A Structural and Functional Model for the 1-Aminocyclopropane-1-carboxylic Acid Oxidase

Article abstract of DOI:10.1002/anie.201502529

The hitherto most realistic low-molecular-weight analogue for the 1-aminocyclopropane-1-carboxylic acid oxidase (ACCO) is reported. The ACCOs 2-His-1-carboxylate iron(II) active site was mimicked by a TpFe moiety, to which the natural substrate ACC could be bound. The resulting complex [Tp^Me,PhFeACC] (1), according to X-ray diffraction analysis performed for the nickel analogue, represents an excellent structural model, featuring ACC coordinated in a bidentate fashion - as proposed for the enzymatic substrate complex - as well as a vacant coordination site that forms the basis for the first successful replication also of the ACCO function: 1 is the first known ACC complex that reacts with O₂ to produce ethylene. As a FeOOH species had been suggested as intermediate in the catalytic cycle, H₂O₂ was tested as the oxidant, too, and indeed evolution of ethylene proceeded even more rapidly to give 65 % yield.

Full text of DOI:10.1002/anie.201502529

Angewandte
Chemie
International Edition: DOI: 10.1002/anie.201502529
German Edition: DOI: 10.1002/ange.201502529
Oxidase Mimetics
A Structural and Functional Model for the 1-Aminocyclopropane-1-
carboxylic Acid Oxidase**
Madleen Sallmann, Fabio Oldenburg, Beatrice Braun, Marius RØglier, A. Jalila Simaan,* and
Christian Limberg*
Abstract: The hitherto most realistic low-molecular-weight
analogue for the 1-aminocyclopropane-1-carboxylic acid ox-
idase (ACCO) is reported. The ACCOs 2-His-1-carboxylate
iron(II) active site was mimicked by a TpFe moiety, to which
the natural substrate ACC could be bound. The resulting
complex [TpMe,PhFeACC] (1), according to X-ray diffraction
analysis performed for the nickel analogue, represents an
excellent structural model, featuring ACC coordinated in
a bidentate fashion—as proposed for the enzymatic substrate
complex—as well as a vacant coordination site that forms the
basis for the first successful replication also of the ACCO
function: 1 is the first known ACC complex that reacts with O2
to produce ethylene. As a FeOOH species had been suggested
as intermediate in the catalytic cycle, H2O2 was tested as the
oxidant, too, and indeed evolution of ethylene proceeded even
more rapidly to give 65% yield.
also requires CO2 (or bicarbonate ions) to develop its
reactivity.
The crystal structure of the ACCO from Petunia hybrida
has been determined in 2004 by Schofield and co-workers.[3] It
belongs to the family of non-heme iron enzymes featuring
a structural motif termed the 2-His-1-carboxylate facial triad:
The active center contains an iron(II) ion coordinated by two
histidine moieties and one aspartate residue in a facial
arrangement, leaving the other three sites of the iron ion
vacant for the binding of ACC and potentially O2. The
mechanism, by which the complex conversion shown in
Scheme 1 is realized, is still discussed controversially, but it is
generally assumed that the oxidation of ACC into ethylene
follows a radical mechanism including two successive single-
electron oxidation steps.[4] Spectroscopic studies have shown
that in the first step of the reaction the substrate ACC binds to
the FeII ion in a bidentate mode most likely followed by O2
activation to generate an FeIII-superoxide intermediate.[5] For
the events after O2 binding several suggestions have been
E
thylene is a gaseous hormone for plants that induces fruit
ripening.[1] It is produced from the amino acid 1-amino-
cyclopropane-1-carboxylic acid, through oxidation with O2 as
the oxidant and ascorbate as the coreductant, mediated by the
enzyme 1-aminocyclopropane-1-carboxylic acid oxidase
(ACCO).[2] CO2 and HCN are generated concomitantly
(Scheme 1). For not well-understood reasons, the enzyme
made that all involve an FeIV O intermediate but differ in the
=
sequence of the electron transfer steps with ascorbate and in
the nature of the intermediates responsible for the successive
oxidations of ACC. Rocklin et al. have proposed that binding
of ACC and O2 is followed by one electron reduction to
provide an iron(III)-hydroperoxide intermediate that per-
=
forms the first ACC oxidation step leading to an FeIV
O
species responsible for the second oxidation step.[6] One
electron from ascorbate would then complete the catalytic
cycle. Alternatively, Mirica et al. have proposed an ascorbate-
dependent formation of the FeIV O intermediate that would
=
act as the first oxidizing species.[7]
Scheme 1. The last step of the biosynthesis of ethylene.
Many questions thus remain unanswered regarding the
interaction mode with the substrate, the role of the different
cofactors/co-substrates (ascorbic acid, dioxygen, and carbon
dioxide) and the catalytic mechanism.
[*] Dr. M. Sallmann, B. Sc. F. Oldenburg, Dr. B. Braun,
Prof. Dr. C. Limberg
Humboldt-Universität zu Berlin, Institut für Chemie
Brook-Taylor-Strasse 2, 12489 Berlin (Germany)
E-mail: Christian.limberg@chemie.hu-berlin.de
Homepage: http://www.chemie.hu-berlin.de/aglimberg
Bioinorganic model studies can provide valuable infor-
mation, both with regard to structural issues and require-
ments for reactivity. However, there are hardly any com-
pounds known which may be regarded as models of ACCO,
applying the minimum requirement that they contain ACC
and that they display substantial ACCO-like activities.[8,9]
One iron complex has been reported, which is dinuclear,
though, contains iron(III) ions and two bridging ACC
ligands.[8] Whereas iron(II) complexes have long remained
unknown, mononuclear 1:1 metal–ACC complexes have been
accessed for copper(II).[9] An FeII–ACC complex has very
recently been described, but it can hardly be considered as
a functional replicate of the ACCO: the iron center is
coordinated octahedrally by six donor atoms, and hence
Dr. M. RØglier, Dr. A. J. Simaan
Aix Marseille UniversitØ, Centrale Marseille, CNRS, iSm2 UMR
7313, 13397, Marseille, (France)
E-mail: jalila.simaan@univ-amu.fr
Homepage: http://ism2.univ-amu.fr/pages-bleues/index2.htm
[**] We are grateful to the Humboldt-Universität zu Berlin for financial
support. We also thank the Cluster of Excellence “Unifying Concepts
in Catalysis” funded by the DFG for helpful discussions. The EU-
COST Network (CM1305) on spin state reactivity is acknowledged
for support.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201502529.
Angew. Chem. Int. Ed. 2015, 54, 12325 –12328
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
12325

.
Angewandte
Communications
oxidants have no access to the iron center. This explains that
no reactivity with O2 was observed and that the yield of
ethylene produced in reactions with H2O2 was only 7% higher
as compared to the blank experiments.[10]
characterize the complex. The spectrum of 1 shows a doublet
typical of high-spin iron(II) species with an isomeric shift d =
1.0906 mmsÀ1 and an electronic quadrupole splitting of DEq =
2.7 mmsÀ1.
To mimic mononuclear non-heme iron dioxygenases
based on a 3-His structural motif, we have in recent years
successfully employed the tris(pyrazolyl)borato (Tp) ligand
system,[11] which in the past has also been employed to
simulate representatives based on the 2-His-1-carboxylate
structural motif.[12] It thus appeared a promising choice also
for the development of replicates of the ACCO. One problem
that has emerged in the past with regard to the modeling of
this enzyme is the tendency of ACC to bridge two metal
centers, as both polynuclearity and ACC in bridging coordi-
nation modes rather limit the model character of the resulting
complexes (see above). Hence, we chose phenyl residues at
the 3-position of the pyrazole donors within Tp to create
a shielding reaction pocket approaching the situation in the
enzyme. A suitable starting material was therefore the
complex [TpMe,PhFeCl][13] (TpMe,Phe = 3-Phenyl-5-methylhy-
drido-trispyrazol-1-ylborato), which after dissolution in
dichloromethane was reacted with the potassium salt of the
To obtain structural information we have prepared
a corresponding nickel complex [TpMe,PhNiACC] (2) in the
same way setting out from [TpMe,PhNiBr] (see SI). After work-
up of the blue-green reaction solution a light green solid could
be isolated, which showed an almost identical IR spectrum as
compared to the one displayed by 1, with respect to band
shape, pattern, and intensities (Figure 1). Since this indicated
that the structure of 2 is rather similar to the one of 1, too,
a structural investigation of 2 was of high interest.
Compound 2.thf could be crystallized through overlaying
a THF solution with hexane, and Figure 2 shows the
amino
acid
1-amino-cyclopropane-1-carboxylic
acid
(Scheme 2).
Figure 1. Comparison of the IR spectra of 1 (blue) and 2 (red) as
recorded from KBr discs.
Scheme 2. Synthesis of [TpMe,PhFeACC] (1).
After work-up a light yellow solid was isolated, which was
1
characterized by elemental analysis, IR, and H NMR spec-
troscopy (see the Supporting Information, SI). The IR
spectrum showed characteristic bands for the NH2 stretching
vibrations as well as for the carboxylate absorptions, and the
n(BH) band, which is typical for Tp complexes and sensitive
to the environment, appeared at 2549 cmÀ1. While all
analytical and spectroscopic data pointed to the envisaged
complex [TpMe,PhFeACC] (1), all attempts to crystallize this
compound led to crystals of [TpMe,Ph2Fe].[20] To clarify whether
the latter complex represents an impurity that has favorable
crystallization properties, or whether it forms through ligand
exchange during attempts to crystallize 1, it was prepared
1
independently (see SI) and characterized by H NMR spec-
troscopy. Comparison showed that [TpMe,Ph2Fe] was not
present in freshly prepared samples of 1 nor in 7 h old ones;
however, after storing the solutions for 5 days, [TpMe,Ph2Fe]
could be detected in significant amounts. This clearly showed
that 1 is stable enough in solution for reactivity studies but
does not readily crystallize, and crystallization can neither be
achieved by allotting additional time as this leads to ligand
exchange. Mçssbauer spectroscopy was used to further
Figure 2. Molecular structure of 2.thf (all hydrogen atoms have been
omitted for clarity). Selected bond lengths [Š] and angles [8]: Ni–O1
1.9585(11), Ni–N1 2.0859(14), Ni–N2 2.0993(13), Ni–N5 2.0698(13),
Ni–N7 2.0510(13); O1-Ni-N1 82.19(5), N2-Ni-N5 87.29(5), N2-Ni-N7
90.03(5), N5-Ni N7 95.17(5), O1-Ni-N2 95.59(5), O1-Ni-N5 146.53(5),
O1-Ni-N7 118.12(5), N1-Ni-N2 175.66(5), N1-Ni-N5 92.51(5), N1-Ni-
N7 94.31(5).
12326 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 12325 –12328

Angewandte
Chemie
molecular structure determined by single-crystal X-ray anal-
ysis.
showed that even at RT after O2 addition the main changes
(general increase of absorption in the range of 300–500 nm)
occur within 0–5 min (see SI). After this time the spectrum
remains nearly constant displaying only discrete changes,
while ethylene yield increases within ca. 2 h. Hence, there is
probably an initial fast step (perhaps only partially as part of
an equilibrium), followed by a much slower reaction produc-
ing ethylene. At RT ethylene yield reaches a plateau at ca.
7%. This suggests that a decomposition reaction competes
with the productive path, which, however, can be accelerated
to a significant extent by warming to 708C (resulting in 17%
yield).
As expected the ACC ligand binds as a chelating ligand.
The nickel(II) center is thus coordinated by the three N atoms
of the TpMe,Ph ligand as well as by the amine and carboxylate
functions of the O-deprotonated ACC creating a ligand
sphere that is in-between trigonal bipyramidal and square
À
pyramidal (t = 0.48). Although in previous models the C C
bond of the cyclopropyl unit appeared shortened,[9] such an
effect was not noticeable in case of 2: The C3–C4 bond length
in 2 (1.496(3) Š) was almost identical with the one found for
free ACC[14] (1.490–1.497 Š).
Based on the striking resemblance of the IR spectra of
1 and 2 a corresponding structure is suggested also for 1,
which thus features one free coordination site for the
potential binding and activation of O2 or alternative oxi-
dants.[18] Accordingly, reactivity studies were carried out. It
has been suggested that a prerequisite for O2 reactivity of
For many O2 activating non-heme iron enzymes and
models thereof the initial O2 binding step with formation of
iron(III) superoxide intermediates has been found to be
endergonic in theoretical investigations. The subsequent steps
are usually exothermic but in case of low-molecular-weight
analogues often characterized by substantial barriers, which—
together with the endergonic O2 binding step—decrease the
reaction rates. If enzymes are considered, which—as the
ACCO—in parallel to O2 consume electrons to reach
iron(II) complexes is E1/2(FeIII/II) < À0.1 V versus Fc+/Fc.[16]
A
cyclovoltammetry (CV) investigation of 1 dissolved in
dichloromethane indicated a reversible redox event, but the
oxidation and reduction peaks were separated by 410 mV
indicating reversible structural changes upon oxidation/
reduction (see SI). The oxidation peak occurred at 0.03 V,
which appeared not negative enough for O2 reactivity. On the
other hand reactivity trends observed in previous work for
TpFe-based Dke1 models could not be rationalized on the
basis of redox potentials,[17] and indeed 1 was found to react
with O2.
Complex 1 was dissolved in DMF and added to O2-
saturated DMF placed in hermetically sealed vials. Analysis
of the headspace gas by GC revealed the presence of
ethylene, and the conversion yield reached a maximum of
ca. 17% after a few minutes at 708C. Importantly, no ethylene
was detected when ACC alone (in form of its NBu4+ salt) was
placed in aerated DMF for one hour indicating that the
observed activity is not due to free ACC (potentially released
from the complex) in solution. Furthermore, ethylene pro-
duction was found to be three times lower (conversion yields
of ca. 6%) when Fe(ClO4)2·xH2O in combination with
(NBu4)ACC was employed instead of complex 1 under the
same conditions. Notably, the nickel complex 2 does not react
with O2, which suggests an initiation of the reaction by
binding of O2 at the iron center in the first step.
Under single turn-over conditions and in the absence of
ascorbate, Rocklin et al. found that only 0.35 mol of ethylene
per mol of ACCO are formed, and hence they proposed that
the electrons needed for catalysis are provided by a fraction of
the initial ACCO enzyme.[6] Also in case of our experiments it
is conceivable that the required electrons are provided by
a fraction of complex 1, therefore limiting the reaction yield
as observed for the enzymatic system. Attempts to provide
electrons by adding different reductants (ascorbate or ben-
zoine) remained unsuccessful, though.
To obtain information about the fate of the residual
skeleton of 1 we have performed GC/MS studies with the gas
phase after the reaction, which revealed CO2 as a further
product, as one should expect.[19] Monitoring the UV/Vis
spectrum of a DMF solution of 1 with time at RT and at 708C
substrate oxidation and feature peroxide or FeIV O inter-
=
mediates, this problem can be circumvented by employing
reduced forms of dioxygen, that is, for instance H2O2, or O-
atom transfer reagents. Significant ethylene production was
observed when complex 1 was placed in the presence of PhIO
or mCPBA (meta-choroperoxybenzoic acid). However, the
latter reagents significantly oxidize ACC already in the
absence of the iron complex (ca. 75% and 35% yield within
2–3 h respectively), so that their employment does not
provide any information (see SI). Hence, different concen-
trations of hydrogen peroxide were tested, and interestingly,
for a 0.5 mm solution of complex 1 the ACC conversion into
ethylene reaches 65% yield employing 10–20 mm of hydrogen
peroxide after 20 min. Under the same conditions, ACC alone
was hardly oxidized to produce ethylene (less than 2%
conversion yield) confirming that the observed activity is not
due to released ACC in solution. Also the reactivity of 2
toward H2O2 was found to be rather limited (less than 2%
conversion yield), emphasizing the importance of the nature
of the metal center. Finally, ethylene production from
a 0.5 mm solution of Fe(ClO4)2·xH2O/(NBu4)ACC in DMF
in the presence of 10–20 mm H2O2 was found to be six times
lower (conversion yields of ca. 10–12%) than that observed
for complex 1, which demonstrates the importance also of the
co-ligand to control the reactivity.
Altogether, these results imply that the observed ethylene
production from complex 1 after reaction with O2 or hydro-
gen peroxide crucially depends on the suitability of the
TpMe,PhFe moiety to mimic the 2-His-1-carboxylate iron core
within the ACCO. It has been rather well established that
H2O2 can react with FeII or FeIII complexes to provide the
[15]
corresponding, reactive FeIII OOH intermediates.
Reac-
À
tion of H2O2 with complex 1 can thus be expected to first
result in the oxidation of the FeII to FeIII and then in the
formation of the FeIII OOH intermediate. In the presence of
À
dioxygen, the formation of the FeIII OOC intermediate
À
followed by one-electron reduction by a further equivalent
of 1 and protonation (for instance through residual water in
Angew. Chem. Int. Ed. 2015, 54, 12325 –12328
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12327

.
Angewandte
Communications
DMF) could lead to a FeIII OOH intermediate as well. Our
À
[12] a) W. H. Armstrong, S. J. Lippard, J. Am. Chem. Soc. 1983, 105,
4837 – 4838; b) W. H. Armstrong, A. Spool, G. C. Papaefthy-
miou, R. B. Frankel, S. J. Lippard, J. Am. Chem. Soc. 1984, 106,
3653 – 3667; c) E. H. Ha, R. Y. N. Ho, J. F. Kisiel, J. S. Valentine,
Inorg. Chem. 1995, 34, 2265 – 2266; d) S. Hikichi, T. Ogihara, K.
Fujisawa, N. Kitajima, M. Akita, Y. Moro-oka, Inorg. Chem.
1997, 36, 4539 – 4547; e) E. L. Hegg, R. Y. N. Ho, L. Que, Jr., J.
Am. Chem. Soc. 1999, 121, 1972 – 1973; f) M. P. Mehn, K.
Fujisawa, E. L Hegg, L. Que, Jr., J. Am. Chem. Soc. 2003, 125,
7828 – 7842; g) T. K. Paine, H. Zheng, L. Que, Jr., Inorg. Chem.
2005, 44, 474 – 476; h) N. Burzlaff, Angew. Chem. Int. Ed. 2009,
48, 5580 – 5582; Angew. Chem. 2009, 121, 5688 – 5690; i) A.
Mukherjee, M. Martinho, E. L. Bominaar, E. Münck, L.
Que, Jr., Angew. Chem. Int. Ed. 2009, 48, 1780 – 1783; Angew.
Chem. 2009, 121, 1812 – 1815.
results thus suggest that a FeIII OOH species is indeed
À
involved in the catalytic cycle of the ACCO and that its
formation probably precedes the oxidation of the bound
substrate, in agreement with the two proposed mechanisms
for the enzymatic system.
In conclusion we have described here the first FeII-based
low-molecular-weight analogue which faithfully mimics the
structure of the ACCO active site and at the same time also
simulates the function. Indeed, this FeII complex in contact
with O2 shows an oxidase activity (17%), which is remarkable
as the enzymatic one is only ca. 35% under single turn-over
conditions. Future studies, including various alternative oxy-
genation reagents, electron and proton donors as well as
kinetic and theoretical calculations, will focus on the further
elucidation of the mechanism of this biomimetic oxidation.
[13] T. Tietz, C. Limberg, R. Stçßer, B. Ziemer, Chem. Eur. J. 2011,
17, 10010 – 10020.
[14] a) M. C. Pirrung, J. Org. Chem. 1987, 52, 4179 – 4184; b) G. Valle,
M. Crisma, C. Toniolo, E. M. Holt, M. Tamura, J. Bland, C. H.
Stammer, Int. J. Pept. Protein Res. 1989, 34, 56 – 65.
Keywords: aminocyclopropane-1-carboxylic acid ·
[15] a) J. J. Girerd, F. Banse, A. J. Simaan, Struct. Bonding (Berlin)
2000, 97, 145 – 177; b) M. Costas, M. P. Mehn, M. P. Jensen, L.
Que, Jr., Chem. Rev. 2004, 104, 939 – 986; c) M. R. Bukowski, P.
Comba, A. Lienke, C. Limberg, C. L. de Laorden, R. Mas-
BallestØ, M. Merz, L. Que, Jr., Angew. Chem. Int. Ed. 2006, 45,
3446 – 3449; Angew. Chem. 2006, 118, 3524 – 3528; d) M. R.
Bukowski, P. Comba, C. Limberg, M. Merz, L. Que, Jr., T.
Wistuba, Angew. Chem. Int. Ed. 2004, 43, 1283 – 1287; Angew.
Chem. 2004, 116, 1303 – 1307; e) Y. M. Kim, K.-B. Cho, J. Cho, B.
Wang, C. Li, S. Shaik, W. Nam, J. Am. Chem. Soc. 2013, 135,
8838 – 8841; f) L. V. Liu, S. Hong, J. Cho, W. Nam, E. I. Solomon,
J. Am. Chem. Soc. 2013, 135, 3286 – 3299; g) J. Cho, S. Jeon, S. A.
Wilson, L. V. Liu, E. A. Kang, J. J. Braymer, M. H. Lim, B.
Hedman, K. O. Hodgson, J. S. Valentine, E. I. Solomon, W. Nam,
Nature 2011, 478, 502 – 505.
[16] Y. M. Badiei, M. A. Siegler, D. P. Goldberg, J. Am. Chem. Soc.
2011, 133, 1274 – 1277.
[17] H. Park, M. M. Bittner, J. S. Baus, S. V. Lindeman, A. T. Fiedler,
Inorg. Chem. 2012, 51, 10279 – 10289.
[18] For the proposal/detection of iron(III) superoxide intermediates,
compare: a) Y.-M. Lee, S. Hong, Y. Morimoto, W. Shin, S.
Fukuzumi, W. Nam, J. Am. Chem. Soc. 2010, 132, 10668 – 10670;
b) S. Hong, K. D. Sutherlin, J. Park, E. Kwon, M. A. Siegler, E. I.
Solomon, W. Nam, Nat. Commun. 2014, 5, 5440 – 5447; F.
Oddon, Y. Chiba, J. Nakazawa, T. Ohta, T. Ogura, S. Hikichi,
Angew. Chem. Int. Ed. 2015, 54, 7336 – 7339; Angew. Chem.
2015, 127, 7444-7447.
enzyme models · iron · oxidase · oxygen
How to cite: Angew. Chem. Int. Ed. 2015, 54, 12325–12328
Angew. Chem. 2015, 127, 12501–12505
[1] F. Abeles, Ethylene in Plant Biology, Academic Press, New York,
1992.
[2] D. O. Adams, S. F. Yang, Proc. Natl. Acad. Sci. USA 1979, 76,
170 – 174.
[3] Z. Zhang, J.-S. Ren, I. J. Clifton, C. J. Schofield, Chem. Biol.
2004, 11, 1383 – 1394.
[4] M. C. Pirrung, Acc. Chem. Res. 1999, 32, 711 – 718.
[5] a) A. M. Rocklin, D. L. Tierney, V. Kofman, N. M. W. Brun-
huber, B. M. Hoffman, R. E. Christoffersen, N. O. Reich, J. D.
Lipscomb, L. Que, Jr., Proc. Natl. Acad. Sci. USA 1999, 96,
7905 – 7909; b) D. L. Tierney, A. M. Rocklin, J. D. Lipscomb, L.
Que, Jr., B. M. Hoffman, J. Am. Chem. Soc. 2005, 127, 7005 –
7013; c) J. Zhou, A. M. Rocklin, J. D. Lipscomb, L. Que, Jr., E. I.
Solomon, J. Am. Chem. Soc. 2002, 124, 4602 – 4609.
[6] A. M. Rocklin, K. Kato, H. W. Liu, L. Que, Jr., J. D. Lipscomb, J.
Biol. Inorg. Chem. 2004, 9, 171 – 182.
[7] L. M. Mirica, J. P. Klinman, Proc. Natl. Acad. Sci. USA 2008, 105,
1814 – 1819.
[8] W. Ghattas, Z. Serhan, N. El Bakkali-Taheri, M. RØglier, M.
Kodera, Y. Hitomi, A. J. Simaan, Inorg. Chem. 2009, 48, 3910 –
3912.
[9] W. Ghattas, C. Gaudin, M. Giorgi, A. Rockenbauer, A. J.
Simaan, M. RØglier, Chem. Commun. 2006, 1027 – 1029.
[10] Y. Roux, W. Ghattas, F. Avenier, R. Guillot, A. J. Simaan, J.-P.
Mahy, Dalton Trans. 2015, DOI: 10.1039/c5dt00347d.
[11] a) I. Siewert, C. Limberg, Angew. Chem. Int. Ed. 2008, 47, 7953 –
7956; Angew. Chem. 2008, 120, 8071 – 8074; b) M. Sallmann, I.
Siewert, L. Fohlmeister, C. Limberg, C. Knispel, Angew. Chem.
Int. Ed. 2012, 51, 2234 – 2237; Angew. Chem. 2012, 124, 2277 –
2280; c) M. Sallmann, S. Kumar, P. Chernev, J. Nehrkorn, A.
Schnegg, D. Kumar, H. Dau, C. Limberg, S. P. de Visser, Chem.
Eur. J. 2015, 21, 7470 – 7479; d) M. Sallmann, B. Braun, C.
Limberg, Chem. Commun. 2015, 51, 6785 – 6787.
[19] Cyanide—a further product of the natural process—could not be
detected through characteristic bands in IR spectra recorded for
the solution after the reaction or the solid product mixture after
removal of all volatiles.
[20] A. L. Rheingold, R. L. Ostrander, B. S. Haggerty, S. Trofimenko,
Inorg. Chem. 1994, 33, 3666 – 3676.
Received: March 18, 2015
Revised: May 22, 2015
Published online: July 17, 2015
12328 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 12325 –12328