Sesquiterpenes from Cupressus macrocarpa foliage

Article abstract of DOI:10.1016/j.phytochem.2004.11.002

Ten sesquiterpenes, many with unusual carbon skeletons, were identified in foliage of Cupressus macrocarpa. These are (-)-10-epi-β-acoradiene; ent-widdra-2,4(14)-diene; (E)-iso-γ-bisabolene, i.e., (4E)-4-(1,5- dimethylhex-5-enylidene)-1-methylcyclohexene; (-)-cumacrene, i.e., (4S)-4-[(1R,2S)-2-isopropenyl-1-methylcyclobutyl]-1-methylcyclohexene; (-)-α-chamipinene, i.e., (1S,6S,7S)-2,2,6,8-tetramethyltricyclo[5.3.1. 0^1,6]undec-8-ene; and five sesquiterpenes with a 3,3,4′- trimethyl-1,1′-bi(cyclohexyl) skeleton for which the trivial name macrocarpane is proposed. The possible single-enzyme biogenesis of these sesquiterpenes is discussed.

Full text of DOI:10.1016/j.phytochem.2004.11.002

PHYTOCHEMISTRY
Phytochemistry 66 (2005) 249–260
www.elsevier.com/locate/phytochem
Sesquiterpenes from Cupressus macrocarpa foliage
*
Laurence G. Cool
University of California College of Natural Resources, 1301 S. 46th St., Richmond, CA 94804, USA
Received 6 August 2004; received in revised form 3 November 2004
Available online 19 December 2004
Abstract
Ten sesquiterpenes, many with unusual carbon skeletons, were identified in foliage of Cupressus macrocarpa. These are (ꢀ)-10-
epi-b-acoradiene; ent-widdra-2,4(14)-diene; (E)-iso-c-bisabolene, i.e., (4E)-4-(1,5-dimethylhex-5-enylidene)-1-methylcyclohexene;
(ꢀ)-cumacrene, i.e., (4S)-4-[(1R,2S)-2-isopropenyl-1-methylcyclobutyl]-1-methylcyclohexene; (ꢀ)-a-chamipinene, i.e., (1S,6S,7S)-
2,2,6,8-tetramethyltricyclo[5.3.1.01,6]undec-8-ene; and five sesquiterpenes with a 3,3,40-trimethyl-1,10-bi(cyclohexyl) skeleton for
which the trivial name macrocarpane is proposed. The possible single-enzyme biogenesis of these sesquiterpenes is discussed.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Keywords: Cupressus macrocarpa; Cupressaceae; Monterey cypress; Foliage; Sesquiterpenes; Macrocarpene; Cumacrene; a-Chamipinene; epi-b-
Acoradiene; ent-Widdradiene; iso-c-Bisabolene
1. Introduction
All of the north-coastal Cupressus species produce an
essential oil that is low in sesquiterpenoids (Zavarin
et al., 1971). Briggs and Sutherland (1942) analyzed
C. macrocarpa foliar oil by fractional distillation and re-
ported only ca. 0.7% for the sesquiterpene fraction of
the oil. Our examination of 40 C. macrocarpa specimens
(20 planted and 20 from the Point Lobos population) by
gas chromatography–mass spectrometry (GC–MS) con-
firmed that sesquiterpenes are generally a minor compo-
nent of the volatile oil. However, some individuals
produced oil with up to 9% sesquiterpenoids, which in-
cluded a number of apparently novel hydrocarbons. The
identity of ten of these sesquiterpenes is the subject of
this report.
Cupressus macrocarpa Hartw. ex Gord. (Monterey
cypress) is a narrow endemic native only on Point Lobos
and Cypress Point, Monterey Co., California. It is a
large and picturesque tree, and despite its limited natural
range it is one of the best-known and most widely
planted conifers in the state.
Within Cupressus, Monterey cypress has been consid-
ered most closely related to four other coastal Califor-
nian cypress taxa with non-glandular foliage: C.
abramsiana C.B. Wolf, C. goveniana ssp. goveniana
Gord., C. goveniana ssp. pygmaea (Lemm.) Bartel, and
C. sargentii Jeps., although Wolf (1948) saw it as mor-
phologically somewhat distant from the others. The dis-
tinction of C. macrocarpa within this north-coastal
group of cypresses was supported by analysis of their fo-
liar monoterpene composition (Zavarin et al., 1971).
2. Results and discussion
A number of sesquiterpenoids were ubiquitous in
C. macrocarpa foliage, though in widely varying
amounts: a-cubebene, b-funebrene, caryophyllene, ger-
macrenes A and B, cedrol, a-bisabolol, nerolidol, and
farnesol were detected by GC–MS in nearly all trees.
*
Present address: 2408 McKinley Avenue, Berkeley, CA 94703,
USA. Tel.: +1 510 549 3972.
E-mail address: larry.cool@sbcglobal.net.
0031-9422/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2004.11.002

250
L.G. Cool / Phytochemistry 66 (2005) 249–260
There was also a distinct chemophenotype, represented
in three of the 40 trees, which produced sizable amounts
of germacrene C (detected as the Cope rearrangement
product d-elemene), (+)-guaia-6,9-diene 1, and an un-
known that is tentatively identified as (1R,4S,5R)-
guaia-6,10(15)-diene 2 (see Section 3). The chirality of
1 was determined from its retention time on enantiose-
lective GC (EGC hereinafter), which was different from
that of authentic (ꢀ)-guaia-6,9-diene. (+)-Guaia-
6,10(15)-diene has been reported from the coral Neph-
thea chabrolii (Rao et al., 2000), though the absolute
stereochemistry was not determined. The absolute
stereochemistry of putative 2 from C. macrocarpa is as-
sumed to be the same as 1.
fourth having presumably been incorporated during bio-
genesis of the dimethylcyclohexane ring.
The structures of a- and b-macrocarpene (3 and 4),
which were the major sesquiterpenes in this chemotype,
were deduced from the 1D NMR data (see Section 3).
1
The H and 13C spectra of both included seven signals
almost identical to those of the corresponding non-isop-
1
ropenyl atoms of limonene, while other H signals of 3
and 4 closely matched data for the ring protons of
1,3,3- and 1,5,5-trimethylcyclohexene, respectively
(Snowdon et al., 1987). HMQC experiments permitted
1
assignment of all H and 13C signals, and an HMBC
experiment on 4 confirmed the macrocarpane carbon
skeleton (correlations in Table 1).
A second, and more common, C. macrocarpa chemo-
type (29 of 40 trees) was typified by presence of over 25
unknown sesquiterpenes, ten of which were in suffi-
ciently high concentration for isolation and character-
ization. Known compounds (ꢀ)-b-sesquiphellandrene,
d-cuprenene, acora-3,7- and -4,7-diene (Kaiser and Nae-
geli, 1972; Marx and Norman, 1975), (ꢀ)-dauca-8,
11-diene (Cool, 2001), and the rare (ꢀ)-cuparene and
(ꢀ)-a-cuprenene, were also detected in small amounts
during isolation of the major sesquiterpenes in the oil.
Both 3 and 4 are levorotatory, and based on studies
on the optical rotation of 4- substituted 1-methylcyclo-
hexenes (Mills, 1952), their absolute stereochemistry
can be assumed to be that of (ꢀ)-limonene, i.e., 1S.
The structures of non-chiral double bond isomers
(Z)- and (E)-c-macrocarpene (5 and 6, respectively),
which occur in ca. 1:5 ratio in the oil, were inferred from
their 1H, 13C, 1H–1H NOESY, and (for 5) HMBC NMR
data (see Table 1). The spectra of 5 and 6 were very sim-
ilar and absence of a chiral center in these compounds
Compounds 3–7 proved to be irregular sesquiter-
penes with a 3,3,40-trimethyl-1,10-(bicyclohexyl) skele-
ton, for which the trivial name macrocarpane is
proposed. Greek letters and ar-, E- and Z-prefixes indi-
cate the various double-bond and geometric macrocar-
pene isomers (see Section 3). These sesquiterpenes are
remarkable in having only three terminal carbons, the
was evident from magnetic equivalence of the 1H or
13C signals from both atoms of each geminal pair.
NMR data of terpinolene (1H and 13C) and 3,3-dimeth-
ylmethylenecyclohexane (1H; Bernard et al., 1974) had
the pertinent methyl and ring signals very close to those
of 5 and 6. The macrocarpane carbon skeleton of the
predominant E isomer 6 was confirmed by hydrogena-

Table 1
13C and 1H NMR spectroscopic dataa of macrocarpenes 3–6, d from TMSb
C. No. 3 (CDCl3)
4 (CDCl3)
5 (C6D6)
13C d
6 (C6D6)
NOESYc HMBC (C. No.) 13C d
13C d
1H d
13C d
1H d
HMBC (C. No.)
3/6,7,8,12
1H d
1H d
NOESYc
1
2
41.1
27.9
1.94 m
1.42 dddd (6,12,12,13)
1.70 m
41.3
27.7
1.96 m
l.44 dddd (6,12,12,13) 1,3/6,4
ꢁ127.6
–
2.39 t (6)
–
12
–
1,3,4,6?
126.7
26.8
–
2.34 t (6)
–
8
26.9
1.72 m
1.89 m, 2.05 m
–
1,3/6,4,7?
1,2,4,5
3
4
30.7d 1.86 m, 1.98 m
30.7
133.7
120.9
30.7
32.4
134.0
121.4
29.8d 2.79 br s
1.96 br t (5)
–
5.40 br m
2?,4,5
32.3
133.9
121.5
1.97 br t (5)
–
5.39 br m
133.7
121.0
30.9d 2H, 1.98 m
–
5.37 br m
–
1,6
–
–
–
–
5
5.39 br m
2H, 1.96 m
–
–
6
1,4,5
–
8
29.7d 2.83 br s
12
–
7
139.4
130.1
31.4
–
5.09 s
140.8
40.5
29.0
128.6
43.8
–
–
128.5
43.5
33.3
40.3
24.3
–
8
1.705 s, 1.709 s
–
1?,7,9?,10,12,13/14
–
1.97 s
–
6,13/14
–
13/14
1,7,9,10,13/14
–
2.01 s
–
2,13/14
–
13/14
9
–
ꢁ33.5
40.4
24.7
30.0d 2.15 br t (6)
10
11
12
13
14
15
37.5
20.2
2H, 1.35 t (7)
2H, 1.58 m
1.846 t, 1.850 t
0.912 s
35.3
23.1
2H, 1.30 t (7)
2H, 2.01 m
5.36 br m
8,9,11,12,13,14
7,9,10,12
1/8,10,11
8,9/14,10
8,9/13,10
3,4,5
1.30 m
1.50 m
8,9,11,12?,13/14
7,9,12
1.30 m
1.46 m
26.6
117.8
2
1,7,8,10,11
8,9,10,14
8,9,10,13
3,4,5
30.4d 2.09 br t (6)
6
30.3e
30.4e
23.5
28.0d 0.890 s
28.4d 0.882 s
28.6
28.6
23.6
0.88 s
0.88 s
1.63 br m
8,10
8,10
3,5
28.7
28.7
23.6
0.89 s
0.89 s
1.64 br m
8,10
8,10
3,5
0.915 s
1.62 br s
23.5
1.64 br s
a
b
c
1H–1H coupling constants (Hz) in parentheses; query (?) = weak or indistinct 2D signal; slash (/) = and/or.
Solvent references, d from TMS: CDCl3 1H 7.26, 13C 77.0; C6D6 1H 7.15, 13C 128.0.
Two- and three-bond correlations not listed.
d,e
Column entries with the same superscript may be reversed.

252
L.G. Cool / Phytochemistry 66 (2005) 249–260
tion, which yielded the same two isomeric Mr = 208
compounds as were obtained from hydrogenation of 4.
The geometry of the tetrasubstituted double bond was
determined by NOESY experiments, in which the deci-
sive cross-peaks were from protons 2 ! 12 and 6 ! 8
in 5 vs. from 2 ! 8 and 6 ! 12 in 6 (Table 1).
equatorial H-6a to H-13 in one conformation or to
H-15 in the other, while for the C-1 epimer neither of
these interactions would be expected. The NOESY
experiment in CDCl3 was not definitive due to overlap
of key signals, but in C6D6, where the broad H-6a mul-
tiplet as well resolved, there was a cross-peak indicating
an H-13/H-6a correlation. Furthermore, cross-peaks
from H-15 to H-6b and -6a appeared which suggested
the participation of both conformations of 8 in the ob-
served NOEs.
It should be noted that 5 and 6 were very unstable
when purified, being oxidized and isomerised within
minutes unless preserved with an antioxidant.
Minor component ar-macrocarpene 7 (Mr = 202) had
1H NMR signals (see Section 3) that indicated its basic
structural features: a bicyclic sesquiterpene consisting of
a 1,4-disubstituted benzene ring (4H, d 7.06) with a
methyl group (3H, d 2.18) and attached to a saturated
cycloalkyl ring bearing a geminal methyl pair (two sing-
lets, d 0.91 and 0.93, 3H each). The macrocarpane struc-
ture suggested by these features was confirmed by GC
comparison of the hydrogenation products of 7 with
those of 4 and 6. All three starting materials produced
the same two isomeric macrocarpanes, though in differ-
ent ratios (see Section 3). Since 7 is chiral and was found
to be dextrorotatory (aD = +7.2ꢁ), its enantiomeric pur-
ity was investigated by EGC. Neither 7 itself nor its
hydrogenation products showed evidence of separable
enantiomers. However, EGC analysis of the hydrogena-
tion product mix from 6, which must contain both enan-
tiomers of each diastereomer, also showed no separation
of the enantiomeric pairs. Thus there is no assurance
that 7 is optically pure. However, the fact that it is not
racemic suggests that its biosynthesis involves a stereo-
specific enzyme-mediated hydride shift from C-1 to
C-7 before aromatization. The absolute stereochemistry
shown is based on the fact that 3,3-dimethylcyclohex-
anes with achiral 1R substituents are reported to be dex-
trorotatory (Ansari, 1973).
The depicted absolute stereochemistry of 8 is based
on the assumption that the compound derives from
the same (S)-bisabolyl precursor as congeners 3 and 4
(see discussion below).
A sesquiterpenoid carboxylic acid with the 1-(2-iso-
propyl-1-methyl-cyclobutyl)-4-methylcyclohexane (‘‘dun-
niane’’) carbon skeleton has been described from
Illicium dunnianum (Sy and Brown, 1998). Besides the
C-14 carboxyl functionality, this compound differs from
8 in having cis relative stereochemistry at C-7 and -8.
Thus the cumacrane and dunniane carbon skeletons
bear the same stereochemical relationship as the cyclo-
butane monoterpenoids grandisol 9 and fragranol 10,
respectively. In fact, co-occurrence of 8 with the macro-
carpenes in C. macrocarpa suggests a biosynthetic paral-
lel between sesquiterpene 8 and monoterpenoid 9, in
that the latter also occurs naturally with corresponding
monoterpenoid dimethylcyclohexane congeners, e.g.,
2-(3,3-dimethylcyclohexylidene)ethanol
et al., 1971).
(Tumlinson
Compound 11, another minor component (ꢁ0.5% of
total sesquiterpenes), had a MS quite similar to that of
thujopsene. 1D NMR of the purified compound showed
that it is a tricyclic sesquiterpene with four methyls (one
of them vinyl) on quaternary carbons. Unlike thujop-
sene, there were no high-field signals indicative of a
cyclopropane ring. The NMR spectroscopic data (in
C6D6, see Section 3) were very similar to the literature
values (in CDCl3) for one candidate structure, 2,2,6,8-
tetramethyltricyclo[5.3.1.01,6]undec-8-ene (Naegeli and
Wetli, 1981). A 1H NMR experiment with the C. macro-
carpa unknown in CDCl3 gave spectral data identical to
the literature values, confirming its identification.
The compound reported by Naegeli and Wetli (1981)
was a by-product in a synthesis of the 1,1,5,9-tetrameth-
ylspiro[5.5]undecane (chamigrane) skeleton. The
authors note that acid-catalyzed scission of the C-1/C-
10 bond of 11 leads directly to the chamigrenes. This
relationship and the presence of a pinene substructure
prompt the proposal of a-chamipinene as the trivial
name for 11. None of the related compounds found by
Naegeli and Wetli (which by analogy would be called
b-chamipinene, chamicamphene, and a-chamifenchene)
was detected in C. macrocarpa.
The structure of cyclobutanoid sesquiterpene 8,
which is given the trivial name cumacrene, was deter-
mined by a series of 2D NMR experiments (data in
1
Table 2). H–1H TOCSY defined the coupled proton
system H-8 ! H-9ab ! H-10ab, and HMQC gave the
proton-carbon assignments. HMBC gave 2- and 3-bond
heteronuclear correlations that were only consistent
with a cyclobutane ring substituted as shown. The trans
relative stereochemistry at C-7 and C-8 was clear from a
strong NOESY peak between H-8 and H-15. The
attachment of the 4-methylcyclohexene ring at C-1 was
required by the TOCSY and HMBC data.
The C-1/C-7 relative stereochemistry was investigated
by comparing results of NOESY experiments (Table 2)
with minimum-energy (MM2 force field) conformation
calculations for 8 and its C-1 diastereomer. The calcula-
tions for 8 found two low-energy conformations differ-
ing by just 0.18 kcal molꢀ1, while for the C-1 epimer
the steric energy difference between the two lowest-
energy conformations was 1 kcal molꢀ1. Potentially
informative close contacts for 8 would be from quasi-
Naegeli and Wetli (1981) synthesized racemic mix-
tures in their work, so there was no information on

Table 2
13C and 1H NMR spectroscopic dataa of sesquiterpenes 8 and 12, d from TMSb
C. No. 8 (CDCl3)
8 (C6D6)
12 (CDCl3)
13C d 1H d
TOCSYc NOESYd
HMBC (C. No.)
13C d 1H d
NOESYd
13C d 1H d
COSYe
NOESYd,e
–
1
2
37.8 1.69 m
24.3 b 1.20 m
a 1.73 m
31.3 b 1.91 br m
a 2.00 br m
2b
15?
15
2,3,7,15
1,3,6
38.2 1.79 m
24.7 b 1.20 m
a 1.70 m
31.6 1.86 br dm
1.98 br m
15?
15
44.6 –
39.1 eq 1.93 br dm
–
3,14
1,3ab
3ab?
2a?b
2b
3,4
4
ax 2.01 br dm 3,14
7
3
14
14
–
14
14
–
120.9 5.37 br m
2ax,eq;14
4?
4
5
133.7 –
121.4 5.33 br d (4)
–
–
1,3,6,14
133.3 –
121.9 5.43 br d (4)
133.9
–
–
–
15
6ab
14
14
28.8 ax 1 .93 br dm
eq 2.01 br dm
24.2 2H, 1.46 m
6
6
5ax,eq
6
26.5 b 1.69 m
a 1.93 m
46.2 –
52.5 2.56 t (9)
5?
5
15?
15?
–
27.0 b 1.77 m
15
12Z,13,15
2ax,13?
a 2.05 br dm (14) 13,15
46.4 –
52.8 2.48 t (8.9)
7
8
–
–
1,7,9,10,11,12,13,15
–
13,15
58.0 2.18 m
27.9 b 1.68 m
a 1.79 m
31.4 a 1.35 m
b 1.73 m
44.0 1.75 m
8ab
9ab
13,15
7,9ab
7,9ab
8ab,10
8ab,10
9a,15
12Z,15?
15
9
19.9 b 1.86 m
a 1.94 m
29.6 b 1.55 dt (9,11) 9ab
a 1.79 m
8,10b
8,10b
7,8,10
7,8,10,11
1,7,8,9,15
1,7,8,9
–
20.2 b 1.83 m
a 1.94 m
29.9 b 1.46 dt (9,11)
a 1.76 m
12Z
15
12Z
15
10
11
12
146.8
–
–
–
13
146.6
–
–
13
147.1 –
111.8 (Z) 4.72 br s
–
–
110.0 (E) 4.80 br s
(Z) 4.71 br s
8,13
110.5 (E) 4.86 br s
(Z) 4.82 br s
7?
(E) 4.83 br m 13
6,8a,15
13
9a
8,12E,15
9a
13
14
15
23.8 1.77 br s
23.4 1.63 br s
25.2 1.11 s
8,12
3,4,5
23.9 1.70 br s
23.6 1.64 br s
25.4 1.03 s
6a,8,12E, 15
3ab,5
l?,2b,6ab,8,10b,13
24.1 1.77 s
23.3 1.61 br s
15.8 0.94 d (7)
6,7?,12E
–
3ab,5
2ax,eq
10
l?,2b,6a?b?, 8,10b,13 1,7,8,10
8a?,9a,5ax, 6,12Z
a
b
c
1H–1H coupling constants (Hz) in parentheses; query (?) = weak or indistinct 2D signal.
Solvent references, d from TMS: CDCl3 1H 7.26, 13C 77.0; C6D6 1H 7.15, 13C 128.0.
24 ms mixing time used to limit correlations to 2 and 3 bonds; two-bond correlations not listed.
Two- and three-bond correlations not listed.
d
e
Two-bond correlations not listed; assignments for H-2 and -5 confirmed by COSY (or NOESY) experiment in C6D6.

254
L.G. Cool / Phytochemistry 66 (2005) 249–260
the optical rotation of the enantiomers of these
compounds. But since the absolute stereochemistry of
a- and b-chamigrene is known, it seemed possible to
determine that of 11 (aD ꢀ4.3ꢁ) by acid-catalyzed rear-
rangement followed by enantioselective analysis of any
chamigrenes produced. When this was done, 11 com-
pletely rearranged to two major products that were iden-
tical to (ꢀ)-a- and (ꢀ)-b-chamigrene when analyzed by
GC–MS and dual-column GC-FID, but differed in
retention time from the (ꢀ) enantiomers on EGC.
Although (+)- and (ꢀ)-a-chamigrene were only slightly
separated, the b-chamigrene enantiomers were resolved
to the baseline, showing that the enantiomeric purity
of the (+)-b-chamigrene, and therefore of 11 itself, is
very high.
to ent-14c/d (see Section 3). An unexpected finding in
this experiment was that although no unreacted 12 re-
mained, a large peak corresponding to ent-c-acoradiene
[acora-4,7(11)-diene] was seen.
A search through our in-house library of plant terpe-
noid GC–MS analyses showed that the mass spectrum
of a sesquiterpene from the liverwort Pellia epiphylla
(L.) Corda, collected in Humboldt Co., CA, very closely
matched that of 12. GC–MS, dual-column GC-FID,
and EGC comparisons under identical conditions
showed that the P. epiphylla component, present at a
concentration of ca. 2% in the oil, had MS and retention
data exactly the same as that of 12 from C. macrocarpa.
Thus the former is probably also (ꢀ)-10-epi-b-
acoradiene.
Acoradiene 12 is the C-10 epimer of (+)-b-acoradi-
ene. Its GC retention time was slightly longer than that
of b-acoradiene, while its mass spectrum was very simi-
lar, the only significant difference being the somewhat
larger ratio of ion m/z 108 to ion m/z 107. The structure
and absolute stereochemistry of 12 were elucidated by a
combination of 2D NMR, molecular mechanics (MM2)
calculation, hydrogenation experiments, and EGC.
A compound with the same structure as 12 (or its
enantiomer) has been reported as an aggregation phero-
mone in the flour beetle Gnatocerus cornutus (Tebayashi
et al., 1998). However, the published NMR and MS
data are clearly different from those of 12 . Since no con-
formational analysis was done, the authorsꢀ interpreta-
tion of the NOESY data for the beetle pheromone
may be open to question. Of the four acora-3,11- and
-4,11-diene diastereomers, only 12 and a-acoradiene
are consistent with the reported NOESY cross-peak
from H-15 to non-allylic H-6 or -2. But since the a-aco-
radiene NMR spectroscopic data referred to in the pa-
per was from a low-resolution experiment run in CCl4
rather than CDCl3, the possibility that the unknown (re-
ported aD = +37.1ꢁ) might actually be the enantiomer of
(ꢀ)-a-acoradiene 13 was missed.
1
1H–1H COSY and H–13C HSQC experiments indi-
cated the acoradiene structure and allowed assignment
of the 1H and 13C signals (Table 2). Position of the endo-
1
cyclic double bond at C-3 was inferred from the 39
ppm 13C signal for C-2, which is significantly different
from that of the corresponding allylic carbon of a-
and b-acoradiene (d ꢁ31 ppm). This downfield chemical
shift can be explained by absence of a c substituent
interaction between Me-15 and C-2. Examination of
the MM2 minimum-energy conformations of all four
diastereomers of acora-3,11- and -4,11-diene shows that
To check this, a high-resolution (400 MHz) 1H NMR
spectrum was run on authentic 13 in CDCl3. The result-
ing signals were identical to those reported by Tebayashi
et al. (1998), with two exceptions: vinyl H-5 is at d 5.30,
not 5.60 as reported; and vinyl Me-14 is at d 1.63, not
1.67. Both of these discrepancies are presumably due
to typographical errors: the depicted COSY spectrum
clearly shows Me-14 at d 1.63, and the authors charac-
˚
in three of them H-15 is within 2.45 A of the nearest
allylic proton (H-2 or -6). Only in 12 is this distance
˚
large enough (ca. 4.7 A) to preclude the upfield shift typ-
ical for this allylic carbon.
GC–MS, dual-column GC-FID, and EGC compari-
son of the hydrogenation products of 12 and authentic
(ꢀ)-a-acoradiene 13 gave final proof of the relative
and absolute stereochemistry at C-7 and C-10. Hydroge-
nation of 13 gave two major saturated (Mr = 208) com-
pounds 14a/b, along with small amounts of isopropyl
epimers 14c/d. Compound 12 gave nearly equal amounts
of ent-14c/d and ent-14a/b. The assumption that ent-14a/
b were artefacts due to acid-catalysed isomerization at
C-7 was tested by repeating the hydrogenation of 12
with addition of a small amount of pyridine. Though
hydrogenation was less complete (there were now sizable
amounts of monounsaturated Mr = 206 compounds),
the amount of ent-14a/b was greatly reduced compared
1
terize the H signals of the G. cornutus acoradiene as
‘‘similar to those in the literature’’ for a- and b-acorad-
iene, both of which have the vinyl methyl at d 5.25 (CCl4
data). Thus the identity of the G. cornutus aggregation
pheromone is revised to (+)-a-acoradiene.
A small amount of compound 15, with a widdrane
carbon skeleton, was isolated and identified by 1D
and 2D NMR spectroscopic experiments (Table 3).
The essential features of the molecule were suggested
by the 1D spectrum: a bicyclic sesquiterpene with
three methyl singlets and a disubstituted double bond
flanked by (a) methine proton H-1 (d 1.75, doublet) at
a bridgehead carbon; and (b) exo-methylene-bearing
quaternary carbon C-4. A TOCSY experiment con-
firmed three correlated proton systems: the allylic pro-
tons at C-5 were coupled to H-6a,b; three methylenes
at C-8, -9, and -10 formed a second coupled system;
1
In this discussion, numbering of the acoradienes begins at spiro C-
1, with C-2 being the adjacent carbon in the cyclohexene ring that is
trans to the isopropenyl group.

L.G. Cool / Phytochemistry 66 (2005) 249–260
255
Table 3
13C and 1H NMR spectroscopic dataa of sesquiterpenes 15 and 16, d from TMSb
C. No. 15 (C6D6)
16 (C6D6)
13C d 1H d
58.2 1.75 d (8)
TOCSYc
2
NOESYd
HMBC (C. No.) 13C d 1H d
TOCSYc NOESYd HMBC (C. No.)
1
10b,13,15
2,6/7,8?,11,
12,13,15
1?,4,7/11
1,5,14
128.7
–
–
–
–
2
3
4
5
131.1 5.55 dd (13,8) 1,3
13
26.9 2.34 br t (6)
32.1 1.95 br m
3
2
–
6
8,9
14
–
1,3,4,6,7
1,2,4,5,14?
–
132.6 6.28 br d (13)
148.0
32.2
2
–
14Z
–
–
–
133.9 –
121.3 5.39 br s
b 2.48 tm (14) 6ab
15
14E
14
a 2.30 ddd
(15,5,4)
b 1.08 m
a 2.16 ddd
(14,12,4)
–
6ab
6/7
6
35.7
5ab
5ab
15?
9a,12
1,7?,8?,5/15
5?,7,15
30.2 2.75 br s
5,14?
15
1,4,5,7
7
8
ꢁ36
–
–
15?
–
126.0
33.9 2.07 br t (8)
–
–
9
–
2,10,15
–
1,6,9,10,15
44.6
b 1.11 ddd
(13,13,4)
a 1.37 ddd
(13,3,3)
b 1.29 m
a 1.56 qt
(14,3)
9ab
9ab
15
6/7,9?,15
9
18.6
42.0
8a/10a,8b/10b
8a/10a,8b/10b 6a,12
27.0 1.53 m
8,10
9
2
7,8,10,11
10?
10
b 1.11 ddd
(13,13,4)
a 1.37 ddd
(13,3,3)
–
9ab
9ab
–
1
8,9?,12?,13?
1,8?,12,13?
38.0 1.97 br m
8,12
8,9,11,12,13
12
11
12
13
14
36.7
23.3
33.3
–
–
145.8
–
–
–
–
a 0.93 s
b 0.91 s
6a,9a,10a
1,2
3
1,10,11,13
1,10,11,12
3,5
110.3 4.82 br s
22.5 1.66 br s
23.6 1.63 br s
10,13
12
3,5
10,11,13
10,11,12
3,4,5
114.2 (Z) 4.83 br s
(E) 4.92 br s
6?
5a
3,5
15
30.7
1.01 s
1,5b,6b?,8ab? 1,6/7,8
18.5 1.61 br s
6,8
1,7,8
a
b
c
1H–1H coupling constants (Hz) in parentheses; query (?) = weak or indistinct signal; slash (/) = and/or.
Solvent references, d from TMS: CDCl3 1H 7.26, 13C 77.0; C6D6 1H 7.15, 13C 128.0.
24 ms mixing time used to limit correlations to 2 and 3 bonds; two-bond correlations not listed.
Two- and three-bond correlations not listed.
d
and vinyl H-2 (d 5.55, dd) was coupled to H-3 (d 6.30,
d) and H-1. Since H-1 was coupled only to H-2, the
other two carbon atoms adjacent to C-1 must be qua-
ternary. HMQC accounted for 12 protonated carbons,
so the three methyl singlets were obviously attached to
these two quaternary carbons as a geminal pair and
an angular methyl at the other bridgehead. The widd-
rane skeleton suggested by these data was confirmed
by an HMBC experiment, which showed key correla-
tions from H-12 and -13 to C-1, -10, and -11; H-6 to
C-15; and H-15 to C-1, -6/7, and -8. A cis ring fusion
was found to be the desired cis-fused stereoisomer
(see Section 3). This was then dehydrated to
two monoenes, which mixture was in turn hydroge-
nated to produce the cis-widdrane epimers. These
had retention times identical to those from 15 on
dual-column capillary GC but different on EGC,
which established the 7R absolute stereochemistry of
15.
1D NMR of compound 16 showed it to be an achi-
ral monocyclic sesquiterpene with three double bonds
– one tetrasubstituted, one trisubstituted, and one
disubstituted terminal methylene – and three vinyl
methyl singlets. HMQC and HMBC experiments (Ta-
ble 3) allowed all signals to be unambiguously as-
signed, proving the iso-c-bisabolene structure. The
1E geometry was determined by a 1H–1H NOESY
experiment, which showed a cross peak from H-15
to bis-allylic H-6. Though it is a new natural product,
this compound has been synthesized in a 60:40 mix-
ture with the 1Z isomer, and the reported NMR data
(in CDCl3) and MS data are very close to that found
for 16 (Braun et al., 2003).
1
was evident from a H–1H NOESY experiment, which
gave a clear cross-peak from H-15 to H-1. The abso-
lute stereochemistry was determined by EGC compar-
ison of the two epimeric saturated hydrogenation
products of 15 with those derived from authentic
(+)-widdrol (7S absolute stereochemistry). Generation
of reference (7S)-widdranes required
a somewhat
roundabout procedure, since dehydration of widdrol
leads to the methyl migration product pseudowiddrene
rather than widdradienes (Ito et al., 1974). (+)-Wid-
drol was first hydrogenated to dihydrowiddrol, which

256
L.G. Cool / Phytochemistry 66 (2005) 249–260
Scheme 1. Proposed mechanism for biosynthesis of novel C. macrocarpa sesquiterpenes via an (S)-bisabolyl carbocation.
Production of very complex mixtures of constitutive
monoterpenoid pheromones in Anthonomus grandis.)
Deprotonation of the resulting macrocarpene carbo-
cation at C-6, C-8, or at the newly incorporated meth-
ylene (C-8 in macrocarpene numbering) gives 3–6,
while a C-6, C-7 hydride shift followed by aromatiza-
tion accounts for ar-macrocarpene 7.
Path (b), involving C-7, C-10 cyclization of the origi-
nal (S)-bisabolyl carbocation with proton loss from C-
12, leads to 8. Alternatively, path (c), C-7, C-11 cycliza-
tion with a well-attested C-6, C-10 hydride shift (Cane,
1981, and references therein; Nabeta et al., 1993), gener-
ates an ent-cuparenyl carbocation. Deprotonation of
this cation accounts for the rare (7S)-cuparenes ob-
served, while C-7, C-11 bond scission with deprotona-
tion at C-12 [path (c1)] is a plausible mechanism
leading to bisabolene 16, with its unusual C-12
unsaturation.
An alternative reaction of the ent-cuparenyl carbocat-
ion is C-7, C-6 skeletal rearrangement (c2), which gener-
ates an ent-chamigrenyl carbocation. This opens the way
to 11 via path (c3), C-2, C-7 closure with C-4 deprotona-
tion; and to 15 by path (c4), C-6, C-7 skeletal rearrange-
ment and a C-1, C-6 hydride and double bond shift, with
deprotonation at the C-3 methyl.
sesquiterpenes by a single enzyme catalyst has been
demonstrated in Abies grandis stem tissue (Steele
et al., 1998). It is proposed that biosynthesis of the
novel sesquiterpenes in C. macrocarpa foliage
likewise involves a single ‘‘macrocarpene synthase’’
and can be rationalized as in Scheme 1. The first step
(not shown) is cyclase-catalyzed isomerization of
(E,E)-farnesyl diphosphate to (3S)-nerolidyl diphos-
phate (NPP) in the preferred anti-endo conformation
2
(Cane, 1985, 1999). Ionization and C-1, C-6 closure
then generates an (S)-bisabolyl carbocation that is
the proposed precursor of all product sesquiterpenes.
Macrocarpenes 3–7 can be considered to arise via
path (a), the first step of which is deprotonation of
the bisabolyl carbocation to (S)-b-bisabolene. This
neutral intermediate is not released from the enzyme
active site but is immediately reprotonated at C-10
with ring closure from the C-7 methylene to C-11.
(A similar mechanism has been adduced by Erman
(1985) in the biogenesis of the dimethylcyclohexylidene
2
NPP numbering is used throughout this discussion.

L.G. Cool / Phytochemistry 66 (2005) 249–260
257
Finally, biosynthesis of acoradiene 12 [route (d)] re-
quires a 1,2 hydride shift from C-6 to C-7 of the bisab-
olyl carbocation (Nabeta et al., 1990) followed by C-6,
C10 cyclization and proton loss from C-12.
Me]+ (31), 161 (100), 147 (20), 133 (48), 119 (61), 105
(87), 91 (92), 81 (55), 67 (38), 55 (37), 41 (59).
For isolation of the remaining sesquiterpenes, ca.
4000 g of liq. N2-ground foliage from the chosen hor-
ticultural tree was hydrodistilled in 800 g batches,
yielding ca. 20 g oil. Additional oil was obtained by
9· batchwise hexane extraction of ꢀ20 mesh ground
air-dried foliage (1700 g), extract solvent removal,
and subsequent hydrodistillation of the dark green res-
idue (89 g) with 400 ml satd NaCl and 10 g NaHCO3,
yielding another 20 g of oil. The combined oils (40 g)
consisted of ca. 86% monoterpenoids, 9% sesquiterpe-
noids, and 5% diterpenoids. Before silica gel LC, most
of the monoterpenes were removed by evaporation un-
der N2 flow at ca. 120 ꢁC, giving 13 g of monoter-
pene-depleted oil.
3. Experimental
3.1. General experimental procedures
Survey analyses of C. macrocarpa trees were done by
grinding ca. 2 g of liq. N2-frozen foliage from each tree,
extracting overnight with n-pentane, and analyzing the
extracts by GC–MS (SE-54 WCOT column) and dual-
column GC-FID (SE-54 and OV-17 WCOT columns)
under these conditions: splitless (0.7 min) injection,
injector 220 ꢁC, FID detector 250 ꢁC, column(s) 35 ꢁC
(0.7 min), 6 ꢁC minꢀ1 program to 250 ꢁC. The same
GC conditions were used for retention time determina-
tions of purified sesquiterpenes.
Hydrodistillation of bulk samples of liq. N2-ground
foliage was from satd NaCl with NaHCO3 as previously
described (Kim et al., 1994).
LC of hydrodistilled oil (silica gel; hexane–EtOAc
eluents) and FTIR of purified compounds were de-
scribed before (Kim et al., 1994), as was EGC on a
10% permethyl-b-cyclodextrin column (Cool, 2001).
The first LC fraction, eluted with n-hexane, consisted
solely of hydrocarbons. This fraction (8 g) was separated
by repeated complexation LC of 1.5 g portions on 50 g
of 20% AgNO3/silica gel using hexane–toluene step gra-
dients, giving 11 fractions. Minor fractions 1–3, 8 and 10
(each <0.1 g, complex mixtures of trace compounds)
and major fraction 5 (1.6 g, mostly 3), were not further
analyzed. Part of major fraction 6 (3 g), consisting
mainly of 3 and 4, was subjected to prep. GC (see below)
to give analytical samples (ca. 20 mg each) of 3 and 4.
Fraction 4 (0.6 g), consisting of ca. 100:3:1:1 3, 11, 7
and 15 was re-chromatographed with 1% toluene/hex-
ane to give a sub-fraction (ca. 0.1 g) enriched in 7, 11
and 15. In the case of 5–8, 11, 12 and 15, further sepa-
ration was by isocratic HPLC (hexane–toluene eluents;
RI detection) of portions of fractions rich in these com-
pounds on a 10 · 250 mm column packed with 15%
AgNO3/silica gel (5 l), using the following toluene con-
centrations: 1.8% for fraction 4 sub-fraction (7, 11 and
15), 3.5% for fraction 7 (1.3 g) (5 and 6), 7% for fraction
9 (0.2 g) (8 and 12). Final purification of all except 16
was by prep. GC on 4 mm i.d. packed columns (OV-
17 column, 170 ꢁC for 3, 4, 8 and 12; SE-30 column,
185 ꢁC for 7, 11 and 15; Carbowax 20M column, 165
ꢁC for 5 and 6). Compounds 5 and 6 were very prone
to rearrangement and oxidation. Therefore, they were
collected in capped 3 mm OD glass U-tubes which were
treated with a solution of 100 lg each of antioxidant (Io-
nox 330, Ethyl Corp.) and pyridine-d5 in 5 ll of pentane,
applied in the condensation zone of the U-tubes just be-
fore use. b-Sesquiphellandrene and 16, in argentation
LC fraction 11 (0.9 g; predominantly diterpenes), were
inseparable by GC or AgNO3/silica gel HPLC, but nor-
mal-phase silica gel HPLC (5 l particle size; 10 · 250
mm column; RI detection) with n-hexane eluent par-
tially resolved them, yielding >97% pure samples of
3.2. Plant material
Small survey samples of foliage were taken from 20
widely-separated trees in the Point Lobos population
of C. macrocarpa in March 1996. Planted specimens
(20 trees) were similarly sampled from the University
of California Richmond Field Station in July 2001,
and the horticultural tree with the highest yield of mac-
rocarpenes was used for compound isolation.
3.3. Sesquiterpene isolation and identification
The n-pentane extract (ca. 2 ml) of foliage from a tree
containing germacrene C and the guaiadienes was con-
centrated 10-fold under N2. This was applied to a col-
umn of silica gel in a Pasteur pipette and the
hydrocarbon fraction eluted with 2 column volumes of
n-hexane. (+)-Guaia-6,9-diene 1 was identified by GC–
MS, dual-column GC-FID, and EGC comparison of
the hydrocarbon fraction with authentic (ꢀ)-guaia-6,9-
diene from commercial rose geranium (Pelargonium
graveolens) oil. The MS of the standard and the C. mac-
rocarpa component were identical, as were the GC reten-
tion times under all conditions except on EGC.
Guaia-6,10(15)-diene 2 was tentatively identified by
the similarity of its MS with that of the compound from
N. chabrolii (Venkateswarlu, personal communication).
GC–MS 70 eV, m/z (rel. int.): 204 [M]+ (57), 189 [M–
16 (12 mg) and b-sesquiphellandrene (9 mg;
22
D
½aꢂ ꢃ ꢀ15ꢄ, c5; CDCl3).
Chirality of (ꢀ)-cuparene and (ꢀ)-a-cuprenene was
determined by EGC comparison of prep. GC or HPLC

258
L.G. Cool / Phytochemistry 66 (2005) 249–260
fractions rich in the appropriate compound with cedar-
wood (Juniperus sp.) oil hydrocarbon fractions enriched
in (+) enantiomers of these compounds.
H-3), 1.99 (2H, s, H-8), 2.10 (2H, br t, J = 6.1 Hz,
H-12), 2.33 (2H, t, J = 6.3 Hz, H-2), 2.80 (2H, br s, H-
6), 5.37 (1H, br m, H-5); 13C NMR (100.6 MHz, CDCl3,
solv. ref. 77.0, d from TMS): 23.4 (q, C-15), 23.9 (t, C-
11), 26.3 (t, C-2), 28.5 (2C, q, C-13, -14), 29.2 (t, C-3),
30.1 (t, C-12), 31.9 (t, C-6), 33.2 (s, C-9), 40.1 (t, C-
10), 43.2 (t, C-8), 121.0 (d, C-5), 126.2 (s, C-1), 128.6
(s, C-7), 134.2 (s, C-4); multiplicities by DEPT.
Hydrogenation experiments on 4, 6, and 7 were done
under 70 atm H2 with ca. 2 mg sesquiterpene and cata-
lytic amounts of PtO2 in 200 ll hexane. The same two
Mr = 208 products, inseparable on a 5% phenyl meth-
ylpolysiloxane WCOT column but separable on 50%
phenyl methylpolysiloxane, were produced from each
starting material, in the following ratios (earlier-eluting
to later): 43:57 for 4, 32:68 for 6, and 20:80 for 7.
All IUPAC names were obtained using the ACD/I-
Lab Web service (ACD/IUPAC Name Free 7.06).
3.8. (+)-Ar-macrocarpene, 1-[(1R)-3,3-
dimethylcyclohexyl]-4-methylbenzene (7)
22
D
Oil. ½aꢂ +7.2ꢁ (c 5.1; n-hexane); HREIMS: m/z
found 202.1728, Calc. for C15H22 202.1722; GC–MS
70 eV, m/z (% rel. int.): 202 [M]+ (93), 187 (11), 159
(53), 131 (49), 118 (39), 117 (23), 105 (100), 91 (19), 57
(14), 41 (14). 1H NMR (400.13 MHz, C6D6, solvent
ref. 7.15, d from TMS): 0.91 (3H, s), 0.93 (3H, s), 1.09
(1H, ddd, J = 4.5, 13.1, 13.1 Hz), 1.23 (1H, dddd,
J = 4.5, 12.6, 12.6, 12.6 Hz), 1.26 (1H, dd, J = 12.8,
12.8 Hz), 1.33 (1H, br d, J = 12.9 Hz), 1.46 (1H, ddddd,
J = 3.5, 3.5, 13.2, 13.2, 13.2 Hz), 1.49–1.58 (2H, m), 1.83
(1H, br d, J = 12.8 Hz), 2.18 (3H, s), 2.62 (1H, dddd,
J = 3.4, 3.4, 12.7, 12.7 Hz), 7.06 (4H, s); 13C NMR
(100.6 MHz, C6D6, solvent ref. 128.0, d from TMS):
21.0 (q), 23.1 (t), 24.7 (q), 31.3 (s), 33.6 (q), 34.6 (t),
39.2 (t), 39.9 (d), 48.0 (t), 127.2 (2C, d), 129.3 (2C, d),
135.2 (s), 145.0 (s); multiplicities by DEPT.
3.4. (ꢀ)-a-Macrocarpene, (10S)-3,3,40-trimethyl-1,
10-bi(cyclohexane)-1,30-diene (3)
22
D
KBr
max
Oil. ½aꢂ ꢀ81ꢁ (c 3.9; n-hexane); IR m
cmꢀ1: 2930,
1453, 1377, 1358, 1207, 1147, 1035, 939, 914, 858, 789;
HREIMS: m/z found 204.1883, Calc. for C15H24
204.1878; GC–MS 70 eV, m/z (rel. int.): 204 [M]+ (16),
189 (19), 176 (4), 162 (3), 148 (6), 136 (64), 121 (100),
107 (26), 95 (86), 93 (53), 92 (33), 79 (27), 67 (18), 55
(20), 41 (28).
3.5. (ꢀ)-b-Macrocarpene, (10S)-40,5,5-trimethyl-1,
10-bi(cyclohexane)-1,30-diene (4)
22
D
KBr
max
Oil. ½aꢂ ꢀ72ꢁ (c 2.2; n-hexane); IR m
cmꢀ1: 2911,
1471, 1363, 1217, 1146, 948, 914, 833, 796; HREIMS: m/
z found 204.1883, Calc. for C15H24 204.1878; GC–MS
70 eV, m/z (rel. int.): 204 [M]+ (33), 189 (22), 175 (7),
162 (6), 148 (13), 136 (53), 121 (56), 107 (53), 105 (18),
95 (20), 94 (33), 93 (100), 92 (40), 91 (23), 80 (41), 79
(50), 67 (17), 55 (19), 41 (38).
3.9. (ꢀ)-Cumacrene, (4S)-4-[(1R,2S)-2-isopropenyl-
1-methylcyclobutyl]-1-methylcyclohexene (8)
22
D
Oil. ½aꢂ ꢀ77ꢄ (c 1.7; n-hexane); HREIMS: m/z found
204.1877, Calc. for C15H24 204.1878; GC–MS 70 eV, m/
z (% rel. int.): 204 [M]+ (2), 189 (3), 175 (17), 161 (9), 147
(4), 136 (8), 121 (49), 109 (23), 107 (38), 105 (14), 94 (43),
93 (100), 79 (31), 68 (84), 67 (41), 55 (17), 53 (21), 41
(32).
3.6. (Z)-c-Macrocarpene, (1Z)-30,30,4-trimethyl-1,
10-bi(cyclohexane)-1(10),3-diene (5)
Oil. HREIMS: m/z found 204.1883, Calc. for C15H24
204.1878; GC–MS 70 eV, m/z (rel. int.): 204 [M]+ (59),
189 (10), 161 (19), 135 (25), 133 (18), 121 (20), 119
(26), 107 (36), 105 (43), 95 (27), 94 (100), 93 (79), 92
(28), 91 (50), 81 (19), 79 (57), 69 (31), 55 (31), 41 (53).
3.10. (ꢀ)-a-Chamipinene, (1S,6S,7S)-2,2,6,8-
tetramethyltricyclo[5.3.1.01,6]undec-8-ene (11)
22
D
Oil. ½aꢂ ꢀ4:3ꢄ (c 4.4, n-hexane); HREIMS: m/z
found 204.1877, Calc. for C15H24 204.1878; GC–MS
70 eV, m/z (rel. int.): 204 [M]+ (6), 189 (2), 161 (2),
147 (3), 133 (22), 123 (17), 121 (16), 119 (100), 111
(36), 105 (33), 94 (27), 93 (40), 92 (30), 91 (25), 81
3.7. (E)-c-Macrocarpene, (1E)-30,30,4-trimethyl-1,
10-bi(cyclohexane)-1(10),3-diene (6)
1
(13), 69 (44), 55 (27), 41 (28); H NMR (400.13 MHz,
Oil. HREIMS: m/z found 204.1883, Calc. for C15H24
204.1878; GC–MS 70 eV, m/z (rel. int.): 204 [M]+ (59),
189 (10), 161 (13), 135 (26), 133 (22), 121 (30), 119
(35), 107 (38), 105 (43), 95 (30), 94 (100), 93 (86), 92
(30), 91 (51), 81 (19), 79 (58), 69 (29), 55 (35), 41 (51);
1H NMR (400.13 MHz, CDCl3, solvent ref. 7.26, d from
TMS): 0.86 (6H, s, H-13,-14), 1.34 (2H, m, H-10), 1.48
(2H, m, H-11), 1.66 (3H, br s, H-15), 1.98 (2H, br t,
C6D6, solvent ref. 7.15, d from TMS): 0.69 (3H, s),
0.79 (3H, s), 0.89 (3H, s), 1.16 (1H, d, J = 8.4 Hz),
1.18 (1H, dm, J ꢃ 12 Hz), 1.67 (3H, m), 1.74 (1H, dd,
J = 5.7, 1.6 Hz), 1.98 (1H, dm, J = 17.2 Hz), 2.07 (1H,
dm, J = 17.0 Hz), 2.24 (1H, dd, J = 8.3, 5.7 Hz), 5.29
(1H, br m, J = 1.6 Hz); 13C NMR (100.6 MHz, C6D6,
solvent ref. 128.0, d from TMS): 18.2 (t), 18.7 (q), 21.9
(q), 22.5 (q), 24.0 (q), 30.8 (t), 31.4 (s), 31.5 (t), 34.8

L.G. Cool / Phytochemistry 66 (2005) 249–260
259
(t), 38.1 (t), 38.2 (s), 44.4 (d), 47.2 (s), 117.0 (d), 142.0 (s);
multiplicities by DEPT.
(100), 94 (80), 93 (78), 91 (95), 79 (63), 69 (58), 55
(42), 41 (72).
Acid-catalyzed rearrangement of (ꢀ)-a-chamipinene –
3 mg 11 and 1 mg Cu(OAc)2 Æ H2O were dissolved in
100 ll of HOAc in a 1 ml vial, which was briefly flushed
with argon then capped. After 18 h in a 110ꢁ oven, the
vial was cooled and the contents diluted 20· with water
and shaken 3· with 1 ml pentane. The combined pen-
tane extracts were washed with dil. aq NaHCO3, dried
over Na2SO4, and analyzed by GC (2 columns), GC–
MS, and EGC. Major products were (+)-a-chamigrene
(44%) and (+)-b-chamigrene (38%), identical by GC
and GC–MS (but differing in r.t. on EGC) with (ꢀ)-a-
and (ꢀ)-b-chamigrene produced by acid-catalyzed rear-
rangement of authentic (ꢀ)-thujopsene under the same
conditions (Daeniker et al., 1972).
Hydrogenation of (+)-widdrol and 15; determination
of absolute stereochemistry: 30 mg authentic (+)-widdrol
(7S absolute stereochemistry), dissolved in 0.4 ml AcOH
and with a catalytic amount of PtO2, was hydrogenated
for 18 h by bubbling in H2 at atmospheric pressure (Enz-
ell, 1962). Contaminating hydrocarbons were removed
from the single oxygenated product by passing the mix-
ture in hexane through a small silica gel column, then
eluting the cis-dihydrowiddrol [(4aS,7S,9aS)-1,1,4a,7-
tetramethyldecahydro-1H-benzo[7]annulen-7-ol] with
25% EtOAc/hexane (yield ca. 20 mg). NMR: 1H
(500.13 MHz, CDCl3, solv. ref. 7.26, d from TMS):
0.89 (3H, s, H-13), 1.00 (3H, s, H-12), 1.07 (1H, d,
J = 10 Hz, H-1), 1.10 (3H, s, H-15), 1.12 (1H, m, H-
2b), 1.13 (1H, m, H-8b), 1.17 (1H, m, H-10b), 1.18
(1H, m, H-6a), 1.24 (3H, s, H-14), 1.35 (2H, m, H-8a,
H-10a), 1.36 (1H, m, H-6b), 1.41 (1H, m, H-3a), 1.46
(1H, m, H-9b), 1.51 (1H, m, H-5b), 1.53 (1H, m, H-
2a), 1.58 (1H, qt, J = 6.1, 1.8 Hz, H-9a), 1.76 (1H, dd,
J = 13.5, 11.7 Hz, H-5a), 1.87 (1H, ddd, J = 13.1, 7.7,
2.4 Hz, H-3b); 13C NMR (125.77 MHz, CDCl3, solv.
ref. 77.0, d from TMS): 17.9 (t, C-9), 22.2 (t, C-2),
27.8 (s, C-14), 29.0 (s, C-13), 30.3 (s, C-15), 32.1 (s, C-
12), 34.4 (s, C-7), 35.1 (t, C-8), 35.6 (t, C-10), 36.4 (s,
C-11), 37.1 (t, C-5), 39.8 (t, C-6), 46.0 (t, C-3), 55.3 (d,
C-1), 73.2 (s, C-4); assignments by HMQC and HMBC,
cis ring fusion confirmed by NOESY cross-peaks from
H-15 to H-12 (both a-oriented) and from H-12 to H-1,
in reference to minimum-energy conformation deter-
mined by MM2 calculation. 10 mg of cis-dihydrowid-
drol was dehydrated for 1 h with 8 drops POCl3 in 0.3
ml pyridine, excess POCl3 destroyed by adding ice,
and the mixture of monoenes (2 isomers, ca. 4:1 ratio)
extracted with pentane. About 2 mg of the product
hydrocarbon mix was directly hydrogenated with a cat-
alytic amount of PtO2 in 0.4 ml MeOH for 2 h, yielding
2 fully saturated epimers in ca. 2:1 ratio, with some of
the major monoene remaining unreacted. 1 mg 15 was
similarly hydrogenated, yielding 2 fully saturated epi-
mers in ca. 1:1 ratio. Retention times of the widdrol-de-
rived widdranes were identical with those from 15 on
conventional dual-column GC but longer on EGC.
3.11. (ꢀ)-10-epi-b-Acoradiene, (1S,4S,5R)-1-
isopropenyl-4,8-dimethylspiro[4.5]dec-7-ene (12)
22
D
Oil. ½aꢂ ꢀ4:1ꢄ (c 1.8, n-hexane); HREIMS: m/z
found 204.1877, Calc. for C15H24 204.1878; GC–MS
70 eV, m/z (rel. int.): 204 [M]+ (7), 189 (5), 175 (2),
161 (16), 148 (10), 147 (21), 134 (8), 133 (12), 121 (52),
119 (100), 108 (21), 107 (23), 105 (40), 93 (62), 91 (27),
1
79 (30), 68 (16), 67 (17), 55 (18), 53 (12), 41 (28); H
NMR (500.17 MHz, C6D6, solvent ref. 7.15, d from
TMS): 0.93 (3H, d, J = 6.7 Hz, H-15), 1.29 (1H, m, H-
9a), 1.44 (2H, dd, J = 6.8, 6.0 Hz, H-6), 1.62 (3H, br s,
H-14), 1.64 (1H, m, H-8b), 1.68 (1H, m, H-10), 1.69
(1H, m, H-9b), 1.74 (3H, s, H-13), 1.78 (1H, m, H-8a),
1.88 (2H, br dm, J ꢃ 17 Hz, H-2eq, H-5ax), 2.00 (2H,
br dm, J ꢃ 17 Hz, H-2ax, H-5eq), 2.14 (1H, dd,
J = 9.8, 8.3 Hz, H-7), 4.80 (1H, s, (Z)-H-12), 4.90 (1H,
s, (E)-H-12), 5.38 (1H, br s, H-5); 13C NMR (125.8
MHz, C6D6, approx. d from TMS, protonated carbons,
indirect detection; multiplicities by HMQC): 15.8 (q, C-
15), 23.3 (q, C-14), 24.0 (q, C-13), 24.0 (t, C-2), 27.8 (t,
C-9), 28.9 (t, C-3), 31.5 (t, C-9), 39.1 (t, C-6), 44.1 (d,
C-10), 58.1 (d, C-7), 112.2 (t, C-12), 121.2 (d, C-5).
Hydrogenation of 12 and (ꢀ)-a-acoradiene 13 – ca. 1
mg of sesquiterpene in ca. 300 ll hexane was hydroge-
nated over Pd/C catalyst at 70 atm H2 pressure for (a) 1
h (13), giving Mr 208 products 14a–d in ratios 23:69:2:6;
(b) 12 h (12), giving Mr 208 products ent-14a–d in ratios
8:43:6:43; (c) 2 h (12 with a trace of pyridine), giving Mr
208 products ent-14a–d in ratios 1:9:19:71.
Acknowledgements
3.12. (ꢀ)-Widdra-2,4(14)-diene, (4aR,9aR)-1,1,4a-
trimethyl-7-methylene-2,3,4,4a,5,6,7,9a-octahydro-1
H-benzo[7]annulene (15)
I thank Robert Flath for a sample of authentic (ꢀ)-
guaia-6,9-diene; Y. Venkateswarlu for MS data on
guaia-6,10(15)-diene; Robert P. Adams for GC–MS
data of a-cuprenene; Eugene Zavarin for encourage-
ment and helpful discussions; and Rudi Nunlist and
Herman van Halbeek for their capable help with many
of the 2D NMR experiments. I am especially grateful
22
Oil. ½aꢂ ꢃ ꢀ18ꢄ (c 0.27, benzene-d6); UV
D
hexane
max
k
nm ðlog eÞ: 233 (3.8); GC–MS 70 eV, m/z (rel.
int.): 204 [M]+ (18), 189 (39), 161 (49), 148 (23), 133
(47), 123 (37), 121 (54), 119 (61), 111 (78), 105

260
L.G. Cool / Phytochemistry 66 (2005) 249–260
Kim, Y.K., Cool, L.G., Zavarin, E., 1994. cis-Calamenene-related
sesquiterpenoids from Cupressus bakeri foliage. Phytochemistry 36,
961–965.
to David E. Cane for generously offering guidance in
developing the biosynthetic scheme. Deo gratias.
Marx, J.N., Norman, L.R., 1975. Synthesis of (ꢀ)-acorone and related
spirocyclic sesquiterpenes. J. Org. Chem. 40, 1602–1606.
Mills, J.A., 1952. Correlations between monocyclic and polycyclic
unsaturated compounds from molecular rotation differences. J.
Chem. Soc., Abstracts, 4976–4985.
References
Ansari, H.R., 1973. Cyclisation of optically active dihydromyrcenes
(2,6-dimethyl-2,7-octadiene). Tetrahedron 29, 1559–1564.
Bernard, M., Canuel, C., St-Jacques, M., 1974. A comparative study of
the rates and mechanisms of conformational interconversion in
derivatives of cyclohexanone and methylenecyclohexane. JACS 96,
2929–2936.
Nabeta, K., Ara, Y., Aoki, Y., Miyake, M., 1990. Biosynthesis of
monoterpenes and sesquiterpenes in Larix leptolepis callus from
deuterated mevalonates. J. Nat. Prod. 53, 1241–1248.
Nabeta, K., Kawakita, K., Yada, Y., Okuyama, H., 1993. Biosynthesis
of sesquiterpenes from deuterated mevalonates in Perilla callus.
Biosci. Biotech. Biochem. 57, 792–798.
Braun, N.A., Meier, M., Schmaus, G., Ho¨lscher, B., Pickenhagen, W.,
2003. Enantioselectivity in odor perception: synthesis and olfactory
properties of iso-b-bisabolol, a new natural product. Helv. Chim.
Acta 86, 2698–2708.
Naegeli, P., Wetli, M., 1981. Cationic p-cyclisation of a,b-unsaturated
carbonyl compounds. Tetrahedron 37 (Suppl. 1), 247–255.
Rao, M. Rama, Sridevi, K.V., Venkatesham, U., Rao, T. Prabhaker,
Lee, S.S., Venkateswarlu, Y., 2000. Four new sesquiterpenoids
from the soft coral Nephthea chabrolii. J. Chem. Res., Synopses,
245–247.
Briggs, L.H., Sutherland, M.D., 1942. The essential oil of Cupressus
macrocarpa. J. Org. Chem. 7, 397–407.
Cane, D.E., 1981. Biosynthesis of sesquiterpenes. In: Porter, J.W.,
Spurgeon, S.L. (Eds.), Biosynthesis of Isoprenoid Compounds.
Wiley, New York, pp. 283–374.
Snowdon, R.L., Linder, S.M., Muller, B.L., Schulte-Elte, K.H., 1987.
b-Cleavage of bis(homoallylic) potassium alkoxides. Two-step
preparation of propenyl ketones from carboxylic esters. Synthesis
of ar-turmerone, a-damascone, b-damascone, and b-damascenone.
Helv. Chim. Acta 70, 1858–1878.
Cane, D.E., 1985. Isoprenoid biosynthesis. Stereochemistry of the
cyclization of allylic pyrophosphates. Acc. Chem. Res. 18, 220–226.
Cane, D.E., 1999. Sesquiterpene biosynthesis: cyclization mechanisms.
In: Barton, D., Nakanishi, K., Meth-Cohn, O. (Eds.), Compre-
hensive Natural Products Chemistry, vol. 2. Elsevier, Amsterdam,
pp. 155–200.
Steele, C.L., Crock, J., Bohlmann, J., Croteau, R., 1998. Sesquiterpene
synthases from grand fir (Abies grandis). J. Biol. Chem. 273, 2078–
2089.
Sy, L.-K., Brown, G.D., 1998. A sesquiterpene class from Illicium
dunnianum. Phytochemistry 47, 301–302.
Cool, L.G., 2001. ent-Daucane and acorane sesquiterpenes
from · Cupressocyparis leylandii foliage. Phytochemistry 58, 969–
972.
Tebayashi, S., Hirai, N., Suzuki, T., Matsuyama, S., Nakakita, H.,
Nemoto, T., Nakanishi, H., 1998. Identification of (+)-acoradiene
as an aggregation pheromone for Gnatocerus cornutus (F.) (Col-
optera: Tenebrionidae). J. Stored Prod. Res. 34, 99–106.
Tumlinson, J.H., Gueldner, R.C., Hardee, D.D., Thompson, A.C.,
Hedin, P.A., Minyard, J.P., 1971. Identification and synthesis of
the four compounds comprising the boll weevil sex attractant. J.
Org. Chem. 36, 2616–2621.
Daeniker, H.U., Hochstetler, A.R., Kaiser, K., Kitchens, G.C., 1972.
The acid-catalyzed isomerization of thujopsene. J. Org. Chem. 37,
1–5.
Enzell, C., 1962. Chemistry of the natural order Cupressales. XLVII.
Structure and absolute stereochemistry of widdrol and widdrol a-
epoxide. Acta Chem. Scand. 16, 1553–1568.
Erman, W.F., 1985. Chemistry of the Monoterpenes: an Encyclopedic
Handbook. Marcel Dekker, New York.
Wolf, C.B., 1948. Taxonomic and distributional studies of the New
World cypresses: Cupressus macrocarpa. El Aliso 1, 181–193.
Zavarin, E., Lawrence, L., Thomas, M.C., 1971. Compositional
variations of leaf monoterpenes in Cupressus macrocarpa, C.
pygmaea, C. goveniana, C. abramsiana and C. sargentii. Phyto-
chemistry 10, 379–393.
Ito, S., Endo, K., Narita, H., 1974. a-Pseudowiddrene,
a new
sesquiterpene hydrocarbon from Thujopsis dolobrata. Tetrahedron
Lett., 1041–1043.
Kaiser, R., Naegeli, P., 1972. Biogenetically significant components in
vetiver oil. Tetrahedron Lett., 2009–2012.