A novel approach for construction of the naturally occurring dihydroagarofuran sesquiterpene skeleton

Article abstract of DOI:10.1016/S0040-4039(01)02203-1

A general and efficient approach for the synthesis of a kind of dihydrofuran sesquiterpenes extensively present in the Celastraceae family of plants has been developed by a series of transformations from santonin. The key creative and versatile steps invo

Full text of DOI:10.1016/S0040-4039(01)02203-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 627–630
A novel approach for construction of the naturally occurring
dihydroagarofuran sesquiterpene skeleton
Wu Jiong Xia, De Run Li, Lei Shi and Yong Qiang Tu*
Department of Chemistry & National Laboratory of Applied Organic Chemistry, Lanzhou University, Lanzhou 730000, PR China
Received 11 October 2001; accepted 16 November 2001
Abstract—A general and efficient approach for the synthesis of a kind of dihydrofuran sesquiterpenes extensively present in the
Celastraceae family of plants has been developed by a series of transformations from santonin. The key creative and versatile steps
involve the strategic acid-catalyzed double-bond shifting affording 6, the novel base-promoted epoxide rearrangement of 7
generating two key functions (the C5ꢀOH and the D7,11 double bond) and the stereoselective construction of a tetrahydrofuran
moiety without particularly controlling the stereochemistry of C7. In particular, this approach can introduce any number of
hydroxyls at the required positions for synthesis of many natural dihydroagarofuran sesquiterpenes. © 2002 Elsevier Science Ltd.
All rights reserved.
Numerous agarofuran sesquiterpenes with a different
number of polyol esters bearing the skeleton 1, isolated
from the Celastraceae plants have attracted a great deal
of interest from chemists on account of their wide range
of biological activities, such as important cytotoxic,1
antitumor,2 immunosuppressive,3 insecticidal4 and anti-
HIV activities.5 However, few successful synthetic
works on agarofuran sesquiterpenes have been reported
up to now, because the multi-hydroxylation on the
skeleton and construction of the tetrahydrofuran ring is
challenging work. In 1997, White and co-workers
reported the first successful total synthetic work.6 How-
ever, it was very complicated and was only used for the
synthesis of a few target compounds with some limited
substitution fashions. In connection with our investiga-
tion of this field, our goal was to develop a general and
efficient procedure for the synthesis of the agarofuran
sesquiterpenes with a different number of hydroxyl
Scheme 1.
* Corresponding author. Tel.: +86 0931 8912410; fax: +86 0931 8913569; e-mail: tuyq@lzu.edu.cn
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)02203-1

628
W. J. Xia et al. / Tetrahedron Letters 43 (2002) 627–630
substituents. For this purpose, we chose the rich and
synthetically versatile natural source a-santonin7 as the
starting material because of the possible generation of
multiple hydroxyls from its multi-functionality such as
carbonyl, double bond and hydroxyl. Here, as an exam-
ple of our synthetic program to illustrate the generality
of this approach, we want to report the synthesis of a
natural product skeleton, the 8,9-dimethyletherified-
2,14-dideoxyalatol 2, which was the nucleus of many
natural dihydroagarofuran sesquiterpenes, such as
angulatueoidand,8 several congeners,9 and so on.
of 9 with acetone successfully led to the single product
10 with C6ꢀOH and C12ꢀOH protected in 90% yield.
The successive oxidation of 10 with SeO2/tBuO2H gave
the exclusive product 11 in 95% yield.
All attempts to cyclize 11 to the tetrahydrofuran deriva-
tive 12 with many reagents, such as NBS,11 Hg(OAc)2/
NaBH4,12 m-CPBA/NaBr,13 and so on, unfortunately,
were not successful. The possible reason was steric
hindrance of the quaternary acetonide. Subsequently,
we had to change the protective group by protection of
triol 9 with PhCHO/ZnCl2 followed by oxidation to
obtain the b-alcohol 14 in 75% yield (Scheme 3), and
the other 8a-isomer was not detected. The stereoselec-
tive tetrahydrofuran ring closure of compound 14 with
NBS/CaCO3 readily afforded the product 15 in 95%
yield. The two characteristic downfield 13C NMR sig-
nals of 15 as l 92.8(C5) and 84.7(C11) indicated the
successful cyclization of the tetrahydrofuran ring as
nearly all compounds of this kind have the similar
shifting values.14 Debromination of compound 15 by a
radical reaction with tri-n-butyltin hydride followed by
dehydration gave the product 16, which was then con-
verted to the 8b,9b-diol 17 by oxidation with OsO4/
NMO in acetone.15 Furthermore, the proton coupling
constants, J=4.0 Hz (H7–H8) and J=4.8 Hz (H8–H9),
suggested the dihydroxylation was performed from the
b-face. Several reagents, such as acetone and tert-
butyldimethylsilyl chloride, were used for protection of
8,9-diol, but the protective groups at C8, C9 and C3 of
these obtained compounds were easily removed in the
next steps. Thus, we protected 8b,9b-diol of 17 as
methyl ethers with sodium hydride and methyl iodide in
From the retro-synthetic analysis (Scheme 1), it was
found that the key problems to be solved were focused
on the construction of the tetrahydrofuran moiety by
the conversion of the configuration on C7 and the
multi-hydroxylation
on
the
dihydroagarofuran
sesquiterpene skeleton. In our previous paper, we
reported the synthesis of the basic dihydroagarofuran
sesquiterpene skeleton from a-santonin.10 However, the
reported ring-closure method with a Hg(OAc)2/NaBH4
system failed to synthesize the polyhydroxylated agaro-
furan sesquiterpene after many experiments. Thus, we
need to develop a new method for the construction of
the tetrahydrofuran moiety. As shown in Scheme 2, the
triol 9 was first synthesized from a-santonin in a previ-
ous procedure.10 However, direct oxidation of com-
pound 9 with SeO2/tBuO2H only gave the complicated
product possibly due to the existence of two allylic
hydroxyls. Thus, we thought C6ꢀOH and C12ꢀOH of 9
should be protected before oxidation. From the molec-
ular model inspection, it was found that the C12ꢀOH
was very adjacent to the C6ꢀOH in comparison to the
C5ꢀOH and C5ꢀOH was a tertiary alcohol. Protection
Scheme 2. Reaction conditions: (a) H2, Raney Ni, PhH; Glycol (5 equiv.), PTS, toluene, 53%; (b) m-CPBA, CH2Cl2, 68%; (c)
t
NaOMe (50 equiv.), MeOH, 83%; (d) LiAlH4, THF, −78°C to rt, 95%; (e) acetone, PTS, rt, 90%; (f) SeO2, BuO2H, dioxane, rt,
95%.

W. J. Xia et al. / Tetrahedron Letters 43 (2002) 627–630
629
t
Scheme 3. Reaction conditions: (a) PhCHO, ZnCl2, rt, 70%; (b) SeO2, BuO2H, 75%; (c) NBS, CaCO3, THF, 0°C, 95%; (d)
t
nBu3SnH, PhH; (e) SOCl2, Py, 60% (two steps); (f) OsO4, NMO, acetone, rt, 84%; (g) NaH, MeI, rt; (h) K, BuNH2, THF, rt,
68% (two steps).
Scheme 4. Reaction conditions: (a) benzoyl chloride, Py, 0°C; (b) TBSOTf, 2,6-lutidine, CH2Cl2, rt; (c) NaOMe (2 equiv.), MeOH,
rt, 81% (three steps); (d) PDC, CH2Cl2; (e) NH2NH2·H2O, K2CO3, diethylene glycol, 150–220°C, 72% (two steps); (f) PTS,
acetone, rt; (g) p-MeC6H4SO2NHNH2, MeOH; CH3Li (3 equiv.), rt, 50% (two steps); (h) m-CPBA, CH2Cl2, rt, 88%; (i) i-Pr2NLi
n
(5 equiv.), THF, −78°C to rt, 72%; (j) HCl, THF, rt; (k) 10% Pd/C, H2; Bu4N+F−, THF, rt, 47% (three steps).

630
W. J. Xia et al. / Tetrahedron Letters 43 (2002) 627–630
THF followed by hydrogenation with K/tBuNH2/
THF smoothly afforded the diol 18 in 68% yield.
Having the diol in hand, we tried to remove
C12ꢀOH. Nevertheless, some attempts to dehydroxyl-
ate using many reagents, such as Ph3P/NBS/
nBu3SnH, MsCl/Py/LiBHEt3 and so on, failed. There-
fore, the C12ꢀOH of 18 was first selectively protected
with benzoyl chloride (Scheme 4), then we protected
the hindered secondary alcohol at C6 with tert-
References
1. Kuo, Y.-H.; King, M.-L.; Chen, C.-F.; Chen, H.-Y.;
Chen, C.-H.; Chen, K.; Lee, K.-H. J. Nat. Prod. 1994,
57, 263–269.
2. Takaishi, Y.; Ujita, K.; Tokuda, H.; Nishino, H.;
Iwashima, A.; Fujita, T. Cancer Lett. 1992, 65, 19.
3. Zheng, Y. L.; Xu, Y.; Lin, J. F. Act Pharm. Sin. 1989,
24, 568–572.
4. Wu, W. J.; Tu, Y. Q.; Liu, H. X.; Zhu, J. B. J. Nat.
Prod. 1992, 55, 1294–1298.
butyldimethylsilyl
trifluoro-methanesulfonate
and
finally removed the benzoyl group with NaOMe in
MeOH to obtain compound 19 in 81% yield. Oxida-
tion of 19 with PDC afforded the product 20 in
quantitative yield, and the decarbonylation of com-
pound 20 using a Huang-Minlon reduction was very
effective to give 21 in 72% yield.16 Thereafter, treat-
ment of 21 with p-toluene sulphonic acid in acetone
gave a carbonyl compound, hydrozonization of which
with TsNHNH2 followed by CH3Li led to compound
22 in 43% yield (total yield from 21).17 To hydroxyl-
ate on C1, compound 22 was subjected to oxidation
with SeO2, but this did not lead to the desired
product even when the reaction conditions were
varied. Thus, compound 22 was epoxidized with m-
CPBA to give 23 in 88% yield. We rearranged the
epoxide 23 with a strong base to obtain the allylic
alcohol 24 in 72% yield, which was further rearranged
to another allylic alcohol 25 in the presence of acid.18
Reduction of the D2,3 double bond by catalytic hydro-
genation with 10% Pd/C and deprotection of the silyl
group yielded the final product 2 in 47% yield. The
structure of 2 was confirmed by 1D and 2D NMR
technique and mass spectroscopy.19
5. Duan, H.; Takaishi, Y.; Imakura, Y.; Jia, Y.; Li, T.;
Cosentino, L. M.; Lee, K.-H. J. Nat. Prod. 2000, 63,
357–361.
6. White, J. D.; Shin, H.; Kim, T.-S.; Cutshall, N. S. J.
Am. Chem. Soc. 1997, 119, 2404–2419.
7. (a) Banerjee, A. K.; Vera, W. J.; Gonzalez, N. C. Tet-
rahedron 1993, 49, 4761; (b) Hernandez, R.; Velazquez,
S. M.; Suarez, E.; Kodriguez, M. S. J. Org. Chem.
1994, 59, 6395.
8. Wu, D. G.; Liu, J. K.; Cheng, C. Q. Phytochemistry
1992, 31, 4219–4222.
9. (a) Takaishi, Y.; Ohshima, S.; Nakano, K.; Tomimatsu,
T. J. Nat. Prod. 1993, 56, 815–824; (b) Takaishi, Y.;
Tokura, K.; Tamai, S.; Ujita, K.; Nakano, K.; Tomi-
matsu, T. Phytochemistry 1991, 30, 1567–1572; (c) Tu,
Y. Q.; Hu, Y. J.; Wu, W. J.; Chen, N. Y.; Pan, X. F.
Phytochemistry 1992, 31, 3633–3634.
10. Tu, Y. Q.; Sun, L. D. Tetrahedron Lett. 1998, 39,
7935–7938.
11. Frederickson, M.; Grigg, R. Org. Prep. Proc. Int. 1997,
29, 35.
12. Evans, D. A.; Dow, R. L.; Shin, T. L.; Takacs, J. M.;
Zahler, R. J. Am. Chem. Soc. 1990, 112, 5290–5313.
13. Srebnik, M.; Mechoulam, R. J. Chem. Soc., Chem.
Commun. 1984, 1070–1071.
In conclusion, the approach reported here is indeed a
new route for the synthesis of the polyhydroxylated
dihydroagarofuran sesquiterpenenes, which can be
used for the synthesis of other agarofuran sesquiter-
pene derivatives. The application of this approach to
the synthesis of other hydroxylated agarofurans for
the biological activities test is ongoing and will be
reported in due course.
14. Wu, W. J.; Tu, Y. Q.; Liu, H. X.; Zhu, J. B. J. Nat.
Prod. 1992, 55, 1294–1298.
15. Bates, H. A.; Fanina, J.; Tong, M. J. Org. Chem. 1986,
51, 2641.
16. (a) Huang-Minlon J. Am. Chem. Soc. 1946, 68, 2487;
(b) Huang-Minlon J. Am. Chem. Soc. 1949, 71, 3301.
17. Watanabe, M.; Yoshikoshi, A. J. Chem. Soc., Perkin
Trans. 1 1987, 2833.
18. Ando, M.; Akahane, A.; Takase, K. Bull. Chem. Soc.
Jpn. 1978, 51, 284.
19. The spectral data of 2: 1H NMR: 4.75 (brs, 1H), 4.14
(m, 1H), 3.82 (dd, J 4.4, 4.8 Hz, 1H), 3.45 (s, 3H), 3.41
(s, 3H), 3.27 (d, J 4.8 Hz, 1H), 2.41 (m, 2H), 1.60 (m,
4H), 1.53 (s, 3H), 1.32 (s, 3H), 1.12 (s, 3H), 1.11 (d, J
7.2 Hz, 3H); 13C NMR: 92.5, 85.1, 81.0, 77.6, 73.4,
70.5, 58.1, 52.0, 44.2, 40.3, 34.3, 31.6, 31.6, 25.1, 24.0,
16.9, 10.9; HRMS for C17H30O5: calcd 337.1986 [M+
Na]+, found 337.1985.
Acknowledgements
We are grateful for financial support from NSFC
(No. 29972019, 29925205 and QT program), FUK-
TME of China, the Young Teachers’ Fund of the
Ministry of Education and the Fund of the Ministry
of Education (No. 99209).