Polyoxypregnane steroids from the stems of marsdenia tenacissima

Article abstract of DOI:10.1021/np500385b

A new polyoxypregnane aglycone, tenacigenin D (1), and seven new C₂₁ steroid glycosides, tenacissimosides D-J (2-8), were isolated from the stems of Marsdenia tenacissima. Their structures were determined by interpretation of their 1D and 2D NMR and other spectroscopic data, as well as by comparison with published values for related known compounds. Compound 1 was found to circumvent P-glycoprotein (P-gp)-mediated multidrug resistance through an inhibitory effect on P-gp with a similar potency to verapamil. In addition, compound 1 potentiated the activity of erlotinib and gefitinib in epidermal growth factor receptor tyrosine kinase inhibitor (EGFR TKI)-resistant non-small-cell lung cancer cells.

Full text of DOI:10.1021/np500385b

Article
pubs.acs.org/jnp
Polyoxypregnane Steroids from the Stems of Marsdenia tenacissima
Sheng Yao,†,‡,∥ Kenneth Kin-Wah To,‡,§,∥ Ya-Zhou Wang,†,‡ Chun Yin,‡,⊥ Chunping Tang,†,‡
Stella Chai,‡,⊥ Chang-Qiang Ke,†,‡ Ge Lin,*,‡,⊥ and Yang Ye*,†,‡
†State Key Laboratory of Drug Research and Natural Products Chemistry Department, Shanghai Institute of Materia Medica, Chinese
Academy of Sciences, 555 Zu-Chong-Zhi Road, Zhangjiang Hi-Tech Park, Shanghai 201203, People’s Republic of China
‡Joint Research Laboratory of Promoting Globalization of Traditional Chinese Medicines between Shanghai Institute of Materia
Medica, Chinese Academy of Sciences and The Chinese University of Hong Kong, People’s Republic of China
§School of Pharmacy, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong SAR, People’s Republic of China
⊥School of Biomedical Sciences, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong SAR, People’s Republic of
China
S
* Supporting Information
ABSTRACT: A new polyoxypregnane aglycone, tenacigenin
D (1), and seven new C21 steroid glycosides, tenacissimosides
D−J (2−8), were isolated from the stems of Marsdenia
tenacissima. Their structures were determined by interpretation
of their 1D and 2D NMR and other spectroscopic data, as well
as by comparison with published values for related known
compounds. Compound 1 was found to circumvent P-
glycoprotein (P-gp)-mediated multidrug resistance through
an inhibitory effect on P-gp with a similar potency to
verapamil. In addition, compound 1 potentiated the activity
of erlotinib and gefitinib in epidermal growth factor receptor tyrosine kinase inhibitor (EGFR TKI)-resistant non-small-cell lung
cancer cells.
ultidrug resistance (MDR), the ability of cancer cells to
treatment of cancer.11 Chemical investigations have revealed
Macquire resistance to anticancer drugs, severely hinders that this plant is rich in polyoxypregnane glycosides. The
polyoxypregnane glycosides isolated from M. tenacissima have
been identified as tenacigenin B derivatives with an
oligosaccharide chain at C-3, acylated groups such as acetyl,
benzoyl, 2-methylbutyryl, and tigloyl esters at C-11 and/or C-
12,12−14 and an epoxide ring at C-8 and C-14. Several of the
polyoxypregnane compounds isolated from this plant showed
cytotoxic activity against cancer cell lines.13,14c,14d A recent
study reported that tenacissoside C from M. tenacissima
inhibited tumor growth in nude mice bearing K562 human
erythroleukemia xenografts.15 In addition to their cytotoxic
activity for cancer cells, two derivatives of tenacigenin B isolated
from this plant have also been reported to increase drug
sensitivity in P-gp-overexpressing human hepatoma cells.16
Furthermore, a recent in vitro study reported that an extract of
M. tenacissima restored gefitinib sensitivity in resistant non-
small-cell lung cancer (NSCLC) cells.17
the effectiveness of different forms of anticancer chemotherapy.
Many cellular mechanisms are known to cause MDR, including
increased drug efflux, reduced drug uptake, activation of drug-
metabolizing enzymes, activation of DNA repair, mutation of
drug acting targets, and disruption in apoptotic signaling.1,2
Among them, P-glycoprotein (P-gp), which belongs to a family
of ATP-binding cassette (ABC) transporters, is known to play a
crucial role in MDR by increasing drug efflux out of cancer
cells.1,2
On the basis of the strategy of inhibiting drug efflux, three
generations of P-gp inhibitors have been developed to enhance
the effect of chemotherapeutic drugs in MDR cancer cells in
vitro and in vivo during the past few decades.3−6 Nevertheless,
attempts to translate the inhibitors of P-gp transporters into
useful drug targets have been unsuccessful, and no significant
benefit of P-gp inhibition has been achieved yet.
In an in-depth search for potential constituents and MDR-
reversal agents from natural sources, eight new polyoxypreg-
nane compounds (1−8) were isolated and evaluated from M.
tenacissima. Among these eight compounds, only compound 1
was found to significantly circumvent anticancer drug resistance
by inhibiting the P-gp-mediated efflux of substrate anticancer
Natural products and their derivatives are valuable
therapeutic agents, including those obtained from traditional
Chinese medicines (TCM).7 Numerous classes of compounds
isolated from TCM, including coumarins, terpenoids, and
steroids, have been reported to exhibit MDR-modulating
activity.8−10 Marsdenia tenacissima (Roxb.) Wight et Arn.,
belonging to the family Asclepiadaceae, is distributed mainly in
the southwest region of mainland China. Its rhizomes and roots
have been recorded in medicinal texts as being useful for the
Received: May 6, 2014
Published: September 12, 2014
© 2014 American Chemical Society and
American Society of Pharmacognosy
2044
dx.doi.org/10.1021/np500385b | J. Nat. Prod. 2014, 77, 2044−2053

Journal of Natural Products
Article
drugs in the P-gp-overexpressing cancer cell lines. Compound 1
was found to potentiate the activity of two P-gp substrate
epidermal growth factor receptor (EGFR) tyrosine kinase
inhibitors (TKIs), erlotinib and gefitinib, in two resistant
NSCLC cell lines, via a dual action on reversal of both the P-gp-
overexpression and K-ras-mutation-mediated MDR mecha-
nisms.
Analysis of the 1D NMR spectroscopic data of 1 showed this
compound closely resembles 11α-O-2-methylbutyryl-12β-O-
acetyltenacigenin B (1a), which was isolated from the same
plant.16 The significant differences found were that the signal at
δH 3.48 (m) assigned to H-3 in 1a and the signal at δC 70.4
were absent, but a resonance for a carbonyl group appeared at
δC 210.5 in 1. This suggested the hydroxy group at C-3 in 1a
was replaced by a carbonyl group in 1. This deduction was
supported by the cross-peaks of H-2/C-3 and H-4/C-3 in the
HMBC spectrum of 1 (see Supporting Information Figure 1A).
The 2-methylbutyryl and acetyl ester groups were located at C-
11 and C-12, respectively, by interpretation of the HMBC
correlations of H-11/C-1′ (δC 175.5) and H-12/C-1″ (δC
170.6). Therefore, compound 1 was determined to be 11α-O-
2-methylbutyryl-12β-O-acetyl-3-oxo-17α-tenacigenin B.
RESULTS AND DISCUSSION
■
A 95% EtOH extract of air-dried M. tenacissima stems (23.0 kg)
was suspended in water and partitioned successively with
EtOAc and n-BuOH. The EtOAc fraction was subjected to
column chromatography over silica gel, MCI gel, Sephadex LH-
20 gel, and finally preparative HPLC, affording eight new
polyoxypregnane compounds, 1−8 (Figure 1).
Tenacissimoside D (2) gave a molecular formula of
C56H82O24 from the quasimolecular ion peak at m/z
1161.5045 [M + Na]+ in its HRESIMS. The IR absorption
bands at 3419, 1724, 1662, and 1452 cm−1 suggested the
presence of hydroxy, carbonyl, and aromatic functional groups.
1
The H NMR spectrum of compound 2 (Tables 1 and 3)
showed four anomeric proton signals at δH 4.37 (d, J = 7.7 Hz),
4.38 (d, J = 7.4 Hz), 4.60 (d, J = 8.5 Hz), and 4.66 (d, J = 7.8
Hz), with the corresponding carbon resonances occurring at δC
105.1, 106.6, 99.1, and 102.6, indicating four sugar moieties
with a β-orientation. The signals for two methyls at δH 1.27 and
1.28 (d, J = 5.8 Hz) and two methoxy groups at δH 3.36 and
1
3.58 (s) in the H NMR spectrum and the signals at δC 19.3,
18.6, 57.9, and 62.5 suggested that compound 2 possesses a 2,6-
didexoy-3-O-methylpyranosyl moiety and a 6-deoxy-3-O-
methylpyranosyl unit, as supported by TOCSY and HMBC
correlations (see Supporting Information Figure 2A).
Enzymatic hydrolysis of 2 with β-cellulase yielded only one
sugar unit in the aqueous extract, which was identified as β-D-
glucose by comparing the Rf value on thin-layer chromatog-
raphy (TLC) with an authentic sample, as well as with its
specific rotation measurement. The sequence of the sugar chain
of 2 was deduced as β-D-glucopyranosyl-(1→4)-β-D-glucopyr-
anosyl-(1→4)-3-O-methyl-6-deoxy-β-D-allopyranosyl-(1→4)-β-
D-oleandropyranoside by HMBC correlations. Further analysis
on the hydrolysate in the organic extract revealed a single
product from enzymatic hydrolysis of 2, which was identified as
tenacissoside I (2a) from its 1D NMR spectra, a steroid
glycoside isolated from the title plant in 1999.14a Therefore, the
structure of compound 2 was established as 3-O-β-D-
glucopyranosyl-(1→4)-β-D-glucopyranosyl-(1→4)-6-deoxy-3-
O-methyl-β-D-allopyranosyl-(1→4)-β-D-oleandropyranosyl
11α-O-benzoyl-12β-O-acetyl-17α-tenacigenin B.
Figure 1. Structures of 1−8 isolated from Marsdenia tenacissima.
Tenacissimoside E (3) gave a quasimolecular ion at m/z
733.3817 [M + Na]+ in its HRESIMS, suggesting a molecular
formula of C37H58O13. In the 1H NMR spectrum, two anomeric
proton signals at δH 4.56 (d, J = 8.8 Hz) and 4.78 (d, J = 8.2
Hz) (Table 3), combined with resonances at δC 96.8 and 99.2
(Table 4) in the 13C NMR spectrum, suggested two sugar units
to be present with a β-linkage. Interpretation of the TOCSY
and HSQC spectra of compound 3 revealed the presence of the
partial moieties −C(6′)H3−C(5′)H−C(4′)H−C(3′)H−C(2′)-
H2−C(1′)H− and −C(6″)H3−C(5″)H−C(4″)H−C(3″)H−
C(2″)H−C(1″)H−. HMBC correlations were observed from
the methoxy group signal at δH 3.36 (s) to the oxygenated
methine carbon at δC 78.7 (C-3′) and from the methoxy group
signal at δH 3.64 (s) to the oxygenated methine carbon at δC
81.0 (C-3″), indicating the two methoxy groups to be attached
Tenacigenin D (1) was obtained as a colorless, amorphous
powder. HREIMS (m/z 488.2763 [M]+) analysis revealed a
molecular formula of C28H40O7 with nine degrees of
unsaturation. IR absorptions at 1726 and 1708 cm−1 indicated
1
the presence of carbonyl groups. In the H NMR spectrum
(Table 1), characteristic protons for three methyls at δH 1.08
(s), 1.28 (s), and 2.20 (s) and two oxygenated methines at δH
5.40 (t, J = 10.0 Hz) and 5.00 (d, J = 10.0 Hz) were observed.
The proton signals at δH 1.97 (s) and [δH 2.20 (m), 1.42 (m),
and 1.63 (m), 0.88 (t, J = 7.5 Hz), 1.04 (d, J = 7.0 Hz)],
together with carbon resonances at δC 170.6 and 20.8 and at δC
175.5, 41.2, 26.1, 11.7, and 15.3, suggested the presence of
acetyl and 2-methylbutyryl groups, respectively.
2045
dx.doi.org/10.1021/np500385b | J. Nat. Prod. 2014, 77, 2044−2053

Journal of Natural Products
Article
1
Table 1. H NMR Data of the Aglycones of Compounds 1−7 (400 MHz, δ in ppm, J in Hz)
a
b
a
a
a
b
b
position
1
1
2
3
4
5
6
7
1.65, m
1.88, m
2.20, m
2.35, m
1.30, m
1.26, m
1.54, m
1.47, m
1.80, m
3.62, m
1.30, m
1.66, m
1.29, m
1.40, m
1.58, m
1.26, m
1.68, m
1.21, m
1.53, m
1.45, m
1.74, m
3.62, m
1.30, m
1.67, m
1.29, m
1.40, m
1.58, m
1.27, m
1.70, m
1.26, m
1.54, m
1.44, m
1.78, m
3.63, m
1.31, m
1.66, m
1.30, m
1.40, m
1.59, m
1.27, m
1.69, m
1.26, m
2.27, m
1.46, m
1.78, m
3.63, m
1.22, m
1.72, m
1.30, m
1.38, m
1.49, m
1.28, m
1.68, m
1.25, m
1.46, m
1.38, m
1.67, m
3.62, m
1.24, m
1.68, m
1.35, m
1.40, m
1.54, m
1.24, m
1.80, m
1.56, m
1.31, m
1.57, m
3.61, m
1.26, m
1.72, m
1.43, m
1.44, m
1.55, m
1.28, m
1.84, m
2
3
4
2.24, m
2.38, m
5
6
1.79, m
1.46, m
1.64, m
7
1.40, m
1.90, m
9
2.10, d (10.0)
5.40, t (10.0)
5.00, d (10.0)
1.46, m
2.10, d (10.2)
5.55, t (10.2)
5.12, d (10.2)
1.60, m
1.76, d (9.8)
5.04, t (9.8)
3.40, m
1.81, d (9.7)
5.15, t (9.7)
3.42, d (9.7)
1.70, m
1.96, d (8.6)
5.35, t (8.6)
3.36, m
1.51, d (9.5)
3.44, t (9.5)
3.28, d (9.5)
1.58, m
1.90, d (9.7)
5.10, t (9.7)
3.49, d (9.7)
1.62, m
11
12
15
1.69, m
1.71, m
1.65, m
2.00, m
1.98, m
1.98, m
1.99, m
1.96, m
2.00, m
16
1.62, m
1.66, m
2.01, m
2.01, m
2.01, m
1.78, m
1.82, m
1.98, m
2.23, m
2.13, m
2.13, m
2.12, m
2.11, m
2.10, m
17
18
19
21
C-11
2′
2.92, d (7.6)
1.08, s
3.02, d (7.3)
1.12, s
2.65, dd (5.9, 11.9)
0.70, s
2.65, dd (6.2, 11.8)
0.99, s
2.67, dd (5.5, 12.6)
1.04, s
2.92, dd (11.6, 6.1)
0.82, s
3.07, dd (11.7, 6.4)
0.90, s
1.28, s
1.06, s
1.03, s
1.03, s
1.08, s
1.02, s
1.00, s
2.20, s
2.14, s
2.24, s
2.23, s
2.22, s
2.30, s
2.24, s
mBu
Bz
Ac
Tig
Bz
Tig
2.20, m
2.04, s
3′
1.42, m
7.92, br d (7.4)
6.85, d (8.7)
8.04, d (7.4)
6.85, q (8.7)
1.63, m
4′
5′
6′
7′
0.88, t (7.5)
1.04, d (7.0)
7.47, br d (7.4)
7.60, t (7.4)
7.47, br d (7.4)
7.92, br d (7.4)
Ac
1.84, t (8.7)
1.78, d (8.7)
7.44, d (7.4)
7.51, t (7.4)
7.44, d (7.4)
8.04, d (7.4)
1.82, d (8.7)
1.79, s
C-12
Ac
2″
1.97, s
b
1.55, s
a
In CDCl3. In methanol-d4.
to C-3′ and C-3″ of a sugar moiety, respectively. Consequently,
the oligosaccharide chain of compound 3 was deduced as
pachybiose (2,6-dideoxy-4-O-(6-deoxy-3-O-methyl-β-D-allopyr-
anosyl)-3-O-methyl-β-D-arabinohexopyranose), which was
identical to that of tenacigenoside A (3a).18 This deduction
was confirmed by the fact that the mild acid hydrolysate of
compounds 3 and 3a gave the same sugar unit, which was
identified as pachybiose by TLC comparison with the authentic
sample. The 1H and 13C NMR spectroscopic data of the
aglycone of compound 3 resembled closely those of compound
3a, suggesting that they share the same carbon skeleton. The
methyl protons at δH 2.04 (s) and the carbon signals at δC 21.6
and 171.4 supported the presence of an acetyl group, placed at
the C-11 hydroxy group from the HMBC correlation from H-
11 to the carbonyl of the acetyl unit. The β-orientation of H-11
was inferred from the ROESY correlation of H-11/H3-19. The
α-orientation of H-12 was deduced by the coupling constant of
H-11/H-12. The signal of H-17 appeared at δH 2.65 (dd, J = 5.9
and 11.9 Hz), different from that in compound 1, suggesting
that H-17 possesses an α-orientation instead of a β-orientation,
which was supported by the ROESY correlation of H3-18/H3-
21. Therefore, compound 3 was established as 3-O-pachybiosyl-
11-O-acetyl-17β-tenacigenin B.
respectively. Analysis of the NMR data (Tables 1−4) of
compounds 4 and 5 indicated that their structures are similar to
that of compound 3. The observed differences revealed that the
acetyl group in compound 3 is replaced by a tigloyl unit in
compound 4 and by a benzoyl unit in compound 5,
respectively. The location of the acylated groups was
ascertained by HMBC correlations from δH 5.15 (H-11 of
the aglycone) to δC 168.2 (CO of the tigloyl unit) in
compound 4 and from δH 5.35 (H-11 of the aglycone) to δC
166.7 (CO of the benzoyl moiety) in compound 5. Thus, the
aglycones of compounds 4 and 5 were determined to be 11-O-
tigloyl-17β-tenacigenin B and 11-O-benzoyl-17β-tenacigenin B,
respectively. The sugar moiety of compounds 4 and 5 was
identical to that of compound 3, which was confirmed by co-
TLC and the specific rotations of the products after mild acid
hydrolysis of compounds 3−5. Further, the same glycosidation
shift of C-3 was observed for compounds 4 and 5 when
comparing the 13C NMR data with compound 3a, indicating
the same linkage of the sugar moiety. Therefore, the structures
of compounds 4 and 5 were established as shown.
Tenacissimoside H (6) gave a molecular formula of
C41H66O17 based on the quasimolecular ion at m/z 853.4233
[M + Na]+ in its HRESIMS. The 1H NMR spectrum (Tables 1
and 3) of 6 displayed three anomeric signals at δH 4.34 (d, J =
7.6 Hz), 4.64 (dd, J = 9.7, 1.3 Hz), and 4.69 (d, J = 8.3 Hz),
The molecular formulas of tenacissimosides F (4) and G (5)
were determined by HRESIMS as C40H62O13 and C42H60O13,
2046
dx.doi.org/10.1021/np500385b | J. Nat. Prod. 2014, 77, 2044−2053

Journal of Natural Products
Article
Table 2. 13C NMR Data (100 MHz) of the Aglycones of Compounds 1−7 (δ in ppm)
a
b
a
a
a
b
b
carbon
1
2
3
4
5
6
7
1
38.9, CH2
44.7, CH2
210.5, C
37.4, CH2
46.2, CH
26.8, CH2
31.2, CH2
66.4, C
39.3, CH2
28.4, CH2
78.4, CH
36.4, CH2
45.6, CH
30.7, CH2
33.4, CH2
68.7, C
36.9, CH2
28.9, CH2
76.2, CH
34.3, CH2
44.2, CH
26.9, CH2
32.3, CH2
60.8, C
37.3, CH2
28.9, CH2
76.0, CH
34.3, CH2
44.3, CH
27.0, CH2
32.5, CH2
65.8, C
37.4, CH2
28.8, CH2
76.1, CH
34.4, CH2
44.4, CH
27.0, CH2
32.5, CH2
65.9, C
39.9, CH2
30.7, CH2
78.9, CH
36.3, CH2
46.3, CH
28.7, CH2
34.0, CH2
68.2, C
39.1, CH2
30.6, CH2
78.4, CH
36.2, CH2
45.8, CH
28.6, CH2
33.9, CH2
68.0, C
2
3
4
5
6
7
8
9
50.6, CH
38.9, C
53.3, CH
40.7, C
51.8, CH
39.2, C
52.1, CH
39.3, C
52.1, CH
39.4, C
55.4, CH
40.8, C
53.9, CH
40.8, C
10
11
12
13
14
15
16
17
18
19
20
21
C-11
1′
2′
3′
4′
5′
6′
7′
C-12
1″
2″
68.4, CH
74.9, CH
45.7, C
71.3, CH
76.5, CH
47.6, C
70.2, CH
78.8, CH
47.1, C
70.2, CH
79.3, CH
47.4, C
71.0, CH
79.0, CH
47.3, C
69.7, CH
82.0, CH
49.4, C
72.0, CH
80.4, CH
49.5, C
71.5, C
73.6, C
70.5, C
70.7, C
70.6, C
73.4, C
73.3, C
26.5, CH2
24.8, CH2
60.1, CH
16.7, CH3
12.0, CH3
210.6, C
29.6, CH3
mBu
28.2, CH2
26.5, CH2
61.4, CH
17.4, CH3
13.7, CH3
213.6, C
30.8, CH3
Bz
27.9, CH2
25.4, CH2
63.5, CH
10.2, CH3
12.8, CH3
212.1, C
30.6, CH3
Ac
28.0, CH2
25.5, CH2
63.9, CH
10.3, CH3
12.8, CH3
211.9, C
30.8, CH3
Tig
28.0, CH2
25.5, CH2
63.5, CH
10.3, CH3
12.9, CH3
211.9, C
30.6, CH3
Bz
29.2, CH2
26.6, CH2
64.5, CH
11.5, CH3
13.8, CH3
214.5, C
32.6, CH3
29.3, CH2
26.8, CH2
64.1, CH
11.4, CH3
13.8, CH3
214.2, C
32.4, CH3
Tig
175.5, C
41.2, CH
26.1, CH2
11.7, CH3
15.3, CH3
167.9, C
131.8, C
131.0, CH
130.3, CH
135.1, CH
130.3, CH
131.0, CH
Ac
171.4, C
21.6, CH3
168.2, C
128.8, C
138.0, CH
12.2, CH3
14.6, CH3
166.7, C
130.4, C
129.7, CH
128.3, CH
132.9, CH
128.3, CH
129.8, CH
169.9, C
130.6, CH
139.4, CH
12.7, CH3
14.9, CH3
Ac
170.6, C
20.8, CH3
b
172.6, C
20.8, CH3
a
Measured in CDCl3. Measured in methanol-d4.
together with the corresponding 13C NMR signals at δC 106.7,
99.1, and 102.6 (Table 4), indicating the presence of three
sugar moieties with β-linkages. By cellulase hydrolysis,
compound 6 afforded a sugar unit and a steroidal glycoside.
The sugar unit was established as β-D-glucose by comparison
with the authentic sample on TLC, as well as the specific
rotation. This steroidal glycoside was deduced to share the
same skeleton with tenacissoside A (3a) from their same Rf
value on TLC and similar NMR data.18 The sugar sequence was
determined to be β-D-glucopyranosyl-(1→4)-6-deoxy-3-O-
methyl-β-D-allopyranosyl-(1→4)-β-D- oleandropyranosyl by an
HMBC experiment. The oligosaccharide chain could be located
at C-3 via the HMBC correlation from the anomeric proton at
δH 4.64 to C-3. Therefore, the structure of compound 6 was
constructed as shown.
hydroxy group at C-11. Therefore, the structure of compound 7
was established as 3-O-β-D-glucopyranosyl-(1→4)-6-deoxy-3-
O-methyl-β-D-allopyranosyl-(1→4)-β-D-oleandropyranosyl
11α-O-tigloyl-17β-tenacigenin B.
The molecular formula of tenacissimoside J (8) was deduced
as C37H58O13 from the HRESIMS (733.3764 [M + Na]+). Two
anomeric protons at δH 4.56 (d, J = 8.8 Hz) and 4.78 (d, J = 8.2
1
Hz) in the H NMR spectrum and the corresponding carbons
at δC 97.1 and 99.1 suggested the presence of two sugar units
with a β-linkage. The oligosaccharide chain of compound 8 was
identified as pachybiose from the NMR data and confirmed by
comparing the hydrolysis products of compound 8 with those
1
of compound 3 under mild acidic conditions. The H and 13C
NMR data (Table 5) of the aglycone of compound 8 compared
closely with those of tenacigenin A (8a),19 indicating the same
aglycone skeleton in these two compounds. The methyl
protons at δH 2.10 (s) and the carbon resonances at δC 21.4
and 170.2 supported the occurrence of an acetyl group. The
HMBC correlation from H-12 [δH 5.26 (d, J = 4.0 Hz)] to the
carbonyl carbon of the acetyl unit suggested the acetyl group to
be located at C-12. The oligosaccharide chain was assigned at
C-3 on the basis of the long-range correlation from the
anomeric proton at δH 4.56 to C-3 in the HMBC spectrum.
Consequently, the structure of compound 8 was established as
shown.
Tenacissimoside I (7) was obtained as a colorless,
amorphous powder. The molecular formula was determined
1
to be C46H72O18 by HRESIMS. The H and 13C NMR data
(Tables 1−4) closely resembled those of compound 6,
indicating that they share the same nucleus and sugar moiety.
By comparison of its 1H NMR data with those of compound 6,
a set of additional signals at δH 6.85 (q, J = 8.7 Hz), 1.82 (d, J =
8.7 Hz), and 1.79 (s), ascribed to a tigloyl moiety, suggested the
presence of this ester group in compound 7. The HMBC
correlation from H-11 at δH 5.10 (t, J = 9.7 Hz) to the carbonyl
carbon at δC 169.9 revealed the tigloyl group to be linked to the
2047
dx.doi.org/10.1021/np500385b | J. Nat. Prod. 2014, 77, 2044−2053

Journal of Natural Products
Article
1
Table 3. H NMR Data of Sugar Moieties of Compounds 2−7 (400 MHz, δ in ppm, J in Hz)
a
b
b
b
a
a
position
2
3
4
5
6
7
I
D-ole
D-ole
D-ole
D-ole
D-ole
D-ole
1
2
4.60, d (8.5)
1.32, m
2.19, m
3.34, m
3.12, m
3.32, m
1.28, d (5.8)
3.36, s
4.56, d (8.8)
1.48, m
4.56, d (9.7)
1.30, m
4.54 d (8.8)
1.32, m
4.64 dd (1.3, 9.7)
1.33, m
4.62, d (9.3)
1.34, m
2.29, m
2.29, m
2.29, m
2.23, m
2.22, m
3
3.39, m
3.38, m
3.39, m
3.37, m
3.36, m
4
3.33, m
3.34, m
3.34, m
3.16, m
3.15, m
5
3.34, m
3.31, m
3.32, m
3.33, m
3.34, m
6
1.36, d (5.8)
3.36, s
1.34, d (5.4)
3.36, s
1.35, d (5.2)
3.36, s
1.34, d (6.2)
3.38, s
1.34, d (6.2)
3.39, s
OMe-3
II
D-allo
D-allo
D-allo
D-allo
D-allo
D-allo
1
4.66, d (7.8)
3.30, m
3.90, m
3.30, m
3.82, m
1.27, d (6.0)
3.58, s
4.78, d (8.2)
3.46, m
4.78, d (8.3)
3.46, m
4.77, d (8.3)
3.45, m
4.69, d (8.3)
3.30, m
4.70, d (7.8)
3.28, m
2
3
3.79, t (2.8)
3.17, dd (2.8, 9.5)
3.55, m
3.78, t (3.1)
3.17, dd (9.3, 3.1)
3.52, m
3.78, t (2.8)
3.16, dd (2.8, 9.5)
3.53, m
3.92, m
3.93, m
4
3.31, m
3.32, m
5
3.80, m
3.80, m
6
1.25, d (6.2)
3.64, s
1.24, d (6.2)
3.65, s
1.24, d (6.2)
3.64, s
1.28, d (6.0)
3.57, s
1.28, d (6.0)
3.57, s
OMe-3
III
1
D-glc
D-glc
D-glc
4.37, d (7.7)
3.24, m
3.50, m
3.51, m
3.42, m
3.83, m
3.91, m
D-glc
4.34, d (7.6)
3.18, m
4.32, d (7.8)
3.16, m
2
3
3.34, m
3.32, m
4
3.24, m
3.26, m
5
3.21, m
3.22, m
6
3.63, m
3.63, m
3.89, m
3.88, m
IV
1
4.38, d (7.4)
3.21, m
3.33, m
3.32, m
3.28, m
3.64, m
2
3
4
5
6
3.84, m
a
b
In methanol-d4. In CDCl3.
The isolated compounds were screened initially for potential
dependent manner. Importantly, compound 1 was shown to
exhibit a P-gp inhibitory effect similar to verapamil, a known P-
gp inhibitor, at the same concentration (50 μM).
reversal of P-gp-mediated MDR using parental SW620 and P-
gp overexpressing resistant SW620 Ad300 cells. The SW620
Ad300 cells have been fully characterized and proven to be
suitable for evaluating P-pg-mediated MDR reversal activity.21
Among the major transporters responsible for MDR, only P-gp
is overexpressed in SW620 Ad300. As demonstrated in Table 6,
resistant SW620 Ad300 cells are remarkably resistant to
doxorubicin (166-fold), a commonly used P-gp substrate
anticancer drug in the clinic. Among all compounds tested,
only compound 1 (tested at 2 or 10 μM) was found to
significantly potentiate the cytotoxicity of doxorubicin in
SW620 Ad300 cells (Table 6). Its effect was found to be
specific because compound 1 did not affect the cytotoxic
activity of doxorubicin in the parental SW620 cells. Moreover,
compound 1 also did not affect the cytotoxic effect of cisplatin,
a non-P-gp substrate anticancer drug, in both parental SW620
and the P-gp overexpressing resistant SW620 Ad300 cells
(Table 6).
EGFR tyrosine kinase inhibitors are promising molecular-
targeted agents used for the treatment of lung cancer. However,
their usefulness is affected greatly by acquired resistance due to
various mechanisms, including secondary mutations in EGFR,
mutation of the oncogene, K-ras, and overexpression of P-gp. A
panel of three NSCLC cell lines, H292 (sensitive), H460
(resistant; K-ras-mutated), and H460 Vbl (resistant; K-ras-
mutated and P-gp overexpression), was used to investigate the
potentiation effect of compound 1 on the activity of erlotinib
and gefitinib.
In the sensitive NSCLC cell line H292, the concurrent
treatment of compound 1 (10 or 20 μM) did not affect the
activity of either erlotinib or gefitinib (Table 7). On the other
hand, in the resistant NSCLC cells, H460 (with a K-ras
mutation), compound 1 significantly potentiated the growth
inhibitory activity of both erlotinib and gefitinib in a
concentration-dependent manner (Table 7). Since compound
1 was previously found to inhibit P-gp, it was reasonable to
speculate that P-gp inhibition was also involved in this
potentiation. Verapamil, a proven P-gp inhibitor, was also
tested against the H460 cells, but no potentiation of activity of
the EGFR TKIs was observed (Table 7).
Since P-gp mediates MDR by actively effluxing substrate
anticancer drugs out of the cells, the inhibition of P-gp efflux
activity by compound 1 was evaluated subsequently. As
indicated in Figure 2, compound 1 was found to inhibit the
efflux of Rh123 (a fluorescent P-gp probe substrate) in the P-gp
overexpressing resistant SW620 Ad300 cells in a concentration-
2048
dx.doi.org/10.1021/np500385b | J. Nat. Prod. 2014, 77, 2044−2053

Journal of Natural Products
Article
1
Table 4. 13C NMR Data (100 MHz) of Sugar Moieties of
Table 5. H (400 MHz) and 13C NMR (100 MHz) Data of
Compounds 2−7 (δ in ppm)
Compound 8 in CDCl3
a
b
b
b
a
a
carbon
2
3
4
5
6
7
position δH (J in Hz) δC, type position
δH (J in Hz)
1.03 s
δC, type
I
D-ole
D-ole
D-ole
D-ole
D-ole
D-ole
1
1.19, m
1.66, m
1.48, m
1.87, m
3.65, m
1.27, m
1.65, m
1.30, m
1.28, m
1.31, m
2.28, m
38.5, CH2 18
19
16.2, CH3
15.7, CH3
99.7, C
1
2
3
4
5
6
99.1,
CH
96.8,
CH
96.6,
CH
96.7,
CH
99.1,
CH
99.0,
CH
1.03 s
1.19 s
2.10 s
2
28.5, CH2 20
21
38.4,
CH2
36.0,
CH2
36.0,
CH2
36.0,
CH2
38.4,
CH2
38.4,
CH2
24.2, CH3
170.2, C
21.4, CH3
3
4
77.2, CH 1′
33.5, CH2 2′
81.0,
CH
78.7,
CH
78.8,
CH
78.8,
CH
81.0,
CH
81.0,
CH
D-ole
84.4,
CH
79.1,
CH
79.0,
CH
79.2,
CH
84.5,
CH
84.4,
CH
5
6
7
45.8, CH
27.7, CH2
34.1, CH2
1
4.56 d (8.8)
1.48 m
97.1, CH
36.1, CH2
73.0,
CH
71.2,
CH
71.3,
CH
71.3,
CH
73.0,
CH
73.0,
CH
2
2.29 m
19.3,
CH3
18.5,
CH3
18.6,
CH3
18.6,
CH3
19.4,
CH3
19.3,
CH3
3
4
5
6
3.39 m
78.8, CH
79.2, CH
71.3, CH
18.6, CH3
55.6, CH3
8
77.8, C
57.4, CH
35.3, C
3.33 m
OMe-3 57.9,
CH3
55.6,
CH3
55.6,
CH3
55.6,
CH3
58.0,
CH3
57.9,
CH3
9
2.38 m
3.34 m
10
11
1.36 d (5.8)
II
1
D-allo
D-allo
D-allo
D-allo
D-allo
D-allo
4.29, dd
68.3, CH OMe-3 3.36 s
D-allo
102.6,
CH
99.2,
CH
99.2,
CH
99.1,
CH
102.6,
CH
102.6,
CH
(4.0, 1.8)
2
3
4
5
6
73.5,
CH
71.7,
CH
71.7,
CH
71.7,
CH
73.4,
CH
73.4,
CH
12
5.26, d
(4.0)
71.8, CH
1
4.78 d (8.2)
99.1, CH
83.7,
CH
81.0,
CH
78.8,
CH
81.0,
CH
83.7,
CH
83.7,
CH
13
14
15
44.0, C
2
3
4
3.46 m
71.8, CH
81.0, CH
72.8, CH
81.0, C
3.79 t (2.8)
84.5,
CH
72.8,
CH
72.8,
CH
72.8,
CH
84.4,
CH
84.4,
CH
1.58, m
34.3, CH2
3.17 dd (2.8,
9.5)
70.6,
CH
71.2,
CH
71.3,
CH
71.3,
CH
70.6,
CH
70.6,
CH
2.38, m
1.46, m
1.86, m
5
6
3.55 m
71.4, CH
17.8, CH3
61.9, CH3
16
17
22.9, CH2
55.1, CH
1.25 d (6.2)
18.6,
CH3
17.8,
CH3
17.9,
CH3
17.8,
CH3
18.6,
CH3
18.6,
CH3
OMe-3 3.64 s
1.97, d
(6.0)
OMe-3 62.5,
CH3
61.9,
CH3
62.0,
CH3
61.9,
CH3
62.5,
CH3
62.4,
CH3
III
1
D-glc
D-glc
D-glc
105.1,
CH
106.7,
CH
106.7,
CH
Table 6. Effects of Compounds 1−8 on Reversal of P-gp-
Mediated Resistance to Doxorubicin
2
3
4
5
6
75.7,
CH
76.0,
CH
76.0,
CH
a
IC50 SD (μM)
76.8,
CH
78.4,
CH
78.4,
CH
b
treatment
SW620
SW620 Ad300 fold resistance
doxorubicin alone
+ PSC833 (400 nM)
+ 1 (2 μM)
0.17 0.05
0.10 0.06
0.11 0.06
0.22 0.09
0.16 0.06
0.18 0.06
0.12 0.05
0.17 0.04
0.15 0.08
0.18 0.10
0.14 0.03
10.2 1.32
11.5 0.98
9.4 0.76
28.2 0.52
0.22 0.03**
2.1 0.21**
0.85 0.11**
26.8 4.16
23.4 3.41
25.6 3.25
23.4 4.19
24.8 4.38
23.9 3.98
27.7 5.16
11.8 1.53
12.0 0.59
10.2 0.72
166
2
81.6,
CH
72.3,
CH
72.4,
CH
77.0,
CH
78.5,
CH
78.5,
CH
19
+ 1 (10 μM)
+ 2 (10 μM)
+ 3 (10 μM)
+ 4 (10 μM)
+ 5 (10 μM)
+ 6 (10 μM)
+ 7 (10 μM)
+ 8 (10 μM)
cisplatin alone
+ 1 (2 μM)
4
62.8,
CH2
63.5,
CH2
63.5,
CH2
168
130
213
138
165
133
198
IV
1
D-glc
106.6,
CH
2
3
4
5
6
75.4,
CH
78.4,
CH
1.2
71.9,
CH
1.0
1.1
+ 1 (10 μM)
78.6,
CH
a
Values represent the mean SD of three independent experiments
62.9,
CH2
performed in quadruplicate. **p < 0.01, compared with doxorubicin
b
alone in SW620 Ad300 cells. Fold resistance was calculated by the
a
b
In methanol-d4. In CDCl3.
IC50 values of doxorubicin/cisplatin in resistant SW620 Ad300 cells
divided by the IC50 values of doxorubicin/cisplatin in SW620 cells in
the presence or absence of PSC833 or individual compounds tested.
Furthermore, by prolonged exposure to vinblastine, H460/
Vbl cells were derived from H460, and the cells were known to
overexpress P-gp to mediate MDR. Compared with H292
(sensitive) and resistant H460 (K-ras mutation), H460/Vbl was
found to be much less responsive to both erlotinib and gefitinib
(>20-fold) (Table 7), indicating that such additional resistance
was due to its P-gp overexpression. Consistent with the P-gp
inhibitory effect, compound 1 was found to reverse the
resistance of both EGFR TKIs in the H460/Vbl cells, and its
activity at 20 μM was similar or even higher than that of the
known P-gp inhibitor verapamil at 50 μM for the two TKIs
tested (Table 7). Collectively, the results in H460 (K-ras
mutated) and H460/Vbl (K-ras mutated and P-gp over-
expressing) suggest that compound 1 has a dual action and is
2049
dx.doi.org/10.1021/np500385b | J. Nat. Prod. 2014, 77, 2044−2053

Journal of Natural Products
Article
EXPERIMENTAL SECTION
■
General Experimental Procedures. Optical rotations were
measured on a PerkinElmer 341 MC polarimeter. UV spectra were
recorded with a Beckman DU-7 spectrometer. IR spectra were
recorded on a Nicolet Magna FT-IR 750 spectrophotometer using
1
KBr disks. H NMR and 13C NMR spectra were recorded on Bruker
AM-300 or AM-400 and INVOR-600 NMR spectrometers. The
chemical shift (δ) values are given in ppm with TMS as internal
standard, and coupling constants (J) in Hz. EIMS and HREIMS were
recorded on a Finnigan MAT-95 mass spectrometer, while ESIMS and
HRESIMS were recorded on a Micromass Q-Tof Global mass
spectrometer. All solvents used were of analytical grade (Shanghai
Chemical Plant). Column chromatographic separations were carried
out using silica gel (200−300 mesh; Qingdao Marine Chemical Ltd.),
Sephadex LH-20 (25−100 μm; Pharmacia Fine Chemicals), and MCI
gel CHP 20P (high porous polymer, 75−150 μm; Mitsubishi
Chemical Industrials) as packing materials. TLC was carried out on
precoated silica gel GF254 plates (Yantai Chemical Industrials).
Analytical HPLC was performed on a Waters 2690 instrument with
a 996 PAD (photodiode array detector) and an Alltech ELSD 2000
detector. Preparative HPLC was preformed on a Varian Pro-star
solvent delivery module with a Varian Pro-star UV−vis detector.
Chromatographic separations were carried out on C18 columns (220 ×
25 mm, 10 μm, Waters), using a gradient solvent system comprising
H2O and MeCN, with a flow rate of 15.0 mL/min, respectively. β-D-
Glucose (anhydrous, 98%), α-L-glucose (99%), and cellulase from
Aspergillus niger (powder, ≥ 0.3 units/mg solid) were purchased from
Sigma-Aldrich.
Figure 2. Reduced efflux of the P-gp substrate rhodamine 123 by
compound 1 (10, 20, and 50 μM) in P-gp overexpressing resistant
SW620 Ad300 cells.
able to circumvent EGFR TKI resistance mediated by both K-
ras mutation and P-gp overexpression.
In order to confirm the effects on EGFR TKI, an apoptosis
assay was performed in the three NSCLC cell lines after
treatment with gefitinib. Consistent with the responsiveness of
the NSCLC cells to the EGFR TKIs, the induction of apoptosis
by gefitinib (1 μM) was found as follows: H292 > H460 >
H460/Vbl (Figure 3A). While compound 1 alone did not
exhibit any appreciable apoptotic effect, the concurrent
treatment of gefitinib with compound 1 (10 μM) was found
to increase the percentage of apoptotic cells in H460 and
H460/Vbl cells to a considerable extent and to almost
completely circumvent the resistance of gefitinib in both
resistant cell lines (Figure 3B).
In summary, the new polyoxypregnane compound 1 from M.
tenacissima exhibited reversal activity on P-gp-mediated MDR
by inhibiting P-gp. The results also revealed that compound 1
produces sensitization effects on two P-gp substrate anticancer
TKIs, gefitinib and erlotinib, in two resistant NSCLC cell lines
via a dual action on reversal of both P-gp overexpression and K-
ras-mutation mediated MDR mechanisms. The results obtained
have demonstrated that compound 1 generates a promising
MDR-reversing effect via inhibition of different resistant
mechanisms and suggest that further investigation of its
concurrent use with anticancer drugs for MDR cancer therapy
may be worthwhile.
Plant Material. The stems of M. tenacissima were collected in
March 2005 from Simao County, Yunnan Province, People’s Republic
of China. A voucher specimen (no. 200503) was deposited at the
Shanghai Research Center for Modernization of Traditional Chinese
Medicine, Shanghai Institute of Materia Medica, Shanghai.
Extraction and Isolation. The dried stems of M. tenacissima (20.3
kg) were extracted three times with 95% EtOH at room temperature
for 3 days each, which afforded a dark residue (3.4 kg) after
evaporation. The residue was suspended in water and then partitioned
with EtOAc and n-butanol, successively, to give EtOAc (460 g) and
butanol (500 g) fractions, respectively. A portion of EtOAc fraction
(270 g) was subjected to CC (silica gel, petroleum ether−EtOAc, 3:1
to 1:2, 100% EtOAc, and then EtOAc−MeOH, 20:1 to 4:1) to give
Frs. A−H. Compound 1 (23 mg) was recrystallized from Fr. C in a
petroleum ether−EtOAc mixture. Fr. D was subjected to CC over
MCI gel (gradient MeOH−H2O, 65% to 100%, then acetone) and
then silica gel (CHCl3−MeOH, 100:1) to afford compound 8 (38
mg). Fr. E was separated over a MCI gel column (MeOH−H2O, 60%
to 100%, then acetone) and then by preparative HPLC (MeCN−H2O,
20% to 50%, 15 mL/min, 0 to 192 min), yielding compounds 4 (78
mg) and 5 (25 mg). Fr. F was subjected to CC (silica gel, CHCl3−
MeOH, 10:1), then purified by preparative HPLC (MeCN−H2O, 15%
a
Table 7. Sensitization of Cytotoxicity of Erlotinib and Gefitinib by Compound 1 in Human NSCLC Cell Lines
cell line
IC50 (μM) of erlotinib with/without compound 1
EGFR status K-ras status P-gp expression erlotinib alone erlotinib + 1 (10 μM) erlotinib + 1 (20 μM) erlotinib + verapamil (50 μM)
H292
WT
WT
WT
WT
Mut
Mut
low
low
high
1.0 0.19
0.84 0.14
5.4 0.65#
11.1 2.16##
0.60 0.20
3.1 0.42##
6.5 0.80##
1.2 0.15
10.3 1.78
13.3 1.96#
H460
9.5 1.96**
21.5 2.57**
H460/Vbl
cell line
IC50 (μM) of gefitinib with/without compound 1
EGFR status K-ras status P-gp expression
gefitinib alone
gefitinib + 1 (10 μM) gefitinib + 1 (20 μM) gefitinib + verapamil (50 μM)
H292
WT
WT
WT
WT
Mut
Mut
low
low
high
0.19 0.19
0.14 0.14
11.3 2.19#
32.2 3.56
0.16 0.20
7.2 1.63##
21.8 2.54
0.18 0.15
19.6 1.87
25.7 2.76
H460
21.0 2.67**
b
H460/Vbl
ND
a
Data represent mean SD from four independent and reproducible experiments. WT: wild type; Mut = mutant (K-ras Q61H). **p < 0.01,
b
compared with TKI alone in H292 cells, #p < 0.05 and ##p < 0.01 compared with TKI alone in the same cell line. ND: not accurately determined
due to the limitation of gefitinib solubility; its highest concentration tested did not reach 50% eradication of cells.
2050
dx.doi.org/10.1021/np500385b | J. Nat. Prod. 2014, 77, 2044−2053

Journal of Natural Products
Article
Figure 3. Restoration of gefitinib-mediated apoptosis by compound 1 in gefitinib-resistant cell lines. (A) Representative set of data obtained from the
annexin V-7-AAD double staining apoptosis assay. Abscissa: cells positive for annexin-V; ordinate: cells positive for 7-AAD uptake. Lower left
quadrant shows viable cells; lower right, annexin-V positive cells (early apoptotic); upper right, cells positive for both annexin-V and 7-AAD (late
apoptotic). (B) Percentage of apoptosis determined from three independent experiments presented as mean SD (a confidence level of p < 0.05
was considered to be statistically significant).
to 40%, 15 mL/min, 0 to 192 min), to give compounds 3 (38 mg) and
3a (21 mg). Fr. G (2.0 g) was separated by CC (silica gel, CHCl3−
MeOH, 10:1) and then preparative HPLC (MeCN−H2O, 15% to
54%, 15 mL/min, 0 to 192 min) to yield compounds 6 (14 mg) and 7
(30 mg). Fr. H (3.2 g) was separated by CC over silica gel (EtOAc−
MeOH, 5:1), then MCI gel (gradient MeOH−H2O, 40% to 100%,
then acetone), and finally preparative HPLC (MeCN−H2O, 30% to
55%, 15 mL/min, 0 to 192 min), to give compound 2 (38 mg).
Tenacissimoside F (4): colorless, amorphous powder; [α]20D −16 (c
0.21, CHCl3); IR (KBr) νmax 3448, 2933, 1700, 1648, 1450, 1269,
1163, 1082 cm−1; 1H NMR, see Tables 1 and 3; 13C NMR, see Tables
2 and 4; ESIMS m/z 773.3 [M + Na]+, 749.4 [M − H]−; HRESIMS
m/z 773.4056 [M + Na]+ (calcd for C40H62O13Na, 773.4088).
Tenacissimoside G (5): colorless, amorphous powder; [α]20D +14 (c
0.10, CHCl3); IR (KBr) νmax 3434, 2933, 1714, 1627, 1452, 1276,
1068 cm−1; 1H NMR, see Tables 1 and 3; 13C NMR, see Tables 2 and
4; ESIMS m/z 795.3 [M + Na]+, 771.4 [M − H]−; HRESIMS m/z
795.3943 [M + Na]+ (calcd for C42H60O13Na, 795.3932).
Tenacigenin D (1): colorless, amorphous powder; [α]20 +32 (c
D
0.32, CHCl3); IR (KBr) νmax 2934, 1726, 1708, 1452, 1277, 1161,
1072 cm−1; 1H NMR, see Table 1, 13C NMR, see Table 2; EIMS m/z
488 [M]+ (10), 386 (24), 344 (78), 327 (16), 283 (100), 230 (57),
156 (46); HREIMS m/z 488.2763 [M]+ (calcd for C28H40O7,
488.2774).
Tenacissimoside H (6): colorless, amorphous powder; [α]20D +2 (c
0.40, MeOH); IR (KBr) νmax 3423, 2931, 1691, 1637, 1452, 1383,
1
1163, 1078, 617 cm−1; H NMR, see Tables 1 and 3; 13C NMR, see
Tables 2 and 4; ESIMS m/z 853.4 [M + Na]+, 829.4 [M − H]−;
HRESIMS m/z 853.4233 [M + Na]+ (calcd for C41H66O17Na,
853.4198).
Tenacissimoside D (2): colorless, amorphous powder; [α]20D −5 (c
0.25, MeOH); IR (KBr) νmax 3419, 2953, 1724, 1662, 1452, 1375,
1277, 1161, 1072 cm−1; H NMR, see Tables 1 and 3, 13C NMR, see
Tenacissimoside I (7): colorless, amorphous powder; [α]20 0 (c
1
D
Tables 2 and 4; ESIMS m/z 1161.6 [M + Na]+, 1137.7 [M − H]−;
HRESIMS m/z 1161.5045 [M + Na]+ (calcd for C56H82O24Na,
1161.5094).
0.07, MeOH); UV (MeOH) λmax (log ε) 210 (4.12) nm; IR (KBr)
1
νmax 3423, 2933, 1651, 1452, 1383, 1267, 1161, 1078, 989 cm−1; H
NMR, see Tables 1 and 3; 13C NMR, see Tables 2 and 4; ESIMS m/z
935.4 [M + Na]+, 911.6 [M − H]−; HRESIMS m/z 935.4586 [M +
Na]+ (calcd for C46H72O18Na, 935.4661).
Tenacissimoside E (3): colorless, amorphous powder; [α]20 −4 (c
D
0.10, MeOH); IR (KBr) νmax 3425, 2929, 1705, 1628, 1735, 1163,
1082 cm−1; 1H NMR, see Tables 1 and 3; 13C NMR, see Tables 2 and
4; ESIMS m/z 733.5 [M + Na]+, 709.6 [M − H]−; HRESIMS m/z
733.3817 [M + Na]+ (calcd for C37H58O13Na, 733.3775).
Tenacissimoside J (8): colorless, amorphous powder; [α]20 −2 (c
D
0.30, CHCl3); IR (KBr) νmax 3457, 2937, 1739, 1637, 1454, 1384,
1
1236, 1062 cm−1; H and 13C NMR, see Table 5; ESIMS m/z 773.4
2051
dx.doi.org/10.1021/np500385b | J. Nat. Prod. 2014, 77, 2044−2053

Journal of Natural Products
Article
[M + Na]+, 709.3 [M − H]−; HRESIMS m/z 733.3764 [M + H]+
(calcd for C37H58O13Na, 733.3775).
were allowed to incubate with the fluorescent P-gp substrate probe
rhodamine 123 in the presence or absence of compound 1. The
retention of a greater fluorescence signal inside the cells was used as
the readout to measure the inhibition of P-gp efflux activity. The
known P-gp inhibitor verapamil (50 μM) was used as the positive
control in a parallel study.
Apoptosis Assay. H292, H460, and H460/Vbl cells were grown
on a 60 mm tissue culture dish at a density of about 2.0 × 105 cells/
well. They were treated for 48 h with 1 μM gefitinib in the presence or
absence of compound 1 (10 μM). At the end of the treatment, both
floating and attached cells were collected and washed twice with ice-
cold phosphate buffer solution. The extent of apoptosis was
determined by using the APC annexin V apoptosis kit (BD Bioscience,
San Jose, CA, USA) according to the manufacturer’s instructions. Cells
positive for both annexin V and 7-AAD were considered apoptotic.
One-way analysis of variance (one-way ANOVA) with Bonferroni’s
multiple comparison test was used to compare the differences among
various treatment groups. A confidence level of p < 0.05 was
considered significant.
Mild Acid Hydrolysis of Compounds 3, 3a, 4, 5, and 8. MeOH
(15 mL) and 0.1 M H2SO4 (5 mL) were added to compound 3 (15
mg), and the mixture was kept at 60 °C for 30 min. After evaporation
of MeOH under reduced pressure, H2O (15 mL) was added and the
mixture was kept around 60 °C for another 30 min, then cooled to
room temperature. The solution was then extracted with Et2O (10 mL
× 3), and the Et2O layer was washed with water (5 mL × 4), dried
(Na2SO4), and evaporated to dryness. The residue was dissolved in
EtOAc, and compound 3 was crystallized from a petroleum ether−
EtOAc mixture. The aqueous acidic layer of the hydrolysate solution
was neutralized with 5% Ba(OH)2 aqueous solution. The sediment
was filtered and the solution was evaporated. The residue was
identified as pachybiose by TLC comparison with an authentic sample
[chromatography conditions: upper layer of n-BuOH−AcOH−H2O
(4:1:5, v/v), Rf 0.78; CHCl3−MeOH (9:1, v/v), Rf 0.35], and its
specific optical rotation {[α]20 −6.8 (c 0.23, H2O)} was comparable
D
with the published value in the literature {[α]20 −10 (c 0.47,
D
H2O)}.20 The same mild acid hydrolysis procedure was applied to
compounds 3a (4 mg), 4 (5 mg), 5 (8 mg), and 8 (5 mg), and the
aqueous acidic layer of their hydrolysates all contained pachybiose, as
determined by TLC comparison with an authentic sample.
ASSOCIATED CONTENT
* Supporting Information
■
S
Key HMBC and ROESY correlations of 1 (Figure 1A), key
TOCSY, HMBC, and ROSEY correlations of compound 2
Cellulase Treatment of Compounds 2 and 6. Compounds 2 (5
mg) and 6 (7 mg) were treated with cellulase (20 mg) (Sigma-Aldrich,
C1184) in a 0.1 M AcOH−NaOAc (pH 5.0) buffer solution (1 mL) at
37 °C for 7 days. Each reaction mixture was then extracted with
CHCl3. Tenacissoside I (2a, 2 mg) and tenacigenoside A (3a, 3 mg)
were both obtained from the organic layer by preparative TLC. The β-
D-glucose was identified by TLC comparison with the authentic
sample and the specific optical rotation: [α]20D +52 (c 0.10, H2O) [β-
D-glucose: [α]20 +52.7 (c 0.16, H2O); α-L-glucose: [α]20 −56.6 (c
(Figure 2A), and available copies of H and 13C NMR, 1H−1H
1
COSY, HSQC, HMBC, and ROESY spectra for compounds
1−8. This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
*(G. Lin) Tel: 852-39436824. Fax: 852-26035139. E-mail:
linge@cuhk.edu.hk.
■
D
D
0.11, H2O)].
Cell Culture. All cell lines used (except H292) were generous gifts
provided by Dr. Susan Bates (National Cancer Institute, NIH,
Bethesda, MD, USA). A human colon parental SW620 cell line and a
drug selected resistant SW620 Ad300 (P-gp overexpressing) subline
were used to screen for the reversal effect of P-gp-mediated multidrug
resistance. The SW620 Ad300 subline was developed from parental
SW620 cells by stepwise selection in increasing concentrations of
doxorubicin.21 These cells have been fully characterized and proven to
be appropriate for studying MDR-mediated resistance and its reversal.
Among the most common drug transporters, only P-gp is overex-
pressed in SW620 Ad300 cells. The human NSCLC cell lines H292
(ATCC, Manassas, VA, USA), H460, and H460/Vbl were used to test
for potentiation of EGFR TKI activity. H460/Vbl is a resistant subline
developed from H460 by prolonged selection in vinblastine.22 H460/
Vbl is highly resistant to vinblastine (>2000-fold), and it was found to
overexpress only P-gp among all major MDR transporters. All cells
were cultured in RPMI-1640 medium supplemented with 10% fetal
bovine serum, 100 units/mL streptomycin sulfate, and 100 units/mL
penicillin G sulfate and were incubated at 37 °C in 5% CO2.
Growth Inhibition Assay. The growth inhibitory effects of various
standard anticancer drugs (including cisplatin, doxorubicin, erlotinib,
and gefitinib), with or without the concomitant treatment of the
individual test compounds (1−8), were evaluated using a sulforhod-
amine B assay.23 Briefly, cells were seeded into 96-well plates in 100
μL at a seeding density of 3000−5000 cells/well and allowed to
incubate overnight. The cells were then treated with the anticancer
drugs at a range of concentrations in the presence or absence of a fixed
concentration of individual compound tested (2, 10, or 20 μM) and
allowed to incubate at 37 °C in 5% CO2 for 72 h. Each drug
concentration was evaluated in quadruplicate, and controls were tested
in replicates of eight. Each experiment was carried out independently
at least three times. To determine whether differences between IC50
values were significant, Student’s t-test was performed, with p < 0.05
being considered statistically significant.
*(Y. Ye) Tel: 86-21-50806726. Fax: 86-50806726. E-mail:
yye@mail.shcnc.ac.cn.
Author Contributions
∥S. Yao and K. K.-W. To contributed equally to the work and
share first authorship.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support from the National Science and Technology
Major Project “Key New Drug Creation and Manufacturing
Program” (No. 2012ZX09301001-001), the National Natural
Science Funds of China (No. 81302657), the Ministry of
Science and Technology (2010DFA30980), the Chinese
Academy of Sciences (KSZD-EW-Z-004-01), the Shanghai
Commission of Science and Technology (11DZ1970700,
12JC1410300), the Health and Medical Research Fund
(HHSRF Project No. 08090481) from the Food and Health
Bureau, HKSAR, and the Chinese University of Hong Kong
Direct Grant (2041448) is gratefully acknowledged.
REFERENCES
■
(1) Gottesman, M. M.; Fojo, T.; Bates, E. Nat. Rev. Cancer 2002, 1,
48−58.
(2) Krishna, R.; Mayer, L. D. Eur. J. Pharm. Sci. 2000, 11, 265−283.
(3) Robert, J.; Jarry, C. J. Med. Chem. 2003, 46, 4805−4817.
(4) Tan, B.; Piwnica-Worms, D.; Ratner, L. Curr. Opin. Oncol. 2000,
12, 450−458.
(5) Dantzig, A. H.; de Alwis, D. P.; Burgess, M. Adv. Drug Delivery
Rev. 2003, 55, 133−150.
(6) Teodori, E.; Dei, S.; Martelli, C.; Scapecchi, S.; Gualtieri, F. Curr.
Drug Targets 2006, 7, 893−909.
Flow Cytometry-Based Substrate Efflux Assay. A flow
cytometry-based assay was employed to study the inhibition of P-gp
efflux activity by compound 1, as described previously.24 Briefly, cells
2052
dx.doi.org/10.1021/np500385b | J. Nat. Prod. 2014, 77, 2044−2053

Journal of Natural Products
Article
(7) Newman, D. J.; Cragg, G. M. J. Nat. Prod. 2007, 70, 461−477.
(8) Chearwae, W.; Anuchapreeda, S.; Nandigama, K.; Ambudkar, S.
V.; Limtrakul, P. Biochem. Pharmacol. 2004, 68, 2043−2052.
(9) Jain, S.; Laphookhieo, S.; Shi, Z.; Fu, L. W.; Akiyama, S.; Chen, Z.
S.; Youssef, D. T.; van Soest, R. W.; El Sayed, K. A. J. Nat. Prod. 2007,
70, 928−931.
(10) Wang, C.; Zhang, J.; Shen, X. L.; Wan, C. K.; Tse, A. K.; Fong,
W. F. Biochem. Pharmacol. 2004, 68, 843−855.
(11) Jiangsu Institute of New Medicine. Zhong Yao Da Cidian
(Encyclopedia of Chinese Materia Medica); Shanghai Science and
Techonology Press: Shanghai, 1977; p 1976.
(12) (a) Abe, F.; Okabe, H.; Yamauchi, T.; Honda, K.; Hayashi, N.
Chem. Pharm. Bull. 1999, 47, 869−875. (b) Abe, F.; Yamauchi, T.;
Honda, K.; Hayashi, N. Chem. Pharm. Bull. 2000, 48, 154−156.
(13) Luo, S. Q.; Lin, L. Z.; Cordell, G. A.; Xue, L.; Johnson, M. E.
Phytochemistry 1993, 34, 1615−1620.
(14) (a) Chen, J. J.; Zhang, Z. X.; Zhou, J. Acta Bot. Yunnan 1999, 21,
369−377. (b) Deng, J.; Liao, Z. X.; Chen, D. F. Phytochemistry 2005,
66, 1040−1051. (c) Zhang, H.; Tan, A. M.; Zhang, A. Y.; Chen, R.;
Yang, S. B.; Huang, X. Steroids 2010, 75, 176−183. (d) Xue, H. L.;
Huang, X. D.; He, D.; Lin, S. J.; Wang, S.; Niu, T. J. Sichuan Univ.
(Med. Sci. Ed.) 2012, 43, 147−149.
(15) Ye, B. G.; Yang, J. R.; Li, J.; Niu, T.; Wang, S. Planta Med. 2014,
80, 29−38.
(16) Hu, Y. J.; Shen, X. L.; Lu, H. L.; Zhang, Y. H.; Huang, X. A.; Fu,
L. C.; Fong, W. F. J. Nat. Prod. 2008, 71, 1049−1051.
(17) Han, S. Y.; Zhao, M. B.; Zhuang, G. B.; Li, P. P. Lung Cancer
2012, 75, 30−37.
(18) Wang, X. L.; Li, Q. F.; Yu, K. B.; Peng, S. L.; Zhou, Y.; Ding, L.
S. Helv. Chim. Acta 2006, 89, 2738−2744.
(19) Deng, J.; Liao, Z. X.; Chen, D. F. Helv. Chim. Acta 2005, 88,
2675−2682.
(20) Gupta, V.; Kumar, A.; Khare, A.; Khare, N. K. Nat. Prod. Res.
2011, 25, 959−973.
(21) Lai, G. M.; Chen, Y. N.; Mickley, L. A.; Fojo, A. T.; Bates, S. E.
Int. J. Cancer 1991, 49, 696−703.
(22) Caceres, G.; Robey, R. W.; Sokol, L.; McGraw, K. L.; Clark, J.;
Lawrence, N. J.; Sebti, S. M.; Wiese, M.; List, A. F. Cancer Res. 2012,
72, 1−10.
(23) Skehan, P.; Stornet, R.; Scudiero, D.; Monks, A.; McMahon, J.;
Vistica, D.; Warren, J. T.; Bokesch, H.; Kenney, S.; Boyd, M. R. J. Natl.
Cancer Inst. 1990, 82, 1107−1112.
(24) To, K. K.; Robey, R.; Zhan, Z.; Bangiolo, L.; Bates, S. E. Mol.
Cancer Res. 2011, 9, 516−527.
2053
dx.doi.org/10.1021/np500385b | J. Nat. Prod. 2014, 77, 2044−2053