Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 May 29;65(29):e8078898. doi: 10.1002/anie.8078898

A Non‐Covalent [4Fe–4S]/[2Fe] Interface in HydF Guides [FeFe]‐Hydrogenase Maturation

Giorgio Caserta 1,✉, Princess R Cabotaje 2,3,4, Armel T Waffo 1, Deepak Prajapat 5, Ilya Sergueev 5, Stefan Frielingsdorf 1, Gustav Berggren 2,✉
PMCID: PMC13360658  PMID: 42212729

ABSTRACT

Maturation of [FeFe]‐hydrogenase depends on the HydF maturase, a [4Fe–4S]‐containing scaffold protein that assembles and delivers the organometallic [2Fe] subsite of the H‐cluster. Despite extensive research, the function of the HydF [4Fe–4S] cluster and its interaction with the [2Fe] cofactor remains unresolved, with conflicting evidence regarding cyanide linkage isomerism and the functional role of the cubane. Using 57Fe nuclear resonance vibrational spectroscopy in combination with selective isotopic labeling, we show that the [2Fe] subsite binds adjacent to the [4Fe–4S] cluster without forming a covalent cyanide bridge, while remaining electronically coupled. Complementary protein structure predictions support this arrangement and reconcile prior spectroscopic, mutagenesis, and structural studies. Extending this framework to earlier biosynthetic steps, additional structure predictions suggest that the [4Fe–4S] cluster contributes to assembly of the CH2–NH–CH2 bridge of the [2Fe] site via interactions with the lipoate cofactor of the aminomethyl‐lipoyl H‐protein, thereby positioning reactive components in proximity to the [2Fe] precursor. Together, these results provide a coherent structural and mechanistic framework for HydF function across distinct stages of H‐cluster biosynthesis.

Keywords: biosynthesis, hydrogenase, iron‐sulfur clusters, metalloenzymes, nrvs

Bridged or not? 57Fe NRVS reveals a non‐covalent, electronically coupled interaction between the HydF [4Fe–4S] cluster and the [2Fe] precursor. Boltz‐2 structure predictions suggest that the cubane organizes lipoate‐dependent Hmet chemistry, positioning intermediates for assembly of the CH2–NH–CH2 bridge during H‐cluster biosynthesis.

graphic file with name ANIE-65-e8078898-g001.jpg

1. Introduction

[FeFe]‐hydrogenases catalyze the reversible interconversion of protons and dihydrogen using a sophisticated organometallic active site known as the H‐cluster [1, 2]. This cofactor comprises a canonical [4Fe–4S] cubane coordinated by four conserved cysteine residues and a diiron catalytic subsite ([2Fe]H) [3]. The [2Fe]H subsite is covalently linked to the [4Fe–4S] cluster via a cysteine thiolate (Figure 1) and is coordinated by three carbon monoxide (CO) ligands, two cyanide (CN–) ligands, and an azadithiolate (ADT, also referred to as dithiomethylamine, DTMA) bridge [4]. Whereas the [4Fe–4S]H cluster and additional accessory [Fe–S] centers (F‐clusters) are assembled by the housekeeping iron–sulfur cluster machinery [5], biosynthesis of the [2Fe]H subsite requires a dedicated maturation pathway involving the HydE, HydF, and HydG proteins (Figure 1) [6, 7]. Maturation is initiated by the radical S‐adenosyl‐L‐methionine (rSAM) enzyme HydG, which catalyzes the conversion of tyrosine into CO and CN– ligands and generates an FeII(κ3‐Cys)(CO)2(CN) synthon [8, 9, 10]. This iron precursor, commonly referred to as complex B, is transferred from HydG to a second rSAM enzyme [11, 12], HydE, which mediates reductive activation of the FeII synthon to form an adenosylated low‐spin FeI intermediate [13, 14]. This species subsequently dimerizes to yield the [FeIFeI] complex [Fe2(HS)2(CO)4(CN)2]2–, denoted [2Fe]E [15]. Rauchfuss, Britt, and coworkers further showed that synthetic [2Fe]E efficiently matures apo‐HydA1 from Chlamydomonas reinhardtii in the presence of the third maturase HydF and Escherichia coli cell lysate [16]. Building on this work, Broderick and coworkers identified a fourth essential component of the maturation pathway: the aminomethyl‐lipoyl H‐protein (Hmet) of the glycine cleavage system. Hmet catalyzes installation of the CH2–NH–CH2 moiety across the two bridging sulfides of [2Fe]E, generating the fully functional diiron precursor [Fe2(CO)4(CN)2(adt)]2–, termed [2Fe]F [17, 18, 19].

FIGURE 1.

FIGURE 1

Open in a new tab

Schematic overview of the [FeFe]‐hydrogenase maturation pathway and proposed interaction between the HydF cofactors. Maturation is initiated by the rSAM enzyme HydG, which converts tyrosine into CO and CN– ligands and generates an FeII(κ3‐Cys)(CO)2(CN) synthon (complex B). Complex B is then transferred to the rSAM enzyme HydE, where reductive activation yields a low‐spin FeI intermediate that dimerizes to form [2Fe]E. In the presence of HydF and the aminomethyl‐lipoyl H‐protein (Hmet), [2Fe]E is further converted into the fully functional [2Fe]F precursor via installation of the DTMA bridge. Earlier 15N HYSCORE and 13C ESEEM studies on [2Fe(pdt)]F(13CN)‐HydF suggested a close interaction between the 13C nucleus of a labeled cyanide ligand and the unpaired electron of the [4Fe–4S] cluster, originally interpreted as a cyanide‐mediated contact potentially involving linkage isomerism [20].

Importantly, installation of the azadithiolate bridge occurs exclusively when the [2Fe]E precursor is bound to HydF, underscoring the central role of this protein in [FeFe]‐hydrogenase maturation [21]. HydF is a guanosine triphosphatase (GTPase) that harbors a [4Fe–4S] cluster, which has been extensively characterized [22, 23, 24, 25]. An X‐ray crystal structure of HydF from Thermosipho melanesiensis (TmeHydF) revealed that this cluster is coordinated by three cysteine residues and a glutamate [26]. Despite these structural insights, the functional role of the [4Fe–4S] cluster remains the subject of ongoing debate [17, 27]. In 2013, we showed that synthetic [2Fe]F analogues bearing dithiolate bridges with different bridgehead groups (CH2, NH, or O) can be loaded onto HydF from Thermotoga maritima (TmHydF) and subsequently support maturation of apo‐CrHydA1 [20]. Advanced EPR studies suggested a close interaction between the synthetic diiron cofactor and the HydF [4Fe–4S] cluster, which was interpreted as involving a cyanide‐mediated contact, potentially consistent with linkage isomerism (Figure 1) [20, 26]. In this context, the Darensbourg group demonstrated that cyanide flipping can occur in cyanide‐bridged diiron model systems such as Fe–CN/NC–CuI/II complexes [28, 29]. Later, Happe and coworkers reported that the cubane is dispensable for the final stages of the maturation, as apo‐HydF loaded with synthetic [2Fe]F was still able to activate apo‐CrHydA1 [30]. Conversely, Broderick and coworkers concluded that the [4Fe–4S] cluster is essential, demonstrating that only cluster‐containing HydF supports the final stages of [2Fe]F biosynthesis [31]. These latter findings are consistent with earlier work by King and coworkers [32], which likewise established the importance of the cluster for in vivo H‐cluster assembly. Thus, functional studies have yielded conflicting conclusions regarding the importance of the HydF [4Fe–4S] cluster.

Structurally, a binding site for the [2Fe]F precursor was first identified by the Fontecave group [26, 33]. The picture was later refined by Happe and coworkers [34], which pinpointed residues critical for binding and assembly of [2Fe]F. However, neither these static structural models nor prior spectroscopic data could directly address whether the interaction between the [4Fe–4S] cluster and the [2Fe]F subsite involves a direct covalent (cyanide‐bridged) linkage or a non‐covalent association. This distinction is critical, as it defines fundamentally different functional roles for the cluster, ranging from direct coordination of the [2Fe] subsite to a purely structural and electronic contribution, and is therefore essential for defining the function of HydF within the broader framework of H‐cluster maturation. To directly probe the interaction between the [4Fe–4S] cluster and the [2Fe]F subsite of HydF, we employed nuclear resonance vibrational spectroscopy (NRVS) in combination with structure predictions and protein–ligand docking calculations. NRVS is a synchrotron‐based technique that selectively probes vibrational modes of Mössbauer‐active nuclei (e.g., 57Fe) and is exquisitely sensitive to changes in bonding, geometry, and oxidation state of iron atoms [35, 36]. The method has previously provided detailed structural insights into a wide range of iron‐containing enzymes, including hydrogenases [37, 38, 39], nitrogenases [40], heme, [41] and nonheme systems [42, 43]. As such, NRVS provides a unique experimental handle to directly assess whether—and how—the [4Fe–4S] cluster of HydF interacts with the [2Fe]F subsite, beyond the reach of static structural models and conventional ensemble spectroscopies.

2. Results and Discussion

Here, we report NRVS spectra of holo‐TmHydF in which either the [4Fe–4S] cluster or the diiron subsite was selectively labeled with 57Fe. Apo‐TmHydF was isolated as described previously [44], and its [Fe–S] cluster was semi‐synthetically reconstituted under strictly anaerobic conditions (see methods in Supporting Information). UV–vis spectroscopy (Figure S1) and iron quantification (4.7 ±  0.2 Fe/protein) following size‐exclusion chromatography confirmed the assembly of a [4Fe–4S] cluster in TmHydF, hereafter termed FeS‐HydF. To distinguish isotopically enriched samples from those prepared with iron of natural abundance, the former are denoted 57FeS−HydF, whereas the latter are referred to as FeS‐HydF. 57FeS‐HydF was analyzed as prepared, treated with an excess of sodium dithionite to probe redox‐dependent spectral changes, and a third aliquot was incubated anaerobically with [2Fe]F to yield [2Fe]F–57FeS‐HydF, denoted holo‐HydF. Unbound [2Fe]F was removed prior to spectroscopic analyses. Infrared spectroscopy of holo‐HydF (Figure S2) displayed the characteristic CO/CN– stretching vibrations associated with protein‐bound [2Fe]F, while UV–visible spectra revealed the appearance of a new absorption band at 350 nm, diagnostic of the diiron subsite within the HydF scaffold (Figure S3). Figure 2a compares the NRVS spectra of as‐prepared 57FeS‐HydF and holo‐HydF. Both spectra exhibit vibrational features characteristic of a [4Fe–4S] cluster in the 2+ oxidation state, including intense Fe–S stretching modes of the bridging (Sb) sulfides and terminal thiolate ligands (St) in the ∼250–400 cm−1 region [35]. Upon reduction these modes shift by approximately 10–30 cm−1 to lower energies (Figure S4), consistent with weakening of Fe–S bonds, whereas lower‐energy S–Fe–S bending and twisting (100–200 cm−1), and acoustic modes (<100 cm−1) remain largely invariant with respect to the cluster oxidation state.

FIGURE 2.

FIGURE 2

Open in a new tab

NRVS characterization of selectively labeled [4Fe–4S] and [2Fe]F sites in HydF. (a) NRVS spectra of 57FeS‐TmHydF (black trace) and [2Fe]F–57FeS‐TmHydF (blue trace) exclude the presence of a covalent cyanide bridge or cyanide linkage isomerization between the two iron sites. (b) NRVS spectra of [257Fe]F‐FeS‐TmHydF (holo‐HydF, black trace) and [257Fe]F‐TmHydF (apo‐HydF, cyan trace) reveal a similar binding mode of [2Fe]F in holo‐ and apo‐HydF.

Notably, the NRVS spectrum of holo‐HydF does not exhibit absorptions in the frequency range expected for FeFeS–CN2Fe stretching modes, which would be diagnostic of cyanide linkage isomerization involving direct coordination to the [4Fe–4S] cluster. The close similarity between the holo‐HydF and 57FeS‐HydF spectra therefore argues against the presence of a covalent cyanide bridge between the [2Fe]F subsite and the cubane (Figure 1). Instead, comparison of the spectra reveals additional low‐frequency features in holo‐HydF at 85 and 200 cm−1, which fall within the region associated with torsional and breathing modes of the [4Fe–4S] cluster, as well as collective motions of the cubane coupled to the protein backbone, consistent with weak, non‐covalent coupling between the two metal sites. In addition, Fe–S stretching vibrations in 57FeS‐HydF are slightly red‐shifted upon [2Fe]F loading (356/351, 384/383 cm−1), consistent with weakened Fe─S bonds and increased electron density at the cubane. Holo‐HydF displays also an enhanced absorption at ∼318 cm−1 within the Fe─S stretching region. This increased intensity is consistent with a redistribution of vibrational modes within the Fe─S stretching manifold upon [2Fe]F binding. Proximity of the diiron subsite likely perturbs the local electrostatic and structural environment of the cubane, altering normal‐mode mixing and enhancing the 57Fe contribution to specific Fe─S stretches without major frequency shifts. Consistent with this interpretation, [2Fe]F binding does not substantially perturb the EPR signature of the [4Fe–4S]+ cluster [21, 26], yet reduction of holo‐HydF induces a slight blueshift of the dominant CO stretching band of [2Fe]F [20, 26], indicative of electronic communication between the two metal sites in the absence of covalent linkage.

FeS−HydF was then reacted with [257Fe]F to monitor the complete set of Fe–ligand vibrations of the di‐iron core when bound to the protein scaffold. As a control, we also reacted apo‐HydF with [257Fe]F. Iron quantitation, 1.6 ± 0.1 Fe/protein, and UV–visible spectroscopy (Figure S5) confirmed incorporation of [257Fe]F. NRVS spectra for these two HydF samples are shown in Figure 2b. Holo‐HydF exhibits intense bands at 558, 591, and 616 cm−1, primarily associated with Fe─CO vibrational modes, accompanied by weaker absorptions at ca. 450 and 500 cm−1 attributable mainly to Fe–CN vibrations [45]. Owing to selective 57Fe labeling of the [2Fe] core, additional low‐frequency features at 123, 196, 297, and 337 cm−1 could be resolved that would otherwise be masked by contributions from the [4Fe–4S] cluster.

Importantly, the NRVS data of apo‐HydF closely resemble those of holo‐HydF, demonstrating that [2Fe]F can be successfully loaded into apo‐HydF in a binding mode analogous to that in holo‐HydF [30, 34]. Evidently, this mode of incorporation is consistent with the absence of cyanide linkage isomerism, as apo‐HydF lacks a [4Fe–4S] cluster. FeFeS–NC2Fe coordination in holo‐HydF appears also unlikely [46], as such a linkage would be expected to induce substantial perturbations in both the [4Fe–4S] and [2Fe]F vibrational regions as well as measurable differences between apo‐ and holo‐HydF spectra (Figure 2), neither of which is observed.

However, holo‐HydF spectra display sharper and better‐resolved vibrational features, particularly in the Fe–CO stretching region, which may be indicative of a more homogeneous [2Fe]F binding. Taken together, these results suggest that the [4Fe–4S] cluster could contribute to the formation of a more rigid binding cavity for the diiron subsite.

While recent docking studies provided valuable initial insight into [2Fe]E/F binding in HydF [31, 34], these models were based on x‐ray coordinates of TmeHydF lacking the [4Fe–4S] cluster. Given the proximity of the cubane to the proposed [2Fe] binding cavity, we sought a more realistic representation of the [4Fe–4S]:[2Fe]F interaction. To this end, we employed the recently released Boltz‐2 machine‐learning model [47], which achieves AlphaFold3‐level accuracy in protein–ligand interactions [48], including systems containing complex metal cofactors such as [Fe–S] clusters. A structural model of TmeHydF containing bound GTP and a [4Fe–4S] cluster is shown in Figure S6. The model reveals a well‐defined, positively charged pocket suitable for [2Fe]E/F binding, closely matching the binding site inferred from the crystallographic coordinates of TmeHydF (Figure 3ab) [26, 31, 34]. Notably, whereas this cavity appears solvent‐exposed in the x‐ray structure, the loop spanning residues V188–I200 adopts an alternative conformation in the Boltz‐2 model that partially shields the pocket (Figure 3b). Conformational variability of this loop, which may facilitate opening and closing of the binding cavity, was further explored using ColabFold‐AF2 (methods in Supporting Information) and can be visualized in Figure S7 [49, 50]. On this basis, the Boltz‐2 model was used for blind docking simulations with CB‐Dock2 [51], which has recently proven effective in identifying binding sites of maturation intermediates in [NiFe]‐hydrogenases [52]. The most favorable docking site (Vina score –6.8) places the [2Fe]F site in proximity of the [4Fe–4S] cluster near residues Ile200, Thr241, Glu242, Ser243, Gln244, Phe262, Ser263, Asp306, Ile307, and Tyr376 (Figure 3c), some of which were recently identified by site‐directed mutagenesis as critical for [2Fe]E/F binding and biosynthesis of the DTMA bridge [34]. These residues form a network of electrostatic, van der Waals/hydrophobic and hydrogen‐bonding interactions stabilizing the diiron subsite within the HydF scaffold. Notably, Glu305—which coordinates the [4Fe–4S] cluster in FeS‐TmeHydF (Figure 3a)—is displaced away from the cubane in the holo‐HydF model (Figs. 3c), consistent with conformational changes in the binding pocket possibly inferred from NRVS‐detected perturbations of Fe–S vibrational modes upon [2Fe]F binding (Figure 2a). This rearrangement provides a structural rationale for previous mutagenesis studies in which replacement of the conserved Glu305(/Asp) showed negligible effects on diiron core binding and assembly [30, 31, 53], explaining why the residue is dispensable for [2Fe]F incorporation despite its conserved position in the cubane‐ligating motif. Importantly, one CN ligand of [2Fe]F is positioned approximately 4 Å from the [4Fe–4S] cluster (Figure 3c), consistent with a non‐covalent, through‐space interaction.

FIGURE 3.

FIGURE 3

Open in a new tab

Structural representations of TmeHydF and the proposed [2Fe]E/F binding pocket. Electrostatic potential surface representations (±5 kBT/e) of the x‐ray structure of TmeHydF (a) and the Boltz‐2 model (b) including the [4Fe–4S] cluster (SF4), GTP (panel b), the flexible loop V188–I200 (see Figure S7), and the loop carrying E305 (cartoon representation). The positively charged binding site of the [2Fe] precursor is shown in blue. (c) Close‐up view of the proposed [2Fe]F binding site. The [4Fe–4S] cluster coordination sphere and residues in proximity to the diiron subsite are shown as sticks (C, grey; N, blue; S, yellow; O, red). (d) Close‐up view of the proposed Hmet:[2Fe]E‐HydF complex, highlighting coordination of the lipoate group to the [4Fe–4S] cluster.

This spatial arrangement is in full agreement with our NRVS data, which exclude Fe–CN vibrational signatures expected for cyanide bridging while revealing subtle perturbations of low‐frequency [4Fe–4S] modes. Next, we combined this structural framework with the recent identification of Hmet as an essential component of H‐cluster biosynthesis and performed additional structure predictions of HydF (Figure S8) in the presence of GTP, its [4Fe–4S] cluster, the [2Fe]E precursor, and lipoate‐functionalized Hmet [54], proposed to carry the CH2–NH3 moiety of the DTMA bridge [17, 19]. The resulting models identify a plausible assembly scenario (Figures 3d and S8) comprising a HydF–Hmet complex, in which one thiolate of lipoate coordinates to the non‐cysteine‐ligated iron of the cubane, while the second thiolate remains positioned in proximity to the [2Fe]E precursor, consistent with a role in formation of the bridging moiety.

3. Conclusion

Based on these predictions, we outline a tentative mechanistic scenario (Figure 4) that integrates the x‐ray structure of FeS‐HydF (panel a) [26], prior work on the role of Hmet (panels b–d) [18, 19], and site‐directed mutagenesis studies highlighting the importance of strictly conserved HydF residues Ser243 and Ser263 for [2Fe]E/F binding and Asp306 for biosynthesis of the CH2–NH–CH2 bridge [34]. This framework incorporates our revised model of the [2Fe]F:[4Fe–4S] interaction (panel f) and proposes plausible intermediates (panels c and d), in which Hmet coordinates to the [4Fe–4S] cluster and delivers DTMA components at the [2Fe]E precursor via its lipoate functionality. In this unified model, the [4Fe–4S] cluster does not coordinate the diiron subsite but instead stabilizes and regulates the binding cavity, controlling access of Hmet to the fragile [2Fe] precursor during the final stages of H‐cluster assembly. While still preliminary, our analysis provides the first structural framework linking HydF function to earlier maturation stages [31, 32], illustrating how auxiliary Fe–S clusters can guide metallocofactor maturation through structural and electronic preorganization. These findings also establish a basis for future studies aimed at identifying transient intermediates in H‐cluster biosynthesis including the proposed lipoate–[4Fe–4S] cluster interaction in HydF.

FIGURE 4.

FIGURE 4

Open in a new tab

Proposed catalytic cycle of HydF leading to assembly of the fully functional [2Fe]F precursor. (a) [4Fe–4S] cluster of HydF coordinated by a labile glutamate [26]. (b) Transfer of the [2Fe]E precursor by HydE to the diiron‐binding cavity of HydF with concomitant release of the Glu305 [4Fe–4S] ligand. (c) Proposed HydF–Hmet complex in which one thiolate of lipoate coordinates to the [4Fe–4S] cluster, while the second thiol positions the first CH2–NH3 moiety in proximity to [2Fe]E. (d) HydF–Hmet complex positioning the second CH2–NH3 moiety in proximity to the [2Fe]E CH2NH2 intermediate. (e) Intramolecular condensation of the two aminomethyl groups of the [2Fe]E (CH2NH2)2 intermediate [55]. (f) Holo‐HydF complex showing a noncovalent, through‐space interaction between the [2Fe]F subsite and the cubane. At this stage [2Fe]F is transferred to apo‐HydA, where one CO ligand is released giving rise to the characteristic geometry of the matured H‐cluster. Critical residues Ser243, Ser263, Glu305, and Asp306 of TmeHydF at different stages of the cycle are highlighted.

Author Contributions

Giorgio Caserta: investigation, conceptualization, software, data curation, methodology, validation, formal analysis, resources, project administration, supervision, visualization, writing – original draft, funding acquisition, writing – review and editing. Princess Cabotaje: investigation, data curation, formal analysis, writing – review and editing, methodology, validation, visualization. Armel T. Waffo: writing – review and editing, formal analysis, data curation, software. Deepak Prajapat: formal analysis, software, methodology. Ilya Sergueev: software, formal analysis, supervision, methodology. Stefan Frielingsdorf: methodology, software, visualization, writing – review and editing, formal analysis, resources, data curation. Gustav Berggren: funding acquisition, project administration, writing – review and editing, resources, investigation, formal analysis, supervision, visualization, validation. The article was written through contributions of all authors.

Conflicts of Interest

The authors declare no conflict of interest

Supporting information

Supporting File: anie72892‐sup‐0001‐SuppMat.docx.

ANIE-65-e8078898-s001.docx (4.8MB, docx)

Acknowledgments

This work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany´s Excellence Strategy‐EXC2008‐390540038 (UniSysCat), and the Swedish Research Council, VR (grant no. 2021–04471, to G.B.) We acknowledge DESY (Hamburg, Germany), a member of the Helmholtz Association HGF, for the provision of experimental facilities. Parts of this research were carried out at Petra III (photon beamline P01). Beamtime was allocated for proposals I‐20241074 and I‐20250741. G.C. acknowledges Ingo Zebger and Oliver Lenz for generous support by using their laboratories. Prof. Marc Fontecave is gratefully acknowledged for providing the TmHydF plasmid and for initiating the pioneering studies on HydF hybrids that inspired this work.

Open access funding enabled and organized by Projekt DEAL.

Contributor Information

Giorgio Caserta, Email: giorgio.caserta@tu-berlin.de.

Gustav Berggren, Email: gustav.berggren@kemi.uu.se.

Data Availability Statement

Experimental NRVS data; the Boltz‐1 model of TmeHydF; CB‐Dock2 blind docking results for the [2Fe]F cofactor within the Boltz‐1 TmeHydF model; TmeHydF AF2 models (flex configuration); and the Boltz‐2 model of the Hmet–TmeHydF complex in the presence of GTP, the [4Fe−4S] cluster, and the [2Fe]E precursor are available on Zenodo (10.5281/zenodo.18669547) [56]. All other data supporting the findings of this study are provided within the manuscript and the Supporting Information.

References

  • 1. Lubitz W., Ogata H., Rüdiger O., and Reijerse E., “Hydrogenases,” Chemical Reviews 114 (2014): 4081–4148, 10.1021/cr4005814. [DOI] [PubMed] [Google Scholar]
  • 2. Simmons T. R., Berggren G., Bacchi M., Fontecave M., and Artero V., “Mimicking Hydrogenases: From Biomimetics to Artificial Enzymes,” Coordination Chemistry Reviews 270 (2014): 127–150, 10.1016/j.ccr.2013.12.018. [DOI] [Google Scholar]
  • 3. Lampret O., Esselborn J., Haas R., et al., “The Final Steps of [FeFe]‐Hydrogenase Maturation,” Proceedings of the National Academy of Sciences of the United States of America 116 (2019): 15802–15810, 10.1073/pnas.1908121116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Britt R. D., Rao G., and Tao L., “Bioassembly of Complex Iron–Sulfur Enzymes: Hydrogenases and Nitrogenases,” Nature Reviews Chemistry 4 (2020): 542–549, 10.1038/s41570-020-0208-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Marlott A., Pagnier A., Shepard E. M., et al., “In Vitro Maturation of Fully Active [FeFe]‐Hydrogenase in a Defined System Including the Iron Carrier NfuA,” Proceedings of the National Academy of Sciences of the United States of America 122 (2025): e2517347122, 10.1073/pnas.2517347122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Britt R. D., Rauchfuss T. B., and Rao G., “The H‐Cluster of [FeFe] Hydrogenases: Its Enzymatic Synthesis and Parallel Inorganic Semisynthesis,” Accounts of Chemical Research 57 (2024): 1941–1950, 10.1021/acs.accounts.4c00231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Rubach J. K., Brazzolotto X., Gaillard J., and Fontecave M., “Biochemical Characterization of the HydE and HydG Iron‐Only Hydrogenase Maturation Enzymes From Thermatoga Maritima,” FEBS Letters 579 (2005): 5055–5060, 10.1016/j.febslet.2005.07.092. [DOI] [PubMed] [Google Scholar]
  • 8. Shepard E. M., Duffus B. R., George S. J., et al., “[FeFe]‐Hydrogenase Maturation: HydG‐Catalyzed Synthesis of Carbon Monoxide,” Journal of the American Chemical Society 132 (2010): 9247–9249, 10.1021/ja1012273. [DOI] [PubMed] [Google Scholar]
  • 9. Kuchenreuther J. M., Myers W. K., Suess D. L. M., et al., “The HydG Enzyme Generates an Fe(CO)2(CN) Synthon in Assembly of the FeFe Hydrogenase H‐Cluster,” Science 343 (2014): 424–427, https://www.science.org/doi/10.1126/science.1246572. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Pagnier A., Martin L., Zeppieri L., Nicolet Y., and Fontecilla‐Camps J. C., “CO and CN− Syntheses by [FeFe]‐Hydrogenase Maturase HydG Are Catalytically Differentiated Events,” Proceedings of the National Academy of Sciences of the United States of America 113 (2016): 104–109, 10.1073/pnas.1515842113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Omeiri J., Martin L., Usclat A., Cherrier M. V., and Nicolet Y., “Maturation of the [FeFe]‐Hydrogenase: Direct Transfer of the (κ3‐cysteinate)FeII (CN)(CO)2 Complex B From HydG to HydE,” Angewandte Chemie International Edition 62 (2023): e202314819, 10.1002/anie.202314819. [DOI] [PubMed] [Google Scholar]
  • 12. Rohac R., Martin L., Liu L., et al., “Crystal Structure of the [FeFe]‐Hydrogenase Maturase HydE Bound to Complex‐B,” Journal of the American Chemical Society 143 (2021): 8499–8508, 10.1021/jacs.1c03367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Tao L., Pattenaude S. A., Joshi S., Begley T. P., Rauchfuss T. B., and Britt R. D., “Radical SAM Enzyme HydE Generates Adenosylated Fe(I) Intermediates en Route to the [FeFe]‐Hydrogenase Catalytic H‐Cluster,” Journal of the American Chemical Society 142 (2020): 10841–10848, 10.1021/jacs.0c03802. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Chen N., Rao G., Tao L., Britt R. D., and Wang L.‐P., “HydE Catalytic Mechanism Is Powered by a Radical Relay With Redox‐Active Fe(I)‐Containing Intermediates,” Journal of the American Chemical Society 147 (2025): 4800–4809, 10.1021/jacs.4c12668. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Rao G., Tao L., Yu X., Rauchfuss T. B., and Britt R. D., “The Radical S‐Adenosyl‐l‐Methionine Enzyme HydE Forms an Fe(I)Fe(I) Dimer En Route to the [FeFe] Hydrogenase H‐Cluster,” Journal of the American Chemical Society 147 (2025): 32737–32744, 10.1021/jacs.5c08533. [DOI] [PubMed] [Google Scholar]
  • 16. Zhang Y., Tao L., Woods T. J., Britt R. D., and Rauchfuss T. B., “Organometallic Fe2(μ‐SH)2(CO)4(CN)2 Cluster Allows the Biosynthesis of the [FeFe]‐Hydrogenase With Only the HydF Maturase,” Journal of the American Chemical Society 144 (2022): 1534–1538, 10.1021/jacs.1c12506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Pagnier A., Balci B., Shepard E. M., Broderick W. E., and Broderick J. B., “Hydrogenase In Vitro Maturation,” Angewandte Chemie International Edition 61 (2022): e202212074, 10.1002/anie.202212074. [DOI] [PubMed] [Google Scholar]
  • 18. Pagnier A., Balci B., Shepard E. M., et al., “Hydrogenase: Defined Lysate‐Free Maturation Reveals a Key Role for Lipoyl‐H‐Protein in DTMA Ligand Biosynthesis,” Angewandte Chemie International Edition 61 (2022): e202203413, 10.1002/anie.202203413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Pagnier A., Balci B., Shepard E. M., et al., “Role of Ammonia‐Lyases in the Synthesis of the Dithiomethylamine Ligand During [FeFe]‐Hydrogenase Maturation,” Journal of Biological Chemistry 300 (2024): 107760, 10.1016/j.jbc.2024.107760. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Berggren G., Adamska A., Lambertz C., et al., “Biomimetic Assembly and Activation of [FeFe]‐Hydrogenases,” Nature 499 (2013): 66–69, 10.1038/nature12239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Balci B., O'Neill R. D., Shepard E. M., et al., “Semisynthetic Maturation of [FeFe]‐Hydrogenase Using [Fe2(μ‐SH)2(CN)2(CO)4]2− : Key Roles for HydF and GTP,” Chemical Communications 59 (2023): 8929–8932, 10.1039/D3CC02169F. [DOI] [PubMed] [Google Scholar]
  • 22. Berggren G., Garcia‐Serres R., Brazzolotto X., et al., “An EPR/HYSCORE, Mössbauer, And Resonance Raman Study Of The Hydrogenase Maturation Enzyme Hydf: A Model For N‐Coordination To [4Fe–4S, ] Clusters,” Journal of Biological Inorganic Chemistry 19 (2014): 75–84, 10.1007/s00775-013-1062-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Berto P., Di Valentin M., Cendron L., et al., “The [4Fe–4S]‐Cluster Coordination of [FeFe]‐Hydrogenase Maturation Protein HydF as Revealed by EPR and HYSCORE Spectroscopies,” Biochimica et Biophysica Acta (BBA) ‐ Bioenergetics 1817 (2012): 2149–2157, 10.1016/j.bbabio.2012.09.004. [DOI] [PubMed] [Google Scholar]
  • 24. Shepard E. M., McGlynn S. E., Bueling A. L., et al., “Synthesis of the 2Fe Subcluster of the [FeFe]‐Hydrogenase H Cluster on the HydF Scaffold,” Proceedings of the National Academy of Sciences of the United States of America 107 (2010): 10448–10453, 10.1073/pnas.1001937107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Németh B., Land H., Magnuson A., Hofer A., and Berggren G., “The Maturase HydF Enables [FeFe] Hydrogenase Assembly via Transient, Cofactor‐Dependent Interactions,” Journal of Biological Chemistry 295 (2020): 11891–11901, 10.1074/jbc.RA119.011419. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Caserta G., Pecqueur L., Adamska‐Venkatesh A., et al., “Structural and Functional Characterization of the Hydrogenase‐Maturation HydF Protein,” Nature Chemical Biology 13 (2017): 779–784, 10.1038/nchembio.2385. [DOI] [PubMed] [Google Scholar]
  • 27. Németh B., Esmieu C., Redman H. J., and Berggren G., “Monitoring H‐Cluster Assembly Using a Semi‐Synthetic HydF Protein,” Dalton Transactions 48 (2019): 5978–5986, 10.1039/C8DT04294B. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Sil D., Martinez Z., Ding S., et al., “Cyanide Docking and Linkage Isomerism in Models for the Artificial [FeFe]‐Hydrogenase Maturation Process,” Journal of the American Chemical Society 140 (2018): 9904–9911, 10.1021/jacs.8b04189. [DOI] [PubMed] [Google Scholar]
  • 29. Lunsford A. M., Beto C. C., Ding S., et al., “Cyanide‐Bridged Iron Complexes as Biomimetics of Tri‐Iron Arrangements in Maturases of the H Cluster of the Di‐Iron Hydrogenase,” Chemical Science 7 (2016): 3710–3719, 10.1039/C6SC00213G. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Haas R., Engelbrecht V., Lampret O., et al., “The [4Fe‐4S]‐Cluster of HydF is not Required for the Binding and Transfer of the Diiron Site of [FeFe]‐Hydrogenases,” ChemBioChem 24 (2023): e202300222, 10.1002/cbic.202300222. [DOI] [PubMed] [Google Scholar]
  • 31. Balci B., Shepard E. M., Marlott A., et al., “The [4Fe‐4S] Cluster of HydF Is Essential for [FeFe]‐Hydrogenase Maturation,” Journal of the American Chemical Society 147 (2025): 47773–47782, 10.1021/jacs.5c18286. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. King P. W., Posewitz M. C., Ghirardi M. L., and Seibert M., “Functional Studies of [FeFe] Hydrogenase Maturation in an Escherichia coli Biosynthetic System,” Journal of Bacteriology 188 (2006): 2163–2172, 10.1128/JB.188.6.2163-2172.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Artero V., Berggren G., Atta M., et al., “From Enzyme Maturation to Synthetic Chemistry: The Case of Hydrogenases,” Accounts of Chemical Research 48 (2015): 2380–2387, 10.1021/acs.accounts.5b00157. [DOI] [PubMed] [Google Scholar]
  • 34. Haas R., Lampret O., Yadav S., Apfel U.‐P., and Happe T., “A Conserved Binding Pocket in HydF Is Essential for Biological Assembly and Coordination of the Diiron Site of [FeFe]‐Hydrogenases,” Journal of the American Chemical Society 146 (2024): 15771–15778, 10.1021/jacs.4c01635. [DOI] [PubMed] [Google Scholar]
  • 35. Lorent C., Caserta G., Schünemann V., and Zebger I., “Vibrational and Mössbauer Spectroscopic Techniques to Study Iron–Sulfur Clusters,” in Iron‐Sulfur Clusters (eds Leimkühler S., Schwarz G., Lenz O. and Einsle O.), Wiley‐VCH; (2025), Chap. 17; 10.1002/9783527843596.ch17. [DOI] [Google Scholar]
  • 36. Wang H., Pelmenschikov V., Yoda Y., and Cramer S. P., “NRVS of Fe S Cluster Proteins & Models—A Bestiary of Nifty Normal Modes,” Journal of Inorganic Biochemistry 270 (2025): 112935, 10.1016/j.jinorgbio.2025.112935. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Caserta G., Hartmann S., Van Stappen C., et al., “Stepwise Assembly of the Active Site Of [NiFe]‐Hydrogenase,” Natural Chemical Biology 19 (2023): 498–506, 10.1038/s41589-022-01226-w. [DOI] [PubMed] [Google Scholar]
  • 38. Caserta G., Laun K., Oudsen J.‐P., Sergueev I., Zebger I., and Lenz O., “NRVS Spectroscopy Resolves Distinct Bridging Hydride Intermediates in [NiFe]‐Hydrogenase,” Journal of the American Chemical Society 147 (2025): 41216–41220, 10.1021/jacs.5c15408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Ogata H., Krämer T., Wang H., et al., “Hydride Bridge in [NiFe]‐Hydrogenase Observed by Nuclear Resonance Vibrational Spectroscopy,” Nature Communications 6 (2015): 7890, 10.1038/ncomms8890. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Van Stappen C., Benediktsson B., Rana A., et al., “Structural Correlations of Nitrogenase Active Sites Using Nuclear Resonance Vibrational Spectroscopy and QM/MM Calculations,” Faraday Discussions 243 (2023): 253–269, 10.1039/D2FD00174H. [DOI] [PubMed] [Google Scholar]
  • 41. Galinato M. G. I., Kleingardner J. G., Bowman S. E. J., et al., “Heme‐Protein Vibrational Couplings in Cytochrome c Provide a Dynamic Link That Connects the Heme‐Iron and the Protein Surface,” Proceedings of the National Academy of Sciences of the United States of America 109 (2012): 8896–8900, 10.1073/pnas.1200345109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Folgosa F., Pelmenschikov V., Caserta G., et al., “Hydroxo‐Bridged Active Site of Flavodiiron NO Reductase Revealed by NRVS and DFT,” Proceedings of the National Academy of Sciences of the United States of America 123 (2026): e2512429123, 10.1073/pnas.2512429123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Jacobs A. B., Banerjee R., Deweese D. E., et al., “Nuclear Resonance Vibrational Spectroscopic Definition of the Fe(IV)2 Intermediate Q in Methane Monooxygenase and Its Reactivity,” Journal of the American Chemical Society 143 (2021): 16007–16029, 10.1021/jacs.1c05436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Brazzolotto X., Rubach J. K., Gaillard J., Gambarelli S., Atta M., and Fontecave M., “The [Fe‐Fe]‐Hydrogenase Maturation Protein HydF From Thermotoga Maritima Is a GTPase With an Iron‐Sulfur Cluster,” Journal of Biological Chemistry 281 (2006): 769–774, 10.1074/jbc.M510310200. [DOI] [PubMed] [Google Scholar]
  • 45. Pelmenschikov V., Birrell J. A., Gee L. B., et al., “Vibrational Perturbation of the [FeFe] Hydrogenase H‐Cluster Revealed by 13C2H‐ADT Labeling,” Journal of the American Chemical Society 143 (2021): 8237–8243, 10.1021/jacs.1c02323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Scott A. G., Szilagyi R. K., Mulder D. W., et al., “Compositional and Structural Insights Into the Nature of the H‐Cluster Precursor on HydF,” Dalton Transactions 47 (2018): 9521–9535, 10.1039/C8DT01654B. [DOI] [PubMed] [Google Scholar]
  • 47. Wohlwend J., Corso G., Passaro S., et al., “Boltz‐1 Democratizing Biomolecular Interaction Modeling,” BioRxiv (2024), 624167, 10.1101/2024.11.19.624167. [DOI] [Google Scholar]
  • 48. Abramson J., Adler J., Dunger J., et al., “Accurate Structure Prediction Of Biomolecular Interactions With Alphafold 3,” Nature 630 (2024): 493–500, 10.1038/s41586-024-07487-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Mirdita M., Schütze K., Moriwaki Y., Heo L., Ovchinnikov S., and Steinegger M., “ColabFold: Making Protein Folding Accessible to All,” Nature Methods 19 (2022): 679–682, 10.1038/s41592-022-01488-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Kim G., Lee S., Karin E. L., et al., “Easy and Accurate Protein Structure Prediction Using ColabFold,” Nature Protocols 20 (2025): 620–642, 10.1038/s41596-024-01060-5. [DOI] [PubMed] [Google Scholar]
  • 51. Liu Y., Yang X., Gan J., Chen S., Xiao Z.‐X., and Cao Y., “CB‐Dock2: Improved Protein–Ligand Blind Docking by Integrating Cavity Detection, Docking and Homologous Template Fitting,” Nucleic Acids Research 50 (2022): W159–W164, 10.1093/nar/gkac394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52. Kwiatkowski A., Caserta G., Schulz A.‐C., et al., “ATP‐Triggered Fe(CN)2 CO Synthon Transfer From the Maturase HypCD to the Active Site of Apo‐[NiFe]‐Hydrogenase,” Journal of the American Chemical Society 146 (2024): 30976–30989, 10.1021/jacs.4c09791. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53. Papini C., Sommer C., Pecqueur L., et al., “Bioinspired Artificial [FeFe]‐Hydrogenase With a Synthetic H‐Cluster,” ACS Catalysis 9 (2019): 4495–4501, 10.1021/acscatal.9b00540. [DOI] [Google Scholar]
  • 54. Okamura‐Ikeda K., Hosaka H., Maita N., et al., “Crystal Structure of Aminomethyltransferase in Complex With Dihydrolipoyl‐H‐Protein of the Glycine Cleavage System,” Journal of Biological Chemistry 285 (2010): 18684–18692, 10.1074/jbc.M110.110718. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Yu X., Rao G., Britt R. D., and Rauchfuss T. B., “Final Stages in the Biosynthesis of the [FeFe]‐Hydrogenase Active Site,” Angewandte Chemie International Edition 63 (2024): e202404044, 10.1002/anie.202404044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56. Caserta G., Cabotaje P. R., Tadjoung Waffo A. F., et al., “A Non‐Covalent [4Fe–4S]/[2Fe] Interface in HydF Guides [FeFe]‐Hydrogenase Maturation,” [Dataset], Zenodo (2026), 10.5281/zenodo.18669547. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting File: anie72892‐sup‐0001‐SuppMat.docx.

ANIE-65-e8078898-s001.docx (4.8MB, docx)

Data Availability Statement

Experimental NRVS data; the Boltz‐1 model of TmeHydF; CB‐Dock2 blind docking results for the [2Fe]F cofactor within the Boltz‐1 TmeHydF model; TmeHydF AF2 models (flex configuration); and the Boltz‐2 model of the Hmet–TmeHydF complex in the presence of GTP, the [4Fe−4S] cluster, and the [2Fe]E precursor are available on Zenodo (10.5281/zenodo.18669547) [56]. All other data supporting the findings of this study are provided within the manuscript and the Supporting Information.

Articles from Angewandte Chemie (International Ed. in English) are provided here courtesy of Wiley

RESOURCES