A dehydratase domain in ambruticin biosynthesis displays additional activity as a Pyran-forming Cyclase

Article abstract of DOI:10.1002/anie.201407979

Hydropyran rings are a common structural motif in reduced polyketides. Information on their biosynthetic formation and particularly the biochemical characterization of the responsible enzymes has only been reported in few cases. The dehydratase domain AmbDH3 from the ambruticin polyketide synthase was investigated. Through in vitro assay of the recombinant domain with synthetically-derived substrate surrogates, it was shown that it has a second catalytic activity as a cyclase that performs oxa-conjugate addition. Probing AmbDH3 with synthetic substrate analogues revealed stereo-selectivity and substrate tolerance in both substeps. This is the first characterization of a pyran-forming cyclase from a cis-AT PKS system and the first report of a polyketide synthase domain with this kind of dual activity. Finally, it was revealed that this domain shows potential for application in chemo-enzymatic synthesis.

Full text of DOI:10.1002/anie.201407979

.
Angewandte
Communications
DOI: 10.1002/anie.201407979
Enzyme Catalysis
A Dehydratase Domain in Ambruticin Biosynthesis Displays
Additional Activity as a Pyran-Forming Cyclase**
Gesche Berkhan and Frank Hahn*
Abstract: Hydropyran rings are a common structural motif in
reduced polyketides. Information on their biosynthetic forma-
tion and particularly the biochemical characterization of the
responsible enzymes has only been reported in few cases. The
dehydratase domain AmbDH3 from the ambruticin poly-
ketide synthase was investigated. Through in vitro assay of the
recombinant domain with synthetically-derived substrate sur-
rogates, it was shown that it has a second catalytic activity as
a cyclase that performs oxa-conjugate addition. Probing
AmbDH3 with synthetic substrate analogues revealed stereo-
selectivity and substrate tolerance in both substeps. This is the
first characterization of a pyran-forming cyclase from a cis-AT
PKS system and the first report of a polyketide synthase
domain with this kind of dual activity. Finally, it was revealed
that this domain shows potential for application in chemo-
enzymatic synthesis.
domains that carry out unconventional transformations,
which finally lead to the formation of alternative structural
elements.[2] Although biosynthetic mechanisms for these
processes have frequently been proposed based on gene
cluster analysis, few have been confirmed by in vitro studies
on the enzyme level.
Oxygen-containing heterocycles are present in many
polyketides, including some with exceptional biological
activity like salinomycin (1) and bryostatin (2; Scheme 1a).[3]
It is known that these rings are formed by oxidation/
P
olyketides are important natural products with great
potential for drug discovery. The products of reducing type I
polyketide synthases (PKSs) are structurally highly diverse
and represent ambitious synthetic targets. Bacterial type I
PKSs are multimodular multi-enzyme complexes that con-
duct the standard processing of the polyketide backbones
during biosynthesis.[1] They catalyze the elongation of a grow-
ing polyketide chain by decarboxylative Claisen-like conden-
sations. In each module, a so-called reductive loop establishes
the individual backbone functionalization pattern. Ketore-
ductase (KR) domains reduce 3-oxo-thioesters to 3-hydroxy-
thioesters, dehydratase (DH) domains eliminate water, and
enoylreductase (ER) domains reduce the resulting a,b-
unsaturated thioesters to fully saturated chains. During the
whole process, the substrate remains tethered to the PKS
through the phosphopantetheine linker on the acyl carrier
proteins (ACPs). PKS assembly lines can also contain
Scheme 1. a) Structures of pharmacologically relevant pyran-containing
polyketides; b) In vitro activity of a PS domain from the pederin PKS
was previously demonstrated with structurally reduced substrate
surrogates 4a and 4b. These differ considerably from the natural
precursor 6.[2b,3,5h] PS=pyran synthase, SNAC=N-acetylcysteamine.
cyclization or epoxidation/epoxide-opening cascades.[4] In
addition, domains with sequence similarities to D5-3-ketoste-
roid isomerases, hydrolases, or dehydratases have been
proposed to act as cyclases that perform oxa-conjugate
addition, based on gene cluster analysis and bioconversion
experiments.[5]
[*] G. Berkhan, Dr. F. Hahn
Institut fꢀr Organische Chemie, Leibniz Universitꢁt Hannover
Schneiderberg 1 B, 30167 Hannover (Germany)
E-mail: frank.hahn@oci.uni-hannover.de
Homepage: http://uhw3dev.uni-hannover.de/oci/de/arbeitskreise/
hahn/index.php
[**] Funding from the Emmy Noether program of the DFG (HA 5841/2-
1), the Marie Curie program of the European Union (project number
293430) and the Graduate School MINAS (doctoral stipend to G.B.)
is gratefully acknowledged. We thank Dr. Gerald Drꢁger and Dr.
Anna Kromm for their help with X-ray crystal structure analysis as
well as Claudia Holec for preliminary synthetic efforts. We thank the
NMR facility and the mass facility of the OCI Hannover as well as
Prof. Andreas Kirschning for general support.
Recently, the first in vitro characterization of a so-called
pyran synthase (PS) domain was reported.[2b,5e] The PS
domain from module 7 of the pederin PKS-NRPS hybrid
(PedPS7) converted simple substrate surrogates 4a and 4b
into the respective cyclized forms 5a and 5b with ee values of
67 and 89% in favor of the natural 3-l-configuration
(Scheme 1b; NRPS = nonribosomal peptide synthetase).
Since the proposed natural PedPS7 precursor 6 differs
considerably from the surrogate substrates 4a and 4b, this
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201407979.
14240
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 14240 –14244

Angewandte
Chemie
result demonstrates substrate tolerance of PedPS7, but also
leaves open questions about its actual stereoselectivity.
Further experiments with more realistic substrate surrogates
are thus required.
Gene cluster analyses revealed that PS domains are
usually coded in DH-containing modules and act downstream
of the latter domain by cyclizing its a,b-unsaturated product
(Scheme 1b). All PS-containing systems known to date
belong to the class of trans-AT PKS. For the much more
abundant cis-AT PKS, nature seems to have developed
alternative strategies, which have not yet been extensively
studied.[5h,6]
Although the gene clusters of several pyran-containing
cis-AT PKS products, for example, the ionophoric polyethers
salinomycin (1) and nigericin, have been sequenced, a unifying
proposal for the responsible catalytic entity is still mis-
sing.[5h,6a]
The ambruticins (13) are important lead structures for the
development of antimycotic agents because of their unique
interaction with the high-osmolarity glycerol protein kinase
pathway (Scheme 2).[6b,7] They are assembled by a type I cis-
AT PKS and contain a dihydropyran ring in the eastern part
of the molecule. This ring was proposed to be formed by an
electrophilic methyl group originating from S-adenosyl-l-
methionine (Scheme 2b).[6b] The latter step was proposed to
be catalyzed by the C-methyltransferase AmbM. Usually,
standard PKS domains process intermediates only on C-2 and
C-3. Thus we favored a mechanisms in which oxa-1,4-
conjugate addition in module 3 leads to tetrahydrofuran 9
(Scheme 2a). Elongation to the a,b-unsaturated system in
module 4, followed by methylation and a double-bond shift
would then yield the b,g-unsaturated intermediate 12.[8] Since
the ambruticin gene cluster does not harbor a PS-like domain,
we hypothesized that the DH domain of module 3
(AmbDH3) is responsible for carrying out both transforma-
tions: dehydration and the following cyclization.
To test our biosynthetic hypothesis, we aimed to assay the
recombinant domain AmbDH3 in vitro with suitable sub-
strate surrogates.[9] In order to learn about mechanistic details
and to assign the exact constitution and configuration of the
biosynthetic precursor, we envisaged chemically synthesizing
all potential precursors, intermediates, and products of the
process. The number of putative substrates could gratifyingly
be reduced by bioinformatic analysis.[10] The KR domains of
the ambruticin PKS modules 1 and 3 both showed the
characteristic LDD motif and thus indicated a 3-d,7-d-
configuration of precursor 7.[10a,b]
Application of another recently
reported algorithm, however, did
not give a sufficiently clear predic-
tion for the configurations at C-2
and C-6, so we decided to synthesize
all possible configurational iso-
mers.[10c]
Surrogates for 3-hydroxy-ACP
thioester 7, enoate-ACP thioester 8,
and tetrahydropyran-ACP thioester
9 were prepared in the form of N-
acetylcysteamine (SNAC) thioest-
ers, which are common surrogates
for the ACP-bound form of DH
Scheme 2. Two hypothetical pathways for hydropyran ring formation in the eastern part of the
substrates (Figure 1a).[9] Two series
of compounds with 6-l,7-d-config-
uration (14–18) and 6-d,7-d-config-
ambruticins as proposed by us (a) and Reeves et al. (b).[6]
oxa-conjugate addition, followed by dehydrogenation by the
iron-sulfur Rieske protein AmbO at a later point of the
biosynthesis. The configuration of the intermediately formed
stereocenter at C-8 (in 12) is unknown. The respective 20,21-
dihydroambruticins were isolated from an ambO deletion
mutant and structurally analyzed. However, both epimers of
20,21-dihydroambruticin were shown in the respective manu-
script and the analytical data were not presented. A clear
assignment of the absolute configuration at C-21 was thus not
possible.[6b]
Gene cluster analysis suggested that the pyran cyclization
must occur during processing by PKS modules 3 or 4.[6b] Both
modules contain ACP, KS, AT, KR, and DH domains,
consistent with twofold ketide elongation and ultimate
processing to a 2,4-dienoyl thioester. Reeves et al. suggested
a biosynthetic route to 12 that proceeds through a 1,6-
conjugate addition on the vinylic Michael acceptor 10 and
subsequent trapping of the resulting enolate 11 with an
uration (19–22) were synthesized and their absolute config-
urations were determined by single-crystal X-ray analysis and
NMR spectroscopy (Scheme S1, Figure S16–18, and Table S2
in the Supporting Information).[11] The domain boundaries of
AmbDH3 were determined through alignment to previously
characterized DH domains.[9a,b] The identified gene was
cloned into a pET28a(+) vector, heterologously expressed
in E. coli, and used in the assays after affinity purification
(Figure S1).
HPLC–MS analysis after individual overnight incubations
of the 3-hydroxy thioesters 14 and 15 with AmbDH3 showed
conversion only for the 2-d,3-d-configured precursor 15
(Figure 1b, Figure S3,S4). This finding is in agreement with
previous reports of a conserved high substrate specificity for
DH domains regarding the configurations at C-2 and C-3.[9]
The two more hydrophobic products of this reaction co-eluted
with the synthetic standards 16 and 17, thus suggesting
a dehydration–cyclization cascade (Figure 1b,c). To confirm
Angew. Chem. Int. Ed. 2014, 53, 14240 –14244
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 14241

.
Angewandte
Communications
excluding spontaneous, non-enzymatic side reactions (Fig-
ure S3a,b–8a,b,S9). To ensure that the observed enzymatic
conversion was not caused by trace amounts of another
protein that was purified along with AmbDH3, we generated
a specific mutant in which the active site histidine was
replaced with an alanine residue. No conversion was obtained
if 19 or 20 were incubated with this mutant, thus showing that
AmbDH3 was indeed responsible for dehydration (Fig-
ure S10). Finally, we aimed to show that the cyclization
activity of AmbDH3 is an intrinsic feature of the domain and
not just an artifact from applying this particular kind of
substrate. Therefore, we incubated the competent substrate
19 with the recombinant domain BorDH3 from the borrelidin
PKS. We had shown before that this domain accepts 2-d,3-d-
configured precursors and transforms them stereoselectively
into E-configured enoates with broad substrate specificity.[9a,b]
BorDH3 performed the expected dehydration to the E-
configured enoate 20, but no further cyclization to tetrahy-
drofuran rings like 21 or 22 was observed (Figure S11).
In order to unambiguously prove the configurational
assignment of the products and furthermore explore the
potential of AmbDH3 for synthetic purposes, we repeated all
of the assays with substrates 15, 16, 19, and 20 on a semi-
preparative scale (> 8 mg substrate in each case) and
analyzed the outcome by NMR spectroscopy (Figure 2 and
Figure S12–15). In all cases, we observed similar results to the
assays analyzed by HPLC–MS. AmbDH3 formed products 17
and 21 from substrates 16 and 20, respectively, with con-
versions of 18 and 87% (determined from the NMR spectra).
Incubation of AmbDH3 with 8.0 mg of the precursor 20,
followed by extractive workup gave virtually pure product 21.
This result is a promising starting point for the further
development of AmbDH3 as a useful chemoenzymatic tool
for the stereoselective synthesis of tetrahydropyran rings.
Known PS domains show relevant protein sequence
homology to DH domains, but fall into a distinct phylogenetic
clade.[2b] Furthermore, the DH-characteristic active-site motif
His-Ser-Asp is mutated to His-Ser-His/Asn, in agreement
with the fact that an acidic residue located close to C-3 of the
a,b-unsaturated PS precursor would be counterproductive for
the proposed conjugate addition mechanism. We aligned the
AmbDH3 amino acid sequence against a collection of PS
domains and DH domains from both cis- and trans-AT PKSs
and found that AmbDH3 bears the DH-characteristic His-
Ser-Asp active-site motif (Figure S19). Phylogenetic analysis
also revealed that AmbDH3 falls into a clade with typical DH
domains from cis-AT PKS systems (Figure S20).
Figure 1. a) Structures of the synthetic substrate and product surro-
gates 14–22 that were applied in the assays. HPLC–MS analyses are
shown for the AmbDH3 overnight incubations with 15 (b) and 16 (d),
as well as with compounds 19 (e) and 20 (g). The individual traces for
the synthetic reference compounds 15–22 are combined in panels (c)
and (f) for clarity (for unprocessed data, see Figure S5–9).
the nature of E-configured olefin 16 as a competent inter-
mediate, we also individually incubated this compound with
the AmbDH3 domain. We observed the formation of only
one product, which also co-eluted with the 2d,3D-configured
pyran 17 (Figure 1d and Figure S5). Usually, DH-catalyzed
reactions exist in an equilibrium between the hydrated and
dehydrated forms. Interestingly, in our case no 3-hydroxy
thioester like 15 was obtained, thus suggesting that the
position of the equilibrium was located far on the side of the
unsaturated (16) and cyclized (17) forms.[9]
The presence of detectable quantities of starting material
15, intermediate 16, and product 17 (amounts decreasing in
this order) following long-term incubation of AmbDH3 with
compound 15 points towards low specificity of the domain for
the applied substrate, particularly in the cyclization step. By
contrast, incubation of 2-d,3-d,6-d,7-d-configured compound
19 with AmbDH3 gave only minor amounts of the starting
material 19 and the a,b-unsaturated intermediate 20, but high
quantities of tetrahydropyran 21 (Figure 1e,f and Figure S7).
Incubation of intermediate 20 gave nearly complete con-
version into compound 21 and no rehydration to 19 (Fig-
ure 1g and Figure S8). On the basis of the differences in
conversion, we concluded that the natural intermediates of
the pathway must be 2-d,6-d-7, (2E)-6-d-8, 2-d,6-d-9 and
8-d-12 (Scheme 2).
In summary, we describe the unprecedented discovery of
a PKS domain that shows two catalytic activities: as
a dehydratase and as a cyclase that performs oxa-conjugate
1,4-addition. Our work is the first characterization of a pyran-
forming cyclase from a cis-AT PKS system and the first report
of a PKS domain with this kind of dual activity.[2b] Stereose-
lective synthesis of surrogate substrates allowed us to inter-
rogate the AmbDH3 domain with realistic precursor, inter-
mediate, and product analogues. As a result, we were able to
show that both sub-steps naturally proceed from 2-d,6-d-7
with high stereoselectivity, thereby leading to the intermedi-
ate (2E),6-d-8 and finally to the tetrahydropyran product
To definitely assign the cyclase activity to AmbDH3, we
performed a series of control experiments. Enzyme-free
overnight incubations of the synthetic surrogates 14–16 and
19–22 under assay conditions showed no conversion, thus
14242 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 14240 –14244

Angewandte
Chemie
Figure 2. 1H NMR analysis of a semipreparative-scale conversion of compound 19 with AmbDH3 (middle panel), as well as for starting material
19 (upper panel) and product 21 (lower panel; all spectra recorded in CDCl3 at 400 MHz). Characteristic signals are highlighted with a gray
background.
Keywords: biocatalysis · cyclases · dehydratases · heterocycles ·
polyketides
.
2-d,6-d-9. The substrate specificity of AmbDH3 is narrow
regarding the configurations at C-2 and C-3, but configura-
tional changes at C-6 are tolerated. Pronounced differences in
the conversion of the diastereomeric substrates 15 and 19, as
well as 16 and 20, clearly indicate that the previously
[1] a) J. Staunton, K. J. Weissman, Nat. Prod. Rep. 2001, 18, 380 –
uncertain configuration at C-8 in the biosynthetic intermedi-
ate 12 is d.
416; b) K. J. Weissman, P. F. Leadlay, Nat. Rev. Microbiol. 2005,
3, 925 – 936; c) S. Dutta, J. R. Whicher, D. A. Hansen, W. A.
Hale, J. A. Chemler, G. R. Congdon, A. R. H. Narayan, K.
Enzymes from the biosynthetic pathways of secondary
metabolites have great potential to find application as tools
for the chemoenzymatic synthesis of natural products.[12] We
have shown that AmbDH3 is able to conduct clean, semi-
preparative-scale reactions under neutral conditions. Thus,
the stage is set for a further investigation of AmbDH3
regarding its applicability to the chemoenzymatic synthesis of
pyran-containing natural products. Extensive investigation of
the substrate specificity of AmbDH3 will be the next step, as
well as the further elucidation of mechanistic and structural
details in order to understand how two catalytic activities are
combined in one PKS domain.
Hakansson, D. H. Sherman, J. L. Smith, G. Skiniotis, Nature
2014, 510, 512 – 517; d) J. R. Whicher, S. Dutta, D. A. Hansen,
W. A. Hale, J. A. Chemler, A. M. Dosey, A. R. H. Narayan, K.
Hakansson, D. H. Sherman, J. L. Smith, G. Skiniotis, Nature
2014, 510, 560 – 564; e) C. Khosla, D. Herschlag, C. E. Cane, C. T.
Walsh, Biochemistry 2014, 53, 2875 – 2883; f) C. Hertweck,
Angew. Chem. Int. Ed. 2009, 48, 4688 – 4716; Angew. Chem.
2009, 121, 4782 – 4811.
[2] a) T. Bretschneider, J. B. Heim, D. Heine, R. Winkler, B. Busch,
B. Kusebauch, T. Stehle, G. Zocher, C. Hertweck, Nature 2013,
502, 124 – 128; b) P. Pçplau, S. Frank, B. I. Morinaka, J. Piel,
Angew. Chem. Int. Ed. 2013, 52, 13215 – 13218; Angew. Chem.
2013, 125, 13457 – 13460; c) T. A. M. Gulder, M. F. Freeman, J.
Piel, Top. Curr. Chem. 2010, 297, 149 – 201.
[3] a) P. B. Gupta, T. T. Onder, G. Jiang, K. Tao, C. Kuperwasser,
R. A. Weinberg, E. S. Lander, Cell 2009, 138, 645 – 659; b) B. A.
DeChristopher, B. A. Loy, M. D. Marsden, A. J. Schrier, J. A.
Zack, P. A. Wender, Nat. Chem. 2012, 4, 705 – 710.
Received: August 5, 2014
Revised: September 4, 2014
Published online: October 19, 2014
Angew. Chem. Int. Ed. 2014, 53, 14240 –14244
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 14243

.
Angewandte
Communications
[4] a) M. E. A. Richter, N. Traitcheva, U. Knꢀpfer, C. Hertweck,
Angew. Chem. Int. Ed. 2008, 47, 8872 – 8875; Angew. Chem.
2008, 120, 9004 – 9007; b) S. M. Cooper, W. Laosripaiboon, A. S.
Rahman, J. Hothersall, A. K. El-Sayed, C. Winfield, J. Crosby,
R. J. Cox, T. J. Simpson, C. M. Thomas, Chem. Biol. 2005, 12,
825 – 833; c) A. Minami, M. Shimaya, G. Suzuki, A. Migita, S. S.
Shinde, K. Sato, K. Watanabe, T. Tamura, H. Oguri, H. Oikawa,
J. Am. Chem. Soc. 2012, 134, 7246 – 7249; d) P. F. Leadlay, J.
Staunton, M. Oliynyk, C. Bisang, J. Cortꢁs, E. Frost, Z. A.
Hughes-Thomas, S. G. Kendrew, J. B. Lester, P. F. Long, H. A. I.
MacArthur, E. L. McCormick, Z. Oliynyk, C. B. W. Stark, C. J.
Wilkinson, J. Ind. Microbiol. Biotechnol. 2001, 27, 360 – 367.
[5] a) A. J. Woo, W. R. Strohl, N. D. Priestley, Antimicrob. Agents
Chemother. 1999, 43, 1662 – 1668; b) W. C. Smith, L. Xiang, B.
Shen, Antimicrob. Agents Chemother. 2000, 44, 1809 – 1817; c) J.
Piel, Proc. Natl. Acad. Sci. USA 2002, 99, 14002 – 14007; d) J.
Piel, D. Hui, G. Wen, D. Butzke, M. Platzer, N. Fusetani, S.
Matsunaga, Proc. Natl. Acad. Sci. USA 2004, 101, 16222 – 16227;
e) S. Sudek, N. B. Lopanik, L. E. Waggoner, M. Hildebrand, C.
Anderson, H. Liu, A. Patel, D. H. Sherman, M. G. Haygood, J.
Nat. Prod. 2006, 70, 67 – 74; f) C. Li, K. E. Roege, W. L. Kelly,
ChemBioChem 2009, 10, 1064 – 1072; g) H. Irschik, M. Kopp,
K. J. Weissman, K. Buntin, J. Piel, R. Mꢀller, ChemBioChem
2010, 11, 1840 – 1849; h) M. E. Yurkovich, P. A. Tyrakis, H.
Hong, Y. Sun, M. Samborskyy, K. Kamiya, P. F. Leadlay,
ChemBioChem 2012, 13, 66 – 71; i) M. A. Matilla, H. Stçckmann,
F. J. Leeper, G. P. C. Salmond, J. Biol. Chem. 2012, 287, 39125 –
39138.
Antibiot. 2000, 53, 1182 – 1190; c) K. Kojima, Y. Takano, A.
Yoshimi, C. Tanaka, T. Kikuchi, T. Okuno, Mol. Microbiol. 2004,
53, 1785 – 1796; d) P. J. Westfall, D. R. Ballon, J. Thorner, Science
2004, 306, 1511 – 1512.
[8] a) F. Taft, M. Brunjes, T. Knobloch, H. G. Floss, A. Kirschning, J.
Am. Chem. Soc. 2009, 131, 3812 – 3813; b) J. Moldenhauer,
D. C. G. Gçtz, C. R. Albert, S. K. Bischof, K. Schneider, R. D.
Sꢀßmuth, M. Engeser, H. Gross, G. Bringmann, J. Piel, Angew.
Chem. Int. Ed. 2010, 49, 1465 – 1467; Angew. Chem. 2010, 122,
1507 – 1509; c) B. Kusebauch, B. Busch, K. Scherlach, M. Roth,
C. Hertweck, Angew. Chem. Int. Ed. 2010, 49, 1460 – 1464;
Angew. Chem. 2010, 122, 1502 – 1506.
[9] a) O. Vergnolle, F. Hahn, A. Baerga-Ortiz, P. F. Leadlay, J. N.
Andexer, ChemBioChem 2011, 12, 1011 – 1014; b) N. Kandziora,
J. N. Andexer, S. J. Moss, B. Wilkinson, P. F. Leadlay, F. Hahn,
Chem. Sci. 2014, 5, 3563 – 3567; c) C. R. Valenzano, Y.-O. You,
A. Garg, A. Keatinge-Clay, C. Khosla, D. E. Cane, J. Am. Chem.
Soc. 2010, 132, 14697 – 14699; d) X. Guo, T. Liu, C. R. Valen-
zano, Z. Deng, D. E. Cane, J. Am. Chem. Soc. 2010, 132, 14694 –
14696; e) D. Gay, Y.-O. You, A. Keatinge-Clay, D. E. Cane,
Biochemistry 2013, 52, 8916 – 8928.
[10] a) R. Reid, M. Piagentini, E. Rodriguez, G. Ashley, N. Viswa-
nathan, J. Carney, D. V. Santi, C. R. Hutchinson, R. McDaniel,
Biochemistry 2003, 42, 72 – 79; b) P. Caffrey, ChemBioChem
2003, 4, 654 – 657; c) A. Kitsche, M. Kalesse, ChemBioChem
2013, 14, 851 – 865.
[11] CCDC 1017520 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via www.
ccdc.cam.ac.uk/data_request/cif.
[6] a) B. M. Harvey, T. Mironenko, Y. Sun, H. Hong, Z. Deng, P. F.
Leadlay, K. J. Weissman, S. F. Haydock, Chem. Biol. 2007, 14,
703 – 714; b) B. Julien, Z.-Q. Tian, R. Reid, D. Reeves, Chem.
Biol. 2006, 13, 1277 – 1286.
[12] A. Kirschning, F. Hahn, Angew. Chem. Int. Ed. 2012, 51, 4012 –
4022; Angew. Chem. 2012, 124, 4086 – 4096.
[7] a) D. T. Connor, R. C. Greenough, M. von Strandtmann, J. Org.
Chem. 1977, 42, 3664 – 3669; b) P. Knauth, H. Reichenbach, J.
14244 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 14240 –14244