, NifB(E)N10. The active site metallocluster of Nif comprises a six iron (Fe) and nine sulfur (S) atom core with Fe symmetrically coordinating a central carbon atom; the metallocluster is capped by Fe and Mo atoms [7Fe-Mo-9S] (termed the FeMo-cofactor10,11). In addition to FeMo-cofactor(s), Nif requires the complex P-cluster that consists of eight Fe atoms and seven S atoms [8Fe-7S]12. A FeMo-cofactor is housed within each NifD structural protein, whereas the P-cluster is found at the interface of each NifD and NifK, ultimately forming the heterotetramer, NifD2K213. Dinitrogenase reductase, NifH, modulates the ATP-dependent transfer of electrons to NifD2K2 and it harbors an additional [4Fe-4S] cluster for each of two NifH subunits14,15. Thus, cells performing N2-fixation via Nif have a high demand for Fe, S, Mo, ATP, and reducing equivalents.

Phylogenetic analyses of a concatenation of NifHDK proteins indicate that the earliest evolving lineages of Nif are found in hydrogenotrophic methanogens6,7,8,9,16,17. These observations are corroborated by other data indicating NifHDK evolved from a series of duplications of genes encoding an ancestor of CfbCD8, proteins that are required to synthesize cofactor F43018,19. That F430 (and CfbCD encoding genes) are exclusively found in archaeal methanogens (and archaeal alkanotrophs)20 is further evidence indicating an origin for Nif among ancestors of these Archaea. These data have been used to suggest an origin for Nif among an ancestor of anaerobic methanogens during the mid-Paleoproterozoic ~1.8–2.1 billion years ago (Ga)6,9,17, although isotopic data of organic matter preserved in shales dated to >3 Ga suggest an even earlier origin21,22. Regardless of the actual date of its origin, nitrogenase is interpreted to be an ancient enzyme that originated in an anoxic environment on early Earth. An anoxic origin for this enzyme is consistent with the oxygen-sensitivity of the metal clusters required for Nif function23. Nif then diversified among anaerobes and only late in its evolutionary history was it acquired via horizontal gene transfer among organisms capable of integrating oxygen (O2) into their energy metabolism or capable of producing O2 in the case of Cyanobacteria. Expanded biological productivity associated with the proliferation of Cyanobacteria and O2 production would have increased demand for existing abiotic pools of fixed N24, which may have been the selective pressure to evolve a biological mechanism to reduce atmospheric N2 and relieve N limitation7,25.

Ferruginous conditions (anoxic and ferrous iron (Fe(II))-rich) likely dominated anoxic environments during the Archean (>2.4 Ga)26,27. This is due to the circulation of hydrothermal fluids through iron-rich mid-ocean basalts, which led to input of a greater amount of Fe(II) into anoxic ocean basin waters than sulfide (HS-)28. In the absence of oxygen, the excess Fe(II) would have been stable, and free Fe(II) concentrations are estimated to have ranged from 0.05 to 0.5 mM29. However, the proliferation of Cyanobacteria and the production of O2 during the late Archean drove the oxidative weathering of continental sulfide minerals that increased the flux of sulfate into oceans. When combined with increased productivity near ocean margins30,31, this would have stimulated heterotrophic sulfate reduction and HS- production. In turn, this led to stratified coastal oceans that were oxygenated at the surface and were euxinic (anoxic and HS--rich) at depth30,31,32. In contrast, deeper ocean waters and those more distal from continental margins remained ferruginous26, due to lower productivity in the overlying water column, decreased availability of sulfate and subsequent heterotrophic sulfate reduction, and hydrothermal input of Fe26,33,34. HS- has a high affinity for Fe(II), resulting in the formation of low-solubility iron-sulfide minerals, including pyrite (FeS2)35,36,37. As such, in euxinic environments, concentrations of HS- exceed that of Fe(II), resulting in the titration and precipitation of Fe(II) as FeS2. Under these conditions, excess HS- is also available to complex with other thiophilic metals (e.g., Mo, Co, Ni), potentially rendering them less bioavailable33,34,38.

How might N2-fixing methanogens have met their concurrent demands for Fe, S, and Mo for the synthesis of Nif cofactors during the Paleoproterozoic or even earlier, when one or more of these elements were likely to be of limited availability due to the tendency for metal sulfide formation? Potential clues come from recent studies of contemporary methanogens, specifically Methanococcus voltae and Methanosarcina barkeri, that reveal their ability to reductively dissolve FeS2 and utilize dissolution products to meet Fe and S nutritional demands39,40,41. M. voltae and M. barkeri grown with FeS2 had similar growth rates and cell yields when compared to traditional sources of Fe and S used to grow methanogens (Fe(II) and HS- and/or cysteine)39,40. However, these experiments were conducted on non-N2-fixing methanogen cells, which would be expected to have a lower demand for Fe, S, and Mo than those actively fixing N2, in particular those growing with Mo-nitrogenase. Such observations point to ferruginous conditions as being possibly more favorable than euxinic conditions for the origin and early proliferation of Nif.

Here, we sought to evaluate the effect of euxinic and ferruginous conditions on the growth and activity of the methanogen, Methanococcus maripaludis S2 (MmS2), to provide new insights into the environment type most conducive for the origin of Nif. Cells were first grown with Fe(II) and HS- or FeS2 as a primary Fe and S source under N2-fixing or NH3-amended conditions, to establish whether FeS2 could serve as a Fe and S source for biosynthesis of nitrogenase and to determine whether HS- drives Mo limitation. Cultivation assays were then focused on FeS2-grown cells, since this condition allows for excess Fe(II) or HS- to be included in the cultivation medium thereby permitting an evaluation of the effect of ferruginous or euxinic conditions, respectively, on metal availability and N2 fixation activity. Nif is the only nitrogenase encoded in MmS242, mitigating the confounding variable associated with cells switching to more recently evolved and alternative forms of nitrogenase (i.e., Fe-nitrogenase Anf; V-nitrogenase, Vnf9) if, or when, Mo becomes limiting5,43. Further, the Nif encoded by MmS2 belongs to the earliest evolving lineage of nitrogenase6,8,9,16. To place additional Mo demands on MmS2, cells were grown with formate, requiring the expression of the molybdopterin-containing formate dehydrogenase44,45 and Mo- or tungsten (W)-dependent formylmethanofuran dehydrogenase (Fmd and Fwd, respectively)46,47. Results are discussed as they relate to the potential environmental condition (i.e., euxinic or ferruginous) that would have allowed the O2-sensitive Nif enzyme to evolve and function in a methanogen, and ultimately other anaerobes, in the anoxic habitats of early and present-day Earth.

MmS2 grew in both N2-fixing and non-N2-fixing (NH3-amended) growth conditions when synthetic nanoparticulate FeS2 or Fe(II)/HS- were provided as the sole Fe/S sources and with formate as the electron donor and methanogenesis substrate (Fig. 1a–d). Growth did not occur in cultures that were not provided with an Fe/S source, regardless of whether cells were provided with N2 or NH3. Likewise, cultures grown in the absence of NH3 and under a headspace of argon (Ar) did not grow, regardless of the Fe/S source provided (Fig. 1a, c). Importantly, MmS2 was also able to grow under N2-fixing conditions with specimen (63–150 µm grain size) FeS2, indicating that a natural (not laboratory-synthesized) form of FeS2 provides Fe and S during N2 fixation (Fig. 1a, b). Cell counts and CH4 measurements were performed infrequently in cultures with specimen FeS2 since preliminary experiments showed they were sensitive to mechanical disturbance. This may be related to the different geometries of the synthetic nanoparticulate FeS2 that is composed of small framboids with high surface area compared to the large and flat facies of the mostly cubic specimen FeS2, which could affect the ability for cells to attach (Supplementary Fig. 1). Nonetheless, these data indicate that both synthetic and natural forms of FeS2 can serve as the sole Fe/S source in N2-fixing MmS2 cells.

Growth and activity were monitored by quantifying cell density (a, b) and methane production (c, d) over time, respectively. Panels a and b and panels c and d are plotted using the same y-axes, respectively. The data presented are the mean and standard deviation of three biological replicates per condition. ammonia NH3, argon Ar, dinitrogen N2, ferrous iron Fe(II), methane CH4, pyrite FeS2, sulfide HS-.

The total number of cells produced in the NH3-amended cultures was higher relative to N2-fixing cultures, regardless of the source of Fe/S provided (Fig. 1a, b). Despite differences in final cell densities, cultures grown under all conditions (except for negative controls) produced similar amounts of CH4 by 54 h incubation (Fig. 1c, d). As such, the cell yield (cells per mmol CH4) in N2-fixing cultures was significantly (p < 0.01) lower than in NH3-amended cultures (Supplementary Fig. 2). This finding is consistent with a nearly three-fold decrease in the cell yield of the methanogen, Methanothermobacter thermolithotrophicus, when grown under N2-fixing conditions relative to NH3-amended conditions48. The source of Fe and S provided to MmS2 cultures also had a marked effect on the rate of cell production (Supplementary Table 1) and the total number of cells produced for N2-fixing conditions but had minimal effect on NH3-amended cultures (Fig. 1a, b). Specifically, N2-fixing cultures provided with FeS2 had a nearly three-fold higher (p < 0.01) cell yield than those provided with Fe(II)/HS- (Supplementary Fig. 2). In contrast to past reports that showed NH3-amended cultures of M. voltae and M. barkeri (both strains MS and Fusaro) achieved similar cell yields when provided with FeS2 or Fe(II)/HS-40,41, FeS2-grown MmS2 had a yield that was 29% lower than Fe(II)/HS--grown cells provided with NH3. However, the reduction in cell yield was even more dramatic when N2-fixing MmS2 cells were compared, with those provided with Fe(II)/HS- exhibiting a 92% reduction in yield relative to those provided NH3. This demonstrates the energetic burden N2 fixation places on MmS2 cells and the dependency of this on the Fe/S source provided.

To better understand differences in the growth kinetics and yields observed in cultures of MmS2 grown under N2-fixing or NH3-amended conditions with FeS2 or Fe(II)/HS- as the sole Fe/S source, the size of cells during the log phase in each growth condition was determined using field-emission scanning electron microscopy (FEM). The average size of MmS2 cells was found to be significantly different (p < 0.05) for each growth condition, where NH3-amended cells grown with Fe(II)/HS- were the largest followed by N2-fixing cells grown with Fe(II)/HS-. The NH3-amended cells grown with FeS2 were smaller yet and N2-fixing cells grown with FeS2 were the smallest in size among all treatments (Supplementary Table 1).

Cell size is suggested to be positively correlated with cell growth rate (cell divisions hr−1) and/or nutrient availability, although the regulatory mechanisms for cell size are poorly understood across all domains of life49,50. The “growth law” suggests cell size is positively correlated with growth rate50; however, this fails to explain the patterns observed here. Diazotrophic MmS2 cells grown on Fe(II)/HS- had the second largest size but the lowest of the observed growth rates across all four conditions (Supplementary Table 1). Rather, these results indicate that growth on FeS2 imparts a greater reduction in cell size than N2 fixation, and that these effects are semi-additive. This is consistent with recent data that suggest that cell size is more of a function of nutrient availability than growth rate alone49, with nutrient-limited conditions leading to smaller cell sizes possibly due to the physiological advantage of increased surface area to volume ratios for nutrient transport across the membrane. For example, the diazotrophic marine cyanobacterium, Crocosphaera watsonii, was demonstrated to undergo a ~2.2-fold reduction in cell volume in response to Fe limitation51. Similarly, M. voltae cells showed a ~3.2-fold reduction in cell volume when grown on FeS2 relative to Fe(II)/HS-39. Proteomic analysis of M. voltae cells revealed an up-expression of iron uptake proteins (Feo) when grown with FeS2, indicating cells may incorrectly sense Fe(II) limitation due to assimilation of Fe(II) complexed with sulfide (FeS(aq)), despite FeS2-grown cells having higher Fe content39. In comparison, N limitation has also been shown to lead to decreased cell size, with ~1.5-to-2.5-fold reduction in cell volume in heterotrophic bacterioplankton isolates grown under N-limiting conditions52,53. The semi-additive effect of Fe, S, and N sources and availability on MmS2 cell size and growth rate suggest that MmS2 cells are responding to real or perceived limitations of one or more of these elements.

The isotopic composition of C in biomass and CH4 has been shown to vary with growth rate, substrate availability, and environmental conditions in a variety of methanogen cultures54,55,56,57,58. To further evaluate the effect of N availability and Fe and S source on the phenotype of MmS2, and to confirm N2 fixation activity, stable isotopes of C (δ13C) and N (δ15N) in CH4 and/or biomass were examined in cultures grown under N2-fixing or NH3-amended conditions with either FeS2 or Fe(II)/HS- provided as the sole Fe/S source.

MmS2 cell biomass had distinctly different δ15N values when grown under N2-fixing vs. NH3-amended conditions (Fig. 2a). Cells grown under N2-fixing conditions had similar δ15N values of −3.22 ± 0.14 and −3.16 ± 0.06‰ when provided with Fe(II)/HS- and FeS2, respectively. These slight differences in δ15N values are not statistically significant (p = 0.55) and are similar to δ15N values of −4‰ previously reported for N2-fixing Methanocaldococcus and Methanothermococcus strains isolated from hydrothermal vents59. NH3-amended MmS2 cells grown with Fe(II)/HS- and FeS2 had nearly identical δ15N values of −21.63 ± 0.23 and −21.61 ± 0.47‰, respectively. Assimilation of NH3 by microorganisms such as E. coli is known to produce biomass with δ15N values ranging from −16 to −21‰ under NH3 replete conditions60. The NH4Cl salt used to cultivate MmS2 had a δ15N value of -3.06‰ (Fig. 2a); as such, the fractionation of δ15N in biomass would be closer to ~ −18.6‰ and therefore is within the range of expected values for cells assimilating NH3.

Mid-log-phase cells (biomass) were analyzed for their carbon-to-nitrogen (C/N) mass ratios and δ15N values (a). Biomass and methane were also analyzed for δ13C values (b). The legend in a is the same for b. The data presented are the mean and standard deviation of three replicate cultures for all conditions. Isotope results are presented in the delta notation as per mille (‰) values vs. Air (δ15N) or VPDB (δ13C). See Supplementary Data 1 for additional data. Ammonia NH3, dinitrogen N2, ferrous iron Fe(II), methane CH4, pyrite FeS2, sulfide HS- .

The δ13C values for MmS2 biomass were similar for all growth conditions tested (−48 to −50‰), with the exception of biomass grown under N2-fixing conditions with Fe(II)/HS- (−46‰; Fig. 2b). Previous studies have interpreted decreased fractionation of biomass C isotopes in methanogens to indicate lower turnover of C substrates54, and this is consistent with the slow growth rate and low cell densities observed in the N2-fixing condition with Fe(II)/HS-. Nonetheless, all biomass δ13C values are within the range of values of biomass (range −30 to −49‰) previously determined for a variety of methanogens when grown with various electron donors/carbon sources across a range of cultivation conditions54. The δ13C values of CH4 were overall similar (−58 to −64‰) with the exception of N2-fixing cells provided with Fe(II)/HS-, which was heavier (−58‰) but not significantly different (p = 0.07) than that of N2-fixing cells grown with FeS2 (−60‰) (Fig. 2b). Like the δ13C biomass, the δ13C of CH4 is within the range of values (−50 to −100‰) measured for methanogen cultures when grown with a variety of electron donors/methanogenesis substrates across a range of cultivation conditions54,55,61. Together, these results suggest that N2 fixation impacts the carbon and energy metabolism of MmS2, and this effect is greater when N2-fixing cells are provided with Fe(II)/HS- when compared to FeS2. This could be due to competition for molybdate between formate dehydrogenases, nitrogenases, and potentially formylmethanofuran dehydrogenases, which may be amplified by HS--promoted decreases in the availability of Mo (discussed more below). This would be expected to decrease the rate of formate oxidation and CO2 production, allowing for the passive exchange of intracellular and extracellular CO2 to have a larger influence on the isotopic composition of CH4.

In addition to investigating C and N isotopes, the masses of C and N and C/N ratios were determined for biomass samples (Fig. 2a). The C/N ratios were similar across all conditions (range of 3.4 to 3.6), except N2-fixing MmS2 cells grown on Fe(II)/HS- that had a C/N ratio of 4.3. Previous studies have shown that C/N ratios in cultures of heterotrophic marine bacterioplankton increase by ~40% when limited to fixed N52. Furthermore, the C/N ratio of N2-fixing Nostoc cultures increased by 80% under MoO42- limitation62. The elevated C/N ratio of N2-fixing MmS2 cells provided with Fe(II)/HS- (Fig. 2a), when considered in light of the low cell yields in this condition (Supplementary Fig. 2), points to N and/or MoO42- limitation in these cells. Importantly, differences in the autotrophic pathways used by methanogens (Wood-Ljungdahl) and oxygenic phototrophs (Calvin Cycle) may influence the partitioning of carbon into biomass or CH4 and may influence the C/N ratios in biomass that are being compared here.

Shotgun transcriptomics was used to generate gene expression profiles from log-phase cultures of N2-fixing and NH3-amended cells provided with FeS2 or Fe(II)/HS- as the sole Fe/S source (see Supplementary Table 2 for growth data). Differences in gene expression were then used to begin to identify the metabolic/physiological processes that were impacted by the N, Fe, and S sources that were provided. Overall, 1737 unique transcripts were detected among the 1742 protein-coding genes encoded by MmS2 (99.7% of genes had detectable transcripts). Principal component analysis of gene expression profiles revealed that samples clustered according to cultivation conditions (Supplementary Fig. 3). A variety of cellular processes (215 genes) were significantly (p < 0.05) differentially regulated between N2-fixing and NH3-amended cells grown with either Fe(II)/HS- or FeS2 (Fig. 3a). The majority of the genes differentially expressed between N2-fixing and NH3-amended cells had similar log2-fold-change values for either Fe(II)/HS- or FeS2 conditions (i.e., these genes fall on a 1:1 line in Fig. 3a), further emphasizing that N2 fixation induces a significant physiological response regardless of the Fe and S source. Of these genes, those involved in N2 fixation, molybdenum and iron transport, and metal binding were significantly upregulated under N2-fixing conditions regardless of Fe and S source. Conversely, ribosomal genes, genes involved in cellular biosynthesis (e.g., acetyl-CoA synthase), and genome replication (e.g., DNA polymerase) were repressed under N2-fixing conditions. These results are consistent with decreased growth rates and yields observed for N2-fixing cells when compared to NH3-amended cells, regardless of the Fe and S source provided (Fig. 1).

a Transcripts that were significantly differentially expressed (p < 0.05, Wald test) for nitrogen-fixing relative to ammonia-amended cells when grown on ferrous iron and sulfide (x-axis) or pyrite (y-axis) shown on a log2-fold-change scale. Positive values indicate higher expression under nitrogen-fixing conditions, and negative values indicate higher expression under ammonia-amended conditions. A gray dashed 1:1 line is provided to ease comparisons based on iron and sulfur source. Mean transcript expression is shown for select genes related to nitrogen fixation (b) and molybdenum transport (c) that are organized in the order they appear in the genome. Each gene is represented by a color-coded point (a) or arrow (b, c) with locus tags designated in parentheses (“MMP_RS” has been truncated from the locus designation to conserve space). Genes are colored according to function in the legend to the right of each panel. Each point represents the mean normalized expression of three culture replicates per growth condition. Statistically significant (p < 0.05, Wald test) differences in expression for nitrogen-fixing vs. ammonia-amended conditions with ferrous iron and sulfide (a) or pyrite (b) or between nitrogen-fixing conditions grown with either iron and sulfur source (c) are shown above each gene for panels b and c. ferrous iron Fe(II), pyrite FeS2, sulfide HS-, dinitrogen N2, ammonia NH3, no significance, n.s.

Core nif genes are encoded in a single operon in MmS263 that is transcriptionally regulated by the repressor protein NrpR and 2-oxoglutarate, a signal of N limitation64,65,66. The nif transcriptional regulator, NrpR, was not differentially transcribed based on the N source, which is consistent with previous reports that the activity of this repressor is independent of its levels in the cell (Fig. 3b)67. Genes encoding the core structural components of Nif (nifHDK), maturases (nifENX), and regulatory proteins (nifI1, nifI2) were upregulated in N2-fixing MmS2 cells relative to NH3-amended cells and patterns of their expression were generally similar regardless of whether cells were provided with Fe(II)/HS- or FeS2 (Fig. 3b). A single nif gene, nifB, falls outside of the nif operon and was found to be significantly differentially expressed between N2-fixing conditions, with Fe(II)/HS--grown cells having slightly lower levels of expression than FeS2-grown cells. NifB is responsible for the assembly of the FeMo-cofactor precursor, NifB-co, which is then transferred to NifEN where final cofactor maturation takes place through the addition of Mo and homocitrate68. It is not clear why the expression of NifB was lower in N2-fixing conditions in Fe(II)/HS--grown cells. In addition to nif genes, expression of glutamine synthase (glnA) was elevated in N2-fixing cells relative to NH3-amended cells, but this difference was only significant for FeS2-grown cells. GlnA is an important enzyme for NH3 assimilation in MmS264, and the higher expression in N2-fixing FeS2-grown cells may have contributed to their lower C/N ratios. Collectively, the transcriptomic profiles demonstrate that MmS2 upregulates the expression of genes encoding Nif maturase, Nif structural, and NH3 assimilation (i.e., GlnA) proteins when fixing N2 and expression of these genes is generally similar between the Fe/S sources.

Nif is the only nitrogenase encoded by MmS2 and this enzyme is Mo-dependent42. In addition, MmS2 encodes a Mo-dependent formate dehydrogenase (Fdh)44,45 and both Mo- and W-dependent forms of formylmethanofuran dehydrogenase (Fmd and Fwd, respectively) that are required for methanogenesis with formate46,47. Expression of fdh was overall similar among the conditions, with only one subunit encoding gene (fdhB; MMP_RS00800) that was significantly upregulated in NH3-amended conditions relative to N2-fixing conditions when grown with Fe(II)/HS- (Supplementary Table 3). Interestingly, transcription of fwd and fmd differed among growth conditions, with three out of seven fwd genes (MMP_RS06415-MMP_RS06425) being significantly upregulated under NH3-amended conditions relative to N2-fixing conditions in Fe(II)/HS- grown cells; expression of fwd genes was similar in NH3-amended conditions relative to N2-fixing conditions in FeS2-grown cells. In the case of fmd, the expression of an entire cluster of five fmd genes (MMP_RS02690-MMP_RS02710) was significantly upregulated (log2 fold-change of ~1.5) in N2-fixing cells relative to NH3-amended conditions when grown with FeS2; aside from a single gene fmdE (MMP_RS02690), fmd transcripts did not differ significantly in N2-fixing cells relative to NH3-amended cells when grown with Fe(II)/HS- (Supplementary Table 3). It is unclear why fwd was up-expressed under NH3-amended conditions with Fe(II)/HS- but it could be due to the very high affinity of MoO42- for HS- and its low solubility69. This, in turn, would increase the bioavailability of tungstate (WO42-) relative to MoO42-, that when combined with the higher growth yields in the NH3-amended conditions with Fe(II)/HS-, leads to up-expression of fwd transcripts. Similarly, the increased transcription of fmd genes under N2-fixing conditions relative to NH3-amended conditions on FeS2 could indicate Mo is more available in FeS2 growth conditions, with an expression of fwd transcripts being unaffected.

Members of both Archaea and Bacteria acquire Mo in the form of MoO42- using MoO42--specific ABC transporters called Mod70. Transcripts for three mod-related gene clusters encoded by MmS2 were compared in N2-fixing and NH3-amended cells grown with FeS2 or Fe(II)/HS- (Fig. 3c). Expression of genes comprising all three mod gene clusters was upregulated in N2-fixing cells relative to NH3-amended cells regardless of Fe/S source. Furthermore, expression of genes comprising two of the three mod gene clusters (MMP_RS01135-MMP_RS01145 and MMP_RS02670-MMP_RS02685) and an annotated NifC-like transporter (MMP_RS08475) were significantly upregulated (p < 0.05) in N2-fixing cells grown with Fe(II)/HS- relative to those grown with FeS2. These results indicate that N2 fixation increases the expression of putative MoO42- transporters, and this expression is higher in N2-fixing cells grown with Fe(II)/HS- when compared to those grown with FeS2. Collectively, these data suggest that MoO42- transport is influenced by not only demands associated with Mo-dependent enzymes (Fdh, Fmd, Nif) but also by Fe/S source. More specifically, the results suggest Mo limitation in N2-fixing cells grown with Fe(II)/HS-, possibly due to HS- complexing and/or reacting with Mo38,71.

A previous study reported that HS- at a concentration of >6 mM inhibits N2 fixation in stream sediments72. While this effect was attributed to the toxicity of HS- in cells responsible for N2 fixation, it is plausible that the added HS- influenced the availability of trace metals (e.g., Fe, Mo) required for N2-fixing cells62,73,74,75. A similar effect of HS- on metal availability in N2-fixing MmS2 cultures may explain the lower cell yields and increased expression of Mod genes observed in Fe(II)/HS--grown cells relative to those grown with FeS2. MoO42- is a thiophilic molecule that readily reacts with HS- to form soluble thiomolybdate (MoO4-nSn2-) species that can ultimately be incorporated in sulfide minerals, such as FeS271,76. Such thiolation reactions would be expected to occur under the sulfidic (2 mM) cultivation conditions typically used to culture methanogens77 and that were utilized herein for the Fe(II)/HS--grown cultures. These euxinic cultivation conditions potentially limited the availability of MoO42- such that it did not meet cellular demands.

To begin to examine this hypothesis, the speciation of Mo was tracked in abiotic reactors containing anoxic, base salts medium without added NH3 or metals. Reactors were amended with various combinations of HS- (2 mM), Fe(II) (25 µM), or FeS2 (2 mM) to mimic cultivation conditions used in experiments up to this point. To these, Mo was added as MoO42- and the concentration of MoO42- and its conversion to MoO4-nSn2- were monitored via colorimetric assays78 and UV–Vis spectroscopy71, respectively. Experiments were performed with 10 µM of MoO42- (as opposed to 4 µM used to cultivate MmS2 herein) to increase signal sensitivity such that it was within the dynamic range of the assays that were used.

MoO42- concentrations remained at ~8–10 µM in reactors not amended with HS- in the presence or absence of FeS2 (Supplementary Fig. 4a). As expected, the product of complete thiolation of MoO42-, tetrathiomolybdate (MoS42-), was not detected in these conditions throughout the course of the experiment (Supplementary Fig. 4b). In contrast, MoO42- concentrations decreased rapidly (within the first 12 h of incubation) when HS- alone or HS- and Fe(II) were added to reactors (Supplementary Fig. 4a). In reactors containing HS-, MoS42- was rapidly produced (within the first hour of incubation) and MoO42- was completely converted to MoS42- by 72 h incubation. In reactors containing both HS- and Fe(II), the conversion of MoO42- to MoS42- was accelerated, and complete conversion occurred within the first 3 h of incubation (Supplementary Fig. 4b). UV–Vis spectroscopy revealed production of MoO4–nSn2- intermediates in reactors amended with HS- with or without Fe(II) (Supplementary Fig. 5). Ions of MoO42-/thiomolybdate have a strong affinity for Fe, including Fe in iron sulfides79,80, and it has been shown that these ions readily react to form stable Fe-Mo-S cubane-like structures in sulfur-rich environments80. It is possible that the acceleration of MoO42- thiolation by Fe(II) in the experiments described here is similarly attributable to the formation of stable Fe-Mo-S cubane-like clusters. Regardless, these data indicate that MoO42- is readily available in cultures grown with FeS2, while the predominant forms of Mo in cultures grown with Fe(II)/HS- are a mixture of MoS42- and MoO4-nSn2-. This could potentially explain poor growth and up-expression of mod genes in N2-fixing cultures provided with Fe(II)/HS- relative to FeS2. It is unknown if N2-fixing MmS2 cells in the Fe(II)/HS- conditions are using sub-micromolar levels of MoO42- that are below the detection limit (~1 µM) of the colorimetric assay used herein or if they can use MoO4-nSn2- (albeit to a limited extent based on growth data) to meet their Mo nutritional demands.

To further test the hypothesis that euxinic conditions (i.e., excess HS-) limit the availability of MoO42- in MmS2 cultures, the growth of N2-fixing cells provided with Fe(II)/HS- was compared to N2-fixing cells provided with FeS2 in the presence or absence of 2 mM added HS-. An NH3-amended MmS2 culture provided with FeS2 and 2 mM added HS- was also tested. As was observed previously, MmS2 grown with FeS2 exhibited faster growth and CH4 production kinetics and generated higher cell yields than cultures provided with Fe(II)/HS- when fixing N2 (Fig. 4a, b). In comparison, FeS2-grown cultures with 2 mM added HS-, regardless of N source, exhibited significantly slower rates of cell and CH4 production, demonstrating that HS- decreases the growth of MmS2 in cells provided with FeS2. While both N2-fixing and NH3-amended cultures provided with FeS2 and 2 mM added HS- conditions grew poorly, N2-fixing cells were more negatively impacted.

Two mM HS- was added to nitrogen-fixing or ammonia-amended cultures provided with pyrite as the primary iron and sulfur source to test its effect on the production of cells (a) and methane (b). The effect of different molybdate concentrations on the growth rates (c) and yields (d) of cells grown with either pyrite or ferrous iron and sulfide as the sole iron and sulfur source under nitrogen-fixing conditions are shown. Maximal growth rates (calculated from measured cell densities) were determined over a 5-day period for each replicate. The data presented are the mean and standard deviation of three biological replicates per condition. Ammonia NH3, dinitrogen N2, ferrous iron Fe(II), methane CH4, pyrite FeS2, sulfide HS-.

If HS- negatively impacts cellular growth under N2-fixing conditions through its effect on MoO42-, it would be expected that cells grown on FeS2 would have lower requirements for MoO42- in solution than Fe(II)/HS--grown cells since relatively small amounts of HS- are released into solution during FeS2 reduction. Previous experiments performed with H2-grown, N2-fixing MmS2 cells provided with Fe(II)/HS- found that >400 nM MoO42- was required for growth with an optimum of 4000 nM42. These values are much higher than those required of other N2-fixing organisms. For example, aerobic, N2-fixing cyanobacteria such as Anabaena variabilis81, as well as anaerobic N2-fixing purple sulfur bacteria (PSB) inhabiting the interface of oxic/euxinic waters and that serves as a modern analog of the productive continental margins of Proterozoic oceans82, can be supported by less than 10 nM MoO42-. Interestingly, maximum N2 fixation for the PSB was maximal above the chemocline where HS- was near zero82. Thus, it seems incongruous that methanogens would require 40-fold more MoO42- than these organisms, despite ancestors of methanogens likely being where Nif originated6,9.

To further test this, formate-grown and N2-fixing MmS2 cells were provided with a range of MoO42- concentrations with Fe(II)/HS- or FeS2 as the sole Fe/S source. Consistent with previous work42, MmS2 grown on Fe(II)/HS- showed little to no growth response in conditions with <100 nM MoO42- but growth rates increased substantially when ≥500 nM MoO42- was provided (Fig. 4c). Cell yields did not reach maximum levels in Fe(II)/HS--grown cells until >1000 nM MoO42- was provided (Fig. 4d). Strikingly, cells provided with FeS2 grew well under all Mo concentrations tested, including when 0 µM MoO42- was added (Fig. 4c). After this experiment, abiotic reactors containing media and FeS2 (no MoO42- added) were analyzed by ICP-MS and were determined to have background (contaminant) Mo ranging from 10 to 30 nM despite using ACS-grade chemicals and acid-washed glassware. Thus, Fe(II)/HS--grown N2-fixing cells required >500 nM MoO42- to achieve near-optimal growth rates and yields whereas FeS2-grown N2-fixing cells required <30 nM MoO42-.

Collectively, these results suggest that MmS2 cells grown with FeS2 (in the absence of high concentrations of free HS-) can access MoO42- at nM concentrations that are >10-fold lower than previously reported for this strain42. Further, MmS2 cells can achieve optimal growth rates and yields at MoO42- concentrations that are at levels ~100-fold lower than previously reported42. Despite similar growth rates for the FeS2 conditions at different MoO42- concentrations, there was a positive trend in the cell yield with increasing MoO42- that indicates higher MoO42- concentrations are still beneficial to these cells (Fig. 4d). These findings are consistent with other studies that have shown N2 fixation at low Mo concentrations with cells grown under non-sulfidic conditions81,82. Notably, the low levels (10-30 nM) of MoO42- shown to support N2 fixation herein are similar to the inferred Mo concentrations of Proterozoic oceans83,84,85). Other studies investigating Mo requirements for other methanogen species fixing N2 found 1 to 10 µM MoO42- to be the optimal concentrations86,87. There is one other example of methanogens fixing N2 at low (<10 nM) MoO42- concentrations; Nishizawa et al. showed that isolates of Methanocaldococcus and Methanothermococcus from a hydrothermal vent could fix N2 with as low as 5 nM or 1 µM MoO42-, respectively59. This indicates that different methanogen species within the same hydrothermal environment (at different temperatures) can have very different Mo requirements. Importantly, these past experiments were all performed in the presence of >1 mM HS-, necessitating a re-evaluation of methanogens’, and other anaerobes’, ability to access MoO42- in the absence of high HS-. These data support the hypothesis that N2 fixation via Nif is more efficient in low HS- conditions where Mo is more bioavailable.

Excess HS- repressed growth of MmS2 when grown with FeS2 (euxinic conditions), regardless of the N source provided (Fig. 4). Next, the growth of N2-fixing cells was compared in FeS2-grown cultures with medium formulations designed to mimic mildly (100 µM added Fe(II)) and highly (1000 µM added Fe(II)) ferruginous conditions and to mimic mildly (100 µM added HS-) and highly (1000 µM added HS-) euxinic conditions. Control cultures contained FeS2 as the sole source of Fe and S. Cell densities and CH4 concentrations were monitored as growth proxies, and HS- was measured as a proxy for FeS2 reduction.

MmS2 cells grew better under the mildly ferruginous condition relative to mildly and highly euxinic conditions, although the highly ferruginous condition caused a longer lag phase (Fig. 5a). Growth of MmS2 cells under euxinic conditions, especially with 1000 µM added HS-, was greatly diminished relative to cells grown with FeS2 as the control. Despite the differences in the overall densities of cells at the end of the experiment, all conditions produced similar amounts of CH4 (Fig. 5b). This led to pronounced differences in final cell yields (Fig. 5d), as well as yields during log phase (Supplementary Fig. 6). In particular, cell yields for the mildly ferruginous condition were the highest even relative to the control condition (FeS2 only). Again, yields of cells grown in euxinic conditions were significantly lower than those grown in either ferruginous conditions or in the control (FeS2 only) condition. Together, these data suggest that mildly ferruginous conditions are preferred for N2-fixing MmS2 cells.

Production of cells (a), methane, CH4 (b), and HS- (c) was monitored in cultures of MmS2 provided with 2 mM pyrite (FeS2) as the sole Fe/S source, with FeS2 and 100 or 1000 µM of Fe(II), or with FeS2 and 100 or 1000 µM HS-. Cell yields (d) in each growth condition were calculated using the initial and maximum cell and CH4 concentrations for each replicate. The legend at the bottom of the figure applies to all four panels. Statistically significant (p < 0.05, t-test) differences were determined pairwise in panel d and are depicted above each bar using the symbols from the legend to denote the significantly different pairs. The data presented are the mean and standard deviation of three biological replicates per condition. Ferrous iron Fe(II), methane CH4, pyrite FeS2, sulfide HS-.

Total sulfide (HS-, H2S, and acid labile iron-sulfide) was measured in non-filtered culture samples as an indicator of FeS2 reduction41. Despite all conditions requiring the mobilization of FeS2 to access either Fe (euxinic conditions), S (ferruginous conditions), or both Fe and S (control condition), the production of total sulfide differed among conditions (Fig. 5c). For FeS2 alone (control conditions), 34 µmol total sulfide was observed by the end of the experiment, while the mildly euxinic condition produced 46 µmol total sulfide. Accounting for the added HS- (100 µM, 7.5 µmol), 38 µmol of total sulfide was produced minus what was assimilated by the cells in mildly euxinic conditions. Further, this indicates a similar amount of cellular FeS2 reduction took place in these two growth conditions; however, more FeS2 was reduced per cell in the mildly euxinic condition given low cell production. Similarly, in the highly (1000 µM, 75 µmol added HS-) euxinic condition, 148 µmol total sulfide was measured at the end of the experiment, indicating that more FeS2 was reduced on a per cell basis than control conditions considering low cell production in the former condition (Fig. 5a). Total sulfide production in cultures grown in ferruginous conditions was far less than control cultures (FeS2-only). Specifically, mildly (100 µM added Fe(II)) ferruginous conditions produced only 4 µmol total sulfide, while total sulfide remained undetectable in highly (1000 µM added Fe(II)) ferruginous conditions until the final time point where 0.3 µmol was detected. Thus, the amount of FeS2 that must be reductively dissolved to meet biosynthetic demands for Fe or S is lower in ferruginous conditions than in euxinic conditions, which may also help to explain better growth under these conditions.

The impact that HS- had on FeS2-grown cells is likely attributable to the complex conditions created when HS-, FeS2, and trace metals (i.e., Mo, Fe(II)) co-occur. High HS- concentrations decrease the thermodynamic favorability of FeS2 reduction41, which might make FeS2 reduction the rate-limiting process in these cultures and possibly explain the slower production of CH4 observed in FeS2-grown cultures with added HS- (Figs. 4b and 5b). In addition, high HS- may reduce the metal availability of not only MoO42-, but also Fe(II) that is released during FeS2 reduction. Spietz et al. describe a model for reductive dissolution of FeS2 by methanogens wherein FeS2 reduction releases HS- into solution and pyrrhotite (Fe1-xS) is precipitated as a secondary phase on the mineral surface41. Dissolution of Fe1-xS results in the release of Fe(II) (but not HS-) that can then react with soluble HS- to form FeS(aq) clusters, the presumed source of Fe/S for these cells39,41. In abiotic experiments with Fe1-xS sequestered in 100 kDa dialysis membranes, it was shown that dissolution and/or diffusion of Fe from Fe1–xS to the bulk medium was significantly decreased in the presence of 500 µM HS- when compared to no added HS-41. This may indicate that HS- increases the rate of FeS(aq) nucleation and re-precipitation as mackinawite (FeSmack) nanoparticles that are too large to diffuse across the cell membrane. Such particles would be expected to limit Fe (and S) availability to cells41. These effects of HS- are presumed to not occur under ferruginous conditions because Fe(II) effectively scrubs HS- or because cells minimize FeS2 reduction such that all HS- released from the process is assimilated. It is also important to consider the tendency for trace metals to become adsorbed onto FeS2 surfaces that, in turn, reduces their concentration in solution. This is the case for MoS42-, the product of complete MoO42- thiolation (see above), which more strongly adsorbs to FeS2 surfaces than MoO42-80.

Experiments conducted herein were aimed at identifying the most plausible conditions (euxinic or ferruginous) that would have enabled N2 fixation via Nif during the early Proterozoic, when phylogenetic data indicate this enzyme evolved6,7,8, or even in the late Archean, when isotopic data suggest this enzyme or its predecessor evolved21,22. Experiments were conducted with the marine methanogen, MmS2, which harbors a Nif homolog that belongs to the deepest branching Nif lineage6,8,9. Data showed that euxinic conditions (excess of HS- relative to free Fe(II)) negatively affected the growth of N2-fixing MmS2 cells, regardless of whether they were grown with Fe(II)/HS- or FeS2. This was shown to be due to the effect of HS- on the speciation and availability of Mo. However, under ferruginous conditions, excess Fe(II) titrates HS- leading to its precipitation as FeSmack and FeS2 which, in turn, allows Mo to remain bioavailable as MoO42-. Indeed, the growth rate and yield of N2-fixing MmS2 cells were significantly higher when cells were grown with FeS2 and excess Fe(II) (ferruginous conditions) than when grown with i) FeS2 and excess HS- (euxinic conditions), ii) FeS2 alone, or iii) under canonical conditions typically used to culture methanogens, excess HS- and Fe(II) (also a euxinic condition).

Collectively, these data indicate that MmS2 prefers FeS2 over Fe(II)/HS- during growth with formate as methanogenesis substrate and with N2 as the sole N source, conditions that increase the demand of Mo via Fdh, Nif, and Fmd. Additionally, cells grew more efficiently with FeS2 when MoO42- was ~100-fold lower in concentration than cells grown with Fe(II)/HS-. These data suggest that a ferruginous environment would have favored the function and potentially the origin of Nif on early Earth. In contrast, euxinic conditions pose significant challenges to access thiophilic metals required for N2-fixing cells in both ancient and modern environments, thus limiting N2 fixation via Nif. Since metal availability is a major driver of the evolution of metalloenzymes (reviewed in83), we suggest that Nif likely evolved in an environment that favored the availability of Fe and Mo.

Synthetic FeS2 was synthesized in an anaerobic chamber by mixing separate solutions of 16.7 g FeSO4 · 7 H2O and 17.28 g Na2S · 9 H2O each dissolved in 50 mL of anoxic, deionized MilliQ H2O (MQ) in a 500 mL media bottle. 2.1 g elemental sulfur was then added to the reactor, which was then sealed with a rubber stopper and removed from the anaerobic chamber. The reactor was then incubated for four days at 65 °C followed by another four days at 85 °C. The reactor was then cooled to room temperature and was opened in a chemical fume hood where the resulting mineral was collected into centrifuge tubes and washed several times aerobically with MQ, 1 M HCl, 6 M HCl, and acetone. Finally, the mineral was washed in sterile, anoxic MQ several times within an anaerobic chamber and transferred into a sterile glass serum bottle as a slurry40. Dried specimen FeS2 was crushed and sieved (63–150 µm particle size) before it was subjected to the same wash series as the synthetic FeS2. The specimen FeS2 was then dried under a stream of N2 and stored in a sterile glass serum bottle until it was weighed in an anaerobic chamber and then added directly to culture bottles. Synthetic FeS2 was added to prepared culture bottles as a slurry. Minerals were characterized by X-ray diffraction prior to their use in experiments (Supplementary Fig. 7).

M. maripaludis strain S2 (MmS2), kindly provided by Dr. Kyle Costa, was grown in Fe- and S-free basal medium that contained (g L−1): NaCl, 21.98; MgCl2 · 6H2O, 5.10; NaHCO3, 5.00; K2HPO4, 0.14; KCl, 0.33; CaCl2 · 2H2O, 0.10. For NH3-amended cultures, NH4Cl was added to a final concentration of 0.50 g L−1. The basal medium was amended with 25 µM FeCl2 · 4H2O and 2 mM Na2S · 9H2O for Fe(II)/HS--grown cells. Sulfide addition experiments tested the effect of 2 mM HS- during growth on FeS2. Iron and HS- were added from anoxic, autoclaved stock solutions 30 min. prior to inoculation. For FeS2 growth conditions, the basal medium was amended with synthetic FeS2 to a final concentration of 2 mM Fe (0.018 g per 75 mL of media) or with 1.5 g of specimen FeS2. The higher amount of specimen FeS2 was added to account for differences in the estimated surface area of in specimen vs. synthetic FeS240. Basal medium was amended (each 1% v/v final concentration) with a trace element, vitamin, formate, and acetate solutions (Supplementary Table 4). The vitamin solutions were filter-sterilized (0.22 µm) whereas the trace element, formate, and acetate solutions were sterilized individually by autoclaving. All solutions were sparged with filtered (0.22 µm) N2. N2 and other gases used in this study were passed over a heated column (300 °C) containing H2-reduced copper shavings.

To prepare the growth medium, all components, except for NaHCO3, were dissolved in MQ H2O and then boiled for less than 1 min. After boiling, the bottle containing the medium was removed from the heat and sealed with a butyl rubber stopper with a metal cannula and vent needle and sparged with N2 gas for 1 h per liter of medium. After degassing, the bottle was sealed, brought into an anaerobic chamber, and NaHCO3 was added after cooling. After adding these components, the pH of the medium was adjusted to 7.00 using anoxic 2 M HCl or 1 M NaOH. Seventy-five milliliters of media were dispensed into 165 mL serum bottles that were then sealed with new and freshly boiled 20 mm blue butyl rubber (Bellco, Vineland, NJ) stoppers and aluminum crimp caps. The headspace of the medium bottles was purged with 80:20 N2:CO2 for 20 min. The bottles were then autoclaved for 20 min at 123 °C. After autoclaving, FeS2, Fe(II), HS-, trace elements, vitamins, and organics were added to the concentrations indicated above.

MmS2 was maintained by twice-weekly transfers in a defined medium with Fe(II)/HS-, N2 gas, and formate. Cultures were inoculated on a 5% v/v basis with cells grown under N2-fixing conditions and that were provided with Fe(II)/HS-. All cells were washed in basal, NH3-free medium (no additions) by centrifuging at 4696 x g for 20 min at 20 °C in a swing-out bucket rotor for all experiments. The density of an aliquot of the washed cells was then immediately enumerated before they were used as inoculum. Cultures were pressurized to 2.87 atm with either 80:20 N2:CO2 or Ar:CO2. All cultures were incubated statically on their sides at 38 °C in the dark.

Headspace gas from microcosms was sampled with an N2-flushed syringe and stopcock and diluted with ultra-high purity N2 into CaliBond bags (Calibrated Instruments Inc., Manhasset, NY) prior to CH4 determination. Liquid samples were taken on the benchtop using an N2-flushed syringe and needle. CH4 was quantified by gas chromatography and total sulfide was quantified colorimetrically via the methylene blue assay, as described previously40. Cells for direct counting were first fixed for four hrs in 2.5% v/v (final concentration) glutaraldehyde at 4 °C and were then counted using a Petroff-Hausser counting chamber on a Nikon YS100 light microscope with a 100x oil objective lens (Nikon, Tokyo, Japan). Fixed cell samples were concentrated by centrifugation at 15,000 x g for 15 min and resuspended in the basal medium of a sufficient volume to facilitate cell counting

To test the stability of MoO42- in various medium formulations used herein, 75 mL of NH3-free base salts medium containing trace elements without MoO42-, formate, acetate, or vitamins were added to 165 mL serum bottles. Medium preparation was as described above for cultures of MmS2. The headspace composition was 80:20 N2:CO2. The medium was amended with Na2MoO4 · 4H2O (10 µM final concentration), Na2S · 9H2O (2 mM), FeCl2· 4H2O (25 µM), or synthetic FeS2 (2 mM). Reactors were incubated at 38 °C in the dark and were transferred to an anaerobic chamber (97.5% N2, 2.5% H2) and subsamples for spectroscopy were collected via syringe and needle and placed into disposable plastic cuvettes (1 cm path length) that were sealed with plastic cuvette caps before removing from the anaerobic chamber. The concentration of MoO42- was determined colorimetrically using a modified catechol assay78. The assay was modified to increase the concentrations of working solution components four-fold and the sample-to-working solution volume ratio was adjusted to 4:1, with a 1 mL total assay volume that was measured at 400 nm with a Genesys 10 S Vis Spectrophotometer (Thermo Scientific, Waltham, MA). The working solution was prepared and stored in the anaerobic chamber 2-weeks prior to the experiment. For tetrathiomolybdate quantification, 1 mL of the sample was directly measured at 467 nm. UV–Vis scans (300 nm to 525 nm) were performed on a Cary UV–Vis-NIR 6000i spectrophotometer (Agilent Technologies Inc., Santa Clara, CA) using the same samples prepared for tetrathiomolybdate quantification. Sample absorbance values for each assay were compared against a standard curve that was prepared with fresh reagents (Na2MoO4· 4H2O or MoS4(NH4)2).

To characterize the size of cells grown under different conditions, field-emission electron microscopy (FEM) was performed. Five milliliters of cell culture was taken as a subsample from mid-log-phase cultures used in transcriptomics experiments (see below). Cells were fixed for two hrs at room temperature in 2.5% EM-grade glutaraldehyde (Electron Microscopy Sciences, Hatfield, PA) in basal medium. The fixed cells were then filtered onto a previously Au-sputtered 0.2 µm black Isopore polycarbonate filter (MilliporeSigma, Burlington, MA) using gentle vacuum filtration. The cells were then washed by the addition of 10 mL basal medium to the filter. After washing, the cells were subjected to an ethanol dehydration series (by filtration) using molecular grade ethanol (25%, 50%, 70%, 85%, 95%, 100%) and were then stored at 4 °C until FEM was performed at the Imaging and Chemical Analysis Laboratory at Montana State University. Samples were mounted on the FEM holder using double-sided carbon tape and sputtered with a thin film of iridium for conductivity before imaging. FEM was performed using a high-resolution Supra 55VP electron microscope (Zeiss, Thornwood, NY) with a primary electron beam energy of 1 keV at different magnifications. Cell size was determined using ImageJ to manually measure individual cells by recording the length of two normal measurements calibrated to the image scale bar.

Shotgun transcriptomics was performed on cells of MmS2 grown under N2-fixing or NH3-amended conditions with either Fe(II)/HS- or FeS2 as the sole Fe/S source. Fifty milliliters of culture were harvested during the mid-log phase in an anaerobic chamber by centrifugation at 4696 x g for 30 min at 4 °C. The supernatant was then discarded, and the tubes were removed from the anaerobic chamber and immediately flash-frozen in liquid N2. Cell pellets were stored at −80 °C until RNA extraction.

Total RNA was extracted from cell pellets using TRIzol reagent (Invitrogen, Carlsbad, CA) following the manufacturer’s protocol with minor modifications. One milliliter of TRIzol was added to cell pellets and the resuspended cells were transferred to Lysis E tubes (MP Biomedicals, Irvine, CA) on ice. The cell samples were mechanically lysed by three cycles of 40 sec of bead beating and 5 min of rest at room temperature (~20 °C). After the lysis procedure, 200 µL of molecular grade chloroform was added, the tubes were mixed by inversion and allowed to incubate for 3 min at room temperature. The tubes were then centrifuged for 15 min at 12,000 x g at 4 °C and the upper aqueous phase containing RNA was transferred by pipette into a clean 2-mL tube. RNA was precipitated by the addition of 0.5 mL of 100% molecular grade isopropanol that had been pre-chilled to 4 °C followed by 10 min incubation on ice. RNA was then pelleted by centrifugation for 10 min at 12,000 x g at 4 °C. The supernatant was removed by pipette, and the RNA was washed in 1 mL of 75 % molecular-grade ethanol. RNA was pelleted by centrifugation for 5 min at 7500 x g at 4 °C, the supernatant was removed, and the RNA pellet was air dried for 10 min. Once dried, the RNA was resuspended in 50 µL of RNA-free H2O (Fisher Scientific, Waltham, MA) by incubating at 55 °C in a heat block for 10 min. RNA was treated to remove residual DNA by the addition of Turbo DNase (Invitrogen, Waltham, MA) according to the manufacturer’s instructions. The RNA was then subjected to a second round of precipitation, washing, drying, and resuspension steps as described above. RNA was quantified using a Qubit 2.0 fluorometer with a Qubit BR RNA kit (Invitrogen) and the quality was assayed using a NanoDrop ND-1000 spectrometer (Thermo Scientific). DNA contamination was checked by performing PCR amplification using archaeal-specific 16S rRNA primers and ensuring no amplification had occurred by gel electrophoresis. The RNA was then sent to the University of Wisconsin’s Genome Expression Center for quality control, rRNA depletion using custom Methanococcus-specific oligonucleotides (Supplementary Table 5), and sequencing on an Illumina NovaSeq 2×150 bp (Illumina, San Diego, CA).

RNA paired-end reads were processed using default settings in TrimGalore! (https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/), a wrapper that implements CutAdapt88 to remove adapter sequences and FastQC (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/) to filter reads. Reads were aligned to the reference MmS2 genome (ASM1158v1) using BowTie289. Transcriptional profiles were generated by read counting for each locus using HTSeq90, then normalized and analyzed in DESeq291 implemented in R v3.6.0. The RNA-sequencing data reported in this paper have been deposited in the NCBI GEO database (GSE216895).

Duplicate cultures of MmS2 were grown (see above) and pooled to provide sufficient sample mass for isotope analyses of cell biomass and CH4. Cultures were grown with Fe(II)/HS- or FeS2 under N2-fixing or NH3-amended conditions and harvested during the late-log phase of growth. Concentrations of CH4 were determined from each duplicate and cell densities were determined after the duplicates were combined. Cultures were harvested in an anaerobic chamber after reaching > ~3 × 107 cells per mL by centrifugation in 50 mL centrifuge tubes (Globe Scientific, Mahwah, NJ) for 30 min at 4696 x g at 4 °C. The supernatant was removed using a line and needle under low-vacuum and the cells were resuspended in 8 mL of basal medium and transferred to a 15 mL centrifuge tube (Globe Scientific).

To remove bulk FeS2 (if present), the sample was underlaid with a Percoll (GE Healthcare, Chicago, IL) working solution with 0.4 M NaCl according to the manufacturer’s instructions using a syringe and cannula. The samples were then spun at 2000 x g in a swing-out bucket rotor for 20 min at 4 °C. The overlaying aqueous phase containing cells was removed by pipette and transferred to a clean 50 mL centrifuge tube, to which ~40 mL of fresh basal medium was added and the tubes were mixed by vortex. The tubes were then spun for 30 min at 4696 x g at 4 °C to pellet the cells away from any residual Percoll. The supernatant was then removed by vacuum, and the cell pellets were resuspended in 1.5 mL of basal medium and transferred into 2.0 mL screw-top microcentrifuge tubes (Thermo Scientific). A subsample (10 µL) of cells was taken and diluted for cell enumeration after processing. Biomass was concentrated by centrifugation for 20 min at 15,000 x g at 4 °C in a fixed-angle rotor, and the supernatant was removed by pipette.

Cell pellets were frozen at −80 °C until shipment to Utah State University Geosciences for C and N stable isotope analyses. Cell pellets were dried aerobically by incubating for 16 h at 60 °C in a drying oven. Once dried, biomass was scraped out of the tubes, weighed, and further prepared for analysis by grinding to a powder and transferring into tin capsules. Isotope analysis of δ15N (vs. AIR) and δ13C (vs. VPDB) values was then performed by combustion and gas chromatography of the biomass samples at 1000 °C using a Costech 4010 Elemental Analyzer coupled with a Thermo Scientific Delta V Advantage isotope ratio mass spectrometry (GC-IRMS). For δ13C analysis of CH4, gas samples from cultures used in the above isotope experiments were shipped to Northern Arizona University and analyzed on a Picarro G2201-I cavity ring-down spectrometer. A commercial δ13C −60‰ (vs. VPDB) CH4 standard (Airgas, Plumsteadville, PA) served as a calibration standard in addition to methane-free air “zero air” for Picarro analyses.

All data shown are mean values of at least three independent replicates, with error bars showing the standard deviation of the replicates. Independent Student’s t-tests were used to calculate p-values reported herein for all experiments except for transcriptomics experiments, for which p-values were determined by a Wald test calculated in the DEseq2 package implemented in R. Results were considered significantly different when p < 0.05.

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

All data generated or analyzed during this study are included in this published article and its Supplementary Materials. Numerical and replicate data used for all figures are provided in Supplementary Data 1. Sequencing data from transcriptomics experiments are available through the NCBI Gene Expression Omnibus under the accession ID GSE216895. All other data are available from the corresponding author upon reasonable request.

LeBauer, D. S. & Treseder, K. K. Nitrogen limitation of net primary productivity in terrestrial ecosystems is globally distributed. Ecology 89, 371–379 (2008).

Article PubMed Google Scholar

Yung, Y. L. & McElroy, M. B. Fixation of nitrogen in the prebiotic atmosphere. Science 203, 1002–1004 (1979).

Article CAS PubMed Google Scholar

Kasting, J. F. & Walker, J. C. G. Limits on oxygen concentrations in the prebiological atmosphere and the rate of abiotic nitrogen fixation. J. Geophys. Res. 86, 1147–1158 (1981).

Article CAS Google Scholar

Howarth, R. W. Nutrient limitation of net primary production in marine ecosystems. Annu. Rev. Ecol. Evol. Syst. 19, 89–110 (1988).

Article Google Scholar

Joerger, R. D. & Bishop, P. E. Bacterial alternative nitrogen fixation systems. Crit. Rev. Microbiol. 16, 1–14 (1988).

Article CAS PubMed Google Scholar

Boyd, E. S. et al. A late methanogen origin for molybdenum-dependent nitrogenase. Geobiology 9, 221–232 (2011).

Article CAS PubMed Google Scholar

Raymond, J., Siefert, J. L., Staples, C. R. & Blankenship, R. E. The natural history of nitrogen nixation. Mol. Biol. Evol. 21, 541–554 (2004).

Article CAS PubMed Google Scholar

Boyd, E. & Peters, J. New insights into the evolutionary history of biological nitrogen fixation. Front. Microbiol. 4, 201 (2013).

Article PubMed PubMed Central Google Scholar

Boyd, E. S., Hamilton, T. L. & Peters, J. W. An alternative path for the evolution of biological nitrogen fixation. Front. Microbiol. 2, 205 (2011).

Article PubMed PubMed Central Google Scholar

Rubio, L. M. & Ludden, P. W. Biosynthesis of the iron-molybdenum cofactor of nitrogenase. Annu. Rev. Microbiol. 62, 93–111 (2008).

Article CAS PubMed Google Scholar

Lancaster, K. M. et al. X-ray emission spectroscopy evidences a central carbon in the nitrogenase iron-molybdenum cofactor. Science 334, 974–977 (2011).

Article CAS PubMed PubMed Central Google Scholar

Chan, M. K., Kim, J. & Rees, D. C. The nitrogenase FeMo-cofactor and P-cluster pair: 2.2 Å resolution structures. Science 260, 792–794 (1993).

Article CAS PubMed Google Scholar

Seefeldt, L. C., Hoffman, B. M. & Dean, D. R. Mechanism of Mo-dependent nitrogenase. Annu. Rev. Biochem. 78, 701–722 (2009).

Article CAS PubMed PubMed Central Google Scholar

Einsle, O. & Rees, D. C. Structural enzymology of nitrogenase enzymes. Chem. Rev. 120, 4969–5004 (2020).

Article CAS PubMed PubMed Central Google Scholar

Danyal, K., Dean, D. R., Hoffman, B. M. & Seefeldt, L. C. Electron transfer within nitrogenase: evidence for a deficit-spending mechanism. Biochemistry 50, 9255–9263 (2011).

Article CAS PubMed Google Scholar

Boyd, E. S. et al. Evolution of molybdenum nitrogenase during the transition from anaerobic to aerobic metabolism. J. Bacteriol. 197, 1690–1699 (2015).

Article CAS PubMed PubMed Central Google Scholar

Shock, E. L. & Boyd, E. S. Principles of geobiochemistry. Elements 11, 395–401 (2015).

Article Google Scholar

Zheng, K., Ngo, P. D., Owens, V. L., Yang, X. & Mansoorabadi, S. O. The biosynthetic pathway of coenzyme F430 in methanogenic and methanotrophic archaea. Science 354, 339–342 (2016).

Article CAS PubMed Google Scholar

Moore, S. J. et al. Elucidation of the biosynthesis of the methane catalyst coenzyme F430. Nature 543, 78–82 (2017).

Article CAS PubMed PubMed Central Google Scholar

Allen, K. D., Wegener, G. & White, R. H. Discovery of multiple modified F430 coenzymes in methanogens and anaerobic methanotrophic archaea suggests possible new roles for F430 in nature. Appl. Environ. Microbiol. 80, 6403–6412 (2014).

Article PubMed PubMed Central Google Scholar

Stüeken, E. E., Buick, R., Guy, B. M. & Koehler, M. C. Isotopic evidence for biological nitrogen fixation by molybdenum-nitrogenase from 3.2 Gyr. Nature 520, 666–669 (2015).

Article PubMed Google Scholar

Garvin, J., Buick, R., Anbar, A. D., Arnold, G. L. & Kaufman, A. J. Isotopic evidence for an aerobic nitrogen cycle in the latest Archean. Science 323, 1045–1048 (2009).

Article CAS PubMed Google Scholar

Dean, D. R., Bolin, J. T. & Zheng, L. Nitrogenase metalloclusters: structures, organization, and synthesis. J. Bacteriol. 175, 6737–6744 (1993).

Article CAS PubMed PubMed Central Google Scholar

Canfield, D. E., Glazer, A. N. & Falkowski, P. G. The evolution and future of Earth’s nitrogen cycle. Science 330, 192–196 (2010).

Article CAS PubMed Google Scholar

Mus, F., Colman, D. R., Peters, J. W. & Boyd, E. S. Geobiological feedbacks, oxygen, and the evolution of nitrogenase. Free Radical Biol. Med. 140, 250–259 (2019).

Article CAS Google Scholar

Canfield, D. E. et al. Ferruginous conditions dominated later Neoproterozoic deep-water chemistry. Science 321, 949–952 (2008).

Article CAS PubMed Google Scholar

Poulton, S. W. & Canfield, D. E. Ferruginous conditions: a dominant feature of the ocean through Earth’s history. Elements 7, 107–112 (2011).

Article CAS Google Scholar

Walker, J. C. & Brimblecombe, P. Iron and sulfur in the pre-biologic ocean. Precambrian Res. 28, 205–222 (1985).

Article CAS PubMed Google Scholar

Holland, H. D. The oceans; a possible source of iron in iron-formations. Econ. Geol. 68, 1169–1172 (1973).

Article CAS Google Scholar

Planavsky, N. J. et al. Widespread iron-rich conditions in the mid-Proterozoic ocean. Nature 477, 448–451 (2011).

Article CAS PubMed Google Scholar

Reinhard, C. T., Raiswell, R., Scott, C., Anbar, A. D. & Lyons, T. W. A late Archean sulfidic sea stimulated by early oxidative weathering of the continents. Science 326, 713–716 (2009).

Article CAS PubMed Google Scholar

Fakhraee, M., Hancisse, O., Canfield, D. E., Crowe, S. A. & Katsev, S. Proterozoic seawater sulfate scarcity and the evolution of ocean–atmosphere chemistry. Nat. Geosci. 12, 375–380 (2019).

Article CAS Google Scholar

Arnold, G. L., Anbar, A. D., Barling, J. & Lyons, T. W. Molybdenum isotope evidence for widespread anoxia in mid-Proterozoic oceans. Science 304, 87–90 (2004).

Article CAS PubMed Google Scholar

Anbar, A. D. & Knoll, A. H. Proterozoic ocean chemistry and evolution: a bioinorganic bridge? Science 297, 1137–1142 (2002).

Article CAS PubMed Google Scholar

Rickard, D. & Luther, G. W. Chemistry of iron sulfides. Chem. Rev. 107, 514–562 (2007).

Article CAS PubMed Google Scholar

Berner, R. A. Sedimentary pyrite formation: an update. Geochim. Cosmochim. Acta 48, 605–615 (1984).

Article CAS Google Scholar

Schoonen, M. A. A. & Barnes, H. L. Reactions forming pyrite and marcasite from solution: II. Via FeS precursors below 100 °C. Geochim. Cosmochim. Acta 55, 1505–1514 (1991).

Article CAS Google Scholar

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).

Article CAS Google Scholar

Payne, D. et al. Examining pathways of iron and sulfur acquisition, trafficking, deployment, and storage in mineral-grown methanogen cells. J. Bacteriol. 203, 1–18 (2021).

Article Google Scholar

Payne, D., Spietz, R. L. & Boyd, E. S. Reductive dissolution of pyrite by methanogenic archaea. ISME J. 15, 3498–3507 (2021).

Article CAS PubMed PubMed Central Google Scholar

Spietz, R. L. et al. Investigating abiotic and biotic mechanisms of pyrite reduction. Front. Microbiol. 13, 878387 (2022).

Article PubMed PubMed Central Google Scholar

Kessler, P. S., McLarnan, J. & Leigh, J. A. Nitrogenase phylogeny and the molybdenum dependence of nitrogen fixation in Methanococcus maripaludis. J. Bacteriol. 179, 541–543 (1997).

Article CAS PubMed PubMed Central Google Scholar

Hamilton, T. L. et al. Transcriptional profiling of nitrogen fixation in Azotobacter vinelandii. J. Bacteriol. 193, 4477–4486 (2011).

Article CAS PubMed PubMed Central Google Scholar

May, H. D., Schauer, N. L. & Ferry, J. G. Molybdopterin cofactor from Methanobacterium formicicum formate dehydrogenase. J. Bacteriol. 166, 500–504 (1986).

Article CAS PubMed PubMed Central Google Scholar

Wood, G. E., Haydock, A. K. & Leigh, J. A. Function and regulation of the formate dehydrogenase genes of the methanogenic archaeon Methanococcus maripaludis. J. Bacteriol. 185, 2548–2554 (2003).

Article CAS PubMed PubMed Central Google Scholar

Bertram, P. A., Schmitz, R. A., Linder, D. & Thauer, R. K. Tungstate can substitute for molybdate in sustaining growth of Methanobacterium thermoautotrophicum. Identification and characterization of a tungsten isoenzyme of formylmethanofuran dehydrogenase. Arch Microbiol. 161, 220–228 (1994).

Article CAS PubMed Google Scholar

Schmitz, R. A., Albracht, S. P. & Thauer, R. K. A molybdenum and a tungsten isoenzyme of formylmethanofuran dehydrogenase in the thermophilic archaeon Methanobacterium wolfei. Eur. J. Biochem. 209, 1013–1018 (1992).

Article CAS PubMed Google Scholar

Belay, N., Sparling, R. & Daniels, L. Dinitrogen fixation by a thermophilic methanogenic bacterium. Nature 312, 286–288 (1984).

Article CAS PubMed Google Scholar

Vadia, S. & Levin, P. A. Growth rate and cell size: a re-examination of the growth law. Curr. Opin. Microbiol. 24, 96–103 (2015).

Article PubMed PubMed Central Google Scholar

Schaechter, M., MaalØe, O. & Kjeldgaard, N. O. Dependency on medium and temperature of cell size and chemical composition during balanced growth of Salmonella typhimurium. Microbiology 19, 592–606 (1958).

CAS Google Scholar

Jacq, V., Ridame, C., L'Helguen, S., Kaczmar, F. & Saliot, A. Response of the unicellular diazotrophic cyanobacterium Crocosphaera watsonii to iron limitation. PLoS ONE 9, e86749–e86749 (2014).

Article PubMed PubMed Central Google Scholar

Vrede, K., Heldal, M., Norland, S. & Bratbak, G. Elemental composition (C, N, P) and cell volume of exponentially growing and nutrient-limited bacterioplankton. Appl. Environ. Microbiol. 68, 2965–2971 (2002).

Article CAS PubMed PubMed Central Google Scholar

Mårdén, P., Tunlid, A., Malmcrona-Friberg, K., Odham, G. & Kjelleberg, S. Physiological and morphological changes during short term starvation of marine bacterial isolates. Arch. Microbiol. 142, 326–332 (1985).

Article Google Scholar

House, C. H., Schopf, J. W. & Stetter, K. O. Carbon isotopic fractionation by Archaeans and other thermophilic prokaryotes. Org. Geochem. 34, 345–356 (2003).

Article CAS Google Scholar

Botz, R., Pokojski, H.-D., Schmitt, M. & Thomm, M. Carbon isotope fractionation during bacterial methanogenesis by CO2 reduction. Org. Geochem. 25, 255–262 (1996).

Article CAS Google Scholar

Fuchs, G., Thauer, R., Ziegler, H. & Stichler, W. Carbon isotope fractionation by Methanobacterium thermoautotrophicum. Arch. Microbiol. 120, 135–139 (1979).

Article CAS Google Scholar

Kleikemper, J., Schroth, M. H., Bernasconi, S. M., Brunner, B. & Zeyer, J. Sulfur isotope fractionation during growth of sulfate-reducing bacteria on various carbon sources. Geochim. Cosmochim. Acta 68, 4891–4904 (2004).

Article CAS Google Scholar

Topçuoğlu, B. D., Meydan, C., Nguyen, T. B., Lang, S. Q. & Holden, J. F. Growth kinetics, carbon isotope fractionation, and gene expression in the hyperthermophile Methanocaldococcus jannaschii during hydrogen-limited growth and interspecies hydrogen transfer. Appl. Environ. Microbiol. 85, e00180–00119 (2019).

Article PubMed PubMed Central Google Scholar

Nishizawa, M., Miyazaki, J., Makabe, A., Koba, K. & Takai, K. Physiological and isotopic characteristics of nitrogen fixation by hyperthermophilic methanogens: Key insights into nitrogen anabolism of the microbial communities in Archean hydrothermal systems. Geochim. Cosmochim. Acta 138, 117–135 (2014).

Article CAS Google Scholar

Vo, J., Inwood, W., Hayes, J. M. & Kustu, S. Mechanism for nitrogen isotope fractionation during ammonium assimilation by Escherichia coli K12. Proc. Natl Acad. Sci. USA 110, 8696 (2013).

Article CAS PubMed PubMed Central Google Scholar

Miller, H. M. et al. Large carbon isotope variability during methanogenesis under alkaline conditions. Geochim. Cosmochim. Acta 237, 18–31 (2018).

Article CAS Google Scholar

Glass, J. B., Wolfe-Simon, F., Elser, J. J. & Anbar, A. D. Molybdenum-nitrogen co-limitation in freshwater and coastal heterocystous cyanobacteria. Limnol. Oceanogr. 55, 667–676 (2010).

Article CAS Google Scholar

Kessler, P. S., Blank, C. & Leigh, J. A. The nif gene operon of the methanogenic archaeon Methanococcus maripaludis. J. Bacteriol. 180, 1504–1511 (1998).

Article CAS PubMed PubMed Central Google Scholar

Lie, T. J. & Leigh, J. A. Regulatory response of Methanococcus maripaludis to alanine, an intermediate nitrogen source. J. Bacteriol. 184, 5301–5306 (2002).

Article CAS PubMed PubMed Central Google Scholar

Cohen-Kupiec, R., Marx, C. J. & Leigh, J. A. Function and regulation of glnA in the methanogenic archaeon Methanococcus maripaludis. J. Bacteriol. 181, 256–261 (1999).

Article CAS PubMed PubMed Central Google Scholar

Lie, T. J., Wood, G. E. & Leigh, J. A. Regulation of nif expression in Methanococcus maripaludis: roles of the euryarchaeal repressor NrpR, 2-oxoglutarate, and two operators. J. Biol. Chem. 280, 5236–5241 (2005).

Article CAS PubMed Google Scholar

Lie, T. J. & Leigh, J. A. A novel repressor of nif and glnA expression in the methanogenic archaeon Methanococcus maripaludis. Mol. Microbiol. 47, 235–246 (2003).

Article CAS PubMed Google Scholar

Hu, Y. & Ribbe, M. W. Biosynthesis of the metalloclusters of nitrogenases. Annu. Rev. Biochem. 85, 455–483 (2016).

Article CAS PubMed Google Scholar

Stiefel, E. L. in Metal Ions in Biological Systems. Vol. 39 (eds. Sigel, A. & Sigel, H.) 29 (CRC Press, Boca Raton, FL, 2002).

Zhang, Y., Rump, S. & Gladyshev, V. N. Comparative genomics and evolution of molybdenum utilization. Coord. Chem. Rev. 255, 1206–1217 (2011).

Article CAS PubMed PubMed Central Google Scholar

Erickson, B. E. & Helz, G. R. Molybdenum(VI) speciation in sulfidic waters: Stability and lability of thiomolybdates. Geochim. Cosmochim. Acta 64, 1149–1158 (2000).

Article CAS Google Scholar

Tam, T., Mayfield, C. I., Inniss, W. E. & Knowles, R. Effect of sulfide on nitrogen fixation in a stream sediment-water system. Appl. Environ. Microbiol. 43, 1076–1079 (1982).

Article CAS PubMed PubMed Central Google Scholar

Tuit, C., Waterbury, J. & Ravizza, G. Diel variation of molybdenum and iron in marine diazotrophic cyanobacteria. Limnol. Oceanogr. 49, 978–990 (2004).

Article CAS Google Scholar

Saito, M. A. et al. Iron conservation by reduction of metalloenzyme inventories in the marine diazotroph Crocosphaera watsonii. Proc. Natl Acad. Sci. USA 108, 2184 (2011).

Article CAS PubMed PubMed Central Google Scholar

Naito, T., Sachuronggui, Ueki, M. & Maeda, I. Light-enhanced bioaccumulation of molybdenum by nitrogen-deprived recombinant anoxygenic photosynthetic bacterium Rhodopseudomonas palustris. Biosci. Biotechnol. Biochem. 80, 407–413 (2016).

Article CAS PubMed Google Scholar

Phillips, R. & Xu, J. A critical review of molybdenum sequestration mechanisms under euxinic conditions: Implications for the precision of molybdenum paleoredox proxies. Earth Sci. Rev. 221, 103799 (2021).

Article CAS Google Scholar

Liu, Y., Beer, L. L. & Whitman, W. B. Methanogens: a window into ancient sulfur metabolism. Trends Microbiol. 20, 251–258 (2012).

Article CAS PubMed Google Scholar

Seifter, S. & Novic, B. Colorimetric determination of molybdate with catechol. Anal. Chem. 23, 188–189 (1951).

Article CAS Google Scholar

Xu, N., Christodoulatos, C. & Braida, W. Adsorption of molybdate and tetrathiomolybdate onto pyrite and goethite: effect of pH and competitive anions. Chemosphere 62, 1726–1735 (2006).

Article CAS PubMed Google Scholar

Bostick, B. C., Fendorf, S. & Helz, G. R. Differential adsorption of molybdate and tetrathiomolybdate on pyrite (FeS2). Environ. Sci. Technol. 37, 285–291 (2003).

Article CAS PubMed Google Scholar

Zerkle, A. L., House, C. H., Cox, R. P. & Canfield, D. E. Metal limitation of cyanobacterial N2 fixation and implications for the Precambrian nitrogen cycle. Geobiology 4, 285–297 (2006).

Article CAS Google Scholar

Philippi, M. et al. Purple sulfur bacteria fix N2 via molybdenum-nitrogenase in a low molybdenum Proterozoic ocean analogue. Nat. Commun. 12, 4774 (2021).

Article CAS PubMed PubMed Central Google Scholar

Anbar, A. D. Elements and evolution. Science 322, 1481–1483 (2008).

Article CAS PubMed Google Scholar

Scott, C. et al. Tracing the stepwise oxygenation of the Proterozoic ocean. Nature 452, 456–459 (2008).

Article CAS PubMed Google Scholar

Collier, R. W. Molybdenum in the Northeast Pacific Ocean. Limnol. Oceanogr. 30, 1351–1354 (1985).

Article CAS Google Scholar

Scherer, P. Vanadium and molybdenum requirement for the fixation of molecular nitrogen by two Methanosarcina strains. Arch. Microbiol. 151, 44–48 (1988).

Article Google Scholar

Maslać, N., Sidhu, C., Teeling, H. & Wagner, T. Comparative transcriptomics sheds light on remodeling of gene expression during diazotrophy in the thermophilic methanogen Methanothermococcus thermolithotrophicus. mBio 13, e02443–02422 (2022).

Article PubMed PubMed Central Google Scholar

Martin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. Embnet J. 17, 3 (2011).

Article Google Scholar

Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).

Article CAS PubMed PubMed Central Google Scholar

Anders, S., Pyl, P. T. & Huber, W. HTSeq−a Python framework to work with high-throughput sequencing data. Bioinformatics 31, 166–169 (2015).

Article CAS PubMed Google Scholar

Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).

Article PubMed PubMed Central Google Scholar

Download references

This work was supported by the Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences of the U.S. Department of Energy through Grant DE-SC0020246 to R.L.S. and E.S.B.

Department of Microbiology and Cell Biology, Montana State University, Bozeman, MT, 59717, USA

Devon Payne, Rachel L. Spietz & Eric S. Boyd

Department of Geosciences, Utah State University, Logan, UT, 84322, USA

Dennis L. Newell

Center for Ecosystem Science and Society and Department of Biological Sciences, Northern Arizona University, Flagstaff, AZ, 86011, USA

Paul Dijkstra

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

D.P.: contributed to the experimental design, conducting experiments, and writing of the manuscript. R.L.S.: contributed to the experimental design, conducting experiments, and writing of the manuscript. D.L.N.: contributed to the experimental design, conducting experiments, and writing of the manuscript. P.D.: contributed to the experimental design, conducting experiments, and writing of the manuscript. E.S.B.: contributed to the experimental design and writing of the manuscript.

Correspondence to Eric S. Boyd.

The authors declare no competing interests.

Communications Biology thanks Masaru Nobu, Tom Jilbert, and the other anonymous reviewer for their contribution to the peer review of this work. Primary handling editors: Anna Heintz-Buschart and David Favero.

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

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

Payne, D., Spietz, R.L., Newell, D.L. et al. Influence of sulfide on diazotrophic growth of the methanogen Methanococcus maripaludis and its implications for the origin of nitrogenase. Commun Biol 6, 799 (2023). https://doi.org/10.1038/s42003-023-05163-9

Download citation

Received: 10 December 2022

Accepted: 21 July 2023

Published: 31 July 2023

DOI: https://doi.org/10.1038/s42003-023-05163-9

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.