DNA-damaging steroidal alkaloids from Eclipta alba from the suriname rainforest

Article abstract of DOI:10.1021/np970561c

Bioassay-guided fractionation of the MeOH extract of Eclipta alba using three yeast strains (1138, 1140, and 1353) resulted in the isolation of eight bioactive steroidal alkaloids (1-8), six of which are reported for the first time from nature. The major alkaloid was identified as (20S)(25S)-22,26- imino-cholesta-5,22(N)-dien-3β-ol (verazine, 3), while the new alkaloids were identified as 20-epi-3-dehydroxy-3-oxo-5,6-dihydro-4,5-dehydroyerazine (1), ecliptalbine [(20R)-20-pyridyl-cholesta-5-ene-3β,23-diol] (4), (20R)- 4β-hydroxyverazine (5), 4β-hydroxyverazine (6), (20R)-25β-hydroxyverazine (7), and 25β-hydroxyverazine (8). Ecliptalbine (4), in which the 22,26- imino ring of verazine was replaced by a 3-hydroxypyridine moiety, had comparable bioactivity to verazine in these assays, while a second alkaloid (8) showed good activity against Candida albicans. All the alkaloids showed weak cytotoxicity against the M-109 cell line.

Full text of DOI:10.1021/np970561c

1
202
J . Nat. Prod. 1998, 61, 1202-1208
DNA-Da m a gin g Ster oid a l Alk a loid s fr om Eclipta a lba fr om th e Su r in a m e
Ra in for est1
†
†
‡
§
⊥
⊥ ,|
Maged S. Abdel-Kader, Brian D. Bahler, Stan Malone, Marga C. M. Werkhoven, Frits van Troon, David,
∇
O
O
O
,†
J an H. Wisse, Isia Bursuker, Kim M. Neddermann, Stephen W. Mamber, and David G. I. Kingston*
Department of Chemistry, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061-0212,
Stichting Conservation International Suriname, Gravenstraat 17, Paramaribo, Suriname, The National Herbarium of
Suriname, P.O. Box 9212, Paramaribo, Suriname, Saramaka Village, Brownsweg, Kadju, Suriname, Bedrijf Geneesmiddelen
Voorziening Suriname, Commissaris Roblesweg, Geyersvlijt, Suriname, and Bristol-Myers Squibb, Pharmaceutical Research
Institute, 5 Research Parkway, Wallingford, Connecticut 06492-7660
Received December 16, 1997
Bioassay-guided fractionation of the MeOH extract of Eclipta alba using three yeast strains (1138, 1140,
and 1353) resulted in the isolation of eight bioactive steroidal alkaloids (1-8), six of which are reported
for the first time from nature. The major alkaloid was identified as (20S)(25S)-22,26-imino-cholesta-
5
,22(N)-dien-3â-ol (verazine, 3), while the new alkaloids were identified as 20-epi-3-dehydroxy-3-oxo-5,6-
dihydro-4,5-dehydroverazine (1), ecliptalbine [(20R)-20-pyridyl-cholesta-5-ene-3â,23-diol] (4), (20R)-4â-hydroxy-
verazine (5), 4â-hydroxyverazine (6), (20R)-25â-hydroxyverazine (7), and 25â-hydroxyverazine (8).
Ecliptalbine (4), in which the 22,26-imino ring of verazine was replaced by a 3-hydroxypyridine moiety,
had comparable bioactivity to verazine in these assays, while a second alkaloid (8) showed good activity
against Candida albicans. All the alkaloids showed weak cytotoxicity against the M-109 cell line.
In continuation of our search for anticancer and other
bioactive agents from the Suriname rainforest,2,3 an extract
of the leaves and bark of Eclipta alba (L.) Haask. (Aster-
aceae) was found to show reproducible activity in a mech-
anism-based bioassay utilizing genetically engineered mu-
tants of the yeast Saccharomyces cerevisiae.4 This plant
is known as “totobia” in the Saramacca language and
was thus selected for fractionation studies as a potential
DNA-damaging agent or as an antifungal agent. Frac-
tionation was carried out by a combination of liquid-liquid
partition and chromatography as described in the Experi-
mental Section and yielded the eight bioactive compounds,
1-8.
The major active compound was identified as verazine
“luisawiwiri” in Sranan Tongo; it is used in Surinamese
(
3) [(20S,25S)-22,26-iminocholesta-5,22(N)-dien-3â-ol] by
traditional medicine, and its collection was made as part
of an ethnobotanical collection strategy.5 E. alba is also
used in Indian traditional medicine for the treatment of
leukoderma, night blindness, and catarrhal jaundice; as an
emetic, a purgative, and an antiseptic; and to treat hepatic
and spleen enlargements. It is reported to improve hair
comparing its spectral data (see Experimental Section and
Table 2) with the literature data.7 Among flowering plants,
the occurrence of verazine is restricted to Veratrum species
8
in the family Liliaceae, although it has also been isolated
from the fern Zygadenus sibiricus.9 This is the first report
of the isolation of verazine from the Asteraceae. The (20R)
epimer of verazine (2) was isolated as a minor component
and identified by comparing its spectral data (see Experi-
6
growth and color.
7
Resu lts a n d Discu ssion
mental Section and Table 2) with the literature data.
4
Reduction of verazine (3) with NaBH resulted in the
formation of veramiline (9), identified by comparing its
A sample of the leaves of E. alba was subjected to
bioassay at Virginia Tech and found to be active against
the three yeast strains S. cerevisiae 1138, 1140, and 1353.
Activities in these assays are recorded as IC12 values, which
are the concentrations (in µg/mL) required to give an
inhibition zone 12 mm in diameter around a 100-µL well
in a 4-mm agar layer plated with the yeast strain in
question. Topoisomerase 1 inhibitors show inhibition of
strains 1138 and 1140 only, topoisomerase 2 inhibitors
show activity against strain 1138 only, while general DNA-
damaging agents and antifungal agents show activity
against all three strains. A methanol extract of E. alba
leaves gave IC12 values of 16, 26, and 9 µg/mL against the
strains 1138, 1140, and 1353, respectively, and this extract
spectral data (see Experimental Section) with the literature
10
13
data. Its C NMR spectrum and assignments (Table 2)
are reported for the first time.
Compound 1 was assigned the structure shown based
on a comparison of its 1 C NMR chemical shifts with those
3
of 3- and 6-oxo steroids1
1,12
and those of verazine (3), as
well as on the basis of the observation of HMBC correla-
tions from H-4 to C-2, C-6, and C10 at δ 34.0, 32.9, and
8.6, respectively. The H NMR chemical shift of CH -21
0.97 ppm indicated that 1 is the 20(R) epimer. The
C
1
3
3
at δ
H
2
0(S) epimer of compound 1 has previously been prepared
8
a
from 3, but it has not previously been isolated from a
natural source. The absolute stereochemistry of 1 and of
the other alkaloids described below is assigned on the basis
of their assumed relationships to verazine (3).
*
To whom enquiries should be addressed: Tel.: (540) 231-6570. Fax:
(
540) 231-7702. E-mail: dkingston@vt.edu.
†
Virginia Polytechnic Institute and State University.
Conservation International Suriname.
The National Herbarium of Suriname.
Saramaka Village.
Deceased April 11, 1994.
Bedrijf Geneesmiddelen Voorziening Suriname.
2
Compound 4 had a composition of C27H39NO as shown
‡
1
13
§
by HREIMS. The H and C NMR chemical shifts of its
steroidal skeleton (Tables 1 and 2) showed a close similarity
to those of 2 and 3, indicating it to be a member of the
same class of compounds. In contrast to the tetrahydro-
⊥
|
∇
O
Bristol-Myers Squibb, Pharmaceutical Research Institute.
1
0.1021/np970561c CCC: $15.00
© 1998 American Chemical Society and American Society of Pharmacognosy
Published on Web 09/25/1998

Bioactive Steroidal Alkaloids from Eclipta
J ournal of Natural Products, 1998, Vol. 61, No. 10 1203
Ta ble 1. Selected 1H NMR Data (δ values) for Compounds 1-9a
position
3
1
2
3
4
5
5a
3.48 (m)
3.05 (m)
3.38 (m)
3.52 (dt, J )
.3, 12.2 Hz)
4.09 (d, J )
.8 Hz)
5.62 (br d, J )
3.7 Hz)
4.73 (dt, J )
3.3, 12.2 Hz)
5.49 (d, J )
2.7 Hz)
5.80 (br d, J )
3.6 Hz)
3
4
6
5.71 (br s)
2
5.33 (br d, J )
5.32 (br d, J )
5.0 Hz)
5.22 (br d, J )
4.9 Hz)
5
.3 Hz)
1
1
2
8
9
1
0.74 (s)
1.16 (s)
0.97 (d, J )
0.70 (s)
0.98 (s)
0.97 (d, J )
7.8 Hz)
0.69 (s)
0.99 (s)
1.07 (d, J )
6.7 Hz)
0.73 (s)
0.92 (s)
1.14 (d, J ) 6.6)
0.69 (s)
1.15 (s)
0.96 (d, J )
6.9 Hz)
0.61 (s)
1.11 (s)
1.14 (d, J )
6.9 Hz)
6
.9 Hz)
2
2
2
3
5.11 (t, J )
3
.4)
2
2
2
4
6ax
6eq
6.79 (br s)
7.71 (br s)
2.92 (m)
3.69 (m)
2.93 (m)
3.69 (m)
2.96 (m)
3.65 (dd, J )
2.92 (m)
3.68 (m)
6
.7, 16.6 Hz)
2
7
0.91 (d, J )
0.90 (d, J )
6.6 Hz)
0.88 (d, J )
6.7 Hz)
2.12 (s)
0.90 (d, J )
6.6 Hz)
0.96 (d, J )
6.6 Hz)
6
.6 Hz)
CH
3
CO
2.00 (s), 2.06 (s),
2
.15 (s)
position
3
6
6a
7
7a
8
8a
9
3.55 (dt, J )
.6, 11.6 Hz)
4.11 (d, J )
.2 Hz)
5.65 (br d, J )
.5 Hz)
4.74 (dt, J )
3.6, 11.6 Hz)
5.49 (d, J )
2.8 Hz)
3.34 (m)
4.61 (m)
3.34 (m)
4.62 (m)
3.52 (m)
3
4
6
3
5.80 (m)
5.20 (d, J ) 5.48 (br d, J )
4.9 Hz) 5.0 Hz)
0.62 (s) 0.63 (s)
1.05 (s) 1.28 (s)
0.91 (d, J ) 1.18 (d, J ) Hz) 1.01 (d, J ) 1.11 (d, J )
6.9 Hz)
5.20 (d, J ) 5.37 (br d, J )
4.9 Hz) 4.8 Hz)
0.60 (s) 0.69 (s)
0.88 (s) 1.25 (s)
5.33 (br d, J )
4.7 Hz)
0.68 (s)
1.00 (s)
0.85 (d, J )
6.4 Hz)
3
1
1
2
8
9
1
0.69 (s)
1.17 (s)
0.67 (s)
1.13 (s)
1.12 (d, J )
5.7 Hz)
1.07 (d, J )
6
.6 Hz)
6.8 Hz)
Hz)
2
2
2
3
2.65 (br d, J )
1
1.4 Hz)
5.18 (m)
5.09 (t, J ) 3.1)
5.16 (t, J )
3
.2)
2
2
2
4
6ax
6eq
2.97 (m)
3.66 (m)
3.23 (m)
3.23 (m)
3.23 (m)
3.23 (m)
2.26 (m)
3.22 (br d, J )
1
1.4 Hz)
2
7
0.90 (d, J ) 6.9 Hz) 0.94 (d, J )
.6 Hz)
2.01 (s), 2.06 (s),
.14 (s)
0.91 (s)
1.00 (s)
1.08 (s)
1.02 (s)
1.03 (d, J )
6.7 Hz)
6
CH3CO
2.01 (s), 2.22 (s)
2.02 (s), 2.20 (s)
2
a
Measured in CDCl3; chemical shifts in ppm from internal TMS, coupling constants in Hz.
pyridine rings of compounds 1-3, however, compound 4
required by their composition. The presence of the ad-
ditional oxygen atom was confirmed by the formation of
the triacetyl derivatives 5a and 6a of both 5 and 6 upon
acetylation. HMQC and HETCOR experiments for 5 and
contained a substituted pyridine ring, as evidenced by the
appearance of two broad proton singlets at δ
ppm, a singlet for an aromatic methyl group at δ
H
6.79 and 7.71
2.12 ppm
), and its UV spectra in neutral and
H
(assigned to 27-CH
3
6 correlated the CH group at δ
C
77.2 ppm to the protons
alkaline media (λmax 290, 307 nm). The 13C NMR chemical
shifts for C-22 to C-26 (Table 2) and the presence of
diagnostic peaks at m/ z 136, 137, and 150 all indicated
at δ 4.09 (d, J ) 2.8 Hz) and 4.11 (d, J ) 3.2 Hz) in 5 and
H
6, respectively. The COSY correlation of these protons with
H-3 at 3.52 (dt, J ) 3.3, 12.2 Hz) in 5 and 3.55 (dt, J ) 3.6,
11.6 Hz) in 6 and the coupling pattern of H-3 and H-4
supported the assignment of the additional OH group to
C-4. HMBC correlations of H-4 with C-2, C-6, and C-10
1
3,14
the presence of a 3-hydroxy-5-methylpyridine moiety.
The position of the hydroxyl and methyl groups were
further confirmed by a NOEDIFF experiment in which
irradiation of CH
9
7
3
-27 at δ
% enhancements of the signals for H-26 and H-24 at δ
.71 and 6.79 ppm, respectively. The HMBC correlations
H
2.12 ppm resulted in 17 and
C
at δ 25.3, 128.3, and 36.0 ppm were in complete agreement
H
with that assignment.
1
The H NMR chemical shift of CH
3
-19 at δ
H
1.15 in 5
of H-24 and H-26 (Figure 1) were in complete agreement
with the proposed structure.
and 1.17 ppm in 6 (0.17 and 0.16 ppm downfield in
comparison with the corresponding signals in 2 and 3)
indicated that the C-4 OH group has the axial, â-configu-
Compounds 5 and 6 were isomers, with compositions of
27 2
C H43NO
as determined by HREIMS. Their 1H NMR
ration. Both 1H and C NMR chemical shifts of CH-3
15
13
spectra (Table 1) were essentially identical except for the
chemical shift of CH -21 (δ 0.96 and 1.07 ppm), which
indicated that they are (20R) and (20S) epimers. Their
NMR spectra (Table 2), in comparison with those of 2 and
, revealed the replacement of the CH group at δ 42.3
ppm in 2 and 3 with an oxygenated CH at δ 77.2 ppm in
and 6, consistent with the additional oxygen atom
and CH-4 were in complete agreement with those of 3â,4â-
confirming the structural and ster-
eochemical assignments as indicated.
The HREIMS of 8 showed a molecular ion at m/ z
2
413.329, consistent with the molecular formula C27H43NO ,
indicating that 8 is a structural isomer of 5 and 6. The
position of the second OH group was assigned to C-25 based
dihydroxy steroids,1
6,17
3
H
1
3
C
3
2
C
C
5

1
204 J ournal of Natural Products, 1998, Vol. 61, No. 10
Abdel-Kader et al.
Ta ble 2. 13C NMR Data (δ values) for Compounds 1-9a
carbon
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
35.7
34.0
199.7
123.7
171.6
32.9
32.0
35.6
53.8
38.6
21.0
37.9
42.3
55.4
24.0
27.8
53.5
11.8
17.4
46.6
18.0
174.4
26.4
27.4
27.7
56.9
19.5
37.3
31.6
71.7
42.3
140.8
121.6
31.8
31.9
50.1
36.5
21.1
38.1
42.2
56.3
24.0
26.4
53.6
11.8
19.4
46.6
18.1
174.5
27.3
27.8
27.6
56.9
19.5
37.3
31.6
71.6
42.3
140.9
121.5
31.8
31.8
50.1
36.5
21.0
39.7
42.4
56.4
24.3
27.7
53.1
12.0
19.3
47.0
18.3
175.3
26.5
27.2
27.4
56.4
19.1
37.0
31.4
71.2
41.7
140.7
121.4
31.6
31.7
49.9
36.3
20.9
39.5
42.2
56.2
24.0
27.2
29.6
12.0
19.3
54.3
19.3
151.7
151.7
122.9
130.9
139.9
30.6
37.0
25.3
72.2
77.2
142.8
128.3
32.0
31.9
50.2
36.0
20.5
38.0
42.2
56.4
24.0
27.3
53.5
11.8
20.1
46.3
18.1
175.0
26.5
27.8
27.6
56.6
19.4
37.0
25.3
72.3
77.2
142.8
128.3
32.0
31.8
50.3
36.0
20.5
39.6
42.4
56.7
24.3
27.2
53.1
12.0
20.9
46.8
18.3
175.5
26.6
27.6
27.4
56.3
19.1
36.9
31.0
71.1
41.8
140.5
121.1
31.7
31.5
49.9
36.3
20.1
39.7
42.2
56.0
23.8
26.7
53.2
11.3
19.1
46.1
17.5
178.1
31.7
30.8
65.1
59.1
26.6
37.0
31.1
71.1
41.8
140.6
121.1
31.7
31.6
49.9
36.3
20.1
39.7
42.3
56.3
24.1
26.8
52.9
11.6
19.1
46.3
17.6
178.2
31.7
30.9
65.4
59.1
26.7
37.2
29.7
71.7
42.3
140.7
121.6
31.8
31.8
50.0
36.4
21.1
39.7
42.2
56.4
24.2
27.9
53.9
11.6
19.4
40.3
13.6
59.6
31.7
33.4
31.6
53.9
19.2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
a
Obtained in CDCl3. Assignments made by a combination of DEPT, HETCOR, or HMQC, HMBC data and comparison with the spectrum
6
of verazine (3).
Ta ble 3. Antifungal and Cytotoxic Activities of Compounds 1-6, 8, 9
MIC (µg/mL) in strain
C. Albicans
IC12 values (µg/mL) in S. cerevisiae strain:
IC50 values (µg/mL)
in the M-109 cell line
compound
1138
1140
1353
S. cerevisiae Sc7
SC5314
1
2
3
4
5
6
8
9
14
45
50
0.69
0.14
0.40
24
NTa
6.25
6.25
20
200
100
6.25
NTa
12.5
1.6
NTa
12.5
25
16.8
NT
12.5
>20
a
0.88
0.14
0.04
10
12
0.35
3
1.12
0.25
0.70
23
10
2.5
3
20
400
200
<3.1a
NT
1.6
<0.8
10.5
15.5
31.2
>50
4
4
a
NT
amphotericin B
ketoconozale
8.1
2.5
15.6
4.8
25
9.4
7.5
17
a
NT ) not tested.
F igu r e 1. HMBC correlations of H-24 and H-26 in 4.
F igu r e 2. Partial structures of compound 8 (left) and its C-27 epimer
on the appearance of CH
along with a quaternary carbon resonance at δ
3
-27 as a singlet at δ
H
1.08 ppm,
65.4 ppm
(right) showing the CH
for both isomers.
3 3
-27 to H-26ax and CH -27 to H-26eq distances
C
assigned to the oxygenated C-25. Additional evidence was
CH
show NOE effects. Thus, distances of 2.55, 3.05, and 3.76
Å between the protons of CH -27 and H-26ax and distances
of 2.67, 3.22, and 3.81 Å between the protons of CH -27
and H-26eq were calculated for one conformation of CH
27eq, while similar calculations for CH -27ax gave distances
3
group would be close enough to both H-26 protons to
derived from an HMBC experiment (J ) 9 Hz), where the
H-26 proton at δ
H
3.23 ppm showed two bond correlations
3
to C-25. Although the two H-26 protons were overlapped
at 3.23 ppm when the spectrum was recorded in CDCl
3
3
3
-
(Table 1), the two resonances were well resolved when the
3
spectrum was measured in pyridine-d
5
(see Experimental
of 3.79, 4.34, and 3.83 for H-26ax and 2.58, 3.79, and 2.97
for H-26eq. The results of this experiment thus suggested
H
Section). The two doublets at δ 3.76 and 3.97 ppm were
assigned to H-26ax and H-26eq, respectively, based on
3
that 8 had an R-equatorial CH group and a â-axial OH
group. Upon acetylation 8 produced the diacetyl derivative
8a .
1
7
comparison with the H NMR spectra of 2 and 3 as well
1
as the H NMR spectrum of 3 in pyridine (see Experimental
Section). In a NOESY experiment, the CH
ppm showed a strong NOE correlation with the R-oriented
H-26ax and a weaker correlation with the â-oriented H-26eq
Molecular modeling (Figure 2) showed that an R-equatorial
3
-27 at δ
H
1.08
In addition to pure compound 8, an epimeric mixture of
compounds 7 and 8 was also obtained. This mixture could
not be separated by normal chromatographic methods, but
the acetylated mixture of 7a and 8a could be separated by
.

Bioactive Steroidal Alkaloids from Eclipta
J ournal of Natural Products, 1998, Vol. 61, No. 10 1205
HPLC using a C18 Si gel column. The H and 13C NMR
spectra of 7 (Tables 1 and 2) were obtained by subtracting
the resonances due to compound 8 from the spectra of the
1
Sch em e 1. Synthesis of Cyclic Imines 12a -d a) RMgBr, b)
TFA, 3 h, c) NaOH.
1
mixture. A comparison of the H NMR data for 7, 7a , 8,
and 8a indicated that 7 and 8 are epimeric at C-20, with 7
being the (20R) epimer and 8 being the (20S) epimer.
ing, for there seemed to be no obvious structural feature
of these compounds that would suggest such activity. The
only obviously reactive group was the imino function. In
the initial stages of this work the pyridine alkaloid 4 had
not been isolated, and so we carried out the synthesis of
the model compounds 12a -d to determine whether such
simple analogues might show a similar bioactivity. The
2
1
syntheses were carried out by literature procedures and
involved the addition of a Grignard reagent to the lactam
1
-(tert-butoxycarbonyl)-1-azacyclohexan-2-one (10) to give
the ketoamides 11a -d (Scheme 1). These amides were
then cyclized by treatment with TFA and deprotected to
give the imines 12a -d . These compounds were, however,
completely inactive, indicating that the steroidal moiety
is necessary for bioactivity.
All the acetylated derivatives were inactive. The large
decrease in activity by small modifications in the steroidal
skeleton, as in 1, 5, and 6 (Table 4), indicate that this
skeleton with a 3â-OH is required. Introduction of an OH
group at C-25 as in 8 or reduction of the double bond
between C-22 and C-26 as in 9 resulted in less dramatic
loss of activity. Finally, conversion of the imine to a
pyridine moiety as in 4 resulted in an increase of both
activity and selectivity in comparison with verazine (3).
Exp er im en ta l Section
Gen er a l Exp er im en ta l P r oced u r es. NMR spectra were
recorded in CDCl
3
1
on a Varian Unity 400 NMR instrument at
13
3
99.951 MHz for H and 100.578 MHz for C, using standard
Varian pulse sequence programs. LRMS and exact mass
measurements were taken on a VG 7070 E-HF at the Virginia
Polytechnic Institute and State University. Molecular model-
ing studies to calculate the internuclear distances discussed
for compound 8 were carried out using the MacSpartan
program (Wavefunction, Inc., Irvine, CA). Other conditions
were as previously described.22
The activities of compounds 1-6 and 8 in the three yeast
assays and against the Sc7 yeast strain18 are shown in
Table 3. The most active compounds are 2-4 and 8, with
the pyridine-containing alkaloid 4 being the most active
in the yeast assay. Interestingly, compound 8 showed good
activity against Candida albicans, with an MIC value only
slightly less than those of the clinically used antifungal
drugs amphotericin B and ketoconazole. These activities
apparently do not translate well into cytotoxicity, however,
because all the compounds showed IC50 values of >10 µg/
mL against the M-109 cell line, and this suggests that the
primary activity of the alkaloids is as antifungal agents.
Regrettably, the slight but real cytotoxicity of these com-
pounds was sufficient to preclude further development as
antifungal agents. Verazine (3) has previously been re-
ported to have antifungal activity19 and to inhibit DNA
formation by hepatoma and S-180 cells.20
Yea st Bioa ssa y. The yeast bioassay was carried out using
the following strains: 1140 ) J N392 ) ise1rad52; 1353 )
+
J N2-291 ) ISE rad52top1; 1138 ) J N394 ) aISE2ura3-
5
1leu2-3,112trp1-289 rad52::LEU2. All strains were obtained
from Dr. Nitiss through one of us (S. M.). Cultures were grown
to stationary phase in YEPD broth (Difco), centrifuged,
resuspended in 0.9% saline to 25% light transmission at 600
nm, and refrigerated at 4 °C. The yeasts were grown on YEPD
Agar (Difco) as base agar with an overlay of 0.8% Noble Agar
(Scott); plates were prepared and flooded with 2.5 mL of
inoculum. After a brief interval the excess inoculum was
pipeted off the plates (approximately 2 mL is removed) and
the plates allowed to dry under a biological hood leaving a
uniform lawn of yeast cells. After drying, wells of 6-7-mm
diameter were cut in the agar layer, and samples to be tested
dissolved in DMSO-MeOH (1:1) and added to the wells in 100-
mL aliquots. Nystatin (Sigma) was used as a positive control
at 20 µg/mL. The plates were incubated at 30 °C for 36-48
The observation that steroidal alkaloids such as verazine
3) showed good activity in our yeast bioassay was intrigu-
(

1
206 J ournal of Natural Products, 1998, Vol. 61, No. 10
Abdel-Kader et al.
Ta ble 4. 1H and 13C NMR Data (δ values) for Compounds 12a -d a
2a 12b
1
12c
12d
position
1H
13C
1H
13C
1H
13C
1H
13C
2
3
4
5
6
1
2
3
4
5
6
7
8
9
170.8
21.7
28.8
19.6
48.9
42.9
19.4
13.7
165.7
21.8
27.0
19.6
49.7
130.4
128.1
125.8
129.4
125.8
128.1
164.4
21.7
26.8
19.6
49.9
135.5
128.3
127.2
138.6
127.2
128.3
2.06 (4H, m)
1.49 (4H, m)
1.60 (2H, m)
3.49 (2H, m)
2.06 (4H, m)
1.49 (4H, m)
1.21 (6H, b)
1.21 (6H, b)
1.21 (6H, b)
1.21 (6H, b)
1.21 (6H, b)
1.21 (6H, b)
21.8
28.9
19.5
49.0
41.1
31.9
29.5
29.5
29.5
29.2
28.9
22.6
14.0
2.01 (4H, m)
1.49 (4H, m)
1.60 (2H, m)
3.42 (2H, m)
2.06 (4H, m)
1.49 (4H, m)
2.60 (2H, m)
1.65 (2H, m)
1.81 (2H, m)
3.81 (2H, m)
2.56 (2H, m)
1.64 (2H, m)
1.81 (2H, m)
3.80 (2H, m)
′
′
′
′
′
′
′
′
′
7.35-7.69 (5H, m)
7.35-7.69 (5H, m)
7.35-7.69 (5H, m)
7.35-7.69 (5H, m)
7.35-7.69 (5H, m)
7.68 (2H, d, J ) 8.6 Hz)
7.31 (2H, d, J ) 8.6 Hz)
7.31 (2H, d, J ) 8.6 Hz)
7.68 (2H, d, J ) 8.6 Hz)
0.82 (3H, t, J ) 7.0 Hz)
a
Obtained in CDCl3. b J ) Hz. c Carbon type as determined by DEPT spectra: 0 ) quaternary, 1 ) methine, 2 ) methylene, 3 )
methyl.
h, and the resulting zones of inhibition were measured in
millimeters. Activity was determined from a dose-response
curve and is reported as an IC12 value, which is the dose (in
µg/mL) required to produce a zone of inhibition 12 mm in
diameter.
Antifungal assays against S. cerevisiae Sc7 and Candida
albicans SC5314 and cytotoxicity assays against the M-109
cell line were performed at Bristol-Myers Squibb Pharmaceu-
tical Research Institute, Wallingford, CT, using standard
protocols.
P la n t Ma ter ia ls. The leaves of E. alba (L.) Hassk. (Aster-
aceae) were collected from Akisiamau, near Asindopo,
Suriname, on J une 11, 1994. A herbarium specimen was
deposited in the National Herbarium of Suriname under
voucher number CI 0018.
P la n t Extr a ction . E. alba leaves were extracted with
EtOAc and MeOH at Bedrijf Geneesmiddelen Voorziening
Suriname (BGVS). Extraction of 1 kg of dried plant material
with MeOH yielded 63 g of extract as BGVS M-940116.
Isola tion of Bioa ctive Alk a loid s. The MeOH extract of
E. alba was active against the mutant strains 1138, 1140, and
Me
eluting by 3% MeOH in CHCl
MeOH to afford verazine (3, 230 mg). The more polar active
fractions (18-20, 1.6 g), eluted with 35% Me CO in CHCl
were rechromatographed on a Si gel column (50 g) eluting with
10% MeOH in CHCl The active fractions (200 mg) were
subjected to preparative TLC (Si gel, 8% MeOH in CHCl , 5
developments) to afford 4 (2 mg, R 0.81), 5, (18 mg, R 0.68),
and 6 (30 mg, R 0.65). The most polar active fractions (23-
25, 0.3 g), eluted with 50% Me CO in CHCl , were further
purified on a Si gel column (10 g) eluting by 10% MeOH in
2
CO in CHCl
3
, were purified on a Si gel column (50 g)
3
followed by crystallization from
2
3
,
3
.
3
f
f
f
2
3
CHCl
CHCl
3
followed by preparative TLC (Si gel, 10% MeOH in
, 5 developments) to afford a mixture of 7 and 8 (7 mg,
3
R
f
0.62) and pure 8 (20 mg, R
20-epi-3-Deh yd r oxy-3-oxo-5,6-d ih yd r o-4,5-d eh yd r over -
a zin e [(20R,20S)-3-oxo-22,26-im in och olest a -4,22(N)-d i-
en e] (1): colorless amorphous matrix, [R]26
+68° (c 1.0,
MeOH); UV (MeOH) λmax (log ꢀ) 240 (5.1), 209 (4.9) nm; IR
f
0.59).
D
-
1
1
(film) νmax 2946, 1668 (CO), 1614, 1455, 1229, 753 cm ; H
NMR and 13C NMR, see Tables 1 and 2; EIMS m/ z 395 (M ,
10), 381 (12), 366 (6), 272 (4), 164 (18), 152 (29), 140 (37), 125
+
1
353. The three strains were used to determine the IC12
(C
8
H15N, 100), 111 (64), 91 (48), 81 (47), 67 (35), 55 (75);
+
values of the total extract and pure isolates, while the activity
was monitored during the fractionation procedures using the
1
2
respectively, 63.5 g) was fractionated between BuOH and H O.
2
The H
HREIMS m/ z 395.317 (M ), calcd for C27
H
41NO 395.318.
20-epi-Ver a zin e
[(3S,20R,25S)-22,26-im in och olesta -
2
6
138 mutant strain. The bioactive MeOH extract (IC12, 9,16,
6 µg/mL on the mutant yeast strains 1353, 1138, and 1140,
5,22(N)-d ien -3-ol] (2): colorless amorphous matrix, [R]
D
+6.5° (c 1.9, MeOH); UV (MeOH) λmax (log ꢀ) 206 (4.8) nm; IR
(film) νmax 3328 (OH), 2936, 1653 (CdN), 1569, 1415, 1377,
-
1
1
13
13
2
O fraction (18.5 g) was inactive. The active BuOH (44.9
1123 cm ; H NMR and C NMR, see Tables 1 and 2;
C
+
+
g) fraction was dissolved in 80% aqueous MeOH and extracted
3
NMR, see Table 2; EIMS m/ z 397 (M , 10), 382 (M - CH ,
11), 379 (9), 364 (13), 164 (15), 125 (C
(20), 81 (15), 55 (30).
with hexane (750 mL × 3). The activity was retained in the
8
H15N, 100), 111 (55), 91
2
aqueous MeOH fraction (33.7 g), which was diluted with H O
to 60% MeOH and fractionated with CHCl
CHCl fraction (20 g) was the most active, showing an IC12
.5 µg/mL, while the 60% aqueous MeOH fraction was much
weaker: IC12 114 µg/mL. The CHCl fraction (20 g) was
purified by chromatography on Sephadex LH-20 (5 × 100 cm)
using the solvents hexane-CH Cl (1:4, 3 L), CH Cl -Me CO
3:2, 3 L), CH Cl -Me CO (4:1, 3 L), and MeOH (2 L). The
active fraction (IC12 2.8 µg/mL, 15.0 g), eluted with hexane-
CH Cl (1:4), was separated on 500 g Si gel using the VLC
technique eluting with CHCl -Me CO mixtures.
The active fractions (7-9) eluted with 30% Me CO in CHCl
5.8 g) were refractionated using flash Si gel column (200 g,
CHCl -Me CO mixtures), and 30 fractions were collected.
Fractions 5-7 (0.2 g), eluted with 20% Me CO in CHCl , were
further purified on a Si gel column (15 g) eluting by CHCl
followed by preparative TLC (C18 Si gel, 10% H O in MeOH)
to afford 20-epi-3-dehydroxy-3-oxo-5,6-dihydro-4,5-dehydro-
3
(750 mL × 3). The
Ver a zin e [(3S,20S,25S)-22,26-im in och olest a -5,22(N)-
d ien -3-ol] (3): colorless needles, mp 175-176 °C (MeOH), lit.
3
8
26
26
3
mp 176-178 °C, [R]
D
-65° (c 1.5, MeOH), [lit. [R]
D
-89.7°
8
3
(EtOH)]; UV (MeOH) λmax (log ꢀ) 206 (4.6) nm; IR (film) νmax
3336 (OH), 2928, 1659 (CdN), 1567, 1413, 1370, 1118 cm-1;
1
1
2
2
2
2
2
H NMR in CDCl
(3H, s, CH -18), 0.82 (3H, d, J ) 6.6 Hz, CH
CH -19), 1.14 (3H, d, J ) 6.9 Hz, CH -21), 3.08 (1H, dd, J )
9.8, 16.8 Hz, H-26ax), 3.83 (1H, dd, J ) 4.4, 23.9 Hz, H-26eq),
3
, see Table 1; H NMR (pyridine-d
5
) δ 0.67
(
2
2
2
3
3
-27), 1.05 (3H, s,
3
3
2
2
1
3
3
2
3.84 (1H, m, H-3), 5.41 (1H, bd, J ) 4.8 Hz, H-6); C NMR,
+
+
2
3
see Table 2; EIMS m/ z 397 (M , 12), 382 (M - CH
3
, 13), 379
H15N, 100), 111 (60), 91 (13),
(
(12), 364 (10), 164 (11), 125 (C
81 (12), 55 (25).
8
3
2
2
3
Eclip ta lbin e [(3S,20R)-20-(3-h yd r oxy-5-m eth yl-2-p yr -
id yl)ch olest-5-en e-3,23-d iol] (4): white powder, mp 275-278
3
°C (MeOH), [R]26
-77° (c 0.41, MeOH); UV (MeOH) λmax (log
2
D
ꢀ) 290 (3.8), 203 (4.1) nm, (NaOH) λmax (log ꢀ) 307 (4.1), 243
1
13
verazine (1, 10 mg, R
5% Me CO in CHCl
f
0.64). Fraction 9 (100 mg), eluted with
, was again purified on a Si gel column
(4.1), 201 (4.5) nm; H NMR and C NMR, see Tables 1 and
1
3
2
2
3
2; C NMR (pyridine-d
(C-26), 151.1 (C-22), 153.1 (C-23); EIMS m/ z 409 (M , 20), 394
5
) δ 122.7 (C-24), 131.1 (C-25), 140.8
+
(
10 g) eluting by 3% MeOH in CHCl
TLC (Si gel, hexane-Me CO-MeOH, 60:40:4, 5 developments)
to obtain 20-epiverazine (2, 11 mg, R 0.78) and verazine (3,
3 mg, R 0.75). Fractions 10-14 (1.5 g), eluted with 25%
3
followed by preparative
+
2
(M - CH
100), 136 (C
2
(25); HREIMS m/ z 409.297 (M ), calcd for C27H39NO 409.298.
3
, 21), 150 (C
9 8
H12NO, 11), 138 (95), 137 (C H11NO,
f
8
H
10NO, 38), 117 (39), 91 (50), 77 (30), 67 (25), 55
+
2
f

Bioactive Steroidal Alkaloids from Eclipta
J ournal of Natural Products, 1998, Vol. 61, No. 10 1207
Red u ction of Ver a zin e (3) to Ver a m ilin e (9). Compound
2
0-ep i-4â-H yd r oxyver a zin e
[(3S,4R,20R,25S)-22,26-
im in och olesta -5,22(N)-d ien e-3,4-d iol] (5): white powder,
3 (15 mg) in 0.5 mL MeOH was treated with NaHB
for 30 min at room temperature; H O (5 mL) was added to
the reaction mixture, which then was extracted with CHCl
4
(75 mg)
2
6
mp 107-109 °C (MeOH), [R]
D
-20.5° (c 2.0, MeOH); UV
2
(
2
MeOH) λmax (log ꢀ) 206 (3.4) nm; IR (film) νmax 3316 (OH),
3
-
1 1
933, 1657 (CdN), 1572, 1424, 1380, 1067, 841 cm ; H NMR
(4 × 5 mL). The reaction gave 9 (11 mg) as single product:
1
3
+
10
and C NMR, see Tables 1 and 2; EIMS m/ z 413 (M , 14),
colorless crystals, mp 199-202 °C (Me CO) [lit. 198-200 °C];
10 1 13
2
3
(
95 (M+ - H
20), 139 (15), 125 (C
1 (17), 79 (21), 67 (21), 55 (40); HREIMS m/ z 413.329 (M ),
calcd for C27 413.329.
2
O, 11), 380 (M - H
+
2
O - CH
3
, 10), 164 (18), 150
[R]
NMR, see Tables 1 and 2; EIMS m/ z 156 (7), 99 (15), 98
26
D
-43° (c 0.7, MeOH) (lit. [R]
26
D
-49°); H NMR and
C
8
H
15N, 100), 121 (42), 111 (53), 93 (19),
+
+
+
9
(C
6
H
12N, 100); CIMS m/ z 400 (M + 1, 20), 398 (M - 1, 11),
H
43NO
2
98 (C H
6
12N, 100).
Acetyla tion of 20-epi-4â-Hyd r oxyver a zin e (5) to 20-epi-
Gen er a l P r oced u r e for th e Syn th esis of Cyclic Im in es.
A solution of 2-piperidone (990 mg, 10 mmol) in dry THF (30
mL) was cooled to -78 °C, then BuLi (2.0 M, 4 mL, 10 mmol)
was added dropwise with stirring, and after 30 min di-tert-
butyl dicarbonate (2.2 g, 10 mmol) in dry THF (10 mL) was
added slowly and the stirring was continued for 3 h at -78
°C. A solution of the chosen Grignard reagent (15 mmol) was
then added over 15 min, and the mixture was stirred at -78
°C for 3 h more. The reaction mixture was then quenched with
2N HCl (10 mL) and extracted with ether (4 × 60 mL). The
combined organic phase was washed with 10% aqueous
NaHCO , then with brine, and finally dried over Na SO . The
residue left after evaporation of the ether was purified by
crystallization from MeOH. TFA (2 mL) was added slowly at
0 °C with stirring to each of the pure crystalline products
11a -d (100 mg), and after 3 h 30% aqueous NaOH was added
to pH 10-11, and the reaction mixture was extracted with
ether (4 × 15 mL). The final product was obtained after
evaporation of the ether under atmospheric pressure.
4
â-Hyd r oxyver a zin e Tr ia ceta te (5a ). Compound 5 (2.0 mg)
2
in pyridine (0.5 mL) was treated with Ac O (0.2 mL) for 24 h
at room temperature. Evaporation of the resulting solution
under a stream of argon yielded chromatographically homo-
geneous 5a (2.0 mg): colorless amorphous matrix; UV (MeOH)
1
λ
max (log ꢀ) 228 (3.7), 206 (3.7) nm; H NMR, see Table 1.
4
â-H yd r oxyver a zin e
[(3S,4R,20S,25S)-22,26-im in o-
ch olesta -5,22(N)-d ien e-3,4-d iol] (6): colorless crystals, mp
2
6
1
λ
(
63-165 °C (MeOH), [R]
D
-12.8° (c 0.7, MeOH); UV (MeOH)
max (log ꢀ) 207 (3.4) nm; IR (film) νmax 3319 (OH), 2939, 1654
-
1
1
13
CdN), 1569, 1430, 1377, 1062, 838 cm ; H NMR and
C
3
2
4
+
+
NMR, see Tables 1 and 2; EIMS m/ z 413 (M , 4), 395 (M
-
+
H
(
(
2
O, 8), 380 (M - H
2
O - CH
3
, 8), 164 (14), 150 (19), 125
8
C H15N, 100), 121 (45), 111 (75), 91 (28), 79 (26), 67 (26), 55
+
53); HREIMS m/ z 413.329 (M ), calcd for C27
Acetyla tion of 4â-Hyd r oxyver a zin e (6) to 4â-Hyd r oxy-
ver a zin e Tr ia ceta te (6a ). Compound 6 (4.0 mg) in pyridine
0.5 mL) was treated with Ac O (0.2 mL) for 24 h at room
2
H43NO 413.329.
(
2
temperature. Evaporation of the resulting solution under a
stream of argon yielded chromatographically homogeneous 6a
2
-Non yl-3,4,5,6-tetr a h yd r op yr id in e (12a ): colorless oil;
1
13
+
H and C NMR, see Table 4; CIMS m/ z 210 (M + H, 22),
73 (19), 172 (24), 144 (23), 100 (24), 97 (21), 71 (45), 57 (100);
EIMS m/ z 144 (23), 143 (32), 124 (C 14N, 14), 110 (C
(
λ
4.0 mg): colorless crystals, mp 97-98 °C (MeOH); UV (MeOH)
1
1
max (log ꢀ) 228 (3.7), 207 (3.7) nm; H NMR, see Table 1.
0-ep i-25â-H yd r oxyver a zin e [(3S,20R,25R)-22,26-im -
in och olesta -5,22(N)-d ien e-3,25-d iol] (7): compound 7 could
8
H
7
H12N,
2
1
5), 97 (C
7
H13, 100).
2
-P r op yl-3,4,5,6-tetr a h yd r op yr id in e (12b): colorless oil;
1
13
not be separated from compound 8, and the H NMR and
C
1
13
H and C NMR, see Table 4.
NMR data for 7 (Tables 1 and 2) were thus obtained by
subtraction of the data for 8 from the corresponding spectrum
of the mixture of 7 and 8.
2
-P h en yl-3,4,5,6-tetr a h yd r op yr id in e (12c): colorless oil;
1
13
H and C NMR, see Table 4.
2
-(p -Ch lor op h en yl)-3,4,5,6-t et r a h yd r op yr id in e (12d ):
2
0-epi-25â-Hyd r oxyver a zin e Dia ceta te (7a ). A mixture
of 7 and 8 (4.0 mg) in pyridine (0.5 mL) was treated with Ac
0.2 mL) for 24 h at room temperature. After evaporation of
the resulting solution the mixture was separated by HPLC (C18
Si gel, MeOH-H O, 90:10) to afford 1.5 mg 7a and 2 mg 8a .
Com p ou n d 7a : amorphous matrix, [R]
MeOH); UV (MeOH) λmax (log ꢀ) 232 (4.6), 207 (4.6) nm; IR
film) νmax 3408 (OH), 2962, 1713 (CdO), 1669, 1385, 1239
2
3
1
13
colorless crystals, mp 54-55 °C, (lit. mp 54 °C); H and
NMR, see Table 4.
C
2
O
(
Ack n ow led gm en t. This work was supported by an Inter-
national Cooperative Biodiversity Group Award, number U01
TW/CA-00313, from the Fogarty Center, NIH, and this finan-
cial support, together with support and encouragement from
Dr. J oshua Rosenthal of the Fogarty Center, is gratefully
acknowledged. Mass spectra were obtained by Mr. Kim Harich
of Virginia Polytechnic Institute and State University. Ms.
Yvette Merton, Mr. Reggy Nelson, and Mr. Richard Autar of
Conservation International Suriname and Ms. Lisa Famolare
and Ms. Regina de Souza of Conservation International USA
are thanked for their administrative assistance and support.
2
2
6
D
-134° (c 1.5,
(
-
1
1
+
+
cm ; H NMR, see Table 1; EIMS m/ z 497 (M , 35), 483 (M
CH , 13), 437 (25), 420 (15), 375 (25), 360 (19), 183 (57), 168
50), 141 (54), 121 (51), 91 (100), 79 (77), 67 (65), 55 (92);
-
(
3
+
HREIMS m/ z 497.352 (M ), calcd for C31
5â-Hyd r oxyver a zin e [(3S,20S,25R)-22,26-im in och oles-
ta -5,22(N)-d ien e-3,25-d iol] (8): colorless crystals, mp 179-
82 °C (MeOH), [R]26
-28° (c 0.5, MeOH); UV (MeOH) λmax
log ꢀ) 211 (3.6) nm; IR (film) ν -m ax 3315 (OH), 2930, 1653 (Cd
4
H47NO 497.351.
2
1
(
D
1
1
N), 1569, 1415, 1377, 1123 cm ; H NMR (CDCl
3
), see Table
-18), 1.04 (3H, s,
-21), 1.36 (3H, s, CH
7), 3.76 (1H, d, J ) 17.3 Hz, H-26ax), 3.85 (1H, m, H-3), 3.97
1H, d, J ) 17.3 Hz, H-26eq), 5.40 (1H, br d, J ) 4.8 Hz, H-6);
Refer en ces a n d Notes
1
1
; H NMR (pyridine-d
5
) δ 0.68 (3H, s, CH
3
(1) Biodiversity Conservation and Drug Discovery in Suriname, Part 3.
For Part 2, see Abdel-Kader et al.3
CH -19), 1.19 (3H, d, J ) 6.6 Hz, CH
3
3
3
-
2
(
2) Zhou, B.-N.; Baj, N. J .; Glass, T. E.; Malone, S.; Werkhoven, M. C.
M.; van Troon, F.; David, M.; Wisse, J . H.; Kingston, D. G. I. J . Nat.
Prod. 1997, 60, 1287-1293.
(
1
3
+
+
C NMR, see Table 2; EIMS m/ z 413 (M , 5), 398 (M - CH
), 142 (C 16NO, 47), 141 (C 15NO, 100), 127 (94), 121 (51),
1 (31), 79 (33), 67 (34), 55 (58); HREIMS m/ z 413.328 (M ),
413.329.
3
,
(
(
(
3) Abdel-Kader, M. S.; Wisse, J . H.; Evans, R.; van der Werff, H.;
Kingston, D. G. I. J . Nat. Prod. 1997, 60, 1294-1297.
4) Nitiss, J .; Wang, J . C. Proc. Natl. Acad. Sci. USA 1988, 85, 7501-
7505.
5) Plant collections in Suriname were made on a “random” basis by the
Missouri Botanical Garden and on an ethnobotanical basis by
Conservation International. This collection was made by Conservation
International.
3
9
8
H
8
H
+
calcd for C27H43NO
2
Acetyla tion of 25â-Hyd r oxyver a zin e to 25â-Hyd r oxy-
ver a zin e Dia ceta te (8a ). Compound 8 (5.0 mg) in pyridine
0.5 mL) was treated with Ac O (0.2 mL) for 24 h at room
2
(
temperature. Evaporation of the resulting solution under a
stream of argon yielded chromatographically homogeneous 8a
(6) (a) Kritikar, K. R.; Basu, B. D. Indian Medicinal Plants, Chronica
Botanica: New Delhi, 1975. (b) Chopra, R. N.; Nayar, S. L.; Chopra,
I. C. Glossary of Indian Medicinal Plants. C.S.I.R.: New Delhi, 1955.
(5.0 mg), which was identical with 8a obtained by HPLC
(
7) Han, X.; Ruegger, H. Planta Med. 1992, 58, 449-453.
separation of the acetylated mixture of 7 and 8: colorless
(8) (a) Adam, G.; Schreiber, K.; Tomko, J .; Vassova, A. Tetrahedron 1967,
23, 167-171. (b) Bondarenko, N. V. Khim. Prir. Soedin. 1973, 132.
2
6
needle crystals, mp 160-163 °C (MeOH); [R]
D
+16° (c 1.0,
(c) Bondarenko, N. V. Khim. Prir. Soedin. 1979, 415. (d) Bondarenko,
MeOH); UV (MeOH) λmax (log ꢀ) 232 (4.6), 209 (4.6) nm; IR
film) νmax 3423 (OH), 2939, 1723 (CdO), 1639, 1385, 1246,
N. V. Khim. Prir. Soedin. 1984, 263. (e) Bondarenko, N. V. Khim.
Prir. Soedin. 1984, 801-2. (f) Zhao, W.; Tezuka, Y.; Kikuchi, T.; Chen,
J .; Guo, Y. Chem. Pharm. Bull. 1989, 37, 2920-2928.
(
-
1
1
1
030, 753 cm ; H NMR, see Table 1.

1
208 J ournal of Natural Products, 1998, Vol. 61, No. 10
Abdel-Kader et al.
(
9) Taskhanova, E. M.; Shakirov, R.; Yunusov, S. Y. Chem. Nat. Comp.
985, 21, 343-344.
10) Vassova, A.; Voticky, Z.; Tomko, J . Collect. Czech. Chem. Commun.
(18) This strain was selected for screening and assay purposes because it
demonstrated hypersensitivity to a variety of known antitumor and
antifungal agents relative to that of other S. cerevisiae strains
evaluated. (S. W. Mamber, Bristol-Myers Squibb Pharmaceutical
Research Institute, personal communication.).
(19) Kusano, G.; Takahashi, A.; Sugiyama, K.; Nozoe, S. Chem. Pharm.
Bull. 1987, 35, 4862-4867.
(20) Lisheng, L.; Xianhua, X.; Longdi, Z.; Rongliang, Z.; Zhongjian, J .; Yu,
L.; Wenkai, L.; Liming, Y. Lanzhou Daxue Xuebao, Ziran Kexueban
1985, 21, 86-90.
(21) Giovannini, A.; Savoia, D.; Umani-Ronchi, A. J . Org. Chem. 1989,
54, 228-234.
1
(
(
(
(
(
1
977, 42, 3643-3645.
11) Reich, H. J .; J autelat, M.; Messe, M. T.; Weigert, F. J .; Roberts, J . D.
J . Am. Chem. Soc. 1969, 91, 7445-7454.
12) Hanson, J . R.; Siverns, M. J . Chem. Soc., Perkin Trans. 1 1975, 1956-
1
958.
13) Ali, E.; Chakravarty, A. K.; Dhar, T. K.; Pakrashi, S. C. Tetrahedron
Lett. 1978, 3871-3874.
14) Mosher, H. S. In Heterocyclic Compounds; Elderfield, R. C., Ed.; J ohn
Wiley & Sons: New York, 1959; Vol. 1, p 442.
15) Zurcher, R. F. Helv. Chim. Acta 1963, 46, 2054-2088.
16) Albrecht, H. P.; Reichling, P. Liebigs Ann. Chem. 1975, 2211-2215.
17) Gong, Y.-Z. C NMR Spectra of Natural Products; Yun-Nan Press of
Sciences & Technology: Yun-Nan, Peoples’ Republic of China, 1985;
pp 243-244; the assignments of C-3 and C-4 chemical shifts in this
reference should be interchanged.
(
(
(
(22) Valente, L. M. M.; Gunatilaka, A. A. L.; Kingston, D. G. I.; Patitucci,
M. L.; Pinto, A. C. J . Nat. Prod. 1997, 60, 478-481.
(23) Salathiel, R.; Burch, J . M.; Hixon, R. M. J . Am. Chem. Soc. 1937, 59,
984-986.
1
3
NP970561C