Isolation, identification, and biological evaluation of HIF-1-modulating compounds from Brazilian green propolis

Article abstract of DOI:10.1016/j.bmc.2011.07.060

The tumor microenvironment is characterized by hypoxia, low-nutrient levels, and acidosis. A natural product chemistry-based approach was used to discover small molecules that modulate adaptive responses to a hypoxic microenvironment through the hypoxia-inducible factor (HIF)-1 signaling pathways. Five compounds, such as baccharin (3), beturetol (4), kaempferide (5), isosakuranetin (6), and drupanin (9), that modulate HIF-1-dependent luciferase activity were identified from Brazilian green propolis using reporter assay. Compounds 3, 9 and 5 reduced HIF-1-dependent luciferase activity. The cinnamic acid derivatives 3 and 9 significantly inhibited expression of the HIF-1α protein and HIF-1 downstream target genes such as glucose transporter 1, hexokinase 2, and vascular endothelial growth factor A. They also exhibited significant anti-angiogenic effects in the chick chorioallantoic membrane (CAM) assay at doses of 300 ng/CAM. On the other hand, flavonoids 4 and 6 induced HIF-1-dependent luciferase activity and expression of HIF-1 target genes under hypoxia. The contents (g/100 g extract) of the HIF-1-modulating compounds in whole propolis ethanol extracts were also determined based on liquid chromatography-electrospray ionization mass spectrometry as 1.6 (3), 14.2 (4), 4.0 (5), 0.7 (6), and 0.7 (9), respectively. These small molecules screened from Brazilian green propolis may be useful as lead compounds for the development of novel therapies against ischemic cardiovascular disease and cancer based on their ability to induce or inhibit HIF-1 activity, respectively.

Full text of DOI:10.1016/j.bmc.2011.07.060

Bioorganic & Medicinal Chemistry 19 (2011) 5392–5401
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Isolation, identification, and biological evaluation of HIF-1-modulating
compounds from Brazilian green propolis
Hisanori Hattori a, Kensuke Okuda a, Tetsuji Murase a, Yuki Shigetsura a, Kosuke Narise a,
Gregg L. Semenza b,c,d,e,f,g,h, Hideko Nagasawa a,
⇑
a Laboratory of Medicinal & Pharmaceutical Chemistry, Gifu Pharmaceutical University, 1-25-4 Daigaku-nishi, Gifu 501-1196, Japan
b Vascular Program, Institute for Cell Engineering, The Johns Hopkins University School of Medicine, Baltimore, MD 21205, USA
c Department of Pediatrics, The Johns Hopkins University School of Medicine, Baltimore, MD 21205, USA
d Department of Medicine, The Johns Hopkins University School of Medicine, Baltimore, MD 21205, USA
e Department of Oncology, The Johns Hopkins University School of Medicine, Baltimore, MD 21205, USA
f Department of Radiation Oncology, The Johns Hopkins University School of Medicine, Baltimore, MD 21205, USA
g Department of Biological Chemistry, The Johns Hopkins University School of Medicine, Baltimore, MD 21205, USA
h McKusick-Nathans Institute of Genetic Medicine, The Johns Hopkins University School of Medicine, Baltimore, MD 21205, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The tumor microenvironment is characterized by hypoxia, low-nutrient levels, and acidosis. A natural
product chemistry-based approach was used to discover small molecules that modulate adaptive
responses to a hypoxic microenvironment through the hypoxia-inducible factor (HIF)-1 signaling path-
ways. Five compounds, such as baccharin (3), beturetol (4), kaempferide (5), isosakuranetin (6), and drup-
anin (9), that modulate HIF-1-dependent luciferase activity were identified from Brazilian green propolis
using reporter assay. Compounds 3, 9 and 5 reduced HIF-1-dependent luciferase activity. The cinnamic
Received 12 July 2011
Revised 28 July 2011
Accepted 28 July 2011
Available online 4 August 2011
Keywords:
acid derivatives 3 and 9 significantly inhibited expression of the HIF-1a protein and HIF-1 downstream
HIF-1 modulator
Angiogenesis inhibitor
Brazilian green propolis
Drupanin
target genes such as glucose transporter 1, hexokinase 2, and vascular endothelial growth factor A. They
also exhibited significant anti-angiogenic effects in the chick chorioallantoic membrane (CAM) assay at
doses of 300 ng/CAM. On the other hand, flavonoids 4 and 6 induced HIF-1-dependent luciferase activity
and expression of HIF-1 target genes under hypoxia. The contents (g/100 g extract) of the HIF-1-modu-
lating compounds in whole propolis ethanol extracts were also determined based on liquid chromatog-
raphy–electrospray ionization mass spectrometry as 1.6 (3), 14.2 (4), 4.0 (5), 0.7 (6), and 0.7 (9),
respectively. These small molecules screened from Brazilian green propolis may be useful as lead com-
pounds for the development of novel therapies against ischemic cardiovascular disease and cancer based
on their ability to induce or inhibit HIF-1 activity, respectively.
Tumor microenvironment
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
tumor microenvironment. Hypoxia-inducible factor (HIF)-1 has
emerged as a key mediator of tumor adaptation and survival in
The tumor microenvironment has attracted much attention as a
target for the development of novel cancer treatment strategies
because it may be a general target to treat solid tumors regardless
of cancer types. Therefore, we have been focusing on the solid
tumor microenvironment, characterized by hypoxia, low nutrient
availability, and acidosis due to inefficient perfusion for drug
discovery of tumor-selective anticancer agents.1,2 The tumor
microenvironment has been recognized as a major factor that not
only influences the response to conventional anticancer therapies
but also promotes invasion and metastasis.3,4 In particular, hypoxia
is now considered a fundamentally important characteristic of
such a deprived microenvironment.5 HIF-1 modulating compounds
should affect adaptive responses in the tumor microenvironment
such as angiogenesis, metabolic reprogramming, and metastasis.
Natural products continue to provide promising lead com-
pounds and drug candidates in modern antitumor drug discovery
and also serve as useful molecular probes to elucidate signal trans-
duction pathways that control the expression of genes in physio-
logical and pathological states.6 Most compounds that modulate
HIF-1 are natural products or synthetic compounds derived from
natural product leads.7–9 Propolis, produced by honeybees from
various plant sources, is a natural resinous hive product and a good
source of bioactive polyphenols. It has a broad spectrum of biolog-
ical activities.10 Especially, extracts of Brazilian green propolis and
Chinese red propolis and their polyphenolic constituents, such as
artepillin C and caffeoylquinic acid derivatives, have angiostatic
⇑
Corresponding author. Tel./fax: +81 58 230 8112.
E-mail address: hnagasawa@gifu-pu.ac.jp (H. Nagasawa).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.07.060

H. Hattori et al. / Bioorg. Med. Chem. 19 (2011) 5392–5401
5393
effects through suppression of the vascular endothelial growth fac-
tor (VEGF) signaling pathway.11–13 Furthermore, because flavo-
noids/polyphenols from natural products belong to the largest
group of HIF-1 inhibitors,14–16 we recognized propolis as a useful
natural source to be investigated for exploring drug seeds targeting
the tumor microenvironment. We therefore evaluated the effects
of components of Brazilian green propolis extract, which contains
various bioactive polyphenolic compounds, on cellular responses
to hypoxia for the development of a novel anticancer drug target-
ing the tumor microenvironment. In this study, we report the iso-
lation and identification of HIF-1-modulating compounds from
Brazilian green propolis extract by using a cell-based luciferase re-
porter assay17–19 to assess their effect on HIF-1 transcriptional
activity. We also performed chick chorioallantoic membrane
(CAM) assay to evaluate anti-angiogenic effects.
the luciferase coding sequences controlled by a hypoxia response
element from the human enolase 1 gene.20 At first, an ethanol ex-
tract was evaluated for a HIF-1-modulating effect in cells exposed
to hypoxia (1% O2). The extract of Brazilian green propolis in-
creased luciferase activity induced by hypoxia by one half com-
pared with that of the control at a concentration of 20 lg/mL
(Fig. 1A). Then, the extract was subjected to a silica gel column
chromatography to separate 14 fractions (frs.); among them, frs.
A3, A8, and A9 exhibited a significantly increased luciferase activ-
ity, whereas frs. A6 and A10–A13 resulted in decreased luciferase
activity (Fig. 1B). Further the luciferase assay-guided chromato-
graphic separation of these positive fractions (Fig. 2) gave nine
known compounds as shown in Figure 3. On the basis of spectral
data in comparison with spectra in the literature, the structures
of the compounds were identified as lupeol-3-(30R-hydroxy)-hexa-
decanoate (1),21 artepillin C (2),22 baccharin (3),23 beturetol (4),24
kaempferide (5),25 isosakuranetin (6),26 (E)-4-(2,3-dihydrocinna-
moyloxy)cinnamic acid (7),27 dihydrokaempferide (8),26 and drup-
anin (9)28 (Fig. 3; see also the chemical and spectral data in the
Supplementary data). All isolated compounds (except for com-
pound 1, whose solubility in dimethylsulfoxide (DMSO) was extre-
mely poor) were tested for their effects on HIF-1 transcriptional
2. Results and discussion
2.1. Isolation and identification of HIF-1-modulating
compounds
To evaluate the effects of propolis constituents on cellular adap-
tation to a hypoxic microenvironment, we established human
embryonic kidney (HEK) 293 stable reporter cell lines expressing
activity under hypoxic conditions at a concentration of 50
lM
(Fig. 4). Treatment of cells with compound 3, 5, or 9, which were
A
1.5
1.0
0.5
0.0
Hypoxia (1% O2)
+
-
+
+
+
5.0
+
10
+
20
1.25 2.5
Whole Ethanol Extract (µg/mL)
B
1.5
1.0
0.5
0.0
Hypoxia (1% O2)
Fraction
+
-
-
-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
A1 A2 A3 A4 A5 A6 A7 A8 A9 A10A11A12A13A14
20 µg/mL
Figure 1. Effect of the extract and the separated fractions of Brazilian green propolis on HIF-1 transactivation. (A) The whole ethanol extract of Brazilian green propolis was
evaluated by luciferase assay. (B) Fourteen fractions separated by silica gel column chromatography described in Figure 2 were each evaluated at 20 g/mL. Averages from
l
one representative experiment performed in triplicate are shown, and the error bars indicate standard deviations. An asterisk indicates the statistical significance compared
with the hypoxic control; ⁄p <0.05.

5394
H. Hattori et al. / Bioorg. Med. Chem. 19 (2011) 5392–5401
Brazilian Propolis
Extracted with EtOH, then evaporated
EtOH extract (59.0 g)
A
A3 (1.7 g) A5 (2.5 g) A6 (3.4 g) A8 (5.1 g)
A10 (2.0 g)
G
A11-13 (11.4 g)
A
B, C
B, D
B, E
G
B
1
2
3
4
5
C4
C6
B7
B, F
B8
B, E
F
B, F
A; CC over silica gel eluted with CHCl3-acetone
B; CC over silica gel eluted with n-hexane-acetone
C; Recrystallization (acetone)
6
7
8
9
D; Recrystallization (n-hexane-EtOAc)
E; Recrystallization (n-hexane-acetone)
F; Recrystallization (CHCl3)
G; Filtration and wash using CHCl3
Figure 2. Flowchart for the isolation of chemical constituents from Brazilian green propolis.
O
H
R2
R4
R1
O
H
OH
O
R3
O
12
2: R1 = H, R2 = prenyl, R3 = prenyl, R4 = OH
O
1
3: R1 = H, R2 = prenyl, R3 = H, R4 =
O
Ph
O
7: R1 = H, R2 = H, R3 = H, R4 =
O
Ph
9: R1 = H, R2 = prenyl, R3 = H, R4 = OH
OCH3
OCH3
HO
O
O
HO
O
O
R
OH
R
OH
OH
4
5
: R = OCH3
: R = H
6
8
: R = H
: R = OH
Figure 3. Structures of the compounds isolated from Brazilian green propolis.
isolated from fractions with inhibitory effects, resulted in signifi-
cantly reduced luciferase activity. On the other hand, treatment
of cells with flavonoids 4 and 6 resulted in a significant increase
in luciferase activity under hypoxic condition. Furthermore, these
HIF-1-modulating effects occurred in a dose-dependent manner
(Fig. 5). The cytotoxicity of the active compounds was evaluated
by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
(MTT) and clonogenic assays. At the maximum concentration of
the compounds that dissolved in the medium, all active com-
pounds had weak or no cytotoxic effects (Table 1). Although the
MTT and clonogenics assays were also performed under hypoxic
conditions, there were no differences in the results compared with
those observed under normoxic conditions (Table 1).
Among the 4-hydroxycinnamic acid derivatives 2, 3, 7, and 9,
compounds 3 and 9 showed HIF-1 inhibitory effects. To evaluate
the pharmacophore of the 4-hydroxycinnamate derivatives, we

H. Hattori et al. / Bioorg. Med. Chem. 19 (2011) 5392–5401
5395
µ
Figure 4. Effects of isolated compounds from Brazilian green propolis on hypoxia-
induced HIF-1-dependent luciferase reporter gene activity. Compounds 2–9 were
examined at
a concentration of 50 lM. Mean data from one representative
experiment performed in triplicate are shown, and the error bars indicate standard
deviations. The asterisks indicate the statistical significance compared with the
hypoxic control; ⁄p <0.05, ⁄⁄p <0.01.
Figure 5. Dose–response relationship of 3–6 and 9 in the HIF-1-dependent reporter
assay. Data shown are means from one experiment performed in triplicate, and the
bars represent standard deviations. An asterisk indicates statistical significance
compared with the hypoxic control; ⁄p <0.05.
modified
9 to obtain the 4-methoxy-3-prenyl cinnamic acid
methyl ester (10), 4-hydroxy-3-prenyl cinnamic acid methyl ester
(11), and 4-methoxy-3-prenyl cinnamic acid (12), and synthesized
cinnamic acid derivatives 17 and 18 from commercially available
reagents, respectively (Table 2; see also Scheme S1 in Supplemen-
tary data). The methyl ester derivatives 10 and 11 bearing conju-
gated ester as a Michael acceptor were highly cytotoxic, also as
expected from their higher hydrophobicity as evidenced by the
c log D values (4.31 and 4.03, respectively) (Table 2). Methyl ether
12 was more hydrophilic and non-cytotoxic at a maximum concen-
Interestingly, we found that flavonoids 4 and 6 enhanced HIF-1
transcriptional activity even under hypoxic condition. Because
PHDs were inactivated under hypoxic condition, their inducing ef-
fects on HIF-1 transcriptional activities in hypoxia may not be
caused by inhibition of PHDs but by some direct affects on the
tration of 100
lM. But, it did not show a significant inhibiting ef-
structure or functions of HIF-1a protein. While, compound 5 re-
fect at a concentration of 50
l
M in the HIF-1-dependent reporter
duced HIF-1-dependent luciferase activities moderately as in the
case of quercetine.15 Common structure of HIF-1 inhibiting flavo-
noids shown in the previous report15 and 5 is 3,5,7-trihydroxy-
1,4-benzopyrone moiety. In contrast, with the presence of only
6-methoxy group, flavonoid 4 showed a significant inducing effect
of HIF-1 transcriptional activities.
Consequently, hydroxycinnamic acid and flavonoid seem to be
interesting and unique scaffolds of HIF-1 modulator, while it is
necessary to further investigate the structure-activity relationship.
assay, suggesting the hydrogen donor ability of 4-hydroxy group
was necessary for their activity. Compound 3 may become active
after metabolic hydrolysis of dihydrocinnamate ester to reveal 4-
hydroxyl group. 3-Prenyl group was also important because nei-
ther corresponding analogs with 3-allyl (17), t-butyl (18) group
nor 3-desprenyl analog (13) showed any HIF-1 inhibition. Com-
pound 9 showed the most potent HIF-1 inhibitory effect among
all polyphenols tested in this study (Table 2). These data suggested
that hydrogen bond donor abilities of phenol, conjugated carbox-
ylic acid moiety of 4-hydroxycinnamic acids, and monoprenyl
group were important for HIF-1 inhibitory effect.
2.2. Inhibition of expression of HIF-1
a
protein and HIF-1 target
genes
Next, the essential moiety for modulating effects of flavonoids
(4–6 and 8) on HIF-1 activity are quite interesting. The common
feature among them is the chelating moiety at 5-hydroxyl 4-oxo
group.29 It is known that flavonoids with such a coordination site
Immunoblot analyses of the HIF-1
a
protein expression in hu-
man colorectal cancer cells treated with compounds 3 and 9 are
shown in Figure 6A and B. HCT116 cells were exposed to the com-
pounds under normoxic (20% O2) or hypoxic conditions for 4 h.
induce accumulation of HIF-1
their iron-chelating properties. They would inhibit HIF-1
hydroxylases (PHDs), which play a key role for HIF-1 degradation
and contain Fe(II) in their catalytic center, resulting in stabilization
of HIF-1
in normoxia.30,31 On the other hand, Hasebe et al.
a
under normoxic condition due to
a
prolyl
Compounds 3 and 9 reduced the expression of the HIF-1a protein
a
induced by hypoxia in a dose-dependent manner (Fig. 6A and B) at
concentrations that are comparable with those that inhibited HIF-
1-dependent luciferase activity as already shown in Figure 5. Their
effects on HIF-1-dependent induction of the glucose transporter 1
(GLUT1), hexokinase 2 (HK2) and VEGFA gene expression in hypoxic
HCT116 cells were also investigated by real-time reverse transcrip-
tion-polymerase chain reaction (RT-PCR). Treatment of cells with
compounds 3 and 9 for 24 h equally significantly reduced the
a
showed that some flavonoids including quercetin and homoisofl-
avonoids which possessed 5-hydroxyl 4-oxo group inhibited
HIF-1-dependent reporter activity under hypoxic condition.15
These apparently discrepant reports indicate that the HIF-1 modu-
lating effects of the flavonoids depend on oxygen concentration.

5396
H. Hattori et al. / Bioorg. Med. Chem. 19 (2011) 5392–5401
Table 1
Cytotoxicity and hydrophobicity of isolated compounds under hypoxic (1% O2) or normoxic (20% O2) conditions
Compounds
Concn (
l
M)
MTT assaya
Hypoxiab
Clonogenic assaya
Hypoxiab
c log D (pH 7.0)c
Normoxia
Normoxia
3
4
5
6
9
200
50
50
50
200
69.4 2.0
86.8 3.3
173.3 25.6
NDd
58.2 4.7
69.0 2.4
102.5 16.0
103.4 9.7
127.1 9.4
74.0 6.2
90.3 6.6
87.6 6.8
104.7 7.3
101.2 2.0
74.8 2.3
96.8 2.6
79.9 5.2
99.0 7.1
105.8 2.2
3.75
3.21
2.96
2.57
1.17
146.0 4.6
a
b
c
Viability (%) at indicated concentrations.
1% O2.
Caluculated by Accord for Excel ver. 7.1.5 (Accelrys, Inc.).
Not determined.
d
Table 2
Biological activity and hydrophobicity of cinnamic acid derivatives
O
R2
R4
R1
O
R3
Compounds
R1
H
R2
R3
R4
HIF-1-dependenta
Luciferase activity
MTT assayb
Viavility ( M)
c log D (pH 7.0)c
l
2
Prenyl
Prenyl
OH
1.09 0.03
0.52 0.06
53.3d (200)
2.53
O
3
7
H
H
Prenyl
H
H
H
58.2 (200)
86.0d (100)
3.19
1.27
77.4 25.1e
1.08
O
9
H
Prenyl
H
OH
0.36 0.03
25.6 12.4e
1.17 0.23f
NDg
0.82 0.08
0.93 0.07
0.90 0.10
0.77 0.04
127.1 (200)
1.12
10
11
12
13
17
18
CH3
CH3
H
H
H
Prenyl
Prenyl
Prenyl
H
Allyl
tBu
H
H
H
H
H
tBu
OCH3
OH
OCH3
OH
OH
OH
37.6 3.7e
10.6 1.2e
95.7(100)
99.3 (100)
98.4 (100)
108.4 (100)
4.31
4.03
1.63
ꢀ0.59
0.73
H
2.84
a
b
c
d
e
f
Show the ratio of hypoxic control using HEK293 p2.1 #3 cells at concentration of 50
Using HCT116 cells.
Caluculated by Accord for Excel ver. 7.1.5.
Using HEK293 p2.1 #3 cells.
lM.
IC50 valus (
Activity at 30
l
M).
M.
Not determined.
l
g
expression of GLUT1, HK2, and VEGFA mRNA induced by hypoxia at
concentrations of 50 M (Fig. 6C–E). On the other hand, the HIF-1-
inducing flavonoids 4 and 6 showed a tendency to induce all HIF-1
target genes both under hypoxic and normoxic conditions. Com-
pound 5 did not show significant suppression of the HIF-1a protein
and 9 suppressed angiogenesis similarly to that of artepillin C (2)
at concentrations of 300 ng/CAM. However, compound 2 did not
inhibit HIF-1-dependent luciferase activity (Fig. 4). On the other
hand, flavonoids 4, 5, and 6 did not show significant effects
(Table 3).
l
(data not shown), but GLUT1 induction by hypoxia was reduced as
much as those by compounds 3 and 9. However, expression of the
VEGFA gene was most potently induced by compound 5 among
them by twice as much as that of hypoxic control. As mentioned
above, the effect of flavonoid 5 on the expression of HIF-1 related
genes were not consistent with those of cinnamate analog HIF-1
inhibitors (3 and 9), suggesting that flavonoid and hydroxycin-
namic acid might modulate HIF-1 via different pathway based on
their unique scaffold.
2.4. Quantitative analysis of Brazilian green propolis by liquid
chromatography–electrospray ionization mass spectrometry
(LC–ESI-MS)
To estimate the contents of the HIF-1-modulation constituents
from Brazilian green propolis, quantitative analysis by LC–ESI-MS
was performed. Reversed-phase high performance LC of the Brazil-
ian green propolis ethanol extract gave good separation of 3–6 and
9 using the modified conditions in de Sousa’s method,32 and the
compounds were identified by ESI-MS (Fig. 8; see also ESI-MS data
in the Supplementary data). Their contents (g/100 g extract), esti-
mated using an internal standard calibration method, were 1.6
(3), 14.2 (4), 4.0 (5), 0.7 (6), and 0.7 (9), respectively. Veratralde-
hyde was used as the internal standard. Compound 4, which signif-
icantly enhanced HIF-1 transcriptional activity (Fig. 4), had the
2.3. Angiogenesis inhibition by CAM assay
The anti-angiogenic effects of the compounds were evaluated
in vivo by a CAM assay. Because artepillin C (2) was previously
identified to be an angiogenesis inhibitor,11 it was also examined.
As shown in Figure 7, the hydroxycinnamic acid derivatives 3

H. Hattori et al. / Bioorg. Med. Chem. 19 (2011) 5392–5401
5397
highest proportion in the extract. Thus, compound 4 may have sig-
nificantly contributed to the net increased HIF-1-dependent lucif-
erase activity of the whole extract (Fig. 1A).
3. Conclusion
We have identified five compounds from Brazilian green prop-
olis that modulate HIF-1 activity. The results of biological evalua-
tions of the HIF-1-modulating compounds are summarized in
Table 3. None of the compounds showed significant cytotoxicity.
We also found that the hydroxycinnamic acid derivatives 3 and 9
from Brazilian green propolis inhibited not only HIF-1 transcrip-
tional activity but also the hypoxia-induced expression of the
HIF-1a protein and downstream target genes, such as GLUT1,
HK2, and VEGFA. Furthermore, the HIF-1 inhibitory compounds
showed angiogenesis inhibition. Additional studies are warranted
to determine whether they have antitumor effects in xenograft
models as demonstrated for other HIF-1 inhibitors.17–19 On the
other hand, the flavonoid 4 induced HIF-1 transcriptional activity
under normoxia as well as hypoxia. This compound may be useful
in the treatment of ischemic cardiovascular disease.33
4. Materials and methods
4.1. General experimental procedures
1H and 13C NMR spectra were recorded using a JEOL JNM-AL400
or JNM-ECA500 spectrometer at 400 or 500 MHz (1H NMR) and 100
or 125 MHz (13C NMR) in deuterated chloroform (CDCl3), DMSO-d6
or acetone-d6. Chemical shifts for 1H NMR were referenced to tet-
ramethylsilane (0.00 ppm). Chemical shifts for 13C NMR were cali-
brated to the solvent signals (CDCl3: 77.0 ppm, DMSO-d6:
39.5 ppm, and acetone-d6: 29.8 ppm). The electron impact (EI)-
MS, high resolution (HR)EI-MS, direct analysis in real time
(DART)-MS and HRDART-MS measurements were performed on a
JEOL JMS-SX102A and JMS-T100TD. Column chromatography was
performed with normal-phase silica gel (AP-300; Daico Trading
Co., Ltd.). LC-MS spectra were measured using a HP 1100 Series
and a HP LC/MSD (Agilent Technologies). Unless otherwise noted,
all chemicals were obtained from commercial sources and used
without further purification.
4.2. Chemicals
4.2.1. Materials
Brazilian green propolis ethanol extract (#Y080527, collected in
Minas Gerais, Brazil) was obtained from Yamada Bee Farm (Oka-
yama, Japan). 4-Hydroxycinnamic acid (13) was purchased from
Sigma-Aldrich Japan and used for biological studies as well as
starting material of 17 without further purification.
4.2.2. Isolation and identification
The residue from the propolis ethanol extract (59.0 g) was first
applied to silica gel column eluted with chloroform (CHCl3)–ace-
tone (CHCl3?20:1?5:1?1:1) and acetone (each volume; 2 L, Col-
umn A, Fig. 2) to separate 14 fractions. Fraction A3 (1.7 g) was
purified by repeated silica gel chromatography using n-hexane–
acetone (40:1?10:1? 5:1?1:1) and recrystallized (acetone) to
give 1 (153.6 mg). Fraction A5 (2.5 g) was purified by silica gel col-
umn chromatography using n-hexane–acetone (20:1?10:1?
5:1?1:1) and recrystallized (n-hexane–ethyl acetate) to give 2
(207.4 mg). Fraction A6 (3.4 g) was separated by silica gel column
chromatography using n-hexane–acetone (20:1?10:1?5:1?1:1)
Figure 6. Effects of the test compounds on HIF-1 pathway in HCT116 human
colorectal cancer cells. HIF-1 protein expression availability in HCT116 cells.
Immunoblot assays of HIF-1 and b-actin proteins following 4 h treatment with the
a
a
test compounds, 3 (A), 9 (B), and under hypoxic (1% O2) or normoxic (20% O2)
conditions. Compounds 3–6 and 9 regulated the induction of the HIF-1 target gene
expression in HCT116 cells. Quantitative real-time RT-PCR analysis of GLUT1 (C),
HK2 (D), and VEGFA (E) mRNA levels upon treatment with each compound for 24 h
under hypoxic or normoxic conditions. The data (mean SD) were normalized to an
internal control (18S rRNA), and the relative expression levels were determined by
the DDCT method (see Section 4.3.8). The p values are shown where there is a
statistically significant difference (⁄p <0.05, ⁄⁄p <0.01) between the hypoxic control
and the compound-treated samples.
and recrystallized (n-hexane–acetone) to give
3 (102.4 mg).

5398
H. Hattori et al. / Bioorg. Med. Chem. 19 (2011) 5392–5401
A
B
C
D
Figure 7. Anti-angiogenic effects of compounds 2, 3, and 9. Each compound was loaded on CAMs of chick embryos at 5 days of age. After 48-h incubation, fat emulsion was
injected under the CAMs to make the vascular network clear for photography. (A) control (2% DMSO); (B) 2 (300 ng/CAM); (C) 3 (300 ng/CAM); and (D) 9 (300 ng/CAM). Each
angiogenesis inhibition ratio (%; mean SD) is also shown under photography.
Table 3
Summary of biological evaluations of HIF-1-modulating compounds from Brazilian green propolis
Compounds
Reporter assaya
HIF-1
Western blot
Real-time RT-PCR
CAM assayb
HIF-1
a
GLUT1
HK2
VEGFA
c
3
4
5
6
9
;
;
50
;
50
;
;
83.2 8.0
46.7 8.1
39.5 9.6
38.1 5.7
84.3 5.0
77.4 25.1
l
l
l
M
lM
lM
lM
lM
lM
d
e
f
f
g
"
"
"
"
30
M
M
30
;
;
"
30
30
f
"
NDh
"
30
;
50
;
;
50
;
;
25.6 12.4
l
M
a
b
c
IC50 value (lM).
Inhibition % of angiogenesis at 300 ng/CAM.
;, Suppression under hypoxia.
", Induction under hypoxia.
, Difference is not significant.
Under hypoxia and normoxia.
Under normoxia.
d
e
f
g
h
Not determined.
was purified by silica gel column chromatography using n-hex-
ane–acetone (5:1?1:1) and recrystallized (CHCl3) to give
6
(20.0 mg). Fraction B8 (132.5 mg) was recrystallized (n-hexane–
acetone) to give 7 (10.4 mg). Fractions A11–13 (11.4 g) were sepa-
rated by silica gel chromatography using CHCl3–acetone
(40:1?5:1?1:1) (each volume; 1 L, Column C) to give 7 fractions.
Fraction C4 (712.0 mg) was recrystallized (CHCl3) to give
8
(180.0 mg). Fraction C6 (4.6 g) was purified by silica gel column
chromatography using n-hexane–acetone (5:1?1:1) and recrystal-
lized (CHCl3) to give 9 (188.0 mg). The structures of the isolated
compounds were identified on the basis of physicochemical prop-
erties and spectral data comparison with spectra published in the
literatures (see the chemical and spectral data in the Supplemen-
tary data).
4.2.3. Syntheses of cinnamic acid derivatives
Figure 8. Chromatographic profile of the Brazilian green propolis at 280 nm
absorption revealed the following compounds (g/100 g) in extract of Brazilian green
propolis: 3, 1.6; 4, 14.2; 5, 4.0; 6, 0.7; and 9, 0.7. Internal standard (IS):
veratraldehyde.
4.2.3.1. 3-(4-Methoxy-3-prenylphenyl)acrylic acid methyl ester
(10)34
. Compound 9 (26.2 mg, 0.11 mmol) was dissolved in
0.5 mL of methanol, then TMSCH2N2 (0.5 mL) was added and stir-
red at room temperature for overnight. Purification by column
chromatography on silica gel using n-hexane–acetone (20:1) as
eluted yield 24.0 mg (82%) of 10. Colorless oil; 1H NMR (CDCl3): d
1.71 (3H, s), 1.76 (3H, s), 3.31 (2H, d, J = 7.6 Hz), 3.79 (3H, s), 3.86
(3H, s), 5.29 (1H, t, J = 7.6 Hz), 6.30 (1H, d, J = 16.1 Hz), 6.83 (1H,
d, J = 8.4 Hz), 7.32 (1H, d, J = 2.3 Hz), 7.35 (1H, dd, J = 8.4, 2.3 Hz),
7.64 (1H, d, J = 16.1 Hz); 13C NMR (CDCl3): d 167.8, 159.2, 145.0,
133.1, 130.1, 128.9, 127.7, 126.7, 121.7, 114.8, 110.2, 55.4, 51.4,
Fraction A8 (5.1 g) was suspended in CHCl3, and the residue was
filtered and washed to give 4 (30.7 mg). Fraction A10 (2.0 g) was
suspended in CHCl3, the residue filtered and washed to give 5
(350.3 mg), and then subjected to silica gel column chromatogra-
phy using n-hexane–acetone (10:1?5:1?1:1) (each volume;
200 mL, Column B) to give 10 fractions. Fraction B7 (145.4 mg)

H. Hattori et al. / Bioorg. Med. Chem. 19 (2011) 5392–5401
5399
28.2, 25.7, 17.7; DARTMS [M+H]+ m/z 261; HRDARTMS [M+H]+ m/z:
261.1488 (calcd for C16H21O3: 261.1491).
4.2.3.6. 3-(4-Hydroxy-3-allylphenyl)acrylic acid methyl ester
(16)39
Compound 15 (54.2 mg, 0.25 mmol) was dissolved in
.
3.0 mL of N,N-diethylaniline, and the solution was heated to reflux.
After reaction completion (5 h) the mixture was brought to pH 4 by
the addition of 2 N HCl and extracted three times with Et2O and the
combined organic layer was washed with brine and dried over
anhydrous Na2SO4. The solvent was evaporated under reduced
pressure. The residue was purified by prep. TLC (n-hexane–
EtOAc = 5:1) to afford 16 as a brown oil (35.3 mg, 65%). 1H NMR
(CDCl3): d 3.42 (2H, d, J = 6.1 Hz), 3.80 (3H, s), 5.17 (1H, dd,
J = 7.6, 1.5 Hz), 5.20 (1H, t-like, J = 1.5 Hz), 5.52 (1H, s), 5.97–6.00
(1H, m), 6.31 (1H, d, J = 16.1 Hz), 6.83 (1H, d, J = 7.6 Hz), 7.31 (1H,
br s), 7.32 (1H, dd, J = 7.6, 3.1 Hz), 7.63 (1H, d, J = 16.1 Hz); 13C
NMR (CDCl3): d 168.2, 156.4, 145.0, 135.8, 130.5, 128.2, 127.1,
126.2, 116.9, 116.1, 114.9, 51.7, 34.7; DARTMS [M+H]+ m/z 219;
HRDARTMS [M+H]+ m/z: 219.0997 (calcd for C13H15O3: 219.1021).
4.2.3.2. 3-(4-Hydroxy-3-prenylphenyl)acrylic acid methyl ester
(11)34
. Compound 9 (91.0 mg, 0.39 mmol) was dissolved in
0.5 mL of methanol, then TMSCH2N2 (0.5 mL) was added and stir-
red at 0 °C for 30 min. Purification by column chromatography on
silica gel using n-hexane–EtOAc (5:1) as eluted yield 63.7 mg
(66%) of 11. White powder; mp 86 °C (lit. 78–82 °C)35 1H NMR
;
(CDCl3): d 1.78 (3H, s), 1.79 (3H, s), 3.36 (2H, d, J = 7.6 Hz), 3.80
(3H, s), 5.31 (1H, t, J = 7.6 Hz), 5.68 (1H, br s), 6.29 (1H, d,
J = 16.1 Hz), 6.81 (1H, d, J = 8.4 Hz), 7.29 (1H, br s), 7.30 (1H, dd,
J = 8.4, 2.3 Hz), 7.63 (1H, d, J = 16.1 Hz); 13C NMR (CDCl3): d
168.3, 156.6, 145.3, 135.0, 134.9, 130.1, 127.7, 126.9, 121.2,
116.0, 114.6, 51.7, 29.2, 25.8, 17.8; DARTMS [M+H]+ m/z 247;
HRDARTMS [M+H]+ m/z: 247.1330 (calcd for C15H19O3: 247.1334).
4.2.3.3. 3-(4-Methoxy-3-prenylphenyl)acrylic acid (12)34
pound 10 (47.9 mg, 0.18 mmol) was dissolved in 3.0 mL of THF,
then KOH (25.0 mg, 0.45 mmol) and H2O (50 L) were added, and
.
Com-
4.2.3.7. 3-(4-Hydroxy-3-allylphenyl)acrylic acid (17)38
pound 16 (23.4 mg, 0.11 mmol) was dissolved in 3.0 mL of THF,
then KOH (20.0 mg, 0.36 mmol) and H2O (50 L) were added,
. Com-
l
l
the solution was heated to reflux. After reaction completion (3 h)
the mixture was brought to pH 4 by the addition of 2 N HCl and ex-
tracted three times with Et2O and the combined organic layer was
washed with brine and dried over anhydrous Na2SO4. The solvent
was evaporated under reduced pressure to afford 12 as white solid
(44.6 mg, 99%). Mp 153–154 °C; 1H NMR (acetone-d6): d 1.72 (6H,
s), 3.32 (2H, d, J = 7.6 Hz), 3.89 (3H, s), 5.32 (1H, t, J = 7.6 Hz), 6.36
(1H, d, J = 16.1 Hz), 6.99 (1H, d, J = 8.4 Hz), 7.47 (1H, d, J = 2.3 Hz),
7.49 (1H, dd, J = 8.0, 2.3 Hz), 7.62 (1H, d, J = 16.1 Hz); 13C NMR (ace-
tone-d6): d 168.3, 160.2, 145.7, 133.0, 131.3, 129.8, 128.8, 127.7,
123.0, 116.2, 111.4, 55.9, 29.0, 25.9, 17.8; DARTMS [M+H]+ m/z
247; HRDARTMS [M+H]+ m/z: 247.1338 (calcd for C15H19O3:
247.1334).
and the solution was heated to reflux. After reaction completion
(3 h) the mixture was brought to pH 4 by the addition of 2 N
HCl and extracted three times with Et2O and the combined organ-
ic layer was washed with brine and dried over anhydrous Na2SO4.
The solvent was evaporated under reduced pressure. The residue
was recrystallized (CHCl3) to afford 17 as a white solid (14.7 mg,
67%). Mp 178–180 °C; 1H NMR (acetone-d6): d 3.41 (2H, d,
J = 6.8 Hz), 5.01–5.05 (1H, m), 5.10 (1H, dq, J = 17.1, 1.8 Hz),
5.98–6.09 (1H, m), 6.33 (1H, d, J = 15.9 Hz), 6.91 (1H, d,
J = 8.2 Hz), 7.39 (1H, dd, J = 8.2, 2.4 Hz), 7.45 (1H, d, J = 2.4 Hz),
7.60 (1H, d, J = 15.9 Hz); 13C NMR (acetone-d6): d 170.9, 158.0,
145.7, 137.5, 130.9, 128.7, 128.0, 127.1, 116.1, 115.8, 115.6,
34.7; DARTMS [M+H]+ m/z 205; HRDARTMS [M+H]+ m/z:
205.0881 (calcd for C12H13O3: 205.0865).
4.2.3.4. 4-Hydroxyphenyl-acrylic acid methyl ester (14)36
. 4-
Hydroxycinnamic acid (13, 300.0 mg, 1.83 mmol) was dissolved
4.2.3.8.
3-(3,5-Di-tert-butylphenyl-4-hydroxy)acrylic
acid
0
in 1.5 mL of methanol, catalytic amount of concd H2SO4 and 4 ÅA
MS were added, and the solution was heated to reflux. After reac-
tion completion (29 h) the solvent was evaporated in vacuo. The
residue was extracted three times with EtOAc and the combined
organic layer was washed with brine and satd NaHCO3, dried over
anhydrous Na2SO4, and then evaporated in vacuo to afford 14 as a
white solid (236.0 mg, 72%). Mp 122–125 °C (lit. 124–130 °C)37; 1H
NMR (acetone-d6): d 3.72 (3H, s), 6.35 (1H, d, J = 16.1 Hz), 6.90 (2H,
d, J = 8.4 Hz), 7.54 (2H, d, J = 8.4 Hz), 7.61 (1H, d, J = 16.1 Hz); 13C
NMR (acetone-d6): d 167.8, 160.5, 145.3, 130.8, 126.9, 116.6,
115.2, 51.5; DARTMS [M+H]+ m/z 179; HRDARTMS [M+H]+ m/z:
179.0701 (calcd for C10H11O3: 179.0708).
(18)40
. 3,5-Di-tert-butylphenyl-4-hydroxybenzaldehyde (300.0 mg,
1.28 mmol) and malonic acid (199.6 mg, 1.92 mmol) were dissolved
in 6.0 mL of pyridine, catalytic amount of piperidine was added and
stirred at 40 °C for overnight. After reaction completion the mixture
was brought to pH 4 by the addition of 2 N HCl and extracted three
times with EtOAc and the combined organic layer was washed with
brine and dried over anhydrous Na2SO4. The solvent was evaporated
under reduced pressure. The residue was purified by recrystallization
(CHCl3/n-hexane) to afford 18 as a white solid (130.9 mg, 37%). Mp
203–204 °C (lit. 206–208 °C)40 1H NMR (CDCl3): d 1.46 (18H, s),
;
5.56 (1H, br s), 6.32 (1H, d, J = 16.1 Hz), 7.40 (2H, s), 7.76 (1H, d,
J = 16.1 Hz); 13C NMR (CDCl3): d 172.8, 156.6, 148.2, 136.4, 125.8,
125.5, 113.7, 34.3, 30.1; DARTMS [M+H]+ m/z 277; HRDARTMS
[M+H]+ m/z: 277.1815 (calcd for C17H25O3: 277.1804).
4.2.3.5.
(15)38
3-(4-Allyloxyphenyl)acrylic
Compound 14 (99.5 mg, 0.56 mmol) was dissolved in
acid
methyl
ester
.
1.5 mL of acetone, 108.1 mg (0.78 mmol) K2CO3 and 0.10 mL
(1.17 mmol) allyl bromide were added, and the solution was
heated to reflux. After reaction completion (1.5 h) the solvent
was evaporated in vacuo. The residue was extracted three times
with CHCl3 and the combined organic layer was washed with
brine, dried over anhydrous Na2SO4, and evaporated in vacuo.
The residue was recrystallized (n-hexane-EtOAc) to afford 15 as a
4.3. Biological studies
4.3.1. Preparation of test compounds
Compounds were prepared as stock solutions in DMSO and
stored in aliquots at ꢀ20 °C. The final concentration of DMSO
was 1.0% (v/v) in the biological assays unless otherwise noted.
white solid (112.1 mg, 92%). Mp 53–56 °C (lit. 65.5 °C)38
;
1H NMR
4.3.2. Cell culture, transfection of the reporter gene, and cloning
of stable transformants
HEK293 cell lines were maintained in Eagle’s minimum essen-
tial medium with 1% (v/v) nonessential amino acids (GIBCO), sup-
plemented with 10% (v/v) fetal bovine serum (FBS; Nichirei), 50
units/mL penicillin, 50 lg/mL streptomycin, and 50 lg/mL kana-
mycin (Meiji). Reporter plasmid p2.1 contained a 68-bp hypoxia
response element from the ENO1 gene inserted upstream of an
(CDCl3): d 3.79 (3H, s), 4.57 (2H, d, J = 5.3, 1.4 Hz), 5.31 (1H, dq,
J = 10.5, 1.5 Hz), 5.42 (1H, dq, J = 17.1, 1.5 Hz), 6.00–6.10 (1H, m),
6.31 (1H, d, J = 15.9 Hz), 6.91 (2H, d, J = 8.7 Hz), 7.47 (2H, d,
J = 8.7 Hz), 7.65 (1H, d, J = 15.9 Hz); 13C NMR (CDCl3): d 167.8,
160.4, 144.5, 132.8, 129.7, 127.3, 118.0, 115.4, 115.1, 68.8, 51.6,
29.7; DARTMS [M+H]+ m/z 219; HRDARTMS [M+H]+ m/z:
219.0997 (calcd for C13H15O3: 219.1021).

5400
H. Hattori et al. / Bioorg. Med. Chem. 19 (2011) 5392–5401
SV40 promoter in the luciferase reporter plasmid pGL2-promoter
(Promega).20 The stable transformants of the HEK293 cells for the
HIF-1-dependent promoter assay were established by the transfec-
tion of p2.1 along with a pcDNA 3.1 empty vector containing a neo-
mycin resistant gene using TransFection™ lipid reagent (Bio-Rad),
followed by selection with Geneticin (G418) sulfate.
1:500 dilution of one of the mouse monoclonal antibodies against
human HIF-1 (Novus Biologicals Inc.) or goat polyclonal anti-hu-
a
man b-actin (1:3000 dilution, Santa Cruz). After washing, the
membranes were incubated for 1 h at room temperature with
1:1000 or 1:3000 dilutions of appropriate horseradish peroxi-
dase-labeled secondary antibody (Sigma), and bound antibody
was visualized and quantified by chemiluminescence detection
using a LAS 3000 mini imager and Multi Gauge software (Science
Lab., version 2005, version 3.0; Fuji Photo).
4.3.3. Hypoxia treatment
Hypoxic incubation (1% O2) was performed in an air-tight mod-
ular incubator chamber (Billups-Rothenberg, Inc.) flushed with a
gas mixture containing 1% O2, 94% N2 and 5% CO2. The chamber
was placed in a 37 °C incubator. For normoxic conditions (20%
O2), cells were incubated in a standard tissue culture incubator
in 95% air and 5% CO2.
4.3.8. RNA extraction and real-time RT-PCR
HCT116 cells were plated at a density of 1 ꢁ 106 cells/well (
u10
dish) in 8 mL of McCoy’s 5A medium with 10% FBS and antibiotics,
incubated for 24 h, and exposed to compounds under normoxic or
hypoxic conditions for 24 h. Total RNA samples were extracted
immediately following treatments with the Isogen (Nippongene).
Synthesis of the first strand cDNAs was performed using ReverTra
AceÒ qPCR RT kit (TOYOBO) per the manufacturer’s instructions.
Quantitative real-time PCR was performed in triplicate using the
Thermal Cycler DiceÒ Real Time system (TaKaRa Bio). Independent
PCRs were performed using the same cDNA samples for the gene
of interest and 18S rRNA using SYBRÒ Premix Ex Taq (TaKaRa Bio).
The following gene-specific pairs were used: GLUT1, 50-ACTGCAA
CGGCTTAGACTTCGAC-30, 50-ACAAACTAGGGACAATGGGTCTCT-30;
VEGFA, 50-TGCTTCTGAGTTGCCCAGGA-30, 50-GATACAGGAGTGTGG
TAACTTTGGT-30; HK2, 50-ATGCCTAGATGACTTCCGCACA-30, 50-ACA-
TGGACCCACTCTAACAGGC-30; and 18S rRNA, 50-CGTTGATCCTGCCAG-
TAGC-30, 50-CAATACCAAGGAAACCAGCG-30. The PCR conditions
consisted of denaturation at 95 °C for 30 s, followed by 40 cycles of
denaturation at 95 °C for 5 s, and annealing/extension at 60 °C for
30 s. Data generated from each PCR reaction were analyzed
using Thermal Cycler Dice RealÒ Time System Software ver. 3.0
(TaKaRa Bio). The CT value of 18S rRNA was subtracted from
4.3.4. Cell-based luciferase reporter assay
HEK293 clone cells were plated into 24-well plates (TPP Techno
Plastic Products AG) at an initial density of 8 ꢁ 104 cells/well and
cultured for 24 h with regular growth medium. After 24 h, the
medium was replaced with a fresh medium containing the test
compounds. After 1 h, they were incubated for 24 h under normox-
ic or hypoxic conditions. The assay was conducted according to the
luciferase assay kit instructions (Roche). The activity was mea-
sured using a luminometer (Berthold Detection Systems, Sirius.).
The final concentration of DMSO is 0.5% (v/v).
4.3.5. MTT assay
HCT116 cells were plated at a density of 6.0 ꢁ 103 cells/well
(96-well plate, TPP Techno Plastic Products AG) in 100 lL of
McCoy’s 5A medium with 10% FBS and antibiotics, incubated for
24 h, and exposed to the test compounds under normoxic or hyp-
oxic conditions for 24 h. The MTT reagent (0.5 mg/mL, Sigma-Al-
drich Japan) was added to the media. After 4 h incubation at
37 °C, the media were removed, and the cells were lysed with
DMSO. Absorbance at 570 nm was measured on a Multiskan JX
plate reader (Thermo Fisher Scientific).
that of the gene of interest to obtain a
an arbitrary calibrator (e.g., untreated sample in the case of upregu-
lated genes) was subtracted from the CT value to obtain a DDCT va-
lue. The gene expression level relative to the calibrator was
DCT value. The CT value of
D
DDCT
expressed as 2ꢀ
4.3.6. Clonogenic assay
.
HCT116 cells were plated at a density of 3.0 ꢁ 102 cells in a
u6
dish (TPP Techno Plastic Products AG) and treated with the test
compounds for 24 h under normoxic or hypoxic conditions. On
day 7 after colonies were formed and washed with PBS, the cells
were fixed in methanol for 30 min at room temperature, stained
with Giemsa’s solution over 3 h, and washed with water. These col-
onies were then counted.
4.3.9. CAM assay in fertilized chicken eggs
The effect of the test compounds on angiogenesis was deter-
mined by CAM assay.41 Briefly, fertilized chicken eggs were incu-
bated at 37.5 °C in
a humidified incubator with forced air
circulation. Ovalbumin (3 mL) was removed from 3-day-old
embryonated eggs. Thereafter, windows were opened for each
egg, coated with caps, and eggs were incubated at 39.5 °C. On
day 5, samples in saline solution with 2.0% DMSO and 1.0% meth-
ylcellulose were applied in the center of silicon rings (outer diam-
eter: 5 mm; inner diameter: 3 mm; height: 1 mm) on the CAM, and
the eggs were incubated at 39.5 °C for 2 days. The assay was scored
and photographed on the 7th embryonic day. Saline solution with
2.0% DMSO and 1.0% methylcellulose was used as a vehicle. Ten
eggs were used in total for each data point. Inhibition point was
judged by estimation of area of the avascular zone. The inhibition
ratios were calculated from the following formula; inhibition ratio
(%) = [1 ꢀ (control point/drug point)] ꢁ 100.
4.3.7. Immunoblot assays
HCT116 cells were plated at a density of 4 ꢁ 106 cells/well (
u10
dish, TPP Techno Plastic Products AG) in 8 mL of McCoy’s 5A med-
ium with 10% FBS and antibiotics described above, incubated for
24 h, and exposed to compounds under normoxic or hypoxic con-
ditions for 4 h. Treated cells were rinsed twice with ice-cold PBS
and lysed in appropriate extraction buffer [50 mM Tris–HCl, pH
8; 150 mM sodium chloride; 0.1% (v/v) sodium dodecyl sulfate
(SDS); 1% (v/v) Tergitol solution; 5 mM ethylenediaminetetraacetic
acid; 0.5% (w/v) sodium deoxycholate; 0.1 mM dithiothreitol;
1 mM phenylmethylsulfonyl fluoride; 1 mM NaVO4; 10 mM so-
dium fluoride; and one-half of tablet CompleteÒ (Roche)] for
30 min on ice. The extracts were centrifuged for 30 min at
15000 rpm. The protein concentrations were determined using a
4.4. Quantitative analysis by LC–ESI-MS
Quantitative chromatographic analysis of the Brazilian green
propolis extract was performed using a HP 1100 Series and a HP
LC/MSD (Agilent Technologies). A GL-Cart cartridge column ODS-
micro BCA assay kit (Pierce). Each sample was loaded at 120 lg
protein/lane on 7% SDS–polyacrylamide gel. After electrophoresis,
proteins were electrotransferred to nitrocellulose membranes.
The membranes were blocked for 1 h at room temperature in
Tris-buffered saline plus 0.1% Tween 20 containing 5% skim milk
(Wako), and then incubated for 1.5 h at room temperature with a
80A and Waters SymmetryÒ C18 3.5
were used. The mobile phase consisted of buffer solution in reser-
voir A (93.9% water, 0.8% acetic acid, 0.3% ammonium acetate, and
5% methanol) and acetonitrile in reservoir B. The elution was
l
m column (4.6 ꢁ 75 mm)

H. Hattori et al. / Bioorg. Med. Chem. 19 (2011) 5392–5401
8. Nagle, D. G.; Zhou, Y. D. Curr. Pharm. Des. 2006, 12, 2673–2688.
5401
performed using a linear gradient of 25–90% B over 90 min at a
flow rate of 0.5 mL/min. The column was operated at a constant
temperature of 40 °C. Detection was performed 254 and 280 nm
using a diode array detector. MS analysis was performed in the
negative mode on a HP 1100 MSD Series equipped with an ESI
interface and LC/MSD Chemstation software (Agilent Technolo-
gies). Veratraldehyde was used as the internal standard.32 Conven-
tional ESI-MS data were recorded using a scan range of m/z 150–
500. Active compounds were identified by retention time and
molecular ion peak (see the MS data in the Supplementary data).
9. Semenza, G. L. Curr. Pharm. Des. 2009, 15, 3839–3843.
10. Khalil, M. L. Asian Pac. J. Cancer Prev. 2006, 7, 22–31.
11. Ahn, M. R.; Kunimasa, K.; Ohta, T.; Kumazawa, S.; Kamihira, M.; Kaji, K.; Uto, Y.;
Hori, H.; Nagasawa, H.; Nakayama, T. Cancer Lett. 2007, 252, 235–243.
12. Chikaraishi, Y.; Izuta, H.; Shimazawa, M.; Mishima, S.; Hara, H. Mol. Nutr. Food
Res. 2010, 54, 566–575.
13. Izuta, H.; Shimazawa, M.; Tsuruma, K.; Araki, Y.; Mishima, S.; Hara, H. BMC
Complement. Altern. Med. 2009, 9, 1–10.
14. Anso, E.; Zuazo, A.; Irigoyen, M.; Urdaci, M. C.; Rouzaut, A.; Martinez-Irujo, J. J.
Biochem. Pharmacol. 2010, 79, 1600–1609.
15. Hasebe, Y.; Egawa, K.; Yamazaki, Y.; Kunimoto, S.; Hirai, Y.; Ida, Y.; Nose, K. Biol.
Pharm. Bull. 2003, 26, 1379–1383.
16. Liu, Y.; Veena, C. K.; Morgan, J. B.; Mohammed, K. A.; Jekabsons, M. B.; Nagle, D.
G.; Zhou, Y. D. J. Biol. Chem. 2009, 284, 5859–5868.
17. Lee, K.; Qian, D. Z.; Rey, S.; Wei, H.; Liu, J. O.; Semenza, G. L. Proc. Natl. Acad. Sci.
U.S.A. 2009, 106, 2353–2358.
4.5. Statistical analysis
18. Lee, K.; Zhang, H.; Qian, D. Z.; Rey, S.; Liu, J. O.; Semenza, G. L. Proc. Natl. Acad.
Sci. U.S.A. 2009, 106, 17910–17915.
19. Zhang, H.; Qian, D. Z.; Tan, Y. S.; Lee, K.; Gao, P.; Ren, Y. R.; Rey, S.; Hammers, H.;
Chang, D.; Pili, R.; Dang, C. V.; Liu, J. O.; Semenza, G. L. Proc. Natl. Acad. Sci. U.S.A.
2008, 105, 19579–19586.
20. Semenza, G. L.; Jiang, B. H.; Leung, S. W.; Passantino, R.; Concordet, J. P.; Maire,
P.; Giallongo, A. J. Biol. Chem. 1996, 271, 32529–32537.
21. Furukawa, S.; Takagi, N.; Ikeda, T.; Ono, M.; Nafady, A. M.; Nohara, T.;
Sugimoto, H.; Doi, S.; Yamada, H. Chem. Pharm. Bull. 2002, 50, 439–440.
22. Uto, Y.; Hirata, A.; Fujita, T.; Takubo, S.; Nagasawa, H.; Hori, H. J. Org. Chem.
2002, 67, 2355–2357.
For the result from the luciferase assay (first screening) and
real-time RT-PCR assays, statistical differences between cells trea-
ted with control and with compounds were evaluated by Student’s
t-test as indicated in the figures (⁄p <0.05, ⁄⁄p <0.01).
Dose-response results of the luciferase assay are presented as
the mean SD of three independent experiments, with each treat-
ment performed in triplicate. Statistical significance was deter-
mined by a post-hoc Dunnett’s test for comparisons of treated
samples versus controls, as indicated in the figures (⁄p <0.05).
23. Aga, H.; Shibuya, T.; Sugimoto, T.; Kurimoto, M.; Nakajima, S. Biosci. Biotechnol.
Biochem. 1994, 58, 945–946.
24. Horie, T.; Shibata, K.; Yamashita, K.; Kawamura, Y.; Tsukayama, M. Chem.
Pharm. Bull. 1997, 45, 446–451.
Acknowledgements
25. Shin, D.; Kinoshita, K.; Koyama, K.; Takahashi, K. J. Nat. Prod. 2002, 65, 1315–
1318.
26. Parmar, V.; Vardhan, A.; Nagarajan, G. R.; Jain, R. Phytochemistry 1992, 31,
2185–2186.
27. Tazawa, S.; Warashina, T.; Noro, T. Chem. Pharm. Bull. 1999, 47, 1388–1392.
28. Tani, H.; Hasumi, K.; Tatefuji, T.; Hashimoto, K.; Koshino, H.; Takahashi, S.
Bioorg. Med. Chem. 2010, 18, 151–157.
29. Leopoldini, M.; Russo, N.; Chiodo, S.; Toscano, M. J. Agric. Food Chem. 2006, 54,
6343–6351.
This work was supported in part by Kowa Life Science Founda-
tion (K.O.), the Research Foundation for the Electrotechnology of
Chubu (K.O.), Japan Society for the Promotion of Science (Grant-
in-Aid for Scientific Research (B), 20790100 to H.N.), and Api Co.,
Ltd (H.N.). We thank Yamada Bee Farm for providing us with Bra-
zilian green propolis. We also thank Mr. Tatsuya Watarai, Ms. Min-
ae Noda, and Ms., Emi Inaba for their technical assistance.
30. Jeon, H.; Kim, H.; Choi, D.; Kim, D.; Park, S.-Y.; Kim, Y.-J.; Kim, Y. M.; Jung, Y.
Mol. Pharmacol. 2007, 71, 1676–1684.
31. Park, S.-S.; Bae, I.; Lee, Y. J. J. Cell. Biochem. 2008, 103, 1989–1998.
32. de Sousa, J. P.; Bueno, P. C.; Gregorio, L. E.; da Silva Filho, A. A.; Furtado, N. A.; de
Sousa, M. L.; Bastos, J. K. J. Sep. Sci. 2007, 30, 2656–2665.
33. Rey, S.; Luo, W.; Shimoda, L. A.; Semenza, G. L. Blood 2011, 117, 4988–4998.
34. Bisogno, F.; Mascoti, L.; Sanchez, C.; Garibotto, F.; Giannini, F.; Kurina-Sanz, M.;
Enriz, R. J. Agric. Food Chem. 2007, 55, 10635–10640.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2011.07.060.
35. Menon, S. R.; Patel, V. K.; Mitscher, L. A.; Shih, P.; Pillai, S. P.; Shankel, D. M. J.
Nat. Prod. 1999, 62, 102–106.
36. De, P.; Baltas, M.; Lamoral-Theys, D.; Bruyere, C.; Kiss, R.; Bedos-Belval, F.;
Saffon, N. Bioorg. Med. Chem. 2010, 18, 2537–2548.
37. Huang, W.-J.; Chen, C.-C.; Chao, S.-W.; Lee, S.-S.; Hsu, F.-L.; Lu, Y.-L.; Hung, M.-
F.; Chang, C.-I. ChemMedChem 2010, 5, 598–607.
38. Mills, L. E.; Adams, R. J. Am. Chem. Soc. 1923, 45, 1842–1854.
39. Cairns, N.; Harwood, L. M.; Astles, D. P.; Orr, A. J. Chem. Soc., Perkin Trans. 1
1994, 3095–3100.
40. Katsumi, I.; Kondo, H.; Fuse, Y.; Yamashita, K.; Hidaka, T.; Hosoe, K.; Takeo, K.;
Yamashita, T.; Watanabe, K. Chem. Pharm. Bull. 1986, 34, 1619–1627.
41. Kasai, S.; Nagasawa, H.; Shimamura, M.; Uto, Y.; Hori, H. Bioorg. Med. Chem.
Lett. 2002, 12, 951–954.
References and notes
1. Nagasawa, H. J. Pharmacol. Sci. 2011, 115, 446–452.
2. Nagasawa, H.; Uto, Y.; Kirk, K. L.; Hori, H. Biol. Pharm. Bull. 2006, 29, 2335–2342.
3. Ebbesen, P.; Pettersen, E. O.; Gorr, T. A.; Jobst, G.; Williams, K.; Kieninger, J.;
Wenger, R. H.; Pastorekova, S.; Dubois, L.; Lambin, P.; Wouters, B. G.; Van Den
Beucken, T.; Supuran, C. T.; Poellinger, L.; Ratcliffe, P.; Kanopka, A.; Gorlach, A.;
Gasmann, M.; Harris, A. L.; Maxwell, P.; Scozzafava, A. J. Enzyme Inhib. Med.
Chem. 2009, 24, 1–39.
4. Yang, J. D.; Nakamura, I.; Roberts, L. R. Semin. Cancer Biol. 2011, 21, 35–43.
5. Semenza, G. L. Oncogene 2010, 29, 625–634.
6. Kingston, D. G. J. Nat. Prod. 2011, 74, 496–511.
7. Nagle, D. G.; Zhou, Y. D. Curr. Drug Targets 2006, 7, 355–369.