X-ray Emission Spectroscopy Evidences a Central Carbon in the Nitrogenase Iron-Molybdenum Cofactor

See allHide authors and affiliations

Science  18 Nov 2011:
Vol. 334, Issue 6058, pp. 974-977
DOI: 10.1126/science.1206445


Nitrogenase is a complex enzyme that catalyzes the reduction of dinitrogen to ammonia. Despite insight from structural and biochemical studies, its structure and mechanism await full characterization. An iron-molybdenum cofactor (FeMoco) is thought to be the site of dinitrogen reduction, but the identity of a central atom in this cofactor remains unknown. Fe Kβ x-ray emission spectroscopy (XES) of intact nitrogenase MoFe protein, isolated FeMoco, and the FeMoco-deficient ∆nifB protein indicates that among the candidate atoms oxygen, nitrogen, and carbon, it is carbon that best fits the XES data. The experimental XES is supported by computational efforts, which show that oxidation and spin states do not affect the assignment of the central atom to C4–. Identification of the central atom will drive further studies on its role in catalysis.

Nitrogenase (N2ase), found in symbiotic and free-living diazotrophs, catalyzes the reduction of dinitrogen (N2) to ammonia (NH3) using eight electrons, eight protons, and 16 MgATPs (ATP, adenosine triphosphate) (1). Industrially, the same reaction is performed by the Haber-Bosch process that produces more than 100 million tons of NH3 each year, thereby accounting for ~1.4% of global energy consumption. Understanding how nature activates the strongest homodinuclear bond in chemistry, the triple bond of N2, is the key for the future design of molecular catalysts.

The high-resolution crystal structure of N2ase determined by Einsle et al. (2) showed that the active site of the molybdenum-iron (MoFe) protein component of N2ase binds a complex cluster consisting of seven iron ions, one molybdenum ion, and nine sulfides (Fig. 1A); this cluster is referred to as the iron-molybdenum cofactor (FeMoco) and is thought to be the site of dinitrogen activation. For each FeMoco (of which there are two in the α2β2 tetrameric MoFe protein) there is an additional cluster that consists of eight irons and seven sulfides (Fig. 1B); this cluster is referred to as the P cluster. The P clusters serve as electron-transfer sites. Several reaction intermediates in nitrogenase catalysis have recently been observed (3, 4). However, despite the progress in the experimental and theoretical analysis of the FeMoco (47), neither the reaction that occurs at the FeMoco nor the structure of FeMoco has been fully clarified. In 2002, Einsle et al. identified a light atom in the center of FeMoco that could be attributed to a single, fully ionized C, N, or O atom (2). No consensus has since emerged concerning the nature of this key atom. Study of FeMoco by electron paramagnetic resonance and related techniques is complicated by complex spin-couplings between the open-shell ions, which are not fully understood. Mössbauer spectroscopy suffers from spectral crowding, and neither nuclear resonance vibrational spectroscopy nor extended x-ray absorption fine structure are sufficiently conclusive (8).

Fig. 1

The FeMoco (A) and P cluster (B) of nitrogenase (adapted from the Protein Data Bank: identification number 1MIN). Orange, Fe; yellow, S; light blue, Mo; black, C4–, N3-, or O2–; dark blue, nitrogen; gray, carbon. For clarity, the homocitrate and histidine ligands to the Mo have been omitted.

Herein, we report iron Kβ valence-to-core (V2C) x-ray emission spectroscopy (XES) of N2ase and demonstrate that these data provide a signature for the presence and identity of the central atom. Kα and Kβ XES monitor the emission of photons after ionization of a metal 1s electron. The Kβ1,3 emission line (~7040 to 7070 eV) corresponds to an electric dipole allowed 3p → 1s transition. To higher emission energies, valence-electron transitions into the metal 1s core hole are observed (referred to as the Kβ2,5/Kβ′′ or V2C region). These transitions have previously been assigned as ligand np → metal 1s (Kβ2,5, ~7102 to 7112 eV) and ligand ns → metal 1s (Kβ′′ or “satellite,” ~7090 to 7102 eV) transitions (9). V2C XES studies of Cr and Mn complexes have shown that the Kβ′′ features provide a signature for the identity of the directly coordinating ligands, because energies of the observed features depend primarily on the ligand 2s ionization energies (10, 11). We recently developed an experimental and theoretical protocol for the analysis of V2C XES spectra and applied it to mono- (1214) and multinuclear (15) iron complexes. Of particular relevance is a study of a six-iron cluster with a central μ6-C4– (15). These data show a feature at 7099 eV that is attributed to a transition originating from the μ6-C4– 2s orbital. Computationally, this feature is predicted to shift to 7094 eV for a μ6-N3– and 7088 eV for a μ6-O2–. These trends closely parallel previous observations for infinite lattice complexes and mononuclear molecular complexes, thus highlighting the general applicability of this method (11, 12, 16).

Figure 2A presents the normalized V2C XES data of the isolated FeMoco of N2ase, together with a representative fit to the data. Based on previous studies, the features observed with maxima at ~7108 and ~7100 eV are assigned to ligand np and ns contributions, respectively (12). To assess the contribution of the sulfur ligands relative to the interstitial atom X, data were also obtained for the ∆nifB MoFe protein (Fig. 2B). This mutant contains only the two P clusters (17, 18). Based on their structural similarity, it can be assumed that the P cluster and the FeMoco have similar sulfur contributions to their XES spectra; this assumption is also supported computationally (see below). Data were obtained for the intact MoFe protein (containing both clusters) (Fig. 2B). The MoFe protein spectra map well onto an average of the spectra of the P cluster (represented by the ∆nifB MoFe protein) and the isolated FeMoco (fig. S1).

Fig. 2

(A) Normalized V2C XES spectra of isolated FeMoco (red) and a representative fit to the data (black dashed line). (B) Comparison of the normalized V2C XES data for FeMoco (red), the MoFe protein (gray), and the ∆nifB MoFe protein (black). (Inset) V2C satellite region for Fe2O3 (red), Fe3N (blue), and MoFe protein (gray).

Comparison between the data of the isolated FeMoco and that of the P clusters in the ∆nifB MoFe protein allowed us to assess the relative contributions of these clusters to the spectra. The V2C XES data of the P clusters showed only a weak satellite at 7098.8 with 0.30 ± 0.03 units of integrated intensity. In contrast, isolated FeMoco exhibited a well-resolved satellite feature to higher energy (7100.2 eV) with an approximately sixfold increase in the integrated intensity of the satellite feature (1.78 ± 0.18 units). To better understand the origin of these satellite features, we also compared the data for ∆nifB MoFe protein to the XES data for a [Fe4S4(SPh)4]2– model complex (19). Both the P clusters and the Fe4S4 cubane have very similar XES spectra (figs. S2 and S3). Thus, the weak 7098.8-eV feature must be attributed to a S 3s → Fe 1s transition. Under the plausible assumption that the S 3s contributions to the P cluster and FeMoco V2C XES spectra are similar, we can model the satellite region with two features: one fixed at 7098.8 eV (corresponding to the S 3s contributions) and a second to higher energy (7100.2 eV) with increased intensity (1.6 units) (Fig. 2A and Table 1). The higher-energy feature is attributed to the presence of the interstitial light atom. Comparison of the energy of the 7100.2-eV satellite feature in FeMoco to the O 2s → Fe 1s (~7092 eV) and N 2s → 1s (~7095 eV) transitions observed in Fe2O3 and Fe3N, respectively (Fig. 2B, inset), indicates that this feature arises from a ligand with 5- and 8-eV lower ionization potential than O or N, respectively. Therefore, this comparison argues against either N or O 2s contributions and strongly supports a C 2s → Fe 1s assignment.

Table 1

V2C XES fit parameters. n/a, not applicable.

View this table:

Fe V2C XES spectra can be predicted surprisingly well within a simple scheme based on density functional theory (12). To complement the experimental data, we performed detailed calculations on the FeMoco, the P cluster, and the [Fe4S4(SPh)4]2– model complex. The FeMoco was modeled by a structure containing 152 atoms obtained from the high-resolution crystal structure of Einsle et al. (2), which incorporates the key structural and electronic features of the system. The oxidation state of this iron-sulfur cluster in the resting state of the enzyme has not been determined unambiguously. Thus, we performed calculations for the two oxidation states that have been shown to be most likely (2022). Although these two states differ by two units of charge, the differences in the calculated V2C XES transition energies and intensities are much smaller than the experimental resolution (fig. S6). The broken symmetry approach was used to approximate the effects of magnetic coupling of the various spin centers in the FeMoco calculations (23). However, calculations reveal that the ligand-to-metal crossover region of the predicted V2C XES spectra is largely unaffected by magnetic coupling (fig. S7). This finding is understandable considering the large linewidth of the experimental spectra and the rather subtle differences in orbital energies arising from different magnetic coupling schemes. More importantly, the predicted V2C spectra were highly sensitive to the identity of the interstitial ion. Figure 3A presents the calculated spectra of the FeMoco, assuming interstitial O2–, N3–, and C4– ions together with the calculated spectrum for the P cluster. As expected, all four spectra exhibit a relatively strong feature at ~7099.3 eV, corresponding to transitions from S 3s orbitals to the Fe 1s orbitals. The only exception is FeMoco with a central C4– (Fig. 3B), where the maximum is slightly shifted to higher energies due to contributions from C4–-related transitions in the same region. Hence, our presented data, along with analogous calculations on [Fe4S4(SPh)4]2– (fig. S8 and S9), support the aforementioned assumption that the S peak in the V2C region appears at the same position of the spectrum for all measured species.

Fig. 3

(A) Comparison of the calculated V2C XES spectra of FeMoco with an interstitial C4– (black), N3– (blue), and O2– (red) and of the spectra of the P clusters (gray). (B) Calculated V2C XES spectra of FeMoco with an interstitial C4– (black) and the P clusters (gray). (C) Experimental difference spectrum of FeMoco with the P clusters (gray), as well as calculated difference spectra of the P clusters with FeMoco containing interstitial C4– (black), N3– (blue), and O2– (red).

Subtraction of the calculated P-cluster spectrum from the calculated spectrum of the three FeMoco species yields the contributions from the respective interstitial ions (Fig. 3C). Analysis of the difference spectra reveals that the interstitial ions give rise to two features in the V2C spectrum associated with transitions from the ligand 2s and 2p orbitals, respectively. These features occur at 7096.1 and 7105.1 eV for N3– and at 7091.0 and 7104.0 eV for O2–. When a C4– ion is placed in the center of the FeMoco, the two features are observed at 7100.2 and 7107.9 eV (Fig. 3B). For N3– and C4–, the higher-energy feature is not distinguishable from the large peak at ~7107 eV that is dominated by transitions originating from the S 3p orbitals.

Taken together, the experimental and theoretical results support assignment of the interstitial species as a C4–. The calculated position of the C4– 2s → Fe 1s peak matches the experimentally determined position at 7100.2 eV. Both N3– and O2– are unlikely, as their respective calculated spectra show strong features at 7096.1 eV (N3– 2s) and 7091.0 eV (O2– 2s). In addition, the measured spectra do not exhibit any features at lower energies than the S 3s peak, whereas such features have been observed experimentally in other N3– and O2– systems (as shown in the inset of Fig. 2B). The assignment is further supported by our previous studies that have shown that features with a calculated intensity of more than 10 to 15 units of intensity are experimentally observable (12). These studies also showed that the integrated intensities of experimental and calculated V2C agree strongly, with a 19% error for crystallographic structures. Even considering this error, the calculated low-energy features related to the N3– (31 units of intensity) and O2– (26 units of intensity) ions considerably exceed this threshold. In addition, several other studies on O2–and N3– have shown features at the corresponding energy offsets (912). This finding raises interesting questions about both the role of the central atom and the possible pathways for biosynthesis of such an organometallic cluster.

Supporting Online Material

SOM Text

Figs. S1 to S11

Tables S1 and S2

References (2435)

References and Notes

  1. Acknowledgments: S.D. thanks Cornell Univ. for financial support and the Alfred P. Sloan Foundation for a fellowship; F.N. acknowledges financial support from the Univ. of Bonn, the Max Planck Society, and the SFB 813; M.W.R. thanks the NIH for funding (grant R01-GM 67626). Portions of this research were carried out at the Stanford Synchrotron Radiation Lightsource (SSRL), a U.S. Department of Energy (DOE), Basic Energy Sciences user facility. The SSRL Structural Molecular Biology program is supported by DOE, Biological and Environmental Research, and NIH, National Center for Research Resources, Biomedical Technology Program.
View Abstract

Stay Connected to Science

Navigate This Article