Total synthesis and revision of C6 stereochemistry of (+)-amphidinolide W

Article abstract of DOI:10.1021/jo052181z

An enantioselective first total synthesis and structural revision of the cytotoxic natural product amphidinolide W is described. We initially investigated a ring-closing metathesis based synthetic strategy to form the 12-membered macrocycle. This strategy was unsuccessful as it led to formation of a 17-membered macrocycle. Subsequently, we explored an alternative strategy that involved cross-metathesis followed by a Yamaguchi macrolactonization reaction sequence utilizing the same key intermediates. This strategy led to the synthesis of amphidinolide W. The synthesis was carried out in a convergent manner, and four of the five stereogenic centers in amphidinolide W were set by asymmetric synthesis. The synthesis features Sharpless asymmetric dihydroxylation, diastereoselective alkylation, efficient cross-metathesis of functionalized substrates, and novel functional group transformations using selective lipase-catalyzed hydrolysis of the primary acetate group. Of particular note, the C6 absolute stereochemistry of amphidinolide W has now been revised through our synthesis.

Full text of DOI:10.1021/jo052181z

Total Synthesis and Revision of C6 Stereochemistry of
(+)-Amphidinolide W
Arun K. Ghosh* and Gangli Gong
Departments of Chemistry and Medicinal Chemistry, Purdue UniVersity, West Lafayette, Indiana 47907,
and Department of Chemistry, UniVersity of Illinois at Chicago, Chicago, Illinois 60607
akghosh@purdue.edu
ReceiVed October 19, 2005
An enantioselective first total synthesis and structural revision of the cytotoxic natural product
amphidinolide W is described. We initially investigated a ring-closing metathesis based synthetic strategy
to form the 12-membered macrocycle. This strategy was unsuccessful as it led to formation of a 17-
membered macrocycle. Subsequently, we explored an alternative strategy that involved cross-metathesis
followed by a Yamaguchi macrolactonization reaction sequence utilizing the same key intermediates.
This strategy led to the synthesis of amphidinolide W. The synthesis was carried out in a convergent
manner, and four of the five stereogenic centers in amphidinolide W were set by asymmetric synthesis.
The synthesis features Sharpless asymmetric dihydroxylation, diastereoselective alkylation, efficient cross-
metathesis of functionalized substrates, and novel functional group transformations using selective lipase-
catalyzed hydrolysis of the primary acetate group. Of particular note, the C6 absolute stereochemistry of
amphidinolide W has now been revised through our synthesis.
Introduction
Amphidinolides are a group of structurally unique macrolides
obtained from marine dinoflagellates of the genus Amphidinium,
which are symbionts of Okinawan marine acoel flatworms
Amphiscolops sp.1 Amphidinolides are extremely scarce, and
as a result, subsequent biological studies have been very limited.
In many cases, progress toward complete structural assignments
has been hampered because of lack of supply.2 The mechanisms
of action of amphidinolides have not been established due to
its scarce natural abundance. The unique structural features,
structural resemblence among members of the family, and potent
antitumor properties have generated considerable interest in their
synthesis.3-5 Amphidinolide W (1b, Figure 1) was isolated by
FIGURE 1. Amphidinolide W: proposed structure 1a and revised
structure 1b.
Kobayashi in 2002 from strain Y-42 of the genus Amphidinium
dinoflagellate.6 It is a new cytotoxic 12-membered macrolide.
(4) (a) Williams, D. R.; Kissel, W. S. J. Am. Chem. Soc. 1998, 120,
11198. (b) Williams, D. R.; Meyer, K. G. J. Am. Chem. Soc. 2001, 123,
765. (c) Williams, D. R.; Myers, B. J.; Mi, L. Org. Lett. 2000, 2, 945. (d)
Trost, B. M.; Papillon, J. P. N. J. Am. Chem. Soc. 2004, 126, 13618.
(5) (a) Ghosh, A. K.; Liu, C. J. Am. Chem. Soc. 2003, 125, 2374. (b)
Furstner, A.; Aissa, C.; Riveiros, R.; Ragot, J. Angew. Chem., Int. Ed. 2002,
41, 4763. (c) Aiessa, C.; Riveiros, R.; Ragot, J.; Fuerstner, A. J. Am. Chem.
Soc. 2003, 125, 15512. (d) Colby, E. A.; O’Brien, K.; Jamison, T. F. J.
Am. Chem. Soc. 2004, 126, 998. (e) Lepage, O.; Kattnig, E.; Fuerstner, A.
J. Am. Chem. Soc. 2004, 126, 15970. (f) Colby, E. A.; O’Brien, K. C.;
Jamison, T. F. J. Am. Chem. Soc. 2005, 127, 4297.
* Corresponding author. Phone: 765-494-5323. Fax: 765-496-1612.
(1) Kobayashi, J.; Shimbo, K.; Kubota, T.; Tsuda, M. Pure Appl. Chem.
2003, 75, 337.
(2) Kobayashi, J.; Tsuda, M. Nat. Prod. Rep. 2004, 21, 77.
(3) (a) Lam, H. W.; Pattenden, G. Angew. Chem., Int. Ed. 2002, 41,
508. (b) Maleczka, R. E.; Terrell, L. R.; Geng, F.; Ward, J. S. Org. Lett.
2002, 4, 2841. (c) Trost, B. M.; Chisholm, J. D.; Wrobleski, S. T.; Jung,
M. J. Am. Chem. Soc. 2002, 124, 12420. (d) Trost, B. M.; Harrington, P.
E. J. Am. Chem. Soc. 2004, 126, 5028.
10.1021/jo052181z CCC: $33.50 © 2006 American Chemical Society
Published on Web 12/24/2005
J. Org. Chem. 2006, 71, 1085-1093
1085

Ghosh and Gong
SCHEME 1. Synthesis of Enoate 14
cyclization reaction to form the 12-membered macrolactone ring
for 1a. The syn-diol functionality along with the trisubstituted
E-olefin would be derived from optically active lactone 5. The
syn-diol stereochemistry was planned to be introduced by
Sharpless asymmetric dihydroxylation8 of readily available
dienoate 6. The subunit 4 would be prepared by a Horner-
Wadsworth-Emmons reaction9 between ketophosphonate 7 and
aldehyde 8a. Phosphonate 7 was planned to be derived from
diastereoselective methylation of the acyl oxazolidinone 9a.10
Ring-Closing Metathesis-Based Strategy. As shown in
Scheme 1, synthesis of C10-C17 fragment 3 started with an
ortho ester Claisen rearrangement of penta-1,4-dienol 10 using
triethyl orthoacetate to provide ethyl (E)-hepta-4,6-dienoate 6.11
Sharpless asymmetric dihydroxylation8 of 6 furnished the
γ-lactone 11 in 43% yield along with the other regioisomer 12,
which was obtained in 30% yield.12 Both of these isomers were
readily separated by silica gel chromatography. Mosher ester
anlysis of lactone 11 revealed the enantiomeric purity of 11 to
be 94% ee.13 Protection of alcohol 11 using TIPSOTf gave the
silyl ether 13. Methylation of the lactone 13 resulted in
methylated product 5 as a major diastereomer (ratio 11:1). The
diastereomers were readily separated by column chromatography
and the major trans-lactone 5 was isolated in 81% yield. The
trans relationship between the two substituents on the five-
membered-ring lactone 5 was confirmed by NOESY. Lactone
5 was reduced by DIBAL to the corresponding hemiacetal.
Wittig reaction of the resulting hemiacetal with (carbethoxy-
ethylidene)phosphorane provided the E-trisubstituted olefin 3
as a single isomer in 92% yield for two steps. The resulting
FIGURE 2. Retrosynthesis of proposed amphidinolide W (1a) via
RCM-based strategy.
Kobayashi and co-workers elucidated the gross structure of 1a
on the basis of the spectroscopic data including 13C-13C
correlation studies. The absolute configurations at C11, C12,
and C14 of 1a were assigned by J-based configuration analysis
of a modified Mosher ester. The configurations at C2 and C6
were elucidated from NMR data of MTPA esters of a reduction
product of amphidinolide W. Amphidinolide W exhibited
cytotoxicity against murine lymphoma L1210 cells in vitro with
an IC50 value of 3.9 µg/mL.6 Recently, we reported the first
total synthesis of amphidinolide W and revised the C6 absolute
stereochemistry of the proposed amphidinolide W structure
(1a).7 We employed a combination of synthesis and NMR
spectroscopy as tools to determine the correct structure of natural
amphidinolide W (1b). Herein, we provide a complete account
of our synthetic studies.
Results and Discussion
Our initial synthetic plan is outlined in Figure 2. Strategic
disconnection of the proposed amphidinolide W (1a) at the C9-
C10 double bond and a choice of the C12-MOM protecting
group provide the appropriate precursor 2 for the synthesis of
proposed amphidinolide W structure 1a. We planned to assemble
amphidinolide W by ring-closing metathesis followed by
removal of the C12-MOM protecting group. The acyclic ester
precursor 2 could be derived from esterification of an ap-
propriately functionalized alcohol derived from intermediates
3 and an acid derived from 4. We elected to install the C15-
C18 diene functionality prior to macrocyclization. The synthons
3 and 4 can also be utilized for our alternative synthetic strategy
in which the C9-C10 double bond will be introduced first by
a cross-metathesis reaction followed by an appropriate macro-
(8) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. ReV.
1994, 94, 2483.
(9) (a) Horner, L.; Hoffmann, H.; Wippel, J. H. G.; Klahre, G. Ber. 1959,
92, 2499. (b) Wadsworth, W. S., Jr.; Emmons, W. D. J. Am. Chem. Soc.
1961, 83, 1733.
(10) Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem. Soc. 1982,
104, 1737.
(11) Baillie, L.; Batsanov, A.; Bearder, J.; Whiting, D. J. Chem. Soc.,
Perkin Trans. 1 1998, 20, 3471.
(6) Shimbo, K.; Tsuda, M.; Izui, N.; Kobayashi, J. J. Org. Chem. 2002,
67, 1020.
(7) Ghosh, A. K.; Gong, G. J. Am. Chem. Soc. 2004, 126, 3704.
(12) Evans, P. A.; Murthy, V. S. Tetrahedron Lett. 1999, 40, 1253.
(13) Dale, J. A.; Dull, D. L.; Mosher, H. S. J. Org. Chem. 1969, 34,
2543.
1086 J. Org. Chem., Vol. 71, No. 3, 2006

Total Synthesis of (+)-Amphidinolide W
SCHEME 2. Synthesis of Triene 18
SCHEME 4. Synthesis of Acid 25
SCHEME 3. Synthesis of Enone 22
methylphosphonate to the amide 20 at -78 °C afforded
ketophosphonate 7 in excellent yield. For preparation of
aldehyde 8a, commercial methyl (R)-3-hydroxy-2-methylpro-
pionate 21 was protected as the TIPS ether. DIBAL reduction
of the resulting ester provided aldehyde 8a.17 The Horner-
Wadsworth-Emmons reaction between ketophosphonate 7 and
aldehyde 8a using barium hydroxide as the base furnished R,â-
unsaturated enone 22 without any epimerization.18
Enone 22 was converted to acid 25 as shown in Scheme 4.
Selective conjugated reduction of enone 22 using a Red-Al and
CuBr complex at -20 °C provided 1,4-reduction product 4 in
90% yield along with 5% of the 1,2-reduction byproduct.19 This
byproduct was readily separated by silica gel chromatography.
Deprotection of the TIPS ether in 4 with TBAF furnished
primary alcohol 23 and hemiketal 24 as a mixture. Exposure of
this mixture to PDC in DMF afforded keto acid 25 in 83%
isolated yield.
With both acid 25 and alcohol 18 in hand, we investigated
the esterification and subsequent ring-closing metathesis strat-
egy. As shown in Scheme 5, esterification of alcohol 18 and
acid 25 in the presence of DCC and DMAP gave the ester 2.
We then examined ring-closing metathesis,20 and the results are
summarized in Table 1. As can be seen, there was no reaction
when the first generation Grubbs’ catalyst (26) was utilized
(entry 1). Addition of Lewis acid Ti(OiPr)4 with catalyst 26
provided the 17-membered ring macrolactone 28 in 60% yield
(entry 2).21 Furthermore, the second-generation Grubbs’ catalyst
(27)22 provided the same 17-membered ring macrolactone 28
alcohol was then protected as MOM ether to afford 14 in 94%
yield.
Introduction of the C15-C18-conjugated diene functionality
was carried out as shown in Scheme 2. DIBAL reduction of
ester 14 afforded alcohol 15. It was converted to allylic bromide
16 by exposure to triphenylphosphine and carbon tetrabromide.14
Reaction of bromide 16 with tributylphosphine provided the
corresponding phosphonium salt. Treatment of this salt with
tBuOK formed the corresponding ylid. Reaction of the resulting
ylid with propionaldehyde furnished the conjugated olefin 17.
Only E-olefin 17 was isolated, in 75% yield for three steps (from
15). Removal of the TIPS protecting group provided the alcohol
18.
Synthesis of the C1-C9 fragment 5 is outlined in Scheme 3.
Diastereoselective methylation of known imide 9a15 with
NaHMDS and MeI afforded 19a as the major diastereomer in
87% yield (ratio 15:1 by 1H NMR). Evans and co-workers have
previously demonstrated the stereochemical outcome of such
diastereoselective alkylation process.10 Hydrolysis of 19a with
lithium hydroperoxide provided the corresponding acid in
quantitative yield. The resulting acid was converted to Weinreb
amide 20 in 92% yield.16 Addition of the lithiated dimethyl
(17) Ishiwata, A.; Sakamoto, S.; Noda, T.; Hirama, M. Synlett 1999, 6,
692.
(18) Paterson, I.; Yeung, K.; Watson, C.; Ward, R. A.; Wallace, P. A.
Tetrahedron 1998, 54, 11935.
(19) For a general reaction protocol, see: Semmelhack, M. F.; Stauffer,
R. D.; Yamashita, A. J. Org. Chem. 1977, 42, 3180.
(20) (a) Furstner, A. Angew. Chem., Int. Ed. 2000, 39, 3012. (b) Grubbs,
R. H.; Miller, S. J.; Fu, G. C. Acc. Chem. Res. 1995, 28, 446. (c) Grubbs,
R. H.; Chang, S. Tetrahedron 1998, 54, 4413. (d) Maier, M. E. Angew.
Chem., Int. Ed. 2000, 39, 2073.
(21) For the effect of Lewis acid on RCM, see: (a) Ghosh, A. K.;
Cappiello, J.; Shin, D. Tetrahedron Lett. 1998, 39, 4651. (b) Ghosh, A. K.;
Bilcer, G. Tetrahedron Lett. 2000, 41, 1003.
(22) Scholl, M.; Ding, S.; Lee, C.; Grubbs, R. H. Org. Lett. 1999, 1,
953.
(14) Hayashi, H.; Nakanishi, K.; Brandon, C.; Marmur J. Am. Chem.
Soc. 1973, 95, 8749.
(15) Rivero, R. A.; Greenlee, W. J. Tetrahedron Lett. 1991, 32, 2453.
(16) Nahm, S.; Weinreb, S. M. Tetrahedron Lett. 1981, 22, 3815.
J. Org. Chem, Vol. 71, No. 3, 2006 1087

Ghosh and Gong
SCHEME 6. Preparation of the Substrates for
SCHEME 5. Ring-Closing Metathesis of 2
Cross-Metathesis
TABLE 1. RCM Strategy Study
entry
catalyst
additive
product (yield, %)
1
2
3
26
26
27
no
no reaction
28 (60)
28 (82)
cross-metathesis of substrates derived from 3 and 4. The second-
generation Grubbs’ catalyst 27 is generally superior to first-
generation 26 for the cross-metathesis reactions.23 However, we
found that the conjugated diene in 2 was vulnerable to the
second-generation Grubbs’ catalyst in the ring-closing metathesis
reaction, which led to the formation of 17-membered macro-
lactone 28 in excellent yield. Therefore, we elected to modify
our substrate for the cross-metathesis. Since the trisubstituted
internal olefin in R,â-unsaturated ester 3 is much more inert
toward the second-generation Grubbs’ catalyst for steric and
electronic reasons, one can expect the cross-metathesis reaction
involving electron-rich monosubstituted terminal olefins in 3
and 4 will proceed preferentially.
Ti(OiPr)4
no
However, the choice of the protecting group in substrate 3
may be critical to yields and olefin geometry. We, therefore,
planned to investigate cross-metathesis involving enoate 3 and
its derivatives (31-33) with different protecting groups. Prepa-
ration of various substrates for cross-metathesis is outlined in
Scheme 6. Removal of the TIPS in 3 followed by protecting
the diol as the acetonide gave 31. Removal of the TIPS in 14
provided alcohol 32. Protection of 32 with acetic anhydride and
triethylamine provided acetate 33. We planned to protect the
ketone in 4 as its dioxolane derivative and examine cross
metathesis with this substrate. Thus, 4 was reacted with ethylene
glycol and triethyl orthoformate in the presence of PTSA to
give dioxolane 34 in 90% yield.25
As shown in Scheme 7, using olefin 34 and a range of C10-
C17 enoates (3, 31-33) in hand, we investigated cross-
metathesis reactions under a variety of reaction conditions, and
the results are shown in Table 2. Cross-metathesis of 3 and 34
provided diene 35 with a very good E/Z ratio of 10:1; however,
the reaction yield (25%) was far from satisfactory (entry 1).
The presence of the bulky allylic TIPS group in 3 may have
been the reason for this inefficient reaction. To examine the
effect of a hydroxyl group or other protecting group, we
removed the TIPS group in 3 to provide the alcohol 32.
However, no cross-metathesis product was obtained between
FIGURE 3. Retrosynthesis of 1a via macrolactonization-based
strategy.
in 82% yield (entry 3). No desired 12-membered ring lactone
29 resulting from ring-closing metathesis of terminal olefins
was obtained. Formation of the 17-membered ring is definitely
more favorable than the 12-membered ring due to both enthalpic
and entropic reasons.20e It is possible to form a 12-membered
macrolactone involving a methathesis substrate that lacks the
C15-C18 diene functionality. However, in our synthetic
strategy, we planned to avoid installation of this functionality
at a late stage of the synthesis. Therefore, we elected to explore
an alternative strategy utilizing the same synthons 3 and 4.
Cross-Metathesis and Macrolactonization-Based Strategy.
We then turned to a cross-metathesis23 and macrolactonization
sequence for assembly of the macrolide 1a. As shown in Figure
3, we planned to form the 12-membered macrolactone by
Yamaguchi macrolactonization24 of acyclic seco acid 30a. The
C9-C10 trans-olefin in this seco acid would be installed by
(23) (a) Connon, S. J.; Blechert, S. Angew. Chem., Int. Ed. 2003, 42,
1900. (b) Blackwell, H. E.; O’Leary, D. J.; Chatterjee, A. K.; Washenfelder,
R. A.; Bussmann, D. A.; Grubbs, R. H. J. Am. Chem. Soc. 2000, 122, 58.
(c) Chatterjee, A. K.; Morgan, J. P.; Scholl, M.; Grubbs, R. H. J. Am. Chem.
Soc. 2000, 122, 3783. (d) Chatterjee, A. K.; Choi, T.; Sanders, D. P.; Grubbs,
R. H. J. Am. Chem. Soc. 2003, 125, 11360.
(24) Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. Bull.
Chem. Soc. Jpn. 1979, 52, 1989.
(25) Caserio, F. F., Jr.; Roberts, J. D. J. Am. Chem. Soc. 1958, 80, 5837.
1088 J. Org. Chem., Vol. 71, No. 3, 2006

Total Synthesis of (+)-Amphidinolide W
SCHEME 7. Cross-Metathesis Reactions of Various
Substrates
SCHEME 8. Synthesis of Seco Acid 30a
TABLE 2. Results of Cross Metathesis Studies
entry
olefins
product
yield (%)
E/Z
1
2
3
4
3 + 34
35
36
37
38
25
0
73
86
10:1
N/A
2:1
32 + 34
31 + 34
33 + 34
11:1
32 and 34 (entry 2). Allylic alcohol in 32 may have sequestered
the Ru-carbene complex through strong chelation with the
hydroxyl group. Appropriate protection of the allylic alcohol
was deemed necessary for effective cross-metathesis reaction.
Cross-metathesis of isopropylidene derivative 31 and dioxolane
34 proceeded with good yield; however, the E/Z selectivity was
only 2:1 (entry 3). To improve the E/Z ratio, acetyl and MOM
groups in 33 were investigated. As shown, cross-metathesis of
33 and 34 provided product 38 in excellent yield (86%) and
very good E/Z selectivity (up to 11:1). The E/Z mixture could
not be separated by silica gel chromatography. Therefore, the
mixture was used for the subsequent reaction.
tively providing the primary alcohol 42 in excellent yield (89%
over three steps).26 Oxidation of alcohol 42 with PDC in DMF
furnished the corresponding acid 43. Saponification of the
secondary acetate group with K2CO3 in MeOH furnished seco
acid 30a in 67% yield over two steps.
With the optimized cross-metathesis product 38 in hand, we
then focused on the synthesis of seco acid 30a (Figure 3) for
macrolactonization. As shown in Scheme 8, reduction of 38 by
DIBAL afforded the readily separable E-diol 39 in 88% yield.
It was converted to allylic bromide 40 in a four step sequence
involving the following: (1) regioselective protection of the
primary alcohol with pivaloyl chloride and pyridine; (2)
protection of the secondary alcohol with TIPSOTf; (3) depro-
tection of the pivalate with DIBAL; (4) conversion of the
resulting allylic alcohol to bromide 40 with carbon tetrabromide
and triphenylphosphine.14 Bromide 40 was isolated in 85%
overall yield. Reaction of this bromide with tributylphosphine
provided the corresponding tributylphosphonium salt. Wittig
olefination of this salt with propionaldehyde in the presence of
KOtBu in THF/Benzene provided exclusively E-olefin 41 in
90% yield. Selective removal of the primary TIPS group was
unsuccessful. Therefore, both TIPS groups were removed using
TBAF to furnish the corresponding diol. Protection of the diol
with acetic anhydride and triethylamine afforded the corre-
sponding diacetate. Selective basic hydrolysis of the primary
acetate was attempted using LiOH, Ba(OH)2, or aqueous KCN.
None of the conditions was effective as both acetates were
hydrolyzed under the reaction conditions. However, lipase PS-
30 catalyzed hydrolysis of the diacetate proceeded very selec-
Synthesis of the Proposed Structure of Amphidinolide W
(1a). With seco acid 30a in hand, we then proceeded to form
the 12-membered macrolactone. As shown in Scheme 9,
Yamaguchi macrolactonizaion24 of 30a afforded two diastere-
omeric macrolactones 44a and 44c as a 3:1 mixture. Presumably,
these lactones were formed due to base-catalyzed epimerization
of the C2 stereogenic center during the reaction. Our many
attempts to prevent this epimerization under different macro-
lactonization conditions including DCC/DMAP as reported by
Borden and Keck27 as well as acid-catalyzed macrolactonization
protocol reported by Trost and Chisholm28 were unsuccessful.
The isomers were readily separated by silica gel chromatogra-
phy. Treatment of the major isomer 44a with PPTS in aqueous
acetone removed the dioxolane group. Subsequent removal of
the MOM group with BF3‚OEt2 furnished the presumed natural
amphidinolide W (1a). However, 1H and 13C NMR spectra for
1a did not match with those reported for natural amphidinolide
(26) (a) Ghosh, A. K.; Chen, Y. Tetrahedron Lett. 1995, 36, 505. (b)
Ghosh, A. K.; Kincaid, J. F.; Walters, D. E.; Chen, Y.; Chaudhuri, N. C.;
Thompson, W. J.; Culberson, C.; Fitzgerald, P. M. D.; Lee, H. Y.; McKee,
S. P.; Munson, P. M.; Duong, T. T.; Darke, P. L.; Zugay, J. A.; Schleif, W.
A.; Axel, M. G.; Lin, J.; Huff, J. R. J. Med. Chem. 1996, 39, 3278.
(27) Borden, E. P.; Keck, G. E. J. Org. Chem. 1985, 50, 2394.
(28) Trost, B. M.; Chisholm, J. D. Org. Lett. 2002, 4, 3743.
J. Org. Chem, Vol. 71, No. 3, 2006 1089

Ghosh and Gong
SCHEME 9. Synthesis of Proposed Amphidinolide W (1a)
SCHEME 10. Synthesis of Amphidinolide W (1b) and Its
Isomer 1d
W. Furthermore, deprotection of both dioxolane and MOM
groups from the minor isomer 44c afforded amphidinolide W
diastereomer 1c. Unfortunately, neither 1H nor 13C NMR spectra
for 1c were identical with those of the natural product.6 It should
be noted that removal of dioxolane and MOM protecting groups
proceeded in 50% yield in a two-step sequence, and no
epimerization was observed under the reaction conditions.
Structural Revision of Amphidinolide W. A thorough
1
examination of H and 13C NMR spectra of 1a and 1c and
comparison of reported spectra6 for natural amphidinolide W
revealed that the discrepancies of chemical shifts are mostly
1
located in the ring instead of the side chain. The H and 13C
THF at 0 °C; DBU in CH2Cl2 or in MeOH) resulted in the
decomposition of the starting material. While it is possible to
epimerize the C-6 stereocenter in 5, we elected to synthesize
6-R-stereoisomer diastereoselectively to avoid any ambiguity.
Therefore, for introduction of R-stereochemistry at C6, diaste-
reoselective methylation was carried out with oxazolidinone
derivative 9b (enantiomer of 9a). Stereochemical outcome of
such alkylation process is now well established.10 As shown in
Scheme 10, deprotonation of 9b with NaHMDS followed by
alkylation with MeI afforded 19b as the major diastereomer in
87% yield (ratio 12:1 by 1H NMR). The alkylated oxazolidinone
derivative 19b was converted to seco acid 30b following the
reactions sequence described for 30a. Yamaguchi macrolac-
tonization24 of seco acid 30b as described previously furnished
two diastereomeric macrolactones, 44b and 44d, as a mixture
(1:1 ratio) in 50% yield. Both isomers were separated by silica
gel chromatography. Our many attempts to improve this ratio
were unsuccessful. Removal of dioxolane and MOM protecting
groups from diastereomer 44b provided synthetic amphidinolide
W (1b) in 50% yield for two steps. The spectral data (1H and
NMR chemical shifts of the C12-C20 side-chain region of
isomers 1a and 1c matched very closely with the reported values
of the natural amphidinolide W. However, there were significant
discrepancies of chemical shifts in the C2-C11 ring region,
1
particularly H NMR chemical shifts at C2-C8 and the 13C
NMR chemical shifts at C2, C4, C5, C9, and C11.6 The actual
structure of amphidinolide W appeared to be epimeric to the
proposed structure at one or more stereocenters in the ring
system. The 2S-configuration of natural amphidinolide W was
deduced from NMR data of 1,5,11,12-tetrakis-MTPA esters of
a reductive product of amphidinolide W.6 The absolute stere-
ochemistry at C6 was elucidated to be the S configuration on
the basis of NMR data of 6,11,12-tris-MTPA esters of the C6-
C20 segment obtained by the Baeyer-Villiger reaction of
amphidinolide W.6 Among all the stereocenters, the C2 and C6
configuration assignments were more liable to error since neither
of the stereocenters is in close proximity to other stereocenters.
As a consequence, no stereochemical correlation could be
posited to conclusively assign the relative and absolute stere-
ochemistry. Since we had already prepared both C2 epimers
(1a and 1c) and neither of them corresponded to natural
amphidinolide W, we hypothesized that the C6 stereocenter of
originally proposed amphidinolide W structure may require
revision.
13C NMR) for 1b ([R]23 +8.1, c 1, CHCl3)7 were identical
D
with those reported for the natural amphinolide W.6 Deprotection
of dioxolane and MOM protecting groups of diastereomer 44d
furnished the C2 diastereomer of amphidinolide W (1d). The
stereochemical identifications at the C2 positions of 1b and 1d
were established by NOESY as shown in Figure 4. A significant
NOE effect was observed between HA and HB protons in 1b.
On the other hand, there was no observable NOE between HA
and HB protons in 1d.
Thus, we turned to synthesis of the C6-epimer (R-configu-
ration) of the originally proposed structure of amphidinolide
W (1a). Our attempted direct epimerization of C6 stereocenter
using a variety of conditions (KOtBu in THF at 0 °C; KH in
1090 J. Org. Chem., Vol. 71, No. 3, 2006

Total Synthesis of (+)-Amphidinolide W
SCHEME 13. Enantioselective Synthesis of Diastereomer 1d
FIGURE 4. NOE effect of diastereomes 1b and 1d.
SCHEME 11. Effect of Dioxolane on Macrocyclization
pure seco acid 30d as described above. Yamaguchi macrolac-
tonization of 30d afforded a single macrolactone 44d in 85%
yield. There was no epimerization at the C2 stereogenic center.
Removal of dioxolane and MOM protecting groups from
macrolactone 44d afforded 1d as a single isomer. These
experiments further provided evidence in support of the assign-
ment of C2 and C6 absolute stereochemistry of amphidinolide
W structure 1b.
SCHEME 12. Enantioselective Synthesis of Diastereomer 1c
Conclusion
In summary, we have achieved the first total synthesis of
amphidinolide W (1b) along with three diastereomers (1a, 1c,
1d), thus leading to a revision of the structure 1a originally
proposed for natural amphidinolide W. The convergent synthesis
features Sharpless asymmetric dihydroxylation, diastereoselec-
tive alkylation, efficient cross metathesis of functionalized
substrates, and novel functional group transformations using
selective lipase-catalyzed hydrolysis of a primary acetate group.
Of particular note, the C6 absolute stereochemistry of amphi-
dinolide W (1b) has now been revised through our synthesis.
Also, optical rotation of amphidinolide W was reported for the
first time after total synthesis. It is important to note that
amphidinolide W and its diastereomers are very unstable and
prone to decomposition even when they were stored at low
temperature.
We then examined the effect of the dioxolane protecting
group on macrolactonization. As shown in Scheme 11, treatment
of 44b with PPTS in aqueous acetone resulted in the removal
of the dioxolane group. The resulting keto seco acid 45 was
subjected to Yamaguchi macrolactonization. This condition also
provided a 1:1 mixture of macrolactones. Removal of the MOM
group by exposure to BF3‚OEt2 and Me2S afforded 1b and 1d
in 30% yield for two steps. Therefore, it is evident that the
dioxolane ring on 44b did not play any role for epimerization
of the C2 stereogenic center during the macrocyclization.
To further confirm the C2 stereochemistry of amphidinolide
W diastereomers 1c and 1d, we synthesized the corresponding
enantiomerically pure seco acids and investigated Yamaguchi
macrolactonization. As shown in Scheme 12, following the same
procedures for the synthesis of 30a, enantiomeric aldehyde 8b
was converted to enantiomerically pure seco acid 30c. Interest-
ingly, cyclization of 30c under the same Yamaguchi condition
furnished a single macrolactone 44c in excellent yield. There
was no epimerizaion at C2. Further deprotection of dioxolane
and MOM groups furnished 1c as a single isomer.
Experimental Section
(2S,6S)-((3S,4S,6R,7E,9E)-4-(Methoxymethoxy)-6,8-dimeth-
yldodeca-1,7,9-trien-3-yl)-2,6-dimethyl-5-oxodec-9-enoate (2). To
a solution of alcohol 18 (137 mg, 0.51 mmol) and acid 25 (110
mg, 0.52 mg) in CH2Cl2 (4 mL) was added DCC (118 mg, 0.57
mmol) followed by DMAP (7 mg, 0.05 mmol). The mixture was
stirred at room temperature overnight, filtered, and concentrated
in vacuo. Purification by flash column (5% EtOAc/hexanes)
afforded ester 2 (182 mg, 77%) as a clear oil: [R]23D -5.3 (c 1.25,
CHCl3); IR (neat) 1736, 1714, 1043; 1H NMR (500 MHz, CDCl3)
δ 6.00 (d, J ) 16.0 Hz, 1H), 5.86 (ddd, J ) 17.0, 11.0, 6.0 Hz,
1H), 5.75 (ddd, J ) 17.0, 10.5, 6.5 Hz, 1H), 5.59 (dt, J ) 15.5, 6.5
Hz, 1H), 5.39 (dd, J ) 6.0, 5.0 Hz, 1H), 5.28 (d, J ) 17.5 Hz,
1H), 5.26 (d, J ) 11.0 Hz, 1H), 5.05 (d, J ) 9.5 Hz, 1H), 5.00 (d,
J ) 17.0 Hz, 1H), 4.96 (d, J ) 10.0 Hz, 1H), 4.64 (AB, JAB ) 6.5
Hz, ∆νAB ) 26.0 Hz, 2H), 3.50 (m, 1H), 3.38 (s, 3H), 2.70 (m,
1H), 2.45-2.54 (m, 4H), 2.09 (m, 2H), 2.01 (m, 2H), 1.88 (m,
1H), 1.76 (m, 2H), 1.72 (s, 3H), 1.51 (m, 1H), 1.38 (m, 2H), 1.18
(d, J ) 7.0 Hz, 3H), 1.06 (d, J ) 7.0 Hz, 3H), 1.00 (t, J ) 7.5 Hz,
As shown in Scheme 13, we also synthesized 1d from the
corresponding seco acid 30d. Thus, enantiomeric aldehyde 8b
and acyl oxazolidinone 19b were converted to enantiomerically
J. Org. Chem, Vol. 71, No. 3, 2006 1091

Ghosh and Gong
3H), 0.96 (d, J ) 6.5 Hz, 3H); 13C NMR (125 MHz, CDCl3) δ
213.7, 174.9, 138.0, 135.7, 133.7, 133.3, 129.8, 118.0, 115.1, 97.5,
77.9, 75.0, 56.0, 45.6, 39.1, 38.8, 38.7, 31.9, 31.4, 28.9, 27.2, 25.9,
21.8, 17.4, 16.4, 13.9, 12.7; HRMS (ESI) [M + Na]+ calcd for
C28H46O5Na 485.3243, found 485.3267.
) 15.4, 6.5 Hz, 1H), 5.43 (dd, J ) 15.4, 7.1 Hz, 1H), 5.31 (dd, J
) 6.4, 5.4 Hz, 1H), 4.63 (m, 2H), 4.18 (q, J ) 7.1 Hz, 2H), 3.91
(m, 4H), 3.43-3.52 (m, 3H), 3.37 (s, 3H), 2.72 (m, 1H), 2.17 (m,
1H), 2.05 (s, 3H), 1.97 (m, 1H), 1.83 (s, 3H), 1.50-1.70 (m, 8H),
1.29 (t, J ) 7.2 Hz, 3H), 1.24 (m, 1H), 1.08 (m, 1H), 1.04, (m,
21H), 1.01 (d, J ) 6.8 Hz, 3H), 0.91 (d, J ) 7.0 Hz, 3H), 0.89 (d,
J ) 6.7 Hz, 3H); 13C NMR (125 MHz, CDCl3) δ 170.3, 168.6,
147.2, 136.5, 127.8, 124.7, 114.3, 97.6, 78.5, 75.8, 69.0, 68.9, 65.6,
65.5, 60.9, 56.3, 39.7, 38.6, 36.7, 31.8, 31.1, 30.7, 30.1, 26.9, 21.6,
20.9, 18.5, 17.3, 14.7, 14.5, 12.9, 12.4; HRMS (ESI) [M + Na]+
calcd for C38H70O9SiNa 721.4687, found 721.4673.
Macrolactone (44a). To a solution of hydroxy acid 30a (23 mg,
0.046 mmol) in THF (2 mL) were added diisopropylethylamine
(0.32 mL, 1.85 mmol) and 2,4,6-trichlorobenzoyl chloride (0.14
mL, 0.93 mmol). The reaction was stirred at room temperature
overnight. The mixture was diluted by addition of benzene (8 mL)
and added slowly to a solution of DMAP (283 mg, 2.3 mmol) in
benzene (50 mL) at 80 °C by syringe pump over 8 h. The syringe
was rinsed with benzene (2 mL), and the solution was combined
to the reaction mixture to stir for an additional 1 h. After being
cooled to room temperature, the reaction mixture was poured into
saturated aqueous NaHCO3 solution, and the aqueous solution was
extracted with EtOAc twice. The combined organic layers were
RCM Product (28). To a solution of 2 (25 mg, 0.054 mmol) in
DCM (27 mL) was added second-generation Grubbs’ catalyst 27
(4.6 mg, 0.0054 mmol). The reaction was heated to 45 °C for 2 h
and then cooled to room temperature. The mixture was concentrated
in vacuo and purified by column chromatography (10% EtOAc/
hexanes) to afford 28 (18 mg, 82%) as a clear oil: [R]23 +92 (c
D
1
1.0, CHCl3); IR (neat) 1737, 1709, 1041; H NMR (500 MHz,
CDCl3) δ 5.90 (d, J ) 16.0 Hz, 1H), 5.88 (ddd, J ) 17.0, 10.5, 5.0
Hz, 1H), 5.49 (m, 1H), 5.30 (ddd, J ) 15.0, 10.0, 5.0 Hz, 1H),
5.26 (d, J ) 17.5 Hz, 1H), 5.25 (d, J ) 10.0 Hz, 1H), 5.02 (d, J )
10.0 Hz, 1H), 4.74 (AB, JAB ) 7.0 Hz, ∆νAB ) 34.5 Hz, 2H), 3.42
(s, 3H), 3.34 (m, 1H), 2.59 (m, 1H), 2.48 (m, 1H), 2.32-2.39 (m,
3H), 2.16 (m, 1H), 1.93 (m, 2H), 1.76 (m, 1H), 1.67 (s, 3H), 1.58
(m, 1H), 1.49 (m, 1H), 1.38 (m, 2H), 1.13 (d, J ) 6.5 Hz, 3H),
1.03 (d, J ) 7.0 Hz, 3H), 0.99 (d, J ) 6.5 Hz, 3H); 13C NMR (125
MHz, CDCl3) δ 214.4, 175.4, 137.0, 135.8, 132.2, 131.9, 128.5,
117.4, 96.8, 78.2, 73.4, 55.9, 46.1, 42.0, 39.2, 39.1, 33.3, 32.8, 29.4,
29.0, 22.3, 18.8, 17.0, 13.0; HRMS (ESI) [M + Na]+ calcd for
C24H38O5Na 429.2617, found 429.2614.
washed with pH 4 buffer (aqueous solution of NaH2PO4
+
NaHSO4), dried over Na2SO4, and concentrated in vacuo. Purifica-
tion by flash column (15% EtOAc/Hexanes) afforded the macro-
lactone 44a (7.5 mg, 35%) as a clear oil along with the diastereomer
Cross-Metathesis Product (35). To a mixture of 34 (120 mg,
0.30 mmol) and 3 (60 mg, 0.15 mmol) in CH2Cl2 (1 mL) was added
second-generation Grubbs’ catalyst 27 (7 mg, 0.008 mmol). The
reaction was heated to 45 °C for 15 h and then cooled to room
temperature. The mixture was concentrated in vacuo and purified
by column chromatography (5-10% EtOAc/Hexanes) to afford 35
(28 mg, 25%, inseparable mixture of 10:1 E/Z isomers) as a clear
oil along with the dimer of 34 (100 mg) and recovered 3 (42 mg):
1
44c (2.5 mg, 12%): IR (neat) 1720; H NMR (500 MHz, CDCl3)
δ 6.14 (dt, J ) 16.0, 7.2 Hz, 1H), 6.04 (d, J ) 15.4 Hz, 1H), 5.64
(dt, J ) 15.8, 7.0 Hz, 1H), 5.61 (dd, J ) 15.4, 6.4 Hz, 1H), 5.15
(dd, J ) 8.1, 7.8 Hz, 1H), 5.09 (d, J ) 9.2 Hz, 1H), 4.66 (d, J )
6.6 Hz, 1H), 4.60 (d, J ) 6.6 Hz, 1H), 3.88 (m, 4H), 3.72 (m, 1H),
3.36 (s, 3H), 2.74 (m, 1H), 2.62 (m, 1H), 2.12 (m, 3H), 1.92 (m,
1H), 1.75 (s, 3H), 1.50-1.76 (m, 5H), 1.43 (m, 2H), 1.34 (m, 1H),
1.26 (m, 1H), 1.06 (d, J ) 7.0 Hz, 3H), 1.04 (t, J ) 7.5 Hz, 3H),
0.98 (d, J ) 6.7 Hz, 3H), 0.86 (d, J ) 6.7 Hz, 3H); 13C NMR (125
MHz, CDCl3) δ 176.1, 140.1, 136.3, 134.2, 133.7, 130.1, 128.7,
113.2, 97.9, 78.3, 78.0, 65.7, 65.0, 56.2, 40.4, 39.6, 36.7, 35.7, 30.0,
29.4, 28.9, 27.2, 26.3, 22.2, 14.3, 14.1, 13.8, 13.1; HRMS (ESI)
[M + Na]+ calcd for C28H46O6Na 501.3192, found 501.3193.
Proposed Amphidinolide W (1a). To a solution of macrolactone
44a (7.5 mg, 0.016 mmol) in acetone/H2O (3 mL/0.2 mL) was
added PPTS (79 mg, 0.31 mmol). The mixture was heated to 40
°C and stirred overnight. After being cooled to room temperature,
the mixture was poured into H2O and extracted with CH2Cl2 three
times. The combined organic layers were dried over MgSO4 and
concentrated in vacuo. Purification on silica gel (15% EtOAc/
hexanes) provided ketolactone (5.9 mg, 87%): 1H NMR (500 MHz,
CDCl3) δ 6.01 (d, J ) 15.7 Hz, 1H), 5.64 (dt, J ) 15.6, 6.5 Hz,
1H), 5.56 (dt, J ) 15.7, 6.9 Hz, 1H), 5.46 (dd, J ) 15.6, 7.4 Hz,
1H), 5.11 (dd, J ) 7.4, 7.2 Hz, 1H), 5.02 (d, J ) 9.6 Hz, 1H), 4.66
(d, J ) 6.5 Hz, 1H), 4.59 (d, J ) 6.6 Hz, 1H), 3.51 (m, 1H), 3.36
(s, 3H), 2.71 (m, 1H), 2.41-2.50 (m, 4H), 2.25 (m, 1H), 2.12 (m,
2H), 2.08 (m, 2H), 1.93 (m, 1H), 1.74 (s, 3H), 1.61 (m, 1H), 1.44
(m, 1H), 1.38 (m, 1H), 1.25 (m, 1H), 1.17 (d, J ) 6.9 Hz, 3H),
1.04 (d, J ) 7.0 Hz, 3H), 1.03 (t, J ) 7.5 Hz, 3H), 0.96 (d, J ) 6.7
Hz, 3H); 13C NMR (125 MHz, CDCl3) δ 214.2, 175.4, 137.0, 136.2,
134.2, 133.7, 130.1, 128.0, 98.0, 78.2, 75.7, 56.2, 45.6, 41.7, 40.0,
39.1, 32.5, 30.1, 29.5, 27.3, 26.3, 22.2, 18.1, 17.5, 14.3, 13.1.
To a solution of above product (5.9 mg, 0.01 mmol) in CH2Cl2
(2 mL) was added dimethyl sulfide (1 mL) followed by BF3‚OEt2
(17 µL, 0.14 mmol) at -20 °C. After 15 min at this temperature,
the reaction was quenched with saturated aqueous NaHCO3 solution
and warmed to room temperature. The aqueous layer was extracted
with CH2Cl2 twice. The combined organic layers were dried over
MgSO4 and concentrated in vacuo. Purification by preparative TLC
(20% EtOAc/hexanes) provided pure 1a (3.1 mg, 60%) as a
1
IR (neat) 3500, 1712, 1099; H NMR (500 MHz, CDCl3) δ 6.46
(d, J ) 8.5 Hz, 1H), 5.59 (dt, J ) 15.5, 6.5 Hz, 1H), 5.38 (dd, J
) 15.5, 8.5 Hz, 1H), 4.18 (q, J ) 7.0 Hz, 2H), 3.87-3.93 (m,
5H), 3.44-3.54 (m, 2H), 3.22 (m, 1H), 2.89 (m, 1H), 2.17 (m,
1H), 1.97 (m, 1H), 1.88 (s, 3H), 1.37-1.72 (m, 8H), 1.29 (t, J )
7.0 Hz, 3H), 1.26 (m, 1H), 1.01-1.64, (m, 46H), 0.92 (d, J ) 6.5
Hz, 3H), 0.90 (d, J ) 6.5 Hz, 3H); 13C NMR (125 MHz, CDCl3)
δ 168.5, 147.1, 134.7, 130.0, 127.4, 113.9, 78.7, 73.1, 68.6, 65.3,
60.4, 39.7, 39.4, 36.4, 31.7, 30.6, 30.4, 29.9, 29.7, 26.6, 20.9, 18.2,
18.1, 16.9, 14.3, 14.1, 12.5, 12.4, 12.0; HRMS (ESI) [M + Na]+
calcd for C43H84O7Si2Na 791.5654, found 791.5660.
Cross-Metathesis Product (36). To a mixture of 34 (127 mg,
0.32 mmol) and 31 (45 mg, 0.16 mmol) in CH2Cl2 (1 mL) was
added second-generation Grubbs’ catalyst 27 (8 mg, 0.009 mmol).
The reaction was heated to 45 °C for 15 h and then cooled to room
temperature. The mixture was concentrated in vacuo and purified
by column chromatography (5-10% EtOAc/hexanes) to afford 36
(76 mg, 73%, inseparable mixture of 2:1 E/Z isomers) as a clear
oil along with the dimer of 34 (79 mg): IR (neat) 1712, 1462,
1106; 1H NMR (500 MHz, CDCl3) δ 6.45 (m, 1H), 5.66-5.77 (m,
1H), 5.35 (m, 1H), 4.18 (m, 2H), 3.83-3.91 (m, 5H), 3.43-3.53
(m, 3H), 2.80 (m, 1H), 1.94-2.38 (m, 2H), 1.85 (s, 3H), 1.50-
1.70 (m, 8H), 1.39 (s, 3H), 1.36 (s, 3H), 1.29 (m, 3H), 1.01-1.15
(m, 26H), 0.89-0.91 (m, 6H); 13C NMR (125 MHz, CDCl3) major
isomer δ 168.4, 146.4, 137.1, 127.5, 126.4, 113.9, 108.1, 82.9, 78.4,
68.6, 65.3, 65.2, 60.5, 39.5, 38.5, 36.3, 31.5, 30.7, 30.6, 30.4, 27.4,
26.9, 26.6, 20.7, 18.1, 16.9, 14.3, 14.2, 12.5, 12.0; HRMS (ESI)
[M + Na]+ calcd for C37H68O7SiNa 675.4632, found 675.4628.
Cross-Metathesis Product (38). To a mixture of 34 (369 mg,
0.927 mmol) and 33 (145 mg, 0.442 mmol) in CH2Cl2 (1 mL) was
added second-generation Grubbs’ catalyst 27 (34 mg, 0.04 mmol).
The reaction was heated to 45 °C for 15 h and then cooled to room
temperature. The mixture was concentrated in vacuo and purified
by column chromatography (15% EtOAc/hexanes) to afford 38 (262
mg, 85%, inseparable mixture of 11:1 E/Z isomers) as a clear oil
1
along with the dimer of 34 (191 mg): IR (neat) 1742, 1711; H
colorless oil: [R]23 +27 (c 0.23, CHCl3); IR (neat) 3468, 1729;
D
NMR (500 MHz, CDCl3) δ 6.46 (d, J ) 10.1 Hz, 1H), 5.74 (dt, J
1H NMR (400 MHz, CDCl3) δ 6.04 (d, J ) 15.4 Hz, 1H), 5.64 (dt,
1092 J. Org. Chem., Vol. 71, No. 3, 2006

Total Synthesis of (+)-Amphidinolide W
J ) 15.6, 6.6 Hz, 1H), 5.56 (dt, J ) 15.3, 6.6 Hz, 1H), 5.50 (dd,
J ) 15.9, 6.6 Hz, 1H), 5.01 (d, J ) 9.8 Hz, 1H), 4.86 (dd, J ) 7.2,
7.2 Hz, 1H), 3.60 (m, 1H), 2.84 (m, 1H), 2.45 (m, 4H), 2.24 (m,
1H), 2.13 (m, 2H), 2.01 (m, 2H), 1.92 (m, 1H), 1.78 (s, 3H), 1.68
(m, 1H), 1.39 (m, 2H), 1.28 (m, 1H), 1.19 (d, J ) 6.8 Hz, 3H),
1.04 (t, J ) 7.5 Hz, 3H), 1.04 (d, J ) 7.0 Hz, 3H), 0.96 (d, J ) 6.8
Hz, 3H); 13C NMR (100 MHz, CDCl3) δ 213.8, 175.3, 136.9, 135.5,
133.8, 133.5, 129.8, 127.6, 77.3, 70.5, 45.4, 41.3, 40.4, 38.6, 32.1,
30.2, 28.9, 26.5, 25.9, 21.8, 17.9, 17.3, 13.9, 12.7; HRMS (ESI)
[M + Na]+ calcd for C24H38O4Na 413.2668, found 413.2651.
Macrolactone (44b). According to the same procedures for
preparation of 44a from acid 30a, acid 30b (110 mg, 0.22 mmol)
was converted to macrolactone 44b (26 mg, 25%) as a clear oil
along with the diastereomer 44d (26 mg, 25%): 1H NMR (300
MHz, CDCl3) δ 6.03 (d, J ) 15.3 Hz, 1H), 5.87 (m, 1H), 5.61 (dt,
J ) 15.3, 6.6 Hz, 1H), 5.33-5.36 (m, 2H), 5.07 (d, J ) 9.1 Hz,
1H), 4.62 (dd, J ) 18.9, 6.6 Hz, 2H), 3.86 (m, 4H), 3.55 (m, 1H),
3.36 (s, 3H), 2.73 (m, 1H), 2.06-2.26 (m, 4H), 1.75 (s, 3H), 1.26-
1.76 (m, 9H), 1.14 (d, J ) 6.9 Hz, 3H), 1.01 (t, J ) 7.8 Hz, 3H),
0.98 (d, J ) 6.6 Hz, 3H), 0.87 (d, J ) 6.3 Hz, 3H); 13C NMR (75
MHz, CDCl3) δ 174.4, 138.4, 135.7, 133.5, 133.0, 129.5, 127.0,
112.1, 100.9, 97.2, 75.6, 65.0, 63.9, 55.5, 41.2, 39.3, 34.5, 31.8,
31.6, 30.2, 30.1, 28.8, 25.6, 21.5, 16.9, 13.7, 13.6, 12.4. HRMS
(ESI) [M + Na]+ calcd for C28H46O6Na 501.3192, found 501.3189.
Synthetic Amphidinolide W (1b). According to the same
procedures for preparation of 1a from 44a, macrolactone 44b (26
mg, 0.054 mmol) was converted to synthetic amphidinolide W (1b)
(12 mg, 50% for two steps) as a clear oil: [R]23D +8.1 (c 1, CHCl3);
IR (neat) 3468, 1729; 1H NMR (500 MHz, CDCl3) δ 6.03 (d, J )
15.7 Hz, 1H), 5.64 (dt, J ) 15.4, 6.4 Hz, 1H), 5.62 (dt, J ) 15.6,
6.6 Hz, 1H), 5.52 (dd, J ) 15.4, 8.5 Hz, 1H), 5.04 (d, J ) 9.7 Hz,
1H), 4.96 (dd, J ) 8.3, 6.6 Hz, 1H), 3.56 (ddd, J ) 10.0, 6.7, 2.2,
1H), 2.83 (m, 1H), 2.64 (m, 1H), 2.49 (dt, J ) 18.3, 6.3 Hz, 1H),
2.38 (m, 1H), 2.34 (m, 1H), 2.31 (m, 1H), 2.17 (m, 1H), 2.11 (m,
2H), 1.89 (m, 2H), 1.77 (s, 3H), 1.65 (m, 1H), 1.50 (m, 1H), 1.41
(ddd, J ) 14.0, 10.5, 3.5, 1H), 1.26 (ddd, 13.9, 11.5, 2.2, 1H),
1.15 (d, J ) 6.9 Hz, 3H), 1.03 (d, J ) 7.1 Hz, 3H), 1.02 (t, J ) 7.1
Hz, 3H), 0.96 (d, J ) 6.7 Hz, 3H); 13C NMR (125 MHz, CDCl3)
δ 212.8, 175.3, 138.2, 135.5, 133.7, 133.5, 129.8, 127.2, 79.0, 70.6,
45.8, 41.0, 39.3, 35.9, 32.3, 32.3, 28.8, 26.5, 25.8, 21.7, 18.6, 16.4,
13.8, 12.7; HRMS (ESI) [M + Na]+ calcd for C24H38O4Na
413.2668, found 413.2662.
Acknowledgment. This research was supported in part by
the National Institutes of Health.
Supporting Information Available: Full experimental details
and characterization; copies of 1H and 13C NMR spectra of selected
compounds; tables of comparison of the NMR data of synthetic
amphidinolide W (1b) and its isomers 1a, 1c, and 1d with those
reported in the literature. This material is available free of charge
via the Internet at http://pubs.acs.org.
JO052181Z
J. Org. Chem, Vol. 71, No. 3, 2006 1093