Full stereochemical determination of ajudazols a and b by bioinformatics gene cluster analysis and total synthesis of ajudazol B by an asymmetric ortholithiation strategy

Article abstract of DOI:10.1021/ja309685n

The stereochemical determination of the potent respiratory chain inhibitors ajudazols A and B and the total synthesis of ajudazol B are reported. Configurational assignment was exclusively based on biosynthetic gene cluster analysis of both ketoreductase

Full text of DOI:10.1021/ja309685n

Communication
pubs.acs.org/JACS
Full Stereochemical Determination of Ajudazols A and B by
Bioinformatics Gene Cluster Analysis and Total Synthesis of Ajudazol
B by an Asymmetric Ortholithiation Strategy
Sebastian Essig,† Sebastian Bretzke,† Rolf Muller,‡ and Dirk Menche*,§
̈
†Institut fur Organische Chemie, Ruprecht-Karls Universitat Heidelberg, Im Neuenheimer Feld 270, 69120 Heidelberg, Germany
̈
̈
‡Helmholtz-Institut fur Pharmazeutische Forschung Saarland (HIPS) and Institut fur Pharmazeutische Biotechnologie, Universitat
̈
̈
̈
des Saarlandes, Gebaude C 2.3, 66123 Saarbrucken, Germany
̈
̈
§Kekule-
́
Institut fur Organische Chemie und Biochemie, Universitat Bonn, Gerhard-Domagk-Str. 1, 53121 Bonn, Germany
̈
̈
S
* Supporting Information
mechanism for aerobic production of energy,6 and dysfunctions
contribute to various diseases7 and were recently mentioned in
the context of Parkinson’s and Alzheimer’s diseases.8 This
renders the detailed molecular understanding and development
of effective and synthetically accessible inhibitors important
research goals. The unique architecture of the ajudazols is
characterized by an unusual isochromanone heterocycle bearing
two vicinal anti-configured hydroxyl groups (C8 and C9) that is
connected to an extended side chain incorporating an oxazole, a
(Z,Z)-diene, and a terminal methoxybutenoic acid methylamide
as characteristic features. While 1 bears an exomethylene group
next to the oxazole, 2 has a methyl group at this position (C15).
The ajudazols contain up to four stereocenters of unknown
absolute configuration. Their potent biological properties,
natural scarcity, and unique and intriguing molecular architecture
render the ajudazols attractive targets, and several fragment
syntheses have been reported.9 However, these efforts have been
severely hampered by apparent difficulties in establishing an
efficient route to the isochromanone core and the lack of full
stereochemical knowledge. In particular, the assignment of
isolated methyl-bearing stereocenters10 such as C15 in 2 poses a
particular analytical challenge.
Herein we report the determination of the full stereochemistry
of 1 and 2 and the first total synthesis of 2, the more potent and
less abundant ajudazol. Stereochemical assignment prior to the
synthesis was based on a bioinformatics approach employing one
of the first predictive enoylreductase (ER) alignments for the
assignment of methyl groups. The total synthesis was based on an
innovative approach to the isochromanone core that included an
asymmetric ortholithiation strategy, modular oxazole formation,
and a late-stage Z,Z-selective Suzuki coupling of elaborate
substrates.
ABSTRACT: The stereochemical determination of the
potent respiratory chain inhibitors ajudazols A and B and
the total synthesis of ajudazol B are reported. Configura-
tional assignment was exclusively based on biosynthetic
gene cluster analysis of both ketoreductase domains for
hydroxyl-bearing stereocenters and one of the first
predictive enoylreductase alignments for methyl-bearing
stereocenters. The expedient total synthesis resulting in
unambiguous proof of the predicted stereochemistry
involves a short stereoselective approach to the challenging
isochromanone stereotriad by an innovative asymmetric
ortholithiation strategy, a modular oxazole formation, and
a late-stage Z,Z-selective Suzuki coupling.
he varied architectures of natural products continue to
T
present formidable inspirations for new methods of
structure elucidation and directed synthesis. In recent years,
innovative combinations of genetic and chemical tools have
become more and more important in enabling novel analytical
and synthetic techniques.1 In particular, secondary metabolites
from myxobacteria have become increasingly well understood
objects of study2 for such approaches.3 Ajudazols A (1) and B (2)
(Figure 1) are structurally unique polyketides from the
Stereochemically, the relative anti,anti configuration of C8−
C10 has been proposed on the basis of conformational NMR
studies.4 In contrast, determination of the absolute configuration
has been thwarted by the scarcity of similar natural products, the
notable lability,4 and the general difficulties associated with
assigning isolated methyl-bearing stereocenters. Therefore, we
turned our attention to a complementary approach relying on an
analysis of the biosynthetic gene cluster.11 In detail, the hydroxyl-
Figure 1. Ajudazols A (1) and B (2), potent mitochondrial respiratory
chain inhibitors from the myxobacterium C. crocatus.
myxobacterium Chondromyces crocatus, stain Cm c54 that have
proved to be highly effective inhibitors of the mitochondrial
respiratory chain through selective binding to complex I
(NADH-dehydrogenase).5 The NADH oxidation level in beef
heart submitochondrial particles was inhibited at an IC50 value of
13.0 ng/mL (22.0 nM) for 1 and 10.9 ng/mL (18.4 nM) for 2.
The mitochondrial respiratory chain represents the key
Received: October 1, 2012
Published: November 5, 2012
© 2012 American Chemical Society
19362
dx.doi.org/10.1021/ja309685n | J. Am. Chem. Soc. 2012, 134, 19362−19365

Journal of the American Chemical Society
Communication
bearing stereocenter at C9 is derived by a ketoreductase (KR)-
mediated reduction of the corresponding ketone intermediate.
The stereochemistry at the methyl-bearing C10 and C15 centers
is dictated by ER-catalyzed reduction. The groups of McDaniel12
and Caffrey13 proposed a model for the observed selectivity of
KRs that allows the configuration of secondary alcohols to be
determined in a simple fashion by analysis of the amino acid
sequence in the conserved core regions of the enzymes. In brief,
the presence or absence of one amino acid, an aspartate (D)
residue, corresponds to D-configured or L-configured alcohols,
respectively. This method for configurational assignment of
hydroxyl-bearing stereocenters has proved to be successful in a
number of cases.3a,b Accordingly, analysis of the respective
ajudazol gene cluster revealed a D residue in the KR core region
responsible for the reduction to the alcohol at C9, suggesting that
this stereocenter should be D (or R)-configured (Figure 2 top).
Figure 3. Retrosynthetic analysis of (+)-ajudazol B (2).
cyclodehydration with 3 for oxazole formation and Z,Z-selective
Suzuki coupling16 for connection with 5 was envisioned. A
variable metathesis strategy would provide a basis for 5. For
construction of the labile isochromanone core, a conceptually
novel approach relying on an innovative ortholithiation strategy
was devised.17 The Clayden group published pioneering
studies18 on an asymmetric variant19 of this reaction. To
generate the preferred atropisomer to mediate the electrophilic
attack, they used a chiral amidosulfoxide as a temporary
stereogenic center next to the directing metalation group. After
cleavage of this stereocenter by tBuLi, the amide axis retains the
chiral information in the sense of chiral memory. Capturing the
resulting lithiated species with an aldehyde leads finally to self-
regeneration of the stereocenter SRS-principle.20 However, the
applicability of this method has been thwarted by difficulties
resulting from cleavage of the amide residue (required for
asymmetric induction), the lability of the newly generated
benzylic alcohol toward epimerization, and the limitation to five-
membered rings,19b rendering this part of our route particularly
challenging but also generally rewarding.
Figure 2. Bioinformatics-based configurational assignment of 2 by
sequence alignment of (top) KR and (bottom) ER domains. The crucial
amino acids for the prediction of stereochemistry are marked in blue
boxes. The crucial tyrosine (Y) is absent in the ajudazol ERs, suggesting
that the methyl-bearing stereocenters should be R-configured.
As the starting material for 3, we chose commercially available
3-methylsalicylic acid (6), which already incorporates most of the
aromatic substitution pattern. For protection of the phenolic
hydroxyl, we used an allyl group, which proved to be stable
enough during the subsequent ortholithiation and could be
selectively removed later in the synthesis. After amide formation,
the sterically hindered amide axis was fixed by ortholithiation and
capture with the Andersen reagent21 (7) to yield sulfoxide 8
(78% over four steps; Scheme 1, left). An orthogonal protecting
group strategy was likewise required for OH-8 to secure exclusive
formation of the six-membered lactone. The best results were
obtained with orthogonal silyl protecting groups. In detail
(Scheme 1 right), alcohol 10 was readily available from aldehyde
9 by a Brown crotylation (70%, 90% ee).22 The absolute
configuration was determined by Mosher ester analysis and
independently at a later stage by X-ray structure analysis (see
below). After triethylsilyl (TES) protection (99%), homologa-
tion of 10 involving hydroboration, oxidation, and Wittig
reaction (80% over three steps) afforded ester 11, which was
transformed to the required slightly volatile aldehyde 12 by
standard interconversions (86% over two steps). Gratifyingly, we
were able to achieve the pivotal asymmetric coupling of lithiated
8 with 12 in high yield (80%) after slight modifications of the
originally reported conditions, giving the desired anti,anti-
configured product 1323 with high selectivity.24 As anticipa-
ted,19b conversion of 13 into the desired isochromanone 14
proved challenging because of apparent difficulties in removing
the stable tertiary amide without translactonizing the labile
More recently, the group of Leadlay14 analyzed ERs in an
analogous fashion. Their study revealed that a tyrosine (Y)
residue in the catalytic ER domain plays a critical role in the
stereochemical outcome of this process. According to their
model, the presence of this amino acid results in the formation of
an S-configured methyl-bearing center, while its absence results
in an R configuration. However, in contrast to existing studies on
KRs, only a very limited number of partially contradictory3c
examples have been analyzed, mainly from actinomycetes.14
Therefore, we first evaluated the applicability of this method for
stereochemical determination of myxobacterial metabolites by
analyzing the biosynthetic gene cluster in direct comparison with
the known stereochemical outcome (Figure 2 bottom). In all
cases, perfect agreement was observed.15 This enabled a
confident assignment of the methyl-bearing C10 and C15
stereocenters. In detail, the absence of Y in the respective ER core
regions suggested that these two centers should be R-configured.
The full configurational assignment for 2 was therefore proposed
to be 8S,9R,10R,15R, and on the basis of their common
biogenesis, the stereochemistry of 1 was assigned accordingly
(Figure 1).
Efforts were then directed to a validation of our assignment by
the first total synthesis of 2. As shown in Figure 3, our
retrosynthetic analysis relied on two main subunits, a western
fragment (3) and an eastern fragment (5). A modular late-stage
introduction of the central methyl-bearing subunit (4) by
19363
dx.doi.org/10.1021/ja309685n | J. Am. Chem. Soc. 2012, 134, 19362−19365

Journal of the American Chemical Society
Communication
realized in reasonable yields based on concomitant recovery of
starting material. After removal of the terminal TBS group, the
free alcohol was oxidized, and the resulting aldehyde was
homologated by an Ohira−Bestmann reaction30 with 18 to give
alkyne 19 in 55% yield over four steps. Finally, a Rh-catalyzed
trans-hydroboration following the protocol of Miyaura31 gave
the desired eastern fragment 5 in reasonable yield and selectivity.
As shown in Scheme 3, completion of the total synthesis was
initiated by amide coupling of amino alcohol 3 and acid 432 with
Scheme 1. Synthesis of Western Fragment 3
Scheme 3. Completion of the Total Synthesis
isochromanone and epimerizing the benzylic alcohol. Finally, the
successful strategy was initiated by tert-butyldimethylsilyl (TBS)
protection of the benzylic hydroxyl group (77%). Pd-mediated
deprotection of the phenolic group then proved to be essential to
enable the subsequent amide cleavage,25 which after evaluation
of diverse reagents and strategies was realized by refluxing a
mixture of acetic acid in toluene. These conditions conserved the
TBS protecting group and led to exclusive formation of the
desired six-membered lactone with complete configurational
retention. Notably, simultaneous cleavage of the TES group was
achieved, adding to the effectiveness of this protocol. Microwave
assistance (150 °C) improved the yield to 90% and reduced the
reaction time from 7 days to 3 h. After unification of the
protecting groups (TBSOTf, 2,6-lutidine, 95%), the absolute and
relative configurations of 14 were confirmed by X-ray structure
analysis of a structurally homologated product prepared in an
analogous fashion.26 Finally, 14 was transformed into the desired
western fragment 3 by dihydroxylation of the terminal alkene,
mono-TBS protection of the resulting primary alcohol, azide
substitution27 of the secondary alcohol, and hydrogenation (71%
over four steps).
DEPBT33 followed by selective removal of the primary TBS
group using catalytic amounts of TMSCl in water.34 Oxazole
formation by 2,4-cyclodehydration with IBX followed by
treatment with DTBMP, PPh3, C2Br2Cl4, and DBU relied on a
modified Wipf protocol35 (78% over three steps). Stereoselective
attachment of the side chain was then effected by iodination
(NIS, AgNO3) of alkyne 20, transformation into the
corresponding (Z)-vinyl iodide 21 by syn reduction with
NBSH36 (84% over two steps), and subsequent stereoselective
Suzuki coupling16 with 5 under mild conditions ([Pd(dppf)Cl2],
Ba(OH)2).37 Concomitantly, cleavage of the phenolic TBS
group was realized. Importantly, no signs of isomerization were
observed. Finally, careful deprotection with buffered TASF38
provided synthetic (+)-ajudazol B (2) in 74% yield over two
1
steps. The H and 13C NMR data and specific rotation of our
synthetic material were in agreement with those published for an
authentic sample of ajudazol B (synthetic [α]2D1 = +7.9°, c = 0.9,
MeOH; natural [α]2D1 = +6.1°, c = 1.34, MeOH), thus confirming
the relative and absolute configuration of 2 and validating our
bioinformatics-based assignment.
In contrast to previous approaches,9a,b our synthesis of 5 was
based on an ambitious cross-metathesis28 of allylic amine 16 with
terminal alkene 17 (Scheme 2) to enable facile modifications of
the side chain in the context of scheduled SAR studies. The
synthesis of 16 was accomplished by conversion of methyl
acetonate (15) into the corresponding enoic acid9a and amide
coupling with N-allylmethylamine (68% yield over three steps).
After optimization, the challenging coupling29 with 17 was
In conclusion, we have assigned the full stereochemistry of the
ajudazols purely by an innovative bioinformatics approach. This
involved gene cluster analyses of a ketoreductase domain for
assignment of a hydroxyl-bearing center and one of the first
applications of predictive enoylreductase-based determination of
methyl-bearing stereocenters. This assignment was unequiv-
ocally validated by a total synthesis of ajudazol B (2), which
represents the first total synthesis of the ajudazols in general and
the first total synthesis based exclusively on stereochemical
bioinformatics techniques. Our study documents the high
reliability of modern genetic tools for structure elucidation, in
particular for the determination of isolated methyl groups that
are very difficult to assign by other means. Furthermore, our
synthetic approach involves one of the first applications of an
asymmetric ortholithiation strategy in complex target synthesis.
Along these lines, we have reported effective substrate design and
protocol developments that circumvent various problems
previously associated with this method, including epimerizations,
translactonizations, and eliminations of the product alcohols.
Scheme 2. Synthesis of Eastern Fragment 5
19364
dx.doi.org/10.1021/ja309685n | J. Am. Chem. Soc. 2012, 134, 19362−19365

Journal of the American Chemical Society
Communication
(13) Caffrey, P. ChemBioChem 2003, 4, 654.
These results will generally advance the use of asymmetric
ortholithiations in functional target syntheses.
(14) Kwan, D. H.; Sun, Y.; Schulz, F.; Hong, H.; Popovic, B.; Sim-
Stark, J. C. C.; Haydock, S. F.; Leadlay, P. F. Chem. Biol. 2008, 15, 1231.
(15) In all cases, the presence of Tyr resulted in an S-configured center,
while its absence led to an R configuration. Here an Arg residue was
observed, in contrast to Val, Ala, or Phe in the examples reported by
Leadlay14 [see the Supporting Information (SI)].
(16) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.
(17) (a) Beak, P.; Snieckus, V. Acc. Chem. Res. 1982, 15, 306.
(b) Snieckus, V. Pure Appl. Chem. 1990, 62, 671. (c) Snieckus, V. Pure
Appl. Chem. 1990, 62, 2047. (d) Snieckus, V. Chem. Rev. 1990, 90, 879.
(e) Snieckus, V. Pure Appl. Chem. 1994, 66, 2155.
(18) (a) Bowles, P.; Clayden, J.; Helliwell, M.; McCarthy, C.;
Tomkinson, M.; Westlund, N. J. Chem. Soc., Perkin Trans. 1 1997, 2607.
(b) Clayden, J.; Davies, R. P.; Hendy, M. A.; Snaith, R.; Wheatley, A. E.
H. Angew. Chem., Int. Ed. 2001, 40, 1238. (c) Clayden, J. Chem. Commun.
2004, 127.
(19) (a) Clayden, J.; Mitjans, D.; Youssef, L. H. J. Am. Chem. Soc. 2002,
124, 5266. (b) Clayden, J.; Stimson, C. C.; Keenan, M. Synlett 2005,
1716.
ASSOCIATED CONTENT
* Supporting Information
Experimental procedures and characterization data. This material
is available free of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
dirk.menche@uni-bonn.de
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the Fonds der Chemischen Industrie (stipend to S.E.)
for generous financial support, Dr. Kathrin Buntin for
introduction into ClustalW alignments, Dr. Frank Rominger
for X-ray structure analyses, Andreas J. Schneider for HPLC
support, and Dr. Rolf Jansen for providing original spectra of
natural ajudazol B.
(20) Seebach, D.; Sting, A. R.; Hoffmann, M. Angew. Chem., Int. Ed.
Engl. 1996, 35, 2708.
(21) Andersen, K. K. Tetrahedron Lett. 1962, 3, 93.
(22) (a) Brown, H. C.; Bhat, K. S. J. Am. Chem. Soc. 1986, 108, 293.
(b) Brown, H. C.; Bhat, K. S. J. Am. Chem. Soc. 1986, 108, 5919.
(23) It was unclear whether a second product that could not be
separated from the sulfoxide byproducts was a diastereomer or an
atropisomer. Thus, we could not determine the exact diastereomeric
ratio. The given yield is for pure isolated diastereomer 13, as
diastereomers (potentially resulting from epimerization) can be
differentiated by their NMR spectra (see section 2.11 in the SI).
(24) Without the chiral sulfoxide, product ratios of 2:1 were observed,
presumably as a result of moderate Felkin−Anh control. In agreement
with previous mechanistic studies,18 the selectivites may be rationalized
by the fixed amide axis (preoriented by the chiral sulfoxide), which
controls the attack of the aldehyde from the preferred half-space by
shielding the other half-space.
REFERENCES
■
(1) (a) Bode, H. B.; Muller, R. Angew. Chem., Int. Ed. 2005, 44, 6828.
̈
(b) Udwary, D. W.; Zeigler, L.; Asolkar, R. N.; Singan, V.; Lapidus, A.;
Fenical, W.; Jensen, P. R.; Moore, B. S. Proc. Natl. Acad. Sci. U.S.A. 2007,
104, 10376. (c) Laureti, L.; Song, L.; Huang, S.; Corre, C.; Leblond, P.;
Challis, G. L.; Aigle, B. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 6258.
(2) (a) Wenzel, S. C.; Muller, R. Nat. Prod. Rep. 2009, 26, 1385.
̈
(b) Weissman, K. J.; Muller, R. Nat. Prod. Rep. 2010, 27, 1276.
̈
(3) (a) Janssen, D.; Albert, D.; Jansen, R.; Muller, R.; Kalesse, M.
Angew. Chem., Int. Ed. 2007, 46, 4898. (b) Menche, D.; Arikan, F.;
̈
Perlova, O.; Horstmann, N.; Ahlbrecht, W.; Wenzel, S. C.; Jansen, R.;
Irschik, H.; Muller, R. J. Am. Chem. Soc. 2008, 130, 14234. (c) Jahns, C.;
̈
(25) Various previously published procedures for cleavage of tertiary
amides resulted in no conversion or five-membered-ring formation.
(26) The stereochemistry of 14 was further confirmed by two
additional X-ray structure analyses of deprotected derivatives of 13 (for
full details, see the SI).
Hoffmann, T.; Muller, S.; Gerth, K.; Washausen, P.; Hofle, G.;
̈
̈
Reichenbach, H.; Kalesse, M.; Muller, R. Angew. Chem., Int. Ed. 2012, 51,
̈
5239.
(4) Jansen, R.; Kunze, B.; Reichenbach, H.; Hofle, G. Eur. J. Org. Chem.
̈
2002, 917.
(27) Lal, B.; Pramanik, B. N.; Manhas, M. S.; Bose, A. K. Tetrahedron
Lett. 1977, 18, 1977.
(5) Kunze, B.; Jansen, R.; Hofle, G.; Reichenbach, H. J. Antibiot. 2004,
̈
57, 151.
(28) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18.
(29) Unsaturated amines are well-known to be poor substrates for
metathesis because of their ability to coordinate to metal−alkylidene
complexes through their lone pair, thereby inhibiting the catalytic
activity. See: Compain, P. Adv. Synth. Catal. 2007, 349, 1829.
(6) (a) Saraste, M. Science 1999, 283, 1488. (b) Hirst, J. Biochem. J.
2010, 425, 327.
(7) (a) Wallace, D. C. Science 1999, 283, 1482. (b) Leonard, J. V.;
Schapira, A. H. Lancet 2000, 355, 299. (c) Leonard, J. V.; Schapira, A. H.
Lancet 2000, 355, 389. (d) Smeitink, J.; van den Heuvel, L.; DiMauro, S.
Nat. Rev. Genet. 2001, 2, 342.
(30) (a) Ohira, S. Synth. Commun. 1989, 19, 561. (b) Muller, S.;
̈
Liepold, B.; Roth, G. J.; Bestmann, H. J. Synlett 1996, 521.
(31) Ohmura, T.; Yamamoto, Y.; Miyaura, N. J. Am. Chem. Soc. 2000,
122, 4990.
(32) The synthesis of 4 relied on a Myers alkylation (see the SI).
(33) Li, H.; Jiang, X.; Ye, Y.-h.; Fan, C.; Romoff, T.; Goodman, M. Org.
Lett. 1999, 1, 91.
(34) Grieco, P. A.; Markworth, C. J. Tetrahedron Lett. 1999, 40, 665.
(35) (a) Wipf, P.; Miller, C. P. J. Org. Chem. 1993, 58, 3604. (b) Wipf,
P.; Venkatraman, S. J. Org. Chem. 1995, 60, 7224.
(36) Myers, A. G.; Zheng, B.; Movassaghi, M. J. Org. Chem. 1997, 62,
7507.
(37) Gopalarathnam, A.; Nelson, S. G. Org. Lett. 2006, 8, 7.
(38) Scheidt, K. A.; Chen, H.; Follows, B. C.; Chemler, S. R.; Coffey, D.
S.; Roush, W. R. J. Org. Chem. 1998, 63, 6436.
(8) (a) Navarro, A.; Boveris, A. J. Bioenerg. Biomembr. 2009, 41, 517.
(b) Muller, W. E.; Eckert, A.; Kurz, C.; Eckert, G. P.; Leuner, K. Mol.
̈
Neurobiol. 2010, 41, 159.
(9) (a) Krebs, O.; Taylor, R. J. Org. Lett. 2005, 7, 1063. (b) Ganame,
D.; Quach, T.; Poole, C.; Rizzacasa, M. A. Tetrahedron Lett. 2007, 48,
5841. (c) Hobson, S. J.; Parkin, A.; Marquez, R. Org. Lett. 2008, 10,
2813. (d) Egan, B. A.; Paradowski, M.; Thomas, L. H.; Marquez, R. Org.
Lett. 2011, 13, 2086. (e) Birkett, S.; Ganame, D.; Hawkins, B. C.;
Meiries, S.; Quach, T.; Rizzacasa, M. A. Org. Lett. 2011, 13, 1964.
(f) Egan, B. A.; Paradowski, M.; Thomas, L. H.; Marquez, R. Tetrahedron
2011, 67, 9700. (g) Brikett, S. L.; Loits, D. A.; Wimala, S.; Rizzacasa, M.
A. Pure Appl. Chem. 2012, 84, 1421.
(10) Menche, D. Nat. Prod. Rep. 2008, 25, 905.
(11) (a) Buntin, K.; Rachid, S.; Scharfe, M.; Blocker, H.; Weissman, K.;
̈
Muller, R. Angew. Chem., Int. Ed. 2008, 47, 4595. (b) Buntin, K.;
̈
Weissman, K. J.; Muller, R. ChemBioChem 2010, 11, 1137.
̈
(12) Reid, R.; Piagentini, M.; Rodriguez, E.; Ashley, G.; Viswanathan,
N.; Carney, J.; Santi, D. V.; Hutchinson, C. R.; McDaniel, R. Biochemistry
2003, 42, 72.
19365
dx.doi.org/10.1021/ja309685n | J. Am. Chem. Soc. 2012, 134, 19362−19365