Structural characterization and assessment of the cytotoxicity of 2,3-dihydro-1H-indene derivatives and coumarin glucosides from the bark of Streblus indicus

Article abstract of DOI:10.1021/acs.jnatprod.6b00306

A pair of enantiomers and a pair of 2,3-dihydro-1H-indene epimers, rac-indidene A (rac-1), indidenes B and C (2, 3); four new coumarin glucosides (4-7); and four known coumarin glucosides (8-11) were isolated from the bark of Streblus indicus (Bur.) Corner. The structures of 1-11 were defined by physical data analyses, including MS, NMR, and single-crystal X-ray diffraction. The absolute configurations of the 2,3-dihydro-1H-indene derivatives were defined via experimental and calculated ECD data. rac-Indidene A and indidenes B and C showed inhibitory activity against A549 and MCF-7 tumor cells with IC₅₀ values in the range of 2.2 ± 0.1 to 7.2 ± 0.9 μM.

Full text of DOI:10.1021/acs.jnatprod.6b00306

Article
pubs.acs.org/jnp
Structural Characterization and Assessment of the Cytotoxicity of
2,3-Dihydro‑1H‑indene Derivatives and Coumarin Glucosides from
the Bark of Streblus indicus
Ruijie He,†,‡ Xishan Huang,§ Yanjun Zhang,† Liangdeng Wu,† Hui Nie,† Dexiong Zhou,† Buming Liu,⊥
Shengping Deng,† Ruiyun Yang,† Shuai Huang,† Zhijie Nong,† Jun Li,*,† and Yan Huang*,⊥
†State Key Laboratory for the Chemistry and Molecular Engineering of Medicinal Resources, College of Chemistry and Pharmacy,
Guangxi Normal University, Guilin 541004, People’s Republic of China
‡Guangxi Key Laboratory of Functional Phytochemicals Research and Utilization, Guangxi Institute of Botany, Guilin 541006,
People’s Republic of China
§School of Chemistry and Chemical Engineering, Sun Yat-sen University, Guangzhou 510275, People’s Republic of China
⊥Guangxi Key Laboratory of Traditional Chinese Medicine Quality Standards, Nanning 530022, People’s Republic of China
S
* Supporting Information
ABSTRACT: A pair of enantiomers and a pair of 2,3-dihydro-
1H-indene epimers, rac-indidene A (rac-1), indidenes B and C
(2, 3); four new coumarin glucosides (4−7); and four known
coumarin glucosides (8−11) were isolated from the bark of
Streblus indicus (Bur.) Corner. The structures of 1−11 were
defined by physical data analyses, including MS, NMR, and
single-crystal X-ray diffraction. The absolute configurations of
the 2,3-dihydro-1H-indene derivatives were defined via
experimental and calculated ECD data. rac-Indidene A and
indidenes B and C showed inhibitory activity against A549 and
MCF-7 tumor cells with IC50 values in the range of 2.2 0.1
to 7.2 0.9 μM.
RESULTS AND DISCUSSION
Streblus indicus (Bur.) Corner, a perennial arbor of the
Moraceae family, is widespread in South Asia, particularly in
southern China.1 The bark of S. indicus, “Zhi Xie Shu Pi” in
Chinese, is often used for hemostasis and the treatments of
inflammation and various rheumatoid diseases.2,3 The S. indicus
plant contains large amounts of coumarins, which have been
demonstrated as bioactive substances corresponding to the
traditional application of this plant.2,3 The coumarin derivatives
have also been reported to have potent cytotoxic activities,4−6
prompting further investigation of more structurally attractive
or bioactive coumarin derivatives from this plant. As part of an
ongoing project to identify the bioactive metabolites from this
genus,7−18 the extracts of S. indicus were examined, and the
results of bioactive screening indicated that the EtOAc extract
exhibited significant cytotoxic inhibitory activity of less than 50
μg/mL. These findings led to the isolation of four 2,3-dihydro-
1H-indene derivatives and eight coumarin glucosides. To date,
no 2,3-dihydro-1H-indene derivatives have been obtained from
prior phytochemical work on Streblus asper plants.2,19 In the
present study, the isolation and structural elucidation of eight
new compounds and cytotoxic assessment of these isolates are
reported in detail.
■
The EtOAc fractionation and purification of the 75% EtOH
extract of S. indicus bark using various chromatographic
methods yielded four 2,3-dihydro-1H-indene derivatives and
eight coumarin glucosides. The structures of 1−7 are shown in
Chart 1. The four known coumarin glucosides 8−11 were
identified as 7-O-(6-O-syringoyl-β-D-glucopyranosyl)-6-me-
thoxycoumarin,20 skimmin,21 scopolin,22 and 7-O-(6-O-sinapo-
yl-β-D-glucopyranosyl)-6-methoxycoumarin,20 respectively.
Indidene A (1) was obtained as colorless crystals (MeOH)
25
with [α]D 0.01 (c 0.4, MeOH). A sodium adduct ion at m/z
395.14560 [M + Na]+ (calcd for C21H24O6Na, 395.14706)
observed in the positive-ion HRESIMS indicated a molecular
1
formula of C21H24O6. The H NMR data (Table 1) revealed a
1,2,4-trisubstituted aromatic moiety at δH 6.47 (d, J = 2.1 Hz,
H-3‴), 6.42 (dd, J = 8.5, 2.1 Hz, H-5‴), and 7.67 (d, J = 8.5 Hz,
H-6‴); two ortho-coupled aromatic protons at δH 6.55 (d, J =
8.5 Hz, H-6′) and 6.44 (d, J = 8.5 Hz, H-7′); one methoxy
group at δH 3.84 (s, 2‴-OMe); two methyl groups at δH 1.06 (s,
Me-1″) and 1.04 (s, Me-3″); two methylene groups at δH 2.98
(dd, J = 17.0, 8.9 Hz, H-3′a), 2.82 (dd, J = 17.0, 3.7 Hz, H-3′b),
Received: April 7, 2016
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.6b00306
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
The structure of indidene A was confirmed using X-ray
crystallographic diffraction (Figure 2). The centrosymmetric
Chart 1. Structures of 1−7
space group P1 (n = 2) indicated a racemic mixture, congruent
̅
with the near zero optical rotation. The chiral HPLC separation
of rac-1 using a Chiralpak AD-H column revealed the presence
of two enantiomers, (+)-1 and (−)-1, with opposite Cotton
effects (Figure 3) and opposite optical rotations. To determine
the absolute configurations of (+)-1 and (−)-1, the ECD
spectra of (+)-1 and (−)-1 were measured in MeOH and
compared with the calculated ECD spectra of the enantiomers.
The experimental ECD spectrum of (+)-1 was consistent with
the calculated ECD spectrum of (1′R, 2′S)-1. Thus, the
absolute configurations of (+)-1 and (−)-1 were elucidated as
1′R,2′S and 1′S,2′R, respectively. The complete structures of
(+)-indidene A (1) and (−)-indidene A (1) were established as
depicted in Chart 1.
Indidene B (2) was isolated as pale red powder. The positive-
ion HRESIMS data (m/z, 425.15618 [M + Na]+, calcd for
C22H26O7Na, 425.15762) suggested a molecular formula of
C22H26O7, that is, one more CH2O unit than 1. Comparison
between the NMR data of 2 and 1 (Table 1) revealed that the
4-hydroxy-2-methoxyphenyl group at C-1 in 1 was replaced by
a methoxy group in 2, and the 4-hydroxy-2-methoxyphenyl
group shifted to C-7′ in 2. This was confirmed by the
HRESIMS data and the correlations of the methoxy protons
(δH 3.51) with the ester carbonyl carbon (δC 175.0), H-6‴ (δH
6.90) with C-7′ (δC 127.9), and H-6′ (δH 6.45) with C-1‴ (δC
122.4) in the HMBC spectrum (Figure 1), as well as the
chemical shifts of the CO group and C-7′ that shifted from
δC 202.5 and 115.8 in 1 to δC 175.0 and 127.9 in 2, respectively.
Therefore, the 2D structure of 2 was established. Based on the
ROESY correlations (Figure 1) of H2-2 with H3-1″/3″, the C-
1′/C-2′ relative configuration was determined as cis. The
absolute configuration of 2 was determined via the
experimental and calculated ECD data. The atropisomerism
of 2 was investigated by a 1D potential energy surface scan of
the C-6′−C-7′−C-1‴−C-6‴ dihedral angle. The minimum
conversion energy for P and M atropisomers was less than 20
kcal/mol (Figure S49, Supporting Information), suggesting that
2 exists as a single compound with interconverting P and M
atropisomers at room temperature.23 The experimental and
calculated ECD data permitted assignment of the absolute
configuration of the C-1′ and C-2′ stereogenic centers of 2 as
1′S,2′S (Figure 4). Thus, the structure of indidene B (2) was
established as methyl 2-[(1S,2S)-4,5-dihydroxy-7-(4-hydroxy-2-
methoxyphenyl)-2-(2-hydroxypropan-2-yl)-2,3-dihydro-1H-
inden-1-yl]acetate.
Based on the HRESIMS data analysis, the molecular formula
of indidene C (3) was established as C22H26O7, the same as
that of indidene B (2). It exhibited NMR data similar to that of
indidene B (2) (Table 1), suggesting that 3 shared the same 2D
structure with 2. The ROESY correlations of H-1′ with H3-1″/
3″ defined the 1′,2′-trans relative configuration of 3. Similar to
2, the minimum energy for the conversion of P and M
atropisomers of 3 was also below 20 kcal/mol (Figure S50,
Supporting Information), indicating the fast interconversion of
its P and M conformers. The experimental and calculated ECD
data (Figure 5) suggested that the absolute configuration of 3
was 1′S,2′R. Therefore, the structure of indidene C (3) was
established as methyl 2-[(1S,2R)-4,5-dihydroxy-7-(4-hydroxy-2-
methoxyphenyl)-2-(2-hydroxypropan-2-yl)-2,3-dihydro-1H-
inden-1-yl]acetate.
and δH 3.26 (m, H-2); and two methine groups at δH 3.64 (td, J
= 6.8, 3.4 Hz, H-1′) and 2.26 (td, J = 8.9, 3.7 Hz, H-2′). The
13C NMR data (Table 1) displayed 21 carbon resonances that
were identified by HSQC as nine nonprotonated carbons (one
ketocarbonyl carbon, seven olefinic carbons, and one sp3
oxygen-bearing carbon); seven methine carbons (five olefinic
carbons and two sp3 carbons); two sp3 methylene carbons; and
1
three methyl carbons. The H−1H COSY cross-peaks (Figure
1) of H2-2/H-1′/H-2′/H2-3′ and the HMBC correlations of H-
7′/C-1′; Ha-3′/C-4′, C-8′, C-9′, C-2″; and H-2′/C-8′, C-2″, C-
9′ indicated that C-1′, C-2′, and C-3′ were linked to C-8′, C-2″,
and C-9′, respectively. The H-1′/H-6‴ correlations with the
carbonyl carbon (δC 202.5) in the HMBC spectrum indicated
that C-2 and C-1‴ were linked via a carbonyl carbon (Figure
1). The connectivity of the methoxy group with C-2‴ was
demonstrated by the methoxy protons (δH 3.84) /C-2‴ (δC
163.0) and H-5‴ (δH 6.42)/C-4‴ (δC 165.0) HMBC
correlations. The HMBC correlations of H-1″/C-2′, C-2″, C-
3″ and H-3″/C-2′, C-1″, C-2″ revealed the linkages of C-1″ and
C-3″ with C-2″. The hydroxy groups at C-4′, C-5′, C-2″, and
C-4‴ were verified via the HRESIMS data and supported by
the chemical shifts of C-4′ (δC 141.9), C-5′ (δC 144.7), C-2″
(δC 74.0), and C-4‴ (δC 165.0). Based on these results, the
structure of compound 1, indidene A, was defined as 2-[4,5-
dihydroxy-2-(2-hydroxypropan-2-yl)-2,3-dihydro-1H-inden-1-
yl]-1-(4-hydroxy-2-methoxyphenyl)ethanone.
B
DOI: 10.1021/acs.jnatprod.6b00306
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 1. NMR Data (500 MHz, Methanol-d4, δ in ppm) of Indidenes A−C (1−3)
indidene A (1)
δH (J in Hz)
indidene B (2)
δH (J in Hz)
indidene C (3)
δH (J in Hz)
position
δC, type
δC, type
δC, type
CO
202.5, C
175.0, C
175.0, C
2a
52.8, CH2
3.26, m
40.63, CH2
2.16, dd (15.1, 3.4)
2.06, dd (15.1, 10.3)
3.47, ddd (10.3, 3.4, 1.6)
2.22, dt (5.3, 1.6)
2.98, d (5.3)
40.60, CH2
2.14, dd (15.0, 3.5)
2.02, dd (15.0, 10.2)
3.47, ddd (10.2, 3.5, 1.5)
2.22, dt (5.3, 1.5)
2.96, d (5.3)
2b
1′
2′
3′a
3′b
4′
5′
6′
7′
8′
9′
1″
2″
3″
1‴
2‴
3‴
4‴
44.2, CH
57.0, CH
31.2, CH2
3.64, td (6.8, 3.4)
2.26, dt (8.9, 3.7)
2.98, dd (17.0, 8.9)
2.82, dd (17.0, 3.7)
44.61, CH
56.0, CH
30.8, CH2
44.57, CH
56.0, CH
30.8, CH2
141.9, C
144.7, C
115.2, CH
115.8, CH
140.6, C
130.7, C
27.5, CH3
74.0, C
140.9, C
144.9, C
117.3, CH
127.9, C
137.8, C
130.6, C
27.0, CH3
74.1, C
140.9, C
144.9, C
117.3, CH
127.9, C
137.8, C
130.6, C
27.0, CH3
74.1, C
6.55, d (8.5)
6.44, d (8.5)
6.45, s
6.43, s
1.06, s
1.04, s
1.13, s
1.04, s
1.13, s
1.04, s
26.45, CH3
120.9, C
163.0, C
99.9, CH
165.0, C
109.0, CH
133.9, CH
56.0, CH3
26.1, CH3
122.4, C
158.7, C
100.1, CH
159.0, C
108.0, CH
132.6, CH
55.6, CH3
51.9, CH3
26.1, CH3
122.4, C
158.7, C
100.1, CH
159.0, C
108.0, CH
132.6, CH
55.6, CH3
51.9, CH3
6.47, d (2.1)
6.50, d (1.7)
6.48, d (1.7)
5‴
6‴
CH3O-2‴
CH3O-
6.42, dd (8.5, 2.1)
7.67, d (8.5)
3.84, s
6.42, dd (8.1, 1.7)
6.90, d (8.1)
3.73, s
6.41, dd (8.1, 1.7)
6.89, d (8.1)
3.72, s
3.51, s
3.51, s
1
Figure 1. HMBC (arrows), H−1H COSY (bonds), and ROESY (double arrows) correlations of 1−3.
Figure 2. ORTEP drawing of 1.
The molecular formula of indidene D (4) was C24H24O12
based on the sodium adduct ion at m/z 527.11545 [M + Na]+
(calcd for C24H24O12Na 527.11655) in its HRESIMS.
Comparison of its NMR data (Tables 2 and 3) with those of
7-O-(6-O-syringoyl-β-D-glucopyranosyl)-6-methoxycoumarin
(8)20 revealed that the C-6 methoxy signal (δH 3.79, δC 56.1) in
C
DOI: 10.1021/acs.jnatprod.6b00306
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 2. 1H NMR Data (500 MHz, DMSO-d6, δ in ppm) for
a
Indidenes D−G (4−7)
indidene D
indidene E
indidene F
indidene G
(4)
(5)
(6)
(7)
position
δH (J in Hz)
δH (J in Hz)
δH (J in Hz)
δH (J in Hz)
3
6.30, d (9.5)
7.95, d (9.5)
7.50, d (8.6)
6.31, d (9.5)
7.93, d (9.5)
7.27, s
6.32, d (9.5)
7.99, d (9.5)
7.31, s
6.25, d (9.5)
7.89, d (9.5)
7.27, s
4
5
6
Figure 3. Experimental ECD spectra of (+)-1 and (−)-1 and
calculated ECD spectrum of (1′R,2′S)-1.
7.00, dd (8.6,
2.3)
8
7.04, d (2.3)
5.18, d (7.5)
3.31*
7.21, s
7.27, s
7.16, s
1′
2′
3′
4′
5.26, d (7.4)
3.34, m
5.24, d (7.4)
3.36, m
5.18, d (7.3)
3.32*
3.20, m
3.35, m
3.36, m
3.38, m
3.26, dd (8.8, 3.29 dd (8.5,
4.9) 4.9)
3.25, dd (9.1, 3.13, dd (8.6,
5.2) 4.8)
5′
3.89, ddd (8.8, 3.92, ddd (8.5, 3.95, ddd (9.1, 3.80, ddd (8.6,
7.1, 1.7) 6.3, 1.7) 8.0, 1.6) 7.0, 1.7)
6′a
6′b
4.63, d (11.8) 4.60, d (12.0) 4.66, d (11.8) 4.31, d (11.8)
4.19, dd (11.8, 4.24, dd (12.0, 4.08, dd (11.8, 4.07, dd (11.8,
7.1)
6.3)
8.0)
7.0)
Figure 4. Experimental ECD spectrum of 2 and calculated ECD
spectrum of (1′S,2′S)-2.
2″
5″
6″
7.16, d (2.0)
7.16, d (2.0)
7.44, d (2.0)
7.19, d (8.4)
6.39, d (2.0)
7.16, d (2.0)
7.16, d (2.0)
7.57, dd (8.4, 6.39, d (2.0)
2.0)
7″
8″
1‴
2‴
2.70, m
2.61, m
5.11, d (7.3)
5.05, d (7.4)
4.80, d (7.4)
3.22, dd (9.4, 3.28, dd (8.6, 3.20, m
5.3) 5.1)
3‴
3.06, dd (9.4, 3.39, dd (8.6, 3.02, ddd (9.0,
5.3) 5.1) 5.4, 2.0)
4‴
3.14, ddd (9.4, 3.25, dd (9.1, 3.19, m
8.7, 1.9) 5.1)
a
5‴
3.20, dd (8.7, 3.30, dd (9.3, 3.35
4.8) 3.5)
Figure 5. Experimental ECD spectrum of 3 and calculated ECD
spectrum of (1′S,2′R)-3.
6‴a
6‴b
CH3O-6
3.55, dd (11.7, 3.53, dd (9.3, 3.59, dd (9.8,
5.6)
5.9)
4.8)
3.40, dd (11.7,
5.6)
3.43, m
8 was replaced by an aromatic proton signal (δH 7.00) in 4. The
molecular formula, the 32.8 ppm upfield shift of C-6 (δC
113.2), the 10.1 ppm downfield shift of C-7 (δC 159.7), and the
correlation of H-6 (δH 7.00) with C-7 (δC 159.7) in the HMBC
spectrum (Figure 6) supported these changes in 4. The
hydrolysis of 4 in 1 M HCl yielded D-glucose, as determined by
TLC and GC analyses (see Experimental Section). The
coupling constant between H-1′ and H-2′ of 7.5 Hz indicated
the β-anomeric configuration for 4. The structure of 4 was
therefore established as 7-O-(6-O-syringoyl-β-D-
glucopyranosyl)coumarin and named indidene D.
Indidene E (5) gave a sodium adduct ion at m/z 719.17989
[M + Na]+ (calcd for C31H36O18Na, 719.17993) in the positive-
ion HRESIMS, which suggested a molecular formula of
C31H36O18. The NMR data (Tables 2 and 3) of 5 were highly
similar to those of 7-O-(6-O-syringoyl-β-D-glucopyranosyl)-6-
methoxycoumarin (8),20 except for the additional signals for
one glucosyl moiety at δH 5.11 and δC 102.6, 74.2, 77.3, 69.8,
76.6, and 60.7. The shielded C-4″ resonance (δC 138.7),
deshielded C-3″/C-5″ resonances (δC 152.2), and the
correlation of Glc-H-1‴ (δH 5.11) with C-4″ (δC 138.7) in
the HMBC spectrum (Figure 6) supported this conclusion.
The acid hydrolysis and subsequent TLC and GC analyses
revealed that 5 contained two D-glucopyranosyl moieties, the
coupling constants of H-1′ (7.4 Hz) and H-1‴ (7.3 Hz)
reminiscent of β-anomeric configurations. Thus, the structure
3.79, s
3.74, s
3.74, s
3.82, s
3.77, s
3.81, s
3.66, s
3.66, s
CH3O-3″ 3.74, s
CH3O-5″ 3.74, s
a
Asterisk indicates overlapped signals.
of 5, indidene E, was established as 7-[6-(4-O-β-D-glucopyr-
anosyloxy-3,5-dimethoxybenzoyl)]-O-β-D-glucopyranosyloxy-6-
methoxycoumarin.
The molecular formula of C30H34O17 of indidene F (6) was
deduced from its sodium adduct ion in the positive-ion
HRESIMS at m/z 689.16847 [M + Na]+ (calcd for
C30H34O17Na, 689.16937), which was 30 mass units less than
that of 5. The NMR data of 6 (Tables 2 and 3) were similar to
those of 5, except for the absence of the signals for the C-5″
methoxy group, which was supported by its HRESIMS data, the
11.8 ppm downfield shift of C-4″ (δC 150.5), the 38.3 ppm
upfield shift of C-5″ (δC 113.2), and the correlations of H-5″
with C-1″/C-4″ in the HMBC spectrum (Figure 6). The acid
hydrolysis revealed that 6 had the same sugar units as 5.
Compound 6, indidene F, was therefore characterized as 7-[6-
(4-O-β-D-glucopyranosyloxy-3-methoxybenzoyl)]-O-β-D-gluco-
pyranosyloxy-6-methoxycoumarin.
The molecular formula of indidene G (7) was C33H40O18, a
C2H4 unit more than that of 5, based on its HRESIMS data.
Comparison between the NMR data of 7 and 5 (Tables 2 and
D
DOI: 10.1021/acs.jnatprod.6b00306
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 3. 13C NMR Data (125 MHz, DMSO-d6, δ in ppm) for
Indidenes D−G (4−7)
the deshielded C-1″ resonance (ΔδC +11.6) and the deshielded
ester carbonyl resonance (ΔδC +6.4), the H-7″ (δH 2.70)/H-8″
(δH 2.61) TOCSY correlation, and the H-7″/C-1″, C-6″, C-8″,
C-9″; H-8″/C-1″, C-7″, C-9″ correlations in the HMBC
spectrum (Figure 6). Thus, the structure of 7, indidene G, was
unambiguously established as 7-[6-(4-O-β-D-glucopyranosylox-
ydihydrosinapoyl)]-O-β-D-glucopyranosyloxy-6-methoxycou-
marin.
All compounds isolated from the bark of S. indicus were
subjected to MTT assays to assess the cytotoxic activities of
these isolates against human lung epithelial A549 and human
breast adenocarcinoma MCF-7 cells. As shown in Table 4, rac-
1, 2, and 3 exhibited cytotoxic activities against A549 and MCF-
7 cells with IC50 values ranging from 2.2 0.1 to 7.2 0.9 μM.
indidene D
indidene E
indidene F
indidene G
(4)
(5)
(6)
(7)
position
δC, type
δC, type
δC, type
δC, type
2
160.0, C
113.1, CH
144.0, CH
113.6, C
129.3, CH
113.2, CH
159.7, C
103.0, CH
154.5, C
99.4, CH
73.0, CH
76.1, CH
70.1, CH
73.9, CH
64.0, CH2
119.2, C
107.0, CH
147.5, C
140.9, C
147.5, C
107.0, CH
165.4, C
160.3, C
160.7, C
113.3, CH
144.3, CH
112.4, C
110.1, CH
145.9, C
149.7, C
102.8, CH
148.9, C
99.2, CH
72.8, CH
76.4, CH
70.0, CH
73.1, CH
64.5, CH2
122.6, C
112.5, CH
148.5, C
150.5, C
113.9, CH
123.1, CH
165.3, C
160.4, C
3
113.9, CH
144.0, CH
112.3, C
113.3, CH
144.0, CH
112.3, C
4
4a
5
110.3, CH
145.9, C
109.7, CH
145.9, C
6
7
149.9, C
149.5, C
8
103.1, CH
148.8, C
103.0, CH
148.8, C
8a
1′
2′
3′
4′
5′
6′
1″
2″
3″
4″
5″
6″
7″
8″
9″
101.9, CH
72.9, CH
76.3, CH
69.84, CH
73.7, CH
64.2, CH2
124.4, C
99.1, CH
72.9, CH
76.5, CH
70.0, CH
73.7, CH
63.7, CH2
136.0, C
Table 4. Cytotoxicity Data for the Compounds Isolated from
a
S. indicus (IC50 SD, μM)
compound
A549
MCF-7
(+)-1
(−)-1
2
2.5 0.3
2.2 0.1
5.6 0.5
6.0 0.4
0.02 0.01
3.9 0.7
3.3 0.4
6.7 0.6
7.2 0.9
0.1 0.03
107.3, CH
152.2, C
106.2, CH
152.3, C
3
paclitaxel
138.7, C
132.8, C
a
Other compounds were inactive (IC50 > 10 μM).
152.2, C
152.3, C
107.3, CH
165.7, C
106.2, CH
30.4, CH2
35.0, CH2
172.1, C
EXPERIMENTAL SECTION
■
General Experimental Procedures. Melting points were
acquired with an X-6 micromelting point apparatus and are
uncorrected. The optical rotations in MeOH were obtained with an
ADP440+ polarimeter (λ = 589 nm, path length = 1.0 cm). The UV
spectra were measured in MeOH on a TU-1901 spectrophotometer.
The ECD spectra were recorded in MeOH on a JASCO J-180
spectropolarimeter. The IR spectra were obtained using a Nicolet
Avatar 360 FT-IR spectrometer. The NMR spectra were acquired on a
Bruker Advance 500 spectrometer, and the residual solvent peaks of
methanol-d4 (δH 3.31 and δC 49.00) or DMSO-d6 (δH 2.50 and δC
39.52) were used as references. The ESIMS and HRESIMS data were
carried out on a Bruker HCT mass spectrometer and MAT 95XP mass
spectrometer, respectively. GC was run on an Agilent 7890AGC
system. Analytical HPLC was conducted on an Agilent 1200 HPLC
system using a 4.6 i.d. × 150 mm Agilent Zorbax SB-C18 (5 μm)
column. Semipreparative HPLC was run on an Agilent 1260 HPLC
system using a 9.4 i.d. × 250 mm Agilent Zorbax SB-C18 (5 μm)
column. A 4.7 i.d. × 250 mm Daicel Chiralpak AD-H (5 μm) column
was employed for the chiral HPLC preparations of 1−3. A silica gel
column (200−300 mesh, Qingdao Marine Chemical Co. Ltd., China)
and a reversed-phase C18 column (50 μm, Merck, Germany) were
used for the preliminary separation of crude samples, and a Sephadex
LH-20 column (Amersham Pharmacia Biotech AB, Sweden) was used
for the final purification.
1‴
2‴
3‴
4‴
5‴
6‴
CH3O-6
CH3O-3″
CH3O-5″
102.6, CH
74.2, CH
77.3, CH
69.8, CH
76.6, CH
60.7, CH2
56.3, CH3
56.0, CH3
56.0, CH3
99.0, CH
73.7, CH
76.9, CH
69.2, CH
76.7, CH
60.3, CH2
56.0, CH3
55.5, CH3
102.9, CH
74.2, CH
77.1, CH
69.9, CH
76.5, CH
60.7, CH2
56.0, CH3
56.2, CH3
56.2, CH3
56.0, CH3
56.0, CH3
Plant Material. The dried bark (20 kg) of S. indicus was collected
in Qinzhou (Guangxi Province, China) in September 2012 and was
authenticated by Professor Fa-Nan Wei, Guangxi Institute of Botany.
A reference voucher specimen (No. SIC20120925) was deposited at
the Laboratory of Natural Products, Guangxi Normal University,
China.
Extraction and Isolation. Air-dried, powdered bark (20 kg) of S.
indicus was refluxed in 75% EtOH (3 × 2 h) to afford a crude extract
(1.3 kg) that was subsequently suspended in H2O (3 L) and
successively extracted with petroleum ether (4 × 5 L), EtOAc (4 × 5
L), and n-BuOH (4 × 5 L). The bioactive screening of these extracts
indicated that the EtOAc extract was the most cytotoxic against A549
and MCF-7 cells. The EtOAc extract (400 g) was divided into six
fractions (A−F) by silica gel column chromatography (CC, 12 i.d. ×
35 cm) eluting with a gradient of CH2Cl2/MeOH (100:0, 95:5, 90:10,
Figure 6. HMBC (arrows) correlations of 4−7.
3) revealed that C-1″ of 7 was linked to the ester carbonyl
carbon via a CH2CH2 unit. This conclusion was confirmed by
E
DOI: 10.1021/acs.jnatprod.6b00306
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
80:20, 50:50, 0:100, v/v). Fraction B (60 g) was loaded onto a silica
gel column (8 i.d. × 30 cm) and eluted with a gradient of CH2Cl2/
MeOH (100:0−30:70, v/v) to afford 20 subfractions (B1−B20).
Subfraction B1 (4 g) was separated on another silica gel column (4 i.d.
× 15 cm) with a gradient of CH2Cl2/MeOH (100:0−10:90, v/v) to
afford nine subfractions (B1-1−B1-9). Subfraction B1-1 (0.5 g) was
fractionated by CC over Sephadex LH-20 with CHCl3/MeOH (50:50,
v/v) to yield 1 (45.0 mg) and 8 (8.5 mg). Racemic mixtures 1 and
subfractions (B1-2, 40 mg) were further separated over a Chiralpak
AD-H column with 80% n-hexane-isopropyl alcohol (0.8 mL/min) to
afford a pair of enantiomers (−)-1 (tR = 9.4 min, 6.9 mg) and (+)-1 (tR
= 15.6 min, 7.1 mg) and a pair of epimers 2 (tR = 16.9 min, 6.3 mg)
and 3 (tR = 15.3 min, 6.1 mg), respectively. Subfraction B2 (20.0 g)
was fractionated by CC over RP-C18 and eluted with mixtures of
MeOH/H2O with decreasing polarity (65:35−100:0, v/v) to afford
seven fractions B2-(1−7). Separation of subfraction B2-1 (2.2 g) was
done by semipreparative HPLC eluting with a gradient of MeOH/
H2O (20:80−25:75, v/v) and Sephadex LH-20 CC eluting with
MeOH to yield 6 (7.9 mg), 7 (8.2 mg), and 9 (4.6 mg). Compounds 4
(6.2 mg), 5 (6.6 mg), 10 (5.3 mg), and 11 (7.8 mg) were obtained
from subfraction B2-2 (1.5 g) via semipreparative HPLC eluting with a
gradient of MeOH/H2O (24:76−32:68, v/v).
pyridine/EtOH/H2O (8:1:1:2) as the mobile phase. Glucose (Rf 0.37)
was identified as the sugar component of 4 by comparison with
authentic D-glucose (Rf = 0.36) and L-glucose (Rf = 0.39). Then, L-
cysteine methyl ester hydrochloride (4 mg) and the neutral residue of
the aqueous layer were dissolved in anhydrous pyridine (2 mL) and
heated at 80 °C for 1 h. The reaction mixture was vacuum-dried and
treated with N-trimethylsilylimidazole (0.4 mL) at 80 °C for 1 h. The
reaction mixture was partitioned between water and n-hexane (each 3
mL). The n-hexane extract was analyzed by GC using a 0.32 mm i.d. ×
25 m L-Chirasil-Val column. The detector temperature was kept at
280 °C, and the injector temperature was 250 °C. A temperature
gradient was programmed to start at 160 °C and hold for 5 min,
increase to 280 °C at 5 °C·min−1, and hold for 10 min. The authentic
samples were analyzed in the same way. Retention time for authentic
silylated D-glucose and L-glucose were 19.09 and 19.25 min,
respectively. The results of the GC analysis indicated that the sugar
component of 4 was D-glucose (tR = 19.08 min). The sugar
components of 5−7 were established using the same procedure.
Computational Section. Conformational analysis was carried out
using the VEGA ZZ3.0 program with Molecular Merck force field in
Spartan 14 software.24 A 1D potential energy surface scan of the C-6′−
C-7′−C-1‴−C-6‴ dihedral angle in 2 and 3 was carried out at the
semiempirical AM1 level. The conformers with relative energies within
2 kcal/mol (Table S1, Supporting Information) were optimized by
density functional theory (DFT) at the B3LYP/6-31G (d) level. The
B3LYP/6-31G(d) frequency calculations were performed to confirm
the stability of each optimized conformer. The ECD spectra for the
stable conformers were calculated by time-dependent (TD)-DFT at
the B3LYP/6-311+G(2d,p) level using the polarizable continuum
model in MeOH. The calculated ECD curves were generated using
SpecDis 1.53 software (σ = 0.3 eV).25 The final ECD spectra of
(1′R,2′S)-1, (1′S,2′S)-2, and (1′S,2′R)-3 were obtained based on the
Boltzmann statistical contribution of each conformer and the
experimental data. All DFT and TD-DFT calculations were conducted
with Gaussian 09 program.26
rac-Indidene A (1): colorless crystals (MeOH); mp 156−157 °C;
25
[α]D +0.01 (c 0.4, MeOH); UV (MeOH) λmax (log ε) 212 (4.30),
222 (4.11), 273 (3.92), 304 (3.83) nm; IR (KBr) νmax 3403, 2968,
1
1715, 1601, 1472, 1382, 1163 cm−1; for H and 13C NMR data, see
Table 1; HRESIMS m/z 395.14560 (calcd for C21H24O6Na,
395.14706).
25
(+)-Indidene A (1): yellow gum; [α]D +31 (c 0.4, MeOH); ECD
(MeOH) λmax (Δε) 289 (−7.8), 267 (0.43), 240 (2.7), 223 (−6.8),
207 (5.0).
25
(−)-Indidene A (1): yellow gum; [α]D −31 (c 0.4, MeOH); ECD
(MeOH) λmax (Δε) 289 (7.8), 267 (−0.43), 240 (−2.7), 223 (6.8),
207 (−5.0).
25
Indidene B (2): pale red powder; [α]D +42 (c 0.4, MeOH); UV
(MeOH) λmax (log ε) 211 (3.9), 250 (3.81), 284 (3.60) nm; ECD
(MeOH) λmax (Δε) 288 (11.4), 268 (14.1), 245 (−6.5), 218 (14.7);
IR (KBr) νmax 3410, 2939, 1741, 1730, 1690, 1503, 1398, 1147 cm−1;
X-ray Crystallographic Analysis of Indidene A (1). Crystals of
indidene A (1) were obtained from its MeOH solution. Diffraction
intensity data were collected on a Bruker APEX-II CCD diffractometer
with Cu Kα radiation (λ = 1.541 84 Å). Structure solution was
performed with program SHELXS-97 (direct method).27 Structure
refinement was done by full-matrix least-squares on F2 (SHELXL-
97),27 with non-hydrogen atoms treated anisotropically. H atoms
bonded to C and O were inserted in the ideal geometrical positions
and refined as riding with d(C−H) = 0.93−0.98 Å, d(O−H) = 0.82 Å,
and Uiso(H) = 1.2 Ueq (C) or 1.5 Ueq (O). Crystallographic data for
indidene A (1) reported in the present study have been deposited at
the Cambridge Crystallographic Data Centre (deposit number: CCDC
1471448). The data can be obtained free of charge at www.ccdc.cam.
ac.uk or from the Cambridge Crystallographic Data Centre (12 Union
Road, Cambridge CB2 1EZ, UK; Fax: (+44) 1223-336-033; or e-mail:
desposit@ccdc.cam.ac.uk).
1
for H and 13C NMR data, see Table 1; HRESIMS m/z 425.15618
(calcd for C22H26O7Na, 425.15762).
Indidene C (3): yellow gum; [α]D −32 (c 0.3, MeOH); UV
25
(MeOH) λmax (log ε) 210 (3.8), 250 (3.75), 284 (3.58) nm; ECD
(MeOH) λmax (Δε) 286 (−8.1), 271 (2.8), 237 (−3.5), 227 (−1.8),
220 (−4.1), 210 (5.0); IR (KBr) νmax 3409, 2938, 1740, 1730, 1691,
1504, 1398, 1146 cm−1; for 1H and 13C NMR data, see Table 1;
HRESIMS m/z 425.15618 (calcd for C2225H26O7Na, 425.15762).
Indidene D (4): white powder; [α]D −84 (c 0.3, MeOH); UV
(MeOH) λmax (log ε) 254 (3.82), 290 (4.25), 330 (3.92) nm; IR
(KBr) νma1x 3372, 2988, 1738, 1725, 1630, 1605, 1512, 1241, 1180
cm−1; for H and 13C NMR data, see Tables 2 and 3; HRESIMS m/z
527.11545 (calcd for C24H24O12Na, 527.11655).
25
Indidene E (5): white powder; [α]D −95 (c 0.4, MeOH); UV
Crystal Data of indidene A (1): Moiety formula, 2(C21H24O6),
(MeOH) λmax (log ε) 262 (4.20), 286 (4.06), 342 (3.96) nm; IR
(KBr) νmax 3408, 2979, 1742, 1732, 1611, 1564, 1245, 1170 cm−1; for
1H and 13C NMR data, see Tables 2 and 3; HRESIMS m/z 719.17989
(calcd for C31H36O18 Na, 719.17993).
CH O, M = 776.85, triclinic, space group P1 (no. 2), Z = 2, a =
̅
4
r
8.9728(4) Å, b = 13.5806(6) Å, c = 16.6376(7) Å, α = 93.952(2)°, β =
103.777(2)°, γ = 90.502(2)°, V = 1963.72(15) Å3, T = 100 K, μ(Cu
Kα) = 0.800 mm−1, 35 146 reflections measured, 6401 unique (Rint
=
25
Indidene F (6): white powder; [α]D −75 (c 0.4, MeOH); UV
0.0301), which were used in all calculations. The final R1 was 0.0478,
wR2 was 0.1390, and s was 1.05 (all data).
(MeOH) λmax (log ε) 260 (4.12), 294 (3.97), 344 (3.86) nm; IR
(KBr) νmax 3421, 2945, 1736, 1730, 1615, 1460, 1245, 1072 cm−1; for
1H and 13C NMR data, see Tables 2 and 3; HRESIMS m/z 689.16847
(calcd for C30H34O17Na, 689.16937).
Cytotoxicity Assay. The cytotoxic activities of the isolates were
investigated using the MTT method.28 The 1 × 104 cells suspended in
190 μL of media were added to each well of a 96-well microplate,
incubated in 5% CO2 atmosphere at 37 °C for 12 h, treated with a
series of diluted compounds or paclitaxel for 48 h, and then cultured
with 20 μL of 5 mg/mL MTT reagent for another 4 h. The formazan
crystals in each well were then solubilized with 150 μL of DMSO. The
absorbance was measured at 570 nm with a Bio-Rad 680 microplate
reader. All experiments were repeated three times and each time in
triplicate. The half-maximal inhibitory concentration (IC50) values
were acquired by fitting sigmoid curves with GraphPad Prism software
(version 5.0).29
25
Indidene G (7): white powder; [α]D −63 (c 0.4, MeOH); UV
(MeOH) λmax (log ε) 260 (3.73), 294 (3.80), 342 (4.03) nm; IR
(KBr) νmax 3429, 2968, 1742,1731, 1640, 1610, 1453, 1238,1069 cm−1;
for 1H and 13C NMR data, see Tables 2 and 3; HRESIMS m/z
747.21074 (calcd for C33H40O18Na, 747.21123).
Acid Hydrolysis of 4−7. Compound 4 (2 mg) was dissolved in 1
M HCl (4 mL) and reacted at 80 °C for 4 h and extracted with EtOAc
(3 × 5 mL). The aqueous layer was evaporated under a stream of N2
to afford a neutral residue that was analyzed using TLC with EtOAc/
F
DOI: 10.1021/acs.jnatprod.6b00306
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(16) Li, L. Q.; Li, J.; Huang, Y.; Wu, Q.; Deng, S. P.; Su, X. J.; Yang,
R. Y.; Huang, J. G.; Chen, Z. Z.; Li, S. Fitoterapia 2012, 83, 303−309.
(17) Li, J.; Meng, A. P.; Zhu, S. J.; Tang, M. T.; Li, L. Q.; Li, S.; Chen,
Z. Z.; Huang, J. G. CN Patent HZ201110094274.8, 2012.
(18) Li, J.; Meng, A. P.; Guan, X. L.; Li, J.; Wu, Q.; Deng, S. P.; Su, X.
J.; Yang, R. Y. Bioorg. Med. Chem. Lett. 2013, 23, 2238−2244.
(19) Rastogi, S.; Kulshreshtha, D. K.; Rawat, A. K. S. Evid.-Based
Compl. Alt. Med. 2006, 3, 217−222.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.6b00306.
■
S
NMR and HRESIMS spectra of the new compounds and
calculation data of 1−3 (PDF)
X-ray crystallographic data for compound 1 (CIF)
(20) Fan, L.; Wang, Y.; Liang, N.; Huang, X. J.; Li, M. M.; Fan, C. L.;
Wu, Z. L.; Li, Y. L.; Ye, W. C. Planta Med. 2013, 79, 1558−1564.
(21) Iqbal, P. F.; Bhat, A. R.; Azam, A. Eur. J. Med. Chem. 2009, 44,
2252−2259.
AUTHOR INFORMATION
Corresponding Authors
*(J. Li) Tel: +86-773-5803907. Fax: +86-773-2120958. E-mail:
lijun9593@gxnu.edu.cn.
*(Y. Huang) Tel: +86-771-5883405. Fax: +86-771-5867737. E-
mail: hy2002-2006@163.com.
■
(22) Tsukamoto, H.; Hisada, S.; Nishibe, S. Chem. Pharm. Bull. 1985,
33, 396−399.
(23) Laplante, S. R.; Edwards, P. J.; Fader, L. D.; Jakalian, A.; Hucke,
O. ChemMedChem 2011, 6, 505−513.
(24) Spartan 14; Wavefunction, Inc.: Irvine, CA.
(25) Bruhn, T.; Hemberger, Y.; Schaumloffel, A.; Bringmann, G.
̈
SpecDis, version 1.53; University of Wuerzburg, 2011.
(26) Gaussian 09, revision C.1; Gaussian, Inc.: Wallingford, CT,
2010. A full list of authors can be found in the Supporting Information.
(27) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr.
2008, A64, 112−122.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(28) Mosmann, T. J. J. Immunol. Methods 1983, 65, 55−63.
(29) GraphPad Software, San Diego, CA.
J.L. acknowledges the following grants for funding this project:
The Program for Changjiang Scholars and Innovative Research
Team in University (IRT1225); the Natural Science
Foundation of Guangxi Province (2014GXNSFDA118008);
the Open Research Fund program of Guangxi Key Laboratory
of Traditional Chinese Medicine Quality Standards
(GZZK201405); the Open Research Fund program of the
Key Laboratory for the Chemistry and Molecular Engineering
of Medicinal Resources (CMEMR2012−A21); and National
Natural Science Foundation of China (21662004).
REFERENCES
■
(1) Zhang, S. S.; Wu, C. Y.; Cao, Z. Y. Flora Reipublicae Popularis
Sinciae; Science Press: Beijing, 1998; Vol. 23, pp 30−35.
(2) Zhao, A. H.; Yang, L. C.; Liu, G.; Wei, J. Chin. Pharm. J. 1999, 34,
368−369.
(3) Chen, J. M.; Qin, Y. N.; Cang, H. D. Acta Acad. Med. Xuzhou
1983, 4, 20−21.
(4) Wu, T. S.; Leu, Y. L.; Hsu, H. C.; Ou, L. F.; Chen, C. C.; Chen,
C. F.; Ou, J. C.; Wu, Y. C. Phytochemistry 1995, 39, 383−385.
(5) Yang, X. W.; Xu, B.; Ran, F. X.; Wang, R. Q.; Wu, J. Zhongxiyi
Jiehe Xuebao 2007, 5, 56−60.
(6) Panno, M. L.; Giordano, F.; Palma, M. G.; Bartella, V.; Rago, V.;
Maggiolini, M.; Sisci, D.; Lanzino, M.; De Amicis, F.; Ando, S. Curr.
Cancer Drug Targets 2009, 9, 469−481.
(7) Li, J.; Jin, B. F.; Zhang, Y. J.; Su, X. J.; He, X. C.; Huang, Y.;
Huang, X. S.; Liao, R. Q.; Long, B. CN Patent ZL2006100 22463.3,
2011.
(8) Li, J.; Zhang, Y. J.; Jin, B. F.; Su, X. J.; Tao, Y. W.; She, Z. G.; Lin,
Y. C. Magn. Reson. Chem. 2008, 46, 497−500.
(9) Lu, X. W.; Li, J.; Huang, C. P.; Meng, A. P.; Zhu, S. J. J. Guangxi
Normal Univ. (Nat. Sci. Ed.) 2009, 27, 61−64.
(10) Zhu, S. J.; Li, J.; Meng, A. P.; Liu, Q. Y.; Huang, C. P.; Lu, X. W.
J. Guangxi Normal Univ. (Nat. Sci. Ed.) 2010, 28, 33−36.
(11) Chen, Z. Z.; Li, J.; Wu, Q.; Yang, R. Y.; Li, L. Q.; Li, S.; Huang,
J. G. Guihaia 2011, 31, 849−852.
(12) Huang, J. G.; Li, J.; Wu, Q.; Yang, R. Y.; Li, S.; Chen, Z. Z.; Li,
L. Q. Nat. Prod. Res. Dev. 2012, 24, 780−783.
(13) Li, J.; Huang, Y.; Guan, X. L.; Li, J.; Deng, S. P.; Wu, Q.; Zhang,
Y. J.; Su, X. J.; Yang, R. Y. Phytochemistry 2012, 82, 100−109.
(14) Chen, H.; Li, J.; Wu, Q.; Niu, X. T.; Tang, M. T.; Guan, X. L.;
Li, J.; Yang, R. Y.; Deng, S. P.; Su, X. J. Fitoterapia 2012, 83, 643−649.
(15) Li, J.; Tang, M. T.; Wu, Q.; Chen, H.; Niu, X. T.; Guan, X. L.;
Li, J.; Deng, S. P.; Su, X. J.; Yang, R. Y. Nat. Prod. Commun. 2012, 7,
599−602.
G
DOI: 10.1021/acs.jnatprod.6b00306
J. Nat. Prod. XXXX, XXX, XXX−XXX