Sequential and differential interaction of assembly factors during nitrogenase MoFe protein maturation

Abstract

Nitrogenases reduce atmospheric nitrogen, yielding the basic inorganic molecule ammonia. The nitrogenase MoFe protein contains two cofactors, a [7Fe-9S-Mo-C-homocitrate] active-site species, designated FeMo-cofactor, and a [8Fe-7S] electron-transfer mediator called P-cluster. Both cofactors are essential for molybdenum-dependent nitrogenase catalysis in the nitrogen-fixing bacterium Azotobacter vinelandii We show here that three proteins, NafH, NifW, and NifZ, copurify with MoFe protein produced by an A. vinelandii strain deficient in both FeMo-cofactor formation and P-cluster maturation. In contrast, two different proteins, NifY and NafY, copurified with MoFe protein deficient only in FeMo-cofactor formation. We refer to proteins associated with immature MoFe protein in the following as "assembly factors." Copurifications of such assembly factors with MoFe protein produced in different genetic backgrounds revealed their sequential and differential interactions with MoFe protein during the maturation process. We found that these interactions occur in the order NafH, NifW, NifZ, and NafY/NifY. Interactions of NafH, NifW, and NifZ with immature forms of MoFe protein preceded completion of P-cluster maturation, whereas interaction of NafY/NifY preceded FeMo-cofactor insertion. Because each assembly factor could independently bind an immature form of MoFe protein, we propose that subpopulations of MoFe protein-assembly factor complexes represent MoFe protein captured at different stages of a sequential maturation process. This suggestion was supported by separate isolation of three such complexes, MoFe protein-NafY, MoFe protein-NifY, and MoFe protein-NifW. We conclude that factors involved in MoFe protein maturation sequentially bind and dissociate in a dynamic process involving several MoFe protein conformational states.

Keywords: FeMo-cofactor; MoFe protein; P-cluster; electron paramagnetic resonance (EPR); nitrogen fixation; nitrogenase; protein assembly; protein purification; reductase.

Figure 1.

Schematic representation of molybdenum-dependent nitrogenase and its associated metal-containing cofactors. Fe protein subunits (encoded by nifH) are shown in light brown, the MoFe protein α-subunit (encoded by nifD) is shown in green, and the MoFe protein β-subunit (encoded by nifK) is shown in blue. Atoms in metal-containing cofactors are indicated as follows: iron (rust), sulfur (yellow), molybdenum (magenta), carbon (gray), and oxygen (red).

Figure 2.

Organization of the 55 genes associated with the molybdenum-dependent nitrogen fixation system from A. vinelandii. Numbers refer to gene designations used in the original annotation of the A. vinelandii genome (54), and arrows indicate transcription units. The nitrogenase-encoding genes (nifH encoding the Fe protein; nifD and nifK encoding the MoFe protein α and β subunits, respectively) are filled in blue. The nafH, nifW, and nifZ genes are filled in yellow; and the nifY and nafY genes are filled in light green. The seven genes whose products are strictly required to sustain molybdenum-dependent nitrogen fixation in A. vinelandii are indicated by red dots. Genes whose products are involved in mobilizing iron and sulfur for nitrogenase-associated metallo-cluster formation are indicated by gray dots, and the gene encoding homocitrate synthase, which supplies the organic acid portion of FeMo-cofactor, is indicated by a green dot. The gene designations nif (nitrogen fixation), naf (nitrogenase-associated factor), rnf (rhodobacter nitrogen fixation), and nfa (nitrogen fixation associated) either have historical origins or have been given formal genetic designations here. Transcriptome analyses have revealed the elevated expression of all of these genes in response to molybdenum-dependent nitrogen-fixing conditions (11, 16). Gene letters do not necessarily indicate similar functions. For example, there is no structural or functional similarity between the products of nifH and nafH.

Figure 3.

UV-visible spectra of Strep-tagged MoFe protein prepared from different genetic backgrounds. Samples in A were prepared in the absence of the reducing agent Na₂S₂O₄, and samples in B were prepared in the presence of 2 mm Na₂S₂O₄. Spectra in A could be reversibly converted to the spectra shown in B by the addition of excess Na₂S₂O₄, and spectra in B could be reversibly converted to the spectra shown in A by the addition of an excess of the oxidizing reagent indigo disulfonate. All proteins were purified using the Strep-tag affinity method.

Figure 4.

X-band EPR spectra of resting state Strep-tagged MoFe proteins purified from different A. vinelandii strains. Black trace, Strep- tagged MoFe protein produced by WT; red trace, Strep-tagged MoFe protein produced by a ΔnifB strain; green trace, Strep-tagged MoFe protein produced by a ΔnifH strain; blue trace, Strep-tagged MoFe protein produced by a ΔnifH ΔnifB strain; purple trace, Strep-tagged MoFe protein produced by a ΔnifH ΔvnfH strain. All samples are Na₂S₂O₄-reduced. The top panel shows the full EPR spectrum of each protein with EPR inflections associated with FeMo-cofactor indicated by a horizontal arrow. The bottom panel shows low-field region spectra, highlighting the EPR signatures (g = 4.32 and 3.64) of the FeMo-cofactor in this region. All spectra were normalized to a final protein concentration of 43.5 μm. EPR parameters are described under “Materials and methods.”

Figure 5.

SDS-PAGE of Strep-tagged MoFe protein samples prepared from various genetic backgrounds using Streptactin affinity columns. Protein samples shown here and in other figures were separated using a 4% acrylamide stacking gel and 15% acrylamide running gel and then stained with Coomassie Brilliant Blue. Protein standards in the left panel include phosphorylase B (97.4 kDa), BSA (66.2 kDa), ovalbumin (45.0 kDa), carbonic anhydrase (31.0 kDa), soybean trypsin inhibitor (21.5 kDa), and lysozyme (14.4 kDa). The inset located adjacent to the sample showing proteins that co-purify with MoFe protein produced in a ΔnifH background represents an overloaded sample to clearly show co-purification of NafH. Arrows indicate MoFe protein subunits and co-purifying nitrogen fixation-related proteins and also indicate prominent biotin-binding proteins unrelated to nitrogen fixation, acetate carboxylase subunit (AccA) and pyruvate carboxylase subunit (PycA), that are independently captured by the affinity-purification method. Higher levels of PycA and AccA in certain samples reflect the relatively lower abundance of MoFe protein subunits in extracts of those samples. Very light bands recognized in some samples were identified as various MoFe protein subunit degradation products. All proteins were identified by MS as described under “Materials and methods.” Neither MoFe protein nor any other nitrogen fixation-related protein binds to the Streptactin column if there is no Strep-tag on the MoFe protein.

Figure 6.

Susceptibility of the α-Cys²⁷⁵ residue of MoFe protein produced in different genetic backgrounds and co-purification of NifY and NafY with MoFe protein having Ala or Gln substituted for either α-Cys²⁷⁵ or α-His⁴⁴², respectively. A, structure of FeMo-cofactor showing anchoring α-Cys²⁷⁵ and α-His⁴⁴² residues. B (top), MoFe protein samples purified from different genetic backgrounds and treated with the fluorescent alkylating reagent I-AEDANS before SDS-PAGE and visualized by UV light illumination before Coomassie Brilliant Blue staining (18). Note that B is a composite of a single gel for which a lane located between the ΔnifH sample and ΔnifB sample has been excised. B (bottom), same samples as shown in the top after staining with Coomassie Brilliant Blue. Identities of affinity tags used to assist purification, Strep-tag or His-tag, are indicated by superscripts. C, co-purification of NifY and NafY with Strep-tag affinity-purified MoFe protein produced in different genetic backgrounds.

Figure 7.

SDS-PAGE of affinity-purified MoFe protein using immobilized Strep-tagged NafY or Strep-tagged NifY as bait. Purified Strep-tagged NafY or NifY was immobilized on a Streptactin column, and crude extracts prepared from strains deleted for nifH, nifE, or nifB and expressing MoFe protein that does not have a Strep-tag were separately passed over NafY- or NifY-charged columns, washed with buffer, and eluted using biotin-containing buffer, as described under “Materials and methods.” A displays the results of one-step NafY-directed affinity purification of MoFe protein subunits without further processing. The prominent band located below NifK in the ΔnifB sample is a subunit of the biotin-binding protein pyruvate carboxylase (PycA), which is enriched as a result of the relatively low level of MoFe protein accumulated in crude extracts of that sample. B corresponds to the same ΔnifE sample shown in A, but after further purification to remove excess NafY using gel exclusion chromatography as described under “Materials and methods.” C displays the result of one-step NifY-directed affinity purification of MoFe protein subunits without further processing.

Figure 8.

Sequential and differential interaction of NafH, NifW, and NifZ with MoFe protein during the process of its maturation. A, summary of proteins that co-purify with Strep-tagged MoFe proteins produced in different genetic backgrounds. B, an example of experimental results used to generate the data summarized in A. B, first lane, only NafH co-purifies with Strep-tagged MoFe protein produced by a ΔnifW strain (AccA is a subunit of the biotin-binding protein acetate carboxylase, unrelated to nitrogen fixation, that also separately binds to the Streptactin column used for purification). Second lane, NafH and NifZ co-purify with Strep-tagged MoFe protein produced by a ΔnifH ΔnifW strain. Although no NifW is present in these samples, the position of where NifW would migrate on the gel, if present, is indicated. C, pathway of NafH, NifW, NifZ, and NifH involvement in MoFe protein maturation deduced from data summarized in A.

Figure 9.

SDS-PAGE of affinity-purified MoFe proteins using NifW as bait. Samples were prepared and processed as described under “Materials and methods” as well as in the legend to Fig. 6. A, one-step NifW-directed affinity purification of nontagged MoFe protein prepared from extracts of either a ΔnifH strain or a ΔnifB strain. B corresponds to the same ΔnifH sample shown in A, but after further purification by gel exclusion chromatography. Note that the C terminus of NifW produced in E. coli is subject to C-terminal proteolytic cleavage, giving rise to multiple truncated species. This phenomenon has also been previously observed for NifW samples prepared from A. vinelandii (42).