Infrared Spectroscopy Elucidates the Inhibitor Binding Sites
in a Metal-Dependent Formate Dehydrogenase
Konstantin Laun+,[a] Benjamin R. Duffus+,[b] Stefan Wahlefeld,[a, c] Sagie Katz,[a] Dennis Belger,[a]
Peter Hildebrandt,[a] Maria Andrea Mroginski,*[a] Silke Leimkühler,*[b] and Ingo Zebger*[a]
Abstract: Biological carbon dioxide (CO2) reduction is an
important step by which organisms form valuable energy-
richer molecules required for further metabolic processes. The
Mo-dependent formate dehydrogenase (FDH) from Rhodo-
bacter capsulatus catalyzes reversible formate oxidation to
CO2at a bis-molybdopterin guanine dinucleotide (bis-MGD)
cofactor. To elucidate potential substrate binding sites
relevant for the mechanism, we studied herein the interaction
with the inhibitory molecules azide and cyanate, which are
isoelectronic to CO2and charged as formate. We employed
infrared (IR) spectroscopy in combination with density func-
tional theory (DFT) and inhibition kinetics. One distinct
inhibitory molecule was found to bind to either a non-
competitive or a competitive binding site in the secondary
coordination sphere of the active site. Site-directed muta-
genesis of key amino acid residues in the vicinity of the bis-
MGD cofactor revealed changes in both non-competitive and
competitive binding, whereby the inhibitor is in case of the
latter interaction presumably bound between the cofactor
and the adjacent Arg587.
Introduction
In light of the impact of carbon dioxide on the climate, catalysts
capable of converting CO2into fuels are of great interest.[1]
Metal-containing formate dehydrogenases (FDH) perform the
reversible oxidation of formate to yield CO2, one proton, and
two electrons under physiological conditions.[2] Thus, these
enzymes are also candidates for potential biotechnological
applications and have been extensively studied in the field of
(photo)electrochemical generation of carbon-based fuels with
regard to bio-fuel cells and photoelectrochemical tandem
devices as well as for the regeneration of natural and non-
natural redox cofactors.[3] The herein studied FDH from
Rhodobacter capsulatus (RcFDH) consists of a (αβγδ)2dimer of
heterotetramers harboring a bis-molybdopterin guanine dinu-
cleotide (bis-MGD) cofactor, in addition to five [4Fe-4S] clusters,
two [2Fe-2S] clusters, and a flavin mononucleotide (FMN)
prosthetic group.[4] Nicotinamide adenine dinucleotide (NAD+)
is the physiological electron acceptor.[2a,4,5] In its resting state,
the MoVI ion is coordinated by four sulfurs from two MGD
dithiolenes, a sulfido ligand and a cysteine (Cys386) ligand from
the protein backbone.[2a,4] Upon two electron reduction, the
MoIV ion is proposed to be pentacoordinated due to the
displacement of Cys386.[6]
Furthermore, important amino acid residues in the second
coordination sphere, present in all metal-containing FDH
enzymes, include a highly conserved histidine (His387) and an
arginine (Arg587). Both have been proposed to ensure optimal
substrate binding to and proton transfer away from the active
site (Figure 1).[6a,7] The exact catalytic mechanism is currently still
under debate.[6b,7] In previous studies, azide and cyanate (N3
and OCN), both isoelectronic and isostructural molecules to
CO2
[8] but charged as formate, have been considered as
analogues to putative intermediate states of the catalytic
cycle.[9] Moreover, they have been identified as inhibitors of
FDH enzymes.[10] In addition, azide is also used as a protectant
during the purification of such proteins to prevent oxidative
inactivation of the enzyme caused by the replacement of the
essential sulfido ligand by an oxo ligand, as recently shown by
EXAFS spectroscopy.[12] However, the exact inhibitor binding
site remained elusive. As basis for deeper mechanistic studies,
such inhibitors were chosen to elucidate potential substrate
binding sites in the bis-MGD-containing FDHH from Escher-
ichia coli by employing in depth protein film electrochemical
investigations, that suggested an oxidation state-dependent
[a] K. Laun,+Dr. S. Wahlefeld, Dr. S. Katz, Dr. D. Belger, Prof. P. Hildebrandt,
Prof. M. A. Mroginski, Dr. I. Zebger
Institut für Chemie
Max-Volmer-Laboratorium für Biophysikalische Chemie, PC14
Technische Universität Berlin
Strasse des 17. Juni 135, 10623 Berlin (Germany)
E-mail: [email protected]
[b] Dr. B. R. Duffus,+Prof. S. Leimkühler
Institut für Biochemie und Biologie, Molekulare Enzymologie
Universität Potsdam,
Karl-Liebknecht-Strasse 24–25
14476 Potsdam (Germany)
E-mail: [email protected]
[c] Dr. S. Wahlefeld
Institut für Technische Biokatalyse
Technische Universität Hamburg
Denickestr. 15, 21073 Hamburg (Germany)
[+]These authors contributed equally to this manuscript.
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/chem.202201091
© 2022 The Authors. Chemistry - A European Journal published by Wiley-VCH
GmbH. This is an open access article under the terms of the Creative
Commons Attribution License, which permits use, distribution and re-
production in any medium, provided the original work is properly cited.
Chemistry—A European Journal
www.chemeurj.org
Research Article
doi.org/10.1002/chem.202201091
Chem. Eur. J. 2022,28, e202201091 (1 of 7) © 2022 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
Wiley VCH Dienstag, 20.09.2022
2254 / 256353 [S. 74/80] 1
competitive inhibition by various compounds including azide
and cyanate, which act supposedly as π-electron donors to the
MoVI ion in a coordinate bond.[7c] However, no structural
evidence could be provided by this technique.
In the present study, infrared (IR) spectroscopy was
employed in combination with density functional theory (DFT)
and inhibition kinetic studies to elucidate the specific inter-
action of such inhibitory CO2analogues with RcFDH. The
negative charge and hence the large transition dipole moment
of the inhibitory anions causes high extinction coefficients of
the antisymmetric stretching modes, which are exceptionally
well-suited for IR measurements as they are located in a spectral
region free of any protein or water absorptions.[12] Moreover, as
these particular vibrational reporters are highly sensitive to
changes in the electronic structure in their molecular environ-
ment, they can be used as marker bands.[13] By characterizing
the binding of these anions and their selected isotopologues to
the RcFDH and associated variants, potential interaction sites at
the bis-MGD and in its secondary coordination sphere could be
identified. In addition, we have investigated the azide inhibition
of the inactive oxo-ligated bis-MGD that was produced in
absence of its maturation precursor, RcFDHΔFdsC.[14] In order to
elucidate the modality of the bis-MGD dependent binding, the
calculated spectra derived from conceivable DFT models were
compared to the experimentally obtained vibrational bands of
protein-bound azide and cyanate.
Results and Discussion
Following the procedure described in Figure S1, all baseline
corrected and to amide II band normalized IR spectra shown
here, were obtained from RcFDH proteins incubated with 10-
fold excess inhibitor in which contributions of free inhibitors in
solution were subtracted. The spectra of the as isolated protein
sample treated with azide and cyanate (Figure 2 top traces)
exhibit each two major absorption bands. Both bands are red
shifted compared to those of the free inhibitors in water
(Figure S1). The first band, found at 2040 cm1for azide- and at
2160 cm1for cyanate-treated samples, was also observed in
the apoprotein (RcFDHΔbisMGD) (Figure 2, gray traces).
The small frequency shift (Δν�8 cm1) and the detected
band shape suggest a conformation of the azide, which is more
similar to that of free azide in water (Figure S1). The second
distinct band, observed at 2031 cm1for azide, and 2153 cm1
for cyanate treatment, could only be detected in the bis-MGD-
containing enzyme (Figure 2). Thus, these bands could be
assigned to a specific interaction site of an inhibitor molecule
with the bis-MGD cofactor. Moreover, all bands were not
observed in corresponding control experiments, comparing
both azide binding to the diaphorase subcomplex FdsGB that
Figure 1. Active site and ligand environment of the bis-MGD cofactor
derived from the recently published corresponding cryo-EM structure of the
“as isolated” form of the formate dehydrogenase from Rhodobacter
capsulatus (PDB entry 6TGA).[4] Here, the hexacoordinated molybdenum ion
is ligated by a sulfido group, Cys386 and four sulfur atoms belonging to the
dithiolenes from two MGDs. The highlighted conserved amino acids Arg587
and His387 play a crucial role for inhibitor binding in the second
coordination sphere.
Figure 2. Normalized IR spectra of RcFDHWT samples in different redox states
treated with 10 mM azide (left panel) and 10 mM cyanate (right panel) after
subtraction of the spectral contributions from the free inhibitors in solution
and baseline correction. Further details are given in the Figure S1. The
binding of azide to the RcFDHWT gives rise to two major bands at 2031 and
2040 cm1. Cyanate incubation results in the formation of two bands at 2153
and 2160 cm1. The obtained spectra of inhibited “as isolated” samples are
plotted in black. Spectra of the apoenzyme (RcFDHΔbisMGD) incubated with
the respective inhibitors are displayed in grey, exhibiting a broader
absorption at higher frequencies. The incorporation of the bis-MGD cofactor
in the protein is reflected the presence of a second band at slightly lower
wavenumbers (Δν�7–9 cm1). No change in the respective band positions
was observed upon reduction or oxidation of the protein sample with
10 mM sodium formate or thionine (depicted as red and blue traces),
respectively.
Chemistry—A European Journal
Research Article
doi.org/10.1002/chem.202201091
Chem. Eur. J. 2022,28, e202201091 (2 of 7) © 2022 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
Wiley VCH Dienstag, 20.09.2022
2254 / 256353 [S. 75/80] 1
lacks the α-subunit harboring the bis-MGD cofactor, and bovine
serum albumin (BSA). Based on the spectral results of these
control experiments, we can exclude unspecific interactions of
azide with globular proteins in general and in particular with
other subunits of the RcFDH (Figure S2).
To elucidate the specific role of the different binding sites,
the concentrations of both types of bound azide molecules
were estimated from their band intensities (Aint) within a semi-
quantitative evaluation (Figures S3 and S4). They were found to
be in a similar order of magnitude as the protein concentration,
yielding in sum values about 0.5 anions per RcFDH molecule. As
this observation was consistent for all azide-incubated samples,
it rather points to a modified extinction coefficient due to the
interaction with the protein environment[12a] and thus an
underestimation of the amount of bound inhibitor. Previous
studies have reported that azide acts as a protectant in a
stoichiometric fashion.[7a,11b] Therefore, it is appropriate to
assume a stochiometric 1:1 ratio between inhibitor and protein
molecules. The bis-MGD cofactor (Mo) loading of approximately
50% was also considered.
Moreover, this result indicates that in presence of the bis-
MGD cofactor, the azide molecule from the interaction site,
related to the absorption at 2040 cm1, forms the binding site
indicated by the band at 2031 cm1. This finding is supported
by the fact that the overall integrated absorbance Aint of both
bands (Aint(2031+2040 cm1)) in the RcFDHWT sample is equiv-
alent to the total integral Aint of the band at 2040 cm1in the
apoenzyme, RcFDHΔbisMGD. The calculated difference spectrum,
in which spectral contribution of the apoenzyme with respect
to the Mo loading are subtracted, shows exclusively the
remaining bis-MGD dependent band centered at 2031 cm1
(see Figure S4). This observation suggests that the two
observed bands represent two populations of potential binding
sites within the sample (apo- and holoenzyme), to which in
total only one inhibitory molecule can bind.
Complementary DFT calculations were performed to clarify
the modality of the bis-MGD dependent binding. For this
purpose, DFT calculations on structural models comprising a
bis-MGD cofactor (MoVI), the ligated Cys386, and an additionally
incorporated inhibitor anion were conducted (Figure S5). The
corresponding Mo coordination geometry was derived from the
oxidized structure of the related bis-MGD of the nitrate-
inducible formate dehydrogenase (FDHN) from E. coli.[15] Due
to the inherent error of the spectra calculation, the absolute
positions of IR bands do not reflect experimental data
accurately. However, the calculated trends related to changes
of the structure or the electronic environment can be directly
correlated with the corresponding experimental data. The
calculations refer to two possible inhibitor interaction scenarios.
In one case, the Mo ion remained hexacoordinated (i.e., Cys386
is still bound) in a MoVI oxidation state and the respective
inhibitor was kept in an energetically favored position near the
MoVI ion but excluding a covalent MoN bond (see Figure S6
and Table 1 in the Supporting Information). The second
scenario was based on the inhibitor binding according to a
proposed catalytic mechanism where the substrate binds at the
metal center.[7c] Here, the Mo ion resided in a reduced MoIV state
and the thiolate bond to Cys386 was replaced by a covalent
bond to the respective azide or cyanate molecule (MoIV-N/O).
DFT calculations were also performed for the corresponding
interactions with isotopically labeled azide (15N=N=N) and
cyanate (O=13C=15N) for a proper comparison with the
experimental spectra (see Figures 3, S6 and S7). The shifts of the
two adjacently and partially overlapping experimentally derived
IR bands relative to the band of azide in solution are in good
agreement, both in terms of direction and magnitude, with an
interaction site that does not involve direct coordination to the
Mo ion. This is also in line with former investigations on azide
binding in Cu,Zn superoxide dismutase as discussed in more
detail previously[16] and supported by the fact that no significant
band shifts were observed upon oxidation or reduction of
RcFDH samples with an excess of thionine and formate,
respectively (Figure 2, middle and bottom traces).
Enzyme reduction is illustrated in the corresponding UV-
visible absorption spectra by the respective intensity loss of
bands characteristic for FeS clusters and FMN (Figure S8). In
addition, previous XAS and EPR spectroscopic studies have
proven that such treatments change the oxidation state of the
Mo ion to MoVor MoIV.[11] So far, our DFT and IR data suggest
that the bis-MGD-independent band at higher frequencies
represents an inhibitor species solely in the RcFDHΔbisMGD
(apoenzyme), whereas the lower frequency band appears to
refer to a specific interaction site in the RcFDHWT (holoenzyme).
Notably, the observed binding-site-specific bands remain
independent from the cofactor’s oxidation state, indicating that
none of the inhibitors coordinate directly the Mo ion. This has
been also recently proposed within a combined quantum
mechanical and molecular mechanic calculations.[17]
Moreover, previous studies have suggested that the con-
served His387 and Arg587 residues, located in the second
Table 1. Formate oxidation constants and inhibition kinetic data[a] for R. capsulatus FDH.
FDHWT FDHWT FDHH387M FDHH387M FDHR587K FDHR587K
DCPIP NAD+DCPIP NAD+DCPIP NAD+
Kformate [μM] 151�6 29275 �1647 401750�12884
kcat [min1] 1804�44 956�41 812 �13
N3Ki[μM] 47�3 48�3 517�37 476�14 7384�458 4840�226
αKi[μM] 1378�16 1372�17 5656�84 6391�38 – –
OCNKi[μM] 783�48 780�81 7841�276 8079�265 8137�418 5571 �613
αKi[μM] 11755�88 12451�131 27497�340 30369�386 – –
[a] Further information are given in the Supporting Information.
Chemistry—A European Journal
Research Article
doi.org/10.1002/chem.202201091
Chem. Eur. J. 2022,28, e202201091 (3 of 7) © 2022 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
Wiley VCH Dienstag, 20.09.2022
2254 / 256353 [S. 76/80] 1
coordination sphere of bis-MGD, are involved in both substrate
binding and stabilization in the protein scaffold.[7a,9,11a] To
evaluate the role of these key residues for binding of inhibitors,
two active protein variants RcFDHH387M and RcFDHR587K (see
Figures 4 and S9) were prepared and biochemically character-
ized including the respective inhibition kinetics (Table 1).
In catalytic assays for formate oxidation, using NAD+as an
electron acceptor to interact with the FMN or DCPIP (2,6-
dichlorophenolindophenol) as an electron acceptor at the bis-
MGD-site, kcat values of about 1800 min-1 were obtained for
RcFDHWT, while lowered kcat values were observed for the
RcFDHH387M and RcFDHR587K variants.[11a] For the latter variant, a
large KM(formate) value was obtained, displaying a significantly
lower binding affinity. While RcFDHWT and RcFDHH387M revealed a
mixed-type inhibition for azide and cyanate, the amino acid
exchange in the RcFDHR587K variant resulted in a strictly
competitive inhibition, which underlines the importance of
Arg587 for both substrate and competitive inhibitor binding.
Notably, significant differences were observed for the
inhibition constants of azide and cyanate. In comparison to Ki
(azide) of RcFDHWT, the Ki(azide) of the active site variants
RcFDHH387M and RcFDHR587K increased by a factor of ~10 and 120,
respectively (Table 1). By comparison, the Ki(cyanate) of the
RcFDHWT has increased by a factor of ~15 relative to Ki(azide).
This suggests a stronger influence of these amino acids on the
binding of azide compared to cyanate. Furthermore, azide is
also a more efficient inhibitor as reflected by the lowered Kiand
αKivalues.
Additionally, no influence on the Kivalues was observed in
experiments using DCPIP (2,6-dichlorophenolindophenol) as
electron acceptor at the bis-MGD, confirming that both forms
of inhibition rather occur near the active site instead of being
related to an interruption of the intramolecular electron
transfer.[2b]
Complementary to the biochemical assays, IR spectra of the
azide- and cyanate-treated variants were recorded (Figure 4).
While the IR characteristic band doublet was observed in the
spectrum of the RcFDHH387M variant for each of the two
inhibitors, albeit slightly shifted compared to the WT protein,
only the higher frequency component, as found for the
RcFDH~bisMGD, was detected in the spectrum of the RcFDHR587K
variant at 2042 and 2157 cm1for azide and cyanate,
respectively. These data suggest that the binding site of the
inhibitor, giving rise to the lower frequency IR bands at 2031
and 2153 cm1for azide and cyanate, respectively, is localized
between the bis-MGD cofactor, presumably the sulfido ligand,
Figure 3. Comparison of DFT-predicted (cal) and experimental (exp) IR absorption bands related to the OCN, NNNinhibitors and their isotopologues. The
corresponding asymmetric stretching frequencies of the inhibitors were calculated for a covalent MoN/15N/O bond (bound, dashed back lines) and a non-
covalent, pure electrostatic interaction near the bis-MGD cofactor (unbound, solid black lines) and compared to the corresponding band positions of free
inhibitors in water (solid dark blue line). A non-covalent interaction would thus result in a red shift of the inhibitor bands as depicted by red arrows. The
experimental results of the RcFDHWT incubated with the respective inhibitors exhibit two red shifted absorption bands (black solid trace, exp) relative to the
inhibitor bands in water (blue solid trace, exp) for A) NNNand B) OCN. The same results were obtained by using isotopically labelled azide and thiocyanate,
again in good agreement with C, D) the DFT predictions. Asterisk represents non-substracted free O13C15Nin H2O (D).
Chemistry—A European Journal
Research Article
doi.org/10.1002/chem.202201091
Chem. Eur. J. 2022,28, e202201091 (4 of 7) © 2022 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
Wiley VCH Dienstag, 20.09.2022
2254 / 256353 [S. 77/80] 1
and the Arg587 in a more defined manner. This binding site
would correspond to the competitive inhibition mode compo-
nent as derived from the kinetic experiments of RcFDHWT and
RcFDHH387M.
The interaction site at the bis-MGD was further investigated
by comparing an inactive oxo-ligated bis-MGD embedded in
RcFDHΔFdsC with the active RcFDHWT enzyme (Figures 5 and S10).
In the presence of the oxo ligand (Mo=O), the bis-MGD
dependent band related to competitive binding azide, shifts
from 2031 cm1to 2033 cm1, indicating the interaction of the
azide molecule with the oxo-ligated Mo ion. The addition of
formate leads to the displacement of azide from the binding
site reflected by the replacement of the lower frequency IR
band at 2033 cm1with a band at about 2042 cm1(Figure 5).
This translocation of the inhibitor was found to be reversible,
which is consistent with the observed mixed-type inhibition
observed in RcFDHWT . These findings imply that azide can
remain in the active site pocket under non-turnover conditions
and also in the presence of the substrate which would explain
the non-competitive component of azide inhibition, probably
by interrupting the rapid flow and alignment of substrate and
product to and from the active site, respectively.
Nevertheless, the exact location of the non-competitive
binding site of the inhibitor could not be conclusively inferred
from our data. It is reasonable to assume that this site is located
in the proximity of the competitive binding site but rather
exhibits differences in its conformation, that is more similar to
those of free azide in water (νfreeN3 =2049 cm1) (for further
explanation, see Figure S4). Presumably, it is bound to the
protein in a manner, which provides a higher degree of
translational freedom. This assumption would explain the small
spectral shifts in the inhibitor-treated RcFDHWT,RcFDHR587K and
RcFDHH387M holoenzyme variants (Figure 4). The respective
apoenzyme variants without the bis-MGD cofactor confirm
these small spectral shifts of the stretching vibrations of the
azide molecule at higher frequencies (Figure S11). Notably, also
the non-competitive binding site seems to be relevant for
catalysis. The azide molecule remains in the pocket during
catalysis and, hence, would contribute to a mixed-type
Figure 4. Normalized IR spectra of inhibited RcFDH variants. Azide (left
panel) and cyanate (right panel) inhibition was performed with RcFDHWT,
RcFDHH387M and RcFDHR587K. Spectrum displayed in grey was obtained from
apoprotein, RcFDHΔbisMGD, and is assigned to the non-competitive inhibition
interaction site. A small intensity loss as well as a shift (Δν=5 cm1) is
detected for the band assigned to the bis-MGD-dependent inhibition site in
RcFDHH387M samples treated with azide. A more significant loss of intensity
and a shift of Δν=6 cm1is observed for cyanate inhibition. When Arg587 is
replaced (RcFDHR587K and RcFDHH387M/R587T), an absorption assigned to an
apparent competitive interaction of the respective inhibitor can be detected,
due to changes in competitive inhibitor binding relative to RcFDHWT. Further
details related to RcFDHH387M/R587T are given in the Figure S9. Spectra were
measured in 100 mM Tris-HCl pH 9.0 (10°C) and 10 mM of the respective
inhibitor. The asterisk labels residual spectral contribution of free cyanate in
water.
Figure 5. Normalized IR spectra of azide inhibited RcFDHΔFdsC variants. The
inactive oxo-ligated bis-MGD cofactor embedded in RcFDH was character-
ized in the “as isolated” form (dark blue), in presence of formate (red) and
after removal of formate (purple). Solid lines represent the fits of individual
band contributions, whereas the dashed line display the respective
experimental spectrum. Similar to RcFDHWT, two main bands can be
deconvoluted. The absorption of the competitively bound azide at
2033 cm1(dark blue) is shifted towards slightly higher frequencies
compared to RcFDHWT. This could be explained by the exchange of the
sulfido ligand, which was previously identified as putative interaction site of
the competitively inhibiting azide, to an oxo ligand. The competitively
inhibiting azide in this position was displaced in the presence of formate,
thereby, forming the second observed, apo-protein-like bound azide at
about 2042 cm-1 (red). Upon removal of the substrate formate, the azide flips
back from the non-competitive to the competitive binding site (purple).
Chemistry—A European Journal
Research Article
doi.org/10.1002/chem.202201091
Chem. Eur. J. 2022,28, e202201091 (5 of 7) © 2022 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
Wiley VCH Dienstag, 20.09.2022
2254 / 256353 [S. 78/80] 1
inhibition for RcFDHWT and RcFDHH387M as well as an apparent
competitive inhibition for RcFDHR587K.
Conclusion
By means of a combined approach including inhibition kinetics,
site directed mutagenesis, DFT calculations and IR spectroscopy,
we have elucidated the binding motifs of inhibitory azide and
cyanate anions in the Mo-dependent formate dehydrogenase. It
was found, that, independent of the oxidation state of the Mo
ion, one inhibitor can electrostatically interact with two differ-
ent binding sites in the secondary coordination sphere of the
bis-MGD Mo center. These interaction sites were correlated with
the inhibition modes suggested by enzyme kinetic assays and
were assigned to a competitive and a non-competitive binding
site, whereby the latter is presumably located more remote
from the bis-MGD cofactor.
Since the inhibitors are isoelectronic and isostructural with
CO2, but also charged as formate, they may serve as models for
substrate or product binding to FDH. Hereby, the competitive
binding site could contribute to a stabilization of formate/CO2
molecules between the crucial Arg587 and the sulfido ligand of
the bis-MGD cofactor during catalysis. This particular binding
motif might also explain the increased oxygen stability of azide-
or cyanate-inhibited enzymes by sterically shielding the sulfido
ligand against oxidative damage. In addition, also the adjacent
His387 plays a supportive role for inhibitor binding. The
importance of both amino acid residues in the binding of
formate/CO2has been discussed in a very recent computational
study on the active site structure of EcFDHN.[6a,11b] Moreover,
the specific interaction site was confirmed by an experiment in
which the inactive, oxo-ligated bis-MGD in RcFDHΔFdsC was
incubated with azide, exhibiting clearly its displacement from
the competitive binding site accompanied by a migration
towards the non-competitive binding site. However, a direct
binding of CO2or formate to the Mo ion during catalysis cannot
be entirely excluded from our static studies, but it seems to be
unfavored.[9,17] In summary, the present study provides the basis
for future mechanistic investigations of Mo/W-dependent
formate dehydrogenases and other metalloenzymes, which
may be deepened by (ultrafast) time-resolved, non-linear IR
methods targeting at the transitional state as demonstrated by
the work of Cheatum and co-workers.[18] Such results could
potentially provide valuable information for the optimization of
biotechnological applications using for example electro- or
photocatalytic reduction of formate in bio-fuel cells and photo-
electrochemical tandem devices,[3a,b] as well as for future blue-
prints for molecular catalysts, which may be easier produced on
an industrial relevant scale.
Acknowledgements
This work was funded by the DFG under Germany's Excellence
Strategy – EXC 2008/1-390540038 – UniSysCat. In addition, K. L.
and S.W. were supported by the Einstein Foundation Berlin
(Einstein Center of Catalysis) and the BIG-NSE, respectively. P.H.
and I.Z. are grateful for funding from the EU’s Horizon 2020
research and innovation programme under grant agreement
No. 810856. We thank Christian Teutloff (FU Berlin) and, in
particular, Giorgio Caserta and Stefan Frielingsdorf (TU Berlin)
for fruitful discussions. Open Access funding enabled and
organized by Projekt DEAL.
Conflict of Interest
The authors declare no conflict of interest.
Data Availability Statement
The data that support the findings of this study are available
from the corresponding author upon reasonable request.
Keywords: CO2reduction ·DFT ·formate oxidation ·inhibition
kinetics ·IR spectroscopy ·molybdoenzyme
[1] a) A. M. Appel, J. E. Bercaw, A. B. Bocarsly, H. Dobbek, D. L. Dubois, M.
Dupuis, J. G. Ferry, E. Fujita, R. Hille, P. J. A. Kenis, C. A. Kerfeld, R. H.
Morris, C. H. F. Peden, A. R. Portis, S. W. Ragsdale, T. B. Rauchfuss, J. N. H.
Reek, L. C. Seefeldt, R. K. Thauer, G. L. Waldrop, Chem. Rev. 2013,113,
6621–6658; b) S. Sultana, P. Chandra Sahoo, S. Martha, K. Parida, RSC
Adv. 2016,6, 44170–44194; c) Y. Li, M. Gomez-Mingot, T. Fogeron, M.
Fontecave, Acc. Chem. Res. 2021,54, 4250–4261; d) M. Yuan, M. J.
Kummer, S. D. Minteer, Chem. Eur. J. 2019,25, 14258–14266.
[2] a) T. Hartmann, S. Leimkühler, FEBS J. 2013,280, 6083–6096; b) T.
Hartmann, N. Schwanhold, S. Leimkühler, Biochim. Biophys. Acta -
Proteins and Proteomics 2015,1854, 1090–1100.
[3] a) M. Miller, W. E. Robinson, A. R. Oliveira, N. Heidary, N. Kornienko, J.
Warnan, I. A. C. Pereira, E. Reisner, Angew. Chem. Int. Ed. 2019,58, 4601–
4605; Angew. Chem. 2019,131, 4649–4653; b) E. Edwardes Moore, V.
Andrei, A. R. Oliveira, A. M. Coito, I. A. C. Pereira, E. Reisner, Angew.
Chem. Int. Ed. 2021,60, 26303–26307; c) K. P. Sokol, W. E. Robinson, A. R.
Oliveira, J. Warnan, M. M. Nowaczyk, A. Ruff, I. A. C. Pereira, E. Reisner, J.
Am. Chem. Soc. 2018,140, 16418–16422; d) R. Abu, J. M. Woodley,
ChemCatChem 2015,7, 3094–3105.
[4] C. Radon, G. Mittelstädt, B. R. Duffus, J. Bürger, T. Hartmann, T. Mielke, C.
Teutloff, S. Leimkühler, P. Wendler, Nat. Commun. 2020,11, 1912.
[5] X. Guo, X. Wang, Y. Liu, Q. Li, J. Wang, W. Liu, Z. K. Zhao, Chem. Eur. J.
2020,26, 16611–16615.
[6] a) C. S. Mota, M. G. Rivas, C. D. Brondino, I. Moura, J. J. G. Moura, P. J.
González, N. M. F. S. A. Cerqueira, J. Biol. Inorg. Chem. 2011,16, 1255–
1268; b) H. C. A. Raaijmakers, M. J. Romão, J. Biol. Inorg. Chem. 2006,11,
849–854.
[7] a) T. Hartmann, P. Schrapers, T. Utesch, M. Nimtz, Y. Rippers, H. Dau,
M. A. Mroginski, M. Haumann, S. Leimkühler, Biochemistry 2016,55,
2381–2389; b) D. Niks, J. Duvvuru, M. Escalona, R. Hille, J. Biol. Chem.
2016,291, 1162–1174; c) W. E. Robinson, A. Bassegoda, E. Reisner, J.
Hirst, J. Am. Chem. Soc. 2017,139, 9927–9936; d) M. Meneghello, A. R.
Oliveira, A. Jacq-Bailly, I. C. Pereira, C. Léger, V. Fourmond, Angew. Chem.
Int. Ed. 2021,60, 9964–9967.
[8] J. S. Blanchard, W. W. Cleland, Biochemistry 1980,19, 3543–3550.
[9] P. E. M. Siegbahn, J. Phys. Chem. B 2022, 126, 1728–1733.
[10] a) R. K. Thauer, G. Fuchs, B. Käufer, Hoppe-Seyler’s Z. Physiol. Chem. 1975,
356, 653–662; b) M. J. Barber, H. D. May, J. G. Ferry, Biochemistry 1986,
25, 8150–8155; c) M. J. Axley, A. Böck, T. C. Stadtman, Proc. Nati. Acad.
Sci. 1991,88, 8450–8454.
[11] a) P. Schrapers, T. Hartmann, R. Kositzki, H. Dau, S. Reschke, C. Schulzke,
S. Leimkühler, M. Haumann, Inorg. Chem. 2015,54, 3260–3271; b) B. R.
Duffus, P. Schrapers, N. Schuth, S. Mebs, H. Dau, S. Leimkühler, M.
Haumann, Inorg. Chem. 2020,59, 214–225.
Chemistry—A European Journal
Research Article
doi.org/10.1002/chem.202201091
Chem. Eur. J. 2022,28, e202201091 (6 of 7) © 2022 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
Wiley VCH Dienstag, 20.09.2022
2254 / 256353 [S. 79/80] 1
[12] a) A. Marie Woys, S. S. Mukherjee, D. R. Skoff, S. D. Moran, M. T. Zanni, J.
Phys. Chem. B 2013,117, 5009–5018; b) R. Bloem, K. Koziol, S. A.
Waldauer, B. Buchli, R. Walser, B. Samatanga, I. Jelesarov, P. Hamm, J.
Phys. Chem. B 2012,116, 13705–13712.
[13] C. G. Bazewicz, M. T. Liskov, K. J. Hines, S. H. Brewer, J. Phys. Chem. B
2013,117, 8987–8993.
[14] N. Böhmer, T. Hartmann, S. Leimkühler, FEBS Lett. 2014,588, 531–537.
[15] M. Jormakka, S. Törnroth, B. Byrne, S. Iwata, Science (1979) 2002,295,
1863–1868.
[16] M. Leone, A. Cupane, V. Militello, M. E. Stroppolo, A. Desideri,
Biochemistry 1998,37, 4459–4464.
[17] G. Dong, U. Ryde, J. Biol. Inorg. Chem. 2018,23, 1243–1254.
[18] a) J. N. Bandaria, S. Dutta, S. E. Hill, A. Kohen, C. M. Cheatum, J. Am.
Chem. Soc. 2007,130, 22–23; b) P. Pagano, Q. Guo, C. Ranasinghe, E.
Schroeder, K. Robben, F. Häse, H. Ye, K. Wickersham, A. Aspuru-Guzik,
D. T. Major, L. Gakhar, A. Kohen, C. M. Cheatum, ACS Catal. 2019,9,
11199–11206; c) J. N. Bandaria, S. Dutta, M. W. Nydegger, W. Rock, A.
Kohen, C. M. Cheatum, Proc. Nat. Acad. Sci. 2010,107, 17974–17979.
Manuscript received: April 9, 2022
Accepted manuscript online: June 5, 2022
Version of record online: August 3, 2022
Chemistry—A European Journal
Research Article
doi.org/10.1002/chem.202201091
Chem. Eur. J. 2022,28, e202201091 (7 of 7) © 2022 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
Wiley VCH Dienstag, 20.09.2022
2254 / 256353 [S. 80/80] 1
