Clerodane Diterpenoid Glucosides from the Stems of Tinospora sinensis

Article abstract of DOI:10.1021/acs.jnatprod.6b00976

Ten new clerodane diterpenoid glucosides (1-10) and three known analogues (11-13) were isolated from an EtOAc extract of the stems of Tinospora sinensis. Spectroscopic analyses and chemical methods were used to elucidate the structures of these isolates. The absolute configurations of tinosinenosides A-C (1-3) were established by using experimental and calculated ECD data. Their cytotoxicity against the human epithelioid cervical carcinoma (HeLa) cell line and the nitric oxide production inhibitory activity of lipopolysaccharide-activated N9 microglial cells were tested. 1-Deacetyltinosposide A (12) exhibited mild cytotoxicity against HeLa cells, with an IC₅₀ value of 8.35 ± 0.60 μM.

Full text of DOI:10.1021/acs.jnatprod.6b00976

Article
pubs.acs.org/jnp
Clerodane Diterpenoid Glucosides from the Stems of Tinospora
sinensis
Huan Jiang,† Gui-Jie Zhang,‡ Yan-Fei Liu,§ Heng-Shan Wang,† and Dong Liang*,†
†State Key Laboratory for Chemistry and Molecular Engineering of Medicinal Resources, School of Chemistry and Pharmaceutical
Sciences, Guangxi Normal University, Guilin 541004, People’s Republic of China
‡College of Pharmacy, Guilin Medical University, Guilin 541004, People’s Republic of China
§State Key Laboratory of Bioactive Substance and Function of Natural Medicines, Institute of Materia Medica,
Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100050, People’s Republic of China
S
* Supporting Information
ABSTRACT: Ten new clerodane diterpenoid glucosides (1−10) and
three known analogues (11−13) were isolated from an EtOAc extract
of the stems of Tinospora sinensis. Spectroscopic analyses and chemical
methods were used to elucidate the structures of these isolates.
The absolute configurations of tinosinenosides A−C (1−3) were
established by using experimental and calculated ECD data. Their
cytotoxicity against the human epithelioid cervical carcinoma (HeLa)
cell line and the nitric oxide production inhibitory activity of lipo-
polysaccharide-activated N9 microglial cells were tested. 1-Deacetylti-
nosposide A (12) exhibited mild cytotoxicity against HeLa cells, with
an IC50 value of 8.35 0.60 μM.
lerodane diterpenoids are commonly found in the
that 1-deacetyltinosposide A (12) exhibited mild cytotoxicity
against HeLa cells.
C
genus Tinospora,1−4 and a number of those have been
reported to possess diverse biological activities such as
cytotoxic,5,6 immunomodulatory,7 antifeedant,8 and anti-inflam-
matory effects.9 The biological activities and chemical struc-
tures of clerodanes have led to extensive efforts toward their
synthesis.10−12
RESULTS AND DISCUSSION
■
The HRESIMS data of tinosinenoside A (1) showed a sodium
adduct ion at m/z 559.1780 [M + Na]+, consistent with the
molecular formula C26H32O12. The presence of hydroxy groups
(3416 cm−1) and a δ-lactone moiety (1738 cm−1) was evident
from their characteristic IR absorptions. The presence of a
furan moiety was indicated by typical signals at δH 6.54 (dd,
J = 1.5, 0.5 Hz, H-14), 7.49 (t, J = 1.5 Hz, H-15), and 7.58 (br s,
Tinospora sinensis (Lour.) Merr. is widely distributed and
cultivated in the South of China.13 The stems of this species,
commonly known as “Kuan-Jin-Teng” in Chinese, have been
used as a traditional Yao medicine to treat rheumatism, lumbar
muscle strain, and sciatica.14 The crude extracts of T. sinensis
exhibited potent immunomodulatory and anti-inflammatory
activities.15,16 A wide array of clerodane diterpenoids have been
reported to be the main components of T. sinensis.1,2 Owing to
its fascinating physiological properties and in an ongoing search
for the bioactive agents of traditional Chinese medicines,17 the
clerodane diterpenoids of T. sinensis have been investigated.
Ten new diterpenoid glucosides (1−10) and three known
compounds (11−13) were obtained from the stems of
T. sinensis. The rare presence of a C-4 carbonyl group was
demonstrated in tinosinenoside B (2).1,18−21 Seven new
compounds (4−10) exhibited a dinorclerodane diterpenoid
scaffold attached to a cis- or trans-feruloyl group. Herein, the
isolation and structural elucidation of 1−10 are discussed. The
cytotoxicity against the human epithelioid cervical carcinoma
(HeLa) cell line and the nitric oxide (NO) production inhib-
itory activity of lipopolysaccharide (LPS)-activated N9 micro-
glial cells were also tested. Cytotoxicity evaluations revealed
1
H-16) in the H NMR spectrum (Table 1). There were 20
carbon signals aside from those of the sugar moiety (δC 101.5,
75.2, 78.3, 71.3, 78.1, and 62.6) in the 13C NMR spectrum
(Table 1), which were classified by the HSQC data as two
methyls (δC 20.7 and 25.1), three methylenes (δC 28.0, 28.2,
and 37.2), three methines (δC 48.5, 73.0, and 74.7), two
oxygenated tertiary carbons (δC 74.0 and 87.9), two sp3
quaternary carbons (δC 40.1 and 41.1), one double bond (δC
131.8 and 133.1), a furan moiety (δC 109.7, 127.0, 141.4, and
145.0), and two lactone carbonyls (δC 175.2 and 175.9). The
lactone moieties were distinguished via the HMBC cross-peaks
from H-1 (δH 5.36) to C-18 (δC 175.2) and from H-12 (δH
5.58) to C-17 (δC 175.9), respectively. According to the litera-
ture, the NMR signals of tinosinenoside A (1) and palmatoside
Received: October 24, 2016
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.6b00976
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Chart 1
1
Table 1. H and 13C NMR Data of Compounds 1−3
a
b
c
1
2
3
no.
δC
δH (J in Hz)
δC
δH (J in Hz)
δC
δH (J in Hz)
1
2
74.7
131.8
133.1
5.36, br d (5.0)
73.1
145.9
127.1
4.38, d (4.4)
71.8
74.4
30.82
3.88, dd (8.4, 3.2)
4.99, m
6.63, dd (8.0, 5.0)
6.83, dd (8.0, 1.0)
6.97, dd (10.0, 4.4)
5.91, d (10.0)
3a
3b
4
2.28, ddd (15.6, 4.4, 1.6)
1.77, dt (15.6, 4.4)
3.99, m
87.9
40.1
28.0
202.5
42.3
29.2
74.6
38.6
26.8
5
1.69, m
6a
6b
7a
7b
8
1.92, dd (14.0, 8.0)
1.66, m
2.20, dt (14.0, 3.6)
0.92, td (14.0, 3.6)
2.03, dt (14.0, 3.6)
1.38, td (14.0, 3.6)
1.67, overlapped
1.57, m
28.2
3.03, m
26.9
30.78
2.33, m
1.72, m
1.67, overlapped
74.0
41.1
48.5
37.2
74.3
40.0
47.9
33.2
76.7
41.1
40.1
36.5
9
10
11a
11b
12
13
14
15
16
17
18
19
20
2-Ac
1.81, br s
2.44, br s
2.19, dd (10.4, 8.4)
2.55, dd (14.0, 4.4)
2.46, dd (14.0, 12.2)
5.71, dd (12.2, 4.4)
2.43, dd (14.6, 11.5)
2.12, dd (14.6, 5.5)
5.58, dd (11.5, 5.5)
2.27, dd (14.4, 12.8)
2.07, dd (14.4, 3.6)
5.71, dd (12.8, 3.6)
73.0
127.0
109.7
145.0
141.4
175.9
175.2
25.1
70.2
125.3
108.9
144.0
140.3
171.6
73.7
127.3
109.7
144.8
141.1
175.2
6.54, dd (1.5, 0.5)
7.49, t (1.5)
7.58, br s
6.56, d (1.4)
7.68, t (1.4)
7.72, br s
6.50, dd (2.0, 0.8)
7.47, dd (2.0, 1.6)
7.55, m
1.10, s
1.18, s
31.0
17.9
1.20, s
0.61, s
20.7
15.3
173.7
21.6
101.3
75.0
78.3
72.0
78.1
63.1
1.14, s
2.10, s
Glc-1′
2′
3′
4′
5′
101.5
75.2
78.3
71.3
78.1
62.6
4.70, d (7.5)
105.1
73.5
76.6
70.0
77.0
61.2
4.35, d (7.6)
2.95, m
4.24, d (7.6)
3.16, dd (9.2, 7.6)
3.32, m
3.38, overlapped
3.38, overlapped
3.38, overlapped
3.29, m
3.12, m
3.03, m
3.24, m
3.15, m
3.20, m
6′a
6′b
3.81, dd (12.0, 2.0)
3.66, m
3.84, dd (11.8, 2.2)
3.66, dd (12.0, 5.0)
3.43, dd (12.0, 5.6)
3.62, dd (11.8, 5.8)
c
a
b
In MeOH-d4, 1H NMR at 500 MHz, 13C NMR at 125 MHz. In DMSO-d6, 1H NMR at 400 MHz, 13C NMR at 100 MHz. In MeOH-d4, 1H NMR
at 400 MHz, 13C NMR at 125 MHz.
C are similar, except for the presence of a hydroxy group at C-8
HMBC cross-peaks from H2-6 (δH 1.66, 1.92) and H3-20 (δH
1.18) to C-8 (δC 74.0) verified this relation. Following acid
in 1 replacing the proton in the known analogue.22 The key
B
DOI: 10.1021/acs.jnatprod.6b00976
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
hydrolysis, the sugar moiety was identified as D-glucose by using
data (Table 1) showed the presence of a furan moiety (δH
6.56, d, J = 1.4 Hz; 7.68, t, J = 1.4 Hz; and 7.72, br s), an
α,β-unsaturated carbonyl moiety (δH 6.97, dd, J = 10.0, 4.4 Hz;
5.91, d, J = 10.0 Hz; δC 202.5), two methyls (δH 0.61, 1.20, each
3H, s), a lactone moiety (δC 171.6), and a β-glucopyranosyl
group (δH 4.35, d, J = 7.6 Hz). The NMR data analysis
indicated that 2 was structurally related to tinosporaside,27
differing only in the presence of a hydroxy group at C-8 (δC
74.3) and an interchange of the substituents at C-1 and C-4.
This hypothesis was shown to be correct by the HMBC cross-
peaks from H-1 (δH 4.38) to C-5 (δC 42.3), C-9 (δC 40.0), and
C-1′ (δC 105.1) and from H3-19 (δH 1.20) to C-4 (δC 202.5).
The NOESY cross-peaks of H-10 (δH 2.44)/H3-19 and
H-10/H-12 (δH 5.71) suggested that H-10, H-12, and H3-19
were cofacial, and they were assigned randomly as α-oriented.
The correlation of H-1/H3-20 (δH 0.61) supported the β-orien-
tation of H-1 (Figure 3). The strong cross-peak of H-10/H-12
an HPLC equipped with an optical rotation detector.23,24 The
3
β-configuration was based on the large J1,2 value (7.5 Hz) and
the chemical shift of its anomeric carbon at δC 101.5.
In the NOESY spectrum, the cross-peaks of H-1 (δH 5.36)/
H-10 (δH 1.81), H-10/H3-19 (δH 1.10), H-1′/H3-19, and
H-10/H-12 (δH 5.58) indicated that H-1, H-10, H-12, H3-19,
and the β-D-glucopyranosyl moiety were on the same side, and
they were assigned randomly as α-oriented, while the furan
moiety was assigned as β-oriented (Figure 1). The NOESY
Figure 1. Selected NOESY correlations of 1.
cross-peaks of H3-19/H-6α (δH 1.92) and H-6β (δH 1.66)/
H3-20 suggested that H3-20 was β-oriented. These data sug-
gested that 1 shared the same 2D structure and relative config-
uration as palmatoside C, except for the oxygenated tertiary
C-8. The configuration of C-8 was determined based on the
structure elucidated via computer-aided simulation and critical
NOESY information obtained from H-10.25 The strong cross-
peak of H-10/H-12 and the absence of a cross-peak of H-10/
H2-11 (δH 2.12, 2.43) in 1 suggested that OH-8 was β-oriented.
This was further confirmed by the γ-gauche effect induced by
OH-8,26 which caused the decreased chemical shifts of C-11
(δC 37.2) and C-20 (δC 20.7) in 1 relative to palmatoside C (δC
40.2, C-11; δC 23.89, C-20).22 By comparing the experimental
and calculated ECD spectra predicted by time-dependent
density functional theory (TDDFT) at the B3LYP/6-311G**
level, the absolute configuration of 1 was established. The experi-
mental electronic circular dichroism (ECD) spectrum of 1
(Figure 2) showed a negative Cotton effect (CE) at 203 nm
Figure 3. Selected NOESY correlations of 2.
and the absence of a cross-peak of H-10/H2-11 (δH 2.07, 2.27)
in 2 suggested that OH-8 was β-oriented.25 This was verified by
the clear cross-peak of OH-8 (δH 6.03)/H3-20. According to
empirical rules,28,29 the positive CE at 232 nm (Δε = +6.9)
caused by the π → π* electronic transition and the negative CE
at 337 nm (Δε = −0.8) resulting from the n → π* transition of
an α,β-unsaturated carbonyl moiety both indicated that the
configuration of 2 was 1S, 5R, 8S, 9S, 10S, 12S. The experi-
mental ECD spectrum corresponded well with the calculated
ECD spectrum for (1S,5R,8S,9S,10S,12S)-2 (Figure 4), which
Figure 4. Comparison of experimental and calculated ECD spectra of 2.
Figure 2. Comparison of experimental and calculated ECD spectra of 1.
(Δε = −19.3) and a positive CE at 230 nm (Δε = +8.0), fitting well
with the calculated ECD spectrum for (1R,4R,5R,8S,9S,10S,12S)-1.
Hence, the structure of 1 was established.
The HRESIMS data of tinosinenoside B (2) produced a
chloride adduct ion at m/z 543.1643 [M + Cl]−, which was
consistent with a molecular formula of C25H32O11. The NMR
authenticated the configuration. Thus, the structure of 2 was
shown to be an 18-norclerodane diterpenoid with a C-4 car-
bonyl group, which is rare because the carbonyl group is usually
at C-1 or C-2.21
Tinosinenoside C (3) had a molecular formula of C26H36O13,
as determined by HRESIMS (m/z 591.1857 [M + Cl]−, calcd
C
DOI: 10.1021/acs.jnatprod.6b00976
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
for 591.1850) and 13C NMR data. A β-glucopyranosyl moiety
was recognized from the anomeric proton signal at δH 4.24 (d,
J = 7.6 Hz). The H NMR data (Table 1) of 3 displayed
The 1H NMR spectrum showed a cluster of typical signals at δH
7.17 (d, J = 8.5 Hz), 7.21 (dd, J = 8.5, 1.7 Hz), and 7.29 (d, J =
1.7 Hz), representing a 1,2,4-trisubstituted aromatic moiety,
two trans-olefinic protons at δH 6.40 (d, J = 16.0 Hz) and 7.65
(d, J = 16.0 Hz), and one methoxy group at δH 3.84 (s). On
the basis of the above data, the structure of 4 was similar to
compound 3 but carrying a trans-feruloyl group, which was
confirmed by 2D NMR data analysis. In the HMBC spectrum,
the cross-peaks from H-4 (δH 4.85) to C-1′ (δC 168.0) located
the feruloyl group at C-4, and the correlation from H-1″ (δH
4.97) to C-7′ (δC 150.1) linked the β-D-glucopyranosyl moiety
to C-7′. The relative configuration of the diterpenoid moiety of
4 was established as being identical to that of 3 by analyz-
ing their NMR data (Tables 1−3) and NOESY interactions
1
diagnostic signals for a typical furan moiety (δH 6.50, 7.47, and
7.55), a methyl group (δH 1.14), and a lactone moiety (δH 5.71,
δC 175.2). The aforementioned data showed similarities to
tinosineside A (11).30 The substituents at C-1 and C-2 in 3
were hydroxy and acetoxy, respectively, compared to 1-acetoxy
and 2-hydroxy in 11. This regiochemistry was supported by the
COSY cross-peaks of H-10 (δH 2.19)/H-1 (δH 3.88)/H-2 (δH
4.99)/H2-3 (δH 1.77, 2.28) and HMBC cross-peaks from H-2
to C-10 (δC 40.1) and CH3CO-2 (δC 173.7).
The relative configuration of 3 was constructed based on the
3
coupling constants and the NOESY spectrum. The J1,2 (3.2
Hz) and 3J1,10 (8.4 Hz) values indicated that the orientations of
H-1, H-2, and H-10 were β-axial, β-equatorial, and α-axial,
respectively (Figure 5).1 The cross-peaks of H-1/H-5 (δH 1.69)
3
(Figure 7). The J2′,3′ value (16.0 Hz) suggested a 2′,3′-trans
olefinic bond. The structure of compound 4 was thus
established as shown.
Compound 5, 4-epi-tinosinenoside D, shared the same molec-
ular formula as 4, C36H44O16, as was evident from an HRESIMS
ion at m/z 755.2508 [M + Na]+ (calcd for 755.2522) and
13C NMR data. The NMR data (Tables 2 and 3) showed that the
structure of compound 5 was similar to 4 except for the C-4
3
3
configuration. The J3β,4 (11.0 Hz) and J4,5 (11.0 Hz) values
indicated that H-4 had an α-axial orientation, as could also be
inferred from the NOESY cross-peak of H-4 (δH 4.79)/H-6α
(δH 0.99) (Figure 8). Detailed 2D NMR data analyses indicated
that the rest of the structure of 5 was the same as 4. Hence, the
structure of 5 was established as the 4-epimer of 4.
Figure 5. Selected NOESY correlations of 3.
Tinosinenoside E (6) also shared the molecular formula
C36H44O16 with 4, as was established via 13C NMR data and an
HRESIMS sodium adduct ion at m/z 755.2501 [M + Na]+
(calcd for 755. 2522). The NMR data (Tables 2 and 3) showed
that compound 6 was also similar to 4, except for a 2′,3′-cis
and H-5/H3-20 (δH 1.14) suggested that H-5 and H3-20 were
β-oriented. The cross-peak of H-1′ (δH 4.24)/H-3α (δH 2.28)
suggested that the glucopyranosyl moiety was α-oriented and
that H-4 was β-oriented. The cross-peak of H-10/H-12 (δH
5.71) and the absence of the cross-peak of H-10/H2-11 (δH
2.46, 2.55) revealed that H-12 was α-oriented and that OH-8 was
β-oriented.25 The experimental ECD spectrum of 3 conformed
to the calculation spectrum for (1R,2S,4R,5R,8S,9S,10R,12S)-3
(Figure 6). Therefore, the structure of 3 was defined as shown.
3
olefinic unit as indicated by a distinctly smaller J2′,3′ value
(13.0 Hz). The relative configurations of the C-1, C-2, C-4, C-5,
C-8, C-9, C-10, and C-12 stereocenters of 6 were consistent with
those of 4 according to the NOESY data and their similar NMR
patterns. Thus, the structure of 6 was established as shown.
The molecular formula of tinosinenoside F (7) was C36H44O16
based on the 13C NMR data and an HRESIMS ion at m/z
755.2512 [M + Na]+ (calcd for 755.2522). The NMR data
(Tables 2 and 3) revealed that the structure of 7 was similar to
4, but with interchanged acetoxy and hydroxy groups at C-1
(δC 77.1) and C-2 (δC 68.1), respectively. The COSY cross-
peaks of H-10 (δH 2.71)/H-1 (δH 5.01)/H-2 (δH 4.10)/H2-3
(δH 2.24, 1.90) and HMBC cross-peaks from H-1 to C-9 (δC
40.7) and CH3CO-1 (δC 171.6) all supported this structural
relation. The 3J1,2 (3.0 Hz) and 3J1,10 (10.5 Hz) values suggested
that the orientations of H-1, H-2, and H-10 were β-axial,
β-equatorial, and α-axial, respectively. The orientation of H-4
3
3
was β-equatorial, as indicated by the J4,3 and J4,5 values (both
3.0 Hz) and the NOESY cross-peak of H-4 (δH 4.97)/H-6β (δH
1.56). The structure of 7 was thus defined as shown.
The molecular formula of 2-deacetyltinosinenoside D (8)
was defined as C34H42O15 based on the 13C NMR data and an
HRESIMS sodium adduct ion at m/z 713.2405 [M + Na]+
(calcd for 713.2416). Compound 8 was determined to be a
deacetylated derivative of 4, as its molecular mass was 42 amu
fewer than that of compound 4. A hydroxy group replacing the
acetoxy group in 4 was attached to C-2 (δC 71.6) in 8, which
was verified by the chemical shift of H-2 (δH 3.93) and HMBC
cross-peaks from H-2 to C-1 (δC 74.2), C-4 (δC 73.5), and
C-10 (δC 37.5). Analysis of the coupling constants of H-1
Figure 6. Comparison of experimental and calculated ECD spectra of 3.
Tinosinenoside D (4) had a molecular formula of C36H44O16
based on its 13C NMR and HRESIMS data (m/z 755.2505
[M + Na]+, calcd for 755.2522). In the 13C NMR spectrum of 4,
in addition to the signals similar to those of compound 3, 10
extra carbon signals were present, including one methoxy group
(δC 56.8), eight sp2 carbons (δC 113.0, 117.3, 117.4, 123.3, 130.3,
146.7, 150.1, and 151.0), and one ester moiety (δC 168.0).
D
DOI: 10.1021/acs.jnatprod.6b00976
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
E
DOI: 10.1021/acs.jnatprod.6b00976
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
a
Table 3. 13C NMR Data of Compounds 4−10
position
4
5
6
7
8
9
10
1
72.7
75.0
72.0
75.4
72.5
75.0
77.1
68.1
74.2
71.6
73.6
72.4
72.0
32.2
2
3
32.1
34.5
32.1
34.8
35.0
37.2
29.4
4
73.1
74.3
72.8
72.9
73.5
74.5
74.2
5
39.1
40.4
38.9
39.3
39.2
40.7
39.4
6
26.4
27.3
26.6
26.5
26.6
27.4
27.2
Figure 8. Selected NOESY correlations of 5.
7
30.3
30.3
30.4
30.2
30.4
30.4
30.4
8
76.7
76.7
76.6
76.6
76.7
76.8
76.6
The NMR data (Tables 2 and 3) of 9 were similar to those of
compound 8, except for a difference in the chemical shifts
around C-4, which suggested opposite relative configurations of
C-4. The J3β,4 (10.5 Hz) and J4,5 (10.5 Hz) values and the
NOESY cross-peak of H-4 (δH 4.98)/H-6α (δH 0.99) con-
firmed the α-axial orientation of H-4. The structure of 9 was
thus established as shown.
The molecular formula of 2-deacetoxytinosinenoside D (10)
was defined as C34H42O14 by the 13C NMR data and an
HRESIMS ion at m/z 697.2453 [M + Na]+ (calcd for
697.2467), which was 16 amu fewer than that of 8. Its NMR
data (Tables 2 and 3) revealed that compound 10 was struc-
turally related to 8, the difference being the shielded H2-2 (δH
1.81) and C-2 (δC 32.2) resonances, suggesting that 10 was the
C-2 deoxy derivative of 8. This was verified by an analysis of the
COSY cross-peaks of H-10 (δH 1.93)/H-1 (δH 3.79)/H2-2/H2-3
(δH 1.98, 1.65). Thus, the structure of 10 was defined as shown.
Three known diterpenoid glucosides, tinosineside A (11),30
1-deacetyltinosposide A (12),2 and tinosineside B (13),30 were
also isolated. Similar NMR and MS data analyses and comparisons
with literature data were used to establish their structures.
All the isolates and the positive control, doxorubicin (IC50
9
40.9
41.2
41.0
40.7
40.9
41.0
41.4
10
11
12
13
14
15
16
17
20
1′
2′
3′
4′
5′
6′
7′
8′
9′
1-Ac
38.3
42.4
38.5
35.7
37.5
41.2
44.5
36.9
36.8
36.9
37.1
37.0
36.8
37.2
3
3
74.1
74.1
74.0
73.3
74.1
74.0
74.2
127.4
109.8
144.9
141.1
175.0
14.9
127.3
109.7
144.7
141.0
175.0
15.5
127.3
109.8
144.8
141.2
175.1
15.0
127.2
109.7
145.1
141.3
174.7
15.3
127.5
109.8
144.7
141.2
175.2
15.2
127.4
109.8
144.7
141.2
175.3
15.7
127.4
109.7
144.8
141.1
175.2
15.5
168.0
117.3
146.7
130.3
113.0
151.0
150.1
117.4
123.3
168.3
117.0
146.3
130.5
112.6
151.1
150.2
117.5
123.5
167.6
118.5
144.6
130.7
115.2
145.0
150.0
116.7
125.9
168.5
117.6
146.4
130.6
112.8
151.0
150.0
117.4
123.3
171.6
21.6
168.6
117.8
146.3
130.7
113.0
151.0
150.0
117.5
123.3
168.3
117.4
146.1
130.5
112.5
151.0
150.1
117.5
123.5
168.3
117.3
146.4
130.4
112.9
151.0
150.1
117.4
123.4
2-Ac
172.8
21.5
56.8
102.1
74.8
77.8
71.2
78.2
62.4
172.7
21.2
56.8
102.2
74.8
77.9
71.3
78.3
62.5
173.0
21.4
56.6
102.3
74.8
77.8
71.3
78.2
62.5
0.32
0.04 μM), were assessed in terms of their cytotoxic
effects against HeLa cells. Only 1-deacetyltinosposide A (12)
showed mild cytotoxicity, with an IC50 value of 8.35
0.60 μM; no signs of cytotoxicity were found for the other
compounds (IC50 ≥ 10 μM). Additionally, the NO production
inhibitory activity of compounds 1−13 on LPS-activated N9
microglial cells was also tested; the results showed that none of
these compounds had inhibitory activity (IC50 ≥ 100 μM).
6′-OMe
Glc-1″
2″
3″
4″
56.7
102.2
74.8
77.8
71.3
78.3
62.5
56.8
102.3
74.8
77.9
71.3
78.3
62.5
56.8
102.2
74.8
77.9
71.3
78.3
62.5
56.8
102.2
74.8
77.8
71.3
78.3
62.5
5″
6″
a
13C NMR were measured in MeOH-d4 at 125 MHz.
EXPERIMENTAL SECTION
■
General Experimental Procedures. Optical rotations were
recorded on a PerkinElmer 341 polarimeter. UV absorption spectra
were acquired on a PerkinElmer 650 spectrophotometer, and ECD
spectra were measured in MeOH with a Chirascan spectrometer. IR
spectra were obtained with KBr disks on a PerkinElmer Spectrum Two
FT-IR spectrometer. NMR spectra were recorded on a 500 or
400 MHz Bruker AVANCE apparatus. HRESIMS data were acquired
on a Thermo-Scientific Exactive mass spectrometer. Silica gel (200−
300 mesh, Qingdao Marine Chem. Co., Ltd., China), C18 reversed-
phase silica gel (50 μm, YMC, Japan), MCI gel (CHP20, 75−150 μm,
MCC, Japan), and Sephadex LH-20 gel (Pharmacia Biotech, Sweden)
were used for column chromatography (CC). Precoated silica gel
plates (Qingdao Marine Chem. Co., Ltd., GF254) were used for
analytical TLC. After spraying with a color reagent (10% H2SO4 in
EtOH), heating revealed the spots. Semipreparative HPLC was
performed on a YMC-Pack ODS-A column (250 × 20 mm, S-5 μm,
Japan) with an LC3000 instrument (Chuang Xin Tong Heng Science
and Technology Co., Ltd., Beijing, China) equipped with a UV3000
detector (Chuang Xin Tong Heng Science and Technology).
Figure 7. Selected NOESY correlations of 4.
(dd, J = 10.0, 3.0 Hz), H-2 (q, J = 3.0 Hz), and H-10 (dd, J =
11.0, 10.5 Hz) suggested that the orientations of H-1, H-2, and
H-10 were β-axial, β-equatorial, and α-axial, respectively.1 The 3J4,3
and 3J4,5 values (both 3.0 Hz) and the NOESY cross-peak of H-4
(δH 4.94)/H-6β (δH 1.52) revealed the β-equatorial orientation
of H-4. Hence, the structure of compound 8 was elucidated.
The 13C NMR data and HRESIMS ion at m/z 713.2405 [M +
Na]+ (calcd for 713.2416) showed that 4-epi-2-deacetyltinosine-
noside D (9) had the same molecular formula C34H42O15 as 8.
Plant Material. The stems of T. sinensis were collected in May
2014 in Guilin, Guangxi Province, China. The specimen (No.
TS-2014016) was identified by Professor Shao-Qing Tang of the
College of Life Science, Guangxi Normal University, and stored in the
F
DOI: 10.1021/acs.jnatprod.6b00976
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
State Key Laboratory for Chemistry and Molecular Engineering of
Medicinal Resources, Guangxi Normal University.
see Table 3; (+) HRESIMS m/z 755.2508 [M + Na]+ (calcd for
C36H44O16Na 755.2522).
Tinosinenoside E (6): white, amorphous powder; [α]20 −55
Extraction and Isolation. The air-dried stems of T. sinensis
(17.3 kg) were extracted with 95% aqueous EtOH (3 × 100 L) under
reflux. The extract (2.2 kg) was dispersed in H2O and partitioned with
petroleum ether, EtOAc, and n-BuOH. The EtOAc fraction was
concentrated to yield a dark brown gum (226.1 g), which was
separated into seven fractions (Fr1−Fr7) by silica gel CC and gradient
elution first with petroleum ether/acetone (10:1 to 1:1) and then with
CH2Cl2/MeOH (8:1 to 2:1). Fraction Fr6 (24.8 g) was further
fractionated by MCI CC (50:50 to 100:0 MeOH/H2O) to afford eight
subfractions (Fr6.1−Fr6.8). Subfraction Fr6.3 (3.4 g) was subjected to
reversed-phase C18 (RP-C18) CC (20:80 to 100:0 MeOH/H2O) to
yield eight subfractions (Fr6.3.1−Fr6.3.8). Subfraction Fr6.3.2
(232.7 mg) was purified by Sephadex LH-20 CC (MeOH) and then
subjected to silica gel CC (CH2Cl2/MeOH, 20:1) to yield compound 2
(9.0 mg). The purification of subfraction Fr6.3.3 (282.0 mg) by
semipreparative HPLC with 20:80 MeCN/H2O (6 mL/min) as the
isocratic solvent system yielded compound 1 (14.7 mg, tR 43.7 min).
Subfraction Fr6.3.6 (1.2 g) was fractionated by Sephadex LH-20 CC
(MeOH) and purified by semipreparative HPLC (20:80 MeCN/H2O,
6 mL/min) to yield compounds 12 (3.2 mg, tR 23.9 min), 3 (18.6 mg,
tR 27.6 min), and 11 (173.2 mg, tR 53.9 min). Subfraction Fr6.4 (1.2 g)
was separated into four subfractions (Fr6.4.1−Fr6.4.4) via RP-C18 CC
(30:70 to 100:0 MeOH/H2O). Subfraction Fr6.4.2 (134.2 mg) was
subjected to semipreparative HPLC (21:79 MeCN/H2O, 6 mL/min)
to yield compound 8 (16.8 mg, tR 36.1 min). Compound 13 (33.2 mg,
tR 37.4 min) was purified from Fr6.4.4 (96.9 mg) by semipreparative
HPLC (23:77 MeCN/H2O, 8 mL/min). Subfraction Fr6.5 (5.0 g) was
subjected to RP-C18 CC (30:70 to 100:0 MeOH/H2O) to yield
11 subfractions (Fr6.5.1−Fr6.5.11). Subfractions Fr6.5.4 (270.7 mg)
and Fr6.5.6 (376.7 mg) were purified via semipreparative HPLC
(6 mL/min) and eluted with 25:75 MeCN/H2O + 0.1% TFA to yield
compounds 4 (47.4 mg, tR 36.3 min) and 6 (30.7 mg, tR 48.2 min).
Subfraction Fr6.5.8 (244.4 mg) was subjected to Sephadex LH-20 CC
(MeOH) and purified by semipreparative HPLC (25:75 MeCN/H2O,
6 mL/min) to yield compound 9 (8.4 mg, tR 25.1 min). Fr6.5.10
(557.9 mg) and Fr6.5.11 (292.6 mg) were purified via semipreparative
HPLC (6 mL/min) and eluted with 25:75 MeCN/H2O + 0.1% TFA
to obtain compounds 7 (47.6 mg, tR 52.1 min) and 10 (11.1 mg, tR
58.7 min). Subfraction Fr6.6 (2.5 g) was subjected to silica gel CC
(EtOAc/EtOH/H2O, 64:2:1 to 8:2:1) and subsequently to semi-
preparative HPLC (45:55 MeOH/H2O, 8 mL/min) to yield
compound 5 (15.1 mg, tR 65.1 min).
D
(c 0.3, MeOH); UV (MeOH) λmax (log ε) 204 (4.51), 291 (4.13),
320 (4.13) nm; IR (KBr) νmax 3390, 2942, 1717, 1382, 1264,
1
1136, 1066 cm−1; H NMR (MeOH-d4 and DMSO-d6), see Table 2;
13C NMR (MeOH-d4), see Table 3; (+) HRESIMS m/z 755.2501
[M + Na]+ (calcd for C36H44O16Na 755.2522).
Tinosinenoside F (7): white, amorphous powder; [α]20 +3
D
(c 0.3, MeOH); UV (MeOH) λmax (log ε) 205 (4.71), 290 (4.33),
320 (4.31) nm; IR (KBr) νmax 3437, 2939, 1717, 1378, 1254, 1140,
1071 cm−1; 1H NMR (MeOH-d4), see Table 2; 13C NMR (MeOH-d4),
see Table 3; (+) HRESIMS m/z 755.2512 [M + Na]+ (calcd for
C36H44O16Na 755.2522).
2-Deacetyltinosinenoside D (8): white, amorphous powder; [α]20
D
−17 (c 0.3, MeOH); UV (MeOH) λmax (log ε) 205 (4.30), 292 (4.24),
317 (4.25) nm; IR (KBr) νmax 3424, 2937, 1704, 1380, 1258, 1136,
1
1072 cm−1; H NMR (MeOH-d4), see Table 2; 13C NMR (MeOH-
d4), see Table 3; (+) HRESIMS m/z 713.2405 [M + Na]+ (calcd for
C34H42O15Na 713.2416).
4-epi-2-Deacetyltinosinenoside D (9): white, amorphous powder;
[α]20D +17 (c 0.3, MeOH); UV (MeOH) λmax (log ε) 205 (4.35), 293
(4.22), 319 (4.22) nm; IR (KBr) νmax 3427, 2941, 1701, 1383, 1258,
1
1174, 1074 cm−1; H NMR (MeOH-d4 and DMSO-d6), see Table 2;
13C NMR (MeOH-d4), see Table 3; (+) HRESIMS m/z 713.2405
[M + Na]+ (calcd for C34H42O15Na 713.2416).
2-Deacetoxytinosinenoside D (10): white, amorphous powder;
[α]20D −20 (c 0.2, MeOH); UV (MeOH) λmax (log ε) 203 (4.35), 293
(4.22), 320 (4.23) nm; IR (KBr) νmax 3428, 2941, 1707, 1380, 1257,
1131, 1070 cm−1; 1H NMR (MeOH-d4), see Table 2; 13C NMR
(MeOH-d4), see Table 3; (+) HRESIMS m/z 697.2453 [M + Na]+
(calcd for C34H42O14Na 697.2467).
Acid Hydrolysis of Compounds 1−10. Each of 1−10 (each
1.0 mg) was added to 1.0 mL of 6% hydrochloric acid. The reaction
mixture was refluxed at 80 °C for 4 h and extracted with EtOAc (3 ×
2 mL) to remove the aglycone. The aqueous layer was subjected to
silica gel CC (EtOAc/EtOH/H2O, 7:4:1) to obtain the sugar fraction.
The sugar fraction was subjected to HPLC (Jasco LC-4000) under the
following conditions: a Shodex Asahipak NH2P-50 4E (250 mm ×
4.6 mm, 5 μm) column; a Jasco OR-4090 optical rotation detector;
and a 78:22 MeCN/H2O mobile phase (1 mL/min). The retention
time (tR 11.8 min) and positive optical rotation of the sugar were
compared with those of an authentic sample, confirming the sugar to
be D-glucose.
ECD Calculations. The conformational analysis was initially
performed using Confab31 with the MMFF94 force field. The
conformers chosen for the ECD calculations were from above 1% of
the Boltzmann population. The selected conformer was optimized at
B3LYP/6-311G** using DFT. It was further optimized in MeOH
using the conductor-like polarizable continuum model calculation
model. The ECD calculations were conducted using the TDDFT
method at the B3LYP/6-311G** level in MeOH. After overlapping
the Gaussian functions for each transition, the ECD spectrum was
simulated in SpecDis. All calculations were performed with the
Gaussian 09 program package.
Cytotoxicity Assay. Cytotoxicity was measured using the MTT
assay.32 In short, 8 × 103 HeLa cells per well (in 100 μL of culture
medium) were seeded in 96-well plates (Corning). Cells were
incubated with five concentrations (2.5, 5, 10, 20, and 50 μM) of
each compound in triplicate at 37 °C for 48 h, and doxorubicin was
used as a positive control. The MTT solution (20 μL, 5 mg/mL) was
directly dropped into the proper wells. After 4 h, the formazan crystals
of the surviving cells were dissolved by adding 150 μL of DMSO to
each well. The absorbance values of each well at 570 nm were
measured using a microplate spectrophotometer (iMark, Bio-Rad,
USA). The IC50 values were calculated by the Logit method.
NO Production Measurement and Cell Viability Assay. The
Griess reaction was used to measure both the accumulation of nitrite
in the culture supernatants and the NO synthase activity.33 The
viability of the microglial cells was evaluated by the MTT assay.33
Tinosinenoside A (1): white, amorphous powder; [α]20 +2 (c 0.3,
D
MeOH); UV (MeOH) λmax (log ε) 203 (4.17) nm; IR (KBr) νmax
1
3416, 2931, 1738, 1381, 1161, 1068 cm−1; H NMR and 13C NMR
(MeOH-d4), see Table 1; (+) HRESIMS m/z 559.1780 [M + Na]+
(calcd for C26H32O12Na 559.1786).
Tinosinenoside B (2): white solid; [α]20D +47 (c 0.04, MeOH); UV
(MeOH) λmax (log ε) 210 (4.03) nm; IR (KBr) νmax 3428, 2924, 1641,
1425, 1066, 593 cm−1; 1H NMR and 13C NMR (DMSO-d6), see
Table 1; (−) HRESIMS m/z 543.1643 [M + Cl]− (calcd for
C25H32O11Cl 543.1639).
Tinosinenoside C (3): white, amorphous powder; [α]20 −14
D
(c 0.09, MeOH); UV (MeOH) λmax (log ε) 209 (3.73) nm; IR (KBr)
1
νmax 3323, 2942, 1716, 1373, 1264, 1031, 603 cm−1; H NMR and
13C NMR (MeOH-d4), see Table 1; (−) HRESIMS m/z 591.1857
[M + Cl]− (calcd for C26H36O13Cl 591.1850).
Tinosinenoside D (4): white, amorphous powder; [α]20 +28
D
(c 0.2, MeOH); UV (MeOH) λmax (log ε) 209 (5.12), 283 (4.69),
323 (4.37) nm; IR (KBr) νmax 3437, 2944, 1712, 1375, 1260, 1133,
1071 cm−1; 1H NMR (MeOH-d4 and DMSO-d6), see Table 2;
13C NMR (MeOH-d4), see Table 3; (+) HRESIMS m/z 755.2505
[M + Na]+ (calcd for C36H44O16Na 755.2522).
4-epi-Tinosinenoside D (5): white, amorphous powder; [α]20 +7
D
(c 0.3, MeOH); UV (MeOH) λmax (log ε) 203 (4.42), 294 (4.23), 318
(4.22) nm; IR (KBr) νmax 3434, 2932, 1709, 1383, 1258, 1128,
1074 cm−1; 1H NMR (MeOH-d4), see Table 2; 13C NMR (MeOH-d4)
G
DOI: 10.1021/acs.jnatprod.6b00976
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(18) Atta-ur-Rahman; Ahmad, S.; Ali, S.; Shah, Z.; Choudhary, M. I.;
Clardy, J. Tetrahedron 1994, 50, 12109−12112.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnatprod.6b00976.
■
S
(19) Pierre, T. H.; Kamdem, W. J.; Ayafor, F.; Sterner, O.
Phytochemistry 1997, 46, 165−167.
(20) Huang, X. Z.; Cheng, C. M.; Dai, Y.; Fu, G. M.; Guo, J. M.; Yin,
Y.; Liang, H. Molecules 2010, 15, 8360−8365.
Computational section and the ECD spectra of com-
pounds 1−3 as well as the 1D NMR, 2D NMR, and
HRESIMS spectra of compounds 1−10 (PDF)
(21) Li, R.; MN, S. L.; Lee, K. H. Nat. Prod. Rep. 2016, 33, 1166−
1226.
(22) Yonemitsu, M.; Fukuda, N.; Kimura, T.; Komori, T. Liebigs Ann.
Chem. 1987, 1987, 193−197.
(23) Yoshikawa, M.; Morikawa, T.; Kashima, Y.; Ninomiya, K.;
Matsuda, H. J. Nat. Prod. 2003, 66, 922−927.
AUTHOR INFORMATION
Corresponding Author
*E-mail (D. Liang): liangdonggxnu@163.com.
ORCID
Dong Liang: 0000-0002-9765-7548
Notes
The authors declare no competing financial interest.
■
(24) Yoshikawa, M.; Morikawa, T.; Zhang, Y.; Nakamura, S.;
Muraoka, O.; Matsuda, H. J. Nat. Prod. 2007, 70, 575−583.
(25) Li, W.; Huang, C.; Li, S.; Ma, F.; Li, Q.; Asada, Y.; Koike, K.
Planta Med. 2012, 78, 82−85.
́
(26) Bautista, E.; Toscano, A.; Calzada, F.; Díaz, E.; Yepez-Mulia, L.;
Ortega, A. J. Nat. Prod. 2013, 76, 1970−1975.
(27) Khan, M. A.; Gray, A. I.; Waterman, P. G. Phytochemistry 1989,
28, 273−275.
ACKNOWLEDGMENTS
(28) Liu, Z. G.; Li, Z. L.; Bai, J.; Meng, D. L.; Li, N.; Pei, Y. H.; Zhao,
F.; Hua, H. M. J. Nat. Prod. 2014, 77, 792−799.
■
Grants from the following projects are appreciated: the
National Natural Science Foundation of China (21462006,
81402817, 21562017, and 21431001), the Innovation Fund of
the Ministry of Education (IRT_16R15), the Natural Science
Foundation of Guangxi (2016GXNSFGA380005 and
2014GXNSFBA118189), the Department of Education of
Guangxi (YB2014282), the Innovation Project of Guangxi
Graduate Education (YCSZ2015091), and the project of the
State Key Laboratory for Chemistry and Molecular Engineering
of Medicinal Resources (CMEMR2016-A02).
(29) Hao, Z. Y.; Liang, D.; Luo, H.; Liu, Y. F.; Ni, G.; Zhang, Q. J.;
Li, L.; Si, Y. K.; Sun, H.; Chen, R. Y.; Yu, D. Q. J. Nat. Prod. 2012, 75,
1083−1089.
(30) Yonemitsu, M.; Fukuda, N.; Kimura, T.; Isobe, R.; Komori, T.
Liebigs Ann. 1995, 1995, 437−439.
(31) O'Boyle, N. M.; Vandermeersch, T.; Flynn, C. J.; Maguire, A. R.;
Hutchison, G. R. J. Cheminf. 2011, 3, 1−9.
(32) Mosmann, T. J. Immunol. Methods 1983, 65, 55−63.
(33) Hou, Y.; Wu, C. F.; Yang, J. Y.; He, X.; Bi, X. L.; Yu, L.; Guo, T.
Prog. Neuro-Psychopharmacol. Biol. Psychiatry 2006, 30, 1523−1528.
REFERENCES
■
(1) Li, W.; Wei, K.; Fu, H.; Koike, K. J. Nat. Prod. 2007, 70, 1971−
1976.
(2) Dong, L. P.; Chen, C. X.; Ni, W.; Xie, B. B.; Li, J. Z.; Liu, H. Y.
Nat. Prod. Res. 2010, 24, 13−17.
(3) Lam, S. H.; Ruan, C. T.; Hsieh, P. H.; Su, M. J.; Lee, S. S. J. Nat.
Prod. 2012, 75, 153−159.
(4) Atta-ur-Rahman; Ahmad, S. Phytochemistry 1988, 27, 1882−1884.
(5) Liu, C. P.; Xu, J. B.; Zhao, J. X.; Xu, C. H.; Dong, L.; Ding, J.;
Yue, J. M. J. Nat. Prod. 2014, 77, 1013−1020.
(6) Shen, Y. C.; Wang, C. H.; Cheng, Y. B.; Wang, L. T.; Guh, J. H.;
Chien, C. T.; Khalil, A. T. J. Nat. Prod. 2004, 67, 316−321.
(7) Sharma, U.; Bala, M.; Kumar, N.; Singh, B.; Munshi, R. K.;
Bhalerao, S. J. Ethnopharmacol. 2012, 141, 918−926.
(8) Sivasubramanian, A.; Narasimha, K. K. G.; Rathnasamy, R.;
Campos, A. M. F. O. Nat. Prod. Res. 2013, 27, 1431−1436.
(9) Abdelwahab, S. I.; Koko, W. S.; Taha, M. M. E.; Mohan, S.;
Achoui, M.; Abdulla, M. A.; Mustafa, M. R.; Ahmad, S.; Noordin, M. I.;
Yong, C. L.; Sulaiman, M. R.; Othman, R.; Hassan, A. A. Eur. J.
Pharmacol. 2012, 678, 61−70.
(10) Zhou, M.; Geng, H. C.; Zhang, H. B.; Dong, K.; Wang, W. G.;
Du, X.; Li, X. N.; He, F.; Qin, H. B.; Li, Y.; Pu, J. X.; Sun, H. D. Org.
Lett. 2013, 15, 314−317.
(11) Grossman, R. B.; Rasne, R. M. Org. Lett. 2001, 3, 4027−4030.
(12) Nozawa, M.; Suka, Y.; Hoshi, T.; Suzuki, T.; Hagiwara, H. Org.
Lett. 2008, 10, 1365−1368.
(13) Ma, J. S.; Wu, C. Y. Flora of China (Zhongguo Zhiwu Zhi);
Science Press: Beijing, 1996; Vol. 30, Issue 1, p 305.
(14) Qin, X. Y.; Luo, J. Y; Gao, Z. G. Yao Ethnic Medicinals in China;
Ethnic Publish House: Beijing, 2002; p 50.
(15) Manjrekar, P. N.; Jolly, C. I.; Narayanan, S. Fitoterapia 2000, 71,
254−257.
(16) Li, R. W.; David, L. G.; Myers, S. P.; Leach, D. N. J.
Ethnopharmacol. 2003, 85, 61−67.
(17) Pan, Q. M.; Li, Y. H.; Hua, J.; Huang, F. P.; Wang, H. S.; Liang,
D. J. Nat. Prod. 2015, 78, 1683−1688.
H
DOI: 10.1021/acs.jnatprod.6b00976
J. Nat. Prod. XXXX, XXX, XXX−XXX