The Fe protein is required for the biosynthesis of the two unique metalloclusters of the MoFe protein: the P-cluster [8Fe:7S] and the active site FeMo-co ([Mo:7Fe:9S:C]-R-homocitrate ) group. The -3 oxidation state is most common in biological compounds, such as nucleic and amino acids[1]. The most abundant form of nitrogen is atmospheric dinitrogen, which makes up ~78% of the Earth's atmosphere[2]. Complex formation with the MoFe protein results in a displacement of ∼5 Å of the [4Fe:4S] cluster to the plane of interaction with the MoFe protein, as well as a ∼13 degree tilt of the protein subunits toward the dimer two- folding axis (in total a 26 degree movement of one subunit relative to the other).
Multiple associations and dissociations of the Fe protein with the MoFe protein in the presence of ATP and external reductants (physiologically flavodoxin or ferredoxin, in vitro with dithionite) give. An electron from a ferredoxin is transferred to the Fe protein, which binds ATP in the presence of the MoFe protein. Additionally, the affinity of Fe protein for MoFe protein was demonstrated to be equivalent in the 1+ and 0+ states[19].
Examining the distribution of electrons in the Fe protein cluster is an important first step in understanding the relationship to activity. The signal intensity ratio S = 3/2 to 1/2 varies depending on the nucleotide state and the perturbing buffer conditions (eg, glycerol or urea), reflecting the solvent accessibility of the group or conformational state.[20, 21] Notably, the g = 4.3 signal may also arise from ferrous iron contamination (in the case of Lindahl et al., the signal was removed by stimulation of ATP hydrolysis, suggesting that the signal did not originate from ferrous iron, but a state of depending on the bound nucleotide) . Studies using model compounds have characterized many properties of synthetic FeS compounds: density functional theory (DFT) calculations and spectroscopic techniques (Mössbauer and EPR) show that the [4Fe4S] cluster can be split into two-[2Fe- 2S] pairs. [20,22] The 2+ state is modeled as a pair of ferromagnetically coupled high-spin Fe and a second pair of two antiferromagnetically coupled low-spin Fe, with the sixth d electron delocalized between the irons in each substrate.
The Fe-protein is essential for the biosynthesis of two unique metal clusters in the MoFe-protein, the P-group [8Fe:7S]. Maturase activity of the Fe-protein was determined by isolating the MoFe-protein in the ΔnifH deletion strain. This construction demonstrates that each atom in the crystal lattice contributes to the intensity of the reflection.
The 𝑓_(𝜆) and 𝑓__(𝜆) terms are respectively the real and imaginary terms of the scattering factor, where 𝑓_(𝜆) is the dispersive component and 𝑓__(𝜆) is the anomalous component of the atomic scattering factor. The method was tested in the case of the [2Fe:2S] cluster of a ferredoxin from Aquifex aeolicus[34]. SITE-SPECIFIC OXIDATION STATE ASSIGNMENTS OF THE IRONS IN THE [4FE:4S]2+/1+/0 STATES OF THE NITROGENASE FE-PROTEIN [4FE:4S]2+/1+/0 STATES OF THE NITROGENASE FE-PROTEIN.
Absorption curves were compiled for each iron site in different forms of the Fe protein [4Fe:4S] cluster and overlaid with the reference spectra (Figures 29, 30). To confirm that the electronic state of the [4Fe:4S]1+ cluster in the crystal reflected the properties of the solution, perpendicular-mode electron paramagnetic resonance (EPR) was used. A polycrystalline sample of the [4Fe:4S]1+ nucleotide-free protein crystals (used in SpReAD experiments) was prepared and compared to the respective protein solution sample using EPR (Figure 31, see Methods).
The quiescent state of Fe protein is generally considered to be a mixture of S = 1/2 and S = 3/2 spin states[20], and the lack of S = 3/2 state signal may reflect solvent dependence. of equilibrium, with PEG favoring the S = 1/2 state as observed with ethylene glycol[20].
SOLUTION OF AN ANAEROBIC CRYO-EM STRUCTURE OF A NITROGENASE MATURATION COMPLEX
Here we report the cryo-EM structure of the intermediate complex NifENH-ADP•AlF4-, formed in the FeMo-co maturation pathway, using an anaerobic lattice preparation technique. The structure is superimposable with the crystal structure, 3U7Q[12], with the notable exception of the N-terminal 50 residues, which were disordered in the EM structure. At lower contour levels, the N-terminal element of the α subunit is present, but the density is significantly weaker compared to the core of the protein.
The protein density agrees well with the 2.3-Å resolution crystal structure of the complex, 1M34[11], and importantly, all six metal clusters were identified. The density was well modeled by each of the constituent subunit structures (NifEN, 3PDI[84] and NifH, 1M34[11]). Similarly, the NifEN part of the model has an RMSD of about 1 Å compared to the published crystal structure.
A pairwise alignment of the monomers indicates that the NifH dimers in the EM structure most closely match the ADP-AlF4 conformation, but the EM model has a larger displacement of the hinge motion, resulting in the ∼1 Å deviation with the NifH from 1M34 (Figure 39). Importantly, RMSD deviations in the P-loop (Lys10-Lys15) and the Switch II region (Asp129-Val130), which reflect the nucleotide state of the protein, are very low and the backbones are superimposable. Further detailed interpretation of the structural differences is limited by the resolution of the structures.
One region is at the N-terminus of the NifE domain, near the position of the putative L-cluster (Figure 41). A second region of unmodeled density is in the core of the protein, extending from the C-terminus of NifN (Figure 42). Unassigned density in the core of the protein, corresponding to the N-terminal region of NifB.
In our structure of NifENH-ADP•AlF4-, the arrangement of the proteins is analogous to the nitrogenase complex NifDKH ADP•AlF4- [11], where the position of the auxiliary clusters (non-L) supports an electron transfer role through NifEN to the putative L cluster. While the role of electron transfer is not fully understood in the NifENH complex, the L cluster may require reduction for removal of the 8th iron. While there are no complex structures of vanadium nitrogenase (VFe protein) and the MoFe protein from Clostridium pasteurianum, electrostatic potential map analysis of the structures reveals similar motifs observed in Av NifDK (MoFe protein) and NifEN.
STRUCTURAL STUDIES OF A P-CLUSTER MATURATION INTERMEDIATE
A crystal structure of the MoFe protein isolated from the single ΔnifB deletion strain contained two mature P clusters and was FeMo co-deficient [ 108 ], representing the final stage of P cluster maturation. Interestingly, the ΔBNifDK structure adopted an open conformation in the α subunit to accommodate the mature FeMo-co. A map containing a fully occupied α-subunit could not be obtained, suggesting that an α-subunit is missing or disordered in a fraction of the ΔBΔZNifDK molecules and could not be resolved.
We refined a model of the protein in real space (Table 8), using the crystal structure of the FeMo co-defect ΔBNifDK as starting coordinates (Figure 48). Major deviations between the cofactor-deficient structures are localized to the α-subunit, although the main helix in the β-subunit (βTyr487-βArg510) that constitutes the core of the tetramer has displacements on the order of 2–3 Å. Among the crystal structures, the RMSD between the α-subunit of the high-resolution MoFe protein and ΔBNifDK is 0.39 Å.
By omitting the α subunit, the Ca positions of the holostructure and ΔBΔZNifDK are very similar, but deviations do occur in the surface helix βLys400-βAla411 and the helix loop βGlu202-βLys222. The displacement of the helices ~2 Å at a site distal and opposite to the FeMo co-binding site suggests that a region of flexibility could be related to the maturation of the P cluster (Figure 51). The maturation of the complex nitrogenase cofactors, the FeMo-co ([7Fe:9S:C:Mo:R-homocitrate]) and the P-cluster ([8Fe:7S]), is a crucial cellular process that provides an intriguing synthetic process. challenge, as well as a delicate logistical operation involving 15 gene products[108].
The initial steps in the maturation process of apo-MoFe-protein into a functional component of the nitrogenase complex are dominated by the stepwise generation of P groups. We solved a 4.4-Å resolution cryo-EM structure of the MoFe protein isolated from a strain ΔnifB ΔnifZ. Deletion of the NifZ gene product confers asymmetric maturation of the P pool in the NifB deletion background, raising important questions about the fusion mechanism.
Our results provide structural evidence for the spectroscopic prediction of asymmetric P cluster maturation and identify a disorder in the α subunit that did not contain a mature P cluster. Further experiments will be needed to investigate the origin of the asymmetric 2[4Fe:4S] fusion, including how it is identified partially matured MoFe protein for further processing. 602,236 particles contributed to the 4.4 Å resolution structure of the ΔBΔZNifDK tetramer with an incompletely occupied α subunit.
