Total synthesis of the marine toxin polycavernoside a via selective macrolactonization of a trihydroxy carboxylic acid [1]

Full text of DOI:10.1021/ja011256n

J. Am. Chem. Soc. 2001, 123, 8593-8595
Communications to the Editor
Scheme 1a
8593
Total Synthesis of the Marine Toxin Polycavernoside
A via Selective Macrolactonization of a Trihydroxy
Carboxylic Acid
James D. White,* Paul R. Blakemore, Cindy C. Browder,
Jian Hong, Christopher M. Lincoln, Pavel A. Nagornyy,
Lonnie A. Robarge, and Duncan J. Wardrop
Department of Chemistry, Oregon State UniVersity
CorVallis, Oregon 97331-4003
ReceiVed May 22, 2001
Naturally occurring macrolactones (macrolides) are found in
a wide variety of ring sizes, yet little is known of the propensity
of their precursor seco acids for forming a particular ring size
when more than one macrocyclization pathway is available.1 There
does, nevertheless, appear to be a preponderance of macrolides
with 14- and 16-membered rings, reflecting a bias by the synthase
which fabricates these systems toward rings of this size. Our plan
for the total synthesis of polycavernoside A (1), a macrolide
isolated by Yasumoto from the red alga PolycaVernosa tsudai
and shown to possess lethal toxic properties,2 incorporated a step
which hinged critically upon macrolactonization of a trihydroxy
carboxylic acid to create a 14-membered ring. We now report
the successful realization of a scheme leading to a total synthesis
of 1 which accomplishes this key reaction in highly selective
fashion.3
a (i) (i-Bu)2AlH, CH2Cl2, -78 °C f rt, 100%; (ii) (COCl)2, DMSO,
Et3N, CH2Cl2, -78 °C f rt, 96%; (iii) (S)-TBDPSOCH2CH(CH3)CH2-
SO2Ph (5), n-BuLi, THF, -50 °C; (iv) Dess-Martin periodinane, CH2Cl2,
rt; (v) SmI2, THF, -78 °C, 90% from 4; (vi) DDQ, CH2Cl2-H2O (10:
1), 0 °C f rt, 89%; (vii) Me4NBH(OAc)3, AcOH-MeCN (1:1), -20
°C; (viii) Me2C(OMe)2, PPTS (cat.), CH2Cl2, rt, 76% for 2 steps; (ix)
n-Bu4NF, THF, rt, 94%; (x) Dess-Martin periodinane, CH2Cl2, rt, 76%.
Scheme 2a
Our strategy from the outset envisioned assembly of the aglycon
of 1 from subunits which would be coupled first at the C9-C10
bond before closure of the macrocycle by lactonization. An
important feature of this route was concealment of the C9 ketone
and a C16 aldehyde, the latter for eventual elaboration of the
conjugated triene chain of 1, as alkenes until exposure via
ozonolysis at a late stage of the synthesis. Construction of the
C10-C16 portion of 1 commenced from (S)-pantolactone (2)
which was transformed to olefin 3 by the method of Mukaiyama
(Scheme 1).4 Reduction of the benzylidene acetal with DIBAL-H
gave a p-methoxybenzyl ether in which the liberated primary
alcohol was oxidized to aldehyde 4. Julia coupling of 4 with the
lithio anion of (S)-sulfone 5 afforded a mixture of stereoisomers
which were oxidized to a pair of keto sulfones. Reductive cleavage
of the sulfonyl substituent with samarium diiodide5 then gave 6
a (i) CH2dCHCH2MgBr, (-)-Ipc2BOMe, Et2O, -100 °C f -78 °C,
82%, 85% ee; (ii) TBSCl, imidazole, DMF, rt, 90%; (iii) O3, CH2Cl2,
-78 °C, then Ph3P, 93%; (iv) (4R)-Benzyl-3-propionyloxazolidin-2-one
(11), n-Bu2BOTf, Et3N -78 °C f rt; (v) MeNH(OMe)‚HCl, Me3Al,
CH2Cl2, 0 °C f rt, 75% from 10; (vi) TIPSOTf, 2,6-lutidine, CH2Cl2,
-78 °C f 0 °C, 100%; (vii) (i-Bu)2AlH, THF, -78 °C, 81%; (viii)
(CF3CH2O)2P(O)CH2CO2Me (14), KHMDS, 18-crown-6, THF, -78 °C
f rt, 94%; (ix) PPTS, MeOH-THF, 55 °C, then K2CO3, 70%; (x) Dess-
Martin periodinane, CH2Cl2, rt, 85%; (xi) MeCOC(N2)PO(OMe)2 (17),
K2CO3, MeOH, rt, 84%; (xii) 9-Br-9-BBN (19), CH2Cl2, 0 °C f rt, 77%.
as a single stereoisomer. Removal of the p-methoxybenzyl ether
from 6 and reduction of the resulting â-hydroxy ketone with
tetramethylammonium triacetoxyborohydride6 yielded the anti-
1,3-diol, subsequently protected as its acetonide 7. The primary
silyl ether of 7 was unmasked to expose a primary alcohol which
was oxidized to aldehyde 8.
(1) For a review of marine macrolides, see: Norcross, R. D.; Paterson, I.
Chem. ReV. 1995, 95, 2041.
(2) Yotsu-Yamashita, M.; Haddock, R. L.; Yasumoto, T. J. Am. Chem.
Soc. 1993, 115, 1147.
(3) For total syntheses of 1, see: (a) Fujiwara, K.; Murai, A.; Yotsu-
Yamashita, M.; Yasumoto, T. J. Am. Chem. Soc. 1998, 120, 10770. (b)
Paquette, L. A.; Barriault, L.; Pissarnitski, D.; Johnston, J. N. J. Am. Chem.
Soc. 2000, 122, 619.
Synthesis of the C1-C9 subunit of 1 began with the protected
aldehyde 9. This was first subjected to Brown asymmetric
(4) The synthesis of ent-3 from (R)-pantolactone has been reported very
recently: Shiina, I.; Shibata, J.; Ibuka, R.; Imai, Y.; Mukaiyama, T. Bull.
Chem. Soc. Jpn. 2001, 74, 113.
(5) Molander, G. A.; Hahn G. J. Org. Chem. 1986, 51, 1135.
(6) Evans, D. A.; Chapman, K. T.; Carreira, E. M. J. Am. Chem. Soc. 1988,
110, 3560.
10.1021/ja011256n CCC: $20.00 © 2001 American Chemical Society
Published on Web 08/10/2001

8594 J. Am. Chem. Soc., Vol. 123, No. 35, 2001
Communications to the Editor
Scheme 3a
a (i) CrCl2, NiCl2 (cat.), DMF, rt, 79%, dr ) 1:1; (ii) PPTS, MeOH, 35 °C, then NaOH, rt, 86%; (iii) 2,4,6-Cl3C6H2COCl, Et3N, THF, rt, then DMAP,
PhMe, ∆, 75%; (iv) TESOTf, 2,6-lutidine, CH2Cl2, 0 °C, 95%; (v) O3, CH2Cl2-MeOH (3:1), -78 °C, then Ph3P, -78 °C f rt, 81%; (vi) CHI3, CrCl2,
THF, 0 °C f rt, 76%; (vii) PPTS, MeOH-CH2Cl2 (4:1), rt, 88%; (viii) Dess-Martin periodinane, pyr-CH2Cl2, 0 °C, then HF‚pyr, THF, rt, 88%; (ix)
29, NBS, MeCN, -35 °C f rt, 27%; (x) DDQ, CH2Cl2-H2O (20:1), rt, 70%; (xi) 32, PdCl2(MeCN)2, DMF, rt, 74%.
allylation using (-)-diisopinylcampheylmethoxyborane (Scheme
2),7 and the resulting (S) homoallylic alcohol8 was protected as
its tert-butyldimethylsilyl ether before ozonolysis of the terminal
alkene furnished aldehyde 10. Condensation of 10 with the di-
n-butylboron enolate of (R)-4-benzyl-3-propionyloxazolidin-2-one
(11)9 afforded aldol product 12, a substance that could not be
purified without causing retroaldol fragmentation. Oxazolidinone
12 was therefore transformed directly to its easily purified
Weinreb amide by reaction with trimethylaluminum and N-
methoxymethylamine hydrochloride.10 After protection of the
secondary alcohol of this amide as its triisopropylsilyl ether,
reduction with DIBAL-H yielded aldehyde 13. Condensation of
13 with the Gennari-Still phosphonate 1411 gave the expected
Figure 1. Energy-minimized conformation (Spartan, PM3) of the
prelactonization complex from 22 and DMAP. The model is constrained
cis R,â-unsaturated ester 15, and subsequent removal of both tert-
to mimic a Burgi-Dunitz approach trajectory of the C15 hydroxyl at the
butyldimethylsilyl ethers under acidic conditions produced a diol
which underwent clean cyclization in the presence of methanolic
potassium carbonate to produce tetrahydropyran 16. Oxidation
of primary alcohol 16 to an aldehyde, followed by exposure to
Ohira’s reagent (17),12 furnished alkyne 18 which was reacted
with 9-bromo-9-borabicyclo[3.3.1]nonane (19)13 to give vinyl
bromide 20.
Coupling of aldehyde 8 with bromide 20 was carried out under
Nozaki-Hiyama-Kishi conditions14 and gave allylic alcohol 21
as a 1:1 mixture of epimers (Scheme 3). These were separated
by column chromatography, and the less polar (10S) alcohol was
treated with acidic methanol and then saponified to yield
trihydroxy carboxylic acid 22. Macrolactonization of 22 under
Yamaguchi conditions15 afforded exclusively 23 resulting from
closure at the C15 hydroxyl group, with no trace of lactones
activated acyl group.
derived from the hydroxyl substituents at either C10 or C13. The
structure of 23 was fully confirmed by X-ray crystallographic
analysis. A molecular modeling study of the cationic acyl DMAP
adduct derived from acid 22, a putative intermediate in the
lactonization, revealed that the reactive conformer leading to 23
is >8.5 kcal mol-1 lower in energy than alternative assemblies
leading to possible nine- and twelve-membered lactones (Figure
1). Although analogous theoretical treatment of the epimeric (10R)
acyl DMAP complex indicated a difference in energy between
competing lactonization pathways, the (10R) epimer corresponding
to trihydroxy acid 22 yielded only 46% of the 14-membered
lactone together with 16% of the 12-membered macrocycle
derived from lactonization at the C13 hydroxyl group.
Dihydroxy lactone 23 was protected as its bis-triethylsilyl ether
24 before ozonolysis to furnish keto aldehyde 25. A Takai
reaction16 of 25 with iodoform gave iodoalkene 26 in which the
C10 alcohol was unmasked selectively with acidic methanol.
Oxidation of R-hydroxy ketone 27 to a diketone, followed by
exposure to hydrogen fluoride, resulted in removal of both silyl
groups with concomitant cyclization to hemiketal 28. The latter
was identical in all respects to the substance prepared indepen-
dently by Murai3a and Paquette3b in the course of their syntheses
of 1.17 The C10 epimer of 23 was successfully elaborated to 28
(7) Racherla, U. S.; Brown, H. C. J. Org. Chem. 1991, 56, 401.
(8) Paterson, I; Wallace, D. J.; Gibson, K. R. Tetrahedron Lett. 1997, 38,
8911.
(9) Gage, J. R.; Evans, D. A. Organic Syntheses; John Wiley & Sons: New
York, 1993; Collect. Vol. 8, p 339.
(10) (a) Levin, J. I.; Turos, E.; Weinreb, S. M. Synth. Commun. 1982, 12,
989. (b) Evans, D. A.; Bender, S. L.; Morris, J. J. Am. Chem. Soc. 1988, 110,
2506.
(11) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405.
(12) (a) Ohira, S. Synth. Commun. 1989, 19, 561. (b) Mu¨ller, S.; Liepold,
B.; Roth, G. J.; Bestmann, H. J. Synlett 1996, 521.
(13) Hara, S.; Dojo, H.; Takinami, S.; Suzuki, A. Tetrahedron Lett. 1983,
24, 731.
(14) (a) Jin, H.; Uenishi, J.; Christ, W. J.; Kishi, Y. J. Am. Chem. Soc.
1986, 108, 5644. (b) Takai, K.; Tagashira, M.; Kuroda, T.; Oshima, K.;
Uchimoto, K.; Nozaki, H. J. Am. Chem. Soc. 1986, 108, 6048.
(15) Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. Bull.
Chem. Soc. Jpn. 1979, 52, 1989.
(16) Takai, K.; Nitta, K.; Utimoto, K. J. Am. Chem. Soc. 1986, 108, 7408.
(17) Data for 28: [R]22 -35.6 (c 0.09, CHCl3); lit.3a [R]22 -38 (c 0.12,
D
D
CHCl3).

Communications to the Editor
J. Am. Chem. Soc., Vol. 123, No. 35, 2001 8595
Acknowledgment. We are indebted to Professor Leo Paquette, Ohio
State University, for NMR spectra of 28 and to Professor Alexandre
Yokochi for the X-ray crystal structure of 23. Financial support was
provided by the National Institutes of Health (GM50574).
in 23% overall yield by the same route described for 23.
Glycosidation of alcohol 28 with 3-[4-O-benzyl-2,3-di-O-methyl-
R-L-fucopyranosyl]-2,4-di-O-methyl-1-thiophenyl-â-D-xylopyra-
nose (29)18 gave glycoside 30 as a single â-anomer, albeit in low
yield. Oxidative debenzylation of 30 followed by Stille coupling
of vinyl iodide 31 with dienylstannane 323b furnished polycav-
1
ernoside A (1), identical by comparison of the H NMR spectra
Supporting Information Available: Experimental procedures and
characterization data for all compounds; X-ray crystallographic data for
23; molecular modeling data and energy calculations for lactonization of
22 (PDF). This material is available free of charge via the Internet at
http://pubs.acs.org.
with those of the natural material.
In summary, a convergent and efficient route to polycaverno-
side A has been achieved which employs a remarkably selective
macrolactonization of trihydroxy carboxylic acid 22. Modeling
studies provide a plausible explanation for this selectivity.
(18) Disaccharide 29 is the thiophenyl anomer of the sugar used in Murai’s
synthesis of 1.3a
JA011256N