(±)-3′-O, 4′-O-dicynnamoyl-cis-khellactone, a derivative of (±)-praeruptorin A, reverses P-glycoprotein mediated multidrug resistance in cancer cells

Article abstract of DOI:10.1016/j.bmc.2006.06.066

P-glycoprotein (Pgp) is an ATP-driven membrane exporter for a broad spectrum of hydrophobic xenobiotics. Pgp-overexpression is a common cause of multidrug resistance (MDR) in cancer cells and could lead to chemotherapeutic failure. Through an extensive he

Full text of DOI:10.1016/j.bmc.2006.06.066

Bioorganic & Medicinal Chemistry 14 (2006) 7138–7145
( )-30-O, 40-O-dicynnamoyl-cis-khellactone, a derivative
of ( )-praeruptorin A, reverses P-glycoprotein mediated
multidrug resistance in cancer cells
Xiaoling Shen,a Guangying Chen,b Guoyuan Zhua and Wang-Fun Fonga,*
aBioactive Products Research Group, Department of Biology and Chemistry, City University of Hong Kong, Hong Kong SAR, China
bDepartment of Chemistry, Hainan Normal University, Haikou, Hainan, China
Received 19 May 2006; revised 29 June 2006; accepted 30 June 2006
Available online 27 July 2006
Abstract—P-glycoprotein (Pgp) is an ATP-driven membrane exporter for a broad spectrum of hydrophobic xenobiotics. Pgp-over-
expression is a common cause of multidrug resistance (MDR) in cancer cells and could lead to chemotherapeutic failure. Through
an extensive herbal drug screening program we previously showed that ( )-praeruptorin A (PA), a naturally existing pyranocumarin
isolated from the dried root of Peucedanum praeruptorum Dunn., re-sensitizes Pgp-mediated MDR (Pgp-MDR) cancer cells to
cancer drugs. A number of PA derivatives were synthesized and one of these, ( )-30-O, 40-O-dicynnamoyl-cis-khellactone
(DCK), was more potent than PA or verapamil in the reversal of Pgp-MDR. In Pgp-MDR cells DCK increased cellular accumu-
lation of doxorubicin without affecting the expression level of Pgp. In Pgp-enriched membrane fractions DCK moderately stimu-
lated basal Pgp-ATPase activity, suggesting some transport substrate-like function. However, DCK also inhibited Pgp-ATPase
activity stimulated by the standard substrates verapamil or progesterone with decreased Vmaxs but Kms were relatively unchanged,
suggesting a primarily non-competitive mode of inhibition. While the binding of substrates to active Pgp would increase the
reactivity of the Pgp-specific antibody UIC2, DCK decreased UIC2 reactivity. These results suggest that DCK could bind simulta-
neously with substrates to Pgp but perhaps at an allosteric site and thus affect Pgp–substrate interactions.
ꢀ 2006 Elsevier Ltd. All rights reserved.
1. Introduction
antagonists, such as curcumin, may act by modulating
Pgp expression.5 Direct inhibitors of Pgp belong to
various groups of chemicals including calcium channel
blockers, calmodulin inhibitor, coronary vasodilators,
hormones, cyclosporines and indole alkaloids, quino-
lines, surfactants, and antibodies.6 Several pharmaco-
phore models of Pgp inhibitors have been built
through three-dimensional quantitative structure–activi-
ty relationship analysis of various Pgp substrates and
inhibitors.7–10 In general, these Pgp models consist of
hydrophobic centers (aromatic rings), hydrogen bond
acceptors (O or N atom), and hydrogen bond donors
(OH or NH group) in certain spatial separation.
Multidrug resistance (MDR) in cancer cells could lead
to chemotherapeutic failure. The molecular mechanisms
leading to MDR include the activation of transport and
detoxification systems, enhancement of target repair
activities, alterations of drug targets, and disregulation
of cell death pathways.1 MDR may be resulted from
the over-expression of transporter proteins such as
P-glycoprotein (Pgp), multidrug resistance associated
protein (MRP1), lung resistance protein (LRP), and
breast cancer resistance protein (BRCP).2,3 Pgp is a
170-kDa ATP-dependent membrane transporter protein
encoded by MDR1 gene and is active against a spectrum
of functionally unrelated hydrophobic drugs.4 Pgp
antagonists have been shown to reverse MDR by
increasing intracellular drug availability. A few MDR
( )-Praeruptorin A (PA), a naturally existing pyrano-
cumarin isolated from the dried root of a medicinal
plant Peucedanum praeruptorum Dunn., was shown to
re-sensitize Pgp-overexpressing MDR (Pgp-MDR) cells
by transiently depleting ATP and/or suppressing Pgp
expression.11 DCK was derived from PA by replacing
the two aliphatic acyloxys at C-30 and C-40 with two
cynnamoyloxys that could form hydrophobic centers
in a more planar configuration. Our data show that
Keywords: ( )-Praeruptorin A; ( )-30-O, 40-O-dicynnamoyl-cis-khel-
lactone; P-glycoprotein; Multidrug resistance.
*
Corresponding author. Tel.: +852 2788 9789; fax: +852 2788 7406;
e-mail: bhwffong@cityu.edu.hk
0968-0896/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.06.066

X. Shen et al. / Bioorg. Med. Chem. 14 (2006) 7138–7145
7139
DCK is a potent MDR reversal agent and may directly
inhibit Pgp by an non-competitive mechanism which
interferes with the binding of transport substrates
and/or the transport process.
HepG2/Dox and K562/Dox were employed for testing
MDR reversal activity of DCK. Compared to their
parental drug-sensitive cells by IC50 values (concentra-
tion inhibiting 50% of cell growth), K562/Dox cells were
227.3 and 129.2 times more resistant to Pgp substrate
anticancer drugs vinblastine and doxorubicin, whereas
HepG2/Dox cells were 1174.2 and 211.8 times more
resistant to the two drugs, respectively. Both HepG2/
Dox and K562/Dox showed no drug resistance to non-
Pgp substrate drug cisplatin (substrate of MRP2,
MRP3), camptothecin (substrate of BCRP), and arsenic
(substrate of MRP1), indicating that the overexpressed
Pgp was the main reason for the MDR phenotype
in the two cell lines used in this study (Table 1). DCK
by itself showed little growth inhibitory effect
(IC50 > 18 lM). However, the addition of 4 lM DCK
to treatments with Pgp substrate drugs vinblastine,
doxorubicin, puromycin or paclitaxel increased the drug
sensitivity of HepG2/Dox cells, as indicated by changes
in IC50, by 28.2- 20.8- 17.6- and 23.5-fold, respectively
(Table 2). Under similar conditions the increase in drug
sensitivity of K562/Dox cells was 28.6- 15.0- 11.3- and
108.3-fold, respectively (Table 3). DCK showed a much
stronger MDR reversing activity than PA and was
2. Results and discussion
2.1. Preparation and identification of DCK
DCK was prepared by a two-step reaction (see Scheme
1). Basic hydrolysis of PA by 0.5 M KOH in dioxane
provided the intermediate ( )-cis-khellactone (1), and
dicynnamoylation of (1) catalyzed by dimethylamino-
pyridine (DMAP) in CHCl2 gave DCK.12 The NMR
and MS data of DCK were identical with those of
(+)-30-O,40-O-dicynnamoyl-cis-khellactone reported by
Wu et al.12 Its zero specific rotation indicates that the
product was ( )-30-O,40-O-dicynnamoyl-cis-khellactone.
2.2. DCK resensitized Pgp-MDR cells
In this study, human hepatoma cell line HepG2, leuke-
mia cell line K562, and their Pgp-MDR sublines
O
O
O
O
OAc
O
O
O
O
O
O
O
OH
OR
OH
OR
b
a
+
+
+
O
O
O
O
O
O
O
O
O
OH
OR
OAc
OH
OR
O
O
PA
O
1
DCK R=
Scheme 1. Reagents and conditions: (a) 0.5 M KOH, dioxane, 60 ꢁC, 20 min, 15%; (b) cynnamic acid, DMAP, DCC, CHCl2, reflux, 2.5 h, 61%.
Table 1. Cytotoxicity of anticancer agents, PA and DCK
RRc
HepG2 IC50
HepG2/Dox IC50
RRc
a
a
b
b
Treatment
K562 IC50
K562/Dox IC50
Vinblastine
Doxorubicin
Cisplatin
Camptothecin
Arsenic
8.80 2.07 (·10ꢀ4
0.24 0.11
2.58 0.01
)
0.20 0.07
31.02 17.07
2.01
227.3
129.2
0.8
2.64 0.67 (·10ꢀ4
0.17 0.09
3.00 0.79
0.02 0.01
4.52 0.97
>50
)
0.31 0.05
36.00 2.46
2.07 1.12
0.02 0.01
2.88 0.60
14.42 2.52
34.67 5.39
1174.2
211.8
0.7
0.10 0.00
3.34 0.25
0.09 0.01
3.47 0.61
30.14 6.09
52.59 8.74
0.9
1.0
1.0
0.6
PA
DCK
>50
30.61 5.43
—
—
1.7
18.4 4.28
1.8
Cells were exposed to drugs for 72 h and IC50 (lM drug concentration inhibiting 50% growth) was determined. Values presented as means SD of
three experiments.
a IC50 was determined by MTT assay.
b IC50was determined by SRB assay.
c RR = Resistant ratio = IC50 (MDR cell)/IC50(parental cell)
.

7140
X. Shen et al. / Bioorg. Med. Chem. 14 (2006) 7138–7145
Table 2. Effects of PA, DCK, and verapamil on drug sensitivity of HepG2 and HepG2/Dox cells
HepG2/Dox
HepG2
IC50 (lM)
Sensitivity increaseda
IC50 (lM)
Sensitivity increaseda
Vinblastine
With PA
0.31 0.05
0.13 0.05
2.64 0.67 (·10ꢀ4
2.41 0.88 (·10ꢀ4
2.45 0.52 (·10ꢀ4
3.38 1.07 (·10ꢀ4
)
)
)
)
2.4
28.2
4.9
1.1
1.1
0.8
With DCK
With verapamil
0.011 0.004
0.063 0.023
Doxorubicin
With PA
36.00 2.46
25.43 4.23
1.73 0.52
6.28 3.33
0.17 0.09
0.15 0.03
0.15 0.04
—
1.4
20.8
5.7
1.1
1.1
—
With DCK
With verapamil
Puromycin
With PA
104.41 10.87
61.03 6.32
5.93 0.90
0.21 0.03
0.20 0.03
0.17 0.07
—
1.7
17.6
7.7
1.1
1.2
—
With DCK
With verapamil
13.49 2.49
Paclitaxel
With PA
4.23 0.05
1.33 0.54
0.18 0.14
0.86 0.12
5.94 2.68 (·103)
5.42 1.91(·103)
9.78 1.22 (·103)
—
3.2
23.5
4.9
1.1
0.6
—
With DCK
With verapamil
HepG2 and HepG2/Dox cells were exposed to anticancer drugs in the presence of 4 lM PA, DCK or verapamil for 72 h and IC50 (lM drug
concentration inhibiting 50% growth) was determined by SRB assay. Values are means SD of three experiments.
a Folds increased in drug sensitivity = (IC50 of control)/(IC50 in the presence of modulator).
Table 3. Effects of PA, DCK, and verapamil on drug sensitivity of
K562/Dox cells
a complete G2/M arrest at 0.2 lM but in Pgp-MDR
HepG2/Dox cells the concentration required was
Drug
IC50
Sensitivity
increaseda
50 lM. DCK at concentrations as high as 20 lM had no
effect on the cell cycle of HepG2/Dox cells. If DCK revers-
es MDR by maintaining cellular doxorubicin concentra-
tion as proposed, the effectiveness of doxorubicin on cell
cycle arrest should be restored by DCK. Indeed, DCK
significantly enhanced doxorubicin-induced G2/M arrest
in Pgp-MDR cells but not in drug-sensitive cells. For
example, 4 lM DCK reduced the required doxorubicin
concentration from 50 to 2 lM. This result supported
the suggestion that DCK was functioning as a Pgp
modulator.
Vinblastine
With PA
0.20 0.07
0.057 0.020
0.007 0.003
0.017 0.005
3.5
28.6
11.8
With DCK
With verapamil
Doxorubicin
With PA
31.02 17.07
19.70 2.44
2.08 0.72
3.36 0.93
1.6
15.0
9.3
With DCK
With verapamil
Puromycin
With PA
47.69 9.00
30.13 6.50
4.23 1.08
6.62 1.49
1.6
11.3
7.2
2.4. DCK increased drug uptake and reduced drug efflux
in MDR cells
With DCK
With verapamil
Paclitaxel
With PA
3.25 0.45
1.44 0.18
0.03 0.01
0.12 0.02
Flow cytometry analysis showed that DCK increased
cellular concentration of doxorubicin, a Pgp substrate.
After an 1-h incubation with doxorubicin, the presence
of 10 lM DCK increased cellular doxorubicin accumu-
lation by 30%, compared to a 25% increase caused by
15 lM of verapamil (Fig. 2A). The increase in drug
accumulation was probably due to a decreased drug ef-
flux since DCK inhibited doxorubicin efflux from
HepG2/Dox cells pre-loaded with doxorubicin as effi-
cient as verapamil (Fig. 2B). DCK also decreased the ef-
flux of another fluorescent Pgp substrate Rh-123 in
HepG2/Dox cells but in this case DCK was less active
than verapamil (data not shown). The differential inhib-
itory effect is conceivable if these drugs are transported
by different transport sites on Pgp.13–18
2.2
108.3
27.1
With DCK
With verapamil
K562/Dox cells were exposed to anticancer drugs in the presence of
4 lM PA, DCK or verapamil for 72 h and IC50 (lM drug concentra-
tion inhibiting 50% growth) was determined by MTT assay. Values are
means SD of three experiments.
a Folds increased in drug sensitivity = (IC50 of control)/(IC50 in the
presence of modulator).
active at concentrations as low as 2 lM. The decreases
in IC50 values were not observed in drug-sensitive cells
(Table 2), implying that DCK mainly acted on the over-
expressed Pgp.
2.3. DCK enhanced doxorubicin-induced G2/M arrest in
HepG2/Dox cells
2.5. DCK did not affect Pgp expression
Doxorubicin is a topoisomerase II inhibitor which induc-
es G2/M arrest in cell cycle. As shown in Figure 1, in
drug-sensitive HepG2 cells doxorubicin achieved almost
It was previously noted that PA might decrease both
MDR1 mRNA level and Pgp level in Pgp-overexpress-
ing KB V1 cells. Pgp expression was monitored by

X. Shen et al. / Bioorg. Med. Chem. 14 (2006) 7138–7145
7141
HepG2/Dox
HepG2/Dox + 2µM Dox
23.36 0.41
23.01 2.86%
HepG2
+10µM Dox
38.47 3.98%
+2µM DCK
65.83 3.79%
22.59 0.26 %
+50µM Dox
69.77 4.34%
+4µM DCK
70.83 2.31%
+0.2µM Dox
76.57 1.32%
1000
400
600
800
200
0
800
1000
200
400
600
800
1000
200
400
600
FLH-2
Figure 1. DCK enhanced doxorubicin-induced G2/M arrest in HepG2/Dox cells. Cells were incubated with doxorubicin with or without addition of
DCK for 48 h. Flow cytometry results of one typical experiment are shown. Numerical data are percent of G2/M cells and are presented as
means SD of three independent experiments.
Western blot analysis and our results showed that both
40
A
*
K562/Dox cells and HepG2/Dox cells expressed high le-
vel of MDR1, and a 72 h treatment with 20 lM DCK
did not affect the expression level (Fig. 3).
Dox accumulation
*
*
30
20
10
*
2.6. DCK inhibited transport substrate-stimulated Pgp-
ATPase activity
Pgp hydrolyzes ATP to support drug transport activity
therefore the Pgp-catalyzed ATP hydrolysis rate reflects
indirectly the transport activity of Pgp. We studied Pgp-
ATPase activity catalyzed by the membrane fractions
prepared from Pgp-MDR cells while suppressing the
activities of other major membrane ATPases. As shown
in Figure 4, verapamil and progesterone, two standard
Pgp transport substrates, stimulated ATP hydrolysis in
a dose dependent manner. DCK by itself also slightly
stimulated basal Pgp-ATPase activity (Fig. 4). But in
the presence of 5 lM DCK, the verapamil- and proges-
terone-stimulated ATPase activity was notably inhibited
with decreased Vmaxs but Kms were decreased or
unchanged.
0
40
B
*
30
20
10
0
Dox efflux
*
*
2µM
4µM
10µM
20µM VRP (15µM)
Figure 2. Effect of DCK on doxorubicin retention in HepG2/Dox cells.
(A) Cellular doxorubicin accumulation assay. Cells were incubated
with 5 lM doxorubicin with or without addition of DCK or verapamil
(VRP) for 1 h. Effect of DCK and verapamil on cellular doxorubicin
accumulation was analyzed by flow cytometry. Increase of cellular
fluorescence is shown in percentage. (B) Doxorubicin efflux assay. Cells
were treated with 5 lM doxorubicin for 1 h followed by an additional
hour in doxorubicin-free medium with DCK or verapamil. Cellular
fluorescence was analyzed by flow cytometry and increase of cellular
fluorescence is shown in percentage. Results are mean values of three
experiments. Significantly different at *P < 0.05 compared with base-
line control.
2.7. DCK affecting UIC2 antibody reactivity
The mAb UIC2 is a conformation-sensitive antibody
which preferentially binds to an external epitope of
Pgp that is associated with substrate or competitive
inhibitor.18 UIC2 reactivity was monitored by the use
of a fluorescent secondary antibody. Figure 5 shows that
cyclosporin A, a Pgp substrate, increased UIC2 labeling

7142
Sensitive cell
X. Shen et al. / Bioorg. Med. Chem. 14 (2006) 7138–7145
A
DCK
Normal
IgG2a
Na3VO4
Cyclosporine A
Resistant cell
K562/Dox
37ºC
HepG2/Dox
DCK (µM)
B
Control
DCK
0
0
4
10
20
4ºC
Figure 3. DCK did not affect Pgp expression within MDR cells.
HepG2/Dox cells and K562/Dox cells were incubated with or without
DCK for 72 h. Lysed cell extracts containing 50 lg of total cellular
protein were separated on 10% SDS–PAGE. Pgp was detected by
antibody labeling. Samples in lane 1 were from untreated K562 and
HepG2 cells.
100
101
102
103
104
FL2-H
Figure 5. Effect of DCK on mAb UIC2 binding reactivity to Pgp. (A)
HepG2/Dox cells reacted with mAb UIC2 at 37 ꢁC, in the presence or
in the absence (filled histogram) of 1 mM Na3VO4 (thin line), 5 lM
DCK (thick line), or 5 lM cyclosporine A (broken line). Normal IgG2a
was used as negative control (dotted line). UIC2 binding affinity to Pgp
can be detected by labeling with a fluorescent secondary antibody and
analyzed by flow cytometry. Pictures from a typical experiment
showed the cellular fluorescence variation in HepG2/Dox cells when
cells were pretreated with 1 mM sodium vanadate, or 5 lM DCK or
5 lM cyclosporin A. (B) HepG2/Dox cells reacted with mAb UIC2 at
4 ꢁC in the presence or in the absence (filled histogram) of 5 lM DCK
(dotted line), the cellular fluorescence kept unaltered.
of HepG2/Dox cells. The blocking of the ATP site by
sodium vanadate (Na3VO4), and thus blocking the
transportation process, also decreased UIC2 labeling.
Unlike substrates, DCK decreased UIC2 reactivity like
Na3VO4.
2.8. DCK enhanced growth inhibitory effect of doxorubi-
cin in soft-agar colony-forming assay
Cancer cell proliferation in a chronic cytotoxicity envi-
ronment can be evaluated by the soft-agar colony
formation assay. As shown in Figure 6, 4 lM DCK or
2 lM doxorubicin individually had no appreciable effect
on colony formation of HepG2/Dox cells. However,
when the two drugs were combined (4 lM DCK and
2 lM doxorubicin), few HepG2/Dox colonies could be
formed.
DCK stimulates the basal Pgp ATPase activity and this
has some substrate-like activity but it is an effective
non-competitive inhibitor that decreases the Vmaxs of
verapamil and progesterone with the Kms relatively
unchanged. As a non-competitive inhibitor it is likely
that DCK might bind to Pgp simultaneously with
other substrates. The interaction would cause Pgp
conformational changes that could hinder intramolecu-
lar movements necessary for the transportation process
as well as transport-associated ATP hydrolysis.
3. Conclusion
Although the mother compound PA suppresses MDR1
gene expression, our results show that DCK is a more
potent MDR modulator that inhibits Pgp directly.
DCK was derived from PA by replacing the two acyls at
C-30-O and C-40-O site with two cynnamoyls. The new
derivative has two bulky polar groups on the same plane
and is more active than PA and verapamil in resensitiz-
ing Pgp-MDR cells. Our results provide a positive foun-
dation for future PA based drug development.
50
Progesterone
progesterone
Vmax=60.01 3.79
Km=31.95 5.35
Progesterone
µM DCK
25
Vmax=32.11 2.98
Km=15.26 5.04
4. Experimental
4.1. General
0
0
25
50
75
100
Verapamil
Vmax=42.58 1.06
Km=5.36 0.60
50
verapamil
Doxorubicin, vinblastine, puromycin, paclitaxel,
cisplatin, camptothecin, arsenic, verapamil, sulforhoda-
mine B (SRB), 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphen-
yl tetrazolium bromide (MTT), MgATP, cell culture
grade agarose, and other reagent grade chemicals were
purchased from Sigma/Aldrich Co. (St. Louis, MO,
USA). Propidium iodide (PI) was from Molecular
Probes (OR, USA). Cell culture media and supplements
were products of GibcoBRL (MD, USA). Nitrocellulose
membranes and secondary antibody (horseradish–
Verapamil
Vmax=16.97 1.33
Km=4.75 0.48,
µM DCK
25
0
DCK
Vmax=9.12 0.78
Km=0.24 0.55
0
20
40
60
Concentration (µM)
Figure 4. Effect of DCK on basal and substrate-stimulated ATPase
activity of Pgp. Membrane vesicles were prepared from drug-resistant
Pgp-overexpressing K562/Dox cells. Pgp ATPase activity was mea-
sured with other major membrane ATPases suppressed.

X. Shen et al. / Bioorg. Med. Chem. 14 (2006) 7138–7145
7143
A
1000
800
600
400
200
0
*
B
HepG2/Dox
Control
2µM Doxorubicin
2µM Doxorubicin
+ 4µM DCK
4µM DCK
Figure 6. DCK enhanced growth inhibitory effect of doxorubicin in soft-agar colony-forming assay. (A) Colony numbers from three independent
experiments are shown as means SD. Significantly different to the untreated control group at *P < 0.05. (B) Pictures from a typical experiment
showing the colonies formed in untreated plate (control), and the plates treated with 2 lM doxorubicin, or 4 lM DCK, or combination of 2 lM
doxorubicin and 4 lM DCK.
peroxidase-conjugated anti-rabbit IgG) were from Bio-
Rad (CA, USA), and the anti-Pgp antibody was from
Calbiochem (La Jolla, USA). Anti-MDR1 monoclonal
UIC2 (UIC2 mouse monoclonal anti-human MDR1)
was from Immunotech (PA, USA). Goat anti-mouse
IgG2a-PE and normal IgG2a were from Santa Cruz Bio-
technology (CA, USA).
Cinnamic acid (220 mg or 15 mmol) was added to a mix-
ture of (1) (80 mg, 0.31 mmol), dichloromethane (5 mL),
N, N0-dicyclohexylcarbodiimide (206 mg, 1 mmol), and
4-dimethylaminopyridine (4 mg, 0.032 mmol). The mix-
ture was stirred/refluxed for 3 h, cooled to room temper-
ature, filtered, and the filtrate was separated and purified
by a repeated flash silica gel 60 column chromatograph
(petroleum ether/EtOAc, 4:1). Fractions were monitored
by TLC at 365 nm and purified DCK (81 mg, 52% yield)
was obtained.
Nuclear magnetic resonance spectra (NMR) were
recorded on a Varian NMR-300 MHz spectrometer in
CDCl3. Optical rotation was measured on a PE343
polarimeter. Mass spectra (MS) were recorded on a
Thermo-Finnigan LCQ advantage mass spectrometer.
DCK appeared to be white amorphous solid with a zero
specific rotation. 1H NMR dppm (CDCl3-300 MHz):
7.68 (2H, d, J = 15.9 Hz, –CO–CH@), 7.61 (1H, d,
J = 9.5 Hz, 4-H), 7.47–7.42 (4H, m, Ar–H), 7.39 (1H,
d, J = 8.6 Hz, 5-H), 7.36–7.30 (6H, m, Ar–H), 6.86
(1H, d, J = 8.4 Hz, 6-H), 6.78 (1H, d, J = 4.9 Hz,
40-H), 6.45 (2H, d, J = 15.9 Hz, Ar–CH@), 6.22 (1H,
d, J = 9.6 Hz, 3-H), 5.53(1H, d, J = 4.9 Hz, 30-H), 1.57
and 1.48(3H each, s, C-20-Me). MS (ESI) for
C32H26O7: 545 [M+Na]+.
4.2. Preparation of DCK
Basic hydrolysis was carried-out by adding 60 mL of
0.5 M KOH to PA (1.0 g in 150 mL of dioxane) and the
mixture was stirred at 60 ꢁC for 20 min. After cooling,
pH was adjusted to 2–3 by 10% H2SO4 and the mixture
wasstirred for2 hatroomtemperatureandextractedwith
CHCl3 and trice (100, 100, and 60 mL). The combined
CHCl3 fraction was washed with 100 mL of saturated
NaHCO3 followed by 100 mL of water and dried over
Na2SO4. After solvent removal, the residue was separated
by repeated flash chromatography on silica gel (CHCl3/
EtOAc, 3:1) and fractions were monitored by a thin-layer
chromatograph (TLC) at 365 nm. The yield was 150 mg
(15%) of (1). Compound (1) appeared to be white amor-
4.3. Cell lines and cell culture
Human hepatoma cell line HepG2 and its doxorubicin-
selected Pgp-overexpressing subline HepG2/Dox were
kindly provided by Dr. Judy Chan, Chinese University
of Hong Kong.19 Human leukemia cell line K562 was
from Gibco (ATCC No. CCL-243), whereas its doxo-
rubicin-selected Pgp-overexpressing subline K562/Dox
was kindly provided by Dr. Morjani, University of
Reims Champagne-Ardenne, France.20,21 All cell lines
were grown in RPMI 1640 medium containing 10%
FBS and 100 U antibiotics, at 37 ꢁC in a humidified
5% CO2 incubator. To maintain MDR phenotype,
1
phous solid with zero specific rotation. H NMR d ppm
(CDCl3-300 MHz): 7.67 (1H, d, J = 9.5 Hz, 4-H), 7.33
(1H, d, J = 8.6 Hz, 5-H), 6.81 (1H, d, J = 8.6 Hz, 6-H),
6.23(1H, d, J = 9.6 Hz, 3-H), 5.21 (1H, d, J = 4.9 Hz, 40-
H), 3.88 (1H, d, J = 5.0 Hz, 30-H), 1.48 and 1.41 [3H each,
s, 20-(Me)2]. MS (ESI) for C14H14O5: 285 [M+Na]+.

7144
X. Shen et al. / Bioorg. Med. Chem. 14 (2006) 7138–7145
1.2 and 0.1 lM doxorubicin were added to HepG2/
Dox and K562/Dox cultures, respectively. All MDR
cells were maintained in drug-free medium for at least
7 days before used.
ice-cold PBS. Cellular fluorescence intensity was moni-
tored on a FACSCAN flow cytometer.
4.8. Cell cycle distribution
4.4. Cell growth inhibition assay
About 1 · 106 cells in 1 mL medium were seeded in a
35-mm culture dish and incubated overnight. One milli-
liter of medium with or without drugs of various con-
centrations was added and incubated for 48 h. Cells
were harvested, washed with PBS twice, and fixed in
70% EtOH at ꢀ20 ꢁC overnight. After that, the cells
were washed with PBS once, resuspended in 1 mL of
PBS containing 100 lg/mL Rnase A, and incubated at
37 ꢁC for 30 min. PI staining solution was added at a fi-
nal concentration of 40 lg/mL and incubated at RT for
5–10 min. Samples were then analyzed by a FACSCAN
flow cytometer.
Drug concentrations inhibiting 50% growth (IC50) were
determined in 72 h cultures in the 96-well format. The cell
number of anchorage dependentHepG2 and HepG2/Dox
cells was estimated by SRB cellular protein assay.22 Each
well was seeded with 5000 cells in 50 ll medium per well
and incubated overnight. Cells were incubated with fresh
medium containing drugs for 72 h, fixed in 50 ll ice-cold
15% trichloroacetic acid, and washed with triple-distilled
water five times. Cellular protein was stained by adding
50 ll of 0.4% SRB in 1% acetic acid for 10 min, rinsed with
1% acetic acid five times, and air-dried. After adding
100 ll of 10 mM Tris base (pH 10.5), optical density
(OD) at 515 nm was obtained.
4.9. Western blot analysis for detection of Pgp expression
HepG2/Dox and K562/Dox cells were treated for 72 h
with 5, 10 and 20 lM, DCK in cell growth medium
and were then incubated in ice-cold lysis buffer
(50 mM Tris, pH 7.4, 100 mM NaCl, 1 mM EDTA,
5% glycerol, 0.5 mM MgCl2, 0.5% NP 40, 0.05% SDS,
1 mM PMSF, 1 lg/mL leupeptin, 1% aprotinin, 1%
pepstatin A, 2 mM Na3VO4, and 50 mM NaF) for
20 min and collected by centrifugation (10,000g) at
4 ꢁC for 15 min. Protein concentration of supernatants
was determined by Bradford assay. Samples containing
50 lg of protein were subjected to 10% SDS–PAGE
and electro-transferred to nitrocellulose membranes.
Membranes were blocked with 3% non-fat milk/0.05%
Tween 20/TBS (10 mM Tris, pH 7.5, 100 mM NaCl),
incubated with anti-Pgp antibody for 1 h at room temper-
ature followed by horseradish–peroxidase-conjugated
secondary antibody for another 1 h at room tempera-
ture. Protein bands were detected by the ECL method.
Drug-sensitive HepG2 and K562 cell extracts were also
included as basal expression level control.
MTT cytotoxic assay23 was employed to monitor
viable cells in suspended K562 and K562/Dox cell
cultures. Cells were seeded and grown in 96-well plates
as above, treated with drugs for 68 h, and 10 ll of
5 mg/mL MTT solution was added to each well. After
4 h, 100 ll of a stop solution (50% isobutanol, 0.04 N
HCl, and 10% SDS) was added and plates were
incubated overnight at room temperature. OD was
determined at 570 nm. Each experiment was repeated
three times and results were expressed as means stan-
dard deviation (SD). Solvents and media were included
as blank control.
4.5. MDR reversing activity
MDR reversing activity of a test compound was evaluated
by comparing IC50 values of an anticancer drug in the
presence and absence of 4 lM of the test compound.
Verapamil, a Pgp modulator, was used as the positive con-
trol and medium control was subtracted as background.
All experiments were repeated at least three times.
4.10. Pgp-enriched microsomal membrane preparation
4.6. Cellular doxorubicin accumulation
Pgp-overexpressing K562/Dox cells (109–1010) were
washed with ice-cold PBS containing 0.1 mM PMSF
and vesicle buffer (0.01 M Tris–HCl, pH 7.5, 0.25 M
sucrose, and 0.2 mM CaCl2) at 4 ꢁC and collected by
centrifugation (4000g, 10 min). Membrane vesicles were
prepared by following the steps as described previous-
ly24 and stored at ꢀ80 ꢁC before use.
About 2 · 106 cells were suspended in 2 mL of medium
containing 5 lM doxorubicin. Various concentrations
of a test substance were added and incubated at 37 ꢁC
for 40 min. Cells were washed with ice-cold PBS twice
and resuspended in 1 mL of ice-cold phosphate-buffered
saline (PBS). Verapamil was included as a positive con-
trol. Cellular doxorubicin concentration was monitored
by a FACSCAN flow cytometer (Becton–Dickinson
Immunocytometry Systems, San Jose, CA, USA).
4.11. ATP hydrolysis24–26
Membrane preparations were thawed on ice and diluted
by ice-cold ATPase buffer (50 mM Tris–Cl, pH 7.4,
50 mM KCl, 2.5 mM MgSO4, and 3 mM DTT) contain-
ing 3 mM sodium azide (to inhibit the mitochondrial
ATPase), 0.5 mM EGTA (to inhibit Ca-ATPase), and
2 mM ouabain (to inhibit the Na/K-ATPase). The reac-
tion was carried out in the 96-well format and reaction
mixture (60 ll per well) contained membrane prepara-
tion (10 lg of protein), drugs, and 4 mM MgATP in
ATPase buffer. Plates were kept in ice and the reaction
4.7. Drug efflux assay
About 2 · 106 cells were suspended in 2 mL of medium
containing 5 lM doxorubicin and incubated at 37 ꢁC
for 1 h. Cells were washed with ice-cold PBS twice,
resuspended in fresh medium containing the test modu-
lator, and incubated at 37 ꢁC for 1 h. Cells were washed
with ice-cold PBS twice and resuspended in 1 mL of

X. Shen et al. / Bioorg. Med. Chem. 14 (2006) 7138–7145
7145
3. Gottesman, M. M.; Fojo, T.; Bates, S. E. Nat. Rev. Cancer
2002, 1, 48–58.
was started by bringing the plates to 37 ꢁC. After
30 min, the reaction was terminated by the addition of
30 ll of 10% SDS solution, 200 ll of a 4:1 mixture of
10% fresh ascorbic acid (pH 5.0) and 35 mM ammonium
molybdate in 15 mM zinc acetate (pH 5.0). After 30 min
at 37 ꢁC in the dark, absorption at 750 nm was deter-
mined and the free phosphate content was determined
by comparing to a standard curve. Base line controls
contained 100 lM Na3VO4, pH 10.0, and/or ethanol
(drug solvent) at a maximum final concentration of
2% v/v. All experiments were repeated three times.
4. Ambudkar, S. V.; Kimchi-Sarfaty, C.; Sauna, Z. E.;
Gottesman, M. M. Oncogene 2003, 22, 7468–7485.
5. Anuchapreeda, S.; Leechanachai, P.; Smith, M. M.;
Amubudkar, S. V.; Limtrakul, P. N. Biochem. Pharmacol.
2002, 64, 573–582.
6. Krishna, R.; Mayer, L. D. Eur. J. Pharm. Sci. 2000, 11,
265–283.
7. Cianchetta, G.; Singleton, R. W.; Zhang, M.; Wildgoose,
M.; Giesing, D.; Fravolini, A.; Cruciani, G.; Vaz, R. J.
J. Med. Chem. 2005, 48, 2927–2935.
8. Ekins, S.; Kim, R. B.; Leake, B. F.; Dantzig, A. H.;
Schuetz, E. G.; Lan, L. B.; Yasuda, K.; Shepard, R. L.;
Winter, M. A.; Schuetz, J. D.; Wikel, J. H.; Wrighton, S.
A. Mol. Pharmacol. 2002, 61, 974–981.
9. Pajeva, I. K.; Wiese, M. J. Med. Chem. 2002, 45, 5671–
5686.
10. Seelig, A. Eur. J. Biochem. 1998, 251, 252–261.
11. Wu, J. Y.; Fong, W. F.; Zhang, J. X.; Leung, C. H.;
Kwong, H. L.; Yang, M. S.; Li, D.; Cheung, H. Y. Eur. J.
Pharmacol. 2003, 473, 9–17.
4.12. Soft-agar colony formation
Each sample (5000 cells and drugs in 1.5 mL of 0.3%
agar-medium) was plated onto a 1.5-mL underlayer
(0.6% agar in RPMI-1640 medium and 10% FBS) in a
35-mm dish. Plates were incubated at 37 ꢁC in humidi-
fied 5% CO2 for 2 weeks.
12. Wu, X. L.; Kong, L. Y.; Min, Z. D. Yao Xue Xue Bao
(Acta Pharmaceutica Sinica) 2002, 37, 527–534.
13. Loo, T. W.; Bartlett, M. C.; Clark, D. M. J. Biol. Chem.
2003, 278, 39706–39710.
14. Shapiro, A. B.; Ling, V. Eur. J. Biochem. 1997, 250, 130–
137.
15. Shapiro, A. B.; Ling, V. Acta Physiol. Scand. Suppl. 1998,
643, 227–234.
16. Shapiro, A. B.; Fox, K.; Lam, P.; Ling, V. Eur. J.
Biochem. 1999, 259, 841–850.
4.13. UIC2 binding ability
This assay was performed as described27 with minor
modifications. HepG2/Dox Cells were washed with
PBS and resuspended in UIC2 binding buffer (PBS +
1 % BSA). Approximately 1 · 106 cells in 800 ll binding
buffer were pre-warmed at 37 ꢁC for 10 min, incubated
with drugs at 37 ꢁC for another 10 min, and 1 lg of
the monoclonal antibody UIC2 was added. After
15 min at 37 ꢁC, 700 ll of ice-cold UIC2 buffer was add-
ed to stop the reaction. Cell samples were washed with
cold buffer twice, resuspended in 500 ll ice-cold UIC2
buffer and 2 ll of goat anti-mouse IgG2a-PE was added.
After 15 min at 4 ꢁC in the dark, samples were washed,
resuspended in 1 mL ice-cold UIC2 binding buffer, and
analyzed by using a FACSCalibur flow cytometer. The
fluorescence intensity associated with the cells was
expressed on a log scale. Normal mouse IgG2a served
as a negative control. UIC2 reactivity at 4 ꢁC in DCK-
treated HepG2/Dox cells was also performed as control.
17. Lugo, M. R.; Sharom, F. J. Biochemistry 2005, 44, 14020–
14029.
18. Mechetner, E. B.; Schott, B.; Morse, B. S.; Stein, W. D.;
Druley, T.; Davis, K. A.; Tsuruo, T.; Roninson, I. B. Proc.
Natl. Acad. Sci. U.S.A. 1997, 94, 12908–12913.
19. Chan, J. Y.; Chu, A. C.; Fung, K. P. Life Sci. 2000, 67,
2117–2124.
20. Morjani, H.; Belhoussine, R.; Lahlil, R.; Manfait, M. Eur.
J. Haematol. 1998, 61, 240–249.
21. Cuvier, C.; Roblot-Treupel, L.; Millot, J. M.; Lizard, G.;
Chevillard, S.; Manfait, M.; Couvreur, P.; Poupon, M. F.
Biochem. Pharmacol. 1992, 44, 509–517.
22. Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.;
McMahon, J.; Vistica, D.; Warren, J. T.; Bokesch, H.;
Kenney, S.; Boyd, M. R. J. Natl. Cancer Inst. 1990, 82,
1107–1112.
Acknowledgments
23. Chou, T. C.; Talalay, P.. In Advances in Enzyme Regula-
tion; Weber, G., Ed.; Pergamon Press: Oxford, 1984; Vol.
22, pp 27–55.
24. Wang, C.; Zhang, J. X.; Shen, X. L.; Wan, C. K.; Tse, K.
W.; Fong, W. F. Biochem. Pharmacol. 2004, 68, 843–855.
25. Schwab, D.; Fischer, H.; Tabatabaei, A.; Poli, S.; Huwy-
ler, J. J. Med. Chem. 2003, 46, 1716–1725.
26. Tampal, N. M.; Robertson, L. W.; Srinivasan, C.;
Ludewigc, G. Toxicol. Appl. Pharmacol. 2003, 187, 168–
177.
We thank Mr. Lai Hau-Yun, our department technician
for NMR measurement. We also thank Prof. Wei Xiao-
Yi in South China Botanical Garden, Chinese Academy
of Sciences, for optical rotation measurement.
References and notes
1. Stavrovskaya, A. A. Biochemistry (Moscow) 2000, 65, 95–
106, Translated fromBiokhimiya 2000,65, 112–126.
2. Tan, B.; Piwnica-Worms, D.; Ratner, L. Curr. Opin.
Oncol. 2000, 12, 450–458.
27. Nagy, H.; Goda, K.; Arceci, R.; Cianfriglia, M.; Mechet-
ner, E.; Szabo, G., Jr. Eur. J. Biochem. 2001, 268, 2416–
2420.