An efficient formal synthesis of (-)-clavosolide a featuring a "mismatched" stereoselective oxocarbenium reduction

Article abstract of DOI:10.1021/ol8028302

(Chemical Equation Presented) The enantioselective formal synthesis of the polyketide marine natural product (-)-clavosolide A is presented. The construction of the β-C-glycoside subunit is highlighted by a one-pot oxocarbenium cation formation/reduction

Full text of DOI:10.1021/ol8028302

ORGANIC
LETTERS
2009
Vol. 11, No. 3
769-772
An Efficient Formal Synthesis of
(-)-Clavosolide A Featuring a
“Mismatched” Stereoselective
Oxocarbenium Reduction
Jesse D. Carrick and Michael P. Jennings*
Department of Chemistry, The UniVersity of Alabama, Tuscaloosa, Alabama
35487-0336
jenningm@bama.ua.edu
Received December 8, 2008
ABSTRACT
The enantioselective formal synthesis of the polyketide marine natural product (-)-clavosolide A is presented. The construction of the ꢀ-C-
glycoside subunit is highlighted by a one-pot oxocarbenium cation formation/reduction sequence. Yamaguchi dimerization afforded the diolide
aglycon in 18 steps (longest linear sequence).
Secondary metabolites of marine origins continue to garner
significant attention from the synthetic, as well as, medicinal
community. The dimeric marine glycoside (-)-clavosolide
A and related structures were recently isolated from the
marine sponge M. claVosa in 2002 by Faulkner and co-
workers who initially proposed the relative and absolute
configuration of all clavosolides based on the extensive usage
of 2-D NMR techniques.1 Thus, the initial absolute config-
uration of the stereocenters surrounding the cyclopropyl
subunit were described as 9S,9′S,10S,10′S,11S,11′S as shown
in Scheme 1. A second disclosure by Erickson independently
confirmed the absolute and relative configuration of (-)-
clavosolides A and B as originally proposed by Faulkner
and also described the presence of two subsequent class
members, clavosolides C and D.2
from the synthetic community. The first synthesis of the
proposed clavosolide A structure was reported by the Willis
group. However, the synthesized compound was a diaste-
reomer of the originally proposed structure.3 Close inspection
of the cyclopropyl region of the 1H NMR spectrum revealed
the origin of the discrepancy between the synthesized
structure and that of the natural product. This observation
subsequently led to an amended structure (1) being proposed
by Willis with the 9S,9′S,10R,10′R,11R,11′R absolute ster-
eochemistry for the cyclopropane rings. The synthesis of 1
by Lee and co-workers yielded material that was spectro-
scopically identical to the originally described structure and
hence constituted the first total synthesis of 1.4 The initial
efforts of Chakraborty also yielded a synthesis of a diaste-
reomer of 1. However, a follow-up report provided the
This novel, highly functionalized 16-membered dimeric
polyketide macrocyclic has attracted considerable interest
(3) Barry, C. S.; Bushby, N.; Charmant, J. P. H.; Elsworth, J. D.;
Harding, J. R.; Willis, C. L. Chem. Commun. 2005, 40, 5097.
(4) Son, J. B.; Kim, S. N.; Kim, N. Y.; Lee, D. H. Org. Lett. 2006, 8,
661.
(1) Rao, M. R.; Faulkner, D. J. J. Nat. Prod. 2002, 65, 386.
(2) Erickson, K. L.; Gustafson, K. R.; Pannell, L. K.; Beutler, J. A.;
Boyd, M. R. J. Nat. Prod. 2002, 65, 1303.
10.1021/ol8028302 CCC: $40.75
Published on Web 01/06/2009
2009 American Chemical Society

Scheme 1
.
Original and Revised Structures of (-)-Clavosolide
Scheme 2. Retrosynthetic Analysis of (-)-Clavosolide A
A
correct structure.5 Similarly, Gurjar and co-workers synthe-
sized the incorrect monomeric subunit of 1.6 The efforts of
Smith confirmed the relative and absolute stereochemistry
of 1 through total synthesis and ultimately via X-ray
analysis.7 A revised total synthesis of 1 by the Willis group
also reconfirmed Smith’s preceding efforts.8 While 1 has no
reported biological activity, the polyfunctionality of the
monomeric subunit made it an attractive target to investigate
a “mismatched” oxocarbenium cation formation/reduction
protocol.9 We have successfully demonstrated this synthetic
strategy in a variety of previous total synthetic endeavors.10
Our retrosynthetic analysis of 1 followed previous discon-
nections in that the carbohydrate moiety, a permethylated
derivative of D-xylose, would be attached at the last step as
shown in Scheme 2. Completion of the dimerized aglycon 2
would be contingent upon a successful synthesis of the
monomeric subunit 3 which would be obtained from a bis-
Yamaguchi macrolactonization reaction process. Thus, seco
acid 3 would be derived from a functionalization of the
alkylation product of lactone 4 followed by a stereoselective
mismatched oxocarbenium cation formation/reduction se-
quence to afford the necessary ꢀ-C-glycoside precursor. We
envisaged the generation of lactone 4 from a functionalized
Evans’ aldol adduct which, in turn, would be produced from
the syn-dioxolane aldehyde 5. The required aldehyde 5 could,
in principle, be derived in eight steps from 6 by means of
an asymmetric acetate aldol reaction with (E)-crotonaldehyde.
Scheme 3. Synthesis of the ꢀ-Hydroxy Ketone 12
(5) (a) Chakraborty, K. T.; Reddy, V. R. Tetrahedron Lett. 2006, 47,
2099. (b) Chakraborty, T. K.; Reddy, V. R.; Chattopadhyay, A. K.
Tetrahedron Lett. 2006, 47, 7435.
(6) Yakambram, P.; Puranik, V.; Gurjar, M. Tetrahedron Lett. 2006,
22, 3781.
(7) Simov, V.; Smith, A. B., III. Org. Lett. 2006, 8, 3315.
(8) Barry, C. S.; Elsworth, J. D.; Seden, P. T.; Bushby, N.; Harding,
J. R.; Alder, R. W.; Willis, C. L. Org. Lett. 2006, 8, 3319.
(9) Lewis, M. D.; Cha, J. K.; Kishi, Y. J. Am. Chem. Soc. 1982, 104,
4976.
With this initial retrosynthetic plan in mind, focus was
placed on the synthesis of the ꢀ-hydroxy-ꢀ,γ-unsaturated
ketone 12. As shown in Scheme 3, the synthesis of the
required aldol adduct 6 relied on the treatment of the
oxazolidinethione 7 with TiCl4 to afford the necessary enolate
which upon quenching with E-crotonaldehyde (8) at -78
(10) (a) Jennings, M. P.; Clemens, R. T. Tetrahedron Lett. 2005, 46,
2021. (b) Ding, F.; Jennings, M. P. Org. Lett. 2005, 7, 2321. (c) Clemens,
R. T.; Jennings, M. P. Chem. Commun. 2006, 2720. (d) Sawant, K. B.;
Jennings, M. P. J. Org. Chem. 2006, 71, 7911. (e) Sawant, K. B.; Ding, F.;
Jennings, M. P. Tetrahedron Lett. 2006, 47, 939. (f) Sawant, K. B.; Ding,
F.; Jennings, M. P. Tetrahedron Lett. 2007, 48, 5177. (g) Ding, F.; Jennings,
M. P. J. Org. Chem. 2008, 73, 5965. (h) Martinez-Solorio, D.; Jennings,
M. P. Org. Lett. 2009, 11, 189.
770
Org. Lett., Vol. 11, No. 3, 2009

°C provided the desired aldol adduct in 85% yield as a 10:1
ratio of inseparable diastereomers as described by Phillips.11
An ensuing transamidation of 6 to the Weinreb amide under
standard condition of the MeN(H)OMe·HCl salt and imida-
zole furnished chiral amide 9 in 91% yield.12 With amide 9
in hand, a directed Simmons-Smith cyclopropanation under
the Charette protocol afforded cyclopropyl carbinol 10 in
91% over two steps from 6 as a 20:1 ratio of diastereomers.13
A subsequent Mitsunobu inversion of the free hydroxyl
moiety of 10, parallel to the efforts of Smith, afforded the
correct C9 acetate 11 in 74% yield as a 12:1 mixture of
diastereomers. Careful treatment of 11 with excess allyl
magnesium bromide concomitantly alkylated the Weinreb
amide moiety and deprotected the secondary alcohol to afford
the necessary ꢀ-hydroxy-ꢀ,γ-unsaturated 12 in 61% over the
two transformations within the same reaction flask. With the
necessary substrate for directed reduction in hand, our focus
turned toward the construction of the necessary dioxolane
aldehyde 5.
Scheme 4. Synthesis of the ꢀ-C-Glycoside 19
Hence, treatment of ketone 12 under the method of Prasad
(Et2BOMe and NaBH4) afforded the required 1,3-syn-diol
as a 10:1 mixture of inseparable diastereomers.14 The crude
diol was then dissolved in CH2Cl2 and treated under standard
ketalization conditions with DMP and PPTS to afford the
acetonide-protected alkene. An ensuing buffered ozonolysis
of the olefinic moiety afforded dioxolane aldehyde 5 in 61%
yield over three steps from 12 and set the stage for a reagent-
controlled Evans aldol reaction.15 Thus, stereoselective (Z)-
enolate formation of the propionyl oxazolidinone 13 was
accomplished under the standard Evans conditions (n-
Bu2BOTf, Et3N); quenching with aldehyde 5 readily afforded
aldol adduct 14 in 86% yield (12:1 dr by 1H NMR) as shown
in Scheme 4. Ensuing protection of the resulting secondary
alcohol 14 was attempted under a variety of conditions.
Treatment of 14 under standard protection conditions with
silyl chlorides and imidazole led to no conversion, even with
heating. The use of silyl triflates and 2,6-lutidine, unfortu-
nately, did not allow for silicon ether formation. However,
the protection of 14 was successful with MOMCl; it afforded
acetal 15 in 91% yield after extensive optimization. Conver-
sion of the oxazolidinone aldol adduct 15 into the desired
lactone 4 was initiated by initially transforming 15 to the
benzyl ester followed by hydrolysis of the acetonide and
subsequent lactonization.16 Along this line, the lithium
alkoxide of benzyl alcohol led to the formation of the
corresponding ester which was used immediately without
purification. Subsequent lactonization was accomplished
under standard hydrolysis conditions by treating the crude
product with aqueous trifluoroacetic acid (TFA) in THF. The
hydroxy lactone 4 was obtained in 66% yield from 15.
With the synthesis of 4 complete, focus shifted toward
the construction of the ꢀ-C-glycoside subunit. Hence, alky-
lation of lactone 4 with allylmagnesium bromide afforded
the intermediate hemiketal 16 which readily underwent
tandem stereoselective oxocarbenium cation formation/
reduction with TFA and Et3SiH to afford the ꢀ-C-glycoside
(as determined by NOE) 19 in 65% yield over the three step
sequence from 4 (in >20:1 dr). Presumably, reductions of
oxocarbenium cations occur via axial addition of Et3SiH to
afford the ꢀ-C-glycoside.17 Of the two possible reactive
conformers, and based on the isolated ꢀ-C-glycoside, the
proposed conformer 18 places all of the substituents at C2,
C3, and C5 in the pseudoequatorial positions. A majority of
our stereoselective endocyclic oxocarbenium reductions have
placed the C3 hydroxyl moiety in the axial position and the
C5 substituent in the pseudo equatorial geometry. Typically,
these oxocarbenium formation/reduction reactions are com-
(11) Guz, N. R.; Phillips, A. J. Org. Lett. 2002, 4, 2253.
(12) Nahm, S.; Weinreb, S. M. Tetrahedron Lett. 1981, 22, 3815. (b)
Rychnovsky, S. D.; Sinz, C. J. Tetrahedron. 2002, 58, 6561. (c) Sibi, M. P.
Org. Prep. Proced. Int. 1993, 25, 15. (d) Mentzel, M.; Hoffman, H. M. R.
J. Prakt. Chem. 1997, 339, 517.
(13) (a) Charette, A.; Label, H. J. Org. Chem. 1995, 60, 2966. (b)
Charette, A. B.; Beauchemin, A. Org. React. 2001, 58, 1.
(14) Chen, K-M.; Hardtmann, G. E.; Prasad, K.; Repic, O.; Shapiro,
M. J. Tetrahedron Lett. 1987, 28, 155.
(15) (a) Evans, D. A.; Glorius, F.; Burch, J. D. Org. Lett. 2005, 7, 3331.
(b) Evans, D. A.; Trenkle, W. C.; Zhang, J.; Burch, J. D. Org. Lett. 2005,
7, 3335.
(16) (a) Damon, R. E.; Coppola, G. M. Tetrahedron Lett. 1990, 31, 2849.
(b) Evans, D. A.; Britton, T. C.; Ellman, J. A. Tetrahedron Lett. 1987, 28,
6141. (c) Prashad, M.; Har, D.; Kim, H.-Y.; Repic, D. Tetrahedron Lett.
1998, 39, 7067.
(17) Romero, J. A. C.; Tabacco, S. A.; Woerpel, K. A. J. Am. Chem.
Soc. 2000, 122, 168.
Org. Lett., Vol. 11, No. 3, 2009
771

specific case regarding the reduction of the oxocarbenium
cation 18, no possibility existed that would allow for such a
silylation process due to the protected pseudoequatorial C3
hydroxyl moiety. Interestingly, no silicon ether formation
at the C9 alcohol of the ansa chain was observed. Thus, the
concomitant directing effect and intramolecular silicon
capture seems to be confined to hydroxyl functionality
exclusively on the six-membered ring.10
Scheme 5. Synthesis of 2 via a Yamaguchi Dimerization
With the functionalized ꢀ-C-glycoside in hand, our focus
turned toward completion of the formal synthesis of 1 as
delineated in Scheme 5. Consequently, buffered ozonolysis
of the olefin moiety resident in 19 afforded aldehyde 20 in
77% yield. Subsequently, a Pinnick oxidation provided the
necessary seco acid 3 in 95% yield which was used without
further purification.18 Dimerization of the monomeric acid
3 was completed under the previously reported Yamaguchi
macrolactonization conditions to afford the protected glycon
22 in a modest 42% yield.19 Final, deprotection of the bis-
MOM acyclic acetals was accomplished using bromocatechol
borane to afford the aglycon 2 in a 44% yield over two
deprotection steps, thus constituting a formal synthesis of
1.20 The spectral data (1H NMR, 600 MHz; 13C NMR, 125
MHz), optical rotation, and HRMS data of synthetic 2 were
in complete agreement with those previously reported.7,8
Given the challenges of diastereoselective glycosylation of
2 with 24, the final reaction was not attempted.7,8
In summary, an efficient formal synthesis of 1 has been
successfully achieved in 18 transformation sequences (23
total reactions) with an overall yield of ∼1% from com-
mercially available (E)-crotonaldehyde. Additionally, we
have observed a “mismatched” stereoselective oxocarbenium
reduction to afford the ꢀ-C-glycoside subunit leading to an
efficient synthesis of 2.
Acknowledgment. Support for this project was provided
by the University of Alabama. We thank Cody B. Smith
(UA) and William A. Kilgo (UA) for the synthesis of chiral
auxiliary 7.
plete at -78 f -40 °C within 0.25-0.5 h. Based on
Woerpel’s observations and our prior investigations, these
C3 axial and C5 pseudoequatorial oxocarbenium conforma-
tions represent a “matched” geometry. However, the reactive
oxocarbenium conformer 18 places the C3 and C5 substituents
in the pseudoequatorial positions. Interestingly, the stereo-
selective oxocarbenium formation/reduction reaction required
>12 h for completion at -20 °C. Hence, it is suggested that
conformer 18 represents a “mismatched” geometry that
nonetheless allowed for a highly stereoselective oxocarbe-
nium reduction.
Supporting Information Available: Full experimental
procedures and spectroscopic data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL8028302
(18) Overman, L. E.; Paone, D. V. J. Am. Chem. Soc. 2001, 123, 623.
(19) Yamaguchi, M. A.; Katsuki, T.; Sacki, H.; Hirata, K.; Inanaga, J.
Bull. Chem. Soc. Jpn. 1979, 52, 1989.
Previous efforts in our group have demonstrated the ability
to silylate the axial alcohol moiety in the C3 position of a
six membered endocyclic oxocarbenium cation.10 In this
(20) Paquette, L. A.; Sun, L-Q.; Friedrich, D.; Savage, P. B. J. Am.
Chem. Soc. 1997, 119, 8438.
772
Org. Lett., Vol. 11, No. 3, 2009