Biogenetic-Type Synthesis of (+)-Cymbodiacetal, a Constituent of Cymbopogon martinii

Article abstract of DOI:10.1021/np030338h

A biogenetic-type synthesis of (+)-cymbodiacetal (1), a novel bismonoterpenoid dihemiacetal, is described.

Full text of DOI:10.1021/np030338h

7
00
J . Nat. Prod. 2004, 67, 700-702
Biogen etic-Typ e Syn th esis of (+)-Cym bod ia ceta l, a Con stitu en t of Cym bopogon
m a r tin ii†
‡
§
⊥
⊥
,‡
Asha M. D’Souza, Shashikumar K. Paknikar, Vasu Dev, Philip S. Beauchamp, and Shrivallabh P. Kamat*
Department of Chemistry, Goa University, Goa 403 206, India, Siddharth Chemicals, Kundai Industrial Estate, Kundai,
Goa 403 115, India, and Department of Chemistry, California State Polytechnic University, Pomona, California 91768
Received J uly 29, 2003
A biogenetic-type synthesis of (+)-cymbodiacetal (1), a novel bismonoterpenoid dihemiacetal, is described.
Cymbodiacetal (1) was isolated from the essential oil of
the aerial parts of flowering Cymbopogon martinii (Grami-
Sch em e 1. Possible Biogenetic Pathway for Cymbodiacetal (1)
1
naceae). The structure of 1 was established by spectro-
scopic methods (MS, IR, 1H and C NMR) and further
13
confirmed by X-ray diffraction studies of its 1:1 solvate with
CDCl , which also established its absolute stereochemistry.
3
The stereochemistry at C-2 and C-7 of 1 was shown to be
the same as that in R-(+)-limonene (2), the other compo-
nent isolated from the essential oil.1
The biogenesis of 1 involving the key intermediate 3
(
Scheme 1) looked more attractive than the earlier pro-
Sch em e 2. Synthetic Route to Cymbodiacetal (1)
1
posal especially in view of the fact that a large number of
natural dimers, including 3, have been reported to be
formed by heteroatom Diels-Alder self-dimerization.2
We have now achieved biogenetic-type synthesis of (+)-
cymbodiacetal (1) starting with (+)-limonene oxide (4) via
the key intermediate 3. This is the first report of the
synthesis of 1.1
-11
Our synthetic strategy was based on the assumption that
the dimer 3 should be accessible through simple Diels-
Alder dimerization of the conjugated enone 5, which, in
turn, can be obtained from epoxide 4 (Scheme 2). The dimer
3
could then be transformed into monoepoxide 6 on
treatment with 1 equiv of moist peroxy acid. The free acid
formed in the reaction could open the epoxide 6 in the
desired manner by the participation of ether and carbonyl
oxygens followed by attack of water to give 1 (Scheme 2).
Reaction of (+)-limonene oxide (4) with lithium diiso-
propylamide (LDA)12 followed by oxidation of the resultant
allylic alcohols (exocarveols 7) with pyridiniumchlorochro-
mate (PCC)13 gave the expected enone 5 in 68% yield. In
the mobile phase gave 3 of 97% purity, and it was used
immediately for the acquisition of its H, C, and other
1
2
view of the unstable nature of 5, no other physical data
except GC/MS were recorded for further characterization.
R-Alkylidenecyclanones are reported to undergo regiose-
lective heteroatom Diels-Alder dimerization on standing
1
13
1
3
D NMR data. The H chemical shift values agree closely
10
with the previously reported values for 3, and the
structure is further supported by the C assignments being
reported here. The latter were derived from the C DEPT
and H- C COSY data. The GC/MS of 3 showed promi-
nent peaks at m/z 300 (M , 55), 150 (23), and 107 (100).
13
at room temperature3 to give spirochroman derivatives.
,6
11
13
Therefore, the enone 5 was kept in a stoppered flask at
room temperature. A week later, GC showed the appear-
1
13
+
ance of a peak (≈10%) at retention time t
considerable reduction in the peak size of 5 at t
min, which disappeared almost completely at the end of
0 days. Careful column chromatographic separation over
R
) 66.5 min and
The genesis of the major fragment ions in the mass
spectrum of 3 is shown in Scheme 3.
R
) 29.9
The transformation of 3 into (+)-cymbodiacetal (1) took
place in an unusually simple and unexpected manner,
supporting the proposed biogenetic pathway (Scheme 1).
Interesting observations were made while handling 3. It
was difficult to obtain 3 in pure form, and the GC/MS of
1
silica gel using 5% diethyl ether in hexane gave fractions
rich in dimer 3 (77%), as indicated by GC. Further
purification by HPLC using 5% diethyl ether in hexane as
9
7% pure 3 showed an additional peak at m/z 316. A
†
Dedicated to the late Professor Albert T. Bottini, Department of
reasonable explanation for these observations could be a
slow transformation of 3 into 6 by air oxidation.
Therefore, the flask containing 3 was exposed to diffused
daylight expecting to obtain the desired epoxy intermediate
6 without using peroxy acid. Most likely 6 was formed but
Chemistry, University of California, Davis, CA 95616.
*
1
4,15
To whom correspondence should be addressed. Tel: + 91-832-2454317.
Fax: + 91-832-2452889. E-mail: srikamat@sancharnet.in.
‡
Goa University.
Siddharth Chemicals.
California State Polytechnic University.
§
⊥
1
0.1021/np030338h CCC: $27.50
© 2004 American Chemical Society and American Society of Pharmacognosy
Published on Web 02/13/2004

Notes
J ournal of Natural Products, 2004, Vol. 67, No. 4 701
NaCl (50 mL), and dried over anhydrous MgSO . Evaporation
Sch em e 3. Mass Spectral Fragmentation of Dimer 3
4
of the solvent gave a viscous oil (2.38 g, 80.4%). TLC of this
oil (hexane/diethyl ether, 9:1) showed a single spot under UV.
However, GC showed a major peak (67.6%) at t ) 29.9 min
R
in addition to some minor peaks. It was purified by column
chromatography over silica gel (hexane/diethyl ether, 9:1) to
give 512 (2.013 g, 68%) as a colorless oil: m/z (rel intensity)
+
1
50 (M , 15), 135 (30), 122 (36), 107 (60), 91 (48), 79 (100), 67
(92), 53 (32), 41 (15).
Dim er iza tion of En on e 5 to 3. The enone 5 was kept in
a loosely stoppered flask for a period of one week. GC showed
a peak (≈10%) at t ) 66.5 min (3) and a decrease in size of
the peak at t ) 29.9 min, corresponding to 5. At the end of
0 days, the peak at t ) 29.9 min almost completely
disappeared, while the size of the peak at t ) 66.5 min
increased correspondingly. TLC (hexane/diethyl ether, 19:1)
showed a faint spot different from and just above that of 5.
Column chromatography over silica gel (hexane/diethyl ether,
R
R
1
R
R
could not be characterized because silica gel chromatog-
raphy of the contents of the flask led directly to the
isolation of crystalline cymbodiacetal (1), identified by the
correspondence of its mp, optical rotation, and other
spectral data (IR, H and C) with those reported for 1
isolated from C. martinii.1
1
R
9:1) gave fractions rich in peak at t ) 66.5 min (GC), which
were combined. GC/MS of the major peak (77% intensity)
showed m/z (rel intensity) 300 (M , 55), 151 (23), 135 (28), 107
1
13
+
(
100), 95 (50), 79 (47), 67 (32), 55 (36), 41 (30), indicating it to
1
0
be the dimer 3, molecular formula C20
28 2
H O . Further puri-
Exp er im en ta l Section
fication of a small sample (<50 mg) by preparative HPLC
(hexane/diethyl ether, 95:5) gave 97.4% pure (GC) 3 (11 mg).
Gen er a l Exp er im en ta l P r oced u r es. Chemical shifts are
expressed in parts per million (δ) relative to TMS as the
The purified sample did not solidify. 1H NMR (CDCl
, 300
and 2CH),
-CO), 4.58 (2H, s, -Cd
); C NMR (CDCl , 75 MHz) δ 20.3
, C-9 or C-9′), 20.8 (CH , C-9 or C-9′), 22.6 (CH , C-10),
5.4 (CH , C-6′), 27.3 (CH , C-10′), 27.7 (CH , C-5), 28.5 (CH
, C-6), 38.9 (CH , C-3′), 41.7 (CH, C-4′), 43.4
, C-3), 48.5 (CH, C-4), 79.3 (C, C-1), 105.5 (C, C-1′), 108.7
, C-8 or C-8′), 109.9 (CH , C-8 or C-8′), 143.8 (C, C-2′),
3
1
13
internal standard. H and C NMR spectra were recorded in
MHz) δ 1.65 (6H, s, 2CH
3
), 1.4-2.3 (16Hs, m, 7CH
2
CDCl at 300 and 75 MHz, respectively, on Varian Gemini 300
3
2.8 (2H, t, J ) 12.1, 11.6 Hz, -CH
2
1
3
and Bruker WT 300 FT-NMR spectrometers. IR spectra were
recorded on a Shimadzu 8101A FT-IR spectrophotometer. Gas
chromatographic analyses were performed on a Varian 3700
GC with a 30 m × 0.25 mm HP-5 capillary column. An
identical column was used for the GC/MS system, which
consisted of an Agilent 6865 GC interfaced with a 5973
Network Mass Selective Detector, with the mass spectrometer
operating at 70 eV. High-performance liquid chromatography
was carried out using a Waters 3000 HPLC system with a 25
cm × 1.0 cm semipreparative 10 µm silica gel column under
isochratic conditions. Column chromatography utilized 60-
CH
2
), 4.72 (2H, s, -CdCH
2
3
(CH
3
3
2
2
2
2
2
2
,
C-5′), 33.0 (CH
2
2
(
(
CH
CH
2
2
2
1
47.4 (C, C-7 or C-7′), 149.2 (C, C-7 or C-7′), 212.4 (C, C-2);
GC/MS of the minor (10%) peak showed m/z (rel intensity) 316
+
(
M , 43), 163 (26), 149 (58), 135 (72), 120 (100), 107 (86), 95
48), 79 (57), 67 (40), 55 (53), 41 (38), indicating it to be the
(
epoxide 6, molecular formula C20
28 3
H O .
Cym bod ia ceta l (1). The remaining portion of impure 3 (1.0
g) on chromatography over silica gel in diffused daylight
1
20 mesh silica gel (Acme Synthetic Chemicals). TLC was
performed with silica gel impregnated with 13% calcium
sulfate (Merck India).
(
2
hexane/diethyl ether, 4:1) gave colorless needles (0.288 g), mp
1
25
1
13 °C [lit. mp 206-7 °C]; [R]
D
+24.2° (c 3.0, CHCl
3
) [lit.
(
5R )-2-Me t h yle n e -5-(1-m e t h yle t h e n yl)-1-cycloh e x-
25
[
2
R]
D
+26° ( 5° (c 0.12, CHCl
3
)]; IR (KBr) νmax 3379 (OH),
a n ol (7). A solution of n-butyllithium in hexane (0.12 mol, 85.5
mL, 1.4 M) was added to 11.1 g (0.11 mol) of diisopropylamine
-1 1
941, 1649, 1450, 1180, 1128, 1080, 1006, 891 cm ; H NMR
, 300 MHz) δ 1.68 (3H, s, CH ), 1.49-1.92 (6H, m, C
, C11-Hs), 2.09 (2H, m, C -Hs), 4.66 (2H, s, dCH
OD, 75 MHz) δ 21.1 (CH ), 27.3 (CH , C-1 or C-3),
, C-1 or C-3), 34.1 (CH , C-11), 41.9 (CH or CH , C-2
or C-4), 42.4 (CH or CH , C-2 or C-4), 72.8 (C, C-4a), 98.9 (C,
C-10a), 109.1 (CH , dCH ), 150.7 (C, -CdCH ).
(
C
CDCl
9
3
3
8
,
C
in anhydrous diethyl ether (300 mL) at 0 °C under a N
2
13
1
2
);
atmosphere, and the solution was stirred. R-(+)-Limonene
oxide (4) [15.2 g, 0.10 mol, purchased from Aldrich Chem. Co.
NMR (CD
7.5 (CH
3
3
2
2
2
2
2
(Aldrich 21832-4) as a mixture of cis and trans isomers] in
2
anhydrous diethyl ether (60 mL) was added dropwise over a
period of 30 min. The resulting reaction mixture was allowed
to warm to room temperature and then stirred for another 12
h. The reaction mixture was cooled in an ice-bath, and water
2
2
2
Ack n ow led gm en t. The authors thank Professor Emeritus
Robert B. Bates, Chemistry Department, University of Ari-
zona, Tucson, for his critique of the manuscript and to one of
the referees for drawing our attention to ref 10, and S.P.K.
wishes to thank the Chemistry Department, California State
Polytechnic University, Pomona, for a Visiting Scholar Lec-
tureship.
(
300 mL) was added. The ether phase was separated and
washed successively with 100 mL of 2 N HCl, water, saturated
aqueous NaHCO , and saturated NaCl. The aqueous phase
3
and each washing was extracted two times each with 50 mL
portions of diethyl ether. The ether extracts were combined,
4
dried over anhydrous MgSO , and distilled under reduced
pressure through a short distillation head to yield 712 (81%).
GC analysis of 7 indicated it to be a mixture of two diastere-
Refer en ces a n d Notes
omers with t
R
) 27.95 and 30.1 min in the ratio of 1:1.82,
(1) Bottini, A. T.; Dev, V.; Garfagnoli, D. J .; Hope, H.; J oshi, P.; Lohani,
respectively. This mixture was used for the next reaction
H.; Mathela, C. S.; Nelson, T. E. Phytochemistry 1987, 26, 2301-
2
302.
without further purification. IR (film) νmax 3370 (OH), 2930,
(
2) Carreiras, M. C.; Rodriguez, B.; Lopez-Garcia, R. E.; Rabanal, R. M.
2
850, 1640 (CdC), 1430, 890 cm-1
.
Phytochemistry 1987, 26, 3351-3353.
(3) Kakiuchi, K.; Ue, M.; Takeda, M.; Tadaki, T.; Kato, Y.; Nagashima,
(
5R )-2-Me t h yle n e -5-(1-m e t h yle t h e n yl)-1-cycloh e x-
a n on e (5). A solution of the diastereomeric mixture of 7 (3.0
g, 20 mmol) in CH Cl (40 mL) was added to a suspension of
PCC (3.0 g) in CH Cl (75 mL) and stirred at room temperature
for 2 h. The reaction mixture was diluted with diethyl ether
110 mL), stirred for 1 min, and allowed to stand overnight.
T.; Tobe, Y.; Koike, H.; Ida, N.; Odaira, Y. Chem. Pharm. Bull. 1987,
2
5, 617-631.
2
2
(
(
4) Richer, J . C.; Arlotto, R. Can. J . Chem. 1975, 53, 3294-3298.
5) Whittaker, D.; Banthrope, D. V. Chemistry of Thujone Derivatives:
Chem. Rev. 1972, 72, 305-313.
2
2
(
(6) Hikino, H.; Aota, K.; Takemoto, T. Chem. Pharm. Bull. 1967, 15,
1
929-1933.
The liquid phase was decanted from the residue (tarry mass),
(
(
7) Nakajima, T. J . Pharm. Soc. J pn. 1962, 82, 1278-1281.
8) Klinck, R. E.; De Mayo, P.; Stothers, J . C. Chem. Ind. (London) 1961,
471-472.
successively extracted with 7% NaOH (3 × 65 mL), 10% HCl
(3 × 10 mL), saturated NaHCO (2 × 50 mL), and saturated
3

7
02 J ournal of Natural Products, 2004, Vol. 67, No. 4
Notes
(
9) Takemoto, T.; Nakajima, T. J . Pharm. Soc. J pn. 1957, 77, 1157-
158.
10) Bessiere, Y.; Derguini-Boumechal, F. J . Chem. Res. (M) 1976, 3522-
540.
(13) Ghisalberti, E. L. Aust. J . Chem. 1979, 32, 1627-1630.
(14) Ngo, K.; Wong, W.; Brown, G. D. J . Nat. Prod. 1999, 62, 549-553.
15) L o¨ sing, G.; Degener, M.; Matheis, G. Dragoco Rep. 1998, 4, 181-187.
1
(
(
3
(
(
11) Desimoni, G.; Tacconi, G. Chem. Rev. 1975, 75, 651-692.
12) Wang, Q.; Fan, S. Y.; Wong, N. C.; Li, Z.; Fung, B. M.; Twieg, R. J .;
Nguyen, H. T. Tetrahedron 1993, 49, 619-638, and ref 16 therein.
NP030338H