Determination of the absolute configuration of marine oxylipin topsentolide A1 by the synthesis of the enantiomer of the natural product

Article abstract of DOI:10.1016/j.tetlet.2010.03.071

Two possible stereoisomers of topsentolide A₁, a cytotoxic oxylipin against human solid tumor cell lines, were prepared in order to determine the stereochemistry of natural product. That is, the enantiomer of topsentolide A₁, (8S,11S,12R)-isomer, and its diastereomer was efficiently synthesized in a stereoselective manner. The stereochemistry of topsentolide A₁ was determined to be 8R,11R,12S by comparing NMR spectra and specific rotations of the synthetic isomers and the natural product.

Full text of DOI:10.1016/j.tetlet.2010.03.071

Tetrahedron Letters 51 (2010) 2762–2764
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Determination of the absolute configuration of marine oxylipin topsentolide
A1 by the synthesis of the enantiomer of the natural product
*
Munetaka Kobayashi, Ken Ishigami, Hidenori Watanabe
Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Two possible stereoisomers of topsentolide A1, a cytotoxic oxylipin against human solid tumor cell lines,
were prepared in order to determine the stereochemistry of natural product. That is, the enantiomer of
topsentolide A1, (8S,11S,12R)-isomer, and its diastereomer was efficiently synthesized in a stereoselective
manner. The stereochemistry of topsentolide A1 was determined to be 8R,11R,12S by comparing NMR
spectra and specific rotations of the synthetic isomers and the natural product.
Received 24 February 2010
Revised 12 March 2010
Accepted 18 March 2010
Available online 21 March 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Topsentolide A1
Nine-membered lactone
Determination of the absolute configuration
Marine oxylipin
A number of oxylipins have been isolated as secondary metab-
olites of marine organisms such as cyanobacteria, alga, hydrozoa,
and sponges. These oxylipins have interesting activities and syn-
thetic studies on them are meaningful, as well as biological studies,
because their bioactivities are frequently concerned with structure.
Halicholactone (2, isolated from Halichondoria okadai)1 is an inhib-
itor of lipoxygenase; mueggelone (3, isolated from Aphanizomenon
flos-aquae)2 is an inhibitor of fish development; and solandelactone
D (4, isolated from Solanderia secunda)3 is an inhibitor of farnesyl
protein transferase. Previously in our group, the syntheses of these
marine oxylipins had been achieved in stereoselective manner and
the results had been already reported (Fig. 1).4–8
(8S,11S,12R)-isomer (6) stereoselectively via epoxide formation,
where the stereochemistry of the epoxide could be controlled by
changing elimination direction.4,5 Compound 7 would be con-
structed by Horner–Wadsworth–Emmons (HWE) reaction of a side
chain moiety 8 with a lactone moiety 9. Both the phosphonate 8
and the lactone 9 would be derived from the known aldehyde
10,10,11 which could be easily prepared from
L-malic acid.
Synthesis of the phosphonate 8 is shown in Scheme 2. The
known aldehyde 1010,11 was subjected to Wittig reaction with
the known phosphonium salt 1112 to afford Z-olefin 12 selectively
(>98%, checked by GC analysis). The Z-olefin 12 was simply trans-
formed into the corresponding primary alcohol 14 in four steps,
which was then oxidized to give aldehyde 15. Oxidation of the
aldehyde 15 with NaClO2 and subsequent esterification using
CH2N2 afforded the desired methyl ester 16 only in poor yield
(ꢀ20%), and undesired migration of TBS group was observed. On
the other hand, oxidation of 15 with NIS13 in methanol proceeded
smoothly to give ester 16 in good yield without any side reactions.
In 2006, Jung’s group isolated topsentolide A1 (1)9 from the ex-
tract of a marine sponge Topsentia sp. as a cytotoxic oxylipin against
human solid tumor cell lines. Topsentolide A1 has a nine-membered
lactone with an epoxide side chain. Stereochemistry of the epoxide
was proposed to be cis by spectroscopic analysis, however, the rela-
tive as well as the absolute configuration of the three stereocenters
remained unknown. Therefore, we undertook the synthesis of possi-
ble stereoisomers of topsentolide A1 in order to determine the
absolute configuration of the natural product. In this Letter, we re-
port the synthesis of (8S,11R,12S)-isomer (5) and (8S,11S,12R)-iso-
mer (6), and the determination of the absolute configuration of
topsentolide A1.
OH
O
H
H
H
H
O
O
O
O
cis
OH
Topsentolide A1 (1)
Halicholactone (2)
Our synthetic strategy of (8S)-isomers (5 and 6) is illustrated in
Scheme 1. We decided to utilize a key intermediate 7, which could
be transformed into both of (8S,11R,12S)-isomer (5) and
OH
H
O
H
H
H
O
O
O
O
OH
Mueggelone (3)
Solandelactone D (4)
* Corresponding author. Fax: +81 3 5841 8019.
E-mail address: ashuten@mail.ecc.u-tokyo.ac.jp (H. Watanabe).
Figure 1. Bioactive marine oxylipins.
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.03.071

M. Kobayashi et al. / Tetrahedron Letters 51 (2010) 2762–2764
2763
O
O
H
H
H
H
OH
O
O
O
a
b, c
O
O
O
HO
10
CO2Me
CO2H
CO2H
18
17
(8S,11R,12S)-isomer (5)
(8S,11S,12R)-isomer (6)
O
OH
O
d, e
f
g, h
OPMB
epoxide formation
PMBO
9
HO OTBS
11
19
20
O
O
Scheme 3. Synthesis of the lactone 9. Reagents and conditions: (a) (4-carboxybu-
tyl)triphenylphosphonium bromide, NaHMDS, THF, ꢁ78 °C to rt; (b) CH2N2, Et2O,
MeOH, 0 °C, 69% in two steps; (c) 1 N HCl, THF, rt, 97%; (d) PMBOC(NH)CCl3,
TsOHꢂH2O, CH2Cl2, rt, 63%; (e) LiOHꢂH2O, H2O, MeOH, rt, 98%; (f) 2,4,6-trichloro-
benzoyl chloride, Et3N, THF, rt, then DMAP, toluene, reflux, 81%; (g) DDQ, H2O,
CH2Cl2, 87%; (h) (COCl)2, DMSO, Et3N, CH2Cl2, ꢁ78 °C, 69%.
7
HWE reaction
O
P
O
O
OTBS
O
CHO
MeO
MeO
8
9
O
OTBS
a
b
O
O
+
O
7
8
7
7
9
O
CHO
10
21
MsO OTBS
Scheme 1. Retrosynthetic analysis of two possible (8S)-stereoisomers of topsen-
c
d
O
(8S,11R,12S)-isomer
O
O
tolide A1.
(5)
22
Finally, ester 16 was reacted with anion of dimethyl methylphos-
phonate to afford the desired b-ketophosphonate 8, one of the sub-
strates for HWE reaction.
EEO OMs
e, f, c
g, h
O
(8S,11S,12R)-isomer
(6)
Synthesis of the lactone 9 is shown in Scheme 3. Wittig reaction
of the known aldehyde 1010,11 and a commercially available phos-
phonium salt afforded Z-olefin 17 selectively (E-isomer was not
observed in 1H NMR spectrum), which was transformed into the
corresponding hydroxycarboxylic acid 19 in four steps. Lactoniza-
tion of 19 was successfully achieved by Yamaguchi’s method14 to
provide nine-membered lactone 20 in satisfactory yield, together
with a small amount of a dimeric lactone (3%). After the removal
of PMB group, 20 was oxidized into the desired aldehyde 9, another
substrate for HWE reaction.
23
Scheme 4. Synthesis of possible (8S)-stereoisomers of topsentolide A1. Reagents
and conditions: (a) DBU, LiCl, MeCN, rt, 67%; (b) NaBH4, CeCl3ꢂ7H2O, MeOH, rt, 96%;
(c) MsCl, Et3N, DMAP, CH2Cl2, rt; (d) TBAF, THF, rt, 85% in two steps; (e) EVE, PPTS,
CH2Cl2, rt, 83%; (f) TBAF, THF, rt, 91%; (g) 0.5 N HCl, THF, rt; (h) TBAF, THF, rt, 29% in
three steps.
Table 1
1H NMR (300 MHz, CDCl3) data for (S)- and (R)-MTPA ester of 7
Now that both the side chain moiety 8 and the lactone moiety 9
were obtained enantioselectively, HWE reaction was examined
(Scheme 4). Under Masamune’s conditions employing DBU-LiCl,15
HWE reaction of the aldehyde 9 and the b-ketophosphonate 8 pro-
ceeded smoothly to give unsaturated ketone 21, with excellent
E selectivity without any epimerization (other isomers were not ob-
served in 1H NMR spectrum). Stereoselective reduction of the ketone
21 using Luche’s conditions16 successfully afforded the key interme-
diate 7, the precursor for epoxide formation. Reduction of 21 pro-
ceeded under Felkin–Anh control, and the stereochemistry at C-11
was confirmed to be S by modified Mosher’s method17 after conver-
sion to the corresponding MTPA esters (Table 1). In order to
MTPAO OTBS
9
O
O
11 12
10
16
MTPA ester of 7
Position
1H chemical shift (d)
dS–dR
(S)-MTPA
(R)-MTPA
9
10
12
16
5.79
5.68
3.79
2.70
5.87
5.81
3.77
2.67
ꢁ0.08
ꢁ0.13
+0.02
+0.03
construct botha-epoxideand b-epoxidefrom 7, the stereochemistry
of the epoxide was controlled as described below by changing
elimination direction.4,5 At first, alcohol 7 was converted to the
corresponding mesylate 22, which was treated with TBAF to provide
b-epoxide, (8S,11R,12S)-isomer (5). On the other hand, alcohol 7
was converted to another mesylate 23 by changing the position of
the leaving group. Finally, the removal of EE group was followed
O
O
a
b, c
O
O
CHO
P+Ph3 I–
10
OH
12
11
OTBS
d, e
f
PivO
OHC
HO
by treatment under basic conditions to afford
a-epoxide,
13
14
OTBS
(8S,11S,12R)-isomer (6), successfully.
Having succeeded in the construction of two possible stereoiso-
mers of topsentolide A1, we analyzed the NMR spectra and specific
rotations carefully.18,19 As for the 13C NMR, there was no difference
between both the diastereomers (5 and 6) to distinguish them. In
the 1H NMR, the signals of both diastereomers (5 and 6) appeared
at almost similar chemical shifts to those of natural topsentolide
A1, however, slight differences of the chemical shifts and the
shapes of peaks were observed between 5 and 6. As shown in
OTBS
g
h
8
MeO2C
15
16
Scheme 2. Synthesis of the phosphonate 8. Reagents and conditions: (a) 11,
KHMDS, THF, ꢁ78 °C to rt, 84%; (b) 1 N HCl, THF, rt, 90%; (c) PivCl, Pyr., rt, 89%; (d)
TBSOTf, 2,6-lutidine, CH2Cl2, rt, 96%; (e) DIBAL, CH2Cl2, ꢁ50 °C, 97%; (f) (COCl)2,
DMSO, Et3N, CH2Cl2, ꢁ78 °C to rt, 97%; (g) NIS, K2CO3, MeOH, rt, 78%; (h)
(MeO)2P(O)Me, n-BuLi, THF, ꢁ78 °C, 52%.

2764
M. Kobayashi et al. / Tetrahedron Letters 51 (2010) 2762–2764
16-H
2-H
13-Hb
13-Ha
3-H
of topsentolide A1 to be 8R,11R,12S. Our work is under way to pre-
4-H
pare (8R)-isomers including the natural form of topsentolide A1 (1)
and to submit these isomers to further biological assay. Results will
be reported in a full account.
19-H
Natural Topsentolide A 1 (1)
Acknowledgment
The authors sincerely thank Professor J. H. Jung, Pusan National
University, for a kind gift of the spectral charts of natural topsento-
lide A1.
O
H
H
O
O
O
O
identical
(8S,11S,12R)-isomer (6)
References and notes
1. Niwa, H.; Wakamatsu, K.; Yamada, K. Tetrahedron Lett. 1989, 30, 4543–4546.
2. Papendorf, O.; König, G. M.; Wright, A. D.; Chorus, I.; Oberemm, A. J. Nat. Prod.
1997, 60, 1298–1300.
3. Seo, Y.; Cho, K. W.; Rho, J.-R.; Shin, J.; Kwon, B.-M.; Bok, S.-H.; Song, J.-I.
Tetrahedron 1996, 52, 10583–10596.
O
H
H
not identical
(8S,11R,12S)-isomer (5)
4. Ishigami, K.; Motoyoshi, H.; Kitahara, T. Tetrahedron Lett. 2000, 41, 8897–8901.
5. Motoyoshi, H.; Ishigami, K.; Kitahara, T. Tetrahedron 2001, 57, 3899–3908.
6. Takahashi, T.; Watanabe, H.; Kitahara, T. Heterocycles 2002, 58, 99–104.
7. Takahashi, T.; Takahashi, T.; Watanabe, H.; Kitahara, T. Abstracts of Papers, 44th
Symposium on the Chemistry of Natural Products, Tokyo, Japan, 2002, pp 1–6.
8. Ishigami, K. Biosci., Biotechnol., Biochem. 2009, 73, 971–979.
9. Luo, X.; Li, F.; Hong, J.; Lee, C.-O.; Sim, C. J.; Im, K. S.; Jung, J. H. J. Nat. Prod. 2006,
69, 567–571.
Figure 2. 1H NMR signals of natural topsentolide A1 and synthesized stereoisomers
(500 MHz, in CD3OD).
10. Mori, K.; Takigawa, T.; Matsuo, T. Tetrahedron 1979, 35, 933–940.
11. Saito, S.; Hasegawa, T.; Inaba, M.; Nishida, R.; Fujii, T.; Nomizu, S.; Moriwake, T.
Chem. Lett. 1984, 1389–1392.
12. Kojima, K.; Koyama, K.; Amemiya, S. Tetrahedron 1985, 41, 4449–4462.
13. McDonald, C.; Holcomb, H.; Kennedy, K.; Kirkpatrick, E.; Leathers, T.; Vanemon,
P. J. Org. Chem. 1989, 54, 1213–1215.
14. Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. Bull. Chem. Soc. Jpn.
1979, 52, 1989–1993.
MeO OH
12S
O
H
H
17
O
O
O
O
8R 11R 12S
Topsentolide C1 (24)
Topsentolide A1 (1)
Topsentolide C2 (25): Δ17
15. Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfeld, A. P.; Masamune, S.; Roush,
W. R.; Sakai, T. Tetrahedron Lett. 1984, 25, 2183–2186.
16. Luche, J.-L. J. Am. Chem. Soc. 1978, 100, 2226–2227.
17. Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem. Soc. 1991, 113,
Figure 3. Absolute configuration of topsentolide A1.
4092–4096.
18. Analytical and spectral data of synthesized 5: ½a D24
ꢃ
ꢁ104 (c 0.11, CH3OH). 1H NMR
Figure 2, some signals of (8S,11R,12S)-isomer (5) were not identical
with those of natural topsentolide A1 [especially at 2.74–2.86 ppm
(16-H), 2.32–2.42 ppm (2-H and 13-Hb), 2.23–2.30 (13-Ha) and
2.03–2.12 ppm (3-H, 4-H and 19-H)], while 1H NMR spectrum of
(8S,11S,12R)-isomer (6) was completely identical with that of
natural topsentolide A1. Furthermore, specific rotations of the
(500 MHz, CD3OD) d (ppm) 0.96 (3H, t, J = 7.5 Hz, 20-H), 1.76 (1H, m, 3-Ha),
2.03–2.29 (7H, m, 2-Ha, 3-Hb, 4-Ha, 7-Ha, 13-Ha, 19-H), 2.32–2.55 (4H, m, 2-Hb,
4-Hb, 7-Hb, 13-Hb), 2.73–2.85 (2H, m, 16-H), 3.13 (1H, dt, J = 4.5, 6.5 Hz, 12-H),
3.47 (1H, br dd, J = 6.5, 4.5 Hz, 11-H), 5.24–5.31 (2H, m, 8-H, 17-H), 5.35–5.52
(5H, m, 5-H, 6-H, 14-H, 15-H, 18-H), 5.76 (1H, ddd, J = 15.5, 7.0, 1.5 Hz, 10-H),
6.03 (1H, ddd, J = 15.5, 5.5, 1.0 Hz, 9-H). 13C NMR (125 MHz, CD3OD) d (ppm)
14.7, 21.5, 26.3, 26.6, 27.1, 27.5, 34.4, 35.4, 57.2, 59.4, 73.9, 124.9, 125.5, 127.2,
127.9, 132.0, 133.0, 135.2, 136.3, 175.6. IR (film)
969 cmꢁ1 ESI-HRMS m/z calcd for C20H28NaO3 [M+Na]+ 339.1931, found
339.1942.
m 2962, 1742, 1218, 1136,
synthetic diastereomers, ½a D24
ꢃ
ꢁ104 (c 0.11, MeOH) for 5 and ½a D24
ꢃ
.
ꢁ88 (c 0.11, MeOH) for 6, showed opposite sign to ½a D24
ꢃ
+59.4
19. Analytical and spectral data of synthesized 6: ½a D24
ꢃ
ꢁ88 (c 0.11, CH3OH). 1H NMR
(c 0.11, MeOH)9 for the natural material. On the basis of these re-
sults, we propose the (8R,11R,12S)-stereochemistry, the enantio-
mer of 6, for natural topsentolide A1 (1) as shown in Figure 3.
Actually, Jung’s group also isolated topsentolide C1 (23) and C2
(24) together with 1.9 They suspected these methyl ether ana-
logues to be artifacts formed during the process of extraction with
MeOH, and determined the absolute configuration at C-12 as S by
Mosher’s method. Their determination of stereochemistry at C-12
also supports our results.
(500 MHz, CD3OD) d (ppm) 0.97 (3H, t, J = 7.5 Hz, 20-H), 1.76 (1H, m, 3-Ha),
2.03–2.30 (7H, m, 2-Ha, 3-Hb, 4-Ha, 7-Ha, 13-Ha, 19-H), 2.32–2.55 (4H, m, 2-Hb,
4-Hb, 7-Hb, 13-Hb), 2.74–2.86 (2H, m, 16-H), 3.12 (1H, dt, J = 4.5, 6.5 Hz, 12-H),
3.48 (1H, br dd, J = 7.0, 4.5 Hz, 11-H), 5.25–5.32 (2H, m, 8-H, 17-H), 5.35–5.52
(5H, m, 5-H, 6-H, 14-H, 15-H, 18-H), 5.76 (1H, ddd, J = 15.5, 7.0, 1.5 Hz, 10-H),
6.04 (1H, ddd, J = 15.5, 5.5, 1.0 Hz, 9-H). 13C NMR (125 MHz, CD3OD) d (ppm)
14.7, 21.5, 26.3, 26.7, 27.1, 27.5, 34.4, 35.3, 57.3, 59.4, 73.8, 124.9, 125.5, 127.3,
127.9, 132.0, 133.0, 135.3, 136.3, 175.6. IR (film)
969 cmꢁ1 ESI-HRMS m/z calcd for C20H28NaO3 [M+Na]+ 339.1931, found
339.1933.
m 2961, 1743, 1218, 1137,
.
In summary, we succeeded in the synthesis of ent-topsentolide
A1 (6) and its stereoisomer 5, and determined the stereochemistry