全 文 ：
380 Journal of Chinese Pharmaceutical Sciences http://www.jcps.ac.cn
An UPLC-MS/MS application to investigate the chemical composition of
the ethanol extract from Anoectochilus chapaensis and its hypoglycemic
activity in insulin-resistant HepG2 cells
Cai Jinyan1*, Ni Jun1, Zhao Lin2
1. School of Pharmacy, Guangdong Pharmaceutical University, Guangzhou 510006, China;
2. School of Life Science and Bio-pharmaceutical, Guangdong Pharmaceutical University, Guangzhou 510006, China
Abstract: Anoectochilus chapaensis Gagnep. (Orchidaceae) was named as the “king of medicine” because of its excellent efficacy
for the treatment of diabetes. However, the bioactive constituents are unknown. An ethanol extract from A. chapaensis showed
significant stimulating effect on glucose consumption in HepG2 cells. The chemical composition was investigated by UPLC-MS/MS
in negative electrospray ionization (ESI) mode, and 63 compounds including flavonoids, triterpenoids, and aliphatic acids were
tentatively identified by accurate mass and characteristic fragments. Moreover, the method of hypoglycemic screening with insulin
resistant HepG2 cells and UPLC-MS/MS might be potentially useful in rapid and efficient characterization and primary prediction of
natural products prior to traditional isolation.
Keywords: UPLC-MS/MS, Hypoglycemic, HepG2 cells, Anoectochilus chapaensis
CLC number: R284 Document code: A Article ID: 1003–1057(2016)5–380–07
Received: 2016-02-23, Revised: 2016-03-25, Accepted: 2016-04-12.
Foundation items: Pearl River Nova Program of Guangzhou 2015
(Grant No. 201506010061), Foundation for Distinguished Young
Teachers in Higher Education of Guangdong (Grant No. YQ2015097),
and National Natural Science Foundation of China (Grant No.
81001628).
*Corresponding author. Tel.: +86-020-39352140,
E-mail: caijy928@163.com
http://dx.doi.org/10.5246/jcps.2016.05.043
1. Introduction
Anoectochilus chapaensis is distributed in Guangxi,
Guangdong, Hainan, Guizhou, Sichuan, and Yunnan,
and has been widely used in China to treat diabetes,
nephritis, and other diseases. In our previous study,
A. chapaensis herb extract exhibited a potent effect on
slowing the progression of high-fat diet associated
insulin resistance in rats[1]. However, its bioactive
constituents are unknown. Conventional isolation and
identification of individual compounds from the complex
mixture by nuclear magnetic resonance (NMR), mass
spectrometry (MS), and other spectroscopic techniques
are time-consuming and difficult in many cases[2]. Ultra
performance liquid chromatography (UPLC) coupled







with MS/MS has been proven to be an effective method
in rapid identification of the constituents in a mixture[3].
In this way, accurate mass and molecular formulas of
the compounds could be analyzed and tentatively
identified[4]. Moreover, the fragmentation behavior of
some natural compounds has been investigated exten-
sively and summarized, and the unknown compounds
could be identified according to the characteristic
fragments even without reference standards[5].
Type 2 diabetes (T2DM) is one of the major public
health challenges in the 21st century. IDF estimated
that by 2040 as many as 642 million people worldwide
would develop T2DM. A useful treatment for type 2
diabetes is needed urgently. In vitro, HepG2 cells are
usually employed to evaluate the hypoglycemic activity
of drugs as food is metabolized primarily in liver[6].
HepG2 cells were used in this study due to their common
physiological function in lipid or glucose metabolism
compared to normal hepatic cells[7]. Prior to dosing, the
cells should be made to be insensitive to insulin by
applying a culture medium with high concentrations of
insulin for about 24 h.

381 Cai, J.Y. et al. / J. Chin. Pharm. Sci. 2016, 25 (5), 380–386
2. Experimental
2.1. Plant Material
The herbs of A. chapaensis were collected in
Sept. of 2010 from Yunnan Province, in southwest
China. The plant was authenticated by Hongyan Ma,
Associate Professor in School of Traditional Chinese
Medicine, Guangdong Pharmaceutical University. A
voucher specimen (2010-DY1001) has been deposited
in the herbarium of School of Pharmacy, Guangdong
Pharmaceutical University.
2.2. Reagents and chemicals
Insulin and MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide) were purchased from
Sigma-Aldrich Co. LLC. (San Francisco, American).
DMEM high glucose medium, FBS (Fetal Bovine
Serum), PBS (phosphate buffered saline), and penicillin-
streptomycin were obtained from Hyclone (Utah,
American). Trypsin (0.25%) was purchased from Gibco
(California, American). Glucose assay kits were obtained
from Zhong Shen Bai Kong Biotechnology Co. Ltd.
(Shenzhen, China). Final concentrations of DMSO for
dissolving in medium were below 0.1% (v/v). MTT was
diluted with PBS buffer and filtered for sterilization.
2.3. Preparation of the samples
Dry powdered herbs (1.6 kg) of A. chapaensis were
refluxed with 95% (v/v) ethanol for 2 h, and the filtrate
was concentrated to dryness in vacuo to render the
total ethanol extract, which was then redissolved in
DMSO.
2.4. Cell culture
Human HepG2 cells were maintained in DMEM
supplemented with 10% fetal bovine serum (FBS),
100 U/mL penicillin, 100 mg/mL streptomycin, 2 mM
L-glutamine (Invitrogen, CA, USA), and kept at 37 °C
in a humidified atmosphere of 5% CO2 in air. Cells
were grown to 70% confluence and then preincubated
in serum-free medium for 24 h before treatment.
2.5. Insulin-resistant HepG2 cell model
The HepG2 cells were seeded into 96 multi-well
plates in DMEM supplemented with 10% FBS, 100 U/mL
penicillin, and 100 mg/mL streptomycin. The cells
were cultured in a humidified incubator (5% CO2) at
37 ºC, and were allowed to attach for 24 h. Insulin-
resistant cells were induced according to a previous
method with a slight modification[8]. In brief, HepG2 cells
were incubated with fresh medium containing 1% FBS and
5×10–7 mol/L bovine insulin for 24 h. Subsequently,
the medium was exchanged with medium containing
10–9 mol/L insulin and a test sample or metformin, and
then incubated for 12 h.
2.6. Bioassay of hypoglycemic activity in insulin
resistant HepG2 cells
The hypoglycemic activity was analyzed by measuring
the glucose consumption in HepG2 cells. The HepG2 cells
in 96-well plates was pre-incubated with DMEM high
glucose medium containing different concentrations of
the extract of A. chapaensis. and dimethyldiguanide
(100 μL) at 37 ºC for 24 h. This was repeated three
times. Glucose consumption was measured at 510 nm by
using a glucose assay kit on an automatic biochemical
analyzer. The media was supplemented MTT for 4 h.
The absorbance (OD) retained by the cell lysates was
determined at 490 nm by a Multiskan Spectrum[9].
2.7. Qualitative analysis by UPLC-MS/MS
An ACQUITY UPLC system, equipped with a
quaternary solvent delivery system, an auto-sampler, and a
column compartment was used. The chromatographic
separation was achieved on an ACQUITY UPLCTM
BEH C18 column (2.1 mm×50 mm, 1.7 μm). Different
gradient programs were compared to optimize the
UPLC separation condition. The final gradient elution
was as follows: 15% methanol in 0–2.5 min, 30%
methanol in 2.5–5 min, 50% methanol in 5–9 min,
70% methanol in 9–19 min, 80% methanol in 19–24 min,
and 100% methanol in 24–30 min. The flow rate was
0.4 mL/min and the sample injection volume was 5 μL.

382 Cai, J.Y. et al. / J. Chin. Pharm. Sci. 2016, 25 (5), 380–386
2.8. Data analysis
Data are shown as the mean±standard deviation (SD).
The statistical analysis was performed by oneway analysis
of variation (ANOVA) followed by Dunnett t-test for
multiple comparisons. A difference with a P value below
0.05 was considered statistically different.
3. Results
3.1. Hypoglycemic activity in vitro
As shown in Table 1, the glucose consumption (GC)
and viabilty of HepG2 cells were significantly different
in the tested groups compared with the model group
(P<0.05), which was not treated with any A. chapaensis
extract. When the extract concentration was in the
range of 6.25–25 g/L, the GC was improved compared
to the model group, however the difference was less
significant. The number of cells that survived increased
with the concentration of extract. The greatest GC and
maximum cell survival was achieved at a concentration
of 50 g/L. At the highest concentration of 200 g/L, the
GC was 4.66, whereas the MTT was 1.38, indicating
the lowest cell survival, which also showed the best
GC/MTT of 3.42. Thus, increasing extract concentra-
tions showed a dose-dependent hypoglycemic effect
in insulin-resistant HepG2 cells.
3.2. UPLC-MS/MS analysis
When a reference standard was available, the com-
pound was identified by comparing its retention time
and MS/MS spectra with those of the standard. In contrast,
the identification without an available standard was
mainly based on the MS/MS spectra and literature
information. Four reference standard compounds were
investigated with the established analysis method first.
Figure 1 showed the total ion chromatograms of the
crude extract from A. chapaensis in the negative ESI
mode. The separation was completed and all peaks
were well separated from each other within 30 min. In
our study, flavonoids, triterpenes, and aliphatic acids
were found to be the main constituents in A. chapaensis.
among 63 major constituents identified or tentatively
characterized, including 22 flavonoids, 14 coumarins,
and 27 aliphatic acids (Tables 2–4).
Table 2 illustrates 22 kinds of flavonoids in the crude
extract of A. chapaensis. By comparing the retention time
and MS/MS fragmentation pattersn with the reference
standard, peaks 9 and 11 were identified as quercetin
and kaempferol, respectively.
At tR of 2.271 min, peak 9, with a [M-H]
– of 301.0334
and the MS2 data displayed a profile consistent with
quercetin. Thus the ions 463.0874, 585.0912, 609.1443,
593.1488 and 639.1357 were also derived from quercetin.
The sugar fraction is lost easily, and so it presented
with the same base peak of 301. At tR of 9.799 min,
162 Da and 301 Da indicated the structure of quercetin-
hexosyl. At 1.518 min, the loss of 169 Da corresponded
to the galloyl group. The other fragments of 433 and 301
confirmed the structure of quercetin-galloyl-pentose. At
6.144 min, the base peak of 301 was clear. Combined
with the lost fractions of the 162 Da and 146 Da, the
structure should be quercetin-hexosyl-deoxy hexosyl.
Group Concentrations of sample GC (mmol/L) MTT/A490 GC/MTT
Model 0 2.40±0.03 1.80±0.17 1.33±0.07
Drug 6.25 g/L 2.71±0.52* 1.62±0.01 1.67±0.18*
12.5 g/L 2.53±0.59* 1.63±0.31 1.61±0.28*
25.0 g/L 2.71±0.52* 1.71±0.04 1.59±0.18*
50.0 g/L 5.02±0.52* 1.79±0.38 2.93±0.40*
100 g/L 3.83±0.43* 1.61±0.11 2.39±0.18*
200 g/L 4.66±0.43* 1.38±0.17* 3.42±0.30*
Metformin 10 mg/L 3.25±0.08* 1.75±0.01 1.86±0.17*
Data were expressed as mean±SD (n = 4), analysed by IBM SPSS Statistics 19.0. *P<0.05 compared with model group.
Table 1. Hypoglycemic activity of the extract of A. chapaensis in HepG2 cells.

383 Cai, J.Y. et al. / J. Chin. Pharm. Sci. 2016, 25 (5), 380–386
Figure 1. Total ion chromatograms of the crude extract from A. chapaensis in negative ESI mode.
Table 2. Characterisation of flavonoids in the extract of A. chapaensis by UPLC-MS/MS.
Peak tR (min) Formula [M-H]
– Error (ppm) Major and important MS2 ions Identification
1 0.668 C22H21O11 461.1100 3.5 315, 301 MQ-dHx
2 1.139 C29H31O16 635.1643 4.9 285, 257, 229, 162, 146 MK-Hex-dHx
3 1.390 C29H33O14 605.1895 4.1 299, 285, 229, 146 MK-OCH3-2dHx
4 1.430 C16H11O6 299.0536 –4.7 285, 257, 229 MK
5 1.518 C27H21O15 585.0912 5.5 433, 301, 255, 179, 151 Q-galloyl-pen
6 1.622 C28H31O16 623.1603 –1.4 477, 315, 301, 271, 255 MQ-Hex-dHx
7 1.643 C22H21O12 477.1013 –4.2 315, 301 MQ-Hex
8 2.054 C28H23O15 599.1030 –1.2 285, 257, 229, 179 K-galloyl-Hex
9 2.271 C15H9O7 301.0334 –2.9 255, 179, 151 Q
10 2.454 C21H21O12 465.1029 –0.9 285, 257, 229, 179 K-Hex
11 3.467 C15H9O6 285.0415 –1.1 257, 229, 151 K
12 6.144 C27H29O16 609.1443 –2.1 463, 301, 255, 179 Q-Hex-dHx
13 6.525 C22H21O12 477.1050 3.6 315, 301 MQ-Hex
14 6.527 C28H31O17 639.1541 –3.1 477, 315, 301 MQ-2Hex
15 6.530 C27H29O15 593.1488 –3.0 447, 285 K-dHx-Hex
16 6.581 C31H27O15 639.1357 1.1 477, 315, 301 Q-feruloyl-Hex
17 6.924 C34H41O21 785.2128 –1.5 639, 477, 315, 301 MQ-2Hex-dHx
18 6.924 C16H9O8 329.0296 –0.3 315, 301 MQ-OCH3
19 8.427 C16H11O7 315.0485 –4.5 300, 271, 151 MQ
20 8.525 C22H21O12 477.1050 3.6 315, 301 MQ-Hex
21 9.382 C18H15O7 343.0799 –3.9 329, 315, 301 MQ-2OCH3
22 9.799 C21H19O12 463.0874 –0.6 301, 271, 255, 162 Q-Hex
Note: tR, retention time; Q, quercetin; MQ, methyl-quercetin; K, kaempferol; MK, methyl-kaempferol; Hex, hexose; Pen, pentose; dHx, deoxyhexose.
100

90

80

70

60

50

40

30

20

10

0
UPLC-MS
%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
t (min)
0.49
1.05
1.30
1.43
1.57
2.25
4.45 5.59
5.94 6.62
7.97
9.23
10.20
11.15
11.57
10.55
14.13
14.22
14.97 16.71
17.94
20.85
21.80
25.17
25.94
17.17
19.40
20.03
20.30

384 Cai, J.Y. et al. / J. Chin. Pharm. Sci. 2016, 25 (5), 380–386
Table 3. Characterisation of triterpenoids in crude extract of A. chapaensis by UPLC-MS/MS.
Table 4. Characterisation of aliphatic acids in crude extract of A. chapaensis by UPLC-MS/MS.
Peak tR (min) Formula [M-H]
– Error (ppm) Identification
23 8.686 C30H47O 423.3631 0.9 Anoectosterol
24 13.611 C29H49O3 445.3671 –2.5 Campesterol-formic
25 17.606 C31H51O3 471.3857 4.0 Friedelin
26 20.410 C30H47O3 455.3527 0.4 Oleanolic acid
27 23.262 C36H57O8 617.4053 2.3 Oleanolic acid-Hex
28 23.373 C35H57O7 589.4107 0.5 Ursolic acid-dHx
29 23.452 C36H57O8 617.4067 2.3 Lanosterol-Hex-formic
30 23.868 C36H59O5 571.4370 1.2 Amyain-dHx
31 24.632 C29H47O 411.3609 –4.4 Anoectosterol
32 25.466 C29H47O3 443.3526 0.2 Ursolic acid
33 25.754 C30H49O 425.3769 –3.3 Sorghumol
34 26.346 C36H57O7 601.4086 –3.0 Oleanolic-dHx
35 27.316 C36H59O6 587.4287 –4.3 Amyain-Hex
36 27.944 C39H55O3 571.4137 –2.5 Sorghumol-coumarate
Peak tR (min) Formula [M-H]– Error (ppm) Identification
37 16.980 C14H27O2 227.2004 –3.1 Myristic acid
38 16.983 C18H29O2 277.2169 0.4 Calendic acid
39 17.230 C17H35O2 271.2654 4.3 Methyl hexadecanoate
40 17.838 C16H29O2 253.2167 –0.4 Hexadecenoic acid
41 19.290 C15H29O2 241.2165 –1.2 Pentadecanoate acid
42 19.345 C18H31O2 279.2335 3.9 Octadecadienoic acid
43 19.483 C17H31O2 267.2317 –2.6 Heptadecenoic acid
44 19.881 C16H31O2 255.2333 3.5 Palmitic acid
45 20.488 C18H33O2 281.2495 5.0 Jeceric acid
46 20.906 C17H33O2 269.2483 0.7 Heptadecanoic acid
47 21.096 C19H35O2 295.2650 4.4 Jecoleic acid
48 21.174 C19H37O2 297.2783 –3.7 Nonadecanoic acid
49 21.800 C18H35O2 283.2647 3.5 Octadecanoic acid
50 22.126 C20H37O2 309.2797 1.0 Eicosenoic acid
51 24.484 C20H39O2 311.2959 2.9 Arachic acid
52 25.056 C21H41O2 325.3125 5.5 Heneicosanoic acid
53 25.383 C22H43O2 339.3263 0.0 Behenic acid
54 25.631 C23H45O2 353.3401 –5.4 Tricosanoic acid
55 25.975 C24H47O2 367.3589 3.5 Lignoceric acid
56 26.128 C25H49O2 381.3741 2.1 Pentacosanoic acid
57 26.793 C26H51O2 395.3902 3.3 Hexacosanic acid
58 27.001 C27H53O2 409.4042 –1.0 Heptacosanoic acid
59 27.479 C28H55O2 423.4206 0.9 Octacosanoic acid
60 28.069 C29H57O2 437.4348 –2.5 Nonacosanoic acid
61 28.624 C30H59O2 451.4516 0.2 Melissic acid
62 29.177 C31H61O2 465.4672 0.0 Hentriaconta acid
63 29.804 C32H63O2 479.4802 –5.4 Dotriacontanoic acid

385 Cai, J.Y. et al. / J. Chin. Pharm. Sci. 2016, 25 (5), 380–386
The fragment ions of 301, 271 and 151 were observed
at tR of 8.427 min and m/z was 315.0845, therefore, it
should be methyl-quercetin. 301 Da was due to the
loss of a methyl group (15 Da). Furthermore, [M-H]–
623.1603 at 1.622 min, [M-H]– 461.1100 at 0.668 min,
[M-H]– 477.1013 and 639.1541 at 6.527 min, [M-H]-
785.2128 at 6.924 min, [M-H]– 329.0296 at 6.924 min,
[M-H]– 477.105 at 8.525 min, and [M-H]– 343.0799
at 9.382 min all showed the base peak of 315 Da.
[M-H]– 329.0296 was consistent with a methyl group
and methyl-quercetin from its MS2 spectra. In addition,
dimethyl-quercetin and another methyl was trimethyl-
quercetin (343.0799). In the MS2 spectra, 162 Da
(hexosyl), 146 Da (deoxy hexosyl), and 315 Da
(quercetin) could be found at 8.427 min, thus, it
was tentatively identified as quercetin-hexosyl-deoxy
hexosyl.
At the retention time (tR) of 3.467 min, kaempferol
was characterized. The [M-H]– was 285.0415 with a
formula of C15H10O6. And its MS
2 spectrum showed
characteristic fragment ions of 257, 229 and 151. The
fragments of 257 and 229 could be observed as the
result of losing CO ions gradually. The constituents at
tR of 2.054 min and 2.454 min were presumed to be
kaempferol-galloyl-pentose and kaempferol-hexosyl,
respectively, which had the same moiety as kaempferol,
and the fragment ions were produced by loss of a hexosyl
group (162 Da), a hexosyl plus a galloyl group (285 Da).
The fragments of 257 and 229 could be found at tR of
2.054 min, and the fragment of 162 Da could easily be
observed at 2.454 min.
A new analogue with [M-H]– 299.0536 was found at
1.430 min, and it had the same base fragment as
kaempferol after loss of a methyl group. It was identified
as methyl-kaempferol. [M-H]– 635.1643 and 605.1895
indicated the structures of methyl-kaempferol-hexosyl-
deoxy hexosyl and dimethyl-kaempferol-dideoxy hexosyl
with the same base peak of 285.
Table 3 showed 14 kinds of triterpenoids and
triterpenoid saponins. As in the ESI mode, a formic
group (46 Da) was usually attached to the [M-H]–.
Peak 25 gave an [M-H]– ion at m/z 471.3857 (C31H51O3)
and was directly identified as friedelin. Peak 26 gave
an [M-H]– ion at m/z 455.3527 (C30H47O3) and was
identified as oleanolic acid.
Table 4 showed 27 kinds of aliphatic acids in crude
extract of A. chapaensis by UPLC-MS/MS. Peaks 37
and 44 were identified as myristic acid and palmitic
acid upon comparison with their authentic standards.
4. Discussion
In conclusion, the method of hypoglycemic screening
in insulin-resistant HepG2 cells coupled with UPLC-
MS/MS might be potentially useful in the systematic
analysis and pilot study of natural products prior to
traditional isolation. A total of 63 compounds were
identified or tentatively characterized on the basis of
their retention times, exact mass measurement for
molecular ions, and subsequent product ions. Flavo-
noids, terpenoids, and aliphatic acids were the major
constituents in A. chapaensis, and the hypoglycemic
activity of its extract was observed in insulin-resistant
HepG2 cells. As a type of promising plant resource,
A. chapaensis had the hypoglycemic activity for diabetic
treatment. Therefore, our work might play a guiding
role in the future studies.
Acknowledgements
This work was supported by Pearl River Nova
Program of Guangzhou 2015 (Grant No. 201506010061),
Foundation for Distinguished Young Teachers in Higher
Education of Guangdong (Grant No. YQ2015097), and
National Natural Science Foundation of China (Grant
No. 81001628) for financial support.

386 Cai, J.Y. et al. / J. Chin. Pharm. Sci. 2016, 25 (5), 380–386

滇越金线兰醇提物对胰岛素抵抗的HepG2细胞降血糖活性
及其化学成分的UPLC-MS/MS研究
蔡金艳1*, 倪俊1, 赵林2
1. 广东药科大学 药学院, 广东 广州 510006
2. 广东药科大学 生命科学与生物制药学院, 广东 广州 510006
摘要: 民间俗称“药王”的滇越金线兰对治疗糖尿病有很好的疗效, 但其活性成分的系统分析还未见报道。本研究将
滇越金线兰粗提物作用于胰岛素抵抗的HepG2细胞, 可显著促进细胞的葡萄糖摄取。应用超高效液相色谱串联质谱法
(UPLC-MS/MS)快速分析滇越金线兰中的化学成分, 通过保留时间、质谱精确分子量和特征碎片推定出63种成分, 包括
黄酮类、三萜和脂肪族化合物, 结合胰岛素抵抗的HepG2对提取物进行降血糖活性筛选, 采用UPLC-MS/MS快速进行
成分分析, 证明是一种可行且可靠的分离和鉴定中药中活性成分的方法。
关键词: 降血糖活性; HepG2细胞; 高效液相色谱串联质谱法; 滇越金线兰
References
[1] Cai, J.Y.; Zhao, L.; Zhu, E.; Wang, C.X. J. Chin. Pharm.
Sci. 2014, 23, 262–267.
[2] Vukics, V.; Guttman, A. Mass Spectrom. Rev. 2010, 29,
1–16.
[3] Stalmach, A.; Edwards, C.A.; Wightman, J.D.; Crozier, A.
J. Agric. Food Chem. 2011, 59, 9512–9522.
[4] de la Torre-Carbot, K.; Jauregui, O.; Gimeno, E.; Castellote,
A.I.; Lamuela-Raventós, R.M.; López-Sabater, M.C. J.
Agric. Food Chem. 2005, 53, 4331–4340.
[5] Gu, D.Y.; Yang, Y.; Bakri, M.; Chen, Q.B.; Xin, X.L.;
Aisa, H.A. Phytochem. Anal. 2013, 24, 661–670.
[6] Xie, W.D.; Wang, W.; Su, H.; Xing, D.M.; Pan, Y.;
Du, L.J. Comparative Biochemistry and Physiology Part C:
Toxicology & Pharmacology. 2006, 143, 429–435.
[7] Xu, J.S.; Ma, M.W.; Purcell, W.M. Toxicol. Appl.
Pharmacol. 2003, 189, 100–111.
[8] Chen, Q.; Xia, Y.P.; Qiu, Z.Y. Chin. Pharmacol. Bull.
2006, 22, 248–251.
[9] Zhu, X.K.; Ye, X.L.; Song, L. Med. Chem. Res. 2013, 22,
2228–2234.