Characterization of chlorinated valepotriates from Valeriana jatamansi

Article abstract of DOI:10.1016/j.phytochem.2012.08.015

HPLC-PDA-MS and TLC analysis were used to look for minor cytotoxic chlorinated valepotriates from whole plants of Valeriana jatamansi (syn. Valeriana wallichii DC.). This resulted in isolation of 15 chlorinated valepotriates, designated as chlorovaltrates A-O, together with six known analogues, (1S,3R,5R,7S,8S,9S)-3,8-epoxy-1,5-dihydroxyvalechlorine, volvaltrate B, chlorovaltrate, rupesin B, (1S,3R,5R,7S,8S,9S)-3,8-epoxy-1-O-ethyl-5- hydroxyvalechlorine, and (1R,3R,5R,7S,8S,9S)-3,8-epoxy-1-O-ethyl-5- hydroxyvalechlorine. Their structures were elucidated by spectroscopic methods including homo- and heteronuclear two-dimensional NMR experiments. Chlorovaltrates K-N, chlorovaltrate and rupesin B showed moderate cytotoxicity against lung adenocarcinoma (A 549), metastatic prostate cancer (PC-3M), colon cancer (HCT-8) and hepatoma (Bel 7402) cell lines with IC₅₀ values of 0.89-9.76 μM.

Full text of DOI:10.1016/j.phytochem.2012.08.015

Phytochemistry 85 (2013) 185–193
Contents lists available at SciVerse ScienceDirect
Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem
Characterization of chlorinated valepotriates from Valeriana jatamansi
Sheng Lin a,b,1, Zhong-Xiao Zhang a,1, Tao Chen c, Ji Ye a, Wei-Xing Dai a, Lei Shan a, Juan Su a,
a,
Yun-Heng Shen a, Hui-Liang Li a, Run-Hui Liu a, Xi-ke Xu a, Hui Wang c, , Wei-dong Zhang
⇑
⇑
a Department of Natural Product Chemistry, School of Pharmacy, Second Military Medical University, 325 Guohe Road, Shanghai 200433, People’s Republic of China
b 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
c Key Laboratory of Nutrition and Metabolism, Institute for Nutritional Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences
and Graduate School of the Chinese Academy of Sciences, Shanghai 200031, People’s Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 April 2012
Received in revised form 26 June 2012
Available online 1 October 2012
HPLC–PDA–MS and TLC analysis were used to look for minor cytotoxic chlorinated valepotriates from
whole plants of Valeriana jatamansi (syn. Valeriana wallichii DC.). This resulted in isolation of 15 chlori-
nated valepotriates, designated as chlorovaltrates A-O, together with six known analogues,
(1S,3R,5R,7S,8S,9S)-3,8-epoxy-1,5-dihydroxyvalechlorine, volvaltrate B, chlorovaltrate, rupesin B,
(1S,3R,5R,7S,8S,9S)-3,8-epoxy-1-O-ethyl-5-hydroxyvalechlorine, and (1R,3R,5R,7S,8S,9S)-3,8-epoxy-1-O-
ethyl-5-hydroxyvalechlorine. Their structures were elucidated by spectroscopic methods including
homo- and heteronuclear two-dimensional NMR experiments. Chlorovaltrates K-N, chlorovaltrate and
rupesin B showed moderate cytotoxicity against lung adenocarcinoma (A 549), metastatic prostate can-
Keywords:
Valeriana jatamansi
Valerianaceae
Chlorinated valepotriates
Dereplication
Cytotoxicity
cer (PC-3M), colon cancer (HCT-8) and hepatoma (Bel 7402) cell lines with IC50 values of 0.89–9.76 lM.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
identified in the presence of major metabolites of bioactive frac-
tions. For example, a desired class of compounds, such as the hal-
ogenated secondary metabolites, could be readily located in the
fractions by HPLC–MS analysis due to their characteristic MS iso-
tope patterns (Oh et al., 2006; Chlipala et al., 2010; Allard et al.,
2012).
Natural products continue to play an important role in the dis-
covery and development of new chemical entities as drug-lead
compounds (Cragg et al., 1997; Newman et al., 2003; Newman
and Cragg, 2007). The remaining major challenge in the isolation
and purification of bioactive natural products is the efficient detec-
tion of a desired class of compounds. Although bioactivity-directed
fractionation is effective, it is usually problematic when multiple,
potentially interfering active or nonspecifically active compounds
are present. Dereplication, the rapid identification of known com-
pounds present in a mixture, is crucial to the fast discovery of no-
vel natural products. Dereplication processes typically combine
chromatographic and spectroscopic methods with database
searching. A hyphenated technique of HPLC coupled to UV photo-
diode array detection (HPLC/UV) and to mass spectrometry
(HPLC/MS or HPLC/MSn) has been extensively used as effective tool
for the online identification of natural products in plant extract
analysis (Gao et al., 2010; Stobiecki et al., 2010; Li et al., 2006).
The advantage of this method is that minor components can be
Valepotriates, a family of iridoid esters exhibiting a variety of
properties of biological significance, have been attracting great
interest in natural products. Phytochemical studies on the plants
of the genus Valeriana (Valerianaceae) have resulted in a series of
valepotriates, some of which showed potent cytotoxic and antitu-
mor activities (Stahi and Schütz, 1980; Tang et al., 2002; Yu et al.,
2005; Becker et al., 1984; Bounthanh et al., 1981). A recent inves-
tigation of Valeriana jatamansi (syn. Valeriana wallichii DC.) led to
the isolation of 29 previously unreported iridoids and 17 known
ones, including four chlorinated valepotriates (Lin et al., 2009,
2010a,b). Interestingly, the major chlorinated valepotriate, volval-
trate B (17), induced a significant percentage of definitive remis-
sions of ovarian tumors in the female mice (Zhang et al., 2010).
This thus extended our attention to the chlorinated valepotriates
of V. jatamansi. A large-scale extract of V. jatamansi (25 kg) was
then undertaken to enrich the minor chlorinated valepotriates.
Application of bioactivity-directed fractionation combined with
HPLC–PDA–MS and TLC analysis has resulted in discovery of fifteen
minor chlorinated valepotriates (1–15) and six known analogues.
Reported herein are the isolation, structure elucidation, and biolog-
ical activities of the isolates.
⇑
Corresponding authors. Tel.: +86 21 54920941; fax: +86 21 54920291 (H.
Wang), tel./fax: +86 21 81871244 (W.-d. Zhang).
E-mail addresses: huiwang@sibs.ac.cn (H. Wang), wdzhangy@hotmail.com
(W.-d. Zhang).
1
These authors have contributed equally to this work.
0031-9422/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.phytochem.2012.08.015

186
S. Lin et al. / Phytochemistry 85 (2013) 185–193
Table 2
2. Results and discussion
Cytotoxicity of Compounds 1–15, 18 and 19 to A 549, PC-3M, HCT-8, and Bel 7402 cell
lines.
2.1. Biological and chemical screening of the minor chlorinated
valepotriates from the whole plants of V. jatamansi
Compound
IC50 (lM)
A 549
PC-3M
HCT-8
Bel 7402
In previous work, LC/( )ESI-MS analysis of chlorinated valepo-
triates showed typical isotopic patterns for monochlorinated sec-
ondary metabolites ([M + Na]+:[M + Na + 2]+ = 3:1, [M + Cl]ꢀ:
[M + Cl + 2]ꢀ = 3:2). In particular, using TLC (CHCl3–MeOH, 20:1),
the chlorinated valepotriates displayed a characteristic green col-
ored spot on spraying with 10% H2SO4 reagent followed by heating.
The fractionation and isolation procedures were then guided by LC/
UV, LC/ESI-MS/MS, TLC, and cytotoxic assays (Table 1 and Sec-
tion 4). The subfraction C3 had the greatest cytotoxic activity
among the C subfractions of the EtOAc fraction. However, after
removing the major compounds volvaltrate B and IVHD-valtrate,
the residue (fraction C3-2) of subfraction C3 still retained the high-
test cytotoxic activity. Thus, fraction C3–2 was selected for further
analysis by HPLC-DAD/ESIMSn to obtain detailed structural infor-
mation (Figs. 1 and 2 and Figures S1–S17 in Supplementary Infor-
mation). The constituents of this subfraction were tentatively
characterized based on their retention times, UV spectra, and
MSn fragmentation information (Table 3). It was evident from
HPLC-MS profile of the fraction C3–2 that monochloriated com-
pounds were present, because at least 20 components contained
distinct peaks with monochloriated isotope patterns. Correspond-
ingly, TLC analysis of fraction C3–2 showed at least five minor spots
under UV light that turned green, with a series of other green and
reddish brown spots (non-chlorinated valepotriates), on spraying
with 10% H2SO4 reagent and heating. These results guided the
targeted isolation of the chlorinated valepotriates in fraction
1
2
3
4
5
6
7
8
>10
>10
>10
>10
6.58
4.27
>10
>10
>10
7.52
3.17
8.26
9.76
4.70
>10
5.70
3.71
0.39
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
4.56
2.35
>10
>10
>10
>10
2.35
7.23
2.70
2.11
>10
3.54
2.03
1.16
4.92
2.34
3.48
6.25
7.16
3.69
2.04
3.83
4.79
1.06
6.25
2.59
2.24
1.54
7.15
4.22
2.71
9.51
8.62
6.02
2.90
2.32
1.93
0.89
9.51
3.28
1.41
1.63
9
10
11
12
13
14
15
18
19
PABa
a
Pseudolaric acid B was used as a positive control.
rine, and volvaltrate B (17), respectively, by comparing their
retention times with those of authentic samples in HPLC experi-
ments and their Rf values in TLC analysis.
Analysis of the literature showed that only four chlorinated
valepotriates, rupesins A-C and chlorovaltrate, have been previ-
ously reported from the other Valerianaceae species (Yang et al.,
2006; Fuzzati et al., 1996). However, only peak 19 possessed the
same molecular weight as rupesin B or chlorovaltrate. These sug-
gested that at least fifteen chlorinated valepotriates present in V.
jatamansi should be new. Targeted isolation of these chlorinated
valepotriates by HP-20 macroporous adsorbent resin, ODS, Sepha-
dex LH-20, preparative TLC, and preparative HPLC led to identifica-
tion of fifteen new chlorinated valepotriates, chlorovaltrates AO
(1–15), and two known ones, chlorovaltrate (18) and rupesin B (19).
C3–2. HPLC-DAD analysis indicated that the UV spectra of peaks
14 and 17–20 exhibited two broad maximum absorption bands
at k 200 and 250–260 nm, which were characteristic for the
presence of diene valepotriates. On the other hand, the UV spectra
(kmax ꢁ 204 nm) of peaks 1–13, 15 and 16 indicated the presence of
monoene valepotriates (Fig. 3). These characteristic UV spectra can
be used to differentiate between diene and monoene valepotriates.
Moreover, all 20 peaks except for peaks 1–3, and 5 yielded a frag-
ment ion at m/z 102 or m/z 204 in the positive MS2 mode, which
was characteristic for the loss of one or two isovaleroxy units (Fig-
ures S1–S17, Supplementary Information). Therefore, peaks 4, and
6–21 could be characterized as monoene and diene chlorinated
valepotriates containing one or two isovaleroxy units, while peaks
1–3, and 5 could be elucidated as monoene chlorinated valepotri-
ates with no isovaleroxy units. Peaks 1–3, and 13 were readily
assigned as (1S,3R,5R,7S,8S,9S)-3,8-epoxy-1,5-dihydroxyvalechlorine
(16), (1S,3R,5R,7S,8S,9S)-3,8-epoxy-1-O-ethyl-5-hydroxyvalechl-
orine, (1R,3R,5R,7S,8S,9S)-3,8-epoxy-1-O-ethyl-5-hydroxyvalechlo-
2.2. Structure elucidation of obtained constituents
Chlorovaltrate A (1) was obtained from peak 7 and gave a
molecular formula of C17H23ClO7 on the basis of HRESIMS and
NMR spectroscopic data. Its IR, 1H and 13C NMR spectra are similar
to those of (1S,3R,5R,7S,8R,9S)-3,8-epoxy-1,5-dihydroxyvalechlo-
rine (16) except for the presence of an isovaleroxy substituent
(dC 170.7, 43.2, 25.7, 22.3 and 22.2) (Lin et al., 2010a). Online LC-
DAD/MSn analyses suggested the presence of one isovaleroxy
group in 1 (Figure S1, Supplementary Information). This substitu-
ent was readily assigned at C-1 since the signal for H-1 was signif-
icantly deshielded due to the acylation effect. Further proof that
the isovaleroxy group was attached to C-1 was provided by an
Table 1
Cytotoxicity of the various fractions of the whole plants of V. jatamansi.
HMBC correlation from H-1 to the carbonyl carbon at 170.7.
C
Fraction
IC50
(
l
g mLꢀ1
)
Based on biogenetic considerations, chlorovaltrate A (1) was as-
sumed to possess the same absolute configuration as in 16.
Therefore, 1 was elucidated as (1S,3S,5R,7S,8S,9S)-3,8-epoxy-5-hy-
droxy-1-isovaleroxyvalechlorine.
Chlorovaltrate B (2), corresponding to peak 4, was deduced as
18H27ClO8 from its HRESIMS. This formula required 5° of unsatu-
A 549
PC-3M
HCT-8
Bel 7402
Alcohol extract
EtOAc fraction
Fr. A
Fr. B
Fr. C
Fr. D
Fr. C1
Fr. C2
22.3
16.3
>50
38.1
12.6
23.4
38.4
6.9
18.8
12.2
48.1
25.4
7.8.
18.9
25.3
6.1
29.4
15.8
42.3
31.2
11.5
21.1
41.2
8.2
25.2
12.1
>50
41.4
12.0
16.7
>50
6.4
C
ration, one less than that of chlorovaltrate A (1). Analysis of the 1H
and 13C NMR spectroscopic data (Tables 4 and 6) showed reso-
nances similar to those of 1, but also indicated that the oxo bridge
connecting C-3 and C-8 had been converted to 3-methoxy and 8-
hydroxy functionalities, and this appeared to explain the 32 Da
mass reduction found between the two compounds. HMBC correla-
tion from methoxy protons to C-3 and no observed correlation be-
Fr. C3
6.3
6.8
7.6
6.1
Fr. C3–2
Fr. C4
Fr. C5
5.2
13.4
18.6
4.6
15.8
28.4.
6.1
21.7
19.5
4.8
25.4
16.6

S. Lin et al. / Phytochemistry 85 (2013) 185–193
187
OR11
O
OR11
O
11
O
OH
O
OH
R5
H
5
OR3
3
O
O
H
O
R7O
Cl
R7O
Cl
9
7
O
1
O
8
Cl
Cl
H
H
H
10
OH
R1
Iv
Iv
Iv
Iv
Iv
Iv
OH
OH
OR1
OR1
OR1
OR1
R1 R5
R7 R11
2 R1=Iv, R3=Me
R7
Iv
R11
Hiv
Ac
Ac
H
Ac
Iv
16 R1=H
5
Iv
Iv
Iv
Iv
Iv
Iv
Iv
Iv
OH
H
liv
Iv
H
12
13
14
3
4
R1=Iv, R3=Et
1
R1=Iv
6
OH
OH
OH
OH
H
Ac
Ac
Ac
Ac
H
Abv
Mv
Iv
R1=R3=Et
7
4
8
Me 15
O
CH3
O
9
18
19
Et
Iv
Iv
Ac
2
10
11
17
3
1
CH3
1
CH3
2
5
H
Ac
Iv
9
OH Ac
liv
Ac
Iv
O
CH3
7
CH3
7
8
6
6
6
O
CH3
4
OH
10
CH3
C4H3
O
O
5
CH3
1
CH3
O
2
3
4
1
2
3
2
3
1
O
CH3
5
3
O
CH3
1
CH3
O
CH3
2
5
4
Abv 5
Mv
liv
Hiv
Fig. 1. Compounds 1–19.
Fig. 2. HPLC/ESIMSn analysis of the cytotoxic fraction C3 from V. jatamansi: (a) total ion chromatogram, all mass (b) UV chromatogram (k = 204 nm).
tween H-3 and C-8 supported the deduction. The relative configu-
ration of 2 was assigned by analysis of NOESY NMR spectroscopic
data. Analogous correlations from protons on the cyclopentane
ring confirmed that 2 has the same configuration as 1 at C-5, C-7,
close on the same face of the tetrahydro-2H-pyran ring (Figure S29,
Supporting Information). On the basis of these data and biogenetic
considerations, 2 was assigned as (1S,3R,5R,7S,8S,9S)-5,8-dihy-
droxy-3-methoxy-1-isovaleroxyvalechlorine.
C-8, and C-9. NOESY correlations from H-3 to H-6
cated that these protons, opposite of H-9 and MeO-3, were spatially
a
and H-1 indi-
Chlorovaltrate C (3) from peak 6 was an analogue of 2.
Comparison of the NMR and MS data between 2 and 3

188
S. Lin et al. / Phytochemistry 85 (2013) 185–193
Table 3
MS characteristics of compounds from fraction C3-2 of V. jatamansi.
Peak (tR, min)
Compounds
[M + Na]+, m/z
[M + Cl]ꢀ, m/z
HPLC/ESIMSn, m/z (Positive mode)
Monoene/diene form shown by UV
4 (30.8)
5 (31.6)
6 (33.5)
7 (37.8)
8 (38.5)
9 (41.3)
10 (42.1)
11 (44.8)
12 (46.0)
13 (47.2)
14 (51.4)
15 (54.9)
16 (56.4)
17 (58.2)
18 (58.6)
19 (60.5)
20 (60.8)
2
4
3
1
5
7
10
8
9
17
15
6
11
12
14
18, 19
13
429
387
443
397
557
415
441
429
443
599
439
499
483
539
495
481
539
441
399
455
409
569
427
453
441
455
611
451
511
495
551
507
–
MS2, 327, 291
Monoene
Monoene
Monoene
Monoene
Monoene
Monoene
Monoene
Monoene
Monoene
Monoene
Diene
Monoene
Monoene
Diene
Diene
Diene
MS2, 351, 305, 245
MS2, 341, 305
MS2, 295
MS2, 355, 253, 235
MS2, 379, 313, 295, 213, 177
MS2, 339, 237
MS2, 393, 327, 295, 227, 191
MS2, 407, 341, 295, 241, 205
MS2, 397, 295
MS2, 337, 235, 213, 177
MS2, 397, 295, 217
MS2, 381
MS2, 465, 335, 277
MS2, 277, 219
MS2, 379, 319, 277, 219
MS2, 437, 379, 277, 219
551
Diene
Fig. 3. Characteristic UV spectra for monoene chlorinated valepotriates (D D
4,11, a; 3,4, b) and diene chlorinated valepotriates (c).
demonstrated that the only difference was replacement of
resonances for OMe in 2 by those for OEt in 3. Accordingly, 3
was established as (1S,3R,5R,7S,8S,9S)-5,8-dihydroxy-3-ethoxy-1-
isovaleroxyvalechlorine.
the presence of one (R)-a-isovaleroxyisovaleroxy group instead of
an isovaleroxy group at C-11 in 6 (Tables 4 and 6). Further inter-
pretation of the 2D NMR spectroscopic data of 6 confirmed this
assignment and completed the full assignments the 1H and 13C
NMR signals. Therefore, 6 was determined as (1S,5R,7S,8S,9S)-
7-acetoxy-10-chloro-5-hydroxy-1,11-diisovaleroxy-5,6-dihydroval-
trate chlorohydrin.
Chlorovaltrate G (7) from peak 9 was found to possess the
molecular formula, C17H25ClO8, as shown from its HRESIMS. The
1H and 13C NMR spectroscopic data (Tables 4 and 6) showed signals
similar to those of 6, but H2-11 was significantly shielded and one
isovaleryl group was absent. These spectroscopic differences sug-
gested that chlorovaltrate G (7) was an 11-O-deisovaleryl deriva-
tive of 6. The 2D NMR spectroscopic data were also consistent
Chlorovaltrate D (4) from peak 5 was inferred as C16H25ClO7 by
HRESIMS. Analysis of 1H and 13C NMR data suggested that chlor-
ovaltrate D (4) had the same iridoid nucleus as chlorovaltrate B
(2), but lacked the resonances corresponding to the isovaleroxy
and O-methyl units and instead contained two O-ethyl groups at
C-1 and C-3, respectively. Further interpretation of 2D NMR spec-
troscopic data allowed structure 4 to be assigned. NOESY NMR
experiments with 4 displayed identical correlations to those ob-
served in 2 and 3. Hence, 4 was identified as (1S,3R,5R,7S,8S,9S)-
1,3-diethoxy-5,8-dihydroxyvalechlorine.
Chlorovaltrate E (5) from peak 8 was analyzed for C25H39ClO10
by HRESIMS and NMR spectroscopic data. Since the NMR spectra
of 5 showed the absence of one acetyl signal, it was clear that com-
pound 5 was the deacetyl analogue of volvaltrate B (17) (Tables 4
and 6). As expected, H-7 of 5 was significantly shifted upfield due
to the deacylation effect when compared with the 1H NMR spectro-
scopic data of 17 (Table 4). Thus, chlorovaltrate E (5) was assigned
as the 7-O-deacetyl derivative of 17, which was confirmed by the
with those observed for 7. Thus, 7 was assigned to be
(1S,5R,7S,8S,9S)-7-acetoxy-10-chloro-5,11-dihydroxy-1-isovaler-
oxy-5,6-dihydrovaltrate chlorohydrin.
Chlorovaltrate H (8) from peak 11 was assigned the molecular
formula of C18H27ClO8 by HRESIMS. The 1H and 13C NMR spectro-
scopic data of
8 closely resembled those of 6 except that
11-isovaleroxy group was replaced by a methoxy unit [dH 3.37
(3H, s); dC58.2]. This was supported by the downfield shift of C-11
from dC 71.4 in 8 to dC 61.3 in 6, as well as by gHSQC and gHMBC
experimentsof8. Consequently, 8wasproposedtobe(1S,5R,7S,8S,9S)-
7-acetoxy-10-chloro-5-hydroxy-1-isovaleroxy-11-methoxy-5,6-
dihydrovaltrate chlorohydrin.
detailed 2D NMR experiments and negative ½a D20
ꢂ
value of 5. Thus,
5 was deduced as (1S,5R,7S,8S,9S)-10-chloro-5,7-dihydroxy-11-
[(R)-a-(isovaleroxy)isovaleroxy]-1-isovaleroxy-5,6-dihydroval-
trate chlorohydrin.
Chlorovaltrate F (6) from peak 15 gave a molecular formula of
22H33ClO9 on the basis of HRESIMS. The NMR spectroscopic fea-
Chlorovaltrate I (9) from peak 12 was assigned as the 11-O-
ethyl derivative of 7 by comparison of the NMR and MS data be-
C
tures of 6 were similar to those of volvaltrate B (17), except for
tween 7,
8 and 9 (Tables 4–6). Hence, 9 was defined as

S. Lin et al. / Phytochemistry 85 (2013) 185–193
189
Table 4
1H NMR spectroscopic data (d) for compounds 1–8.a
No.
1 (CDCl3)
2 (CDCl3)
3 (CDCl3)
4 (Me2CO-d6)
5 (CDCl3)
6 (CDCl3)
7 (Me2CO-d6)
8 (CDCl3)
1
3
6
6.43 d (3.6)
5.34 s
6.25 d (2.4)
5.20 s
6.26 d (6.0)
5.31 s
5.02 d (3.6)
5.29 s
2.57 dd (15.0, 7.2)
6.33 d (4.8)
6.55 s
6.55 d (1.8)
6.54 s
6.45 d (2.4)
6.34 s
6.50 d (1.2)
6.39 s
2.61 dd (13.8, 2.70 dd (14.4, 2.88 dd (15.0, 6.0)
7.2) 6.0)
2.06 dd (13.8, 2.04 dd (14.4, 2.01 dd (15.0, 2. 4)
2.4)
4.98 dd (7.2,
2.4)
2.41 dd (13.8, 2.58 dd (13.2, 2.61 dd (13.8, 2.58 dd (13.8,
4.2) 6.0) 5.4) 6.0)
2.10 dd (13.8, 2.11 dd (13.2, 2.06 dd (13.8, 2.10 dd (13.8,
4.2)
4.03 dd (4.2,
4.2)
1.84 dd (15.0, 6. 0)
5.15 dd (7.2, 6.0)
6.0)
5.03 dd (6.0,
6.0)
7.2)
4.92 dd (6.0,
7.2)
6.0)
4.96 dd (5.4,
6.0)
5.4)
4.89 dd (6.0,
5.4)
7
5.13 dd (6.0, 2.4)
9
10
2.67 d (3.6)
3.82 d (11.4)
3.79 d (11.4)
5.42 s
5.16 s
2.20 m
2.57 d (2.4)
3.86 d (11.4)
3.77 d (11.4)
5.45 s
5.39 s
2.22 m
2.52 d (6.0)
3.85 d (11.4)
3.81 d (11.4)
5.49 s
5.36 s
2.22 m
2.42 d (3.6)
4.08 d (11.4)
3.96 d (11.4)
5.34 s
5.29 s
3.84 m, 3.45 m
1.13 t (6.6)
2.59 d (4.8)
4.11 d (11.4)
3.80 d (11.4)
4.81 d (12.0)
4.75 d (12.0)
2.20 m
2.71 d (1.8)
3.77 d (11.4)
3.70 d (11.4)
4.75 d (12.6)
4.67 d (12.6)
2.23 mb
2.65 d (2.4)
3.94 d (11.4)
3.90 d (11.4)
4.31 brd (12.6) 4.18 d (12.0)
4.17 brd (12.6) 3.91 d (12.0)
2.21 m
2.08 m
0.96 d (6.6)
0.96 d (6.6)
2.06 s
2.68 d (1.2)
3.84 d (11.4)
3.74 d (11.4)
11
R1
2
3
4
5
2
2
3
4
2.25 m
2.11 m
0.97 d (6.6)
0.97 d (6.6)
2.09 s
2.09 m
2.11 m
2.10 m
2.10 m
2.10 m
0.96 d (6.6)b
0.95 d (6.6)b
2.09 s
0.96 d (6.6)b
0.95 d (6.6)b
2.10 s
0.96 d (6.6)b
0.95 d (6.6)b
2.09 s
0.98 d (6.6)b
0.97 d (6.6)b
0.98 d (6.6)c
0.97 d (6.6)c
2.11 s
R7
R11
2.04 s
4.75 d (4.8)
2.23 m
2.22 mb
2.10 m
1.01 d (7.2)c
1.00 d (7.2)c
2.28 m
0.98 d (6.6)c
0.97 d (6.6)c
5
7
8
9
2.10 m
0.98 d (6.6)b
0.97 d (6.6)b
10
OMe/
OEt
3.50 s
/3.96 m, 3.57 m; 1.25 t /3.93 m, 3.59 m; 1.22 t
(6.6) (6.6)
3.37 s
a
1H NMR data () were measured in 600 MHz NMR instrument. Proton coupling constants (J) in Hz are given in parentheses. The assignments were based on DEPT, 1H–1H
COSY, NOESY, HSQC, and HMBC experiments.
b,c
Assignment may be interchanged in each column.
(1S,5R,7S,8S,9S)-7-acetoxy-10-chloro-11-ethoxy-5-hydroxy-1-iso-
valeroxy-5,6-dihydrovaltrate chlorohydrin.
and OH-8 indicated that 12 has the same configuration as 6 at C-
1, C-7, C-8 and C-9. Hence, 12 was deduced as (1S,7S,8S,9S)-10-
chloro-11-(3-hydroxyisovaleroxy)-1,7-diisovaleroxyvaltrate
chlorohydrin.
Chlorovaltrate J (10) from peak 10 was determined to be the
5,6-dihydrovaltrate chlorohydrin valepotriate with two isovaler-
oxy groups from its spectroscopic data and 2D NMR data analysis.
The two isovaleroxy groups were located at C-1 and C-11 by the
HMBC correlations from H-1 and H2-11 to the corresponding car-
bonyl carbon signals at dC 171.4 and 173.0, respectively. The con-
figuration of 10 was identical with that of didrovaltrate acetoxy
hydrin on the basis of the NOE correlation of H-5/H-9/H2-10 and
the coupling constant (9.6 Hz) between H-5 and H-9 and the neg-
Chlorovaltrate M (13) from peak 20 was formulated as C24H33-
ClO10 from HRESIMS. The 1H and 13C NMR spectra indicated that
chlorovaltrate M (13) possessed the same iridoid nucleus as 12.
Three acyloxy substituents were readily assigned as an isovaleroxy,
a b-(acetoxy)isovaleroxy, and an acetoxy group by analysis of the
1H and 13C NMR spectroscopic data of 13 (Tables 5 and 6). Finally,
the observed HMBC correlations from H-1, H-7 and H2-11 to the
respective carbonyl carbon signals at dC 170.5, 168.9, and 170.8
served to locate the three acyloxy substituents at C-1, C-7 and C-
11, respectively. Thus, 13 was defined as (1S,7S,8S,9S)-11-acet-
oxy-7-[b-(acetoxy)isovaleroxy]-10-chloro-1-isovaleroxyvaltrate
chlorohydrin.
ative ½a 2D0
ꢂ
value. Accordingly, 10 was assigned as (1S,5S,7S,8S,9S)-
10-chloro-7-hydroxy-1,11-diisovaleroxy-5,6-dihydrovaltrate
chlorohydrin.
Spectroscopic analysis of chlorovaltrate J (11) from peak 16
indicated that it is an acetyl derivative of 10 due to the presence
of an acetyl unit in 11 (Tables 5 and 6). This was supported by
its molecular formula, showing 58 Da more than that of 10. The rel-
atively downfield proton chemical shift for H-7 in 11 compared to
H-7 in 10 suggested that 11 was the 7-O-acetyl derivative of 10.
Therefore, 11 was determined as (1S,5S,7S,8S,9S)-7-acetoxy-10-
chloro-1,11-diisovaleroxy-5,6-dihydrovaltrate chlorohydrin.
The molecular formula of chlorovaltrate K (12) from peak 17
was inferred as C25H37ClO9 by HRESIMS. The 1H NMR spectrum dis-
played characteristics of a valtrate chlorohydrin nucleus (Lin et al.,
2009), exhibiting signals for two olefinic methines [dH 6.85 (s, H-3)
and 5.81 (dd, J = 3.0, 2.4 Hz, H-6)] and an oxymethylene [dH 4.81
and 4.70 (each 1H, J = 12.6 Hz, H2-11)]. Characteristic resonances
for three acyloxy moieties found in 12, including two isovaleroxy
groups and a 3-hydroxyisovaleroxy moiety, were observed in the
1H, 13C, and DEPT NMR spectra. Further 2D NMR data analysis al-
lowed the three acyloxy moieties to be positioned at C-1, C-7,
and C-11, respectively. In the NOESY spectrum, the observed
NOESY correlations between H-9 and H2-10 and between H-7
Chlorovaltrate N (14) from peak 18 was established the molec-
ular formula C23H33ClO8 by its HRESIMS. The 1H and 13C NMR spec-
tra of chlorovaltrate N (14) were very similar to those of 13, with
the same iridoid nucleus and 1-isovaleroxy and 11-acetoxy again
present; however, two methyls at d 0.88 (3H, t, J = 7.2 Hz) and
0.93 (3H, d, J = 6.6 Hz) in the 1H NMR spectrum of 14 indicated
the presence of a 3-methylvaleroxy residue in place of the b-(acet-
oxy)isovaleroxy at C-7 in 14. The 13C and 2D NMR data closely
matched those observed for 14 and confirmed the position of the
3-methylvaleroxy group at C-7. The absolute configuration of 3-
methylvaleroxy in 14 was determined by basic hydrolysis. Basic
hydrolysis of 14 yielded 3-methylvaleric acid (Section 4). Compar-
ison of the optical rotation data of 3-methylvaleric acid
{½a 2D0
ꢀ 6:3(c 0.15, CH2Cl2)} and (R)-3-methylvaleric acid
ꢂ
½
a 2D5
ꢂ
ꢀ 8:4 (c 1.0, CH2Cl2)} demonstrated the R configuration of
the 3-methylvaleric acid (Belzecki et al., 2000). Therefore, 14 was
determined as (1S,7S,8S,9S)-11-acetoxy-10-chloro-1-isovaleroxy-
7-[(R)-3-methylvaleroxy]-valtrate chlorohydrin.

190
S. Lin et al. / Phytochemistry 85 (2013) 185–193
Table 5
1H NMR spectroscopic data (d) for compounds 9–15.a
No.
9 (Me2CO-d6)
10 (Me2CO-d6)
11 (CDCl3)
12 (Me2CO-d6)
13 (CDCl3)
14 (Me2CO-d6)
15 (Me2CO-d6)
1
3
5
6
6.46 d (1.8)
6.37 s
6.14 d (6.0)
6.46 s
3.05 m
2.04 m
1.97 m
6.19 d (5.4)
6.46 s
2.93 m
2.24 m
2.10 m
6.21 d (9.6)
6.85 s
6.24 d (10.8)
6.67 s
6.21d (9.6)
6.83 s
6.17 d (10.8)
6.65 s
2.61 dd (13.8, 6.0)
2.02 dd (13.8, 5.4)
4.97 dd (6.0, 5.4)
2.66 d (1.8)
3.94 d (11.4)
3.90 d (11.4)
4.13 d (12.0)
4.04 d (12.0)
2.21 m
5.81 dd (3.0, 2.4)
5.76 dd (2.4, 2.4)
5.79 dd (2.4, 2.4)
5.80 dd (3.0, 2.4)
7
9
10
4.08 brd (2.4)
2.39 dd (9.6, 6.0)
4.04 d (11.4)
3.80 d (11.4)
4.62 d (12.0)
4.44 d (12.0)
2.23 mb
5.02 dd (4.2, 3.6)
2.91 dd (9.6, 5.4)
3.84 d (11.4)
3.73 d (11.4)
4.61 d (12.0)
4.43 d (12.0)
2.19 mb
5.49 d (3.0)
3.00 dd (9.6, 2.4)
4.08 d (11.4)
3.86 d (11.4)
4.81 d (12.6)
4.70 d (12.6)
2.38 m
5.42 d (2.4)
2.91 dd (10.8, 2.4)
3.98 d (11.4)
3.78 d (11.4)
4.70 d (12.0)
4.62 d (12.0)
2.29 m
5.49 d (2.4)
3.01 dd (9.6, 2.4)
4.09 d (11.4)
3.87 d (11.4)
4.75 d (12.0)
4.66 d (12.0)
2.37 m
5.48 d (3.0)
2.96 dd (10.8, 2.4)
4.07 d (11.4)
3.86 d (11.4)
4.22 s
4.21 s
2.37 m
2.13 m
11
R1
R7
2
3
4
5
2
2.08 m
2.08 mc
2.10 mc
2.13 m
2.15 m
2.13 m
0.94 d (6.6)b
0.93 d (6.6)b
2.05 s
0.98 d (6.6)d
0.97 d (6.6)d
0.96 d (6.6)d
0.95 d (6.6)d
2.07 s
1.00 d (6.6)b
0.99 d (6.6)b
2.19 (m)
1.00 d (6.6)
1.00 d (6.6)
3.01 d (14.4)
2.81 d (14.4)
1.00 d (6.6)
1.00 d (6.6)
2.30 m, 2.13 m
1.00 d (6.6)
1.00 d (6.6)
2.19 (m)
3
4
5
6
2
3
4
5
2.08 (m)
1.87 m
2.08 (m)
0.96 d (6.6)b
0.95 d (6.6)b
1.50 sb
1.53 sb
1.95 s
2.04 s
1.39 m, 1.22 m
0.88 t (7.2)
0.93 d (6.6)
1.98 s
0.96 d (6.6)b
0.94 d (6.6)b
R11
2.17 mb
2.17 mb
2.45 s
2.07 mc
2.07 mc
0.96 d (6.6)d
0.95 d (6.6)d
0.96 d (6.6)d
0.95 d (6.6)d
1.24 s
1.24 s
OEt
3.51 m, 1.15 t (7.2)
a
1H NMR data () were measured in 600 MHz NMR instrument. Proton coupling constants (J) in Hz are given in parentheses. The assignments were based on DEPT, 1H–1H
COSY, NOESY, HSQC, and HMBC experiments.
b,c,d
Assignment may be interchanged in each column.
Table 6
13C NMR spectroscopic data (d) for compounds 1–15, 18 and 19.
No.
1a
2a
3a
4b
5a,c
6a
7b
8a
9b
10b
11a
12b
13a
14b
15b
18a
19a
1
3
4
5
6
7
8
9
10
11
1
2
3
4
5
1
2
3
4
5
6
7
1
2
3
4
5
89.5
94.4
89.9
101.0 99.2
89.0
98.2
99.1
148.7
78.8
43.7
80.9
83.0
60.5
51.6
110.9
64.3
15.8
90.8
89.3
90.4
89.5
90.4
91.2
89.4
93.5
92.3
93.5
93.5
92.4
92.4
144.5 143.1 139.4 140.6 140.7
114.8 113.6 119.2 114.5 116.3
141.2 141.0 148.9 148.1 148.8 145.8 148.0 148.0
115.5 113.0 110.2 108.6 110.3 115.1 108.7 108.8
151.0 142.5 145.4
77.5
46.5
74.0
83.1
45.5
45.2
76.1
40.3
79.4
81.2
57.5
49.1
79.1
43.7
80.4
82.5
58.8
49.5
75.4
43.8
78.2
83.0
55.6
50.5
62.6
70.9
41.0
79.8
80.3
54.4
49.6
61.3
72.7
43.0
80.1
81.8
55.9
50.9
60.8
71.9
42.3
79.8
80.9
53.9
49.5
71.4
72.2
43.1
80.1
81.8
55.8
51.0
69.0
33.8
38.6
77.8
83.9
46.2
50.8
64.1
32.7
35.0
80.0
81.8
45.4
49.1
63.3
140.5 139.5 140.5 141.1 139.3 139.5
118.6 117.1 118.5 117.9 117.4 117.2
83.6
81.5
50.5
49.4
61.3
83.3
80.2
49.3
48.4
60.8
83.6
81.5
50.5
49.4
61.4
83.9
81.4
50.7
49.5
59.5
83.2
80.3
49.3
48.4
60.9
83.5
80.3
49.3
48.5
60.5
108.9 114.1 114.8
170.7 171.5 171.2
43.2
25.7
22.3
22.2
R1
171.0 171.8 170.7 171.6 170.6
171.4 171.1 171.1 170.5 171.1 171.1 170.5 170.5
43.4
25.4
22.4
22.4
43.7
25.7
22.6
22.6
43.4
25.9
22.6
22.6
43.4d 43.7
25.7
22.4
22.4
170.8 171.8 171.0 171.7
20.9
43.4
25.8
22.6
22.6
43.7
26.3
22.6
22.6
43.9
26.5
22.7
22.7
43.4d 43.6
25.7 26.4
43.2
25.7
44.0
26.4
44.0
26.4
43.4d 43.4d
25.7e 25.7
26.3
22.6
22.6
22.4e 22.8d 22.3d 22.6d 22.7d 22.4f
22.3e 22.7d 22.2d 22.5d 22.6d 22.3f
22.4e
22.3e
R7
169.5 170.5 170.0
21.0
169.8 172.1 168.9 172.3 172.1 171.8 169.7
21.0
21.0
21.3
21.0
21.2
21.0
44.0
26.4
44.2
79.3
43.6
43.6
43.5d 21.0
25.8e
32.8e 26.4
22.6d 26.6e 30.5e
22.5d 26.5e 11.6
170.4 19.6
22.6d 22.4f
22.5d 22.3f
22.2
R11
173.0
43.2d
25.7
22.4
22.4
173.0 172.9 172.3 170.8 171.0
170.8 172.9
20.9
43.9
26.5
22.7
22.7
43.3d 48.3
20.9
20.9
43.2d
25.7
25.7
69.5
22.4e 29.9
22.3e 29.9
22.4e
22.3
e
OMe/
OEt
55.9
/64.8,
15.1
/64.8,
15.9
58.2
65.8,
15.5
a
b
c
13C NMR data () were measured in CDCl3 for 1–3, 5, 6, 8, 11, 13, 18 and 19 at 150 MHz. The assignments were based on DEPT, 1H–1H COSY, HSQC, and HMBC experiments.
13C NMR data () were measured in acetone-d6 for 4, 7, 9, 10, 12, 14 and 15 at 150 MHz. The assignments were based on DEPT, 1H–1H COSY, HSQC, and HMBC experiments.
Data of the a-isovaleroxyisovaleroxy group at C-11 of 5: d 170.0 (CR11-1), 77.1 (CR11-2), 30.3 (CR11-3), 19.0 (CR11-4), 17.6 (CR11-5), 173.6 (CR11-6), 43.3 (CR11-7), 25.9 (CR11-8),
22.6 (CR11-9), 22.5 (CR11-10).
d,e,f
Assignment may be interchanged in each column.
The molecular formula of chlorovaltrate L (15) from peak 14
son with data from 12, the 1H NMR, 13C NMR, and mass spectro-
was found to be C20H29ClO7 on the basis of HRESIMS. In compari-
scopic data established the lack of a 3-hydroxyisovaleryl moiety

S. Lin et al. / Phytochemistry 85 (2013) 185–193
191
in 15. Furthermore, the relatively upfield proton chemical shift for
H2-11 in 15 compared to H2-11 in 12 suggested that the loss of the
3-hydroxyisovaleryl moiety from OH-11, leaving two isovaleroxy
groups to be located at C-1 and C-7, respectively. This conclusion
was supported by analysis of 2D NMR data. Thus, 15 was estab-
lished as (1S,7S,8S,9S)-10-chloro-11-hydroxy-1,7-diisovaleroxy-
valtrate chlorohydrin.
The known compounds 18 and 19 from peak 19 were identified
as chlorovaltrate and rupesin B, respectively, by comparing the UV,
MS, and NMR data with those reported in the literature and analyz-
ing their 2D NMR data (Yang et al., 2006; Fuzzati et al., 1996; Beck-
er et al., 1984).
Although compounds 6, 8, and 11 have been presented in sepa-
rated patents, no spectroscopic data have been reported (Thies and
David, 1983; Zhang et al., 2009). By using the MTT method com-
pounds 1–15, 18 and 19 were tested in vitro against several human
cancer cell lines including lung adenocarcinoma (A 549), meta-
static prostate cancer (PC-3M), colon cancer (HCT-8) and hepatoma
(Bel 7402) cell lines, and pseudolaric acid B was used as a positive
control (Lin et al., 2009). Compounds 11–14, 18 and 19 showed
moderate cytotoxicity against all tested cell lines with IC50 values
barium specimen was deposited in School of Pharmacy, Second
Military Medical University, PR China (herbarium No. 2009–03-1).
4.3. HPLC-DAD/ESIMSn Analyses of Constituents in Fractions
For online HPLC-DAD/ESIMSn analyses, an Agilent 1100 Series
liquid chromatography system chromatography was utilized
equipped with a quaternary pump, degasser, column oven, auto-
sampler, and diode array detector operating at 204 and 254 nm,
which was coupled to MSD Trap XCT ESI mass spectrometer. The
positive ion ESIMSn experiments were conducted using conditions
as follows: drying temperature, 350 °C; drying gas (N2), 10 L/min;
nebulizer gas, high-purity nitrogen (N2), 35 psi; collision gas, ultra-
high-purity helium (He); skimmer, 40 V; injection rate, 5.0
HPLC separation performed on an Eclipse C18 XDB column
m) using a mobile phase of MeOH and H2O
lL/min.
(4.6 ꢃ 150 mm, 5
l
(flow rate, 0.8 mL/min; temperature, 25 °C). The mobile phase gra-
dient program was 10:90 (t = 0 min), 60:40 (t = 40 min), 95: 5
(t = 70 min), and 95:5 (t = 80 min). Each fractions (2.0 mg) was dis-
solved in MeOH (1 mL), filtered through a 0.45 lm microporous
membrane, and stored at 4 °C until analysis.
of 0.89–9.76
lM. Except for compounds 14 the remaining com-
pounds exhibited selective cytotoxicity against metastatic prostate
4.4. Extraction and isolation
cancer (PC-3M) cell line with IC50 values of 0.89–9.51 lM (Table 2).
The extraction and isolation procedures were guided by LC-MS
screening and cytotoxic assays. The air-dried whole plants of V. jat-
amansi (25 kg) were powdered and extracted with EtOH–H2O,
(40.0 L & 95:5, v/v) at room temperature for 3 ꢃ 48 h. The EtOH ex-
tract was evaporated under reduced pressure to yield a residue
(2640 g). The residue was suspended in H2O (2000 mL) and then
partitioned with EtOAc (5 ꢃ 2000 mL). The EtOAc fraction
(1050 g) was applied to a HP-20 macroporous adsorbent resin col-
umn. Successive elution of the column with EtOH–H2O (1:9, 3:7.
7.3 & 95:5) (5000 mL each) yielded four fractions (A–D, 125, 286,
319, and 168 g, respectively) after removing the solvents. A portion
of cytotoxic fraction C (300 g) was separated by Sephadex LH-20
CC eluting with MeOH to give five subfractions (C1–C5), which
were subjected to the cytotoxic assay. Subfractions C2 (55 g) and
C3 (61 g) showed significant cytotoxicity. Subfraction C2 mainly
contained volvaltrate B (17) and IVHD-valtrate by TLC analysis.
However, after removing volvaltrate B (17) and IVHD-valtrate by
CC over Sephadex LH-20 eluting with MeOH, the residue (fraction
C3–2, 42 g) of subfraction C3 still remained the highest cytotoxic
activity. HPLC-MSn and TLC analysis of fraction C3–2 indicated that
fraction C3–2 was rich in chlorinated valepotriates. A portion of
subfraction C3–2 (35 g) was then separated by medium-pressure li-
quid chromatography over reversed-phase silica gel eluting with a
gradient of increasing MeOH (10–100%) in H2O to give ten fractions
(C3–2–1–C3–2–10). Subfraction C3–2–3 (3.5 g), including mainly peaks
1–6 was fractioned by CC over Sephadex LH-20 using CHCl3–MeOH
(1:1) to afford six subfractions (C3–2–3–1–C3–2–3–6). Subfraction
C3–2–3–2 (125 mg) was separated by reversed-phase preparative
3. Conclusion
Target isolation of minor chlorinated valepotriates from the
whole plants of V. jatamansi was conducted by application of bio-
activity-directed fractionation combined with dereplication princi-
ple. This resulted in the fast discovery of fifteen chlorinated
valepotriates (1–15) and six known analogues, most of which
exhibited cytotoxicity against A 549, PC-3M, HCT-8 and Bel 7402
cell lines with IC50 values of 0.89–9.76 lM. With such an approach,
the time-consuming isolation of common natural products is
avoided and the success rate of discovering new and/or bioactive
compounds is increased dramatically.
4. Experimental
4.1. General experimental procedures
Optical rotations were measured on a Perkin-Elmer 341 polar-
imeter. IR and UV spectra were recorded on a Bruker Vector-22
and Shimadzu UV-2550 UV–visible spectrophotometers, respec-
tively. NMR spectra were acquired on a Bruker Avance 600 MHz
or Avance 300 MHz NMR spectrometer in CDCl3 or Me2CO-d6 with
TMS as an internal standard. ESIMS and HRESIMS were acquired on
an Agilent LC/MSD Trap XCT and a Q-TOF micro mass spectrometer
(Waters, USA), respectively. Column chromatography (CC) was
performed using silica gel (100–200, 300–400 mesh, and 10–
40
and Sephadex LH-20 (40–70
AB, Uppsala, Sweden). Semipreparative HPLC was conducted on
an ODS column (Kromasil, 5
lm; Huiyou Silical Gel Development Co. Ltd., Yantai, PR China)
HPLC (RP18, 5 lm, 25010 mm, 208 nm, MeCN–H2O, 45:55) to give
l
m; Amersham Pharmacia Biotech
2 (6.2 mg) and 3 (5.8 mg), respectively. Subfractions C3–2–3–3
and C3-2-3-4 mainly contained 16, (1S,3R,5R,7S,8S,9S)-3,8-epoxy-
1-O-ethyl-5-hydroxyvalechlorine, and (1R,3R,5R,7S,8S,9S)-3,8-
epoxy-1-O-ethyl-5-hydroxyvalechlorine, respectively. Subfraction
C3-2-3-5 (160 mg) was separated by reversed-phase preparative
lm, 300 ꢃ 10 mm) using a PDA UV
detector at 208 nm. Preparative TLC (0.4–0.5 mm) was performed
on precoated silica gel GF254 plates (Yantai). Zones were visualized
under UV light (254 nm) or by spraying with 10% H2SO4 followed
by heating.
HPLC (RP18, 5
l
m, 250 ꢃ 10 mm, 208 nm, MeCN–H2O, 42:58) to
give 4 (13.2 mg). Subfraction C3–2–4 (346 mg), mainly consisting
of peaks 7 and 8, was isolated after preparative TLC developed
with CHCl3–MeOH (20:1) and then reversed-phase preparative
4.2. Plant material
HPLC (RP18,
5 lm, 208 nm, MeCN–H2O, 45:55) to give 1
Whole plants of V. jatamansi were collected in Gaopo, Guizhou
Province, PR China, in July 2009, and identified by Prof. Shun-zhi
He in Guiyang University of Traditional Chinese Medicine. A her-
(10.3 mg) and 5 (2.8 mg). Subfraction C3–2–5 (5.6 g), consisting of
peaks 9–12, was separated by CC over Sephadex LH-20 using
petroleum ether-CH3Cl-MeOH (5:5:1) to afford six subfractions

192
S. Lin et al. / Phytochemistry 85 (2013) 185–193
(C3–2–5–1–C3–2–5–6). Subfractions C3–2–5–2 (248 mg) and C3–2–5–3
(157 mg) were purified by preparative TLC developed with
4.10. Chlorovaltrate F (6)
CHCl3–MeOH (20:1) to give
Separation of subfraction C3–2–5–4 (202 mg) by reversed-phase
preparative HPLC (RP18, 5
7
(16.2 mg) and 10 (4.1 mg).
Colorless oil, ½a D20
ꢂ
ꢀ 76:0 (c 0.10, MeOH); UV (MeOH) max (log
e)
204 (3.92) nm; IR (KBr)
2970, 1738, 1366, 1229, 1217,
max
lm, 250 ꢃ 10 mm, 208 nm, MeCN–H2O,
1092 cmꢀ1; for 1H NMR (CDCl3, 600 MHz) and 13C NMR (CDCl3,
150 MHz) spectroscopic data, see Tables 4 and 6; ESIMS m/z 499
[M + Na]+, 475 [M ꢀ H]ꢀand 511 [M + Cl]ꢀ; HRESIMS m/z
499.1722 [M + Na]+ (calcd for C22H33ClO9Na, 499.1711).
48:52) afforded 8 (7.5 mg) and 9 (8.4 mg). Using the same proce-
dure as described above for the isolation of 1 and 5, subfractions
C3–2–6 (1.8 g) yielded a majority of known 17 and a little of 6
(9.0 mg) and 15 (4.2 mg). Subfractions C3–2–7 (1.6 g) and C3–2–8
(2.8 g) were separated by Sephadex LH-20 CC using petroleum
ether–CHCl3–MeOH (5:5:1) and then by reversed-phase prepara-
4.11. Chlorovaltrate G (7)
tive HPLC (RP18, 5
l
m, 250 ꢃ 10 mm, 208 nm, MeCN–H2O, 55:45)
Colorless oil, ½a D20
ꢂ
ꢀ 39:0 (c 0.30, MeOH); UV (MeOH) max (log
e)
to give 11 (4.6 mg) and 12 (22.5 mg), respectively. Subfraction
C3–2–9 (6.5 g), including peaks 18–20, was fractioned by
Sephadex LH-20 CC using petroleum ether-CHCl3–MeOH (5:5:1)
to afford five subfractions (C3–2–9–1–C3–2–9–5). Subfractions C3–2–9–2
(362 mg) and C3–2–9–3 (269 mg) were purified by preparative TLC
developed with CHCl3–MeOH (20:1) to yield 13 (23.8 mg), 14
(14.6 mg), 18 (6.5 mg) and 19 (5.9 mg), respectively.
203 (3.90) nm; IR (KBr)
3400, 2969, 1737, 1372, 1231,
max
1082 cmꢀ1; for 1H NMR (Me2CO-d6, 600 MHz) and 13C NMR (Me2-
CO-d6, 150 MHz) spectroscopic data, see Tables 4 and 6; ESIMS m/z
415 [M + Na]+, 391 [M ꢀ H]ꢀ and 427 [M + Cl]ꢀ; HRESIMS m/z
415.1146 [M + Na]+ (calcd for C17H25ClO8Na, 415.1136).
4.12. Chlorovaltrate H (8)
4.5. Chlorovaltrate A (1)
Colorless oil, ½a D20
ꢂ
ꢀ 43:8 (c 0.19, MeOH); UV (MeOH) max (log
e)
204 (3.92) nm; IR (KBr)
3418, 2960, 1734, 1372, 1294,
Colorless oil, ½a D20
ꢂ
+ 79.8 (c 0.39, MeOH); UV (MeOH) max (log e)
max
1082 cmꢀ1; for 1H NMR (CDCl3, 600 MHz) and 13C NMR (CDCl3,
150 MHz) spectroscopic data, see Tables 4 and 6; ESIMS m/z 429
[M + Na]+ and 441 [M + Cl]ꢀ; HRESIMS m/z 429.1303 [M + Na]+
(calcd for C18H27ClO8, 429.1292).
204 (3.84) nm; IR (KBr) max 2969, 1738, 1366, 1217, 940 cmꢀ1; for
1H NMR (CDCl3, 600 MHz) and 13C NMR (CDCl3, 150 MHz) spectro-
scopic data, see Tables 4 and 6; ESIMS m/z 397 [M + Na]+ and 409
[M + Cl]ꢀ; HRESIMS m/z 397.1038 [M + Na]+ (calcd for C17H23ClO7-
Na, 397.1030).
4.13. Chlorovaltrate I (9)
4.6. Chlorovaltrate B (2)
Colorless oil, ½a D20
ꢂ
ꢀ 49:6 (c 0.15, MeOH); UV (MeOH) max (log
e)
Colorless oil, ½a D20
ꢂ
þ 83:5 (c 0.31, MeOH); UV (MeOH) max (log
e)
204 (3.93) nm; IR (KBr)
3422, 2962, 1736, 1370, 1292,
max
1080 cmꢀ1; for 1H NMR (Me2CO-d6, 600 MHz) and 13C NMR (Me2-
CO-d6, 150 MHz) spectroscopic data, see Tables 5 and 6; ESIMS m/z
443 [M + Na]+ and 455 [M + Cl]ꢀ; HRESIMS m/z 443.1463 [M + Na]+
(calcd for C19H29ClO8Na, 443.1449).
203 (3.82) nm; IR (KBr)
3447, 2959, 1736, 1373, 1242, 1086,
max
964 cmꢀ1; for 1H NMR (CDCl3, 600 MHz) and 13C NMR (CDCl3,
150 MHz) spectroscopic data, see Tables 4 and 6; ESIMS m/z 429
[M + Na]+, 405 [M ꢀ H]- and 441 [M + Cl]ꢀ; HRESIMS m/z
405.1332 [M ꢀ H]- (calcd for C18H26ClO8, 405.1316).
4.14. Chlorovaltrate J (10)
4.7. Chlorovaltrate C (3)
Colorless oil, ½a D20
ꢂ
ꢀ 46:0 (c 0.27, MeOH); UV (MeOH) max (log
e)
Colorless oil, ½a D20
ꢂ
þ 69:5 (c 0.15, MeOH); UV (MeOH) max (log
e)
203 (3.92) nm; IR (KBr)
3489, 2960, 1735, 1370, 1294,
max
1094 cmꢀ1; for 1H NMR (Me2CO-d6, 600 MHz) and 13C NMR (Me2-
CO-d6, 150 MHz) spectroscopic data, see Tables 5 and 6; ESIMS m/z
441 [M + Na]+, 417 [M ꢀ H]ꢀ and 453 [M + Cl]ꢀ; HRESIMS m/z
441.1670 [M + Na]+ (calcd for C20H31ClO7Na, 441.1656).
204 (3.85) nm; IR (KBr) max 2966, 1736, 1368, 1222, 935 cmꢀ1; for
1H NMR (CDCl3, 600 MHz) and 13C NMR (CDCl3, 150 MHz) spectro-
scopic data, see Tables 4 and 6; ESIMS m/z 443 [M + Na]+ and 455
[M + Cl]ꢀ; HRESIMS m/z 443.1165 [M + Na]+ (calcd for C19H29ClO8-
Na, 443.1449).
4.15. Chlorovaltrate K (11)
4.8. Chlorovaltrate D (4)
Colorless oil, ½a D20
ꢂ
ꢀ 57:6 (c 0.20, MeOH); UV (MeOH) max (log
e)
Colorless oil, ½a D20
ꢂ
þ 100:4 (c 0.85, MeOH); UV (MeOH)
(log
max
203 (3.90) nm; IR (KBr)
3511, 2962, 1730, 1361, 1314,
max
1110 cmꢀ1;Colorless oil; for 1H NMR (CDCl3, 600 MHz) and 13C
NMR (CDCl3, 150 MHz) spectroscopic data, see Tables 5 and 6;
ESIMS m/z 483 [M + Na]+ and 495 [M + Cl]ꢀ. HRESIMS m/z
483.1771 [M + Na]+ (calcd for C22H33ClO8Na, 483.1762).
e
) 204 (3.86) nm; IR (KBr)
3459, 2977, 1731, 1376, 1245,
max
1072 cmꢀ1; for 1H NMR (Me2CO-d6, 600 MHz) and 13C NMR (Me2-
CO-d6, 150 MHz) spectroscopic data, see Tables 4 and 6; ESIMS m/z
387 [M + Na]+ and 399 [M + Cl]ꢀ; HRESIMS m/z 387.1196 [M + Na]+
(calcd for C16H25ClO7Na, 387.1187).
4.16. Chlorovaltrate L (12)
4.9. Chlorovaltrate E (5)
Colorless oil, ½a D20
ꢂ
þ 143:6 (c 0.45, MeOH); UV (MeOH)
(log
max
3446, 2963, 1735,
max
Colorless oil, ½a D20
ꢂ
ꢀ 28:0 (c 0.16, MeOH); UV (MeOH) max (log
e
)
e
) 200 (4.12), 254 (3.82) nm; IR (KBr)
203 (3.92) nm; IR (KBr)
3467, 2970, 1736, 1367, 1235,
1640, 1614, 1430, 1370, 1195, 1101 cmꢀ1; for 1H NMR (Me2CO-
d6, 600 MHz) and 13C NMR (Me2CO-d6, 150 MHz) spectroscopic
data, see Tables 5 and 6; ESIMS m/z 539 [M + Na]+ and 551
[M + Cl]ꢀ; HRESIMS m/z 539.2030 [M + Na]+ (calcd for C25H37ClO9-
Na, 539.2024).
max
1091 cmꢀ1; for 1H NMR (CDCl3, 600 MHz) and 13C NMR (CDCl3,
150 MHz) spectroscopic data, see Tables 4 and 6; ESIMS m/z 557
[M + Na]+ and 569 [M + Cl]ꢀ; HRESIMS m/z 557.2138 [M + Na]+
(calcd for C25H39ClO10Na, 557.2129).

S. Lin et al. / Phytochemistry 85 (2013) 185–193
193
4.17. Chlorovaltrate M (13)
search for special environments, PLA, Shanghai Engineering Re-
search Center for the Preparation of Bioactive Natural Products
(10DZ2251300) and the Scientific Foundation of Shanghai China
(09DZ1975700, 09DZ1971500, 10DZ1971700).
Colorless oil, ½a D20
ꢂ
þ 227:3 (c 0.59, MeOH); UV (MeOH) max (log
e)
200 (4.10), 255 (3.82) nm; IR (KBr)
3461, 2963, 1736, 1641,
max
1612, 1429, 1370, 1245, 1148, 1022 cmꢀ1; for 1H NMR (CDCl3,
600 MHz) and 13C NMR (CDCl3, 150 MHz) spectroscopic data, see
Tables 5 and 6; ESIMS m/z 539 [M + Na]+ and 551 [M + Cl]ꢀ;
HRESIMS m/z 539.1676 [M + Na]+ (calcd for C24H33ClO10Na,
539.1660).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.phytochem.2012.
08.015.
4.18. Chlorovaltrate N (14)
References
Colorless oil, ½a D20
ꢂ
þ 142:0 (c 0.15, MeOH); UV (MeOH) max (log
e)
Allard, P.M., Martin, M.T., Tran Huu Dau, M.E., Leyssen, P., Guéritte, F., Litaudon, M.,
2012. Trigocherrin A, the first natural chlorinated daphnane diterpene
orthoester from Trigonostemon cherrieri. Org. Lett. 14, 342–345.
Becker, H., Chavadej, S., Thies, P.W., Finner, E., 1984. Die Struktur neuer valepotriate
aus colchicinbehandelten zellkulturen von Valeriana wallichii. Planta Med. 44,
245–248.
201(4.12), 254 (3.83) nm; IR (KBr)
3467, 2962, 1738, 1639,
max
1612, 1428, 1366, 1230, 1101 cmꢀ1; for 1H NMR (Me2CO-d6,
600 MHz) and 13C NMR (Me2CO-d6, 150 MHz) spectroscopic data,
see Tables 5 and 6; ESIMS m/z 495 [M + Na]+ and 507 [M + Cl]ꢀ;
HRESIMS m/z 495.1769 [M + Na]+ (calcd for C23H33ClO8Na,
495.1762).
Belzecki, C., Trojnar, J., Chmielewski, M., 2000. Practical synthesis of (+)-
alloisoleucine. Synth. Commun. 30, 2245–2252.
Bounthanh, C., Bergmann, C., Beck, J.P., Haag-Berrurier, M., Anton, R., 1981.
Valepotriates, a new class of cytotoxic and antitumor agents. Planta Med. 41,
21–28.
4.19. Chlorovaltrate O (15)
Chlipala, G.E., Tri, P.H., Hung, N.V., Krunic, A., Shim, S.H., Soejarto, D.D., Orjala, J.,
2010. Nhatrangins A and B, aplysiatoxin-related metabolites from the marine
cyanobacterium Lyngbya majuscula from Vietnam. J. Nat. Prod. 73, 784–787.
Cragg, G.M., Newman, D.J., Snader, K.M., 1997. Natural products in drug discovery
and development. J. Nat. Prod. 60, 52–60.
Fuzzati, N., Wolfender, J.L., Hostettmann, K., 1996. Isolation of antifungal
valepotriates from Valeriana capense and the search for valepotriates in crude
Valerianaceae extracts. Phytochem. Anal. 7, 76–85.
Gao, H.M., Zehl, M., Kaehlig, H., Schneider, P., Stuppner, H., Banuls, L.M.Y., 2010.
Rapid structural identification of cytotoxic bufadienolide sulfates in toad venom
from Bufo melanosticus by LC-DAD-MSn and LC-SPE-NMR. J. Nat. Prod. 73, 603–
608.
Li, B., Abliz, Z., Tang, M.J., Fu, G.M., Yu, S.S., 2006. Rapid structural characterization of
triterpenoid saponins in crude extract from Symplocos chinensis using liquid
chromatography combined with electrospray ionization tandem mass
spectrometry. J. Chromatogr. A. 1101, 53–62.
Lin, S., Shen, Y.H., Li, H.L., Yang, X.W., Chen, T., Lu, L.H., Huang, Z.S., Liu, R.H., Xu, X.K.,
Zhang, W.D., Wang, H., 2009. Acylated iridoids with cytotoxicity from Valeriana
jatamansi. J. Nat. Prod. 72, 650–655.
Lin, S., Chen, T., Liu, X.H., Shen, Y.H., Li, H.L., Shan, L., Liu, R.H., Xu, X.K., Zhang, W.D.,
Wang, H., 2010a. Iridoids and lignans from Valeriana jatamansi. J. Nat. Prod. 73,
632–638.
Lin, S., Shen, Y.H., Zhang, Z.X., Li, H.L., Shan, L., Liu, R.H., Xu, X.K., Zhang, W.D., 2010b.
Revision of the structures of 1,5-dihydroxy-3,8-epoxyvalechlorine, volvaltrate
B, and valeriotetrate C from Valeriana jatamansi and V. officinalis. J. Nat. Prod. 73,
1723–1726.
Newman, D.J., Cragg, G.M., Snader, K.M., 2003. Natural products as sources of new
drugs over the period 1981–2002. J. Nat. Prod. 66, 1022–1037.
Newman, D.J., Cragg, G.M., 2007. Natural products as sources of new drugs over the
last 25 years. J. Nat. Prod. 70, 461–477.
Colorless oil, ½a D20
ꢂ
þ 226:0 (c 0.18, MeOH); UV (MeOH) max (log
e)
201 (4.13), 254 (3.84) nm; IR (KBr)
3455, 2961, 1737, 1642,
1610, 1432, 1369, 1291, 1110 cmꢀm1;axfor 1H NMR (Me2CO-d6,
600 MHz) and 13C NMR (Me2CO-d6, 150 MHz) spectroscopic data,
see Tables 5 and 6; ESIMS m/z 439 [M + Na]+ and 451 [M + Cl]ꢀ;
HRESIMS m/z 439.1514 [M + Na]+ (calcd for C20H29ClO7Na,
439.1500).
4.20. Basic Hydrolysis of Chlorovaltrate N (14)
To a solution of 14 (9.0 mg) in THF (2.0 mL) and H2O (6 mL) was
added NaOH (120 mg), and then the mixture was stirred at room
temperature for 1 h. After neutralization with 4 N HCl, the solution
was evaporated under reduced pressure to give a residue. The later
was resolved by Sephadex LH20 chromatography eluting with
petroleum etherCHCl3–MeOH (5:5:1) and then preparative TLC
using CHCl3–MeOH (25:1) as developing solvent to yield (R)-3-
methylvaleric acid (1.5 mg): colorless oil, ½a D20
ꢀ 6:3 (c 0.15, CH2-
ꢂ
Cl2); 1H NMR (CDCl3, 600 MHz) d 2.34 (1H, m, H-2a), 2.15 (1H, m,
H-2b), 1.89 (1H, m, H-3), 1.43 (1H, m, H-4a), 1.17 (1H, m, H-4b),
0.90 (3H, t, J = 7.2 Hz, H3-5), 0.96 (3H, d, J = 6.6 Hz, H3-6); 13C
NMR (CDCl3, 150 MHz) d 180.2 (C, C-1), 41.2 (CH2, C-2), 31.7 (CH,
C-3), 29.2 (CH, C-4), 11.2 (CH3, C-5), 19.2 (CH3, C-6); ESIMS m/z
139 [M + Na]+.
Oh, D.C., Williams, P.G., Kauffman, C.A., Jensen, P.R., Fenical, W., 2006.
Cyanosporasides A and B, chloro- and cyano-cyclopenta[a]indene glycosides
from the marine Actinomycete ‘‘Salinispora pacifica’’. Org. Lett. 8, 1021–1024.
Stahi, E., Schütz, E., 1980. Extraktion labiler naturstoffe mit überkritischen gasen 4.
Mitt.: extraktion der valepotriate aus Valeriana wallichii. Planta Med. 40, 262–
270.
4.21. Cells, culture conditions, and cell proliferation assay
Stobiecki, M., Staszków, A., Piasecka, A., Garcia-Lopez, P.M., Zamora-Natera, F.,
Kachilicki, P., 2010. LC-MSMS profiling of flavonoid conjugates in wild Mexican
lupine, Lupinus reflexus. J. Nat. Prod. 73, 1254–1260.
Lin et al. (2009). Pseudolaric acid B was used as a positive
control.
Tang, Y.P., Liu, X., Yu, B., 2002. Iridoids from the rhizomes and roots of Valeriana
jatamansi. J. Nat. Prod. 65, 1949–1952.
Acknowledgments
Thies, P.W., David, S., 1983. 2,9-Dioxa tricyclo [4,3,1,03,7] decane compounds and
process of making same. U. S. Patent 4,391,819.
We acknowledge the Pharmaceutical Analysis Center of Second
Military Medical University for measuring the IR and NMR spectra.
We also thank the Shanghai Institute of Materia Medica, Chinese
Academy of Sciences for the optical rotation determination. The
work was supported by program NCET Foundation, NSFC
(30725045), partially supported by Global Research Network for
Medicinal Plants (GRNMP) and King Saud University, Shanghai
Leading Academic Discipline Project (B906), FP7- PEOPLE-IRSES-
2008 (TCMCANCER Project 230232), Key laboratory of drug re-
Yang, X.P., Li, E.W., Zhang, Q., Yuan, C.S., Jia, Z.J., 2006. Five new iridoids from
Patrinia rupestris. Chem. Biodiversity 3, 762–770.
Yu, L.L., Huang, R., Han, C.R., Lv, Y.P., Zhao, Y., 2005. New iridoid triesters from
Valeriana jatamansi. Helv. Chem. Acta. 88, 1059–1062.
Zhang, W.D., Lin, S., Shan, L., Su, J., Liu, R.H., Li, H.-L., Shen, Y.H., Xu, X.K., 2009. Uses
of iridoid compounds extracted from Valeriana officinalis as antitumor agents.
Faming Zhuanli Shenqing, CN 101444500.
Zhang, W.D., Su, J., Lin, S., Shan, L., Liu, R.H., Li, H.L., Xu, X.K., 2010. Application of
iridoid compounds in preparation of medicine for treating ovarian cancer.
Faming Zhuanli Shenqing, CN 101829080.