Please be patient while the structures in the left frame load. In order to display all of the structures in the tour properly, press 'View' buttons below in order (from 1 to the end).
The enzyme nitrogenase is found in certain bacteria and blue-green algae, which can reduce N2 to NH3 (nitrogen fixation). Some of these bacteria are free-living while others are symbiotic (in the anaerobic environment of roots of legume plants). This bacterial reaction is the key step in the nitrogen cycle, which maintains a balance between two reservoirs of the nitrogen compounds: the Earths atmosphere and the biosphere. The plants cannot extract nitrogen directly from the atmosphere. N2 must be fixed by these bacteria (as NH3) or converted by lightning to NOx and then NO3- in the reaction with water. The fixed nitrogen is lost through leaching of the soil, harvesting of crops and the action of the denitrifying bacteria.
The overall net reaction that occurs at this enzyme can be presented with the following equation:
Note that in addition to the dinitrogen reduction, the reduction of protons to dihydrogen also takes place so nitrogenase is also a hydrogenase.
In order to "pump" the eight electrons required for reduction of the substrate (dinitrogen), ATP is hydrolysed to ADP with the stoichiometry of two ATP for each electron (16 equivalents of ATP per 1 equivalent of N2).
Nitrogenase actually consists of two proteins that work in tandem: the iron (Fe) protein (currently shown in the left frame as wireframe model) and the molybdenum-iron (MoFe) protein. [2] During the catalytic reduction of dinitrogen, the electrons are transferred from the Fe-protein to the MoFe-protein.
The Iron Protein
The Fe-protein accepts the electrons from a flavedoxin or a ferredoxin and transfers them further to the MoFe-protein. This electron transfer is enabled by (probably) simultaneous hydrolysis of ATP to ADP. The protein has a molecular weight of 60 kDaltons and 578 amino acid residues.
The Fe-protein is a dimer composed of two identical subunits (white and blue).
The two subunits are connected via a [4Fe - 4S] cluster. This cluster accepts the electrons and transfers them further to the MoFe-protein.
The cluster is ligated with four cysteine residues, two from each subunit.
Each subunit binds one ATP (or in the structure on the left ADP are present instead of ATP and are shown as spacefilling models). Magnesium ions (shown in green) provide the necessary charge balance. The binding of MgATP causes a change in both the protein conformation and the redox potential of the cluster that facilitates binding and electron transfer to the FeMo-protein. The Fe-protein "docks" to the MoFe-protein and transfers one electron each time two ATP are converted to 2ADP and 2(HPO4)2-. This is followed by the dissociation of the protein, preventing the back-electron-transfer reaction. This dissociation likely represents the rate-limiting step in the reduction of N2. MgATP hydrolysis occurs only when the dimer complex is formed. Therefore, the non-productive electron loss is prevented and the mechanism only allows the unidirectional flow of the electrons to the FeMo-protein.
In this display the protein backbone is removed and only the [4Fe - 4S] cluster and two MgADP molecules are shown.
The MoFe-Protein
This is the MoFe-protein shown in wireframe. This protein has a molecular weight of 220 kDaltons.
This protein is a α2β2 tetramer. It consists of four subunits, two alpha colored blue and two beta subunits colored red.
The MoFe-protein contains two copies each of two very unique metal clusters: a so-called P-cluster, Fe8S7 and the FeMo-cofactor cluster (sometimes referred to as an M-cluster or simply FeMoCo), Fe7MoS9N. While the FeMo-cofactor is located in the alpha subunit, the P-cluster sits at the interface between the alpha and beta subunits. Both are buried in the protein to prevent access of H2O, which would be converted to H2 by these very reducing metal centres.
The role of the P-cluster is to transfer the electrons from the Fe-protein to the FeMo-cofactor. This is the structure of the oxidized P-cluster. It is connected to the backbone via 7 amino acid residues: six cysteines and one serine. Three cysteine residues are provided by alpha subunit while the rest come from the beta subunit.
This display shows the oxidized P-cluster highlighting some of its important structural features. It can be described as the one containing the [4Fe-4S] and [4Fe-3S] clusters bridged via the four-coordinate sulphur S1 atom. (Note the distance between S1 and Fe6 atoms and also that Fe5 is not bonded to S1 in the oxidized state.) The sulphur atom from Cys88(alpha) bridges Fe5 and one Fe atom from 4Fe-4S cluster while its backbone amide nitrogen is coordinated to the Fe5. The Fe6 atom, apart from being coordinated by one cysteine residue, is bonded to the oxygen atom from Ser188 in the beta subunit.
This is the structure of the P cluster after the two electron reduction. Several important structural changes compared to the oxidised form might be observed. First, the S1 sulphur atom now coordinates to the Fe5 and moves closer to the Fe6 atom. As a result, the S1 atom adopts a distorted octahedral geometry. Additionally, the amide nitrogen from Cys88(beta) and the alkoxide oxygen from Ser188(beta) are no longer bonded to the Fe5 and the Fe6 atoms respectively. Since both of these ligands are protonated in their free states and should be deprotonated in their bound states, this means that the two-electron oxidation of P-cluster is accompanied by the release of two protons necessary for dinitrogen reduction (a proton for each electron, the ratio required for the substrate reduction). Hence, the P-cluster might also serve as a "proton pump".
This is the structure of the FeMo-cofactor cluster. The protein holds this cluster with only two residues: His442 bonded to the molybdenum atom and Cys275 bonded to the iron atom located opposite from Mo. The alpha subunit provides both residues. It is interesting to note that all the iron atoms, except the one bonded to Cys275, have a distorted trigonal pyramidal rather than tetrahedral arrangement of ligands around them. The six-coordinate molybdenum atom has a homocitrate ligand bonded via the hydroxyl and carboxylate oxygens. The homocitrate group also has two dangling carboxyl groups which might be part of a relay system to bring the protons to the active site. The most surprising structural feature of this cluster is presence of an atom in the middle of the FeMo-cofactor. The identity of this central atom has been established only recently as carbon. This hexacoordinate, trigonal prismatic carbon atom is almost equidistant from six Fe atoms surrounding it (average Fe - C distance is 2.00 Å ± 0.05 Å). While the P-cluster undergoes significant structural changes during the reduction of dinitrogen substrate, it is not clear yet what conformational changes occur at MoFe-cofactor.
Electrons coming from the P-cluster have to travel approximately 14 Å to reach the MoFe-cofactor. It is interesting to note here that the net charge of either of the clusters has not been yet established unambiguously. However, there is evidence that the majority of the Fe ions are in the Fe(II) state.
This is the structure of the FeMo-cofactor cluster.
If you would like to learn more about the nitrogenase and its model compounds see the following:
Much could be learned from the synthetic models for nitrogenase's cluster and cofactors (their function, mechanism of dinitrogen reduction, possible applications as industrial catalysts etc). However, the design and preparation of such compounds presents a major challenge for synthetic chemists. The most notable examples, described below, come from Holm's research group.
A Model For the P Cluster
Holm's model complex: The structure in the left frame presents a model cluster of the reduced form of the P cluster - [(Tp)2 Mo2Fe6S9(SH)2]3- [3] [Tp = hydrido(trispyrazolyl)borate]. The bridging pattern is Mo2Fe6( µ2-S)2( µ3-S)6( µ6-S). The structure can be described as one consisting of two MoFe3( µ3-S)3 cuboidal clusters linked by two µ2-S atoms from HS- ligands (simulating µ2-SCYS bridges) and one µ6-S atom (simulating S1 atom in the cluster). Each iron atom has distorted tetrahedral FeS4 geometry. The experimental results suggest the following oxidation states: Mo(III)2Fe(II)5Fe(III). The complex shows three oxidation steps in the cyclic voltamogram.
A Model For the MoFe Cofactor
The frame on the left displays the structure of the complex anion [MoFe3S4(SEt)3(Cl4cat)(CN)]3-, (Cl4cat = C6Cl4O2) [4]. It has a mixed Mo-Fe cuboidal core which resembles one part of the MoFe cofactor. The molybdenum atom has an octahedral geometry with Cl4cat2- ligand replacing homocitrate. The interesting part of the structure is the presence of a cyano (CN-) ligand on molybdenum atom. Since dinitrogen molecule and CN- anion are isoelectronic, this complex suggests that N2 substrate might be coordinated to the Mo atom of the MoFe cofactor during the reduction.
A Theoretical Model
Even though nitrogenase has been extensively studied many important questions still remain unanswered, for example: How is the substrate (dinitrogen) binding to the MoFe cofactor? What is the mechanism of dinitrogen reduction? In the cases when the experiments are unable to provide the answers to such a questions, the theoretical approach can be very useful. The structure on the left is a calculated structure of the N2-MoFe cofactor complex [5]. In order to make the calculations easier, the system is usually simplified. So in the case of the MoFe cofactor, an imidazole ring and glycolate have been used to replace the histidine and homocitrate ligands and an SH group instead of the cysteine terminal ligand. Of the three N2 binding modes (end-on bonding to the iron atom, bridging between two Fe atoms or end-on bonding to the Mo atom), the one with dinitrogen end-on bonded to an Fe atom (left) was shown to be the most probable structure. The calculations also indicate that the central N atom adds to the cluster flexibility by offering a variable number of bonds to the Fe atoms (these calculations were done before it was established that the central atom is actually a carbon). These calculations have also provided a plausible reaction mechanism.
[1] Picture taken from Whitten, Davis & Larry Peck "General Chemistry", 6th edition, Saunders College Publishing (2000), pg. 959.
[2] Nitrogenases in which molybdenum is replaced with vanadium are also known (VFe protein).
[3] Zhang, Y. and Holm, R.H. (2003) J. Am. Chem. Soc. 125(13) 3910. See also Zhang, Y. et al. (2002) J. Am. Chem. Soc. 124(48) 14292.
[4] Palermo, R.E. et al. (1984) J. Am. Chem. Soc. 106(9) 2600. See also Huang, J.S. et al. (1997) J. Am. Chem. Soc. 119(37) 8662.Note that the compounds in these articles have been prepared before the existence of the central N atom in the MoFe cofactor was established.
[5] Schimpl, J. et al. (2003) J. Am. Chem. Soc. 125(51) 15772.