Synthesis of the fused polyether core of hemibrevetoxin B by two-directional ring-closing metathesis

Article abstract of DOI:10.1021/ol0630651

(Chemical Equation Presented) The tetracyclic fused polyether core of the marine natural product hemibrevetoxin B has been prepared in an efficient manner by using a strategy in which ring-closing metathesis (RCM) reactions were employed for ring synthesi

Full text of DOI:10.1021/ol0630651

ORGANIC
LETTERS
2007
Vol. 9, No. 6
1033-1036
Synthesis of the Fused Polyether Core
of Hemibrevetoxin B by Two-Directional
Ring-Closing Metathesis
J. Stephen Clark,*,† Damian M. Grainger,‡ Alexandra A.-C. Ehkirch,‡
Alexander J. Blake,‡ and Claire Wilson‡
WestCHEM, Department of Chemistry, Joseph Black Building, UniVersity of Glasgow,
UniVersity AVenue, Glasgow G12 8QQ, United Kingdom, and School of Chemistry,
UniVersity of Nottingham, UniVersity Park, Nottingham NG7 2RD, United Kingdom
stephenc@chem.gla.ac.uk
Received December 19, 2006
ABSTRACT
The tetracyclic fused polyether core of the marine natural product hemibrevetoxin B has been prepared in an efficient manner by using a
strategy in which ring-closing metathesis (RCM) reactions were employed for ring synthesis. Simultaneous construction of the A and D rings
was accomplished by double two-directional RCM of a tetraene.
Hemibrevetoxin B was first isolated from cultured cells of
the marine dinoflagellate Karenia breVis (previously known
as Gymnodinium breVe) by Shimizu and Prasad in 1989 and
is the smallest member of the fused polycyclic ether family
of marine natural products (Scheme 1).1 Even though it is
significantly smaller than the other brevetoxins and structur-
ally related compounds, such as the ciguatoxins and yesso-
toxins, hemibrevetoxin B is a formidable synthetic target
possessing four trans-fused cyclic ethers (A-D) and 10
stereogenic centers. It also presents many of the synthetic
challenges found in larger congeners such as brevetoxin B
(e.g., rings D-G and I-K). Although hemibrevetoxin B has
been a very popular synthetic target, only four rather lengthy
total syntheses and four formal syntheses have been pub-
lished to date.2,3 A short and highly efficient total synthesis
of hemibrevetoxin B has yet to be completed.
the D-ring methyl substituent gives ketone ii and discon-
nection of the D-ring hexadiene side chain and the oxygen-
containing substituents from the A-ring leads to the tetra-
cyclic intermediate iii. A double ring-closing metathesis
(RCM) disconnection then reveals the bicyclic tetraene iv.
Removal of the allyl group, the enone substituent, and the
terminal methylene of the vinyl group produces the bicyclic
triol v. Subsequent opening of the B-ring and acetal
formation leads to the C-ring diol vi. Further disconnection
of the methyl group and the hydroxypropyl side chain
delivers the simplified C-ring unit vii and a further RCM
(2) For total syntheses, see: (a) Nicolaou, K. C.; Reddy, K. R.; Skokotas,
G.; Sato, F.; Xiao, X.-Y. J. Am. Chem. Soc. 1992, 114, 7935-7936. (b)
Nicolaou, K. C.; Reddy, K. R.; Skokotas, G.; Sato, F.; Xiao, X.-Y.; Hwang,
C.-K. J. Am. Chem. Soc. 1993, 115, 3558-3575. (c) Kadota, I.; Jungyoul,
P.; Koumura, N.; Pollaud, G.; Matsukawa, Y.; Yamamoto, Y. Tetrahedron
Lett. 1995, 36, 5777-5780. (d) Morimoto, M.; Matsukura, H.; Nakata, T.
Tetrahedron Lett. 1996, 37, 6365-6368. (e) Kadota, I.; Yamamoto, Y. J.
Org. Chem. 1998, 63, 6597-6606. (f) Zakarian, A.; Batch, A.; Holton, R.
A. J. Am. Chem. Soc. 2003, 125, 7822-7824.
(3) For formal syntheses, see: (a) Yamamoto, Y.; Kadota, I. Bull. Soc.
Chim. Belg. 1994, 103, 619-628. (b) Mori, Y.; Yaegashi, K.; Furukawa,
H. J. Org. Chem. 1998, 63, 6200-6209. (c) Rainier, J. D.; Allwein, S. P.;
Cox, J. M. J. Org. Chem. 2001, 66, 1380-1386. (d) Fujiwara, K.; Sato,
D.; Watanabe, M.; Morishita, H.; Murai, A.; Kawai, H.; Suzuki, T.
Tetrahedron Lett. 2004, 45, 5243-5246.
Our retrosynthetic analysis of hemibrevetoxin B is shown
in Scheme 1. Removal of the enal-containing side chain from
the A-ring reveals the late-stage intermediate i. Cleavage of
† University of Glasgow.
‡ University of Nottingham.
(1) Prasad, A. V. K.; Shimizu, Y. J. Am. Chem. Soc. 1989, 111, 6476-
6477.
10.1021/ol0630651 CCC: $37.00
© 2007 American Chemical Society
Published on Web 02/21/2007

Scheme 1
disconnection gives the enone viii, which then suggests the
enolate was extremely problematic and poor yields of the
required product were obtained. To circumvent these prob-
lems, the ketone 6 was converted into the corresponding N,N-
dimethylhydrazone 7 (Scheme 2).8 Sequential deprotonation
of the hydrazone 7 with tert-butyllithium, alkylation with
3-benzyloxy-1-iodopropane, and subsequent hydrazone hy-
drolysis afforded the diastereomeric alkylated ketones 8 and
9 as a 1:1 mixture in reasonable yield. After partial separation
of the isomeric alkylation products, material enriched in
unrequired ketone 8 (1:2.5, 9:8) was epimerized by treatment
of the mixture with DBU in benzene at rt to give predomi-
nantly the required ketone 9 (4.5:1, 9:8). Treatment of the
ketone 9 with methylmagnesium bromide at low temperature
gave a tertiary allylic alcohol in excellent yield as a single
isomer. Alkene reduction and hydrogenolysis of the benzyl
ether were then accomplished simultaneously by treatment
of the tertiary allylic alcohol with hydrogen in the presence
of Pearlman’s catalyst. The required diol 10 was isolated as
a crystalline solid in 99% yield and its structure was
confirmed by X-ray crystallography.7
Installation of the B-ring was accomplished by using the
route shown in Scheme 3. Dehydration of the side chain in
the diol 10 to give the alkene 11 was achieved in a one-pot
fashion by conversion of the primary hydroxyl group into
the corresponding 2-nitrophenylselenide, treatment of this
selenide with hydrogen peroxide, and subsequent in situ
thermal elimination.9 Conversion of the hindered tertiary
alcohol 11 into the alkynyl ether 12 was achieved by using
Greene’s procedure10 in the manner previously described by
glucose acetal ix as a chiral pool starting material.
The central feature of our synthetic strategy was to be the
deployment of a double two-directional RCM reaction for
simultaneous construction of the A and D rings of hemi-
brevetoxin B from a precursor bearing an enone and an allyl
ether (iv f iii). In principle, a two-directional strategy could
be adopted for construction of the B and C rings, but for the
purposes of this study we chose to assemble this bicyclic
system sequentially. We have previously reported that double
two-directional RCM is a powerful method for the prepara-
tion of a variety of fused tricyclic systems possessing various
combinations of rings,4 and have recently applied iterative
two-directional RCM to the synthesis of pentacyclic F-J
fragments of the gambieric acids.4b However, prior to
embarking on this study we had not explored the feasibility
of using enones as reaction partners in two-directional RCM
reactions.
The starting material for our synthesis was the com-
mercially available glucose derivative 1 (Scheme 2). Perio-
date cleavage and immediate Wittig methylenation of the
resulting aldehyde afforded the alcohol 2, which was
converted into the RCM precursor 4 by etherification with
3-chloro-2-oxopropylidene triphenylphosphorane (3) and
reaction of the resulting stabilized phosphonium ylide with
formaldehyde under buffered conditions.5,6 Ring-closing
metathesis of the enone 4 using the ruthenium complex 5 (3
mol %) in dichloromethane at reflux provided the crystalline
oxepenone 6 in 94% yield.6 The structure of this compound
and stereochemical assignments were confirmed by X-ray
crystallography.7
(7) Crystallographic data (excluding structure factors) for the compounds
6, 10, 14, and 23 have been deposited (6 CCDC 626520; 10 CCDC 626521;
14 CCDC 626522; 23 CCDC 626523) with the Cambridge Crystallographic
Data Centre. Copies of the data can be obtained, free of charge, on
application to CCDC, 12 Union Road, Cambridge, CB2 1EZ, UK [fax +44-
(0) 1223 336033, e-mail deposit@ccdc.cam.ac.uk]. CIF files are available
as Supporting Information.
Introduction of a side chain by direct deprotonation of the
R,â-unsaturated ketone 6 and alkylation of the resulting
(4) (a) Clark, J. S.; Hamelin, O. Angew. Chem., Int. Ed. 2000, 39, 372-
374. (b) Clark, J. S.; Kimber, M. C.; Robertson, J.; McErlean, C. S. P;
Wilson, C. Angew. Chem., Int. Ed. 2005, 44, 6157-6162. (c) Clark, J. S.
Chem. Commun. 2006, 3571.
(5) Hudson, R. F.; Chopard, P. A. J. Org. Chem. 1963, 28, 2446-2447.
(6) Cossy, J.; Taillier, C.; Bellosta, V. Tetrahedron Lett. 2002, 43, 7263-
7266.
(8) Ulven, T.; Sørbye, K. A.; Carlsen, P. J. H. Acta Chem. Scand. 1997,
51, 1041-1044.
(9) Grieco, P. A.; Gilman, S.; Nishizawa, M. J. Org. Chem. 1976, 41,
1485-1486.
(10) Moyano, A.; Charbonnier, F.; Greene, A. E. J. Org. Chem. 1987,
52, 2919-2922.
1034
Org. Lett., Vol. 9, No. 6, 2007

Scheme 2
us for the synthesis of related systems.4,11 Ring-closing enyne
which had been obtained from the enyne metathesis reaction
already, in excellent yield.14
metathesis mediated by the ruthenium complex 5 then
afforded the tricyclic diene 13. Unfortunately, alkynyl ether
formation and the RCM reaction were complicated by the
propensity of the alkynyl ether to undergo a thermal retro-
ene reaction resulting in fragmentation to give an allylic ether
and ketene.12 Alkynyl ethers possessing highly branched
alkoxy groups (such as ours) show enhanced susceptibility
to the retro-ene reaction,12b and in our case, highly variable
yields were obtained depending on the reaction scale and
workup procedure employed.
To circumvent the retro-ene problem, an alternative
sequence was investigated. The diol 10 was first converted
into the crystalline lactone 14 by using the oxidative
procedure first described by Anelli and co-workers and
susbequently applied to a related system by Yamamoto and
co-workers.13 The structure of the lactone 14 was confirmed
by X-ray crystallography.7 Deprotonation of the lactone 14
and immediate enolate trapping with N-phenyltrifluoromethane-
sulfonimide provided the ketene acetal 15. Subsequent Stille
coupling of the triflate 15 with tributylvinyltin, using tetrakis-
(triphenylphosphine)palladium(0), afforded the diene 13,
Selective epoxidation of the diene at the electron-rich enol
ether site was accomplished by using a freshly prepared
solution of dimethyldioxirane at 0 °C. Subsequent regiose-
lective reduction of the resulting vinyl epoxide at the
anomeric position was performed by using lithium triethyl-
borohydride to afford the alcohol 16.15 At this stage, it was
necessary to invert the configuration at the hydroxyl-bearing
stereogenic center to give the alcohol 18; this operation was
accomplished by sequential Dess-Martin oxidation16 and
diastereoselective ketone reduction with sodium borohydride.
The synthesis of the complete tetracyclic framework of
hemibrevetoxin B was completed as shown in Scheme 4.
The alcohol 18 was converted into the corresponding allyl
ether by using standard etherification conditions. Removal
of the acetal provided the diol 19 in high yield and
subsequent formation of the bis-TBS ether followed by
selective deprotection of the primary hydroxyl site gave the
alcohol 20. Parikh-Doering oxidation17 followed by Wittig
methylenation of the resulting aldehyde gave the triene 21.
Removal of the TBS group, alkylation with the chloride 3,
Scheme 3
Org. Lett., Vol. 9, No. 6, 2007
1035

Scheme 4
and reaction of the resulting stabilized phosphonium ylide
with formaldehyde gave the double ring-closing metathesis
precursor 22 in modest yield.
alization of rings A and D to complete the total synthesis of
the natural product is in progress and the results of this work
will be reported in due course.
It was now possible to effect the key two-directional
double RCM reaction. Exposure of the tetraene 22 to the
Grubbs second generation ruthenium catalyst (5) resulted in
simultaneous RCM of both the enone and allylic ether with
their proximal vinyl groups and afforded the fused polyether
23 in good yield. The tetracyclic compound was obtained
as a crystalline solid and its structure and stereochemical
assignments were confirmed by X-ray crystallography.7
Elaboration of the fused tetracyclic compound 23 to give
hemibrevetoxin B by sequential D-ring alkylation via the
hydrazone, introduction of the D-ring methyl substituent, and
subsequent A-ring elaboration should be possible. Function-
Acknowledgment. We thank the EPSRC (Grant no. GR/
S15761/01) for directly funding this work and for funding
of the X-ray diffractometers. Financial support by Novartis
through the Novartis European Young Investigator Award
in Chemistry to J.S.C. and by Merck Research Laboratories
is also gratefully acknowledged.
Supporting Information Available: Spectroscopic and
other data for the key compounds 6, 10, 11, 13, 14, 18, 22,
and 23 plus X-ray data (CIF files) for compounds 6, 10, 14,
and 23. This material is available free of charge via the
Internet at http://pubs.acs.org.
(11) (a) Clark, J. S.; Elustondo, F.; Trevitt, G. P.; Boyall, D.; Robertson,
J.; Blake, A. J.; Wilson, C.; Stammen, B. Tetrahedron 2002, 58, 1973-
1982. (b) Clark, J. S.; Elustondo, F.; Kimber, M. C. Chem. Commun. 2004,
2470-2471.
(12) (a) Ficini, J. Bull. Soc. Chim. Fr. 1954, 1367-1371. (b) Moyano,
A.; Pericas, M. A.; Serratosa, F.; Valent´ı, E. J. Org. Chem. 1987, 52, 5532-
5538. (c) Moslin, R. M.; Jamison, T. F. J. Am. Chem. Soc. 2006, 128,
15106-15107.
(13) (a) Anelli, P. L.; Banfi, S.; Montanari, F.; Quici, S. J. Org. Chem.
1989, 54, 2970-2972. (b) Kadota, I.; Takamura, H.; Sato, K.; Yamamoto,
Y. J. Org. Chem. 2002, 67, 3494-3498.
OL0630651
(14) Nicolaou, K. C.; Sato, M.; Miller, N. D.; Gunzner, J. L.; Renaud,
J.; Untersteller, E. Angew. Chem., Int. Ed. Engl. 1996, 35, 889-891.
(15) Inoue, M.; Yamashita, S.; Tatami, A.; Miyazaki, K.; Hirama, M. J.
Org. Chem. 2004, 69, 2797-2804.
(16) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277-
7287.
(17) Parikh, J. R.; Doering, W. v. E. J. Am. Chem. Soc. 1967, 89, 5505-
5507.
1036
Org. Lett., Vol. 9, No. 6, 2007