Nitrogenases are complex metalloenzymes that catalyze the reduction of nitrogen gas (N2) to form bioavailable ammonia (NH3) under ambient conditions. These enzymes can also reduce carbon monoxide (CO), a potent pollutant gas, into small hydrocarbon products. The molybdenum (Mo) and vanadium (V) nitrogenases are two homologous members of the nitrogenase family that are both comprised of a multi-subunit protein scaffold and a complex active site metallocofactor made up of an iron-sulfur core and an organic ligand.

The Mo- and V-nitrogenases demonstrate different capabilities with respect to the reduction of N2 and CO. Specifically, the Mo-nitrogenase is about twice as active as V-nitrogenase at producing NH3 from N2. Interestingly, the V-nitrogenase is ~600-fold more active than the Mo-nitrogenase in reducing CO to hydrocarbon products. These reactivity differences likely stem from dissimilarities in the protein scaffolds and active site cofactor properties of the Mo- and V-nitrogenase variants, and the work described in this dissertation probes the roles of these aspects through biochemical and spectroscopic characterization. A hybrid protein made up of the V-nitrogenase protein scaffold and the Mo-nitrogenase cofactor was created in vivo through the manipulation of bacterial growth conditions. The substrate reduction activity of the purified hybrid protein indicates that the protein scaffold of V-nitrogenase is the primary contributor to the observed differences in its CO-reducing capabilities. Additional work was conducted with a mutant of V-nitrogenase that produced an active site cofactor with an alternative organic ligand. The modified V-nitrogenase and its isolated cofactor demonstrated an increased production of NH3 from N2, which indicates that the organic ligand plays an important role in protonation of various small molecule substrates. The results of this work highlight the importance of the V-nitrogenase protein scaffold and its organic ligand in the production of hydrocarbon products from CO and NH3 from N2.

Nitrogenase catalyzes the enzymatic reduction of dinitrogen (N2) into ammonia (NH3) at ambient temperatures and pressures. N2 reduction by molybdenum (Mo) nitrogenase requires the catalytic molybdenum-iron protein (NifDK) and its reductase, the iron protein (NifH). In addition to its role in N2 reduction, NifH proteins can independently reduce carbon dioxide (CO2) to carbon monoxide (CO) and short-chain hydrocarbons. Nitrogenase proteins contain iron-sulfur cluster cofactors that are crucial to these activities. NifH uses a [Fe4S4] cluster for transferring electrons to NifDK and for independently reducing C1, whereas the site of N2 reduction on NifDK is the complex [MoFe7S9C-homocitrate] M-cluster. M-cluster, however, is synthesized ex-situ by other proteins from [Fe4S4] cluster precursors. The radical S-adenosylmethionine enzyme NifB uses [Fe4S4] cluster units for the biosynthesis of the [Fe8S9C] L-cluster, the immediate precursor to M-cluster. NifH and NifB rely on [Fe4S4] clusters to perform very different functions, mediated by their protein environments in ways that aren’t well understood. The focus of this dissertation is on how the distinct protein environments of NifH and NifB proteins modulate the reactivity of their [Fe4S4] clusters.

Three [Fe4S4] cluster modules, and their amino acid ligands, were identified on NifB by spectroscopic and biochemical characterization of mutant proteins. Interestingly, a histidine residue that serves as a transient nitrogen ligand to a [Fe4S4] precursor cluster was crucial for L-cluster assembly. Mutation of this ligand was found to interfere with the structural transformations of iron-sulfur clusters of NifB.

To understand how the [Fe4S4] cluster at the active site of NifH catalyzes C1 substrate reductions, I solved the structures of two NifH proteins by X-ray crystallography. The first was NifH in the all-ferrous [Fe4S4]0 state from Azotobacter vinelandii (AvNifH). This form, produced by treating the protein with a strong reductant, is the most active in substrate reduction. The structure solved from a crystal in this form revealed two arginine residues that have an asymmetric arrangement that can serve to capture and protonate substrates. Another NifH protein, Methanosarcina acetivorans (MaNifH), can reduce CO2 and CO to short-chain hydrocarbons. MaNifH in the [Fe4S4]1+ state, produced by the presence of dithionite, demonstrates a similar asymmetric arrangement of a pair of arginines. These crystallization solution also contained bicarbonate, an alternative CO2 source. Excitingly, additional density, potentially a captured substrate, was observed near the [Fe4S4] cluster. These structures provided parameters for density functional theory (DFT) calculations provided details that enabled propose mechanisms for CO2 and CO reduction by [Fe4S4] clusters.

Together, these results contribute towards building a mechanistic model of a novel [Fe4S4]-based system for converting greenhouse gases into hydrocarbon fuels, and for understanding how [Fe4S4] clusters are used to synthesize more complex metallocofactors.

Nitrogenases are complex metalloenzymes capable of catalyzing two of the most challenging reactions in Nature: the reduction of atmospheric dinitrogen to ammonia and the reduction of carbon monoxide (CO) to hydrocarbons. The Azotobacter vinelandii molybdenum (Mo) and vanadium (V)-nitrogenases are homologous systems consisting of two components: the reductase component (Fe protein) and the catalytic component (MoFe or VFe protein, respectively). The reductase component contains a [Fe4S4]-cluster, whereas the catalytic component contains two unique metal clusters, the P-cluster and the M or V cluster for Mo- and V-nitrogenase, respectively.

This dissertation focuses on the conversion of small carbon compounds such as the toxic exhaust CO and the greenhouse gas carbon dioxide (CO2) by nitrogenase proteins. The CO2 reducing capability of Mo- and V-nitrogenase was investigated under physiologically relevant assay conditions. Indeed, both nitrogenases reduce CO2 to CO, but only V nitrogenase reduces CO2 to hydrocarbons. These studies demonstrated that V-nitrogenase directly couples CO2 or CO2-derived species to hydrocarbon chains.

Building on this, the efficiency of this reaction was improved by using the strong reductant europium(II) diethylenetriaminepentaacetate, rendering the ATP-dependent electron transfer obsolete and increasing the hydrocarbon chain length as well as the yield.

Excitingly, the reductase component of V- and Mo-nitrogenase alone can reduce CO2 to CO. Strikingly, the reverse reaction, the oxidation of CO to CO2, is catalyzed under oxidizing conditions by Fe protein. The interconversion of CO and CO2 establishes the reductase component housing a [Fe4S4]-cluster as simple Fe/S-based mimic of CO dehydrogenase. The physiological relevance of this reduction is suggested by the observation that A. vinelandii strains expressing only the Fe protein component form CO2-derived CO in vivo.

These discoveries led to the question of whether V-nitrogenase expressing A. vinelandii strains could also in vivo form hydrocarbons from CO or CO2. Indeed, this strain can convert CO to hydrocarbons in vivo. Furthermore, this process is a secondary metabolism since the carbon from CO is not incorporated into the cell mass, but released. Coupling the two in vivo processes could result in a biofuel process that catalytically converts CO2 into combustible hydrocarbon fuels.

Nitrogen fixation, the reduction of N2 to NH3, is a key step in the global nitrogen cycle. Biological nitrogen fixation (BNF) is carried out by diazotrophic microbes, which utilize the metalloenzyme nitrogenase to perform this chemical transformation. BNF by Mo-nitrogenase requires both the catalytic component, NifDK, and the reductase component, NifH. NifDK is the site of N2 reduction and contains two of the most complex metallocofactors known to nature: the [F8S7] P-cluster and the [MoFe7S9C-homocitrate] M-cluster. NifH is a homodimer that contains a single [Fe4S4] cluster and plays a crucial role in the transfer of electrons to NifDK and the assembly of NifDK’s metallocofactors and catalyzes the reduction of C1 substrates. Section 1 of this dissertation details progress toward the heterologous expression of functional nitrogenase and related proteins. Section 2 of this dissertation focuses on a pair of studies that probe the reactivity of NifH homologs.Endowing cereals with nitrogen fixation activity is considered a holy grail. However, previous attempts in the literature fail to demonstrate the expression of functional nitrogenase and related proteins in a heterologous host. Section 1 details work done toward heterologously expressing functional nitrogenase and related proteins in the model organism Escherichia coli. NifH is a multipurpose actor that plays an essential role in nitrogen fixation and may have its own industrially relevant applications independent of NifDK. However, the relationship between NifH’s reactivity and redox properties has not been thoroughly studied. Section 2 details a pair of studies that employ novel methods, which can be applied broadly to the study of [Fe4S4] containing proteins, to probe the reactivity of NifH and the redox properties of its metallocluster. In study 1, the disparate reactivities of two homologs of NifH toward the reduction of C1 substrates in the all-ferrous state are explored and related to differences in the redox potential of the all-ferrous state. In study 2, a synthetic [Fe4Se4] compound is incorporated into NifH, showing redox properties distinct from its [Fe4S4] counterpart that can be used to distinguish the differential redox requirements of nitrogenase catalysis and assembly.

The iron (Fe) protein is one of two components involved in the biological fixation of dinitrogen (N2) to bioavailable ammonia (NH3) by nitrogenase. The physiological roles of the Fe protein include transferring electrons to the catalytic component and assisting in the bioassembly of the two metalloclusters found within the catalytic component. The Fe protein has also been shown to adventitiously reduce CO2. The goal of this work has been to heterologously express and characterize Fe protein homologs according to their physiological and adventitious functions. This dissertation is divided into two sections, comprising multiple chapters, based on these functions.

Section 1 begins by comparing the molybdenum (Mo-) and vanadium (V-) nitrogenase Fe proteins from Methanosarcina acetivorans (MaNifH and MaVnfH, respectively) to the Fe proteins from the model organism Azotobacter vinelandii (AvNifH and AvVnfH, respectively). Characterization includes comparing the size of the proteins, the behavior of the [Fe4S4] cluster of the protein, and the ability of the homologs to assist in the bioassembly of the catalytic component’s metalloclusters. Next, there is a discussion about a hybrid nitrogenase system, created by pairing MaVnfH with the catalytic component of A. vinelandii, which resulted in the ability to trap the alternate V-nitrogenase substrate, CO, on the catalytic cofactor. Also included are a brief chapter that reports the MgADP-bound crystal structure of MaVnfH and a miscellaneous chapter, which contains preliminary results regarding the characterization of five additional Mo-nitrogenase Fe protein homologs.

The first chapter of Section 2 reports that MaNifH is able to reduce CO2 beyond CO and also generate hydrocarbons, which is further than the A. vinelandii counterparts. Consequently, a CO reduction assay was created, and this section focuses on the ability of Fe protein homologs to reduce both CO2 and CO. It was determined that CO2 and CO reduction is an intrinsic property of [Fe4S4] clusters. The final chapter compares the product profiles of different Fe protein homologs and compares the crystal structures of these Fe proteins to gain insights into what may cause these differences.