Prenylated coumarins: Natural phosphodiesterase-4 inhibitors from toddalia asiatica

Article abstract of DOI:10.1021/np401040d

Bioassay-guided fractionation of the ethanolic extract of the roots of Toddalia asiatica led to the isolation of seven new prenylated coumarins (1-7) and 14 known analogues (8-21). The structures of 1-7 were elucidated by spectroscopic analysis, and their absolute configurations were determined by combined chemical methods and chiral separation analysis. Compounds 1-5, named toddalin A, 3t′-O-demethyltoddalin A, and toddalins B-D, represent an unusual group of phenylpropenoic acid-coupled prenylated coumarins. Compounds 1-21 and four modified analogues, 10a, 11a, 13a, and 17a, were screened by using tritium-labeled adenosine 3,5-cyclic monophosphate ([³H]-cAMP) as substrate for their inhibitory activity against phosphodiesterase-4 (PDE4), which is a drug target for the treatment of asthma and chronic obstructive pulmonary disease. Compounds 3, 8, 10, 10a, 11, 11a, 12, 13, 17, and 21 exhibited inhibition with IC₅₀ values less than 10 μM. Toddacoumalone (8), the most active compound (IC₅₀ = 0.14 μM), was more active than the positive control, rolipram (IC₅₀ = 0.59 μM). In addition, the structure-activity relationship and possible inhibitory mechanism of the active compounds are also discussed.

Full text of DOI:10.1021/np401040d

Article
pubs.acs.org/jnp
Prenylated Coumarins: Natural Phosphodiesterase‑4 Inhibitors from
Toddalia asiatica
Ting-Ting Lin,† Yi-You Huang,† Gui-Hua Tang, Zhong-Bin Cheng, Xin Liu, Hai-Bin Luo,*
and Sheng Yin*
School of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou, Guangdong 510006, People’s Republic of China
S
* Supporting Information
ABSTRACT: Bioassay-guided fractionation of the ethanolic extract of the roots of Toddalia asiatica led to the isolation of seven
new prenylated coumarins (1−7) and 14 known analogues (8−21). The structures of 1−7 were elucidated by spectroscopic
analysis, and their absolute configurations were determined by combined chemical methods and chiral separation analysis.
Compounds 1−5, named toddalin A, 3‴-O-demethyltoddalin A, and toddalins B−D, represent an unusual group of
phenylpropenoic acid-coupled prenylated coumarins. Compounds 1−21 and four modified analogues, 10a, 11a, 13a, and 17a,
were screened by using tritium-labeled adenosine 3′,5′-cyclic monophosphate ([3H]-cAMP) as substrate for their inhibitory
activity against phosphodiesterase-4 (PDE4), which is a drug target for the treatment of asthma and chronic obstructive
pulmonary disease. Compounds 3, 8, 10, 10a, 11, 11a, 12, 13, 17, and 21 exhibited inhibition with IC50 values less than 10 μM.
Toddacoumalone (8), the most active compound (IC50 = 0.14 μM), was more active than the positive control, rolipram (IC50
=
0.59 μM). In addition, the structure−activity relationship and possible inhibitory mechanism of the active compounds are also
discussed.
been isolated from this plant, some of which exhibited anti-
inflammatory,6 antiplatelet aggregation,7 and antinitric oxide
generation activities.8
he phosphodiesterases (PDEs) are an 11-membered
Tfamily of enzymes that catalyze the hydrolysis of the
secondary signal messengers cyclic adenosine monophosphate
(cAMP) and cyclic guanosine monophosphate (cGMP).1
Phosphodiesterase-4 (PDE4), which specifically catalyzes the
hydrolysis of cAMP, is a therapeutic target of high interest for
central nervous system (CNS), inflammatory, and respiratory
diseases.2 Although a number of chemically diverse molecules
have been developed as PDE4 inhibitors over the last decades,
roflumilast is the sole PDE4 inhibitor recently approved in both
the United States and Europe for the treatment of chronic
obstructive pulmonary disease (COPD).3 As the efficacy of
roflumilast may be restricted by the dose-limiting side effects of
nausea, diarrhea, weight loss, and headaches, the search for
novel PDE4 inhibitors continues unabated.
In our continuing search for PDE4 inhibitors from medicinal
plants, a fraction of the ethanolic extract of T. asiatica showed
an inhibitory activity of 46.2% at a concentration of 10 μM by
using tritium-labeled adenosine 3′,5′-cyclic monophosphate
([3H]-cAMP) as the substrate toward PDE4. Subsequent
chemical investigation led to the isolation of seven new
prenylated coumarins (1−7) together with 14 known analogues
(8−21). Compounds 1−5 represented an unusual group of
phenylpropenoic acid-coupled prenylated coumarins. Com-
pounds 1−21 together with four modified analogues, 10a, 11a,
13a, and 17a, were screened for their inhibitory activity against
PDE4D2, and 10 compounds were identified as PDE4
inhibitors, with IC50 values ranging from 0.14 to 9.98 μM.
Herein, details of the isolation, structural elucidation, inhibitory
Toddalia asiatica (L.) Lam. (Rutaceae), a woody climber,
grows widely in south China. Its barks and roots have been
extensively used in Traditional Chinese Medicine (TCM) for
the treatment of rheumatic arthritis, traumatic injury, and
pyogenic infections.4,5 In the past decades, a number of
prenylated coumarins and benzophenanthridine alkaloids have
Received: December 12, 2013
Published: March 5, 2014
© 2014 American Chemical Society and
American Society of Pharmacognosy
955
dx.doi.org/10.1021/np401040d | J. Nat. Prod. 2014, 77, 955−962

Journal of Natural Products
Article
Chart 1
activities, structure−activity relationship, and possible inhib-
retention times. As the 13C NMR data of 15 were not reported
before, the full NMR spectrum of 6 was assigned in the current
research. Thus, 6 was given the trivial name ent-toddalolactone.
Compound 1 exhibited a molecular formula of C33H38O14 as
determined by 13C NMR data and the HRESIMS ion at m/z
657.2191 [M − H]− (calcd for C33H37O14, 657.2189). The IR
spectrum exhibited absorption bands for OH (3443 cm−1),
ester (1709 cm−1), and benzene ring (1608, 1516, and 1458
itory mechanism of these compounds are described.
RESULTS AND DISCUSSION
■
The air-dried powder of the roots of T. asiatica was extracted
with 95% EtOH at room temperature to give a crude extract,
which was suspended in H2O and successively partitioned with
petroleum ether, EtOAc, and n-BuOH. Various column
chromatographic separations of the EtOAc extract afforded
compounds 1−21.
1
cm−1) functionalities. The H NMR spectrum showed two
methyl singlets [δH 1.24 (H3-5′) and 1.31 (H3-4′)], three
methoxy groups (δH 3.86, 3.87, and 3.93), two trans-olefinic
protons [δH 6.26 (1H, d, J = 16.0 Hz, H-8‴) and 7.57 (1H, d, J
= 16.0 Hz, H-7‴)], two cis-olefinic protons [δH 6.12 (1H, d, J =
9.6 Hz, H-3) and 7.89 (1H, d, J = 9.6 Hz, H-4)], a 1,2,4-
trisubstituted benzene ring [δH 6.82 (d, J = 8.2 Hz, H-5‴), 7.07
(d, J = 8.2 Hz, H-6‴), and 7.20 (s, H-2‴)], and four
oxymethines [δH 3.57, 4.05, 5.21, and 5.28]. The 13C NMR
spectrum in combination with DEPT experiments resolved 33
carbon resonances attributable to three carbonyl carbons, two
benzene rings, two vinylic groups, two oxygenated quaternary
carbons, four oxygenated methines, three methoxy groups,
three methylenes, and two methyls. The collective data implied
Compound 6, a white powder, was isolated as a major
component by using HPLC equipped with a chiral column. The
molecular formula of C16H20O6 was determined by 13C NMR
data and the HRESIMS ion at m/z 331.1152 [M + Na]+ (calcd
331.1158). The 1H NMR spectrum of 6 comprising the
prenylated coumarin features was identical to that of
(+)-toddalolactone (15),9 a major component previously
reported from the same plant. However, the optical rotation
of 6 ([α]2D0 −69) was opposite that of 15 ([α]2D0 +69),10
indicating 6 was the enantiomer of 15. This was supported by
the co-injection of 6 and 15 on HPLC equipped with a chiral
column, which gave two well-resolved peaks with different
956
dx.doi.org/10.1021/np401040d | J. Nat. Prod. 2014, 77, 955−962

Journal of Natural Products
Article
that 1 comprised the structural features of ferulic and quinic
acid moieties linked to compound 6. Analysis of 2D NMR data
permitted structure 1 to be proposed. In particular, HMBC
correlations of H-5″/C-9‴ and H-2′/C-7″ confirmed the
linkage of the three fragments. The structure of 1 was further
confirmed by alkaline hydrolysis, which gave a mixture of three
products. Ferulic and quinic acids were identified by co-TLC of
the reaction mixture with authentic samples, while the
coumarin was assigned as 6 by co-injection of the reaction
mixture with 6 on HPLC equipped with a chiral column.
Compound 1 was given the trivial name toddalin A.
Compound 2 displayed a molecular ion at m/z 667.1991 [M
+ Na]+, consistent with a molecular formula of C32H36O14, 14
mass units less than that of 1. The 1H and 13C NMR data of 2
(Table 1) were similar to those of 1 except for the absence of a
methoxy group, indicating 2 was a demethylated derivative of 1.
Analysis of 2D NMR data led to the proposal of structure of 2,
which revealed the replacement of the ferulic acid unit in 1 by a
caffeic acid moiety in 2. The absolute configuration of 2 was
confirmed by using the same methods as described for 1. Thus,
2 was given the trivial name 3‴-O-demethyltoddalin A.
Table 1. NMR Data for Toddalins A (1) and B (2) in
Methanol-d4 (δ in ppm)
a
1
2
no.
δH (J in Hz)
δC, type
δH (J in Hz)
δC, type
2
163.0, C
113.0, CH
140.8, CH
157.7, C
117.9, C
163.7, C
96.5, CH
156.5, C
108.3, C
163.1, C
113.0, CH
140.8, CH
157.6, C
117.9, C
163.6, C
96.4, CH
156.4, C
108.3, C
24.6, CH2
3
6.12, d (9.6)
7.89, d (9.6)
6.13, d (9.6)
7.89, d (9.6)
4
5
6
7
8
6.64, s
6.63, s
9
10
1′a
1′b
2′
2.92, dd (13.6,
2.3)
24.5, CH2 2.92, d (13.6)
3.20, dd (13.6,
11.2)
3.18, dd (13.6,
11.1)
5.28, dd (11.2,
2.3)
80.2, CH
72.9, C
5.26, d (11.1)
80.3, CH
3′
72.9, C
4′
5′
1″
2″a
2″b
1.31, s
1.24, s
26.0, CH3 1.31, s
26.4, CH3 1.24, s
76.9, C
26.0, CH3
26.4, CH3
76.9, C
Compound 3 was assigned a molecular formula of C32H38O10
as established on the basis of 13C NMR data and the HRESIMS
ion at m/z 605.2341 [M + Na]+ (calcd 605.2357). The IR
spectrum exhibited absorption bands for hydroxy (3445 cm−1),
ester (1709 cm−1), and aromatic (1606, 1514, and 1456 cm−1)
1.77, d (13.6)
37.9, CH2 1.76, d (14.4)
37.9, CH2
1.96, dd (13.6,
2.1)
1.95, dd (14.4,
2.5)
3″
4″
4.05, m
71.5, CH
73.8, CH
4.05, m
71.6, CH
73.7, CH
1
functionalities. The H NMR spectrum showed four methyl
3.57, dd (9.4,
2.6)
3.58, dd (9.3,
2.5)
singlets [δH 1.34 (H3-5′), 1.38 (H3-4′), 1.64 (H3-4‴), and 1.72
(H3-5‴)], four methoxy groups [δH 3.59, 3.71, 3.80, and 3.84],
two trans-olefinic protons [δH 6.65 (d, J = 16.0 Hz, H-3″) and
7.78 (d, J = 16.0 Hz, H-4″)], two cis-olefinic protons [δH 6.10
(d, J = 9.6 Hz, H-3) and 7.73 (d, J = 9.6 Hz, H-4)], two
aromatic singlets [δH 6.21 (s, H-8″) and 6.49 (s, H-8)], and an
olefinic proton [δH 5.09 (t, J = 6.4 Hz, H-2‴)]. The 13C NMR
spectrum, in combination with DEPT experiments, showed 32
carbon resonances attributable to two carbonyl carbons, two
benzene rings, three double bonds, one oxygenated sp3
quaternary carbon, one sp3 oxymethine, four methoxy groups,
two methylenes, and four methyls. The aforementioned data
implied that 3 comprised the structural features of a prenylated
cinnamic acid moiety linked to compound 6. 2D NMR analysis
allowed structure 3 to be postulated as depicted in Figure 1. In
particular, HMBC correlations of H-1‴/C-5″ and C-7″
confirmed the prenyl group at C-6″ of the cinnamic acid
moiety, and HMBC correlations from H-2′ to C-2″ connected
this prenylated cinnamic acid moiety to the C-2′ hydroxy group
of 6. The absolute configuration of 3 was confirmed by co-
injection of the alkaline hydrolysis products of 3 with an
authentic sample of 6 on HPLC equipped with a chiral column.
Thus, 3 was given the trivial name toddalin B.
5″
5.21, ddd (10.0,
9.4, 4.7)
71.6, CH
5.21, ddd (9.8,
9.3, 4.5)
71.6, CH
39.4, CH2
6″a
6″b
2.0, m
39.5, CH2 2.07, m
1.46, dd (12.4,
11.3)
1.52, dd (12.2,
11.4)
7″
1‴
2‴
3‴
4‴
5‴
6‴
7‴
174.2, C
174.2, C
127.7, C
127.7, C
7.20, s
111.7, CH
149.4, C
7.05, s
115.2, CH
146.8, C
150.8, C
150.0, C
6.82, d (8.2)
7.07, d (8.2)
7.57, d (16.0)
6.26, d (16.0)
116.5, CH
124.3, CH
147.0, CH
115.6, CH
168.5, C
6.80, d (8.2)
6.96, d (8.2)
7.51, d (16.0)
6.18, d (16.0)
116.7, CH
123.0, CH
147.1, CH
115.2, CH
168.6, C
8‴
9‴
5-OMe
3.86, s
3.87, s
63.8, CH3 3.85, s
56.8, CH3 3.86, s
56.5, CH3
63.8, CH3
56.8, CH3
7-OMe
3‴-OMe 3.93, s
a
1H and 13C NMR were recorded at 400 and 100 MHz, respectively.
Compound 4, a colorless oil, gave the molecular formula
C32H40O12, as determined by 13C NMR and HRESIMS data.
The 1H and 13CNMR data (Table 2) of 4 were similar to those
of 3, with notable differences being the absence of an olefinic
proton in 3 and the presence of two additional oxygenated
carbons (δC 78.7 and 73.0) in 4. HMBC correlations from two
methyl singlets to the two oxygenated carbons along with the
molecular formula suggested that the prenyl group was
dihydroxylated in 4. 2D NMR analysis established the
molecular structure of 4. Alkaline hydrolysis of 4 followed by
acid cyclization yielded only a major product, which was
identified as 6 by comparison with an authentic sample on a
Figure 1. Key HMBC (H→C) and 1H−1H COSY () correlations of
1 and 3.
957
dx.doi.org/10.1021/np401040d | J. Nat. Prod. 2014, 77, 955−962

Journal of Natural Products
Article
a
Table 2. NMR Data for Compounds 3, 4, and 5 in CDCl3 (δ in ppm)
3
4
5
no.
δH (J in Hz)
6.10, d (9.6)
δC, type
δH (J in Hz)
6.10, d (9.6)
δC, type
δH (J in Hz)
6.13, d (9.6)
δC, type
2
161.6, C
111.9, CH
139.1, CH
156.1, C
117.0, C
162.1, C
95.3, CH
155.0, C
106.9, C
24.0, CH2
161.4, C
112.0, CH
139.1, CH
156.2, C
117.0, C
162.1, C
95.4, CH
155.1, C
107.0, C
24.1, CH2
161.3, C
112.1, CH
138.9, CH
156.1, C
116.9, C
162.1, C
95.4, CH
155.1, C
106.9, C
24.2, CH2
3
4
7.73, d (9.6))
7.75, d (9.6)
7.75, d (9.6)
5
6
7
8
6.49, s
6.50, s
6.52, s
9
10
1′a
2.95, dd (13.6, 2.5)
2.94, dd (13.5, 2.2)
2.95, dd (13.7, 2.8)
1′b
2′
3′
3.15, dd (13.6, 10.1)
5.30, dd (10.1, 2.5)
3.15, dd (13.5, 10.1)
5.30, dd (10.1, 2.2)
3.16, dd (13.7, 10.3)
5.31, dd (10.3, 2.8)
78.3, CH
72.9, C
78.4, CH
72.8, C
78.2, CH
72.8, C
4′
5′
2″
3″
4″
5″
6″
7″
8″
9″
10″
1‴a
1‴b
2‴
3‴
4‴
5‴
1.38, s
1.34, s
25.1, CH3
26.5, CH3
168.8, C
117.0, CH
137.4, CH
159.3, C
115.6, C
160.5, C
95.7, CH
156.9, C
108.1, C
22.4, CH2
1.38, s
1.34, s
25.2, CH3
26.6, CH3
168.5, C
117.8, CH
137.0, CH
159.8, C
112.7, C
160.3, C
96.1, CH
157.5, C
108.4, C
25.7, CH2
1.38, s
1.34, s
25.2, CH3
26.7, CH3
168.4, C
117.7, CH
136.9, CH
159.8, C
109.5, C
160.1, C
96.0, CH
157.9, C
108.3, C
35.7, CH2
6.65, d (16.0)
7.78, d (16.0)
6.59, d (16.0)
7.67, d (16.0)
6.58, d (16.0)
7.70, d (16.0)
6.21, s
6.15, s
6.20, s
3.70, s
3.20, d (6.4)
5.09, t (6.4)
2.57, dd (13.6, 3.3)
2.80, dd (13.6, 10.1)
3.50, m
123.3, CH
131.0, C
78.7, CH
73.0, C
214.2, C
2.78, m
40.5, CH
18.5, CH3
18.5, CH3
63.3, CH3
56.2, CH3
61.6, CH3
55.4, CH3
1.64, s
1.72, s
3.84, s
3.80, s
3.59, s
3.71, s
25.6, CH3
17.7, CH3
63.2, CH3
56.2, CH3
61.5, CH3
55.5, CH3
1.27, s
1.26, s
3.84, s
3.82, s
3.60, s
3.67, s
26.0, CH3
23.8, CH3
63.3, CH3
56.2, CH3
61.5, CH3
55.6, CH3
1.16, d (6.9)
1.16, d (6.9)
3.84, s
5-OMe
7-OMe
5″-OMe
7″-OMe
3.83, s
3.54, s
3.61, s
a
1H and 13C NMR were recorded at 400 and 100 MHz, respectively.
chiral column. Thus, the absolute configuration of 4 was
depicted as shown, and the compound was named toddalin C.
Compound 5 was obtained as a colorless oil. The molecular
formula of 5 was found to be C32H38O11 by 13C NMR data and
the HRESIMS ion at m/z 621.2303 [M + Na]+ (calcd
81.9) in 7 compared to 6 (δC 72.8). Acid hydrolysis of 7
generated a coumarin/glucose mixture. The coumarin was
confirmed as 6 by co-injection of the reaction mixture with an
authentic sample on HPLC equipped with a chiral column. The
D-configuration of the glucosyl unit was determined by HPLC
analysis.11,12 Thus, the structure of 7 was defined as
(−)-toddalolactone 3′-O-β-D-glucopyranoside.
1
621.2306). The H and 13C NMR data (Table 2) of 5 were
similar to those of 3 except for the replacement of a prenyl
group in 3 by a 3-methylbutan-2-one group in 5. This was
confirmed by HMBC correlations from the two methyl
doublets [δH 1.16 (6H, d, J = 6.9 Hz, H-4‴ and H-5‴)] to a
methine [δC 40.5] and a carbonyl carbon [δC 214.2]. The
absolute configuration of 5 was defined by using the same
method as for 3. Compound 5 was given the trivial name
toddalin D.
The known compounds toddacoumalone (8),13 toddalosin
(9),9,14 5-methoxyseselin (10),15 braylin (11),16 norbraylin
(12),17 toddaculin (13),9 toddanone (14),9 toddalolactone
(15),9 5,7,8-trimethoxycoumarin (16),18 coumurrayin (17),9
gleinadiene (18),19 cis-dehydrocoumurrayin (19),20 toddale-
none (20),9 and toddacoumaquinone (21)21 were identified by
comparison of their observed and reported NMR data.
The HRESIMS data of compound 7 showed a molecular ion
at m/z 493.1674 [M + Na]+ (calcd 493.1680), which was 162
mass units more than that of 6. The 1H and 13C NMR data of 7
were similar to those of 6 except for the presence of additional
signals attributable to a β-glucopyranose residue [δH 4.56 (d, J
= 7.7 Hz, H-1″) and δC 98.6 (C-1″)], indicating 7 was a
glucosylated derivative of 6 or 15. The glucopyranose moiety
was located at C-3′ by HMBC correlation from H-1″ to C-3′,
and this was supported by the deshielded C-3′ resonance (δC
To enhance the structural diversity of the prenylated
coumarin library for subsequent screening, compounds 10,
11, 13, and 17 were modified to the corresponding hydro-
genated analogues, 5-methoxydihydroseselin (10a),22 3′,4′-
dihydrobraylin (11a),23 5,7-dimethoxy-6-(3-methylbutyl)-
coumarin (13a),9 and 5,7-dimethoxy-8-(3-methylbutyl)-
coumarin (17a),9 respectively. The library was screened for
inhibitory activity against PED4D2 by using our reported
methods.24−27 Rolipram, a well-known PDE4 inhibitor, was
958
dx.doi.org/10.1021/np401040d | J. Nat. Prod. 2014, 77, 955−962

Journal of Natural Products
Article
used as the reference compound (IC50 0.59 μM), comparable
to the reported value of 1 μM.1 The bioassay results showed
that compounds 3, 8, 10, 10a, 11, 11a, 12, 13, 17, and 21 had
strong activity, with IC50 values less than 10 μM toward
PDE4D2 (Table 3). The inhibitory curves of the two most
active compounds (8 and 11, IC50 = 0.14 and 0.96 μM,
respectively) are represented in Figure 2.
It was found that compounds with an angular tricyclic system
derived from the coumarin or quinolone core coupled with a
pyran moiety exhibited the strongest inhibition, as exemplified
by compounds 8, 10, 10a, 11, 11a, and 12; compounds with a
nonoxygenated or nonconjugated prenyl moiety also showed
good activity, e.g., 13, 13a, 17, and 17a. Thus, toddacoumalone
(8), possessing both an angular tricyclic system and an
unmodified prenylated moiety, showed the most potent
activity. It is noteworthy that oxidation or conjugation of the
prenyl moiety caused a significant decrease of the activity, such
as in 1, 2, 4, 5, 14, 18, and 20.
Table 3. IC50 Values of the Active Compounds against
PDE4D2
compound
IC50 (μM)
compound
IC50 (μM)
To further explore their inhibitory mechanism, the binding
modes of 8, 11, and 12 with PDE4D were simulated by using
the molecular docking approach CDOCKER.28 The reliability
of this method was validated by the root-mean-square deviation
(RMSD) values for the top 10 redocked poses of roflumilast,
which ranged from 0.8 to 1.5 Å relative to the crystal
counterpart. It is considered that a successful docking holds the
2
16.65 1.20
7.81 0.40
0.14 0.02
1.87 0.12
2.20 0.27
0.96 0.10
1.53 0.23
12
2.38 0.14
9.98 0.63
11.49 0.95
6.82 0.34
10.22 0.15
5.51 0.15
0.59 0.05
3
13
8
13a
17
10
10a
11
11a
17a
21
a
rolipram
́
RMSD value of the optimum pose below a threshold of 1 Å in
a
reference to the crystal pose.22,23 Under identical conditions,
compounds 8, 11, and 12 were docked into the PDE4D
catalytic pocket, and the resulting poses for each ligand were
ranged according to the “−CDOCKER_INTERACTION_E-
NERGY” scores. Several poses with high scores were further
determined by the common scheme of inhibitors binding to
PDE4, that is, the hydrophobic interactions and hydrogen bond
interactions formed between the conserved residues (Phe372
and Gln369) and ligands.29
Positive control.
As shown in Figure 3, although these three active compounds
comprised a similar angular tricyclic feature, the binding
patterns of the simple coumarins 11 and 12 and compound 8
were different due to their reversed poses of the pyran ring in
the tricyclic system. Compounds 11 and 12 could form two key
hydrogen bonds with Gln369 via the oxygen atoms at C-6 and
C-7 (2.9 Å/3.4 Å and 2.9 Å/3.2 Å, respectively) and generate
Figure 2. Inhibitory curves of compounds 8, 11, and rolipram
(positive control) against PDE4D2.
Figure 3. Binding modes of compounds 8, 11, and 12 with PDE4D derived from docking simulations (red dashed lines for hydrogen bond and
yellow dashed lines for stacking interaction, respectively). (A) Binding mode of compound 8. (B) The similar binding patterns of 11 (orange), 12
(yellow), and roflumilast (cyan). (C) Binding mode of compound 11. (D) Binding mode of compound 12.
959
dx.doi.org/10.1021/np401040d | J. Nat. Prod. 2014, 77, 955−962

Journal of Natural Products
Article
Extraction and Isolation. The air-dried powder of the roots of T.
asiatica (1 kg) was extracted with 95% EtOH (3 × 10 L) at room
temperature (rt) to give 85 g of crude extract. The extract was
suspended in H2O (1 L) and successively partitioned with petroleum
ether (PE, 3 × 1 L) and EtOAc (3 × 1 L), respectively. The EtOAc
extract (63 g) was subjected to MCI gel CC eluted with a MeOH/
H2O gradient (3:7 → 10:0) to afford three fractions (I−III). Fraction I
(10.5 g) was subjected to silica gel CC (PE/EtOAc, 2:1 → 0:1) to give
three fractions (Ia−Ic). Fr. Ia (2.1 g) was separated by silica gel CC
(PE/EtOAc, 2:1), followed by semipreparative HPLC equipped with a
chiral column (CH3OH/H2O, 7:3, 3 mL/min), to give 6 (83 mg) and
15 (52 mg). Fr. Ic (5.3 g) was separated by Rp-C18 silica gel CC
(MeOH/H2O, 6:4 → 10:0) to yield 1 (65 mg), 2 (70 mg), and 7 (200
mg). Fraction II (36.5 g) was subjected to silica gel CC (PE/CHCl3,
2:1 → 0:1) to give three fractions (IIa−IIc). Fr. IIa (4.1 g) was
separated by Rp-C18 silica gel CC (MeOH/H2O, 7:3 → 10:0)
followed by Sephadex LH-20 CC using EtOH to give 10 (34 mg), 12
(21 mg), 13 (48 mg), 17 (37 mg), and 19 (54 mg). Fr. IIb (16.5 g)
was subjected to silica gel CC (PE/CHCl3, 1:1 → 0:1) to give three
fractions (IIb1−IIb4). Fr. IIb1 (2.9 g) was subjected to Rp-C18 CC
(MeOH/H2O, 6:4 → 10:0), followed by silica gel CC (PE/acetone,
12:1 → 0:1), to afford 11 (31 mg) and 14 (22 mg). Fr. IIb2 (3.0 g)
was applied to silica gel CC (PE/EtOAc, 8:1 → 0:1) and Sephadex
LH-20 eluted with CHCl3/MeOH, 1:1, to yield 16 (12 mg) and 18
(22 mg). Fr. IIb3 (4.6 g) was subjected successively to silica gel CC
(PE/EtOAc, 4:1 → 0:1), Rp-C18 CC (MeOH/H2O, 6:4 → 10:0), and
Sephadex LH-20 (EtOH) chromatography to yield 8 (32 mg), 9 (22
mg), 20 (17 mg), and 21 (15 mg). Fr. IIb4 (3.2 g) was subjected to
Rp-C18 CC (MeOH/H2O, 6:4 → 10:0), silica gel CC (PE/acetone,
6:1 → 1:2), and Sephadex LH-20 (EtOH) to yield 3 (50 mg), 4 (10
mg), and 5 (6 mg).
favorable stacking interactions with Phe372 via the 1-
benzopyran-2-one ring system (4.1 Å/5.1 Å/16° and 2.9 Å/
3.2 Å/18°, respectively, Figure 3), which shared a similar
binding pattern to roflumilast (Figure 3B), as shown in the
crystal structure of 1XOQ.29 Interestingly, compound 8 forms
only one hydrogen bond with Gln369 via the ester carbonyl
group. However, the stacking interactions might be the
predominant forces contributing to the binding of 8 with
PDE4. For the coumarin moiety, it could form favorable
interactions with hydrophobic residue Phe340 (4.7 Å and 79°)
apart from interacting with the conserved Phe372 (4.4 Å and
7.8°). For the angular tricylic system, it could form two extra
favorable stacking interactions with Phe372 (4.4 Å/4.9 Å and
35°), which might explain its relatively high inhibitory
potencies despite the lack of one hydrogen bond. The absence
of the pyran ring in 13, 13a, 17, and 17a decreased the stacking
interactions between the ligands and Phe372, which led to a
more moderate activity of this group of compounds, while their
side-chain-modified analogues (14−16 and 18−20) lost the
activity probably due to the steric effects caused by the
oxygenated or conjugated prenyl tails (see Supporting
Information).
Natural PDE4 inhibitors are rare, and the current study
revealed a new group of PDE4 inhibitors from T. asiatica, which
may explain the anti-inflammatory efficacy of this plant in
Traditional Chinese Medicine. It is possible that the peculiar
prenylated coumarin features confer on these compounds
potent PDE4 inhibitory activity, which makes them promising
lead structures for the development of PDE4 inhibitors. Studies
toward their selectivity versus other PDE members are in
progress.
Toddalin A (1): colorless oil; [α]2D0 −150 (c 0.1, MeOH); UV
(MeOH) λmax (log ε) 207 (4.71), 222 (4.46), 327 (4.52) nm; IR
1
(KBr) νmax 3443, 1709, 1608, 1516, 1458 cm−1; H and 13C NMR
data, see Table 1; ESIMS m/z 657.2 [M − H]−, HRESIMS m/z
657.2191 [M − H]− (calcd for C33H37O14, 657.2189).
3‴-O-Demethyltoddalin A (2): colorless oil; [α]2D0 −140 (c 0.2,
MeOH); UV (MeOH) λmax (log ε) 207 (4.75), 222 (4.52), 329 (4.48)
EXPERIMENTAL SECTION
■
General Experimental Procedures. Optical rotations were
measured on a Perkin-Elmer 341 polarimeter. UV spectra were
recorded on a Shimadzu UV-2450 spectrophotometer. IR spectra were
determined on a Bruker Tensor 37 infrared spectrophotometer with
KBr disks. NMR spectra were measured on a Bruker AM-400
spectrometer at 25 °C. ESIMS and HRESIMS were carried out on a
Finnigan LC QDECA instrument. A Shimadzu LC-20 AT equipped with
an SPD-M20A PDA detector was used for HPLC, and a YMC-pack
ODS-A column (250 × 10 mm, S-5 μm, 12 nm) was used for
semipreparative HPLC separation. A chiral column (Phenomenex Lux,
cellulose-2, 250 × 10 mm, 5 μm) was used for chiral separation. Silica
gel (300−400 mesh, Qingdao Haiyang Chemical Co. Ltd.), C18
reversed-phase silica gel (Rp-C18) (12 nm, S-50 μm, YMC Co.
Ltd.), Sephadex LH-20 gel (Amersham Biosciences), and MCI gel
(CHP20P, 75−150 μm, Mitsubishi Chemical Industries Ltd.) were
used for column chromatography (CC). All solvents used were of
analytical grade (Guangzhou Chemical Reagents Company, Ltd.).
Expression and purification of PDE4D were carried out by using a
Hielscher UP200S ultrasonic cell disruption processor, a Sigma 6K15
centrifugal machine, an Eppendorf BioPhotomer spectrophotometer,
and a Qiagen nickel-nitriloacetic acid (Ni-NTA) column. The
radioactivity of the samples was measured on a PerkinElmer Tricarb
2910 liquid scintillation counter. The yeast extract and tryptone
prepared for LB medium were purchased from Oxoid Ltd., and the
substrate [3H]-cAMP was from Waukesha GE Healthcare. Other
reagents such as ampicillin and rolipram were purchased from Sigma.
Plant Material. Roots of T. asiatica were collected in October 2012
in Yunnan Province, P. R. China, and were authenticated by Prof. You-
Kai Xu of Xishuangbanna Tropical Botanical Garden, Chinese
Academy of Sciences. A voucher specimen (accession number:
FLZX201210) has been deposited at the School of Pharmaceutical
Sciences, Sun Yat-sen University.
1
nm; IR (KBr) νmax 3444, 1708, 1610, 1515, 1458 cm−1; H and 13C
NMR data, see Table 1; ESIMS m/z 643.2 [M − H]−; HRESIMS m/z
667.1991 [M + Na]+ (calcd for C32H36O14Na, 667.1997).
Toddalin B (3): colorless oil; [α]2D0 −116 (c 0.1, CHCl3); UV
(CHCl3) λmax (log ε) 250 (4.29), 325 (4.59) nm; IR (KBr) νmax 3445,
1
1709, 1606, 1514, 1456 cm−1; H and 13C NMR data, see Table 2;
ESIMS m/z 581.3 [M − H]−, HRESIMS m/z 605.2341 [M + Na]+
(calcd for C32H38O10Na, 605.2357).
Toddalin C (4): colorless oil; [α]2D0 −92 (c 0.1, CHCl3); UV
(CHCl3) λmax (log ε) 252 (4.16), 323 (4.45) nm; IR (KBr) νmax 3446,
1
1709, 1615, 1454 cm−1; H and 13C NMR data, see Table 2; ESIMS
m/z 615.1 [M − H]−; HRESIMS m/z 639.2432 [M + Na]+ (calcd for
C32H40O12Na, 639.2412).
Toddalin D (5): colorless oil; [α]2D0 −107 (c 0.1, CHCl3); UV
(CHCl3) λmax (log ε) 251 (4.05), 324 (4.34) nm; IR (KBr) νmax 3445,
1
1707, 1606, 1455 cm−1; H and 13C NMR data, see Table 2; ESIMS
m/z 597.2 [M − H]−; HRESIMS m/z 621.2303 [M + Na]+ (calcd for
C32H38O11Na, 621.2306).
ent-Toddalolactone (6): white powder; [α]2D0 −69 (c 0.1, CHCl3);
UV (CHCl3) λmax (log ε) 255 (3.94), 329 (4.11) nm; IR (KBr) νmax
3446, 1610 cm−1; 1H NMR (CDCl3, 400 MHz) δH 7.79 (1H, d, J = 9.6
Hz, H-4), 6.58 (1H, s, H-8), 6.17 (1H, d, J = 9.6 Hz, H-3), 3.84 (3H, s,
7-OCH3), 3.83 (3H, s, 5-OCH3), 3.57 (1H, d, J = 10.2, 2.0 Hz, H-2′),
2.85 (1H, dd, J = 13.6, 2.0 Hz, H-1′b), 2.72 (1H, dd, J = 13.6, 10.2 Hz,
H-1′a), 2.57 (OH, s), 2.32 (OH, s), 1.25 (3H, s, H-4′), 1.24 (3H, s, H-
5′); 13C NMR (CDCl3, 100 MHz) δC 160.9 (C, C-2), 112.4 (CH, C-
3), 138.7 (CH, C-4), 155.9 (C, C-5), 117.9 (C, C-6), 161.5 (C, C-7),
95.5 (CH, C-8), 154.8 (C, C-9), 107.1 (C, C-10), 26.0 (CH2, C-1′),
77.7 (CH, C-2′), 72.8 (C, C-3′), 26.0 (CH3, C-4′), 23.5 (CH3, C-5′);
ESIMS m/z 309.1 [M + H]+; HRESIMS m/z 331.1152 [M + Na]+
(calcd for C16H20O6Na, 331.1158).
960
dx.doi.org/10.1021/np401040d | J. Nat. Prod. 2014, 77, 955−962

Journal of Natural Products
Article
(−)-Toddalolactone 3′-O-β-D-glucopyranoside (7): colorless oil;
[α]2D0 −44 (c 0.1, MeOH); UV (MeOH) λmax (log ε) 206 (4.53), 225
to those we described previously.21−24 More details about the
experimental procedures are provided in the Supporting Information.
Molecular Modeling. The crystal structure of the catalytic domain
of human PDE4D2 with bound roflumilast (PDB code: 1XOQ29) was
used here for the docking studies. The crystallographic water
molecules were removed except those coordinated with the two
metal ions Mg2+ and Zn2+. Hydrogen atoms and charges were added to
the systems using the CHARMm force field and the Momany-rone
partial charge method, which were implemented in Accelrys Discovery
Studio 2.5.5.30 All ionizable residues in the systems were set to their
protonation states at a neutral pH. The bound roflumilast was used as
a reference compound to define the active site of PDE4 into which the
active compounds were docked by using CDOCKER.28 The radius of
the input site sphere was set as 10 Å from the center of the binding
site, and 10 random conformations were generated for each ligand.
Other docking parameters were set to default values.
1
(4.16), 329 (4.07) nm; IR (KBr) νmax 3444, 1609 cm−1; H NMR
(CDCl3, 400 MHz) δH 8.02 (1H, d, J = 9.6 Hz, H-4), 6.75 (1H, s, H-
8), 6.23 (1H, d, J = 9.6 Hz, H-3), 4.56 (1H, d, J = 7.7 Hz, H-1″), 3.91
(3H, s, 7-OCH3), 3.90 (3H, s, 5-OCH3), 3.82 (1H, m, H-6″b), 3.82
(1H, dd, J = 10.2, 2.5 Hz, H-2′), 3.65 (1H, m, H-6″a), 3.40 (1H, m, H-
3″), 3.30 (1H, m, H-5″), 3.29 (1H, m, H-4″), 3.21 (1H, m, H-2″),
2.91 (1H, m, H-1′b), 2.78 (1H, dd, J = 13.6, 2.5 Hz, H-1′a), 1.36 (3H,
s, H-5′), 1.36 (3H, s, H-4′); 13C NMR (CDCl3, 100 MHz) δC 163.4
(C, C-2), 112.6 (CH, C-3), 141.2 (CH, C-4), 157.7 (C, C-5), 120.2
(C, C-6), 163.7 (C, C-7), 96.3 (CH, C-8), 156.2 (C, C-9), 108.4 (C,
C-10), 27.2 (CH2, C-1′), 77.4 (CH, C-2′), 81.9 (C, C-3′), 21.8 (CH3,
C-4′), 23.9 (CH3, C-5′), 98.6 (CH, C-1″), 75.2 (CH, C-2″), 78.1
(CH, C-3″), 71.7 (CH, C-4″), 77.7 (CH, C-5″), 62.7 (CH2, C-6″);
ESIMS m/z 493.1 [M + Na]+; HRESIMS m/z 493.1674 [M + Na]+
(calcd for C22H30O11Na, 493.1680).
ASSOCIATED CONTENT
Acid Hydrolysis of 7 and Determination of the Absolute
Configuration of Sugar and Aglycone. Compound 7 (2 mg) was
refluxed with 2 mL of 2 M HCl (dioxane/H2O, 1:1) for 4 h. After
removing the dioxane under vacuum, the solution was diluted with
H2O and extracted with EtOAc (3 × 1 mL). The aqueous layer was
evaporated under vacuum, diluted repeatedly with H2O, evaporated
under vacuum to obtain a neutral residue, and analyzed by TLC on
silica gel (acetone/n-BuOH/H2O, 6:3:1) with an authentic sugar
sample (D-glucose, Rf = 0.49). The remaining residue was dissolved in
pyridine (200 μL), to which 2 mg of L-cysteine methyl ester
hydrochloride was added. The mixture was stirred for 1 h; 50 μL of o-
tolyl isothiocyanate was added, and the mixture was stirred at 60 °C
for another 1 h. The mixture was directly analyzed by standard C18
HPLC [a YMC-pack ODS-A column (250 × 10 mm, S-5 μm, 12 nm),
CH3CN/H2O, 25:75, 3 mL/min]. The peak (tR = 19.0 min) coincided
with a derivative of D-glucose, as compared with authentic D-glucose
with tR at 19.1 min. In addition, the EtOAc layer was evaporated under
vacuum to get the corresponding toddalolactone fragment, which was
analyzed by HPLC equipped with a chiral column (CH3OH/H2O,
70:30, 3 mL/min). The S absolute configuration of the aglycone of 7
was confirmed by comparison of the retention time (tR = 13.5 min) of
the segments with that of (−)-toddalolactone in a similar way
[(+)-toddalolactone tR = 19.5 min and (−)-toddalolactone tR = 13.5
min].
Hydrolysis of Compounds 1−5 and Determination of the
Absolute Configuration of Constituent Units. Compounds 1−3
and 5 (each 2 mg) were refluxed with 4 mL of 1 M NaOH (MeOH/
H2O, 3:1) for 2 h. After cooling, each solution was neutralized with 1
mL of 1 M HCl, and the resin was removed by filtration. The filtrate
was extracted with EtOAc to obtain the corresponding toddalolactone
fragments, which were analyzed by HPLC equipped with a chiral
column (CH3OH/H2O, 70:30, 3 mL/min). The S absolute
configuration of the corresponding toddalolactone segments was
confirmed by comparison of the retention time (tR = 13.5 min) of the
■
S
* Supporting Information
This material (1D and 2D NMR spectra of 1−7, 1D NMR
spectra of 8−21, expression and purification of PDE4D2, and
enzymatic assays) is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Authors
*(H.-B. Luo) E-mail: luohb77@mail.sysu.edu.cn.
*(S. Yin) Tel: +86-20-39943090 or +86-20-39943031. Fax:
+86-20-39943090. E-mail: yinsh2@mail.sysu.edu.cn.
Author Contributions
†T.-T. Lin and Y.-Y. Huang contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank the National Natural Science Foundation of
China (Nos. 81102339, 21103234, and 81373258), Guangdong
Natural Science Foundation (Nos. S2011040002429,
S2011030003190, and S2013010014867), and the Fundamen-
tal Research Funds for the Central Universities (11ykzd05) for
providing financial support to this work. We cordially thank
Prof. H. Ke in Department of Biochemistry and Biophysics at
University of North Carolina, Chapel Hill, for his help with the
expression, purification, and enzymatic assay of PDE4.
REFERENCES
■
segments with that of (−)-toddalolactone [(+)-toddalolactone tR
=
(1) Bender, A. T.; Beavo, J. A. Pharmacol. Rev. 2006, 58, 488−520.
(2) Burgin, A. B.; Magnusson, O. T.; Singh, J.; Witte, P.; Staker, B. L.;
Bjornsson, J. M.; Thorsteinsdottir, M.; Hrafnsdottir, S.; Hagen, T.;
Kiselyov, A. S.; Stewart, L. J.; Gurney, M. E. Nat. Biotechnol. 2010, 28,
63−72.
(3) Fabbri, L. M.; Beghe, B.; Yasothan, U.; Kirkpatrick, P. Nat. Rev.
́
Drug Discovery 2010, 9, 761−762.
19.5 min and (−)-toddalolactone tR = 13.5 min]. Compound 4 (2 mg)
was hydrolyzed under the above-mentioned basic conditions; then 4
mL of 2 M HCl was added. The mixture was refluxed until the
phenylpropenoic acid moiety of 4 was converted to 6 as monitored by
HPLC analysis.
Preparation of 11a, 12a, 13a, and 17a. To solutions of 10 (8.2
mg), 11 (5.8 mg), 13 (10 mg), and 17 (7.5 mg) in MeOH (4 mL) was
added 10% Pd/C (0.8 mg). The mixtures were stirred under H2 at rt
for 1 h. The catalyst was filtered off, and the resulting products were
subjected to flash chromatography eluting with MeOH/CH2Cl2 (20:1
→ 10:1) to give 10a (6 mg), 11a (4.5 mg), 13a (7 mg), and 17a (5
(4) Shi, L.; Li, D.; Kang, W. Y. J. Chin. Pharm. 2011, 22, 666−668.
(5) Chu, D. H. J. Anhui Agric. Sci. 2010, 38, 19374−19375.
(6) Hao, X. Y.; Peng, L.; Ye, L.; Huang, N. H.; Shen, Y. M. J. Chin.
Integr. Med. 2004, 2, 450−452.
(7) Tsai, I. L.; Wun, M. F.; Teng, C. M.; Ishikawa, T.; Chen, I. S.
Phytochemistry 1998, 48, 1377−1382.
1
mg), respectively. Analytical data: H NMR spectra of compounds
10a, 11a, 13a, and 17a (see Supporting Information) were identical to
reported data;9,22,23 ESIMS m/z 10a (261.1 [M + H]+), 11a (261.1
[M + H]+), 13a (277.1 [M + H]+), and 17a (277.1 [M + H]+).
PDE4D Inhibitory Screening Assays. The protocols for
expression, purification, and enzymatic assays of PDE4D2 were similar
(8) Nakamura, T.; Kodama, N.; Oda, M.; Tsuchiya, S.; Arai, Y.;
Kumamoto, T.; Ishikawa, T.; Ueno, T.; Yano, S. J. Nat. Med. 2009, 63,
15−20.
(9) Ishii, H.; Kobayashi, J. I.; Ishikawa, M.; Haginiwa, J.; Ishikawa, T.
Yakugaku Zasshi 1991, 111, 365−375.
961
dx.doi.org/10.1021/np401040d | J. Nat. Prod. 2014, 77, 955−962

Journal of Natural Products
Article
(10) Ishii, H.; Kobayashi, J. I.; Sakurada, E.; lshikawa, T. J. Chem. Soc,
Perkin Trans. 1 1992, 13, 1681−1684.
(11) Tanaka, T.; Nakashima, T.; Ueda, T.; Tomii, K.; Kouno, I.
Chem. Pharm. Bull. 2007, 55, 899−901.
(12) Chan, H. H.; Hwang, T. L.; Reddy, M. V. B.; Li, D. T.; Qian, K.;
Bastow, K. F.; Lee, K. H.; Wu, T. S. J. Nat. Prod. 2011, 74, 796−802.
(13) Ishii, H.; Kobayashi, J. I.; Ishikawa, T. Tetrahedron Lett. 1991,
32, 6907−6910.
(14) Ishii, H.; Kobayashi, J. I.; Yamaguchi, K. I.; Ishikawa, T. Chem.
Pharm. Bull. 1993, 41, 1655−1656.
(15) Xie, L.; Takeuchi, Y.; Cosentino, L. M.; Lee, K. H. J. Med. Chem.
1999, 42, 2662−2672.
(16) Randrianarivelojosia, M.; Mulholland, D. A.; Mc Farland, K.
Biochem. Syst. Ecol. 2005, 33, 301−304.
(17) Wei, H.; Zeng, F. J.; Lu, M. Y.; Tang, R. J. Acta Pharm. Sin.
1998, 33, 688−692.
(18) Maes, D.; Riveiro, M. E.; Shayo, C.; Davio, C.; Debenedetti, S.;
De Kimpe, N. Tetrahedron 2008, 64, 4438−4443.
(19) Kumar, V.; Reisch, J.; Wickremaratne, D. B. M.; Hussain, R. A.;
Adesina, K. S.; Balasubramaniam, S. Phytochemistry 1987, 26, 511−
514.
(20) Kinoshita, T.; Firman, K. Chem. Pharm. Bull. 1996, 44, 1261−
1262.
(21) Ishii, H.; Kobayashi, J. I.; Seki, H.; Ishikawa, T. Chem. Pharm.
Bull. 1992, 40, 1358−1359.
(22) Trost, B. M.; Toste, F. D.; Greenman, K. J. Am. Chem. Soc. 2003,
125, 4518−4526.
(23) Um, B. H.; Lobstein, A.; Weniger, B.; Caroline, S.; Yice, F.;
Rakotoarison, O.; Andriantsitohaina, R.; Anton, R. Fitoterapia 2003,
74, 638−642.
(24) Chen, S. K.; Zhao, P.; Shao, Y. X.; Li, Z.; Zhang, C. X.; Liu, P.
Q.; He, X. X.; Luo, H.-B.; Hu, X. P. Bioorg. Med. Chem. Lett. 2012, 22,
3261−3264.
(25) Li, Z.; Cai, Y. H.; Cheng, Y. K.; Lu, X.; Shao, Y. X.; Li, X. S.; Liu,
M.; Liu, P. Q.; Luo, H.-B. J. Chem. Inf. Model. 2013, 53, 972−981.
(26) Zhao, P.; Chen, S. K.; Cai, Y. H; Lu, X.; Li, Z.; Cheng, Y. K.;
Zhang, C. X.; Hu, X. P; He, X. X.; Luo, H.-B. Biochim. Biophys. Acta
Proteins Proteom. 2013, 1834, 2089−2096.
(27) Zhong, J. S.; Huang, Y. Y.; Ding, W. J.; Wu, X. F.; Wan, J. Z.;
Luo, H.-B. Fitoterapia 2013, 91, 159−165.
(28) Wu, G.; Robertson, D. H.; Brooks, C. L.; Vieth, M. J. Comput.
Chem. 2003, 24, 1549−1562.
(29) Card, G. L.; England, B. P.; Suzuki, Y.; Fong, D.; Powell, B.; Lee,
B.; Luu, C.; Tabrizizad, M.; Gillette, S.; Ibrahim, P. N.; Artis, D. R.;
Bollag, G.; Milburn, M. V.; Kim, S. H.; Schlessinger, J.; Zhang, K. Y. J.
Structure 2004, 12, 2233−2247.
(30) Accelrys Discovery Studio, version 2.5.5; Accelrys Software Inc.:
San Diego, CA, 2010.
NOTE ADDED AFTER ASAP PUBLICATION
■
This paper was published ASAP on March 5, 2014, with an
error to Compound 17a in the Results and Discussion section.
The corrected version was reposted on April 2, 2014.
962
dx.doi.org/10.1021/np401040d | J. Nat. Prod. 2014, 77, 955−962