C8-selective biomimetic transformation of 5,7-dihydroxylated flavonoids by an acid-catalysed phenolic Mannich reaction: Synthesis of flavonoid alkaloids with quercetin and (–)-epicatechin skeletons

Article abstract of DOI:10.1016/j.tet.2017.01.068

We hereby report the biomimetic synthesis of three flavonoid alkaloids, namely 8-(2″-pyrrolidinon-5″-yl)quercetin, 6-(2″-pyrrolidinon-5″-yl)-(–)-epicatechin and 8-(2″-pyrrolidinon-5″-yl)-(–)-epicatechin. These known natural products were prepared via an a

Full text of DOI:10.1016/j.tet.2017.01.068

Accepted Manuscript
C8-selective biomimetic transformation of 5,7-dihydroxylated flavonoids by an acid-
catalysed phenolic Mannich reaction: Synthesis of flavonoid alkaloids with quercetin
and (–)-epicatechin skeletons
Viktor Ilkei, András Spaits, Anita Prechl, Judit Müller, Árpád Könczöl, Sándor Lévai,
Eszter Riethmüller, Áron Szigetvári, Zoltán Béni, Miklós Dékány, Ana Martins, Attila
Hunyadi, Sándor Antus, Csaba Szántay, Jr., György Tibor Balogh, György Kalaus,
Hedvig Bölcskei, László Hazai
PII:
S0040-4020(17)30113-8
10.1016/j.tet.2017.01.068
TET 28438
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 22 July 2016
Revised Date: 14 January 2017
Accepted Date: 27 January 2017
Please cite this article as: Ilkei V, Spaits A, Prechl A, Müller J, Könczöl E, Lévai S, Riethmüller E,
Szigetvári E, Béni Z, Dékány M, Martins A, Hunyadi A, Antus S, Szántay Jr. C, Balogh GT, Kalaus G,
Bölcskei H, Hazai L, C8-selective biomimetic transformation of 5,7-dihydroxylated flavonoids by an acid-
catalysed phenolic Mannich reaction: Synthesis of flavonoid alkaloids with quercetin and (–)-epicatechin
skeletons, Tetrahedron (2017), doi: 10.1016/j.tet.2017.01.068.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
C8-Selective biomimetic transformation of
5,7-dihydroxylated flavonoids by an acid-
catalysed phenolic Mannich reaction:
synthesis of flavonoid alkaloids with
quercetin and (–)-epicatechin skeletons
Leave this area blank for abstract info.
Viktor Ilkeia, * , András Spaitsa , Anita Prechlb , Judit Müllerb , Árpád Könczölb , Sándor Lévaib , Eszter
Riethmüllerb , Áron Szigetvárib , Zoltán Bénib , Miklós Dékányb , Ana Martinsc, # , Attila Hunyadid , Sándor
Antuse , Csaba Szántay Jr.b , György Tibor Baloghb , György Kalausa, † , Hedvig Bölcskeia and László Hazaia
aDepartment of Organic Chemistry and Technology, Budapest University of Technology and Economics, Szt. Gellért tér 4, Budapest, H-1111, Hungary
bGedeon Richter Plc., Gyömrői út 19-21, Budapest, H-1103, Hungary
cDepartment of Medical Microbiology and Immunobiology, University of Szeged, Dóm tér 9, Szeged, H-6720, Hungary
#Permanent address: Synthetic Systems Biology Unit, Institute of Biochemistry, Biological Research Centre, Temesvári krt. 62, Szeged, H-6726, Hungary
dInstitute of Pharmacognosy, Faculty of Pharmacy, University of Szeged, Eötvös utca 6, Szeged, H-6720, Hungary
eDepartment of Organic Chemistry, University of Debrecen, Egyetem tér 1, Debrecen, H-4032, Hungary

ACCEPTED MANUSCRIPT
Tetrahedron
journal homepage: www.elsevier.com
C8-Selective biomimetic transformation of 5,7-dihydroxylated flavonoids by an acid-
catalysed phenolic Mannich reaction: synthesis of flavonoid alkaloids with quercetin
and (–)-epicatechin skeletons
Viktor Ilkeia, * , András Spaitsa , Anita Prechlb , Judit Müllerb , Árpád Könczölb , Sándor Lévaib , Eszter
Riethmüllerb , Áron Szigetvárib , Zoltán Bénib , Miklós Dékányb , Ana Martinsc, # , Attila Hunyadid , Sándor
Antuse , Csaba Szántay Jr.b , György Tibor Baloghb , György Kalausa, † , Hedvig Bölcskeia and László
Hazaia
aDepartment of Organic Chemistry and Technology, Budapest University of Technology and Economics, Szt. Gellért tér 4, Budapest, H-1111, Hungary
bGedeon Richter Plc., Gyömrői út 19-21, Budapest, H-1103, Hungary
cDepartment of Medical Microbiology and Immunobiology, University of Szeged, Dóm tér 9, Szeged, H-6720, Hungary
#Permanent address: Synthetic Systems Biology Unit, Institute of Biochemistry, Biological Research Centre, Temesvári krt. 62, Szeged, H-6726, Hungary
dInstitute of Pharmacognosy, Faculty of Pharmacy, University of Szeged, Eötvös utca 6, Szeged, H-6720, Hungary
eDepartment of Organic Chemistry, University of Debrecen, Egyetem tér 1, Debrecen, H-4032, Hungary
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
We hereby report the biomimetic synthesis of three flavonoid alkaloids, namely 8-(2”-
pyrrolidinon-5”-yl)quercetin, 6-(2”-pyrrolidinon-5”-yl)-(–)-epicatechin and 8-(2”-pyrrolidinon-
5”-yl)-(–)-epicatechin. These known natural products were prepared via an acid-catalysed
regioselective phenolic Mannich reaction involving the electrophilic attack of an N-acyliminium
ion on the corresponding flavonoidal precursors. The products were purified by preparative
HPLC. The reactions showed high C8-regioselectivity. The major isomers of the synthesized
flavonoid alkaloids were further characterized in terms of their medicinal-chemical properties.
Available online
Keywords:
Flavonoid alkaloids
2009 Elsevier Ltd. All rights reserved
.
Biomimetic synthesis
Regioselective phenolic Mannich reaction
Flavonoid chemistry
N-acylaminocarbinol
Iminium chemistry
1. Introduction
unexploited potential that lies in this family of natural products in
terms of bioactivity and structural diversity, it can be rightfully
stated that the isolation and synthesis of flavonoid alkaloids and
their derivatives is a highly important field of research today.
Flavonoid alkaloids constitute a unique group of natural
products. They contain – in almost all cases – a five- or six-
membered nitrogen heterocycle which is connected to the C6 or
C8 position of the A-ring of the flavonoid skeleton. Almost all
molecules in this group are found in plants which are used in folk
medicine as herbal remedy for the treatment of a wide range of
conditions. Some compounds were reported to possess valuable
pharmacological activity themselves.1 Research into the field of
flavonoid alkaloids gained even more attention when a synthetic
flavonoid alkaloid derivative, namely flavopiridol2 (brand name:
Alvocidib, by Tolero Pharmaceuticals®) performed promisingly
in phase II clinical trials and was granted the orphan drug
designation by the FDA for the treatment of acute myeloid
The flavonoid alkaloid 8-(2”-pyrrolidinone-5”-yl)quercetin (1,
Fig. 1) was isolated as a racemate from Senecio argunensis by N.
Li and co-workers in 2008.4 This molecule has a 2-pyrrolidone
moiety attached to the C8 position of the A-ring of quercetin (4,
Fig. 1). Senecio argunensis is a perennial herb distributed in
northeast and northwest China. It is used in traditional Chinese
medicine as a remedy for the treatment of, among other
conditions, sore throat, dysentery and snake bite.
leukaemia in the United States in 2014.3 Based on the still
———
Corresponding author. Tel.: +36 1 463 1277; e-mail address: vilkei@mail.bme.hu (V. Ilkei).
† We regret to report that our co-author prof. György Kalaus died in 2014.

2
Tetrahedron
ACCEPTED MANUSCRIPT
Figure 1. Structure of flavonoid alkaloids 1-3 and their flavonoid precursors (4, 5)
Scheme 1. A plausible biosynthetic route to pyrrolidone-substituted flavonoids
6-(2”-Pyrrolidinone-5”-yl)-(–)-epicatechin (2, Fig. 1) and 8-
to yield the C6- and C8-isomers of pyrrolidone-substituted
flavonoids (Scheme 1). Plausible explanations of this
biosynthetic route for similar cases are given by E. Leete6 and by
Tanaka et al.7
(2”-pyrrolidinone-5”-yl)-(–)-epicatechin (3, Fig. 1) are both
derivatives of the flavan-3-ol (–)-epicatechin (5, Fig. 1). They
were isolated from the roots of Actinidia arguta by D. S. Jang
and co-workers in 2009.5 Actinidia arguta (or hardy kiwi) is a
perennial vine native to northern China, Japan, Korea and
Siberia. It produces an edible kiwi-like fruit used in traditional
Chinese medicine to improve general health. In in vitro
biological tests compounds 2 and 3 were both found to exhibit
significant inhibitory effect against the formation of advanced
glycation end products (AGEs), which play a key role in the
pathogenesis of diabetic complications, cataracts, atherosclerosis
and other neurodegenerative diseases.5
The aim of our work was to synthetically prepare flavonoid
alkaloids 1-3 from their flavonoid precursors (4, 5) in order to
further investigate the desired products from a medicinal
chemistry point of view.
2. Results and discussion
We have devised a feasible methodology for the synthesis of
lactam-containing flavonoid alkaloids, utilizing a cyclic N-
acylaminocarbinol reagent,8 which can be easily prepared from
succinimide in one step. This N-acylaminocarbinol is allowed to
react with the appropriate flavonoid compound in a suitable
solvent to yield the desired flavonoid alkaloids via an acid-
catalysed regioselective phenolic Mannich reaction.9 This type of
transformation was applied in recent works by Tanaka et al. in
the synthesis of a black tea polyphenol-type flavonoid alkaloid,7
and also by Nguyen and co-workers in the selective synthesis of
the C6- and C8-isomers of chrysin-derived flavonoid alkaloids
and other pyrrolidine- and piperidine-substituted chrysin
The biosynthesis of these flavonoid-pyrrolidone conjugates is
hypothesized6,7 to proceed via an aminocarbinol-derived iminium
intermediate, which presumably arises through the Strecker-
degradation of the corresponding amino acids present in the
plants – in the case of molecules 1-3, L-glutamine (6). The
spontaneous cyclization of Strecker aldehyde 7 gives rise to 5-
hydroxypyrrolidin-2-one (8), which is transformed into the N-
acyliminium ion 9 upon protonation and subsequent loss of
water. This N-acyliminium ion is a reactive electrophile, which
attacks the activated A-ring of the flavonoids present in the plants

derivatives.10 Complementarily to their work, our biomimetic
ACCEPTED MANUSCRIPT
synthetic route provides a convenient way to pyrrolidone-
substituted flavonoids and can be generally applied to various
flavonoid substrates.
First we prepared the N-acylaminocarbinol reagent 5-
ethoxypyrrolidin-2-one (11) by the partial reduction of
succinimide (10) with sodium borohydride (Scheme 2), as
described in the literature.11
Scheme 2. Synthesis of 5-ethoxypyrrolidin-2-one (11)
the pertinent isomeric ratio of the natural forms of 2 and 3
isolated from Actinidia arguta was found to be 2 : 3 = 22:78.5
Quercetin (4) was allowed to react with reagent 11 with acid
catalysis in refluxing THF (Scheme 3). The crude product was
subjected to preliminary purification in order to dispose of the
remaining quercetin by refluxing its suspension in ethyl acetate
(dissolving unreacted 4), and filtering the purified product. The
obtained solid was purified by preparative HPLC, after which the
desired flavonoid alkaloid 1 was obtained together with its C6-
isomer (12) in a ratio of 94:6, indicating an electrophilic attack
predominantly directed towards C8 in the reaction. We note that
the C8-substituted regioisomer was the only isomer of this
natural product which was isolated from its natural source
(Senecio argunensis).4
A plausible mechanism of these phenolic Mannich reactions
(Scheme 5) is basically the same as that of the proposed
biosynthetic route (Scheme 1). It involves the electrophilic attack
of the N-acyliminium ion 9 on the aromatic A-ring of the
flavonoid substrate at C6 or C8. The reactive iminium ion 9
arises from 5-ethoxypyrrolidin-2-one (11) after protonation and
loss of ethanol.
2.1. Structure elucidation of flavonoid alkaloids 1, 2, 3 and 12
Literature NMR and MS data4,5 are available for compounds
1, 2 and 3. Li et al4 elucidated structure 1 with high-resolution
mass spectrometry (HRMS), infrared (IR) spectroscopy, and
Scheme 3. Synthesis of 8-(2”-pyrrolidinone-5”-yl)quercetin (1) and the minor C6-isomer (12)
Scheme 4. Synthesis of 8-(2”-pyrrolidinone-5”-yl)-(–)-epicatechin (3) and 6-(2”-pyrrolidinone-5”-yl)-(–)-epicatechin (2)
In the next step, (–)-epicatechin (5) was allowed to react with
5-ethoxypyrrolidin-2-one (11) in the presence of an acid catalyst
in refluxing THF (Scheme 4). A suspension of the crude product
in ethyl acetate was refluxed (dissolving unreacted 5) and
filtered, in order to get rid of the remaining starting material
contaminating the product. The obtained solid was purified by
preparative HPLC, which gave the epimeric mixtures of the C8-
(3) and the C6-substituted isomers (2) in a ratio of 87:13. The
epimers of the regioisomers could not be separated by the applied
method. The C8-regioselectivity observed in the reaction
involving (–)-epicatechin is somewhat lower than in the case of
quercetin, but nonetheless relatively high. One should note that
various one-dimensional (1D) and two-dimensional (2D) NMR
1 1
techniques, such as H-1H correlated spectroscopy (COSY), H-
13C heteronuclear multiple quantum coherence (HMQC; one-
bond 1H-13C connectivity data), and 1H-13C heteronuclear
multiple bond connectivity (HMBC). Jang et al5 elucidated
structures 2 and 3 with the techniques mentioned above along
with an additional NMR method, 1H-1H nuclear Overhauser
effect spectroscopy (NOESY), and circular dichroism (CD)
spectroscopy. They used CD to determine the absolute
configuration at C2 and they analyzed the 1H-1H coupling
constants to determine relative configuration between C2 and C3,
concluding that the configuration of epicatechin derivatives 2 and
3 is (2R,3R).

4
Tetrahedron
ACCEPTED MANUSCRIPT
Scheme 5. Proposed mechanism for the formation of flavonoid alkaloids (dashed lines indicate variations in the flavonoid core)
Figure 2. Qualitative representation of the observed 2D ROESY correlations of compounds 1, 2, 3 and 12. For clarity, 1H chemical shift values
are given for only one of the two epimers of compound 3. In the case of compound 2, the 1H NMR peaks of the two epimers overlap, except for
the proton C(5)-OH, which facilitates the determination of the epimeric ratio of 2.
1
Despite the detailed spectroscopic data available in the
literature, a direct comparison of the NMR data of our synthetic
samples with those reported in the literature4,5 proved to be
unsuitable for unambiguous structure identification. This is
because the C6- and C8-substituted regioisomers have very
similar H and 13C NMR spectra with only slight differences in
the pertinent chemical shift values. The small experimental
variations in these values with pH, temperature, concentration,
etc., are in the same order of magnitude as the regioisomeric
differences, making the comparison with the literature data

#
#
TPSA# [Å2]
unreliable. The cases of epicatechin derivatives 2 and 3 are
ACCEPTED MANUSCRIPT
clogP
clogD7.4/clogD6.5
Cpnd.
further complicated by the epimeric nature of these compounds,
because the configuration of C5” is unspecified relative to C2
and C3, resulting in two 1H NMR and 13C NMR signal sets.
1
3
1.93
1.48
0.11/0.87
0.79/1.01
159.38
156.55
For the above reasons, we conducted an ab initio NMR-based
structure determination for all synthetic samples, also with a view
to giving full 1H and 13C NMR assignments where possible. This
provided a solid basis for a post factum comparison of our
assignments to those available in the literature (1 in DMSO-d6, 2
and 3 in CD3OD). Note that for the purposes of a thorough and
reliable spectral and structural assignment the choice of DMSO-
d6 as the solvent was critical in order to observe the OH and NH
proton resonances, even though Jang et al assigned 2 and 3 in
methanol-d4. The application of 1H NMR, 13C NMR, COSY,
HSQC (heteronuclear single quantum coherence; similarly to
# Physicochemical parameters were predicted for 1 and 3 with MarvinSketch
v6.0.2 software.
The above data are in agreement with the results of the
parallel artificial membrane permeability assay (PAMPA)
measurements shown in Table 2. As a general rule of thumb, the
total polar surface area (TPSA) cut-off values in the case of
membrane permeability by passive diffusion are TPSA < 120 Å2
for gastrointestinal absorption and TPSA < 70 Å2 for BBB
penetration.12 Thus it comes as no surprise that none of the two
compounds were permeable in the blood–brain barrier (BBB)- or
gastrointestinal (GI)-PAMPA systems. However, the difference
between the membrane retention values (MR%) showed a good
correlation with the difference between the calculated logP
(clogP) values, indicating that both tissue specific lipid
membranes (GI and also BBB) bind compound 1 to a greater
extent, which is slightly more lipophilic than the epimeric
mixture of compound 3.
1
HMQC, this technique provides one-bond H-13C connectivity
data, but with the advantage that it can differentiate methylene
groups from CHs and CH3s), and HMBC measurements was
necessary for proving the structure of the flavon skeleton, the
pyrrolidinone ring, and the connection between these moieties.
Even so, the regioisomers could not be identified unambiguously
from these experiments in the cases of the substituted epicatechin
samples 2 and 3. To determine whether a particular sample
contained a C6- or a C8-substituted flavonoid, 1H-1H spatial
proximity measurements (2D ROESY: two-dimensional rotating
frame Overhauser effect spectroscopy) were therefore essential
(Fig. 2).
Table 2. Brain (BBB) and gastrointestinal (GI) specific
permeability parameters of 1 and 3 (NP: non-permeable)
BBB-PAMPA
Cpnd.
Pe (cm/s*10-6)
MR%
45.3
10
SD (MR%)
In the cases of the C8-substituted flavonoids 1 and 3, the
regioisomers could be identified by the spatial proximity of the
pyrrolidinone moiety and the catechol ring of the flavone
skeleton (ring B). The C6-substituted flavonoids (12 and 2) give
weak correlation peaks in their 2D ROESY spectra between the
catechol ring and the C8-H proton. The application of 2D
ROESY resulted in unambiguous structures for all synthetic
samples. Subsequently we compared our NMR data with those of
Li et al. and Jang et al. In order to facilitate that comparison, we
also acquired 1H NMR spectra for compounds 2 and 3 in CD3OD,
and a 13C NMR spectrum in CD3OD for the synthetic material
before the separation of the regioisomers (this sample was
available in a larger quantity, and contained 2 and 3 in ca. 1:6
ratio). We found a good agreement with the literature NMR data,
except for a minor discrepancy in the 1H NMR literature
assignment of 14: the revised chemical shift values for C(3)-OH
and C(4’)-OH are 9.42 ppm and 9.60 ppm, which are proven with
the corresponding three-bond H-O-C-C HMBC correlations.
1
3
NP
NP
3.5
2.4
GI-PAMPA
1
3
NP
NP
20.4
3.3
3.3
2.3
2.2.2. Antioxidant activity
1 and 3 had a strong antioxidant effect on the DPPH radical
scavenging test (IC50=12.1±1.8 and 7.5±1.1 µg/mL respectively),
compared to the positive control quercetin (IC50=9.4±0.7 µg/mL).
2.3. Anticancer activity and effect on the ABCB1 efflux
transporter
In connection with the above studies on their BBB
penetration, the cytotoxicity of the major isomers of the
synthesized flavonoid alkaloid samples 1 and 3 was evaluated on
the SH-SY5Y cancer cell line. None of the above compounds
exerted cytotoxicity on the tested cell line (IC50>150 ꢀM for all
the compounds). Compounds 1 and 3 were also evaluated for
their potential to interfere with the ABCB1 transporter
(commonly referred to as P-glycoprotein or P-gp), which plays a
major role in the function of BBB through active transport.13
Neither compound inhibited this efflux transporter as determined
by measuring the intracellular accumulation of Rhodamin 123.
The molar ratio of the two epimers of 2 can be estimated by
integration of the singlets (linewidth at 50% peak height: 7 Hz) at
1
8.26 ppm and 8.29 ppm (C(5)-OH). This is the only pair of H
NMR peaks which appear separately for the two epimers at 800
MHz. The epimeric ratio for 2 is ca. 50%:50%. The C(5)-OH
peaks for the epimers of 3 are broad and they overlap, thus the
molar ratio of the two epimers of 3 is calculated by integration of
the H-3 multiplets, because they separate well at 800 MHz
(epimer A: ca. 58%, 4.02–4.04 ppm; epimer B: ca. 42%, 3.99–
4.01 ppm). The estimated margin of measurement error is ca. 3%.
3. Conclusion
In the present work, we have developed a convenient way to
2.2. Medicinal-chemical properties of compounds 1 and 3
synthesize
pyrrolidone-containing
flavonoid
alkaloids,
2.2.1. Physicochemical parameters
employing an easily accessible N-acylaminocarbinol reagent, and
successfully synthesized 8-(2”-pyrrolidinone-5”-yl)quercetin (1),
6-(2”-pyrrolidinone-5”-yl)quercetin (12), 6-(2”-pyrrolidinone-5”-
yl)-(–)-epicatechin (2) and 8-(2”-pyrrolidinone-5”-yl)-(–)-
epicatechin (3) via an acid-catalysed regioselective phenolic
Mannich reaction. The crude products were obtained as isomeric
mixtures, which were separated by preparative HPLC. The pure
Typical predicted physicochemical parameters of compound 1
and the epimeric mixture of compound 3 are shown in Table 1.
Table 1. Predicted physicochemical parameters of compounds
1 and 3

6
Tetrahedron
ACCEPTED MANUSCRIPT
4.2.3. Preparative separation procedure
major isomers of the synthesized flavonoid alkaloids were
further characterized in terms of their medicinal-chemical
properties.
The sample (500 µL/injection) was loaded to a Kinetex 5 µm
Phenyl-Hexyl AXIA packed 21.2×150 mm column in the
concentration of 20 mg/mL. The separation was achieved at 25
°C. Phase A was 0.1% TFA in water, phase B contained 0.1%
TFA in acetonitrile. The developed linear gradient program
increased the ratio of phase B from 5 to 30 percent by volume in
15 min. The flow rate was 21 mL/min. The fractions were
collected based on UV absorption at 220 nm wavelength.
4. Experimental
4.1. General
Melting points were measured on a SANYO Gallenkamp
apparatus and are uncorrected. IR spectra were recorded on a
Bruker FT-IR instrument and a PerkinElmer Spectrum 100 FT-IR
Spectrometer equipped with Universal ATR (diamond/ZnSe)
accessory. NMR measurements (1H, 13C, gCOSY, 1D NOESY,
1D ROESY, 2D ROESY, gHSQCAD, gHMBCAD) were
performed on Varian 400 MHz (equipped with 5 mm OneNMR
15N-31P/{1H-19F} PFG Probe), Varian 500 MHz (equipped with
1H{13C/15N} 5 mm PFG Triple Resonance 13C Enhanced Cold
4.3. Medicinal-chemical properties
4.3.1. Antioxidant activity
The radical scavenging capacities of 1 and 3 were tested in the
microplate format of the 2,2-diphenyl-1-picrylhydrazyl (DPPH)
assay.14
Quercetin
was
obtained
from
PhytoLab
(Vestenbergsgreuth, Germany), and used as control.
Probe) and Varian 800 MHz (equipped with H{13C/15N} Triple
1
Resonance 13C Enhanced Salt Tolerant Cold Probe)
4.3.2. Permeability measurements
1
The PAMPA-BBB method was previously published.15
Briefly, a 96-well acceptor plate and a 96-well filter plate are
assembled into a sandwich. The hydrophobic filter material of the
96 well filter plate is coated with 5 ꢀL of a 2.6% (w/v) dodecane:
hexane (25:75 v/v%) solution of porcine brain lipid (PBL).
Subsequently, the acceptor wells at the bottom of the sandwich
are filled with 300 ꢀL of 10 mM PBS solution with 5% DMSO
adjusted to pH 7.4. The donor wells at the top of the sandwich
are hydrated with 150 ꢀL of test compound solution. The test
compound solution is prepared by diluting ×100 from a 10 mM
stock solution in DMSO using PBS solution at pH 7.4 with 5%
DMSO followed by filtration through a MultiScreen Solubility
filter plate. The resulting sandwich is then incubated at 37 oC for
4 h. After incubation, the PAMPA sandwich plates are separated,
and compound concentrations in donor and acceptor solutions are
determined by HPLC-DAD.
spectrometers. H chemical shifts are given on the delta scale as
parts per million (ppm) with tetramethylsilane (TMS) as the
internal standard (0.00 ppm). 13C chemical shifts are given on the
delta scale as parts per million (ppm) with tetramethylsilane
(TMS) or dimethylsulfoxide-d6 as the internal standard (0.0 ppm
and 39.5 ppm, respectively). HRMS and MS analyses were
performed on a Finnigan MAT 95 XP, and a Thermo LTQ FT
Ultra as well as a Thermo LTQ XL (Thermo Fisher Scientific,
Bremen, Germany) system. The ionization method was EI
operated in positive ion mode on a Finnigan MAT 95 XP.
Electron energy was 70 eV and the source temperature was set at
220°C. The ionization method was ESI operated in positive ion
mode on the other two systems. For the CID experiment helium
was used as the collision gas, and normalized collision energy
(expressed in percentage), which is a measure of the amplitude of
the resonance excitation RF voltage applied to the endcaps of the
linear ion trap, was used to induce fragmentation. The protonated
molecular ion peaks were fragmented by CID at a normalized
collision energy of 35%. Data acquisition and analysis were
accomplished with Xcalibur software version 2.0 (Thermo Fisher
Scientific). TLC was carried out on TLC Silica gel 60 F254 on
20×20 cm aluminium sheets (Merck), and preparative TLC was
carried out using Silica gel 60 PF254+366 (Merck) coated glass
plates. Column chromatography was performed using Silica gel
60 (0.063-0.200 mm) (Merck). Flavonoids (quercetin and (–)-
epicatechin) were purchased from Sigma-Aldrich and used
without further purification.
A slightly modified version of PAMPA16 was used to
determine the gastrointestinal permeability (PAMPA-GI) as
described previously.17 Gradient-pH (donor: 6.5 → acceptor: 7.4)
conditions were utilized. The hydrophobic filter material of the
96 well filter plate was coated with 5 ꢀL of a 4% (w/v) dodecane
solution of phosphatidylcholine:cholesterol (2:1).
4.4. Cytotoxicity on SH-SY5Y cancer cell line
4.4.1. Cell lines
SH-SY5Y neuroblastoma cells were cultured in EMEM media
supplemented with non-essential amino acids, 1 mM Na-pyruvate
and 10% inactivated fetal bovine serum, Nystatin, 2 mM L-
glutamine, 100 U penicillin and 0.1 mg streptomycin, purchased
from Sigma. Cells were cultured at 37 °C and 5% CO2.
4.2. Preparative HPLC
4.2.1. Chemicals and reagents
The acetonitrile used in the preparative chromatographic
separation was gradient grade LiChrosolv purchased from Merck.
The trifluoroacetic acid (Uvasol) was obtained from Merck.
Water was purified with a Milli-Q system. The sample solvent
was spectroscopy grade dimethyl sulfoxide (Uvasol) purchased
from Merck.
Two mouse lymphoma cell lines were used to test for the
capacity of the compounds to inhibit the function of the ABCB1
transporter: a parental cell line, L5178 mouse T-cell lymphoma
cells (ECACC catalog no. 87111908, U.S. FDA, Silver Spring,
MD, U.S.), and a multi-drug resistant (L5178MDR) cell line
derived from L5178 by transfection with pHa MDR1/A
retrovirus.18 Cells were cultured in McCoy’s 5A media
supplemented inactivated horse serum and antibiotics as above.
MDR cell line was selected by culturing the infected cells with
60 ꢀg/L colchicine (Sigma).
4.2.2. Apparatus
The separation of the isomeric mixtures of flavonoid alkaloids
was performed with a Shimadzu chromatograph equipped with
an LC-8A pump unit, SPD-M20A Photodiode detector and
CBM-20A system controller. The samples were introduced via
an SIL-10AP sample injector and the chromatograms were
processed using the LCsolution software. The fractions were
collected by an FRC10A fraction collector module. The
temperature was controlled by a CTO-20AC prominence column
oven.
4.4.2. Materials and methods
Compounds 1 and the epimeric mixture of 3 were dissolved in
DMSO at a stock concentration of 10 mM.
4.4.2.1. Cytotoxicity on SH-S5Y5 cells

0.814 mmol) was added and the mixture was refluxed for 4 h.
Ten thousand cells per well were seededAoCverCniEghPt.TSEerDial MANUSCRIPT
dilutions of the compounds were prepared and added the
following day to the plate. Cells were then incubated for 48 h,
after which 10% MTT was added to each well. After 4 h, SDS
was added to the medium and the results were read after o/n
incubation. Fifty percent inhibitory concentrations (IC50) were
calculated using nonlinear regression curve fitting of log
(inhibitor) versus normalized response and variable slope with a
least squares (ordinary) fit of GraphPad Prism 5 software, for
three independent samples.
The solvent was evaporated in vacuo and the solid residue was
suspended in ethyl acetate, and refluxed for 5 min. Then the
suspension was left to cool to room temperature and it was
filtered. The filtrate was discarded and the filtered solid was
subjected to the above procedure two more times to dispose of
the quercetin remaining in the product. A crude product (1.922 g,
61%) was obtained, a sample of which (98 mg) was subjected to
further purification by preparative HPLC, which gave title
compound 1 (32.46 mg, 94.74% purity) as a yellow solid, mp
244–246 °C; Rf (17% CH3OH/CH2Cl2) 0.46; νmax (KBr) 3536,
3296, 3120, 3003, 2828, 2723, 1688, 1646, 1603, 1555, 1513,
1436, 1375, 1319, 1272, 1201, 1177, 1135, 1020, 1008 cm−1; δH
(799.7 MHz, DMSO-d6) 2.16–2.21 (1H, m), 2.30–2.41 (3H, m):
H2-3”, H2-4”, 5.37–5.40 (1H, m, H-5”), 6.28 (1H, s, H-6), 6.87
(1H, d, J 8.5 Hz, H-5’), 7.49 (1H, dd, J 8.5, 2.2 Hz, H-6’), 7.68
(1H, d, J 2.2 Hz, H-2’), 7.78 (1H, s, NH-1”), 9.25 (1H, s, C(3’)-
OH), 9.42 (1H, s, C(3)-OH), 9.62 (1H, s, C(4’)-OH), 10.99 (1H,
s, C(7)-OH), 12.69 C(5)-OH); δC (125.7 MHz, DMSO-d6) 25.7
(C-4”), 30.8 (C-3”), 47.2 (C-5”), 98.1 (C-6), 102.9 (C-10), 106.2
(C-8), 115.1 (C-2’), 115.6 (C-5’), 119.9 (C-6’), 121.9 (C-1’),
135.4 (C-3), 144.9 (C-3’), 146.9 (C-2), 147.6 (C-4’), 153.7 (C-9),
159.4 (C-5), 162.0 (C-7), 176.0 (C-4), 176.8 (C-2”); HRMS (EI):
M=385.07892 (C19H15O8N; delta=-0.8 ppm) and title compound
12 (2.13 mg, 83.95% purity), also a yellow solid, δH (499.9 MHz,
DMSO-d6) 2.06–2.27 (2H, m, Hx-3”, Hx-4”), 2.31–2.43 (2H, m,
Hy-3”, Hy-4”), 5.16–5.21 (1H, m, H-5”), 6.46 (1H, s, H-8), 6.89
(1H, d, J 8.5 Hz, H-5’), 7.55 (1H, dd, J 8.5, 2.2 Hz, H-6’), 7.57
(1H, br s, NH-1”), 7.67 (1H, d, J 2.2 Hz, H-2’), 9.32 (1H, s,
C(3’)-OH), 9.42 (1H, s, C(3)-OH), 9.60 (1H, s, C(4’)-OH), 11.06
(1H, br s, C(7)-OH), 13.12 (1H, s, C(5)-OH); δC (125.7 MHz,
DMSO-d6) 25.3 (C-4”), 30.6 (C-3”), 46.5 (C-5”), 93.0 (C-8),
102.7 (C-10), 111.4 (C-6), 114.9 (C-2’), 115.6 (C-5’), 120.0 (C-
6’), 121.8 (C-1’), 135.7 (C-3), 145.1 (C-3’), 146.8 (C-2), 147.7
(C-4’), 154.7 (C-9), 158.6 (C-5), 162.1 (C-7), 176.0 (C-4), 177.0
(C-2”); HRMS (EI): M=385.07907 (C19H15O8N; delta=-0.4 ppm).
4.4.2.2. Effect on Rhodamine 123 accumulation
Rhodamine 123 (Rd123) is a fluorescent dye and a substrate
of the ABCB1 transporter whose overexpression prevents the
intracellular accumulation of this dye.19 Rd123 concentration
inside the cells was determined by flow cytometry. Briefly, 2 ×
106 cells/mL were treated with 2 and 20 ꢀM of each compound
and incubated for 10 min at RT. Rhodamine 123 (Sigma,
Germany) was added to a final concentration of 5.2 ꢀM. The
samples were incubated for 20 min at 37 °C in water bath and
then centrifuged (2000 rpm, 2 min). The pellet was resuspended
in 0.5 mL of phosphate buffer saline (PBS) (Sigma, Germany).
The washing step was repeated twice. The fluorescence of the
samples was measured by flow cytometry (Becton Dickinson
FACScan, BD, U.S.). Tariquidar at 2 ꢀM was used as positive
control. Fluorescence activity ratio measures the capacity of
inhibition (accumulation of Rd123) and it is equal to the ratio
between the FL-1 values of the L5178MDR cells treated and
untreated.
4.5. Experimental procedures
4.5.1. 5-Ethoxypyrrolidin-2-one (11)
To a solution of succinimide (10) (7.157 g, 72.23 mmol) in
ethanol (300 mL) sodium borohydride (4.00 g, 105.74 mmol)
was added in one portion at 0 °C. The reaction mixture was
stirred at 0 °C for 4 h, during which time five drops of 2 M
ethanolic hydrogen chloride solution were added every 15 min.
Then the reaction mixture was acidified to pH=3 with 2 M
ethanolic hydrogen chloride solution over 30 min, after which it
was stirred at 10 °C for 90 min. Then the reaction mixture was
neutralised (pH = 7) with 5% ethanolic potassium hydroxide
solution and evaporated in vacuo to give a syrupy solid, which
was suspended in chloroform (80 mL), filtered, and the
precipitate washed with chloroform (3×20 mL). The filtrate was
evaporated in vacuo to give a colourless oil, which was dissolved
in dichloromethane (80 mL) and washed with water (3×10 mL).
The aqueous phase was extracted with dichloromethane
(6×20 mL), then the organic phases were unified, dried (MgSO4)
and the solvent evaporated in vacuo to give the title compound 11
(5.788 g, 62%) as a colourless oil, which crystallized on standing,
giving white crystals, mp 54–58 °C (lit.: 48–53 °C);11 Rf
(acetone) 0.72; νmax (KBr) 3200, 2978, 1707, 1689, 1668, 1457,
1282, 1250, 1067, 986 cm−1; δH (399.8 MHz, DMSO-d6) 1.10
(3H, t, J 7.0 Hz, CH3), 1.78–1.89 (1H, m, Hx-4), 1.95–2.05 (1H,
m, Hx-3), 2.11–2.30 (2H, m, Hy-3, Hy-4), 3.27–3.35 (1H, m),
3.44–3.54 (1H, m): OCH2, 4.83–4.89 (1H, m, H-5), 8.62 (1H, s,
NH-1); δC (100.5 MHz, DMSO-d6) 15.1 (CH3), 27.7 (C-4), 28.1
(C-3), 61.6 (OCH2), 85.0 (C-5), 177.4 (C-2); MS(ESI):
M+H=130 (C6H12NO2); ESI-MS-MS (cid=35) (rel. int. %):
84(100).
4.5.3. 6-(2”-Pyrrolidinone-5”-yl)-(–)-epicatechin
(2) and 8-(2”-pyrrolidinone-5”-yl)-(–)-epicatechin
(3)
To a mixture of (–)-epicatechin (5) (440 mg, 1.52 mmol) and
5-ethoxypyrrolidin-2-one (11) (235 mg, 1.82 mmol) in THF (30
mL) 10 mol% p-toluenesulfonic acid monohydrate (29 mg, 0.152
mmol) was added and the mixture was refluxed for 2 h. The
solvent was evaporated in vacuo and the solid residue was
suspended in ethyl acetate, and refluxed for 5 min. Then the
suspension was left to cool to room temperature and it was
filtered. The filtrate was discarded and the filtered solid was air-
dried by suction. A crude product (765 mg) was obtained, a
sample of which (103 mg) was subjected to further purification
by preparative HPLC, which gave an epimeric mixture of title
compound 2 (5.30 mg, 95.95% purity) as a white solid, mp 183–
189 °C (lit.: 195–197 °C);5 Rf (17% CH3OH/CH2Cl2) 0.27; νmax
(diamond/ZnSe) 3272, 1652, 1620, 1516, 1451, 1344, 1282,
1199, 1159, 1118, 1058 cm−1; δH for epimer A (799.7 MHz,
DMSO-d6) 2.03–2.08 (1H, m, Hx-4”), 2.13–2.18 (1H, m, Hx-3”),
2.22–2.27 (1H, m, Hy-4”), 2.30–2.34 (1H, m, Hy-3”), 2.48–2.50
(1H, m, Hx-4), 2.71–2.74 (1H, m, Hy-4), 4.01–4.03 (1H, m, H-3),
4.70–4.72 (1H, m, H-2), 4.71–4.77 (1H, br m, C(3)-OH), 5.12
(1H, dd, J 9.1, 5.3 Hz, H-5”), 5.88 (1H, s, H-8), 6.63–6.65 (1H,
m, H-6’), 6.65–6.67 (1H, m, H-5’), 6.88–6.89 (1H, m, H-2’),
7.38* (1H, s, NH-1”), 8.29 (1H, s, C(5)-OH), 8.74 (1H, br s,
C(4’)-OH), 8.81 (1H, br s, C(3’)-OH), 9.12 (1H, s, C(7)-OH); δC
for epimer A (201.1 MHz, DMSO-d6) 25.9 (C-4”), 28.6 (C-4),
30.9 (C-3”), 47.5 (C-5”), 64.6 (C-3), 77.6 (C-2), 95.0 (C-8), 99.5
(C-10), 109.0 (C-6), 114.6 (C-5’), 114.7 (C-2’), 117.8 (C-6’),
4.5.2. 8-(2”-Pyrrolidinone-5”-yl)quercetin (1) and
6-(2”-pyrrolidinone-5”-yl)quercetin (12)
To a mixture of quercetin (4) (2.459 g, 8.14 mmol) and 5-
ethoxypyrrolidin-2-one (11) (1.261 g, 9.76 mmol) in THF (70
mL) 10 mol% p-toluenesulfonic acid monohydrate (155 mg,

8
Tetrahedron
ACCEPTED MANUSCRIPT
7. (a) Tanaka, T.; Watarumi, S.; Fujieda, M.; Kouno, I. Food Chem.
130.3 (C-1’), 144.3 (C-4’), 144.4 (C-3’), 154.0 (C-9), 154.4 (C-
2005, 93, 81–87.; (b) Tanaka, T.; Matsuo, Y.; Kouno, I. Int. J.
Mol. Sci. 2010, 11, 14–40.
5), 154.6 (C-7), 176.8 (C-2”); δH for epimer B (799.7 MHz,
DMSO-d6) 2.03–2.08 (1H, m, Hx-4”), 2.13–2.18 (1H, m, Hx-3”),
2.22–2.27 (1H, m, Hy-4”), 2.30–2.34 (1H, m, Hy-3”), 2.50–2.52
(1H, m, Hx-4), 2.70–2.73 (1H, m, Hy-4), 4.01–4.03 (1H, m, H-3),
4.70–4.72 (1H, m, H-2), 4.71–4.77 (1H, br m, C(3)-OH), 5.12
(1H, dd, J 9.1, 5.3 Hz, H-5”), 5.88 (1H, s, H-8), 6.63–6.65 (1H,
m, H-6’), 6.65–6.67 (1H, m, H-5’), 6.88–6.89 (1H, m, H-2’),
7.38* (1H, s, NH-1”), 8.26 (1H, s, C(5)-OH), 8.74 (1H, br s,
C(4’)-OH), 8.81 (1H, br s, C(3’)-OH), 9.12 (1H, s, C(7)-OH), *:
in the epimeric mixture of 2, the NH-1” peaks of the two epimers
are separated by 4 ppb; δC for epimerB (201.1 MHz, DMSO-d6)
25.9 (C-4”), 28.6 (C-4), 30.9 (C-3”), 47.6 (C-5”), 64.6 (C-3),
77.6 (C-2), 95.0 (C-8), 99.4 (C-10), 109.0 (C-6), 114.6 (C-5’),
114.7 (C-2’), 117.8 (C-6’), 130.3 (C-1’), 144.3 (C-4’), 144.4 (C-
3’), 154.0 (C-9), 154.3 (C-5), 154.5 (C-7), 176.7 (C-2”); HRMS
(ESI+): M+H=374.12344 (delta=0.03 ppm; C19H20O7N). HR-
ESI-MS-MS (CID=35%; rel. int. %): 356(100), 339(9), 311(2),
246(27), 234(32), 222(20), 217(2), 206(2), 205(3); and an
epimeric mixture of title compound 3 (37.02 mg, 97.89% purity),
also a white solid, mp 178–186 °C (lit.: 195–197 °C);5 Rf (17%
CH3OH/CH2Cl2) 0.27; νmax (diamond/ZnSe) 3277, 1648, 1610,
1521, 1455, 1357, 1280, 1199, 1159, 1099, 1067 cm−1; δH for
epimer A (799.7 MHz, DMSO-d6) 2.07–2.13 (2H, m, Hx-3”, Hx-
4”), 2.17–2.24 (2H, m, Hy-3”, Hy-4”), 2.51–2.54 (1H, m, Hx-4),
2.69–2.73 (1H, m, Hy-4), 4.03–4.04 (1H, m, H-3), 4.32–4.90 (1H,
br, C(3)-OH), 4.74–4.75 (1H, m, H-2), 5.14–5.18 (1H, m, H-5”),
6.00 (1H, s, H-6), 6.66 (1H, d, J 8.1 Hz, H-5’), 6.71 (1H, dd, J
8.1, 2.0 Hz, H-6’), 6.88 (1H, d, J 2.0 Hz, H-2’), 7.37 (1H, br s,
NH-1”), 8.59–8.87 (2H, br, C(3’)-OH, C(4’)-OH), 9.03 (1H, br s,
C(7)-OH), 9.11 (1H, br s, C(5)-OH); δC for epimer A (201.1
MHz, DMSO-d6) 25.8 (C-4”), 28.5 (C-4), 30.8 (C-3”), 47.1 (C-
5”), 64.5 (C-3), 77.9 (C-2), 95.2 (C-6), 98.4 (C-10), 106.6 (C-8),
114.4 (C-2’), 114.7 (C-5’), 117.5 (C-6’), 130.4 (C-1’), 144.1 (C-
4’), 144.3 (C-3’), 153.7 (C-9), 154.5 (C-7), 155.1 (C-5), 176.5
(C-2”); δH for epimer B (799.7 MHz, DMSO-d6) 2.03–2.07 (1H,
m, Hx-3”), 2.15–2.20 (2H, m, H2-4”), 2.17–2.21 (1H, m, Hy-3”),
2.47–2.50 (1H, m, Hx-4), 2.67–2.71 (1H, m, Hy-4), 3.99–4.01
(1H, m, H-3), 4.32–4.90 (1H, br, C(3)-OH), 4.76–4.77 (1H, m,
H-2), 5.13–5.16 (1H, m, H-5”), 5.99 (1H, s, H-6), 6.66–6.68 (2H,
m, H-5’, H-6’), 6.87 (1H, m, H-2’), 7.30 (1H, br s, NH-1”), 8.59–
8.87 (2H, br, C(3’)-OH, C(4’)-OH), 9.03 (1H, br s, C(7)-OH),
9.11 (1H, br s, C(5)-OH); δC for epimer B (201.1 MHz, DMSO-
d6) 25.6 (C-4”), 28.1 (C-4), 30.6 (C-3”), 47.2 (C-5”), 64.3 (C-3),
78.1 (C-2), 95.0 (C-6), 98.4 (C-10), 106.4 (C-8), 114.5 (C-2’),
114.7 (C-5’), 117.7 (C-6’), 130.4 (C-1’), 144.2 (C-4’), 144.3 (C-
3’), 153.8 (C-9), 154.2 (C-7), 155.0 (C-5), 176.4 (C-2”); HRMS
(ESI+): M+H=374.12337 (delta=-0.16 ppm; C19H20O7N). HR-
ESI-MS-MS (CID=35%; rel. int. %): 356(87), 339(1), 246(5),
234(13), 222(100), 206(1), 205(8).
8. For reviews on the chemistry of N-acyliminium ions and the
synthesis of their precursors see: (a) Speckamp, W. N.; Hiemstra,
H. Tetrahedron 1985, 41, 4367–4416.; (b) Speckamp, W. N.;
Moolenaar, M. J. Tetrahedron 2000, 56, 3817–3856.; (c) Marson,
C. M. ARKIVOC (Gainesville, FL, U.S.) 2001, 1–16.; (d)
Maryanoff, B. E.; Zhang, H.; Cohen, J. H.; Turchi, I. J.;
Maryanoff, C. A. Chem. Rev. (Washington, DC, U.S.) 2004, 104,
1431–1628.; (e) Royer, J.; Bonin, M.; Micouin, L. Chem. Rev.
(Washington, DC, U.S.) 2004, 104, 2311–2352.; (f) Yazici, A.;
Pyne, S. G. Synthesis 2009, 339–368.; (g) Yazici, A.; Pyne, S. G.
Synthesis 2009, 513–541.
9. For relevant studies on regioselective phenolic Mannich reactions
see: (a) Tramontini, M.; Angiolini, L. Mannich Bases – Chemistry
and Uses, CRC Press: Boca Raton, 1994; pp 32–38.; (b) Omura,
Y.; Taruno, Y.; Irisa, Y.; Morimoto, M.; Saimoto, H.; Shigemasa,
Y. Tetrahedron Lett. 2001, 42, 7273–7275.
10. (a) Nguyen, T. B.; Wang, Q.; Guéritte, F. Eur. J. Org. Chem.
2011, 7076–7079.; (b) Nguyen, T. B.; Lozach, O.; Surpateanu, G.;
Wang, Q.; Retailleau, P.; Iorga, B. I.; Meijer, L.; Guéritte, F. J.
Med. Chem. 2012, 55, 2811–2819.
11. Hubert, J. C.; Wijnberg, J. B. P. A.; Speckamp, W. N.
Tetrahedron 1975, 31, 1437–1441.
12. Kelder, J.; Grootenhuis, P. D. J.; Bayada, D. M.; Delbressine, L.
P. C.; Ploemen, J. P. Pharm. Res. 1999, 16, 1514–1519.
13. Schinkel, A. H. Adv. Drug Deliv. Rev. 1999, 36, 179–194.
14. Hu, C.; Kitts, D. D. Phytomedicine 2005, 12, 588–597.
15. Müller, J.; Esső, K.; Dargó, G.; Könczöl, Á.; Balogh, G. T. Eur. J.
Pharm. Sci. 2015, 79, 53–60.
16. Kansy, M.; Senner, F.; Gubernator, K. J. Med. Chem. 1998, 41,
1007–1010.
17. Borbás, E.; Balogh, A.; Bocz, K.; Müller, J.; Kiserdei, É.; Vigh,
T.; Sinkó, B.; Marosi, A.; Halász, A.; Dohányos, Z.; Szente, L.;
Balogh, G. T.; Nagy, Z. K. Int. J. Pharm. 2015, 419, 180–189.
18. Pastan, I.; Gottesman, M. M.; Ueda, K.; Lovelace, E.; Rutherford,
A. V.; Willingham, M. C. Proc. Natl. Acad. Sci. U.S.A. 1988, 85,
4486–4490.
19. Forster, S.; Thumser, A. E.; Hood, S. R.; Plant, N. PLOS ONE
2012, 7, e33253.
Acknowledgments
The authors are grateful to Gedeon Richter Plc. for financial
assistance.
References and notes
1. Khadem, S.; Marles, R. J. Molecules 2012, 17, 191–206.
2. Jain, S. K.; Bharate, S. B.; Vishwakarma, R. A. Mini-Rev. Med.
Chem. 2002, 12, 632–649.
3. Oncology Times 2014, 36(10), p. 91.
4. Li, N.; Shao, L.; Zhang, C.; Zhang, M. J. Asian Nat. Prod. Res.
2008, 10, 1143–1146.
5. Jang, D. S.; Lee, G. Y.; Lee, Y. M.; Kim, Y. S.; Sun, H.; Kim, D.;
Kim, J. S. Chem. Pharm. Bull. 2009, 57, 397–400.
6. Leete, E. J. Nat. Prod. 1982, 45, 605–607.