Directing protonation in [FeFe] hydrogenase active site models by modifications in their second coordination sphere

Article abstract of DOI:10.1039/c0cc00724b

Subtle changes in the second coordination sphere of [Cl ₂bdtFe₂(CO)₄(Ph₂P-CH ₂-X-CH₂-PPh₂)] (bdt = benzene-1,2-dithiolate, X = NCH₃, NCH₂CF₃, CH<

Full text of DOI:10.1039/c0cc00724b

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Directing protonation in [FeFe] hydrogenase active site models
by modifications in their second coordination spherew
Salah Ezzaher,a Adolf Gogoll,b Clemens Bruhnc and Sascha Ott*a
Received 1st April 2010, Accepted 17th June 2010
DOI: 10.1039/c0cc00724b
Subtle changes in the second coordination sphere of [Cl2bdtFe2-
(CO)4(Ph2P–CH2–X–CH2–PPh2)] (bdt = benzene-1,2-dithiolate,
X = NCH3, NCH2CF3, CH2) that do not influence the electronic
character of the Fe2 center can however direct protonation to three
different sites: the N in the bis-phosphane, the Fe–Fe bond or the
bdt-S.
counter anion of the used acid,10,11 solvent12 or temperature.13,14
In contrast, systems in which the protonation behavior is
programmed into the complex have not been reported. We
have therefore prepared a series of electronically equivalent
Fe2 complexes 1–3 that differ solely in the second coordination
sphere (Chart 1b). Whereas complexes 1 and 2 both contain an
N-heteroatom in the bis-phosphane, complex 3 lacks such a
potential protonation site. The N in 1 is expected to be more
basic than that in 2 by 5 pKa units in analogy to the difference
in pKas of the free amines.15
X-Ray structure elucidation of [FeFe] hydrogenases (H2ase) in
the late 1990s1,2 has revealed the molecular structure of their
active sites (Chart 1a) which continue to fascinate bioinorganic
chemists to date.3,4 Apart from the structural modelling
challenge,5 particular focus has been given to the preparation
of functional complexes that catalyze the reduction of protons
to molecular hydrogen and thus offer an alternative to nobel
metals that are currently used for this purpose. Over the last
decade, more than 250 catalyst candidates have been examined
and many of them have been shown to be homogenous
electrocatalysts6 or function even in photocatalytic schemes.7–9
Despite these efforts, bioinorganic models are still far from
achieving similar high turnover rates and low overpotentials as
the enzymes. This discrepancy suggests that it is insufficient to
model only the primary coordination sphere, but that also the
surrounding protein matrix has to be considered in the design
of improved functional mimics. For this purpose, the second
coordination sphere receives more and more attention in the
modelling community.
Complexes 1–3 were synthesized from [Fe2(Cl2bdt)(CO)6]16
by stirring the latter in the presence of a slight excess of the
bidendate phosphane in toluene or THF. All complexes were
characterized by IR, 1H NMR and 31P NMR, elemental
analysis and in the case of 1 by X-ray crystallography.w In
the solid state structure of 1, a p-stacking interaction between
the bdt and a phenyl ring of the bis-phosphane leads to a
mixed apical–basal (ap,ba) coordination of the latter.w RT
31P NMR spectra of 1–3 (CH2Cl2) all feature exclusively one
singlet at practically identical chemical shift of d = 55.5, 55.6
and 55.6 ppm, respectively, indicating that each of the three
complexes prevails in the same coordination mode.14 At
temperatures lower than À50 1C, the signal in the 31P NMR
spectra of 1 and 2 splits into two singlets.w The coalescence
temperature for 3 is lower than those of 1 and 2, preventing the
observation of two signals even at À80 1C. This behavior is
consistent with inversion barriers that are considerably smaller
for cyclohexane compared to those of N-alkylated piperidines.17
Since the bdt ligand in 1–3 is planar and thus cannot give rise
to isomers in the same way as a more flexible propyldithiolate
(pdt),14 it is clear that the two signals stem from two dissimilar
P-centers and that the bis-phosphane in 1–3 thus coordinates
in an ap,ba fashion also in solution. The IR spectra of the
three complexes are identical in the nCO region, indicating
that the three bidentate phosphanes have an equal electronic
impact on their respective Fe2 sites and that the electron-
density at the iron centers in 1–3 is equivalent. The IR pattern
of 1–3 is similar to that reported for an analogue 412,18 that is
identical to 1 with the exception of a pdt instead of the bdt
bridge in 1.w The three major absorptions of 1–3 at 2032,
1965 and 1899 cmÀ1 are shifted by 13 cmÀ1 towards higher
frequencies compared to those of 4 due to the electron-
withdrawing character of the Cl2bdt bridging ligand.19 In
other words, electron density in 1–3 is substantially lower
compared to that in 4, a situation that may impede the
formation of hydride complexes as described for 4. To study
differences in the protonation behavior of 1–3 caused by
their second coordination sphere, the three complexes
were separately exposed to an equivalent acidic medium
(CH2Cl2, HBF4ÁOEt2).
It has been shown numerous times for [FeFe] H2ase active
site models that a given complex can lead to different proto-
nation products depending on the conditions such as the
a Department of Photochemistry and Molecular Science,
˚
Angstrom Laboratories, Uppsala University, Box 523,
¨
75120 Uppsala, Sweden. E-mail: sascha.ott@fotomol.uu.se;
Fax: +46 18-471-6844
b Department of Biochemistry and Organic Chemistry,
Uppsala University, Box 576, 751 23 Uppsala, Sweden
c Dept of Chemistry, Kassel University, Heinrich-Plett-Straße 40,
34132 Kassel, Germany
w Electronic supplementary information (ESI) available: Experimental
and spectroscopic details, crystal data of 1. CCDC 762982. For ESI
and crystallographic data in CIF or other electronic format see DOI:
10.1039/c0cc00724b
Chart 1
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 5775–5777 | 5775

View Online
Exposure of complex 1 to these conditions gives rise to a
new species that is characterized by IR absorptions at 2044,
1980 and 1908 cmÀ1 (Fig. 1a). Compared to the IR spectrum
of 1 that of the protonated form is shifted by 12 cmÀ1 to higher
frequencies, consistent with a protonation at the nitrogen
1
([1HN]+).12 This assignment is confirmed by H NMR data
which show no signal in the hydride region.w The bis-phosphane
in [1HN]+ presumably still resides in the ap,ba coordination
mode as the signal in its 31P NMR spectrum is only marginally
shifted to 59 ppm. The presence of only one signal also
suggests that the ring-flip of the N-protonated, six-membered
metalloazaheterocycle either proceeds fast on the NMR time-
scale or protonation occurs selectively either in the axial or the
equatorial position. This behavior is in contrast to complex 4,
which showed two N-protonated species that differ in the
stereochemistry at the N.12
In contrast to the IR spectrum of [1HN]+, protonation of
complex 2 results in the formation of a species with nCO
frequencies at 2106, 2058 and 1978 cmÀ1 (Fig. 1b). The much
greater shift of 74 cmÀ1 towards higher energy compared to
that of the non-protonated starting material is consistent with
+
Fig. 2 2D 1H–31P HMBC NMR characterization of [2HFe
]
in
CH2Cl2.
conditions despite the electron-withdrawing character of the
Cl2bdt ligand compared to the pdt analogue 4.
Whereas the difference in reactivity between 1 and 2 can
easily be explained by a simple difference in basicity, i.e. the
amine in 1 is a more basic site than that in 2 which is thus
protonated at the Fe–Fe site instead, the protonation behavior
of 3 is not expectable. Treatment of complex 3 with HBF4Á
OEt2 in CH2Cl2 affords a species with IR absorptions at
2095, 2055 and 1996 cmÀ1 that are thus shifted by 63 cmÀ1
compared to those of 3 (Fig. 1c). Furthermore, no signal in the
typical hydride region of its 1H NMR spectrum can be found.
In the absence of a nitrogen in the bis-phosphane bridge, the
only remaining basic sites are the bdt sulfur ligands. Although
suggested by DFT calculations to be a potential intermediate
in catalysis,20 it is only the second time that S-protonation is
experimentally observed in [FeFe] H2ase models. Interesting to
note are the similarities between the two systems. S-Protonation
was first observed in a Fe2(CO)4(PMe3)2 system which con-
tains a strongly electron withdrawing azadithiolate (adt)
ligand.21 It thus seems that electron deficient Fe2 sites are a
general prerequisite for this unusual reactivity. Protonation of
bdt-based S-ligands has furthermore been described almost
protonation of the Fe–Fe site and the formation of
+
[2HFe]+.12,18 The 1H NMR spectrum of [2HFe
]
shows a
doublet of doublet at À12.3 ppm (J = 7.5, 22.5 Hz, FeHFe)
and a triplet at À10.2 ppm (J = 20 Hz, FeHFe). These two
hydride signals correlate with an AB system and a singlet in
the 31P NMR as demonstrated in a 2-dimensional 1H–31P
NMR correlation experiment (Fig. 2). There are thus two
hydride isomers of [2HFe]+ that differ in the orientation of the
phosphane ligands (ba,ba and ap,ba). These results also prove
that protonation of the Fe–Fe bond is still feasible under these
20 years ago by Sellmann et al. for [FeII(bdt)2]2À 22
The
.
observed shift of the nCO IR-frequencies upon S-protonation
of 3 (63 cmÀ1) lies between that observed for S-protonation of
the above mentioned [(adt)Fe2(CO)4(PMe3)2] (130 cmÀ1 21
)
and that of a S-ethylated [(pdt)Fe2(CO)4(PMe3)2] (40 cmÀ1).23
The 31P NMR spectrum of [3HS]+ shows one singlet at 9.6 ppm,
suggesting a rapid proton exchange between the two S ligands.
Prontonation of the bdt-S greatly weakens the Fe–S bond
which results in a low stability of [3HS]+ and rapid decompo-
sition under ambient conditions.
Since the electronic content of the Cl2bdtFe2(CO)4(Ph2PR)2
cores in 1–3 are identical, their different protonation behavior
has to be explained by their second coordination spheres
(Scheme 1). In the presence of a N-base in the bis-phosphane
bridge in 1 and 2, protonation occurs at the most basic site of
the complexes, i.e. at the N in complex 1 and the Fe–Fe bond
in complex 2. The basicity of the Fe–Fe bond in 1–3 thus has
to lie between that of the two N-containing bis-phosphane
ligands in 1 and 2. Since complex 2 is not protonated at the
bdt-S, the basicity of this site has to be lower than that of the
Fig. 1 Carbonyl region of the IR spectra obtained for solutions
(3 mM) of 1 (a), 2 (b) and 3 (c) in CH2Cl2, before (dotted line) and
after the addition of HBF4ÁOEt2 (solid line).
ꢀc
This journal is The Royal Society of Chemistry 2010
5776 | Chem. Commun., 2010, 46, 5775–5777

View Online
thereby providing higher-energy pathways that may look
unfavorable from a purely thermodynamic viewpoint.
Financial support from the Swedish Research Council (VR),
the Swedish Energy Agency, the Knut and Alice Wallenberg
Foundation, EU (FP7 Energy 212508 ‘‘SOLAR-H2’’), and
COST 0802 PhoSciNet is gratefully acknowledged.
Notes and references
1 J. W. Peters, W. N. Lanzilotta, B. J. Lemon and L. C. Seefeldt,
Science, 1998, 282, 1853–1858.
2 Y. Nicolet, C. Piras, P. Legrand, E. C. Hatchikian and
J. C. Fontecilla-Camps, Structure (London), 1999, 7, 13–23.
3 C. Tard and C. J. Pickett, Chem. Rev., 2009, 109, 2245–2274.
4 F. Gloaguen and T. B. Rauchfuss, Chem. Soc. Rev., 2009, 38,
100–108.
5 C. Tard, X. Liu, S. K. Ibrahim, M. Bruschi, L. D. Gioia,
S. C. Davies, X. Yang, L.-S. Wang, G. Sawers and C. J. Pickett,
Nature, 2005, 433, 610–613.
6 G. A. N. Felton, C. A. Mebi, B. J. Petro, A. K. Vannucci,
D. H. Evans, R. S. Glass and D. L. Lichtenberger,
J. Organomet. Chem., 2009, 694, 2681–2699.
7 R. Lomoth and S. Ott, Dalton Trans., 2009, 9952–9959.
8 D. Streich, Y. Astuti, M. Orlandi, L. Schwartz, R. Lomoth,
¨
L. Hammarstrom and S. Ott, Chem.–Eur. J., 2010, 16, 60–63.
9 P. Zhang, M. Wang, Y. Na, X. Li, Y. Jiang and L. Sun, Dalton
Trans., 2009, 1204–1206.
10 L. Schwartz, G. Eilers, L. Eriksson, A. Gogoll, R. Lomoth and
S. Ott, Chem. Commun., 2006, 520–522.
11 G. Eilers, L. Schwartz, M. Stein, G. Zampella, L. D. Gioia, S. Ott
and R. Lomoth, Chem.–Eur. J., 2007, 13, 7075–7084.
12 S. Ezzaher, J.-F. Capon, F. Gloaguen, F. Y. Petillon,
P. Schollhammer, J. Talarmin and N. Kervarec, Inorg. Chem.,
2009, 48, 2–4.
13 B. E. Barton, M. T. Olsen and T. B. Rauchfuss, J. Am. Chem. Soc.,
2008, 130, 16834–16835.
´
14 S. Ezzaher, J.-F. Capon, F. Gloaguen, F. Y. Petillon,
Scheme 1 Complexes 1–3 and their different protonation products.
Fe–Fe bond. Complex 3 which lacks the N-base possesses
sufficient electron density at the Fe2 site to allow for proto-
nation is however protonated at the bdt-S instead. Complex 3
thus does not follow the order of protonation site preferences
that is deduced from the protonation behavior of 1 and 2, and
[3HS]+ has to be assigned as a kinetic protonation product.
This assignment in turn implies that complex 2 contains a
structural feature that allows the system to find its thermo-
dynamic minimum, i.e. the Fe–Fe site. The N in the bis-
phosphane of 2 is the only difference compared to 3 and
thus has to be made responsible for assisting protonation of
the Fe–Fe bond, presumably by shuttling protons from the
surrounding medium to the metal site. This behavior is in
accordance to that described for related diiron12,13 and nickel24
complexes.
P. Schollhammer, J. Talarmin, R. Pichon and N. Kervarec, Inorg.
Chem., 2007, 46, 3426–3428.
15 In water, pKa (CF3CH2N(CH3)2) = 4.75 (E. R. Bissell and
M. Finger, J. Org. Chem., 1959, 24, 1256), and pKa (N(CH3)3) =
9.76 (H. K. Hall Jr., J. Am. Chem. Soc., 1957, 79, 5441).
16 L. Schwartz, P. S. Singh, L. Eriksson, R. Lomoth and S. Ott, C. R.
Chim., 2008, 11, 875–889.
17 R. K. Harris and R. A. Spragg, J. Chem. Soc. B, 1968, 684–691.
18 N. Wang, M. Wang, T. Zhang, P. Li, J. Liu and L. Sun, Chem.
Commun., 2008, 5800–5802.
19 L. Schwartz, L. Eriksson, R. Lomoth, F. Teixidor, C. Vinas and
S. Ott, Dalton Trans., 2008, 2379–2381.
20 M. Bruschi, P. Fantucci and L. D. Gioia, Inorg. Chem., 2002, 41,
1421–1429.
21 W. Dong, M. Wang, X. Liu, K. Jin, G. Li, F. Wang and L. Sun,
Chem. Commun., 2006, 305–307.
In this work, we could demonstrate that protons can be
directed to three different sites of a Fe2 complex by careful
design of the second coordination sphere. Since perfect control
of the second coordination sphere can be assumed in the
enzyme, it is interesting to hypothesize if similar kinetic
products may be intermediates in the enzymatic catalytic cycle,
22 D. Sellmann, M. Geck and M. Moll, J. Am. Chem. Soc., 1991, 113,
5259–5264.
23 X. Zhao, C.-Y. Chiang, M. L. Miller, M. P. Rampersad and
M. Y. Darensbourg, J. Am. Chem. Soc., 2003, 125, 518–524.
24 A. D. Wilson, R. K. Shoemaker, A. Miedaner, J. T. Muckerman,
D. L. DuBois and M. R. DuBois, Proc. Natl. Acad. Sci. U. S. A.,
2007, 104, 6951–6956.
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 5775–5777 | 5777