Total synthesis of herbimycin A

Article abstract of DOI:10.1021/ol062642i

(Chemical Equation Presented) Hsp90 has recently emerged as a promising biological target for treatment of cancer. Herbimycin A and other members of the benzoquinoid ansamycin class of natural products are known to inhibit Hsp90 activity. The total synthesis of herbimycin A was achieved from the commercially available Roche ester 1 by using allylmetals to control the stereogenic centers at C6, C7, C10, C11, and C12 and a ring-closing metathesis to control the (Z)-double bond of the (E,Z)-dienic moiety.

Full text of DOI:10.1021/ol062642i

ORGANIC
LETTERS
2007
Vol. 9, No. 1
145-148
Total Synthesis of Herbimycin A
Sophie Canova,† Ve´ronique Bellosta,† Antony Bigot,‡ Patrick Mailliet,‡
Serge Mignani,‡ and Janine Cossy*,†
Laboratoire de Chimie Organique, associe´ au CNRS, ESPI, 10 rue Vauquelin,
75231 Paris Cedex 05, France, and Sanofi-AVentis, 13 quai Jules Guesde,
94403 Vitry-sur-Seine, France
janine.cossy@espci.fr
Received October 27, 2006
ABSTRACT
Hsp90 has recently emerged as a promising biological target for treatment of cancer. Herbimycin A and other members of the benzoquinoid
ansamycin class of natural products are known to inhibit Hsp90 activity. The total synthesis of herbimycin A was achieved from the commercially
available Roche ester 1 by using allylmetals to control the stereogenic centers at C6, C7, C10, C11, and C12 and a ring-closing metathesis to
control the (Z)-double bond of the (E,Z)-dienic moiety.
Heat shock protein 90 (Hsp90) is an essential protein that
chaperones multiple growth-regulatory signaling proteins,
including protein kinases1 and transcription factors.2,3 The
benzoquinoid ansamycins, such as herbimycin A and gelda-
namycin (Figure 1), have recently been identified as inhibi-
tors of the Hsp90 folding process. This leads to disruption
of the Hsp90 client protein complexes and subsequent
ubiquitination and degradation by the proteosome of the
client proteins. Moreover, a number of known oncogenic
proteins are Hps90 client proteins. As a result, Hsp90 is a
promising biological target for treatment of cancer, and
geldanamycin and herbimycin A as well as their analogues
have the potential of serving as chemotherapeutic agents in
a number of diseases. Consequently, these compounds are
attractive targets for synthetic chemists. Here, we report the
total synthesis of herbimycin A.
exhibits a broad spectrum of biological activities including
herbicidal, antifungal, antiangiogenic, and antitumor proper-
ties.4 The structure and absolute configuration of herbimycin
1
A was based on H and 13C NMR spectroscopic analysis
and X-ray analysis.4b,c Herbimycin A is a 19-membered
macrocyclic lactam which possesses seven stereogenic
centers, a carbamate, an isolated trisubstituted (E)-double
bond, an (E,Z)-diene, and a benzoquinone ring system.
Despite its interesting properties, only two total syntheses
of herbimycin A are described in the literature.5 One formal
synthesis and the synthesis of advanced fragments have also
Herbimycin A was isolated in 1979 from the fermentation
broth of Streptomyces hygroscopicus strain AM-3672 and
† Laboratoire de Chimie Organique.
‡ Sanofi-Aventis.
(1) Chavany, C.; Mimnaugh, E.; Miller, P.; Bitton, R.; Nguyen, P.; Trepel,
J.; Whitesell, L.; Schnur, R.; Moyer, J. D.; Neckers, L. J. Biol. Chem. 1996,
271, 4974-4977.
(2) Minet, E.; Mittet, D.; Michel, G.; Roland, I.; Raes, M.; Remacle, J.;
Michiels, C. FEBS Lett. 1999, 460, 251-256.
Figure 1. Two inhibitors of Hsp90: herbimycin A and geldana-
mycin.
10.1021/ol062642i CCC: $37.00
© 2007 American Chemical Society
Published on Web 12/05/2006

been reported.6 Other members of the ansamycin benzo-
quinone class of natural products have been synthesized,
including macbecin I7 and geldanamycin.8
Scheme 1. Retrosynthetic Analysis
Our synthetic plan for herbimycin A is depicted in Scheme
1 and involved a macrolactamization applied to seco acid A
at the end of the synthesis. In order to control the (Z)-double
bond of the dienic system of herbimycin A, a ring-closing
metathesis of the dienic ester C was considered. The
formation of the (E,Z)-diene should then be completed by a
Wittig reaction applied to lactol B. Further simplification of
C led to a disconnection of the C15-C16 bond giving
fragments D and E. The C12 center in E should be con-
trolled by using an enantioselective allyltitanation, and the
C10,C11 and C6,C7 stereogenic centers would be controlled
by employing a syn-crotylboration and a syn-γ-alkoxy-
allylboration, respectively. The synthesis will begin from the
commercially available Roche ester 1 possessing the (R)
configuration corresponding to the C14 stereogenic center
of herbimycin A.
Hydroxyester 1 was transformed to aldehyde 2 in four
steps with an overall yield of 80%. After protection of the
primary alcohol as a TBDPS ether (TBDPSCl, imidazole,
rt, quant), the ester function was reduced by DIBAL-H (Et2O,
-78 °C, 1.5 h, 93%) to yield the corresponding aldehyde
directly, which was then homologated to 2 in two steps using
the methoxymethylphosphonium ylide followed by an acidic
hydrolysis of the enol ether intermediately formed. Aldehyde
2 was then subjected to an enantioselective allyltitanation9
with the highly face-selective (S,S)-I complex (Et2O, -78 °C,
(3) Segnitz, B.; Gehring, U. J. Biol. Chem. 1997, 272, 18694-18701.
(4) (a) Omura, S.; Iwai, Y.; Sakadane, N.; Nakagawa, A.; Oiwa, H.;
Hasegawa, Y.; Ikai, T. J. Antibiot. 1979, 32, 255-261. (b) Omura, S.;
Nakagawa, A.; Sadakane, N. Tetrahedron Lett. 1979, 44, 4323-4326. (c)
Furusaki, A.; Matsumoto, T.; Nakagawa, A.; Omura, S. J. Antibiot. 1980,
33, 781-782. (d) Shibata, K.; Satsumabayashi, S.; Sano, H.; Komiyama,
K.; Nakagawa, A.; Omura, S. J. Antibiot. 1986, 39, 415-423. (e) Lin, L.-
Z.; Blasko, G.; Cordell, G. A. J. Nat. Prod. 1988, 51, 1161-1165. (f)
Yamaki, H.; Iguchi-Ariga, S. M. M.; Ariga, H. J. Antibiot. 1989, 42, 604-
610. (g) Yamashita, T.; Sakai, M.; Kawai, Y.; Aono, M.; Takahashi, K. J.
Antibiot. 1989, 42, 1015-1017. (h) Suzukake-Tsuchiya, K.; Moriya, Y.;
Hori, M.; Uehara, Y.; Takeuchi, T. J. Antibiot. 1989, 42, 1831-1837. (i)
Nagaishi, K.; Adachi, R.; Matsui, S.; Yamaguchi, T.; Kasahara, T.; Suzuki,
K. J. Cell. Phys. 1999, 180, 345-354. (j) Sakagami, M.; Morrison, P.;
Welch, W. J. Cell Stress Chaperones 1999, 4, 19-28. (k) Janin, Y. L. J.
Med. Chem. 2005, 48, 7503-7512.
(5) (a) Nakata, M.; Osumi, T.; Ueno, A.; Kimura, T.; Tatsuta, K.
Tetrahedron Lett. 1991, 32, 6015-6018. (b) Nakata, M.; Osumi, T.; Ueno,
A.; Kimura, T.; Tatsuta, K. Bull. Chem. Soc. Jpn. 1992, 65, 2974-2991.
(c) Carter, K. D.; Panek, J. S. Org. Lett. 2004, 6, 55-57.
(6) (a) Eshelman, J. E.; Epps, J. L.; Kallmerten, J. Tetrahedron Lett.
1993, 34, 749-752. (b) Martin, S. F.; Limberakis, C.; Burgess, L. E.;
Hartmann, M. Tetrahedron 1999, 55, 3561-3572. (c) Centone-Andureau,
S.; Pore´e, F.-H.; Betzer, J.-F.; Brion, J.-D.; Pancrazi, A.; Ardisson, J. Synlett
2005, 981-985.
(7) (a) Baker, R.; Castro, J. J. Chem. Soc., Chem. Commun. 1989, 378-
381. (b) Baker, R.; Castro, J. J. Chem. Soc., Perkin Trans. 1 1990, 47-65.
(c) Evans, D. A.; Miller, S. J.; Ennis, M. D.; Ornstein, P. L. J. Org. Chem
1992, 57, 1067-1069. (d) Evans, D. A.; Miller, S. J.; Ennis, M. D. J. Org.
Chem. 1993, 58, 471-485. (e) Panek, J; Xu, F. J. Am. Chem. Soc. 1995,
117, 10587-10588. (f) Panek, J.; Xu, F.; Rondon, A. C. J. Am. Chem.
Soc. 1998, 120, 4113-4122. (g) Martin, S. F.; Dodge, J. A.; Burgess, L.
E.; Hartmann, M. J. Org. Chem. 1992, 57, 1070-1072. (h) Martin, S. F.;
Dodge, J. A.; Burgess, L. E.; Limberakis, C.; Hartmann, M. Tetrahedron
1996, 52, 3229-3246.
(8) (a) Andrus, M. B.; Meredith, E. L.; Sekhar, B. B. V. S. Org. Lett.
2001, 3, 259-262. (b) Andrus, M. B.; Meredith, E. L.; Simmons, B. L.;
Sekhar, B. B. V. S.; Hicken, E. J. Org. Lett. 2002, 4, 3549-3552. (c)
Andrus, M. B.; Meredith, E. L.; Hicken, E. J.; Simmons, B. L.; Glancey,
R. R. J. Org. Chem. 2003, 68, 8162-8169.
5 h) to afford the homoallylic alcohol 3 in 96% yield (dr
>95/5). After methylation of the secondary alcohol in 3 (KH,
MeI, 97% yield), the terminal double bond of 4 was
isomerized10 by treatment with the second-generation Grubbs’
catalyst GII11 in the presence of the N-allyl-N-tritylamine
and diisopropylethylamine (toluene, reflux, 1.5 h) resulting
in the formation of allylic ether 5 in 98% yield.
After oxidative cleavage of olefin 5 (OsO4, NMO then
NaIO4), treatment of the obtained aldehyde with the
(Z)-crotylboronate12 6 allowed the control of the C10 and
C11 stereocenters. Alcohol 7 was isolated in 70% yield, and
after methylation (KH, MeI, 92%) and oxidative cleavage
(OsO4, NMO then NaIO4), the resulting aldehyde was
submitted to the Corey-Schlessinger olefination conditions13
to afford the R,â-unsaturated aldehyde 9 in 93% yield (E/Z
) 4/1). At this stage, the control of the C6-C7 stereocenters
was examined. Addition of the (Z)-γ-methoxy allylborane
10, derived from (-)-R-pinene and developed by Brown,14
(9) (a) Hafner, A.; Duthaler, R. O.; Marti, R.; Rihs, J.; Rhote-Streit, P.;
Schwarzenbach, F. J. Am. Chem. Soc. 1992, 114, 2321-2336. (b) Duthaler,
R. O.; Hafner, A. Chem. ReV. 1992, 92, 807-832.
(10) Wipf, P.; Rector, S. R.; Takahashi, H. J. Am. Chem. Soc. 2002,
124, 14848-14849.
(11) Scholl, M.; Ding, S.; Woo Lee, S.; Grubbs, R. H. Org. Lett. 1999,
1, 953-956.
(12) Roush, W. R.; Palkowitz, A. D.; Ando, K. J. Am. Chem. Soc. 1990,
112, 6348-6359.
(13) (a) Corey, E. J.; Enders, D.; Bock, M. G. Tetrahedron Lett. 1976,
7-10. (b) Schlessinger, R. H.; Poss, M. A.; Richardson, S.; Lin, P.
Tetrahedron Lett. 1985, 26, 2391-2934.
146
Org. Lett., Vol. 9, No. 1, 2007

acetone, reflux, 72%) and the nitro group was selectively
reduced (Fe, AcOH, H2O, reflux, 94%). Allylation of the
aniline function of 1615 (allyl bromide, Na2CO3, DMF, reflux,
95%) produced the C16-C21 fragment 17 with an overall
yield of 49% from 14 (Scheme 3).
Scheme 2. Synthesis of the C5-C15 Fragment of
Herbimycin A
Scheme 3. Synthesis of the C16-C21 Fragment
The organolithium reagent issued from 17 was obtained
by halogen-metal exchange using n-BuLi (-78 °C then rt,
30 min) and was then reacted with aldehyde 13 (Et2O,
-78 °C, 1.5 h) to produce the two epimers at C15, 18
and 18′, in a 1.6/1 ratio,16 in 87% yield. After separa-
tion, the major epimer 18 (the Felkin-Anh adduct, 54%
yield) was methylated at C15 (NaH, MeI, DMF, quant)
and N-deprotected [Pd(PPh3)4, N,N-dimethylbarbituric acid
(NDMBA), 97%] to give aniline 19. This latter compound
was then transformed to the desired unsaturated lactone 21
in four steps. In order to perform the RCM, the aniline
function should be protected with a deactivating group, and
the synthesis of the N,N-biscarbamate from 19 was performed
(Boc2O, Et3N, DMAP, THF, reflux, 75%). After deprotection
of the hydroxy group at C7 (TBAF, THF) and esterification
using acryloyl chloride (DIPEA, CH2Cl2, 89%), the obtained
dienic ester 20 was subjected to RCM using the second-
generation Grubbs’ catalyst (GII 20 mol %, PhMe, c )
5.10-3 M), and the unsaturated lactone 21 was isolated in
72% yield. The transformation of 21 to the (E,Z)-conjugated
diene 23 was achieved in two steps. At first, the lactone was
reduced to the corresponding lactol using DIBAL-H (Et2O,
-78 °C), and the resulting lactol was treated with the
stabilized Wittig reagent PPh3dC(Me)CO2Et 22 in toluene
to produce the (E,Z)-diene 23 (dr >9/1) in 85% yield
(Scheme 4).17
on aldehyde 9 (THF, -50 °C to 0 °C, 6 h) led to the desired
syn-product (dr ) 4/1), which was directly protected to
produce the TBDMS ether 11 (TBDMSOTf, 2,6-lutidine)
in 69% overall yield for the two steps. The selective
deprotection of the primary alcohol, without affecting the
secondary TBDMS ether at C7, was achieved by using NH4F
in methanol (reflux, 8 h, 82%), and the primary alcohol in
compound 12 was then oxidized to the corresponding
aldehyde 13 by using Dess-Martin periodinane (DMP,
NaHCO3, rt) in quantitative yield. All data obtained for
compounds 12 (1H NMR, 13C NMR, IR, [R]D) and 13 (1H
NMR) were in agreement with those reported in the
literature.5b The known C5-C15 fragment 13 was synthe-
sized in 16 steps with an overall yield of 25% (Scheme 2).
With the C5-C15 fragment in hand, we next turned our
attention to the coupling reaction of 13 with the aromatic
fragment 17 (C16-C21 fragment). The aromatic fragment
17 was prepared in four steps from the commercially
available 4-methoxy-2-nitrophenol 14. After regioselective
ortho-bromination of the aromatic ring (Br2, AcONa, AcOH,
76%), the phenol function was methylated (Me2SO4, K2CO3,
After deprotection of the aniline, and saponification of the
ester, the treatment of the nonpurified seco acid with
bis(2-oxo-3-oxazolidinyl)phosphinic chloride (BOPCl) and
DIPEA gave the macrocycle 24 in 77% yield. Completion
of herbimycin A synthesis was accomplished by formation
(15) The synthesis of compound 16 is described in the literature. See:
Guay, V.; Brassard, P. J. Heterocycl. Chem. 1987, 24, 1649-1652.
(16) Configuration of the major isomer 18 was verified by formation in
two steps of N-trityl compound already described by Nakata and Tatsuta
(ref 5b).
(14) Brown, H. C.; Jadhav, P. K.; Bhat, K. S. J. Am. Chem. Soc. 1988,
110, 1535-1538.
Org. Lett., Vol. 9, No. 1, 2007
147

Scheme 4
of the carbamate at C7 with trichloroacetylisocyanate and
to control the stereogenic centers at C6, C7, C10, C11, and
C12. The synthesis of the C5-C15 fragment of herbimycin
A is shorter and more efficient than the synthesis of the same
fragment reported previously.5a,b Furthermore, the (Z)-double
bond of the (E,Z)-dienic moiety was formed, with excellent
control of configuration, by using ring-closing metathesis.
The total synthesis of herbimycin A was accomplished in
30 steps from 1 with a nonoptimized overall yield of 1.5%.
In terms of steps and overall yield, our synthesis of
herbimycin A was comparable to the shortest of the two
previous reported syntheses.5
K2CO3/MeOH followed by oxidation of the dimethoxy
aromatic core to the quinone ring system with CAN.
Herbimycin A was produced in 36% overall nonoptimized
yield for the last two steps (Scheme 5). The spectroscopic
Scheme 5. Synthesis of Herbimycin A
Acknowledgment. We thank Sanofi-Aventis and the
Centre National de la Recherche Scientifique (CNRS) for a
grant (S.C.). We are indebted to Prof. J. S. Panek (Boston
University) for providing spectra of natural and synthetical
herbimycin A. This paper is dedicated to the memory of Guy
Ourisson.
Supporting Information Available: General experimen-
tal procedure and characterization data of the described
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL062642I
(17) Synthesis of trisubstituted alkenes from lactols: (a) Smith, A. B.,
III; Brandt, B. M. Org. Lett. 2001, 3, 1685-1688. (b) Martin, D. D.; Marcos,
I. S.; Basabe, P.; Romero, R. E.; Moro, R. F.; Lumeras, W.; Rodriguez, L.;
Urones, J. G. Synthesis 2001, 1013-1022. Wittig reaction on unsaturated lac-
tol: Tulshian, D. B.; Fraser-Reid, B. J. Am. Chem. Soc 1981, 103, 474-475.
(18) The 1H NMR spectrum of our synthetical herbimycin A was identical
to those provided by Prof. Panek (natural and synthetical) and identical to
the spectrum of commercial herbimycin A purchased from Sigma-Aldrich.
Optical rotations of our herbimycin A sample and of the commercially
available one were respectively [R]20 +223 (c 0.14, CHCl3) and [R]20
and physical properties were identical in all respects with
those of natural herbimycin A purchased from Sigma-
Aldrich. Optical rotations of our herbimycin A sample and
of the commercially available one were respectively [R]20
D
+223 (c 0.14, CHCl3) and [R]20 +231 (c 0.05, CHCl3).18
D
D
D
+231 (c 0.05, CHCl3). These values are significantly higher than those
The synthesis of herbimycin A was achieved from the
commercially available Roche ester 1 by using allylmetals
described in the literature (ref 1a: [R]20 +137 (c 1.0, CHCl3); ref 5a:
D
[R]25 +122 (c 0.20, CHCl3); ref 5c: [R]23 +127 (c 0.15, CHCl3)).
D
D
148
Org. Lett., Vol. 9, No. 1, 2007