The global biological nitrogen cycle is driven by the reduction of atmospheric nitrogen gas (N2) to ammonia (NH3) (27, 97), a form of nitrogen that can be used directly by many organisms and is easily converted to other forms of nitrogen such as nitrate (NO3− ), the preferred nitrogen source for plants. Biological nitrogen fixation is catalyzed by nitrogenase, a complex oxygen-sensitive metalloenzyme (10). The reduction of the triple bond of N2 is an extremely difficult reaction and is solely a microbial process that is carried out by selected bacteria and archaea. All nitrogen-fixing microbes synthesize a molybdenum (Mo) nitrogenase that has a unique iron (Fe)-Mo cofactor at its active site. Mo nitrogenase is a two-component enzyme consisting of an Fe protein (dinitrogenase reductase, encoded by nifH) and a MoFe protein (dinitrogenase, encoded by nifDK). The MoFe protein is a heterotetramer consisting of two α subunits and two β subunits (Figure 1a)...

Figure 1  A diagram of the three forms of nitrogenase showing the individual subunits and major route of electron transfer. (a) M is Mo, V, or Fe. (b) The structures for the active site FeMo and FeV cofactors as determined by X-ray crystallography and the predicted structure for the FeFe cofactor are shown. (c) The sequence of reactions for hydride reduction at the active site that occurs in all three nitrogenases. Some aspects of the reaction sequence are proposed. Figure adapted from Reference 33.

The overall reaction catalyzed by Mo nitrogenase is



From this formulation one can see that nitrogen fixation is an ATP-intensive reaction, and it also requires large amounts of protons and reductant in the form of energy-rich electrons. N2 reduction to NH3 is difficult to achieve because although it is thermodynamically favorable, there is a large activation energy barrier to catalysis that must be overcome (90) (Figure 2). Because of the complexity of the reaction it catalyzes, nitrogenase has a slow turnover rate (104). For these reasons it needs to be synthesized in large amounts, in some cases as much as 10% of total cellular protein (20), to sustain growth under nitrogen-fixing conditions.

Figure 2  There is a large activation energy barrier to N2 reduction. Shown is the standard free energy change (ΔG°) between intermediates of the N2 reduction pathway. Adapted from Reference 90.

In the 1980s it became clear that the nitrogen-fixing bacterium Azotobacter vinelandii could express forms of nitrogenase that did not contain Mo. This led to the discovery and initial descriptions of vanadium (V) and Fe-only nitrogenases (4–6, 12, 23, 77, 85). These enzymes comprise VnfHDK and AnfHDK subunits that are homologous to the NifHDK subunits of Mo nitrogenase, but they have either an FeV or an FeFe cofactor at their active sites. The alternative nitrogenases also include VnfG and AnfG subunits as additional components (Figure 1a). So far, alternative nitrogenases have been found only in microbes that also have Mo nitrogenase, most likely because they rely for their cofactor biosynthesis and assembly on Nif proteins that are also necessary for Mo nitrogenase synthesis. V and Fe-only nitrogenases are less efficient and less widespread than Mo nitrogenase (23, 41, 64, 109).

Based on recent studies (32, 33), the reactions catalyzed by the alternative nitrogenases are as follows:

Fe-only nitrogenase,



V nitrogenase,



All three nitrogenases produce hydrogen gas (H2) as an obligate aspect of nitrogen fixation. However, the alternative nitrogenases, and in particular Fe-only nitrogenase, produce much more H2 per N2 reduced than does Mo nitrogenase. This has led to the suggestion that nitrogenases might be used to generate H2 as a biofuel, and there have been a large number of studies directed toward this application (30, 59, 62, 78, 93).

Figure 4  A model for the regulation of synthesis and activities of the three forms of nitrogenase in Rhodopseudomonas palustris in response to fixed nitrogen starvation. The predicted activities of regulatory proteins NtrBC, NifA, and DraT in ammonium-grown cells are shown on the left side, whereas the predicted activities of NtrBC, NifA, and DraG in cells that are starved for fixed nitrogen (nitrogen-fixing cells) and expressing Mo nitrogenase are shown in the middle panel. Not shown is how the activities of these proteins are influenced by interactions with PII proteins, which are either unuridylylated or uridylylated, depending on cellular nitrogen status. As shown in the right-hand panel, cells subjected to extreme nitrogen starvation because of an inability to express functional Mo nitrogenase are predicted to express active V nitrogenase (when V is present) as well as Fe-only nitrogenase. Fe-only nitrogenase and V nitrogenase are posttranslationally modified by ADP-ribosylation and inactivated when cells are exposed to ammonium. This reversible reaction occurs as is shown for Mo nitrogenase in the left and middle panels.