Nitrogenase is the only enzyme in nature that can split the extraordinarily stable triple bond of atmospheric nitrogen (N₂) and convert it into ammonia (NH₃), a form that living cells can use to build proteins, DNA, and other essential molecules. This reaction, called biological nitrogen fixation, supplies roughly half of the reactive nitrogen entering Earth’s ecosystems each year, with global biological fixation contributing hundreds of teragrams annually across land and ocean environments.1PubMed Central. The global nitrogen cycle in the twenty-first century Despite decades of study, how the enzyme manages this feat under mild biological conditions remains one of the deeper puzzles in biochemistry, and the details of its structure and mechanism continue to surprise researchers.
What Nitrogenase Actually Does
At its core, nitrogenase performs a deceptively simple-sounding job: it takes nitrogen gas from the air and adds hydrogen to it, producing ammonia. But N₂ is one of the most inert molecules around. Its two nitrogen atoms are held together by a triple bond with a dissociation energy so high that industrial chemistry requires temperatures above 400°C and pressures of 150–300 atmospheres to break it (the Haber-Bosch process). Nitrogenase accomplishes the same transformation at room temperature and atmospheric pressure, inside living cells. The trade-off is energy: the enzyme burns through large amounts of ATP, the cell’s energy currency, to drive each round of electron transfer needed for the reaction.
The overall reaction requires the delivery of eight electrons and eight protons to reduce one molecule of N₂ into two molecules of NH₃, with one molecule of H₂ released as an obligatory byproduct.2PubMed Central. Nitrite and Hydroxylamine as Nitrogenase Substrates: Mechanistic Implications for the Pathway of N 2 Reduction That mandatory hydrogen release is not wasted motion; it appears to be mechanistically linked to how N₂ binds at the active site. Each electron transfer event consumes two molecules of ATP, meaning the full reaction uses at least 16 ATP molecules per N₂ reduced. That makes nitrogen fixation one of the most energetically expensive biochemical processes a cell carries out, which is why organisms tightly regulate when and whether they bother producing nitrogenase at all.
The Two-Component Protein System
Nitrogenase is not a single protein but a two-component system. The smaller component, called the Fe protein (or dinitrogenase reductase), is a dimer of two identical subunits that share a single iron-sulfur cluster containing four iron and four sulfur atoms. This cluster sits on the protein’s surface, exposed between the two subunits, and is coordinated by two cysteine amino acids from each subunit.3PubMed. Crystallographic structure of the nitrogenase iron protein from Azotobacter vinelandii The Fe protein’s job is to act as an electron shuttle: it picks up an electron from the cell’s metabolism, docks with the larger component, delivers the electron, and then undocks to reload.
The larger component, the MoFe protein (or dinitrogenase), is where nitrogen actually gets reduced. It is a tetramer made of two pairs of subunits, and each pair houses two remarkable metal clusters. The first is the P-cluster, an unusual iron-sulfur cluster containing eight iron atoms and seven sulfur atoms. The P-cluster acts as an intermediate electron relay, accepting electrons from the Fe protein and passing them along to the active site.4PubMed. Redox-dependent structural changes in the nitrogenase P-cluster The P-cluster changes shape when it gains or loses electrons: two of its iron atoms shift position, and different amino acid side chains swing in or out to coordinate the metal, essentially acting as a conformational gate that couples electron flow to proton delivery.5PubMed Central. Tyrosine-Coordinated P-Cluster in G. diazotrophicus Nitrogenase: Evidence for the Importance of O-Based Ligands in Conformationally Gated Electron Transfer
The FeMo-Cofactor and Its Hidden Carbon
The real star of the show is the second metal cluster in the MoFe protein: the iron-molybdenum cofactor, or FeMo-co. This is where N₂ binds and gets reduced. FeMo-co is one of the most complex metal clusters known in biology, containing one molybdenum atom, seven iron atoms, nine sulfur atoms, a molecule of homocitrate, and, buried in its center, a single carbon atom.
That central carbon was a mystery for years. When high-resolution X-ray structures first revealed electron density inside the iron cage of FeMo-co, researchers could tell something was there but could not distinguish whether it was nitrogen, oxygen, or carbon.6PubMed. Testing if the interstitial atom, X, of the nitrogenase molybdenum-iron cofactor is N or C: ENDOR, ESEEM, and DFT studies of the S = 3/2 resting state in multiple environments The question was not settled until 2011, when atomic-resolution X-ray data combined with spectroscopic analysis provided direct evidence that the interstitial atom is carbon, likely present as a carbide ion (C⁴⁻).7PubMed Central. Evidence for interstitial carbon in nitrogenase FeMo cofactor This carbide sits at the geometric center of the cluster, bridging all six of the inner iron atoms in a trigonal prismatic arrangement. Chemists trying to recreate FeMo-co in the lab have found that replicating this central six-iron-one-carbon core is one of the hardest parts.8PubMed Central. Synthetic models of the nitrogenase FeMo cofactor
Building FeMo-co inside a cell is no less remarkable. A minimal set of proteins and small molecules can assemble the cofactor in a test tube: the process requires the scaffold protein NifEN, the radical SAM enzyme NifB (which produces an iron-sulfur-carbon precursor), and the Fe protein NifH, along with iron, sulfide, molybdate, homocitrate, and ATP.9PubMed Central. In vitro synthesis of the iron-molybdenum cofactor of nitrogenase from iron, sulfur, molybdenum, and homocitrate using purified proteins NifB produces an eight-iron precursor cluster already containing the interstitial carbon, which then gets further modified on the NifEN scaffold before being inserted into the MoFe protein.10PubMed Central. Nitrogenase cofactor biosynthesis using proteins produced in mitochondria of Saccharomyces cerevisiae
How the Catalytic Cycle Works
Every time nitrogenase reduces one molecule of N₂, the Fe protein must dock and undock with the MoFe protein eight separate times, each time delivering one electron. Each docking event follows a precise four-step sequence. First, the Fe protein, loaded with ATP and an electron, binds to the MoFe protein, and a shape change triggered by ATP binding opens a gate that allows the electron to flow to FeMo-co. This conformationally gated electron transfer is fast, occurring at about 140 times per second at 25°C. Second, ATP is hydrolyzed (split), at about 70 times per second. Third, phosphate is released more slowly, at about 16 per second. Finally, the spent Fe protein, now holding ADP instead of ATP, detaches from the MoFe protein at about 6 per second, which is the slowest step and the bottleneck for the entire cycle.11PubMed Central. Electron transfer precedes ATP hydrolysis during nitrogenase catalysis
A key insight from this ordering is that ATP hydrolysis does not directly power the electron transfer. Instead, the energy of ATP binding and the resulting conformational changes drive the electron across. Hydrolysis happens afterward and serves mainly to weaken the protein-protein complex so the Fe protein can detach and reload. This is a subtler energy-coupling strategy than was originally assumed: the enzyme uses the shape of the ATP-bound state to open an electron gateway, then uses hydrolysis as a kind of release mechanism. An updated kinetic model refines this picture further, treating electron transfer as a probabilistic event whose likelihood depends on what is bound at the active site at any given moment.12ACS Catalysis. Toward a Unified Kinetic Model of Nitrogenase Catalysis
The obligatory release of H₂ during N₂ reduction is one of nitrogenase’s most distinctive quirks. N₂ cannot bind to FeMo-co until the cofactor has accumulated enough electrons and protons. At a critical stage, a reductive elimination step ejects H₂ from the active site, and this event is what creates the open coordination site where N₂ latches on. If you flood the system with H₂ gas, this step reverses and N₂ gets pushed off, which is why H₂ acts as a competitive inhibitor of nitrogen fixation.2PubMed Central. Nitrite and Hydroxylamine as Nitrogenase Substrates: Mechanistic Implications for the Pathway of N 2 Reduction
Beyond Nitrogen: Unexpected Substrates
Nitrogenase is promiscuous. Although its biological purpose is nitrogen fixation, the enzyme also reduces a surprising range of other small molecules, including acetylene, nitrous oxide, cyanide, and azide. More recently, researchers discovered that nitrogenase can reduce carbon monoxide and carbon dioxide to produce hydrocarbons and other carbon-containing products.13PubMed Central. Nitrogenase reduction of carbon-containing compounds This carbon-reducing ability has attracted attention well beyond traditional nitrogen-fixation research, because it suggests nitrogenase-like chemistry could one day be harnessed for converting CO₂ into useful feedstock chemicals.
The iron-only nitrogenase (discussed below) is particularly good at this. In the photosynthetic bacterium Rhodobacter capsulatus, the iron nitrogenase is almost three times more efficient at reducing CO₂ than the conventional molybdenum nitrogenase, and it is less selective for N₂, meaning CO₂ competes effectively as a substrate. Under light-driven conditions, this enzyme facilitates the accumulation of formate and methane outside the cell, products that other microbes or industrial processes can use.14PubMed Central. The iron nitrogenase reduces carbon dioxide to formate and methane under physiological conditions: A route to feedstock chemicals This is still basic research, but the implication that a biological enzyme can convert a greenhouse gas into one-carbon building blocks under ambient conditions is hard to ignore.
Why Oxygen Is the Enemy
Nitrogenase is exquisitely sensitive to oxygen. Both the Fe protein and the MoFe protein lose activity rapidly when exposed to O₂, because oxygen reacts irreversibly with the iron-sulfur clusters. This creates a fundamental conflict for nitrogen-fixing organisms that also need oxygen for energy production (aerobic respiration), and especially for photosynthetic organisms that actively generate oxygen as a waste product.
Different organisms have evolved radically different solutions. Some anaerobic bacteria simply avoid the problem by living where there is no oxygen. Free-living aerobic bacteria like Azotobacter employ a strategy called respiratory protection, ramping up their oxygen consumption rate so dramatically that they burn through O₂ at the cell surface before it can reach nitrogenase in the interior.15PubMed. Respiratory protection of nitrogenase in Azotobacter species: is a widely held hypothesis unequivocally supported by experimental evidence? In the symbiotic relationship between legume plants and rhizobia bacteria, the plant root nodules produce a protein called leghemoglobin, which works much like hemoglobin in blood. Leghemoglobin binds oxygen and delivers it to the bacteria at a precisely controlled trickle, high enough to sustain respiration but low enough to protect nitrogenase.16Journal of Biological Chemistry. Facilitated oxygen diffusion. The role of leghemoglobin in nitrogen fixation by bacteroids isolated from soybean root nodules
Cyanobacteria face the hardest version of this dilemma because they photosynthesize, splitting water to produce O₂ inside the very cells that need to fix nitrogen. Some cyanobacteria solve this by separating the two processes in time (fixing nitrogen only at night) or in space. The most elaborate spatial solution involves heterocysts, specialized thick-walled cells that shut down photosynthesis entirely and devote themselves to nitrogen fixation, receiving sugars from their photosynthetic neighbors in exchange for ammonia.17PubMed Central. Oxygen relations of nitrogen fixation in cyanobacteria
Vanadium and Iron-Only Alternatives
The molybdenum-dependent nitrogenase is the best studied and most widespread, but it is not the only version. Some organisms carry genes for alternative nitrogenases that use vanadium or iron alone in place of molybdenum at the active site. These alternatives tend to be expressed when molybdenum is scarce in the environment, functioning as biological backup systems.
The vanadium nitrogenase replaces molybdenum with vanadium in its active-site cofactor (called FeVco). A high-resolution crystal structure revealed that FeVco is broadly similar to FeMo-co but with one striking difference: one of the bridging sulfide ions in FeMo-co is replaced by a larger ligand, most likely a carbonate ion. This swap pushes two iron atoms slightly further apart and alters the electronic properties of the cluster.18PubMed Central. The structure of vanadium nitrogenase reveals an unusual bridging ligand Computational modeling supports carbonate as the best-fitting identity for this bridging ligand and suggests that the resting electronic state of FeVco already resembles a partially reduced state of FeMo-co, which may help explain why the vanadium enzyme has different substrate preferences.19PubMed Central. Quantum Mechanics/Molecular Mechanics Study of Resting-State Vanadium Nitrogenase: Molecular and Electronic Structure of the Iron–Vanadium Cofactor
The iron-only nitrogenase, which uses an all-iron cofactor (FeFe-co), is the least characterized of the three. Its reductase component, AnfH, adopts the same overall fold as the Fe proteins of the other two systems but shows limited cross-reactivity: it can work with the vanadium nitrogenase to some extent but cannot substitute for the Fe protein of molybdenum nitrogenase.20PubMed. Structural analysis of the reductase component AnfH of iron-only nitrogenase from Azotobacter vinelandii This selectivity suggests that despite their shared architecture, the three nitrogenase systems have diverged enough at the protein-protein interface to prevent free interchangeability.
How Cells Decide When to Fix Nitrogen
Because nitrogen fixation is so expensive, cells control nitrogenase production tightly. The regulatory logic, best understood in the bacterium Klebsiella pneumoniae, centers on a transcriptional activator protein called NifA, which switches on the genes encoding nitrogenase components. NifA activity is kept in check by a partner protein called NifL, a sensor that responds to two key environmental signals: oxygen and ammonium. When oxygen is too high or when the cell already has plenty of fixed nitrogen (ammonium), NifL blocks NifA, and nitrogenase genes stay off.21The Journal of Nutrition. Regulation of Biological Nitrogen Fixation This makes metabolic sense: why spend 16 ATP per nitrogen molecule if usable nitrogen is already available, or if oxygen would just destroy the enzyme anyway?
When conditions do favor fixation, gene expression unfolds in a specific temporal pattern. The activator gene nifA is expressed first in a brief transient pulse, followed by longer-duration waves of expression for the biosynthetic and structural genes that encode the enzyme components themselves.22PubMed Central. Kinetics of Nif gene expression in a nitrogen-fixing bacterium This staggered timing ensures the regulatory machinery is in place before the cell commits resources to building the full enzyme complex.
Engineering Nitrogenase Into Crops
One of the long-standing goals of agricultural biotechnology is to transfer nitrogenase activity into crop plants so they could fix their own nitrogen instead of relying on synthetic fertilizer. This would be transformative: industrial nitrogen fixation via the Haber-Bosch process consumes roughly 1–2% of global energy production and contributes significantly to greenhouse gas emissions. A self-fertilizing cereal crop could reduce both fertilizer costs and environmental damage.
The challenge is formidable. Nitrogenase requires at least a dozen gene products for assembly and function, all of which must be expressed, properly folded, and loaded with their metal clusters inside a compartment that excludes oxygen. Plant mitochondria are a promising target because they have low internal oxygen concentrations and abundant ATP, but researchers discovered a frustrating bottleneck: the NifD protein, which forms part of the MoFe protein’s backbone, was rapidly degraded inside yeast and plant mitochondria. The culprit turned out to be mitochondrial processing peptidases, enzymes that normally trim targeting signals off imported proteins. These peptidases were inadvertently cutting NifD at a conserved site. By identifying the specific residue responsible for susceptibility and engineering variants that resist cleavage without losing enzymatic activity, researchers opened a path toward stable expression of nitrogenase components in eukaryotic organelles.23PubMed Central. Using synthetic biology to overcome barriers to stable expression of nitrogenase in eukaryotic organelles Active nitrogenase in a plant cell has not yet been achieved, but solving the NifD stability problem removed one of the steepest barriers.
Meanwhile, a parallel approach has explored whether nitrogenase cofactor assembly can proceed using proteins produced in yeast mitochondria, effectively testing whether the biosynthetic pathway can function in a eukaryotic host at all. Yeast mitochondria have been shown to support at least the early steps of FeMo-co assembly, providing cautious optimism that the full pathway might eventually work in a plant organelle.10PubMed Central. Nitrogenase cofactor biosynthesis using proteins produced in mitochondria of Saccharomyces cerevisiae
Evolutionary Origins
Nitrogenase appears to be ancient, likely predating the oxygenation of Earth’s atmosphere. Phylogenetic reconstructions of the enzyme’s core genes consistently recover two major lineages: one containing the conventional molybdenum-dependent nitrogenases (divided into aerobic-associated and anaerobic-associated subgroups), and another containing the vanadium and iron-only alternatives along with several poorly characterized nitrogenase-like sequences found in archaea and unusual bacteria.24PubMed Central. Reconstructing the evolutionary history of nitrogenases: Evidence for ancestral molybdenum‐cofactor utilization The deepest branches of the nitrogenase family tree are exclusively bacterial, with archaeal sequences consistently nested inside bacterial clades, suggesting that archaea acquired nitrogenase genes through horizontal transfer rather than inheriting them from the last universal common ancestor.25Molecular Biology and Evolution. Mapping Geological Events and Nitrogen Fixation Evolution Onto the Timetree of the Evolution of Nitrogen-Fixation Genes
Whether the ancestral nitrogenase used molybdenum, vanadium, or iron alone has been debated. Some evidence from gene phylogenies and metal availability in the early ocean points toward molybdenum utilization being ancestral, with vanadium and iron-only versions arising later as adaptations to environments where molybdenum was scarce.24PubMed Central. Reconstructing the evolutionary history of nitrogenases: Evidence for ancestral molybdenum‐cofactor utilization This is still actively investigated, but the picture that is emerging is of a deeply conserved enzyme family that diversified its metal usage to cope with changing geochemistry over billions of years.
Probing the Active Site With Spectroscopy
Much of what we know about how substrates interact with FeMo-co comes from trapping intermediates and studying them with electron paramagnetic resonance (EPR) and related magnetic techniques. These methods detect unpaired electrons in the metal cluster and are sensitive to what is bound nearby. For instance, when a modified version of the MoFe protein (with a specific amino acid change near the active site) is turned over in the presence of acetylene, new magnetic signals appear that are distinct from the resting-state signal. Using isotopically labeled acetylene, researchers showed that at least two acetylene-derived fragments bind simultaneously to the cofactor’s iron atoms, and that a strongly coupled proton on the bound species comes from the substrate itself rather than from the surrounding water.26Journal of the American Chemical Society. Characterization of an Intermediate in the Reduction of Acetylene by the Nitrogenase α-Gln 195 MoFe Protein by Q-band EPR and 13 C, 1 H ENDOR
Similar experiments with cyanide, another small-molecule substrate, showed that it too binds directly to the FeMo-cofactor. In MoFe proteins carrying amino acid substitutions at a different position near the active site, incubation with isotopically labeled cyanide produced a detectable magnetic coupling, confirming direct bonding and indicating that spin density from the cluster extends onto the bound substrate.27PubMed. Interaction of acetylene and cyanide with the resting state of nitrogenase alpha-96-substituted MoFe proteins These spectroscopic fingerprints are building a picture, one intermediate at a time, of how the active site manipulates bound molecules. Because nobody has yet been able to trap a genuine N₂-bound intermediate at high enough occupancy for detailed spectroscopic study, much of the mechanistic insight has come from studying these alternative substrates as stand-ins for the more elusive nitrogen-reduction pathway.