Relative population of S-form and F-form conformers of bryonolic acid and its derivatives in equilibrium in CDCl3 solutions

Article abstract of DOI:10.1248/cpb.50.467

Relative populations of S-form (D-E rings: boat-boat form) and F-form (D-E rings: chair-chair form) conformers, in equilibrium in CDCl₃ solutions, of 20 derivatives (2-21) of bryonolic acid (D:C-friedoolean-8-en-3 β-ol-29-oic acid) (1) were calculated from NMR spectral data (J-values and chemical shifts), with the aid of molecular mechanic calculation using a MM2/CONFLEX program system. The principal deciding factor of the population ratio was found to be whether the functionality at C-29 is trigonal or tetrahedral; the S-form : F-form was 0:100-32:68 for the "trigonal" type and 48:52-100:0 for the "tetrahedral." The reliability of the results is discussed.

Full text of DOI:10.1248/cpb.50.467

April 2002
Chem. Pharm. Bull. 50(4) 467—474 (2002)
467
Relative Population of S-Form and F-Form Conformers of Bryonolic
Acid and Its Derivatives in Equilibrium in CDCl Solutions
3
,
a
a
a
Chie HONDA* Kiyoko SUWA, Shizuyo TAKEYAMA, and Wasuke KAMISAKO
a
School of Pharmaceutical Sciences, Mukogawa Women’s University, 11–68 Kyuban-cho, Koshien, Nishinomiya 663–8179,
Japan. Received October 15, 2001; accepted January 8, 2002
Relative populations of S-form (D–E rings: boat–boat form) and F-form (D–E rings: chair–chair form) con-
formers, in equilibrium in CDCl solutions, of 20 derivatives (2—21) of bryonolic acid (D:C-friedoolean-8-en-3b-
3
ol-29-oic acid) (1) were calculated from NMR spectral data (J-values and chemical shifts), with the aid of molec-
ular mechanic calculation using a MM2/CONFLEX program system. The principal deciding factor of the popu-
lation ratio was found to be whether the functionality at C-29 is trigonal or tetrahedral; the S-form : F-form was
0
: 100—32 : 68 for the “trigonal” type and 48 : 52—100 : 0 for the “tetrahedral.” The reliability of the results is
discussed.
Key words bryonolic acid; conformation; MM2 calculation; CONFLEX
Bryonolic acid (D:C-friedoolean-8-en-3b-ol-29-oic acid) and that their relative populations are greatly affected by the
1
)
3)
(
1) (see Fig. 1) is a D:C-friedooleanane class of triterpene functionality at C-29.
which has been isolated from the roots, cultured cells or cul-
Tabata and co-workers found that 1 had anti-allergic activ-
tured hairy roots of various cucurbitacean plants. Previ- ities. They also showed that the anti-allergic activities were
2
)
4)
ously, Kamisako et al. suggested that two types of conforma- significantly affected by the C-29 functionality; converting
tions (Fig. 2), i.e., the S-form (D–E rings: boat–boat form) the functional group at C-29 from –COOH to –CH OCO-
2
5
)
and the F-form (D–E rings: chair–chair form) are possible (CH ) COONa raised the activities as much as by tenfold.
2
2
for bryonolic acid derivatives, that molecules having these More recently, a variety of biological or pharmacological ac-
6
)
conformations co-exist in equilibrium in CDCl solutions, tivities, such as anti-inflammatory, sebum secretion promot-
3
7
)
8)
9)
ing, antitumor and skin cell growth promoting action,
have been reported by other workers. This led us to conduct
more detailed studies on the conformational feature of 1 and
its derivatives. The physicochemical properties described
above should be taken into account when relationships be-
tween biological activity and structure are discussed.
This paper presents studies on the relationship between the
C-29 functionality and relative populations of the S-form and
F-form conformers in equilibrium in solutions (S : F popula-
tions). Twenty model compounds (2—21) (see Fig. 1) were
examined and their S : F populations were calculated from
the NMR spectral data, i.e., J-values and chemical shifts,
with the aid of molecular mechanics calculation using an
1
0)
11)
MM2 program. CONFLEX, a program utilized together
with the MM2, enabled us to efficiently find all the practical
conformers, such as rotamers around the C(20)—C(29)
bonds of both the S-form and F-form conformers, and to
check the probability of other conformational types corre-
sponding to those previously discussed by Kikuchi and co-
1
2)
workers The MM2 calculation also offered us theoretical
a)
Fig. 2. Conformations of the D–E Ring Systems of D:A—D:C-
Friedooleananes Drawn from the Dreiding Model in a Regular Conforma-
tion
Fig. 1. Structures of Bryonolic Acid (1) and Its Derivatives (2—21)
a) Extra numbers 31—34 are given to the carbons and protons of the substituents at
C-29. b) Compounds 16 and 17 are epimeric to each other with respect to the asymmet-
ric center at C-29.
Open circles, methyl group; closed circles, hydrogen atoms.
∗
To whom correspondence should be addressed. e-mail: hondac@mwu.mukogawa-u.ac.jp
© 2002 Pharmaceutical Society of Japan

4
68
Vol. 50, No. 4
13
a)
Table 1.
C Chemical Shifts of Bryonolic Acid Derivatives
Carbon
2
3
4
5
6
7
8
9
10
11
1
2
3
4
5
6
7
8
9
34.73
24.23
80.89
37.79
50.58
19.08
27.41
134.10
133.87
37.46
20.63
30.85
37.42
41.84
25.00
36.99
30.91
44.32
29.25
44.45
26.37
34.10
28.00
16.68
19.95
22.33
19.04
31.14
205.27
27.62
36.88
19.29
41.73
33.31
51.46
19.49
27.78
133.31
134.80
37.87
20.80
30.19
37.08
42.10
25.62
37.03
30.93
44.64
30.51
40.46
29.61
34.32
33.02
21.89
19.88
21.81
18.18
31.31
185.46
32.73
35.11
27.96
78.99
38.85
50.56
19.25
27.60
133.95
134.08
37.59
20.70
30.40
37.23
41.88
25.10
37.08
30.95
44.84
30.82
40.51
29.98
34.46
28.05
15.57
19.91
22.08
17.17
31.24
179.21
32.87
51.49
35.23
28.09
79.36
38.91
50.53
19.07
28.29
134.21
133.70
37.66
20.75
30.44
37.19
42.15
25.73
37.38
30.78
45.02
33.01
36.09
29.13
34.64
27.76
15.74
19.77
21.70
21.43
31.84
174.25
31.37
34.81
24.24
80.93
37.78
50.66
19.11
27.42
134.22
133.83
37.47
20.69
30.59
37.33
41.72
25.29
36.89
31.06
44.66
30.20
46.44
29.11
34.79
28.01
16.69
19.96
22.53
18.50
31.20
213.95
31.50
25.36
35.39
28.05
79.30
38.88
50.57
19.15
28.05
134.16
133.92
37.62
20.82
30.75
37.09
41.84
25.88
37.37
31.10
45.37
30.64
39.74
30.40
35.29
27.83
15.68
19.81
22.19
19.58
31.88
169.13
32.49
13.84
35.13
28.00
79.07
38.85
50.70
19.27
27.61
134.51
133.89
37.62
20.77
30.74
37.36
41.52
25.59
37.11
31.02
44.96
34.37
34.77
30.14
35.24
28.08
15.58
19.90
23.05
20.58
31.63
151.35
34.30
107.74
34.79
24.27
81.00
37.78
50.77
19.14
27.45
134.60
133.71
37.48
20.77
30.17
37.34
41.52
25.59
37.08
31.00
44.95
34.36
34.73
30.72
35.24
28.02
16.68
19.95
23.07
20.48
31.63
151.25
34.25
107.82
34.82
24.26
81.01
37.77
50.85
19.15
27.38
134.98
133.50
37.50
20.82
30.57
37.50
41.19
26.15
36.77
31.28
44.58
30.97
37.11
31.22
36.58
28.03
16.70
19.93
24.21
18.76
31.50
153.56
32.53
108.53
19.96
35.10
27.99
79.07
38.85
50.78
19.30
27.53
135.16
133.59
37.67
20.84
30.75
37.36
41.02
26.35
36.73
30.84
44.16
34.49
28.35
34.21
36.90
28.09
15.62
19.85
24.77
18.97
31.52
33.06
34.45
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
3
3
-OCOCH3
-OCOCH3
170.88
21.28
170.91
21.28
170.95
21.29
170.94
21.29
grounds for consideration of the reliability of the concluded
S : F populations.
Recording nuclear Overhauser effect (NOE) subtraction
spectra irradiating 26-, 27-, 28- or 30-methyl protons enabled
us to observe the H-18, H-19a spectra of 2—21 clearly
Methods and Results
enough to decide on the experimental J-value, i.e., J (H-
obs
We synthesized 3, 4, 12 and 13 described in previous pa- 18, 19a), of each compound. When the spectra could not be
3
,13)
pers,
and also 16 derivatives, 2, 5—11 and 14—21. C-29 analyzed as being of the 1st order, the ACD/HNMR spectral
functional groups of these model compounds were designed simulation program was utilized to identify definite J-values
to allow us to reveal the effect on the S : F population of C-29 (Table 3). The Jobs (H-18, 19a) varied widely among 2—21,
functionality, i.e., whether the functional groups are trigonal converging to approximately 1.5—1.7 Hz for 2 and 3 and
or tetrahedral and the identity of the substituents linking to 13.1—13.3 for 19—21. The values were beyond the smallest
the C-29 carbon atoms.
value (2.0 Hz) and the largest value (13.0 Hz) possible from
Assignments of carbon chemical shifts of 2—21 were Eq. 1 below.
done by comparing the shift data with those of the related
3
,13)
J(vic)ϭ7Ϫcos fϩ5 cos 2f
(1)
(2)
compounds
and recording distortion enhancement by po-
1
larization transfer (DEPT) and H-detected heteronuclear
J(vic)ϭAϪB cos fϩC cos 2f
1
multiple bond correlation (HMBC) spectra. H-noise decou-
1
3
pled C spectra of several compounds, which were derived in which f is a dihedral angle and parameters corresponding
1
3
14)
from biosynthetically C-enriched species of bryonolic acid, to A, B and C in the Karplus equation (Eq. 2) were decided
1
5)
were also utilized to discriminate signals appearing close to empirically to fit the cyclic compound by Bothner-By.
each other, such as those of C-10 to C-13 and C-4 to C-13 of Based on these findings, as well as the steric energy differ-
3
,13)
6
, 16, 17 and 19.
In order to corroborate the C-10 and C- ences described below, we concluded that 2 takes only the F-
1
3 signals of 10, a lanthanide shift reagent ([Eu(fod) ]) was form, and 20 and 21 only the S-form in CDCl solutions.
3
3
3
)
used. Assignment of the methyl proton signals was done
based on hetero-correlation spectroscopy ( H– C COSY) ular mechanics calculation method. To test whether the
spectral measurement. The results are given in Tables 1 and MM2 calculation program would be suitable for our model
Conformational studies have frequently utilized the molec-
1
13
16)
2
.
compounds, the most favorable conformations calculated for

April 2002
469
Table 1. (Continued)
Carbon
12
13
14
15
16
17
18
19
20
21
1
2
3
4
5
6
7
8
9
34.76
24.23
80.94
37.77
50.89
19.15
27.30
135.46
133.31
37.55
20.81
30.80
37.55
40.68
26.72
36.25
31.02
43.03
29.38
31.74
29.06
37.34
28.03
16.72
19.89
25.81
18.00
31.35
73.61
27.93
171.40
21.00
34.78
24.26
80.97
37.78
50.93
19.15
27.31
135.49
133.34
37.56
20.81
30.75
37.62
40.67
26.78
36.33
31.16
43.04
28.80
33.17
28.91
37.57
28.04
16.73
19.88
25.90
17.92
31.34
72.80
27.59
35.10
27.99
79.06
38.85
50.91
19.31
27.44
135.82
133.25
37.71
20.88
30.85
37.78
40.50
27.22
36.18
31.09
43.02
32.30
31.04
31.61
38.50
28.10
15.65
19.80
26.77
18.18
31.40
37.63
28.84
8.27
34.79
24.26
81.01
37.77
51.05
19.18
27.25
136.03
133.12
37.60
20.84
30.80
37.91
40.26
27.46
35.90
31.03
42.01
28.36
35.53
26.61
38.93
28.05
16.75
19.85
27.48
17.89
31.28
110.83
23.35
65.18
65.42
35.08
27.97
79.07
38.84
50.97
19.32
27.39
136.11
133.13
37.75
20.85
30.94
38.08
40.15
27.68
35.81
31.00
42.06
28.88
35.93
28.20
39.93
28.12
15.66
19.78
28.02
18.06
31.23
76.97
23.73
17.43
35.08
27.97
79.03
38.84
50.97
19.31
27.39
136.16
133.09
37.75
20.85
30.95
38.08
40.12
27.70
35.75
31.00
42.06
29.73
36.02
26.96
39.72
28.11
15.67
19.77
28.11
18.11
31.26
76.49
23.74
17.45
34.77
24.27
81.03
37.78
51.09
19.20
27.24
136.41
132.89
37.62
20.91
30.95
38.18
40.10
27.83
35.81
30.95
42.19
30.83
33.60
29.17
40.34
28.06
16.77
19.83
28.29
18.04
31.29
39.93
24.92
17.08
17.37
34.77
24.26
81.01
37.77
51.08
19.18
27.20
136.41
132.89
37.61
20.88
30.95
38.12
40.02
27.86
35.68
30.85
41.86
28.08
40.25
27.19
40.52
28.05
16.76
19.81
28.46
18.05
31.21
115.88
23.97
65.02
65.02
18.87
35.06
27.97
79.03
38.84
50.99
19.32
27.38
136.33
133.01
37.76
20.87
31.05
38.66
40.05
27.84
35.71
30.79
42.10
27.97
38.27
27.17
41.07
28.12
15.68
19.77
28.48
18.15
31.21
77.12
23.89
24.87
25.11
34.75
24.26
80.97
37.77
51.08
19.18
27.23
136.33
132.92
37.62
20.88
31.02
38.28
40.06
27.82
35.71
30.73
41.98
27.55
39.61
27.34
41.00
28.05
16.76
19.82
28.40
17.99
31.20
89.33
23.46
170.73
22.64
20.14
20.29
170.97
21.28
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
3
3
-OCOCH3
-OCOCH3
170.91
21.28
170.97
21.30
170.97
21.30
170.98
21.31
170.97
21.30
a) Data for 1 are not present because of extreme insolubility in CDCl (see the text).
3
1
Table 2. H Chemical Shifts of the Methyl Groups of Bryonolic Acid and Its Derivatives
Methyl
proton
1
2
3
4
5
6
7
8
9
10
11
23
24
25
26
27
28
30
0.99
0.78
0.95
0.96
0.88
1.03
1.24
0.87
0.86
0.97
0.96
0.74
1.04
0.94
0.87
0.83
0.94
0.95
0.85
1.03
1.22
0.99
0.80
0.95
0.95
0.75
1.03
1.17
1.00
0.80
0.94
0.94
0.72
1.07
1.04
0.87
0.86
0.97
0.97
0.71
1.05
1.09
1.00
0.80
0.94
0.95
0.68
1.09
1.05
0.99
0.80
0.95
0.99
0.91
1.08
0.98
0.87
0.87
0.97
0.99
0.90
1.08
0.98
0.87
0.87
0.97
1.04
0.86
1.12
1.05
1.00
0.80
0.96
1.06
0.99
1.07
0.96
Methyl
proton
1
2
13
14
15
16
17
18
19
20
21
23
24
25
26
27
28
30
0.88
0.88
0.98
1.11
0.93
1.12
1.00
0.88
0.87
0.98
1.11
0.96
1.13
0.99
1.00
0.80
0.96
1.14
0.95
1.13
0.91
0.88
0.88
0.98
1.17
0.95
1.15
1.01
1.00
0.81
0.96
1.19
0.96
1.17
0.98
1.01
0.81
0.96
1.19
0.95
1.17
0.96
0.89
0.88
0.98
1.19
0.94
1.16
0.90
0.89
0.88
0.98
1.20
0.95
1.17
1.07
1.01
0.81
0.96
1.21
0.96
1.18
1.03
0.89
0.88
0.99
1.21
0.95
1.18
1.03
the S-form of 12 and the F-form for 4 were compared with lar with respect to trends and amplitude of distortion and
those elucidated by X-ray crystallographic analyses (Fig. twisting; the distortion forced at the D–E rings of 4 to avoid
1
7)
3
). The theoretical and experimental structures were simi- the severe steric hindrance between the C-27 methyl group

4
70
Vol. 50, No. 4
Table 3. S : F Populations 2—21 in Equilibrium in CDCl Solutions Derived from Calculated Dihedral Angles, f (H18–C18–C19–H19a), Observed and
3
calc
Calculated J-Values, Jobs (H-18, 19a) and Jcalc (H-18, 19a), and C-26 Chemical Shifts
a)
Calculated dihedral angles (degree) and
S : F populations (%) derived from
b)
Observed
calculated J values (Hz)
Compound
J values
(
Hz)
S-form
F-form
Jobs (H-18, 19a)
C-26 chemical shift
c)
d)
e)
c)
d)
Jobs
fcalc
Jcalc
fcalc
Jcalc
PS (J)
PF (J)
Errors (%)
PS (CS) PF (CS)
f )
f )
g)
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1.5Ϯ0.2
1.7Ϯ0.2
2.0Ϯ0.2
2.5Ϯ0.2
2.7Ϯ0.2
2.9Ϯ0.2
3.2Ϯ0.3
3.2Ϯ0.3
5.4Ϯ0.3
7.1Ϯ0.3
8.6Ϯ0.2
9.0Ϯ0.2
10.2Ϯ0.2
11.8Ϯ0.3
12.6Ϯ0.3
12.7Ϯ0.2
12.9Ϯ0.3
13.1Ϯ0.3
13.3Ϯ0.2
13.3Ϯ0.4
176.8Ϯ2.0
176.5Ϯ2.0
176.5Ϯ2.0
177.3Ϯ2.0
176.0Ϯ2.0
176.3Ϯ2.0
176.4Ϯ2.0
176.4Ϯ2.0
177.4Ϯ2.0
176.8Ϯ2.0
176.4Ϯ2.0
176.5Ϯ2.0
177.4Ϯ2.0
176.6Ϯ2.0
176.8Ϯ2.0
176.8Ϯ2.0
177.1Ϯ2.0
176.1Ϯ2.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.0
13.5—13.1
13.5—13.0
80.0
1.7—1.3
2.0—1.1
2.1—1.1
2.0—1.1
2.2—1.2
2.4—1.2
2.1—1.1
2.1—1.1
2.6—1.4
1.8—1.1
1.9—1.1
1.8—1.1
1.9—1.1
1.8—1.1
1.9—1.1
2.0—1.1
2.0—1.1
1.8—1.1
2.0—1.1
1.7—1.1
0
1
4
8
9
100
99
96
92
91
91
86
86
68
52
39
36
26
12
5
—
0
0
0
0
4
100
100
100
100
96
97
88
88
69
60
42
41
27
15
6
79.2Ϯ5.0
79.0Ϯ5.0
79.5Ϯ5.0
77.7Ϯ5.0
75.3Ϯ5.0
78.8Ϯ5.0
78.8Ϯ5.0
73.0Ϯ5.0
83.2Ϯ5.0
81.9Ϯ5.0
83.1Ϯ5.0
80.9Ϯ5.0
82.9Ϯ5.0
81.2Ϯ5.0
80.3Ϯ5.0
80.2Ϯ5.0
82.0Ϯ5.0
79.7Ϯ5.0
84.3Ϯ5.0
Ϯ6
Ϯ6
Ϯ6
Ϯ6
Ϯ7
Ϯ6
Ϯ6
Ϯ5
Ϯ5
Ϯ4
Ϯ5
Ϯ4
Ϯ5
Ϯ5
Ϯ4
Ϯ5
Ϯ5
—
9
3
14
14
32
48
61
64
74
88
95
95
97
99
100
100
12
12
31
40
58
59
73
85
94
95
98
100
100
100
1
1
1
1
1
1
1
1
1
1
2
2
5
3
1
0
5
2
0
0
g)
176.0
176.4Ϯ2.0
0
Ϯ4
0
a) For errors, Ϯ2.0 and Ϯ5.0, see the text. b) Equations 3 and 3Ј were used to calculate the smallest and largest values, respectively. c) Populations of the S-form mole-
cules. The values were calculated using Eq. 5 or 7. The middle values of the upper limit and the smallest side are listed. d) Population of the F-form molecules. To calculate the
upper limit and the smallest side population, Eq. 4 was used; for the upper limit, conditions aϭthe maximum of the Jcalc for S-form, bϭthe maximum of the Jcalc for F-form and
cϭthe minimum of the Jobs, and for the smallest side, conditions aϭthe minimum of Jcalc for S-form, bϭthe minimum of the Jcalc for F-form and cϭthe maximum of the Jobs were
used. Besides these, conditions bϭc for 2 and aϭc for 20 were used. The middle values of the upper limit and the smallest side are listed. e) Differences between the middle val-
ues and the upper limit or smallest side of the populations. f ) Coupling constants (Hz) found in the spectrum measured with a shift reagent at Eu(fod) -D /substrate molar ratios
3
27
of approximately 0.10. g) Standards of the calculated dihedral angles: 80.0° for 2 in F-form and 176.0° for 20 in S-form were defined as corresponding to 1.3—1.7 Hz and
3.1—13.5 Hz, respectively.
1
and the C-29 functional group, and the twisting caused at the the conditions of 176.0° vs. 13.1 Hz and 80.0° vs. 1.3 Hz for
D–E rings of 12 to release the duplicated bow–stern interac- Eq. 3 and 176.0° vs. 13.5 Hz and 80.0° vs. 1.7 Hz for 3Ј were
tions owing to its boat–boat conformation, indicated that used.
MM2 calculation could be used in this study. This calcula-
Jcalc (maximum)ϭ6.9Ϫ1.3 cos fϩ5.3 cos 2f
Jcalc (minimum)ϭ6.6Ϫ1.0 cos fϩ5.5 cos 2f
(3)
tion yielded important information as described below.
Previously, Kikuchi and co-workers claimed that among
eight types of conformations possible for the D–E ring sys-
(3Ј)
tem of pachysandiol B (22) (Fig. 4), a D:A-friedooleanane where A, B and C in the Karplus equation were replaced by
class of triterpenes having a common D–E ring system to parameters convenient for our compounds. These equations
that of bryonolic acid derivatives, two types, the S-form and led us to estimate amount of the error in the S : F populations
F-form, would be far more stable than others based on con- (see experimental section).
sideration of steric hindrance by the Dreiding model. Aided
Calculation of the S : F populations was carried out as fol-
by CONFLEX, the MM2 calculation clearly indicated that lows: conformations with the minimal steric energy of three
the occurrence of conformers other than the S-form and F- possible rotamers around the C(20)—C(29) bond of both the
form in the solutions, if present, would be negligible; they S-form and F-form conformers of 2—21 were found, and
were shown to be less stable by at least 6 kcal/mol than the S- the resulting structural data were searched for their dihedral
form and F-form conformers.
angles, f
(H18–C18–C19–H19a). They were then con-
calc
Differentials of the steric energy between the two con- verted to the theoretical J-values, J (H-18, 19a), and the
calc
formers [DEϭE (S-form)ϪE (F-form)], 3.51 kcal/mol of 2, values derived from the rotamer giving the least strain energy
Ϫ4.02 kcal/mol of 20 and Ϫ3.72 kcal/mol of 21, supported among others were used to calculate S : F populations. Calcu-
the conclusion described above, that 2, 20 and 21 take only lation of the relative populations was done using Eqs. 4 and
the F-form or S-form conformations.
5:
According to these conclusions, 2 and 20 were treated as
standard, which enabled us to find correspondence of the the-
population (%) of F-form moleculesϭ(aϪc)/(aϪb)ϫ100
(4)
oretical dihedral angles, f (H18–C18–C19–H19a), to the
calc
population (%) of S-form molecules
effective J-values between H-18 and H-19a, i.e., 80.0° vs.
ϭ100Ϫthe population (%) of F-form molecules.
(5)
1
.5Ϯ0.2 Hz for 2 and 176.0° vs. 13.3Ϯ0.2 Hz for 20. These
values were used to decide the parameters of Eqs. 3 and 3Ј; where “a” and “b” are theoretical and F-form J-values, Jcalc

April 2002
471
barely affected J-values when the dihedral angles were close
to 180°, we took the Jcalc (H-18, 19a)s of 2—19 in S-form as
being close to that of 20 and 21, i.e., 13.3 Hz. As to the F-
form, converting two-dimensional C-29 functional groups of
2
—10 into three-dimensional ones of 11—21 led to wider
calculated dihedral angles as expected; e.g., 4 (–COOCH3:
9.0) vs. 15 (–CH(ϭO C H ): 82.9), 2 (–CHO: 80.0) vs. 13
7
2
2
4
(
–CH OH: 83.1), 6 (–COCH : 77.7) vs. 16 (–CH(OH)CH :
2
3
3
8
1.2) and 8 (–CHϭCH : 78.8) vs. 14 (–CH CH : 80.9). How-
2 2 3
ever, replacement of the hydrogen atom at C-29 with a
methyl group definitely made the angles narrower, e.g., 5
(
–CHϭNNH : 79.5) vs. 7 (–C(ϭNNH )CH : 75.3), 2
2
2
3
(
–CHO: 80.0) vs. 6 (–COCH : 77.7), 8 (–CHϭCH : 78.8) vs.
3
2
1
8
0 (–C(ϭCH )CH : 73.0), 11 (–CH : 83.2) vs. 14 (–CH CH :
2 3 3 2 3
0.9) vs. 18 (–CH(CH ) : 80.2), 13 (–CH OH: 83.1) vs. 17
3
2
2
(
(
–CH(OH)CH : 80.3) vs. 20 (–C(OH)(CH ) : 79.7) and 15
3
3
2
–CH(ϭO C H ): 82.9) vs. 19 (–C(CH )(ϭO C H ): 82.0).
2
2
4
3
2
2
4
These results suggest that in amplitude, the dihedral angles
in question do not simply increase in parallel with the bulki-
ness of the C-29 functional groups. Taking these findings
into account, errors of the angles calculated were estimated
within plus or minus 2° for the S-form and 5° for the F-form
1
8)
(see Table 3).
Besides the Jobs (H-18, 19a), chemical shifts of several nu-
clei around the D-ring also varied significantly among the
model compounds, 2—21 (see Tables 1, 2). Because these
nuclei are far away from the C-29 functional groups, the shift
Fig. 3. Comparison of Endocyclic Dihedral Angles (Degree) of the D–E data were expected to offer additional information on the
Ring Systems Calculated by MM2 and X-Ray with Respect to the S-Form
and F-Form Conformers, 4 and 12, Respectively
S : F populations. Structural data from the MM2 calculation
indicated that changes in the conformations at the D–E ring
moieties due to changes in the functional groups at C-29 are
extremely small in the S-form conformers, but significant in
the F-form conformers. The changes are most likely to occur
at the C-19—C-21 of their E-rings, but those around the D-
rings are minor (see Fig. 5). This led us to assume that chem-
ical shifts of the nuclei around the D-rings proper to both
conformers of the model compounds are identical to those of
2
and 20; e.g., 22.33 ppm to the C-26 of the F-form conform-
ers and 28.48 ppm to those of the S-form conformers. Ac-
cording to these, the S : F populations were calculated from
the C-26 shift data by Eqs. 6 and 7 in comparison with those
from the J-values.
population (%) of F-form moleculesϭ(28.5Ϫd)/(28.5Ϫ22.3)ϫ100 (6)
Fig. 4. Structures of D:A-Friedooleananes
population (%) of S-form molecules
ϭ100Ϫthe population (%) of F-form molecules
(7)
(
H-18, 19a) of the S-form and F-form conformers in equilib-
rium, respectively, and “c” is an experimental J-value, J
where “d” is the observed C-26 chemical shift.
obs
(
H-18, 19a), found from the NMR spectra of 2—21.
Calculation from other nuclei shift data, i.e., those of car-
Although different kinds of functional groups at C-29 may bons C-8 and C-15 and methyl protons at C-26, which varied
cause different amounts of twisting of the boat–boat struc- significantly among the model compounds, were also carried
tures of the D–E rings in S-form, the calculated dihedral an- out. The results agree well with that from C-26, supporting
gles, f (H18–C18–C19–H19a), of the model compounds the conclusion described above. The C-26 methyl proton
calc
in S-form were equivalent to each other with values close to chemical shifts (0.96 ppm) of 1, which prevented us from de-
1
80° (see Table 3). Studies using a Dreiding model revealed termining both its Jobs (H-18, 19a) and C-26 chemical shift
that the D–E rings of the model in boat–boat conformation because of the extreme insolubility in CDCl , indicated that 1
3
could freely twist, allowing the dihedral angles in question to took solely the F-form conformation, as did 2.
remain almost constant at approximately 180°. This led us to
consider the dihedral angles of their S-form conformers as Discussion and Conclusion
practically the same as that of 20 (176.0°). Since the Eqs. 3
Relative populations calculated from independent spectral
and 3Ј indicated that small changes in the dihedral angles data, i.e., J-values and chemical shifts, are consistent with

4
72
Vol. 50, No. 4
class of triterpenes (Fig. 4), have structures analogous to bry-
onolic acid derivatives at the C–E ring moiety, MM2 calcula-
tion suggested that it tended to have more S-form conformers
than the bryonolic acid derivatives; the steric energy differ-
ence between the S-form and F-form conformers, [DEϭE
(
1
S-form)ϪE (F-form)] calculated for these compound were
.9, 2.0, 0.0, Ϫ0.17, Ϫ1.15 and Ϫ5.0 kcal/mol, respectively.
These differences were significantly smaller than 3.5, 3.4,
.3, 0.28, Ϫ0.05 and Ϫ4.0 kcal/mol compared with those of
1
the bryonolic acid derivatives 2, 3, 11, 13, 16 and 20 corre-
sponding to 23—28, suggesting that Friedelin derivatives
could not solely adopt the F-form in solutions.
This study has shown that by selecting appropriate func-
tional groups at C-29, various types of bryonolic acid deriva-
tives can be designed with respect to relative populations of
the S-form and F-form conformers. We suggest that bryono-
lic acid is an excellent starting material for preparing a series
of experimental compounds which vary from each other
stepwise over wide ranges with respect to relative popula-
tions between two components in solution in equilibrium.
Experimental
The melting points were measured using a Yanagimoto micromelting
point apparatus and were not corrected. Optical rotations were measured at
room temperature with a JASCO P-1020 digital polarimeter. H- and C-
NMR spectra were obtained on a JEOL GSX-500 spectrometer operating at
Fig. 5. Endocyclic Dihedral Angles (Degree) of the D–E Ring Systems
Calculated by MM2 with Respect to the S-Form and F-Form Conformers, 2
and 21, Respectively
1
13
5
00 MHz for protons and 125 MHz for carbon. Homo decoupling for 2—7
each other, indicating that the results are reliable. Studying and NOE experiments were carried out on the JEOL ECP-500 spectrometer
the conformational situation of the D–E ring moiety of bry- operating at 500 MHz for protons. Spectra were recorded on 50—30
Ϫ1
mg ml concentration solutions with a probe temperature of 35 °C in
onolic acid derivatives and then calculating MM2 enabled us
CDCl , and referred to tetramethylsilane (TMS). Eu(fod) -D for Lan-
thanide-induced shift measurements were purchased from E. Merck, and no
further purification or drying of the reagents was attempted before use. Typi-
3
3
27
to evaluate the reliability of the S : F populations derived
from the J-values, offering much more detailed information
on the steric features of bryonolic acid derivatives than those cal conditions for proton spectral measurements were 40—120 transients
3
)
previously reported.
and a spectral width of 4000 Hz with 32 K data points. Proton-decoupled
carbon spectra were obtained under typical conditions: 5000—10000 tran-
sients, 21 ms pulse delay and a spectral width of 32000 Hz with 64 K data
points. In the case of the homo decoupling spectra, 6000 transients, a spec-
tral width of 7500 Hz with 64 K data points and zero filled to 256 K to give a
The principal factor determining the S : F populations was
whether the functional groups at C-29 are trigonal or tetrahe-
dral: the typical values of populations of the “trigonal
group,” 2—10, were 0 : 100—32 : 68, and those of the “tetra- digital resolution of 0.025 Hz. NOE conditions were 6000 transients, a spec-
tral width of 7500 Hz with 16 K data points and zero filled to 64 K to give a
hedral group,” 11—21, were 48 : 52—100 : 0. For the “trigo-
nal group,” substituting a methyl group for a hydrogen atom
on the C-29 functional groups of 2 and 9 led to an increase in
1
13
digital resolution of 0.1 Hz. Two-dimensional (2D) NMR spectra ( H– C
1
1
COSY, H– H COSY and HMBC) were performed on the JEOL GSX-500
spectrometer using standard JEOL CHSHF, DQFN and HMBC pulse se-
S-form conformers by 9 and 18. For the “tetrahedral group,” quences. MS and HR-MS were taken on a JEOL JMS DX-300 mass spec-
the tendency toward S-form conformations increased step- trometer. Unless otherwise specified, MS spectra were recorded under elec-
tron impact (EI) at 70 eV. ACD/HNMR predictor was purchased from Ad-
vanced Chemistry Development, Inc. The conformational analysis was car-
ried out using CAChe WorkSystem version 4.1 software (Fujitsu Co., Ltd.).
For every compound (2—21), rotational analysis was performed with incre-
wise as the number of substituents linking to C-29 increased;
S : F populations of the “methyl group” 11, were 48 : 52;
“monosubstituted group” 12—14, 61 : 39—74 : 26; “disubsti-
tuted group” 15—18, 88 : 12—97 : 3; and the “trisubstituted mental changes in the dihedral angle C19–C20–C29–X (XϭH, alkyl, OH or
group” 19—21, 99 : 1—100 : 0. These results serve as an OAc) or C19–C20–C29ϭX (XϭO or N) in 2 steps from 60° to Ϫ180° to
CONFLEX.
index for predicting the S : F populations of new compounds
with functional groups at C-29. Steric difference at the C-29
of anomeric 16 and 17 little affected their S : F populations.
The significant difference between the C-26 chemical
Bryonolic acid was isolated as crystals from the CHCl extract of cultured
cells of Luffa cylindrica, and its derivatives, 3, 4, 12 and 13 were prepared
by reported methods.
3
3)
ϩ
3: mp 221—222 °C. [a] 5.4° (cϭ1.0, CHCl ). EI-MS m/z: 440 (M ),
D
3
ϩ
ϩ
ϩ
shifts of the S-form and F-form conformers may be inter- 425 (M ϪCH
), 395 (M ϪCOOH), 243 (C18
H
27 containing A, B and C
3
ϩ
2
ϩ
rings), 235 (C H O containing D and E rings), 231 (C H containing A
preted as arising from the g-gauche effect. In addition to
three protons, H-7b, H-12b and H-18, located at the g-
gauche positions of the 26-methyl group, the F-form con-
15 23
17 27
and B rings, base peak), 191. High resolution (HR)-EI-MS m/z: 440.3649
Calcd for C H O : 440.3654).
(
3
0
48
2
ϩ
4
: mp 196—197 °C. [a]D 9.8° (cϭ1.0, CHCl ). EI-MS m/z: 470 (M ,
3
ϩ ϩ ϩ
former has an additional g-gauche proton, H-16b, which base peak), 455 (M ϪCH ), 452 (M ϪH O), 437 (M ϪCH ϪH O), 411
3
2
3
2
ϩ
ϩ
(
(
M ϪCOOCH ), 259 (C H O containing A, B and C rings), 249
3 18 27
ϩ ϩ
may cause significant upfield C-26 chemical shifts of the F-
form conformers.
Bryonolic acid was found to have a unique character, giv-
C H O containing D and E rings), 247 (C H O containing A and B
1
6
25
2
17 27
rings), 189. HR-EI-MS m/z: 470.3759 (Calcd for C H O : 470.3760).
3
1
50
3
ϩ
1
2: mp 212—213 °C. [a] 24.3° (cϭ1.0, CHCl ). EI-MS m/z: 526 (M ),
D 3
ϩ ϩ ϩ
ing derivatives with S : F populations of 0 : 100—100 : 0. Al- 511 (M ϪCH ), 466 (M ϪCH COOH), 453 (M ϪCH OCOCH ), 301
3
3
2
3
ϩ
ϩ
though Friedelin derivatives (23—28), D:A-friedooleanane (C20
H
O
containing A, B and C rings), 289 (C19
H
O
containing A and B
29
2
29
2

April 2002
473
ϩ
2
ϩ
17 27
rings), 263 (C H O containing D and E rings), 203 (base peak), 189. HR- B rings), 231 (C H containing D and E rings, base peak), 189. HR-EI-MS
17
27
EI-MS m/z: 526.4019 (Calcd for C H O : 526.4022).
m/z: 494.4124 (Calcd for C H O : 494.4124).
21: mp 194—195 °C. [a] 38.9° (cϭ1.0, CHCl ). EI-MS m/z: 554 (M ),
D 3
ϩ ϩ ϩ
3
4
54
4
34 54
2
ϩ
ϩ
1
3: mp 245—246 °C. [a] 34.0° (cϭ1.0, CHCl ). EI-MS m/z: 484 (M ),
D
3
ϩ
ϩ
ϩ
4
3
69 (M ϪCH ), 453 (M ϪCH OH, base peak), 424 (M ϪCH COOH), 539 (M ϪCH ), 494 (M ϪCH COOH), 434 (M Ϫ2CH COOH), 301
3 2 3 3 3 3
ϩ ϩ ϩ ϩ
01 (C H O containing A, B and C rings), 289 (C H O containing A (C H O containing A, B and C rings), 289 (C H O containing A and B
20
29
2
19 29
2
20 29
2
19 29
2
ϩ
ϩ
and B rings), 221 (C H O containing D and E rings), 189. HR-EI-MS rings), 231 (C H O containing D and E ringsϪCH COOH, base peak),
15
25
19 31
2
3
m/z: 484.3908 (Calcd for C H O : 484.3916).
189. HR-EI-MS m/z: 554.4332 (Calcd for C H O : 554.4335).
3
2
52
3
36 58 2
Compound 2 Chromium(VI) oxide (2.2 g, 22 mmol) was added step-
wise to a solution of 13 (630 mg, 1.2 mmol) in pyridine (22 g, 178 mmol)
Compound 11 A solution of a mixture of 2 (20 mg, 0.04 mmol), dieth-
ylene glycol (50 ml), hydrazine hydrate (2 ml) and potassium hydroxide (3 g)
under stirring and ice-cooling, and the mixture was stirred for another 1 h at in a mixture of benzene (5 ml) and methanol (10 ml) was heated for 4 h at
room temperature. The reaction mixture was poured into 5% HCl containing 190 °C. The reaction mixture was poured into 5% HCl containing ice and
ice, then shaken with benzene. The combined organic layer was washed with shaken with benzene. The combined organic layer was washed with H O and
2
H O and evaporated to give a residue, which was purified by recrystalliza- evaporated to give a residue, which was purified by recrystallization from
2
tion from methanol. mp 216—217 °C. [a] 25.7° (cϭ1.0, CHCl ). EI-MS methanol. mp 185—186 °C. [a] 28.2° (cϭ0.7, CHCl ). EI-MS m/z: 426
D
3
D
3
ϩ
ϩ
ϩ
ϩ
ϩ
ϩ
ϩ
ϩ
m/z: 482 (M , base peak), 467 (M ϪCH ), 453 (M ϪCHO), 422 (M Ϫ (M ), 411 (M ϪCH ), 393 (M ϪCH ϪH O), 259 (C H O containing A,
3
3
3
2
18 27
ϩ
2
ϩ
2
ϩ
ϩ
CH COOH), 301 (C H O containing A, B and C rings), 289 (C H O
containing A and B rings), 219 (C H O containing D and E rings), 189. taining D and E rings), 189, 44 (base peak). HR-EI-MS m/z: 426.3859
B and C rings), 247 (C H O containing A and B rings), 205 (C H con-
3
20 29
19 29
17 27 15 25
ϩ
1
5
23
HR-EI-MS m/z: 482.3753 (Calcd for C H O : 482.3760).
(Calcd for C H O: 426.3862).
3
2
50
3
30 50
Compound 5 A solution of a mixture of 2 (20 mg, 0.04 mmol), diethyl-
Compound 14 A solution of a mixture of 6 (20 mg, 0.04 mmol), dieth-
ene glycol (10 ml), hydrazine hydrate (3 ml) and potassium hydroxide (1 g) ylene glycol (50 ml), hydrazine hydrate (2 ml) and potassium hydroxide (3 g)
in a mixture of benzene (4 ml) and methanol (10 ml) was heated in a water in a mixture of benzene (5 ml) and methanol (10 ml) was heated for 4 h at
bath for 2 h at 60 °C. The reaction mixture was poured into 5% HCl contain- 190 °C. The reaction mixture was treated as described above to give pure 14.
ϩ
ing ice and shaken with benzene. The combined organic layer was washed mp 192—193 °C. [a] 45.0° (cϭ0.7, CHCl ). EI-MS m/z: 440 (M ), 425
D
3
ϩ
ϩ
ϩ
ϩ
with H O and evaporated to give a residue, which was purified by recrystal- (M ϪCH ), 411 (M ϪCH CH ), 407 (M ϪCH ϪH O), 259 (C H O
lization from methanol. mp 241—243 °C. [a]D 31.7° (cϭ0.8, CHCl ). Posi- containing A, B and C rings), 247 (C H O containing A and B rings), 219
tive-ion FAB-MS (3-nitrobenzyl alcohol) m/z: 455 (M ϩ1), 438 (M Ϫ (C H containing D and E rings, base peak), 189. HR-EI-MS m/z:
2
3
2
3
3
2
18 27
ϩ
3
17 27
ϩ
ϩ
ϩ
27
1
6
ϩ
ϩ
ϩ
NH ), 436 (M ϪH O), 421 (M ϪCH ϪH O), 259 (C H O containing 440.4021 (Calcd for C H O: 440.4018).
2
2
3
2
18 27
31 52
ϩ
A, B and C rings), 247 (C H O containing A and B rings), 189. HR-EI-
Compound 15 A solution of a mixture of 2 (12 mg, 0.025 mmol), p-
1
7
27
MS m/z: 454.3930 (Calcd for C H N O: 454.3923).
TsOH·H O (2 mg, 0.01 mmol) and ethylene glycol (100 mg, 1.61 mmol) in
3
0
50
2
2
Compound 6 Chromium(VI) oxide (1 g, 10 mmol) was added stepwise benzene was refluxed for 10 h. The reaction mixture was poured into satu-
to a solution of 3-acetyl-29-methyl bryonyl alcohol (70 mg, 0.14 mmol) in rated aqueous NaHCO solution and shaken with benzene. The combined
3
pyridine (10 g, 126 mmol) under stirring and ice-cooling, and the mixture organic layer was washed with H O and evaporated to give a residue, which
2
was stirred for another 4 h at room temperature. The reaction mixture was was purified by recrystallization from methanol. mp 238—239 °C. [a]
D
ϩ
ϩ
ϩ
treated as descried for 2 above to give pure 6. mp 196—197 °C. [a] 45.8° 42.0° (cϭ0.5, CHCl ). EI-MS m/z: 526 (M ), 511 (M ϪCH ), 466 (M Ϫ
D
3
3
ϩ
ϩ
ϩ
ϩ
ϩ
ϩ
(
cϭ0.3, CHCl ). EI-MS m/z: 496 (M ), 481 (M ϪCH ), 453 (M Ϫ CH COOH), 453 (M ϪC H O ; 1.3-dioxolan-2-yl), 301 (C H O con-
3 3 3 3 5 2 20 29 2
ϩ ϩ ϩ
COCH ), 436 (M ϪCH COOH), 301 (C H O containing A, B and C taining A, B and C rings), 289 (C H O containing A and B rings), 263
3
3
20 29
2
19 29
2
ϩ
ϩ
ϩ
ϩ
rings), 289 (C H O containing A and B rings), 233 (C H O containing (C H O containing D and E rings), 189, 73 (C H O ; 1,3-dioxolan-2-yl,
19
29
2
16 25
17 27
2
3
5
2
D and E rings), 189, 44 (base peak). HR-EI-MS m/z: 496.3915 (Calcd for base peak). HR-EI-MS m/z: 526.4024 (Calcd for C H O : 526.4022).
3
4
56
4
C H O : 496.3916).
Compound 16, 17 and 3-Acetyl-29-methyl Bryonyl Alcohol A solu-
33
52
3
Compound 7 A solution of a mixture of 6 (20 mg, 0.04 mmol), diethyl- tion of MeMgI in Et O, prepared from Mg (144 mg, 6.0 mmol) and MeI
2
ene glycol (10 ml), hydrazine hydrate (2 ml) and potassium hydroxide (0.5 g) (855 mg, 6.0 mmol), was added dropwise to a solution of 2 (100 mg, 0.21
in a mixture of benzene (4 ml) and methanol (10 ml) was heated for 2 h at mmol) in Et O under gentle reflux. After the mixture was stirred for 1 h at
2
5
0 °C. The reaction mixture was treated as described above to give pure 7.
room temperature, the reaction was quenched with H O. The reaction mix-
2
ϩ
mp 211—212 °C. [a] 79.6° (cϭ0.4, CHCl ). EI-MS m/z: 468 (M ), 453 ture was poured into 5% HCl and shaken with benzene. The combined or-
D
3
ϩ
ϩ
ϩ
ϩ
(
M ϪCH ), 452 (M ϪNH ), 450 (M ϪH O), 435 (M ϪCH ϪH O), 259 ganic layer was washed with H O and evaporated to give a residue. 16 and
3 2 2 3 2 2
ϩ ϩ
(C H O containing A, B and C rings), 247 (C H N containing D and E 17 were isolated by silica gel column chromatographic treatment and then
1
8
27
16 27
2
ϩ
rings and C H O containing A and B rings), 189, 44 (base peak). HR-EI-
MS m/z: 468.4079 (Calcd for C H N O: 468.4079).
recrystallized from methanol.
16: mp 232—233 °C. [a] 49.4° (cϭ0.2, CHCl ). EI-MS m/z: 456 (M ),
17
27
ϩ
3
1
52
2
D
3
1
9)
ϩ
ϩ
ϩ
ϩ
Compound 8 According to the Cainelli procedure based on the Wit- 441 (M ϪCH ), 438 (M ϪH O), 423 (M ϪCH ϪH O), 411 [M Ϫ
3
2
3
2
ϩ
tig reaction, under an argon atmosphere, methylene iodide (66 mg, CH(OH)CH , base peak], 259 (C H O containing A, B and C rings), 247
0
3
18 27
ϩ
ϩ
.24 mmol) in Et O was added to a solution of 2 (100 mg, 0.22 mmol) and (C H O containing A and B rings), 235 (C H O containing D and E
2
1
7
2
7
1
6
2
7
magnesium amalgam (8 g) in Et O, and the entire mixture was stirred under rings), 189. HR-EI-MS m/z: 456.3970 (Calcd for C H O : 456.3967).
2
31 52
2
ϩ
reflux for 3 h. The reaction mixture was poured into 5% HCl and shaken
17: mp 232—233 °C. [a] 48.7° (cϭ0.6, CHCl ). EI-MS m/z: 456 (M ),
D 3
ϩ ϩ ϩ ϩ
with benzene. The combined organic layer was washed with H O and evapo- 441 (M ϪCH ), 438 (M ϪH O), 423 (M ϪCH ϪH O), 411 [M Ϫ
2
3
2
3
2
ϩ
rated to obtain the residue. This residue was purified by HPLC and then re- CH(OH)CH , base peak], 259 (C H O containing A, B and C rings), 247
3
18 27
ϩ
ϩ
crystallized from methanol. mp 168—169 °C. [a] 20.0° (cϭ0.1, CHCl ). (C H O containing A and B rings), 235 (C H O containing D and E
D
3
17 27
16 27
ϩ
ϩ
ϩ
EI-MS m/z: 438 (M , base peak), 423 (M ϪCH ), 420 (M ϪH O), 405 rings), 189. HR-EI-MS m/z: 456.3974 (Calcd for C H O : 456.3967).
3
2
31 52
2
ϩ
ϩ
(
(
M ϪCH ϪH O), 259 (C H O containing A, B and C rings), 247
C H O containing A and B rings), 217 (C H containing D and E PtO (3 mg) in methanol was stirred at room temperature under H atmo-
Compound 18 A suspension of a mixture of 10 (10 mg, 0.02 mmol) and
3
2
18 27
ϩ
ϩ
1
7
27
16 25
2
2
rings), 189. HR-EI-MS m/z: 438.3861 (Calcd for C H O: 438.3862).
sphere for 2 h. The PtO catalyst was filtered off and the filtrate was evapo-
2
3
1
50
Compound 9 This compound was prepared by acetylation of 8 with rated to give a residue, then the residue was recrystallized from methanol to
acetic anhydride in pyridine, followed by recrystallization from methanol. give pure 18. mp 265—267 °C. [a] 50.5° (cϭ0.5, CHCl ). EI-MS m/z: 496
D
3
ϩ
ϩ
ϩ
ϩ
ϩ
mp 181—182 °C. [a] 18.2° (cϭ0.5, CHCl ). EI-MS m/z: 480 (M , base
(M ), 481 (M ϪCH ), 453 [M ϪCH(CH ) , base peak], 436 (M Ϫ
peak), 465 (M ϪCH ), 453 (M ϪCHϭCH ), 420 (M ϪCH COOH), 301 CH COOH), 301 (C H O containing A, B and C rings), 289 (C H O
D
3
3
3 2
ϩ
ϩ
ϩ
ϩ
2
ϩ
2
3
2
3
3
20 29
19 29
ϩ
ϩ
ϩ
(
C H O containing A, B and C rings), 289 (C H O containing A and B containing A and B rings), 233 (C H containing D and E rings), 189. HR-
20 29 2 19 29 2 17 29
ϩ
rings), 217 (C H containing D and E rings), 189. HR-EI-MS m/z: EI-MS m/z: 496.4285 (Calcd for C H O : 496.4280).
16
25
34 56
4
4
80.3961 (Calcd for C H O : 480.3967).
Compound 19 A solution of a mixture of 6 (10 mg, 0.02 mmol), p-
3
3
52
2
Compound 10 and 21 Acetylation of 20 with acetic anhydride in pyri- TsOH·H O (2 mg, 0.01 mmol) and ethylene glycol (100 mg, 1.61 mmol) in
2
dine gave a mixture of 10 and 21. After treatment in the usual manner, these benzene was refluxed for 16 h. The reaction mixture was treated as described
compounds were isolated by HPLC, then purified by recrystallization from above to give pure 19. mp 267—268 °C. [a] 50.5° (cϭ0.5, CHCl ). EI-MS
D
3
ϩ
ϩ
ϩ
ϩ
methanol.
1
4
m/z: 540 (M ), 525 (M ϪCH ), 480 (M ϪCH COOH), 454 (M Ϫ
3 3
ϩ ϩ
ϩ
0: mp 219—220 °C. [a] 11.2° (cϭ1.0, CHCl ). EI-MS m/z: 494 (M ), C H O ; 2-methyl-1,3-dioxolan-2-yl), 301 (C H O containing A, B and
D 3 4 7 2 20 29 2
ϩ ϩ ϩ ϩ ϩ
79 (M ϪCH ), 453 [M ϪC(ϭCH )CH ], 434 (M ϪCH COOH), 301 C rings), 289 (C H O containing A and B rings), 189, 87 (C H O ; 2-
3 2 3 3 19 29 2 4 7 2
ϩ ϩ
(
C H O containing A, B and C rings), 289 (C H O containing A and methyl-1,3-dioxolan-2-yl, base peak). HR-EI-MS m/z: 540.4175 (Calcd for
20 29 2 19 29 2

4
74
Vol. 50, No. 4
C H O : 540.4178).
Compound 20 A solution of MeMgI in Et O, prepared from Mg (2 g,
iol., 25, 1571—1574 (1984); b) Kitajima J., Tanaka Y., Yakugaku
Zasshi, 109, 250—255 (1989); c) Cho H. J., Tanaka S., Fukui H.,
Tabata M., Phytochemistry, 31, 3893—3896 (1992); d) Akihisa T.,
Kokke W. C. M. C., Tamura T., Nambara T., Chem. Pharm. Bull., 40,
1199—1202 (1992); e) Cho H. J., Ito M., Tanaka S., Kamisako W.,
Tabata M., Phytochemistry, 33, 1407—1413 (1993); f ) Takeda T.,
Kondo T., Mizukami H., Ogihara Y., Chem. Pharm. Bull., 42, 730—
732 (1994); g) Kondo T., Inoue M., Mizukami H., Ogihara Y., Biol.
Pharm. Bull., 18, 726—729 (1995); h) Li Z., Luo L., Ling M., Zhong-
caoyao, 30, 409—411 (1999); i) Thatte U., Bagadey S., Dahanukar S.,
Cell Mol. Biol., 46, 199—214 (2000); j) Appendino G., Jakupovic J.,
Belloro E., Marchesini A., Fitoterapia, 71, 258—263 (2000).
3) Kamisako W., Suwa K., Honda C., Isoi K., Nakai H., Shiro M.,
Machida K., Magn. Reson. Chem., 25, 848—855 (1987).
35
56
4
2
8
3.0 mmol) and MeI (12 g, 84.0 mmol), was added dropwise to a solution of
4
(200 mg, 0.43 mmol) in Et O under gentle reflux. The mixture was then
2
treated as described for 16 and 17 to give pure 20. mp 260—261 °C. [a]
D
ϩ
ϩ
ϩ
5
3.0° (cϭ0.6, CHCl ). EI-MS m/z: 470 (M ), 455 (M ϪCH ), 452 (M Ϫ
3 3
ϩ ϩ ϩ
2
H O), 437 (M ϪCH ϪH O), 411 [M ϪC(OH)(CH ) ], 259 (C H O
3
2
3
2
18 27
ϩ
containing A, B and C rings, base peak), 247 (C H O containing A and B
1
7
27
ϩ
rings), 231 (C H O containing D and E ringsϪH O), 189. HR-EI-MS
17
29
2
m/z: 470.4129 (Calcd for C H O : 470.4124).
3
2
51
2
13
Preparation of C-Enriched Bryonolic Acid Fifteen days after the in-
1
3
oculation of L. cylindrica cells, 300 mg of sodium [2- C]acetate was added
to 500 ml flasks containing cells growing in 300 ml of suspension medium
Ϫ7
(
Linsmaier–Skoog medium supplemented with 3% glucose and 10 M 2,4-
D). After a subsequent incubation period of 6—7 d, the cells were harvested.
Dried and powdered cells were extracted with CHCl ; the CHCl solution
was then condensed to yield fine needles of C-enriched bryonolic acid. Re-
4) Tanaka S., Uno C., Akimoto M., Tabata M., Honda C., Kamisako W.,
Planta Med., 57, 527—530 (1991).
5) a) Tabata M., Tanaka S., Kamisako W., Jpn. Kokai Tokkyo Koho,
JP04290846 (1992) [Chem. Abstr., 94321 (1993)]; b) Tabata M.,
Tanaka S., Cho H. J., Uno C., Shimakura J., Ito M., Kamisako W.,
Honda C., J. Nat. Prod., 56, 165—174 (1993).
3
3
1
3
crystallization of the material from CHCl gave 100 mg (from eight flasks).
3
13
13
Preparation of C-Enriched Derivatives 2- C-enriched species, 10,
6, 17 and 19, were prepared in the same manner as the unlabeled com-
1
pounds.
Equations 3 and 3؅ Parameters A, B and C for Eqs. 3 and 3Ј were deter-
6) Nakada K., Naeshiro H., Jpn. Kokai Tokkyo Koho, JP06345630
(1994) [Chem. Abstr., 169722 (1995)].
mined to fit the following conditions:
For Eq. 3
7) Hanamura A., Sato A., Jpn. Kokai Tokkyo Koho, JP07109214 (1995)
[Chem. Abstr., 40736 (1995)].
8
)
Ogiwara Y., Morishige H., Kato A., Bando K., Jpn. Kokai Tokkyo
Koho, JP07252141 (1995) [Chem. Abstr., 45700 (1995)].
1.7 (Hz)ϭAϪB cos(80.0)ϩC cos(80.0ϫ2)
8.2 (Hz)ϭAϪB cos(32.1)ϩC cos(32.1ϫ2)
13.5 (Hz)ϭAϪB cos(176.0)ϩC cos(176.0ϫ2)
9) Ito M., Oosumi K., Kosugi N., Takagi H., Fukunaga I., Ogawa T., Jpn.
Kokai Tokkyo Koho, JP06157280 (1994) [Chem. Abstr., 141248
(
1994)].
and for Eq. 3Ј
10) Osawa E., Musso H., Angew. Chem., Int. Ed. Engl., 22, 1—12 (1983).
1
1) a) Goto H., Osawa E., J. Am. Chem. Soc., 111, 8950—8951 (1989); b)
Idem, J. Chem. Soc., Perkin Trans. 2, 1993, 187—198.
1
8
1
.3 (Hz)ϭAϪB cos(80.0)ϩC cos(80.0ϫ2)
.2 (Hz)ϭAϪB cos(32.1)ϩC cos(32.1ϫ2)
3.1 (Hz)ϭAϪB cos(176.0)ϩC cos(176.0ϫ2)
12) Masaki N., Niwa M., Kikuchi T., J. Chem. Soc., Perkin Trans. 2, 1975,
10—614.
1
6
3) Kamisako W., Suwa K., Morimoto K., Isoi K., Org. Magn. Reson., 22,
Equations 3 and 3Ј were used to determine the maximum and minimum the-
oretical Jcalc (H-18, 19a) of the S-form and F-form molecules for 2—21.
93—100 (1984).
14) Karplus M., J. Am. Chem. Soc., 85, 2870—2871 (1963).
The values 8.2 (Hz) and 32.1 given in the above equations came from the 15) Bothner-By A. A., “Advances in Magnetic Resonance,” Vol. 1, Acade-
observed J-value between H-18 and H-19b and the calculated dihedral angle
for the H18–C18–C19–H19b, respectively.
mic Press, New York, 1965.
16) Burkert U., Allinger N. L., “Molecular Mechanics,” American Chemi-
cal Society, Washington, DC, 1982.
Acknowledgements We are grateful to Dr. Masako Fujiwara (JEOL 17) a) Kamisako W., Isoi K., Nakai H., Shiro M., Acta Crystallogr., Sect.
Datum, Ltd.) for helpful discussion of the molecular mechanics calculation.
C, 40, 1013—1015 (1984); b) Nakai H., Shiro M., Kamisako W.,
Honda C., Isoi K., ibid., 43, 1779—1782 (1987); c) Nakai H., ibid.,
45, 1465—1467 (1989); d) Idem, ibid., 45, 1472—1473 (1989).
18) Liu S. Y., Purvis G. D., “CAChe Molecular Mechanics. Augmented
Force-Field,” CAChe Scientific draft copy, 1991.
References and Notes
1
)
)
a) Biglino G., Cattel L., Caputo O., Nobili G., Gazz. Chim. Ital., 99,
8
3
30—847 (1969); b) Sim K. Y., Lee H. T., Phytochemistry, 11, 3341—
343 (1972).
19) Bertini F., Grasselli P., Zubiani G., Cainelli G., Tetrahedron, 26,
1281—1290 (1970).
2
a) Kamisako W., Morimoto K., Makino I., Isoi K., Plant & Cell Phys-