A genetically encodable ligand for transfer hydrogenation

Article abstract of DOI:10.1002/ejoc.201300340

Simple tripeptides are shown here to be versatile ligands for iridium-catalyzed transfer hydrogenations affording large acceleration effects. A water-soluble iridium complex with Gly-Gly-Phe, for example, catalyzes the reduction of diverse ketones, aldehydes, and imines by formate with turnover frequencies rivaling or outperforming those of established ligand systems. Regioselective reduction of coenzyme NAD⁺ to NADH illustrates the potential utility of this system for biotechnological applications. Because peptides are genetically encodable, they represent an attractive class of foldamer ligands for creating artificial metalloenzymes.

Full text of DOI:10.1002/ejoc.201300340

SHORT COMMUNICATION
DOI: 10.1002/ejoc.201300340
A Genetically Encodable Ligand for Transfer Hydrogenation
Clemens Mayer[a] and Donald Hilvert*[a]
Keywords: Homogeneous catalysis / Water chemistry / Peptides / Metalloenzymes / Ligand design
Simple tripeptides are shown here to be versatile ligands for
gand systems. Regioselective reduction of coenzyme NAD+
iridium-catalyzed transfer hydrogenations affording large ac- to NADH illustrates the potential utility of this system for bio-
celeration effects. A water-soluble iridium complex with Gly-
Gly-Phe, for example, catalyzes the reduction of diverse
ketones, aldehydes, and imines by formate with turnover fre-
quencies rivaling or outperforming those of established li-
technological applications. Because peptides are genetically
encodable, they represent an attractive class of foldamer li-
gands for creating artificial metalloenzymes.
An alternative approach toward artificial metallo-
enzymes involves anchoring an organometallic complex to a
protein scaffold.[2a,2b] Catalysts for alkene hydrogenation,[7]
transfer hydrogenation of ketones and imines,[8] olefin
metathesis,[9] and other activities[10] have been created by
combining the intrinsic reactivity of a metal cofactor with
the specificity of a protein in this way. Nevertheless, such
hybrid catalysts are typically much less effective than their
natural counterparts with respect to turnover number and
selectivity. These shortcomings can largely be attributed to
trade-offs made in the design process. The organometallic
complex is usually attached to the scaffold protein through
a flexible linker that is equipped either with a reactive chem-
ical group for site-selective protein modification (covalent
anchoring) or a small ligand such as biotin with unusually
high affinity for a specific protein pocket (supramolecular
anchoring).[2a,2b] The flexibility of the linker may limit accu-
rate placement of the metal center in a defined environ-
ment, hampering effective preorganization of substrate and
catalyst. Unlike natural metalloenzymes, residues at the
active site of these hybrid constructs rarely interact pro-
ductively with the cofactor (see recent work on C–H acti-
vation for a notable exception[10b]), making fine-tuning of
metal properties difficult. As a consequence, the unbound
metal complex usually displays higher activity than its
bound form. In principle, artificial metalloenzymes can be
improved by evolutionary approaches, but the optimization
process is tedious and often unsuccessful.[11]
Introduction
Nature utilizes metal ions bound to protein scaffolds to
effect a wide range of demanding chemical transformations
with enviable ease.[1] The large rate accelerations and exact-
ing selectivities achieved by such systems have inspired re-
searchers to explore the feasibility of creating artificial
metalloenzymes for reactions lacking biological counter-
parts.[2]
Both the peptidic backbone and amino acid side chains
are useful for coordinating metal ions. Over the last two
decades, computational search algorithms have been devel-
oped to design metal binding sites in structurally charac-
terized proteins that take advantage of such interactions.[3]
These programs have been successfully used to arrange con-
stellations of residues in geometries appropriate for binding
a range of metals ions, including zinc, copper, and iron.[4]
De novo proteins have also been equipped with unique
metal-binding capabilities and, as a result, appreciable hy-
drolytic and redox activities.[5] The ability to incorporate
non-natural, metal-chelating amino acids site-selectively
into proteins is likely to extend these strategies in the fu-
ture.[6] Nonetheless, the design of functional metalloen-
zymes from scratch remains challenging. Avoiding unpro-
ductive interactions of the metal ions with residues else-
where in the protein is one challenge. Tailoring the second
and third coordination environments around the metal ion
to fine-tune reactivity is another. Engineering a substrate
binding site in proximity to the metal ion is a further hur-
dle.[2c]
Short peptidic motifs with the ability to bind metal ions
directly and thereby enhance catalytic prowess could obvi-
ate some of the problems associated with linkers in semisyn-
thetic metalloenzymes.[12] As modular and genetically enco-
dable elements, peptides can be easily embedded as tags in
larger foldameric structures, potentially providing greater
control over metal ion placement and reactivity. Transfer
hydrogenations, important synthetic transformations in
which unsaturated substrates are reduced by hydrogen do-
[a] Laboratory of Organic Chemistry, ETH Zürich,
8093 Zürich, Switzerland
Fax: +41-44-632-1486
E-mail: hilvert@org.chem.ethz.ch
Homepage: www.protein.ethz.ch
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201300340.
Eur. J. Org. Chem. 2013, 3427–3431
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3427

C. Mayer, D. Hilvert
SHORT COMMUNICATION
nors other than molecular hydrogen,[13] are attractive reac- ance (Table 1, entry 9; Supporting Information, Fig-
tions for exploring this possibility. A myriad of transition- ure S4A). Scrambling the tripeptide sequence also dimin-
metal complexes with amino acids, proline derivatives, and ished catalytic efficiency.
pseudopeptides have been reported to promote this reaction
Table 1. Transfer hydrogenation of acetophenone with 1, 2, and 3
under mild conditions in water.[14] In this work, we show
with peptidic ligands.
that short peptide sequences also function as versatile li-
Entry Complex[a]
Ligand
TOF[b]
(h–1)
gands for iridium-based transfer-hydrogenation catalysts,
providing large ligand-induced accelerations. The biocom-
patibility of such complexes could be advantageous for a
range of biotechnological applications.
1
2
3
4
5
6
[Ru(p-cym)(H2O)3]SO4 (1)
none
none
none
Gly-Gly-Phe
Gly-Gly-Phe
Gly-Gly-Phe
Gly-Gly-Phe
Ac-Gly-Gly-Phe
Ͻ1
Ͻ1
Ͻ1
2
[RhCp*(H2O)3]SO4 (2)
[IrCp*(H2O)3]SO4 (3)
[Ru(p-cym)(H2O)3]SO4 (1)
[RhCp*(H2O)3]SO4 (2)
[IrCp*(H2O)3]SO4 (3)
[IrCp*(H2O)3]SO4 (3)
[IrCp*(H2O)3]SO4 (3)
[IrCp*(H2O)3]SO4 (3)
[IrCp*(H2O)3]SO4 (3)
[IrCp*(H2O)3]SO4 (3)
[IrCp*(H2O)3]SO4 (3)
1
142
200
Ͻ1
7[c]
8[d]
9[d]
10
11
12[d]
Results and Discussion
Gly-Gly-Phe-NH2 149
Gly-Phe-Gly
Phe-Gly-Gly
Phe-Gly-Phe
3
92
193
Despite the well-established capabilities of peptides to
bind metals,[15] aqueous transfer hydrogenations have not
been systematically studied with unmodified peptide li-
gands. Here, the tripeptide Gly-Gly-Phe was chosen as a
simple test ligand. It contains a free amino group at its N
terminus, a carboxylate at its C terminus, and an aromatic
chromophore to facilitate purification and concentration
determination. Our model reaction was transfer hydrogena-
tion of acetophenone by formate catalyzed by the readily
accessible, water-soluble, piano-stool complexes [Ru(p-
cym)(H2O)3]SO4 (1), [RhCp*(H2O)3]SO4 (2), and
[IrCp*(H2O)3]SO4 (3) (p-cym = p-cymene, Cp* = penta-
methylcyclopentadiene).
Experiments were carried out in water at neutral pH and
at 40 °C with 50 μm catalyst, 10 mm substrate, and 50 mm
sodium formate (Scheme 1). In the absence of peptide, the
catalysts displayed no detectable activity (Table 1, entries 1–
3; Supporting Information, Figure S1A). Addition of
1.2 equiv. of Gly-Gly-Phe to ruthenium and rhodium com-
plexes 1 and 2 marginally increased the reaction rates
(Table 1, entries 4 and 5). In contrast, adding the tripeptide
to iridium complex 3 gave quantitative conversion and more
than 140 turnovers per hour (Table 1, entry 6; Supporting
Information, Figure S1B). Even higher turnover numbers
were achieved by increasing the formate concentration.
Thus, doubling the number of reducing equivalents af-
forded 200 turnovers per hour at pH 8 (Table 1, entry 7;
Supporting Information, Figure S2). At pH 7 and a sub-
strate/catalyst ratio of 2000:1, the reduction of acetophen-
one proceeded to Ͼ96% conversion after 16 h, corre-
sponding to Ͼ1900 total turnovers (Supporting Infor-
mation, Figure S3). Acetylation of the N terminus of the
peptide ligand completely abrogated activity (Table 1, en-
try 8), whereas replacement of the C-terminal carboxylate
with an amide did not significantly affect catalyst perform-
[a] Acetophenone/formate/complex = 200:1000:1; 1.2 equiv. of li-
gand used in the reaction. [b] Turnover frequencies: determined af-
ter 60 min by reverse-phase HPLC. [c] Carried out at pH 8 with
10 equiv. of formate for 30 min. [d] Determined after 30 min.
Thus, Phe-Gly-Gly exhibits 35% lower activity than Gly-
Gly-Phe, whereas Gly-Phe-Gly affords basal levels only
(Table 1, entries 10 and 11; Supporting Information, Fig-
ure S4B).
Taken together, these results suggest that tripeptides may
act as bifunctional Noyori-type catalysts, binding to 3
through the N-terminal amine and adjacent amide group
(Scheme 2).[13a] Analogous complexes have been previously
characterized by X-ray crystallography but never tested as
catalysts for transfer hydrogenation.[16] The first and third
amino acids are tolerant to substitution, but steric consider-
ations probably dictate the requirement for glycine as the
second amino acid. Additional bulk at this position could
hinder either formation of the peptide–iridium adduct or
substrate binding. Consistent with this model, Phe-Gly-Phe
still gives rise to large rate accelerations, even outper-
forming the original Gly-Gly-Phe ligand (Table 1, entry 12;
Supporting Information, Figure S4A). Even more active
catalysts might be generated by further sequence optimiza-
tion.
Scheme 2. Proposed transition state for the transfer hydrogenation
of ketones by 3 with peptidic ligands. R1 = H, benzyl, AA3 = Gly,
Phe.
Transfer hydrogenation of acetophenone is pH depend-
ent. In the presence of Gly-Gly-Phe and 50 mm sodium for-
mate, the pH–rate profile for the iridium-catalyzed reaction
is bell shaped with a broad maximum around pH 8
(Scheme 3). Turnover frequency (TOF) exceeds 100 h–1 be-
Scheme 1. Transfer hydrogenation of acetophenone in water with
peptidic ligands.
3428
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 3427–3431

A Genetically Encodable Ligand for Transfer Hydrogenation
tween pH 6 and 9. As proposed by Xiao,[17] protonation of clic imine 1-methyl-3,4-dihydroisoquinoline proved to be a
the ligand, the active metal hydride, or formate decreases poor substrate for the peptide–iridium complex under our
activity. Above pH 9, replacement of a catalyst-bound water standard conditions (Ͻ1 h–1). Lowering the pH to 6.25,
molecule by hydroxide prevents formate binding, thereby however, afforded a large increase in activity (197 h–1), pre-
shutting down the catalytic cycle.
sumably due to protonation of the imine. The resulting rate
is comparable to that for reduction of acetophenone under
our standard conditions.
Table 2. Turnover frequencies (TOFs) for transfer hydrogenation of
different substrates catalyzed by 3 in the presence of Gly-Gly-Phe.[a]
Entry
Substrate
TOF Rel. act.[b]
(h–1)
1
2
3
4
5
6
7
8
acetophenone
200
391
361
366
198
154
138
98
1
4Ј-chloroacetophenone
2Ј-chloroacetophenone
2,2,2-trifluoroacetophenone
4Ј-methylacetophenone
4Ј-methoxyacetophenone
2Ј-methylacetophenone
4-phenyl-2-butanone
benzaldehyde
1.96
1.81
1.83
0.99
0.77
0.69
0.49
1.37
0.99
Scheme 3. pH–rate profile for the transfer hydrogenation of aceto-
phenone with the Gly-Gly-Phe peptide–iridium complex.
The ligand acceleration effect observed here is similar in
magnitude to that of highly active ligand complexes with
diamines, aminosulfonamides, and amino alcohols.[14,18]
Unlike many highly stereoselective transfer-hydrogenation
catalysts, however, the peptide–iridium catalysts generate
racemic products despite the presence of chiral phenylala-
nine(s) in the ligand. Embedding the catalytic tripeptide
motif into larger foldameric structures will therefore be nec-
essary to control substrate access to the metal center and
achieve enantioselective hydride transfer. Toward this end,
the catalyst must form and retain high activity in the pres-
ence of other protein functional groups. In fact, when
1 equiv. of 3 is added to a mixture of the tripeptide (60 μm)
and bovine serum albumin (BSA) (0.1 mgmL–1), a catalyti-
cally active complex is formed that converts substrate into
product with a TOF of 77 h–1 (Supporting Information,
Figure S5A). In the absence of tripeptide, no conversion is
detected. If the amount of BSA is increased to 0.5 mgmL–1,
the TOF drops to 11 h–1, indicative of competitive chelation
by the protein (Supporting Information, Figure S6A).[19]
Given that the BSA–iridium complex is inactive, though,
the concentration of iridium can simply be increased to
overcome this problem. For example, addition of 5 equiv.
of 3 to the mixture of tripeptide and 0.5 mgmL–1 BSA
affords a TOF of 88 h–1 (Supporting Information, Fig-
ure S6B). Such protein compatibility is unusual for many
transition-metal catalysts and supports the feasibility of
using this approach to construct artificial metallo-
enzymes.[20]
9
273
10[c]
1-methyl-3,4-dihydroisoquinoline 197
[a] Reaction conditions: substrate/formate/catalyst = 200:2000:1,
Gly-Gly-Phe (1.2 equiv.), pH 8.0, 40 °C, 600 rpm. [b] Relative ac-
tivity compared to acetophenone as substrate. [c] pH 6.25.
Regeneration of dihydronicotinamide adenine dinucleo-
tide (NADH), an essential coenzyme in biological redox re-
actions, from nicotinamide adenine dinucleotide (NAD+) is
a biotechnologically important reduction reaction.[22] Al-
though nicotinamide cofactors can be regenerated enzymat-
ically, nonenzymatic approaches involving organometallic
redox mediators represent increasingly attractive alterna-
tives.[23] The iridium complex with Gly-Gly-Phe efficiently
reduced NAD+ at pH 6.25 with a TOF of 225 h–1
(Scheme 4). The reaction proceeded regioselectively with a
Ͼ10:1 preference for the biologically relevant 1,4-dihydro-
nicotinamide isomer over the 1,2- and 1,6-isomers. The irid-
ium catalyst is thus more than twofold more efficient than
[Rh(bpy)H2O]SO4 (bpy = 2,2Ј-bipyridine), which has pre-
viously been used for the regeneration of NADH (TOF =
96 h–1); however, the rhodium catalyst is somewhat more
selective (20:1 in favor of 1,4-NADH; Supporting Infor-
mation, Figure S7). These results are particularly notable as
iridium catalysts often display significantly lower activities
than the corresponding rhodium and ruthenium complex-
es.[23a] The combination of high turnover frequencies and
good protein compatibility makes this system attractive for
multistep chemoenzymatic processes.
Broad substrate scope is another notable attribute of the
Gly-Gly-Phe peptide–iridium catalyst. Initial TOFs for a
variety of aldehydes, ketones, and imines are summarized
in Table 2. Benzaldehyde (Table 2, entry 9) and activated
acetophenone derivatives containing electron-withdrawing
aryl substituents or a trifluoromethyl group adjacent to the
carbonyl are particularly good substrates (Table 2, en-
tries 1–4), whereas derivatives containing electron-donating
groups and aliphatic ketones are converted less rapidly than
acetophenone itself (Table 2, entries 5–8). Although trans-
Scheme 4. Reduction of NAD+ by formate catalyzed by 3 in the
fer hydrogenation of imines is well documented,[21] the cy- presence of Gly-Gly-Phe.
Eur. J. Org. Chem. 2013, 3427–3431
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3429

C. Mayer, D. Hilvert
SHORT COMMUNICATION
ium catalyst. Then, 250 mm phosphate buffer (200 μL) at the appro-
priate pH containing 500 mm sodium formate and the substrate
Conclusions
Our results show that peptide–iridium catalysts, which (50 mm in water, 200 μL) were added. The reaction mixture was
allowed to react for 30 to 60 min at 40 °C while shaking before
stopping the reaction by cooling the samples to 4 °C. Aliquots were
periodically removed and analyzed by reverse-phase HPLC (5 μL
injection) to monitor reaction progress and estimate initial turnover
frequencies.
are readily formed under mild conditions in aqueous buffer,
efficiently catalyze transfer hydrogenation of diverse
ketones, aldehydes, and imines. Simple tripeptides such as
Gly-Gly-Phe significantly enhance the reactivity of a water
soluble d6-piano-stool iridium complex, providing turnover
frequencies that rival or surpass those of established high-
performance ligand systems.[14] They are thus interesting al-
ternatives to amino acids[24] and pseudopeptides[25] for
transfer hydrogenations under biological conditions.
Transfer Hydrogenation in the Presence of BSA: BSA (10 or 50 μL
of a 10 mgmL–1 stock solution in 250 mm sodium phosphate buffer,
pH 8) was mixed with the Gly-Gly-Phe tripeptide (60 μm). Follow-
ing addition of 3 (25 to 125 μL of a 2.0 mm stock solution in water)
and incubation at 40 °C for 10 min, the resulting mixture was used
for transfer hydrogenation as described for the standard procedure.
The peptides likely bind the iridium complex through
their free N-terminal amine and the adjacent amide to af-
ford bifunctional Noyori-type catalysts. Because the C-ter-
minal carboxylate is not required for activity, it should be
possible to append the tripeptides to the N terminus of vir-
tually any protein or foldameric structure as simple coordi-
nation tags to create artificial transfer hydrogenases. For-
mation of a productive complex, even in the presence of
excess competing protein functionality, augurs well for
transplantation of this catalytic motif into larger, more
complex structures. Although the peptides tested to date
confer no selectivity, placing the organometallic complex
within the chiral environment of a protein would be ex-
pected to enable enantioselective transformations, as ob-
served for biotinylated transfer-hydrogenation catalysts
bound to streptavidin.[8] In contrast to biotinylated systems,
though, a wide range of structures, tailored to specific ap-
plications, can be considered as scaffolds. Utilization of a
short peptide as opposed to a flexible linker to anchor the
iridium complex to a protein is also likely to provide greater
control over the placement of the catalytic center at the
active site as well as a means of fine-tuning affinity and
reactivity through systematic sequence variation. The de-
fined structure of such complexes may facilitate computa-
tional design efforts as well.
Turnover Experiments: The total turnover number for the catalytic
reduction of acetophenone was determined by treating 50 mm sub-
strate with 25 μm 3, 30 μm Gly-Gly-Phe, and 500 mm sodium for-
mate in 250 mm sodium phosphate buffer at pH 7 for 16 h.
Supporting Information (see footnote on the first page of this arti-
cle): Materials and methods, experimental details, representative
HPLC chromatograms, characterization data, and 1H NMR as well
as 13C NMR spectra for synthesized tripeptides.
Acknowledgments
The authors thank Professors Paul Pregosin and Antonio Togni
for helpful discussions. Generous support by the Schweizerischer
Nationalfonds (to D. H.) and Novartis Pharma AG for a Novartis
Graduate Fellowship (to C. M.) is gratefully acknowledged.
[1] A. Messerschmidt, Handbook of Metalloproteins, Wiley, Chi-
chester, 2001.
[2] a) T. Heinisch, T. R. Ward, Curr. Opin. Chem. Biol. 2010, 14,
184; b) F. Rosati, G. Roelfes, ChemCatChem 2010, 2, 916; c) Y.
Lu, N. Yeung, N. Sieracki, N. M. Marshall, Nature 2009, 460,
855; d) V. Nanda, R. L. Koder, Nat. Chem. 2010, 2, 15.
[3] a) N. D. Clarke, S. M. Yuan, Proteins 1995, 23, 256; b) H. Faz-
elinia, P. C. Cirino, C. D. Maranas, Protein Sci. 2009, 18, 180;
c) H. W. Hellinga, F. M. Richards, J. Mol. Biol. 1991, 222, 763.
[4] a) B. S. Der, D. R. Edwards, B. Kuhlman, Biochemistry 2012,
51, 3933; b) N. Yeung, Y. W. Lin, Y. G. Gao, X. Zhao, B. S.
Russell, L. Y. Lei, K. D. Miner, H. Robinson, Y. Lu, Nature
2009, 462, 1079; c) D. E. Benson, M. S. Wisz, H. W. Hellinga,
Proc. Natl. Acad. Sci. USA 2000, 97, 6292.
[5] a) J. Kaplan, W. F. DeGrado, Proc. Natl. Acad. Sci. USA 2004,
101, 11566; b) M. Faiella, C. Andreozzi, R. T. M. de Rosales,
V. Pavone, O. Maglio, F. Nastri, W. F. DeGrado, A. Lombardi,
Nat. Chem. Biol. 2009, 5, 882; c) M. L. Zastrow, A. F. A. Pea-
cock, J. A. Stuckey, V. L. Pecoraro, Nat. Chem. 2012, 4, 118.
The biocompatibility of the peptide–iridium complexes
and the low background activity observed in the absence of
peptide are important prerequisites for in vitro and in vivo
applications. As for biotin-streptavidin-based artificial
transfer hydrogenases,[8] it may be possible to combine these
organometallic complexes with natural biocatalysts to gen-
erate productive reaction cascades. The fact that tripeptides
are genetically encodable ligands also opens the possibility
of generating artificial metalloenzymes in vivo. Coupling [6] a) H. S. Lee, P. G. Schultz, J. Am. Chem. Soc. 2008, 130, 13194;
the activity of such species to the survival of a cell might
b) X. H. Liu, Y. Yu, C. Hu, W. Zhang, Y. Lu, J. Y. Wang, An-
gew. Chem. 2012, 124, 4388; Angew. Chem. Int. Ed. 2012, 51,
then enable genetic selection of highly active and selective
4312; c) C. C. Liu, P. G. Schultz, Annu. Rev. Biochem. 2010, 79,
transfer-hydrogenation catalysts.[26]
413.
[7] M. Skander, N. Humbert, J. Collot, J. Gradinaru, G. Klein, A.
Loosli, J. Sauser, A. Zocchi, F. Gilardoni, T. R. Ward, J. Am.
Chem. Soc. 2004, 126, 14411.
Experimental Section
[8] a) C. Letondor, N. Humbert, T. R. Ward, Proc. Natl. Acad.
Sci. USA 2005, 102, 4683; b) M. Dürrenberger, T. Heinisch,
Y. M. Wilson, T. Rossel, E. Nogueira, L. Knörr, A. Mutschler,
K. Kersten, M. J. Zimbron, J. Pierron, T. Schirmer, T. R. Ward,
Angew. Chem. 2011, 123, 3082; Angew. Chem. Int. Ed. 2011,
50, 3026.
Typical Procedure for Transfer Hydrogenation: Stock solutions of 3
(2.0 mm in water) and the corresponding peptide (2.4 mm in water)
were freshly prepared, and 25 μL of each was added to water
(550 μL) in a 1.5 mL screw-cap Eppendorf tube. The resulting solu-
tion was allowed to incubate in a thermo mixer at 40 °C for 10 min
while shaking (600 rpm) to facilitate formation of the peptide–irid-
[9] a) C. Mayer, D. G. Gillingham, T. R. Ward, D. Hilvert, Chem.
Commun. 2011, 47, 12068; b) C. Lo, M. R. Ringenberg, D.
3430
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 3427–3431

A Genetically Encodable Ligand for Transfer Hydrogenation
Gnandt, Y. Wilson, T. R. Ward, Chem. Commun. 2011, 47,
Y. M. Wilson, M. Dürrenberger, D. Ghislieri, E. Churakova, T.
Quinto, L. Knörr, D. Häussinger, F. Hollmann, N. J. Turner,
T. R. Ward, Nat. Chem. 2013, 5, 93.
12065.
[10]
[11]
a) J. Steinreiber, T. R. Ward, Coord. Chem. Rev. 2008, 252, 751;
b) T. K. Hyster, L. Knörr, T. R. Ward, T. Rovis, Science 2012,
338, 500.
V. Köhler, Y. M. Wilson, C. Lo, A. Sardo, T. R. Ward, Curr.
Opin. Biotechnol. 2010, 21, 744.
[21] C. Wang, B. Villa-Marcos, J. L. Xiao, Chem. Commun. 2011,
47, 9773.
[22] a) H. K. Chenault, G. M. Whitesides, Appl. Biochem. Biotech-
nol. 1987, 14, 147; b) W. A. van der Donk, H. M. Zhao, Curr.
Opin. Biotechnol. 2003, 14, 421.
[12]
[13]
R. Sambasivan, Z. T. Ball, J. Am. Chem. Soc. 2010, 132, 9289.
ˇ
[23] a) J. Canivet, G. Süss-Fink, P. Stepnicˇka, Eur. J. Inorg. Chem.
˘
a) S. Gladiali, E. Alberico, Chem. Soc. Rev. 2006, 35, 226; b)
S. Hashiguchi, A. Fujii, J. Takehara, T. Ikariya, R. Noyori, J.
Am. Chem. Soc. 1995, 117, 7562.
2007, 4736; b) F. Holmann, A. Kleeb, K. Otto, A. Schmid,
Tetrahedron: Asymmetry 2005, 16, 3512; c) E. Steckhan, S.
Herrmann, R. Ruppert, J. Thommes, C. Wandrey, Angew.
Chem. 1990, 102, 445; Angew. Chem. Int. Ed. 1990, 29, 388; d)
F. Hollmann, K. Hofstetter, A. Schmid, Trends Biotechnol.
2006, 24, 163.
[14]
X. F. Wu, J. L. Xiao, Chem. Commun. 2007, 2449.
[15] a) K. Severin, R. Bergs, W. Beck, Angew. Chem. Int. Ed. 1998,
37, 1634; b) H. Sigel, R. B. Martin, Chem. Rev. 1982, 82, 385.
[16] R. Krämer, M. Maurus, K. Polborn, K. Sünkel, C. Robl, W.
Beck, Chem. Eur. J. 1996, 2, 1518.
[17] X. F. Wu, X. G. Li, F. King, J. L. Xiao, Angew. Chem. 2005,
117, 3473; Angew. Chem. Int. Ed. 2005, 44, 3407.
[18] C. Wang, X. F. Wu, J. L. Xiao, Chem. Asian J. 2008, 3, 1750.
[19] M. E. Wilson, G. M. Whitesides, J. Am. Chem. Soc. 1978, 100,
306.
[24] a) H. Y. Rhyoo, H. J. Park, Y. K. Chung, Chem. Commun.
2001, 2064; b) K. Ahlford, H. Adolfsson, Catal. Commun.
2011, 12, 1118.
[25] I. M. Pastor, P. Västilä, H. Adolfsson, Chem. Eur. J. 2003, 9,
4031.
[26] C. Jäckel, P. Kast, D. Hilvert, Annu. Rev. Biophys. 2008, 37,
153.
[20] a) J. Lutz, F. Hollmann, T. V. Ho, A. Schnyder, R. H. Fish, A.
Schmid, J. Organomet. Chem. 2004, 689, 4783; b) V. Köhler,
Received: March 6, 2013
Published Online: April 26, 2013
Eur. J. Org. Chem. 2013, 3427–3431
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3431