A new triterpenoid saponin from Pulsatilla cernua predicted by NMR-based mosaic method

Article abstract of DOI:10.1080/14786419.2018.1538995

A saponin (1) with gypsogenin as aglycone was isolated from the roots of Pulsatilla cernua. The aglycone of compound 1 was considered as gypsogenin which was rarely found in this genus. Its structure was predicted by NMR-based “mosaic” method rapidly, and further confirmed on the basis of spectroscopic data, including 2D NMR spectra and chemical evidence. This work suggested that NMR-based mosaic method is suitable for most of saponins from common species of genus Pulsatilla.

Full text of DOI:10.1080/14786419.2018.1538995

Natural Product Research
Formerly Natural Product Letters
ISSN: 1478-6419 (Print) 1478-6427 (Online) Journal homepage: http://www.tandfonline.com/loi/gnpl20
A new triterpenoid saponin from Pulsatilla cernua
predicted by NMR-based mosaic method
An-Dong Wang, Ying Bao, Xia Wang, Mei-Chen Li, Xu-Hong Ren, Jian-Yu Liu &
Yong-Nan Xu
To cite this article: An-Dong Wang, Ying Bao, Xia Wang, Mei-Chen Li, Xu-Hong Ren, Jian-Yu Liu
& Yong-Nan Xu (2018): A new triterpenoid saponin from Pulsatillaꢀcernua predicted by NMR-based
mosaic method, Natural Product Research, DOI: 10.1080/14786419.2018.1538995
To link to this article: https://doi.org/10.1080/14786419.2018.1538995
View supplementary material
Published online: 03 Dec 2018.
Submit your article to this journal
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=gnpl20

NATURAL PRODUCT RESEARCH
https://doi.org/10.1080/14786419.2018.1538995
RESEARCH ARTICLE
A new triterpenoid saponin from Pulsatilla cernua
predicted by NMR-based mosaic method
b,c,
An-Dong Wanga,b, , Ying Bao
, Xia Wanga,b, Mei-Chen Lia,b, Xu-Hong Renb,c
,
ꢀ
ꢀ
Jian-Yu Liub,c , and Yong-Nan Xub,c
b
aSchool of Traditional Chinese Medicine, Shenyang Pharmaceutical University, Shenyang, China; Key
Laboratory of Structure-Based Drug Design & Discovery, Ministry of Education, Shenyang
c
Pharmaceutical University, Shenyang, China; School of Pharmaceutical Engineering, Shenyang
Pharmaceutical University, Shenyang, China
ABSTRACT
ARTICLE HISTORY
Received 25 July 2018
Accepted 18 October 2018
A saponin (1) with gypsogenin as aglycone was isolated from the
roots of Pulsatilla cernua. The aglycone of compound 1 was con-
sidered as gypsogenin which was rarely found in this genus. Its
structure was predicted by NMR-based “mosaic” method rapidly,
and further confirmed on the basis of spectroscopic data, includ-
ing 2D NMR spectra and chemical evidence. This work suggested
that NMR-based mosaic method is suitable for most of saponins
from common species of genus Pulsatilla.
1. Introduction
As a traditional herbal medicine, the roots of Pulsatilla cernua (Ranunculaceae) are
used for treatment of amoebic dysentery, malaria and chills for a long time. Modern
researches reported that the chemical components of P. cernua showed strong
biological activities, and especially the saponin-enriched fraction presented remarkable
effects in the treatment of Alzheimer’s disease (Han et al. 2007; Seo et al. 2009;
CONTACT Jian-Yu Liu
These two authors contributed to this work equally.
burningice@126.com; Yong-Nan Xu
ynxu@syphu.edu.cn
ꢀ
Supplemental data for this article can be accessed at https://doi.org/10.1080/14786419.2018.1538995.
ß 2018 Informa UK Limited, trading as Taylor & Francis Group

2
A.-D. WANG ET AL.
Figure 1. Mosaics of saccharide residues of most typical saponins from P. cernua as well as their
1
anomeric protons H NMR chemical shifts reigns.
Liu et al. 2012). Phytochemical investigations of this plant led to a number of
triterpenoid saponins (Bang et al. 2005a,b; Fu et al. 2008; Kang 1989; Li et al. 1990; Xu
et al. 2007a,b; Xu et al. 2010; Yang et al. 2010; Zhang et al. 2000a,b; Liu et al. 2012;
Fan et al. 2013; Liu et al. 2015; Wang et al. 2018) and several of them are often
1
comprised of numerous saccharides (Liu et al. 2012; Shu et al. 2013) of which H-NMR
signals are overlapped seriously. As a result, the elucidation of the structure of
a saponin by NMR spectroscopy commonly involves 2D-NMR experiments (HMBC,
HSQC, COSY, NOESY, and TOCSY, etc.) and comparison with previously reported
data, as well as chemical experiment (acid and/or alkaline hydrolysis). Therefore the
elucidation of structure of a saponin is not only a challenging but also a high-cost and
time-consuming work. The mass fragmentation pattern is usually considered as an
efficient method for identification of saponins. However, it is an experience required
work for identification of the fragments of sugar residues. As a powerful technique for
structural elucidation, NMR contains a wealth of structural information. In our present
study, an NMR-based mosaic method was proposed based on our previous work and
literatures. This approach is exemplified in the structural prediction of a new saponin
with gypsogenin as aglycone, as well as three reported ones.
2. Results and discussion
Based on our study and previous literatures, the types and sequence of saccharide
residues of most typical saponins from P. cernua (and even some other species
of genus Pulsatilla) are distinctive and this might be closely related to their biosyn-
thetic pathways (Table S1). The frequent saccharide residues are a-L-arabinopyranosyl

NATURAL PRODUCT RESEARCH
3
Figure 2. Stitching the mosaics of compound 1 to complete and predict the structure Glc1 signal
was overlapped by water when NMR was measured in pyridine-d5 (lower spectrum), while it can
be recognized clearly when a drop of deuterated water was added (upper spectrum).
(ara), b-D-glucopyranosyl (glc), and a-L-rhamnopyranosyl (rha) groups, and their glyco-
sylation sites are position C-2, 4 (ara), C-4, 6 (glc), and C-3 (rha), respectively. Further
study indicated that the NMR profiles of anomeric hydrogens of these saccharide
residues are also characteristic (Table S2). It can be seen that chemical shift of certain
saccharide residue keeps in a certain region. Chemical shifts of ara H-1 linked to C-3
of aglycone are d 4.61–5.20, and those of other saccharide residues are d 4.70–5.20
(glc1), 4.95–5.50 (glc2), 4.85–5.15 (glc3), 6.12–6.29 (rha1), 5.37–5.39 (rha2), 6.14–6.30
(glc4), 4.90–5.50 (glc5), 5.76–5.87 (rha3), respectively (Figure 1 and Figure S1).
Coupling constant of glc is 7–8 Hz, ara is 6.0 Hz and rha is broad singlet based on their
dihedral angle between bonds C1-H1 and C2-H2. Furthermore, remaining signals will
not change significantly when one or more saccharide residues are removed (Table
S2) because of their discrete spin systems. As a result, a rapid and efficient strategy for
1
elucidate a new or known saponin from this plant based on H NMR is feasible. Of
this method, each saccharide residue or aglycone is considered as one single mosaic,
1
and each one can be easily recognized according to its H NMR signal. In order to pre-
dict the structure, these mosaics could be stitched together. For structural prediction
of known saponins, kalopanaxsaponin Lb (16), pulsatillacernuoside Kb (38) and pulsa-
tillacernuoside Ka (44) (Supplementary material) were used as examples. Spectrum of
anomeric proton of 16 was in good agreement with that of 15 which shared same
sugar chain but different aglycone. Compound 37 and compound 38, which have
same anomeric protons, showed that they have same sugar chain but different

4
A.-D. WANG ET AL.
aglycone. Same to anomeric proton of 43, compound 44 have same sugar chain
(Figure S2). These mosaics were then stitched together as Figure S3 to complete the
structures and they are in great agreement with our previous report (Wang
et al. 2018).
Compound 1 was obtained as a white amorphous powder using D101 macroporous
resin, silica gel and recycling preparative HPLC from 70% aqueous ethanol extract
1
(Supplementary material). After analysis by submitting the H NMR data of aglycone
to an on-line structure search engine of the NMR database (Shanghai Micronmr Infor
Technology Co., Ltd, Shanghai, China), the aglycone of 1 was considered as gypsoge-
nin which was rarely found in this genus (Liu et al. 2012). From signals of anomeric
protons, six mosaics of saccharide residues were recognized as ara, rha1, glc1, glc4,
glc5 as well as rha3. These mosaics were then stitched together as Figure 2 to com-
plete the structure. Compared with the signals of compound 56 which shared the
same aglycone shown in Supplementary material, only glc2 was absent in compound
1 (Figure S4). In order to verify this method and confirm the structures of 1, structure
elucidation using various spectra and chemical reactions were further investigated.
The molecular formula C65H104O31 was established on the basis of HR-ESI-MS at m/z
1403.6454 [M þ Na]þ (calcd for C65H104O31Na, 1403.6459). The 1H NMR (pyridine-d5
with one drop of D2O, 600 MHz) showed the present of six anomeric proton signals at
dH 4.60 (1H, d, J ¼ 6.0 Hz), 4.95 (1H, d, J ¼ 7.8 Hz), 5.10 (1H, d, J ¼ 7.8 Hz), 5.80 (1H, brs),
6.06 (1H, brs) and 6.18 (1H, d, J ¼ 8.4 Hz) and their associated 13C resonances assigned
from HSQC experiment were at dC 104.8, 104.6, 106.2, 102.5, 101.3, and 95.4, respect-
ively, which indicated that it is a glycoside with six sugar moieties. Acid hydrolysis of
1 with 1 M HCl gave arabinose, rhamnose and glucose, which were identified by TLC
comparison with authentic samples. The b-anomeric configurations for the D-glucose,
3
and the a-anomeric configurations for L-arabinose were determined by their JH1,H2
coupling constants of 7–8 Hz. The a-anomeric configuration of L-rhamnose was judged
by the chemical shift of C-5 of rhamnose (dC 69–70) (Mahato and Kundu 1994). The
absolute configuration of the saccharides was determined by GC analysis of chiral
derivatives in the hydrolysate.
Six tertiary methyl proton singlets at dH 0.79, 0.81, 0.81, 0.95, 1.13, and 1.39 as well
as two doublets at dH 1.65 (3H, d, J ¼ 6.0 Hz) and 1.66 (3H, d, J ¼ 6.0 Hz) indicated the
aglycone should be a kind of triterpene and two of the sugar moieties might be rham-
nose. Signals at dH 5.30 (1H, s) and 9.67 (1H, s) and their associated 13C resonances at
dC 122.6 and 205.5 indicated that there should be one olefince and aldehyde group in
this structure. The aglycone was then considered as gypsogenin by compared with
data from our previous work (Liu et al. 2012).
The 13C NMR data of C-3 (dC 83.8) and C-28 (dC 176.6) revealed that 1 was a bidesmo-
sides saponin. The linkage of the sugar moiety at C-3 of the aglycone was established
from the HMBC correlations between d 4.60 (1H, d, J ¼ 6.0 Hz, ara H-1) and d 83.8 (C-3),
d 5.10 (1H, d, J ¼ 7.8 Hz, glc1 H-1) and d 79.3 (ara C-4), d 6.06 (1H, brs, rha1 H-1) and d
75.5 (ara C-2). The linkage of the sugar moiety at C-28 of the aglycone was established
from the HMBC correlations between d 6.18 (1H, d, J ¼ 8.4 Hz, glc4 H-1) and 176.6 (C-28),
d 4.95 (1H, d, J ¼ 7.8 Hz, glc5 H-1) and 69.0 (glc4 C-6), 5.80 (1H, brs, rha3 H-1) and 78.0
(glc5 C-4) (Figure S8). Based on the above evidence, the structure of 1 was determined

NATURAL PRODUCT RESEARCH
5
to be gypsogenin 3-O-a-L-rhamnopyranosyl(1!2)[b-D-glucopyranosyl(1!4)]-a-L- arabi-
nopyranosyl-28-O-a-L-rhamnopyranosyl(1!4)-b-D-glucopyranosyl(1!6)-b-D-glucopyrano-
syl ester, and we propose the trivial name pulsatillacernuoside G. The spectra of 1 see
Figures S5-S11 and data in Tables S3 and S4.
In vitro assay indicated compound 1 showed no anti-proliferative activities against
five human cancer cell lines (A549, MDA-MB-231, SK-Hep-1, SNU638, and HCT116), and
all IC50 values of 1 were above 100 lM compared with those of positive control
Etoposide (4.1, 50.1, 0.6, 4.1 and 7.8 lM, respectively), see Table S5 and Figure S12.
3. Conclusion
An NMR-based mosaic method was proposed to predict typical known or even unde-
1
scribed saponins from P. cernua. In this method, complex H NMR profiles (such as
that of 1) can be taken apart to form several pieces of smaller mosaics which can be
stitched together subsequently according to the characteristic of typical saponins from
this plant. It is noteworthy that this method is also suitable for some other species of
this genus such as P. chinensis, P. campanella and P. dahurica, while it should be used
carefully for other species because saponins from these plants might contain different
kind of sugar chains.
Disclosure statement
No potential conflict of interest was reported by the authors.
Funding
This work was supported by the general project of department of education of Liaoning
Province (grant no. 201610163L13) and the Natural Science Foundation of Liaoning Province
(grant no. 201602712).
ORCID
Jian-Yu Liu
http://orcid.org/0000-0003-0714-1319
References
Bang SC, Lee JH, Song GY, Kim DH, Yoon MY, Ahn BZ. 2005a. Antitumor activity of Pulsatilla
koreana saponins and their structure-activity relationship. Chem Pharm Bull. 53(11):
1451–1454.
Bang SC, Kim Y, Lee JH, Ahn BZ. 2005b. Triterpenoid saponins from the roots of Pulsatilla kore-
ana. J Nat Prod. 68(2):268–272.
Fan WH, Liu JY, Gong YX, Ma J, Zhou N, Xu YN. 2013. A new triterpenoid saponin from Pulsatilla
cernua. Nat Prod Sci. 19(2):150–154.
Fu YM, Chen H, Liu DL, Zhang ZM. 2008. Chemical constituents from Pulsatilla cernua. Chin Trad
Herb Drugs. 39(1):26–29.
Han CK, Choi WR, Oh KB. 2007. Cognition-enhancing and neuroprotective effects of hederacol-
chiside-E from Pulsatilla koreana. Planta Med. 73(7):665–669.
Kang SS. 1989. Saponins from the roots of Pulsatilla koreana. Arch Pharm Res. 12(1):42–47.

6
A.-D. WANG ET AL.
Li XC, Wang DZ, Wu SG, Yang CR. 1990. Triterpenoid saponins from Pulsatilla campanella.
Phytochemistry. 29(2):595–599.
Liu JY, Guan YL, Zou LB, Gong YX, Hua HM, Xu YN, Zhang H, Yu ZG, Fan WH. 2012. Saponins
with neuroprotective effects from the roots of Pulsatilla cernua. Molecules. 17(5):5520–5531.
Liu JY, Ma J, Gong YX, Zhou N, Su XY, Fan WH, Xu YN. 2015. Studies on saponins of
Pulsatilla cernua. J. Liaoning Univ TCM. 17(2):56–60.
Mahato SB, Kundu AP. 1994. 13C NMR spectra of pentacyclic triterpenoids – a compilation and
some salient features. Phytochemistry. 37(6):1517–1575.
Seo JS, Kim TK, Lee YH, Lee KW, Park SK, Baek IS, Kim KS, Im GJ, Lee SM, Park YH, Han PL. 2009.
SK-PC-B70M confers anti-oxidant activity and reduces Abeta levels in the brain of Tg2576
mice. Brain Res. 1261:100–108.
Shu Z, Chen Z, Liu Y-L, Zhu W-F, Feng Y-L, Xu Q-M, Li X-R, Yang S-L. 2013. A new oleanane-type
triterpenoidal saponin from Pulsatilla chinensis. Nat Prod Res. 27(23):2196–2201.
Wang AD, Bao Y, Liu D, Wang X, Li MC, Liu JY, Xu YN. 2018. Isolation and structure
determination of new saponins from Pulsatilla cernua based on an NMR-guided method
and their anti-proliferative activities. Phytochem Lett. 27:9–14.
Xu TH, Xu YJ, Han D, Zhao HF, Xie SX, Xu DM. 2007a. Triterpenoid saponins from Pulsatilla
cernua (Thunb.) Bercht. et Opiz. J Integrative Plant Biology. 49(2):202–206.
Xu TH, Xu YJ, Li HX, Han D, Zhao HF, Xie SX, Li Y, Niu JZ, Si YS, Xu DM. 2007b. Two new
triterpenoid saponins from Pulsatilla cernua (Thunb.) Bercht. et Opiz. J Asian Nat Prod Res.
9(6–8):705–711.
Xu YJ, Bai L, Liu YH, Liu Y, Xu TH, Xie SX, Si YS, Zhou HO, Liu TH, Xu DM. 2010. A new
triterpenoid saponin from Pulstilla cernua. Molecules. 15(3):1891–1897.
Yang HJ, Cho YW, Kim SH, Kim YC, Sung SH. 2010. Triterpenoidal saponins of Pulsatilla koreana
roots. Phytochemistry. 71(16):1892–1899.
Zhang Q, Ye W, Yan X, Zhu G, Che CT, Zhao S. 2000a. Cernuosides A and B, two
sucrase inhibitors from Pulsatilla cernua. J Nat Prod. 63(2):276–278.
Zhang QW, Ye WC, Che ZT, Zhao SX. 2000b. Triterpene saponins from Pulsatilla cernua.
Acta Pharm Sin. 35(10):756–759.