Design, synthesis and inhibitory activities of naringenin derivatives on human colon cancer cells

Article abstract of DOI:10.1016/j.bmcl.2012.10.130

Based on the previous result, several naringenin derivatives modified at position 7 with bulky substituents were designed and synthesized, and their inhibitory effects on HCT116 human colon cancer cells were tested using a clonogenic assay. The half maximal inhibitory concentrations (IC₅₀) of five naringenin derivatives ranged between 1.20 μM and 20.01 μM which are much better than naringenin used as a control. In addition, new structural modification at C-4 of flavanone results in improving both the anti-cancer effect and anti-oxidative effect. In vitro cyclin dependent kinase 2 (CDK2) binding assay was carried out based on the previous results. To elucidate the possible interaction between naringenin derivatives and CDK2, in silico docking study was performed. This result demonstrates the rationale for the different inhibitory activities of the naringenin derivatives. These findings could be used for designing cancer therapeutic or preventive flavanone-derived agents.

Full text of DOI:10.1016/j.bmcl.2012.10.130

Bioorganic & Medicinal Chemistry Letters 23 (2013) 232–238
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design, synthesis and inhibitory activities of naringenin derivatives
on human colon cancer cells
Hyuk Yoon a, Tae Woo Kim b, Soon Young Shin c, Mi Joo Park a, Yeonjoong Yong a, Dong Woon Kim d,
Tasneem Islam e, Young Han Lee c, Kang-Yeoun Jung b, Yoongho Lim a,
⇑
a Division of Bioscience and Biotechnology, BMIC, Konkuk University, Seoul 143-701, Republic of Korea
b Department of Biochemical Engineering, Gangneung-Wonju National University, Gangwon 210-702, Republic of Korea
c Department of Biomedical Science and Technology, SMART-Institute of Advanced Biomedical Science, RCTC, Konkuk University, Seoul 143-701, Republic of Korea
d Swine Science Division, National Institute of Animal Science, RDA, Cheonan 330-801, Republic of Korea
e Department of Biological, Chemical and Physical Sciences, Roosevelt University, IL 60173, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Based on the previous result, several naringenin derivatives modified at position 7 with bulky substitu-
ents were designed and synthesized, and their inhibitory effects on HCT116 human colon cancer cells
were tested using a clonogenic assay. The half maximal inhibitory concentrations (IC50) of five naringenin
Received 26 July 2012
Revised 3 October 2012
Accepted 29 October 2012
Available online 5 November 2012
derivatives ranged between 1.20 lM and 20.01 lM which are much better than naringenin used as a con-
trol. In addition, new structural modification at C-4 of flavanone results in improving both the anti-cancer
effect and anti-oxidative effect. In vitro cyclin dependent kinase 2 (CDK2) binding assay was carried out
based on the previous results. To elucidate the possible interaction between naringenin derivatives and
CDK2, in silico docking study was performed. This result demonstrates the rationale for the different
inhibitory activities of the naringenin derivatives. These findings could be used for designing cancer ther-
apeutic or preventive flavanone-derived agents.
Keywords:
Flavanone
Naringenin
Colon cancer
In silico docking
Ó 2012 Elsevier Ltd. All rights reserved.
Colon cancer is the fourth most commonly diagnosed malignant
disease and is prevalent where the people have adopted western
diets and also among the elderly. Its symptom is worsening consti-
pation or bloody stool. When such symptom arises, complete treat-
ment is late and 5 years survival is <60%, and thus, a periodic
colonoscopy is desired.1 Main treatment for colon cancer is sur-
gery, radiation, or chemotherapy which is applied as adjuvant ther-
apy in many cases. Chemotherapeutic agents such as oxaliplatin,
leucovorin, and irinotecan are known, however, they are associated
with severe adverse effects. Therefore, there is a need for more po-
tent and less toxic drugs.2
Flavanones, 2-phenylchroman-4-one, belong to the family of
flavonoids, many of which are produced as secondary metabolites
in the plant kingdom. They have three-ring skeletons, C6-C3-C6,
and the rings are referred to as A-, C-, and B-rings, respectively.
Their functional groups are attached to the main skeleton through
oxygen or carbon linkages.3 The biological activities of flavonoids
depend on the degree of condensation in their structures and the
position and number of substitutions, such as hydroxy groups, glu-
cosides, isoprenyl units, homodimers, and heterodimers.4 Flava-
nones compared to flavones, 2-phenyl-4H-chromen-4-one, have
more molecular flexibility due to the absence of a carbon–carbon
double bond in the C-ring. In our previous studies, we found that
a methoxy or hydroxy substituent at C-7 position of flavanone re-
sulted in better inhibitory effects on HCT116 human colon cancer
cell lines.5 Based on this result, we tried to design flavanone deriv-
atives with bulkier substituents at C-7 position and elucidate their
inhibitory effects.
Since the anti-cancer activity of naringenin, 40,5,7-trihydroxyf-
lavanone against colon cancer cells has been reported, several fla-
vanone derivatives were designed from naringenin.6 Besides,
naringenin plays a key role as an estrogenic substance in humans
and as an endogenous regulator in plants.7 Various flavonoid deriv-
atives modified at C-7 position have been reported, but in flava-
none, especially naringenin, derivatives modified at position 7
have rarely been studied.8,9 Naringenin derivatives designed in this
study have a common moiety, so that they may result in small
changes in their biological activities. Therefore, one of long-term
survival assays, clonogenic assay, was applied here.10 As a control,
naringenin was used.
Naringenin contains three hydroxy groups (Fig. 1). O-substitu-
tions can be easily at carried out at the 40- and 7-hydroxy groups,
however, the 5-hydroxy group forms a hydrogen bond (H-bond)
with the ketone at C-4, making it less accessible. The treatment
of cancer cell lines such as breast cancer cell line Michigan Cancer
Foundation 7 (MCF7) with flavanone derivatives shows different
effects according to the variation in functional groups on C-40
⇑
Corresponding author.
E-mail address: yoongho@konkuk.ac.kr (Y. Lim).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.10.130

H. Yoon et al. / Bioorg. Med. Chem. Lett. 23 (2013) 232–238
233
HCT116 human colon cancer cells (Health Protection Culture
Collections, Salisbury, UK) were obtained from the American Type
Culture Collection (Rockville, MD). Cells were grown in Dulbecco’s
modified Eagle’s medium (DMEM; Invitrogen Life Technologies,
Carlsbad, CA) supplemented with 10% (v/v) heat-inactivated fetal
bovine serum (FBS; HyClone, Logan, UT). HCT116 cells (5 ꢀ 103
cells/well) were seeded onto 24-well tissue culture plates (BD Fal-
con, Franklin Lakes, NJ) in the absence or presence of different con-
centrations of six naringenin derivatives and naringenin, and
incubated for 7 days.10 HCT116 cells were tested on four plates
OH
(A)
B
R
O
9
1'
2
7
C
A
4
OH
O
treated with 0, 10, 20, and 40 lM of samples (Fig. 2). The colonies
derivative
naringenin
R
that formed were fixed with 6% glutaraldehyde and stained with
0.1% crystal violet, as described previously.15 The clonogenic sur-
vival densities were measured using the densitometry (Multi-
Guage, Fujifilm, Japan) and their half maximal inhibitory
concentrations (IC50) were calculated using SigmaPlot software
(SYSTAT, Chicago, IL).
The 1,1-diphenyl-2-picryl-hydrazyl (DPPH) assay was used to
screen for anti-oxidative effects. DPPH radical-scavenging effects
were tested according to the method reported previously.16
In order to confirm whether naringenin derivatives including
naringenin inhibit cyclin dependent kinase 2 (CDK2), in vitro
CDK2 binding assay was performed using EMD Millipore’s Kinase-
Profiler service assay protocol (EMD Millipore Corporation, Bille-
rica, MA).
All docking experiments were carried out on an Intel Core 2
Quad Q6600 (2.4 GHz) Linux PC with Sybyl 7.3 (Tripos, St. Louis,
MO).17 Of many crystallographic structures of CDK2 deposited in
protein data bank, 2r3j.pdb was selected because its crystal
structure was determined with an inhibitor with bicyclic cores
and the structure of the ligand, 3-bromo-5-phenyl-N-(pyridin-3-
ylmethyl)pyrazolo[1,5-a]pyrimidin-7-amine, was similar to
naringenin derivatives modified at position 4 or 7 (Supplementary
data Fig. 1).18 It was originated from Homo sapiens and its resolu-
tion was 1.65 Å. Its apo-protein without its ligand was obtained
from energy minimization using the Sybyl/FlexX Single Receptor
Module. Since the Sybyl program provides flexible docking proce-
dure, the binding pocket was defined first. The docking radius
was set to 6.5 Å and the residues for docking were selected;
Ile10, Ala31, Glu81, Phe82, Leu83, His84, Gln131, and Leu134.
Naringenin derivatives including naringenin were used as ligands.
Their three dimensional (3D) structures were constructed based on
the X-ray crystallographic structure of naringenin contained in
chalcone isomerase as a ligand, 1eyq.pdb deposited in the protein
data bank and subjected to energy minimization using the molec-
ular mechanics algorithms provided by Sybyl 7.3.19 Minimization
was stopped upon convergence of the total energy (0.05 kcal/
mol Å). Systematic conformational searches were performed be-
cause rotational bonds existed in the compounds.15 The conform-
ers with the lowest energy were selected for the docking. They
were docked into apo-protein CDK2. The docking process was iter-
ated 30 times and 30 docking poses were obtained. Among them,
the results showing the similar docking pose comparing to the li-
gand contained in 2r3j.pdb were selected. The residues surround-
ing the derivatives were analyzed using LigPlot provided by the
European Bioinformatics Institute.20 All 3D images were con-
structed using PyMOL program (The PyMOL Molecular Graphics
System, Version 1.0r1, Schrödinger, LLC.).
OH
2"
O
N1
N2
S
1"
1"
O
2"
O
O
O
1"
2"
N3
N4
N5
O
O
1"
O
O
1"
2"
O
OH
(B)
HO
O
N
O
1"
OH
N
O
H
Figure 1. (A) Structures of synthetic naringenin derivatives N1–N5 and (B) the
structure of derivative N6.
and C-7.11 In this study, five naringenin derivatives modified at po-
sition 7 with thiophenecarboxylate (N1), methylbenzoate (N2),
isobutyrate (N3), allyloxy (N4), and phenyl carbonate (N5) groups
were synthesized. All derivatives except 7-allyl substituted
naringenin are novel.12 The aim of this research is to discover po-
tent anti-cancer agents showing better inhibitory effects on human
colon cancer cell lines than naringenin whose anti-cancer activity
on colon cancer has already been reported.
( )-Naringenin was purchased from INDOFINE chemical com-
pany (Hillsborough, NJ, USA). All the O-alkylated naringenin deriv-
atives N1–N5 were efficiently synthesized from the coupling
reaction of naringenin and the corresponding acid halides (deriva-
tives N1, N2, N3, and N5) or alkyl halide (derivative N4) as shown
in Scheme 1. Because of the hydrogen bond between hydroxy
group at 5 position and ketone group at 4 position, alkylation at
7-position was very selective and gave the 7-O-alkylated naringe-
nin products in good yield. Derivative N6 was obtained from the
reaction of naringenin and ethyl carbazate at 110–120 °C in a rea-
sonable yield. The compounds used for synthesis were supplied by
a local company in Korea.
The naringenin derivatives synthesized here were characterized
using NMR spectroscopy and mass spectrometry (MS). The NMR
data of naringenin, which have been reported previously, could
be applied to the derivative N1, 5-hydroxy-2-(4-hydroxyphenyl)-
4-oxochroman-7-yl thiophene-2-carboxylate, since the structure
is similar with the exception of the thiophene-2-carboxylate moi-
ety.21 The 13C peaks in the B- and C-rings were easily assigned
(Supplementary data Fig. 2). Previous reports did not show the
All nuclear magnetic resonance (NMR) measurements were per-
formed according to the methods published previously (Supple-
13,14
mentary data).

234
H. Yoon et al. / Bioorg. Med. Chem. Lett. 23 (2013) 232–238
Scheme 1. (A) Synthesis of naringenin derivatives N1–N5 modified at C-7 and (B) of ethyl carbamate derivative N6 modified at C-4.
to two 13C peaks at 115.4 and 158.0 ppm in the HMBC spectrum,
which were assigned as C-30/C-50 and C-40, respectively (Supple-
mentary data Fig. 4). Thus, the former proton was assigned as 40-
OH and the later as 5-OH. The protons and carbons in the A-ring
showed different chemical shifts compared to naringenin because
of the 7-thiophene-2-carboxylate group. The proton peak of 5-OH
was long-range coupled to the doublet carbon peak at 102.8 ppm
and the singlet peak at 106.1 ppm. They were assigned to C-6
and C-10, respectively. The 1H peak at 6.50 ppm, which was di-
rectly attached to C-6 in the HMQC spectrum (Supplementary data
Fig. 5), showed a cross peak with the 1H peak at 6.51 ppm in the
COSY spectrum (Supplementary data Fig. 6). Thus, it was assigned
as H-8. Since the 13C peak at 162.5 ppm in HMBC was long-range
coupled to 5-OH, it was assigned as C-7. Three protons at 7.29,
8.00, and 8.07 ppm were correlated with each other in COSY and
were assigned as the protons of the 7-thiophene-2-carboxylate
group. Here, the proton at 7.29 ppm showed a doublet of doublet
coupling; thus, it was assigned as H-400. Because H-6 at 6.50 ppm
was long-range coupled to the 13C peak at 159.2 ppm in HMBC, it
was assigned as C-100 of the 7-thiophene-2-carboxylate group.
The 13C peak at 128.9 ppm showed long-range couplings with
H-6 and H-8 in HMBC; thus, it was assigned as C-200. Likewise, the
13C peak at 157.6 ppm showed long-range coupling with H-6 and
was assigned as C-9. A singlet carbon at 162.4 ppm was assigned
as C-5. The chemical shift of C-300 in 7-thiophene-2-carboxylate
should be shifted farther downfield than that of C-500. Therefore,
the first of two undetermined 13C peaks at 135.7 and 135.9 ppm
was assigned as C-500. To confirm the result, high-resolution mass
spectrometry (HRMS) was carried out on a Hybrid LC-Quadru-
pole-TOF Tandem Mass Spectrometer (Applied Biosystems, Carls-
bad, CA). The calculated mass and the experimental mass agree
with each other (Supplementary data Fig. 7). The spectral data
and chemical yield of derivative N1 are listed in the references
and notes section.22
Figure 2. The effects of naringenin derivatives, N1–N6 including naringenin on the
clonogenicity of HCT116 human colon cancer cells. HCT116 cells (5 ꢀ 103 cells/
well) were plated and cultured for seven days in the absence or presence of
different concentrations of each derivative. Similar results were obtained from two
other independent experiments.
Derivatives N2, 5-hydroxy-2-(4-hydroxyphenyl)-4-oxochroman-7-yl
2-methylbenzoate, N3, 5-hydroxy-2-(4-hydroxyphenyl)-4-oxoch-
roman-7-yl isobutyrate, N4, 7-(allyloxy)-5-hydroxy-2-(4-hydroxy-
phenyl)chroman-4-one, and N5, 5-hydroxy-2-(4-hydroxyphenyl)-
4-oxochroman-7-yl phenyl carbonate, were determined based on
the same procedure as derivative N1, and their NMR assignments
and their HRMS data are listed in the references and notes
section.23–27
1H chemical shifts of 40-OH and 5-OH, but we were able to assign
those (Supplementary data Fig. 3). Two hydroxy protons were ob-
served at 9.65 and 12.01 ppm. The former was long-range coupled

H. Yoon et al. / Bioorg. Med. Chem. Lett. 23 (2013) 232–238
235
The IC50 values of five derivatives (N1–N5) on HCT116 human
colon cancer cells were 1.20, 6.03, 15.87, 1.91, and 20.01 M,
their target protein in the cell was not revealed. Cell proliferation
is tightly regulated by cell cycle progression, which is divided into
four distinct phases, G1 (Gap1), S (DNA synthesis), G2 (Gap2) and
M (mitosis) phases. Cell cycle progression is controlled through
the actions of various classes of cyclin and cyclin-dependent pro-
tein kinase (CDK) complexes. During early G1 progression, CDK4
and CDK6 are associated with D-type cyclins (D1, D2, and D3). In
late G1, CDK2 and E-type cyclin (E1 and E2) complexes are re-
quired for entry into S phase. As cells enter into S phase, CDK2
and A-type cyclin (A1 and A2) complexes play important roles in
S phase progression. As the cell moves into G2 phase, CDK1 and
B-type cyclin (B1 and B2) complexes regulate G2/M transition
and M phase progression. Previously, we and others have demon-
strated that naringenin derivatives block G1 cell cycle progression
in HCT116 colon cancer cells5,30 and in vascular smooth muscle
cells.31 Since G1 cell cycle progression is controlled by CDK2 and
the inhibition of CDK2 alone may be sufficient to induce G1 cell cy-
cle arrest,32 we tried to perform in vitro CDK2 binding assay. CDK2/
cyclinE was prepared in 8 mM 3-(N-morpholino)propanesulfonic
acid buffer (MOPS, pH7.0, 0.2 mM EDTA, 0.1 mg/ml histone H1,
l
respectively. Molecular volumes increased from the substituents
at position 7 compared to naringenin were calculated using the
Sybyl 7.3 program (Tripos, St. Louis, MO). The volume differences
between naringenin and derivatives N1–N5 were 69.2, 96.7, 61.2,
41.1, and 86.7 Å3, respectively. This result does not show a correla-
tion with the biological activities mentioned above. However, the
expectation based on our previous results may be proved from
the current results,5 because the IC50 value obtained from the
treatment of naringenin was 36.75 lM. As a result, all derivatives
showed much better activities than naringenin, and the bulky sub-
stituent at C-7 position of flavanone can increase the inhibitory
effect on HCT116 colon cancer cells.
Like most flavonoids, the common biological activity of flava-
none is an anti-oxidative effect.28,29 In this experiment, radical-
scavenging effects were tested for the five naringenin derivatives.
The values for derivatives N1–N5 were 11.2%, 14.2%, 15.4%,
20.7%, and 15.9%, respectively, compared to 89.1% for vitamin C.
Under the same conditions, naringenin showed 59.3% anti-
oxidative effect. Thus, modification at position 7 with thiophene-
carboxylate, methylbenzoate, isobutyrate, allyloxy, and phenyl
carbonate groups decreased the radical-scavenging effects drasti-
cally. As mentioned above, because the 5-hydroxy group of
naringenin forms a H-bond with the ketone at C-4, O-substitution
at C-5 is not easy. We tried to prepare an alternative substitution
and designed another naringenin derivative N6 modified at
position 4 in which the ketone group at C-4 was substituted with N-
methylene ethyl carbamate (Fig. 1B). Its structure was determined
based on the same procedure as other derivatives. Its scavenging
effect was 70.6% which shows that a modification of the ketone
group at C-4 increased the radical-scavenging effect. The same clo-
nogenic assay for the naringenin derivative N6 was performed and
10 mM MgAcetate,
c-
33P-ATP) and its specific activity was
500 cpm/pmol. Its enzymatic activity was detected using scintilla-
tion counting. When any inhibitor was not added into the enzyme,
scintillation count was 12,637 whose activity was set to 100%. Ten
micromolar of ATP was added into the enzyme. Naringenin deriv-
atives were provided as substrates. All experiments were iterated
3 times and standard deviation was within 5%. When naringenin
was added into CDK2, the enzyme activity was 85.1% (scintillation
count, 10,749). That is, naringenin showed the 14.9% inhibitory ef-
fect on the enzymatic activity of CDK2 at 10 lM. The inhibitory
activities for derivatives N1–N6 were 83.6%, 48.9%, 42.1%, 84.0%,
42.0%, and 62.3%, respectively. This result does show a correlation
with the IC50 values of derivatives N1–N6 on HCT116 human colon
cancer cells. Likewise, six derivatives showed better inhibitory
activities than naringenin in in vitro CDK2 binding assay as well
as cell based assay.
its IC50 value was 4.35 lM as shown in Figure 2. This result is sig-
nificant for suggesting new structural modification at C-4 and C-5
to increase both the anti-cancer effect and anti-oxidative effect.
From the cell based clonogenic assay, it was concluded that
naringenin derivatives inhibit effectively HCT116 cancer cells, but
To elucidate the possible interaction between naringenin deriv-
atives including naringenin itself and CDK2, molecular modeling
Figure 3. The LigPlot analyses of (left) naringenin docked into CDK2 (2r3j.pdb) and (right) the derivative N1 modified at position 7 with thiophenecarboxylate where red half
circles and green dashed lines denote hydrophobic interactions and H-bonds, respectively.

236
H. Yoon et al. / Bioorg. Med. Chem. Lett. 23 (2013) 232–238
Figure 4. The 3D images of (A) naringenin derivative N6 modified at position 4 in which the ketone group at C-4 was substituted with N-methylene ethyl carbamate docked
into CDK2 (2r3j.pdb), and (B) H-bonds between the CDK2 protein and derivative N6 and residues participating in hydrophobic interactions. The images were constructed
using PyMOL.
was carried out. Naringenin docked well into the CDK2 apo-protein
and its docking score was ꢁ11.97 kcal/mol (Supplementary data
Fig. 8A). Seven residues residing in the binding site of the CDK2 li-
gand (protein data bank accession id: 2r3j.pdb) had hydrophobic
interactions with naringenin: Ile10, Gly11, Gly13, Gln85, Asp86,
Gln131, and Leu134. Three residues, Glu12, Leu83, and His84,
interacted with naringenin via H-bonds (Supplementary data
Fig. 8B). Naringenin derivatives modified at position 7, N1–N5,
docked well into the CDK2 apo-protein too and their docking
scores were ꢁ11.50, ꢁ12.80, ꢁ9.99, ꢁ12.87, and ꢁ13.85 kcal/mol,
respectively (Supplementary data Figs. 9A–E). The LigPlot analysis
of derivative N1 modified at position 7 with thiophenecarboxylate
provided information about the residues forming H-bonds and
hydrophobic interactions with the ligand as shown in Figure 3; five
residues, Ile10, Leu83, Asp86, Gln131, and Leu134 were found in
both naringenin and N1. The docking poses of two ligands docked
into CDK2 were different with each other as shown in Figure 3.
While naringenin formed three H-bonds at 40-O, 5-O, and 7-O
with three residues such as Glu12, Leu83, and His84, respectively,
derivative N1 formed three H-bonds at 40-O, 5-O, and 7-O with two
residues such as Leu83 (two H-bonds) and Glu81. The thiophene-
carboxylate group of derivative N1 was placed inside the hole
formed by Ile10, Gly11, and Glu12. Docking poses of derivatives
N2–N5 docked into CDK2 showed similar results as that of deriva-
tive N1. As shown in Supplementary data Figs. 9C and E, isobuty-
rate group of derivative N3 and phenyl carbonate group of N5
Figure 5. The LigPlot analyses of (left) naringenin docked into CDK2 (2r3j.pdb) and (right) the derivative N6 modified at position 4 where red half circles and green dashed
lines denote hydrophobic interactions and H-bonds, respectively.

H. Yoon et al. / Bioorg. Med. Chem. Lett. 23 (2013) 232–238
237
2. Bleiberg, H.; Vandebroek, A.; Deleu, I.; Vergauwe, P.; Rezaei Kalantari, H.;
D’Haens, G.; Paesmans, M.; Peeters, M.; Efira, A.; Humblet, Y. Acta Gastroenterol.
Belg. 2012, 75, 14.
3. Harborne, J. B. In The Flavonoids: Advances in Research; Harborne, J. B., Ed.;
Chapmann & Hall: London, 1994.
4. Goto, H.; Terao, Y.; Akai, S. Chem. Pharm. Bull. 2009, 57, 346.
5. Woo, Y.; Shin, S. Y.; Hyun, J.; Lee, S. D.; Lee, Y. H.; Lim, Y. Int. J. Mol. Med. 2012,
29, 403.
6. Frydoonfar, H. R.; McGrath, D. R.; Spigelman, A. D. Colorectal Dis. 2003, 5, 149.
7. Kretzschmar, G.; Vollmer, G.; Schwab, P.; Tischer, S.; Metz, P.; Zierau, O. J.
Steroid Biochem. Mol. Biol. 2007, 107, 114.
8. Szkudelska, K.; Nogowski, L.; Nowicka, E.; Szkudelski, T. J. Anim. Physiol. Anim.
Nutr. 2007, 91, 91.
9. Lee, S.; Shin, S. Y.; Lee, Y.; Park, Y.; Kim, B. G.; Ahn, J. H.; Chong, Y.; Lee, Y. H.;
Lim, Y. Bioorg. Med. Chem. Lett. 2011, 21, 3866.
10. Franken, N. A.; Rodermond, H. M.; Stap, J.; Haveman, J.; van Bree, C. Nat. Protoc.
2006, 1, 2315.
11. Tokalov, S. V.; Henker, Y.; Schwab, P.; Metz, P.; Gutzeit, H. O. Pharmacology
2004, 71, 46.
12. Lättig, J.; Böhl, M.; Fischer, P.; Tischer, S.; Tietböhl, C.; Menschikowski, M.;
Gutzeit, H. O.; Metz, P.; Pisabarro, M. T. J. Comput. Aided Mol. Des. 2007, 21,
473.
13. Yoon, H.; Eom, S.; Hyun, J.; Jo, G.; Hwang, D.; Lee, S.; Yong, Y.; Park, J. C.; Lee, Y.
H.; Lim, Y. Bull. Korean Chem. Soc. 2011, 32, 2101.
14. Yong, Y.; Shin, S. Y.; Lee, Y.; Kim, S. H.; Lee, Y. H.; Lim, Y. J. Korean Soc. Appl. Biol.
Chem. 2010, 53, 657.
15. Hyun, J.; Shin, S. Y.; So, K. M.; Lee, Y. H.; Lim, Y. Bioorg. Med. Chem. Lett. 2012, 22,
2664.
16. Park, Y.; Lee, S.; Woo, Y.; Lim, Y. Bull. Korean Chem. Soc. 2009, 30, 1397.
17. Lee, Y.; Lim, Y. Mol. Simul. 2009, 35, 1242.
18. Fischmann, T. O.; Hruza, A.; Duca, J. S.; Ramanathan, L.; Mayhood, T.; Windsor,
W. T.; Le, H. V.; Guzi, T. J.; Dwyer, M. P.; Paruch, K.; Doll, R. J.; Lees, E.; Parry, D.;
Seghezzi, W. Biopolymers 2008, 89, 372.
19. Jez, J. M.; Bowman, M. E.; Dixon, R. A.; Noel, J. P. Nat. Struct. Mol. Biol. 2000, 7,
786.
20. Wallace, A. C.; Laskowski, R. A.; Thornton, J. M. Protein Eng. 1995, 8, 127.
21. Ibrahim, A. R.; Galal, A. M.; Ahmed, M. S.; Mossa, G. S. Chem. Pharm. Bull. 2003,
51, 203.
were not positioned inside the hole. The docking poses obtained
here may be related with their biological activities. While the
IC50 values of three derivatives N1, N2, and N4 on HCT116 cancer
cells were lower than 7
lM, those of N3 and N5 were upper than
15 M. Naringenin derivative N6 modified at position 4 in which
l
the ketone group at C-4 was substituted with N-methylene ethyl
carbamate docked CDK2 well too as shown in Figure 4A. Its dock-
ing score was ꢁ14.82 kcal/mol. There were two H-bonds between
the protein and derivative N6, and ten residues participate in
hydrophobic interactions (Fig. 4B).
Eight residues, Ile10, Glu12, Leu83, His84, Gln85, Asp86,
Gln131, and Leu134 were observed near both naringenin and
derivative N6. While naringenin formed three H-bonds at 40-O, 5-
O, and 7-O with Glu12, Leu83, and His84, respectively as men-
tioned above, derivative N6 formed two H-bonds at 5-O and 7-O
with Leu83 and His84, respectively (Fig. 5). However, while
seven residues participated in the hydrophobic interactions
with naringenin, ten residues did with derivative N6. Although less
H-bonds were observed for derivative N6, more hydrophobic
interactions existed with CDK2. In addition, as shown in Figure
4A, the N-methylene ethyl carbamate group substituted at position
4 in derivative N6 docked inside the binding pocket. These results
may explain why derivative N6 (4.35
lM) showed stronger inhib-
itory effect than naringenin (36.75 M).
l
In silico docking study performed here demonstrates the ratio-
nale for the different inhibitory activities of six naringenin deriva-
tives including five novel compounds. Their inhibitory effects on
HCT116 human colon cancer cells ranged between 1.20
lM and
20.01 M which are much better than naringenin used as a control.
l
22. Spectral data of 5-hydroxy-2-(4-hydroxyphenyl)-4-oxochroman-7-yl thiophene-2-
Besides, in CDK2 binding assay, six derivatives showed better
inhibitory activities than naringenin too. The inhibitory effects of
flavanones on colon cancer cells have been reported,5 but their
structures are quite different with the naringenin derivatives syn-
thesized here. The study on in vitro CDK2 binding assay of flava-
nones has never been reported yet except Western blot analysis
on two flavanone derivatives, silibinin and tomentodiplacone
B.33,34 In conclusion, the current results of cell-based clonogenic
assay on HCT116 colon cancer cells, in vitro CDK2 binding assay,
and in silico CDK2 docking study for naringenin derivatives are
the first time. As expected based on the previous results,5 the bulky
substituent at C-7 position of flavanone can increase the inhibitory
effect on human colon cancer cells as well as on CDK2 activity. In
addition, new structural modification at C-4 results in improving
both the anti-cancer effect and anti-oxidative effect. These findings
could be used for designing cancer therapeutic or preventive flava-
none-derived agents.
carboxylate (N1). Color: yellow sticky liquid; Yield: 63.4%; 1H NMR (400 MHz,
DMSO-d6)
d
12.01 (s, 1H, 5-OH), 9.65 (s, 1H, 40-OH), 8.07 (d, 1H, H-300,
J = 4.0 Hz), 8.00 (d, 1H, H-500, J = 3.8 Hz), 7.35 (d, 2H, H-20/H-60, J = 8.4 Hz), 7.29
(dd, 1H, H-400, J = 3.8, 4.0 Hz), 6.82 (d, 2H, H-30/H-50, J = 8.4 Hz), 6.51 (d, 1H, H-8,
J = 2.0 Hz), 6.50 (d, 1H, H-6, J = 2.0 Hz), 5.58 (dd, 1H, H-2, J = 2.6, 12.9 Hz), 3.42
13
(dd, 1H, H-3, J = 12.9, 17.2 Hz), 2.80 (dd, 1H, H-3, J = 2.6, 17.2 Hz); C NMR
(100 MHz, DMSO-d6) d 198.4 (C-4), 162.5 (C-7), 162.4 (C-5), 159.2 (C-100), 158.0
(C-40), 157.6 (C-9), 135.9 (C-300), 135.7 (C-500), 131.4 (C-10), 128.9 (C-200), 128.5
(C-400), 128.5 (C-20/C-60), 115.4 (C-30/C-50), 106.1 (C-10), 102.8 (C-6), 101.9 (C-
8), 72.0 (C-2), 42.4 (C-3); HRMS (m/z): Calcd for C17H16O4 (M)+: 405.0709;
Found: 405.0407 (Supplementary data Fig. 7).
23. Spectral data of 5-hydroxy-2-(4-hydroxyphenyl)-4-oxochroman-7-yl 2-
methylbenzoate (N2). Color: yellow sticky liquid; Yield: 66.3%; 1H NMR
(400 MHz, DMSO-d6) d 12.00 (s, 1H, 5-OH), 9.64 (s, 1H, 40-OH), 8.03 (d, 1H,
H-400, J = 7.4 Hz), 7.56 (dd, 1H, H-500, J = 7.4, 7.6 Hz), 7.38 (m, 1H, H-600), 7.38 (m,
1H, H-700), 7.37 (d, 2H, H-20/H-60, J = 8.6 Hz), 6.82 (d, 2H, H-30/H-50, J = 8.6 Hz),
6.53 (d, 1H, H-6, J = 2.1 Hz), 6.52 (d, 1H, H-8, J = 2.1 Hz), 5.60 (dd, 1H, H-2,
J = 2.9, 13.1 Hz), 3.42 (dd, 1H, H-3, J = 13.1, 17.2 Hz), 2.81 (dd, 1H, H-3, J = 2.9,
17.2 Hz), 2.56 (s, 3H, 200-CH3); 13C NMR (100 MHz, DMSO-d6) d 198.4 (C-4),
164.3 (C-100), 162.4 (C-5), 162.4 (C-7), 158.2 (C-9), 158.0 (C-40), 140.5 (C-300),
133.3 (C-500), 132.0 (C-700), 131.0 (C-400), 128.6 (C-20/C-60), 128.5 (C-10), 127.9
(C-200), 126.3 (C-600), 115.3 (C-30/C-50), 106.0 (C-10), 103.0 (C-6), 102.1 (C-8),
79.0 (C-2), 42.4 (C-3), 21.3 (300–CH3); HRMS (m/z): Calcd. for C17H16O4 (M)+:
413.1001; Found: 413.1005 (Supplementary data Fig. 10).
Acknowledgments
24. Spectral data of 5-hydroxy-2-(4-hydroxyphenyl)-4-oxochroman-7-yl isobutyrate
(N3). Color: light yellow sticky liquid; Yield: 68.7%; 1H NMR (400 MHz, DMSO-
This work was supported by the Priority Research Centers Pro-
gram (NRF, 2012-0006686), Agenda program (RDA, 8-21-52), the
NRF funded by MEST (2010-0020966), the next generation Bio-
green21 program (RDA, PJ007982), and the NRF 2011-0010745
for KY Jung. Hyuk Yoon, Tae Woo Kim, and Soon Young Shin con-
tributed equally to this work.
d6)
d
12.13 (s, 1H, 5-OH), 10.87 (bs, 1H, 40-OH), 7.57 (d, 2H, H-20/H-60,
J = 8.5 Hz), 7.17 (d, 2H, H-30/H-50, J = 8.5 Hz), 5.94 (d, 1H, H-6, J = 2.0 Hz), 5.91
(d, 1H, H-8, J = 2.0 Hz), 5.59 (dd, 1H, H-2, J = 2.1, 13.0 Hz), 3.26 (dd, 1H, H-3,
J = 13.0, 17.1 Hz), 2.84 (m, 1H, H-200), 2.81 (dd, 1H, H-3, J = 2.1, 17.1 Hz), 1.23 (d,
6H, 200-CH3, J = 7.0 Hz); 13C NMR (100 MHz, DMSO-d6) d 196.0 (C-4), 175.1 (C-
100), 166.9 (C-7), 163.6 (C-5), 162.8 (C-9), 150.8 (C-40), 136.3 (C-10), 128.0 (C-20/
C-60), 122.0 (C-30/C-50), 101.9 (C-10), 96.1 (C-6), 95.2 (C-8), 78.2 (C-2), 42.2(C-
3), 33.4 (C-200), 18.8 (200–CH3); HRMS (m/z): Calcd for C17H16O4 (M)+: 365.1001;
Found: 365.1005 (Supplementary data Fig. 10).
Supplementary data
25. Spectral data of 7-(allyloxy)-5-hydroxy-2-(4-hydroxyphenyl)chroman-4-one (N4).
Color: yellow sticky liquid; Yield: 59.2%; 1H NMR (400 MHz, DMSO-d6) d 12.10
(s, 1H, 5-OH), 9.60 (s, 1H, 40-OH), 7.32 (d, 2H, H-20/H-60, J = 8.5 Hz), 6.81 (d, 2H,
H-30/H-50, J = 8.5 Hz), 6.15 (d, 1H, H-8, J = 2.2 Hz), 6.09 (d, 1H, H-6, J = 2.2 Hz),
5.98 (m, 1H, H-200), 5.47 (dd, 1H, H-2, J = 2.9, 12.9 Hz), 5.38 (dd, 1H, H-300, J = 1.3,
17.3 Hz), 5.26 (dd, 1H, H-300, J = 1.3, 10.5 Hz), 4.61 (m, 1H, H-100), 3.28 (dd, 1H,
H-3, J = 12.9, 17.1 Hz), 2.72 (dd, 1H, H-3, J = 2.9, 17.1 Hz); 13C NMR (100 MHz,
DMSO-d6) d 196.9 (C-4), 166.2 (C-7), 163.1 (C-5), 162.8 (C-9), 157.7 (C-40),
132.8 (C-200), 128.6 (C-10), 128.4 (C-20/C-60), 118.0 (C-300), 115.1 (C-30/C-50),
102.6 (C-10), 95.2 (C-6), 94.3 (C-8), 78.6 (C-2), 68.7 (C-100), 42.0 (C-3); HRMS
(m/z): Calcd for C17H16O4 (M)+: 335.0895; Found: 335.0892 (Supplementary
data Fig. 10).
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2012.10.
130.
References and notes
1. Cunningham, D.; Atkin, W.; Lenz, H. J.; Lynch, H. T.; Minsky, B.; Nordlinger, B.;
Starling, N. Lancet 2010, 20, 1030.

238
H. Yoon et al. / Bioorg. Med. Chem. Lett. 23 (2013) 232–238
26. Spectral data of 5-hydroxy-2-(4-hydroxyphenyl)-4-oxochroman-7-yl phenyl
carbonate (N5). Color: yellow sticky liquid; Yield: 48.0%; 1H NMR (400 MHz,
DMSO-d6) d 12.15 (s, 1H, 5-OH), 10.88 (s, 1H, 40-OH), 7.63 (m, 2H, H-400/H-600),
7.50 (m, 2H, H-20/H-60), 7.46 (m, 2H, H-300/H-700), 7.45 (m, 2H, H-30/H-50), 7.33
(dd, 1H, H-500, J = 7.3, 7.7 Hz), 5.98 (d, 1H, H-8, J = 2.1 Hz), 5.95 (d, 1H, H-6,
J = 2.1 Hz), 5.63 (dd, 1H, H-2, J = 2.9, 12.8 Hz), 3.28 (dd, 1H, H-3, J = 12.8,
17.1 Hz), 2.83 (dd, 1H, H-3, J = 2.9, 17.1 Hz); 13C NMR (100 MHz, DMSO-d6) d
195.8 (C-4), 166.7 (C-7), 163.5 (C-5), 162.6 (C-9), 151.5 (C-200), 150.6 (C-40),
150.6 (C-100), 136.9 (C-10), 129.7 (C-20/C-60), 128.1 (C-400/C-600), 126.5 (C-500),
121.4 (C-300/C-700), 121.2 (C-30/C-50), 101.7 (C-10), 96.0 (C-6), 95.1 (C-8), 77.8
(C-2), 42.0 (C-3); HRMS (m/z): Calcd for C17H16O4 (M)+: 415.0794; Found:
415.0791 (Supplementary data Fig. 10).
27. Spectral data of (E)-ethyl 2-(5,7-dihydroxy-2-(4-hydroxyphenyl)chroman-4-
ylidene)hydrazinecarboxylate (N6). Color: light yellow sticky liquid; Yield:
83.4%; 1H NMR (400 MHz, DMSO-d6) d 12.63 (s, 1H, 5-OH), 10.50 (s, 1H, 7-OH),
9.74 (bs, 1H, NH), 9.74 (bs, 1H, 40-OH), 7.29 (d, 2H, H-20/H-60, J = 8.5 Hz), 6.81
(d, 2H, H-30/H-50, J = 8.5 Hz), 5.91 (d, 1H, H-6, J = 2.2 Hz), 5.87 (d, 1H, H-8,
J = 2.2 Hz), 5.06 (dd, 1H, H-2, J = 11.7, 2.9 Hz), 4.16 (m, 2H, H-200), 3.23 (dd, 1H,
H-3, J = 2.9, 17.1 Hz), 2.78 (dd, 1H, H-3, J = 11.7, 17.1 Hz), 1.24 (m, 3H, H-300);
13C NMR (100 MHz, DMSO-d6) d 160.9 (C-5), 160.3 (C-7), 158.6 (C-40), 157.6 (C-
9), 154.1 (C-100), 150.3 (C-4), 129.9 (C-10), 128.0 (C-20/C-60), 115.2 (C-30/C-50),
98.3 (C-10), 96.8 (C-6), 95.0 (C-8), 75.8(C-2), 61.0 (C-200), 31.7 (C-3), 14.6 (C-300);
HRMS (m/z): Calcd for
(Supplementary data Fig. 10).
C
17H16O4 (M)+: 381.1062; Found: 381.1064
28. Yao, H.; Liao, Z. X.; Wu, Q.; Lei, G. Q.; Liu, Z. J.; Chen, D. F.; Chen, J. K.; Zhou, T. S.
Chem. Pharm. Bull. 2006, 54, 133.
29. Min, B. S.; Thu, C. V.; Dat, N. T.; Dang, N. H.; Jang, H. S.; Hung, T. M. Chem.
Pharm. Bull. 2008, 56, 1725.
30. Lee, D. E.; Lee, K. W.; Jung, S. K.; Lee, E. J.; Hwang, J. A.; Lim, T. G.; Kim, B. Y.;
Bode, A. M.; Lee, H. J.; Dong, Z. Carcinogenesis 2011, 32, 629.
31. Lee, J. J.; Yi, H.; Kim, I. S.; Kim, Y.; Nhiem, N. X.; Kim, Y. H.; Myung, C. S. J.
Ethnopharmacol. 2012, 139, 873.
32. Akiyama, T.; Ohuchi, T.; Sumida, S.; Matsumoto, K.; Toyoshima, K. Proc. Natl.
Acad. Sci. U.S.A. 1992, 89, 7900.
33. Kollár, P.; Bárta, T.; Závalová, V.; Smejkal, K.; Hampl, A. Br. J. Pharmacol. 2011,
162, 1534.
34. Deep, G.; Singh, R. P.; Agarwal, C.; Kroll, D. J.; Agarwal, R. Oncogene 2006, 25,
1053.