The conversion of nickel-bound CO into an acetyl thioester: Organometallic chemistry relevant to the acetyl coenzyme A synthase active site

Article abstract of DOI:10.1002/anie.201105281

When three become one: Within one nickel-based model system, the three reactants CO, MeI, and PhSH have been assembled to yield an acetyl thioester. The reactivity is of relevance for the functioning of the acetyl coenzyme A synthase active site and provides insights into possible binding sequences. Copyright

Full text of DOI:10.1002/anie.201105281

DOI: 10.1002/anie.201105281
Enzyme Models
The Conversion of Nickel-Bound CO into an Acetyl Thioester:
Organometallic Chemistry Relevant to the Acetyl Coenzyme A
Synthase Active Site**
Bettina Horn, Christian Limberg,* Christian Herwig, and Stefan Mebs
The carbon monoxide dehydrogenase/acetyl coenzyme A
synthase (CODH/ACS) is a bifunctional enzyme which
couples the reduction of CO2 to CO [Eq. (1)] to the
conversion of CO and coenzyme A (HSCoA) to give acetyl
coenzyme A (CH3C(O)SCoA), which is a central metabolite
[Eq (2); CH3-CoIIIFeSP = methylated corrinoid iron sulfur
protein].[1–3]
produces the active state that is capable of creating the acetyl
moiety from CO and a methyl species.[12] For this process, two
II
ꢀ
scenarios are conceivable: 1) a H3C Ni unit is formed first,
ꢀ
followed by CO insertion into the Ni C bond, or 2) CO gets
coordinated at the reduced Ni center in the initial step, and
subsequently the cobalamin transfers CH3+ onto the resulting
ꢀ
Ni CO species. A proposed mechanism favoring the second
option is depicted in Scheme 1.[4]
CODH
CO2 þ 2 Hþ þ 2 eꢀ
CO þ H O
ð1Þ
ð2Þ
G
H
2
ACS H
CH3-CoIIIFeSPþCO þ HSCoA
G
CH3CðOÞSCoA þ CoIFeSP þ Hþ
After CO is produced within the CODH subunit of the
enzyme, it is guided through a channel to the active site of the
ACS subunit (A cluster), which consists of a dinuclear Ni core
linked via a cysteinate bridge to an 4Fe-4S cluster (Scheme 1,
upper left).[3,4] At this site, an acetyl thioester, namely acetyl
coenzyme A CH3C(O)SR (R = CoA), is generated from the
constituents of CO, a thiol (HSR), and a methyl moiety, which
is provided by a corrinoid iron–sulfur protein (ultimately a
methylcobal(III)amine).[1–5] Although protein single-crystal
X-ray diffraction investigations have significantly enhanced
the understanding of the active site structure in recent
years,[4,6,7] the mechanism by which the conversion of
Equation (2) is realized at the bimetallic core has been a
matter of intense debate for over a decade, especially with
respect to the binding sequence of the three substrates.[1,4,8–10]
In the meantime, a consensus has been reached that HSCoA
is the last substrate to bind.[10,11] It is also widely accepted that
the resting state contains both Ni centers in the oxidation
state +II, and that Nid mainly serves the stabilization of the
structure. Reduction of Nip, either to NiI or even to Ni0,
Scheme 1. The ACS catalytic cycle as proposed by Fontecilla-Camps
et al.[4]
However, there are also other suggestions based on
route 1,[12,13] and today one of the prevailing views assumes
a “random binding”[10,14] of either methyl or CO, in which
case 1 and 2 would proceed simultaneously.
Bioinorganic model compounds can reveal important
insights concerning the functioning of enzyme active sites:
they provide information as to which kind of reaction
pathways are plausible from the point of view of molecular
chemistry considering the existence or elusiveness of prece-
dent cases from that area. The CODH/ACS has been a
stimulus for a variety of studies on structural[2,9,15,16] and
functional[2,5,16,17] low-molecular-weight analogues in the past.
Aiming at functional mimics, synthetic efforts have focused
on mononuclear as well as dinuclear Ni complexes that
simulate potential elementary steps of the ACS subunit.[18] In
this context, methyl groups have been successfully transferred
from cobalt to NiI, and there is also evidence for the oxidative
addition of carbon electrophiles to Ni0.[16b] Various examples
[*] Dipl.-Chem. B. Horn, Prof. Dr. C. Limberg, Dr. C. Herwig,
Dr. S. Mebs
Humboldt-Universitꢀt zu Berlin, Institut fꢁr Chemie
Brook-Taylor-Straße 2, 12489 Berlin (Germany)
E-mail: christian.limberg@chemie.hu-berlin.de
Homepage: http://www.chemie.hu-berlin.de/aglimberg
[**] We are grateful to the Cluster of Excellence “Unifying Concepts in
Catalysis” funded by the Deutsche Forschungsgemeinschaft (DFG),
the Fonds der Chemischen Industrie, and the Humboldt-Universitꢀt
zu Berlin for financial support. We gratefully acknowledge valuable
discussions with Prof. Dr. P. S. Pregosin, Prof. Dr. A. Togni, and
Prof. Dr. H. Dobbek. We thank R. Metzinger and Dr. B. Braun for
crystal-structure analyses.
from organonickel chemistry confirm the assumption that CO
[17]
ꢀ
is able to insert into Ni C bonds, and addition of CO to
dinuclear thiolate-coordinated nickel compounds containing
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201105281.
a methylnickel(II) moiety has led to thioester formation.[5,19]
Angew. Chem. Int. Ed. 2011, 50, 12621 –12625
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
12621

Communications
A recent breakthrough has been the successive reaction of a
early stages (first 60 frames) of the measurements allowed for
a solution revealing the molecular structure of 1.[25] Complex
1 shows static disorder for the acetyl ligand (the two different
conformations are shown in the Supporting Information), and
only the major conformer is depicted in Figure 1. Although
the disorder limits a discussion concerning the binding
parameters of the nickel–acetyl moiety, it becomes obvious
that, as expected on basis of the NMR analysis, the acetyl
ligand is bound to the LtBuNiII unit in a side-on fashion, which
has been rarely observed for nickel complexes.[24]
NiIINi0 precursor, first with methylcobaloxime and then with a
thiolate to give a NiIINiII complex that subsequently reacts
with CO to give an CH3C(O)SCoA analogue.[5a] While these
results have contributed evidence that the initial formation of
ꢀ
a H3C Ni unit followed by CO insertion is principally
feasible, to date there is no precedence from the field of
organonickel chemistry supporting route 2, although corre-
sponding investigations have been encouraged[5a] and are
certainly motivated by biochemical results.[14,20]
Herein we provide, to the best of our knowledge, a unique
ꢀ
ꢀ
example of a Ni CO complex that is converted first into a Ni
C(O)CH3 compound by a formal transfer of CH3 and then
+
into a thioester through treatment with a thiol.
Recently, we reported that treatment of a reduced b-
diketiminato-ligated nickel dinitrogen complex[21a] with CO
leads to a complex of the type K2[{LtBuNi(CO)}2] (I, LtBu
=
[HC{C(tBu)NC6H3(iPr)2}2]ꢀ) containing two nickel(0) centers
coordinating one CO ligand each.[21b] Compound I is ther-
mally stable but highly sensitive towards air, and it is also
accessible by reduction of the nickel(I) compound
[LtBuNi(CO)] with KC8.[21b] Considering its low coordination
ꢀ
number and oxidation state, the Ni CO unit in I seemed ideal
for a modeling study concerning the reaction sequence 2,
which, as outlined above, was considered as one mechanistic
option for the construction of the acetyl moiety within the
active site of the CODH/ACS. Methyl iodide was chosen as
the methylating reagent simulating the cobalamin cofactor for
the CH3+ transfer. Correspondingly, compound I was treated
with MeI, and work-up led to a diamagnetic product. A
1H NMR investigation of a C6D6 solution indicated the
presence of an acetyl unit (d(C(O)CH3) = 0.45 ppm), thus
suggesting a nickel(II)–acetyl complex [LtBuNiC(O)CH3] (1)
as the product (Scheme 2). The 13C NMR spectrum of 1
contained a low-field C(O)CH3 resonance at 243.8 ppm,
which is characteristic for h2-acetyl ligands.[22] It is comparable
to that found for the nickel(II)–acetyl complex [(dppp)Ni{h2-
C(O)CH3}]+ (co-existing with [(dppp)Ni(CO)(CH3)]+ as part
of an equilibrium in solution; d(C(O)CH3) = 242.9 ppm)[23]
and also to the resonance of a nickel(II)–acyl complex
Figure 1. Molecular structure of 1. Ellipsoids are set at 50% proba-
bility; hydrogen atoms are omitted for clarity. Because of disorder in
the position of the acetyl ligand, the bond lengths and angles of the
atoms C1, C2, and O1 are not discussed. Selected bond lengths [ꢂ]
and angles [8]: N1–Ni1 1.881(4), N2–Ni1 1.908(4); N1-Ni1-N2
99.61(15).
To elucidate the binding situation, DFT calculations
(Gaussian09, B3LYP/6-31G*)[28] were carried out: Setting
out with the structure shown in Figure 1, geometry optimiza-
tion was performed, which indeed converged with a structure
containing the acetyl entity in a side-on mode, thus supporting
the result of the crystal structure investigation. An natural
bond order (NBO) analysis showed that the bonding can be
rationalized by an interaction of a negatively charged ligand
L
tBu, a NiII ion, and a negatively charged acetyl ligand
[(dtbpe)Ni{h2-C(O)CH2tBu}]+
(d(C(O)CH2tBu) =
(C(O)CH3)ꢀ. Although the NBO analysis assigns a bond
between Ni and the acetyl C atom, the corresponding bond
orbital comprises only 31% of a Ni atomic orbital and 69% of
a C atomic orbital, justifying a treatment of this bond as being
highly polarized and allocating the bond electrons to the
carbon atom of the acetyl group as described. Remarkably,
the stabilizing energy caused by donor–acceptor interactions
between empty valence orbitals of the Ni atom and both the
free electron pairs of the acetyl oxygen atom and p electrons
of the CO bond sums to 320 kJmolꢀ1. This is comparable to
the stabilizing energy obtained by interactions of the lone pair
at each N atom with empty Ni valence orbitals (190 and
350 kJmolꢀ1 for the two different N atoms) in the same
molecule. The small Ni-C-O angle might therefore be due to
the stabilization of the molecule by interaction of oxygen
valence electrons with empty Ni orbitals. Furthermore, the
structure of the hypothetical molecule [LtBuNi(CO)(CH3)] (2)
was optimized and its energy compared with that of 1 to
establish whether 2 could in principle be an intermediate on
248.3 ppm).[24]
Crystallization from various solvents led to single crystals
suitable for X-ray diffraction, which, however, were very
sensitive to the X-ray irradiation and always decomposed
during the data collection, so that the latter could never be
completed. Combination of three data sets belonging to the
Scheme 2. Synthesis of 1.
12622 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 12621 –12625

the way to 1 by analogy to what has been suggested by
Fontecilla-Camps et al. (Scheme 1).[4] Its free energy at room
temperature turned out to be only 39.7 kJmolꢀ1 higher than
that of 1, that is, 2 may well be an intermediate of the
conversion depicted in Scheme 2. However, to definitely
confirm this or to rule it out, respectively, the entire energy
profile would have to be calculated, which is difficult starting
from I and CH3I.
internal standard. The reaction turned out to be quite fast:
Within a few minutes, the set of signals for 1 disappeared
completely, and the yield of the thioester amounted to 40%,
while the yield of the protonated ligand was almost quanti-
tative. Based on this observation, we conclude that the
precipitate formed concomitantly beside elemental nickel
may also contain insoluble nickel(II) thiolates.[17b,30] Interest-
ingly, the reaction of 1 with thiolates KSR (R = Et, Ph) in
CD2Cl2 and [D8]THF, respectively, proved to be extremely
slow (after 4 weeks, the reaction mixture contained only
traces of thioester). This observation argues against a
mechanism involving a direct nucleophilic attack of the
thiol HSPh at the bound acetyl ligand, and at the same time
necessitates thoughts on the role of the proton. We can
exclude that the first step of the reaction consists of a ligand
protonation and that the thioester formation then takes place
at nickel species, which do not contain LtBu: The reaction of 1
with equimolar amounts of the acid 2,6-lutidinium triflate to
give HLtBu proceeds slowly, and after completion addition of
HSPh does not lead to S-phenylthioester formation.[31] Hence,
a more likely scenario is a prebinding of the thiol followed by
a concerted proton shift, umpolung, and reductive elimina-
tion.[13]
It was of interest to assign the absorption of the Ni-h2-
acetyl moiety in the IR spectrum. A band at 1584 cmꢀ1 was
observed in a region where acetyl moieties with a side-on
binding mode commonly absorb.[22,24] To confirm its assign-
ment to n(CO), the 13C isotopologue of 1 was synthesized by
employing 13CO for the synthesis of the precursor I. Indeed,
its IR spectrum differed from that of the 12C isotopologue
only in the band for the n(CO) stretching mode, which was
shifted to 1545 cmꢀ1. This observed isotope shift (Dn(12CO–
13CO) = 39 cmꢀ1) is in agreement with that predicted by
theory (n(12CO) = 1628 cmꢀ1, n(13CO) = 1590 cmꢀ1, Dn(12CO–
13CO) = 38 cmꢀ1).[28]
[21]
ꢀ
Unlike other b-diketiminato Ni X complexes,
com-
pound 1 is quite stable to water, as proved by 1H NMR
spectroscopy (even after 3 days in contact with water, the
major part of the sample still remained undecomposed).
Having created an acetyl group from CO and CH3+ in a way
that may be of relevance to the ACS reactivity, the question
Scheme 4 assembles the essence of the findings made up
to this point. Setting out from a nickel(0) carbonyl compound
(I) generated from a nickel(0) precursor (II) and CO, we have
ꢀ
naturally arose as to whether the next step, C S bond
formation, could also be simulated within the same system, in
analogy to a precedence setting out with thiols.[17a,b] There-
fore, an NMR tube experiment was performed in which a
solution of 10 mg of 1 in 0.6 mL [D8]THF was treated with
1.5 equivalents of thiophenol serving as a HSCoA analogue.
During the course of the reaction, a dark brown suspension
was formed. The NMR tube containing the reaction mixture
was thus centrifuged before recording spectra to separate the
1
solution from a dark brown solid. The H NMR spectrum
obtained indicated complete conversion of the nickel–acetyl
complex 1 to give phenyl thioacetate PhSC(O)CH3 (assigned
through comparison with authentic samples) and the proton-
ated b-diketiminato ligand HLtBu (Scheme 3). Apart from
that, only signals from unreacted thiol were observed.
The conversion of 1 and HSPh in THF was also
investigated by liquid IR spectroscopy: The IR spectrum of
the reaction mixture featured a strong absorption band at
1713 cmꢀ1 caused by the stretching vibration of the carbonyl
group in PhSC(O)CH3.[29] The yield based on the conversion
of [LtBuNi{h2-C(O)CH3}], 1, and HSPh into the thioester was
determined by 1H NMR spectroscopy using DMF as the
Scheme 4. A reaction sequence that mimics the acetyl coenzyme A
synthase function. Reactions include carbonylation of a nickel(0)
precursor (II), methylation of the resulting carbonyl complex (I) to give
an acetyl compound (1), and thioester formation after reaction with a
thiol.
been able to prepare a nickel acetyl complex (1), which in turn
reacts with thiophenol, resulting in HLtBu and thioester
formation. Thus, this reaction sequence combines mimics of
the ACS substrates to give an analogue of its product, and this
raises the question in how far the ligand LtBu resembles the
N2NidS2 metalloligand at Nip beyond the obvious bidenticity.
Apparently, the electronic situations resulting from the
N2NidS2/SCys versus LtBu ligation are quite similar: Treating
the enzyme in the fully oxidized resting state with CO leads to
I
ꢀ
a reduced Ni CO state that shows a n(CO) absorption at
1995 cmꢀ1 in the IR spectrum,[2,9,32] while the NiI pendant of I,
[LtBuNi(CO)], absorbs at 2020 cmꢀ1.[21b] The fact that the ACS
reactivity reported herein involves a Ni0 center might indicate
Scheme 3. Reaction of 1 with thiophenol.
Angew. Chem. Int. Ed. 2011, 50, 12621 –12625
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12623

Communications
[17] For example see: a) P. Stavropoulos, M. Carriꢁ, M. C. Muetter-
ties, R. H. Holm, J. Am. Chem. Soc. 1990, 112, 5385 – 5387; b) P.
Stavropoulos, M. C. Muetterties, M. Carriꢁ, R. H. Holm, J. Am.
Chem. Soc. 1991, 113, 8485 – 8492; c) M. V. Rampersad, S. P.
Jeffery, J. H. Reibenspies, C. G. Ortiz, D. J. Darensbourg, M. Y.
Darensbourg, Angew. Chem. 2005, 117, 1243 – 1246; Angew.
Chem. Int. Ed. 2005, 44, 1217 – 1220; d) D. Sellmann, D. Hꢂus-
singer, F. Knoch, M. Moll, J. Am. Chem. Soc. 1996, 118, 5368 –
5374; e) G. C. Tucci, R. H. Holm, J. Am. Chem. Soc. 1995, 117,
6489 – 6496; f) P. T. Matsunaga, G. L. Hillhouse, Angew. Chem.
1994, 106, 1841 – 1843; Angew. Chem. Int. Ed. Engl. 1994, 33,
1748 – 1749; g) H.-F. Klein, Angew. Chem. 1973, 85, 403; Angew.
Chem. Int. Ed. Engl. 1973, 12, 402; h) Y.-M. Hsiao, S. S.
Chojnacki, P. Hinton, J. H. Reibenspies, M. Y. Darensbourg,
Organometallics 1993, 12, 870 – 875.
that the enzymatic system also intermediately receives a
further electron from its environment to develop its function
as suggested by Fontecilla-Camps et al.[4] and Hall et al.,[13]
although there is still no precedence for Ni0 in a biological
system.
In summary, the results show that an ACS mechanism in
which the Nip site in the A cluster takes up a CO molecule
prior to the transfer of the methyl group is plausible on the
grounds of molecular organometallic chemistry. Insofar they
provide support for a random binding of the ACS substrates,
if not for an exclusive binding of CO in the initial step. Our
model contains a Ni0 center, that is, nickel in an oxidation
state conceivable for an intermediate reduced form of the
ACS, and thus may hint to a NiIINi0 core there. Future studies
will address the relevance of the oxidation state for this kind
of reactivity.
[18] a) C. G. Riordan in Comprehensive coordination chemistry II,
Vol. 8 (Eds: J. A. McCleverty, T. J. Meyer), Elsevier, Oxford,
2003, pp. 677 – 713; b) D. J. Evans, Coord. Chem. Rev. 2005, 249,
1582 – 1595.
[19] W. G. Dougherty, K. Rangan, M. J. OꢃHagan, G. P. A. Yap, C. G.
Riordan, J. Am. Chem. Soc. 2008, 130, 13510 – 13511.
[20] S. J. George, J. Seravalli, S. W. Ragsdale, J. Am. Chem. Soc. 2005,
127, 13500 – 13501.
Received: July 27, 2011
Revised: September 2, 2011
Published online: November 7, 2011
[21] a) S. Pfirrmann, C. Limberg, C. Herwig, R. Stçßer, B. Ziemer,
Angew. Chem. 2009, 121, 3407 – 3411; Angew. Chem. Int. Ed.
2009, 48, 3357 – 3361; b) B. Horn, S. Pfirrmann, C. Limberg, C.
Herwig, B. Braun, S. Mebs, R. Metzinger, Z. Anorg. Allg. Chem.
2011, 637, 1169 – 1174; for the LMe analogues, see: S. Pfirrmann,
C. Limberg, C. Herwig, C. Knispel, B. Braun, E. Bill, R. Stçsser,
J. Am. Chem. Soc. 2010, 132, 13684 – 13691; N. A. Eckert, A.
Dinescu, T. R. Cundari, P. L. Holland, Inorg. Chem. 2005, 44,
7702 – 7704.
[22] L. D. Durfee, I. P. Rothwell, Chem. Rev. 1988, 88, 1059 – 1079.
[23] C. S. Shultz, J. M. DeSimone, M. Brookhart, J. Am. Chem. Soc.
2001, 123, 9172 – 9173.
[24] J. J. Curley, K. D. Kitiachvili, R. Waterman, G. L. Hillhouse,
Organometallics 2009, 28, 2568 – 2571.
Keywords: acetyl coenzyme A synthase · b-diketiminates ·
coordination chemistry · enzyme models · nickel
.
[1] S. W. Ragsdale, M. Kumar, Chem. Rev. 1996, 96, 2515 – 2539.
[2] T. C. Harrop, P. K. Mascharak, Coord. Chem. Rev. 2005, 249,
3007 – 3024.
[3] S. W. Ragsdale, Chem. Rev. 2006, 106, 3317 – 3337.
[4] C. Darnault, A. Volbeda, E. J. Kim, P. Legrand, X. Vernꢀde,
P. A. Lindahl, J. C. Fontecilla-Camps, Nat. Struct. Biol. 2003, 10,
271 – 279.
[5] a) M. Ito, M. Kotera, T. Matsumoto, K. Tatsumi, Proc. Natl.
Acad. Sci. USA 2009, 106, 11862 – 11866; b) T. Matsumoto, M.
Ito, M. Kotera, K. Tatsumi, Dalton Trans. 2010, 39, 2995 – 2997.
[6] V. Svetlitchnyi, H. Dobbek, W. Meyer-Klaucke, T. Meins, B.
Thiele, P. Rçmer, R. Huber, O. Meyer, Proc. Natl. Acad. Sci.
USA 2004, 101, 446 – 451.
[7] a) T. I. Doukov, T. M. Iverson, J. Seravalli, S. W. Ragsdale, C. L.
Drennan, Science 2002, 298, 567 – 572; b) C. L. Drennan, T. I.
Doukov, S. W. Ragsdale, J. Biol. Inorg. Chem. 2004, 9, 511 – 515.
[8] a) J. Seravalli, W. Gu, A. Tam, E. Strauss, T. P. Begley, S. P.
Cramer, S. W. Ragsdale, Proc. Natl. Acad. Sci. USA 2003, 100,
3689 – 3694; b) S. Gencic, D. A. Grahame, J. Biol. Chem. 2003,
278, 6101 – 6110.
[25] Crystallographic data for 1: C37H56N2NiO, 0.3 ꢄ 0.3 ꢄ 0.3 mm3,
Mr = 603.55, monoclinic, P21/n, a = 9.442(2), b = 23.723(5), c =
15.210(3) ꢅ, b = 98.75(3)8, V= 3367.3(12) ꢅ3, Z = 4, T=
100(2) K, F000 = 1312, m = 0.606 mmꢀ1, V= 3.32 – 27.80, collected
reflections 25837, unique reflections 7773 (Rint = 0.1426), GoF =
1.050, R1 = 0.1110, wR2 = 0.2496, largest difference peak and
hole 1.121/ꢀ1.827 eꢅꢀ3. The structures were solved by direct
methods (SHELXS-97)[26] and refined versus F2 (SHELXL-
97)[27] with anisotropic temperature factors for all non-hydrogen
atoms. All hydrogen atoms were added geometrically and
refined by using a riding model. CCDC 831011 (1) contains the
supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[26] G. M. Sheldrick, SHELXS-97, Program for Crystal Structure
Solution, University of Gçttingen, 1997.
[9] E. L. Hegg, Acc. Chem. Res. 2004, 37, 775 – 783.
[10] J. Seravalli, S. W. Ragsdale, J. Biol. Chem. 2008, 283, 8384 – 8394.
[11] A. Volbeda, C. Darnault, X. Tan, P. A. Lindahl, J. C. Fontecilla-
Camps, Biochemistry 2009, 48, 7916 – 7926.
[12] P. A. Lindahl, J. Biol. Inorg. Chem. 2004, 9, 516 – 524.
[13] C. E. Webster, M. Y. Darensbourg, P. A. Lindahl, M. B. Hall, J.
Am. Chem. Soc. 2004, 126, 3410 – 3411.
[27] G. M. Sheldrick, SHELXL-97, Program for Crystal Structure
Refinement, University of Gçttingen, 1997.
[28] M. J. Frisch et al., Gaussian09, RevisionA.02; Gaussian, Inc.,
Wallingford CT, 2009. See the Supporting Information for full
details.
[29] a) P. R. Olivato, M. L. T. Hui, A. Rodrigues, R. Ruiz Filho, R.
Rittner, J. Zukerman-Schpector, G. Distefano, M. Dal Colle, J.
Mol. Struct. 2003, 645, 259 – 271; b) Z. Xi, W. Hao, P. Wang, M.
Cai, Molecules 2009, 14, 3528 – 3537.
[30] For example, see: a) R. G. Hayter, F. S. Humiec, J. Inorg. Nucl.
Chem. 1964, 26, 807 – 810; b) P. Woodward, L. F. Dahl, E. W.
Abel, B. C. Crosse, J. Am. Chem. Soc. 1965, 87, 5251 – 5253;
c) S. A. Ivanov, M. A. Kozee, W. A. Merrill, S. Agarwal, L. F.
Dahl, J. Chem. Soc. Dalton Trans. 2002, 4105 – 4115.
[14] G. Bender, S. W. Ragsdale, Biochemistry 2011, 50, 276 – 286.
[15] a) R. C. Linck, C. W. Spahn, T. B. Rauchfuss, S. R. Wilson, J.
Am. Chem. Soc. 2003, 125, 8700 – 8701; b) P. V. Rao, S. Bhaduri,
J. Jiang, R. H. Holm, Inorg. Chem. 2004, 43, 5833 – 5849; c) T. C.
Harrop, M. M. Olmstead, P. K. Mascharak, Inorg. Chem. 2006,
45, 3424 – 3436; d) T. C. Harrop, M. M. Olmstead, P. K.
Mascharak, J. Am. Chem. Soc. 2004, 126, 14714 – 14715; e) V.
Mathrubootham, J. Thomas, R. Staples, J. McCraken, J. Shearer,
E. L. Hegg, Inorg. Chem. 2010, 49, 5393 – 5406.
[16] a) R. Krishnan, C. G. Riordan, J. Am. Chem. Soc. 2004, 126,
4484 – 4485; b) C. G. Riordan, J. Biol. Inorg. Chem. 2004, 9, 542 –
549.
12624 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 12621 –12625

[31] The system 1/2,6-lutidinium triflate/KSPh leads of course to
thioester formation, as HSPh is produced in the first step; the
addition of water instead of 2,6-lutidinium triflate as a proton
source does not trigger the reaction. See the Supporting
Information for full details.
[32] a) M. Kumar, S. W. Ragsdale, J. Am. Chem. Soc. 1992, 114,
8713 – 8715; b) J. Chen, S. Huang, J. Seravalli, H. Gutzman, Jr.,
D. J. Swartz, S. W. Ragsdale, K. A. Bagley, Biochemistry 2003,
42, 14822 – 14830.
Angew. Chem. Int. Ed. 2011, 50, 12621 –12625
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12625