Advertisement

An in vitro reconstitution system to monitor iron transfer to the active site during the maturation of [NiFe]-hydrogenase

Open AccessPublished:July 19, 2022DOI:https://doi.org/10.1016/j.jbc.2022.102291
      [NiFe]-hydrogenases (Hyds) comprise a small and a large subunit. The latter harbors the biologically unique [NiFe](CN)2CO active-site cofactor. The maturation process includes the assembly of the [Fe](CN)2CO cofactor precursor, nickel binding, endoproteolytic cleavage of the large subunit, and dimerization with the small subunit to yield active enzyme. The biosynthesis of the [Fe](CN)2CO moiety of [NiFe]-Hyd-1 and Hyd-2 occurs on the scaffold complex HybG–HypD (GD), whereas the HypC–HypD complex is specific for the assembly of Hyd-3. The metabolic source and the route for delivering iron to the active site remain unclear. To investigate the maturation process of O2-tolerant Hyd-1 from Escherichia coli, we developed an enzymatic in vitro reconstitution system that allows for the synthesis of Hyd-1 using only purified components. Together with this in vitro reconstitution system, we employed biochemical analyses, infrared spectroscopy (attenuated total reflection FTIR), mass spectrometry (MS), and microscale thermophoresis to monitor the iron transfer during the maturation process and to understand how the [Fe](CN)2CO cofactor precursor is ultimately incorporated into the large subunit. We demonstrate the direct transfer of iron from 57Fe-labeled GD complex to the large subunit of Hyd-1. Our data reveal that the GD complex exclusively interacts with the large subunit of Hyd-1 and Hyd-2 but not with the large subunit of Hyd-3. Furthermore, we show that the presence of iron in the active site is a prerequisite for nickel insertion. Taken together, these findings reveal how the [Fe](CN)2CO cofactor precursor is transferred and incorporated into the active site of [NiFe]-Hyd.

      Keywords

      Abbreviations:

      AM (activation mix), ATR (attenuated total reflection), BV (benzyl viologen), CD (HypC–HypD), GD (HybG–HypD), Hyd (hydrogenase), ICP (inductively coupled plasma), MS (mass spectrometry), MST (microscale thermophoresis analysis)
      Hydrogenases (Hyds) are microbial metalloenzymes that catalyze the reduction of protons to H2 (disposal of excess reducing equivalents) or the oxidation of H2 as an energy source (
      • Vignais P.M.
      • Billoud B.
      Occurrence, classification, and biological function of hydrogenases: an overview.
      ,
      • Thauer R.K.
      • Kaster A.K.
      • Goenrich M.
      • Schick M.
      • Hiromoto T.
      • Shima S.
      Hydrogenases from methanogenic archaea, nickel, a novel cofactor, and H-2 storage.
      ,
      • Lubitz W.
      • Ogata H.
      • Rudiger O.
      • Reijerse E.
      Hydrogenases.
      ). Based on the metal content at the active site, [FeFe]−, [NiFe]−, and iron sulfur cluster-free Hyds are distinguished (
      • Thauer R.K.
      • Kaster A.K.
      • Goenrich M.
      • Schick M.
      • Hiromoto T.
      • Shima S.
      Hydrogenases from methanogenic archaea, nickel, a novel cofactor, and H-2 storage.
      ,
      • Lubitz W.
      • Ogata H.
      • Rudiger O.
      • Reijerse E.
      Hydrogenases.
      ). While [FeFe]-Hyds are typically monomeric, [NiFe]-Hyds are formed by a heterodimer of a large and a small subunit. The large subunit harbors the [NiFe]-cofactor, whereas the small subunit contains 1 to 3 iron sulfur clusters involved in electron transfer (
      • Shafaat H.S.
      • Rudiger O.
      • Ogata H.
      • Lubitz W.
      [NiFe] hydrogenases: a common active site for hydrogen metabolism under diverse conditions.
      ). The [NiFe]-Hyd heterodimer can be associated with additional subunits, for example, for transferring electrons to pyridine nucleotides or for anchoring multimeric enzymes to the membrane (
      • Thauer R.K.
      • Kaster A.K.
      • Goenrich M.
      • Schick M.
      • Hiromoto T.
      • Shima S.
      Hydrogenases from methanogenic archaea, nickel, a novel cofactor, and H-2 storage.
      ).
      Escherichia coli synthesizes 3 membrane-bound [NiFe]-Hyds. Although they carry the same [NiFe]-cofactor, Hyd-1 and Hyd-2 are H2-oxidizing enzymes, whereas Hyd-3 is a H2-evolving enzyme. Each of the 3 Hyds is predicted to contribute to different aspects of cellular metabolism (
      • Sargent F.
      The model [NiFe]-Hydrogenases of Escherichia coli.
      ,
      • Forzi L.
      • Sawers R.G.
      Maturation of [NiFe]-hydrogenases in Escherichia coli.
      ). The O2-tolerant Hyd-1 is able to oxidize H2 in the presence of oxygen (air) and has become one of the most important paradigm systems for understanding O2-tolerant Hyds (
      • Lukey M.J.
      • Roessler M.M.
      • Parkin A.
      • Evans R.M.
      • Davies R.A.
      • Lenz O.
      • et al.
      Oxygen-tolerant [NiFe]-hydrogenases: the individual and collective importance of supernumerary cysteines at the proximal Fe-S cluster.
      ). The O2-sensitive Hyd-2 couples H2 oxidation to quinone reduction during anaerobic respiration (
      • Sargent F.
      The model [NiFe]-Hydrogenases of Escherichia coli.
      ,
      • Sawers R.G.
      • Ballantine S.P.
      • Boxer D.H.
      Differential expression of hydrogenase isoenzymes in Escherichia coli K-12: evidence for a third isoenzyme.
      ,
      • Laurinavichene T.V.
      • Tsygankov A.A.
      H2 consumption by Escherichia coli coupled via hydrogenase 1 or hydrogenase 2 to different terminal electron acceptors.
      ). Hyd-3 forms a complex with the H2-evolving formate hydrogenlyase, which acts to detoxify excess formic acid into CO2 and H2 during fermentation (
      • Ballantine S.P.
      • Boxer D.H.
      Nickel-containing hydrogenase isoenzymes from anaerobically grown Escherichia coli K-12.
      ).
      The [NiFe]-Hyds of E. coli are activated by a specific maturation apparatus. Figure 1 shows the current understanding of the maturation of Hyd-1 and Hyd-2. The complex machinery involved in the synthesis of the [NiFe]-cofactor and its insertion into the apo-form of the large subunit comprises several maturation proteins (
      • Böck A.
      • King P.W.
      • Blokesch M.
      • Posewitz M.C.
      Maturation of hydrogenases.
      ,
      • Lacasse M.J.
      • Zamble D.B.
      [NiFe]-Hydrogenase maturation.
      ). The assembly of the [Fe](CN)2CO moiety (steps 1–3 in Fig. 1) suggests the following sequence of events: (1) the HybG dimer is proposed to deliver both iron and CO2 to HypD (
      • Soboh B.
      • Stripp S.T.
      • Bielak C.
      • Lindenstrauss U.
      • Braussemann M.
      • Javaid M.
      • et al.
      The [NiFe]-hydrogenase accessory chaperones HypC and HybG of Escherichia coli are iron- and carbon dioxide-binding proteins.
      ). Subsequently, the HybG–HypD (GD) complex is formed while a HybG monomer is released (
      • Stripp S.T.
      • Oltmanns J.
      • Muller C.S.
      • Ehrenberg D.
      • Schlesinger R.
      • Heberle J.
      • et al.
      Electron inventory of the iron-sulfur scaffold complex HypCD essential in [NiFe]-hydrogenase cofactor assembly.
      ). (2) The GD complex is proposed to catalyze an ATP-dependent reduction of CO2 (
      • Nutschan K.
      • Golbik R.P.
      • Sawers R.G.
      The iron-sulfur-containing HypC-HypD scaffold complex of the [NiFe]-hydrogenase maturation machinery is an ATPase.
      ); however, definite proof for this enzymatic activity is yet to obtain. (3) The synthesis of the CN ligands from carbamoyl phosphate in an ATP-dependent condensation reaction is catalyzed by HypF and HypE (
      • Paschos A.
      • Bauer A.
      • Zimmermann A.
      • Zehelein E.
      • Böck A.
      HypF, a carbamoyl phosphate-converting enzyme involved in [NiFe] hydrogenase maturation.
      ). Subsequently, cyanide is transferred to the iron ion, which induces the formation of the [Fe](CN)2CO moiety on the GD complex (
      • Blokesch M.
      • Albracht S.P.
      • Matzanke B.F.
      • Drapal N.M.
      • Jacobi A.
      • Böck A.
      The complex between hydrogenase-maturation proteins HypC and HypD is an intermediate in the supply of cyanide to the active site iron of [NiFe]-hydrogenases.
      ,
      • Stripp S.T.
      • Lindenstrauss U.
      • Sawers R.G.
      • Soboh B.
      Identification of an isothiocyanate on the HypEF complex suggests a route for efficient Cyanyl-group channeling during [NiFe]-hydrogenase cofactor generation.
      ,
      • Stripp S.T.
      • Lindenstrauss U.
      • Granich C.
      • Sawers R.G.
      • Soboh B.
      The influence of oxygen on [NiFe]-hydrogenase cofactor biosynthesis and how ligation of carbon monoxide precedes cyanation.
      ). It is unknown how the cofactor precursor is transferred to the large subunit, but once the assembly and insertion of the cofactor are completed (step 4 and 5 in Fig. 1), the C terminus of the large subunit is cleaved off (step 6). This results in conformational changes that allow dimerization with the small subunit, which yields the active enzyme (step 7) (
      • Senger M.
      • Stripp S.T.
      • Soboh B.
      Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
      ,
      • Pinske C.
      • Kruger S.
      • Soboh B.
      • Ihling C.
      • Kuhns M.
      • Braussemann M.
      • et al.
      Efficient electron transfer from hydrogen to benzyl viologen by the [NiFe]-hydrogenases of Escherichia coli is dependent on the coexpression of the iron-sulfur cluster-containing small subunit.
      ).
      Figure thumbnail gr1
      Figure 1Working model for maturation of the large subunit of Hyd-1 and Hyd-2 of Escherichia coli into functional [NiFe]-Hyd. (1) The HybG–HypD complex is formed upon contact with the [Fe]CO2 carrying HybG dimer. (2) The CO2 ligand may undergo an ATP-dependent reduction to CO catalyzed by HybG–HypD complex. (3) The maturation proteins HypF and HypE catalyze the synthesis and transfer of CN ligands to the CO-modified iron ion and formation of the [Fe](CN)2CO moiety on the HybG–HypD complex. (4) It is unclear how the [Fe](CN)2CO moiety is incorporated into the large subunit. (5) Nickel insertion by concerted activity of HypA, HypB, and SlyD. (6) Endoproteolytic cleavage of the C-terminal peptide, associated with conformational changes of the protein. (7) Dimerization of large and small subunits form functional [NiFe]-Hyd. Hyd, hydrogenase.
      The GD complex is the heart of the maturation machinery of Hyd-1 and Hyd-2 (
      • Soboh B.
      • Lindenstrauss U.
      • Granich C.
      • Javed M.
      • Herzberg M.
      • Thomas C.
      • et al.
      [NiFe]-hydrogenase maturation in vitro: analysis of the roles of the HybG and HypD accessory proteins1.
      ,
      • Blokesch M.
      • Magalon A.
      • Böck A.
      Interplay between the specific chaperone-like proteins HybG and HypC in maturation of hydrogenases 1, 2, and 3 from Escherichia coli.
      ). It is the synthesis platform, where the CN and CO ligands are attached to the iron ion of the cofactor. The transfer of each CN ligand to the iron ion on the GD complex requires the input of 2 electrons (
      • Blokesch M.
      • Albracht S.P.
      • Matzanke B.F.
      • Drapal N.M.
      • Jacobi A.
      • Böck A.
      The complex between hydrogenase-maturation proteins HypC and HypD is an intermediate in the supply of cyanide to the active site iron of [NiFe]-hydrogenases.
      ,
      • Blokesch M.
      • Bock A.
      Properties of the [NiFe]-hydrogenase maturation protein HypD.
      ). In addition, 2 electrons are required for CO2 reduction if CO2 is the metabolic precursor of CO (
      • Stripp S.T.
      • Oltmanns J.
      • Muller C.S.
      • Ehrenberg D.
      • Schlesinger R.
      • Heberle J.
      • et al.
      Electron inventory of the iron-sulfur scaffold complex HypCD essential in [NiFe]-hydrogenase cofactor assembly.
      ). HypD must be involved in electron transfer because it is the only maturation enzyme that carries an [4Fe–4S] cluster. In a recent study, we proposed a unique redox cascade between the [4Fe–4S] cluster and a conserved disulfide motif of HypD. This provided comprehensive evidence for an electron inventory fit to drive multi–electron redox reactions required for the attachment of the CN and CO ligands to the iron ion (
      • Stripp S.T.
      • Oltmanns J.
      • Muller C.S.
      • Ehrenberg D.
      • Schlesinger R.
      • Heberle J.
      • et al.
      Electron inventory of the iron-sulfur scaffold complex HypCD essential in [NiFe]-hydrogenase cofactor assembly.
      ).
      Although all 3 Hyds carry the same [NiFe](CN)2CO cofactor, orthologous accessory proteins for the maturation of each Hyd have been described (
      • Böck A.
      • King P.W.
      • Blokesch M.
      • Posewitz M.C.
      Maturation of hydrogenases.
      ). For example, The C-terminal extension of the large subunit of each enzyme is cleaved off by a specific endopeptidase (
      • Rossmann R.
      • Maier T.
      • Lottspeich F.
      • Böck A.
      Characterisation of a protease from Escherichia coli involved in hydrogenase maturation.
      ). Furthermore, the HybG homolog HypC forms the HypC–HypD (CD) complex with HypD that exclusively catalyzes the assembly of the [Fe](CN)2CO moiety of Hyd-3 (
      • Blokesch M.
      • Magalon A.
      • Böck A.
      Interplay between the specific chaperone-like proteins HybG and HypC in maturation of hydrogenases 1, 2, and 3 from Escherichia coli.
      ). Table 1 describes the function of proteins and enzymes involved in the maturation of [NiFe]-Hyd that used in this study.
      Table 1The proposed function of [NiFe]-Hyd proteins used in this study
      ProteinMolecular weight (kDa)Function/propertiesReferences
      HyaB66.5Large subunit of Hyd-1(
      • Ballantine S.P.
      • Boxer D.H.
      Nickel-containing hydrogenase isoenzymes from anaerobically grown Escherichia coli K-12.
      )
      HyaA41Small subunit of Hyd-1(
      • Menon N.K.
      • Chatelus C.Y.
      • Dervartanian M.
      • Wendt J.C.
      • Shanmugam K.T.
      • Peck Jr., H.D.
      • et al.
      Cloning, sequencing, and mutational analysis of the Hyb operon encoding Escherichia coli hydrogenase 2.
      )
      HybC63.5Large subunit of Hyd-2(
      • Ballantine S.P.
      • Boxer D.H.
      Nickel-containing hydrogenase isoenzymes from anaerobically grown Escherichia coli K-12.
      )
      HycE65Large subunit of Hyd-3(
      • Menon N.K.
      • Chatelus C.Y.
      • Dervartanian M.
      • Wendt J.C.
      • Shanmugam K.T.
      • Peck Jr., H.D.
      • et al.
      Cloning, sequencing, and mutational analysis of the Hyb operon encoding Escherichia coli hydrogenase 2.
      )
      HybG8.7Forms the GD complex with HypD, involved in assembly of the [Fe](CN)2CO moiety of Hyd-1 and Hyd-2(
      • Soboh B.
      • Lindenstrauss U.
      • Granich C.
      • Javed M.
      • Herzberg M.
      • Thomas C.
      • et al.
      [NiFe]-hydrogenase maturation in vitro: analysis of the roles of the HybG and HypD accessory proteins1.
      ,
      • Blokesch M.
      • Magalon A.
      • Böck A.
      Interplay between the specific chaperone-like proteins HybG and HypC in maturation of hydrogenases 1, 2, and 3 from Escherichia coli.
      )
      HypC10Forms the CD complex with HypD, involved in assembly of the [Fe](CN)2CO moiety of Hyd-3(
      • Blokesch M.
      • Albracht S.P.
      • Matzanke B.F.
      • Drapal N.M.
      • Jacobi A.
      • Böck A.
      The complex between hydrogenase-maturation proteins HypC and HypD is an intermediate in the supply of cyanide to the active site iron of [NiFe]-hydrogenases.
      ,
      • Soboh B.
      • Stripp S.T.
      • Muhr E.
      • Granich C.
      • Braussemann M.
      • Herzberg M.
      • et al.
      [NiFe]-hydrogenase maturation: Isolation of a HypC-HypD complex carrying diatomic CO and CN- ligands.
      )
      HypD41.4Iron sulfur scaffold protein, involved in assembly of the [Fe](CN)2CO of the 3 Hyds, shows ATPase activity, and catalyzes CO-dependent reduction of MV(
      • Nutschan K.
      • Golbik R.P.
      • Sawers R.G.
      The iron-sulfur-containing HypC-HypD scaffold complex of the [NiFe]-hydrogenase maturation machinery is an ATPase.
      ,
      • Stripp S.T.
      • Soboh B.
      • Lindenstrauss U.
      • Braussemann M.
      • Herzberg M.
      • Nies D.H.
      • et al.
      HypD is the scaffold protein for Fe-(CN)2CO cofactor assembly in [NiFe]-hydrogenase maturation.
      )
      HypE35ATPase, HypE forms the HypEF complex with HypF, involved in synthesis and transfer of the CN ligands(
      • Stripp S.T.
      • Lindenstrauss U.
      • Sawers R.G.
      • Soboh B.
      Identification of an isothiocyanate on the HypEF complex suggests a route for efficient Cyanyl-group channeling during [NiFe]-hydrogenase cofactor generation.
      ,
      • Shomura Y.
      • Higuchi Y.
      Structural basis for the reaction mechanism of S-carbamoylation of HypE by HypF in the maturation of [NiFe]-hydrogenases.
      )
      HypF82Carbamoyl phosphate phosphatase, carbamoyl-transferase, ATPase, involved in CN synthesis(
      • Paschos A.
      • Bauer A.
      • Zimmermann A.
      • Zehelein E.
      • Böck A.
      HypF, a carbamoyl phosphate-converting enzyme involved in [NiFe] hydrogenase maturation.
      )
      HyaD22C-terminal cleavage of the Hyd-1 large subunit(
      • Theodoratou E.
      • Huber R.
      • Böck A.
      [NiFe]-Hydrogenase maturation endopeptidase: structure and function.
      )
      Much is known about the maturation of [NiFe]-Hyd, but the metabolic source(s) and the iron transfer to the active site remains an open question. In the present work, we introduce a novel in vitro reconstitution system for the biosynthesis of Hyd-1 from E. coli and employ it together with biochemical analyses, infrared spectroscopy, MS, and protein interaction studies, to monitor the iron transfer during the maturation process, which allows understanding how the [Fe](CN)2CO cofactor precursor is specifically incorporated into the large subunit of Hyd-1.

      Results

      An in vitro reconstitution system to study the maturation of Hyd-1

      In this study, an enzymatic in vitro maturation assay has been established that allows synthesizing functional [NiFe]-Hyd Hyd-1 of E. coli. This reconstitution system contains only purified components, that is, the apo-form of the large subunit of Hyd-1 (referred to as pre-HyaB), the GD complex, the accessory proteins HypE and HypF, and an activation mix (AM). The reaction was started by addition of the Hyd-1-specific endopeptidase HyaD, which cleaves the C terminus of HyaB by 15 amino acids (
      • Soboh B.
      • Lindenstrauss U.
      • Granich C.
      • Javed M.
      • Herzberg M.
      • Thomas C.
      • et al.
      [NiFe]-hydrogenase maturation in vitro: analysis of the roles of the HybG and HypD accessory proteins1.
      ,
      • Theodoratou E.
      • Huber R.
      • Böck A.
      [NiFe]-Hydrogenase maturation endopeptidase: structure and function.
      ). After incubation for 15 and 30 min, aliquots of the reaction mixture were analyzed by SDS-PAGE for cleavage of pre-HyaB into processed HyaB (Fig. 2, lanes 4 and 5). No cleavage of pre-HyaB could be observed when the reaction was performed in the absence of endopeptidase HyaD (lane 1), under aerobic conditions (lane 2), or in the absence of AM (lane 3). Since HyaB is the only Strep-tagged protein, it can be easily isolated from the reaction mix by Strep-Tactin affinity chromatography. Lane 6 shows both unprocessed pre-HyaB and processed HyaB isolated from the reaction assay after 30 min, whereas a completely processed form of HyaB was isolated from the reaction mix only after 60 min incubation (lane 9). A HisHyaA-enriched fraction (lane 7) was mixed with a fraction of StrepHyaB that was isolated from the reaction mixture after 30 min to reconstitute the active Hyd-1 heterodimer (lane 8). Quality control by attenuated total reflection (ATR) FTIR spectroscopy showed the CO/CN vibrational fingerprint of the cofactor of the in vitro matured Hyd-1 enzyme (Fig. S1). In the absence of the GD complex, HypE, HypF, or AM, the absorption bands of CN and CO were not observed.
      Figure thumbnail gr2
      Figure 2In vitro reconstitution system of Hyd-1 of Escherichia coli. SDS-PAGE analysis of the in vitro assay at different maturation steps. The complete assay includes purified StrepHyaB, HisGD complex, HisHypE and HisHypF, the endopeptidase HisHyaD, and the activation mix (AM). No cleavage of HyaB was observed in the absence of endopeptidase HyaD (lane 1, w/o HyaD) or under aerobic conditions (lane 2, aerobic) or without AM (lane 3, w/o AM). Lanes 4 and 5: aliquots of the reaction mix were analyzed after 15 and 30 min incubation (t15, t30), respectively. Lane 6: unprocessed and processed HyaB isolated from the reaction assay after 30 min incubation. Lane 7: The Hyd-1 small subunit HisHyaA only. Lane 8: The HyaB fraction (t30) was mixed with HisHyaA, and the mixture was applied to Strep-Tactin chromatography. Lane 9: Completely processed StrepHyaB was isolated from the reaction mix after 60 min incubation time. Lane 10: 5 μl of marker proteins of known size (page ruler protein ladder). GD, HybG–HypD; Hyd, hydrogenase.

      Analysis of Hyd-1 activity reconstituted in vitro

      Like all [NiFe]-Hyds, the catalytic large subunit of Hyd-1 (HyaB) alone does not exhibit Hyd activity (
      • Senger M.
      • Stripp S.T.
      • Soboh B.
      Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
      ,
      • Pinske C.
      • Kruger S.
      • Soboh B.
      • Ihling C.
      • Kuhns M.
      • Braussemann M.
      • et al.
      Efficient electron transfer from hydrogen to benzyl viologen by the [NiFe]-hydrogenases of Escherichia coli is dependent on the coexpression of the iron-sulfur cluster-containing small subunit.
      ). Therefore, the purified small subunit of Hyd-1 (HyaA) was added to in vitro reconstituted HyaB yielding the active Hyd-1 heterodimer. Various preparations of in vitro reconstituted Hyd-1 were analyzed for H2 oxidation activity by following H2-dependent reduction of benzyl viologen (BV) photometrically (Fig. 3A). The maximal activity was obtained when all components of the in vitro assay were present (“complete”). The specific activity was 0.75 ± 0.17 μmol H2 oxidized per min−1 mg−1. This value is close to the specific activity of Hyd-1 as isolated from E. coli (
      • Lukey M.J.
      • Parkin A.
      • Roessler M.M.
      • Murphy B.J.
      • Harmer J.
      • Palmer T.
      • et al.
      How Escherichia coli is equipped to oxidize hydrogen under different redox conditions.
      ) and represents the maximum activity of 100%. Our data indicate that the Hyd-1 activity is proportional to the amount of the GD complex added to the in vitro assay, whereas adding higher molar ratios of the GD complex did not increase the Hyd-1 activity. It appears that the optimal amount of GD complex is at a 1:1 stoichiometry (columns 1–6 in Fig. 3A). To define the minimal combination of components that is sufficient to achieve complete maturation, several reactions were conducted in the absence of 1 or more substrates (Fig. 3A). Only background activity was observed when the GD complex, HyaD, HyaA, HyaB, HypF, HypE, or AM were omitted from the assay. Furthermore, only residual activity was measured when the maturation assay was performed under aerobic conditions. Together, these findings indicate that all proteins involved in the maturation assay are necessary and participate directly in the maturation process. Moreover, the in vitro system established in this study was applied to confirm the activity of individual proteins involved in the assay. From a previous study on Hyd-2 of E. coli (
      • Senger M.
      • Stripp S.T.
      • Soboh B.
      Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
      ,
      • Soboh B.
      • Lindenstrauss U.
      • Granich C.
      • Javed M.
      • Herzberg M.
      • Thomas C.
      • et al.
      [NiFe]-hydrogenase maturation in vitro: analysis of the roles of the HybG and HypD accessory proteins1.
      ), increasing amounts of individual components were titrated into reaction assays, and the enzyme activity was only restored when maturation assays were complemented with the missing protein. In this study, a titration of Hyd-1 against the GD complex with all other components (HypE, HypF, and HyaD) in excess was performed. The highest activities of Hyd-1 were obtained when the GD complex was added in equivalent stoichiometric amounts (1:1, Fig. 3). In addition, we observed similar trends by analyzing the activity of Hyd-1 using in-gel activity staining (Fig. 3B). HyaA was added to various compositions of the reaction mix and incubated for 30 min. Then, the mixes were separated by native PAGE, and the H2 oxidation activity was followed by activity staining. The Hyd-1 activity band was only detected in lanes to which all components of the maturation reaction mixture were applied (lanes 2–5). Without HypEF, HyaA, or the AM, no Hyd-1 activity was detected (lanes 6–8). Moreover, the intensity of the activity band significantly decreased when Hyd-1 was reconstituted under aerobic conditions (lane 9). No activity band was observed when HyaB was omitted from reaction mix.
      Figure thumbnail gr3
      Figure 3Kinetic analysis of H2-uptake activity of the in vitro matured Hyd-1. A, the complete assay was composed of purified HyaB, GD complex, HypE, HypF, the endopeptidase HyaD, HyaA, and activation mix (AM). Reactions were conducted in the absence of 1 or more substrates at a time as indicated. The samples were analyzed for activity by following H2-dependent reduction of benzyl viologen photometrically. The maximal attainable activation (100%) was in the presence of all components of the in vitro assay (complete). B, analysis of various reactions of the in vitro matured Hyd-1 using in-gel activity staining. Aliquots of reaction mix (15 μl) were mixed with the small subunit HyaA, separated by native-PAGE and followed by activity stain specific for H2-oxidizing Hyds. Lanes 2 to 5: The Hyd-1 activity band was observed when reaction mixture was complete. No Hyd-1 band was detected in the absence of GD complex (lane 1), without HypEF (lane 6), without HyaA (lane 7), or without AM (lane 8). Note lower intensity of the activity band when Hyd-1 was reconstituted under aerobic conditions (lane 9). GD, HybG–HypD; Hyd, hydrogenase.
      In the next step, we performed in situ ATR FTIR difference spectroscopy (
      • Stripp S.T.
      In Situ infrared spectroscopy for the analysis of gas-processing metalloenzymes.
      ) to confirm the integrity of the [NiFe](CN)2CO active site of reconstituted Hyd-1 enzyme. Figure 4A shows a difference spectrum upon exchange the headspace above the Hyd-1 protein film from N2 to 10% H2. The data clearly comprise CN band features (2100–2050 cm−1) and CO band features (1950–1850 cm−1), indicative of an intact Ni–Fe cofactor. Here, negative bands are observed under N2, whereas positive bands accumulate in the presence of H2. The observed band positions are in excellent agreement with the previously assigned [NiFe]-cofactor states of Hyd-1 from E. coli (
      • Hidalgo R.
      • Ash P.A.
      • Healy A.J.
      • Vincent K.A.
      Infrared spectroscopy during electrocatalytic turnover reveals the Ni-L active site state during H2 oxidation by a NiFe hydrogenase.
      ,
      • Murphy B.J.
      • Hidalgo R.
      • Roessler M.M.
      • Evans R.M.
      • Ash P.A.
      • Myers W.K.
      • et al.
      Discovery of dark pH-dependent H(+) migration in a [NiFe]-hydrogenase and its mechanistic relevance: mobilizing the hydrido ligand of the Ni-C intermediate.
      ,
      • Ash P.A.
      • Hidalgo R.
      • Vincent K.A.
      Proton transfer in the catalytic cycle of [NiFe] hydrogenases: insight from vibrational spectroscopy.
      ) and demonstrate the enrichment of Ni-R2 and Ni-R3 (vCO 1922 and 1914 cm−1, respectively) over Ni-L2 and Ni-L3 (vCO 1876 and 1866 cm−1, respectively) under H2. Smaller negative band features are assigned to cofactor states Ni-C and Ni-SI (vCO 1951 and 1930 cm−1, respectively). These changes result from the oxidation of H2 and the subsequent reduction of the [NiFe]-cofactor. When the enzyme was reduced by deuterium gas (D2) instead, difference spectra show the time-dependent evolution of a broad band at 2500 cm−1 (Fig. S1). Similar traces have been assigned to HDO, which accumulates when D2 is oxidized, and deuterium ions (D+) recombine with OH ions from bulk H2O (
      • Senger M.
      • Stripp S.T.
      • Soboh B.
      Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
      ). This band is a highly specific marker for H2 oxidation activity and proves that Hyd-1 is catalytically active. To probe the inhibition of Hyd-1, we recorded difference spectra in the presence of O2 and CO. Figure 4B demonstrates how the enzymes convert into the Ni-B state (vCO 1942 cm−1) in the presence of 10% O2, whereas Fig. 4C depicts the conversion into Ni-SCO (vCO 1928 cm−1) under 10% CO (note the typical high-frequency Ni-CO band at 2061 cm−1). Based on these data, an experimental band assignment is suggested that agrees well with the in vivo matured Hyd-1 from E. coli (Table 2).
      Figure thumbnail gr4
      Figure 4Infrared spectroscopy of in vitro matured Hyd-1. The CO and CN ligands are annotated in bold and plain fonts, respectively. A, the H2–N2 difference spectrum indicates the conversion of Ni-L2 into Ni-R2 and Ni-R3. Smaller negative bands are assigned to Ni-L3, Ni-C, and Ni-SI. B, in the presence of 10% O2, Hyd-1 converges into the Ni-B state. Negative bands are assigned to Ni-R2 and Ni-R3, with a smaller contribution of Ni-L2. C, the CO-N2 difference spectrum indicates the conversion of Ni-L2 into Ni-SCO (note the typical high-frequency Ni-CO band at 2061 cm−1). Smaller negative bands are assigned to Ni-L3, Ni-C, and Ni-R3. The assignment is detailed in . Hyd, hydrogenase.
      Table 2Experimental band assignment and comparison with the literature (
      • Ash P.A.
      • Hidalgo R.
      • Vincent K.A.
      Proton transfer in the catalytic cycle of [NiFe] hydrogenases: insight from vibrational spectroscopy.
      )
      StateIn vivoIn vitroNi chargeLigand
      Ni-An.o.n.o.3+OOH
      Ni-B19431942OH
      Ni-SI192719302+None
      Ni-SCOn.o1928CO
      Ni-R1n.o.19482+H
      Ni-R219221922
      Ni-R319141914
      Ni-C194919513+H
      Ni-L1189818981+None
      Ni-L218771876
      Ni-L318671866
      Only the CO frequency is given.
      n.o., not observed.

      Iron transfer from the GD complex to the large subunit

      To monitor the iron transfer during the maturation process and to understand how the [Fe](CN)2CO moiety is ultimately incorporated into the large subunit of Hyd-1, we employed our in vitro reconstitution system together with inductively coupled plasma (ICP)–MS, SDS-PAGE, and in-gel activity staining. Our strategy was as follows: (i) perform the in vitro assembly of Hyd-1 in the presence of purified 57Fe-labeled GD complex. (ii) Isolate the catalytic subunit HyaB from the reaction mix. (iii) Analyze SDS-PAGE for cleavage of pre-HyaB into the processed form. (iv) Identification and quantification of 57Fe and metal composition of trapped HyaB preparations by ICP–MS. (v) Supplementation of the in vitro reconstituted HyaB with the small subunit HyaA for activity analysis.
      The results illustrated in Fig. 5, Fig. 6, and Table 3 reveal that in vitro processing of pre-HyaB and the subsequent activation of Hyd-1 correlate very well with the 57Fe and nickel contents of HyaB. SDS-PAGE analysis indicates that only in the presence of all components in the in vitro maturation assay pre-HyaB was completely processed, and the maximal activation of Hyd-1 was achieved (lanes 2–4 in Fig. 5, reaction 1 and 2 in Table 3). Partial cleavage was observed by addition of 0.5 or 0.25 M ratio of the GD complex (lanes 5 and 6 in Fig. 5), and the activity was reduced by approximately 57% and 28%, respectively (reactions 3 and 4 in Table 3). Omission of nickel reduced the restored activity by approximately 22% suggesting that some nickel ions are delivered by the maturation proteins (lane 8 in Fig. 5, reaction 13 in Table 3). In addition, no cleavage and only residual activity were observed in control reactions containing the 57Fe–GD complex that does not harbor the [Fe](CN)2CO moiety. The 57Fe–GD complex in these reactions was isolated either from a ΔhypF-mutant strain or had been treated with EDTA prior to incubation in the reconstitution reaction (lanes 9 and 10 in Fig. 5, reactions 5 and 6 in Table 3). Partial cleavage and a slightly reduced activation were observed when maturation assay was preformed aerobically (lane 11 in Fig. 5, reaction 9 in Table 3). Furthermore, partial cleavage of pre-HyaB was observed when the 57Fe–GD complex was replaced by the 57Fe–HypCD complex (lane 12 in Fig. 5); however, Hyd-1 activity could not be restored (reaction 8 in Table 3).
      Figure thumbnail gr5
      Figure 5Analysis of HyaB processing. SDS-PAGE analysis for the in vitro processing of pre-HyaB. The complete assay includes purified HyaB, GD complex, HypE and HypF, HyaD, and AM. Complete cleavage of pre-HyaB only in the presence of all components of the in vitro system (lanes 2–4). Partial cleavage of pre-HyaB by adding limited amount of 0.5 or 0.25 M ratio of GD complex (lanes 5 and 6). No cleavage was observed without the GD complex or when GD was lacking the [Fe](CN)2CO moiety (lanes 7, 9, and 10), and only partial cleavage can be observed in the absence of nickel (lane 8), or when reaction was performed aerobically (lane 11) or by replacing GD by CD complex (lane 12). AM, activation mix; CD, HypC-HypD; GD, HybG–HypD.
      Figure thumbnail gr6
      Figure 6Correlation of the in vitro activation of Hyd-1 with the 57Fe and nickel contents of HyaB. Titrating the maturation assay with the 57Fe-labeled GD complex (0.25×–2×) increased the 57Fe and nickel content as well as the Hyd-1 activation. Low levels of 57Fe and nickel and only residual activity were observed when the GD complex lacked the [Fe](CN)2CO moiety (GDΔHypF) or by employing truncated HyaB-15AA. Excluding nickel from the maturation assay strongly reduced the activation efficiency; however, the 57Fe content of HyaB was barely affected. In contrast, the presence of iron in the active site seems to be prerequisite for nickel insertion. GD, HybG–HypD; Hyd, hydrogenase.
      Table 3Requirements for Fe transfer into the catalytic subunit during in vitro reconstitution of Hyd-1
      No.Reaction mixture57Fe59Ni(% activation)
      1Complete; 2× 57Fe GD
      A fully 57Fe-labeled GD complex was employed in the maturation assay.
      0.94 ± 0.010.72 ± 0.03100
      2Complete; 1× 57Fe GD0.94 ± 0.020.83 ± 0.03100 ± 6.6
      3Complete; 0.5× 57Fe GD0.46 ± 0.020.60 ± 0.0457 ± 4.6
      4Complete; 0.25× 57Fe GD0.30 ± 0.010.32 ± 0.0328 ± 4.5
      5Complete; 1× 57Fe GDΔhypF
      GD complex with no [Fe](CN)2CO was isolated from a ΔhypF-mutant strain or was desalted after treating with 10 mM EDTA prior to incubation in the reaction.
      0.17 ± 0.060.16 ± 0.000.92 ± 0.03
      6Complete; 1× 57Fe GDEDTA
      GD complex with no [Fe](CN)2CO was isolated from a ΔhypF-mutant strain or was desalted after treating with 10 mM EDTA prior to incubation in the reaction.
      0.20 ± 0.010.01 ± 0.000.88 ± 0.03
      7Complete; 57Fe G + 57Fe D0.25 ± 0.010.03 ± 0.000.74 ± 0.02
      8Complete; 3× 57Fe CD0.38 ± 0.010.12 ± 0.010.17 ± 0.01
      9Complete; aerobic
      Complete reaction mixture was incubated for 60 min in the presence of air prior to activity measurement.
      0.24 ± 0.030.19 ± 0.048.4 ± 1.7
      10Complete; HyaB-15AA0.16 ± 0.010.01 ± 0.000.11 ± 0.04
      11Minus GD complex0.16 ± 0.010.02 ± 0.000.13 ± 0.04
      12Minus AM0.98 ± 0.030.17 ± 0.020.88 ± 0.06
      13Minus nickel1.15 ± 0.0160.33 ± 0.0222.4± 2.8
      14Minus EF0.15 ± 0.050.016 ± 0.001.02 ± 0.04
      15Minus HyaD0.54 ± 0.020.06 ± 0.0021.1 ± 0.06
      The enzyme activities reported are the standard error for 1 representative experiment. The 57Fe and 59Ni metal content was quantified by ICP–MS. The molar ratio (mol metal/mol protein) is calculated based on the molecular mass of the HyaB subunit (66.5 kDa). Metal contents were measured with 2 independent technical replicates. Data are averages of at least 2 independent determinations and reported as the mean ± SD.
      a A fully 57Fe-labeled GD complex was employed in the maturation assay.
      b GD complex with no [Fe](CN)2CO was isolated from a ΔhypF-mutant strain or was desalted after treating with 10 mM EDTA prior to incubation in the reaction.
      c Complete reaction mixture was incubated for 60 min in the presence of air prior to activity measurement.
      Metal analysis of the HyaB preparations isolated from various maturation assays indicates that the completely processed HyaB protein contains 1 mol 57Fe and 1 mol nickel per mole protein (Table 3). The amount of copper and zinc was not significant. Up to a molar ratio of 1:1, the 57Fe content of HyaB increases almost linearly with the amount of the GD complex in the reaction mix (Fig. 6 and reactions 1–4 in Table 3). The same trend is observed for nickel and the activation efficiency of Hyd-1. The 57Fe molar ratio of HyaB isolated from maturation assays in which the GD complex was lacking the [Fe](CN)2CO moiety was less than 0.2 (GDΔhypF in Fig. 6 and reactions 5, 6, and 11 in Table 3). For the first time, these data demonstrate that the source of 57Fe on HyaB and Hyd-1 is the [Fe](CN)2CO moiety of the GD complex.
      When 57Fe-HybG and 57Fe-HypD were added to the reaction mix instead of the 57Fe–GD complex, an 57Fe molar ratio of 0.25 was detected, and the activity of Hyd-1 could not be restored (reaction 7). These results indicate that HybG and HypD individually cannot replace the GD complex and that the physical heterodimer of HybG and HypD is essential for a successful iron transfer. When the 57Fe–GD complex was replaced by 3 equivalents of the 57Fe–CD complex, an 57Fe molar ratio of 0.38 was found; however, only background activity could be observed (reaction 8). When the reaction was performed under aerobic conditions, a 57Fe molar ratio of 0.24 was detected, suggesting that iron transfer is inhibited by O2 (reaction 9). In the absence of HypE and HypF, the 57Fe content of HyaB was very low and Hyd-1 activity could not be restored (reaction 14). HyaB matured in the absence of HyaD protease was found to contain a reduced 57Fe molar ratio of 0.54 and a very low nickel content (reaction 15) suggesting a possible role of HyaD in the insertion of metal ions into HyaB. To determine whether the C-terminal extension of HyaB is required for the iron transfer, pre-HyaB was replaced by HyaB-15AA. This genetic construct resembles pre-HyaB truncated for 15 amino acids at the C terminus. The HyaB-15AA preparation could not restore Hyd-1 activity and was devoid of 57Fe (Fig. 6 and reaction 10). These results are in line with our previous findings that the GD complex was not able to recognize HybC-15AA, the large subunit of Hyd-2 from E. coli (
      • Senger M.
      • Stripp S.T.
      • Soboh B.
      Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
      ). Excluding nickel or the AM from the maturation assay reduced the activation efficiency of Hyd-1 to 22%; however, the 57Fe content of HyaB was barely affected (Fig. 6 and reactions 12 and 13). In contrast, the presence of iron in the active site appears to be a prerequisite for nickel insertion. Our novel finding shown in Fig. 6 and reactions 1 to 14 that nickel insertion takes place only when iron has been inserted before. Together, these data show that iron was exclusively incorporated into HyaB under conditions where a full assembly of the [NiFe]-cofactor was observed, which demonstrates that the [Fe](CN)2CO moiety was specifically transferred into the catalytic subunit HyaB.

      Protein–protein interaction during iron transfer to the active site

      To gain additional insights into the specificity of the GD complex to deliver the [Fe]CN2CO moiety for the activation of Hyd-1 and Hyd-2 but not Hyd-3, the binding affinities of the GD complex to the apo-form of the catalytic subunits of Hyd-1 (pre-HyaB), Hyd-2 (pre-HybC), and Hyd-3 (pre-HycE) were determined. For the microscale thermophoresis analysis (MST), the concentration of the dye-labeled catalytic subunits was kept at 50 nM throughout a titration with increasing concentrations of the GD complex. Figure 7A depicts how the MST signal for the dye-labeled catalytic subunits pre-HyaB and pre-HybC depends on the GD complex. Fitting these data via the Hill equation allowed obtaining the apparent binding affinity (Kd) between the interacting proteins. The data shown in Fig. 7A indicate that the GD complex displayed a higher binding affinity to pre-HybC (Kd = 46 ± 4.2 μM) than to pre-HyaB (Kd = 171 ± 24.4 μM). No binding was observed for the interaction between the GD complex and pre-HycE. It was unclear whether the binding of the GD complex to pre-HyaB and pre-HybC is mediated by the GD complex, HybG, or HypD protein. Figure 7B reveals that HybG interacts equally well with pre-HyaB (30 ± 29 μM) and pre-HybC (23 ± 2.9 μM), whereas no interaction between HypD and pre-HyaB or pre-HybC was detected. Figure 7C shows that the CD complex displayed a higher binding affinity to pre-HycE (Kd = 34 ± 3.2 μM) than to pre-HyaB (Kd = 185 ± 25.1 μM). No detectable binding affinity could be measured for the interaction between the CD complex and pre-HybC. Furthermore, MST experiments suggest a very weak interaction between the GD complex and HyaB-15AA. In summary, these data clearly demonstrate that [Fe](CN)2CO moiety is incorporated into the active site of each Hyd by a specific interaction with the corresponding maturation complex (GD or CD).
      Figure thumbnail gr7
      Figure 7The binding affinities of the catalytic subunit precursors pre-HyaB, pre-HybC, and pre-HycE to GD and HypCD complexes. A, GD complex dependence of the MST signal for the fluorescence-labeled pre-HyaB and pre-HybC. B, HybG-dependent fluorescence signal of pre-HyaB and pre-HybC. C, HypCD complex-dependent fluorescence signal of pre-HycE and pre-HyaB. The MST signals of at least 15 different concentrations of pre-HyaB, pre-HybC, or pre-HycE ranging from 0.1 to 1000 μM were fitted according to Michaelis–Menten kinetics. The insets in panel depict the obtained binding affinities, Kd. Data of at least 2 independent measurements (n) were combined to create 1 dataset for which the averaged data points as the mean ± SD were fitted by NanoTemper analysis software using a Kd model. The data show higher binding affinities of pre-HyaB and pre-HybC to GD complex and HybG, whereas pre-HycE shows the highest binding affinity to HypCD. The Hill coefficients derived from the fits () indicate positive cooperativity (n > 1) in all MST experiments. CD, HypC–HypD; GD, HybG–HypD; MST, microscale thermophoresis analysis.

      Discussion

      In the present study, we report a biochemical in vitro assay for the study of the maturation of the O2-tolerant [NiFe]-Hyd Hyd-1 of E. coli. In previous work, we demonstrated the reconstitution of the O2-sensitive [NiFe]-Hyd Hyd-2 of E. coli (
      • Senger M.
      • Stripp S.T.
      • Soboh B.
      Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
      ). Both approaches exclusively rely on purified and defined components and thus provide the possibility to elucidate the biogenesis of the [NiFe]-cofactor at the molecular level. The properties of the in vitro matured Hyd-1 were comparable to the in vivo matured enzyme, for example, the specific H2 uptake activities (
      • Lukey M.J.
      • Parkin A.
      • Roessler M.M.
      • Murphy B.J.
      • Harmer J.
      • Palmer T.
      • et al.
      How Escherichia coli is equipped to oxidize hydrogen under different redox conditions.
      ), as well as the FTIR signature under N2, H2, O2, or CO (
      • Hidalgo R.
      • Ash P.A.
      • Healy A.J.
      • Vincent K.A.
      Infrared spectroscopy during electrocatalytic turnover reveals the Ni-L active site state during H2 oxidation by a NiFe hydrogenase.
      ,
      • Murphy B.J.
      • Hidalgo R.
      • Roessler M.M.
      • Evans R.M.
      • Ash P.A.
      • Myers W.K.
      • et al.
      Discovery of dark pH-dependent H(+) migration in a [NiFe]-hydrogenase and its mechanistic relevance: mobilizing the hydrido ligand of the Ni-C intermediate.
      ,
      • Ash P.A.
      • Hidalgo R.
      • Vincent K.A.
      Proton transfer in the catalytic cycle of [NiFe] hydrogenases: insight from vibrational spectroscopy.
      ).
      Although major advances in our understanding of the biosynthesis of the [NiFe]-cofactor have been achieved by the group of August Böck (
      • Böck A.
      • King P.W.
      • Blokesch M.
      • Posewitz M.C.
      Maturation of hydrogenases.
      ), we only know certain details about this complex maturation pathway. There is a considerable amount that remains to be understood, for example, neither the metabolic route of CO on the maturation pathway is known nor is anything known about the source of iron in the active site. The map of interactions among the maturation proteins and the sequence of events that occur along the pathway of [NiFe](CN)2CO biosynthesis are not completely understood. Therefore, the development of an in vitro reconstitution system was necessary to close the gaps in our understanding of [NiFe]-Hyd maturation. One of the greatest challenges investigating [NiFe]-Hyd maturation in vivo is that it cannot necessarily distinguish between the phenotypes of mutants that are impaired in the same biochemical function. For example, although hypD, hype, and hypF are 3 distinct maturation genes, a mutation in any one of them gives exactly the same phenotype, that is, an unprocessed catalytic subunit and inactive Hyd (
      • Böck A.
      • King P.W.
      • Blokesch M.
      • Posewitz M.C.
      Maturation of hydrogenases.
      ). The development of a well-defined in vitro reconstitution system allowed us to overcome such problems and exert much greater control over the [NiFe]-Hyd maturation machinery.
      In the present study, our in vitro reconstitution system was employed to monitor the iron transfer during the maturation process. We could demonstrate direct transfer of 57Fe from 57Fe-labeled GD complex to the catalytic subunit of Hyd-1 (pre-HyaB). Several control experiments suggested that the [Fe](CN)2CO moiety was specifically incorporated into the active site of Hyd-1. Transfer of 57Fe from the GD complex to the active site was observed exclusively when all components of the in vitro maturation assay were present; and only very low levels of 57Fe content were found in HyaB when [NiFe]-cofactor assembly was prevented. Our findings are compatible with the assumption that a direct transfer of the [Fe](CN)2CO moiety from the GD complex to pre-HyaB takes place via a thiol-disulfide substitution and redox reactions. It has been suggested that the formation of disulfide by oxidation of the cysteines that usually coordinate the active-site cofactor prevents the incorporation of the [Fe](CN)2CO moiety into the active site (
      • Böck A.
      • King P.W.
      • Blokesch M.
      • Posewitz M.C.
      Maturation of hydrogenases.
      ). It is currently unclear why only low levels of 57Fe were transferred from the GD complex to HyaB when HypEF were excluded from the reaction mix, since both accessory proteins are required in early maturation steps for the synthesis and transfer of the CN ligands to the GD complex (
      • Paschos A.
      • Bauer A.
      • Zimmermann A.
      • Zehelein E.
      • Böck A.
      HypF, a carbamoyl phosphate-converting enzyme involved in [NiFe] hydrogenase maturation.
      ,
      • Stripp S.T.
      • Lindenstrauss U.
      • Sawers R.G.
      • Soboh B.
      Identification of an isothiocyanate on the HypEF complex suggests a route for efficient Cyanyl-group channeling during [NiFe]-hydrogenase cofactor generation.
      ). Our data now indicate that their presence is also important for incorporation of the [Fe](CN)2CO into HyaB. We propose that the HypEF complex may facilitate substrate channeling, as suggested earlier (
      • Stripp S.T.
      • Lindenstrauss U.
      • Sawers R.G.
      • Soboh B.
      Identification of an isothiocyanate on the HypEF complex suggests a route for efficient Cyanyl-group channeling during [NiFe]-hydrogenase cofactor generation.
      ,
      • Soboh B.
      • Lindenstrauss U.
      • Granich C.
      • Javed M.
      • Herzberg M.
      • Thomas C.
      • et al.
      [NiFe]-hydrogenase maturation in vitro: analysis of the roles of the HybG and HypD accessory proteins1.
      ). It is also conceivable that HypE facilitates dissociation of the GD complex from HyaB upon delivering the [Fe](CN)2CO moiety in order to enter a new cycle of biosynthesis.
      The accessory proteins HypA, HypB, and SlyD are specifically involved in the acquisition and insertion of nickel into the catalytic subunit (
      • Böck A.
      • King P.W.
      • Blokesch M.
      • Posewitz M.C.
      Maturation of hydrogenases.
      ,
      • Watanabe S.
      • Kawashima T.
      • Nishitani Y.
      • Kanai T.
      • Wada T.
      • Inaba K.
      • et al.
      Structural basis of a Ni acquisition cycle for [NiFe] hydrogenase by Ni-metallochaperone HypA and its enhancer.
      ). The requirement for these metallochaperones was obviated by the addition of a comparatively high concentration of NiCl2 to the in vitro assay (
      • Senger M.
      • Stripp S.T.
      • Soboh B.
      Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
      ,
      • Soboh B.
      • Lindenstrauss U.
      • Granich C.
      • Javed M.
      • Herzberg M.
      • Thomas C.
      • et al.
      [NiFe]-hydrogenase maturation in vitro: analysis of the roles of the HybG and HypD accessory proteins1.
      ). Omission of the AM or nickel ions from the maturation assay strongly affected the activation of Hyd-1; however, the 57Fe content of HyaB was not affected at all. These findings indicate that iron incorporation does not rely on the presence of nickel in the active site. In contrast, it can be concluded from all reactions in Table 3 that the insertion of nickel occurs after the insertion of the [Fe](CN)2CO moiety. Our conclusion is supported by isolation of a Hyd large subunit from Cupriavidus necator (formerly Ralstonia eutropha) that was loaded with the [Fe](CN)2CO moiety but not nickel (
      • Winter G.
      • Buhrke T.
      • Lenz O.
      • Jones A.K.
      • Forgber M.
      • Friedrich B.
      A model system for [NiFe] hydrogenase maturation studies: purification of an active site-containing hydrogenase large subunit without small.
      ,
      • Loscher S.
      • Zebger I.
      • Andersen L.K.
      • Hildebrandt P.
      • Meyer-Klaucke W.
      • Haumann M.
      The structure of the Ni-Fe site in the isolated HoxC subunit of the hydrogen-sensing hydrogenase from Ralstonia eutropha.
      ). A similar sequence of events in which iron is delivered first was derived from the observation that nickel is not inserted if the iron maturation proteins are deficient (
      • Böck A.
      • King P.W.
      • Blokesch M.
      • Posewitz M.C.
      Maturation of hydrogenases.
      ,
      • Theodoratou E.
      • Paschos A.
      • Magalon A.
      • Fritsche E.
      • Huber R.
      • Böck A.
      Nickel serves as a substrate recognition motif for the endopeptidase involved in hydrogenase maturation.
      ,
      • Blokesch M.
      • Bock A.
      Maturation of [NiFe]-hydrogenases in Escherichia coli: the HypC cycle.
      ).
      Although the use of the in vitro system has yielded significant information concerning the requirements of iron delivery to the active site of [NiFe]-Hyd, the metabolic source of iron as it enters the maturation was not addressed in the current study. The HybG protein (or its homolog HypC) might be involved in recruiting both the iron and CO2 molecules for further reductive synthesis of the CO ligand by HypD (
      • Soboh B.
      • Stripp S.T.
      • Bielak C.
      • Lindenstrauss U.
      • Braussemann M.
      • Javaid M.
      • et al.
      The [NiFe]-hydrogenase accessory chaperones HypC and HybG of Escherichia coli are iron- and carbon dioxide-binding proteins.
      ,
      • Stripp S.T.
      • Oltmanns J.
      • Muller C.S.
      • Ehrenberg D.
      • Schlesinger R.
      • Heberle J.
      • et al.
      Electron inventory of the iron-sulfur scaffold complex HypCD essential in [NiFe]-hydrogenase cofactor assembly.
      ). In future work, we will investigate whether the iron-bound CO2 ligand coordinated by HybG or HypC has only catalytic function or it ends up in the catalytic site.
      It is conceivable that the incorporation of the [Fe](CN)2CO moiety into the active site is based on specific molecular interactions between the GD complex and the catalytic subunit precursor. So far, the exact nature of this interactions and specificity of GD complex to deliver the [Fe](CN)2CO moiety to the catalytic subunit of Hyd-1 and Hyd-2 but not Hyd-3 is poorly understood (
      • Blokesch M.
      • Magalon A.
      • Böck A.
      Interplay between the specific chaperone-like proteins HybG and HypC in maturation of hydrogenases 1, 2, and 3 from Escherichia coli.
      ). In this study, MST was applied to determine the binding affinities of the GD and CD complex to the catalytic subunit of the 3 Hyds from E. coli. The data show that HybG and not HypD interacts specifically with the large subunit precursor of Hyd-1 and Hyd-2 (pre-HybC and pre-HyaB), whereas its homolog HypC interacts with pre-HycE and has a weaker affinity to pre-HyaB. These findings suggest that the interaction surface of pre-HybC and pre-HyaB to HybG has different structural features than pre-HycE. On the other hand, both HybG and HypC form individually a tight complex with HypD suggesting that HybG and HypC have similar structural features at the interaction surface to HypD. The obtained results support a model in which HypD is a central component during [Fe](CN)2CO synthesis (
      • Stripp S.T.
      • Soboh B.
      • Lindenstrauss U.
      • Braussemann M.
      • Herzberg M.
      • Nies D.H.
      • et al.
      HypD is the scaffold protein for Fe-(CN)2CO cofactor assembly in [NiFe]-hydrogenase maturation.
      ), whereas HybG (or its homolog HypC) but not HypD binds specifically to the large subunit precursor of the corresponding Hyd. Overall, the relative affinities of these complexes are on the order of [GD × pre-HybC] > [GD × pre-HyaB], whereas [CD × pre-HycE] > [CD × pre-HyaB] (see Fig. 2 for comparison). It is not completely clear why and how the GD complex specifically binds and mature Hyd-1 and Hyd-2 only while the CD complex is required for the maturation of Hyd-3. Structural comparisons of these precursors will be essential to validate and complement the present results. We suggest that the specificity of the GD and CD complex may determine which [NiFe]-Hyd is matured, depending on the physiological conditions that may require a H2-evolving enzyme (Hyd-3) or H2-oxidizing enzymes (Hyd-1 and Hyd-2) (
      • Sargent F.
      The model [NiFe]-Hydrogenases of Escherichia coli.
      ,
      • Haase A.
      • Sawers R.G.
      Exchange of a single amino acid residue in the HybG chaperone allows maturation of all H2-activating [NiFe]-Hydrogenases in Escherichia coli.
      ).
      Further downstream along the maturation pathway, our data indicate that the C-terminal extension of HyaB is required for an efficient incorporation of [Fe](CN)2CO into pre-HyaB, as a truncated variant without the C-terminal excision (HyaB-15AA) was devoid of 57Fe. This agrees with our MST data that revealed negligible interaction of the GD complex with the truncated variant HyaB-15AA. These findings are in line with previous pull-down interaction experiments (
      • Senger M.
      • Stripp S.T.
      • Soboh B.
      Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
      ) and other studies showing that HybG forms a complex with pre-HybC but not with the truncated variant HybC-15AA (
      • Blokesch M.
      • Magalon A.
      • Böck A.
      Interplay between the specific chaperone-like proteins HybG and HypC in maturation of hydrogenases 1, 2, and 3 from Escherichia coli.
      ,
      • Thomas C.
      • Muhr E.
      • Sawers R.G.
      Coordination of synthesis and assembly of a modular membrane-associated [NiFe]-hydrogenase is determined by cleavage of the C-terminal peptide.
      ). We presume that different conformations of the large subunit upon removal of the C-terminal extension may prevent the interaction with the GD complex and thus the iron transfer (
      • Senger M.
      • Stripp S.T.
      • Soboh B.
      Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
      ). Most likely, the C-terminal extension is maintaining in a “correct” conformation to facilitate interaction with the GD complex and to ensure that the active-site cavity remains open until the cofactor is inserted (
      • Kwon S.
      • Watanabe S.
      • Nishitani Y.
      • Kawashima T.
      • Kanai T.
      • Atomi H.
      • et al.
      Crystal structures of a [NiFe] hydrogenase large subunit HyhL in an immature state in complex with a Ni chaperone HypA.
      ,
      • Pinske C.
      • Thomas C.
      • Nutschan K.
      • Sawers R.G.
      Delimiting the function of the C-terminal extension of the Escherichia coli [NiFe]-hydrogenase 2 large subunit precursor.
      ).
      Taken together, our data demonstrate how the [Fe](CN)2CO cofactor precursor of the GD complex is specifically incorporated into the active site of the corresponding [NiFe]-Hyd. We identified the requirements for the iron incorporation into catalytic subunit during enzyme maturation and provide unprecedented direct evidence for the transfer of iron to the catalytic subunit using 57Fe isotope editing.

      Experimental procedures

      Bacterial strains and plasmids

      All strains and plasmids used are listed in Table 4. To produce the recombinant GD complex, the plasmid pThypDEFCStrepXT (
      • Stripp S.T.
      • Oltmanns J.
      • Muller C.S.
      • Ehrenberg D.
      • Schlesinger R.
      • Heberle J.
      • et al.
      Electron inventory of the iron-sulfur scaffold complex HypCD essential in [NiFe]-hydrogenase cofactor assembly.
      ) was digested with NdeI and BamH1 releasing the hypC gene. The PCR-amplified hybG gene was then ligated into the NdeI and BamH1 restriction sites of pT-hypDEF to generate the pT-hypDEFGStrepXT plasmid encoding a twin-Strep-tag fused to HybG at the C terminus. The plasmid pT-hypDEFGHis was used to produce the GD complex with a His-tag attached to HybG.
      Table 4Strains and plasmids used in this study
      Strains/plasmidsGenotypeReferences
      MC4100F araD139 α(argF-lac)U169 ptsF25 deoC1 relA1 flbB5301 rspL150(
      • Casadaban M.J.
      Transposition and fusion of lac genes to selected promoters in Escherichia-coli using bacteriophage-lambda and bacteriophage-Mu.
      )
      BL21(DE3)F-ompT hsdSB(rB-,mB-)gal dcm (DE3)Novagen
      DHPF-2MC4100 ΔhypF(
      • Paschos A.
      • Bauer A.
      • Zimmermann A.
      • Zehelein E.
      • Böck A.
      HypF, a carbamoyl phosphate-converting enzyme involved in [NiFe] hydrogenase maturation.
      )
      FTD147MC4100 ΔhyaB ΔhybC ΔhycE(
      • Redwood M.D.
      • Mikheenko I.P.
      • Sargent F.
      • Macaskie L.E.
      Dissecting the roles of Escherichia coli hydrogenases in biohydrogen production.
      )
      Plasmids
      pT-hypCDEFpT7-7,hypD, HypE, hypC-strep, hypF, AmpR(
      • Blokesch M.
      • Albracht S.P.
      • Matzanke B.F.
      • Drapal N.M.
      • Jacobi A.
      • Böck A.
      The complex between hydrogenase-maturation proteins HypC and HypD is an intermediate in the supply of cyanide to the active site iron of [NiFe]-hydrogenases.
      )
      pT-hypDEFGStrepXTpT7-7, hypD, hypE, hybG-XT-Strep-tag, hypF, AmpRThis study
      pT-hypDEFGHispT7-7, hypD, hypE, hybG-His-tag, hypF, kmRThis study
      pT-hypDEGHispET30, hypD, hypE, hybG-His-tag, AmpRThis study
      p-hypFpET28 A, encodes His-HypF, kanR(
      • Soboh B.
      • Kruger S.
      • Kuhns M.
      • Pinske C.
      • Lehmann A.
      • Sawers R.G.
      Development of a cell-free system reveals an oxygen-labile step in the maturation of [NiFe]-hydrogenase 2 of Escherichia coli.
      )
      p-hypEpET28 A, encodes His-HypE, kanR(
      • Soboh B.
      • Kruger S.
      • Kuhns M.
      • Pinske C.
      • Lehmann A.
      • Sawers R.G.
      Development of a cell-free system reveals an oxygen-labile step in the maturation of [NiFe]-hydrogenase 2 of Escherichia coli.
      )
      p-hypDpET28 A, encodes His-HypD, kanR(
      • Soboh B.
      • Kruger S.
      • Kuhns M.
      • Pinske C.
      • Lehmann A.
      • Sawers R.G.
      Development of a cell-free system reveals an oxygen-labile step in the maturation of [NiFe]-hydrogenase 2 of Escherichia coli.
      )
      phybGHispET30, hybG-His, AmpRThis study
      PC-HyaDpCA24 N, hyaD, encodes His-HyaD
      pC-hyaBpET28 A, hybB, encodes His-HyaB, kanRThis study
      pASK-HyaBpASK-IBA5, hyaB, encodes Strep-HyaB, AmpR(
      • Soboh B.
      • Kruger S.
      • Kuhns M.
      • Pinske C.
      • Lehmann A.
      • Sawers R.G.
      Development of a cell-free system reveals an oxygen-labile step in the maturation of [NiFe]-hydrogenase 2 of Escherichia coli.
      )
      pASK-HyaB-15AALike pASK-HyaB but without the 15AA C-terminal peptideThis study
      pASK-HybCpASK-IBA5, encodes Strep-HybC, AmpR(
      • Pinske C.
      • Kruger S.
      • Soboh B.
      • Ihling C.
      • Kuhns M.
      • Braussemann M.
      • et al.
      Efficient electron transfer from hydrogen to benzyl viologen by the [NiFe]-hydrogenases of Escherichia coli is dependent on the coexpression of the iron-sulfur cluster-containing small subunit.
      )
      pASK-HycEpASK-IBA5, encodes Strep-HycE, AmpR(
      • Pinske C.
      • Kruger S.
      • Soboh B.
      • Ihling C.
      • Kuhns M.
      • Braussemann M.
      • et al.
      Efficient electron transfer from hydrogen to benzyl viologen by the [NiFe]-hydrogenases of Escherichia coli is dependent on the coexpression of the iron-sulfur cluster-containing small subunit.
      )
      phyaAHispET30, hyaA-His, AmpRThis study

      Culture conditions

      E. coli strain BL21(DE3) (
      • Studier F.W.
      • Moffatt B.A.
      Use of bacteriophage T7 RNA polymerase to direct selective high-level expression of cloned genes.
      ) or strains transformed with the indicated plasmids (
      • Paschos A.
      • Bauer A.
      • Zimmermann A.
      • Zehelein E.
      • Böck A.
      HypF, a carbamoyl phosphate-converting enzyme involved in [NiFe] hydrogenase maturation.
      ,
      • Redwood M.D.
      • Mikheenko I.P.
      • Sargent F.
      • Macaskie L.E.
      Dissecting the roles of Escherichia coli hydrogenases in biohydrogen production.
      ) (Table 4) were used as host for overproduction of recombinant proteins. Cultivation of cells was performed in LB medium at 37 °C. Gene expression was induced by the addition of 0.2 μg/ml anhydrotetracycline or 50 μM IPTG followed by continuing incubation at 25 °C for a further 6 h. Cells were harvested when cultures had reached an absorbance of 0.9 to 1.3 by centrifugation at 10,000g for 30 min. Washed cell pellets were either used immediately or stored at −80 °C. To produce the 57Fe-labeled GD complex, 57Fe-HypD, and 57Fe-HybG, the bacteria were grown anaerobically in modified M9 minimal medium exactly as described previously (
      • Stripp S.T.
      • Oltmanns J.
      • Muller C.S.
      • Ehrenberg D.
      • Schlesinger R.
      • Heberle J.
      • et al.
      Electron inventory of the iron-sulfur scaffold complex HypCD essential in [NiFe]-hydrogenase cofactor assembly.
      ).

      Protein purification

      All steps for protein purification were carried out in an anaerobic chamber (Coy Laboratories) under a N2/H2 mixture (99:1) and at 4 ºC. Wet cell paste containing strep-tagged protein was suspended at a ratio of 1:4 (w/v) in buffer W (100 mM Tris–HCl [pH 8], 150 mM NaCl, 10 mM magnesium acetate, and 1% glycerol) including 2 mM sodium dithionite, 5 μg/ml DNase, and one tablet per 100 ml of Protease Inhibitor Cocktail (Roche). Cells were disrupted by sonication (50 W for 30 min with 0.5 s pulses). Large material and cell debris were removed by centrifugation of the suspension at 15,000g for 30 min. The cell-free extract obtained after centrifugation was used directly for anaerobic protein purification by chromatography on a Strep-Tactin-XT column (IBA Lifesciences), exactly as described (
      • Stripp S.T.
      • Oltmanns J.
      • Muller C.S.
      • Ehrenberg D.
      • Schlesinger R.
      • Heberle J.
      • et al.
      Electron inventory of the iron-sulfur scaffold complex HypCD essential in [NiFe]-hydrogenase cofactor assembly.
      ,
      • Senger M.
      • Stripp S.T.
      • Soboh B.
      Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
      ). Wet cell paste containing His-tagged protein was resuspended at a ratio of 1:6 (w/v) in buffer A (50 mM Tris–HCl [pH 8.0] and 300 mM NaCl) including 5 μg/ml DNase and Protease Inhibitor Cocktail. Target protein was isolated by affinity chromatography using cobalt-charged Talon resins (Clontech). Removal of nonspecifically binding contaminants was achieved by washing with 10 column volumes of buffer A, followed by five column volumes of buffer A supplemented with 10 mM imidazole and then three column volumes of buffer A supplemented with 20 to 30 mM imidazole. His-tagged protein was subsequently eluted with buffer A containing 300 mM imidazole. Imidazole was removed by a desalting column and applied to a Superdex 200 gel filtration column. Proteins were concentrated as described previously. For long-term storage, purified protein drops were directly frozen in liquid nitrogen.

      The in vitro reconstitution system of Hyd-1 of E. coli

      The in vitro maturation reactions were carried out in an anaerobic chamber (Coy Laboratories) under a N2/H2 mixture (99:1). The reaction mixture was prepared in 50 mM Mops KOH (pH 7.0) by combining 4 μM of purified StrepHyaB and 4 μM of HisGD complex in equimolar amounts; and 2 μM of HisHypE and 2 μM of HisHypF in stoichiometric amounts (
      • Senger M.
      • Stripp S.T.
      • Soboh B.
      Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
      ) and an AM consisting of 2.5 mM ATP, 50 μM CP, 100 μM NiCl2, and 2 mM sodium dithionite. The reaction mixtures were incubated at RT for 10 to 15 min. The reaction was initiated by the addition of 1 μM of the endopeptidase HisHyaD in catalytic amounts. After incubation on a shaker at ambient temperature (24–26 °C) for further 30 min, 20 μl aliquots of reaction mixture were analyzed by SDS-PAGE. StrepHyaB was isolated from the reaction mix by Strep-Tactin affinity chromatography applying buffer W containing 300 mM NaCl and 5 mM dithiothreitol, and aliquots of the sample were analyzed for cleavage of HyaB by SDS-PAGE. In the following, the unprocessed large subunit is referred to as pre-HyaB. After puffer exchange, StrepHyaB preparations were concentrated for infrared spectroscopy and metal analysis. The in vitro matured StrepHyaB protein was mixed with the small subunit of Hyd-1 (HisHyaA). Active StrepHyaB–HisHyaA heterodimer was isolated by immobilized metal affinity chromatography via cobalt sepharose resin. The activity of Hyd-1 was determined by following the reduction of BV and infrared spectroscopy as reported earlier (
      • Senger M.
      • Mebs S.
      • Duan J.
      • Wittkamp F.
      • Apfel U.P.
      • Heberle J.
      • et al.
      Stepwise isotope editing of [FeFe]-hydrogenases exposes cofactor dynamics.
      ). Large-scale preparation of maturation reaction mix up to 15 ml was applied to isolate preparative amounts of in vitro matured StrepHyaB. In control experiments, various components of the complete reaction mixture were excluded as indicated in the Results and Discussion sections.

      Determination of Hyd activity

      Hyd activity was assayed by following the H2-dependent reduction of BV at 578 nm using an Uvikon 900 dual wavelength spectrophotometer. The activities were measured under strictly anoxic conditions in 1.5 ml anaerobic cuvettes contained 0.8 ml volumes of 4 mM BV in 50 mM Mops buffer at pH 7.0. Cuvettes sealed with rubber stoppers were flushed with a 100% H2. One unit of H2 oxidation activity corresponds to the reduction of 2 μmol of BV per min.

      PAGE and metal quantification

      Protein samples (5–15 μg) were separated by SDS-PAGE (NuPAGE 4–20% Tris–glycine; Invitrogen). Protein MS was applied to confirm the protein identity as described previously (
      • Seidel K.
      • Kuhnert J.
      • Adrian L.
      The complexome of Dehalococcoides mccartyi reveals its organohalide respiration-complex is modular.
      ). Nondenaturing PAGE was applied to analyze the activity of in vitro matured Hyd-1 using in-gel activity staining. Protein samples and gel electrophoresis running buffer were supplemented with 4 mM n-dodecyl-ß-maltoside. The gel was subsequently stained for enzyme activity as reported previously (
      • Ballantine S.P.
      • Boxer D.H.
      Nickel-containing hydrogenase isoenzymes from anaerobically grown Escherichia coli K-12.
      ). Protein concentration was determined frequently using the Folin phenol reagent by the procedure originally described by Lowry (
      • Lowry O.H.
      • Rosebrough N.J.
      • Farr A.L.
      • Randall R.J.
      Protein measurement with the Folin phenol reagent.
      ). Protein samples were analyzed for iron, nickel, zinc, and copper by ICP–QqQ–MS as described previously (
      • Budhraja R.
      • Ding C.
      • Walter P.
      • Wagner S.
      • Reemtsma T.
      • Gary Sawers R.
      • et al.
      The impact of species, respiration type, growth phase and genetic inventory on absolute metal content of intact bacterial cells.
      ).

      Infrared spectroscopy

      ATR FTIR spectroscopy was carried out in an anaerobic glovebox under a N2/H2 mixture (99:1). All spectroscopic experiments were recorded at ambient temperature (24–26 °C) on a Tensor27 FTIR spectrometer (Bruker Optik) as described previously (
      • Senger M.
      • Stripp S.T.
      • Soboh B.
      Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
      ,
      • Stripp S.T.
      In Situ infrared spectroscopy for the analysis of gas-processing metalloenzymes.
      ,
      • Senger M.
      • Laun K.
      • Soboh B.
      • Stripp S.T.
      Infrared Characterization of the bidirectional oxygen-sensitive [NiFe]-hydrogenase from E-coli.
      ). The protein sample (1 μl) was deposited on the three-reflection silicon crystal of an ATR cell, dried under 100% N2 gas, and rehydrated with buffer solution in the humidified gas stream (
      • Stripp S.T.
      In Situ infrared spectroscopy for the analysis of gas-processing metalloenzymes.
      ). Reactive gases (H2, CO, and O2) were added via digital mass flow controllers (Sierra).

      Protein interaction studies using MST

      Protein-binding experiments were carried out with the Monolith NT.115 (NanoTemper Technologies). Protein labeling was carried out according to the user manuals of the respective labeling kit (RED-NHS or with the dye NT-647). The labeling reaction was carried out for 60 min in the dark at ambient temperature (24–26 °C). Excess dye was removed by a desalting column included in the labeling kit. The buffer used for both protein labeling and MST measurements contained 100 mM Tris–HCl (pH 8), 150 mM NaCl, 10 mM MgCl2, 1% glycerol, and 0.05% Tween-20. Labeled proteins (pre-HyaB, HyaB-15AA, pre-HybC, or pre-HycE) were adjusted to 40 to 50 nM with the buffer mentioned previously. Then, a 16 step 1:1 serial dilution of the “ligand” (GD complex, HybG, HypD, or CD complex) with a final volume of 10 μl was prepared resulting in concentrations from 0.1 to 1000 μM. Next, 10 μl of each dilution was mixed with 10 μl of labeled protein. The mixture was incubated for 30 min and followed by centrifugation at 20,000g for 10 min. Afterward, the samples were loaded onto premium MST capillaries. Experiments were carried out at LED settings of 50% power and 50 to 80% MST-laser power.

      Quality assessment of purified protein samples

      The quality control of purified proteins is the final checkpoint with respect to purity, activity, and integrity. While the H2 oxidation activity of Hyd-1 is easily monitored (see aforementioned), no Hyd activity is associated with any of the purified (dye labeled) maturation proteins or intermediates used in this study (HyaB, HyaA, HyaD, GD complex, HypE, and HypF). Their catalytic competence was therefore confirmed by the maturation assay established in this study. Furthermore, ATR FTIR was routinely applied to assess the integrity of the [Fe](CN)2CO cofactor precursor.

      Data availability

      Strains, mutants, and plasmids that arose from this work will be available on request. All the primary data that are presented in this study can be requested in electronic form by contacting Basem Soboh ( [email protected] ).

      Supporting information

      This article contains supporting information (
      • Senger M.
      • Stripp S.T.
      • Soboh B.
      Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
      ,
      • Ash P.A.
      • Hidalgo R.
      • Vincent K.A.
      Proton transfer in the catalytic cycle of [NiFe] hydrogenases: insight from vibrational spectroscopy.
      ).

      Conflict of interest

      The authors declare that they have no conflicts of interest with the contents of this article.

      Acknowledgments

      Protein mass spectrometry was done at the Centre for Chemical Microscopy (ProVIS) at UFZ Leipzig.

      Author contributions

      B. S. conceptualization; B. S., L. A., and S. T. S. methodology; B. S. and S. T. S. software; B. S. and S. T. S. validation; B. S., L. A., and S. T. S. formal analysis; B. S. and S. T. S. investigation; B. S. and S. T. S. resources; B. S. and S. T. S. data curation; B. S. writing–original draft; B. S. and S. T. S. writing–review & editing; B. S. and S. T. S. visualization; B. S. supervision; B. S. project administration; B. S. and S. T. S. funding acquisition.

      Funding and additional information

      This work was supported by German Research Foundation via the SPP 1927 priority program “Iron-Sulfur for Life,” grant nos.: SO1325/5-2 (to B. S.) and STR1554/5-1 (to S. T. S.).

      Supporting information

      References

        • Vignais P.M.
        • Billoud B.
        Occurrence, classification, and biological function of hydrogenases: an overview.
        Chem. Rev. 2007; 107: 4206-4272
        • Thauer R.K.
        • Kaster A.K.
        • Goenrich M.
        • Schick M.
        • Hiromoto T.
        • Shima S.
        Hydrogenases from methanogenic archaea, nickel, a novel cofactor, and H-2 storage.
        Annu. Rev. Biochem. 2010; 79: 507-536
        • Lubitz W.
        • Ogata H.
        • Rudiger O.
        • Reijerse E.
        Hydrogenases.
        Chem. Rev. 2014; 114: 4081-4148
        • Shafaat H.S.
        • Rudiger O.
        • Ogata H.
        • Lubitz W.
        [NiFe] hydrogenases: a common active site for hydrogen metabolism under diverse conditions.
        Biochim. Biophys. Acta. 2013; 1827: 986-1002
        • Sargent F.
        The model [NiFe]-Hydrogenases of Escherichia coli.
        Adv. Microb. Physiol. 2016; 68: 433-507
        • Forzi L.
        • Sawers R.G.
        Maturation of [NiFe]-hydrogenases in Escherichia coli.
        Biometals. 2007; 20: 565-578
        • Lukey M.J.
        • Roessler M.M.
        • Parkin A.
        • Evans R.M.
        • Davies R.A.
        • Lenz O.
        • et al.
        Oxygen-tolerant [NiFe]-hydrogenases: the individual and collective importance of supernumerary cysteines at the proximal Fe-S cluster.
        J. Am. Chem. Soc. 2011; 133: 16881-16892
        • Sawers R.G.
        • Ballantine S.P.
        • Boxer D.H.
        Differential expression of hydrogenase isoenzymes in Escherichia coli K-12: evidence for a third isoenzyme.
        J. Bacteriol. 1985; 164: 1324-1331
        • Laurinavichene T.V.
        • Tsygankov A.A.
        H2 consumption by Escherichia coli coupled via hydrogenase 1 or hydrogenase 2 to different terminal electron acceptors.
        FEMS Microbiol. Lett. 2001; 202: 121-124
        • Ballantine S.P.
        • Boxer D.H.
        Nickel-containing hydrogenase isoenzymes from anaerobically grown Escherichia coli K-12.
        J. Bacteriol. 1985; 163: 454-459
        • Böck A.
        • King P.W.
        • Blokesch M.
        • Posewitz M.C.
        Maturation of hydrogenases.
        Adv. Microb. Physiol. 2006; 51: 1-71
        • Lacasse M.J.
        • Zamble D.B.
        [NiFe]-Hydrogenase maturation.
        Biochemistry. 2016; 55: 1689-1701
        • Soboh B.
        • Stripp S.T.
        • Bielak C.
        • Lindenstrauss U.
        • Braussemann M.
        • Javaid M.
        • et al.
        The [NiFe]-hydrogenase accessory chaperones HypC and HybG of Escherichia coli are iron- and carbon dioxide-binding proteins.
        FEBS Lett. 2013; 587: 2512-2516
        • Stripp S.T.
        • Oltmanns J.
        • Muller C.S.
        • Ehrenberg D.
        • Schlesinger R.
        • Heberle J.
        • et al.
        Electron inventory of the iron-sulfur scaffold complex HypCD essential in [NiFe]-hydrogenase cofactor assembly.
        Biochem. J. 2021; 478: 3281-3295
        • Nutschan K.
        • Golbik R.P.
        • Sawers R.G.
        The iron-sulfur-containing HypC-HypD scaffold complex of the [NiFe]-hydrogenase maturation machinery is an ATPase.
        Febs Open Bio. 2019; 9: 2072-2079
        • Paschos A.
        • Bauer A.
        • Zimmermann A.
        • Zehelein E.
        • Böck A.
        HypF, a carbamoyl phosphate-converting enzyme involved in [NiFe] hydrogenase maturation.
        J. Biol. Chem. 2002; 277: 49945-49951
        • Blokesch M.
        • Albracht S.P.
        • Matzanke B.F.
        • Drapal N.M.
        • Jacobi A.
        • Böck A.
        The complex between hydrogenase-maturation proteins HypC and HypD is an intermediate in the supply of cyanide to the active site iron of [NiFe]-hydrogenases.
        J. Mol. Biol. 2004; 344: 155-167
        • Stripp S.T.
        • Lindenstrauss U.
        • Sawers R.G.
        • Soboh B.
        Identification of an isothiocyanate on the HypEF complex suggests a route for efficient Cyanyl-group channeling during [NiFe]-hydrogenase cofactor generation.
        PLoS One. 2015; 10e0133118
        • Stripp S.T.
        • Lindenstrauss U.
        • Granich C.
        • Sawers R.G.
        • Soboh B.
        The influence of oxygen on [NiFe]-hydrogenase cofactor biosynthesis and how ligation of carbon monoxide precedes cyanation.
        PLoS One. 2014; 9e107488
        • Senger M.
        • Stripp S.T.
        • Soboh B.
        Proteolytic cleavage orchestrates cofactor insertion and protein assembly in [NiFe]-hydrogenase biosynthesis.
        J. Biol. Chem. 2017; 292: 11670-11681
        • Pinske C.
        • Kruger S.
        • Soboh B.
        • Ihling C.
        • Kuhns M.
        • Braussemann M.
        • et al.
        Efficient electron transfer from hydrogen to benzyl viologen by the [NiFe]-hydrogenases of Escherichia coli is dependent on the coexpression of the iron-sulfur cluster-containing small subunit.
        Arch. Microbiol. 2011; 193: 893-903
        • Soboh B.
        • Lindenstrauss U.
        • Granich C.
        • Javed M.
        • Herzberg M.
        • Thomas C.
        • et al.
        [NiFe]-hydrogenase maturation in vitro: analysis of the roles of the HybG and HypD accessory proteins1.
        Biochem. J. 2014; 464: 169-177
        • Blokesch M.
        • Magalon A.
        • Böck A.
        Interplay between the specific chaperone-like proteins HybG and HypC in maturation of hydrogenases 1, 2, and 3 from Escherichia coli.
        J. Bacteriol. 2001; 183: 2817-2822
        • Blokesch M.
        • Bock A.
        Properties of the [NiFe]-hydrogenase maturation protein HypD.
        Febs Lett. 2006; 580: 4065-4068
        • Rossmann R.
        • Maier T.
        • Lottspeich F.
        • Böck A.
        Characterisation of a protease from Escherichia coli involved in hydrogenase maturation.
        Eur. J. Biochem. 1995; 227: 545-550
        • Theodoratou E.
        • Huber R.
        • Böck A.
        [NiFe]-Hydrogenase maturation endopeptidase: structure and function.
        Biochem. Soc. Trans. 2005; 33: 108-111
        • Lukey M.J.
        • Parkin A.
        • Roessler M.M.
        • Murphy B.J.
        • Harmer J.
        • Palmer T.
        • et al.
        How Escherichia coli is equipped to oxidize hydrogen under different redox conditions.
        J. Biol. Chem. 2010; 285: 3928-3938
        • Stripp S.T.
        In Situ infrared spectroscopy for the analysis of gas-processing metalloenzymes.
        ACS Catal. 2021; 11: 7845-7862
        • Hidalgo R.
        • Ash P.A.
        • Healy A.J.
        • Vincent K.A.
        Infrared spectroscopy during electrocatalytic turnover reveals the Ni-L active site state during H2 oxidation by a NiFe hydrogenase.
        Angew. Chem. Int. Ed. Engl. 2015; 54: 7110-7113
        • Murphy B.J.
        • Hidalgo R.
        • Roessler M.M.
        • Evans R.M.
        • Ash P.A.
        • Myers W.K.
        • et al.
        Discovery of dark pH-dependent H(+) migration in a [NiFe]-hydrogenase and its mechanistic relevance: mobilizing the hydrido ligand of the Ni-C intermediate.
        J. Am. Chem. Soc. 2015; 137: 8484-8489
        • Ash P.A.
        • Hidalgo R.
        • Vincent K.A.
        Proton transfer in the catalytic cycle of [NiFe] hydrogenases: insight from vibrational spectroscopy.
        ACS Catal. 2017; 7: 2471-2485
        • Watanabe S.
        • Kawashima T.
        • Nishitani Y.
        • Kanai T.
        • Wada T.
        • Inaba K.
        • et al.
        Structural basis of a Ni acquisition cycle for [NiFe] hydrogenase by Ni-metallochaperone HypA and its enhancer.
        Proc. Natl. Acad. Sci. U.S.A. 2015; 112: 7701-7706
        • Winter G.
        • Buhrke T.
        • Lenz O.
        • Jones A.K.
        • Forgber M.
        • Friedrich B.
        A model system for [NiFe] hydrogenase maturation studies: purification of an active site-containing hydrogenase large subunit without small.
        Febs Lett. 2005; 579: 4292-4296
        • Loscher S.
        • Zebger I.
        • Andersen L.K.
        • Hildebrandt P.
        • Meyer-Klaucke W.
        • Haumann M.
        The structure of the Ni-Fe site in the isolated HoxC subunit of the hydrogen-sensing hydrogenase from Ralstonia eutropha.
        Febs Lett. 2005; 579: 4287-4291
        • Theodoratou E.
        • Paschos A.
        • Magalon A.
        • Fritsche E.
        • Huber R.
        • Böck A.
        Nickel serves as a substrate recognition motif for the endopeptidase involved in hydrogenase maturation.
        Eur. J. Biochem. 2000; 267: 1995-1999
        • Blokesch M.
        • Bock A.
        Maturation of [NiFe]-hydrogenases in Escherichia coli: the HypC cycle.
        J. Mol. Biol. 2002; 324: 287-296
        • Stripp S.T.
        • Soboh B.
        • Lindenstrauss U.
        • Braussemann M.
        • Herzberg M.
        • Nies D.H.
        • et al.
        HypD is the scaffold protein for Fe-(CN)2CO cofactor assembly in [NiFe]-hydrogenase maturation.
        Biochemistry. 2013; 52: 3289-3296
        • Haase A.
        • Sawers R.G.
        Exchange of a single amino acid residue in the HybG chaperone allows maturation of all H2-activating [NiFe]-Hydrogenases in Escherichia coli.
        Front Microbiol. 2022; 13872581
        • Thomas C.
        • Muhr E.
        • Sawers R.G.
        Coordination of synthesis and assembly of a modular membrane-associated [NiFe]-hydrogenase is determined by cleavage of the C-terminal peptide.
        J. Bacteriol. 2015; 197: 2989-2998
        • Kwon S.
        • Watanabe S.
        • Nishitani Y.
        • Kawashima T.
        • Kanai T.
        • Atomi H.
        • et al.
        Crystal structures of a [NiFe] hydrogenase large subunit HyhL in an immature state in complex with a Ni chaperone HypA.
        Proc. Natl. Acad. Sci. U.S.A. 2018; 115: 7045-7050
        • Pinske C.
        • Thomas C.
        • Nutschan K.
        • Sawers R.G.
        Delimiting the function of the C-terminal extension of the Escherichia coli [NiFe]-hydrogenase 2 large subunit precursor.
        Front Microbiol. 2019; 10: 2223
        • Studier F.W.
        • Moffatt B.A.
        Use of bacteriophage T7 RNA polymerase to direct selective high-level expression of cloned genes.
        J. Mol. Biol. 1986; 189: 113-130
        • Redwood M.D.
        • Mikheenko I.P.
        • Sargent F.
        • Macaskie L.E.
        Dissecting the roles of Escherichia coli hydrogenases in biohydrogen production.
        FEMS Microbiol. Lett. 2008; 278: 48-55
        • Senger M.
        • Mebs S.
        • Duan J.
        • Wittkamp F.
        • Apfel U.P.
        • Heberle J.
        • et al.
        Stepwise isotope editing of [FeFe]-hydrogenases exposes cofactor dynamics.
        Proc. Natl. Acad. Sci. U.S.A. 2016; 113: 8454-8459
        • Seidel K.
        • Kuhnert J.
        • Adrian L.
        The complexome of Dehalococcoides mccartyi reveals its organohalide respiration-complex is modular.
        Front Microbiol. 2018; 9: 1130
        • Lowry O.H.
        • Rosebrough N.J.
        • Farr A.L.
        • Randall R.J.
        Protein measurement with the Folin phenol reagent.
        J. Biol. Chem. 1951; 193: 265-275
        • Budhraja R.
        • Ding C.
        • Walter P.
        • Wagner S.
        • Reemtsma T.
        • Gary Sawers R.
        • et al.
        The impact of species, respiration type, growth phase and genetic inventory on absolute metal content of intact bacterial cells.
        Metallomics. 2019; 11: 925-935
        • Senger M.
        • Laun K.
        • Soboh B.
        • Stripp S.T.
        Infrared Characterization of the bidirectional oxygen-sensitive [NiFe]-hydrogenase from E-coli.
        Catalysts. 2018; 8
        • Menon N.K.
        • Chatelus C.Y.
        • Dervartanian M.
        • Wendt J.C.
        • Shanmugam K.T.
        • Peck Jr., H.D.
        • et al.
        Cloning, sequencing, and mutational analysis of the Hyb operon encoding Escherichia coli hydrogenase 2.
        J. Bacteriol. 1994; 176: 4416-4423
        • Soboh B.
        • Stripp S.T.
        • Muhr E.
        • Granich C.
        • Braussemann M.
        • Herzberg M.
        • et al.
        [NiFe]-hydrogenase maturation: Isolation of a HypC-HypD complex carrying diatomic CO and CN- ligands.
        FEBS Lett. 2012; 586: 3882-3887
        • Shomura Y.
        • Higuchi Y.
        Structural basis for the reaction mechanism of S-carbamoylation of HypE by HypF in the maturation of [NiFe]-hydrogenases.
        J. Biol. Chem. 2012; 287: 28409-28419
        • Casadaban M.J.
        Transposition and fusion of lac genes to selected promoters in Escherichia-coli using bacteriophage-lambda and bacteriophage-Mu.
        J. Mol. Biol. 1976; 104: 541-555
        • Soboh B.
        • Kruger S.
        • Kuhns M.
        • Pinske C.
        • Lehmann A.
        • Sawers R.G.
        Development of a cell-free system reveals an oxygen-labile step in the maturation of [NiFe]-hydrogenase 2 of Escherichia coli.
        FEBS Lett. 2010; 584: 4109-4114