Journal of Chromatographic Science, 2016, Vol. 54, No.

2, 148–157
doi: 10.1093/chromsci/bmv119
Advance Access Publication Date: 26 August 2015
Article

Article

Identification and Characterization of the Major

Chemical Constituents in Fructus Akebiae by

Downloaded from https://academic.oup.com/chromsci/article-abstract/54/2/148/2754759 by guest on 30 October 2018

High-Performance Liquid Chromatography
Coupled with Electrospray Ionization-
Quadrupole-Time-of-Flight Mass Spectrometry
Yun Ling*, Qing Zhang, Dan-dan Zhu, Fei Chen, Xiu-hua Kong,
and Liang Liao*
Department of Pharmaceutical and Life Sciences, Jiujiang University, Jiujiang 332005, People’s Republic of China
*Author to whom correspondence should be addressed. Email: lymanuxt@126.com (Y. Ling); Email: 13970209819@163.com
(L. Liao)
Received 3 January 2014; Revised 23 June 2015

Abstract
Fructus Akebiae (FA), the dry fruit of Akebia quinata (THUNB.) DECNE., possesses potent antidepres-
sant properties. Owing to the structural complexity, high polarity and thermal lability in plants, it is
difficult and time-consuming to analyze the major chemical constituents by traditional strategies that
involve extraction, isolation, purification and identification by chemical manipulations and spectro-
scopic methods. In this study, a high-performance liquid chromatography coupled with electrospray
ionization-quadrupole-time-of-flight mass spectrometry (HPLC–ESI–Q-TOF–MS-MS) method was
established for quickly identifying the chemical constituents in the extract of Fructus Akebiae. The
main saponin components in the extract of Fructus Akebiae were detected with the HPLC–ESI–Q-
TOF–MS-MS in negative-ion mode. These components were further analyzed by MS2 spectra, and
compared with the corresponding reference substances and literature data. Nineteen saponins in the
extract of Fructus Akebiae were well separated in one run. The new method is accurate and rapid. It
can be used to identify the main chemical constituents in the extract of Fructus Akebiae and can be
suitable for the quality control of Fructus Akebiae.

Introduction
The use of traditional Chinese medicine (TCM) has made rapid strides
in the past decade due to its potential benefits to human health. TCM
offers citizen herbs composed of complicated components which were
responsible for their efficacies. Therefore, comprehensive identifica-
tion of the chemical components in TCM is a key issue for therapeutic
effects and safety understanding in treating disease process. However,
due to difficulties regarding isolation and preparation, it is a challeng-
ing task to base all studies on the conventional phytochemical means,
especially for those high polarity, nonchromophorous and low abun-
dant compounds. Therefore, there is an urgent need to establish a
feasible and sensitive method for full-scale qualitative analysis of the
major chemical constituents in TCM.
Fructus Akebiae (FA), the dry fruit of Akebia quinata (THUNB.)
DECNE., is commonly used as an antineoplastic and diuretic agent
and officially documented in the Chinese Pharmacopoeia under the
name “Yu Zhi Zi”. Pharmacological studies have demonstrated that
the extract of FA possesses potent antidepressant properties (1). Triter-
pene glycosides with hederagenin are thought to be the major active
chemical constituents (2, 3). However, the methods for isolation and pu-
rification usually require several column chromatographies, which result
in low recovery. Moreover, the structural characterization and

© The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com 148
Identification and Characterization of the Major Chemical Constituents in FA 149

Downloaded from https://academic.oup.com/chromsci/article-abstract/54/2/148/2754759 by guest on 30 October 2018

Figure 1. MS TIC chromatogram of FA extract.

Figure 2. Chemical structures of compounds identified in FA.

identification of triterpene glycosides are also difficult because the signals Recently, high-performance liquid chromatography coupled with
of the sugar moieties often overlap. Therefore, developing a method to electrospray ionization-quadrupole-time-of-flight mass spectrometry
screen and identify these components in plant extracts is critically impor- (HPLC–ESI-Q-TOF–MS-MS) has demonstrated to be a valuable
tant for understanding of their health benefits and pharmacology. tool for detecting analytes present at trace levels and for screening
 Downloaded from https://academic.oup.com/chromsci/article-abstract/54/2/148/2754759 by guest on 30 October 2018

150
Table I. ESI–TOF–MS Data of Compounds Identified from FA

Peak tR Formula Identification Ion mode Mass (m/z) Fragment ions (m/z) Ref.
(min)

1a 5.750 C16H18O9 Scopolin [2M−H]− 707.1882 353.0848 [M−H]−, 191.0530 [M−H–Glc]− (7)
2 10.105 C59H96O27 3β-[(O-β-D-glucopyranosyl-(1 → 3)- [M−H]− 1235.6159 765.4450 [M−H–Rha–2Glc]−, 469.1532 [2Glc (8)
α-L-arabinopyranosyl)oxy]-23-hydroxyolean- +Rha–H]−
12-en-28-oic acid O-α-L-rhamnopyranosyl-(1 → 4)-
O-β-D-glucopyranosyl-(1 → 6)-β-D-glucopyranosyl ester
3 11.641 C53H86O22 Akemisaponin A [M+HCOO−]− 1119.5312 1073.5176 [M−H]−, 795.4117 [M−H–Xyl–Rha]−, (9)
601.1972 [Xyl+Rha+2Glc–H]−
4 13.976 C53H86O23 2α,3β,23-trihydroxyolean-12-en-28-oic acid O-β-D-xylopyranosyl- [M+HCOO ] 1135.5628 1089.5479 [M−H]−, 601.1966 [Xyl+Rha+2Glc–H]− (9)
− −

(1 → 3)-O-α-L-rhamnopyranosyl-(1 → 4)-O-β-D-glucopyranosyl-
(1 → 6)-β-D-glucopyranosyl ester
5 14.748 C48H78O19 Scheffoleoside A [M+HCOO−]− 1003.5186 957.5048 [M−H]−, 487.3048 [M−H–2Glc–Rha]−, (10)
469.1549 [2Glc+Rha–H]−
6b 16.073 C57H90O26 3β-[(O-β-D- xylopyranosyl -(1 → 2)-O-α-L-arabinopyranosyl)oxy]- [M−H]− 1189.5740 719.3993 [M−H–Rha–2Glc]−, 469.1522 NF
23-hydroxyolean-12,20(29)-dien-28-oic acid O-α-L-rhamnopyranosyl- [Rha+2Glc–H]−, 323.0946 [2Glc–H]−
(1 → 4)-O-β-D-glucopyranosyl-(1 → 6)-β-D-glucopyranosyl ester
7 17.134 C65H106O31 3β-[(O-β-D-glucopyranosyl-(1 → 3)-O-[α-L-rhamnopyranosyl- [M+Cl]− 1417.6400 1381.6628 [M−H]−, 911.4989 [M−H–Rha–2Glc]−, (11)
(1 → 2)]- α-L-arabinopyranosyl)oxy]-23-hydroxyolean-12-en-28-oic acid 749.4482 [M−H–Rha–3Glc]−
O-α-L-rhamnopyranosyl-(1 → 4)-O-β-D-glucopyranosyl-(1 → 6)-
β-D-glucopyranosyl ester
8 17.847 C59H96O26 23-hydroxy-3β-[(O-α-L-rhamnopyranosyl-(1 → 2)- [M+Cl]− 1255.6000 1219.6089 [M−H]−, 749.4464 [M−H–Rha–2Glc]−, (8)
α-L-arabinopyranosyl)oxy]olean-12-en-28-oic acid O-α-L- 469.1533 [Rha+2Glc–H]−
rhamnopyranosyl-(1 → 4)-O-β-D-glucopyranosyl-(1 → 6)-
β-D-glucopyranosyl ester
9 17.906 C64H104O30 3β-[(O-β-D-glucopyranosyl-(1 → 3)-O-[β-D-xylopyranosyl-(1 → 2)]- [M−H]− 1351.6675 881.4907 [M−H–Rha–2Glc]−, 469.1531 (8)
α-L-arabinopyranosyl)oxy]olean-12-en-28-oic acid O-α-L- [2Glc+Rha–H]−
rhamnopyranosyl-(1 → 4)-O-β-Dglucopyranosyl-(1 → 6)- β-D-glucopyranosyl ester
10b 18.365 C58H94O26 23-hydroxy-3β-[(O-β-D- xylopyranosyl-(1 → 2)-O-α-L- [M−H]− 1205.5947 735.4292 [M−H–Rha–2Glc]−, 469.1527 NF
arabinopyranosyl)oxy] olean-12-en-28-oic acid O-α-L-rhamnopyranosyl- [Rha+2Glc–H]−
(1 → 4)-O-β-D-glucopyranosyl-(1 → 6)- β-D-glucopyranosyl ester
11a 19.019 C53H86O22 Dipsacoside B [M+HCOO−]− 1119.5691 1073.5530 [M−H]−, 749.4482 [M−H–2Glc]−, (12)
603.3931 [M−H–2Glc–Rha]−,
471.3558 [M−H–2Glc–Rha–Ara]−, 323.0960
[2Glc–H]−
− −
12a 19.375 C47H76O18 Asperosaponin VI [M+HCOO ] 973.5098 927.4949 [M−H]−, 603.3877 [M−H–2Glc]−, (13)
323.0953 [2Glc–H]−
13b 20.139 C58H92O25 3β-[(O-β-D- xylopyranosyl -(1 → 2)-O-α-L-arabinopyranosyl)oxy]- [M+Cl]− 1223.5739 1187.5810 [M−H]−, 717.4810 [M−H–2Glc–Rha]−, NF
12,20(29)-dien-28-oic acid O-α-L-rhamnopyranosyl-(1 → 4)- 469.1501 [2Glc+Rha–H]−
O-β-D-glucopyranosyl-(1 → 6)-β-D-glucopyranosyl ester
14b 20.496 C53H86O21 3β-[(O-α-L-rhamnopyranosyl-(1 → 2)-α-L-arabinopyranosyl)oxy] olean- [M+Cl]− 1093.5084 1057.5194 [M−H]−, 733.4178 [M−H–2Glc]− NF
12-en-28-oic acid O-β-D-glucopyranosyl-(1 → 6)- β-D-glucopyranosyl ester
15 20.504 C57H90O25 Mutongsaponin E [M+Cl]− 1209.5463 1173.5665 [M−H]−, 703.3998 [M−H–2Glc–Rha]−, (10)

Ling et al.
469.1513 [2Glc+Rha–H]−
 Identification and Characterization of the Major Chemical Constituents in FA 151

large numbers of samples based on accurate mass data, and character-

(11)

(10)
1401.6467 1365.6599 [M−H]−, 895.5037 [M−H–2Glc–Rha]−, (8)

(9)

(8)
istic mass fragments reported in previous literature reports (4–6).
733.4412 [M−H–3Glc–Rha]−, 587.3097 [M−H–

Q-TOF type instruments have better sensitivity and MS accuracy

when performing exact mass neutral loss scan together with MS and
987.5260 941.5184 [M−H]−, 469.1537 [Rha+2Glc–H]−

MS-MS spectra. Furthermore, molecules belonging to a particular

801.4256 765.4414 [M−H]−, 603.4013 [M−H–Glc]−

957.5147 911.4834 [M−H]−, 749.4490 [M−H–Glc]−

class can often be confirmed by the observation of diagnostic frag-

1203.6162 733.4531 [M−H–2Glc–Rha]−, 469.1522

ments and/or neutral losses in MS-MS experiments. The candidates
can be further reduced and focused if exact m/z values, rather than
nominal values, are assigned to characteristic fragments and/or neutral
loss of a certain class of chemicals.
In this study, we developed a simple, rapid and sensitive HPLC–
ESI–Q-TOF–MS-MS method for systematic analysis of the chemical

Downloaded from https://academic.oup.com/chromsci/article-abstract/54/2/148/2754759 by guest on 30 October 2018

[2Glc+Rha–H]−

profile of FA. The results of the present study will assist in careful un-
3Glc–2Rha]−

derstanding of the known and unknown compounds in FA, which

contains herb extract.

Experimental
Materials and reagents
FA was purchased from Bozhou, Anhui Province, PR China, in
September 2013. Scopolin, dipsacoside B and asperosaponin VI
[M+HCOO−]−

[M+HCOO−]−

were purchased from Nanjing Zelang Medical Technology Co. Ltd

(Nanjing, China). The purity of all reference compounds was deter-
[M+Cl]−

[M+Cl]−

[M−H]−

mined to be over 98% by HPLC analysis.

HPLC grade acetonitrile, methanol and formic acid were pur-
chased from Dikma Company (Dikma, USA). Water was deionized
and double-distilled. All other analytical grade reagents were obtained
L-rhamnopyranosyl-(1 → 4)-O-β-D-glucopyranosyl-(1 → 6)- β-D-glucopyranosyl ester

from Sinopharm Chemical Reagent Co. Ltd (Shanghai, China).

22.389 C59H96O25 3β-[(O-α-L-rhamnopyranosyl-(1 → 2)-α-L-arabinopyranosyl)oxy]olean-12-en-28-oic

Sample preparation
An aliquot of 1.0 g FA drug powder was accurately weighed and
acid O-α-L-rhamnopyranosyl-(1 → 4)-O-β-D-glucopyranosyl-(1 → 6)-
21.676 C47H74O19 2α,3β,23-trihydroxy-30-norolean-12,20(29)-dien-28-oic acid O-α-L-

extracted with 10 mL of MeOH : water (70 : 30, v/v) by ultrasonica-

21.260 C65H106O30 3β-[(O-β-D-glucopyranosyl-(1 → 3)-O-[α-L-rhamnopyranosyl-(1 →

(1 → 3)-O-]-α-L-arabinopyranosyl)oxy]-olean-12-en-28-oic acid

tion for 30 min. The solution was centrifuged at 12,000 rpm for
21.727 C41H66O13 3β-[(O-β-D-glucopyranosyl-(1 → 3)-α-L-arabinopyranosyl)oxy]-
2)]- α-L-arabinopyranosyl)oxy]olean-12-en-28-oic acid O-α-

10 min, and the supernatant was filtered through a 0.22-μm filter.

A volume of 2 µL was injected into the HPLC–Q-TOF–MS system
23.306 C47H76O17 3-β-[(β-D-glucopyranosyl-(1 → 2)-O-[β-D-glucopyranosyl-
rhamnopyranosyl-(1 → 4)-O-β-D-glucopyranosyl-(1 →

for the analysis.

HPLC conditions
The chromatographic analysis was performed on an Agilent 1260
Series (Agilent, Santa Clara, CA, USA) LC system equipped with a bi-
23-hydroxyolean-12-en-28-oic acid

nary pump, an online degasser, a well-plate autosampler and a ther-

Ref.: literature referred to structural identification; NF, not found.

mostatically controlled column compartment. The columns were

6)-β-D-glucopyranosyl ester

maintained at 30°C. The separation was carried out on an Agilent po-

β-D-glucopyranosyl ester

roshell 120 EC-C18 column (2.7 × 100 mm, i.d., 2.7 µm), preceded by
a C18 guard column (4.00 × 2.00 mm; Agilent). The mobile phase
consisted of water containing 0.1% (v/v) formic acid (A) and acetoni-
trile (B). A gradient program was used as follows: 0–5 min, 5–20% B;
5–25 min, 20–40% B; 25–30 min, 40–90% B. The composition was
then held at 90% B for 5 min and returned to initial conditions and
maintained 10 min for equilibration. The flow rate was 0.35 mL/min
and sample injection volume was 2 µL.
Identified with standards.

Mass spectrometry conditions

New compounds.

Mass spectrometry was performed using an Agilent 6530 Q-TOF

mass spectrometer (Agilent) equipped with an ESI interface, and
was operated in negative-ion mode with parameters set as follows:
capillary voltage, 3,500 V; fragmentor, 170 V; skimmer, 65 V; OCT
1 RF Vpp, 750 V; pressure of nebulizer, 35 psi; drying gas tempera-
b
a
16

17

18

19

20

ture, 300°C; sheath gas temperature, 350°C. Nitrogen was used as

152
sheath and drying gas at a flow rate of 8.0 and 11.0 L/min, respective-
ly. The collision energy was set 45 V and the mass range recorded m/z
100–2,000. An external calibration solution (Agilent calibration solu-
tion A) was continuously sprayed into the ESI source of the Q-TOF
system, employing the ions with m/z 112.9855 (trifluoroacetic acid
anion) and 1033.9881 [HP-0921 (trifluoroacetic acid adduct)] to re-
calibrate the mass axis, ensuring mass accuracy and reproducibility
throughout the chromatographic run. All operations, acquisition
and analysis of data were monitored by the Agilent HPLC–ESI–Q-
TOF–MS-MS MassHunter Acquisition Software Version A.01.00
(Agilent Technologies) and operated under the MassHunter Acquisi-
tion Software Version B.06.00 (Agilent Technologies).
Results
Optimization of LC and MS conditions
To achieve a good multi-ingredients fingerprint analysis, effects of
the experimental conditions including the mobile phase system,
the parameters of flow rate of gas, gas pressure, spray voltage, capil-
lary temperature and voltage of entrance potential on analysis of FA
were optimized. On the Q-TOF–MS, both positive and negative ioni-
zation modes were investigated. The results indicated that the
negative-ion electrospray mode provided higher sensitivity, cleaner
mass spectra and lower background for components of FA vs. ESI
positive ionization.
Analysis of the chemical constituents of FA
by HPLC–ESI–Q-TOF–MS-MS
Identification of the bioactive ingredients present in FA was per-
formed based on the mass spectral information obtained by
Q-TOF–MS and Q-TOF–MS-MS. As a result, a total of 20 com-
pounds were characterized from the FA sample, 16 of which were
identified by matching the empirical molecular formula with that of
the published components or available reference compounds, and
were further elucidated by lower energy collision-induced dissociation
mass spectra. In addition, four new potential compounds were suc-
cessfully characterized based on the difference of calculated m/z and
fragment ions compared with that of known compounds. The MS
total ion current (TIC) chromatogram of FA extract in negative-ion
mode is shown in Figure 1. The chemical structures of the identified
components are shown in Figure 2 and the details of identified com-
ponents are summarized in Table I.
Peak 1 produced an [2M−H]− adduct ion at m/z 701.1882, which
indicated the molecular formula of C16H18O9. In the MS-MS spec-
trum, the major fragment at m/z 191.0530 was observed indicating
a loss of Glc (162 Da) from the deprotonated molecule [M−H]− of
Peak 1. In comparison with the standard compound, Peak 1 was am-
biguously identified as scopolin. Peak 2 gave an [M−H]− ion at m/z
1235.6159, corresponding to the molecular formula of C59H96O27.
In the MS-MS spectrum (Figure 3A), the ions at m/z 765.4450 and
469.1532 were observed, indicating cleavage of the ester bond at
C-28, as shown in Figure 3B. By searching the known compounds re-
ported in previous literature (8), Peak 2 was tentatively identified as
3β-[(O-β-D-glucopyranosyl-(1 → 3)-α-L-arabinopyranosyl) oxy]-23-
hydroxyolean-12-en-28-oic acid O-α-L-rhamnopyranosyl-(1 →
4)-O-β-D-glucopyranosyl-(1 → 6)-β-D-glucopyranosyl ester. Peak 3,
with an [M+HCOO−]− ion at m/z 1119.5312, produced two charac-
teristic fragment ions at m/z 795.4117 and 601.1972 in the MS2 spec-
trum. The fragment ion at m/z 795.4117 is consistent with loss of
Xyl + Rha units from the parent nucleus, whereas the fragment ion
Ling et al.
at m/z 601.1972 is consistent with cleavage of the ester bond at
C-28 from the parent nucleus. Therefore, Peak 3 was tentatively iden-
tified as akemisaponin A by comparison with the known compounds
reported in literature (9). Peak 11 gave an adducted ion of
[M+HCOO−]− at m/z 1119.5691 and produced fragment ions at m/z
1073.5530 [M−H]−, 749.4482 [M−H–2Glc]−, 603.3931 [M−H–
2Glc–Rha]−, 471.3558 [M−H–2Glc–Rha–Ara]− and 323.0960
[2Glc–H]− in the MS-MS spectrum. Therefore, Peak 11 unambiguous-
ly was identified as dipsacoside B and this identity was confirmed by
matching the retention time, empirical molecular and fragment ions
with those of reference standard. Peak 12 displayed an [M+HCOO−]−
ion at m/z 973.5098 and possessed similar fragmentation pathways
Downloaded from https://academic.oup.com/chromsci/article-abstract/54/2/148/2754759 by guest on 30 October 2018
as Peak 2. On the basis of the available standard, Peak 12 was ambig-
uously identified as asperosaponin VI. Peak 15 produced a [M+Cl]−
ion with high intensity at m/z 1209.5463 (C57H90O25Cl) and
[M−H]− at m/z 1173.5665. The moderately abundant product ions
at m/z 703.3998 and 469.1513 were obtained in the MS-MS spec-
trum, as shown in Figure 4A. It shared a similar pathway to Peak 2,
as shown in Figure 4B, and thus was tentatively assigned as mutong-
saponin E according to the literature (10). Similarly, based on the frag-
ment ions and retention times, and comparison with known
compounds reported in literatures, Peaks 4, 5, 7, 8, 9, 16, 17, 18,
19 and 20 were identified as 2α, 3β, 23-trihydroxyolean-12-en-28-
oic acid O-β-D-xylopyranosyl-(1 → 3)-O-α-L-rhamnopyranosyl-(1 →
4)-O-β-D-glucopyranosyl-(1 → 6)-β-D-glucopyranosyl ester, scheffo
leoside A, 3β-[(O-β-D-glucopyranosyl-(1 → 3)-O-[α-L-rhamnopyra
nosyl-(1 → 2)]-α-L-arabinopyranosyl)oxy]-23-hydroxyolean-12-en-
28-oic acid O-α-L-rhamnopyranosyl-(1 → 4)-O-β-D-glucopyranosyl-
(1 → 6)-β-D-glucopyranosyl ester, 23-hydroxy-3β-[(O-α-L-rhamno
pyranosyl-(1 → 2)- α-L-arabinopyranosyl) oxy]olean-12-en-28-oic
acid O-α-L-rhamnopyranosyl-(1 → 4)-O-β-D-glucopyranosyl-(1 → 6)-
β-D-glucopyranosyl ester, 3β-[(O-β-D-glucopyranosyl-(1 → 3)-O-[β-D-
xylopyranosyl-(1 → 2)]- α-L-arabinopyranosyl)oxy]olean-12-en-28-oic
acid O-α-L-rhamnopyranosyl-(1 → 4)-O-β-D-glucopyranosyl-(1 → 6)-
β-D-glucopyranosyl ester, 3β-[(O-β-D-glucopyranosyl-(1 → 3)-O-[α-L-
rhamnopyranosyl-(1 → 2)]-α-L-arabinopyranosyl) oxy] olean-12-en-
28-oic acid O-α-L-rhamnopyranosyl-(1 → 4)-O-β-D-glucopyranosyl-(1 -
→ 6)- β-D-glucopyranosyl ester, 2α,3β,23-trihydroxy-30-norolean-12,
20(29)-dien-28-oic acid O-α-L-rhamnopyranosyl-(1 → 4)-O-β-D-
glucopyranosyl-(1 → 6)-β-D-glucopyranosyl ester, 3β-[(O-β-D-glucopyr-
anosyl-(1 → 3)-α-L-arabinopyranosyl) oxy]-23-hydroxyolean-12-en-
28-oic acid, 3β-[(O-α-L-rhamnopyranosyl-(1 → 2)-α-L-arabinopyrano-
syl) oxy] olean-12-en-28-oic acid O-α-L-rhamnopyranosyl-(1 → 4)-
O-β-D-glucopyranosyl-(1 → 6)- β-D-glucopyranosyl ester, 3-β-[(β-D-
glucopyranosyl-(1 → 2)-O-[β-D-glucopyranosyl-(1 → 3)-O-]-α-L-arabi-
nopyranosyl)oxy]-olean-12-en-28-oic acid, respectively. Peak 6
displayed an [M−H]− ion at m/z 1189.5740 (C57H89O26), a mass
increase of vs. Peak 15, indicating the addition of oxygen to the C-23
position of Compound 15 to form a hydroxyl group. In the MS-MS
spectrum, cleavage of the ester bond at C-28 that led to the formation
of two characteristic signals at m/z 719.3993 and 469.1522 was
observed, as shown in Figure 5A. In addition, a low abundant ion at
m/z 323.0946 was yielded by the neutral loss of Rha (146 Da) unit
from the fragment ion m/z 469.1522. A possible fragmentation pathway
is proposed in Figure 5B. Thus, Peak 6 was tentatively identified as
23-hydroxy-3β-[(O-β-D- xylopyranosyl-(1 → 2)-O-α-L-arabinopyrano-
syl)oxy] olean-12-en-28-oic acid O-α-L-rhamnopyranosyl-(1 → 4)-
O-β-D-glucopyranosyl-(1 → 6)-β-D-glucopyranosyl ester. Peak 10 pro-
duced an [M−H]− ion at m/z 1205.5947, which was 14 Da less than
that of Peak 8, indicating that the Rha unit at the C-3 sugar chain of
Peak 8 was replaced with the Xyl unit, as shown in Figure 6A. The
Identification and Characterization of the Major Chemical Constituents in FA 153

Downloaded from https://academic.oup.com/chromsci/article-abstract/54/2/148/2754759 by guest on 30 October 2018

Figure 3. MS-MS spectra and proposed fragmentation pathway of Compound 2.

MS-MS spectra gave prominent ions at m/z 735.4292 [M−H–Rha– oxy] olean-12-en-28-oic acid O-α-L-rhamnopyranosyl-(1 → 4)-O-β-D-
2Glc]− and 469.1527 [Rha+2Glc–H]− in the same pathways as glucopyranosyl-(1 → 6)-β-D-glucopyranosyl ester. Peak 13 displayed
that of Peak 8. Therefore, Peak 10 was tentatively identified as 23- an [M+Cl]− ion at m/z 1223.5739, which corresponded to a 14 Da in-
hydroxy-3β-[(O-β-D-xylopyranosyl-(1 → 2)-O-α-L-arabinopyranosyl) crease vs. Peak 15, indicating that the Xyl unit connecting to C-3
 Downloaded from https://academic.oup.com/chromsci/article-abstract/54/2/148/2754759 by guest on 30 October 2018
Ling et al.

Figure 4. MS-MS spectra and proposed fragmentation pathway of Compound 15.

154
Identification and Characterization of the Major Chemical Constituents in FA 155

Downloaded from https://academic.oup.com/chromsci/article-abstract/54/2/148/2754759 by guest on 30 October 2018

Figure 5. MS-MS spectra and proposed fragmentation pathway of Compound 6.

156
Figure 6. Proposed formation pathways of Compounds 10 (A), 13 (B) and 14 (C).
branched chain of Peak 15 was replaced with the Rha unit, as shown in
Figure 6B. The observation of fragment ions at m/z 717.4810 [M−H–
2Glc–Rha]− and 469.1501 [2Glc+Rha–H]− further confirmed that
changes should occur at the C-3 branched chain of Peak 13.
Therefore, the structure of Peak 13 was tentatively identified as 3β-
[(O-β-D-xylopyranosyl -(1 → 2)-O-α-L-arabinopyranosyl)oxy]-12,20
(29)-dien-28-oic acid O-α-L-rhamnopyranosyl-(1 → 4)-O-β-D-glucopyr-
anosyl-(1 → 6)-β-D-glucopyranosyl ester. Peak 14 exhibited an [M+Cl]−
ion at m/z 1093.5084 and showed a similar pathway to Compound 11.
Ling et al.
Downloaded from https://academic.oup.com/chromsci/article-abstract/54/2/148/2754759 by guest on 30 October 2018
The only difference between them is that the molecular weight of
Compound 14 is 14 Da less than that of Compound 11, indicating
that loss of an oxygen atom from the C-23 position of Peak 7 led to
the formation of chemical structure of Compound 14, as shown in Fig-
ure 6C. Hence, Peak 14 was tentatively identified as 3β-[(O-α-L-rhamno-
pyranosyl-(1 → 2)-α-L-arabinopyranosyl)oxy] olean-12-en-28-oic acid
O-β-D-glucopyranosyl-(1 → 6)-β-D-glucopyranosyl ester. To the best of
our knowledge, Compounds 6, 10, 13 and 14 were reported from FA
for the first time and were considered as potential new compounds.
Identification and Characterization of the Major Chemical Constituents in FA
Discussion
Compared with the HPLC–ESI–MS and RP-HPLC methods reported
in the previous literatures (14–16), Q-TOF–MS-MS with pore shell
column has good ability of separation and higher mass resolution, es-
pecially for Peaks 13–20. Based on the fragment information, it was
possible not only to detect 14 known triterpene glycosides in FA,
but also to predict the structures of four previously unreported triter-
pene glycosides in the extract. To our knowledge, this is the first report
of prediction of the structures of novel triterpene glycosides in a crude
plant extract by ESI–Q-TOF–MS-MS. The ability of ESI-Q-TOF–
MS-MS experiments to predict the structures of novel compounds pro-
vides a useful guide for the isolation of compounds and the use of
NMR or X-ray crystallography to confirm the structures of predicted
compounds. These findings suggested that HPLC–ESI–Q-TOF–
MS-MS method was useful in studies of known and unknown com-
pounds in crude plant extracts.
Conclusion
In this study, a rapid and accurate HPLC–ESI–Q-TOF–MS method
was established for identification of the major chemical constituents
in FA. Based on the exact mass, summarized fragmentation behaviors
and the retention time, a total of 20 ingredients were identified in the
crude extract of FA, 4 of which were observed for the first time. The
findings from the present study set a good example for identification of
the known and unknown chemical constituents in herb medicines, and
thereby it is possible to fulfill the safety, efficacy and stability require-
ments for characterizing modern TCM.
Acknowledgments
The authors thank the State Key Program of the National Natural Science Foun-
dation of China (grant nos 31460073 and 31260044) and the performance of
Scientific Planning Project of Jiangxi Province of China (nos: 20142BBF60055
and 20143ACF60010) for the financial support of this work.
Conflict of interest statement. The authors have declared no conflict of interest.
References
1. Zhou, D., Jin, H., Lin, H.B., Yang, X.M., Cheng, Y.F., Deng, F.J., et al.; An-
tidepressant effect of the extracts from Fructus Akebiae; Pharmacology
Biochemistry and Behavior, (2010); 94: 488–495.
2. Jin, Z.L., Gao, N., Zhou, D., Chi, M.G., Yang, X.M., Xu, J.P.; The ex-
tracts of Fructus Akebiae, a preparation containing 90% of the active
ingredient hederagenin: serotonin, norepinephrine and dopamine reup-
take inhibitor; Pharmacology Biochemistry and Behavior, (2012); 100
(3): 431–439.
157
3. Yang, X., Li, G., Chen, L., Zhang, C., Wan, X., Xu, J.; Quantitative deter-
mination of hederagenin in rat plasma and cerebrospinal fluid by ultrafast
liquid chromatography-tandem mass spectrometry method; Journal of
Chromatographic B Analytical Technologies in the Biomedical Life
Sciences, (2011); 879(21): 1973–1979.
4. Zhao, H.Y., Fan, M.X., Wu, X., Wang, H.J., Yang, J., Si, N., et al.;
Chemical profiling of the Chinese herb formula Xiao-Cheng-Qi Decoction
using liquid chromatography coupled with electrospray ionization
mass spectrometry; Journal of Chromatographic Science, (2013); 51:
273–285.
5. Zhao, X., Yang, D.H., Zhou, Q.L., Xu, F., Zhang, L., Liang, J., et al.;
Identification of metabolites in WZS-miniature pig urine after oral
administration of Danshen decoction by HPLC coupled with diode array
Downloaded from https://academic.oup.com/chromsci/article-abstract/54/2/148/2754759 by guest on 30 October 2018
detection with electrospray ionization tandem ion trap and time-of-
flight mass spectrometry; Biomedical Chromatography, (2013); 27:
720–735.
6. Liu, M.H., Tong, X., Wang, J.X., Zou, W., Cao, H., Su, W.W.; Rapid sep-
aration and identification of multiple constituents in traditional Chinese
medicine formula Shenqi Fuzheng Injection by ultra-fast liquid chromatog-
raphy combined with quadrupole-time-of-flight mass spectrometry; Journal
of Pharmaceutical and Biomedical Analysis, (2013); 74: 141–155.
7. Jiang, D., Shi, S.P., Cao, J.J., Gao, Q.P., Yu, P.F.; Triterpene saponins from
the fruits of Akebia quinata; Biochemical Systematics and Ecology, (2008);
36: 138–141.
8. Mimaki, Y., Kuroda, M., Yokosuka, A., Harada, H., Fukushima, M.,
Sashida, Y.; Triterpenes and triterpene saponins from the stems of Akebia
trifoliate; Chemical and Pharmaceutical Bulletin, (2003); 51(8): 960–965.
9. Iwanaga, S., Warashina, T., Miyase, T.; Triterpene saponins from the peri-
carps of Akebia trifoliate; Chemical and Pharmaceutical Bulletin, (2012);
60(10): 1264–1274.
10. Gao, H., Wang, Z.; Triterpenoid saponins and phenylethanoid glycosides
from stem of Akebia trifoliata var. australis; Phytochemistry, (2006); 67:
2697–2705.
11. Mimaki, Y., Doi, S., Kuroda, M., Yokosuka, A.; Triterpene glycosides from
the stems of Akebia quinata; Chemical and Pharmaceutical Bulletin
(Tokyo), (2007); 55(9): 1319–1324.
12. Pasi, S., Aligiannis, N., Pratsinis, H., Skaltsounis, A.L., Chinou, I.B.; Biolog-
ically active triterpenoids from Cephalaria ambrosioides; Planta Medica,
(2009); 75: 163–167.
13. Ji, D., Wu, Y., Zhang, B., Zhang, C.F., Yang, Z.L.; Triterpene saponins
from the roots of Dipsacus asper and their protective effects against the
Aβ25-35 induced cytotoxicity in PC12 cells; Fitoterapia, (2012); 83:
843–848.
14. Meng, Y.H., Huang, H.S., Yu, H.P., He, M.Z., Sun, Y.B., Feng, Y.L., et al.;
Identification of main chemical constituents in extracts of Akebiae Fructus
by HPLC-ESI-MS; Zhongcaoyao, (2013); 44(12): 1562–1567.
15. Wang, J.M., Wang, Z.M., Gao, H.M., Tian, J., Hong, Y.M.; Studies on fin-
gerprint of Fructus akebiae by RP-HPLC; Chinese Pharmaceutical Journal,
(2007); 42(13): 978–980.
16. Wang, J.M., Gao, H.M., Wang, Z.M., Wang, W.H.; Determination of sa-
ponins PK in Fructus akebiae by HPLC; Chinese Pharmaceutical Journal,
(2007); 42(12): 896–898.