Synthesis of sesquiterpene allylic alcohols and sesquiterpene dienes from Cupressus bakeri and Chamaecyparis obtusa

Article abstract of DOI:10.1039/a907752i

Five muurolane (1-5), one nor-seco-muurolane (6) and one nor-muurolane (7) sesquiterpene natural products recently reported from Cupressus bakeri have been obtained by chemical synthesis together with the unnatural 7-epimeric amorphane analogues of four o

Full text of DOI:10.1039/a907752i

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis of sesquiterpene allylic alcohols and sesquiterpene dienes
from Cupressus bakeri and Chamaecyparis obtusa
Koon-Sin Ngo and Geoffrey D. Brown*
1
Chemistry Department, The University of Hong Kong, Pokfulam Rd., Hong Kong
Received (in Cambridge, UK) 27th September 1999, Accepted 1st November 1999
Five muurolane (1–5), one nor-seco-muurolane (6) and one nor-muurolane (7) sesquiterpene natural products
recently reported from Cupressus bakeri have been obtained by chemical synthesis together with the unnatural
7-epimeric amorphane analogues of four of these sesquiterpenes (15, 16, 18 and 19). The conversion of sesquiterpene
allylic tertiary alcohols 1, 2, 15 and 16 into regioisomeric sesquiterpene dienes 3–5 and 17–20 was investigated in vitro
by NMR and the mechanism for such dehydrations and their relevance to the origins of sesquiterpene dienes which
have been reported as natural products (such as 3–5) is discussed.
Introduction
In 1994, Cool and co-workers1 identified the novel sesquiter-
pene alcohols 1 and 2 from the foliage of Cupressus bakeri
together with sesquiterpene dienes 3 and 4, which were sug-
gested to be dehydration products of these alcohols, the known
compound epizonarene 5 and two unusual nor-sesquiterpenes
6 and 7 (Fig. 1). Fully assigned 13C NMR data and partial H
1
NMR data (in CDCl3 solution) were provided to support the
structures of all these natural products with the exception of 4,
which was present in too small amounts to be adequately
characterised and for which the proposed structure was only
tentative. Two years later, Nagahama and co-workers2 also
reported compounds 1 and 2 (referred to as β-hinokienol and
α-hinokienol respectively) as constituents of Chamaecyparis
obtusa leaf oil. Although fully assigned 13C NMR data and
partial 1H NMR data were again provided to support the struc-
tures of both diastereoisomers, confirmation of the identity of
Fig. 1 Sesquiterpenes reported from C. bakeri.
spectroscopy. It is clear from the data in Table 1 that there is a
strong upfield shift in the 13C resonance of the 14-methyl group
in isomers 10 and 11 (∆δC 5–7 ppm), for which the newly intro-
duced 1-alkyl ring-substituent is cis with respect to the 14-
methyl group, as compared to diastereoisomers 6 and 9; the
sizeable upfield shift is the result of a significant gauche
interaction between these two substituents. The resonances for
protons in the 14-methyl group were also shifted upfield in
compounds 10 and 11 (∆δH 0.2–0.3 ppm) when compared to 6
and 9, which confirmed the preferred axial conformation of
this substituent for these two diastereoisomers (the H-1 proton
should therefore be equatorial and indeed was found to be sig-
nificantly downfield, by approximately 0.4 ppm, in both cases).
Treatment of 9 with barium hydroxide under mild conditions
produced the aldol addition product 12 (the cis-decalin isomer
13 was also isolated in very small amounts) which could be con-
verted to the nor-amorphane decalenone 14 in moderately acidic
conditions at low temperature. More forcing conditions induced
epimerisation at the 7-position resulting in the nor-muurolane
decalenone 7 (Scheme 1). Synthetic compound 7 gave identical
13C NMR spectra to the natural product reported from C.
bakeri1 and is expected to be the thermodynamic product of
such Robinson annulation, since the β-isopropyl group can
adopt an axial conformation, thereby avoiding A1,3 strain with
the 5-alkene proton (this unfavourable interaction outweighs the
natural tendency for ring substituents to be equatorial).3 When
comparing the fully assigned NMR spectra of these two deca-
lenone products, it can be seen that there is a significant upfield
shift (δC 4–5 ppm) in the resonances for C-1 and C-9 and a
correspondingly large downfield shift at C-5 for nor-muurolane
7 relative to nor-amorphane 14 (Table 1). This is entirely consis-
these natural products by comparison of the NMR data with
those in the preceding report is difficult because 13C NMR
spectra were acquired in C6D6 and no reference was made to the
earlier publication. More worryingly, some of the 13C assign-
ments made by the latter authors differed significantly from
those in the earlier publication and several are well outside the
range of differences expected for solvent-induced shifts (between
5–10 ppm in the case of C-2, C-3 and C-8 for both compounds).
Results and discussion
In order to resolve the question of the identity of recently iso-
lated natural products 1 and 2 from C. bakeri and C. obtusa, we
set out to synthesise these compounds from the commercially
available menthane monoterpene (ϩ)-menthone (8) and also to
study their dehydration with a view to confirming the tentative
structure proposed for natural product 4. This synthetic strat-
egy also provided a route to nor-sesquiterpene natural products
6 and 7, which arise as synthetic intermediates. All compounds
reported herein were characterised by 2D-NMR (in both
CDCl3 and C6D6 in the case of compounds 1, 2, 15 and 16),
enabling independent verification of stereochemistry and struc-
ture to be made at each step of the synthesis.
Reaction of (ϩ)-menthone (8) with 3-trimethylsilylbut-3-en-
2-one yielded the diketone 9 as the major product together with
small amounts of diastereoisomeric diketones 6, 10 and 11
(Scheme 1). 13C NMR data for 6 (Table 1) was almost identical
with that reported for the natural product from C. bakeri
(except that literature assignments for C-13 and C-14 should
be interchanged);1 the relative stereochemistry of diketones 6
1
and 9–11 was determined by NOESY and H–1H J-resolved
J. Chem. Soc., Perkin Trans. 1, 2000, 189–194
This journal is © The Royal Society of Chemistry 2000
189

View Article Online
Table 1 13C and 1H NMR assignments (CDCl3) for nor-sesquiterpenes 6, 7 and 9–14
δC
δH
6
7
9
10
52.9
11
54.4
12
50.9
13
50.5
14
6
7
9
10
11
12
13
14
1
2
54.2
21.6
41.1
25.9
57.1
20.4
45.9 2.12 1.99
2.03 2.48 2.48 1.29
1.21
1.85
20.6
21.6
25.5
22.2
25.3 1.83 1.70 (α) 1.78 1.58 1.39 1.68 (α) 1.97 (α) 1.81 (α)
1.83 2.21 (β) 1.82 1.89 1.99 2.06 (β) 2.27 (β) 2.17 (β)
34.9 2.36 2.39 (α) 2.34 2.28 2.30 2.44 (α) 2.23 (α) 2.28 (α)
2.52 2.25 (β) 2.55 2.42 2.53 2.29 (β) 2.28 (β) 2.36 (β)
3
41.5
35.7
41.5
41.5
41.9
41.1
36.5
4
5
208.7 199.9 208.9 208.4 209.0 211.0 211.4 200.1
—
—
—
—
—
—
—
—
29.9 125.5
29.8
29.9
29.8
51.9
45.4 121.9 2.12 5.82
2.12 2.11 2.13 2.74 (α) 2.60 (α) 5.86
2.22 (β) 2.32 (β)
6
7
8α
8β
9α
9β
216.0 169.9 213.3 215.9 213.0
78.1
50.9
20.4
80.0 170.0
—
—
—
—
—
—
—
—
57.3
27.3
52.5
28.8
57.2
29.6
56.0
26.3
57.3
25.4
54.6
24.1
51.2 2.04 1.91
29.1 1.72 1.54
1.93 2.00
2.10 2.05 2.06 1.29
1.30 1.65 1.97 1.41
2.10 1.91 1.57 1.57
1.86 1.88 1.95 1.80
1.47 1.49 1.71 1.07
1.55 2.18 2.36 1.42
2.06 2.06 2.10 2.05
0.89 0.82 0.89 0.91
0.86 0.92 0.86 0.88
1.06 0.86 0.73 0.95
1.29
1.19
1.70
1.86
1.17
1.74
2.03
0.96
0.80
1.00
1.88
1.15
2.00
1.87
1.25
1.52
2.01
0.96
0.88
1.04
29.2
29.9
34.9
28.4
32.7
35.4
35.3
35.2 1.68 1.56
1.54 1.40
10
39.2
27.2
20.9
20.0
20.5
39.5
27.3
21.5
20.8
20.2
40.6
26.3
21.5
18.8
20.6
37.5
27.1
21.0
20.0
15.6
37.6
26.1
21.4
18.7
13.3
32.4
25.5
23.6
18.1
20.3
31.3
25.1
24.7
19.3
20.1
39.0 1.68 1.42
26.9 2.02 1.89
22.0 0.80 0.77
18.4 0.92 0.97
20.3 1.04 1.03
11
12a
13a
14
a Assignments interchangeable within column.
Fig.
2
Solution conformations of nor-muurolane
7
and nor-
amorphane 14 as determined by NOESY (critical correlations from 1H
to 1H indicated by arrows).
As expected, addition of a methyl Grignard reagent to
nor-sesquiterpene 7 yielded muurolane sesquiterpene allylic
alcohol natural products
1 and 2; the corresponding
amorphane (7α-isopropyl) alcohols 15 and 16, which are not
known from nature, were obtained from the Grignard reaction
of 14 (Scheme 2). Compounds 1, 2, 15 and 16 were completely
characterised by 2D-NMR in both CDCl3 and C6D6. This
clearly established that 13C assignments (Table 2) in CDCl3 for
the natural products 1 and 2 reported from C. bakeri by Cool
and co-workers.1 were correct. Although the actual values of
13C NMR chemical shifts recorded in C6D6 reported for natural
products 1 and 2 from C. obtusa2 were also correct, over half
of the resonances were wrongly assigned in each case, thus
explaining the apparent discrepancy in spectral data for the
same compounds from two different biological sources. Thus,
the identity of the sesquiterpene alcohols 1 and 2 from C. bakeri
and C. obtusa has now been confirmed by total synthesis; 13C
assignments reported in the literature for 1 and 2 in C6D6
should be revised to those shown in Table 2 in parentheses. The
13C NMR chemical shifts for the C-15 methyl group in alcohols
1 and 15 were slightly upfield (δC 1–1.5 ppm) compared with
their 4-epimers 2 and 16, which is consistent with a pseudo-axial
conformation for this substituent in the half-chair A-ring for
these two isomers (see Fig. 5).
Scheme 1 Synthesis of nor-sesquiterpene natural products 6 and 7
from (ϩ)-menthone 8.
tent with the expectation that the 7β-isopropyl group in 7 adopts
an axial conformation,3 which leads to gauche interactions with
C-1 and C-9; by contrast, the 7α-isopropyl substituent in 14
is equatorial and is involved in a gauche interaction with C-5
(Fig. 2). Inspection of fully assigned NMR data for muurolanes
1–4 and 17 and amorphanes 15, 16 and 18–20 (Tables 2 and 3)
shows that this effect in the B-ring is also apparent for all
five pairs of isomers subsequently obtained by synthesis (C-1:
δC 39.3–40.7 for muurolanes, δC 44.2–45.3 for amorphanes; C-9:
δC 29.1–30.5 for muurolanes, δC 30.3–35.8 for amorphanes),
and we suggest that determination of the 13C chemical shift at
C-1 and C-9 might provide a general and reliable means for dif-
ferentiating between muurolane and amorphane sesquiterpenes.
Konig and co-workers4 have reported that sesquiterpene
dienes such as bicyclosesquiphellandrene, cadina-3,5-diene and
zonarene, which are diastereoisomeric with natural products 3,
4 and 5 at the 10-position can be interconverted simply by
exposure to the trace acid naturally present in CDCl3. Accord-
ingly, we set out to study in vitro the mechanism by which 1 and
2 undergo dehydration to dienes such as 3–5 and the subsequent
isomerisation of these dienes by NMR spectroscopic analysis
of CDCl3 solutions of these compounds. Allylic tertiary alco-
hol 1 underwent complete dehydration in CDCl3 in less than
190
J. Chem. Soc., Perkin Trans. 1, 2000, 189–194

View Article Online
Table 2 13C NMR assignments (CDCl3)a for sesquiterpenes 1–5 and 15–20
15
44.9 (45.1)
1
2
3
4
5
16
44.7 (44.9)
17
18
45.3
27.8
28.4
144.4
119.8
144.6
49.4
28.5
35.8
39.0
27.0
19
20
1
2
40.0 (40.2)
24.6 (25.1)
36.6 (37.0)
69.5 (68.7)
40.5 (40.7)
24.7 (25.0)
36.8 (37.1)
68.5 (67.5)
40.7
28.0
29.5
39.3
28.3
43.2
28.2
31.2
133.3
29.7
27.7
132.5
121.8
128.9
42.7
18.7
29.1
33.1
28.9
20.9
16.8
20.0
23.2
44.2
28.6
116.9
131.0
181.6
135.8
49.8
27.2
35.5
39.2
26.9
22.0
18.4
20.4
21.6
133.4
27.3
28.8
132.2
122.7
126.8
42.7
20.7
30.3
33.6
29.7
21.2
17.5
19.1
23.1
24.2 (24.7)
35.9 (36.3)
69.2 (68.7)
124.6 (125.9)
144.3 (143.1)
49.4 (49.5)
28.5 (28.7)
35.8 (36.0)
39.0 (39.2)
27.0 (27.2)
22.3 (22.5)
18.4 (18.5)
20.1 (20.3)
29.1 (29.5)
24.6 (24.9)
36.1 (36.5)
68.3 (67.7)
124.5 (125.6)
145.1 (144.1)
48.7 (48.8)
27.9 (27.5)
35.6 (35.8)
40.1 (40.1)
26.9 (27.2)
22.3 (22.4)
18.0 (18.2)
20.4 (20.5)
30.1 (30.5)
3
4
5
6
7
8
9
10
11
12b
13b
14
15
116.9
130.9
122.5
128.9
51.0
29.0
30.5
40.2
26.7
21.8
20.9
20.3
21.5
147.4
124.5
144.5
51.8
28.8
30.4
39.7
26.9
21.8
21.0
20.3
107.8
135.3
120.5
127.9
135.0
23.9
31.9
34.6
28.1
21.0
20.4
20.5
24.1
128.7 (130.0)
144.2 (143.2)
51.7 (51.9)
29.0 (29.3)
30.5 (30.8)
39.3 (39.7)
26.4 (26.6)
21.6 (21.9)
20.9 (21.0)
20.1 (20.2)
28.7 (29.2)
128.2 (129.5)
145.4 (144.4)
51.4 (51.8)
28.4 (28.8)
30.1 (30.4)
40.3 (40.5)
26.7 (26.9)
21.5 (21.8)
20.9 (21.0)
20.3 (20.4)
30.2 (30.7)
22.3
18.3
20.3
108.3
a Values in parentheses were obtained in C6D6. b Values interchangeable within column.
Table 3 1H NMR assignments (CDCl3)a for sesquiterpenes 1–5 and 15–20
1
2
3
4
5
15
16
17
18
19
20
1
1.63 (1.45)
1.35 (1.19)
1.93 (1.79)
1.72 (1.61)
1.51 (1.46)
5.32 (5.32)
1.62 (1.55)
1.40 (1.35)
1.87 (1.81)
1.47 (1.32)
1.29 (1.15)
1.18 (1.01)
1.79 (1.65)
0.77 (0.81)
0.90 (0.85)
0.92 (0.82)
1.26 (1.26)
1.56 (1.36)
1.39 (1.34)
1.92 (1.74)
1.75 (1.73)
1.39 (1.27)
5.35 (5.32)
1.61 (1.55)
1.43 (1.42)
1.87 (1.83)
1.48 (1.33)
1.26 (1.17)
1.16 (1.03)
1.74 (1.61)
0.71 (0.72)
0.89 (0.86)
0.96 (0.87)
1.27 (1.28)
1.75
1.33
2.03
2.31
2.12
5.90
1.71
1.48
1.91
1.46
1.26
1.23
1.81
0.77
0.92
0.95
4.68
4.62
1.85
2.00
2.42
5.25
1.58
1.12
2.08
2.12
2.00
6.22
—
1.99
2.03
1.70
1.17
1.18
3.03
0.96
0.95
0.97
1.77
1.50 (1.33)
1.42 (1.28)
1.91 (1.79)
1.69 (1.60)
1.52 (1.50)
5.29 (5.34)
1.62 (1.52)
0.99 (0.92)
1.81 (1.71)
1.79 (1.62)
1.13 (1.00)
1.19 (1.03)
2.01 (1.99)
0.95 (0.96)
0.86 (0.86)
0.92 (0.83)
1.28 (1.28)
1.45 (1.28)
1.43 (1.42)
1.90 (1.74)
1.72 (1.71)
1.43 (1.34)
5.35 (5.39)
1.62 (1.52)
1.05 (0.97)
1.78 (1.66)
1.78 (1.62)
1.14 (0.97)
1.17 (1.03)
2.06 (2.02)
0.95 (0.92)
0.86 (0.87)
0.96 (0.86)
1.29 (1.30)
—
1.62
1.50
1.95
2.32
2.15
5.90
1.60
1.08
1.85
1.72
1.12
1.30
2.00
0.99
0.86
0.95
4.72
4.66
1.70
2.01
2.40
5.29
—
2α
2β
3α
3β
5
1.94
1.94
2.15
2.03
5.61
1.98
1.47
1.47
1.98
1.28
2.03
1.99
0.93
0.68
1.02
1.78
2.00
2.08
2.10
2.07
5.61
1.95
1.38
1.61
1.82
1.33
2.18
1.95
0.92
0.70
0.96
1.79
5.42
1.73
1.48
1.91
1.48
1.30
1.36
1.75
0.80
0.91
0.90
1.68
5.47
1.74
1.12
2.00
1.75
1.20
1.35
2.10
0.97
0.88
0.92
1.71
7
8α
8β
9α
9β
10
11
12b
13b
14
15
a Values in parentheses were obtained in C6D6. b Assignments interchangeable within column.
four days (Fig. 3); the diene 3 was the most rapidly formed
product, accounting for 50% of the mixture at this point in time
(Fig. 3). 13C NMR data for compound 3 agreed well with that
reported for the natural product1 (with the exception that
assignments reported in the literature for C-3–C-9 and C-4–C-6
should be reversed within each pair). Over the next week, peaks
due to 3 were seen to disappear from the 1H NMR spectrum at
a rate roughly equivalent to that at which peaks for 17 (not
yet reported as a natural product) appeared in the mixture;
isomerisation of 3 to 17 involves conversion of an exocyclic
double bond to a thermodynamically more stable endocyclic
double bond. In Fig. 3 it can also be seen that diene 4 is
also formed in the early phases of the dehydration of 1 at a
slower rate than for 3. As for all compounds reported herein,
the structure of 4 was unambiguously established by 2D-NMR
(Tables 2 and 3) thereby rigorously establishing the identity
of this natural product, which was previously only tentatively
identified on mass spectral evidence alone.1 Diene 4 reaches
its maximum concentration after around three days in CDCl3
solution (Fig. 4), after which the signal for this isomer dis-
appears from the 1H NMR spectrum at a rate roughly
equivalent to that for the appearance of regioisomeric diene 5
(Fig. 3), which is the known compound epizonarene. The
course of the reaction followed by tertiary alcohol 2 (epi-
meric at the 4-position) was similar to that described for 1,
although the rate of dehydration was slower in the case of epi-
mer 2. On the basis of the foregoing results we suggest that
diene natural products 3 and 4 were produced directly from the
Fig. 3 Dehydration of muurolane allylic tertiary alcohol 1 to ses-
1
quiterpene dienes 3–5 and 17 as studied by H NMR spectroscopy in
CDCl3 solution.
dehydration of allylic alcohols 1 and/or 2, either as a result
of chemical processes occurring in the plant cell or during
the extraction procedure for C. bakeri, and that natural
product 5 was then formed principally by subsequent isomeris-
ation of diene 4 (Scheme 2).
In vitro study of the course of dehydration of the unnatural
1
amorphane tetiary alcohol 15 in CDCl3 by H NMR showed
that this compound underwent direct dehydration to the 4,6-
diene 5 (epizonarene) at a rate which was appreciably faster
than that for the alternative route yielding the 4(15),5-diene 18
J. Chem. Soc., Perkin Trans. 1, 2000, 189–194
191

View Article Online
Fig. 4 Selected 1H NMR spectra (expansion of alkene region) showing the variation in intensity for signals corresponding to compounds 1 (δH 5.32,
H-5), 3 (δH 5.90, H-5; 4.68 and 4.62, H-15), 4 (δH 5.42, H-5, 5.25 (br), H-3), 5 (δH 6.22, H-5) and 17 (δH 5.61, H-5) against time, used in calculating the
product distribution in Fig. 3.
Fig. 5 Preferred conformations of muurolane allylic alcohols 1 and 2
and amorphane allylic alcohols 15 and 16 as demonstrated by NOESY
(critical correlations from 1H to 1H indicated by arrows).
amounts of the 1(6),4-diene 20, which is formed by further
isomerisation of 18; by contrast, the final product from de-
hydration/rearrangement of muurolanes 1 and 2 consisted pre-
dominantly of the 1(6),4-diene 17 with smaller amounts of 5.
Clearly, the main difference between the dehydration of
amorphane allylic tertiary alcohols 15 and 16 and muurolane
allylic alcohols 1 and 2 in CDCl3 solution is the direct form-
ation of epizonarene 5 in the former case, involving elimination
of H-7. The preference for this elimination in the amorphane
allylic alcohols can be explained as a consequence of the
axial conformation of H-7 in 15 and 16 as compared with the
equatorial conformation of H-7 in 1 and 2. Such conform-
ational preferences have been noted before for muurolane and
Scheme 2 Synthesis of sesquiterpene allylic tertiary alcohols 1, 2, 15
and 16 and conversion into sesquiterpene dienes 3–5 and 17–20.
amorphane sesquiterpenes3 and were explicitly demonstrated in
this study by analysing NOESY spectra for all four isomeric
(which is the amorphane analogue of 3). Compound 18 was in
turn formed more rapidly than the 3,5-diene 19 (which is the
amorphane analogue of 4, Scheme 2). The same observations
held true for the tertiary alcohol 4-epimer 16, which underwent
dehydration more slowly than 15 (cf. slower rate of dehydration
of 2 as compared with 1) but gave a qualitatively similar pattern
of products. On completion of the dehydration/rearrangement
reactions of both 15 and 16 in CDCl3 solution, the reaction mix-
ture consisted predominantly of epizonarene 5, with only small
alcohols (Fig. 5). Assuming an E1 mechanism, direct formation
of diene 5 should involve the initial generation of a C-4–C-6
allylic cation by elimination of water, followed by elimination
of H-7. The latter process is more favoured when this proton
is axial (as is the case for the amorphanes) because there is
a better orbital overlap between the σ(C7–H7) bond and the
π-system of the allylic cation in the intermediate (Fig. 6) than is
the case when H-7 is equatorial (as is found for muurolane
sesquiterpenes 1 and 2, Fig. 5).
192
J. Chem. Soc., Perkin Trans. 1, 2000, 189–194

View Article Online
(1R,6S,7R,10S)-6-Hydroxy-7-(1-methylethyl)-10-methylbicyclo-
[4.4.0]decan-4-one 12
To a solution of 9 in EtOH (243 mg, 15 ml) was added Ba-
(OH)2ؒ8H2O (342 mg) and the solution was stirred for 3 h at
0 ЊC. The reaction was neutralised with HCl (10%), concen-
trated under reduced pressure and extracted with CHCl3
(3 × 20 ml). The organic extract was washed with water (2 × 20
ml), dried and rotary evaporated to yield a crude product (238
mg, 98%) which was shown to consist predominantly of
decalone alcohol 12 by preparative HPLC (15% EtOAC–
Fig. 6 Dehydration of amorphane allylic tertiary alcohol 15 to
epizonarene (5) favoured by the axial nature of H-7.
Experimental
General methods
hexane).
(1R,6S,7R,10S)-6-Hydroxy-7-(1-methylethyl)-10-
Chemical shifts are expressed in ppm (δ) relative to TMS as
internal standard. All NMR experiments were run on a Bruker
DRX 500 instrument with CDCl3 as solvent (C6D6 was also
used for compounds 1, 2, 15 and 16). HSQC and HMBC
experiments were recorded with 2048 data points in F2 and
128 data points in F1. HREIMS were recorded at 70 eV on a
Finnigan-MAT 95 MS spectrometer. FTIR spectra were
recorded in CHCl3 on a Shimadzu FTIR-8201 PC instrument.
TLC plates were developed using p-anisaldehyde. Column
chromatography was performed using silica gel 60–200 µm
(Merck). HPLC separations were performed using a PREP-SIL
20 mm × 25 cm column, flow rate 8 ml minϪ1. Optical rotations
were recorded on a Perkin-Elmer 343 polarimeter at 20 ЊC and
methylbicyclo[4.4.0]decan-4-one 12: oil (120 mg, 51%; Rt 24.8
min); [α]D ϩ51.4 (c 2.6, CHCl3); νmax/cmϪ1 3614, 3422 (br),
3026, 3011, 2961, 2932, 1709, 1466, 1452; δH 2.74 (1H, dd,
J 14.1, 2.4 Hz), 2.44 (1H, dddd, J 14.3, 7.1, 4.6, 2.4 Hz), 2.29
(1H, ddd, J 14.3, 14.2, 6.8 Hz), 2.22 (1H, d, J 14.1 Hz), 0.95
(3H, d, J 6.4 Hz), 0.91 (3H, d, J 6.9 Hz), 0.88 (3H, d, J 6.9 Hz);
13C NMR—see Table 1; m/z (rel. int.) 224.1775 (Mϩ, calc.
224.1776 for C14H24O2) (30), 209 (20), 164 (15), 139 (100), 111
(20). (1R,6R,7R,10S)-6-Hydroxy-7-(1-methylethyl)-10-methyl-
bicyclo[4.4.0]decan-4-one 13: oil (3 mg, 1%; Rt 60.4 min); [α]D
ϩ36.2 (c 0.1, CHCl3); νmax/cmϪ1 3595, 3414 (br), 3022, 2961,
2928, 2872, 1707, 1468, 1450; δH 2.60 (1H, d, J 16.6 Hz), 1.00
(3H, d, J 6.5 Hz), 0.96 (3H, d, J 6.9 Hz), 0.80 (3H, d, J 6.9 Hz);
13C NMR—see Table 1; m/z (rel. int.) 224.1784 (Mϩ, calc.
224.1776 for C14H24O2) (20), 209 (8), 206 (8), 164 (10), 139
(100), 120 (20), 112 (20).
[α]D has units of 10Ϫ1 deg cm2 gϪ1
.
(1S,2R,4R)-1-Methyl-2-(3-oxobutyl)-3-oxo-4-(1-methylethyl)-
cyclohexane 9
To a solution of lithium diisopropylamide (LDA) in THF
(freshly prepared from BuLi (1.6 M; 3.96 ml), diisopropylamine
(0.89 ml) and THF (12 ml)) was added dropwise a solution of
(ϩ)-menthone (8) in THF (722 mg, 1.2 ml) under cooling
(Ϫ78 ЊC). After stirring for 30 min, 3-trimethylsilylbut-3-en-2-
one5 in THF (1.0 g, 2 ml) was added dropwise and stirring
continued at Ϫ78 ЊC for 1 h, then the solution was warmed to
0 ЊC and stirring continued for 2.5 h. The reaction was quenched
by acidification with HCl (10%), neutralised with NaHCO3
(5%) and extracted into EtOAc (3 × 20 ml). The organic extract
was dried and rotary evaporated to give a crude product which
consisted predominantly of diketone 9, which was purified by
preparative HPLC (8% EtOAc–hexane). (1S,2R,4R)-1-Methyl-
2-(3-oxobutyl)-3-oxo-4-(1-methylethyl)cyclohexane 9: oil (335
mg, 32%; Rt 24.8 min); [α]D ϩ46.1 (c 1.5, CHCl3); νmax/cmϪ1
3024, 3011, 2961, 2932, 2874, 1705, 1445, 1367; δH 2.55 (1H,
ddd, J 17.1, 8.8, 5.7 Hz), 2.34 (1H, ddd, J 17.1, 8.1, 4.0 Hz), 2.12
(3H, s), 1.06 (3H, d, J 6.2 Hz), 0.89 (3H, d, J 6.4 Hz), 0.86 (3H,
d, J 6.4 Hz); 13C NMR—see Table 1; m/z (rel. int.) 224.1775
(Mϩ, calc. 224.1776 for C14H24O2) (95), 209 (100), 182 (20), 167
(25), 149 (30), 111 (25), 95 (30). (1S,2R,4S)-1-Methyl-2-(3-
oxobutyl)-3-oxo-4-(1-methylethyl)cyclohexane 6: oil (40 mg,
4%; Rt 36.0 min); [α]D Ϫ62.3 (c 1.1, CHCl3); νmax/cmϪ1 3024,
3011, 2961, 2932, 2872, 1705, 1458, 1371; δH 2.52 (1H, ddd,
J 17.4, 8.2, 6.3 Hz), 2.36 (1H, ddd, J 17.4, 8.1, 7.9 Hz), 2.12 (3H,
s), 1.04 (3H, d, J 6.4 Hz), 0.92 (3H, d, J 6.3 Hz), 0.80 (3H, d,
J 6.3 Hz); 13C NMR—see Table 1; m/z (rel. int.) 224.1782 (Mϩ,
calc. 224.1776 for C14H24O2) (100), 209 (95), 182 (20), 166
(60), 139 (70), 111 (70), 95 (75). (1S,2S,4R)-1-Methyl-2-(3-
oxobutyl)-3-oxo-4-(1-methylethyl)cyclohexane 10: oil (15 mg,
1.5%; Rt 37.9 min); [α]D ϩ60.1 (c 0.5, CHCl3); νmax/cmϪ1 3024,
2963, 2932, 2874, 1701, 1458, 1369; δH 2.11 (3H, s), 0.92 (3H, d,
J 6.4 Hz), 0.86 (3H, d, J 7.1 Hz), 0.82 (3H, d, J 6.4 Hz); 13C
NMR—see Table 1; m/z (rel. int.) 224.1780 (Mϩ, calc. 224.1776
for C14H24O2) (95), 209 (100), 181 (17), 166 (62), 153 (65), 124
(70), 111 (80), 95 (55). (1S,2S,4S)-1-Methyl-2-(3-oxobutyl)-3-
oxo-4-(1-methylethyl)cyclohexane 11: oil (5 mg, 0.5%; Rt 36.5
min); [α]D Ϫ5.4 (c 0.3, CHCl3); νmax/cmϪ1 2955, 2876, 1705;
δH 2.13 (3H, s), 0.89 (3H, d, J 6.5 Hz), 0.86 (3H, d, J 6.5 Hz),
0.73 (3H, d, J 7.2 Hz); 13C NMR—see Table 1; m/z (rel. int.)
224.1779 (Mϩ, calc. 224.1776 for C14H24O2) (90), 209 (100), 181
(15), 167 (55), 153 (60), 139 (45), 124 (50), 95 (50).
(1R,7S,10S)-7-(1-Methylethyl)-10-methylbicyclo[4.4.0]dec-5-
en-4-one 7
A solution of tertiary alcohol 12 in EtOH (70 mg, 15 ml) was
stirred in conc. H2SO4 (15 ml) at room temperature for 3 h. The
reaction was neutralised with NaHCO3 (5%), concentrated
under reduced pressure and extracted with CHCl3 (3 × 15 ml).
The combined organic extracts were washed with water (2 × 10
ml) and brine (20 ml), dried and rotary evaporated to give a
crude product consisting predominantly of decalenone 7 which
was purified by preparative HPLC (5% EtOAc–hexane): oil (38
mg, 59%; Rt 55.5 min); [α]D ϩ49.5 (c 0.4, CHCl3); νmax/cmϪ1
2961, 2930, 2870, 1663; δH 5.82 (1H, d, J 1.7 Hz), 1.03 (3H, d,
J 6.0 Hz), 0.97 (3H, d, J 6.2 Hz), 0.77 (3H, d, J 6.2 Hz); 13C
NMR—see Table 1; m/z (rel. int.) 206.1672 (Mϩ, calc. 206.1671
for C14H22O) (45), 191 (8), 164 (100), 149 (26), 122 (22).
4ꢀ-Hydroxymuurol-5-ene 1 and 4ꢁ-hydroxymuurol-5-ene 2†
To a Grignard reagent freshly prepared from Mg (43 mg),
CH3I (273 mg) and Et2O (15 ml) was added a solution of the
α,β-unsaturated ketone 7 in Et2O (36 mg, 3 ml). The reaction
mixture was refluxed (1 h) and Et2O (20 ml) was added upon
completion. The ethereal layer was washed with water (2 × 5
ml), dried and rotary evaporated to give an oily crude product
consisting of sesquiterpene diastereoisomers 1 and 2 in an
approximately 1:1 ratio which were separated by preparative
HPLC (10% EtOAc–hexane). 4β-Hydroxymuurol-5-ene 1: crys-
tal, mp 75–77 ЊC (11 mg, 29%; Rt 23.9 min); [α]D ϩ33.3 (c 0.1,
CHCl3); νmax/cmϪ1 3429 (br), 2957, 2928, 2868, 1454, 1367;
δH (CDCl3) 5.32 (1H, s), 1.26 (3H, s), 0.92 (3H, d, J 6.4 Hz), 0.90
(3H, d, J 6.6 Hz), 0.77 (3H, d, J 6.6 Hz); δH (C6D6) 5.32 (1H, s),
1.26 (3H, s), 0.85 (3H, d, J 6.6 Hz), 0.82 (3H, d, J 6.5 Hz),
0.81 (3H, d, J 6.6 Hz); 13C NMR—see Table 2; m/z (rel. int.)
222.1983 (Mϩ, calc. 222.1984 for C15H26O) (10), 207 (100), 204
(25), 179 (10), 161 (95), 105 (30). 4α-Hydroxymuurol-5-ene 2:
oil (11 mg, 29%; Rt 25.5 min); [α]D ϩ91.4 (c 0.1, CHCl3);
νmax/cmϪ1 3442 (br), 2928, 2856; δH (CDCl3) 5.35 (1H, s), 1.27
(3H, s), 0.96 (3H, d, J 6.4 Hz), 0.89 (3H, d, J 6.6 Hz), 0.71 (3H,
† IUPAC names for 1 and 2 are (1R,4S,7S,10S)- and (1R,4R,7S,10S)-
4-hydroxy-7-(1-methylethyl)-4,10-dimethylbicyclo[4.4.0]dec-5-ene,
respectively.
J. Chem. Soc., Perkin Trans. 1, 2000, 189–194
193

View Article Online
d, J 6.6 Hz); δH (C6D6) 5.32 (1H, s), 1.28 (3H, s), 0.87 (3H, d,
J 6.4 Hz), 0.86 (3H, d, J 6.4 Hz), 0.72 (3H, d, J 6.4 Hz); 13C
NMR—see Table 2; m/z (rel. int.) 222.1980 (Mϩ, calc. 222.1984
for C15H26O) (5), 207 (100), 204 (15), 161 (30).
calculated in the same way. Muurola-4(15),5-diene 3:§ see ref. 1
for physical data; δH 5.90 (1H, s), 4.68 (1H s), 4.62 (1H, s), 0.95
(3H, d, J 6.4 Hz), 0.92 (3H, d, J 6.6 Hz), 0.77 (3H, d, J 6.6 Hz);
13C NMR—see Table 2. Muurola-3,5-diene 4:§ see ref. 1 for
physical data; δH 5.42 (1H, s), 5.25 (1H, br s), 1.68 (3H, d, J 1.7
Hz), 0.91 (3H, d, J 6.3 Hz), 0.90 (3H, d, J 6.6 Hz), 0.80 (3H, d,
J 6.6 Hz); 13C NMR—see Table 2. Epizonarene (muurola-4,6-
diene) 5:§ see ref. 1 for physical data; δH 6.22 (1H, s), 3.03
(1H, sept, J 6.9 Hz), 1.77 (3H, s), 0.97 (3H, d, J 6.8 Hz), 0.96
(3H, d, J 6.9 Hz), 0.95 (3H, d, J 6.9 Hz); 13C NMR—see Table
2. Muurola-1(6),4-diene 17:§ oil, δH 5.61 (1H, s), 1.78 (3H,
s), 1.02 (3H, d, J 7.0 Hz), 0.93 (3H, d, J 6.6 Hz), 0.68 (3H, d,
J 6.6 Hz); 13C NMR—see Table 2; m/z (rel. int.) 204.1868
(Mϩ, calc. 204.1878 for C15H24) (55), 189 (33), 161 (100), 159
(30), 133 (20), 119 (35), 105 (40).
(1R,7R,10S)-7-(1-Methylethyl)-10-methylbicyclo[4.4.0]dec-5-
en-4-one 14
A solution of tertiary alcohol 12 in EtOH (116 mg, 30 ml) was
stirred with HCl (6 M, 30 ml) for 5 h at 0 ЊC. The reaction was
neutralised with NaHCO3 (5%), concentrated under reduced
pressure and extracted with CHCl3 (3 × 15 ml). The combined
organic extracts were washed with water (2 × 10 ml) and brine
(20 ml), dried and rotary evaporated to give a crude product
consisting predominantly of decalenone 14 which was purified
by HPLC (15% EtOAc–hexane): oil (55 mg, 52%; Rt 17.5 min);
[α]D ϩ3.2 (c 0.4, CHCl3); νmax/cmϪ1 2961, 2932, 2874, 1663;
δH 5.86 (1H, s), 2.36 (1H, ddd, J 17.0, 9.3, 4.6 Hz), 2.28 (1H,
ddd, J 17.0, 14.0, 5.0 Hz), 2.17 (1H, m), 1.04 (3H, d, J 6.4 Hz),
0.96 (3H, d, J 6.7 Hz), 0.88 (3H, d, J 6.7 Hz); 13C NMR—
see Table 1; m/z (rel. int.) 206.1672 (Mϩ, calc. 206.1671 for
C14H22O) (50), 191 (8), 164 (100), 149 (18), 122 (18).
Dehydration of amorphane allylic alcohols (15 and 16) in CDCl3
Compound 15 (10 mg) was dissolved in CDCl3 (0.6 ml) in an
NMR tube wrapped in silver foil (to prevent Diels–Alder type
addition reactions of singlet oxygen with endocyclic dienes
1
formed during the reaction) and left at room temperature. H
NMR spectra were acquired at intervals of several hours during
the first few days of the reaction and then once or twice a day
until all reaction had ceased. Spectra for around 20–30 time
points acquired in this way were then used to create a graph of
the percentage composition of each diene product in the mix-
4ꢀ-Hydroxyamorph-5-ene 15 and 4ꢁ-hydroxyamorph-5-ene 16‡
To a Grignard reagent freshly prepared from Mg (43 mg), CH3I
(273 mg) and Et2O (15 ml) was added a solution of the
α,β-unsaturated ketone 14 in Et2O (36 mg, 3 ml). The reaction
mixture was refluxed (1 h) and Et2O (20 ml) was added upon
completion. The ethereal layer was washed with water (2 × 5
ml), dried and rotary evaporated to give an oily crude product
consisting of sesquiterpene diastereoisomers 15 and 16 in an
approximately 3:2 ratio which were separated by preparative
HPLC (10% EtOAc–hexane). 4β-Hydroxyamorph-5-ene 15:
crystal, mp 46–47 ЊC (15 mg, 38%; Rt 25.5 min); [α]D ϩ37.2
(c 0.3, CHCl3); νmax/cmϪ1 3599, 3420 (br), 3007, 2959, 2930, 2872,
1452, 1371, 1215; δH (CDCl3) 5.29 (1H, s), 1.28 (3H, s), 0.95
(3H, d, J 6.7 Hz), 0.92 (3H, d, J 6.2 Hz), 0.86 (3H, d, J 6.7 Hz);
δH (C6D6) 5.34 (1H, s), 1.28 (3H, s), 0.96 (3H, d, J 6.7 Hz),
0.86 (3H, d, J 6.7 Hz), 0.83 (3H, d, J 6.0 Hz); 13C NMR—see
Table 2; m/z (rel. int.) 222.1984 (Mϩ, calc. 222.1984 for
C15H26O) (5), 207 (100), 202 (50), 189 (10), 179 (7), 161 (50).
4α-Hydroxyamorph-5-ene 16: oil (10 mg, 27%; Rt 22.6 min);
[α]D ϩ69.2 (c 0.4, CHCl3); νmax/cmϪ1 3599, 3429 (br), 3007,
2959, 2930, 2872, 1452, 1375; δH (CDCl3) 5.35 (1H, s), 1.29
(3H, s), 0.96 (3H, d, J 6.2 Hz), 0.95 (3H, d, J 6.7 Hz), 0.86 (3H,
d, J 6.7 Hz); δH (C6D6) 5.39 (1H, s), 1.30 (3H, s), 0.92 (3H, d,
J 6.7 Hz), 0.87 (3H, d, J 6.7 Hz), 0.86 (3H, d, J 6.1 Hz); 13C
NMR—see Table 2; m/z (rel. int.) 222.1978 (Mϩ, calc. 222.1984
for C15H26O) (10), 207 (100), 204 (10), 179 (18), 161 (70).
1
ture against time by calculating the ratio of an integral in H
NMR for an alkene proton which was clearly resolved for each
particular compound (15: H-5, δ 5.29 (1H, s); 16: H-5, δ 5.35
(1H, s); 5: H-5, δH 6.22 (1H, s); 18:¶ H-5, δH 5.90 (1H, s); 19:
H-5, δH 5.47 (1H, s); 20: H-5, δH 5.61 (1H, s) to the sum of
all such integrals. The product distribution for dehydration/
rearrangement of compound 16 (10 mg, 0.6 ml, CDCl3) was
calculated in the same way. Epizonarene 5: physical data as in
previous section. Amorpha-4(15),5-diene 18: δH (CDCl3) 5.90
(1H, s), 4.72 (1H, s), 4.66 (1H, s), 0.99 (3H, d, J 6.4 Hz), 0.95
(3H, d, J 6.2 Hz), 0.86 (3H, d, J 6.4 Hz); 13C NMR—see
Table 2. Amorpha-3,5-diene 19:¶ δH 5.47 (1H, d, J 1.4 Hz),
5.29 (1H, br s), 1.71 (3H, d, J 1.8 Hz), 0.97 (3H, d, J 6.4 Hz),
0.92 (3H, d, J 6.2 Hz), 0.88 (3H, d, J 6.4 Hz); 13C NMR—see
Table 2. Amorpha-1(6),4-diene 20:¶ δH 5.61 (1H, s), 1.79
(3H, s), 0.96 (3H, d, J 6.4 Hz), 0.92 (3H, d, J 6.7 Hz), 0.70 (3H,
d, J 6.7 Hz); 13C NMR—see Table 2.
Acknowledgements
We thank the CRCG for funding this research.
§ IUPAC name for 3 is (1R,7S,10S)-4-methylidene-7-(1-methylethyl)-
10-methylbicyclo[4.4.0]dec-5-ene. IUPAC name for 4 is (1R,7S,10S)-
Dehydration of muurolane allylic alcohols (1 and 2) in CDCl3
7-(1-methylethyl)-4,10-dimethylbicyclo[4.4.0]deca-3,5-diene.
IUPAC
name for 5 is (1R,10S)-7-(1-methylethyl)-4,10-dimethylbicyclo[4.4.0]-
deca-4,6-diene. IUPAC name for 17 is (7S,10S)-7-(1-methylethyl)-4,10-
dimethylbicyclo[4.4.0]deca-1(6),4-diene.
¶ IUPAC name for 18 is (1R,7R,10S)-4-methylidene-7-(1-methylethyl)-
10-methylbicyclo[4.4.0]dec-5-ene. IUPAC name for 19 is (1R,7R,
10S)-7-(1-methylethyl)-4,10-dimethylbicyclo[4.4.0]deca-3,5-diene.
IUPAC name for 20 is (7R,10S)-7-(1-methylethyl)-4,10-dimethyl-
bicyclo[4.4.0]deca-1(6),4-diene.
Compound 1 (10 mg) was dissolved in CDCl3 (0.6 ml) in an
NMR tube wrapped in silver foil (to prevent Diels–Alder type
addition reactions of singlet oxygen with endocyclic dienes
1
formed during the reaction) and left at room temperature. H
NMR spectra were acquired at intervals of several hours during
the first few days of the reaction and then once or twice a day
until all reaction had ceased. Spectra for around 20–30 time
points acquired in this way were then used to create a graph of
the percentage composition of each diene product in the mix-
References
1
ture against time by calculating the ratio of an integral in H
1 Y.-K. Kim, L. G. Cool and E. Zavarin, Phytochemistry, 1994, 36, 961;
L. G. Cool and K. Jiang, Phytochemistry, 1995, 40, 177.
2 T. Hieda, M. Tazaki, Y. Morishita, T. Aoki and S. Nagahama,
Phytochemistry, 1996, 42, 159.
3 S.-M. Hui, K.-S. Ngo and G. D. Brown, J. Chem. Soc., Perkin Trans. 1,
1997, 3435.
NMR for an alkene proton which was clearly resolved for each
particular compound (1: H-5, δ 5.32 (1H, s); 2: H-5, δ 5.35 (1H,
s); 3: H-5, δH 5.90 (1H, s); 4: H-3, δH 5.25 (1H, br s); 5: H-5,
δH 6.22 (1H, s); 17: H-5, δH 5.61 (1H, s)) to the sum of all
such integrals. The product distribution for dehydration/
rearrangement of compound 2 (10 mg, 0.6 ml, CDCl3) was
4 S. Melching, N. Bulow, K. Wihstutz, S. Jung and W. A. Konig,
Phytochemistry, 1997, 44, 1291.
5 X.-X. Xu, J. Zhu, D.-Z. Huang and W.-S. Zhou, Tetrahedron, 1986,
42, 819.
‡ IUPAC names for 15 and 16 are (1R,4S,7R,10S)- and (1R,4R,7R,
10S)-4-hydroxy-7-(1-methylethyl)-4,10-dimethylbicyclo[4.4.0]dec-5-ene,
respectively.
Paper a907752i
194
J. Chem. Soc., Perkin Trans. 1, 2000, 189–194