Electrocatalytic CO₂ reduction catalyzed by nitrogenase MoFe and FeFe proteins

Highlights

• Nitrogenase FeFe and MoFe proteins were immobilized onto electrodes.

• Pyrene was found to stabilize the bioelectrodes.

• CO₂ conversion into formate was achieved by electrocatalysis using these bioelectrodes.

• FeFe protein showed higher activity for CO₂ reduction compared to the MoFe protein.

• FeMo-cofactor and FeFe-cofactor are the active sites for catalytic CO₂ reduction.

Abstract

Nitrogenases catalyze biological dinitrogen (N₂) reduction to ammonia (NH₃), and also reduce a number of non-physiological substrates, including carbon dioxide (CO₂) to formate (HCOO⁻) and methane (CH₄). Three versions of nitrogenase are known (Mo-, V-, and Fe-nitrogenase), each showing different reactivities towards various substrates. Normally, electrons for substrate reduction are delivered by the Fe protein component of nitrogenase, with energy coming from the hydrolysis of 2 ATP to 2 ADP + 2 Pi for each electron transferred. Recently, it has been demonstrated that energy and electrons can be delivered from an electrode to the catalytic nitrogenase MoFe-protein without the need for Fe protein or ATP hydrolysis. Here, it is demonstrated that both the MoFe- and FeFe-protein can be immobilized as a polymer layer on an electrode and that electron transfer mediated by cobaltocene can drive CO₂ reduction to formate in this system. It was also found that the FeFe-protein diverts a greater percentage of electrons to CO₂ reduction versus proton reduction compared to the MoFe-protein. Quantification of electron flow to products exhibited Faradaic efficiencies of CO₂ conversion to formate of 9% for MoFe protein and 32% for FeFe-protein, with the remaining electrons going to proton reduction to make H₂.

Introduction

Nitrogenases catalyze all biological N₂ reduction and are the major contributor of fixed N into the biogeochemical N cycle [1]. There are three known types of nitrogenase, designated the Mo-dependent-, the V-dependent-, and the Fe-dependent-nitrogenase [2], [3], [4], [5], [6]. Each nitrogenase is coded for by distinct genes, resulting in distinct catalytic systems [7], [8]. Nevertheless, the basic architecture for all the nitrogenases is similar; each requiring a distinct, but structurally similar, electron-donating Fe protein that supplies the electrons necessary for N₂ reduction in a series of one-electron transfers. In all of these systems, Fe protein delivers electrons to its corresponding catalytic partner, respectively designated the MoFe-, VFe-, or FeFe-protein, each of which house an electron mediator [8Fe-7S] cluster called P cluster and the active site metal cluster called FeMo-co, FeV-co, and FeFe-co, respectively [2], [4], [5], [9] (Fig. 1). The active site structures for FeMo-co and FeV-co are known and are very similar, with exchange of Mo for V [10], [11], [12], [13], [14], [15]. Further, the FeV-co contains a carbonate that replaces one of the bridging sulfides [10]. No X-ray crystal structure for the FeFe-co is available, but spectroscopic evidence indicates that it is similar to the other two cofactors with Fe substituting for V or Mo found in the other systems [16].

The catalytic cycle of nitrogenase involves the transient association of the Fe protein component with the catalytic component protein followed by the transfer of an electron from the Fe protein into the catalytic component with hydrolysis of two ATP molecules bound in the Fe protein to two ADP and two Pi [2], [9], [17], [18]. The oxidized and ADP containing Fe protein dissociates and the cycle must be repeated for each electron accumulated [18], [19], [20]. Thus, the energy for electron accumulation at the active site for substrate reduction comes from hydrolysis of two ATP (ΔG° 61 kJ/mol) [9], [21].

The reduction of substrates has been investigated for the Mo-dependent nitrogenase in considerable detail [2], [9], [17], [18], [22], [23], while very little is known about the mechanism of the V- and Fe-nitrogenases [3], [4], [5], [24]. For the Mo-nitrogenase, N₂ reduction requires 8 electrons and 8 protons, and thus the hydrolysis of 16 ATP (ΔG° 488 kJ/mol). One H₂ is formed for each N₂ reduced as an essential part of the catalytic mechanism [9], [17], [25], [26], [27], [28], [29]. In addition to reducing protons and N₂, Mo-nitrogenase has been shown to reduce a number of small molecules with double and triple bonds [2], [30]. Among the many molecules reduced by Mo-nitrogenase is carbon dioxide (CO₂) [31], [32], [33], [34], [35], [36]. There is considerable interest in reducing CO₂ as one way to mitigate CO₂ production from the burning of fossil fuels [37], [38], [39], [40], [41]. Mo-nitrogenase was shown to reduce CO₂ by two electrons and two protons to CO [31] and formic acid [33] and by 8 electrons and 8 protons to methane (CH₄) [32], [34] (Eqs. (1), (2), (3)).

$${\mathrm{CO}}_2+{\text{2e}}^{-}+{\text{2H}}^{+}\to \mathrm{CO}+{\mathrm{H}}_2\mathrm{O}$$

$${\mathrm{CO}}_2+{\text{2e}}^{-}+{\text{2H}}^{+}\to \text{HCOOH}$$

$${\mathrm{CO}}_2+{\text{8e}}^{-}+{\text{8H}}^{+}\to {\mathrm{CH}}_4+{\text{2H}}_2\mathrm{O}$$

The V-nitrogenase has also been shown to reduce CO₂ to CH₄ and C₂ and C₃ hydrocarbons [35], [36]. Very recently, both in vivo and in vitro studies with the Fe-nitrogenase indicate that this enzyme shows the highest CO₂ reduction to CH₄ of the three nitrogenases [42]. This study demonstrated the reduction of CO₂ to CH₄ when driven by Fe protein and ATP for both purified Fe-nitrogenase component proteins and in whole cells.

As a way to bypass the energy requirement of ATP hydrolysis necessary for electron delivery by the Fe protein, an electrochemical method has been developed to allow the MoFe-protein to be immobilized in a polymer layer on a glassy carbon electrode with cobaltocene mediated electrocatalysis [43]. This method is a promising new approach to achieve small molecule reduction catalyzed by a nitrogenase component [43] and as a way to probe the mechanism of nitrogenase catalysis [44]. Using the MoFe-protein, it has been demonstrated that protons, nitrite, and azide can be reduced by electrons and energy supplied by an electrode [43], [44]. Here, immobilization of the MoFe- and FeFe-proteins on a pyrene-stabilized bioelectrode was achieved to show catalytic reduction of CO₂ to formate. The FeFe-protein exhibits a higher Faradaic efficiency for CO₂ reduction when compared to the MoFe protein. The electron partition ratios between H₂ and formate production catalyzed by MoFe and FeFe protein are about the same as those obtained in the Fe protein supported reaction using ATP hydrolysis energy. These results provide evidence that the FeMo-co and FeFe-co in the corresponding nitrogenase proteins are the catalytic sites during electrocatalysis.