Synthesis and stereochemical determination of an antifeeding bisabolanoid from Japanese cedar

Article abstract of DOI:10.1016/j.tetlet.2008.02.039

The first enantioselective synthesis of (1S,3R,6R)-1-hydroxy-7(14),10-bisaboladien-4-one, a potent antifeedant isolated from the Japanese cedar, Cryptomeria japonica, was achieved starting from methyl (R)-4-hydroxy-3-methylbutanoate via a stereoselective

Full text of DOI:10.1016/j.tetlet.2008.02.039

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 2438–2441
Synthesis and stereochemical determination of an antifeeding
bisabolanoid from Japanese cedar
Takashi Nakahata, Yohsuke Satoh, Shigefumi Kuwahara *
Laboratory of Applied Bioorganic Chemistry, Graduate School of Agricultural Science, Tohoku University, Tsutsumidori-Amamiyamachi,
Aoba-ku, Sendai 981-8555, Japan
Received 18 January 2008; revised 6 February 2008; accepted 8 February 2008
Available online 13 February 2008
Abstract
The first enantioselective synthesis of (1S,3R,6R)-1-hydroxy-7(14),10-bisaboladien-4-one, a potent antifeedant isolated from
the Japanese cedar, Cryptomeria japonica, was achieved starting from methyl (R)-4-hydroxy-3-methylbutanoate via a stereoselective
carbonyl ene cyclization reaction as the key step. Comparison of the spectral data and specific rotation of the synthetic material with
those of the natural product led to unambiguous stereochemical assignment of the antifeedant as 1S, 3R, and 6R.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Bisabolane; Enantioselective synthesis; Carbonyl ene reaction; Antifeedant
In the course of screening for bioactive products from
the Japanese cedar, Cryptomeria japonica, Kim and co-
workers isolated hydroxy bisabolatrienone 2 as a potent
antifeedant against the snail, Acusta despesta (a well-
known pest of many agricultural crops) (Fig. 1).1 This bisa-
bolanoid 2 had originally been reported by Nagahama
et al. as a chemical constituent of C. japonica without the
assignment of the absolute stereochemistry and with no
mention of biological activity.2 Its absolute configuration
was later determined by Kim et al. as depicted in Figure
1 by converting the natural product into cryptomerione
and comparing its specific rotation with those of both
enantiomers of cryptomerione derived from (R)- and (S)-
carvones.3 They also discovered that sesquiterpenoid 2
exhibited repelling and antifeeding activities against the
pill-bug (Armadillidium vulgare) and the locust (Locusta
migratoria, a notorious pest, which often causes massive
damage to agricultural crops throughout the world) when
mixed with sandaracopimarinol and hydroxy bisaboladie-
none 1, respectively, which were also isolated from C.
O
1
O
3
6
OH
OH
1
2
O
H
H
HO
(R)-cryptomerione
sandaracopimarinol
Fig. 1. Structures of antifeedants from Japanese ceder and cryptomerione.
japonica.4,5 Based on extensive bioassays, they found that
compound 1 was essential for the antifeeding activity, while
compound 2 played a role to support the activity of 1.5
Although the gross structure of 1 was deduced from spec-
troscopic analyses including H–H and C–H COSY experi-
ments, any information on the stereochemistry of 1 was not
provided in their report. We describe herein the enantio-
selective synthesis of one of the stereoisomers of 1, which
culminated in the successful determination of the absolute
stereochemistry of 1.
*
Corresponding author. Tel./fax: +81 22 717 8783.
E-mail address: skuwahar@biochem.tohoku.ac.jp (S. Kuwahara).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.02.039

T. Nakahata et al. / Tetrahedron Letters 49 (2008) 2438–2441
2439
OR1
O
1
a, b, c
d
CHO
3
MeO2C
OH
OTES
6
TBSO
TBSO
3
4
5
OR2
OH
(1S,3R,6R)-1
A
g
e
OR
OAc
TBSO
f
TBSO
OR1
OR1
OR2
6: R = H
7: R = Ac
8
MeO2C
OH
OHC
X
X
3
C
B
h, i
j
4
1
3
6
Scheme 1. Retrosynthetic analysis of (1S,3R,6R)-1.
OHC
OH
9a: X = OAc
9b: X = H
10a: X = OAc
10b: X = H
Assuming that the absolute configurations at the C1 and
C6 chiral centers of 1 would be S and R, respectively, from
the (1S,6R)-stereochemistry assigned to 2 by Kim et al.,3
we decided to synthesize (1S,3R,6R)-1 as a candidate for
the natural stereoisomer of antifeedant 1.6 Our synthetic
plan for (1S,3R,6R)-1 featuring stereoselective installation
of the chiral centers on its cyclohexane ring via an intra-
molecular carbonyl ene reaction (C?B) is shown in
Scheme 1. Enal C was considered to be obtained from
the known hydroxy ester 3 through protections and adjust-
ment of the oxidation levels of the two oxygen functional-
ities of 3 followed by nucleophilic introduction of a prenyl
group. On the other hand, the chain-elongation of B into A
utilizing the double bond of the isopropenyl substituent
and subsequent deprotections and oxidation of A would
furnish the target molecule (1S,3R,6R)-1.
The known hydroxy ester 3, obtained by the reduction
of commercially available (R)-(+)-2-methylsuccinic acid
4-methyl ester with BH3ꢀSMe2,7 was protected as its TES-
ether, and the ester group was reduced and then protected
to give 4 (Scheme 2). Direct oxidation of the TES-oxy
group under the Swern oxidation conditions proceeded
smoothly to afford 5.8 The introduction of a prenyl group
to aldehyde 5 to form 8 was performed by a three-step
sequence of reactions: (1) nonstereoselective addition of
Scheme 2. Reagents and conditions: (a) TESCl, Et3N, DMAP, CH2Cl2,
rt; (b) DIBAL, CH2Cl2, ꢁ78 to ꢁ40 °C; (c) TBSCl, Imid, DMF, rt (95%,
three steps); (d) DMSO, (COCl)2, Et3N, CH2Cl2, ꢁ78 °C; (e) allylmag-
nesium bromide, ether, ꢁ78 °C (66%, two steps); (f) Ac2O, Et3N, DMAP,
CH2Cl2, rt; (g) 2-methyl-2-butene (as solvent), Grubbs cat. (2nd gener-
ation), rt (95%, two steps); (h) TBAF, THF, 0 °C; (i) DMSO, (COCl)2,
Et3N, CH2Cl2, ꢁ78 °C (83%, two steps); (j) ZnBr2, toluene, ꢁ78 to 0 °C
(82%).
stereomers (10a) accompanied by only a small quantity
of the undesired diastereomers (<10%). The ratio of the
two major diastereomers (4-a-10a and 4-b-10a) was ca.
1:1, reflecting the diastereomeric ratio of the starting enal
9a, which means that the stereochemistry of the C4 posi-
tion of 9a bearing an acetoxy substituent had little influ-
ence on the stereochemical course of the cyclization. The
stereochemical assignment of the two diastereomers was
OH
OTES
a, b
c
10a
TBSO
TBSO
11
12
allylmagnesium bromide to give
6
(diastereomeric
separation
ratio = ca. 1:1); (2) protection of the resulting alcohol 6
to acetate 7; and (3) cross-metathesis reaction of the termi-
nal olefin 7 with 2-methyl-2-butene.9 The metathesis step to
furnish 8 proceeded quite efficiently (ca. 95% yield),
although trace amounts of olefinic byproducts (<5%) were
also produced, as judged by 1H NMR analysis.10 Deprotec-
tion of the TBS group of 8 and the Swern oxidation of the
resulting alcoholic intermediate afforded enal 9a, which set
the stage for the stereoselective construction of cyclic inter-
mediate 10a. It is well established that the treatment of 9b
(deacetoxy analog of 9a) with ZnBr2 or ZnI2 induces a
highly stereoselective cyclization via intramolecular car-
bonyl ene reaction, giving 10b with 1,3-cis-1,6-trans relative
stereochemistry.11 According to this protocol, we treated
enal 9a with ZnBr2 in toluene and found that the cycli-
zation reaction proceeded stereoselectively to afford a mix-
ture consisting mainly of the desired 1,3-cis-1,6-trans-dia-
OH
OH
TBSO
11a
TBSO
11b
NOE
J = 10.3 Hz
OTBS
d
H
H
O
HO
Me
H
H
J = 10.3 Hz
11a
TBSO
13
Scheme 3. Reagents and conditions: (a) TBSOTf, 2,6-lutidine, CH2Cl2,
ꢁ78 °C; (b) DIBAL, CH2Cl2, ꢁ78 °C (85%, two steps); (c) TESOTf, 2,6-
lutidine, CH2Cl2, ꢁ78 °C (quant); (d) DMSO, (COCl)2, Et3N, CH2Cl2,
ꢁ78 °C (ca. 90%).

2440
T. Nakahata et al. / Tetrahedron Letters 49 (2008) 2438–2441
31:3
conducted after converting the diastereomeric mixture 10a
into 11 through protection (TBSOTf, 2,6-lutidine) and
reduction (DIBAL) (Scheme 3). The diastereomeric mix-
ture of alcohols (11) could readily be separated by SiO2 col-
umn chromatography to provide pure samples of 11a and
11b, which were subjected to 1H NMR analysis
(500 MHz, CDCl3). In the a-alcohol 11a, the coupling con-
stants between 1-H and 6-H, and 3-H and 4-H were both
10.3 Hz, and a NOE correlation was observed between
4-H and 6-H, supporting the stereochemistry as shown in
Scheme 3.12 On the other hand, the signals for 1-H
and 4-H of 11b were observed as a double triplet
(J1,6 = 10.3 Hz) and seemingly as a broad singlet with a
narrow half-width of 8.5 Hz, respectively.12 Furthermore,
both 11a and 11b converged into the same ketone 13 when
exposed to the Swern oxidation conditions. These results
confirmed our stereochemical assignments for 11a and
11b, and thereby for the ene cyclization products 10a as
well. In our actual synthesis, the mixture of alcohols 11
was treated, without separation, with TESOTf and 2,6-
lutidine to give 12, and the resulting epimeric mixture
was subjected to the next step.
(½aꢂD +38 (c 0.087, CHCl3)).6,16 On the basis of these
results, the stereochemistry of antifeedant 1 was unambi-
guously determined to be 1S, 3R, and 6R.
In conclusion, the enatioselective synthesis of (1S,3R,
6R)-1-hydroxy-7(14),10-bisaboladien-4-one (1) was accom-
plished in 15% overall yield from the known hydroxy ester
3 by a 19-step sequence involving the stereoselective intra-
molecular carbonyl ene reaction of 9a into 10a as the key
step. Good agreement between the synthetic and natural
products in spectral data and specific rotation enabled us
to unambiguously determine the absolute stereochemistry
of the antifeedant as 1S, 3R, and 6R. The synthesis of
the other antifeedant component 2 and related natural
products is currently in progress.
Acknowledgments
We are grateful to Professor Kim (Kochi University) for
providing the copies of the NMR spectra of the natural
antifeedant (1). We also thank Ms. Yamada (Tohoku Uni-
versity) for measuring NMR and MS spectra. This work
was supported, in part, by a Grant-in-Aid for Scientific
Research (B) from the Ministry of Education, Culture,
Sports, Science and Technology of Japan (No. 19380065).
The final stage of our synthesis required the installation
of a prenyl group at the allylic methyl position of 12
(Scheme 4). For this purpose, the olefinic compound 12
was first exposed to ozonolysis conditions to afford ketone
14, which was then prenylated by a conventional method
(LDA, prenyl bromide, THF/HMPA) to furnish 15. Treat-
ment of ketone 15 with the Nysted reagent gave methyl-
enated product 16,13 and subsequent selective deprotection
of its TES group (TBAF, AcOH, DMF)14 followed by oxi-
dation (Dess–Martin’s periodinane) and removal of the
TBS protecting group (aq HF, CH3CN)15 gave
(1S,3R,6R)-1 as a white crystalline solid (mp 79.5–80 °C,
lit.6 75.5–76 °C). The 1H and 13C NMR spectra of
(1S,3R,6R)-1 recorded in CDCl3 at 500 MHz and
125 MHz, respectively, were identical with those of the
natural product,5 and, furthermore, the specific rotation
References and notes
1. Chen, X. H.; Kim, C.-S.; Kashiwagi, T.; Tebayashi, S.; Horiike, M.
Biosci. Biotechnol. Biochem. 2001, 65, 1434–1437.
2. Nagahama, S.; Tazaki, M.; Nomura, H.; Nishimura, K.; Tajima, M.;
Iwasita, Y. Mokuzai Gakkaishi 1996, 42, 1127–1133.
3. Kim, C.-S.; Morisawa, J.; Nishiyama, N.; Kasiwagi, T.; Tebayashi,
S.; Horiike, M. Biosci. Biotechnol. Biochem. 2002, 66, 1997–2000.
4. Morisawa, J.; Kim, C.-S.; Kashiwagi, T.; Tebayashi, S.; Horiike, M.
Biosci. Biotechnol. Biochem. 2002, 66, 2424–2428.
5. Kashiwagi, T.; Wu, B.; Iyota, K.; Chen, X. H.; Tebayashi, S.; Kim,
C.-S. Biosci. Biotechnol. Biochem. 2007, 71, 966–970.
6. Compound 1 had previously been isolated from C. japonica by
Nagahama et al., and its absolute stereochemistry had been proposed
to be 1S, 3R, and 6R. However, their stereochemical assignment was
based only on the chemical shift of its 3-Me (dC 14.2 ppm) and the
sign of its specific rotation (+), and therefore seemed to lack definite
proofs. Nagahama, S.; Iwaoka, T.; Ashitani, T. Mokuzai Gakkaishi
2000, 46, 225–230.
22
of (1S,3R,6R)-1 (½aꢂD +41 (c 0.29, CHCl3)) matched that
reported for the natural product by Nagahama et al.
OTES
OTES
7. Abo, M.; Mori, K. Biosci. Biotechnol. Biochem. 1993, 57, 265–267.
8. (a) Tolstikov, G. A.; Miftakhov, M. S.; Vostrikov, N. S.; Komissar-
ova, N. G.; Adler, M. E.; Kuznetsov, O. M. Zh. Org. Khim. 1988, 24,
224–225; (b) Muzart, J. Synthesis 1993, 11–27.
9. (a) Chatterjee, A. K.; Sanders, D. P.; Grubbs, R. H. Org. Lett. 2002,
4, 1939–1942; (b) Tsukano, C.; Siegel, D. R.; Danishefsky, S. J.
Angew. Chem., Int. Ed. 2007, 46, 8840–8844.
10. It is known that the cross-metathesis reactions of 2-methyl-2-butene
with terminal olefins give, in some cases, small amounts of 1,2-
disubstituted olefins bearing a terminal ethylidene group as well as
desired trisubstituted olefins with a terminal isopropylidene group
such as 8.9 By analogy with those facts, coupled with an additional
experimental result obtained in a cross-metathesis reaction between
the pivaloyl-protected analog of 7 and 2-methyl-2-butene, which gave
considerable amounts (ca. 25%) of byproducts enough for their
structural analysis by NMR (Kuwahara, S., unpublished work), we
consider that the olefinic byproducts produced in the conversion of 7
into 8 would also be 1,2-disubstituted olefins (epimers due to the
a
b
12
TBSO
14
O
TBSO
c
X
15: X = O
16: X = CH2
O
d, e, f
O
BrZn
ZnBr
Zn
OH
(1S,3R,6R)-1
Nysted reagent
Scheme 4. Reagents and conditions: (a) O3, Py, MeOH, CH2Cl2, then
Me2S, ꢁ78 °C to rt (99%); (b) LDA, prenyl bromide, HMPA, THF,
ꢁ78 °C to rt (90%); (c) Nysted reagent, TiCl4, THF, ꢁ78 °C to rt (65%);
(d) TBAF, AcOH, DMF, rt; (e) DMP, Py, CH2Cl2 (95%, two steps); (f) aq
HF, CH3CN, rt (80%).

T. Nakahata et al. / Tetrahedron Letters 49 (2008) 2438–2441
2441
acetoxy-bearing chiral center and, probably, geometrical isomers as
well). Detailed discussions on the structures of the olefinic byproducts
will soon be reported elsewhere.
J = 14.2, 3.4 Hz), 2.34 (1H, ddd, J = 13.5, 10.3, 3.9 Hz), 3.49 (1H, dt,
J = 4.4, 10.3 Hz), 3.76 (1H, br s), 4.76 (1H, br s), 4.76–4.78 (1H, m).
13. (a) Nysted, L.N. U.S. Patent 3,865,848, 1975; Chem. Abstr. 1975, 83,
10406q; (b) Matsubara, S.; Sugihara, M.; Utimoto, K. Synlett 1998,
313–315; (c) Pasetto, P.; Franck, R. W. J. Org. Chem. 2003, 68, 8042–
8060.
11. (a) Nakatani, Y.; Kawashima, K. Synthesis 1978, 147–148; (b)
Imakura, Y.; Yokoi, T.; Yamagishi, T.; Koyama, J.; Hu, H.;
McPhail, D. R.; McPhail, A. T.; Lee, K.-H. J. Chem. Soc., Chem.
Commun. 1988, 372–374.
12. 1H NMR data for 11a and 11b (500 MHz, CDCl3). 11a: d –0.01 (3H,
s), 0.02 (3H, s), 0.84 (9H, s), 1.02 (3H, d, J = 6.3 Hz), 1.15 (1H, dt,
J = 10.3, 12.7 Hz), 1.37–1.46 (1H, m), 1.39 (1H, dt, J = 10.3,
12.7 Hz), 1.53 (1H, br s, OH), 1.68 (3H, s), 1.80–1.87 (2H, m), 2.08
(1H, ddd, J = 12.7, 10.3, 3.4 Hz), 3.21 (1H, dt, J = 3.9, 10.3 Hz), 3.51
(1H, dt, J = 4.4, 10.3 Hz), 4.74 (1H, br s), 4.77 (1H, br s). 11b: d –0.01
(3H, s), 0.02 (3H, s), 0.84 (9H, s), 0.98 (3H, d, J = 6.8 Hz), 1.36 (1H,
d, J = 2.9 Hz, OH), 1.45 (1H, dt, J = 10.3, 12.7 Hz), 1.51 (1H, dt,
J = 2.9, 13.5 Hz), 1.56–1.64 (2H, m), 1.68 (3H, s), 1.77 (1H, dt,
14. Higashibayashi, S.; Shinko, K.; Ishizu, T.; Hashimoto, K.; Shira-
hama, H.; Nakata, M. Synlett 2000, 1306–1308.
15. Treatment of the TBS ether with TBAF/THF brought about a
substantial degree of epimerization at the C3 stereogenic center.
22
16. The specific rotation of (1S,3R,6R)-1 measured in MeOH (½aꢂD +37.1
(c 0.29, MeOH)) was considerably different in magnitude from that
20
reported for the natural product by Kim et al. (½aꢂD +15.0 (c 0.1,
MeOH)).5 We consider that this discrepancy would be ascribable to
impurities contained in the natural sample of 1, judging from its 1H
NMR spectrum provided to us by Professor Kim.