Total synthesis of (-)-(3R,6S,9R)-decarestrictine C2

Article abstract of DOI:10.1016/S0040-4039(99)02138-3

Enantioselective synthesis of the proposed structure Of (-)-(3R,6S,9R)- decarestrictine C₂ (4) has been accomplished using (2S,5S)-1,2,5,6- hexanetenterol (9) as a C₂-symmetric chiral synthon, in which the diastereoselective aldol-type reaction of a tin(II) enolate of 3-acetyl-4(S)- isopropyl-1,3-thiazolidine-2-thione with the α,β-unsaturated aldehyde was used as a key step. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(99)02138-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 1199–1203
Total synthesis of (–)-(3R,6S,9R)-decarestrictine C2
Mayumi Arai, Nobuyasu Morita, Sakae Aoyagi and Chihiro Kibayashi ∗
School of Pharmacy, Tokyo University of Pharmacy & Life Science, Horinouchi, Hachioji, Tokyo 192-0392, Japan
Received 30 August 1999; revised 29 October 1999; accepted 8 November 1999
Abstract
Enantioselective synthesis of the proposed structure of (−)-(3R,6S,9R)-decarestrictine C2 (4) has been accom-
plished using (2S,5S)-1,2,5,6-hexanetetrol (9) as a C2-symmetric chiral synthon, in which the diastereoselec-
tive aldol-type reaction of a tin(II) enolate of 3-acetyl-4(S)-isopropyl-1,3-thiazolidine-2-thione with the α,β-
unsaturated aldehyde was used as a key step. © 2000 Elsevier Science Ltd. All rights reserved.
Keywords: medium-ring heterocycles; aldol reactions; lactonization; X-ray crystal structures.
In 1992, the joint group of Hoechst AG and University of Göttingen reported1 the isolation of a family
of 10-membered lactones, named decarestrictines A–D, as the first representatives of a new class of
fungal metabolites from cultures of Penicillium simplicissimum and P. corylophilum. Subsequent research
by the same group on these cultures led to the further discovery of decarestrictines E–M as minor
components of this class of fungal metabolites.2 These metabolites were found to show more or less
potent inhibitory effects on cholesterol biosynthesis in cell line tests with HEP-G2 liver cells.1,2 Of these
decarestrictines, decarestrictine A and decarestrictine C seemed to be homogeneous on TLC, however,
their 1H NMR spectra revealed that both decarestrictines actually consist of two components; A1 and A2
(1 and 2), and C1 and C2 (3 and 4) in the ratios of 3:1 and 1:1, respectively. The structural elucidation
of both components of each diastereomeric mixture epimeric at C-3 could be possible because of the
well separated signal patterns in their 1H NMR spectra, and the relative and absolute stereochemistries
of these compounds have been presented as shown. The interesting biological activity of decarestrictines
has been responsible for recent synthetic activity, including the total syntheses of decarestrictines D (5),3
J (6),4 and L (7).5
∗
Corresponding author. Fax: +0081-426-764475; e-mail: kibayasi@ps.toyaku.ac.jp (C. Kibayashi)
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02138-3
tetl 16066

1200
Scheme 1.
We have previously published transformation of D-mannitol into (S,S)- and (R,R)-1,2,5,6-
diepoxyhexane (8)6 via a key intermediate C2-symmetric (S,S)-tetrol 9, and the diepoxide 8 has
been used as a multi-purpose chiral building block in the enantioselective syntheses of various natural
products6a,c,7 including decarestrictine L (7).5a In connection with this synthetic strategy based on the
C2-symmetrical chiral synthon approach, we envisioned the use of the (S,S)-tetrol 9 for the synthesis of
decarestrictine C2 (4) revealed by the retrosynthetic approach as briefly outlined in Scheme 1. In this
paper, we report the first total synthesis of (−)-(3R,6S,9R)-decarestrictine C2 according to this synthetic
strategy.
The synthesis began with monotosylation of the primary alcohol in 9,8 followed by protection of the
1,2-diol to give the hydroxy tosylate 12 in 46% overall yield (Scheme 2). Reduction of the tosylate 12
using LiAlH4 followed by protection of the secondary alcohol with the t-butyldiphenylsilyl (TBDPS)
group gave the fully protected triol 13. In selective cleavage of the acetal group in 13, conventional
acid treatment (HCl or CF3CO2H) resulted in unsatisfactory yields of the 1,2-diol 14. Upon treatment
with copper(II) chloride dihydrate9 in acetonitrile, however, 14 was obtained in excellent yield (93%).
Subsequent selective pivaloylation of the primary alcohol yielded 15, which was converted to the alcohol
16 in 90% yield by hydroxyl protection and LiAlH4 reduction. Swern oxidation of 16 followed by
Wittig olefination of the resulting aldehyde afforded the 2(E)-octenal 17 as a single geometrical isomer.
For the enantioselective introduction of the hydroxyl-bearing asymmetric center, 17 was subjected to
diastereoselective aldol-type reaction using tin(II) enolates.10 Thus, the chiral tin(II) enolate, prepared
from (S)-acetyl-1,3-thiazolidine-2-thione 18 and Sn(OTf)2 in the presence of 1-ethylpiperidine, was
allowed to react with 17 to provide the 3(R)-hydroxy-4-decenamide 20 with 95% de, which required
immediate hydroxyl protection because of its lability, and then 21 was isolated in 93% overall yield from
17. The creation of the new stereogenic center with the desired R configuration is understandable by
consideration of the chelated transition state 19.10
Removal of the chiral auxiliary was accomplished effectively by applying a mild aminolysis.11 Thus,
treatment of 21 with imidazole in THF followed by hydrolysis of the resulting imidazolide with 10%
citric acid provided the carboxylic acid 22 in 88% yield (Scheme 3). The TBDPS-protection of 22

1201
Scheme 2.
Scheme 3.
was selectively removed by the use of polymer-supported fluoride12 in refluxing acetonitrile to give
the seco-acid 23 in 61% yield. Subsequent lactonization of 23 by the Yamaguchi method13 afforded the
10-membered lactone 24 in 60% yield. Deprotection of the MOM groups furnished (−)-(3R,6S,9R)-
27
D
decarestrictine C2 (4), mp 182–183.5°C (acetone–CHCl3); [α] −35.0 (c 0.66, MeOH), in 74%
yield. Spectroscopic data14 and the result of X-ray crystallographic analysis (Fig. 1) for our synthetic
material unambiguously confirmed the structure and stereochemistry of 4 proposed for decarestrictine
1
C2, however, there were significant differences between the H and 13C NMR spectral data for this
material and the corresponding data1b for decarestrictine C2 (as well as C1) assigned by NMR analysis
of natural decarestrictines C1/C2. We have previously synthesized15 decarestrictines C1 and C2 (3 and 4)

1202
as an inseparable 1:1 epimeric mixture at C-3 starting with the (S,S)-diepoxide (8). The synthetic material
1b
1
was found to be identical with a diastereomeric mixture of natural decarestrictines C1/C2 on H and
13C NMR spectral comparison. On the basis of these results, it is strongly suggested that the observed
discrepancy in the spectral data may be due to the formation of a 1:1 molecular complex in a solution
resulting from the intermolecular hydrogen bonding between decarestrictines C1 and C2.
Fig. 1. The single-crystal X-ray structure of 4
In conclusion, we have achieved the first enantioselective total synthesis of (−)-(3R,6S,9R)-
decarestrictine C2 (4) utilizing the C2 symmetric (S,S)-tetrol 9 as a simple chiral synthon.
References
1. (a) Grabley, S.; Granzer, E.; Hütter, K.; Ludwig, D.; Mayer, M.; Thiericke, R.; Till, G.; Wink, J.; Philipps, S.; Zeeck, A. J.
Antibiot. 1992, 45, 56. (b) Göhrt, A.; Zeeck, A.; Hütter, K.; Kirsch, R.; Kluge, H.; Thiericke, R. J. Antibiot. 1992, 45, 66.
2. Grabley, S.; Hammann, P.; Hütter, K.; Kirsch, R.; Kluge, H.; Thiericke, R.; Mayer, M.; Zeeck, A. J. Antibiot. 1992, 45, 1176.
3. (a) Andrus, M. B.; Shih, T.-L. J. Org. Chem. 1996, 61, 8780. (b) Pilli, R. A.; Victor, M. M. Tetrahedron Lett. 1998, 39, 4421.
4. Yamada, S.; Tanaka, A.; Oritani, T. Biosci. Biotech. Biochem. 1995, 59, 1657.
5. (a) Machinaga, N.; Kibayashi, C. Tetrahedron Lett. 1993, 34, 5739. (b) Clark, J. S.; Whitlock, G. A. Tetrahedron Lett. 1994,
35, 6381. (c) Nokami, J.; Taniguchi, T.; Ogawa, Y. Chem. Lett. 1995, 43. (d) Solladié, G.; Arce, E.; Bauder, C.; Carreño, M.
C. J. Org. Chem. 1998, 63, 2332.
6. (a) Machinaga, N.; Kibayashi, C. J. Org. Chem. 1991, 56, 1386. (b) Machinaga, N.; Kibayashi, C. Synthesis, 1992, 989. (c)
Machinaga, N.; Kibayashi, C. J. Org. Chem. 1992, 57, 5178.
7. (a) Machinaga, N.; Kibayashi, C. Tetrahedron Lett. 1992, 34, 841. (b) Noda, A.; Aoyagi, S.; Machinaga, N.; Kibayashi, C.
Tetrahedron Lett. 1993, 35, 841.
8. The preparation of 9 has been carried out starting from D-mannitol via hydrogenation of i using Rh–Al2O3 as a catalyst (Ref.
6c). However, this method was found to be less reproducible, often leading to the formation of by-products. We have found
that this problem is solved by the use of PtO2 as a catalyst as follows (unpublished result).
9. Saravanan, P.; Chandrasekhar, M.; Anand, R. V.; Singh, V. K. Tetrahedron Lett. 1998, 39, 3091.
10. Nagao, Y.; Hagiwara, Y.; Kumagai, T.; Ochiai, M.; Inoue, T.; Hashimoto, K.; Fujita, E. J. Org. Chem. 1986, 51, 2391.
11. Nagao, Y.; Kumagai, T.; Nagase, Y.; Tamai, S.; Inoue, Y.; Shiro, M. J. Org. Chem. 1992, 57, 4232 and references cited
therein.
12. Cardillo, G.; Orena, M.; Sandri, S.; Tomasini, C. Chem. Ind. (London) 1983, 643.

1203
13. Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. Bull. Chem. Soc. Jpn. 1979, 52, 1989.
14. NMR data for synthetic 4: 1H NMR (400 MHz, CD3OD) δ 1.13 (3H, d, J=6.5 Hz), 1.43 (1H, dd, J=16.0, 7.1 Hz), 1.62 (1H,
br t, J=12.3 Hz), 1.79–1.93 (2H, m), 2.30 (1H, t, J=10.5 Hz), 2.58 (1H, dd, J=10.2, 5.3 Hz), 4.28 (1H, m), 4.33 (1H, ddd,
J=10.7, 8.7, 5.3 Hz), 4.73 (1H, m), 5.42 (1H, br d, J=16.0 Hz), 5.72 (1H, ddd, J=16.0, 8.7, 1.6 Hz); 13C NMR (100.6 MHz,
CD3OD) δ 22.8, 29.9, 36.9, 48.0, 69.5, 74.6, 75.3, 132.2, 135.2, 173.4.
15. A 1:1 epimeric mixture of decarestrictines C1/C2 (3/4) has been synthesized by the following sequence starting from (S,S)-8
via a nonstereoselective Reformatsky-type cyclization (Machinaga, N. PhD Thesis, Tokyo University of Pharmacy & Life
Science, 1993).