Spasmolytic Effects, Mode of Action, and Structure-Activity Relationships of Stilbenoids from Nidema boothii

Article abstract of DOI:10.1021/np030303h

A CH₂Cl₂-MeOH (1:1) extract prepared from the whole plant of Nidema boothii inhibited spontaneous contractions (IC₅₀ = 6.26 ± 2.5 μg/mL) of the guinea-pig ileum. Bioassay-guided fractionation of the active extract led to t

Full text of DOI:10.1021/np030303h

1
60
J . Nat. Prod. 2004, 67, 160-167
Sp a sm olytic Effects, Mod e of Action , a n d Str u ctu r e-Activity Rela tion sh ip s of
Stilben oid s fr om Nid em a booth ii†
,⊥
‡
§
‡
§
,‡
Yanet Hern a´ ndez-Romero, J uana-Isela Rojas, Rafael Castillo, Alejandra Rojas, and Rachel Mata*
Departamento de Farmacia, Facultad de Qu ´ı mica, Universidad Nacional Aut o´ noma de M e´ xico, Coyoac a´ n, D.F 04510, M e´ xico,
and Facultad de Qu ´ı mica, Universidad Aut o´ noma de Quer e´ taro, Centro Universitario 7610, Quer e´ taro, M e´ xico
Received J uly 1, 2003
2 2
A CH Cl -MeOH (1:1) extract prepared from the whole plant of Nidema boothii inhibited spontaneous
contractions (IC50 ) 6.26 ( 2.5 µg/mL) of the guinea-pig ileum. Bioassay-guided fractionation of the active
extract led to the isolation of the novel spiro compound 1, which was given the trivial name nidemone,
and the new dihydrophenanthrene 3, characterized as 1,5,7-trimethoxy-9,10-dihydrophenanthrene-2,6-
diol. In addition, the known stilbenoids aloifol II (2), 1,5,7-trimethoxyphenanthrene-2,6-diol (4), ephem-
eranthoquinone (5), gigantol (6), ephemeranthol B (7), 2,4-dimethoxyphenanthrene-3,7-diol (8), lusian-
thridin (9), and batatasin III (10) were obtained. The isolates were characterized structurally by
spectroscopic data interpretation. Compounds 2-6, 9, and 10 induced notable concentration-dependent
inhibition of the spontaneous contractions of the guinea-pig ileum with IC50 values that ranged between
0
.14 and 2.36 µM. Bibenzyl analogues 23-35 were synthesized and tested pharmacologically. The results
indicated that for maximum spasmolytic activity the bibenzyls should have oxygenated substituents on
both aromatic rings; on the other hand, methylation of free hydroxyl groups as well as the increment of
oxygenated groups in relation to compounds 6 and 10 decreased the smooth muscle relaxant activity. It
was also demonstrated that bibenzyls 6 and 10 might exert their spasmolytic action not only by a nitrergic
mechanism but also by inhibiting CaM-mediated processes.
Nidema boothii (Lindl.) Schltr. (Orchidaceae), also known
as Epidendrum boothii (Lindl.) L.O. Williams, is an orchid
found from Mexico to Panama, Cuba, and Surinam, and
its habitat is in tropical moist forests up to 1500 m
elevation. A previous phytochemical study on this species
resulted in the isolation and characterization of a novel
rearranged tetracyclic triterpene named nidemin after the
Ch a r t 1. Compounds Isolated from Nidema boothii
1
genus of the plant. N. boothii is not used as a traditional
or alternative remedy; however, the related species Scaphy-
glottis livida (Lindl.) Schltr. and Maxillaria densa Lindl.
are employed for the treatment of stomachache in the State
of Veracruz, Mexico.2 Furthermore, bioactivity-guided
fractionation of the spasmolytic extracts prepared from
these orchids resulted in the isolation and characterization
of several stilbenoids, including 1,5,7-trimethoxyphenan-
threne-2,6-diol (4), gigantol (6), 2,4-dimethoxyphenan-
,3
threne-3,7-diol (8), lusianthridin (9), and batatasin III (10)
Chart 1), among others.2 These natural products induce
,3
(
a concentration-dependent inhibition of the spontaneous
contractions of the rat ileum with a potency higher than
2
or comparable to that of papaverine. It was also demon-
strated that the spasmolytic effect of 6 was mediated by
the nitric oxide (NO)/cyclic guanosine monophosphate
(
cGMP) system.2
In the present investigation N. boothii was selected
initially as potential source of spasmolytic compounds
following a chemotaxonomic approach. Thereafter, a phar-
2
macological test, using the guinea-pig ileum system, of a
2 2
CH Cl -MeOH (1:1) extract of the plant confirmed this
hypothesis. Thus, the first aim of the present study was to
†
Taken in part from the Ph.D. Thesis of Y.H.-R., Posgrado en Ciencias
Qu ´ı micas, Universidad Nacional Aut o´ noma de M e´ xico, 2003.
⊥
Dedicated to the late Dr. Monroe E. Wall and to Dr. Mansukh C. Wani
isolate the spasmolytic principles of N. boothii. In addition,
this investigation was designed to provide more informa-
tion about the mode of action of bibenzyls 6 and 10, also
isolated from N. boothii, by evaluating (i) their in vitro
activity on the regulatory protein calmodulin (CaM) and
of Research Triangle Institute for their pioneering work on bioactive natural
products.
*
Corresponding author. Tel: (525)-55-622-5289. Fax: (525)-55-622-5329.
E-mail: rachel@servidor.unam.mx.
‡
Universidad Nacional Aut o´ noma de M e´ xico.
Universidad Aut o´ noma de Quer e´ taro.
§
1
0.1021/np030303h CCC: $27.50
© 2004 American Chemical Society and American Society of Pharmacognosy
Published on Web 12/02/2003

Stilbenoids from Nidema boothii
J ournal of Natural Products, 2004, Vol. 67, No. 2 161
Ta ble 1. NMR Spectral Data for Nidemone (1) in CDCl3
position
δHa (J in Hz)
δCb
HMBCc
NOESY
1
2
202.7
59.2
25.6
3, 5
3, 4, 3′, 5
4, 5
3
3
4
4
4
5
6
7
8
8
2
3
4
5
5
A
B
B
A
a
2.47 ddd (14, 7.4, 4.0)
2.20 ddd (13.8, 9.3, 4.8)
3.48 brddd (16.7, 9.4, 5.0)
2.87 ddd (16.8, 5.8, 5.8)
3B, 4A, 4B, 5′
3A, 4A, 4B
3A, 3B, 4A, 5
3A, 3B, 4B, 5′A, 5′
32.1
3, 5
145.4
118.9
137.0
115.8
163.4
116.4
189.6
101.9
202.3
39.7
3, 4, 6
3, 4, 7
5, 7
5, 6
6
5, 7
3, 3′, 5′, -OCH3
5′
3′, 5′
3, 3′
6.72 brdd (7.5, 1)
7.39 dd (7.5, 7.5)
6.80 dd (7.5, 1)
4A, 4B
5, 7
6
a
′
′
′
5.27 t (1.2)
-OCH3
′A
′B
OCH3
OH-8
3.29 dd (17.5, 1.5)
2.48 dd (17.5, 1.5)
3.91 s
3A, 4A, 5′B
4A, 5′A
3′
-
-
59.1
12.23 d (0.6)
a
Recorded at 300 MHz. b Recorded at 75 MHz. c Protons showing long-range correlation with indicated carbon.
(
ii) their effects on the spontaneous contraction of the
(δ
C
202.3) system with a polarizable double bond (δ
C
189.6
C
59.1).
guinea-pig ileum in the presence of 1H-[1,2,4]oxadiazolo-
and 101.9) bearing a methoxyl group (δ
H
3.91 s/δ
G
[4,3-R]quinoxalin-1-one (ODQ) and N -nitro-L-arginine meth-
According to the HMQC experiment, the vinyl carbon
resonating at higher field (C-3′) was associated with a
proton resonating at δH 5.27 (H-3′), while the lower one
(C-2′, â-carbon of the enone) was found to bear this
methoxyl group. This evidence defined the conjugated
system as a vinylogous carboxylic acid methyl ester and
yl ester (L-NAME), inhibitors of the enzymes guanylate
cyclase and nitric oxide synthase (NOS), respectively, and
finally to evaluate the spasmolytic effect of several natural
and synthetic analogues of bibenzyls 6 and 10, in an
attempt to establish the influence of the oxygenated
substituents on the pharmacological effects of both these
lead compounds.
1
0,11
explained the unusual polarization of the double bond.
The connection between the three partial structures in 1
was made on the basis of the analysis of the HMBC data
Resu lts a n d Discu ssion
(
Table 1). Thus, the HMBC correlations observed between
the C-4 methylene hydrogens and the aromatic carbons at
118.9 (C-5), 145.4 (C-4a), and 116.4 (C-8a), as well as
A CH
plant of N. boothii inhibited spontaneous contractions
IC50 ) 6.26 ( 2.5 µg/mL) of the guinea-pig ileum when
2 2
Cl -MeOH (1:1) extract prepared from the whole
δ
C
those between the C-3 methylene hydrogens and C-4a and
C-1, clearly indicated that the cyclohexenone unit was
fused to the benzene ring. Similarly, the long-range cor-
(
tested in vitro. Accordingly, this spasmolytic extract was
selected for bioassay-guided fractionation. This process led
to the isolation of the novel spiro-compound 1, which was
given the trivial name nidemone, and the new dihydro-
phenanthrene 3, characterized as 1,5,7-trimethoxy-9,10-
dihydrophenanthrene-2,6-diol (Chart 1). The known stil-
benoids 2 and 4-10 (Chart 1) were also obtained from the
active fraction, and their spectral data were in agreement
with those reported previously.
nidemin1 spontaneously crystallized from some inactive
fractions.
Compound 1 was isolated as an optically active glassy
yellow solid. Its molecular formula was established as
C H O by HREIMS. The nine degrees of unsaturation
15 14 4
relations of the signals at δ
H
2.47 (H-3A), 2.20 (H-3B), 5.20
(H-3′), 2.48 (H-5′B), and 3.29 (H-5′A) with the signal at δ
C
5
9.2 (C-2/1′) connected the cyclohexenone and cyclopen-
tenone systems through a spiro-carbon. The NOESY cor-
relations (Table 1) for H-4A, H-4B, H-5, H-5′A, H-5′B, and
H-3′ provided further support to this proposal. A systematic
conformational search for compound 1 using the Spartan’02
molecular modeling program revealed the two minimum
energy conformations 1a and 1b depicted in Figure 1
2
,4-8
9
In addition, vitexin and
(EMMX ) 39.38 and 39.96 kcal/mol, respectively). In 1b, the
six-membered ring adopts an envelope conformation, with
the interatomic distances between H4A-H5′A and H4A-
H5′B being 2.4084 and 3.1162 Å, respectively. This result
is consistent with the strong NOESY correlations observed
between H-4A-H-5′A and H-4A-H-5′B (Table 1). Finally,
the absolute configuration of compound 1 was proposed as
depicted on the basis of the CD data. The CD spectra of
R,â-unsaturated ketones, including aromatic ketones, typi-
cally exhibit not just two, but three and sometimes four
Cotton effects between 185 and 360 nm. Thus, a weak red-
shifted ketone carbonyl due to an n f π* excitation is
observed at ∼320-350 nm, and this excitation is followed
by one or more π f π* transitions with one corresponding
to the typical UV band between 230 and 260 nm, and a
second often overlapping with the first and lying in the
200-220 nm range; in the particular case of aromatic
ketones a strong band around ∼290 nm (B band) is also
observed. Finally, another, more energetic band is found
near 188 nm, and it is thought to involve a n f σ*
in this formula could be partially accounted for by two
carbonyls, a benzene ring, and one double bond; hence
compound 1 was tricyclic. Three partial structures could
be constructed that fully accounted for all the atoms in
nidemone (1). First, a trisubstituted benzene ring moiety
1
was apparent from the H NMR spectrum data (Table 1),
which showed an ABC system [δ 6.72, br dd, J ) 7.5,
Hz (H-5); 7.39 dd, J ) 7.5, 7.5 Hz (H-6); and 6.80, dd,
J ) 7.5, 1 Hz (H-7)]; these protons correlated with the
signals at δ 118.9 (C-5), 137.0 (C-6), and 115.8 (C-7),
respectively. Additionally, a hydrogen-bonded phenolic
hydroxyl group was determined from the signal at δ 12.23
H
1
C
H
1
in the H NMR spectrum. The second partial structure was
defined by HMBC (Table 1) as a cyclohexenone, as sug-
gested from absorptions in the IR (1692 cm- ) and C NMR
1
13
C
spectra (δ 202. 7, C-1) for a conjugated carbonyl. The third
partial structure consisted of a conjugated cyclopentenone

1
62 J ournal of Natural Products, 2004, Vol. 67, No. 2
Hern a´ ndez-Romero et al.
F igu r e 1. Minimum energy structures of nidemone (1) showing
relevant NOESY correlations.
transition.12 The first two sets of bands are the most useful
for determining absolute stereochemistry, in particular
when inherently dissymmetric chromophores are present
in a given molecule. Compound 1 possesses such a chro-
F igu r e 2. Concentration-response curves showing the relaxant
effects of compounds 2-6, 9, and 10 from Nidema boothii on the
isolated guinea-pig ileum. Values are expressed as the percentages of
inhibition of contractile responses calculated as the mean from six
data ( SEM, p < 0.05.
mophore since the cyclopentenone system is nonplanar.
According to the Snatzke’s helicity rules,1
2,13
a cyclopen-
tenone system has a negative helicity if the CD spectrum
displays a positive Cotton effect for the π f π* transition
and a negative Cotton effect for the n f π*. The CD
spectrum of 1 exhibited a strong positive Cotton effect
around 265 nm and a negative one at 326 nm, correlating
with a negative helicity and thus with an S absolute
stereochemistry at C-2.13 Furthermore, optical rotations of
nidemone (1) and (S)-4-hydroxycyclopent-2-en-1-one11
showed a negative sign, suggesting that both compounds
have the same absolute configuration.
and C-6. On the other hand, the correlations C-9/H-8, C-4b/
H-4, C-4a/H-4, C-1/OMe-1, and C-5/OMe-5 observed in the
HMBC spectrum further supported this proposal. On the
basis of this evidence, stilbenoid 3 was identified as 1,5,7-
trimethoxy-9,10-dihydrophenanthrene-2,6-diol.
Compounds 2-6, 9, and 10 induced notable concentra-
tion-dependent inhibition of the spontaneous contractions
of the guinea-pig ileum. Figure 2 shows the concentration-
response curves for these compounds. All the isolates were
more potent than the crude extract (IC50 ) 6.26 ( 2.5 µg/
mL) and papaverine (IC50 ) 4.23 ( 0.68 µM). The greatest
inhibitory activities were observed for compounds 10
(IC50 ) 0.24 ( 0.11 µM), 6 (IC50 ) 0.26 ( 0.10 µM), 9
(IC50 ) 0.41 ( 0.03 µM), and 4 (IC50 ) 0.45 ( 0.03 µM).
Natural products 1, 7, and 8 were not tested due to scarcity
of the samples; however, the spasmolytic property of
compound 8 was described previously.2
Compound 3 had the composition C17
mined by its MS and C NMR data, differing from
18 5
H O as deter-
1
3
phenanthrene 4 by two mass units. The NMR spectra of 3
(
see Experimental Section) showed the characteristic sig-
3
nals of a 9,10-dihydrophenanthrene and suggested that 3
is the 9,10-dihydro derivative of compound 4. The most
2
obvious differences between the NMR spectra of the two
compounds resulted from the presence of two methylene
signals in 3 [δ
H
/δ
C
2.8 (2H, m, H-10)/22.5 (C-10) and 2.69
To establish the influence of the nature (phenolic vs
phenolic methyl ether) and location of the oxygenated
substituents along the bibenzyl core on the pharmacological
effects of 6 and 10, compounds 23-35 (Chart 2) were
synthesized and tested pharmacologically. All compounds
but 28 and 33 were obtained using the Wittig reaction, a
(2H, m, H-9)/29.4 (C-9)] instead of the aromatic resonances
attributed to H-9/C-9 and H-10/C-10 in 4. In addition, the
chemical shift values for the aromatic protons and carbons
were shifted diamagnetically in comparison to those in 4.
Finally, the HMBC and NOESY correlations supported the
position of the substituents along the dihydrophenanthrene
core. Thus, in the NOESY spectrum the correlations H-8/
1
4-16
method widely used for the synthesis of bibenzyls.
Analogues 28 and 33 were synthesized by catalytic reduc-
1
7,19
OCH
3
-7 and H-9, H-4/OCH
3
-5 and H-3, and H-10/OCH
3
-1
tion of resveratrol and piceatannol, respectively.
Biben-
and H-9 were consistent with the placement of the meth-
oxyl groups at C-7, C-5, and C-1 and the hydroxyl at C-2
zyls 23, 25, 28, and 29 were synthesized as previously
described, and their spectral data were in agreement with

Stilbenoids from Nidema boothii
J ournal of Natural Products, 2004, Vol. 67, No. 2 163
Ch a r t 2. Synthetic Analogues of Gigantol (6) and Batatasin III (10)
Ta ble 2. Inhibition of the Spontaneous Contraction of Isolated
Guinea-Pig Ileum Induced by Gigantol (6), Batatasin III (10),
and Related Compounds
gigantol (6), were the most potent. Furthermore, the
absence of oxygenated substituents in one of the aromatic
rings of the stilbenoid moiety, as in the case of compounds
34 and 35, induced the highest decrease in activity. Hence,
it could be inferred that the presence of oxygenated
substituents in both aromatic rings is essential for activity
within this compound class.
a
potency relative
substance
Emax
IC50 (µM)
to papaverine
2
6
1
2
2
2
2
2
2
2
3
3
3
3
3
3
73.63 ( 1.18 0.56 ( 0.20
81.30 ( 1.84 0.26 ( 0.10
67.50 ( 4.76 0.24 ( 0.11
76.40 ( 4.76 1.39 ( 0.52
84.94 ( 5.38 0.62 ( 0.03
66.55 ( 1.30 0.14 ( 0.08
51.78 ( 6.06 0.70 ( 0.05
66.00 ( 1.30 0.58 ( 0.08
53.65 ( 6.73 0.85 ( 0.04
85.00 ( 1.79 0.14 ( 0.04
58.59 ( 0.81 1.28 ( 0.40
78.50 ( 1.27 2.36 ( 0.77
75.57 ( 0.21 0.96 ( 0.12
64.10 ( 3.62 1.75 ( 0.17
83.28 ( 4.09 1.49 ( 0.39
82.22 ( 2.50 1.29 ( 0.36
91.76 ( 2.65 4.23 ( 0.68
7.65
16.26
17.62
3.05
6.82
31.33
5.99
7.32
4.94
30.85
3.30
1.78
4.41
2.42
2.84
3.28
1
0
3
4
5
6
7
8
9
0
1
2
3
4
5
In our initial studies regarding the mode of action of
compounds 6 and 10 it was demonstrated that their smooth
muscle relaxant effect was blocked by L-NAME when tested
2
at their IC50’s. It was also demonstrated, using a radio-
immunoassay procedure, that compound 6 increased cGMP
content in rat ileal rings. Compound 6-induced elevation
of cGMP has been reported to be inhibited by L-NAME and
ODQ, suggesting that its spasmolytic effect was mediated
2
by the NO/cGMP system. However, on that occasion, the
influences of ODQ and different concentrations of L-NAME
on the spasmolytic effect of both compounds were not
assessed. The results shown in Figure 2 indicated that in
the presence of ODQ (100 µM) and L-NAME (500 µM) the
concentration-response curves of both compounds were
shifted to the right (Figure 3), thus confirming indeed that
they modify the nitrergic system.
papaverine
chlorpromazine 81.80 ( 2.30 0.18 ( 0.12
23.50
a
Values as means ( SEM; n ) 6; p < 0.05. Potency was
obtained by the formula IC50 (µM) papaverine/IC50 (µM) compound
assuming a value of 1.00 for papaverine. Emax indicates the
percentage of maximum inhibition.
To provide more information on the mode of action of
bibenzyls 6 and 10, their effect on the regulatory protein
those reported.15-19 Compound 31 was newly synthesized,
2+
calmodulin (CaM) was investigated. CaM is a major Ca
-
but it was previously isolated as a natural product from
binding protein implicated in a variety of cell functions
through the regulation of CaM-dependent enzymes, such
as cAMP phosphodiesterases (PDEs), protein phosphatase,
2
0
21
22
23
21
Dendrobium plicatile; compounds 25, 27, 28, 29,
2
4
25
3
0, and 33 have also been described as natural products,
while bibenzyls 26 and 32 were previously obtained using
NOS, phosphorylase kinase, kinase I and II, among oth-
ers.3
0,31
Accordingly, CaM influences a number of important
catalytic reduction and methylation of rhapontigenin and
1
O-methylbatatasin III, respectively. The H NMR data of
biological events, and such agents that inhibit its activity
should have profound pharmacological effects. Indeed,
2
0
24
2
6, 32, obtained by the Wittig reaction in the present
2
6
30,32
investigation, and 31, as well as the complete spectral
data for compounds 342
certain antipsychotic drugs, smooth muscle relaxants,
7,28
29
and 35, were in agreement
R-adrenergic blocking agents, cytoprotective compounds,
and neuropeptides inhibit CaM.3
0-32
To study the effect of
with those previously described. The data presented in
Table 2 show that all analogues are potent inhibitors of
the spontaneous contractions of the guinea-pig ileum, with
IC50’s ranging between 0.14 and 2.36 µM. However, struc-
tural differences influenced the potency of this inhibition.
Thus, methylation of one or more of the free hydroxyl
groups and the presence of additional oxygenated groups
in relation to the lead compounds decreased the smooth
muscle relaxant activity. Compounds 29, possessing a free
hydroxyl group at C-4, and 25, in which the location of the
hydroxyl and methoxyl groups in ring B is the opposite of
the bibenzyls on CaM, a sodium dodecyl sulfate-polyacry-
3
3
lamide gel electrophoresis (SDS-PAGE) analysis was first
carried out. The results showed that both natural products
retarded the mobility of bovine-brain CaM in the presence
of Ca2 , suggesting that they are CaM inhibitors. The
synthetic analogues 23-35 provoked the same electro-
phoretic effect, and, as an example, Figure 4 illustrates the
electrophoretogram showing the activity of compounds 6,
10, 25, 29, and 34. Thereafter, to demonstrate if the
binding of the bibenzyls with CaM affected its enzyme
+

1
64 J ournal of Natural Products, 2004, Vol. 67, No. 2
Hern a´ ndez-Romero et al.
Ta ble 3. Effect of Selected Stilbenoids on CaM-Dependent
PDE in the Presence of 0.2 µg of Bovine-Brain CaM
compound
IC50, µM
chlorpromazinea
1
5
6
10.3
3.2
10.9
7.0
1
2
2
3
3
0
5
9
4
5
13.3
9.0
10.1
36.6
21.2
a
Positive control.
Leung and co-workers.35 The PDE reaction was coupled to
the 5′-nucleotidase reaction, and the amount of inorganic
phosphate released represented the activity of the PDE;
the phosphate produced in the assay was measured by the
method of Sumner.36 Bovine-brain CaM was used as
activator of the enzyme. The results summarized in Table
3
indicated that the compounds tested inhibited the activa-
tion of PDE in the presence of CaM with IC50 values
ranging between 3.2 and 36.6 µM. The effect was higher
than or comparable to that of chlorpromazine (IC50 ) 10.3
µM), a well-known CaM inhibitor.3
0-32
Therefore the biben-
zyls tested are CaM inhibitors. Furthermore, when the
smooth muscle relaxant effect of compounds 6 and 10 was
determined in the presence of chlorpromazine (0.1 µM),
their concentration-response curves (Figure 5) were sig-
nificantly shifted to the left. These results are consistent
with chlorpromazine and bibenzyls 6 and 10 being agonists.
Since CaM regulates NOS activity in the myenteric plexus,
it is highly probable that the nitrergic and the CaM-smooth
muscle relaxant effects demonstrated in this investigation
are related. Further work is in progress to confirm this
hypothesis.
Compounds 1 and 5 were also evaluated as potential
CaM inhibitors. The natural products not only modified the
electrophoretic mobility of bovine-brain CaM (Figure 4) but
also inhibited the activity of CaM-dependent PDE with IC50
values of 3.2 and 10.9 µM, respectively. In the case of
compound 5 this effect could be also involved in its
spasmolytic effect.
F igu r e 3. Concentration-response curves showing the relaxant
effects of compounds 6 and 10 on the isolated guinea-pig ileum in the
presence of ODQ (100 µM) and L-NAME (500 µM). Values are expressed
as the percentages of inhibition of contractile responses calculated as
the mean from six data ( SEM, p < 0.05.
To our knowledge the carbocyclic skeleton of nidemone
(1) is new for a naturally occurring product. However, the
4
′-oxo-3′-hydro derivative spiro[cyclopentane-1,2′(1′H)-
naphthalene], which possesses a similar structural core,
has been obtained through the condensation of tetralone
with 1,4-dibromobutane.37 The coexistence of spiro-type
natural products with bibenzyls and phenanthrenes has
been previously described in Cannabis sativa L., although,
the spiro-compounds found in Cannabis possess an inverse
3
8
F igu r e 4. SDS-PAGE of bovine-brain CaM after treatment with
compounds 1, 5, 6, 10, 25, 29, 34, and 35. Electrophoresis of 2 µg
arrangement of the five- and six-membered rings, as the
structure of cannabispiradienone (36) shows in Chart 3. It
has been proposed that the C. sativa spirans originate by
a direct oxidative phenol coupling (p-o fashion) of an
appropriate bibenzyl.38 In the case of compound 1 such a
mechanism is not possible. However, it can be envisaged
that nidemone (1) arises by an oxidative contraction of an
aromatic ring of a suitable dihydrophenanthrene. This
proposal is supported by the fact that one of the several
routes to generate cyclopentane rings from aromatic com-
pounds can be rationalized as an oxidative biological
Bayer-Villiger type of reaction. The overall results seem
to indicate that the orchid bibenzyls exert their spasmolytic
action by inhibiting CaM-mediated processes and/or by a
nitrergic mechanism. Whether both effects are related
remains to be determined. For maximum spasmolytic
samples of bovine-brain CaM in the presence of 1 mM CaCl
2
. Pre-
treatment of the CaM samples: 1.5 h at 30 °C in the presence of CaCl
2
(A); chlorpromazine in DMSO (B); 1 (C); 5 (D); 6 (E); 10 (F); 25 (G); 29
(
H); 34 (I); and CaM in DMSO (J ). In all cases, 0.33 µg of the tested
compounds in DMSO was applied.
regulatory properties in vitro, their effect on the activity
of CaM-dependent PDE was investigated. CaM-sensitive
PDE catalyzes the hydrolysis of cyclic nucleotides to
nucleotide monophosphates. Since CaM-sensitive PDE is
widely used as a tool to discover CaM inhibitors, and in
general to demonstrate the activity of CaM in biochemical
studies, we assessed the effect of 6, 10, the analogues 25,
2
9, and 34, and 35 on this enzyme using the method of
3
4
Sharma and Wang, with the modifications reported by

Stilbenoids from Nidema boothii
J ournal of Natural Products, 2004, Vol. 67, No. 2 165
ments were recorded in CDCl on a Varian Unity Plus 500
3
1
13
spectrometer either at 500 ( H) or 125 ( C) MHz, using
tetramethylsilane (TMS) as an internal standard. Electron-
impact mass spectra were registered on a J EOL SX 102 mass
spectrometer. Open column chromatography was carried out
on silica gel 60 (70-230 mesh, Merck). Analytical and pre-
parative TLC were performed on precoated silica gel 60 F254
plates (Merck). HPLC was carried out with a Waters HPLC
instrument equipped with Waters 996 UV photodiode array
detector (900) set at 209-214 nm, using a µPorasil column
(
19 mm i.d. × 300 mm) at a flow rate of 8.3 mL/min. Control
of the equipment, data acquisition, processing, and manage-
ment of chromatographic information were performed by the
Millennium 2000 software program (Waters). In all cases
purification was achieved using hexane-CHCl -i-PrOH-
3
MeOH (75:23:1:1) as mobile phase.
P lan t Mater ial. The whole plant was collected in Catemaco,
State of Veracruz, Mexico, in October 1996. An authenticated
voucher specimen (G. Carmona-D ´ı az-96-2) is preserved at the
Instituto de Ecolog ´ı a Herbarium (XAL), Xalapa, Veracruz.
Extr a ction a n d Isola tion . The air-dried plant material
(
2 kg) was ground into a powder and extracted exhaustively
by maceration at room temperature with a mixture of MeOH-
CHCl (1:1). After filtration, the extract was concentrated in
3
vacuo to yield 174 g of a brown residue. The extract was
subjected to column chromatography over silica gel (1 kg) and
eluted with a gradient of hexane-EtOAc (10:0 f 0:10) and
EtOAc-MeOH (10:0 f 5:5). Three hundred fractions (1 L each)
were collected and combined to produce 11 pooled fractions
(F-I to F-XI) based on their TLC profiles. According to the
pharmacological testing, F-V was the only fraction that
inhibited (99%) the spontaneous contraction of the guinea-pig
ileum when tested at the IC50 of the original extract.
Active fraction F-V (6.0 g, eluted with hexane-EtOAc 8:2)
was chromatographed on a silica gel column (80 g) using a
gradient of hexane-EtOAc (10:0 f 0:10) to yield eight second-
ary fractions (FV-1 to FV-8). The most active fraction was FV-8
(
600 mg) (100% of inhibition of the spontaneous ileum contrac-
tion) eluted with hexane-EtOAc (8:2). HPLC purification
(hexane-CHCl -i-PrOH-MeOH, 75:23:1:1) of active fraction
FV-8 afforded nidemone (1, 4 mg, t
F igu r e 5. Concentration-response curves showing the relaxant
effects of compounds 6 and 10 on the isolated guinea-pig ileum in the
presence of chlorpromazine (0.1 µM). Values are expressed as the
percentages of inhibition of contractile responses calculated as the
mean from six data ( SEM, p < 0.05.
3
7
R
14.2 min); aloifol II (2,
R
2 mg, t 18 min); 1,5,7-trimethoxy-9,10-dihydrophenanthrene-
2
2
2
,6-diol (3, 25 mg, t
,6-diol (4, 8 mg, t
R
19 min); 1,5,7-trimethoxyphenanthrene-
2
22
R
23 min); ephemeranthoquinone (5, 5.7
Ch a r t 3. Structure of Cannabispiradienone (36)
2
mg, t
R
26 min); gigantol (6, 32 mg, t
R
30 min); ephemeranthol
2
2
6
B
(7, 4 mg, t
8, 3 mg, t 35 min); lusianthridin (9, 55 mg, t
batatasin III (10, 19 mg, t
R
32 min); 2,4-dimethoxyphenanthrene-3,7-diol
8
(
R
R
50 min); and
62 min). From inactive fraction
1
6
R
III crystallized the known triterpenoid nidemin, identical to
1
a standard sample. Finally, from inactive fraction X sponta-
neously crystallized a yellow powder characterized (IR, NMR,
MS) as vitexin.9
Nid em on e (1): vitreous solid; [R] -114° (c 1.6, MeOH);
D
6
5
CD (MeOH) ∆ꢀ (nm) -1.96 × 10 (221), -6 × 10 (246), 9.3 ×
5
6
5
1
(
1
(
0
(265), -1 × 10 (290), -6 × 10 (326); UV (MeOH) λmax
log ꢀ) 343 (3.54), 296 (3.10), 226 (4.30); IR (KBr) νmax 1692,
activity, the bibenzyls should have oxygenated substituents
on both aromatic rings; methylation of free hydroxyl groups
as well as the increment of oxygenated groups in relation
to compounds 6 and 10 decreases the resultant smooth
muscle relaxant activity.
-1
1
13
626, 1596, 1452, 1420, 1358, 1158, 992 cm ; H and C NMR
+
see Table 1); EIMS m/z 258 [M (78)], 241 (12), 229 (10), 226
15), 215 (30), 186 (100), 173 (13), 134 (14); HREIMS m/z
(
2
58.2694 (calcd for C15
14
H O4, 258.2693).
3
,5,7-Tr im eth oxy-9,10-dih ydr oph en an th r en e-2,6-diol (3):
glassy solid; UV (MeOH) λmax (log ꢀ) 265 (3.42), 279 (3.30), 300
Exp er im en ta l Section
(2.44); IR (KBr) νmax 3407, 2928, 1614, 1580, 1503, 1480, 1458,
-1 1
1
363, 1213, 1104, 1047, 995 cm ; H NMR (CDCl
s, D O exchange, -OH), 8.00 (1H, s, D O exchange, -OH), 7.98
1H, d, J ) 8.7 Hz, H-4), 6.88 (1H, d, J ) 8.7 Hz, H-3), 6.59 (s,
H-8), 3.92 (3H, s, CH O-7), 3.81 (3H, s, CH O-1), 3.70 (3H, s,
CH O-5), 2.80 (2H, m, H-9), 2.69 (2H, m, H-10); C NMR
CDCl ) δ 147.3 (C-1), 145.8 (C-7), 144.8 (C-5), 143.6 (C-2),
3
) δ 9.00 (1H,
Gen er a l Exp er im en ta l P r oced u r es. Melting points were
determined on a Fisher-J ohns apparatus and are uncorrected.
Optical rotations were taken on a Perkin-Elmer 241 polarim-
eter. UV spectra were obtained on a Lambda II UV spectrom-
eter in MeOH solution. IR spectra were obtained using KBr
disks on a Perkin-Elmer FT 1605 spectrophotometer. The CD
spectrum of nidemone (1) was recorded on a J ASCO 720
spectropolarimeter at 25 °C in MeOH solution. NMR spectra
including COSY spectra, NOESY, HMBC, and HMQC experi-
2
2
(
3
3
13
3
(
3
137.6 (C-9a), 130.0 (C-4a), 129.0 (C-10a), 125.8 (C-6), 123.9
(C-4), 117.9 (C-4b) 113.1 (C-3), 106.8 (C-8), 29.4 (C-9), 22.5 (C-
+
10); EIMS m/z 302 [M (100)], 287 (15), 255 (23), 244 (5);
HREIMS m/z 302.3217 (calcd for C17
18
H O5, 302.3218).

1
66 J ournal of Natural Products, 2004, Vol. 67, No. 2
Hern a´ ndez-Romero et al.
1204, 1150, 1067, 831 cm- ; H NMR (CDCl
) δ 7.2 (1H, dd,
1
1
Syn th etic In ter m ed ia tes. Resveratrol, piceatannol, 3,4-
3
dimethoxylbenzaldehyde (14), benzaldehyde (15), and m-
anisaldehyde (16) were purchased from Sigma (St. Louis, MO).
J ) 8.4, 7.8, Hz, H-5′), 6.79 (1H, d, J ) 8.1 Hz, H-6′), 6.76-
6.73 (2H, m, H-2′, H-4′), 6.35 (2H, d, J ) 2.4 Hz, H-2, H-6),
6.31 (1H, dd, J ) 2.4, 2.4 Hz, H-4), 3.76 (3H, s, OMe-3′), 3.74
4
-Benzyloxy-3-methoxybenzaldehyde (11), 3-benzyloxy-4-meth-
1
3
oxybenzaldehyde (12), 3,4-dibenzyloxylbenzaldehyde (13), 3-ben-
zyloxybenzaldehyde (17), 4-benzyloxybenzaldehyde (18), 3-ben-
zyloxy-5-methoxybenzyltriphenylphosphonium bromide (19),
(6H, s, OMe-3, OMe-5), 2.8 (4H, m, H-7, H-7′); C NMR
(CDCl ) δ 160.6 (C-5, C-3), 159.5 (C-3′), 144.0 (C-1), 143.2
3
(C-1′), 129.2 (C-4′), 120.7 (C-5′), 114.1 (C-2′), 111.2 (C-3′), 106.4
(C-2, C-6), 97.9 (C-4), 55.2 (OMe-3, OMe-5), 55.1 (OMe-3′), 38.2
3
3
,5-dibenzyloxybenzyltriphenyl-phosphonium bromide (20),
,5-dimethoxybenzyltriphenyl-phosphonium bromide (21), and
+
(C-7), 37.8 (C-7′); EIMS m/z 272 [M (100)].
benzyltriphenylphosphonium chloride (22) were prepared as
previously described.
3,5-Dih yd r oxybiben zyl (34). Compounds 20 (0.64 g, 0.011
µmol) and 15 (0.05 g, 0.09 µmol) were condensed in the
presence of NaH (0.026 g, 0.011 µmol) to yield a mixture of
the corresponding Z- and E-stilbene (3.29 g, 65%), which upon
catalytic hydrogenation as described for 26 gave 34 (0.05 g,
41%). The spectral data of 34 were consistent with those
reported in the literature.27
1
5-17
Biben zyls 23-25, 28, 29, a n d 33. 4′-Hydroxy-3,3′,5-tri-
methoxybibenzyl (23), 3,3′,4′,5-tetramethoxybibenzyl (24), 3,3′-
dihydroxy-4′,5-dimethoxybibenzyl (25), 3,4′-dihydroxy-5-meth-
oxybibenzyl (29), 3,3′,4′,5-tetrahydroxybibenzyl (28), and 3,4′,5-
trihydroxybibenzyl (33) were prepared as previously de-
1
5-19
scribed.
synthetic materials were identical to those described in the
In all cases, the spectroscopic properties of the
3-Hyd r oxy-4-m eth oxybiben zyl (35). Wittig reaction us-
ing as starting materials 22 (0.230 mg, 0.65 µmol), 12 (0.329
g, 2.4 µmol), and NaH (0.0156 g, 0.65) and applying the same
1
5-19
literature.
-Meth oxy-3,3′,5-tr ih yd r oxybiben zyl (26). Compound 20
1 g, 2 µmol) was dissolved in dry tetrahydrofuran (15 mL)
under a N atmosphere; NaH (0.0814 g, 2 µmol) and 11 (0.591
2
9
4
general strategy as for 26 gave compound 35 (0.067 g, 58%).
(
P h a r m a cologica l Testin g. The pharmacological tests
2
employing the guinea-pig ileum model were performed using
2
,39
g, 1.86 µmol) were added to the mixture, which was stirred
during 3 h. After this period of time, water was added and
the product extracted with EtOAc. The extract was washed
with brine and water, dried, and evaporated to give an oil (1.4
g), which was purified by open column chromatography [silica
gel (30 g), hexane-EtOAc (98:2)] to give a mixture (not
purified) of Z- and E-stilbenes (0.54 g, 55%). The mixture was
two different types of experiments, as previously described.
In the first one, the crude extract, primary fractions, and
natural and synthetic compounds were evaluated for their
ability to relax spontaneous ileal contractions. In the second,
the effect of 6 and 10 on the contractions of the ileum in the
presence of L-NAME (500 µM), ODQ (100 µM), and chlorpro-
mazine (100 µM) was investigated. Male guinea-pigs (600-
800 g) were used. The animals were sacrificed by cervical
dislocation. The ileum was dissected and placed in Krebs-
Henseleit (KH) solution, pH 7.4, with the following composition
2
directly hydrogenated at 45 lb/in. (30 °C) in EtOAc (25 mL)
over 10% palladium on carbon for 3 h. Then the catalyst was
filtered off and the filtrate evaporated. The final product was
purified by HPLC [CHCl
3
-MeOH-i-PrOH (98:1:1)] to yield
(in mM): NaCl 119, KCl 4.6, KH
1.5, NaHCO 20, and glucose 11.4. Strips (1 cm long) were
dissected and mounted in organ baths containing KH solution
gassed with a mixture of 5% CO and 95% O and continuously
2 4 4 2
PO 1.2, MgSO 1.2, CaCl
2
6 as a viscous solid (0.22 g, 27%): IR (KBr) νmax 3350, 1600,
3
-1 1
1
590, 1450, 1320, 1295, 965 cm ; H NMR (C
O exchange, -OH), 7.40 (1H, s, D O exchange, -OH), 6.81
1H, d, J ) 8.1 Hz, H-5′), 6.72 (1H, d, J ) 2.1 Hz, H-2′), 6.63
1H, dd, J ) 8.1, 2.1 Hz, H-6′), 6.22 (2H, d J ) 2.4, Hz, H-2
),
O) δ 159.2 (C-3,
C-5), 147.2 (C-4′), 146.5 (C-3′), 145.1 (C-1), 135.8 (C-1′), 120.0
C-6′), 116.1 (C-2′), 112.4 (C-5′), 107.7 (C-2, C-6), 101.2 (C-4),
3 6
D O) δ 8.10 (2H,
s, D
(
(
2
2
2
2
recorded for isometric tension with a Grass 7D polygraph, as
previously described.39 After a stabilization time of 30 min, a
10 min control period was recorded. The test substances
(extract, chromatographic fractions, and compounds), dissolved
in dimethyl sulfoxide (DMSO), were added to the bath in a
volume of 50 µL at different concentrations (one concentration
was used per ileum segment). All the responses were recorded
for a 10 min period. The effects of the plant crude extract,
fractions, pure compounds, and positive controls were deter-
mined by comparing the areas under the curve (AUC) inscribed
by the frequency and the amplitude of the test materials. Areas
were calculated from the polygraph tracings, using an analog-
digital tablet (CPLAB-10) and specially designed software. All
the results are expressed as the mean of six experiments (
SEM. Concentration-response curves for the extract and pure
compounds were plotted and the experimental data adjusted
by the nonlinear curve-fitting program (PRISMA). The sta-
tistical significance (p < 0.05) of differences between means
was assessed by an analysis of variance (ANOVA) followed by
a Dunnet’s test.3
and H-6), 6.18 (1H, dd, J ) 2.4, Hz, H-4), 3.78 (3H, s, OCH
3
2
.71 (4H. m, H-7 and H-7′); 13C NMR (C
3 6
D
(
+
5
6.7 (-CH
3
O), 38.7 (C-7), 37.6 (C-7′); EIMS m/z 260 [M (100)].
-Hyd r oxy-3′,4′,5-tr im eth oxybiben zyl (27). Condensa-
3
tion of Wittig salt 19 (0.6 g, 1.2 µmol), aldehyde 14 (0.170 g, 1
µmol), and NaH (1.2 µmol) gave, after chromatographic workup
[open silica gel column, hexane-EtOAc (98:2)], the corre-
sponding Z- and E-stilbene mixture (0.239 g, 52%), which upon
hydrogenation and purification as described for compound 26
yielded 27 as a viscous solid (0.12 g, 39%). The spectral data
were consistent with those found in the literature.19
3
′-O-Meth ylba ta ta sin III (30). Phosphonium salt 19 (0.650
g, 1.3 µmol), 16 (0.150 g, 1 µmol), and NaH (0.053 g, 1 µmol)
were condensed to yield the mixture of stilbenes (75% yield,
0
.285 g), which was similarly hydrogenated and purified to
9,40
afford 30 (0.117 g, 41%) as a vitreous solid. The spectral
properties were consistent with those of the natural product
In ter a ction of Com p ou n d s w ith Bovin e-Br a in Ca M.
The interaction of the stilbenoids (natural and synthetics) with
bovine-brain CaM (Sigma) was performed using a denaturing
homogeneous electrophoresis (SDS-PAGE) procedure. The
experiment was carried out according to a previously described
procedure30 using a 15% polyacrylamide gel. The interaction
of the compounds with CaM was evaluated by observing the
2
4
isolated from Coelogyne ovalis.
,3′,5-Tr ih yd r oxybiben zyl (31). Wittig reaction using 12
1.16 g, 2 µmol), 17 (0.336 g, 1.6 µmol), and NaH (0.063 g, 1.6
µmol) as described for 26 gave compound 31 as a white powder
0.124 g, 34%): mp 155 °C; IR (KBr) νmax 3350, 1600, 1590,
3
(
(
1
450, 1320, 1295, 965 cm-1; H NMR (DMSO-d
s, D O exchange, -OH), 9.04 (2H, s, D O exchange, -OH), 7.04
1H, dd, J ) 7.8, 7.8 Hz, H-5′), 6.64-6.55 (3H, m, H-2′, H-4′,
H-6′), 6.07 (dd, J ) 1.5, 1.5 Hz, H-2, H-6), 6.0 (1H, dd, J ) 1.5
1
) δ 9.22 (1H,
6
2
+
2
2
difference in electrophoretic mobility in the presence of Ca .
(
Each electrophoretic run was done in triplicate, and chlorpro-
mazine was used as positive control. The experimental condi-
tions are described in the legend of Figure 3.
1
3
Hz, H-4), 2.63 (4H, m, H-7, H-7′); C NMR (DMSO-d
(
6
) δ 157.9
C-5, C-3), 157.0 (C-3′), 143.3 (C-1), 142.9 (C-1′), 128.9 (C-5′),
P DE Assa y. A PDE assay in the presence of bovine-brain
CaM was performed using a modification of the method
1
18.8 (C-6′), 115.0 (C-2′), 112.6 (C-4′), 106.21 (C-2, C-6), 100.0
+
31
(
C-4), 37.0 (C-7), 36.8 (C-7′); EIMS m/z 230 [M (100)], 212 (10),
23 (80), 107 (83), 77 (20).
,3′,5-Tr im eth oxybiben zyl (32). Wittig reaction using 21
1 g, 2.4 µmol), 16 (0.329 g, 2.4 µmol), and NaH (0.057 g, 2.4
described by Sharma and Wang. Bovine-brain CaM (0.2 µg)
1
was incubated with 0.015 units of CaM-deficient-CaM-depend-
ent PDE from bovine brain (Sigma) for 3 min in 800 µL of
assay solution containing 0.3 units of 5′-nucleotidase (from
Crotalus atrox venom, Sigma), 45 mM Tris-HCl, 5.6 mM Mg-
3
(
µmol) and applying the same general strategy as for 26 gave
bibenzyl 32 (0.397 g, 60.9%): IR (KBr) νmax 1600, 1594, 1454,
3 2 2
(CH COO) , 45 mM imidazole, and 2.5 mM CaCl , pH 7.0. The

Stilbenoids from Nidema boothii
J ournal of Natural Products, 2004, Vol. 67, No. 2 167
(8) Majumder, P. L.; Lahiri, S. Phytochemistry 1990, 29, 621-624.
test compounds were then added to the assay medium at 10,
(9) Wollenweber, E.; Dietz, V. H. Phytochemistry 1981, 20, 869-932.
2
0, 40, 60, 80, and 100 µM in DMSO, and the samples were
incubated for 30 min. Then, 100 µL of 10.8 mM cAMP, pH
.0, was added to start the assay. After 30 min, the assay was
(
10) Furness, M. S.; Robinson, T. P.; Goldsmith, J .; Bowen, J . P. Tetra-
hedron Lett. 1999, 40, 459-462.
7
(11) Demir, A. S.; Sesenoglu, O. Tetrahedron Asymm. 2002, 13, 667-670.
(12) Lightner D. A.; Gurst J . E. Organic Conformational Analysis and
Stereochemistry from Circular Dichroism Spectroscopy; Wiley-VCH:
New York, 2000; pp 337-393.
stopped by the addition of 100 µL of 55% trichloroacetic acid
solution. All of the above steps were carried out at 30 °C. The
PDE reaction was coupled to the 5′-nucleotidase reaction, and
the amount of inorganic phosphate released represented the
activity of the PDE. The phosphate produced in the assay was
measured by the method of Sumner. The wavelength used
for the phosphate assay was 660 nm using a CINTRA 5
spectrophotometer. Chlorpromazine was used as a positive
control (IC50 ) 10.2 µM). The results are expressed as IC50
values that were determined from the analysis of the concen-
tration-effect (inhibition of the enzyme activity) curves, where
each point is the mean ((SEM) of at least three experiments.
The concentration-response graphics were analyzed using a
curve-fitting program (Microcal Origin 6.0, Professional Soft-
ware).
(
13) Snatzke, G. Tetrahedron 1965, 21, 421-438.
(14) Gorham, J . The Biochemistry of the Stilbenoids; Chapman & Hall:
London, 1995; pp 128-133.
3
0
(15) Crombie, L.; J amieson, S. V. J . Chem. Soc., Perkin Trans. 1 1982,
1
467-1475.
(
(
16) Hashimoto, T.; Hasegawa, K. Phytochemistry 1974, 13, 2849-2852.
17) Reimann, E. Chem. Ber. 1969, 102, 2881-2888.
(18) Senyavina, L. B.; Sheichenko, V. I.; Sheinker, Y. N.; Dombrovskii,
A. V.; Shevchuk, M. I.; Barsukov, L. I.; Bergelson, L. D. Z. Obshchei
Khim. 1967, 37, 499-506.
(
19) Mannila, E.; Talvitie, A.; Kolehmainen, E. Phytochemistry 1993, 33,
13-816.
(20) Honda, Ch.; Yamaki, M. Nat. Med. 2001, 55, 68-70.
8
(
21) Crombie, L.; J amieson, S. V. J . Chem. Soc., Perkin Trans. 1 1982,
455-1466.
22) Yasuhiro, T.; Hiroyuki, H.; Tohru, K.; Guojun, X. Chem. Pharm. Bull.
1
(
Molecu la r Mod elin g Ca lcu la tion s. Geometry optimiza-
tions and conformational search were carried out using the
MMFF94 force-field calculations as implemented in the Spar-
tan’02 molecular modeling software from Wavefunction, Inc.
1
991, 39, 593-598.
(23) Yishihiko, I.; Masafumi, O.; Hiroshi, T.; Kimiye, B.; Mitsugi, K.; Hideo,
N. Chem. Pharm. Bull. 1991, 39, 3353-3354.
(
(
(
24) Sachdev, K.; Kulshreshtha, D. K. Phytochemistry 1986, 25, 499-502.
25) El-Feraly, F. S. J . Nat. Prod. 1984, 47, 89-92.
(Irvine, CA).
26) Matsuda, H.; Morikawa, T.; Toguchida, I.; Park, J .-Y.; Harima, S.;
Yoshikawa, M. Bioorg. Med. Chem. 2001, 9, 41-50.
Ack n ow led gm en t. This work was supported by a grant
(27) Asakawa, Y.; Kondo, K.; Tori, M.; Hashimoto, T.; Ogawa, S. Phy-
tochemistry 1991, 30, 1, 219-234.
of DGAPA-UNAM and CONACyT. We wish to thank Marisela
Guti e´ rrez for the IR and UV spectra, and Georgina Duarte
and Oscar Ya n˜ ez for recording the MS and NMR spectra,
respectively. The technical assistance of Isabel Rivero-Cruz
and Laura Acevedo is also acknowledged. We are grateful to
Dr. Carlos Cerda-Garc ´ı a-Rojas for valuable suggestions. Y.H.-
R. acknowledges fellowships from Sistema Nacional de Inves-
tigadores (SEP-CONACyT) and Direcci o´ n General Intercambio
Acad e´ mico (DGEP-UNAM) to carry out graduate studies.
(
28) Takasugi, M.; Kawashima, S.; Monde, K.; Katsui, N.; Masamune, T.;
Shirata, A. Phytochemistry 1987, 26, 37-376.
(29) Scheline, R. R. Experientia 1974, 30, 880-881.
(
30) Ovadi J . In Progress in Drug Research; J ucker, E., Ed.; Birkhauser
Verlag: Basel, 1989; Vol. 33, pp 353-395.
(
31) Hook, S.; Means, A. R. Annu. Rev. Pharmacol. Toxicol. 2001, 41, 471-
505.
(32) Cantabrana, B.; Vallina, J . R. P.; Menendes, L.; Hidalgo, A. Life Sci.
995, 57, 1333-1341.
33) Mac ´ı as, M.; Ulloa, M.; Gamboa, A.; Mata, R. J . Nat. Prod. 2000, 63,
57-758.
1
(
(
7
34) Sharma, R. K.; Wang, J . H. In Advances in Cyclic Nucleotide Research;
Greengard, P., Robinson G. A., Eds.; Raven Press: New York, 1979;
Vol. 10, pp 187-198.
Refer en ces a n d Notes
(
(
(
(
(
(
(
1) Estrada, S.; Acevedo, L.; Rodr ı´ guez, M.; Toscano, R. A.; Mata, R. Nat.
Prod. Lett. 2002, 16, 81-86.
2) Estrada, S.; Rojas, A.; Mathison, Y.; Israel, A.; Mata, R. Planta Med.
(
35) Leung, P. C.; Taylor, W. A.; Wang, J . H.; Tripton, C. L. J . Biol. Chem.
1
984, 259, 2742-2747.
(
(
(
36) Sumner, J . B. Science 1944, 100, 413-415.
1
999, 65, 109-114, and references therein.
3) Estrada, S.; Toscano, R. A.; Mata, R. J . Nat. Prod. 1999, 62, 1175-
178.
4) Tezuka, Y.; Hirano, H.; Kikuchi, T.; Xu, G. J . Chem. Pharm. Bull.
991, 39, 593-598.
5) J ujena, R. K.; Sharma, S. C.; Tandon, J . S. Phytochemistry 1987, 26,
37) Ranu, B. C.; J ana, U. J . Org. Chem. 1999, 64, 6380-6386.
38) Crombie, L.; Crombie, M. L. J . Chem Soc., Perkin Trans. I 1982,
1
1
455-1466.
(
39) Rojas, A.; Cruz, S.; Rauch, V.; Bye, R.; Linares, E.; Mata, R.
Phytomedicine 1995, 2, 51-55.
1
(
40) Bailey, N. T. J . Statistical Methods in Biology; Cambridge University
Press: Cambridge, UK, 1995; pp 234-236.
1
123-1126.
6) Tuchinda, P.; Udchachon, J .; Khumtaveeporn, K.; Taylor, W. C.;
Engelhardt, L. M.; White, A. H. Phytochemistry 1988, 27, 3267-3271.
7) Min, Z.-D.; Tanaka, T.; Iinuma, M.; Mizuno, M. J . Nat. Prod. 1987,
5
0, 1189.
NP030303H