Nitrogenase: a draft mechanism

Abstract

Biological nitrogen fixation, the reduction of N(2) to two NH(3) molecules, supports more than half the human population. The predominant form of the enzyme nitrogenase, which catalyzes this reaction, comprises an electron-delivery Fe protein and a catalytic MoFe protein. Although nitrogenase has been studied extensively, the catalytic mechanism has remained unknown. At a minimum, a mechanism must identify and characterize each intermediate formed during catalysis and embed these intermediates within a kinetic framework that explains their dynamic interconversion. The Lowe-Thorneley (LT) model describes nitrogenase kinetics and provides rate constants for transformations among intermediates (denoted E(n), where n is the number of electrons (and protons), that have accumulated within the MoFe protein). Until recently, however, research on purified nitrogenase had not characterized any E(n) state beyond E(0). In this Account, we summarize the recent characterization of three freeze-trapped intermediate states formed during nitrogenase catalysis and place them within the LT kinetic scheme. First we discuss the key E(4) state, which is primed for N(2) binding and reduction and which we refer to as the "Janus intermediate" because it lies halfway through the reaction cycle. This state has accumulated four reducing equivalents stored as two [Fe-H-Fe] bridging hydrides bound to the active-site iron-molybdenum cofactor ([7Fe-9S-Mo-C-homocitrate]; FeMo-co) at its resting oxidation level. The other two trapped intermediates contain reduced forms of N(2). One, intermediate, designated I, has S = 1/2 FeMo-co. Electron nuclear double resonance/hyperfine sublevel correlation (ENDOR/HYSCORE) measurements indicate that I is the final catalytic state, E(8), with NH(3) product bound to FeMo-co at its resting redox level. The other characterized intermediate, designated H, has integer-spin FeMo-co (non-Kramers; S ≥ 2). Electron spin echo envelope modulation (ESEEM) measurements indicate that H contains the [-NH(2)] fragment bound to FeMo-co and therefore corresponds to E(7). These assignments in the context of previous studies imply a pathway in which (i) N(2) binds at E(4) with liberation of H(2), (ii) N(2) is promptly reduced to N(2)H(2), (iii) the two N's are reduced in two steps to form hydrazine-bound FeMo-co, and (iv) two NH(3) are liberated in two further steps of reduction. This proposal identifies nitrogenase as following a "prompt-alternating (P-A)" reaction pathway and unifies the catalytic pathway with the LT kinetic framework. However, the proposal does not incorporate one of the most puzzling aspects of nitrogenase catalysis: obligatory generation of H(2) upon N(2) binding that apparently "wastes" two reducing equivalents and thus 25% of the total energy supplied by the hydrolysis of ATP. Because E(4) stores its four accumulated reducing equivalents as two bridging hydrides, we propose an answer to this puzzle based on the organometallic chemistry of hydrides and dihydrogen. We propose that H(2) release upon N(2) binding involves reductive elimination of two hydrides to yield N(2) bound to doubly reduced FeMo-co. Delivery of the two available electrons and two activating protons yields cofactor-bound diazene, in agreement with the P-A scheme. This keystone completes a draft mechanism for nitrogenase that both organizes the vast body of data on which it is founded and serves as a basis for future experiments.

Fig 1

FeMo-co and key residues

Fig 2

Highly simplified LT kinetic scheme, highlighting, (i) correlated electron/proton delivery in eight steps; (ii) some of the possible pathways for decay by H₂ release are shown; (iii) N₂ binding and H₂ release at either the E₃ or E₄ levels, with the pathway through E₃ deemphasized. LT also denote the protons added to FeMo-co (eg. E₁H₁); for clarity we have omitted this.

Fig 3

ENDOR-derived description of E₄ as containing two Fe-H-Fe moieties, emphasizing our view of the essential role of this key ‘Janus intermediate’, which comes at the halfway point in the LT scheme, and whose properties have implications for the first and second halves of the scheme. Janus image adapted from: Janus12.jpg checkxstarinfinity.blogspot.com/

Fig 4

Formulation of E₁-E₃ derived from consideration of E₄. Note M denotes FeMo-co inorganic core in its entirety; the superscript (+/−) represents difference between core charge and that of core in the resting state.

Fig 5

Alternative formulation of E₁-E₄ under assumption that hydride formation occurs at each stage. As in Fig 4, Note: M denotes FeMo-co inorganic core and superscript (+/−) represents difference between core charge and that of core in the resting state.

Fig 6

Integration of LT kinetic scheme with Alternating (A) pathways for N₂ reduction. again denotes FeMo-co core, and substrate-derived species are drawn to indicate stoichiometry only, not mode of substrate binding. Bold arrow indicates transfer to substrate of hydride remaining after N₂ binding in E₄; P represents ‘Prompt’ transfer. E_n states, n = even, are Kramers states; n = odd are non-Kramers. M^N denotes resting-state FeMo-co. Adapted from ref 29

Fig 7

Visualization of re mechanism for H₂ release upon N₂ (blue) binding to E₄. Shown, the Fe-2,3,6,7 face of resting FeMo-co; the structure of FeMo-co may vary in different stages of catalysis. The Fe shown binding N₂ is presumed to be Fe6, as indicated by studies of α-70^Val variants; when bold, red, formally Fe(0) (see text). The two bridging hydrides of E₄ (green) are positioned as suggested by re mechanism of H₂ release upon N₂ binding. Alternative binding modes for N₂-derived species can be envisaged.

Fig 8

Reversal of re mechanism upon D₂ binding. Details as in Fig 7

Schme 1

Schme 2

Schme 3

Chart 1

Key Constraints on HD Formation under N2/D2