Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Jan 1.
Published in final edited form as: J Inorg Biochem. 2012 Oct 8;118:100–106. doi: 10.1016/j.jinorgbio.2012.10.001

Structure and spectroscopy of a bidentate bis-homocitrate dioxo-molybdenum(VI) complex: Insights relevant to the structure and properties of the FeMo-cofactor in nitrogenase

Zhao-Hui Zhou a,b,c,*, Hongxin Wang a,b, Ping Yu d, Marilyn M Olmstead a, Stephen P Cramer a,b,*
PMCID: PMC3596267  NIHMSID: NIHMS423386  PMID: 23147649

Abstract

Direct reaction of potassium molybdate (with natural abundance Mo or enriched with 92Mo or 100Mo) with excess hydrolyzed homocitric acid-γ-lactone in acidic solution resulted in the isolation of a cis-dioxo bis-homocitrato molybdenum(VI) complex, K2[*MoO2(R,S-H2homocit)2]·2H2O (1) (*Mo = Mo, 1; 92Mo, 2; 100Mo, 3; H4homocit = homocitric acid-γ-lactone·H2O) and K2[MoO2(18O-R,S-H2homocit)2]·2H2O (4). The complex has been characterized by elemental analysis, FT-IR, solid and solution 13C NMR, and single crystal x-ray diffraction analysis. The molybdenum atom in (1) is quasi-octahedrally coordinated by two cis oxo groups and two bidentate homocitrate ligands. The latter coordinates via its α-alkoxy and α-carboxy groups, while the β- and γ-carboxylic acid groups remain uncomplexed, similar to the coordination mode of homocitrate in the Mo-cofactor of nitrogenase. In the IR spectra, the Mo=O stretching modes near 900 cm-1 show 2-3 cm-1 red- and blue-shifts for the 92Mo-complex (2) and 100Mo-complex (3) respectively compared with the natural abundance version (1). At lower frequencies, bands at 553 and 540 cm-1 are assigned to νMo–O vibrations involving the alkoxide ligand. At higher frequencies, bands in the 1700-1730 cm-1 region are assigned to stretching modes of protonated carboxylates. In addition, a band at 1675 cm-1 was observed that may be analogous to a band seen at 1677 cm-1 in nitrogenase photolysis studies. The solution behavior of (1) in D2O was probed with 1H and 13C NMR spectra. An obvious dissociation of homocitrate was found, even though bound to the high valent Mo(VI). This suggests the likely lability of coordinated homocitrate in the FeMo-cofactor with its lower valence Mo(IV).

Keywords: Homocitric acid-γ-lactone, Homocitrate, FeMo-cofactor, Nitrogenase, FT-IR, Nitrogenase, NMR, Crystal structure

1. Introduction

Homocitrate produced by the nifV gene product (homocitrate synthase) is a key ingredient in the complex in vivo biosynthesis of the FeMo-cofactor of nitrogenase [1-6]. X-ray crystallography has shown that homocitrate coordinates the active site Mo in a bidentate fashion through alkoxide and carboxylate ligands from the C-2 position, thus creating a 5-membered ring as shown in Figure 1 [7-9]. Crystallographic analysis of nitrogenase from a nifV mutant shows that citrate can adopt a similar mode of binding [10], but the simple deletion of a single methylene group from the γ-carboxylate tail reduces N2 reduction activity to only 7% that of wild-type enzyme [11]. Removal of homocitrate also results in an enzyme whose H2-evolution activity, unlike the wild-type, is inhibited by CO [12]. Shah et al. investigated the ability of a wide range of homocitrate analogues to reconstitute enzyme activity, and concluded that the minimal requirements were the ‘hydroxy group, the 1- and 2-carboxy groups, and the R-configuration of the chiral center’ [13]. Thus, despite a great deal of evidence favoring Fe as the N2 substrate binding metal [14] and the 4Fe-4S face neighboring the α-V70 sidechain of the Av MoFe protein as the binding location [15-18], there is something special about homocitrate ligation at the Mo end of the FeMo-cofactor that remains critical to the catalytic mechanism. For this reason we have investigated the structural and spectroscopic properties of a homocitrate ligated Mo complex.

Figure 1.

Figure 1

Left: citrate vs. R-homocitrate. Middle: ball and stick view of FeMo cofactor and amino acid side chains in vicinity of homocitrate for Klebsiella pneumoniae (Kp) nitrogenase. Right: Comparison of homocitrate and citrate binding in Kp nitrogenase.

In this work, we report the synthesis, x-ray diffraction structural characterization, FT-IR, and solid and solution NMR spectroscopy of a molybdenum(VI)-homocitrate complex from aqueous solution. Isotopic substitution with 92Mo and 100Mo was used to confirm assignment of the vibrations involving molybdenum and oxygen, and 18O substitution helped in assignment of certain free carboxylate modes. We also observed an equilibrium between coordinated and free homocitrate ligand in solution based on 13C NMR technique, and the potential relevance to properties of homocitrate in nitrogenase is also discussed.

2. Experimental Section

2.1 Materials and synthetic methods

Potassium molybdate (99%) and 18O-water were purchased from Sigma. 92Mo- and 100Mo-enriched molybdenum oxides were purchased from Isoflex. Racemic homocitric acid-γ-lactone was prepared via a reported synthetic route [19]. The lactone compound was further purified from ethyl acetate and n-hexane mixed-solvents.

Homocitrato molybdenum(VI) complexes K2[MoO2(R,S-H2homocit)2]·2H2O (1), and the 92Mo (2) and 100Mo (3) isotopomers were synthesized starting from the total synthesis of R,S-homocitric acid γ-lactone as shown in Scheme 1. The 18O-γ-carboxy (4) isotopomer was prepared starting from the delactonization of R,S-homocitric acid γ-lactone in 18O-water.

Scheme 1.

Scheme 1

Synthesis of Mo homocitrate (1) and isotopomers: Mo* = Mo, 92Mo (2), or 100Mo (3).

The R,S-homocitric acid γ-lactone (0.21 g, 1.1 mmol) was dissolved in a minimal amount of water. The pH was adjusted to 11 by addition of 5.0 M potassium hydroxide to generate the acyclic homocitrate. The mixture was stirred for two days to complete the hydrolysis at room temperature. Potassium molybdate (0.12 g, 0.50 mmol) was then added and the solution was stirred for another 4 hours. The pH value of the solution was adjusted to 2.5 to induce complex formation. Since complex 1 is very soluble in water, potassium chloride could be precipitated first and removed by filtration. Complex 1 was crystallized after evaporation of solvent at room temperature. Yield: 0.15 g (46%). Elemental analyses were acquired using EA 1110 elemental analyzers. Calculated for C28H40O36K4Mo2: C, 22.4%; H, 3.1%, found: C, 22.0%; H, 3.2%.

In a similar procedure, 92Mo- and 100Mo-enriched oxides (0.25 mmol) were dissolved in 5.0 M potassium hydroxide respectively. The resulting isotopically labeled molybdates were reacted with hydrolyzed homocitrate, and then the pH value of the solutions was titrated to 2.5. Complexes 2, 3 and 4 were obtained with yields of 40%, 45%, and 41% respectively.

The preparation of molybdenum homocitrates depends mainly on having the required proportions of molybdate and ligand and the correct pH value of the reaction. The suitable pH value is 2~3. A slight excess and complete hydrolysis of the ligand drive the reaction completely and avoid the formation of undesired homocitrate-γ-lactone complexes.

2.2 X-ray data collections, structure solution and refinements

A crystal of (1) with dimensions 0.1×0.2×0.2 mm suitable for x-ray structural analysis was obtained as a colorless prism by cooling a saturated solution of (1) in a refrigerator (4 °C). X-ray diffraction data were collected on an Bruker Apex II diffractometer with graphite monochromated Mo-Kα radiation at 90 K. A Lorentz-polarization factor, anisotropic decay and empirical absorption corrections were applied. The structure was solved and refined by full-matrix least-squares procedures [20] with anisotropic thermal parameters for all the non-hydrogen atoms. H atoms were located from difference Fourier map and refined with use of a riding model. Crystallographic data (without structure factors) for the structure reported in this paper have been deposited with the Cambridge Crystallographic Data Centre (CCDC) as supplementary publication no. CCDC 871473.

2.3 Spectroscopy

The IR spectra were recorded at a resolution of 2 cm-1 with a Vertex 70v Bruker FT-IR spectrometer on KBr plates at room temperature. A KBr beam-splitter and a LN2 cooled broadband MCT (mercury cadmium telluride) detector were used, allowing measurement over the range of 400-4000 cm-1. The sample chamber was evacuated to 0.5 Torr to minimize the absorption of water vapor.

Solution and solid state NMR spectra were acquired at the UC Davis NMR facility using a Bruker AVANCE500 widebore spectrometer equipped with an 11.7 T magnet field. DSS (sodium 2,2-dimethyl-2-silapentane-5-sulfonate) was used as internal reference in solution 1H NMR and 13C NMR experiments. Solution NMR spectra were taken for measurement after 12 hours to obtain a steady spectrum. The concentration of the complex 1 in solution was 0.2 M.

3. Results and Discussion

3.1 Description of the molecular structure of (1)

The solid state structure of compound (1) comprises discrete potassium cations, bis-homocitrate cis-dioxomolybdenum anions and water molecules, as shown in Figure 2. Each Mo is hexacoordinate and exists in quasi-octahedral geometry. Selected bond distances and angles for (1) are shown in Table 1, while the complete crystallographic data are summarized in the supplemental materials (Table S1).

Figure 2.

Figure 2

Left: Perspective view of the anion structure of K2[MoO2(R,S-H2homocit)2]·2H2O (1) in ΛR configuration, thermal ellipsoids are drawn by ORTEP and represent 50% probability surfaces. Right: Overlay views of the homocitrate coordination in (1) (C in red) and the nitrogenase FeMo-cofactor (C in blue) solved at 1.0 Å resolution [9].

Table 1.

Selected bond distances (Å) and angles (°) for K2[MoO2(R,S-H2homocit)2]·2H2O (1). Mo–O coordination:

Mo(1)–O(1) 2.007(1) Mo(1)–O(2) 2.206(1) Mo(1)–O(8) 1.714(1)
Mo(1)–O(9) 1.706(1) Mo(1)–O(11) 1.961(1) Mo(1)–O(12) 2.211(1)
O(1)–Mo(1)–O(2) 73.75(4) O(1)–Mo(1)–O(8) 89.56(5) O(1)–Mo(1)–O(9) 105.07(5)
O(1)–Mo(1)–O(11) 153.50(4) O(1)–Mo(1)–O(12) 84.76(4) O(2)–Mo(1)–O(8) 160.56(5)
O(2)–Mo(1)–O(9) 89.93(5) O(2)–Mo(1)–O(11) 86.48(4) O(2)–Mo(1)–O(12) 80.97(4)
O(8)–Mo(1)–O(9) 104.26(5) O(8)–Mo(1)–O(11) 105.85(5) O(8)–Mo(1)–O(12) 87.79(5)
O(9)–Mo(1)–O(11) 92.10(5) O(9)–Mo(1)–O(12) 164.28(5) O(11)–Mo(1)–O(12) 74.66(4)
Hydrogen bonding and cation coordination of β-and γ-carboxy or carboxylic acid groups:
O(4)⋯O(13a) 2.520(2) O(4)–H(4)⋯O(13a) 164(3) O(6) ⋯ O(3b) 2.588(2)
O(6)–H(6)⋯O(3b) 168(3) O(14)⋯O(2w) 2.585(2) O(14)–H(14)⋯O(2w) 165(3)
O(16)⋯O(1c) 2.623(2) O(16)–H(16)⋯O(1c) 168(3) O(1w)⋯O(3d) 2.832(2)
O(1w)–H(1w2)⋯O(3d) 170(3) O(2w)⋯O(13e) 2.719(2) O(2w)–H(2w1)⋯O(13e) 158(3)
O(2w)⋯O(2f) 2.918(2) O(2w)–H(2w2)⋯O(2f) 168(3) K(1)⋯O(5g) 2.882(1)
K(1)⋯O(15h) 2.641(2) K(1)⋯O(17i) 2.743(1) K(2)⋯O(5g) 2.838(1)
K(2)⋯O(6j) 2.641(2) K(2)⋯O(7h) 2.755(1) K(2)⋯O(14) 2.893(1)
K(2)⋯O(17i) 2.772(1)

The homocitrate anions chelate to the Mo via α-alkoxy and monodentate α-carboxy groups, while the remaining β- and γ-carboxylic acid groups remain uncomplexed. The homocitrate Mo coordination in (1) thus is similar to the homocitrate coordination in the FeMo-cofactor of nitrogenase, as shown in the overlay of the five-member chelated rings in Figure 2. We note that considerable effort has been expended on the synthesis of chemical models for the FeMo cofactor that emphasize the Mo–S and Fe–S linkages, but relatively less work has been done relevant to the coordination of the tricarboxylic acid [21].

As shown in Table 1, the Mo–O distances in (1) vary systematically in the expected fashion. The terminal Mo=O bond lengths are 1.706(1) and 1.714(1) Å, as expected for double bonds. The 104° O=Mo=O angles, are considerably larger than the 90° nominal value for cis groups. This is expected from the greater O⋯O repulsions between oxygen atoms with short bonds to the metal atom. The Mo–Oalkoxy distances are longer [1.961(1), 2.007(1) Å], but still consistent with deprotonation of the hydroxy group in the homocitrate anion. Those to α-carboxy oxygen atoms are the longest [2.206(1), 2.211(1) Å], which is a result of the strong trans effect of the terminal oxo groups.

As illustrated in Table 2, the Mo–Oalkoxy distances in (1) are ~0.2 Å shorter than those reported in high resolution nitrogenase structures [8,9]. A longer Mo–Oalkoxy bond in the FeMo cofactor seems chemically reasonable, since the oxidation state for Mo is presumed to be Mo(IV) [22], while in (1) the Mo is Mo(VI). In contrast, the Mo–Ocarboxy bond lengths for (1) and for the 1.0 Å Av1 structure are both 2.21 Å. In this case the trans effect from the oxo groups in the Mo(VI) complex apparently produces the same bond lengthening as the lower Mo(IV) oxidation state in nitrogenase.

Table 2.

Comparison of Mo–O distances in (1) with those for FeMo cofactor in nitrogenase.

Complex PDB M–Oalkoxy (Å) M–Ocarboxy\ (Å) ref.
(1) 1.961(1) 2.206(1) this work
2.007(1) 2.211(1)
Azotobacter vinelandii N2ase (1.0 Å) (homocitrate) 3U7Q 2.18 2.21 [9]
Klebsiella pneumoniae N2ase (1.6 Å) (homocitrate) 1QGU 2.35(2) 2.29(1) [8]
Klebsiella pneumoniae N2ase (1.9 Å) (citrate) 1H1L 2.25 2.31 [10]

Referring back to Table 1, it is interesting to note that the β-carboxylic acid group forms a very strong intermolecular hydrogen bond with water molecule [O(14)⋯O(2w), 2.583(2) Å]. Moreover, other β- and γ-carboxylic acid groups form strong intermolecular hydrogen bonds with α-carboxy groups from neighboring homocitrates [O(4)⋯O(13a), 2.520(2) Å; O(6)⋯O(3b), 2.588(2) Å], resulting in the formation of a two dimensional coordination polymer. The detailed interactions among β- and γ-carboxylic acid groups with α-alkoxy, α-carboxy groups, and water molecules are shown in Figure 3. In previous DFT (density functional theory) calculations on an FeMo-cofactor model [23], the γ-carboxy group of R-homocitrate is proposed to undergo intramolecular hydrogen bonding with the imidazole ligand on Mo, while the β-carboxy group of the homocitrate ligand is free. This is in contrast with the previous structural reports of the tetranuclear molybdenum homocitrates (NH4)2K2[Mo4O11(R,S-Hhomocit)2]·6H2O and K5[Mo4O11(R,S-Hhomocit)2]Cl·5H2O [24], where the β-carboxy group is coordinated. Clearly, the conformational and coordination flexibility of the homocitrate ligand deserves further study.

Figure 3.

Figure 3

Environments of β- and γ-carboxylic acid groups in homocitrate portions of (1).

The β- and γ-carboxylic acid groups in (1) are protonated, which is clearly visible in difference maps in the crystallographic data. This is further supported by the charge balance in the complex, and the observed differences in C–O distances (~ 0.1 Å) between protonated and deprotonated carboxylate oxygens [O(4)–C(4), 1.316(2) Å, O(5)–C(4), 1.218(2) Å; O(6)–C(7), 1.321(2) Å, O(7)–C(7), 1.220(2) Å]. IR bands at 1731 and 1714 cm-1 (Figure 4) also confirm the presence of such protonated β- and γ-carboxy groups in complex 1 as will be discussed below.

Figure 4.

Figure 4

FTIR spectra of (1) and isotopomers. (a) region between 1000-1800 cm-1, comparing (1) (—) with isotopomer with 18O γ-carboxylate [4] (---), (b) region between 800-1000 cm-1, comparing (1) (—), (2) (—), and (3) (—), (c) region between 400-600 cm-1, comparing (1) (—), (2) (—), and (3) (—).

3.2 IR Spectra and Isotopic Shifts

IR spectra for (1) and various isotopomers are shown in Figure 4. The 1400~1750 cm-1 region is of course a valuable diagnostic for the protonation state of carboxylic acids [25]. Thus, the bands at 1731 and 1713 cm-1 correspond to C=O stretches of the protonated β- and γ-carboxylic acid groups of (1). The very strong bands of 1701, 1675, 1589 are also associated with carboxylate asymmetric stretches, while the lower frequency features at 1445, 1411, and 1390 cm-1 are presumably carboxylate symmetric stretches.

In the region around 900 cm-1, (1) shows strong bands of 925.6 and 894.8 cm-1 that result from the cis-dioxo Mo cores. The low frequency of symmetric MoO2 stretching can be explained by the coordination of the oxo groups to the potassium cation. Compared to natural abundance 1, complex 2 shows red shifts of 2.9 and 2.0 cm-1 for ΔνMo=O, while complex 3 shows blue shifts of 2.9 and 1.9 cm-1 (Figure 4b). At lower frequencies, bands around 553 and 540 cm-1 are assigned to νMo–O vibrations. In this case the isotope shifts were on the order of 1.0 cm-1 or less and are not considered reliable.

3.3 Solid and solution NMR studies

The 1H NMR spectrum shows partially overlapping patterns for the three magnetically and/or chemically non-equivalent CH2 protons of homocitrate ligands. In theory, two protons in these CH2 groups are inherently magnetically non-equivalent owing to the adjacent α-carbon atom; this is demonstrated by the 1H NMR spectrum of (1) (Figure S2) showing two AB quartets around 2.57 ppm. The formation constant for (1) was determined from this 1H NMR spectrum. Specifically, the methylene peaks between 3.1 and 1.8 ppm were integrated. Based on the relative intensities of bound and free homocitrate features, the estimated value of the cumulative formation constant was given by log K2=5.37, which shows equlirium between coordinated species 1 and free homocitrate.

Due to the overlap of features in the 1H spectrum, a further study of 13C NMR was carried out. The observed peak positions for (1) in the solid state and in solution by 13C NMR spectra are listed in Table 3, and the NMR data for the free homocitric acid-γ-lactone and the potassium homocitrate are also included in the same table for comparisons.

Table 3.

Solid and solution 13C NMR data (in ppm) of (1) and various homocitratesa.

Compound ≡CO/≡COH (CO2)α (CO2)β (CO2)γ (-CH2-)β (-CH2-)γ
Homocitric acid γ-lactone solid 82.8 177.9 177.2 176.2 41.4 29.7, 27.2
(1) – solid 89.1(6.3) 185.4(7.5) 179.2(2.0) 173.7(-2.5) 46.4(5.0) 38.3(8.6),37.9(8.2)
87.0(4.2) 182.1(4.2) 45.8(4.4) 32.0(5.0),29.6(2.4)
Homocitric acid γ-lactone solution 83.5 176.5 172.5 170.5 41.7 31.9, 28.3
Potassium homocitrate solution 79.3 185.8 184.4 182.3 48.8 38.1, 35.0
(1) – solution 90.3(12.8) 186.9(6.2) 181.0 177.2 46.2 36.4, 36.3
180.4 176.6 45.5 32.0, 31.7
K2(NH4)2[Mo4O11(Hhomocit)2] ·6H2O solution [24] 90.0 186.5 181.8 180.7 47.7 37.1, 31.7
a

Δδ values are given in parentheses

Figure 5 shows the solid and solution 13C NMR spectra of (1) and the NMR of homocitric acid γ-lactone in solid and potassium homocitrate in solution. Different peaks in the spectra correspond to different types of carbons. Taking homocitric acid-γ-lactone (Figure 5, top) as an example, the peak at 82.8 ppm corresponds to the carbon in the alkoxy group; while the highest 177.9 ppm peak derives from the α-carboxy group. Those at 177.2 and 176.2 ppm are for β- and γ-carboxy (-CO2) groups respectively. The lowest values at 41.4, 29.7 and 27.2 ppm are the peaks for methylene (-CH2-) groups. These solid-state and solution 13C NMR spectra provide valuable information on the coordination environment and the conversions of 1 in solution.

Figure 5.

Figure 5

Solid state and solution 13C NMR spectra of (1) compared with potassium homocitrate and homocitric acid γ-lactone.

By comparison with the 13C NMR spectrum of the free homocitric acid-γ-lactone, (1) gives only one set of 13C NMR signals in the solid state without decomposition, which can be attributed to the coordinated homocitrate in monomeric form. The spectrum shows an obvious downfield shift with respect to the corresponding carbons of free homocitric acid γ-lactone in the solid (Figure 5, top). As summarized in Table 3, the coordinated α-alkoxy (Δδ 6.3, 4.2 ppm) and α-carboxy (Δδ 7.5, 4.2 ppm) carbons show much larger shifts compared with the corresponding groups in the γ-lactone than do the uncoordinated β- and γ-carboxy carbons (Δδ = 2.0 and -2.5 ppm).

The solution 13C NMR spectrum of (1) is similar to that in solid. However, it gives two sets of resonances. Looking at the region around 75~90 ppm, the stronger set corresponds to that of coordinated (1) in the solid, which can be attributed to the resonances of the monomeric Mo-homocitrate species in solution. The weaker set at high field, not shown in Table 3, could be derived from decomposition of a bound monomeric species, as that for potassium homocitrate solution. The peak of 77.6 ppm in the high field region is assigned to alkoxy carbon of free ligand. Careful analysis of the solution 13C NMR spectrum in the highfield region shows that there is a small peak at 180.7 ppm, which could be assigned to the free α-carboxy group of homocitrate. The uncoordinated β- and γ-carboxy peaks are mixed with the coordinated species peaks.

The dissociated species could be fully assigned in the spectrum, which can thus serve as an internal reference of free ligand. This is further supported by the comparison between the 13C NMR spectrum for complex 1 and that for potassium homocitrate (Figure 5, bottom). The dissociated ligand gives obvious downfield shifts for the β- and γ-carboxy groups relative to the free lactone with two acidic groups. This is attributed to the basic medium of potassium homocitrate. Its hydroxy group appears at 79.3 ppm in high field region.

In general, solution 13C NMR downfield shifts in the monomeric Mo homocitrate 1 with respect to dissociated homocitrate are in the decreasing order of α-alkoxy (12.8 ppm) > α-carboxy (6.2 ppm) > β-carboxy (~ 0 ppm) > γ-carboxy group (~ 0 ppm) as shown in Table 1. Although α-alkoxy and α-carboxy groups are not strictly comparable, this observation is consistent with the increasing trend in their Mo–O lengths (as measured by X-ray structural analysis) and the decreasing trend in the corresponding bond strengths.

In aqueous solution, the Mo6+ complex undergoes acid-base equilibration rapidly and can transform between the complex and free homocitrate ligand, as shown by the 13C NMR measurements in solution. Thus complex 1 also undergoes association-dissociation equilibrium in addition to acid-base equilibrium in solution to give complexed and free homocitrate ligands. The finding here that complex 1 in solution rapidly forms a complex bound with homocitrate ligand as well as free homocitrate may be relevant to homocitrate interactions with molybdenum in nitrogenase. A model for extracted FeMo-co in NMF (N-methylformamide) is thus proposed as shown in Figure 6, which proposes two kinds of equilibriums between bound homocitrato FeMo-co and the substituted NMF solvent or substrate molecules.

Figure 6.

Figure 6

Proposed lability of homocitrate chelate for extracted FeMo-co in NMF.

4. Summary

A mononuclear dioxo molybdenum complex with bidentate coordination from two homocitrate anions has been synthesized and isolated from aqueous solution, under controlled conditions of reactant ratio and pH value using the 13C NMR monitoring as a guide. The complex has been characterized by single crystal x-ray diffraction, FT-IR, and 13C NMR spectroscopy in solution and the solid state, and each technique offered results relevant to homocitrate chemistry in nitrogenase.

The x-ray crystallographic structure revealed that (1) possesses the same type of coordination by the 2-alkoxo group and by the monodentate α-carboxylate group as seen in nitrogenase. It also showed the tendency of the remaining β- and γ- carboxylates to engage in H bonding, in this case yielding a two dimensional polymeric structure. The infrared spectrum showed the strength of this technique for addressing the protonation and or ligation status of the carboxylate side chains. Of special note is a band seen for (1) at 1675 cm-1, which may be related to the bands seen at 1686 and 1677 cm-1 in recent photolysis studies of CO-inhibited nitrogenase [26].

The NMR spectrum of (1) shows that it undergoes an equilibrium between the free ligand and the coordinated species in solution. This suggests that the coordinated α-alkoxy and α-carboxy groups are relatively labile in solution. This finding indicates that the molybdenum homocitrate coordination to the FeMo cofactor will have a similar or even greater lability, since it presumably involves a Mo(IV) rather than Mo(VI) ion. This lability may be a factor in the nitrogenase mechanism [27]. It is certainly relevant to the biosynthesis and insertion of the FeMo-cofactor into nitrogenase.

Supplementary Material

01

Highlights.

► Cis-dioxo bishomocitrato molybdate is bidentately coordinated with α-alkoxy and α-carboxy groups ► Its vibrations at 553 and 540 cm-1 are assigned to νMo–O involving the alkoxide ligand. ► An obvious dissociation of homocitrate was found from 1H and 13C NMR spectra.

Acknowledgments

This work was supported by (US) National Institute of General Medical Sciences (GM-65440 to SPC), US DOE Office of Biological and Environmental Research (SPC) and the National Science Foundation of China (21073150 to ZHZ), The LBNL work is under DOE contract DE-AC02-05CH11231.

Abbreviations

H2homocit

Homocitrate dianion

H4cit

Citric acid

H3mal

Malic acid

NMF

N-methylformamide

DFT

Density functional theory

Footnotes

Supplementary materials.

Structural data, IR and 1H NMR are available in supporting information.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Shah VK, Brill WJ. Proc Natl Acad Sci USA. 1977;74:3249–3253. doi: 10.1073/pnas.74.8.3249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Hu YL, Ribbe MW. Coord Chem Rev. 2011;255:1218–1224. doi: 10.1016/j.ccr.2010.11.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Rubio LM, Ludden PW. Annu Rev Microbiol. 2008;62:93–111. doi: 10.1146/annurev.micro.62.081307.162737. [DOI] [PubMed] [Google Scholar]
  • 4.Hu YL, Ribbe MW. Microbiol Mol Biol Rev. 2011;75:664. doi: 10.1128/MMBR.05008-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Shepard EM, Boyd ES, Broderick JB, Peters JW. Curr Opin Chem Biol. 2011;15:319–327. doi: 10.1016/j.cbpa.2011.02.012. [DOI] [PubMed] [Google Scholar]
  • 6.Dos Santos PC, Dean DR, Hu YL, Ribbe MW. Chem Rev. 2004;104:1159–1173. doi: 10.1021/cr020608l. [DOI] [PubMed] [Google Scholar]
  • 7.Kim JK, Rees DC. Nature. 1992;360:553–560. doi: 10.1038/360553a0. [DOI] [PubMed] [Google Scholar]
  • 8.Mayer SM, Lawson DM, Gormal CA, Roe SM, Smith BE. J Mol Biol. 1999;292:871–891. doi: 10.1006/jmbi.1999.3107. [DOI] [PubMed] [Google Scholar]
  • 9.Spatzal T, Aksoyoglu M, Zhang L, Andrade SLA, Schleicher E, Weber S, Rees DC, Einsle O. Science. 2011;334:940. doi: 10.1126/science.1214025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Mayer SM, Gormal CA, Smith BE, Lawson DM. J Biol Chem. 2002;277:35263–35266. doi: 10.1074/jbc.M205888200. [DOI] [PubMed] [Google Scholar]
  • 11.McLean PA, Dixon RA. Nature. 1981;292:655–656. doi: 10.1038/292655a0. [DOI] [PubMed] [Google Scholar]
  • 12.McLean PA, Smith BE, Dixon RA. Biochem J. 1983;211:589–597. doi: 10.1042/bj2110589. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Imperial J, Hoover TR, Madden MS, Ludden PW, Shah VK. Biochemistry. 1989;28:7796–7799. doi: 10.1021/bi00445a040. [DOI] [PubMed] [Google Scholar]
  • 14.Christie PD, Lee HI, Cameron LM, Hales BJ, Orme-Johnson WH, Hoffman BM. J Am Chem Soc. 1996;118:8707–8709. [Google Scholar]
  • 15.Benton PMC, Mayer SM, Shao J, Hoffman BM, Dean DR, Seefeldt LC. Biochemistry. 2001;40:13816–13825. doi: 10.1021/bi011571m. [DOI] [PubMed] [Google Scholar]
  • 16.Benton PMC, Laryukhin L, Mayer SM, Hoffman BM, Dean DR, Seefeldt LC. Biochemistry. 2003;42:9102–9109. doi: 10.1021/bi034595x. [DOI] [PubMed] [Google Scholar]
  • 17.Yang ZY, Seefeldt LC, Dean DR, Cramer SP, George SJ. Angew Chem. 2011;123:286–289. doi: 10.1002/anie.201005145. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Sarma R, Barney BM, Keable S, Dean DR, Seefeldt LC, Peters JW. J Inorg Biochem. 2010;104:385–389. doi: 10.1016/j.jinorgbio.2009.11.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Chen HB, Chen LY, Huang PQ, Zhang HK, Zhou ZH, Tsai KR. Tetrahedron. 2007;63:2148–2152. [Google Scholar]
  • 20.Sheldrick GM. Acta Crystallogr Sect A. 2008;64:112–122. doi: 10.1107/S0108767307043930. [DOI] [PubMed] [Google Scholar]
  • 21.Rao PV, Holm RH. Chem Rev. 2004;104:527–559. doi: 10.1021/cr020615+. [DOI] [PubMed] [Google Scholar]
  • 22.Doan PE, Telser J, Barney BM, Igarashi RY, Dean DR, Seefeldt LC, Hoffman BM. J Am Chem Soc. 2011;133:17329–17340. doi: 10.1021/ja205304t. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Grönberg KLC, Gormal CA, Durrant MC, Smith BE, Henderson RA. J Am Chem Soc. 1998;120:10613–10621. [Google Scholar]
  • 24.Zhou ZH, Hou SY, Cao ZX, Tsai KR, Chow YL. Inorg Chem. 2006;45:8447–8451. doi: 10.1021/ic061429f. [DOI] [PubMed] [Google Scholar]
  • 25.Max JJ, Chapados C. J Phys Chem A. 2004;108:3324–3337. [Google Scholar]
  • 26.Yan L, Dapper CH, George SJ, Wang HX, Mitra D, Dong WB, Newton WE, Cramer SP. Eur J Inorg Chem. 2011:2064–2074. doi: 10.1002/ejic.201100029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Durrant MC, Francis A, Lowe DJ, Newton WE, Fisher K. Biochem J. 2006;397:261–270. doi: 10.1042/BJ20060102. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01

RESOURCES