Skip to main content
NCBI home page
Acta Crystallographica Section F: Structural Biology Communications logoLink to Acta Crystallographica Section F: Structural Biology Communications
. 2022 Jan 27;78(Pt 2):66–74. doi: 10.1107/S2053230X22000188

Structural and spectroscopic characterization of CO inhibition of [NiFe]-hydrogenase from Citrobacter sp. S-77

Takahiro Imanishi a,‡, Koji Nishikawa a,‡, Midori Taketa a,‡, Katsuhiro Higuchi a, Hulin Tai b,c, Shun Hirota b, Hironobu Hojo d, Toru Kawakami d, Kiriko Hataguchi a, Kayoko Matsumoto a, Hideaki Ogata b,*, Yoshiki Higuchi a,*
PMCID: PMC8805213  PMID: 35102895

The 1.77 Å resolution crystal structure of the CO-bound state of [NiFe]-hydrogenase from Citrobacter sp. S-77 revealed that the exogenous CO ligand binds to the nickel ion in a bent conformation. The CO-bound state was identified as an EPR-silent Ni-SCO state by EPR and FT-IR spectroscopy.

Keywords: [NiFe]-hydrogenases, Citrobacter sp. S-77, inhibitors, carbon monoxide, crystal structure, spectroscopy

Abstract

Hydrogenases catalyze the reversible oxidation of H2. Carbon monoxide (CO) is known to be a competitive inhibitor of O2-sensitive [NiFe]-hydrogenases. Although the activities of some O2-tolerant [NiFe]-hydrogenases are unaffected by CO, the partially O2-tolerant [NiFe]-hydrogenase from Citrobacter sp. S-77 (S77-HYB) is inhibited by CO. In this work, the CO-bound state of S77-HYB was characterized by activity assays, spectroscopic techniques and X-ray crystallography. Electron paramagnetic resonance spectroscopy showed a diamagnetic Ni2+ state, and Fourier-transform infrared spectroscopy revealed the stretching vibration of the exogenous CO ligand. The crystal structure determined at 1.77 Å resolution revealed that CO binds weakly to the nickel ion in the Ni–Fe active site of S77-HYB. These results suggest a positive correlation between O2 and CO tolerance in [NiFe]-hydrogenases.

1. Introduction

Hydrogenases are metalloenzymes that catalyze reversible hydrogen (H2) oxidation and proton reduction (Lubitz et al., 2014 ▸; Ogata et al., 2016 ▸; Nishikawa et al., 2020 ▸; Tai et al., 2018 ▸; Pandelia, Ogata & Lubitz, 2010 ▸; Shafaat et al., 2013 ▸). These enzymes are of great interest for various environmentally friendly applications such as biofuel cells and (bio)­hydrogen production (Lauterbach & Lenz, 2019 ▸; Mazurenko et al., 2017 ▸; Khetkorn et al., 2017 ▸). Numerous chemically synthesized complexes that mimic the function of hydrogenases have also been developed (Ogo et al., 2017 ▸, 2020 ▸; Schilter et al., 2016 ▸; Ogo, 2017 ▸; Dutta et al., 2015 ▸; Caserta et al., 2015 ▸). Hydrogenases can be divided into three classes according to the metal components of the active site: [NiFe]-hydrogenases, [FeFe]-hydrogenases and [Fe]-hydrogenases (Higuchi et al., 1997 ▸; Volbeda et al., 1995 ▸; Peters et al., 1998 ▸; Shima et al., 2008 ▸; Yagi et al., 2014 ▸). The catalytic unit of [NiFe]-hydrogenases consists of two subunits: a large subunit containing the Ni–Fe active site and a small subunit containing the iron–sulfur clusters for the electron-transfer relay (Fig. 1 ▸ a). The Ni–Fe active site comprises two metal ions: nickel and iron. The latter coordinates the nonprotein diatomic ligands, namely one CO molecule and two CN− ions. The active site is co­ordinated by the thiolate side chains of four cysteine residues, of which two bridge the two metal ions and two are ligated to the nickel ion in a terminal fashion. A third bridging ligand is found between the metal ions in the oxidized state but not in the reduced state (Fig. 1 ▸ b). The valence of the nickel ion changes between Ni3+, Ni2+ and Ni+, while the iron ion remains in the redox-inactive low-spin Fe2+ form throughout the catalytic cycle (Fig. 1 ▸ c; Nishikawa et al., 2020 ▸; Ogata et al., 2016 ▸).

Figure 1.

Figure 1

Open in a new tab

(a) 3D structure of [NiFe]-hydrogenase from Citrobacter sp. S-77 (PDB entry 7vxq). The large subunit and the small subunit are shown in light green/dark green and cyan/purple, respectively. (b) Ball-and-stick representations of the Ni–Fe active site in the oxidized state (PDB entry 5xvc, left) and the reduced state (PDB entry 5xvb, right). (c) Schematic representation of the intermediate states of the Ni–Fe active site.

[NiFe]-hydrogenases possess similar active-site coordination structures but can be classified into four groups based on functional aspects, i.e. H2-uptake (groups 1 and 2), bidirectional (group 3) and H2-evolving (group 4). These [NiFe]-hydrogenase groups can be further divided into 22 subgroups according to phylogenetic analysis (Greening et al., 2016 ▸). For example, group 1 membrane-bound H2-uptake [NiFe]-hydrogenases belong to groups 1a–1h. The catalytic activity of the ‘standard’ O2-sensitive [NiFe]-hydrogenases belonging to group 1b drastically decreases upon exposure to oxygen to afford inactive forms. Although the O2-tolerant [NiFe]-hydrogenases belonging to group 1d can maintain their activity under ambient O2 concentrations, extra electrons must be supplied to reduce O2 from the distorted proximal [4Fe–3S] cluster in the super-oxidized state (Shomura et al., 2011 ▸; Volbeda et al., 2012 ▸; Fritsch et al., 2011 ▸). Group 1c (Hyb-type) [NiFe]-hydrogenases are generally considered to be sensitive to O2, with the exception of the recently characterized Citrobacter sp. S-77 [NiFe]-hydrogenase (S77-HYB), which is stable under O2 exposure (Noor et al., 2018 ▸). The S77-HYB enzyme recovers its activity in H2-containing solution in the presence of 1% O2 saturation, indicating that it is partially tolerant to O2. The crystal structure of S77-HYB in the oxidized state suggested that the distorted proximal [4Fe–4S] cluster is probably in the super-oxidized state, as observed for O2-tolerant enzymes. This enables the proximal cluster of S77-HYB to supply extra electrons to reduce the attacking O2 at the Ni–Fe active site for quick reactivation.

The catalytic activity is decreased by the binding of inhibitors at the active site. As mentioned above, standard [NiFe]-hydrogenases are inhibited by O2 to generate inactive states. CO is also a well known competitive inhibitor of [NiFe]-hydrogenases (Purec et al., 1962 ▸). The standard O2-sensitive [NiFe]-hydrogenases are inhibited by CO, indicating the formation of CO-bound states, but their catalytic activity can be recovered in the presence of H2. The CO-adduct state has not been observed for some of the O2-tolerant [NiFe]-hydrogenases; however, spectroscopic results have demonstrated that CO can bind to the Ni–Fe active site of the O2-tolerant membrane-bound [NiFe]-hydrogenase from Aquifex aeolicus (Pandelia, Infossi et al., 2010 ▸). In contrast, H2 oxidation by the O2-tolerant membrane-bound [NiFe]-hydrogenase from Ralstonia eutropha was reported to be unaffected by CO when the enzyme was adsorbed to a graphite electrode (Vincent et al., 2005 ▸). Thus, the relationship between O2 and CO tolerance remains ambiguous. In this study, we aimed to characterize the partially O2-tolerant [NiFe]-hydrogenase S77-HYB in the presence of CO using spectroscopy and X-ray crystallography. The results demonstrate that S77-HYB interacts weakly with CO but can maintain its proton-reduction activity in a CO atmosphere.

2. Materials and methods

2.1. Macromolecule production

Citrobacter sp. S-77 cells were cultivated according to a previously described protocol (Eguchi et al., 2012 ▸). The isolation and purification of S77-HYB were performed as described in a previous report (Noor et al., 2016 ▸). A solution of the enzyme was concentrated to approximately 15 mg ml−1 for subsequent use. The amino-acid composition of S77-HYB was determined using a LaChrom amino-acid analyzer (Hitachi, Tokyo) after hydrolysis with 6 M HCl at 180°C for 25 min in an evacuated sealed tube. The content of the powdered peptides generated by hydrolysis of S77-HYB was estimated by amino-acid analysis. The absorbance coefficient of S77-HYB at 400 nm (ɛ400) was determined to be 13 mM −1 cm−1.

2.2. EPR and FT-IR measurements

S77-HYB was oxidized using K3[Fe(CN)6] and residual reagent was removed using an Amicon Ultracel-10K centrifugal filter. To obtain an H2-reduced sample in an N2 atmosphere, the sample solution was degassed on a vacuum line and purged with 1 bar H2, and subsequently purged with 1 bar N2. To obtain a CO-treated sample, the sample solution was further degassed on the vacuum line and purged with 1 bar CO. EPR spectra were measured at 77 K with an EPR spectrometer (JESFA100N; Jeol, Tokyo, Japan) and averaged over five scans.

For FT-IR measurements, the CO-treated S77-HYB sample (taken from the EPR sample tube) was transferred anaerobically to an infrared cell with CaF2 windows in a glove box prior to freezing in liquid N2. FT-IR spectra of the CO-treated S77-HYB sample were measured before, during and after light irradiation at 123 K with an FT-IR spectrometer (FT-IR 6100V, JASCO, Tokyo, Japan) equipped with a mercury cadmium telluride detector. A cryostat system (CoolSpeK IR USP-203IR-A, Unisoku, Hirakata, Japan) was used to control the cell temperature. Spectral data were collected at 2 cm−1 resolution and averaged over 1024 scans. The corresponding buffer spectrum was collected as a reference spectrum and was subtracted from the sample spectra. Light irradiation was performed at 514.5 nm with an Ar+ laser (Model 2017, Spectra-Physics, Santa Clara, USA). The laser power for the light irradiation was adjusted to 1.15 W cm−2 at the point of sample irradiation.

2.3. Assay for CO inhibition of H2-evolution activity

CO inhibition during H2 evolution by S77-HYB was assayed by gas chromatography. Reaction mixtures consisting of 2.5 mM methyl viologen (MV), 10 mM sodium dithionite and 50 mM MOPS–NaOH pH 7.0 were incubated at 37°C for 30 min under various gaseous mixtures of CO and argon with CO partial pressures of 0, 1, 2, 3 and 5 kPa. The reactions were initiated by the addition of S77-HYB, and the H2 concentrations in the gas phase were then measured every 3 min using an Agilent 490 Micro GC system. One unit of hydrogenase activity was defined as the amount of S77-HYB that catalyzed the evolution of 1 µmol of H2 per minute.

2.4. Crystallization and CO treatment

The crystallization of S77-HYB was achieved using the sitting-drop vapor-diffusion method as described previously (Noor et al., 2016 ▸). To prepare crystals of the CO-bound form, the following operations were conducted in a glove box filled with N2. Firstly, air-oxidized crystals were soaked in a buffer consisting of 100 mM Tris–HCl pH 8.5, 20% PEG 10 000, 200 mM NaCl, 20% glycerol, 1 mM benzyl viologen on a crystallization plate, which was then placed inside a bag composed of gas-barrier film. The gas phase in the bag was replaced with H2 and the plate was then incubated overnight at 283 K. Subsequently, the gas phase was replaced with CO and the plate was incubated for 120 h at 283 K. Finally, the crystals were mounted on cryoloops and cooled in liquid N2 prior to X-ray diffraction measurements.

2.5. X-ray diffraction data collection and structure refinement

The X-ray diffraction data set was collected at 100 K under a nitrogen-gas stream on the BL44XU beamline at SPring-8, Hyogo, Japan. The diffraction data were integrated, scaled and merged using the XDS program package (Kabsch, 2010 ▸). The crystal structure was solved by molecular replacement using the Phaser software (McCoy et al., 2007 ▸) with the atomic coordinates of the reduced S77-HYB structure (PDB entry 5xvb; Noor et al., 2018 ▸) as the initial search model. Model building was performed using Coot (Emsley et al., 2010 ▸). Structure refinement in the early stage was conducted using REFMAC5 (Murshudov et al., 2011 ▸) and phenix.refine (Liebschner et al., 2019 ▸) was used for subsequent refinement. The data-collection and refinement statistics are listed in Table 1 ▸.

Table 1. Data-collection and refinement statistics.

Values in parentheses are for the outer shell.

Data collection
 Beamline BL44XU, SPring-8
 Wavelength (Å) 0.90000
 Space group P21
  a, b, c (Å) 66.39, 122.32, 99.91
 α, β, γ (°) 90.00, 103.12, 90.00
 Resolution range (Å) 34.49–1.77 (1.83–1.77)
 Total No. of reflections 515095 (51632)
 No. of unique reflections 148549 (14786)
 Multiplicity 3.5 (3.5)
 Completeness (%) 98.4 (97.9)
 Mean I/σ(I) 12.6 (2.7)
 CC1/2 (%) 99.7 (87.6)
  R meas (%) 6.7 (63.8)
Refinement
 Resolution range (Å) 34.49–1.77
 No. of reflections (working set) 140959
 No. of reflections (test set) 7420
  R work (95% of data) 0.161
  R free (5% of data) 0.193
 No. of atoms
  Protein 12687
  Ligand 4
  Solvent 624
  B factors (Å2)
  Protein 34.7
  Ligand 26.9
  Solvent 35.8
 Ramachandran plot
  Favored (%) 96.2
  Allowed (%) 3.7
  Outliers (%) 0.1
 R.m.s.d., bond lengths (Å) 0.038
 R.m.s.d., angles (°) 2.25
Open in a new tab

3. Results and discussion

3.1. CO inhibition of hydrogenase activity

In the H2-evolution assay, hydrogenases under reducing conditions in the presence of an electron donor produce hydrogen gas, which can be measured by gas chromatography. In this work, the influence of CO on the evolution of H2 by the partially O2-tolerant S77-HYB was examined. The CO-inhibition curve revealed that S77-HYB was inhibited by CO with an inhibition constant of 7204 Pa (Fig. 2 ▸). The inhibition constant (K i) was calculated using the equation

3.1.

where V 0 is the activity in the absence of CO, V i is the residual activity in the presence of CO and [I] is the concentration of the inhibitor.

Figure 2.

Figure 2

Open in a new tab

CO inhibition of H2 evolution by S77-HYB. The H2-evolution activity of S77-HYB was measured by gas chromatography in the presence of reduced methyl viologen as an electron donor. V i is the residual activity in the presence of CO and V 0 is the activity in the absence of CO. The fitting equation was V 0/V i = 0.1388[pCO] + 1 (R 2 = 0.9894). The inhibition constant was 7204 Pa, which was calculated from the inverse of the slope.

The specific activity of H2 evolution for the O2-sensitive [NiFe]-hydrogenase from Desulfovibrio vulgaris Miyazaki F (DvMF-Hyd) in the presence of MV was reported to be in the range 300–861 U mg−1 in potassium phosphate buffer pH 7.0 (Yagi et al., 2014 ▸), whereas S77-HYB exhibited a lower specific activity of approximately 30 U mg−1 in MOPS–NaOH pH 7.0. In early research in the 1970s, the effects of CO on H2 evolution were examined for several [NiFe]-hydrogenases using the gasometric method. The [NiFe]-hydrogenase from D. desulfuricans (Dd-Hyd, group 1a) was inhibited by CO in the presence of the physiological electron donor cytochrome c 3 and the activity was not recovered upon removing CO (Yagi et al., 1968 ▸). In contrast, when the synthetic electron donor MV was used, 55% of the H2-evolution activity was recovered upon replacing CO with N2 (Yagi et al., 1968 ▸). As another example, H2 evolution by DvMF-Hyd was strongly inhibited by CO, with an inhibition constant (K i) of 1187 Pa. However, the H2 evolution was fully recovered when CO was removed by flushing with N2 (Yagi et al., 1976 ▸). Other O2-sensitive [NiFe]-hydrogenases have also been reported to display inhibition constants in the range 667–4666 Pa (Adams et al., 1980 ▸), compared with the value of 7204 Pa determined for S77-HYB. These results suggest that S77-HYB is approximately 10–30-fold less active with respect to H2 evolution than O2-sensitive [NiFe]-hydrogenases, but is 2.5–10-fold more stable against inhibition by CO.

The O2-tolerant [NiFe]-hydrogenase from Anabaena cylindrica (Ac-Hyd) was also reported to be inhibited by CO in vivo, retaining only 20% of its original activity in an atmosphere composed of 10% CO and 90% argon (Daday et al., 1979 ▸). As discussed above, most O2-sensitive [NiFe]-hydrogenases are easily inhibited by CO; however, some enzymes retain their H2-evolution activity even in the presence of CO. For example, the CO-induced hydrogenase from Rhodo­spirillum rubrum (Rr-Hyd) and the O2-tolerant [NiFe]-hydrogenase from Pyrococcus furiosus (Pf-Hyd1) catalyze the H2-evolution reaction in the presence of CO. Pf-Hyd1 and Rr-Hyd exhibited 3% and 70%, respectively, of their original H2-evolution activities in a 100% CO atmos­phere (Fox et al., 1996 ▸; van Haaster et al., 2008 ▸). Taking these results into account, it was concluded that only a weak correlation exists between O2 and CO tolerance in [NiFe]-hydrogenases.

4. CO binding of S77-HYB hydrogenase

4.1. Spectroscopic characterization of the CO-bound state

4.1.1. EPR spectroscopy of the CO-bound state

The CO-bound states of [NiFe]-hydrogenases have been characterized using various spectroscopic techniques, such as EPR spectroscopy, FT-IR spectroscopy and Raman spectroscopy. EPR spectroscopy can be used to detect paramagnetic states of [NiFe]-hydrogenases, i.e. Ni3+ or Ni+ (S = 1/2). It is well known that there are two distinct CO-bound states denoted Ni-SCO (EPR-silent, Ni2+) and Ni-CO (paramagnetic, Ni+) (Lubitz et al., 2007 ▸).

The S77-HYB samples were oxidized to the Ni-B state (Ni3+) by treatment with potassium ferricyanide (Fig. 3 ▸ a). The obtained oxidized samples were then reduced with H2 (Fig. 3 ▸ b). The reduced samples exhibited a mixture of EPR signals corresponding to Ni-C (Ni3+; g x = 2.189, g y = 2.140 and g z = 2.011) and two unidentified paramagnetic states. The g values (g x = 2.239 and 2.247, g y = 2.089 and g z = 2.052 and 2.058) of the newly detected state were distinct from but similar to those of the Ni-L state (g x = 2.291, g y = 2.116 and g z = 2.047; Noor et al., 2018 ▸), suggesting a similar structure for the newly detected state to that of the Ni-L state. The EPR spectra did not change significantly after the replacement of H2 with N2 (Fig. 3 ▸ c). The EPR signals of the Ni-C state disappeared upon replacement of the atmosphere with CO followed by incubation for 1 h at room temperature (Fig. 3 ▸ d). After incubation for 1 d at 277 K, the EPR signals did not change significantly (Fig. 3 ▸ e). However, after incubation for 5 d at 277 K the EPR signals disappeared, indicating the formation of the EPR-silent CO-bound state (Fig. 3 ▸ f). These results demonstrate that CO binds slowly to S77-HYB. We subsequently performed FT-IR spectroscopy to identify this CO-bound state (see below).

Figure 3.

Figure 3

Open in a new tab

X-band EPR spectra of S77-HYB: (a) in the oxidized state (predominantly Ni-B) after potassium ferricyanide treatment, (b) after reduction with H2, (c) after the replacement of H2 with N2, (d) after the replacement of N2 with CO and incubation for 1 h at room temperature, (e) after incubation in a CO atmosphere for 1 d at 277 K and (f) after incubation in a CO atmosphere for 5 d at 277 K. The EPR spectra were recorded at 77 K.

It has been demonstrated that the O2-sensitive Allochromatium vinosum [NiFe]-hydrogenase in the reduced state exhibits a superhyperfine splitting spectrum when treated with 13CO (13C, I = 1/2), indicating the binding of CO to the Ni+ ion (van der Zwaan et al., 1986 ▸, 1990 ▸). It has also been reported that the reduced Ni-R state of O2-sensitive hydrogenases does not interact with CO (Pandelia, Ogata, Currell et al., 2010 ▸). The paramagnetic Ni-C state would not be expected to interact with CO because the binding of CO to Ni3+ is chemically unfavorable. Instead, the change of the Ni-L state in the dark in the presence of CO (at cryogenic temperature) leads to the paramagnetic Ni-CO state with a formally monovalent Ni+ ion, providing rhombic g-tensors with principal values g x = 2.13, g y = 2.08 and g z = 2.02 (X-band 9.454 GHz, DvMF-Hyd at 40 K; Pandelia, Ogata, Currell et al., 2010 ▸). In the case of S77-HYB, incubation in a CO atmosphere at 277 K for a relatively long time resulted in the formation of the EPR-silent CO-bound state, not the paramagnetic state (Fig. 3 ▸ f).

4.1.2. FT-IR spectroscopy of the CO-bound state

FT-IR spectra of the CO-treated S77-HYB samples, which were EPR-silent, were recorded at 123 K (Fig. 4 ▸ a). In this state, four major bands were observed: an Fe–CO band (1945 cm−1), two Fe–CN− bands (2074 and 2085 cm−1) and a putative CO stretching band (2055 cm−1) from the exogenous CO bound to the nickel ion. To characterize these bands in more detail, we examined the difference spectra recorded at 123 K before and after light irradiation (Fig. 4 ▸ b). When the CO-treated sample was irradiated with laser light at 514.5 nm (1.15 W cm−2 power), the bands observed at 1945, 2074 and 2086 cm−1 in the dark shifted to 1947, 2079 and 2089 cm−1, respectively, indicating that the enzyme was converted to the Ni-SIa state. In addition, the band at 2059 cm−1 disappeared upon light irradiation, suggesting that it could be assigned to the C—O stretching vibration of exogenous CO bound to the nickel ion, which dissociated during irradiation. The spectrum of S77-HYB in the Ni-SCO state was very similar to those of typical O2-sensitive [NiFe]-hydrogenases rather than that of the O2-tolerant Ae-Hyd (Table 2 ▸). The stretching vibration of exogenous CO was previously detected at 2056 cm−1 for DvMF-Hyd in the Ni-SCO state (Pandelia, Ogata, Currell et al., 2010 ▸). The Ni-SCO state is converted to the Ni-SCOred state by electrochemical treatment at low potentials, where no redox transition of the Ni–Fe active site occurs but the proximal FeS cluster is presumably reduced to [4Fe–4S]+ (De Lacey et al., 2002 ▸). Several CO-isotope-sensitive bands have been observed for the Ni-SCO state in resonance Raman (RR) spectra after excitation at 476.5 nm. The RR bands at 375/393 and 430 cm−1 for 12C16O-bound Ni-SCO shifted to 368 and 413 cm−1, respectively, for 13C18O-bound Ni-SCO, where these bands were tentatively assigned to Ni—C stretching and Ni—C—O bending, respectively (Ogata et al., 2002 ▸). The Ni-SCO state is sensitive to light at cryogenic temperatures (Pandelia, Ogata, Currell et al., 2010 ▸). The crystal structure of CO-bound DvMF-Hyd revealed that illumination of the Ni-SCO state induces photolysis of the exogenous CO bound to the Ni atom, resulting in formation of the Ni-SIa state (Ogata et al., 2002 ▸). The Ni–CO bond can be restored in the dark (Pandelia, Ogata, Currell et al., 2010 ▸). Likewise, CO-bound S77-HYB released CO upon irradiation with a green laser, as revealed by FT-IR spectroscopy.

Figure 4.

Figure 4

Open in a new tab

FT-IR spectra of S77-HYB. Left: A, FT-IR spectrum measured in the dark. Bands corresponding to the intrinsic ligands (1945, 2074 and 2086 cm−1) and exogenous CO (2059 cm−1) were observed. B, difference spectrum obtained by subtracting the spectrum of the dark state from that of the irradiated state. C, difference spectrum obtained by subtracting the spectrum of the dark state before irradiation from that of the dark state after irradiation. Right: simulated difference spectrum (blue) corresponding to spectrum B in the left panel. The Ni-SIa and Ni-SCO states are shown in green and red, respectively. The FT-IR spectra were recorded at 123 K. The baseline is shown in cyan.

Table 2. FT-IR bands in the CO-bound Ni-SCO state.
  Temperature (K) νCO (Fe) (cm−1) νCN (Fe) (cm−1) νCO (Ni) (cm−1)
S77-HYB 123 1945 2074, 2086 2059
DvMF† 100 1941 2072, 2086 2061
DvMF† 298 1941 2071, 2084 2056
A. aeolicus ‡ 100 1927 2074, 2086 2072
A. aeolicus ‡ 298 1925 2072, 2082 2066
Open in a new tab
†

Data taken from Pandelia, Ogata, Currell et al. (2010 ▸).

‡

Data taken from Pandelia, Infossi et al. (2010 ▸).

4.2. Structural analysis of CO-bound states

The crystal structure of the CO-bound state of S77-HYB was determined at 1.77 Å resolution. After structure refinement, the R factor and R free converged to values of 0.161 and 0.193, respectively. The refinement statistics are summarized in Table 1 ▸. Two heterodimers were present in the asymmetric unit of space group P21. The difference (r.m.s.d.) between the two heterodimer structures was calculated to be 0.153 Å. Two amino-acid residues (Lys211 in each large subunit) were outliers in the Ramachandran plot.

The overall structure was in accordance with the previously determined structure of S77-HYB (Noor et al., 2018 ▸). The r.m.s.d. of the S77-HYB structures in the H2-reduced (PDB entry 5xvb) and CO-bound states was calculated to be 0.124 Å. There is no obvious structural difference around the CO-binding site at the Ni–Fe active site. The Ni–Fe active site and the surrounding amino-acid residues in the second coordination sphere were well conserved. However, an electron-density omit map of the Ni–Fe active site in the CO-bound state revealed that there was no electron density at the bridging position between the nickel and iron ions (Fig. 5 ▸). A small degree of extra electron density at the nickel ion was also observed in the omit map and was assigned as an exogenous CO molecule, which is in good agreement with the FT-IR results described above. The coordination geometries and the electron-density maps around the exogenous CO at the active site in both heterodimers are very similar, indicating no obvious difference between the heterodimers. The occupancy of the exogenous CO molecule was approximately 0.26, resulting in lower electron density compared with the intrinsic CO ligand on the iron ion. Owing to low occupancy of the CO ligand, it was difficult to observe the electron density of CO in the 2F o − F c map. This explains why CO only weakly inhibits the Hyb-type O2-tolerant [NiFe]-hydrogenases. The geometry of the Ni–Fe active site in the CO-bound state was similar to those reported for CO-bound DvMF-Hyd and the bidirectional F420-reducing [NiFe]-hydrogenase from Methanosarcina barkeri MS (Mb-FRH). In the case of S77-HYB, the distance between the Ni and C atoms was approximately 1.84 Å and the Ni—C—O angle was 158°. The corresponding values reported for the CO-bound state of the O2-sensitive DvMF [NiFe]-hydrogenase were 1.77 Å and 136–161°, respectively (Ogata et al., 2002 ▸). The lack of electron density for the bridging ligand is in good agreement with the spectroscopic analysis. The crystal structure of the Ni-SCO state of Mb-FRH also indicated binding of the exogenous CO to the terminal vacant coordination site at the nickel ion (Ni—C—O angle of 162°), with a CO stretching band at 2048 cm−1 in the corresponding IR spectrum (Ilina et al., 2019 ▸). In all cases, the exogenous CO-binding modes are very similar, with coordination to the nickel ion in a bent conformation. The exogenous CO did not bind to the Fe atom owing to steric hindrance from a conserved arginine residue located in front of the Ni–Fe active site.

Figure 5.

Figure 5

Open in a new tab

CO-bound Ni–Fe active site of S77-HYB shown as a ball-and-stick model with the electron-density polder omit map calculated by Phenix. The exogenous and intrinsic CO ligands were omitted when the omit map was calculated (the electron-density map was contoured at 3.0σ; PDB entry 7vxq). Atoms are shown as spheres: nickel in green, iron in orange, sulfur in yellow, carbon in white, oxygen in red and nitrogen in blue.

The proximal [4Fe–4S] cluster in the oxidized state of S77-HYB has been shown to exist in two distinct forms: the standard cubane shape and a distorted form. In the latter case one of the Fe atoms is dislocated and bound to the side chain of an aspartate residue. In the reduced state, the proximal FeS cluster possesses the standard cubane shape, indicating that the cluster can be reversibly reformed during the oxidation/reduction cycle. In the crystal structure of the CO-bound state of S77-HYB, the proximal FeS cluster has the standard cubane shape (Fig. 6 ▸), similar to the proximal cluster in the reduced state. The active-site structures in the CO-bound state of [NiFe]-hydrogenases are very similar, but the CO affinity might differ among [NiFe]-hydrogenases based on the redox potentials of the metal centers, including the active site and the FeS clusters.

Figure 6.

Figure 6

Open in a new tab

Structure of the proximal [4Fe–4S] cluster of S77-HYB with the composite omit map (contoured at 1.5σ).

5. Summary

CO is a competitive inhibitor of most [NiFe]-hydrogenases. The catalytic activity of O2-sensitive [NiFe]-hydrogenases is easily diminished in the presence of a small amount of CO. In contrast, some O2-tolerant [NiFe]-hydrogenases are not affected by CO, such as the membrane-bound [NiFe]-hydrogenases from R. eutropha and Escherichia coli. The O2-tolerant [NiFe]-hydrogenase from Citrobacter sp. S-77 is clearly inhibited by CO. We determined the crystal structure of the CO-bound state of S77-HYB, which revealed that the exogenous CO ligand binds to the nickel ion in a bent conformation. The CO-bound state of S77-HYB was identified as an EPR-silent Ni-SCO state by EPR and FT-IR spectroscopy, in which the CO stretching IR band of the exogenous CO ligand was observed at 2059 cm−1. The crystal structure demonstrated that the proximal [4Fe–4S] cluster possesses the standard cubane shape, indicating that the proximal FeS cluster is reduced. These spectroscopic and structural analyses of the mechanism of CO inhibition of the Hyb-type O2-tolerant [NiFe]-hydrogenase S77-HYB suggest that the manner of CO binding is similar to that of standard O2-sensitive [NiFe]-hydrogenases but the stability toward CO is 2.5–10-fold higher. These results provide useful insights for understanding the inhibition mechanism of O2-tolerant [NiFe]-hydrogenases.

Supplementary Material

PDB reference: carbon monoxide complex of [NiFe]-hydrogenase (Hyb-type) from Citrobacter sp. S-77, 7vxq

Acknowledgments

We thank the staff of beamline BL44XU for their help during X-ray data collection. The authors declare no conflicts of interest.

Funding Statement

This work was funded by Ministry of Education, Culture, Sports, Science and Technology grant 18H05516 to Yoshiki Higuchi; Japan Society for the Promotion of Science grants 19H00984, JP20K06511, JP21H02060, and JP20H03215 to Yoshiki Higuchi, Koji Nishikawa, Shun Hirota, and Hideaki Ogata; National Natural Science Foundation of China grant 21977125 to Hulin Tai; Education Department of Jilin Province grant JJKH20210566KJ to Hulin Tai.

References

  1. Adams, M. W. W., Mortenson, L. E. & Chen, J. S. (1980). Biochim. Biophys. Acta, 594, 105–176. [DOI] [PubMed]
  2. Caserta, G., Roy, S., Atta, M., Artero, V. & Fontecave, M. (2015). Curr. Opin. Chem. Biol. 25, 36–47. [DOI] [PubMed]
  3. Daday, A., Lambert, G. R. & Smith, G. D. (1979). Biochem. J. 177, 139–144. [DOI] [PMC free article] [PubMed]
  4. De Lacey, A. L., Stadler, C., Fernandez, V. M., Hatchikian, E. C., Fan, H.-J., Li, S. & Hall, M. B. (2002). J. Biol. Inorg. Chem. 7, 318–326. [DOI] [PubMed]
  5. Dutta, A., Lense, S., Roberts, J. A. S., Helm, M. L. & Shaw, W. J. (2015). Eur. J. Inorg. Chem. 2015, 5218–5225.
  6. Eguchi, S., Yoon, K.-S. & Ogo, S. (2012). J. Biosci. Bioeng. 114, 479–484. [DOI] [PubMed]
  7. Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486–501. [DOI] [PMC free article] [PubMed]
  8. Fox, J. D., Kerby, R. L., Roberts, G. P. & Ludden, P. W. (1996). J. Bacteriol. 178, 1515–1524. [DOI] [PMC free article] [PubMed]
  9. Fritsch, J., Scheerer, P., Frielingsdorf, S., Kroschinsky, S., Friedrich, B., Lenz, O. & Spahn, C. M. T. (2011). Nature, 479, 249–252. [DOI] [PubMed]
  10. Greening, C., Biswas, A., Carere, C. R., Jackson, C. J., Taylor, M. C., Stott, M. B., Cook, G. M. & Morales, S. E. (2016). ISME J. 10, 761–777. [DOI] [PMC free article] [PubMed]
  11. Haaster, D. J. van, Silva, P. J., Hagedoorn, P. L., Jongejan, J. A. & Hagen, W. R. (2008). J. Bacteriol. 190, 1584–1587. [DOI] [PMC free article] [PubMed]
  12. Higuchi, Y., Yagi, T. & Yasuoka, N. (1997). Structure, 5, 1671–1680. [DOI] [PubMed]
  13. Ilina, Y., Lorent, C., Katz, S., Jeoung, J. H., Shima, S., Horch, M., Zebger, I. & Dobbek, H. (2019). Angew. Chem. Int. Ed. 58, 18710–18714. [DOI] [PMC free article] [PubMed]
  14. Kabsch, W. (2010). Acta Cryst. D66, 125–132. [DOI] [PMC free article] [PubMed]
  15. Khetkorn, W., Rastogi, R. P., Incharoensakdi, A., Lindblad, P., Madamwar, D., Pandey, A. & Larroche, C. (2017). Bioresour. Technol. 243, 1194–1206. [DOI] [PubMed]
  16. Lauterbach, L. & Lenz, O. (2019). Curr. Opin. Chem. Biol. 49, 91–96. [DOI] [PubMed]
  17. Liebschner, D., Afonine, P. V., Baker, M. L., Bunkóczi, G., Chen, V. B., Croll, T. I., Hintze, B., Hung, L.-W., Jain, S., McCoy, A. J., Moriarty, N. W., Oeffner, R. D., Poon, B. K., Prisant, M. G., Read, R. J., Richardson, J. S., Richardson, D. C., Sammito, M. D., Sobolev, O. V., Stockwell, D. H., Terwilliger, T. C., Urzhumtsev, A. G., Videau, L. L., Williams, C. J. & Adams, P. D. (2019). Acta Cryst. D75, 861–877.
  18. Lubitz, W., Ogata, H., Rüdiger, O. & Reijerse, E. (2014). Chem. Rev. 114, 4081–4148. [DOI] [PubMed]
  19. Lubitz, W., Reijerse, E. & van Gastel, M. (2007). Chem. Rev. 107, 4331–4365. [DOI] [PubMed]
  20. Mazurenko, I., Wang, X., de Poulpiquet, A. & Lojou, E. (2017). Sustain. Energy Fuels, 1, 1475–1501.
  21. McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C. & Read, R. J. (2007). J. Appl. Cryst. 40, 658–674. [DOI] [PMC free article] [PubMed]
  22. Murshudov, G. N., Skubák, P., Lebedev, A. A., Pannu, N. S., Steiner, R. A., Nicholls, R. A., Winn, M. D., Long, F. & Vagin, A. A. (2011). Acta Cryst. D67, 355–367. [DOI] [PMC free article] [PubMed]
  23. Nishikawa, K., Ogata, H. & Higuchi, Y. (2020). Chem. Lett. 49, 164–173.
  24. Noor, N. D. M., Matsuura, H., Nishikawa, K., Tai, H., Hirota, S., Kim, J., Kang, J., Tateno, M., Yoon, K. S., Ogo, S., Kubota, S., Shomura, Y. & Higuchi, Y. (2018). Chem. Commun. 54, 12385–12388. [DOI] [PubMed]
  25. Noor, N. D. M., Nishikawa, K., Nishihara, H., Yoon, K.-S., Ogo, S. & Higuchi, Y. (2016). Acta Cryst. F72, 53–58. [DOI] [PMC free article] [PubMed]
  26. Ogata, H., Lubitz, W. & Higuchi, Y. (2016). J. Biochem. 160, 251–258. [DOI] [PubMed]
  27. Ogata, H., Mizoguchi, Y., Mizuno, N., Miki, K., Adachi, S., Yasuoka, N., Yagi, T., Yamauchi, O., Hirota, S. & Higuchi, Y. (2002). J. Am. Chem. Soc. 124, 11628–11635. [DOI] [PubMed]
  28. Ogo, S. (2017). Coord. Chem. Rev. 334, 43–53.
  29. Ogo, S., Kishima, T., Yatabe, T., Miyazawa, K., Yamasaki, R., Matsumoto, T., Ando, T., Kikkawa, M., Isegawa, M., Yoon, K. S. & Hayami, S. (2020). Sci. Adv. 6, eaaz8181. [DOI] [PMC free article] [PubMed]
  30. Ogo, S., Mori, Y., Ando, T., Matsumoto, T., Yatabe, T., Yoon, K. S., Hayashi, H. & Asano, M. (2017). Angew. Chem. Int. Ed. 56, 9723–9726. [DOI] [PubMed]
  31. Pandelia, M. E., Infossi, P., Giudici-Orticoni, M. T. & Lubitz, W. (2010). Biochemistry, 49, 8873–8881. [DOI] [PubMed]
  32. Pandelia, M. E., Ogata, H., Currell, L. J., Flores, M. & Lubitz, W. (2010). Biochim. Biophys. Acta, 1797, 304–313. [DOI] [PubMed]
  33. Pandelia, M. E., Ogata, H. & Lubitz, W. (2010). Chem. Eur. J. Chem. Phys. 11, 1127–1140. [DOI] [PubMed]
  34. Peters, J. W., Lanzilotta, W. N., Lemon, B. J. & Seefeldt, L. C. (1998). Science, 282, 1853–1858. [DOI] [PubMed]
  35. Purec, L., Krasna, A. I. & Rittenberg, D. (1962). Biochemistry, 1, 270–275. [DOI] [PubMed]
  36. Schilter, D., Camara, J. M., Huynh, M. T., Hammes-Schiffer, S. & Rauchfuss, T. B. (2016). Chem. Rev. 116, 8693–8749. [DOI] [PMC free article] [PubMed]
  37. Shafaat, H. S., Rüdiger, O., Ogata, H. & Lubitz, W. (2013). Biochim. Biophys. Acta, 1827, 986–1002. [DOI] [PubMed]
  38. Shima, S., Pilak, O., Vogt, S., Schick, M., Stagni, M. S., Meyer-Klaucke, W., Warkentin, E., Thauer, R. K. & Ermler, U. (2008). Science, 321, 572–575. [DOI] [PubMed]
  39. Shomura, Y., Yoon, K. S., Nishihara, H. & Higuchi, Y. (2011). Nature, 479, 253–256. [DOI] [PubMed]
  40. Tai, H., Higuchi, Y. & Hirota, S. (2018). Dalton Trans. 47, 4408–4423. [DOI] [PubMed]
  41. Vincent, K. A., Cracknell, J. A., Lenz, O., Zebger, I., Friedrich, B. & Armstrong, F. A. (2005). Proc. Natl Acad. Sci. USA, 102, 16951–16954. [DOI] [PMC free article] [PubMed]
  42. Volbeda, A., Amara, P., Darnault, C., Mouesca, J. M., Parkin, A., Roessler, M. M., Armstrong, F. A. & Fontecilla-Camps, J. C. (2012). Proc. Natl Acad. Sci. USA, 109, 5305–5310. [DOI] [PMC free article] [PubMed]
  43. Volbeda, A., Charon, M. H., Piras, C., Hatchikian, E. C., Frey, M. & Fontecilla-Camps, J. C. (1995). Nature, 373, 580–587. [DOI] [PubMed]
  44. Yagi, T., Honya, M. & Tamiya, N. (1968). Biochim. Biophys. Acta, 153, 699–705. [DOI] [PubMed]
  45. Yagi, T., Kimura, K., Daidoji, H., Sakai, F., Tamura, S. & Inokuchi, H. (1976). J. Biochem. 79, 661–671. [DOI] [PubMed]
  46. Yagi, T., Ogo, S. & Higuchi, Y. (2014). Int. J. Hydrogen Energy, 39, 18543–18550.
  47. Zwaan, J. W. van der, Albracht, S. P. J., Fontijn, R. D. & Roelofs, Y. B. M. (1986). Biochim. Biophys. Acta, 872, 208–215.
  48. Zwaan, J. W. van der, Coremans, J. M. C. C., Bouwens, E. C. M. & Albracht, S. P. J. (1990). Biochim. Biophys. Acta, 1041, 101–110. [DOI] [PubMed]

Associated Data

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

Supplementary Materials

PDB reference: carbon monoxide complex of [NiFe]-hydrogenase (Hyb-type) from Citrobacter sp. S-77, 7vxq

Articles from Acta Crystallographica. Section F, Structural Biology Communications are provided here courtesy of International Union of Crystallography

RESOURCES