Journal of Chromatography B, 969 (2014) 190198

Contents lists available at ScienceDirect

Journal of Chromatography B
journal homepage: www.elsevier.com/locate/chromb

Isolation of two new prenylated avonoids from Sinopodophyllum

emodi fruit by silica gel column and high-speed counter-current
chromatography
Yanjun Sun a,b , Yinshi Sun c , Hui Chen a,b , Zhiyou Hao a,b , Junmin Wang a,b , Yanbin Guan a,b ,
Yanli Zhang a,b , Weisheng Feng a,b, , Xiaoke Zheng a,b,
a
Collaborative Innovation Center for Respiratory Disease Diagnosis and Treatment & Chinese Medicine Development of Henan Province,
Henan University of Traditional Chinese Medicine, Zhengzhou, Henan 450046, China
b
School of Pharmacy, Henan University of Traditional Chinese Medicine, Zhengzhou, Henan 450046, China
c
College of Agronomy, Shandong Agricultural University, Taian, Shandong 271018, China

a r t i c l e

i n f o

Article history:
Received 28 March 2014
Received in revised form 14 June 2014
Accepted 9 August 2014
Available online 19 August 2014
Keywords:
Sinopodophyllum emodi
New prenylated avonoid
Silica gel column chromatography
High-speed counter-current
chromatography

a b s t r a c t
Two new prenylated avonoids, sinoavonoids AB, were isolated from the dried fruits of Sinopodophyllum emodi by silica gel column chromatography (SGCC) and high-speed counter-current chromatography
(HSCCC). The 95% ethanol extract was partitioned with petroleum ether, dichloromethane, ethyl acetate,
and n-butanol in water, respectively. The ethyl acetate fraction was pre-separated by SGCC with a
petroleum etheracetone gradient. The eluates containing target compounds were further separated
by HSCCC with n-hexaneethyl acetatemethanolwater (4:6:4:4, v/v). Finally, 17.3 mg of sinoavonoid
A and 25.9 mg of sinoavonoid B were obtained from 100 mg of the pretreated concentrate. The purities of
sinoavonoid A and sinoavonoid B were 98.47% and 99.38%, respectively, as determined by HPLC. Their
structures were elucidated on the basis of spectroscopic evidences (HR-ESI-MS, 1 H-NMR, 13 C-NMR, HSQC,
HMBC). The separation procedures proved to be efcient, especially for trace prenylated avonoids.
2014 Elsevier B.V. All rights reserved.

1. Introduction
Sinopodophyllum emodi (Wall.) Ying (Berberidaceae) is distributed widely in the southwest of China [1]. The dried ripe fruit
of S. emodi, called Xiaoyelian in Chinese, is used to regulate menstruation and promote the circulation of blood. As a traditional
Tibetan medicine, it is clinically applied to the treatment of amenorrhea, dead fetus, and placental retaining. It has been recorded
in Chinese Pharmacopoeia and Yue Wang Yao Zhen (Somaratsa),
which was compiled in the mid-eighth century as the earliest
literature on traditional Tibetan medicine [2]. Previous chemical investigations on S. emodi revealed the presence of lignans,
avonoids, steroids, phenolics [111]. The prenylated avonoids,
which were representative bioactive constituents from the genus
Sinopodophyllum, possessed multiple biological activities including anti-oxidant, anti-cancer, anti-inammatory, anti-bacterium,

Corresponding author. Tel.: +86 371 65962746; fax: +86 371 65962746.
Co-corresponding author. Tel.: +86 371 65962746; fax: +86 371 65962746.
E-mail addresses: fwsh@hactcm.edu.cn (W. Feng), zhengxk.2006@163.com
(X. Zheng).
http://dx.doi.org/10.1016/j.jchromb.2014.08.017
1570-0232/ 2014 Elsevier B.V. All rights reserved.

anti-osteoporosis, prevention of Alzheimers disease, anti-diabete,

cardiovascular protection, and estrogen-like effect [12].
Conventional isolation strategies for prenylated avonoids from
S. emodi involved multiple chromatographic steps, which were
time-consuming and resulted in a substantial loss of samples due
to irreversible adsorption [2,10]. Therefore, it is indispensible to
develop a rapid and efcient method for the purication of prenylated avonoids from S. emodi. HSCCC is a peculiar liquid-liquid
partition technology, which can eliminate some complications such
as irreversible adsorption and deactivation of target compounds
[13]. HSCCC has been widely used in the preparative isolation and
purication of natural products with high recovery and loading
capacity, low solvent consumption, acceptable efciency, and
the ease of scaling-up [14]. To the best of our knowledge, only
avonoid glycosides [1517], isoavones [18], polymethoxylated
avones [19,20] were successfully isolated from natural sources
by SGCC and HSCCC. Although HSCCC has been developed for the
purication normal avonoid (only kaempferol) [21], there are
no reports on the application of SGCC and HSCCC for the isolation
of prenylated avonoids. The present research established an
optimal preparation method of two new prenylated avonoids
(sinoavonoids A-B) (Fig. 1) from S. emodi with SGCC and HSCCC.

Y. Sun et al. / J. Chromatogr. B 969 (2014) 190198

191

Fig. 1. The chemical structures of two new avonoids from Sinopodophyllum emodi.

2. Experimental

2.3. Preparation of the crude extract

2.1. Apparatus

The fruits of S. emodi (4 kg) were dried constantly at 60 C, then

grilled to powder, and passed through a 40-mesh sieve. The powder
was ultrasonically extracted by 10-fold amounts of 95% ethanol
at 55 C for 60 min. The working frequency and power were xed
at 40 kHz and 200 W, respectively. The extraction procedure was
then repeated twice. After combination and removal of the ethanol
under reduced pressure, the extract was suspended in 5 L distilled
water, and subjected to a series of solvent extractions (1:1, v/v) with
petroleum ether (b.p. 6090 C), dichloromethane, ethyl acetate,
and nally n-butanol. After concentration and freeze-drying, the
dried fractions were stored at 10 C.

Ultrasonic-assisted extraction (UAE) was performed on a

digitally controlled ultrasonic apparatus (KQ 5200V, Kunshan
Ultrasonic Instruments Manufacture Co. Ltd., China).
The HSCCC experiments were carried out by a Model GS-10A
high-speed counter-current chromatography (Beijing Institute of
New Technology Application, Beijing, China). The instrument was
equipped with a polytetrauoroethylene multilayer coil column
(i.d. of the tubing = 1.6 mm, total volume = 230 mL) and a manual
sample injection valve with a 10 mL loop. The revolution radius
between the holder axis and central axis of the centrifuge (R) was
5 cm, and the value of the multilayer coil varied from 0.5 at internal terminal to 0.8 at the external terminal ( = r/R, where r was
the distance from the coil to the holder shaft, and R was the revolution radius or the distance between the holder axis and central
axis of the centrifuge). The revolution speed of the apparatus could
be regulated by a speed controller ranging from 0 to 1000 rpm.
The HSCCC system was also equipped with BF-2002 CT11 signal
collection cell (Chromatography Center of Beifenruili Group Company, Beijing, China), a Model NS-1007A constant-ow pump and
a Model 8823B-UV Monitor (Beijing Institute of New Technology
Application, Beijing, China) at 254 nm. The data were collected with
HW-2000 chromatography workstation (Qianpu Software Co. Ltd.,
Shanghai, China).
The analytical HPLC data were measured on a Waters Alliance
2489 separations module equipped with a Waters 2695 UV/visible
Detector and Empower pro data handling system (Waters Co., Milford, USA).
The structures of the target compounds were determined by
high resolution electrospray ionization mass (HR-ESI-MS) spectrometer (Shimadzu LC-MS 2010, Japan) and nuclear magnetic
resonance (NMR) spectrometer (Bruker AM 500, Switzerland).
2.2. Materials and reagents
Methanol used for HPLC was of chromatographic grade
(Siyou Biology Medical Tech Co. Ltd., Tianjin, China), and water
was puried by means of a water purier (18.2 M) (Wanjie Water Treatment Equipment Co. Ltd., Hangzhou, China). All
organic solvents used for preparation of crude fractions, SGCC
and HSCCC separation were of analytical grade (Fuyu Chemical
Reagent Co. Ltd., Tianjin, China). The chromatographic silica gel
was produced from Qingdao Ocean Chemical Factory (Qingdao,
China).
The fruits were collected in Deqin, Yunnan province, Peoples
Republic of China, in September 2012, and were identied as
the fruits of S. emodi (Wall.) Ying by Professor Chengming Dong
(School of Pharmacy, Henan University of Traditional Chinese
Medicine).

2.4. Pre-separation of the crude extract by SGCC

The ethyl acetate fraction (50 g) was dissolved in acetone
(80 mL), and added to silica gel (60 g, 200300 mesh) by constant
stirring. Acetone was evaporated in a rotary evaporator under vacuum at 45 C. The fraction impregnated silica gel was cooled down
to room temperature and kept in a desiccator until required.
The silica gel chromatography column was packed as follows:
the exit of the chromatography column was plugged with glass
wool to retain solids. Silica gel (1.94 kg, 200300 mesh) was suspended in petroleum ether (7.5 L), and then transferred to the
column (140 cm lenth 8 cm i.d.). The column was rinsed with 7.5 L
of petroleum ether-acetone (10:1, v/v). Prior to sample application,
the level was lowered 20 cm above the stationary phase. The fraction impregnated silica gel was added to the top of the column, and
elution was performed with a petroleum etheracetone gradient
(10:1, 10:3, 10:5, 15 L each) at a constant ow rate of 15 mL min1 .
Fractions of 100 mL each were analyzed by TLC. TLC analyses were
performed on GF254 silica gel plates at room temperature, using
petroleum etheracetone (1:1, v/v) as developing reagent. Spots
were visualized under an ultraviolet lamp at 254 nm. The eluates
containing target compounds were pooled and concentrated under
reduced pressure. The concentrate was stored at 10 C for the
subsequent HSCCC separation.

2.5. Further separation by HSCCC

2.5.1. Determination of the partition coefcient
The partition coefcient (K) of target compounds in different
two-phase solvent systems was determined by HPLC as follows:
25 mg of the pretreated concentrate and 1 mL of the equilibrated
two-phase solvent were added into a 5 mL centrifuge tube. The centrifuge tube was then stoppered and vortically mixed for 1 min to
thoroughly equilibrate the sample between the two phases. Then
an aliquot of each phase (10 L) was analyzed by HPLC. The K value
was expressed as the ratio of the peak area of a given compound in
the upper phase divided that in the lower phase.

192

Y. Sun et al. / J. Chromatogr. B 969 (2014) 190198

Fig. 2. HSCCC chromatogram of the pretreated concentrate from SGCC. Two-phase solvent system: n-hexaneethyl acetatemethanolwater (4:6:4:4, v/v); mobile phase:
the lower phase; stationary phase: the upper phase; ow rate: 1.7 mL min1 ; revolution speed: 800 rpm; detection wavelength: 254 nm; sample size: 100 mg pretreated
concentrate was dissolved in the solvent mixture of n-hexaneethyl acetatemethanolwater (5 mL for each phase); the retention percentage of the stationary phase: 67%.

2.5.2. Preparation of two-phase solvent system and sample

solution
A solvent system consisting of n-hexaneethyl acetate
methanolwater (4:6:4:4, v/v) was prepared by adding the solvents to a separation funnel according to the volume ratios and
thoroughly equilibrated by shaking repeatedly. Then the two
phases were separated and degassed by sonication for 30 min
before use. The sample solution was prepared as follows: 100 mg
pretreated concentrate was dissolved in the solvent mixture of nhexaneethyl acetatemethanolwater (5 mL for each phase).
2.5.3. HSCCC separation procedure
The separations were initiated by lling the coiled column with
the upper phase (stationary phase). The lower phase (mobile phase)
was pumped into the column in a head-to-tail mode at a ow
rate of 1.7 mL min1 , when the apparatus was rotated at 800 rpm.
Approximately 10 mL of the sample solution containing 100 mg of
the pretreated concentrate was injected into the head of the column through the injection valve after hydrodynamic equilibrium
was established in the column, as indicated by the mobile phase
eluting from the tail outlet. The eluates from the column outlet
were continuously monitored by a UV detector at 254 nm. The fractions during 120150 min (peak 1) and 170200 min (peak 2) were
collected respectively according to the elution prole (Fig. 2). All
fractions of the same pure compound were combined and evaporated under reduced pressure. The puried compounds were stored
at 20 C before HPLC and NMR analyses.
2.5.4. HPLC analysis and identication of HSCCC peak fractions
The pretreated concentrate from SGCC and each HSCCC peak
fraction were analyzed by HPLC (Fig. 3AC). Analyses were accomplished on a YMC-Pack ODS A column (5 m, 250 mm 4.6 mm) at
35 C. Methanol (A) 0.1% triuoroacetic acid (v/v) (B) was used as
the mobile phase in gradient elution mode as follows: 3085% A
at 030 min, 85100% A at 3040 min. The ow rate of the mobile
phase was 1.0 mL min1 . The eluates were monitored at 254 nm
by a UV/visible detector. Based on the peak area normalized to all
observed HPLC peak areas, the purities of the isolated avonoids
were determined.

3. Results and discussions

3.1. Pre-separation by SGCC
As seen in the HPLC chromatogram (Fig. 4), the high concentration of avonoids and lignans, and a trace of sinoavonoids A (peak
1, 70.1 min) and B (peak 2, 78.9 min), were present in the ethyl
acetate fraction. To improve the preparative separation efciency,
SGCC was employed for pre-separation, eluting with a petroleum
etheracetone gradient. The chromatographic parameters on separation efciency, including ow rate, mobile phase composition,
and loading amount were investigated to produce optimum separation.
Three kinds of binary solvent systems were tested, includdichloromethanemethanol,
dichloromethaneacetone,
ing
and petroleum etheracetone. With the mixed solvents of
dichloromethanemethanol and dichloromethaneacetone, the
eluates contained a lot of impurities such as citrusinol (peak 3,
15.9 min) and 8-prenylkaempferol (peak 4, 22.0 min) (Fig. 3D and
E), which possessed the similar polarities as target compounds.
Selection for ideal mobile phase depended on the recovery of target
compounds, the amount of eluent required, eluting time, and satisfactory chemical separation [22]. Based on the aforementioned
criterions, the binary solvents of petroleum etheracetone were
selected here. The different ratios of petroleum etheracetone (v/v)
were also investigated systematically. The target compounds could
not be eluted by petroleum ether. With petroleum etheracetone
(10:1), it could be eluted but a large amount of eluent was needed.
With petroleum etheracetone (10:5), the target compounds
moved down too quickly and could not be separated from other
components. Therefore, the target compounds were separated in
a gradient elution mode (10:1, 10:3, 10:5).
The separation efciency on the silica gel column decreased
and the recovery increased with the increase of loading amount
as it exceeded column loading capacity [22]. As the close polarities of prenylated avonoids from S. emodi, their separations are
inherently difcult. In addition, the prenylated avonoids can make
strong interactions on silica gel due to the existence of polar
functional groups (OH). With the decrease of required silica gel,
the adsorption of prenylated avonoids on silica gel decreased

Y. Sun et al. / J. Chromatogr. B 969 (2014) 190198

193

Fig. 3. (A) HPLC chromatogram of the pretreated concentrate from SGCC, which were isolated with the mixed solvents of petroleum etheracetone; (B) HPLC chromatogram
and UV spectrum of HSCCC peak fraction 1 in Fig. 2; (C) HPLC chromatogram and UV spectrum of HSCCC peak fraction 2 in Fig. 2; (D) HPLC chromatogram of the eluates
containing the target compounds from SGCC, which were isolated with the mixed solvents of dichloromethanemethanol; (E) HPLC chromatogram of the eluates containing the
target compounds from SGCC, which were isolated with the mixed solvents of dichloromethaneacetone. Experimental conditions: column, a YMC-Pack ODS A column (5 m,
250 mm 4.6 mm); mobile phase, methanol (A) and 0.1% triuoroacetic acid (B) at the gradient (3085% A at 030 min, 85100% A at 3040 min); ow rate, 1.0 mL min1 ;
detection wavelength, 254 nm; column temperature, 35 C.

194

Y. Sun et al. / J. Chromatogr. B 969 (2014) 190198

Fig. 4. HPLC chromatogram of the ethyl acetate fraction from Sinopodophyllum emodi. Experimental conditions: column, a YMC-Pack ODS A column (5 m, 250 mm 4.6 mm);
mobile phase, methanol (A) and 0.1% triuoroacetic acid (B) at the gradient (2022% A at 05 min, 2225% A at 512 min, 2545% A at 1255 min, 4565% A at 5571 min,
65100% A at 71110 min); ow rate, 1.0 mL min1 ; detection wavelength, 254 nm; column temperature, 35 C.

and their contents in the eluent increased. As it was loaded at

>25 mg g1 , it led to overlapping of neighboring peaks. Normally,
separation efciency will improve by increasing the ratio of adsorbent to sample due to higher interaction surface area available [23].
The satisfactory separation efciency of target compounds was
observed with from 0 to 25 mg g1 . These results showed that the
optimal loading amount on the silica gel column was determined
as 25 mg g1 .
Mobile phase ow rate is an important inuencing factor for
the separation efciency [24]. At ow rates <15 mL min1 , the separation efciency and elution time gradually increased with the
ow rate decreasing, which was attributed to the enhancement of
diffusion capacity. However, the ow rate was too fast to diffuse
effectively, so impurities were prominently present in the collected
fraction. Therefore, an ideal ow rate was selected at 15 mL min1 .

3.2. Selection of two-phase solvent system and preparative

HSCCC separation
The major difculty for successful separation by HSCCC was
to nd a suitable two-phase solvent system. The optimal volume
ratios of the two-phase solvent system were predicted according
to the K values for target compounds (i.e. usually between 0.5 and
2.5) [25]. Small K values resulted in the disappearance of fraction
resolution, while large K values tended to consume excessive solvent and long run time [26]. In view of the polarities of target
compounds, two kinds of relative mid-polar solvent systems at different volume ratio were investigated. The measured K values for
target compounds in different solvent systems were summarized
in Table 1. Started with n-hexaneethyl acetatemethanolwater
(4:4:4:4, v/v), the target compounds inclined to partition well into
the lower phase. The peaks were close to the solvent front, and
two target compounds could not be separated well. The K values
were adjusted by changing the volume ratios of n-hexane versus
ethyl acetate and methanol versus water. If the proportion of ethyl
acetate or methanol was increased, the distribution capacity of target compounds in the upper was stronger. For the solvent systems
composed of n-hexaneethyl acetatemethanolwater (4:4:6:4,
4:4:8:4, and 4:5:4:4 v/v), K values were acceptable, however, two
target compounds were co-eluted in short elution time. Meanwhile, the brown color of lower layer solution would deepen,

Table 1
The partition coefcients (K) of the target compounds in several solvent systems
(compound 1, sinoavonoid A; compound 2, sinoavonoid B).
Type of solvent system

Ratio

n-Hexaneethyl acetatemethanolwater
n-Hexaneethyl acetatemethanolwater
n-Hexaneethyl acetatemethanolwater
n-Hexaneethyl acetatemethanolwater
n-Hexaneethyl acetatemethanolwater
n-Hexaneethyl acetatemethanolwater
n-Hexaneethyl acetatemethanolwater
n-Hexaneethyl acetatemethanolwater
n-Hexaneethyl acetatemethanolwater
Dichloromethane-methanolwater
Dichloromethane-methanolwater

4:8:4:4
4:7:4:4
4:6:4:4
4:5:4:4
4:4:4:4
4:3:4:4
4:2:4:4
4:4:6:4
4:4:8:4
4:3:2
3:2:2

K value
1

3.19
2.58
1.02
0.63
0.35
0.28
0.19
0.55
0.79
0.59
0.31

3.64
2.79
1.44
0.79
0.48
0.37
0.26
0.63
0.85
0.73
0.52

indicating the presence of more pigments. In those systems of nhexaneethyl acetatemethanolwater (4:8:4:4, 4:7:4:4, 4:3:4:4,
and 4:2:4:4 v/v), K values were either too big or too small. In the case
of the solvent system of dichloromethanemethanolwater, the
slight emulsication occurred, and the target compounds could not
be separated from other compounds. In contrast, n-hexaneethyl
acetatemethanolwater at a ratio of 4:6:4:4 provided the suitable
K values and satisfactory separation of the target compounds, and
then was selected for subsequent HSCCC separation.
Using the selected solvent system, sinoavonoids A (17.3 mg)
and B (25.9 mg) with purities of 98.47% and 99.38%, respectively,
were successfully obtained from 100 mg of the pretreated concentrate in less than 240 min. The retention percentage of the
stationary phase was 67%, which was computed from the volume of
the stationary phase collected from the column after the separation
was completed.
3.3. Optimization of HPLC conditions
As shown in Fig. 3A, the HPLC chromatogram of the pretreated
concentrate shows several compounds where the purities of target compounds 1 and 2 were 33% and 49%, respectively, based
on HPLC peak area percentage. Good HPLC analytical conditions
were required to ensure the baseline separation of the target compounds and impurities. Thus, different elution mode, ow rate,

Y. Sun et al. / J. Chromatogr. B 969 (2014) 190198

195

Table 2
1
H NMR and 13 C NMR spectroscopic data for compounds 12.
Position

1a
C

2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
1 
2 
3 
4 
5 
1  
2  
3  
4  
5  
OCH3
OH
a

157.1
139.0
178.6
159.5
98.6
162.7
104.4
155.2
104.9
118.0
121.2
140.5
148.2
115.9
122.6
29.2
73.8
148.3
110.1
17.5
120.6
132.3
75.8
27.5
27.3
60.4

2a
H

6.28 (1H, s)

6.80 (1H, d, J = 8.4 Hz)

6.97 (1H, d, J = 8.4 Hz)
2.73 (2H, d, J = 6.9 Hz)
4.13 (1H, t, J = 6.9 Hz)
4.54 (1H, s), 4.52 (1H, s)
1.52 (3H, s)
6.29 (1H, d, J = 10.0 Hz)
5.80 (1H, d, J = 10.0 Hz)
1.40 (3H, s)
1.39 (3H, s)
3.58 (3H, s)
12.62 (1H, s)

159.8
139.5
178.6
160.1
99.5
159.1
100.1
154.3
105.7
121.4
121.8
143.7
147.5
112.9
121.4
16.0
31.3
76.6
26.6
26.6
26.2
123.5
130.5
17.5
25.5
60.2

6.12 (1H, s)

6.76 (1H, d, J = 8.3 Hz)

6.77 (1H, d, J = 8.3 Hz)
2.59 (2H, t, J = 6.7 Hz)
1.78 (2H, t, J = 6.7 Hz)
1.29 (3H, s)
1.29 (3H, s)
3.28 (2H, d, J = 6.8 Hz)
5.01 (1H, m)
1.26 (3H, s)
1.43 (3H, s)
3.57 (3H, s)
12.45 (1H, s)

NMR spectroscopic data were recorded in DMSO-d6 at 500 MHz (1 H NMR) and 125 MHz (13 C NMR).

column temperature, and detection wavelength were tested. With

the mobile phase of methanolwater, some degree of peak tailing
and broadening was unavoidable. The screening results indicated
the gradient elution of methanol (A) 0.1% triuoroacetic acid (v/v)
(B) (3085% A at 030 min, 85100% A at 3040 min) gave excellent
separation of the target compounds, when the ow rate, column
temperature and detection wavelength were at 1.0 mL min1 , 35 C,
and 254 nm.
3.4. Identication of the separated peaks
The chemical structures of two new prenylated avonoids were
elucidated on the basis of spectroscopic evidences (HR-ESI-MS, 1 HNMR, 13 C-NMR, HSQC, HMBC). The data were given in Table 2.
Compound 1 was obtained as a yellow power and possessed a
molecular formula C26 H26 O8 with 14 of unsaturation, as revealed
from its HR-ESI-MS analysis (m/z 467.1702 [M + H]+ ). The 1 H-NMR
spectrum (Table 2 or Fig. 5) showed signals of two systems of aromatic protons including a single at 6.28 (1H, s) assigned to the A
ring, and two ortho-coupled doublets at 6.97 (1H, d, J = 8.4 Hz) and
6.80 (1H, d, J = 8.4 Hz) assigned to the B ring, indicating the presence
of one pentasubstituted and 1,2,3,4-tetrasubstituted benzene ring.
The 1 H-NMR spectrum contained one aromatic methoxyl group
at 3.58 (3H, s), and one chelated hydroxyl group at 12.62 (1H,
s). The presence of one 2-hydroxy-3-methyl-3-butenyl group was
based on the signals of one exomethylene protons at 4.54 (1H,
s), 4.52 (1H, s), one tertiary methyl protons at 1.52 (3H, s), one
oxymethine protons at 4.13 (1H, t, J = 6.9 Hz), and one methylene protons at 2.73 (2H, d, J = 6.9 Hz). One 2,2-dimethyl-pyran
ring was deduced by a series of signals consisting of one pair of
cis-coupled olenic doublets at 6.29 (1H, d, J = 10.0 Hz), 5.80 (1H,
d, J = 10.0 Hz), and two tertiary methyl groups at 1.40 (3H, s),
1.39 (3H, s). The 13 C-NMR spectrum (Table 2 or Fig. 6) revealed a
skeleton of avone including fourteen olenic carbons, one carbonyl group at 178.6, besides one methoxyl group at 60.4, one
2-hydroxy-3-methyl-3-butenyl group at 29.2, 73.8, 148.3, 110.1,

17.5, and one 2,2-dimethylpyran ring at 120.6, 132.3, 75.8, 27.5,

27.3. These spectrascopic data indicated that compound 1 was a
prenylated avone derivative. The 2-hydroxy-3-methyl-3-butenyl
was further conrmed by the HMBC correlations of exomethylene
proton signals 4.52 (1H, s, H-4 ), 4.54 (1H, s, H-4 ) and methylene
proton signals at 2.73 (2H, d, J = 6.9 Hz, H-1 ) with C-5 ( 17.5),
C-2 ( 73.8), and C-3 ( 148.3), respectively. The location of
2-hydroxy-3-methyl-3-butenyl group was assigned to C-8 from
the HMBC correlations of 2.73 (2H, d, J = 6.9 Hz, H-1 ) with C-7
( 162.7), C-8 ( 104.4), and C-9 ( 155.2). The HMBC spectrum
also showed the correlations of the olenic proton 6.29 (1H, d,
J = 10.0 Hz, H-1 ) with C-1 ( 118.0), C-2 ( 121.2), and C-3 (
140.5), indicating that the 2,2-dimethylpyran ring was attached
to B ring by C-2 and C-3 . The position of the hydroxyl group at
C-4 was deduced by the long-range correlation of the aromatic
proton 6.97 (1H, d, J = 8.4 Hz, H-6 ) with the carbon signal at C-2
( 157.1) in the HMBC spectrum. The methoxyl group was located
at C-3, based on the HMBC correlation between methoxyl group
protons at 3.58 (3H, s) and C-3 ( 139.0). Thus, compound 1 was
established as 6 ,6 -dimethylpyran(2 ,3 :2 ,3 )-8-(2-hydroxy3-methyl-3-butenyl)-5,7,4 -trihydroxyl-3-methoxyavone,
and
named sinoavonoid A.
Compound 2 was obtained as a yellow power and possessed a
molecular formula C26 H28 O7 with 13 of unsaturation, as revealed
from its HR-ESI-MS analysis (m/z 453.1910 [M+H]+ ). The 1 H-NMR
spectrum (Table 2 or Fig. 7) showed signals of two systems of aromatic protons including a single at 6.12 (1H, s) assigned to the A
ring, and two ortho-coupled doublets at 6.77 (1H, d, J = 8.3 Hz), and
6.76 (1H, d, J = 8.3 Hz) assigned to the B ring, indicating the presence
of one pentasubstituted and 1,2,3,4-tetrasubstituted benzene ring.
The 1 H-NMR spectrum displayed one aromatic methoxy group at
3.57 (3H, s), and one chelated hydroxyl group at 12.45 (1H, s). The
existence of one prenyl group was deduced by one olenic proton at
5.01 (1H, m), two tertiary methyl groups at 1.43 (3H, s), 1.26 (3H,
s), and one methylene group at 3.28 (2H, d, J = 6.8 Hz). One 2,2dimethyl-2H-pyran ring was based on a series of signals consisting

196

Y. Sun et al. / J. Chromatogr. B 969 (2014) 190198

Fig. 5.

Fig. 6.

13

H-NMR spectrum of sinoavonoid A.

C-NMR spectrum of sinoavonoid A.

Y. Sun et al. / J. Chromatogr. B 969 (2014) 190198

Fig. 7.

Fig. 8.

13

H-NMR spectrum of sinoavonoid B.

C-NMR spectrum of sinoavonoid B.

197

198

Y. Sun et al. / J. Chromatogr. B 969 (2014) 190198

of two methylene groups at 2.59 (2H, t, J = 6.7 Hz), 1.78 (2H, t,

J = 6.7 Hz), two tertiary methyl groups at 1.29 (6H, s). The 13 C-NMR
spectrum (Table 2 or Fig. 8) revealed a skeleton of avone including
14 aromatic carbons, one carbonyl group at 178.6, along with one
methoxyl group at 60.2, one prenyl group at 26.2, 123.5, 130.5,
17.5, 25.5, one 2,2-dimethyldihydropyran ring at 16.0, 31.3, 76.6,
26.6 (2). These spectrascopic data indicated that compound 2 was
a prenylated avone derivative. The location of the prenyl group
was assigned to C-2 from the HMBC correlation of the methylene
protons 3.28 (2H, d, J = 6.8 Hz, H-1 ) with C-1 ( 121.4), C-2 (
121.8), and C-3 ( 143.7). The locations of the hydroxyl groups at C3 and C-4 were deduced by the long-range correlation of aromatic
proton 6.77 (1H, d, J = 8.3 Hz, H-6 ) with C-2 ( 159.8) in the HMBC
spectrum. The HMBC spectrum also showed the correlations of the
methylene protons 2.59 (2H, t, J = 6.7 Hz, H-1 ) with C-7 ( 159.1),
C-8 ( 100.1), and C-9 ( 154.3), indicating that the 2,2-dimethyl2H-pyran ring was attached to C-7 and C-8. The methoxy group was
located at C-3, based on the HMBC correlation between methoxy
group protons at 3.57 (3H, s) and C-3 ( 139.5). Thus, compound
2 was deduced as 6 ,6 -dimethyldihydropyran(2 ,3 :7,8)-2 -(,dimethylallyl)-5,3 ,4 -trihydroxyl-3- methoxyavone, and named
sinoavonoid B.
4. Conclusions
The perennial plant S. emodi (Berberiaceae) is well known as an
ethnic medicine. Previous literatures have indicated that avonoids
are mainly responsible for the biological activities of the plant
[2,10]. For the rst time, SGCC and HSCCC were developed successfully to separate and purify two new prenylated avonoids
from S. emodi. By the developed method, 17.3 mg of sinoavonoid
A and 25.9 mg of sinoavonoid B were obtained with the purities
of over 98%. Two new prenylated avonoids can be isolated on a
large scale, which may then be used for bioactive research or quality control of traditional Chinese medicine. Many pharmaceutical
researches are often hampered by the problems that the interesting natural products are trace in a complex biological organism.
It is crucial to develop the practical methods for separation and
purication of minor natural products with high purity and potential value. Overall results of our study showed that it is an efcient,
rapid, and economical technique for the extraction, separation, and
purication of trace prenylated avonoids from S. emodi.
Acknowledgements
The authors wish to thank Professor Chengming Dong (Henan
University of Traditional Chinese Medicine, Zhengzhou, Henan,

Peoples Republic of China.) for identication of the plant material.

This work was supported by the National Natural Science Foundation of China (No. 31300284), and Doctoral Science Foundation of
Henan University of Traditional Chinese Medicine (No. BSJJ201113).
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.jchromb.
2014.08.017.
References
[1] C.Q. Zhao, W. Cao, A. Nagatsu, Y. Ogihara, Chem. Pharm. Bull. 49 (2001) 1474.
[2] Y. Kong, J.J. Xiao, S.C. Meng, X.M. Dong, Y.W. Ge, R.F. Wang, M.Y. Shang, S.Q. Cai,
Fitoterapia 81 (2010) 367.
[3] C.Q. Zhao, Y.Y. Zhu, S.Y. Chen, Y. Ogihara, Chin. Chem. Lett. 22 (2011) 181.
[4] X.L. Shi, X.W. Li, J.B. Liu, H.Y. Zhou, H.Q. Zhang, Y.R. Jin, Chromatographia 72
(2010) 713.
[5] C.Q. Zhao, J. Huang, A. Nagatsu, Y. Ogihara, Chem. Pharm. Bull. 49 (2001) 773.
[6] C.Q. Zhao, A. Nagatsu, K. Hatano, N. Shirai, S. Kato, Y. Ogihara, Chem. Pharm.
Bull. 51 (2003) 255.
[7] Y.J. Sun, Z.L. Li, H. Chen, X.Q. Liu, W. Zhou, H.M. Hua, Bioorg. Med. Chem. Lett.
21 (2011) 3794.
[8] Y.J. Sun, Z.L. Li, H. Chen, X.Q. Liu, W. Zhou, H.M. Hua, Planta Med. 78 (2012)
480.
[9] Y.J. Sun, Z.L. Li, H. Chen, W. Zhou, H.M. Hua, J. Chin. Med. Mater. 35 (2012)
1607.
[10] M.Y. Shang, Q.H., Wang, J.J., Xiao, Y.H., Shang, Y., Kong, S.Q. Cai, China patent
CN 102382092 A (2012).
[11] Y.J. Sun, W. Zhou, H. Chen, Z.L. Li, H.M. Hua, Chin. Tradit. Herb. Drugs 43 (2012)
226.
[12] Y.J. Sun, B.Y. Ji, W.S. Feng, H. Chen, Y.Z. Wang, J.H. Gong, Z.Y. Hao, J. Chin. Med.
Mater. 36 (2013) 1537.
[13] I.A. Sutherland, D. Fisher, J. Chromatogr. 1216 (2009) 740.
[14] I.A. Sutherland, P. Hewitson, S. Ignatova, J. Chromatogr. 1216 (2009) 8787.
[15] P. Li, Y.L. Wang, L. Tang, Z.Y. Guo, C.L. Long, Chin. Tradit. Herb. Drugs 44 (2013)
2498.
[16] C.Y. Yang, Y. Yang, H.A. Aisa, X.L. Xin, H.R. Ma, A. Yili, Y.X. Zhao, J. Sep. Sci. 35
(2012) 977.
[17] S.S. Xu, Y. Sun, F. Jing, W.J. Duan, J.H. Du, X. Wang, Chin. J. Chromatogr. 29 (2011)
1244.
[18] H. Jiang, Y.R. Cheng, W. Zheng, Chin. J. Mar. Drugs 26 (2007) 8.
[19] X. Han, X.F. Ma, T.Y. Zhang, Y.B. Zhang, Q.H. Liu, Y. Ito, J. Chromatogr. 1151
(2007) 180.
[20] S.Y. Shi, H.H. Zhou, Y.P. Zhang, Y. Zhao, K.L. Huang, S.Q. Liu, J. Chromatogr. Sci.
47 (2009) 349.
[21] P. Wang, Y.L. Liu, T. Chen, W.H. Xu, J.M. You, Y.J. Liu, Y.L. Li, Phytochem. Anal.
24 (2013) 603.
[22] K.Y. Zhang, X.G. Wang, J.H. Huang, Y.F. Liu, J. Chromatogr. 1220 (2012) 108.
[23] E. Melwita, Y.A. Tsigie, S. Ismadji, Y.H. Ju, J. Liq, J. Liq. Chromatogr. Relat. Technol.
34 (2011) 2462.
[24] S. Gunawan, S. Ismadji, Y.H. Ju, J. Chin. Inst. Chem. Eng. 39 (2008) 625.
[25] S. He, H.Q. Wang, X.J. Yan, P. Zhu, J.J. Chen, R. Yang, J. Chromatogr. 1272 (2013)
15.
[26] H. Oka, K.I. Harada, Y. Ito, J. Chromatogr. 812 (1998) 35.