Reversal of multiple drug resistance in cholangiocarcinoma by the glutathione S-transferase-π-specific inhibitor O 1-hexadecyl-γ-glutamyl-S-benzylcysteinyl-D-phenylglycine ethylester

Article abstract of DOI:10.1124/jpet.103.052696

Cholangiocarcinoma is markedly resistant to chemotherapy and has a dismal prognosis, but its mechanism of drug resistance is unknown. This study examines whether glutathione S-transferase-π (GSTP1-1) is involved in resistance to anticancer drugs in cholangiocarcinoma and whether GSTP1-1-specific inhibitors can overcome this resistance. First, immunohistochemical examination disclosed strong staining of all our 17 cholangiocarcinoma specimens for GSTP1-1, irrespective of histological type. Transfection of the GSTP1-1 antisense expression vector into a human cholangiocarcinoma cell line (HuCCT1) apparently decreased its intracellular GSTP1-1 concentration, and the sensitivity of transfectants to adriamycin (ADR), cisplatin, and alkylating agents such as melphalan and 4-hydroxyperoxycyclophosphamide (4-HC) was increased significantly, compared with that of mock transfectants. We next synthesized GSTP1-1-specific inhibitors by elongating the carbon chain of the ethylester at the N-terminal of γ-glutamyl-S-benzylcysteinyl-phenylglycyl diethylester and performed a pharmacokinetic study on them. Of six GSTP1-1 inhibitors tested, O¹-hexadecyl-γ-glutamyl-S-benzylcysteinyl-D-phenylglycine ethylester (C16C2) showed the smallest volume of central compartment and smallest volume of distribution at steady state and the second smallest clearance, being the most effective inhibitor in vivo. The IC₅₀ value of ADR or 4-HC for HuCCT1 cells decreased greater by treatment with C16C2 in a dose-dependent manner, paralleling the decrease in GSTP1-1 activity, than that of ADR or 4-HC alone. The antitumor activity of ADR or cyclophosphamide was clearly enhanced by combination therapy with C16C2 in a xenograft model. In conclusion, our results demonstrated that GSTP1-1 is a resistance factor for anticancer drugs in cholangiocarcinoma and that C16C2, a GSTP1-1-specific inhibitor, is a potent agent against the resistance.

Full text of DOI:10.1124/jpet.103.052696

0022-3565/03/3063-861–869$7.00
THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS
Copyright © 2003 by The American Society for Pharmacology and Experimental Therapeutics
JPET 306:861–869, 2003
Vol. 306, No. 3
52696/1087468
Printed in U.S.A.
Reversal of Multiple Drug Resistance in Cholangiocarcinoma
by the Glutathione S-Transferase-␲-Specific Inhibitor O1-
Hexadecyl-␥-glutamyl-S-benzylcysteinyl-D-phenylglycine
Ethylester
TAKAHARU NAKAJIMA, TETSUJI TAKAYAMA, KOJI MIYANISHI, ATSUSHI NOBUOKA, TSUYOSHI HAYASHI,
TOMOYUKI ABE, JUNJI KATO, KIYOYUKI SAKON, YOSHIMITSU NANIWA, HIROHUMI TANABE, and YOSHIRO NIITSU
Fourth Department of Internal Medicine, Sapporo Medical University School of Medicine, Sapporo, Japan (T.N., T.T., K.M., A.N., T.H., T.A.,
J.K., Y.N.); and Teijin Ltd., Tokyo, Japan (K.S., Y.N., H.T.)
Received April 30, 2003; accepted May 29, 2003
ABSTRACT
Cholangiocarcinoma is markedly resistant to chemotherapy benzylcysteinyl-phenylglycyl diethylester and performed
a
and has a dismal prognosis, but its mechanism of drug resis- pharmacokinetic study on them. Of six GSTP1-1 inhibitors
tance is unknown. This study examines whether glutathione tested, O1-hexadecyl-␥-glutamyl-S-benzylcysteinyl-D-phenylg-
S-transferase-␲ (GSTP1-1) is involved in resistance to antican- lycine ethylester (C16C2) showed the smallest volume of cen-
cer drugs in cholangiocarcinoma and whether GSTP1-1-spe- tral compartment and smallest volume of distribution at steady
cific inhibitors can overcome this resistance. First, immunohis- state and the second smallest clearance, being the most effec-
tochemical examination disclosed strong staining of all our 17 tive inhibitor in vivo. The IC50 value of ADR or 4-HC for HuCCT1
cholangiocarcinoma specimens for GSTP1-1, irrespective of cells decreased greater by treatment with C16C2 in a dose-
histological type. Transfection of the GSTP1-1 antisense ex- dependent manner, paralleling the decrease in GSTP1-1 activ-
pression vector into a human cholangiocarcinoma cell line ity, than that of ADR or 4-HC alone. The antitumor activity of
(HuCCT1) apparently decreased its intracellular GSTP1-1 con- ADR or cyclophosphamide was clearly enhanced by combina-
centration, and the sensitivity of transfectants to adriamycin tion therapy with C16C2 in a xenograft model. In conclusion,
(ADR), cisplatin, and alkylating agents such as melphalan and our results demonstrated that GSTP1-1 is a resistance factor
4-hydroxyperoxycyclophosphamide (4-HC) was increased sig- for anticancer drugs in cholangiocarcinoma and that C16C2, a
nificantly, compared with that of mock transfectants. We next GSTP1-1-specific inhibitor, is a potent agent against the resis-
synthesized GSTP1-1-specific inhibitors by elongating the car- tance.
bon chain of the ethylester at the N-terminal of ␥-glutamyl-S-
Cholangiocarcinoma is a cancer that is highly resistant to 2000; Isa et al., 2002). In particular, K-ras mutation has been
various anticancer drugs and thereby leads to poor prognosis
(de Groen et al., 1999; Isa et al., 2001; Patel, 2001; Okuda et
al., 2002). However, the mechanism of chemoresistance of
this disease is totally unknown.
It has been reported that there are gene abnormalities in
K-ras, p53, and APC in cholangiocarcinoma (Tada et al.,
1990; Kiba et al., 1993; Ohashi et al., 1995; Tannapfel et al.,
detected in as many as 48 to 80% of these cases. We have
recently reported the close relationship between K-ras muta-
tion and the expression of glutathione-S-transferase-␲
(GSTP1-1), a detoxification enzyme in precancerous lesions
as well as in cancer tissue (Miyanishi et al., 2001). We also
showed that GSTP1-1 is a multidrug-resistant factor for
adriamycin (ADR), cisplatin (CDDP), and alkylating agents
such as melphalan (Ban et al., 1996; Kuga et al., 1997).
Hayes et al. (1991) examined the expression of GSTP1-1 in
cholangiocarcinoma by immunohistochemical staining and
found that it was positive in eight of eight cases, indicating
that GSTP1-1 may be a viable marker for cholangiocarci-
Supported in part by a grant (12217124) from the Ministry of Education,
Science, Sports, and Culture in Japan.
Article, publication date, and citation information can be found at
http://jpet.aspetjournals.org.
DOI: 10.1124/jpet.103.052696.
ABBREVIATIONS: GSTP1-1, glutathione S-transferase P1-1; ADR, adriamycin; CDDP, cisplatin; 4-HC, 4-hydroxyperoxycyclophosphamide; CPA,
cyclophosphamide; 5-FU, 5-fluorouracil; VCR, vincristine; PBS, phosphate-buffered saline; FCS, fetal calf serum; ELISA, enzyme-linked immu-
nosorbent assay; WSC, 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride; LM, lipid microsphere; HPLC, high-performance liquid
chromatography; CDNB, 1-chloro-2,4-dinitrobenzene; V, volume; Vc, volume of central compartment; Vss, volume of distribution at steady state;
CL, total body clearance; GST, glutathione S-transferase; NSAID, nonsteroidal anti-inflammatory drug.
861

862
Nakajima et al.
noma, although they did not refer to its possible role as a digested with EcoRI, and a 0.7-kb EcoRI-EcoRI fragment containing
the whole coding region for GSTP1–1 was recovered. Both ends of
this fragment were then blunted with the Klenow fragment (Takara
Shuzo Co., Ltd., Kyoto, Japan). The pLJ vector described by Korman
et al. (1987) was linearized with BamHI; the blunting of both termi-
nals was similarly performed using the Klenow fragment and was
dephosphorylated with bacterial alkaline phosphatase (Takara
Shuzo Co., Ltd.). The two processed fragments were ligated with T4
ligase, and a clone was selected in which the GSTP1-1 cDNA was
inserted in the reverse direction. This clone was named pLJ/anti-
GSTP.
Gene Transfer. The transfection of the pLJ/antiGSTP into the
HuCCT1 cells was performed by the lipofection method. Briefly,
2.5 ϫ 105 cells were dispersed in a 3.5-cm culture dish and were
incubated for 24 h. The attached cells were then washed three times
with RPMI 1640 medium (Invitrogen), followed by the addition of 3
ml of the same culture medium to the dish. Next, 100 ␮l of plasmid
lipofectin reagent (Invitrogen) was mixed with 3 ␮g of pLJ/antiGSTP
and incubated at room temperature for 15 min. This mixture was
then added to each culture dish, and the dishes were incubated at
37°C for 6 h. RPMI 1640 medium (3 ml) containing 10% FCS was
added to each culture dish, and incubation was continued for another
72 h. G418 (Invitrogen) was added to the culture medium in each
dish to a concentration of 400 ␮g/ml, and the cells were cultured for
approximately 2 weeks at 37°C in an atmosphere of air containing
5% CO2. The G418-resistant clones were obtained and designated
HuCCT1/antiGSTP1 and HuCCT1/antiGSTP2. The pLJ vector with-
out GSTP1-1 antisense cDNA was transfected to HuCCT1 cells to
obtain a control transfectant, HuCCT1/pLJ.
resistance factor to anticancer drugs.
In this study, we first examined the expression of GSTP1-1
in cholangiocarcinoma. Then, to prove that GSTP1-1 is re-
sponsible for chemoresistance, we introduced GSTP1-1 anti-
sense cDNA into a cholangiocarcinoma cell line, HuCCT1,
which expresses GSTP1-1 abundantly, and revealed that the
transfectants became sensitive to ADR, CDDP, melphalan,
and 4-hydroxyperoxycyclophosphamide (4-HC), an active
form of cyclophosphamide (CPA) in in vitro experiments.
Furthermore, we synthesized a GSTP1-1-specific inhibitor,
O1-hexadecyl-␥-glutamyl-S-benzylcysteinyl-D-phenylglycine
ethylester, and examined whether the sensitivities to ADR
and CPA were increased by administration of the inhibitor
simultaneously with anticancer drugs in nude mice inocu-
lated subcutaneously with HuCCT1 cells.
Materials and Methods
Tissue Specimen. This study was approved by the ethics com-
mittee of Sapporo Medical University. Seventeen patients with chol-
angiocarcinoma and 19 patients with hepatocellular carcinoma who
had been admitted to Sapporo Medical University between 1996 and
2001 were included in this study. All the patients were histologically
diagnosed as cholangiocarcinoma or hepatocellular carcinoma by
biopsy or surgery, and the remaining tissue specimens were used in
this study. All the patients gave written informed consent.
GSTP1-1 Inhibitors. GSTP1-1 inhibitors (Fig. 1) were synthe-
sized by elongating the carbon chain of the alkylester (R1) at the N
terminus of ␥-glutamyl-S-benzylcysteinyl-phenylglycine, the active
form of the GSTP1-1-specific inhibitor (Morgan et al., 1996), to
increase their stability in circulation. The inhibitors were synthe-
sized by a conventional method of peptide synthesis as described by
Erickson and Merrifield (1976). In brief, D-phenylglycine-OEt HCl
was synthesized from D-phenylglycine and thionyl chloride (SOCl2)
in ethanol at room temperature (1). To a mixture of 1, Boc-Cys(Bzl)-
Anticancer Drugs. ADR and 5-fluorouracil (5-FU) were pur-
chased from Kyowa Hakko Kogyo Co., Ltd. (Tokyo, Japan), whereas
CDDP, vincristine (VCR), CPA, and 4-HC were obtained from
Shionogi Co., Ltd. (Tokyo, Japan). Melphalan was purchased from
Sigma-Aldrich (St. Louis, MO).
Immunohistochemistry. The paraffin-embedded sections were
deparaffinized in three changes of xylene and rehydrated through
graded alcohol solutions at room temperature. A 10 mM sodium
phosphate buffer containing 0.9% NaCl [phosphate-buffered saline
(PBS), pH 7.4] was used for washes between various steps; incuba-
tions were performed in a humidified chamber. Sections were treated
with 5% normal horse serum (Invitrogen, Carlsbad, CA) in PBS and
then incubated with 1:100 dilution of anti-GSTP1-1-specific mono-
clonal antibody (DAKO, Kyoto, Japan) overnight at 4°C, followed by
incubation with biotinylated horse anti-mouse immunoglobulin G at
room temperature and detection with the ABC kit (Vector Labora-
tories, Burlingame, CA).
Cell Lines and Cell Culture. Human cholangiocarcinoma cell
lines HuCCT1 and HuH28, human cervical carcinoma cell line HeLa,
human mammary carcinoma cell line MCF7, human gastric carci-
noma cell line TMK-1, human hepatocellular carcinoma cell lines
HepG2 and PLC/PRF/5, and human colon carcinoma cell line M7609
were obtained from the Japanese Cancer Research Resources Bank
(Tokyo, Japan). HuCCT1, HuH28, and M7609 were cultured in
RPMI 1640 medium (Invitrogen) and HeLa, MCF7, TMK-1, HepG2,
and PLC/PRF/5 were cultured in Dulbecco’s modified Eagle’s me-
dium containing 10% FCS (Flow Laboratories, North Ryde, Austra-
lia) in tissue culture flasks; incubation was performed at 37°C in an
atmosphere of air containing 5% CO2.
GSTP1-1 Quantitation by ELISA. After washing each cell prep-
aration two times in cold PBS, the cells were adjusted to a concen-
tration of 1 ϫ 106 /ml in the same buffer and were homogenized with
a Dounce homogenizer. The lysates were then centrifuged at 12,000
rpm for 15 min, and the concentration of GSTP1-1 in each superna-
tant was measured by sandwich ELISA established in our laboratory
as described previously (Takahashi et al., 1989; Kura et al., 1996).
Construction of a GSTP1-1 Antisense Vector. The plasmid
pGpi2 (Nakasa et al., 1997) containing GSTP1-1 cDNA was obtained
from the Japanese Cancer Research Resources Bank. pGpi2 was
Fig. 1. Structure of GST-P1-1-specific inhibitors. To increase the stability
of ␥-glutamyl-S-benzylcysteinyl-phenylglycyl diethylester (C2C2) (Mor-
gan et al., 1996) in circulation for in vivo use, the carbon chain of
ethylester (R1) at the N-terminal was elongated from C2 to C18. These
inhibitors were synthesized by the conventional method of peptide syn-
thesis (Erickson and Merrifield, 1976).

Reversal of a Drug Resistance by the GST-␲ Inhibitor
863
OH, hydroxybenzotriazole, and N-methylmorpholine in dimethylfor- droxymethyl benzoate (10 ␮g/ml) was also applied to the HPLC
column as a control.
mamide was added 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide
hydrochloride (WSC) at 0°C, and the mixture was stirred at room
temperature for 80 min. The reaction mixture was poured into water,
extracted with ethyl acetate, washed with saturated aqueous sodium
chloride, dried with anhydrous magnesium sulfate, and concentrated
under reduced pressure. The residue was purified by silica gel col-
umn chromatography to afford Boc-Cys(Bzl)-D-Phg-OEt (2). Com-
pound 2 was treated with hydrogen chloride in dioxane and concen-
trated under reduced pressure to afford Cys(Bzl)-D-Phg-OEt HCl (3).
To a mixture of Boc-Glu(OBzl)-OH and NaHCO3 in dimethylform-
amide was added 1-bromo hexadecane at room temperature. After
stirring for 38 h, the reaction mixture was poured into water, ex-
tracted with ethyl acetate, washed with saturated aqueous sodium
chloride, dried with anhydrous magnesium sulfate, and concentrated
under reduced pressure. The residue was purified by silica gel col-
umn chromatography to afford Boc-Glu(OBzl)-O(Hexadecyl) (4). Boc-
Glu(OBzl)-Oet (5), Boc-Glu(OBzl)-O(Octyl) (6), Boc-Glu(OBzl)-O(Do-
decyl) (7), Boc-Glu(OBzl)-O(Tetradecyl) (8), Boc-Glu(OBzl)-
O(Octadecyl) (9) were synthesized similarly. A mixture of 4 and
catalytic amount of 10% palladium carbon in dioxane was stirred
under an atmosphere of hydrogen at room temperature for 19 h.
After filtration through celite pad, the filtrate was concentrated
under reduced pressure. Dimethylformamide, 3-hydroxybenzotria-
zole, N-methylmorpholine, and WSC were added, and the mixture
was stirred at room temperature for 2 h. The reaction mixture was
poured into water, extracted with ethyl acetate, washed with satu-
rated aqueous sodium chloride, dried with anhydrous magnesium
sulfate, and concentrated under reduced pressure. The residue was
purified by silica gel column chromatography to afford Boc-Glu(Cys-
(Bzl)-D-Phg-OEt)-O(Hexadecyl). Boc-Glu(Cys(Bzl)-D-Phg-OEt)-O(Hexa-
decyl) was treated with 4 N HCl in dioxane at room temperature, and
the reaction mixture was concentrated under reduced pressure.
Diethyl ether was added and the precipitated Glu(Cys(Bzl)-D-Phg-
OEt)-O(Hexadecyl) HCl (C16C2) was collected by filtration.
Similarly, Glu(Cys(Bzl)-D-Phg-OEt)-OEt HCl (C2C2), Glu(Cys-
(Bzl)-D-Phg-OEt)-O(Octyl) HCl (C8C2), Glu(Cys(Bzl)-D-Phg-OEt)-
O(Dodecyl) HCl (C12C2), Glu(Cys(Bzl)-D-Phg-OEt)-O(Tetradecyl)
HCl (C14C2), and Glu(Cys(Bzl)-D-Phg-OEt)-O(Octadecyl) HCl
(C18C2) were synthesized from compounds 5, 6, 7, 8, and 9,
respectively.
Cytotoxicity Assays. The sensitivities of each cultured cell line
to the anticancer drugs ADR, melphalan, 4-HC, VCR, CDDP, and
5-FU were determined by the dye-uptake method. Briefly, 1 ϫ 104
cells in 100 ␮l were dispensed in 96-well culture plates, and GSTP1-1
inhibitors were added at various concentrations. After incubation for
24 h at 37°C, anticancer drugs were added to each well at various
concentrations, and the cells were incubated for another 48 h at
37°C. Next, 25 ␮l of a 25% glutaraldehyde solution was added to each
well to fix the cells, and the plates were then washed with water,
dried, stained with a 0.05% methylene blue solution, and eluted with
0.33 N HCl. The absorbance at 665 nm was measured with an ELISA
reader (MS-3096F; SLT-LAB Instruments Co., Salzburg, Austria).
The cell survival rates were deduced from the relative absorbance
values of the samples to control. Because C16C2 was practically
insoluble in water, it was treated as follows before use. First, 10 mg
of L-␣-lecithin was dissolved in chloroform and air-dried. Ethanol
containing C16C2 at 30 mg/ml was then added to the air-dried
lecithin. After further addition of 10 ml of RPMI 1640 medium, the
mixture was sonicated for 10 min and then added to the culture
solution (Parsaee et al., 2002).
Assay for GSTP1-1 Activity in Cultured Cells. HuCCT1 cells
were seeded at a concentration of 2 ϫ 105/2 ml in a 12-well culture
dish and cultured in RPMI 1640 medium containing 10% FCS for
24 h. C16C2 was added to each well at various concentrations, and
the cells were cultured for another 24 h. They were harvested using
cell scrapers, incubated in hypotonic buffer (pH 7.4) (10 mM Tris, 1.5
mM MgCl2) for 20 min at 4°C, homogenized using Dounce homoge-
nizers, and centrifuged at 10,000g for 30 min at 4°C to collect cyto-
solic proteins. GSTP1-1 activities were measured using 1-chloro-2,
4-dinitrobenzene (CDNB) as a substrate, according to the method of
Habig et al. (1974). In brief, protein samples (10–50 ␮l) were added
to 1 ml of 0.1 M sodium phosphate buffer (pH 6.5) containing 1.3 mM
CDNB and 2.5 mM reduced glutathione (Sigma-Aldrich), and the
absorbance at 343 nm was measured at 25°C.
Xenotransplantation. The experiments were performed in ac-
cordance with the recommendations of the Guide for the Care and
Use of Laboratory Animals of Sapporo Medical University School of
Medicine. Cells (2 ϫ 106) in 100 ␮l of RPMI 1640 medium were
inoculated subcutaneously into the back of each nude mouse (5
weeks old; Sankyo Labo Service Co., Ltd, Tokyo, Japan). When
tumor sizes reached 7 mm in diameter, 36 mice were randomized
into six groups. To three groups, C16C2 was administered into the
tail vein at a dose of 20 mg/kg on days 1, 2, 3, 8, 9, and 10. Of these
three groups, two groups received intraperitoneal administrations of
either 4 mg/kg ADR or 200 mg/kg CPA on days 2 and 9. Of the
remaining three groups, two groups received intraperitoneal admin-
istrations of 4 mg/kg ADR or 200 mg/kg CPA on days 2 and 9. The
tumor size was measured with a sliding caliper every 4 days. Tumor
volume (V) was calculated with the following formula: V ϭ length ϫ
(width)2 ϫ 0.5, in accordance with the protocol of Geran et al. (1972).
To measure the GSTP1-1 activities in tumor tissues of xenografts,
tumors were resected on day 3, minced in hypotonic buffer, and the
cytosol fractions were extracted. Then, GST activities were mea-
sured as described above.
Pharmacokinetic Study of GSTP1-1 Inhibitors. GSTP1-1 in-
hibitors were administered in vivo by formulation with lipid micro-
spheres (LMs) to facilitate their solubility in plasma. In brief, they
were dissolved in a mixture of benzyl alcohol and ethanol (1:4) at 40
mg/ml and then added to a fat emulsion (Intralipos; Nihonseiyaku
Inc., Tokyo, Japan) at a ratio of 1:0.075, mixed well by vortexing, and
subsequently passed through a 1.2-␮m pore-sized filter. The mean
particle size of the emulsified LMs thus synthesized was 243 Ϯ 86
nm (data not shown), which is suitable for in vivo administration
(Yamaguchi and Mizushima, 1994). LM solution containing
GSTP1-1 inhibitors was intravenously injected into a rabbit at 3 to
30 ml/kg. Blood was sampled 5, 10, 15, 30, 60, 120, and 240 min after
injection, and the plasma concentration of inhibitors was determined
by high-performance liquid chromatography (HPLC).
HPLC. A reverse phase HPLC was performed on the HPLC col-
umn (YMC-Pack Pro C18, 4.6 ϫ 25 cm; YMC, Inc., Tokyo, Japan) with
an LC-10A equipped with a UV detector (Shimadzu, Kyoto, Japan).
In Vitro Deesterification of C16C2 by Esterase. C16C2 (200
␮g/ml) was incubated with rabbit liver esterase (134 units/ml) in
PBS containing bovine serum albumin at 37°C for 24 h to hydrolyze
the C16C2 ester bonds. The resultant solution was applied to the
HPLC column to analyze the presence of C16C2 and its active form,
␥-glutamyl-S-benzylcysteinyl-phenylglycine. As an internal stan-
dard, parahydroxymethyl benzoate (10 ␮g/ml) was added to the
solution before application to the reverse phase HPLC column. The
Results
Immunohistochemical Analysis for GSTP1-1 in Chol-
angiocarcinoma and Hepatocellular Carcinoma. We
performed immunohistochemical staining for GSTP1-1 on
specimens of 17 cholangiocarcinoma (four cases of well dif-
ferentiated type, seven of moderately differentiated type, and
six of poorly differentiated type) and 19 hepatocellular carci-
noma cases. Figure 2, A to C, is representative staining
solution of the active form itself (10 ␮g/ml) containing parahy- patterns of cholangiocarcinoma of the well differentiated

864
Nakajima et al.
Fig. 2. Immunohistochemical staining for GSTP1- 1 in a cholangiocarcinoma specimen. The cytoplasms of tumor cells in well differentiated
cholangiocarcinoma (A and D), moderately differentiated cholangiocarcinoma (B and E) and poorly differentiated cholangiocarcinoma (C and F) were
all strongly stained for GSTP1-1. There was essentially no staining for GSTP1-1 in hepatocellular carcinoma (G). Original magnification, 50ϫ in A and
C, 100ϫ in B and G, 200ϫ in D, and 400ϫ in E and F.
TABLE 2
type, moderately differentiated type, and poorly differenti-
ated type, respectively. In all samples, the tumor cells were
Intracellular GSTP1-1 concentration in various cell lines
Cell Lines
GSTP1-1 Concentrationa
strongly stained for GSTP1-1. Figure 2, D to F, is higher
magnification pictures of the corresponding Fig. 2, A to C.
The cytoplasms of tumor cells in all types of cholangiocarci-
noma were strongly stained for GSTP1-1, although some
infiltrating cells showed very weak staining for GSTP1-1. In
contrast, hepatocellular carcinoma showed essentially no
staining for GSTP1-1 (Fig. 2G). The results of immunostain-
ing for GSTP1-1 in 17 cholangiocarcinoma cases and 19 hep-
atocellular carcinoma cases are summarized in Table 1. All
cholangiocarcinoma specimens showed strong staining (pos-
itivity) for GSTP1-1, irrespective of the histological types,
whereas all hepatocellular carcinoma cases were negative for
GSTP1-1.
ng/ml/105 cells
HeLab
25.0 Ϯ 2.2
58.7 Ϯ 6.9
10.0 Ϯ 0.9
76.3 Ϯ 9.7
90.5 Ϯ 9.5
80.7 Ϯ 8.2
168.0 Ϯ 13.6
132.6 Ϯ 14.0
MCF7c
TMK-1d
HepG2e
PLC/PRF/5f
M7609g
HuCCT1h
HuH28i
a GST-p concentration was determined by ELISA. Values are means Ϯ S.D. of
three independent experiments.
b A human cervical carcinoma line; c a human breast carcinoma line; d a human
gastric carcinoma line; e,f human hepatocellular carcinoma lines; g a human colon
carcinoma line; h,i human cholangiocarcinoma lines.
Intracellular GSTP1-1 Concentration in Various Cell
Lines. We measured the concentrations of GSTP1-1 in HeLa
cells from human cervical carcinoma, MCF7 cells from hu-
man mammary carcinoma, TMK-1 cells from human gastric
carcinoma, HepG2 and PLC/PRF/5 cells from human hepa-
tocellular carcinoma, M7609 cells from human colon carci-
noma, and HuCCT1 and HuH28 cells from human cholan-
giocarcinoma using ELISA (Table 2). The GSTP1-1
concentrations in the two cell lines (HuCCT1 and HuH28)
derived from cholangiocarcinoma were clearly higher than
the others, and the HuCCT1 cell line, which had the highest
GSTP1-1 concentration, was used for the following examina-
tion.
Anticancer Drug Sensitivities of HuCCT1/anti-
GSTP1 and HuCCT1/antiGSTP2 Cells. To examine
whether GSTP1-1 expressed in HuCCT1 cells was involved
in the resistance to anticancer drugs, we introduced GSTP1-1
antisense cDNA into HuCCT1 cells and investigated changes
in their drug sensitivity. The intracellular GSTP1-1 concen-
trations in the two clones, HuCCT1/antiGSTP1 and
HuCCT1/ antiGSTP2, were decreased to about half of those
of the parental cell line and the control cells, HuCCT1/PLJ.
Both the HuCCT1/antiGSTP1 and the HuCCT1/antiGSTP2
cells showed significantly elevated sensitivities to ADR, mel-
phalan, 4-HC, and CDDP compared with that of the
HuCCT1/PLJ cells. Neither HuCCT1/antiGSTP1 nor
HuCCT1/antiGSTP2 cells showed statistically significant
changes in their sensitivities to VCR or 5-FU (Table 3).
Pharmacokinetic Study of Synthesized GSTP1-1 In-
hibitors in Rabbit. Because the plasma esterase activity is
similar in rabbits and humans, which is low compared with
the activity in rats or mice, we used rabbits to examine the
pharmacokinetics of GSTP1-1 inhibitors (Fig. 3). The plasma
concentration of C2C2 was already less than 0.1 ␮g/ml at 10
min after administration and decreased sharply within 30
TABLE 1
Summary of immunohistochemical staining for GSTP1-1 in
cholangiocarcinoma and hepatocellular carcinoma
Number of
Cases
Examined
GSTP1-1
Positive
(%)
Cholangiocarcinoma (CCC)
well.
mod.
por.
4
7
4 (100)
7 (100)
6 (100)
17 (100)
0 (0)
6
total
17
19
Hepatocellular carcinoma (HCC)

Reversal of a Drug Resistance by the GST-␲ Inhibitor
865
TABLE 3
Comparison of the sensitivities of HuCCT1, HuCCT1/pLJ, HuCCT1/antiGSTP1, and HuCCT1/anti-GSTP2 cells to various drugs
a
IC50
Cell Line
GSTP1-1 Concentration
ADR
Melphalan
4-HC
CDDP
VCR
5-FU
ng/ml/105 cells
␮M
HuCCT1
172 Ϯ 14.5
168 Ϯ 15.5
88.7 Ϯ 9.0**
90.1 Ϯ 10.2**
3.30 Ϯ 0.27
3.26 Ϯ 0.23
0.86 Ϯ 0.26**
1.01 Ϯ 0.25**
75.0 Ϯ 4.8
8.21 Ϯ 0.96
7.48 Ϯ 0.90
3.58 Ϯ 0.67**
3.98 Ϯ 0.52**
19.8 Ϯ 1.3
495 Ϯ 53
492 Ϯ 46
500 Ϯ 57
530 Ϯ 64
9.6 Ϯ 1.0
9.9 Ϯ 1.0
9.5 Ϯ 0.9
10.4 Ϯ 1.5
HuCCT1/pLJ
88.0 Ϯ 6.1
21.0 Ϯ 1.6
HuCCT1/antiGSTP1
HuCCT1/antiGSTP2
40.0 Ϯ 2.3**
46.5 Ϯ 2.4**
10.0 Ϯ 1.0**
11.2 Ϯ 1.0**
** The value was significantly lower than that of HuCCT1/pLJ cells (p Ͻ 0.01 by Student’s t test).
a IC50, drug concentration that inhibits cell growth by 50%. Values are means Ϯ S.D. of three independent experiments.
steady state (Vss) were the smallest for C16C2, whereas
the total body clearance (CL) was the smallest for C18C2
among the synthesized inhibitors (Table 4). Thus, C16C2
most effectively of all inhibitors maintained a sustained
high blood concentration. In addition, we have confirmed
that C16C2 showed high stability, even in rats and mice
(data not shown).
Deesterification of C16C2 in Vitro. C16C2 was incu-
bated in vitro with an esterase for 24 h to examine whether
an active form of C16C2 could be indeed generated during the
esterase treatment. The control active form of a GSTP1-1-
specific inhibitor, ␥-glutamyl-S-benzylcysteinyl-phenylgly-
cine, was eluted at the 16-min site on a reverse phase HPLC
(Fig. 4B). An analysis of C16C2 before the esterase treatment
demonstrated no peak at the 16-min site (Fig. 4C). After
C16C2 was incubated with an esterase for 24 h, an analysis
of the resultant solution by reverse phase HPLC revealed a
small peak at the 16 min site which represented the elution
of the active form of C16C2 (Fig. 4D).
Inhibitory Effects of C16C2 on GSTP1-1 Activity in
HuCCT1 Cells and Its Effects on Sensitivities to Antican-
cer Drugs in Comparison with Those of C2C2. We exam-
ined whether GSTP1-1 activity in HuCCT1 cells could be in-
deed inhibited by C16C2. As shown in Fig. 5A, GSTP1-1 activity
in HuCCT1 cells was dose dependently inhibited by treatment
with both C16C2 and C2C2, although the inhibition rate of
C16C2 at each dose was slightly lower than that of C2C2 with
no significant difference. Incidentally, there was no significant
cytotoxicity of C16C2 to HuCCT1 cells when it was used at the
concentration of 0 to 200 ␮M (data not shown). We next evalu-
ated the ability of C16C2 to potentiate the killing effect of ADR
or 4-HC on HuCCT1 cells. The IC50 value for both ADR (Fig.
5B) and 4-HC (Fig. 5C) decreased by treatment with C16C2 and
C2C2 in a dose-dependent manner, paralleling the decrease of
Fig. 3. Plasma concentration of GSTP1-1-specific inhibitors after intra-
venous administration in rabbits. GSTP1-1 inhibitors were intravenously
administered into rabbits by formulation with lipid microspheres, as
described under Materials and Methods. Blood was sampled 5, 10, 15, 30,
60, 120, and 240 min after injection, and the plasma concentration was
measured by HPLC. As the number of the carbon chains in the ester at
the N-terminal increased, the plasma concentration of the GSTP1-1 in-
hibitors gradually increased. There was no apparent difference between
the maximal concentrations of C16C2 and C18C2.
min. As the number of the carbons in the alkylester at the N
terminus increased from eight (C8C2) to 18 (C18C2), the
plasma concentration of GSTP1-1 inhibitors gradually in-
creased. The maximal plasma concentration of C16C2 was 17.2
␮g/ml at 5 min. There was no apparent difference between the
maximal plasma concentrations of C16C2 and C18C2.
The plasma inhibitor concentration was analyzed as a
function of time by the nonlinear least-squares program
MULTI, according to a two-compartment model (Metzler, GSTP1-1 activity (Fig. 5A). The decrement rates of IC50
and GSTP1-1 activity with C16C2 was also lower than that
of C2C2 with no significant difference.
1971; Zuideveld et al., 2002) (Table 4). The volume of
central compartment (Vc) and volume of distribution at
Effects of C16C2 on Sensitivities to Anticancer
Drugs in Xenograft Models. We examined whether the
antitumor activity of ADR or 4-HC against HuCCT1 tumors
transplanted into nude mice was enhanced by C16C2. C16C2
itself did not exert any effect on tumor growth (Fig. 6) and
exhibited no apparent adverse effects such as weight loss,
abnormal liver function, or dysfunction of hematopoiesis
(data not shown). When C16C2-treated mice were examined
histologically, liver, kidney, heart, lung, stomach, and intes-
tine showed no apparent abnormalities. Although either
ADR or CPA alone showed some inhibitory activity on tumor
growth, when they were combined with C16C2, the inhibi-
tory activity became more evident. Moreover, the tumor dis-
TABLE 4
Pharmacokinetic parameters obtained in two-compartment model.
The plasma inhibitor concentration versus time curve was analyzed by nonlinear least
squares program.
Compound
Vc
Vss
CL
l/kg
l/kg
l/h/kg
C2C2
C8C2
C12C2
C14C2
C16C2
C18C2
0.799
4.183
0.465
0.080
0.061
0.067
1.063
10.968
0.787
0.140
0.064
0.130
15.274
9.344
2.851
0.292
0.119
0.034

866
Nakajima et al.
Fig. 4. Deesterification of C16C2 in vitro. C16C2 (200 ␮g/ml) was incubated with a rabbit esterase (134 units/ml) for 24 h and applied to a reverse
HPLC. A, PBS containing BSA was applied to a reverse phase HPLC as a blank. B, ␥-glutamyl-S-benzylcysteinyl-phenylglycine (10 ␮g/ml), the active
form of the GSTP1-1-specific inhibitor, was applied to the column and eluted as a peak at 16 min. Parahydroxymethyl benzoate (10 ␮g/ml), an internal
standard (I.S.), was eluted as a peak at 12 min. C and D, elution profiles of C16C2 before and after esterase treatment, respectively. The small peak
at 16 min, which represents the active form, was detected after esterase treatment.
appeared completely in one of the six mice in which CPA was ␮M/min/mg of protein, whereas that of the mice administered
administered in combination with C16C2.
with C16C2 was 6.1 Ϯ 1.1 ␮M/min/mg of protein. The result
GST Activities in Tumor Tissues of the Mice Receiving indicated that the GSTP1-1 activity was inhibited to 45% of the
C16C2. To verify that the augmented antitumor effect of anti- control level by administration of C16C2.
cancer drugs combined with C16C2 is due to the inhibition of
GSTP1-1 activity by the latter agent, we measured GSTP1-1
activity in tumors resected from mice. The mean GSTP1-1 ac-
Discussion
In the present study, we examined the expression of
tivity in tumor tissues of the control mice was 13.6 Ϯ 1.9 GSTP1-1, which is known as a resistance factor for antican-

Reversal of a Drug Resistance by the GST-␲ Inhibitor
867
Fig. 5. Inhibitory effects of C16C2 on GSTP1-1 activity in HuCCT1 cells and its effects on sensitivities to anticancer drugs. Inhibitory effects of C2C2
and C16C2 on GSTP1-1 activity of HuCCT1 cells were examined by treating the cells with these agents for 24 h and measuring their GST activities
by using CDNB and glutathione as substrates (A). The GSTP1-1 activity was dose dependently inhibited by treatment with both agents. The inhibition
rate was slightly greater for C2C2 than C16C2, although the difference was not statistically significant. The IC50 value of HuCCT1 cells for ADR (B)
or 4-HC (C), which was determined by the dye-uptake method, decreased by treatment with both C2C2 and C16C2 in a dose-dependent manner. Again,
C2C2 was slightly more potent than C16C2 in decreasing IC50, although the difference was insignificant. Values are means Ϯ S.D. of three
independent experiments.
Fig. 6. Effects of C16C2 on sensitivities to anti-
cancer drugs in xenograft models. HuCCT1 cells
(2 ϫ 106) were inoculated subcutaneously into
nude mice. When tumor sizes reached 7 mm in
diameter, C16C2, ADR, and CPA were adminis-
tered. There was no significant difference in tu-
mor growth between the group treated with
C16C2 alone and the control group (no treat-
ment). Although ADR and CPA alone showed
some inhibitory activity on tumor growth, when
they were combined with C16C2, the inhibitory
activity became more evident. In one of the six
mice which received CPA in combination with
C16C2, the tumor disappeared completely.
cer drugs in cholangiocarcinoma. In all 17 cholangiocarci- tal cells. These findings suggested that GSTP1-1 plays a role
noma samples, GSTP1-1 was positively stained by immuno- in anticancer drug resistance of cholangiocarcinoma and
histochemistry. When the antisense cDNA of GSTP1-1 was were compatible with the results of previous studies showing
introduced into the cholangiocarcinoma cell line HuCCT1, that GSTP1-1 was associated with the chemoresistance
the intracellular GSTP1-1 concentration was apparently de- against these anticancer drugs in various cell lines (Ban et
creased, and the transfectants showed a significantly higher al., 1996; Penketh et al., 1996; Kuga et al., 1997; Niitsu et al.,
sensitivity to ADR, melphalan, 4-HC, and CDDP than paren- 1998; Goto et al., 2001).

868
Nakajima et al.
There have been several studies to overcome the antican- readily convert into the active form due to a high concentra-
cer drug resistance using GST inhibitors. Tew et al. (1988) tion of esterase (Butterworth et al., 1993; van Ark-Otte et al.,
reported that the sensitivity to chlorambucil increased in 1998), and thus it exerts nearly the same potent anti-
cells of the Walker 256 rat mammary tumor cell line and a GSTP1-1 activity as C2C2. Nevertheless, this property of
human colon cancer cell line treated with pyriplost, an ana- C16C2 is considered to be quite suitable for in vivo use.
log of prostaglandin I, and ethacrynic acid, a diuretic com- Furthermore, for administration of C16C2, we prepared col-
pound as well as a substrate for GST. However, ethacrynic loidal microspheres containing C16C2 to facilitate its trans-
acid has a very problematic side effect in addition to its fer into cancerous tissues. Accordingly, the effect of C16C2
ability to induce the metabolic abnormality and its strong was examined in vivo with tumor-bearing mice. C16C2 was
diuretic activity because it suppresses the bone marrow func- administered three consecutive days (on the day of adminis-
tion markedly when used together with anticancer drugs. tration of an anticancer drug, and on the days before and
Hall et al. (1989) reported that a nonsteroidal anti-inflam- after that) because GSTP1-1 must be inhibited before and
matory drug (NSAID), indomethacin, known to bind to during anticancer drug administration. The applied dose was
GSTP1-1, is useful in overcoming chlorambucil resistance. determined based on the inhibitory activity of C16C2 on
We ourselves found that ketoprofen, one of the NSAIDs that GSTP1-1 in vitro (Fig. 5) and the plasma concentration of
inhibited the GST activity, clearly overcame the ADR resis- C16C2 in vivo (Fig. 3). We found that the combination of
tance (Niitsu et al., 1990). In these reports, however, the C16C2 with ADR and CPA evoked significantly higher anti-
effectiveness of NSAIDs in vivo has not yet been clarified. tumor effects than administration of each drug in isolation.
Maeda et al. (1993) synthesized a calmodulin antagonist We also found that the tumor disappeared completely in one
(Ca2ϩ antagonist), W-77, which blocked GSTP1-1 activity, of the mice in which CPA was administered in combination
and reported that it enhanced the ADR sensitivity through with C16C2. These results clearly indicated that C16C2 is
inhibition of both GSTP1-1 and p-glycoprotein. However, it highly potent in overcoming the chemoresistance caused by
was highly toxic in vivo by itself.
GSTP1-1. Incidentally, in this study we presented the results
Lyttle et al. (1994) recently synthesized a glutathione an- of experiments in which animals were treated with two
alog (␥-glutamyl-S-benzylcysteinyl-phenylglycyl diethylester), courses of combination therapy. The results were much more
which was designed to block GSTP1-1 activity by directly favorable than those of one-course therapy (data not shown),
binding to its glutathione site after entrance into the cells, suggesting the possibility that the efficacy of treatment may
where diethyester is hydrolyzed by esterase to become ␥-glu- increase as the number of therapy course increases. How-
tamyl-S-benzylcysteinyl-phenylglycine, and they demon- ever, multiple therapy courses are not practically feasible in
strated that the analog had extremely high and specific in- this model because of damage to the tail vein, where the
hibitory activity to GSTP1-1 with a Ki value of 0.42 ␮M drugs are injected. This obstacle will be circumvented when
compared with that to GST A1-1 (Ki value of 24.3 ␮M), GST therapy is applied to humans in the future.
M1-1 (Ki value of 57.8 ␮M), and GST M2-2 (Ki value of 184
Because ␥-glutamyl-S-benzylcysteinyl-phenylglycine, the
␮M). Morgan et al. (1996) reported further that the treat- active form of C16C2 has a high selectivity to GSTP1-1, it is
ment with this glutathione analog of a colon cancer cell line not expected to affect the activity of other GST isozymes, and
in which GSTP1-1 was highly expressed resulted in an in- if it does, the affect should not be serious. However, further
crease of sensitivity to chlorambucil. The half-life of this detailed investigations are needed. With regard to toxicity of
compound, however, was so short in vivo that it was difficult C16C2, there was no significant cytotoxicity to HuCCT1 cells
to attain an effective blood concentration. The short half-life when it was used at the concentration of 0 to 200 ␮M. In the
was conceivably attributable to the readiness of two ethyl xenograft experiment, mice administered C16C2 showed no
groups to be hydrolyzed by esterase in the blood. To circum- significant decrease in body weight compared with those
vent this problem, we therefore elongated the ethyl group at administered vehicle alone, and no blood chemical abnormal-
the N terminus of ␥-glutamyl-S-benzylcysteinyl-phenylglycyl ity was found. Histological examinations of liver, kidney,
diethylester with long alkyl carbon chains to form various heart, lung, stomach and intestine showed no apparent ab-
compounds (C8C2, C12C2, C14C2, C16C2, and C18C2), normalities. Moreover, the LD50 value of C16C2 was very
which should be resistant to deesterification in the blood. high, ranging from 800 to 1,000 mg/kg (data not shown).
When we administered the above-mentioned compounds to However, a more detailed examination of C16C2 toxicities
rabbits, the blood concentration of compounds with longer should be performed. In conclusion, C16C2 is considered to
chains (C16C2 and C18C2) was indeed maintained at a be useful for treatment of cholangiocarcinoma in combination
higher level. Because C16C2 showed the smallest volume of with anticancer drugs to which GSTP1-1 is a resistant factor.
central compartment (Vc) and the smallest volume of distri-
bution at steady state (Vss) by a two-compartment model
analysis, we chose C16C2 for further experiments. To verify
the in vivo observation, we then incubated C16C2 with es-
terase in vitro for 24 h. The result that only a small amount
of the deesterified form was detected on HPLC indeed indi-
cated the stable nature of C16C2 against esterase. However,
when in vitro activity of C16C2 on tumor cells (Fig. 5) was
examined, it was almost as high as that of C2C2, although
slightly impaired. Together, it was surmised that C16C2 is
quite stable as an inactive form in circulation despite the
presence of esterase, whereas intracellularly, the agent may
References
Ban N, Takahashi Y, Takayama T, Kura T, Katahira T, Sakamaki S, and Niitsu Y
(1996) Transfection of glutathione S-transferase (GST)-␲ antisense complemen-
tary DNA increases the sensitivity of a colon cancer cell line to adriamycin,
cisplatin, melphalan and etoposide. Cancer Res 56:3577–3582.
Butterworth M, Upshall DG, and Cohen GM (1993) A novel role for carboxylesterase
in the elevation of cellular cysteine by esters of cysteine. Biochem Pharmacol
46:1131–1137.
de Groen PC, Gores GJ, LaRusso NF, Gunderson LL, and Nagorney DM (1999)
Biliary tract cancers. N Engl J of Med 341:1368–1378.
Erickson BW and Merrifield RB (1976) The Proteins, 3rd ed., vol 2, pp 255–292,
Academic Press, New York.
Geran RI, Greenberg NH, Macdonald MM, Schumacher AM, and Abbott BJ (1972)
Protocols for screening chemical agents and natural products against animal
tumors and other biological systems. Cancer Chemother Rep Part 3 3:1–103.

Reversal of a Drug Resistance by the GST-␲ Inhibitor
869
Goto S, Ihara Y, Urata Y, Izumi S, Abe K, Koji T, and Kondo T (2001) Doxorubicin-
induced DNA intercalation and scavenging by nuclear glutathione S-transferase
pi. FASEB J 15:2702–2714.
Habig WH, Pabst MJ, and Jakoby WB (1974) Glutathione S-transferases. The first
enzymatic step in mercapturic acid formation. J Biol Chem 249:7130–7139.
Hall A, Robson CN, Hickson ID, Harris AL, Proctor SJ, and Cattan AR (1989)
Possible role of inhibition of glutathione S-transferase in the partial reversal of
chlorambucil resistance by indomethacin in a Chinese hamster ovary cell. Cancer
Res 49:6265–6268.
Hayes PC, May L, Hayes JD, and Harrison DJ (1991) Glutathione S-transferases in
human liver cancer. Gut 32:1546–1549.
Isa T, Kusano T, Shimoji H, Takeshima Y, Muto Y, and Furukawa M (2001)
Predictive factors for long-term survival in patients with intrahepatic cholangi-
carcinoma. Am J Surg 181:507–511.
Isa T, Tomita S, Nakachi A, Miyazato H, Shimoji H, Kusano T, and Furukawa M
(2002) Analysis of microsatellite instability, K-ras gene mutation and p53 protein
overexpression in intrahepatic cholangiocarcinoma. Hepatogastroenterology 49:
604–608.
Kiba T, Tsuda H, Pairojkul C, Inoue S, Sugimura T, and Hirohashi S (1993) Muta-
tion of the p53 tumor suppressor gene and the ras gene family in intrahepatic
cholangiocarcinomas in Japan and Thailand. Mol Carcinog 8:312–318.
Korman AJ, Frantz JD, Strominger JL, and Mulligan RC (1987) Expression of
human class II major histocompatibility complex antigens using retrovirus vec-
tors. Proc Natl Acad Sci USA 84:2150–2154.
Kuga T, Sakamaki S, Matsunaga T, Hirayama Y, Kuroda H, Takahashi Y, and
Niitsu Y (1997) Fibronectin fragment facilitated retroviral transfer of the gluta-
thione S-transferase pi gene into CD34ϩ cells to protect them against alkylating
agents. Hum Gene Ther 8:1901–1910.
Kura T, Takahashi Y, Takayama T, Ban N, Saito T, Kuga T, and Niitsu Y (1996)
Glutathione S-transferase ␲ is secreted as a monomer into human plasma by
platelets and tumor cells. Biochim Biophys Acta 1209:317–323.
Lyttle MH, Hocker MD, Hui HC, Caldwell CG, Aaron DT, Engqvist-Goldstein A,
Flatgaard JE, and Bauer KE (1994) Isozyme-specific glutathione-S-transferase
inhibitors: design and synthesis. J Med Chem 37:189–194.
Nucleotide sequence of pi class glutathione S-transferase in human fetal liver. Res
Commun Mol Pathol Pharmacol 97:67–78.
Niitsu Y, Ishigaki S, and Takahashi Y (1990) GST-pi assay for serodiagnosis of
malignancy, in Glutathione S-Transferase and Drug Resistance (Heyes JD, Pickett
LB, and Mantle TJ eds) pp 409–417, Taylor & Francis Ltd., London.
Niitsu Y, Takahashi Y, Ban N, Takayama T, Saito T, Katahira T, Umetsu Y, and
Nakajima T (1998) A proof of glutathione S-transferase-␲-related multidrug re-
sistance by transfer of antisense gene to cancer cells and sense gene to bone
marrow stem cells. Chem Biol Interact 24:325–332.
Ohashi K, Nakajima Y, Kanehiro H, Tsutsumi M, Taki J, Aomatsu Y, and Yagura K
(1995) Ki-ras mutations and p53 protein expressions in intrahepatic cholangiocar-
cinomas: relation to gross tumor morphology. Gastroenterology 109:1612–1617.
Okuda N, Nakanuma Y, and Miyazaki
M (2002) Cholangiocarcinoma: recent
progress. Part 1:epidemiology and etiology. J Gastroenterol Hepatol 17:1049–
1055.
Parsaee S, Sarbolouki MN, and Parnianpour M (2002) In vitro release of diclofenac
diethylammonium from lipid-based formulations. Int J Pharm 241:185–190.
Patel T (2001) Increasing incidence and mortality of primary intrahepatic cholan-
giocarcinoma in united states. Hepatology 33:1353–1357.
Penketh PG, Shyam K, and Sartorelli AC (1996) Mechanisms of resistance to alky-
lating agents. Cancer Treat Res 87:65–81.
Tada M, Omata M, and Ohto M (1990) Analysis of ras gene mutations in human
hepatic malignant tumors by polymerase chain reaction and direct sequencing.
Cancer Res 50:1121–1124.
Takahashi Y, Hirata Y, Saito T, Yamashina T, Niitsu Y, Suzuki E, and Hosoda K
(1989) Enzyme immunoassay for human serum glutathione S-transferase ␲ using
monoclonal and polyclonal antibodies. Cancer J 2:225–229.
Tannapfel A, Benicke M, Katalinic A, Uhlmann D, Kockerling F, Hauss J, and
Wittekind C (2000) Frequency of p16(INK4A) alterations and K-ras mutations in
intrahepatic cholangiocarcinoma of the liver. Gut 47:721–727.
Tew KD, Bomber AM, and Hoffman SJ (1988) Ethacrynic acid and piriprost as
enhancers of cytotoxicity in drug resistant and sensitive cell lines. Cancer Res
48:3622–3625.
van Ark-Otte J, Kedde MA, van der Vijgh WJ, Dingemans AM, Jansen WJ, Pinedo
HM, and Boven E (1998) Determinants of CPT-11 and SN-38 activities in human
lung cancer cells. Br J Cancer 77:2171–2176.
Yamaguchi T and Mizushima Y (1994) Lipid microspheres for drug delivery from the
pharmaceutical viewpoint. Crit Rev Ther Drug Carrier Syst 11:215–229.
Zuideveld KP, Rusic-Pavletic J, Maas HJ, Peletier LA, van der Graaf PH, and
Danhof M (2002) Pharmacokinetic-pharmacodynamic modeling of buspirone and
its metabolite 1-(2-pyrimidinyl)-piperazine in rats. J Pharmacol Exp Ther 303:
1130–1137.
Maeda O, Terasawa M, Ishikawa T, Oguchi H, Mizuno K, Kawai M, Kikkawa F,
Tomoda Y, and Hidaka H (1993) A newly synthesized bifunctional inhibitor, W-77,
enhances adriamycin activity against human ovarian carcinoma cells. Cancer Res
53:2051–2056.
Metzler CM (1971) Usefulness of the two-compartment open model in pharmacoki-
netics. J Am Stat Assoc 66:49–53.
Miyanishi K, Takayama T, Ohi M, Hayashi T, Nobuoka A, Nakajima T, and Niitsu
Y (2001) Glutathione S-transferase-␲ overexpression is closely associated with
K-ras mutation during human colon carcinogenesis. Gastroenterology 121:865–
874.
Morgan AS, Ciaccio PJ, Tew KD, and Kauvar LM (1996) Isozyme-specific glutathione
S-transferase inhibitors potentiate drug sensitivity in cultured human tumor cell
lines. Cancer Chemother Pharmacol 37:363–370.
Address correspondence to: Dr. Yoshiro Niitsu, The 4th Department of
Internal Medicine, Sapporo Medical University School of Medicine, South-1,
West-16, Chuo-ku, Sapporo, Japan, 060-8543. E-mail: niitsu@sapmed.ac.jp
Nakasa H, Mera N, Ohmori S, Itahashi K, Igarashi T, Ishii I, and Kitada M (1997)