Stereocontrolled total synthesis of bengazole A: A marine bisoxazole natural product displaying potent antifungal properties

Article abstract of DOI:10.1002/anie.200602050

(Chemical Equation Presented) Candidate for Candida: A high-yielding and versatile synthesis of bengazole A has been developed; the first to provide access to a single stereoisomer of the natural product. Key steps include the construction of the 2,4-disu

Full text of DOI:10.1002/anie.200602050

Communications
Total Synthesis
DOI: 10.1002/anie.200602050
Stereocontrolled Total Synthesis of Bengazole A:
A Marine Bisoxazole Natural Product Displaying
Potent Antifungal Properties**
James A. Bull, Emily P. Balskus, Richard A. J. Horan,
Martin Langner, and Steven V. Ley*
Numerous oxazole-containing natural products have been
isolated from marine sources in recent years[1] and have been
the subject of extensive synthetic investigation.[2] A significant
subsection of these natural products are bisoxazoles. In these
cases, the two oxazole rings may be well separated, for
example in phorboxazole[3] and the disorazoles,[4] or are
directly linked as bisoxazole cores, such as in the henn-
oxazoles,[5] the diazonamides,[6] and in muscoride A.[7]
The bengazoles, on the other hand, are unique in that the
two oxazole rings flank a single carbon atom. In bengazole A
(1), this C10 position carries a myristate ester unit, which
creates a sensitive isolated stereogenic center (Scheme 1).
Moreover, the left hand oxazole ring is a biogenically unusual
5-monosubstituted ring while the central 2,4-disubstituted
oxazole ring links to a stereochemically dense tetraol side
chain.
Bengazole A was originally isolated in 1988 by Crews and
co-workers from a sponge of the genus Jaspis,[8] and to date
the bengazole family consists of 22 members.[9,10] The
structure and absolute stereochemistry of this highly func-
tionalized molecule was determined by a combination of
NMR studies, analysis of degradation products, and circular
dichroism.[8,10] The compound displays potent ergosterol-
dependant antifungal activity comparable to that of ampho-
tericin B, against Candida albicans,[11,12] and is active against
fluconazole-resistant Candida strains.[13] Furthermore, beng-
azole A shows full anthelminthic activity against nematode
Nippostrongylus braziliensis at just 50 mgmLÀ1.[8]
There has been one reported previous synthesis of
bengazole A (1), by Molinski and co-workers,[13] which used
a regioselective metalation/addition strategy starting from
1,3-oxazole.[14,15] However, this gave a 1:1 mixture of the C10
[*] J. A. Bull, E. P. Balskus, Dr. R. A. J. Horan, Dr. M. Langner,
Prof. Dr. S. V. Ley
Department of Chemistry
University of Cambridge
Lensfield Road, Cambridge CB21EW (UK)
Fax: (+44)1223-336-442
E-mail: svl1000@cam.ac.uk
Homepage: http://leygroup.ch.cam.ac.uk
[**] We would like to thank Novartis for a Novartis Research Fellowship
(S.V.L.), the EPSRC for a studentship (J.A.B.) and a postdoctoral
fellowship (R.A.J.H.), the Winston Churchill Foundation of the USA
for a scholarship (E.P.B.), and the DAAD for a postdoctoral
fellowship (M.L.).
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
6714
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6714 –6718

Angewandte
Chemie
epimers, and once the resulting bisoxazole unit was in place,
the C10 epimers could not be separated at any stage of the
synthesis. Shioiri and co-workers achieved a synthesis of
deacylbengazole[16] that was stereochemically enriched at the
C10 position through an enantioselective reduction of a
bisoxazole ketone which resulted in ee values up to 68%.[17]
Owing to the interesting biological profile and the
challenging structural components, we devised a synthetic
route to 1 that would provide both a single stereoisomer of the
natural product and also allow for analogues to be prepared.
We envisaged installing the C10 stereogenic center early in
the synthesis, prior to construction of the bisoxazole unit. This
somewhat risky strategy would require retention of stereo-
chemical integrity at the C10 center during the formation of
the central oxazole ring and on to the end of the synthesis. The
desired bisoxazole unit would be derived from the diastereo-
merically pure serine amide 3 (Scheme 1), which in turn
would be converted into a suitable precursor for a diastereo-
selective nitrile oxide 1,3-dipolar cycloaddition with alkene 4,
which contains a diol protected as a butane-2,3-diacetal
(BDA). Cycloaddition product 2, which contains the com-
plete carbon backbone of 1, would then be elaborated to
reveal the tetraol side chain to complete the total synthesis.
BDA-protected diols have been used extensively by our
research group as stable chiral building blocks.[18] The syn-
thesis of amide fragment 3 relied upon stereocontrol derived
from using our BDA-protected glyceraldehyde derivative
Scheme 2. Synthesis of amide fragment 3. a) TosMIC, K2CO3, MeOH,
reflux, 82%; b) 1. TFA, H2O, RT; 2. TESCl, iPr2NEt, CH2Cl2, À788C,
94% over 2 steps; c) 1. TBDPSCl, Et3N, DMAP, CH2Cl2, RT; 2. PPTS
MeOH, RT, 96% over two steps; d) Jones reagent, acetone, 08C, 49%;
e) TBTU, H-(l)-Ser-OMe·HCl, iPr2NEt, DMSO, RT, 64%. TosMIC=
toluene-4-sulfonylmethylisocyanide, TFA=trifluoroacetic acid, TES=
triethylsilyl, DMAP=4-dimethylaminopyridine, PPTS=pyridiniump-tol
uenesulfonate, TBTU=2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyl
uronium tetrafluoroborate, DMSO=dimethylsulfoxide.
mild conditions in the presence of the sensitive C10 stereo-
genic center. Initial attempts to form this oxazole ring from
serine amide 3 included: 1) oxazoline formation[21] followed
by oxidation,[22] and 2) using conditions modified from those
reported by Shin and co-workers[23] and involving dehydrative
elimination followed by activation using N-bromosuccin-
imide, cyclization, and elimination. Although these
approaches afforded the desired product, they proceeded in
poor yields and caused significant racemization of the
bisoxazole product.
We found that serine amide 3 could be converted into
primary alcohol 10 in two steps (Scheme 3). Oxidation using
buffered Dess–Martin periodinane[24] provided aldehyde 11,
which was subjected to Robinson–Gabriel-type oxazole
formation under conditions reported by Panek and Beresis,[25]
similar to those described previously by Wipf et al.[26] Treating
the crude aldehyde 11 with PPh3, 2,6-di-tert-butylpyridine,
and dibromotetrachloroethane in CH2Cl2 to form the inter-
mediate bromooxazoline, followed by the addition of 1,8-
diazabicyclo[5.4.0]undec-7-ene (DBU) to eliminate HBr,
5
[18d] as a substrate for a Schöllkopf-type oxazole synthesis.[19]
Treatment of aldehyde 5 with TosMIC and K2CO3 provided
the desired 5-substituted oxazole in excellent yield
6
(Scheme 2). Oxazole 6 was isolated exclusively as the
equatorially substituted diastereomer, hence efficiently estab-
lishing the desired stereochemistry at the C10 center for the
natural product. Removal of the BDA group and manipu-
lation of silyl protecting groups afforded the mono-TBDPS-
protected diol 8, with all reactions proceeding in excellent
yields. Oxidation of the primary alcohol 8 proved to be
difficult and was only achieved on using the Jones reagent,
which provided carboxylic acid 9 in moderate yield. Subse-
quent amide coupling performed in DMSO gave the desired
amide 3 as a single diastereomer, thus showing there was no
loss of stereochemical integrity during the preceding steps.[20]
Having gained rapid access to amide 3 as a single
diastereomer, the key challenge in this synthesis was the
construction of the central 2,4-substituted oxazole ring under
Scheme 1. Structure and retrosynthetic analysis of bengazole A (1). TBDPS=tert-butyldiphenylsilyl.
Angew. Chem. Int. Ed. 2006, 45, 6714 –6718ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6715

Communications
trasts significantly with the synthesis of the simple acetonide-
protected alkene derived from ester 13; the known inter-
mediate five-ring acetonide aldehyde[30] is prone to polymer-
ization and the corresponding alkene is difficult to handle
because of its volatility. Furthermore, we hoped the embed-
ded chirality and rigidity of alkene 4 would provide enhanced
side-chain diastereoselectivity in the planned dipolar cyclo-
addition.
Bisoxazole 12 was next converted into oxime 17 in
readiness for the 1,3-dipolar cycloaddition (Scheme 5). This
Scheme 3. Synthesis of bisoxazole 12: a) 1. TBSCl, imidazole, DMAP,
DMF, RT, 90%; 2. NaBH4, LiCl, THF, MeOH, 08C toRT, 80%;
b) Dess–Martin periodinane, NaHCO3, CH2Cl2, 08C; c) PPh3, C2Br2Cl4,
CH2Cl2, 2,6-di-tert-butylpyridine, 08C, then Et3N, MeCN, 08C toRT,
89% over 2 steps. TBS=tert-butyldimethylsilyl.
provided the desired 2,4-disubstituted oxazole 12 in 72%
yield over the two steps.
However, the product obtained from this process still
displayed significant racemization (33–66% ee)[27] at the
crucial C10 center. The cause of this was identified as the
DBU, used to eliminate the proposed bromooxazoline
intermediate.[26b] Replacing DBU with Et3N completely
prevented this racemization and also improved the yield
over the two steps from alcohol 10 to an excellent 89%.
Bisoxazole 12 was thus obtained with greater than 98% ee
and the process was amenable to scale-up.[28,29] We believe this
to be a potentially useful general modification for the
formation of oxazole rings adjacent to highly epimerizable
stereogenic centers.
Scheme 5. Synthesis of isoxazoline 2: a) PPTS, MeOH, RT, 93%;
b) 1. Dess–Martin periodinane, NaHCO3, CH2Cl2, 08C; 2. NH2OH·HCl,
K2CO3, MeOH, H2O, RT, 93% over two steps; c) 1. NCS, CH2Cl2,
pyridine, RT, 2. alkene 4, Cs2CO3, DME, RT; 77% over two steps,
(d.r. 7:2). NCS=N-chlorosuccinimide, DME=1,2-dimethoxyethane.
With the enantiopure bisoxazole unit in hand, we turned
our attention to the alkene fragment. Alkene 4 was readily
synthesized in 71% yield over four steps from the commer-
cially available acetonide 13 (Scheme 4). Transacetalization
using CSA and 2,3-butanedione in methanol at reflux
introduced the BDA protecting group as a 6:1 mixture of
anomerically and non-anomerically stabilized products, which
were then converted exclusively into the fully stabilized
isomer 14 by using BF3·THF. Reduction of ester 14 with
DIBAL-H provided aldehyde 15 in 86% yield as a crystalline
solid, and Wittig methylenation formed the required alkene 4
in quantitative yield. This synthesis was readily scalable
through easily handled crystalline intermediates, which con-
transformation proceeded in three steps, under mild condi-
tions, with an average yield of 95%, and with complete
retention of stereochemical integrity. The required nitrile
oxide was generated from oxime 17 in two stages. Firstly, the
oxime was chlorinated by treatment with NCS, and then
Cs2CO3 was added to a solution of the chlorooxamic acid in
DME in the presence of alkene 4. The resulting 1,3-dipolar
cycloaddition provided isoxazoline 2 in 60% yield of the
major diastereomer.[31] The stereochemistry at C4 of the
major cycloadduct is that found in the natural product and the
result is consistent with that predicted by the “inside alkoxy”
model reported by Houk et al.[32] The minor diastereomer
(isolated in 17% yield) was easily removed by column
chromatography and will be used in the studies towards
analogues of 1.
With the carbon skeleton complete, cycloadduct 2 was
then elaborated to the natural product. Reductive cleavage of
isoxazoline 2 using Raney nickel and boric acid in ethanol/
water[33] gave b-hydroxy ketone 18 in 61% yield, which was
subjected to a syn-selective reduction using diethylmethox-
yborane and sodium borohydride (Scheme 6). Protection of
the resulting diol (d.r. 14:1 by NMR spectroscopy) as
acetonide 19 occurred in 89% yield over the two steps, with
the minor C6 diastereomer being removed at this stage.
Removal of the TBDPS group and installation of the
myristoyl side chain proceeded in excellent yield to afford
the acetal-protected bengazole A 20. Acidic deprotection
Scheme 4. Synthesis of alkene 4: a) 1. 2,3-butanedione, CSA, trimethyl-
orthoformate, MeOH, reflux; 2. BF3·THF, CH2Cl2, RT, 83% over two
steps; b) DIBAL-H, CH2Cl2, À788C, 86%; c) Ph3PCH3Br, KHMDS, THF,
À788C toRT, quant. CSA =camphorsulfonic acid, DIBAL-H=diiso-
butylaluminum hydride, KHMDS=potassium hexamethyldisilazide.
6716
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6714 –6718

Angewandte
Chemie
[9] a) J. Rodrꢂguez, R. M. Nieto,
P. Crews, J. Nat. Prod. 1993,
56, 2034; b) A. Rudi, Y.
Kashman, Y. Benayahu, M.
Schleyer, J. Nat. Prod. 1994,
57, 829; c) A. Groweiss, J. J.
Newcomer, B. R. OꢃKeefe,
A. Blackman, M. R. Boyd,
J. Nat. Prod. 1999, 62, 1691;
d) R. Fernµndez, M. Dherbo-
mez, Y. Letourneux, M.
Nabil, J. F. Verbist, J. F.
Biard, J. Nat. Prod. 1999, 62,
678.
Scheme 6. Completion of the synthesis of bengazole A (1): a) Raney nickel, H2, B(OH)3, EtOH, H2O, 61%;
b) 1. Et2BOMe, NaBH4, THF, À788C, (d.r. 14:1); 2. 2,2-dimethoxypropane, TsOH, 89% over 2 steps;
c) 1. TBAF, THF, 08C, quant.; 2. myristoyl chloride, NEt3, CH2Cl2, 08C, 85%; d) TFA, H2O, RT, 95%.
Ts =toluene-4-sulfonic acid, TBAF=tetrabutylammonium fluoride.
[10] P. A. Searle, R. K. Richter,
T. F. Molinski, J. Org. Chem.
1996, 61, 4073.
[11] T. F. Molinski, J. Nat. Prod.
1993, 56, 1.
[12] J. Antonio, T. F. Molinski, J.
Nat. Prod. 1993, 56, 54.
[13] a) R. J. Mulder, C. M. Shafer, T. F. Molinski, J. Org. Chem. 1999,
64, 4995; b) C. M. Shafer, T. F. Molinski, Tetrahedron Lett. 1998,
39, 2903.
[14] C. M. Shafer, T. F. Molinski, Heterocycles 2000, 53, 1167.
[15] a) J. C. Hodges, W. C. Patt, C. J. Connolly, J. Org. Chem. 1991,
56, 449; b) E. Vedejs, S. D. Monahan, J. Org. Chem. 1996, 61,
5192.
using TFA/water removed both the acetonide and the BDA
groups from 20 in 95% yield to complete the total synthesis of
bengazole A (1) in an overall yield of 3.4%. The data for our
synthetic material were consistent with those reported for the
natural product, and 1HNMR spectroscopic analysis con-
firmed our product to be a single C10 epimer.[8,13,34–36]
In conclusion, we have completed a stereocontrolled total
synthesis of bengazole A by using an oxazole synthesis under
mild conditions and a diastereoselective 1,3-dipolar cyclo-
addition. Our route is the first to provide a single stereoiso-
mer of the product and relies upon BDA building blocks to
establish the stereogenic centers. Ongoing work in our
laboratories is directed toward the synthesis of other mem-
bers of the bengazole family and toward the synthesis of
analogues for studies on structure–activity relationships.
[16] P. Chittari, Y. Hamada, T. Shioiri, Heterocycles 2003, 59, 465.
[17] P. Chittari, Y. Hamada, T. Shioiri, Synlett 1998, 1022.
[18] a) J. S. Barlow, D. J. Dixon, A. C. Foster, S. V. Ley, D. J.
Reynolds, J. Chem. Soc. Perkin Trans. 1 1999, 1627; b) D. J.
Dixon, A. C. Foster, S. V. Ley, D. J. Reynolds, J. Chem. Soc.
Perkin Trans. 1 1999, 1631; c) S. V. Ley, D. K. Baeschlin, D. J.
Dixon, A. C. Foster, S. J. Ince, H. W. M. Priepke, D. J. Reynolds,
Chem. Rev. 2001, 101, 53; d) P. Michel, S. V. Ley, Angew. Chem.
2002, 114, 4054; Angew. Chem. Int. Ed. 2002, 41, 3898; .
[19] a) U. Schöllkopf, R. Schröder, Angew. Chem. 1971, 83, 358;
Angew. Chem. Int. Ed. Engl. 1971, 10, 333; b) A. M. van Leusen,
B. E. Hoogenboom, H. Siderius, Tetrahedron Lett. 1972, 2369.
[20] The yield of the amide coupling to form amide 3 is highly
dependent on the quality of carboxylic acid 9, which is unstable
and decomposes on storage. A more practical route on a large
scale generates acid 9 through hydrolysis of the equivalent non-
enantiopure methyl ester. The coupling with the enantiopure
serine methyl ester resolves the acid into diastereomers, which
are easily separated by column chromatography to provide
amide 3 as a single stereoisomer.
[21] For oxazole-forming reactions using (diethylamino)sulfur tri-
fluoride (DAST), see: a) W. J. Middleton, J. Org. Chem. 1975, 40,
574; b) P. Lafargue, P. Guenot, J.-P. Lellouche, Heterocycles
1995, 41, 947; c) A. J. Phillips, Y. Uto, P. Wipf, M. J. Reno, D. R.
Williams, Org. Lett. 2000, 2, 1165; for reactions in which the
Burgess reagent is used, see: d) G. M. Atkins, Jr., E. M. Burgess,
J. Am. Chem. Soc. 1968, 90, 4744; e) P. Wipf, C. P. Miller,
Tetrahedron Lett. 1992, 33, 907; for selected other methods, see:
f) A. Sakakura, R. Kondo, K. Ishihara, Org. Lett. 2005, 7, 1971;
g) P. Wipf, C. P. Miller, Tetrahedron Lett. 1992, 33, 6267.
[22] D. R. Williams, P. D. Lowder, Y.-G. Gu, D. A. Brooks, Tetrahe-
dron Lett. 1997, 38, 331.
Received: May 23, 2006
Revised: July 31, 2006
Published online: September 20, 2006
Keywords: antifungal agents · cycloaddition · heterocycles ·
.
natural products · total synthesis
[1] a) J. R. Lewis, Nat. Prod. Rep. 2002, 19, 223; b) Z. Jin, Z. Li, R.
Huang, Nat. Prod. Rep. 2002, 19, 454, and references therein.
[2] a) V. S. C. Yeh, Tetrahedron 2004, 60, 11995; b) Y. Hamada, T.
Shioiri, Chem. Rev. 2005, 105, 4441; c) P. Wipf, Chem. Rev. 1995,
95, 2115; d) P. Wipf, S. Venkatraman, Synlett 1997, 1; e) L. O.
Haustedt, I. V. Hartung, H. M. R. Hoffmann, Angew. Chem.
2003, 115, 2815; Angew. Chem. Int. Ed. 2003, 42, 2711; ; f) E.
Riego, D. Hernµndez, F. Albericio, M. lvarez, Synthesis 2005,
1907, and references therein.
[3] P. A. Searle, T. F. Molinski, J. Am. Chem. Soc. 1995, 117, 8126.
[4] R. Jansen, H. Irschik, H. Reichenbach, V. Wray, G. Höfle,
Liebigs Ann. Chem. 1994, 759.
[5] T. Ichiba, W. Y. Yoshida, P. J. Scheuer, T. Higa, D. G. Gravalos, J.
Am. Chem. Soc. 1991, 113, 3173.
[23] N. Endoh, K. Tsuboi, R. Kim, Y. Yonezawa, C.-G. Shin,
Heterocycles 2003, 60, 1567.
[6] N. Lindquist, W. Fenical, G. D. Van Duyne, J. Clardy, J. Am.
Chem. Soc. 1991, 113, 2303.
[24] D. B. Dess, J. C. Martin, J. Am. Chem. Soc. 1991, 113, 7277.
[25] J. S. Panek, R. T. Beresis, J. Org. Chem. 1996, 61, 6496.
[26] a) P. Wipf, C. P. Miller, J. Org. Chem. 1993, 58, 3604; b) P. Wipf, S.
Lim, J. Am. Chem. Soc. 1995, 117, 558.
[7] A. Nagatsu, H. Kajitani, J. Sakakibara, Tetrahedron Lett. 1995,
36, 4097.
[8] M. Adamczeski, E. Quiꢀoꢁ, P. Crews, J. Am. Chem. Soc. 1988,
110, 1598.
Angew. Chem. Int. Ed. 2006, 45, 6714 –6718ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6717

Communications
[27] The ee value of bisoxazole 12 was determined using HPLC on a
chiral stationary phase (Agilent 1100 series, Chiralcel AS,
hexane/2-propanol 99:1, flow rate = 0.5 mLminÀ1, 248C, l =
215 nm, retention times: (S)-C10 isomer 10.35 min, (R)-C10
isomer 11.85 min).
[28] Using Et3N in place of 2,6-di-tert-butylpyridine from the start of
the reaction again provided bisoxazole 12 as a single enantiomer.
However, these conditions gave a yield of only 47% and a
reduced reaction time, thus suggesting a different reaction
pathway.
[29] The ee value was determined by supercritical fluid chromatog-
raphy on a chiral stationary phase following conversion into the
primary alcohol 16 (Berger Minigram, Chiralcel AS column with
7% MeOH/CO2, flow rate = 3 mLminÀ1 at 100 bar, 358C, l =
210 nm, retention times: (S)-C10 isomer 8.66 min, (R)-C10
isomer 9.08 min).
[30] J. M. Gardiner, N. R. Panchal, W. T. Stimpson, J. M. Herbert,
G. J. Ellames, Synlett 2005, 2685.
[31] a) A. P. Kozikowski, A. K. Ghosh, J. Am. Chem. Soc. 1982, 104,
5788; b) A. P. Kozikowski, A. K. Ghosh, J. Org. Chem. 1984, 49,
2762; c) V. Jꢄger, R. Schohe, E. F. Paulus, Tetrahedron Lett. 1983,
24, 5501.
[32] a) K. N. Houk, S. R. Moses, Y.-D. Wu, N. G. Rondan, V. Jꢄger, R.
Schohe, F. R. Fronczek, J. Am. Chem. Soc. 1984, 106, 3880;
b) K. N. Houk, H.-Y. Duh, Y.-D. Wu, S. R. Moses, J. Am. Chem.
Soc. 1986, 108, 2754.
[33] D. P. Curran, J. Am. Chem. Soc. 1983, 105, 5826.
[34] Molinski and co-workers report that the 1HNMR spectra of the
C10 epimers are very similar apart from the C15 (a-CH2 on
ester) and C1 proton signals (see reference [13]). These signals
1
show no such splitting in the HNMR spectra of our synthetic
material.
[35] 1H NMR (600 MHz, CD3OD): d = 0.89(3H, t, J = 7.4 Hz, H27),
1.15 (3H, d, J = 6.4 Hz, H1), 1.28 (20H, m, H17 to H26), 1.63
(2H, t, J = 7.0 Hz, H16), 1.90 (1H, ddd, J = 14.1, 9.2, 7.1 Hz,
H5’), 2.23 (1H, ddd, J = 14.1, 6.9, 2.5 Hz, H5), 2.43 (2H, t, J =
7.3 Hz, H15), 3.18 (1H, dd, J = 6.8, 3.4 Hz, H3), 3.67 (1H, ddd,
J = 9.2, 6.8, 2.5 Hz, H4), 3.92 (1H, dq, J = 6.4, 3.4 Hz, H2), 4.91
(1H, dd, J = 7.1, 6.9 Hz, H6), 7.11 (1H, s, H10), 7.31 (1H, s,
H12), 7.85 (1H, s, H8), 8.25 ppm (1H, s, H13); 13C NMR
(150 MHz, CD3OD): 14.4 (C27), 19.9 (C1), 23.7 (C26), 25.8
(C16), 30.8–30.0, 33.1 (C17 to C25), 34.5 (C15), 40.5 (C5), 62.9
(C10), 66.3 (C6), 67.8 (C2), 71.2 (C4), 78.8 (C3), 127.5 (C12),
138.0 (C8), 145.7 (C7), 147.7 (C11), 154.4 (C13), 159.7 (C9), d =
173.4 ppm (C14); [a]2D5 = + 6 (c = 0.175, CH3OH); HRMS (ESI)
for C27H45N2O8: m/z calcd for [M+H]+: 525.3176, found:
525.3163.
[36] We observed no racemization at the C10 atom on storage of
synthetic bengazole A under argon at À208C over a period of
one month.
6718
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6714 –6718