Hepatoprotective triterpene saponins from the roots of Glycyrrhiza inflata

Article abstract of DOI:10.3390/molecules20046273

Two novel oleanane-type triterpene saponins, licorice-saponin P2 (1) and licorice-saponin Q2 (3), together with nine known compounds 2, 4-11, have been isolated from the water extract of the roots of Glycyrrhiza inflata. The structures of these compounds were elucidated on the basis of spectroscopic analysis, including 2D-NMR experiments (¹H-¹H COSY, HSQC, HMBC and ROESY). In in vitro assays, compounds 2-4, 6 and 11 showed significant hepatoprotective activities by lowering the ALT and AST levels in primary rat hepatocytes injured by D-galactosamine (D-GalN). In addition, compounds 2-4, 6, 7 and 11 were found to inhibit the activity of PLA₂ with IC₅₀ values of 6.9 μM, 3.6 μM, 16.9 μM, 27.1 μM, 32.2 μM and 9.3 μM, respectively, which might be involved in the regulation of the hepatoprotective activities observed.

Full text of DOI:10.3390/molecules20046273

Molecules 2015, 20, 6273-6283; doi:10.3390/molecules20046273
OPEN ACCESS
molecules
ISSN 1420-3049
www.mdpi.com/journal/molecules
Article
Hepatoprotective Triterpene Saponins from the Roots of
Glycyrrhiza inflata
Yun-Feng Zheng 1,2, Juan-Hua Wei 1, Shi-Qi Fang 1, Yu-Ping Tang 1,2, Hai-Bo Cheng 3,
Tian-Lin Wang 1, Cun-Yu Li 1,2 and Guo-Ping Peng 1,2,
*
1
School of Pharmacy, Nanjing University of Chinese Medicine, Nanjing 210023, China;
E-Mails: zyunfeng88@126.com (Y.-F.Z.); weijuanhua1989@163.com (J.-H.W.);
fangsq8866@163.com (S.-Q.F.); yupingtang9@126.com (Y.-P.T.); cpuwtl@gmail.com (T.-L.W.);
licunyuok@163.com (C.-Y.L.)
2
3
Jiangsu Collaborative Innovation Center of Chinese Medicinal Resources Industrialization,
Nanjing 210023, China
Translational Medicine Research Center of Nanjing University of Chinese Medicine,
Nanjing 210023, China; E-Mail: nzychb@163.com
* Author to whom correspondence should be addressed; E-Mail: guopingpeng@sohu.com;
Tel./Fax: +86-25-86798186.
Academic Editor: Derek J. McPhee
Received: 4 March 2015 / Accepted: 2 April 2015 / Published: 9 April 2015
Abstract: Two novel oleanane-type triterpene saponins, licorice-saponin P2 (1) and
licorice-saponin Q2 (3), together with nine known compounds 2, 4–11, have been isolated
from the water extract of the roots of Glycyrrhiza inflata. The structures of these compounds
were elucidated on the basis of spectroscopic analysis, including 2D-NMR experiments
(1H–1H COSY, HSQC, HMBC and ROESY). In in vitro assays, compounds 2–4, 6 and 11
showed significant hepatoprotective activities by lowering the ALT and AST levels in
primary rat hepatocytes injured by D-galactosamine (D-GalN). In addition, compounds 2–4,
6, 7 and 11 were found to inhibit the activity of PLA2 with IC50 values of 6.9 μM, 3.6 μM,
16.9 μM, 27.1 μM, 32.2 μM and 9.3 μM, respectively, which might be involved in the
regulation of the hepatoprotective activities observed.
Keywords: Glycyrrhiza inflate; genus Glycyrrhiza; triterpene saponin; hepatoprotective;
phospholipase A2

Molecules 2015, 20
6274
1. Introduction
The genus Glycyrrhiza consists of about 30 species with a nearly global distribution, of which 18
species are found in China. Among them, three species named Glycyrrhiza uralensis, Glycyrrhiza glabra
and Glycyrrhiza inflata, have been used as traditional Chinese medicine for the treatment of hepatitis,
spasmodic cough, gastric ulcer, and so on. Phytochemical studies have showed that triterpenoid saponins
and flavonoids were the two of major kinds of active substances of Glycyrrhiza, which have a variety of
pharmacological activities, including hepatoprotective [1,2], antiviral [3], anti-inflammatory [4] and
antioxidative [5] effects. Recently, we reported the chemical constituents of G. uralensis and G. glabra,
as well as their cytotoxic or neuraminidase bioactivities [6,7]. As part of our ongoing research on the
genus Glycyrrhiza, an extensive phytochemical investigation on the roots of G. inflata has now led to
the isolation of two new oleanane-type saponins 1, 3 and nine known saponins 2, 4–11. All compounds
were screened for their protective activities against D-galactosamine (D-GlaN) induced toxicity in vitro.
In addition, the inhibitory activities on phospholipase A2 (PLA2) were presented. Herein, we
report the isolation and structural elucidation of these saponins, along with the investigation of their
protective activities.
2. Results and Discussion
The total saponin fraction of G. inflata was prepared by co-application of polyamide and macroporous
resin column chromatography [7]. The resulting extract was subjected to ODS column chromatography
and preparative HPLC to afford two new oleanane-type saponins 1, 3 together with nine known ones 2,
4–11. Their structures were shown in Figure 1.
30
O
29
R5 R6
R2
R4
19
R2
20
17
21
12
11
18
O
O
R3
22
28
O
13
14
25
26
1
4
16
9
2
3
10
8
15
COOH
O O
OH
27
COOH
O O
OH
7
5
6
R1
23
R1
24
HO
COOH
HO
COOH
O
O
R1
R2
R1
R2
βH
βH
R3
H
R4
R5
R6
O
O
OH
1:
CH3
CH3
H
CH2OH COOH
5:
CH3
CH3
COOH
CH3
OH
2:
3:
4:
6:
9:
H
αOH COOH
CH3
7:
8:
CH3
HO
HO
O
CH2OH αH
CH2OH βH
H
H
H
H
H
H
H
CH3
CH3
CH3
CH3
CH3
CH3
COOH
COOH
COOH
CHO
CH2OH
O
H
CH3
CH3
βH OCOCH3
βH OCOCH3
10: CH2OH βH
11: CH3 βH
H
H
COOGlc
COOH
Figure 1. Chemical structures of compounds 1–11.
2.1. Structural Determination
Compound 1 was obtained as a white amorphous powder and showed a protonated peaks in the
low-resolution positive HR-ESI-MS spectrum at m/z 861.3929 [M + Na]+ and m/z 839.4120 [M + H]+.

Molecules 2015, 20
6275
Its molecular formula was thus determined as C42H62O17, requiring 12 degrees of unsaturation. The UV
spectrum showed an absorption maximum at 250.8 nm (MeOH, log ε 4.14), indicating the presence of
1
13
an α,β-unsaturated carbonyl moiety. In the H-NMR and C-NMR spectra, representative signals of
eight tertiary carbons at δC 75.3–85.5, two carboxyl carbons at δC 174.8 and 174.8, and two anomeric
carbons δC 105.9, δH 5.03 (1H, d, J = 8.0 Hz) and 107.3, 5.38 (1H, d, J = 7.5 Hz) suggested the presence
of two monosaccharide residues. The coupling constants of the anomeric protons indicated glycosidic
bonds with β-configurations [8]. Acid hydrolysis and GC comparison with authentic samples indicated
the presence of D-glucuronic acid (GlcA) [9]. This deduction was further supported by the signals of two
fragment ion peaks at m/z 663.3762 [M + H − C6H8O6]+ and m/z 487.3448 [M + H − 2C6H8O6]+ in the
HR-ESI-MS spectrum. The glycosidic site was established unambiguously by a HMBC experiment in
which a long-range correlation between H-1′ (δH 5.03) and C-3 (δC 91.3), H-1′′ (δH 5.38) and C-2′
(δC 85.5). Thus, the carbohydrate sequence of 1 was established as 3-O-β-D-glucuronopyranosyl-(1→2)-
1
β-D-glucuronopyranosyl. The H-NMR spectrum of the aglycone moiety of 1 showed signals
corresponding to six tertiary methyls [δH 1.27, 1.08, 1.07, 0.85, 1.36 and 0.70, (each 3H, s)], one
oxygenated methylene [δH 4.06, 3.98, (2H, d, 10.5)], one oxygenated methine [δH 3.27 (1H, dd, 4.0 and
11.5)], and one unsaturated methine [δH 5.83 (1H, s)], while the 13C NMR and DEPT spectrum displayed
30 carbon resonances, containing six methyls, ten methylenes (including one oxygenated methylene),
five methines (including one oxygenated methine and one unsaturated methine), and nine quaternary
carbons (including one carbonyl quaternary carbon, one unsaturated quaternary carbon and one carboxyl
carbon). Therefore, compound 1 was considered to be an oleanane-type triterpene glucuronide bearing
a 12(13)-double bond and a keto group at C-11. In the HMBC spectrum, correlations of δH 5.03 (H-1′)
to δC 91.3 (C-3) and δH 5.38 (H-1′′) to δC 85.5 (C-2′) could be observed. In addition, the correlations in
the HMBC spectrum from H-1′ at δH 5.03, H-23 at δH 1.27 and H-24 at δH 1.08 to C-3 at δC 91.3 helped
in assigning one oxygenated methine at C-3.
Detailed analysis of the above 1D-NMR data and 2D-NMR correlations indicated that 1 is an
oleanane-type saponin derivative and is structurally related to the known compound licorice-saponin G2
(4). The comparison of the NMR data of 1 with those of 4 suggested that the hydroxyl group at C-24 in
4 was transposed to C-29 in 1. The HMBC correlations from δH 3.98, 4.06 (H-29) to δC 39.1 (C-19) and
δC 180.2 (C-30) and the 1H–1H COSY correlations between the proton signal at δH 2.49 (H-18) and δH
1.98, 2.24 (H-19) confirmed that hydroxyl group was connected to C-29 in compound 1 (Figure 2).
The relative configuration of 1 was established on the basis of NOESY correlations of H-3 with
H-1'α, H-23 and H-5, H-9 with H-5 and H-27, and H-18 with H-28β and H-19β, as well as H-29 with
H-19α, H-21α which revealed that the substituent groups of C-3, H-18 were β-oriented and C-29 was
α-oriented. From these spectroscopic data, compound 1 was deduced to be 3β-O-[β-D-glucuronopyranosyl-
(1→2)-β-D-glucuronopyranosyl]-29-hydroxyglycyrrhetic acid, and named licorice-saponin P2. The
structure of compound 1 was confirmed to be as shown in Figure 1 and the 1H- and 13C-NMR data are
listed in Table 1.

Molecules 2015, 20
6276
HOCH2
COOH
H
H
COOH
H
H
H
H
H
H
O
H
O
H
COOH
H
O
O
H
OH
1
3
HO
COOH
O
O
OH
COSY:
HMBC: H
C
NOESY: H
H
HO
O
Figure 2. Key COSY, NOESY and HMBC correlations of compounds 1 and 3.
Table 1. NMR Chemical Shifts of Compounds 1–3 (C5D5N, 500 MHz).
1
2
3
Position
δC mult
41.4 CH2
28.6 CH2
91.3 CH
41.8 qC
δH (J in Hz)
0.89 *,2.81 (m)
1.84 (m),2.11 (m)
3.27 (dd,4.0,11.5)
δC mult
δH (J in Hz)
1.05 *,3.00 (m)
δC mult
δH (J in Hz)
1
40.5 CH2
40.3 CH2
1.03 *,2.97(m)
2
27.7 CH2 1.98 (m),2.26 (m) 27.3 CH2 2.12(m),2.23(m)
90.3 CH 3.37(dd,4.0,11.5 ) 90.4 CH 3.53(dd,4.5,11.5)
3
4
40.9 qC
56.4 CH
18.6 CH2
34.1 CH2
46.5 qC
45.2 qC
56.7 CH
19.1 CH2
33.9 CH2
46.4 qC
5
57.3 CH
19.5 CH2
34.8 CH2
47.6 qC
0.62 (m)
0.71 (m)
0.87 *
6
1.18 *,1.41 (m)
1.12 *,1.47 (m)
1.25 *,1.49 (m)
1.22 *,1.49 (m)
1.55 *,1.72 *
1.28 *,1.54 *
7
8
9
64.0 CH
34.6 qC
2.31 (s)
5.83 (s)
63.1 CH
38.3 qC
2.44 (s)
5.89 (s)
62.7 CH
37.8 qC
2.43(s)
5.82(s)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
202.4 qC
130.3 CH
173.1 qC
45.5 qC
201.1 qC
129.7 CH
171.7 qC
45.0 qC
200.5 qC
129.4 CH
170.6 qC
44.4 qC
28.6 CH2
28.6 CH2
34.6 qC
1.04 *,1.64 (m)
0.90 *,2.11 (m)
27.9 CH2
31.1 CH2
34.1 qC
1.05 *,1.68 (m)
1.28 *,3.09 (m)
27.4 CH2
27.3 CH2
33.3 qC
1.08 *,1.70 (m)
0.95 *,2.10 (m)
50.3 CH
39.1 CH2
53.0 qC
2.49 (m)
48.2 CH
36.1 CH2
49.1 qC
2.37 (m)
47.5 CH
40.6 CH2
43.4 qC
2.26 (m)
1.98 *,2.24 (m)
1.68 *,3.13 (m)
1.59 *,2.47(m)
28.6 CH2
39.8 CH2
30.1 CH3
18.9 CH3
18.8 CH3
20.8 CH3
25.6 CH3
1.99 *,2.23 (m)
1.42 *,1.61 (m)
1.27 (s)
73.1 CH
44.0 CH2
29.1 CH3
18.0 CH3
17.8 CH3
19.9 CH3
24.1 CH3
4.52 (m)
1.78 *,1.98 (m)
1.35 (s)
30.5 CH2
36.6 CH2
23.7 CH3
1.68 *,2.18(m)
1.38 *,1.53 *
1.49 (s)
1.08 (s)
1.2 (s)
64.2 CH2 3.72 (d, 12),4.40*
1.07 (s)
1.21 (s)
17.4 CH3
19.4 CH3
24.3 CH3
1.21 (s)
1.07 (s)
1.40 (s)
0.85 (s)
1.08 (s)
1.36 (s)
1.46 (s)

Molecules 2015, 20
6277
Table 1. Cont.
1
2
3
Position
δC mult
δH (J in Hz)
0.70 (s)
δC mult
δH (J in Hz)
δC mult
29.3 CH3
20.6 CH3
181.7 qC
105.3 CH
81.6 CH
78.6 CH
78.3 CH
74.1 CH
173.4 qC
105.5 CH
76.5 CH
78.5 CH
78.4 CH
73.9 CH
173.2 qC
δH (J in Hz)
28
29
30
1'
30.8 CH3
30.2 CH3
0.93 (s)
0.88 (s)
1.42 (s)
72.9 CH2 4.06 (d, 10.5), 3.98 (d, 10.5) 181.0 qC
180.2 qC
106.7 CH
85.5 CH
79.4 CH
75.3 CH
79.3 CH
174.8 qC
108.1 CH
78.5 CH
79.4 CH
75.3 CH
80.1 CH
174.8 qC
22.0 CH3
105.9 CH
84.8 CH
78.6 CH
78.2 CH
74.2 CH
173.8 qC
107.3 CH
77.6 CH
78.5 CH
74.4 CH
79.2 CH
173.4 qC
1.44 (s)
5.09 (d, 7.5)
4.26 *
5.03 (d, 8.0)
4.18 *
5.09 (d, 8.0)
4.38 *
2'
3'
4.49 *
4.54 *
4.56 *
4'
4.38 *
4.69 *
4.71 *
5'
4.65 *
4.53 *
4.55 *
6'
1''
2''
3''
4''
5''
6''
5.38 (d, 7.5)
4.14 *
5.45(d, 7.5)
4.22 *
5.77 (d, 8.0)
4.31 *
4.32 *
4.42 *
4.42 *
4.47 *
4.62 *
4.63 *
4.51 *
4.61 *
4.62 *
* where 1H-NMR signals were overlapped, chemical shift data were obtained from 2D correlations.
Compound 3 produced a protonated ion at m/z 839.4069 [M + H]+ by HR-ESI-MS, which indicated
13
its molecular formula C42H62O17. In the C-NMR spectrum eight tertiary carbons at δC 70–85, two
carboxyl carbons at δC 173.4 and 173.2, and two anomeric carbons δC 105.3 and 105.5 suggested the
presence of two glucuronopyranosyl moieties. This conclusion was confirmed by two fragment ion peaks
at m/z 663.3785 [M − C6H8O6 + H]+ and 487.3373 [M − 2C6H8O6 + H]+, as well as acid hydrolysis and
GC analysis.
The NMR spectrum of 3 were very similar to that of 4, and detailed analysis revealed that the main
difference could be seen in the NOESY spectrum. Comparative analysis of this spectrum showed that
correlations of H-18 with H-19α and H-29 in 3 (Figure 2), rather than correlations of H-18 with H-28β
in 4, were present, which indicated that H-18 of 3 was α-oriented. The other correlations of 3, including
H-1β/H-25 and H-26, H-3/H-5 and H-1α, H-23/H-3 and H-1', were the same as those of 4. Accordingly,
compound 3 was identified as 3β-O-[β-D-glucuronpyranosyl-(1→2)-β-D-glucuron-pyranosyl]-24-
hydroxy-18α-glycyrrhetic acid, which was assigned the trivial name licorice-saponin Q2.
The known constituents, namely macedonoside A (2) [10], licorice-saponin G2 (4) [11], licorice-saponin
E2 (5) [12], 22β-acetoxyglycyrrhizin (6) [13], uralsaponin D (7) [7], 24-hydroxylicorice-saponin E2 (8) [14],
22β-acetoxyglycyrrhaldehyde (9) [15], licorice-saponin A3 (10) [12] and glycyrrhizin (11) [15] were
identified by comparison of their NMR data with the literature data.
2.2. Hepatoprotective Activity
All the separated compounds were assessed for their hepatoprotective activities against the increase
of AST and LDH levels in primary rat hepatocytes injured by D-GalN. The maximum nontoxic
concentrations of tested compounds on primary rat hepatocytes were in the range of 120–240 μM. A set
of cells in culture medium treated with D-GalN was used as the model group, and in comparison to the model

Molecules 2015, 20
6278
group, macedonoside A (2), licorice-saponin Q2 (3), licorice-saponin G2 (4), 22β-acetoxy-glycyrrhizin
(6) and glycyrrhizin (11) notably lowered AST (10.3–16.5 U·L−1) and LDH (200.7–242.8 U·L−1) in the
range of concentration 30–120 μM. (Table 2).
Table 2. Hepatoprotective activities of isolated saponins on AST and LDH Levels in primary
cultures of rat hepatocytes injured by D-GalN a.
Groups
Control
Model
Concentration (μM) AST (U·L−1)
LDH (U·L−1)
157.4 ± 11.7
253.5 ± 13.5 b
231.0 ± 18.3 c
213.7 ± 19.6 d
215.9 ± 8.2 d
224.6 ± 21.4 c
208.6 ± 16.0 d
200.7 ± 15.1 d
237.2 ± 16.4
232.7 ± 25.7
224.9 ± 25.2 c
242.8 ± 15.2
228.7 ± 14.0 c
221.7 ± 19.5 d
236.3 ± 9.4 c
219.1 ± 19.3 d
207.8 ± 21.9 d
219.1 ± 10.9 d
6.9 ± 1.7
17.1 ± 2.4 b
30
60
14.6 ± 3.1
13.1 ± 2.1 c
11.2 ± 1.9 d
13.3 ± 2.9 c
12.9 ± 3.2 d
10.8 ± 2.8 d
14.8 ± 2.9
2
3
4
6
120
30
60
120
30
60
13.7 ± 2.6 c
12.6 ± 2.8 c
16.5 ± 2.2
120
30
60
14.6 ± 3.0
120
30
13.2 ± 2.7 c
14.1 ± 2.1 c
13.0 ± 2.2 c
11.6 ± 1.8 d
10.3 ± 2.2 d
11
60
120
Silibinin Meglumine
50
a n = 3, mean ± SD. Control: a set of hepatocytes maintained in culture medium. Model: a set of hepatocytes
maintained in culture medium and treated only with D-GalN. b p < 0.01, compared to control group. c p < 0.05,
compared to model group. d p < 0.01, compared to model group.
Comparing the activities of these saponins, compound 5 and 7 was shown to have significantly weaker
hepatoprotective activities than the compound 2 and 11 owing to presence of a lactone ring at position
22(30). Compound 11 showed stronger activity than 1. That might be because an additional CH2OH
group is preferable to improve the steric hindrance, thus resulting in a decrease in the bonding capacity
with active targets. Interestingly, compound 3 displayed higher activity than compound 4. The reason
might be that compound 3 with a 18α-H group was found to be favorable for the anti-liver injury activity.
On the basis of the above analysis, it seemed that a carboxyl residue at position 29 or 30 was possibly
the necessary group for hepatoprotective activity.
2.3. Enzyme Inhibition Activity
As a regulator associated with the stability of the liver cell membrane, phospholipase A2 (PLA2) is a
promising target for hepatoprotective drug development [16]. To examine whether the compounds
inhibit activities on PLA2, the enzyme inhibitory potency of all isolated compounds was conducted and
the results were summarized in Table 3. Among these, two saponins (compounds 2 and 3) and

Molecules 2015, 20
6279
glycyrrhizin (11) exhibited efficient inhibitory activity with IC50 value of 6.9 μM, 3.6 μM and 9.3 μM,
respectively. Compounds 4, 6 and 7 showed moderate inhibitory activities with IC50 values of 16.9 μM,
27.1 μM and 32.2 μM, respectively.
What was noteworthy, is that analysis of the two assays of 1–11 showed that there was good
relationship between PLA2 inhibitory activities and hepatoprotective effects, leading to the hypothesis
that inhibition of PLA2 was one of the possible mechanisms of the hepatoprotective effect of licorice saponins.
Table 3. Inhibitory activities of isolated saponins on PLA2.
Compounds
IC50 (μM)
>50
1
2
6.9 ± 0.5
3.6 ± 0.3
16.9 ± 0.3
>50
3
4
5
6
27.1 ± 0.9
32.2 ± 0.5
>50
7
8
9
>50
10
>50
11
9.3 ± 0.8
1.8 ± 0.1
Diethylenetriaminepentaacetic acid
3. Experimental Section
3.1. General Procedures
UV-vis spectra were recorded using a UV-2401 spectrophotometer (Shimadzu, Kyoto, Japan). The
1D and 2D-NMR spectra (1H-1H COSY, NOESY, HSQC and HMBC) were obtained using an ASR-500
1
13
spectrometer (Bruker, Fällanden, Switzerland, 500 MHz for H and 500 MHz for C spectra). All
compounds were dissolved in C5D5N and chemical shifts were reported in ppm (δ) relative to TMS.
HR-TOF-MS was recorded on a Bruker MicroTOF-Q spectrometer. Column chromatography was
performed on polyamide resin (100–200 mesh) and macroporous resin (20–40 mesh, AB-8) columns.
Medium pressure liquid chromatography (MPLC) was carried out on a LISURE apparatus
(Chromatography Pump E2-purifier, Lisure Science Co., Ltd, Suzhou, China) with ODS column
(6.5 cm × 50 cm; 25–50 μm, Merck K GaA). Preparative HPLC was performed by Econosil C18 column
(22 × 250 mm; 10 μm, Alltech, Lexington, KY, USA) on a Waters 600 HPLC instrument (Waters,
Milford, MA, USA). HPLC was performed on an Agilent 1100 HPLC instrument (Agilent Technologies
Inc., Palo Alto, CA, USA) connected to a UV detector, which was equipped with a Thermo C18 column
(250 mm × 4.6 mm, I.D. 5 μm). GC was run on Varian CP-3800 Gas Chromatograph (VARIAN, INC.,
Palo Alto, CA, USA) equipped with a CP-sil 5 CB capillary column (30 m, 0.25 mm i.d., 0.25 μm) and
a Saturn 2200 Mass detector. The chemical reagents were supplied by Nanjing Chemical Plant (Nanjing,
China). The fluorescence value of each well in the biological activity assays was read in a ELISA plate
reader (Bio-Tek Instruments, Winooski, VT, USA).

Molecules 2015, 20
6280
3.2. Material
The roots of Glycyrrhiza inflata were collected in Weli County, Xinjiang Uygur Autonomous Region,
China, October 2013. Avoucher sample (No. 20131015) was preserved in Nanjing University of Chinese
Medicine, and identified by Prof. Qi-Nan Wu.
3.3. Extraction and Isolation
The roots of G. inflata (dry weight, 25 kg) were exhaustively extracted two times with boiling water
(200 L × 2, each extraction lasted 2 hours). The combined solutions (about 300 L) were passed over a
polyamide resin column (30 L, 100–200 mesh, 20 × 200 cm) with a flow rate of 120 mL/min, and then
the effluent was chromatographed on a macroporous resin column (25 L, 20–40 mesh, 20 × 200 cm)
using EtOH–H2O (70:30, 30 L, flow rate 100 mL/min) as eluent to afford the crude saponin fraction.
The 70% EtOH combined elution was concentrated in vacuo. The residue (a total of about 350 g) was
then subjected in four portions (90 g of residue each) to MPLC with a ODS column (800 g, 25–50 μm,
6.5 × 50 cm) using a continuous gradient of MeOH–H2O–HCOOH (50:50:1, 55:45:1, 60:40:1, 65:35:1,
v/v, each 4 L) to produce four fractions (Fr. I–IV). Fr. II (about 45 g) was further subjected to MPLC
chromatography on an ODS column with a gradient of MeOH–H2O–HCOOH (50:50:1, 52:48:1, 54:46:1,
56:44:1, 58:42:1, 60:40:1, each 3 L, v/v) as eluent to afford six subfractions (SFr. IIa–IIe). SFr. IIa was
subjected to preparative HPLC with CH3CN–H2O–HCOOH (27:73:1, flow rate 10 mL/min) elution to
give compound 2 (55 mg, tR 19.2 min). SFr. IIb was purified by preparative HPLC using CH3CN–H2O–
HCOOH (27:73:1, flow rate 10 mL/min) to give compound 1 (32 mg, tR 22.6 min) and compound 10
(72 mg, tR 34.2 min). SFr. IIc was subjected to preparative HPLC with CH3CN–H2O–HCOOH (30:70:1,
flow rate 10 mL/min) as eluent to afford compound 3 (38 mg, tR 22.6 min), compound 4 (205 mg, tR
29.1 min) and compound 8 (26 mg, tR 37.5 min). SFr. IId was chromatographed by preparative HPLC
and eluted with CH3CN–H2O–HCOOH (32:68:1, flow rate 10 mL/min) to give compound 5 (81 mg, tR
26.1 min) and compound 7 (34 mg, tR 34.5 min). SFr. IIe was subjected to preparative HPLC with
CH3CN–H2O–HCOOH (35:65:1, flow rate 10 mL/min) to produce compound 6 (87 mg, tR 30.1 min)
and compound 9 (40 mg, tR 39.4 min). Separation of fraction III (about 75 g) by MPLC with ODS column
chromatography yielded 11 (6.2 g) using CH3CN–H2O–HCOOH (40:6:1, flow rate 30 mL/min) as eluent.
3.4. Physical Data of New Compounds
Licorice-Saponin P2 (1)
White amorphous powder; UV (MeOH) λmax (log ε) 250.2 (4.14) nm; IR (KBr) vmax (cm−1): 3463,
1
13
2964, 1751, 1652; H-NMR (C5D5N, 500 MHz) and C-NMR (C5D5N, 125 MHz) spectral data see
Table 1; HR-TOF-MS: m/z 839.4120 [M+H]+ (calcd. for C42H63O17, 839.4065).
Licorice-Saponin Q2 (2)
White amorphous powder; UV (MeOH) λmax (log ε) 250.2 (4.16) nm; IR (KBr) vmax (cm−1): 3438,
2975, 1751, 1652; 13C-NMR (C5D5N, 125 MHz) spectral data see Table 1; HR-TOF-MS: m/z 839.4069
[M+H]+ (calcd for C42H63O17, 839.4065).

Molecules 2015, 20
6281
3.5. Acid Hydrolysis
The configuration of the sugars of compounds 1 and 3 was determined by acid hydrolysis and GC
experiments based of the literature procedure [6,9]. The specific steps were as follows: a solution of
compounds 1–3 (1.0 mg each) in 1 N HCl (1 mL) was stirred at 90 °C for 2 h. After cooling, the solution
was evaporated under a stream of N2. Anhydrous pyridine solutions (0.1 mL) of each residue and
L-cysteine methyl ester hydrochloride (0.06 N) were mixed and warmed at 60 °C for 1 h. The
trimethylsilylation reagent trimethylsilylimidazole (0.15 mL) was added, followed by warming at 60 °C
for another 30 min. After drying the solution, the residue was partitioned between H2O and CH2Cl2
(1 mL, 1:1 v/v). The CH2Cl2 layer was analyzed by GC/MS. The peaks of authentic sample of
D-glucuronic acid after treatment in the same way were detected at 14.23 min. The final result was to
compare the retention times of monosaccharide derivatives with standard sample. The absolute
configuration of sugar was confirmed to be D-glucuronic acid (D-glucuronic acid for compound 1 with
retention time 14.21 min; D-glucuronic acid for compound 3 with retention time 14.22 min).
3.6. Cell Assay
Isolated rat hepatocytes were prepared from male Wistar rats by a collagenase perfusion technique as
described previously [17]. The D-GalN concentration used for cell culture treatment was previously
determined according to a modification of the method of Morikawa et al. [18]. The cultured cells in
logarithmic growth phase were made into a single-cell suspension and seeded in 96-well plates
(1 × 104 cells/well) in the DMEM/F 12 with 2% FBS complete medium for 24 h at 37 °C. Then, the
hepatocytes were exposed to 2 mM D-GalN for 2 h to induce hepatotoxocity. The medium with silibin
meglumine (as positive drug, purity 95.6%, Hunan Xieli Pharmaceutical Co., Ltd., Zhuzhou, China) and
different concentrations of test compounds was mixed in cell medium (final test compounds
concentration were 30 μM, 60 μM and 120 μM, respectively), and incubated for 24 h. The obtained
reacted supernatant was directly used to detect ALT and AST levels. The control group was a set of cells
maintained in culture medium, while the model group was a set of cells maintained in culture medium
and treated only with D-GalN. All data are expressed as the mean ± SD of at least three independent
experiments as indicated. The test for the paired samples was used to determine statistical difference
between parameters. These differences were considered significant for p < 0.05 or 0.01.
3.7. Assay for Inhibition against PLA2
The PLA2 inhibitory assays of compounds 1–11 and the positive drug diethylenetriaminepentaacetic
acid (Purchased from Aladdin, Los Angeles, CA, USA, purity > 98.0%) were carried out according to
the literature [19]. First of all, each tube was added with 1 mL fresh substrate buffer solution (pH = 8.2).
After that, 50 μL tested compounds at various concentrations were placed at reaction tube and blank
tube, respectively. As for control tube, 50 μL deionized water was instead. Then each tube incubated at
40 °C for 10 minutes. The reaction tube and blank tube were followed by the treatment with PLA2
enzyme (5 μL) at the concentration of 5 μg/mL. Before put them into the incubator at the temperature of
40 °C to react 30 minutes, the content of the tube should be fully blending. The optical density value of
each tube was then read in an ELISA plate reader using a wavelength of 495 nm. The IC50 values were

Molecules 2015, 20
6282
calculated from concentration-response curves by means of the GraphPad Prism 5.0 Software
(San Diego, CA, USA). Each experiment was repeated three times to get the mean values.
Supplementary Materials
HRESIMS and NMR spectra of the two new compounds and the 13C-NMR data of compound 2 and
4 are available as Supplementary Materials. Supplementary materials can be accessed at:
http://www.mdpi.com/1420-3049/20/04/6273/s1.
Acknowledgments
The authors greatly appreciate financial support from the National Natural Science Foundation, China
(Program No. 81202881), a project funded by the Priority Academic Program Development of Jiangsu
Higher Education Institutions. We also thank Qi-Nan Wu (Nanjing University of Chinese Medicine,
Nanjing, China) for identification of Glycyrrhiza inflata L. and the Shanghai Institute of Organic
Chemistry (Chinese Academy of Sciences, Shanghai, China) for measuring the NMR spectra.
Author Contributions
The list authors contributed to this work as follows: Y. Z. conceived and designed the experiments,
J. W., S. F., and T. W. performed the experiments and analyzed the data, H. C. and C. L. wrote the paper,
Y. T. and G. P. polished the paper. All authors read and approved the final manuscript.
Conflicts of Interest
The authors declare no conflict of interest.
References
1. Shim, S.B.; Kim, N.J.; Kim, D.H. Beta-glucuronidase inhibitory activity and hepatoprotective effect
of 18 beta-glycyrrhetinic acid from the rhizomes of Glycyrrhiza uralensis. Planta Med. 2000, 66,
40–43.
2. Wu, Y.T.; Shen, C.; Yin, J.; Yu, J.P.; Meng, Q. Azathioprine hepatotoxicity and the protective effect
of liquorice and glycyrrhzic acid. Phytother. Res. 2006, 20, 640–645.
3. Wolkerstorfer, A.; Kurz, H.; Bachhofner, N.; Szolar, O.H. Glycyrrhizin inhibits influenza A virus
uptake into the cell. Antivir. Res. 2009, 83, 171–178.
4. Fujisawa, Y.; Sakamoto, M.; Matsushita, M.; Fujita, T.; Nishioka, K. Glycyrrhizin inhibits the lytic
pathway of complement–possible mechanism of its anti-inflammatory effect on liver cells in viral
hepatitis. Micrrobiol. Immunol. 2000, 44, 799–804.
5. Fu, Y.; Chen, J.; Li, Y.J.; Zheng, Y.F.; Li, P. Antioxidant and anti-inflammatory activities of six
flavonoids separated from licorice. Food Chem. 2013, 141, 1063–1073.
6. Wei, J.H.; Zheng, Y.F.; Li, C.Y.; Tang, Y.P.; Peng, G.P. Bioactive constituents of oleanane-type
triterpene saponins from the roots of Glycyrrhiza glabra. J. Asian Nat. Prod. Res. 2014, 16,
1044–1053.

Molecules 2015, 20
6283
7. Zheng, Y.F.; Qi; L.W.; Cui, X.B.; Peng, G.P.; Peng, Y.B.; Ren, M.T.; Cheng, X.L.; Li, P.
Oleanane-type Triterpene Glucuronides from the Roots of Glycyrrhiza uralensis Fischer.
Planta Med. 2010, 76, 1457–1463.
8. Antonov, A.S.; Avilov, S.A.; Kalinovsky, A.I.; Anastyuk, S.D.; Dmitrenok, P.S.; Kalinin, V.I.;
Taboada, S.; Bosh, A.; Avila, C.; Stonik, V.A. Triterpene Glycosides from Antarctic Sea
Cucumbers. 2. Structure of Achlioniceosides A1, A2, and A3 from the Sea Cucumber Achlionice
Wiolaecuspidata (=Rhipidothuria racowitzai). J. Nat. Prod. 2009, 72, 33–38.
9. Duan, W.J.; Yang, J.Y.; Chen, L.X.; Zhang, L.J.; Jiang, Z.H.; Cai, X.D.; Zhang, X.; Qiu, F.
Monoterpenes from Paeonia albiflora and their inhibitory activity on nitric oxide production by
lipopolysaccharide-activated microglia. J. Nat. Prod. 2009, 72, 1579–1584.
10. Shibano, M.; Nukui, H.; Kita, S.; Kusano, G.; Shibata, T.; Watanabe, H.; Ohashi, H. Studies on natural
medicines index compounds for HPLC Glycyrrhiza macedonica. Nat. Med. 1999, 53, 166–172.
11. Kitagawa, I.; Zhou, J.L.; Sakagami, M.; Uchida, E.; Yoshikawa, M. Licorice-saponins F3, G2, H2,
J2, and K2, five new oleanene-triterpene oligoglycosides from the root of Glycyrrhiza uralensis.
Chem. Pharm Bull. 1991, 39, 244–246.
12. Kitagawa, I.; Zhou, J.L.; Sakagami, M.; Taniyama, T.; Yoshikawa, M. Licorice-saponins A3, B2,
C2, D3 and E2, five new oleanene-type triterpene oligoglycosides from Chinese Glycyrrhizae Radix.
Chem. Pharm Bull. 1988, 36, 3710–3713.
13. Li-Yang, J.; Nakajima, J.I.; Kimura, N.; Saitob, K.; Seo, S. Oleanane-type triterpene glycosides
from Glycyrrhiza uralensis. Nat. Prod. Commun. 2007, 2, 243–248.
14. Zhu, X.M.; Di, Y.T.; Peng, S.L.; Wang, M.K.; Ding, L.S. Chemical constituents from root of
Glycyrrhiza uralensis. Chin. Tradit. Herb Drugs 2003, 34, 198–201.
15. Zhang, R.Y.; Zhang, J.H.; Wang, M.T. Studies on the saponins from the root of Glycyrrhiza
uralensis Fisch. Yao Xue Xue Bao 1986, 21, 510–515.
16. Ohtsuki, K.; Abe, Y.; Shimoyama, Y. Separation of phospholipase A2 in Habu snake venom by
glycyrrhizin (GL)-affinity column chromatography and identification of a GL-sensitive enzyme.
Biol. Pharm Bull. 1998, 21, 574–578.
17. Lee, E.J.; Kim, S.R.; Kim, J.; Kim, Y.C. Hepatoprotective phenylpropanoids from Scrophularia
buergeriana roots against CCl(4)-induced toxicity: action mechanism and structure-activity
relationship. Planta Med. 2002, 68, 407–411.
18. Morikawa ,T.; Ninomiya, K.; Imura, K.; Yamaguchi, T.; Akagi, Y.; Yoshikawa, M.; Hayakawa, T.;
Muraoka, O. Hepatoprotective triterpenes from traditional Tibetan medicine Potentilla anserina.
Phytochemistry 2014, 102, 169–81.
19. Liu, B.; Qi, Y.; Wang, M.; Cai, R.L.; Song, Y. A method for screening Phospolipase A2 inhibitors
in vitro. Chin. Pharma Bull. 2006, 22, 890–892.
Sample Availability: Samples are not available from authors.
© 2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article
distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/4.0/).