Evidence for Functionally Relevant Encounter Complexes in Nitrogenase Catalysis

Abstract

Nitrogenase is the only enzyme that can convert atmospheric dinitrogen (N2) into biologically usable ammonia (NH3). To achieve this multielectron redox process, the nitrogenase component proteins, MoFe-protein (MoFeP) and Fe-protein (FeP), repeatedly associate and dissociate in an ATP-dependent manner, where one electron is transferred from FeP to MoFeP per association. Here, we provide experimental evidence that encounter complexes between FeP and MoFeP play a functional role in nitrogenase catalysis. The encounter complexes are stabilized by electrostatic interactions involving a positively charged patch on the β-subunit of MoFeP. Three single mutations (βAsn399Glu, βLys400Glu, and βArg401Glu) in this patch were generated in Azotobacter vinelandii MoFeP. All of the resulting variants displayed decreases in specific catalytic activity, with the βK400E mutation showing the largest effect. As simulated by the Thorneley-Lowe kinetic scheme, this single mutation lowered the rate constant for FeP-MoFeP association 5-fold. We also found that the βK400E mutation did not affect the coupling of ATP hydrolysis with electron transfer (ET) between FeP and MoFeP. These data suggest a mechanism where FeP initially forms encounter complexes on the MoFeP β-subunit surface en route to the ATP-activated, ET-competent complex over the αβ-interface.

Figure 1

Structures of the nucleotide free, ATP, and ADP bound nitrogenase, where FeP (γ-subunit) is grey, MoFeP α-subunit is blue, and MoFeP β-subunit is red. The location of the β399–401 surface patch is circled, and the inset on the left shows interprotein interactions in DG1. All known FeP conformations in each docking geometry are depicted. For DG1 the only available structure is shown (PDB ID: 2AFH). For DG2, AMPPCP (a nonhydrolyzable ATP analog) (PDB ID: 2AFK), ADP.AlF₄⁻ (PDB ID: 1M34), and ATP/ADP-bound (PDB ID: 4WZA) structures are shown. For DG3, all four ADP-bound FeP conformers (PDB ID: 2AFI) are overlaid. Oxygen, nitrogen, iron, and sulfur are colored red, blue, orange, and yellow, respectively. Hydrogen bonds are marked by dashed lines. Nucleotides and metal clusters are shown as spheres. Only one αβ MoFeP dimer is displayed.

Figure 2

(A) Crystal structure of βK400E-MoFeP (light blue) compared to WT-MoFeP (magenta). (B) SDS-PAGE gels demonstrating that only WT-MoFeP crosslinks with FeP. Full images can be found in Figure S5. The asterisk marks an impurity found in βR401E-MoFeP.

Figure 3

(A) Nitrogenase C₂H₂ reduction assays for WT- and βK400E-MoFeP. The solid lines represent the Thorneley-Lowe simulation. (B) Inhibition of nitrogenase by NaCl. The solid lines represent the best-fit curves to an IC₅₀ equation. (C) Chelation of the [4Fe:4S] cluster from ATP-bound FeP by 2,2-bipyridine in absence of MoFeP or in presence of WT-MoFeP or βK400E-MoFeP. The solid lines represent the single exponential fits of the data. Data for control experiments lacking FeP or ATP are shown in Figure S8. (D) ATP activation of WT- and βK400E-MoFeP. The lines represent the Michaelis-Menten fits of the data. A normalized version of Figure 3D can be found in Figure S9.

Figure 4

AlF₄⁻ inhibition of nitrogenase turnover, where the data are fit to a slow inhibitor model (solid lines).

Figure 5

Catalytic activity under dilution conditions for WT- and βK400E-MoFeP. The solid lines represent the Thorneley-Lowe simulations used to determine k₁.

Figure 6

Model for the initial steps of the complex formation between FeP and MoFeP. Initial encounter of FeP with MoFeP results in loosely bound encounter complexes, which can either dissociate into the component proteins or form the metastable DG1 conformation and subsequently the long-lived DG2 conformation. The rate constants, k_encounter, k_steering and k_transition and their respective reverse rate constants represent the microscopic rates for the forward and reverse reaction, where the magnitude of the individual microscopic rates is not known. The β399-401 patch is represented by the “+” signs and residue γE112 and surrounding residues by a “−” sign.

Scheme 1

Model for nitrogenase turnover, adapted from Rees et al. F^ADP and F^ATP represent FeP with two nucleotides bound. The “n” subscript represents resting state MoFeP, and the “m” subscript, denotes MoFeP after receiving m electrons from FeP. Depending on the substrate, MoFeP is reduced by 2–6 electrons prior to substrate reduction. MoFeP refers to one half of the MoFeP.