Supporting Information
© Wiley-VCH 2009
69451 Weinheim, Germany

Role of the [2Fe]H subcluster environment on the properties of key
intermediates in the catalytic cycle of [FeFe] hydrogenases. Hints for the
rational design of synthetic catalysts.

Maurizio Bruschi,*a Claudio Greco,b Markus Kaukonen,c Piercarlo Fantucci,b Ulf Ryde,*c Luca De
Gioia,*b


a Department of Environmental Science, University of Milano-Bicocca, Piazza della Scienza 1 20126-
Milan (Italy). b Department of Biotechnology and Biosciences, University of Milano-Bicocca, Piazza
della Scienza 2 20126-Milan (Italy). c Department of Theoretical Chemistry, Lund University, P.O. Box
124, Lund S-221 00, Sweden

Supporting Information

2
Computational Details
All quantum mechanics (QM) calculations have been carried out with the TURBOMOLE 5.7 suite of
programsi in the DFT framework by using the BP86ii and B3LYPiii functionals, and an all-electron
valence triple-ζ basis set with polarization functions on all atoms (def-TZVP).iv For the BP86
calculations the resolution-of-the-identity (RI) technique is applied.v The starting structure for the
QM calculations was based on the X-ray geometry of the H-cluster taken from Desulfuvibrio
Desulfuricans (pdb code: 1HFE) [FeFe] hydrogenase,vi in which the water molecule bridging Fed and
Fep (Scheme 1), was replaced by a carbonyl group, following a more recent refinementvii of the
crystal structure. In all QM calculations (Mod-1, Mod-2, and Mod-3) the proteic environment has
been modelled according to the conductor-like screening model (COSMO)viii, by considering a
polarizable continuum medium with ε = 4.
All quantum mechanics/molecular mechanics (QM/MM) optimizations were carried out with the
COMQUM program suite,ix,x in which Turbomole 5.7i for the QM part, and AMBER 8xi (with the Amber
1999 force field)xii for the MM part are used. The QM part in the QM/MM calculations was carried out
within the density functional theory (DFT) framework, using the BP86 functionalii and an all-electron
split-valence basis with polarization functions on all atoms (SVP),xiii and the B3LYPiii functional with the
def-TZVP basis set.iv
A theoretical investigation on the entire H-cluster (Mod-2, Mod-3 and Mod-4) is a challenge because
the Fe4S4 cluster is composed of two Fe2S2 layers of high-spin Fe ions, coupled antiferromagnetically
to give an overall low-spin ground state. The ground-state wave function of such spin-coupled
systems corresponds to linear combinations of multiple determinants that cannot be treated within the
single-determinant DFT approach. However, in the framework of the unrestricted formalism, these
interactions can be modelled with the broken symmetry (BS) approach introduced by Noodleman et
al,xiv which consists in the localisation of opposite spins of the mono-determinant wave function in
different parts of the molecule. In this work the antiferromagnetic coupling of Fe atoms in the Fe4S4
cluster has been treated within the BS formalism. Following this scheme a high spin wave function
was initially obtained by a single point energy calculation. Then, the electron spin density was
polarized in opposite directions to obtain the low-spin (S=0 or 1/2) broken symmetry wave function,
as described in Ref. xv. Spin and charge populations have been computed using the Mulliken scheme.
As noted by Brunold,xvi the Fe atoms of Fe4S4 are not equivalent in the H-cluster, and therefore six
different BS configurations can be generated for each species. However, we only considered one of the
six possible BS wave function variants with the lowest energy. In fact, various QM/MM and QM
calculations on the reduced form of the enzyme (data not shown) have indicated that the structural
differences among H-cluster models characterized by different BS configurations are very small (the

3
differences between corresponding bond lengths in the Fe2S2 cluster were always below 0.02 Å), and the
QM and QM/MM energies never differed by more than about 2.5 kcal mol-1. In view of these
considerations, we only considered the coupling scheme in which the two Fe2S2 layers are defined by the
Fe1-Fe2 and Fe3-Fe4 metal sites (see Scheme S1 for atom labels), in which majority α and β spin orbitals
are localized on the Fe1-Fe2 and Fe3-Fe4 layers, respectively.

Scheme 1. Structure of the H-cluster. Atoms are labelled consistently with Tables S3-S8.

The neutral form of the dtma ligand features two conformations given by the inversion of HN. The dtma-
complexes considered in this study have been modelled with the HN atom in the axial position of the six-
membered ring defined by the atoms FepS2CH2CH2NH.
In the following detailed information for each of the four models considered in the present work will be
reported.
Mod-1. Mod-1 (see Figure S1) corresponds to the bimetallic complexes [FeFe(SHCH3)(CO)3(CN)2-
(dtmaH)]-1, [FeFe(SHCH3)(CO)3(CN)2(dtma)(H)]-x (x = -1, -2), [FeFe(SHCH3)(CO)3(CN)2(dtmaH)(H)]-x
(x = 0, -1), and [FeFe(SHCH3)(CO)3(CN)2(dtma)(H2)]-1, in which the Fe4S4 cluster was simply modelled
by protonation of the sulfur atom of the CH3S group. For each complex a full geometry optimization has
been performed on the terminal- and µ-hydride/CO/H2 isomers using the BP86/def-TZVP and the
B3LYP/def-TZVP computational schemes. Energy differences between catalytically relevant isomers,

4
absolute energy values (BP86 and B3LYP) and selected geometrical parameters (BP86 and B3LYP) are
reported in Tables S1, S2, S3, S6 and S10, respectively.

Figure S1. [(CH3SH)FeFe(CO)3(CN)2(dtma)(H)]1- complex (Mod-1; [tH-dtma]3-) optimized using the
BP86/def-TZVP computational scheme.

Mod-2. Mod-2 (see Figure S2) corresponds to the entire H-cluster, in which the four cysteine residues
coordinated to the Fe4S4 cluster are modelled by CH3S- groups. The complexes investigated are
[(CH3S)4(Fe4S4)FeFe(CO)3(CN)2(dtmaH)]3-, [(CH3S)4(Fe4S4)FeFe(CO)3(CN)2(dtma)(H)]x- (x = -3, -4),
[(CH3S)4(Fe4S4)FeFe(CO)3(CN)2(dtmaH)(H)]x- (x = -2, -3), and [(CH3S)4(Fe4S4)FeFe(CO)3(CN)2-
(dtma)(H2)]3-, both in the terminal-, and µ-hydride/CO/H2 configurations. Geometry optimizations,
performed using the BP86/def-TZVP and B3LYP/def-TZVP computational schemes, have been carried
out by constraining the carbon atoms of the CH3S- groups at the X-ray positions of the Cβ atoms of the
cysteine residues coordinated to the Fe4S4 core (cys179, cys234, cys378, and cys382 in the 1HFE PDB
file). Energy differences between catalytically relevant isomers, absolute energy values (BP86 and
B3LYP), atomic charge and spin densities, and selected geometrical parameters (BP86 and B3LYP) are
reported in Tables S1, S2, S3, S4, S5, S7, and S11, respectively.

5

Figure S2. [(CH3S)4(Fe4S4)FeFe(CO)3(CN)2(dtma)(H)]3- complex (Mod-2; [tH-dtma]3-) optimized using
the BP86/def-TZVP computational scheme. The asterisks indicate the carbon atoms constrained during
geometry optimizations.

Mod-3. Mod-3 (see Figure S3) corresponds to the entire H-cluster as in Mod-2, plus the sidechains of the
residues lys237, glu241 cys178, and the backbone atoms of pro108 and ala109, and pro203 and ile204,
which surround the [2Fe]H cluster (residue numbers from the 1HFE PDB file). In particular, atoms Cγ, Cδ
and Cε of Lys237, atoms Cγ

and Cδ

of Glu241, and Cβ and Sγ of Cys178 are included in the model. In the
case of the backbone atoms, Cα, and N of Ala109, C, O and Cα of Pro108, Cα, and N of Ile204, and C, O
and Cα of Pro203 are included in the model (see Figure S3). The geometry optimizations performed using
the BP86/def-TZVP computational scheme have been carried out by constraining at their X-ray positions
the atoms Cγ of Lys237, Cγ of Glu241, Cα and Cβ of Cys178, Cα of Ala109 and Pro108, and Cα of Ile204
and Pro203 (see Figure S3). The carbon atoms of the CH3S- groups coordinated at the Fe4S4 cluster, are
also constrained as in Mod-2. Energy differences between catalytically relevant isomers, absolute energy
values, atomic charge and spin densities, and selected geometrical parameters are reported in Tables S1,
S2, S4, S5, S8, and S12, respectively.

6

Figure 3S. [(CH3S)4(Fe4S4)FeFe(CO)3(CN)2(dtma)(H)]3- complex (Mod-3; [tH-dtma]3-) optimized using
the BP86/def-TZVP computational scheme. The asterisks indicate the carbon atoms constrained during
geometry optimizations. In the figure only selected hydrogen atoms are shown.

Mod-4. Mod-4 corresponds to the entire protein in the framework of a QM/MM approach. In the
QM/MM calculations the computational system is divided into three subsystems: System 1 is treated at
QM level, and contains the H-cluster atoms and relevant surrounding atoms (see below). System 2
consists of all residues with any atom within 12 Å of any atom in system 1 and it is optimized by a full
MM minimization during QM/MM computations (see below for details on the strategy adopted for
QM/MM optimizations). Finally, the remaining portion of the protein, together with the water molecules
surrounding it are included in system 3, which is kept fixed at the crystallographic coordinates. Apart
from system 1, which is represented by a wave function during the QM/MM geometry optimizations,
each atom is represented by a partial point charge, taken from the Amber libraries.xii All such MM
charges are included in the Hamiltonian of the QM calculations, and thus the quantum chemical system is
polarized by the atoms of system 2 and 3 in a self-consistent way. When the quantum and classical
regions are connected by a chemical bond, the hydrogen link-atom approach is applied,xvii i.e. the QM
system is truncated with hydrogen atoms, the positions of which are linearly related to the corresponding
carbon atom in the protein.
Geometry optimizations were carried out in the following way:

7
(i) First, systems 2 and 3 were frozen and only the quantum system was optimized. (ii) Then, both
systems 1 and 2 were allowed to relax. In the MM optimization of system 2, the charges of the quantum
atoms were updated in each iteration of the QM/MM optimization. This optimization was performed
with the looser convergence criterion of 10-4 a.u. for the change in total QM/MM energy and 10-2 a.u. for
the maximum norm of the Cartesian gradient.x For each couple of protein isomers along a reaction path,
QM/MM energy minimizations with system 1 and 2 free to relax were carried out several times using an
iterative strategy: this means that, after having relaxed both system 1 and system 2 of each species, the
conformation of system 1 in the reactant was modified according to the geometric features of system 1 in
the product, and vice versa; then, the QM/MM optimizations were restarted. This procedure was repeated
until the ∆EQM/MM value of iteration i + 1 differed from that of step i by less than 1 kcal mol-1. (iii) Then,
system 2 was frozen again, and the geometry optimization was continued with default convergence
criteria (10-6 and 10-3 a.u.).
All QM/MM calculations were based on the 1.6-Å resolution structure of the [FeFe] hydrogenase from
Desulfovibrio desulfuricans (PDB code 1HFE).vi This is a hetero-dimer composed of a large subunit that
harbours the H-cluster, and a small subunit. Hydrogen atoms were added to the crystal structure, and the
protein was solvated in a sphere of water molecules with a radius of 48 Å, with the Amber routine leap. In
order to optimize the positions of hydrogen atoms and solvent water molecules, a 90 ps simulated-
annealing molecular dynamics calculation was carried out, followed by 10000 steps of conjugate gradient
energy minimization. The protonation state of histidine side chains was chosen considering solvent
exposure and the hydrogen-bond network around the residues; this means that, for each His side chain, all
possible hydrogen bond donors and acceptors in close proximity of the imidazole nitrogen atoms were
identified, and a congruent disposition of the proton(s) on the ring was established. As a result of such a
procedure, we assigned protonation of Nδ1 for residues S89 (the letter S indicates the small subunit;
residue numbers without a S refer to the large subunit) and 75; protonation of Nε2 for residues 351 and
371; and protonation on both the nitrogen atoms of the imidazole ring for residues S82, S85, S91, 14, 26,
58, 62, 141 and 196. All Lys and Arg residues were considered in their positively charged state, while
Asp and Glu side chains were always included in the anionic form. Finally, the iron-bound Cys residues
(i.e. amino acids 36, 38, 41, 45, 66, 69, 72, 76, 179, 234, 378 and 382) were assumed to be deprotonated.
All the ligands found in the PDB file were included in the QM/MM model, except a water molecule
bridging Fed and Fep, which was replaced by a carbonyl group.
The [FeFe] hydrogenase from D. desulfuricans contains two accessory iron-sulfur clusters in addition to
the H-cluster. For these Fe4S4 sites, we used Merz–Kollman electrostatic potential (ESP) charges,xviii
taken from QM calculations at the B3LYP/6-31G* level for truncated models of each site. The accessory
Fe4S4 clusters can reside in two alternative redox states. Following previous experimental results,xix we

8
adopted the [Fe4S4]1+ redox state of the two accessory clusters for the [dtmaH]3-, [tH-dtma]3-, [µH-
dtma]3-, [tH-dtmaH]2- and [µH-dtmaH]2- species, and the [Fe4S4]2+ redox state for the [tH-dtma]4-,
[µH-dtma]4-, [tH-dtmaH]3-, [tH-dtmaH]3-, and [tH2-dtma]3- species.
The QM system (region 1) of the various [FeFe] hydrogenase models here considered always included
the iron and sulphide ions of the Fe6S6 H-cluster, a DTMA ligand bridging Fed and Fep, three CO groups,
two CN– ligands, and four CH3S- groups that represent the Cys residues connecting the Fe6S6 cluster to
the rest of the enzyme large subunit (Cys179, Cys234, Cys378, Cys382). In addition, the sidechain of
Cys178 was also included in the QM region, in the form of a CH3SH group. This residue is close to the
bidentate ligand of the binuclear subcluster, and it might act as the terminal element of the proton channel
that supplies protons to the H-cluster during the dihydrogen-evolving route.
As for the composition of the various QM/MM models considered in this paper, they can differ in terms
of the nature of ligands coordinated to Fed; moreover, the DTMA protonation state can also vary (see the
main text). The total number of atoms in the QM system is 57-58 for all models.

Comparison between BP86 and B3LYP calculations
The energy differences between µ-hydride and terminal-hydride isomers calculated using the B3LYP/def-
TZVP scheme are reported in Table S1, together with those calculated with the BP86 functional and
reported in the main text. BP86 and B3LYP absolute energies for all of the complexes considered in this
study are reported in Tables S2 and S3, Mulliken spin and atomic charges are reported in Tables S4 and
S5, and relevant geometrical parameters are collected in Tables S6-S11.
It can be noticed from the data reported in table S1, that B3LYP energy differences systematically favor
µ-hydride isomers by about 6-10 kcal mol-1, with respect to the corresponding BP86 values. The only
exception to this trend is represented by the [H-dtmaH]3- species, Mod-1 and Mod-2; in fact, in this case
the µ-hydride/terminal-hydride energy differences computed at BP86 or B3LYP level are very similar.
Taken as a whole, B3LYP results give further support to the scenario that can be drawn from BP86
results, namely that µ-hydrides are significantly more stable than terminal-hydride H-cluster adducts, and
that this stability difference tends to lower upon protonation of DTMA and concomitant reduction of the
binuclear cluster.
It should also be noted that the electronic structures calculated with the two functionals are very similar
for all of the complexes investigated.

9
Table S1. ∆E (kcal mol-1) between catalytically relevant isomers calculated using the BP86/def-TZVP
(Mod-1, Mod-2, and Mod-3), and BP86/SVP (Mod-4), and (in parenthesis) the B3LYP/def-TZVP
schemes (all models; see Computational Details).
Mod-1 Mod-2 Mod-3 Mod-4
[dtmaH]3- → [tH-dtma]3- -9.7 (-9.9)
-14.5
(-15.4)
-11.9
─
-14.3
(-14.4)
[tH-dtma]3- → [µH-dtma]3- -11.4 (-17.5)
-8.9
(-15.4)
-9.3
─
-9.4
(-13.4)
[tH-dtmaH]2-→ [µH-dtmaH]2- -8.7 (-13.7)
-5.9
(-12.3)
-5.3
─
-2.0
(-10.6)
[tH-dtma]4- → [µH-dtma]4- -3.9 (-2.3)
-5.8
(-16.1)
-4.7
─
-0.6
(-9.5)
[tH-dtmaH]3-→ [µH-dtmaH]3- -1.0 (-0.2)
-0.7
(-0.6)
-0.5
─
+0.5
(-8.1)



Molecular and electronic structure of [dtmaH]3-
X-ray crystallographic data of the Hred form of the H-cluster shown that one CO ligand is semi-bridged
between the Fe atoms.vii The crystallographic position of this semi-bridged CO is very difficult to
reproduce by theoretical calculations, probably due to the very flat potential energy profile along the
reaction coordinate corresponding to µ-CO movement.
In the present investigation, two isomers, one with µ-CO and one with a semi-bridging CO, or only one
isomer with a semi-bridging CO have been characterized for the [dtmaH]3- complex, depending on the
functional used and on the modeling of the metal cluster environment. In fact, in the bimetallic complex
(Mod-1), only the semi-bridged CO isomer is stable, both with the BP86 and B3LYP functionals. For the
entire H-cluster models (Mod-2, and Mod-3), the two functionals give different results. In the case of
BP86 the isomers featuring a µ-CO ligand or a semi-bridging COs have been identified, with the latter
slightly more stable than the former (1-2 kcal mol-1). In the case of B3LYP only one isomer featuring a
semi-bridged CO was identified.
The results reported in the Tables below always refer to the semi-bridging CO isomer, which is the lowest
energy isomer and is also in better agreement with the crystallographic structure.

10
Table S2. Energies (in Hartree) of the complexes considered in this study, calculated using the RI-BP86/def-TZVP computational scheme for Mod-
1, Mod-2, Mod-3, and the RI-BP86/SVP computational scheme for Mod-4

Mod-1 Mod-2 Mod-3 Mod-4a
[dtmaH]3- -4424.065576102 -12387.501050313 -13766.686184470 -12839.7897861304
[tH-dtma]3- -4424.081046136 -12387.524082999 -13766.705213573 -12839.8126412242
[tH-dtma]4- -4424.148688269 -12387.541626777 -13766.737184793 -12839.3286525343
[tH-dtmaH]2- -4424.523113859 -12388.023240312 -13767.194846083 -12840.1117395936
[tH-dtmaH]3- -4424.645482522 -12388.085300022 -12367.270512828 -12840.4943453410
[tH2-dtma]3- -4424.654183527 -12388.093878665 -13767.276554842 ─b


[µH-dtma]3- -4424.099165680 -12387.538209043 -13766.719981010 -12839.8276506672
[µH-dtma]4- -4424.154982720 -12387.550916980 -13766.744737802 -12839.3333388840
[µH-dtmaH]2- -4424.536961099 -12388.032664080 -13767.203356984 -12840.1117395936
[µH-dtmaH]3- -4424.647021203 -12388.086364906 -13767.271380148 -12840.4934749944

a) QM/MM energies as reported in Ref. ix and x; b) Energy not calculated.

11
Table S3. Energies (in Hartree) of the complexes considered in this study calculated using the B3LYP/def-TZVP computational scheme

Mod-1a Mod-1b Mod-2a Mod-2b Mod-4c
[dtmaH]3- -4422.645406222 -4422.650807205 -12384.053400859 -12384.086379439 -12839.682407820
[tH-dtma]3- -4422.661254581 -4422.667848482 -12384.07789200 -12384.107446115 -12839.705383004
[tH-dtma]4- -4422.728806870 -4422.744013095d -12384.094195805 -12384.135699927 -12839.227730924
[tH-dtmaH]2- -4423.110093586 -4423.117549047 -12384.578765129 -12384.610931898 -12840.007664894
[tH-dtmaH]3- -4423.22861909 -4423.240396574d -12384.642147298 -12384.673299113d -12840.384528901
[tH2-dtma]3- -4423.238294123 -4423.244650427 -12384.652148331 -12368.6552885783 ─f


[µH-dtma]3- -4422.689138350 -4422.695506146 -12384.102448797 -12384.131538944 -12839.726765007
[µH-dtma]4- -4422.732469338 -4422.747949954d -12384.119775195 -12384.157769941 -12839.243721394
[µH-dtmaH]2- -4422.131988787 -4423.139527453 -12384.598340409 -12384.629015332 -12840.007664894
[µH-dtmaH]3- -4423.22900395 -4423.242784109e -12384.643091671 -12384.672365373 -12840.397400204

a) Energies computed using the B3LYP/def-TZVP scheme on the geometries calculated at the BP86/def-TZVP level of theory; b) energies
computed using the B3LYP/def-TZVP scheme on the geometries calculated at the same level of theory; c) energies computed using the B3LYP/def-
TZVP scheme on the geometries calculated at the BP86/SVP level of theory; d) dissociation of the CH3SH group in the optimized structure; e)
dissociation of one SDTMA-Fed bond in the optimized structure; f) not calculated.

12
Table S4. Atomic spin densities of the Fe atoms in the H-cluster of the [FeFe] hydrogenases
calculated using the BP86/def-TZVP and B3LYP/def-TZVP (in parenthesis) schemes (for Mod-4
BP86/SVP and B3LYP/def-TZVP). Fe1,2 and Fe2,3 refer to the Fe atoms in the [Fe4S4] cluster of the
two layers coupled antiferromagnetically. For the atom labels see Scheme S1


Fe1,2 Fe3,4 Fep Fed
[dtmaH]3-
Mod-2
Mod-3
Mod-4
3.02, 3.02 (3.51, 3.57)
3.00, 3.04
3.14, 3.25
-3.04, -3.00 (-3.54, -3.53)
-2.98, -3.01
-3.17, -3.19
0.01 (-0.01)
0.02
-0.02
0.04 (0.04)
0.02
0.19
[tH-dtma]3-
Mod-2
Mod-3
Mod-4
3.22, 3.02 (3.50, 3.59)
3.00, 3.02
3.23, 3.27
-3.02, -2.98 (-3.53, -3.53)
-2.97, -3.00
-3.26, -3.17
0.03 (0.03)
0.03
0.03
0.01 (0.01)
0.01
0.02
[tH-dtma]4-
Mod-2
Mod-3
Mod-4
3.11, 3.10 (3.49, 3.61)
3.08, 3.03
3.32, 3.31
-2.83, -2.97 (-3.31, -3.39)
-2.95, -2.99
-3.26, -2.83
0.31 (0.05)
0.53
0.18
0.12 (0.00)
0.19
0.07
[tH-dtmaH]2-
Mod-2
Mod-3
Mod-4
2.98, 3.00 (3.47, 3.60)
2.96, 2.98
3.17, 3.26
-2.99, -2.95 (-3.51, -3.52)
-2.95, -2.94
-3.20, -3.15
0.05 (0.04)
0.04
0.05
0.01 (0.00)
0.01
0.02
[tH-dtmaH]3-
Mod-2
Mod-3
Mod-4
3.05, 3.02 (3.53, 3.57)
3.05, 3.01
3.24, 3.29
-3.02, -2.98 (-3.52 -3.54)
-3.01, -2.98
-3.22, -2.88
0.59 (0.66)
0.56
0.23
0.24 (0.24)
0.24
0.20
[tH2-dtma]3-
Mod-2
Mod-3
Mod-4
3.05, 3.03 (3.52, 3.58)
3.05, 3.01
─a
-3.02, -3.00 (-3.53, -3.53)
-3.00, -2.98
─a
0.51 (0.51)
0.46
─a
0.37 (0.41)
0.39
─a


[µH-dtma]3-
Mod-2
Mod-3
Mod-4
3.04, 3.02 (3.52, 3.57)
3.04, 3.00
3.22, 3.26
-3.03, -3.00 (-3.53, -3.53)
-3.01, -2.98
-3.16, -3.23
0.04 (0.02)
0.03
0.04
0.01 (0.01)
0.01
0.02
[µH-dtma]4-
Mod-2
Mod-3
Mod-4
3.14, 3.12 (3.52, 3.61)
3.11, 3.08
3.30, 3.31
-2.80, -2.92 (-3.32, -3.39)
-2.84, -2.94
-3.24, -2.79
0.16 (0.03)
0.26
0.12
0.10 (0.01)
0.18
0.08
[µH-dtmaH]2-
Mod-2
Mod-3
Mod-4
3.00, 3.01 (3.48, 3.59)
2.99, 3.00
3.19, 3.26
-3.00, -2.97 (-3.51, -3.53)
-2.98, -2.96
-3.22, -3.14
0.04 (0.03)
0.04
0.04
0.01 (0.01)
0.01
0.02
[µH-dtmaH]3-
Mod-2
Mod-3
Mod-4
3.07, 3.04 (3.54, 3.57)
3.04, 3.01
3.24, 3.29
-3.04, -2.98 (-3.53, -3.54)
-3.01, -2.97
-3.22, -2.88
0.39 (0.36)
0.38
0.23
0.38 (0.58)
0.39
0.20

a) Not calculated

13
Table S5. Mulliken atomic charges of the Fe atoms in the H-cluster of the [FeFe] hydrogenases
calculated using the BP86/def-TZVP and B3LYP/def-TZVP (in parenthesis) schemes (for Mod-4
BP86/SVP and B3LYP/def-TZVP). Fe1,2 and Fe2,3 refer to the Fe atoms in the [Fe4S4] cluster of the
two layers coupled antiferromagnetically. For the atom labels see Scheme S1


Fe1,2 Fe3,4 Fep Fed
[dtmaH]3-
Mod-2
Mod-3
Mod-4
0.21, 0.15 (0.32, 0.26)
0.21, 0.13
0.20, 0.20
0.14, 0.12 (0.25, 0.26)
0.13, 0.13
0.20, 0.17
-0.42 (-0.33)
-0.41
-0.54
-0.50 (-0.41)
-0.53
-0.51
[tH-dtma]3-
Mod-2
Mod-3
Mod-4
0.21, 0.14 (0.32, 0.26)
0.20, 0.12
0.22, 0.20
0.13, 0.12 (0.25, 0.23)
0.12, 0.13
0.22, 0.16
-0.49 (-0.40)
-0.49
-0.61
-0.67 (-0.58)
-0.67
-0.73
[tH-dtma]4-
Mod-2
Mod-3
Mod-4
0.24, 0.17 (0.36, 0.30)
0.21, 0.14
0.24, 0.21
0.15, 0.15 (0.29, 0.28)
0.14, 0.15
0.22, 0.16
-0.44 (-0.39)
-0.39
-0.60
-0.60 (-0.54)
-0.57
-0.73
[tH-dtmaH]2-
Mod-2
Mod-3
Mod-4
0.20, 0.16 (0.25, 0.31)
0.18, 0.12
0.22, 0.20
0.12, 0.11 (0.24, 0.23)
0.11, 0.12
0.21, 0.17
-0.53 (-0.44)
-0.53
-0.65
-0.70 (-0.61)
-0.69
-0.76
[tH-dtmaH]3-
Mod-2
Mod-3
Mod-4
0.21, 0.14 (0.32, 0.26)
0.19, 0.13
0.22, 0.21
0.14, 0.13 (0.25, 0.25)
0.13, 0.13
0.21, 0.15
-0.40 (-0.30)
-0.42
-0.45
-0.58 (-0.48)
-0.59
-0.79
[tH2-dtma]3-
Mod-2
Mod-3
Mod-4
0.20, 0.15 (0.31, 0.27)
0.20, 0.13
─a
0.14, 0.12 (0.26, 0.24)
0.13, 0.13
─a
-0.45 (-0.35)
-0.45
─a
-0.57 (-0.47)
-0.59
─a


[µH-dtma]3-
Mod-2
Mod-3
Mod-4
0.20, 0.15 (0.31, 0.27)
0.20, 0.13
0.21, 0.22
0.14, 0.13 (0.25, 0.24)
0.13, 0.13
0.20, 0.16
-0.30 (-0.52)
-0.29
-0.60
-0.59 (-0.21)
-0.59
-0.86
[µH-dtma]4-
Mod-2
Mod-3
Mod-4
0.20, 0.15 (0.32, 0.26)
0.20, 0.13
0.24, 0.21
0.14, 0.12 (0.25, 0.24)
0.13, 0.14
0.21, 0.16
-0.47 (-0.35)
-0.45
-0.46
-0.65 (-0.54)
-0.67
-0.80
[µH-dtmaH]2-
Mod-2
Mod-3
Mod-4
0.25, 0.19 (0.37, 0.30)
0.22, 0.16
0.21, 0.21
0.17, 0.16 (0.30, 0.29)
0.14, 0.16
0.21, 0.16
-0.43 (-0.33)
-0.40
-0.49
-0.61 (-0.53)
-0.61
-0.85
[µH-dtmaH]3-
Mod-2
Mod-3
Mod-4
0.18. 0.14 (0.29, 0.26)
0.18, 0.13
0.22, 0.21
0.13, 0.12 (0.24, 0.23)
0.12, 0.12
0.21, 0.15
-0.50 (-0.38)
-0.51
-0.45
-0.69 (-0.59)
-0.70
-0.79

a) Not calculated

14
Table S6. Selected geometrical parameters (Å and degrees) of the complexes considered in this study in corresponding to model 1 (Mod-1),
calculated using the RI-BP86/def-TZVP scheme. For the atom labels see Scheme S1


a) arithmetic mean of the two SDTMA-Fep distances; b) arithmetic mean of the two SDTMA-Fed distances; c) distance between the Fep atom and the
carbon atom of the bridging CO; d) distance between the Fed atom and the carbon atom of the bridging CO.



S1-Fep SDTMA-Fepa SDTMA-Fedb Fep-C(O)c Fed-C(O)d Fep-Fed Fed-HFe HFe—HN
[dtmaH]3- 2.256 2.298 2.328 2.998 1.757 2.566 ─ ─
[tH-dtma]3- 2.267 2.329 2.305 2.157 1.864 2.518 1.517 2.226
[tH-dtma]4- 2.680 2.355 2.354 2.004 1.941 2.608 1.551 1.941
[tH-dtmaH]2- 2.264 2.312 2.317 2.198 1.864 2.527 1.529 1.653
[tH-dtmaH]3- 2.497 2.351 2.363 2.021 1.931 2.629 1.591 1.308
[dtmaH]3- 2.424 2.355 2.355 1.953 1.952 2.630 1.711/1.778 0.823


[µH-dtma]3- 2.263 2.315 2.324 ─ ─ 2.587 1.630/1.706 ─
[µH-dtma]4- 2.265 2.323 2.316 ─ ─ 2.588 1.623/1.716 ─
[µH-dtmaH]2- 2.386 2.350 2.384 ─ ─ 2.743 1.614/1.770 ─
[µH-dtmaH]3- 2.274 2.313 2.324 ─ ─ 2.582 1.622/1.713 ─

15
Table S7. Selected geometrical parameters (Å and degrees) of the complexes considered in this study corresponding to model 2 (Mod-2), calculated
using the RI-BP86/def-TZVP scheme. For the atom labels see Scheme S1


a) arithmetic mean of the six Fe-Fe distances in the Fe4S4 cluster; b) arithmetic mean of the two SDTMA-Fep distances; c) arithmetic mean of the two
SDTMA-Fed distances; d) distance between the Fep atom and the carbon atom of the bridging CO; e) distance between the Fed atom and the carbon
atom of the bridging CO.
Fe-Fea Fe1-S1 S1-Fep SDTMA-Fepb SDTMA-Fedc Fep-C(O)d Fed-C(O)e Fep-Fed Fed-HFe HFe—HN
[dtmaH]3- 2.683 2.321 2.373 2.371 2.330 2.149 1.808 2.582 ─ ─
[tH-dtma]3- 2.681 2.352 2.351 2.338 2.302 2.060 1.906 2.520 1.520 2.210
[tH-dtma]4- 2.690 2.368 2.462 2.353 2.325 2.000 1.943 2.577 1.542 2.047
[tH-dtmaH]2- 2.672 2.356 2.343 2.327 2.307 2.049 1.929 2.522 1.534 1.701
[tH-dtmaH]3- 2.682 2.332 2.572 2.350 2.354 1.973 1.966 2.636 1.593 1.355
[tH2-dtma]3- 2.685 2.320 2.490 2.356 2.356 1.929 1.970 2.638 1.715/1.787 0.823


[µH-dtma]3- 2.682 2.338 2.326 2.323 2.319 ─ ─ 2.605 1.628/1.723 ─
[µH-dtma]4- 2.691 2.378 2.375 2.330 2.340 ─ ─ 2.652 1.625/1.741 ─
[µH-dtmaH]2- 2.671 2.341 2.313 2.312 2.324 ─ ─ 2.603 1.630/1.726 ─
[µH-dtmaH]3- 2.683 2.329 2.420 2.378 2.386 ─ ─ 2.745 1.594/1.829 ─

16
Table S8. Selected geometrical parameters (Å and degrees) of the complexes considered in this study corresponding to model 3 (Mod-3), calculated
using the RI-BP86/def-TZVP scheme. For the atom labels see Scheme S1


a) arithmetic mean of the six Fe-Fe distances in the Fe4S4 cluster; b) arithmetic mean of the two SDTMA-Fep distances; c) arithmetic mean of the two
SDTMA-Fed distances; d) distance between the Fep atom and the carbon atom of the bridging CO; e) distance between the Fed atom and the carbon
atom of the bridging CO.
Fe-Fea Fe1-S1 S1-Fep SDTMA-Fepb SDTMA-Fedc Fep-C(O)d Fed-C(O)e Fep-Fed Fed-HFe HFe—HN
[dtmaH]3- 2.683 2.364 2.381 2.364 2.328 2.170 1.806 2.565 ─ ─
[tH-dtma]3- 2.680 2.384 2.363 2.335 2.302 2.044 1.915 2.522 1.519 2.256
[tH-dtma]4- 2.686 2.357 2.582 2.354 2.347 1.975 1.958 2.615 1.556 1.825
[tH-dtmaH]2- 2.672 2.368 2.353 2.324 2.311 2.031 1.937 2.522 1.535 1.655
[tH-dtmaH]3- 2.681 2.353 2.537 2.353 2.355 1.968 1.968 2.640 1.590 1.402
[tH2-dtma]3- 2.682 2.370 2.475 2.356 2.349 1.920 1.989 2.647 1.710/1.768 0.825


[µH-dtma]3- 2.680 2.362 2.332 2.327 2.321 ─ ─ 2.611 1.634/1.717 ─
[µH-dtma]4- 2.689 2.375 2.402 2.339 2.353 ─ ─ 2.685 1.622/1.752 ─
[µH-dtmaH]2- 2.670 2.357 2.315 2.313 2.323 ─ ─ 2.603 1.632/1.726 ─
[µH-dtmaH]3- 2.680 2.340 2.414 2.335 2.388 ─ ─ 2.748 1.590/1.834 ─

17
Table S9. Selected geometrical parameters (Å and degrees) of the complexes considered in this study corresponding to model 4 (Mod-4),
calculated using the RI-BP86/SVP scheme. For the atom labels see Scheme S1


a) arithmetic mean of the six Fe-Fe distances in the Fe4S4 cluster; b) arithmetic mean of the two SDTMA-Fep distances; c) arithmetic mean of the two
SDTMA-Fed distances; d) distance between the Fep atom and the carbon atom of the bridging CO; e) distance between the Fed atom and the carbon
atom of the bridging CO.


Fe-Fea Fe1-S1 S1-Fep SDTMA-Fepb SDTMA-Fedc Fep-C(O)d Fed-C(O)e Fep-Fed Fed-HFe HFe—HN
[dtmaH]3- 2.649 2.317 2.414 2.389 2.317 1.988 1.830 2.552 ─ ─
[tH-dtma]3- 2.660 2.363 2.390 2.340 2.299 1.990 1.902 2.493 1.523 2.665
[tH-dtma]4- 2.638 2.410 2.430 2.332 2.324 1.983 1.908 2.510 1.542 1.879
[tH-dtmaH]2- 2.650 2.360 2.356 2.324 2.312 1.975 1.924 2.482 1.543 1.693
[tH-dtmaH]3- 2.640 2.358 2.450 2.342 2.341 1.952 1.934 2.538 1.584 1.476


[µH-dtma]3- 2.660 2.350 2.336 2.316 2.311 ─ ─ 2.599 1.633/1.744 ─
[µH-dtma]4- 2.631 2.387 2.368 2.310 2.335 ─ ─ 2.623 1.630/1.755 ─
[µH-dtmaH]2- 2.656 2.351 2.315 2.303 2.301 ─ ─ 2.585 1.623/1.752 ─
[µH-dtmaH]3- 2.632 2.365 2.370 2.311 2.330 ─ ─ 2.646 1.613/1.781 ─

18
Table S10. Selected geometrical parameters (Å and degrees) of the complexes considered in this study in corresponding to model 1 (Mod-1),
calculated using the B3LYP/def-TZVP scheme. For the atom labels see Scheme S1


a) arithmetic mean of the two SDTMA-Fep distances; b) arithmetic mean of the two SDTMA-Fed distances; c) distance between the Fep atom and the
carbon atom of the bridging CO; d) distance between the Fed atom and the carbon atom of the bridging CO; e) dissociation of the CH3SH group
in the optimized structure; f) dissociation of one SDTMA-Fed bond in the optimized structure.


S1-Fep SDTMA-Fepa SDTMA-Fedb Fep-C(O)c Fed-C(O)d Fep-Fed Fed-HFe HFe—HN
[dtmaH]3- 2.324 2.344 2.379 2.887 1.767 2.555 ─ ─
[tH-dtma]3- 2.306 2.333 2.354 2.385 1.839 2.608 1.516 2.075
[tH-dtma]4- 6.156e 2.364 2.389 1.979 2.037 2.564 1.527 2.119
[tH-dtmaH]2- 2.306 2.348 2.375 2.442 1.845 2.639 1.531 1.599
[tH-dtmaH]3- 4.903e 2.362 2.402 1.979 2.047 2.570 1.548 1.573
[tH2-dtma]3- 2.503 2.387 2.391 1.982 1.978 2.681 1.776/1818 0.789


[µH-dtma]3- 2.336 2.365 2.352 ─ ─ 2.612 1.634/1.671 ─
[µH-dtma]4- 6.149e 2.372 2.397 ─ ─ 2.712 1.793/1.618 ─
[µH-dtmaH]2- 2.340 2.362 2.366 ─ ─ 2.615 1.636/1.659 ─
[µH-dtmaH]3- 2.380 2.390 2.314/3.628f ─ ─ 2.808 1.609/1.709 ─

19
Table S11. Selected geometrical parameters (Å and degrees) of the complexes considered in this study corresponding to model 2 (Mod-2),
calculated using the B3LYP/def-TZVP scheme. For the atom labels see Scheme S1


a) arithmetic mean of the six Fe-Fe distances in the Fe4S4 cluster; b) arithmetic mean of the two SPDT-Fep distances; c) arithmetic mean of the two
SPDT-Fed distances; d) distance between the Fep atom and the carbon atom of the bridging CO; e) distance between the Fed atom and the carbon atom
of the bridging CO; f) dissociation of the S1-Fep bond.
Fe-Fea Fe1-S1 S1-Fep SDTMA-Fepb SDTMA-Fedc Fep-C(O)d Fed-C(O)e Fep-Fed Fed-HFe HFe—HN
[dtmaH]3- 2.929 2.402 2.390 2.347 2.381 2.809 1.763 2.631 ─ ─
[tH-dtma]3- 2.926 2.427 2.370 2.374 2.351 2.362 1.848 2.629 1.520 2.063
[tH-dtma]4- 2.967 2.557 2.377 2.349 2.348 2.274 1.866 2.604 1.520 2.090
[tH-dtmaH]2- 2.916 2.442 2.351 2.336 2.368 2.442 1.854 2.675 1.534 1.625
[tH-dtmaH]3- 2.924 2.357 ─f 2.378 2.381 1.983 2.033 2.640 1.566 1.551
[tH2-dtma]3- 2.930 2.381 2.537 2.393 2.384 1.961 1.998 2.687 1.802/1.845 0.786


[µH-dtma]3- 2.929 2.409 2.400 2.370 2.349 ─ ─ 2.630 1.653/1.664 ─
[µH-dtma]4- 2.971 2.526 2.405 2.398 2.351 ─ ─ 2.643 1.665/1.660 ─
[µH-dtmaH]2- 2.915 2.422 2.374 2.365 2.357 ─ ─ 2.631 1.658/1.659 ─
[µH-dtmaH]3- 2.928 2.384 2.515 2.392 2.448 ─ ─ 2.818 1.567/1.902 ─

20
References

i R. Ahlrichs, M. Bar, M. Haser, H. Horn, C. Kolmel Chem. Phys. Lett. 1989, 162, 165.
ii
a) A. D. Becke, Phys. Rev. A 1988, 38, 3098; b) J. P. Perdew, Phys. Rev. B 1986, 33,
8822.
iii
a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648; b) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B
1988, 37, 785.
iv
A. Schafer, C. Huber, R. Ahlrichs J. Chem. Phys. 1994, 100, 5829.
v
a) Eichkorn, K.; Treutler, O.; Öhm, H.; Haser, M.; Ahlrichs, R. Chem. Phys. Lett. 1995, 240, 283;
b) Eichkorn, K.; Weigend, F.; Treutler, O.; Ahlrichs, R. Theor. Chim. Acta 1997, 97, 119.
vi
Y. Nicolet, C. Piras, P. Legrand, E. C. Hatchikian, J. C. Fontecilla-Camps, Structure 1999, 7, 13.
vii
Y. Nicolet,; A. L. de Lacey, X. Vernede, V. M. Fernandez, E. C. Hatchikian, J. C. Fontecilla-
Camps, J. Am. Chem. Soc. 2001, 123, 1596.
viii a) A. Klamt J. Phys. Chem. 1995, 99, 2224; b) A. Klamt J. Phys. Chem. 1996, 100, 3349.
ix
U. Ryde J. Comput.-Aided Mol. Des. 1996, 10, 153.
x
U. Ryde, M. H. M. Olsson, Int. J. Quantum Chem. 2001, 81, 335.
xi
D. A. Case, et al., Amber 8. 2004, University of California: San Francisco, CA.
xii
W. D. Cornell, P. I. Cieplak, C. I. Bayly, I. R. Gould, K. M. Merz, D. M. Ferguson, D. C.
Spellmeyer, T. Fox, J. W. Caldwell, P. A. Kolman J. Am. Chem. Soc. 1995, 117, 5179.
xiii
A. Schaefer, H. Horn, R. Ahlrichs J. Chem. Phys. 1992, 97, 2571.
xiv
a) L. Noodleman, J. G. Norman jr. J. Chem. Phys. 1979, 70, 4903; b) L. Noodleman J. Chem.
Phys. 1981, 74, 5737.
xv M. Bruschi, C. Greco, P. Fantucci, L. De Gioia Inorg. Chem. 2008, 47, 6056.
xvi
A. T. Flieder, T. C. Brunold Inorg. Chem. 2005, 44, 9322.
xvii
N. Reuter, A. Dejaegere, B. Maigret, M. Karplus, J. Phys. Chem. 2000, 104, 1720.
xviii
B. H. Besler, K. M. Merz, P. A. Kollman J. Comput. Chem. 1990, 11, 431.
xix
a) M. W. W. Adams, L. E. Mortenson, J. Biol. Chem. 1984, 259, 7045; b) I. C. Zambrano, A. T.
Kowal, L. E. Mortenson, M. W. W. Adams, M. K. Johnson, M. K. J. Biol. Chem. 1989, 264,
20974; c) F. M. Rusnak, M. W. W. Adams, L. E. Mortenson, E. Munck, J. Biol. Chem. 1987,
262, 38; d) C. V. Popescu, E. Münck, J. Am. Chem. Soc. 1999, 121, 7877; e) A. S. Pereira, P.
Tavares, I. Moura, J. J. G. Moura, B. H. Huynh J. Am. Chem. Soc. 2001, 123, 2771-2782.