Potent in vitro and in vivo antitumor activity of interleukin-4-conjugated
Pseudomonas exotoxin against human biliary tract carcinoma
Kazunori Ishige1,2, Junichi Shoda1, Toru Kawamoto3, Sachiko Matsuda2, Tetsuya Ueda4, Ichinosuke Hyodo1,
Nobuhiro Ohkohchi5, Raj K. Puri6and Koji Kawakami2,7*
1Department of Gastroenterology, Institute of Clinical Medicine, University of Tsukuba Graduate
School of Comprehensive Human Sciences, Tsukuba-Shi, Ibaraki, Japan
2Department of Advanced Clinical Science and Therapeutics, Graduate School of Medicine, The University of Tokyo,
Bunkyo-Ku, Tokyo, Japan
3Department of Medical Oncology, M.D. Anderson Cancer Center, University of Texas, Houston, TX
4Drug Development Service Division, Pharmacodynamics Group, Medi-Chem Business Segment,
Mitsubishi Chemical Medience Corporation, Itabashi-Ku, Tokyo, Japan
5Department of Surgery, Institute of Clinical Medicine, University of Tsukuba Graduate School
of Comprehensive Human Sciences, Tsukuba-Shi, Ibaraki, Japan
6Division of Cellular and Gene Therapies, Center for Biologics Evaluation and Research,
Tumor Vaccines and Biotechnology Branch, Food and Drug Administration, Bethesda, MD
7Department of Pharmacoepidemiology, Graduate School of Medicine and Public Health, Kyoto University, Kyoto, Japan
Targeting cytotoxins or immunotoxins to tumor cell surface recep-
tors represents a new approach for the treatment of cancers. We
tested the antitumor activity of a cytotoxin (IL-4-PE) composed of
an interleukin-4 (IL-4) targeting moiety and a truncated form of
Pseudomonas exotoxin A against human biliary tract carcinoma
(BTC). Immunohistochemical analysis showed that cultured BTC
cell lines and cancerous epithelia in BTC tissue (e.g., gallbladder
carcinoma, extraheaptic cholangiocarcinoma, and intrahepatic
cholangiocarcinoma) expressed receptors for IL-4 in situ at high
densities. However, normal epithelial cells in gallbladder and bile
duct tissues did not express these IL-4 receptors. Eight BTC cell
lines expressed IL-4R on the cell surface as determined by radiola-
beled ligand binding assays. When these cells were treated with
IL-4-PE, significant cytotoxicity was observed as determined by
the inhibition of protein synthesis. The concentration of agent
causing 50% inhibition of protein synthesis (IC50) was found to be
less than 10 ng/mL in 4 of the 8 BTC cell lines studied. The anti-
tumor activity of IL-4-PE was assessed for human BTC cells
implanted subcutaneously in immunodeficient mice. By intratu-
moral injection of IL-4-PE, complete disappearance of the estab-
lished tumors was observed in 40% of animals. Intraperitoneal
administration of IL-4-PE at tolerated doses to animals with peri-
toneally disseminated BTC exhibited significantly prolonged sur-
vival compared to untreated animals (>14 weeks vs. 5 weeks in
treated and untreated mice, respectively). These results indicate
that IL-4 receptor-targeted cytotoxin is a potent agent that may
provide a new therapeutic option for BTC.
' 2008 Wiley-Liss, Inc.
Key words: biotherapy; cytotoxin; cytokine receptor; gallbladder
Biliary tract carcinoma (BTC), that is, gallbladder carcinoma
and cholangiocarcinoma, are relatively rare diseases worldwide,
but are rather aggressive and frequently lethal. The incidence of
BTC varies greatly according to geographic location, and high
standardized mortality ratios for BTC have been found on analysis
of carcinoma registry data from South American countries such as
Chile, Peru and Columbia, and Asian countries such as Japan and
Thailand.1Japan has one of the world’s highest age-adjusted death
rates for carcinoma related to BTC, and the rate seems to be
increasing steadily (rates were 5.7 and 11.5% in 1980 and 1998,
respectively).2There are indications that the incidence of both
gallbladder carcinoma3,4and cholangiocarcinoma5–7may also be
increasing in the United States and United Kingdom.
Despite advances in tumor biology, diagnostic imaging, adju-
vant therapies and surgical techniques, the survival of BTC
patients remains dismal.8–11Without early resection, these tumors
invade adjacent vascular structures and liver parenchyma, extend
beneath the bile duct epithelia, metastasize to the lymph nodes
and disseminate into the peritoneal space. The mainstay of treat-
ment for BTC is complete resection with tumor free surgical mar-
gins. Under these circumstances, however, the role of adjuvant
therapy is not well defined, and the benefits of palliative chemo-
therapy or chemoradiation in patients with unresectable disease
have not been established. An effective new approach against this
aggressive disease is urgently needed.
Interleukin-4 receptor (IL-4R) is expressed in hematological
cells such as B-cells and T-cells, and its ligand, IL-4, is known to
mediate cell proliferation and immune response.12Interestingly, a
number of recent in vitro and in vivo studies have shown that a va-
riety of carcinoma cell lines, including hematological malignan-
cies, overexpress functional IL-4R and its ligand.13,14IL-4 has
been shown to have a modest but direct inhibitory effect on
several carcinoma cells in vitro15,16and to enhance immune
response.17Based on these findings, IL-4 has been tested clinically
as a treatment for various malignancies,18,19but the results have
been disappointing and the expected benefit was not obtained.
With the aim of more directly targeting IL-4R in carcinoma cells,
we have tailored an effective new approach against pancreatic car-
cinoma using circularly permuted interleukin-4 conjugated with a
truncated form of Pseudomonas exotoxin A (IL-4-PE)20or IL-4-
PE in combination with gemcitabine.21Pseudomonas exotoxin A
(PE) is a protein secreted by Pseudomonas aeruginosa, which is
highly toxic to eukaryotic cells. Once PE is internalized into the
cytoplasm of a eukaryotic cell, through enzymatic ADP-ribosyla-
tion of translocation elongation factor 2, it causes arrest of protein
synthesis leading to cell death.22IL-4-PE has been shown to have
strong cytotoxic activity against IL-4R-positive carcinomas in pre-
clinical studies.20,21In addition, several Phase I clinical studies
have been conducted with IL-4-PE in patients with several types
of solid tumors.23,24These clinical trials have established maxi-
mum tolerated dose of IL-4-PE when given by intravenous and
Grant sponsors: Ministry of Education, Culture, Sports, Science and
Technology, Japan; Grant number: 20390339.
*Correspondence to: Department of Pharmacoepidemiology, Graduate
School of Medicine and Public Health, Kyoto University, Sakyoku, Kyoto
606-8501, Japan. Fax: 181-75-753-4469.
Received 14 March 2008; Accepted after revision 8 July 2008
Publishedonline 17September 2008inWileyInterScience(www.interscience.
Abbreviations: Ab, antibody; BTC, biliary tract carcinoma; IL, interleu-
kin; IL-4R, interleukin-4 receptor; mRNA, messenger ribonucleic acid;
PE, Pseudomonas exotoxin A; RT-PCR, reverse transcription-polymerase
Int. J. Cancer: 123, 2915–2922 (2008)
' 2008 Wiley-Liss, Inc.
Publication of the International Union Against Cancer
In this study, we have examined whether IL-4Rs are overex-
pressed in tumor specimens from patients with BTC and BTC cell
lines in vitro. In addition, we have examined the receptor expres-
sion in normal epithelia of gallbladders and intrahepatic bile ducts.
IL-4R overexpressing BTC were targeted in vivo in animals bear-
ing subcutaneously (s.c.) growing human xenografted tumors or
peritoneally disseminated BTC tumors. Administration of toler-
ated dosage of IL-4-PE induced significant regression of estab-
lished BTC tumors and significantly improved the survival of
animals with disseminated tumors.
Material and methods
Specimens from 42 patients (9 males and 33 females) with gall-
bladder carcinoma (3 with pT1, 14 with pT2, 2 with pT3and 23
with pT4carcinoma), 18 patients (16 males and 2 females) with
extrahepatic cholangiocarcinoma (3 with pT1, 1 with pT2and 14
with pT4carcinoma) and 30 patients (18 males and 12 females)
with intrahepatic cholangiocarcinoma (2 with pT1, 6 with pT2, 21
with pT3and 1 with pT4carcinoma) were analyzed in this study.
All patients were diagnosed with BTC and underwent surgery at
the University of Tsukuba Hospital. BTC was diagnosed on the
basis of histological findings and disease was classified according
to the tumor node metastasis (TNM) classification of the Ameri-
can Joint Committee on Cancer (AJCC).25In addition, gallbladder
specimens obtained at hepatectomy from 17 subjects with meta-
static liver disease were used as normal gallbladder specimens.
Liver tissue specimens, including intrahepatic bile ducts, from 12
patients with metastatic liver tumors were used as intact intrahe-
patic bile duct specimens. The study protocol was approved in
accordance to the ethical guidelines of the 1975 Declaration of
Helsinki. All patients provided written informed consents for
Cell lines and culture conditions
The gallbladder carcinoma cell lines Mz-ChA-1 and Mz-ChA-
2,26and the extrahepatic cholangiocarcinoma cell line Sk-ChA-
126were obtained from Dr. A. Knuth (Johannes-Gutenberg
University, Mainz, Germany). The gallbladder carcinoma cell
lines TGBC-1-TKB, TGBC-2-TKB and TGBC-44-TKB27were
obtained from Dr. T. Todoroki (University of Tsukuba, Ibaraki,
Japan). KMBC, an extrahepatic cholangiocarcinoma cell line,28
and KMC-1, an intrahepatic cholangiocarcinoma cell line,29were
obtained from Dr. M. Kojiro (Kurume University School of Medi-
cine, Kurume, Japan). Panc-1, a pancreatic carcinoma cell line,
was purchased from the American Type Culture Collection. Cells
were maintained in Dulbecco’s modified Eagle’s medium
(DMEM) containing 10% heat-inactivated fetal calf serum (Invi-
trogen, Carlsbad, CA) in a humidified atmosphere with 5% carbon
dioxide at 37?C.
Cells from each of the 8 BTC cell lines and 9 frozen surgical
specimens (3 gallbladder carcinoma, 2 extrahepatic cholangiocar-
cinoma, 2 intrahepatic cholangiocarcinoma and 2 normal gallblad-
der) were washed in ice-cold PBS and lysed in NP-40 lysis buffer
(50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EGTA, 1% NP-
40). Lysates for each sample were prepared from 1 3 106cells in
100 ll of lysate buffer. Aliquots of lysate (30 ll) were electropho-
resed through a 4–12% gradient SDS-PAGE gel. Proteins were
transferred onto a PVDF membrane. After the membrane was
blocked using 5% skim milk, it was incubated at 4?C overnight
with a primary monoclonal antibody (mAb) against IL-4R at a
dilution of 1:200 (Santa Cruz Biotechnology, Santa Cruz, CA) or
with a mAb against a-tubulin at a dilution of 1:200 (Sigma-
Aldrich, St. Louis, MO). Proteins were visualized on HyperfilmTM
using an ECL/western blotting system (GE Healthcare, Piscat-
away, NJ) according to the manufacturer’s instructions. Panc-1
cells were used as positive controls, and a-tubulin was used as the
For immunostaining of IL-4R, 2-lm-thick tissue sections were
stained using the indirect immunoperoxidase method with anti-IL-
4R mAb (R & D Systems, Minneapolis, MN) as described in our
previous study.20For a negative control, primary Ab that had been
preincubated with IL-4R peptide was used. IL-4R expression on
membranes was assessed independently by two investigators with
respect to the histopathological characteristics and specific immu-
noreactivity of cells. IL-4R staining seemed to be positive when
more than 10% of the total number of cancerous epithelial cells in
each section expressed IL-4R on their membranes, as described in
our previous study.30
Reverse transcriptase-polymerase chain reaction
First-strand cDNAs were synthesized from total RNA isolated
from pellets of each BTC cell line, using M-MLV RT and the
oligo (dT)20primer method (Invitrogen), to a 20-ll total reaction
volume. A 1-ll aliquot of the product of the reverse transcription
reaction was amplified in a 50-ll final volume of PCR mixture
containing 5 ll of 103 buffer, 4 ll of 25 mM MgCl2, 4 ll of
10 mM dNTP, 1.25 units of AmpliTaq Gold DNA polymerase
(Applied Biosystems, Foster City, CA) and 50 pmol of each of the
human IL-4Ra primers (forward, 50-GACCTGGAGCAACC
CGTATC-30; reverse, 30-CATAGCACAACAGGCAGACG-50) or
each of the human b-actin primers (forward, 50-GGCACCA-
CACCTTCTACAATGAGC-30; reverse, 30-CATTGCCAA TGG
TGATGACCTGGC-50). Beta-actin was used as an internal
control. Amplification was conducted using 35 cycles of 94?C for
30 sec, 58?C for 30 sec and 72?C for 60 sec.
Recombinant IL-4 toxin IL-431,32-PE38KDEL was supplied by
Neurocrine Biosciences, (San Diego, CA), which is identical to
fusion toxin used in our previous study.20
Radioreceptor binding assay
Cells (1 3 106) in 100 ll of binding buffer (DMEM plus 0.2%
human serum albumin and 10 mM HEPES) were incubated with
200 pM125I-IL-4 with or without 100 nM cold IL-4 for 2 hr at
4?C. Cells were washed 3 times with ice-cold bicarbonate buffer
(Sigma-Aldrich). Then the radioactivity of cells was measured
using a gamma counter (Cobra Quantum; PerkinElmer, Wellesley,
MA). CHO-K1 cells served as negative controls. The average
values from quadruplicate wells were used for analysis.
Protein synthesis inhibition assay
The in vitro cytotoxicity of IL-4 or IL-4-PE to human BTC cell
lines was measured based on the inhibition of3H-leucine incorpo-
ration as described previously.33To investigate the specific bind-
ing and activity of IL-4-PE, IL-4 (200 nM) was coincubated with
IL-4-PE and inhibition of IL-4-PE activity was measured in
TGBC-44-TKB cells. All assays were performed in quadruplicate,
and the concentration of IL-4-PE at which 50% inhibition of pro-
tein synthesis occurred (IC50) was calculated.
Subcutaneous xenografted BTC tumor model
All animal experiments were conducted according to the Uni-
versity of Tokyo’s institutional guidelines for the care and use of
laboratory animals in research. Four-week-old female BALB/c nu/
nu athymic mice (CLEA Japan, Tokyo, Japan) were used. An s.c.
xenograft model was prepared by s.c. injection of 1 3 106TGBC-
44-TKB cells in 150 ll PBS into the left flank of each mouse.
Within 3–4 days, established tumors were palpable. Mice were
randomly assigned to the intratumoral (i.t.) or intraperitoneal (i.p.)
administration groups. Mice in the i.t. administration groups
received i.t. injections of vehicle (0.2% human serum albumin in
ISHIGE ET AL.
PBS) or IL-4-PE (200 lg/kg/injection) three times, on Days 5, 7
and 9 after injection of the TGBC-44-TKB cells. Mice in the i.p.
administration groups received i.p. injections of vehicle or IL-4-
PE (100 lg/kg/injection) twice a day (b.i.d.) from Day 5 to Day 9.
Tumor size was measured using Vernier calipers and tumor vol-
ume was calculated as 0.4 3 longest diameter 3 width.2Each of
the 4 groups (i.e., vehicle and treatment groups for each of the i.p.
and i.t. administrations) consisted of 5 mice and all mice com-
pleted the experiments.
Intraperitoneally disseminated BTC tumor model
An intraperitoneal dissemination model was prepared by i.p.
injecting 4-week-old female BALB/c nu/nu athymic mice with 1
3 107TGBC-44-TKB cells in 200 ll of PBS. Tumors were then
allowed to grow for 4 days. Before the experiments, formation of
disseminated tumor nodules in the peritoneal cavity was confirmed
by sacrificing another mouse that had been prepared in the same
way. Ten mice received b.i.d. i.p. administrations of IL-4-PE (100
lg/kg/ injection) from Day 5 to Day 9 after injection of the
TGBC-44-TKB cells. Another 10 mice were administered vehicle
in the same way.
Values are given as means 6 SD (standard deviation). A two-
sided v2test was used for comparison of pathology data between
groups. The difference in tumor volumes in s.c. xenografted model
mice was analyzed using the unpaired Student’s t-test. The sur-
vival of peritoneal dissemination model mice was analyzed using
the Kaplan-Meier method. Differences in the survival of animals
in subgroups were analyzed using the log-rank test. A p value of
<0.05 was defined as statistically significant.
IL-4 receptor expression in BTC tissue specimens
and cultured BTC cells
Immunoblot analysis demonstrated that IL-4R protein is present
in the lysates of all BTC tissue specimens, e.g., gallbladder carci-
noma, extrahepatic and intrahepatic cholangiocarcinoma, but was
not found in the normal gallbladder tissue specimens (Fig. 1a).
Levels of IL-4R protein were not significantly different in the
early and advanced stages of BTC. The results of immunohisto-
chemical assays for IL-4R in BTC, normal gallbladder and intra-
hepatic bile duct tissue specimens are shown in Table I. Reflecting
the protein levels determined using immunoblot analysis (Fig. 1a),
IL-4R immunostaining was found in the BTC cancerous epithelia,
but not in the normal gallbladder or intrahepatic bile duct epithelia
(Fig. 1b). IL-4R expression density was significantly higher in tis-
sue specimens derived from gallbladder carcinoma (60%) and in-
trahepatic cholangiocarcinoma (63%). In sharp contrast, normal
gallbladder and intrahepatic bile duct tissue specimens did not
show any staining for IL-4R. The IL-4R staining density did not
FIGURE 1 – Expression levels of IL-4R in BTC tissues (A & B) and
cultured BTC cells (C & D). mAb raised against the extracellular do-
main of human IL-4R was used for immunoblot analysis and immuno-
histochemistry. (a) IL-4R protein levels in BTC tissues. Tissue lysates
were prepared from 2 normal gallbladder specimens from subjects
with metastatic liver disease who had undergone hepatectomy (NG); 3
gallbladder carcinoma specimens from one patient each with pT1
(well-differentiated), pT2(moderately differentiated) and pT4carci-
noma (moderately differentiated); 2 extrahepatic cholangiocarcinoma
specimens from one patient each with pT2(well-differentiated) and
pT3carcinoma (moderately differentiated); and 2 intrahepatic cholan-
giocarcinoma specimens from one patient each with pT2(well-differ-
entiated) and pT3 carcinoma (moderately differentiated). Tubulin
acted as an internal control, and Panc-1 cells served as positive con-
trols (P.C.). (b) Immunohistochemical localization of IL-4R in BTC
tissues (gallbladder carcinoma and intrahepatic cholangiocarcinoma),
normal gallbladder, and normal intrahepatic bile duct. Bars, 50 lm.
(c) IL-4R protein levels in cultured BTC cells. Cell lysates were pre-
pared from 8 cultured biliary tract carcinoma cell lines: Mz-ChA-1
(well-differentiated adenocarcinoma), Mz-ChA-2 (poorly differenti-
ated adenocarcinoma), TGBC-1-TKB (poorly differentiated adeno-
carcinoma), TGBC-2-TKB (poorly differentiated adenocarcinoma),
TGBC-44-TKB (poorly differentiated adenocarcinoma), KMBC
(well-differentiated adenocarcinoma), Sk-ChA-1 (poorly differenti-
ated adenocarcinoma) and KMC-1 (moderately differentiated adeno-
carcinoma). Tubulin acted as an internal control, and Panc-1 cells
acted as positive controls (P.C.). (d) Steady-state IL-4R mRNA levels
in cultured BTC cells. RT-PCR-assisted amplification of mRNAs was
performed for IL-4Ra and b-actin. b-Actin acted as an internal
IL-4-CONJUGATED PSEUDOMONAS EXOTOXIN
show any significant correlation with the pathological features,
that is, the differentiation status and the pathological stage, of
BTCs. Nevertheless, it should be noted that IL-4R was expressed
not only in well or moderately differentiated carcinomas, but also
in poorly differentiated or adenosquamous carcinomas.
IL-4R protein was also present in the lysates of all cultured
BTC cells (Fig. 1c). The gallbladder carcinoma cell lines Mz-
ChA-1 and Mz-ChA-2, and the extrahepatic cholangiocarcinoma
cell line KMBC yielded thick bands of IL-4R protein. Expression
levels were unrelated to the differentiation status of BTC cells.
We also examined steady-state mRNA expression levels for IL-
4Ra by RT-PCR analysis (Fig. 1d). Panc-1, a pancreatic carci-
noma cell line that reportedly expresses high levels of IL-4R and
is highly sensitive to IL-4-PE,20served as a positive control for
the IL-4Ra. TGBC-1-TKB, a gallbladder carcinoma cell line,
expressed low levels of IL-4Ra mRNA. Although 7 of the 8 BTC
cell lines expressed IL-4Ra mRNA, the levels were not signifi-
cantly different among the BTC cell lines and mRNA levels did
not correlate with protein levels as determined by immunoblot
Binding of radiolabeled IL-4 to cultured BTC cells
A receptor binding assay in BTC cells using125I-IL-4 was per-
formed to verify expression of IL-4R on the surface membranes of
cells, because binding of IL-4 to its receptor is the first and most
important step during the process of IL-4-PE-mediated cytotoxic-
ity. When cells were incubated with125I-IL-4, all BTC cell lines
bound IL-4 with high affinity (black bars in Fig. 2). This binding
was specific as coincubation with an excess amount of unlabeled
IL-4 markedly inhibited the binding of radiolabeled IL-4 to the re-
ceptor (white bars in Fig. 2). Mz-ChA-1 and Mz-ChA-2 cell lines
bound125I-IL-4 at highest level, which is in agreement with high-
est concentration of IL-4R protein determined by immunoblot
analysis. On the other hand, TGBC-1-TKB cell line showed the
lowest radioactivity among all BTC cell lines, even though these
cells expressed a moderate IL-4R protein by immunoblot analysis.
There was no significant difference in radiolabeled IL-4 binding to
other 5 BTC cell lines. These observations indicate that IL-4R tar-
geted agents such as IL-4-PE might be considerably cytotoxic to
BTC cells, especially those showing high levels of IL-4R.
In vitro cytotoxicity of IL-4-PE to cultured BTC cells
To assess the in vitro cytotoxicity of IL-4-PE to cultured BTC
cells, a protein synthesis inhibition assay was performed using
BTC cell lines treated with IL-4-PE (Fig. 3). Inhibition of protein
synthesis was determined by monitoring the incorporation of [3H]-
leucine into cells. IL-4 was found to have no marked effect on the
viability of BTC cells at concentrations of 0.1–100 ng/ml (data not
shown). As shown in Figure 4, 4 BTC cell lines, including TGBC-
44-TKB, Mz-ChA-1, Mz-ChA-2 and KMBC, were highly sensi-
tive to IL-4-PE, with IC50values (concentration of IL-4-PE caus-
ing 50% inhibition of protein synthesis) of less than 5 ng/ml. The
Mz-ChA-1 and Mz-ChA-2 cell lines, which had the highest IL-4
binding activity in the radioreceptor binding assay, were most sen-
sitive to IL-4PE (IC50< 1 ng/ml). The TGBC-1-TKB, TGBC-2-
TKB and Sk-ChA-1 cell lines were least sensitive to IL-4-PE. The
inhibitory effect of IL-4-PE on protein synthesis in these BTC cell
lines was poor to moderate (IC50values of 111–260 ng/ml). This
observation is in agreement with the results of the receptor binding
assay, showing that these cells do not express high levels of IL-
4R. The in vitro cytotoxicity of IL-4-PE to TGBC-44-TKB cells
was neutralized by coincubation with an excess of IL-4, suggest-
ing that the observed in vitro cytotoxic activity of IL-4-PE was
mediated by its binding to IL-4R.
In vivo antitumor activity of IL-4-PE against s.c.
xenografted BTC tumors
Following the observation that IL-4-PE is strongly cytotoxic to
BTC cells in vitro, the in vivo antitumor activity of IL-4-PE was
studied using s.c. tumor-bearing mice. Immunodeficient animals
s.c. implanted with human TGBC-44-TKB tumors were treated by
IL-4-PE by i.t. or i.p. routes. In mice receiving 3 i.t. injections of
IL-4-PE, on Days 5, 7 and 9 (200 lg/kg per day), all s.c. tumors
regressed dramatically in the 2 or 3 days following the treatment
period. In 1 mouse the tumor was completely eradicated, and the
tumor in another mouse regressed to a tiny nodule that did not
change in size again before the end of the observation period. The
FIGURE 2 – Binding of radiolabeled IL-4 to cultured BTC cells.
Recombinant human IL-4 (PeproTech EC, London, UK) was radiola-
beled with iodine-125 (GE Healthcare) using IODO-Gen pre-coated
iodination tubes (Pierce, Rockford, IL) and purified with D-Salt
dextran desalting columns (Pierce) according to the manufacturers’
instructions. The specific activity of labeled IL-4 was estimated to be
124 lCi/lg. Carcinoma cells (1 3 106) were incubated with125I-IL-4
(200 pM) at 4?C for 2 hr with or without unlabeled IL-4 (white col-
umns, with 100 nM unlabeled IL-4; black columns, without unlabeled
IL-4). The results are presented as means 6 SD of quadruplicate
determinations. The assay was repeated twice. CHO cells, which do
not express IL-4R, acted as negative controls.
TABLE I – EXPRESSION OF IL-4 RECEPTOR AND CLINICOPATHOLOGICAL
FEATURES OF BILIARY TRACT CARCINOMAS
(n 5 17)
(n 5 42)
Well2(n 5 15)
Mod (n 5 21)
Poor (n 5 5)
Adeno-sq (n 5 1)
I (n 5 3)
II (n 5 14)
III (n 5 2)
IV (n 5 23)
Extrahepatic bile ducts
(n 5 18) 9 (50)
Well (n 5 7)
Mod (n 5 9)
Poor (n 5 1)
Adeno-sq (n 5 1)
I (n 5 3)
II (n 5 1)
III (n 5 14)
Intrahepatic bile ducts4
(n 5 12)
(n 5 30)
Well (n 5 13)
Mod (n 5 10)
Poor (n 5 3)
Adeno-sq (n 5 4)
I (n 5 2)
II (n 5 6)
III (n 5 21)
IV (n 5 1)
1Values in parentheses represent percentages.–2Well, well-differen-
tiated; mod, moderately differentiated; poor, poorly differentiated;
adeno-sq, adenosquamous.–3pStage, pathological stage.–4p < 0.05,
significantly different between the two groups.
ISHIGE ET AL.
FIGURE 3 – In vitro cytotoxicity of IL-4-PE to cultured BTC cells. Eight BTC cell lines (1 3 104) were cultured in DMEM leucine-free me-
dium with various concentrations of IL-4-PE (0–1000 ng/ml) for 20 hr at 37?C. Then, 1lCi ofL-[4,5-3H] leucine (GE Healthcare) was added to
each well. The cells were cultured for an additional 4 hr and then harvested. After the cells were frozen for 1 night, the radioactivity incorporated
into the cells was measured using a Beta-counter. Competition analysis using an excess of IL-4 was performed with TGBC-44-TKB cells (dotted
line). The growth inhibition data are given as percentages of leucine incorporation in cells cultured without IL-4-PE. The IC50values of each
cell line were as follows: TGBC-44-TKB, 4.0 ng/ml; TGBC-1-TKB, 112 ng/ml; TGBC-2-TKB, 262 ng/ml; Mz-ChA-1, 0.73 ng/ml; Mz-ChA-2,
0.26 ng/ml; Sk-ChA-1, 168 ng/ml; KMBC, 2.3 ng/ml; KMC-1, 22.6 ng/ml. The results are presented as means 6 SD of quadruplicate determina-
tions. The assay was repeated twice.
FIGURE 4 – In vivo antitumor activity of IL-4-PE against s.c. xenografted tumors of BTC cells in mice. Subcutaneous tumors were seeded in
immunodeficient mice using TGBC-44-TKB cells, as described in the Materials and Methods section. Each group consisted of 5 animals. The
arrows show the days on which injections of IL-4-PE took place. Tumor volumes were measured 2 or 3 times a week, and tumor volume (mm3)
was calculated as 0.4 3 longest diameter 3 width2. Tumor volumes are presented as means 6 SD for 5 mice for each group. Significant differ-
ences between the treatment and vehicle groups are indicated by: *p < 0.05; **p < 0.01. (a) Mice received i.t. injections of IL-4-PE (200 lg/kg,
s) or vehicle (PBS plus 0.2% human serum albumin,?) 3 times, on Days 5, 7 and 9. A significant reduction in tumor size was observed 5 days
after the treatment, and 2 of the 5 tumors were completely eradicated. (b) Mice received i.p. injections of IL-4-PE (100 lg/kg, s) or vehicle
(PBS plus 0.2% human serum albumin,?) twice a day for 5 days. A significant reduction in tumor volume was observed in mice in the IL-4-PE
764 mm3in the IL-4-PE and vehicle groups, respectively, and the difference was statistically significant.
group at Day 9, that is, 4 days after the first treatment. At the end of the experiment (Day 47), the average tumor volumes were 243 mm3and
tumors in the other 3 mice began to grow again about 3 weeks
after treatment, but the average tumor volume on day 47 was only
26 mm3. On the other hand, in control mice, which received i.t.
injections of vehicle, the tumors continued to grow throughout the
study period, and the average final tumor volume was 1,300 mm3
(p < 0.0001). The mice in this group had to be euthanized on Day
47 due to tumor burden (Fig. 4a). In mice receiving i.p. injections
of IL-4-PE (100 lg/kg, b.i.d. from Day 5 to Day 9), the implanted
tumors regressed in all 5 mice in the 3-4 days following the treat-
ment period. In this group, no tumor was completely eradicated as
of the end of the observation period; however, the average tumor
volume in this treatment group was just one-third of the tumor vol-
ume in the i.p. vehicle group (p < 0.001) (Fig. 4b). These results
indicate that IL-4-PE has significant antitumor activity in this s.c.
BTC animal model.
In vivo antitumor activity of IL-4-PE against
peritoneally disseminated BTC tumors
To mimic the clinical aggressiveness of BTCs, a peritoneally
disseminated tumor model was developed using immunodeficient
mice that were inoculated with TGBC-44-TKB cells (107/mouse)
into the peritoneal cavity. Each mouse received a total of 10 i.p.
injections of IL-4-PE or vehicle over 5 days (b.i.d). Mice receiving
vehicle showed signs of cachexia and extensive abdominal disten-
sion with formation of bloody ascites within 4 weeks of inocula-
tion (Fig. 5a), and the median survival period of these mice was 5
weeks. One mouse was sacrificed at 4 weeks after the injection of
BTC cells to confirm that the carcinoma cells continued to spread
diffusely and that disseminated nodules existed over the mesente-
rium (Fig. 5a). The cause of death in all vehicle-treated mice was
considered to be peritoneal dissemination. In contrast, 2 of the 10
mice that had received IL-4-PE treatment showed symptoms of
dissemination at 7 weeks and died with bloody ascites at 9 weeks
after inoculation, and another 2 mice showed symptoms at 12
weeks. However, the other 6 mice did not develop any symptoms
(i.e., there was no abdominal distention) and survived for more
than 20 weeks. The survival time was significantly longer in the
treatment group compared with the vehicle group (p < 0.005)
(Fig. 5b). These results indicate that IL-4-PE has significant anti-
tumor activity in mice with peritoneally disseminated BTC
We demonstrate that human BTC tissue specimens express high
levels of IL-4 receptors compared to normal gallbladders. These
observations were confirmed in cultured human BTC cell lines
that showed expression of IL-4R by immunoblot analysis and
receptor binding assay. Immunohistochemical analysis showed
overexpression of IL-4R in a large proportion of human BTC
tissue specimens (in 60% of patients with gallbladder carcinoma,
50% of patients with extrahepatic cholangiocarcinoma and 63%
of patients with intrahepatic cholangiocarcinoma). The IL-4R
expression levels appear to be unrelated to the differentiation sta-
tus or pathological stage of BTC tumors studied. Moreover, there
was no significant relationship between expression levels of IL-4R
in patients with BTC with various pathological features including
histological grade, lymphatic invasion, venous invasion and lymph
node metastasis (data not shown). In contrast to BTC cells, normal
gallbladder and bile duct epithelium tissue specimens did not
express IL-4R. These results identified a potential novel target for
targeted therapy of BTC.
The overexpression of IL-4R in BTC cell lines sensitized them
to cytotoxic effect of IL-4-PE, which correlated with the level of
IL-4R expression. The in vitro cytotoxicity of IL-4-PE was
assessed for 8 cultured BTC cell lines. Mz-ChA-1 and Mz-ChA-2
cell lines showing highest levels of IL-4R expression showed
highest sensitivity (IC50< 1 ng/ml). IL-4, on the other hand, had
no significant growth modulatory or cytotoxic activity on BTC
cell lines expressing the IL-4 receptor (data not shown). These
observations indicate that abundant IL-4R expression in BTC
tumors will facilitate efficient targeting by IL-4-PE.
IL-4-PE mediated significant antitumor activity against BTC tu-
mor in not only s.c. tumor model but also in peritoneal metastasis
model. BTC is localized disease with early metastasis to perito-
neal organs and lymph nodes. In the clinical course, peritoneal dis-
semination is a very unfavorable development, which frequently
occurs in patients with advanced BTC. As systemic chemotherapy
using anticancer drugs has produced few significant benefits for
patients with advanced BTC (e.g., peritoneally disseminated
BTC),34,35administration of anticancer agents into the peritoneal
cavity may be a therapeutic option for this disease. Administration
of IL-4-PE into the peritoneal cavity results in a high local con-
centration of the drug around the disseminated tumors in animal
models. As i.p. administration of IL-4-PE in the current study21or
previous study20in the pancreatic cancer model produced remark-
able antitumor effects at tolerated doses, it is hypothesized that
i.p. administration of IL-4-PE will provide a potential therapeutic
benefit in patients with BTC.
This study does not experimentally cover the immune modula-
tory activity of IL-4-PE. However, we show that IL-4-PE mediates
a potent antitumor activity against BTC by inducing a specific in-
hibition of protein synthesis in cultured human BTC cell lines. As
human IL-4-PE used in this study is composed of human IL-4 and
a truncated form of Pseudomonas exotoxin, IL-4-PE does not bind
FIGURE 5 – In vivo antitumor activity of IL-4-PE against perito-
neally disseminated tumors of BTC cells in mice. Peritoneally disse-
minated tumors were seeded in immunodeficient mice using TGBC-
44-TKB cells, as described in the Materials and Methods section.
Each group consisted of 10 animals. (a) Appearance of a nude mouse
with intraperitoneally disseminated BTC cells (left). Representative
macroscopic view of the peritoneal cavity in a mouse 4 weeks after tu-
mor inoculation (right). Multiple tumorous nodules comprising
TGBC-44-TKB cells were observed over the mesenterium. (b) Rela-
tionship between survival outcome and IL-4-PE treatment in mice
with peritoneally disseminated tumors of BTC cells as assessed using
Kaplan-Meier survival curves. In mice that had received i.p. injections
of 1 3 107TGBC-44-TKB cells, vehicle (PBS plus 0.2% human se-
rum albumin) or IL-4-PE (100 lg/kg/injection) was i.p. injected twice
a day from Day 5 to Day 9. The median survival period of the vehicle
group (dotted line) was 5 weeks, and that in the IL-4-PE treatment
group (solid line) was 14 weeks. The difference was statistically sig-
nificant (p < 0.005).
ISHIGE ET AL.
to murine IL-4R due to species specificity, but it binds to monkey
IL-4R (Ref. 36 and unpublished results). Because of this limita-
tion, we could not assess the immunomodulatory properties of IL-
4-PE on murine T cells. In monkey studies, as human IL-4 can
bind to monkey cells, systemic administration of IL-4-PE did not
deplete T cells as determined by complete blood and lymphocyte
count. In addition, it has been previously demonstrated that IL-4-
PE has no cytotoxic effect on resting T cells.37However, when
cells are activated, they begin to become susceptible to the cyto-
toxic effect of IL-4-PE, but there is no preference of Th1 or Th2
cells. As in the intact host most T cells are in the resting stage, one
will not see selective killing of Th1 or Th2 cells. Furthermore, in a
Phase I clinical trial in which IL-4-PE was administered to
patients with advanced solid tumors, no changes in leukocytes
were reported.23Based on above observations, we do not expect
that IL-4-PE will cause any shift in Th1/Th2 response in intact
The safety of IL-4-PE has been assessed in various preclini-
cal pharmacological and toxicity studies.38As human IL-4-PE
does not bind to murine IL-4R due to species specificity but it
binds to monkey IL-4R; therefore, IL-4-PE was tested in both
species to determine nonspecific and specific toxicity. Mice tol-
erated IL-4-PE well and the LD50 was determined to be 475
lg/kg when given i.v. every day for 3 injections.31Dose limit-
ing toxicity was nonspecific hepatic toxicity. In addition, IL-4-
PE doses up to 200 lg/kg given twice a day i.p. for 5 days and
up to 500 lg/kg given i.t. every other day for 3 injections were
found to be well tolerated without any organ toxicity or mortal-
ity in mice with s.c. xenografted human pancreatic carcinoma.20
In monkeys, intravenous administration of IL-4-PE at doses of
50 and 200 lg/kg was also well tolerated with reversible he-
patic toxicity (Ref. 36 and unpublished results). No other sys-
temic toxicities were observed. In patients with solid tumors,23
systemic administration of IL-4-PE at doses up to 16 lg/m2
daily 3 5 days every 28 days was well tolerated. Similar to
mouse and monkey studies, dose limiting toxicities were eleva-
tion of hepatic enzymes. In contrast, when IL-4-PE was admin-
istered intracranially in patients with recurrent brain tumors, no
systemic toxicities were observed.32Since IL-4-PE serum levels
were not detected, it is hypothesized that intracranial IL-4-PE
did not reach to systemic circulation.
In conclusion, IL-4Rs are overexpressed in both tumor speci-
mens from patients with BTC and BTC cell lines in vitro. The IL-
4Rs that are overexpressed on BTC cells can be successfully tar-
geted with IL-4-PE in vitro and in vivo including peritoneally dis-
seminated tumors in a murine model of BTC. IL-4-PE induces
regression of these established tumors in animal models and pro-
longs the survival of tumor-bearing mice. Therefore, additional
studies should be performed to support a Phase I clinical trial with
IL-4-PE by intraperitoenal route of administration.
1. Henson DE, Albores-Saavedra J, Corle D. Carcinoma of the gallblad-
der. Histologic types, stage of disease, grade, and survival rates.
Statistical white paper on cancer. Japan VSo. Tokyo (Japan): Japanese
Ministry of Health and Welfare: Statistics Assoc; 1998.
Wistuba II, Gazdar AF. Gallbladder cancer: lessons from a rare
tumour. Nat Rev Cancer 2004;4:695–706.
Curley S. Biliary tract cancer. In: Pollock RE, ed. Surgical oncology,
vol. 90. Norwell, MA: Kluwer Academic Publishers, 1997.273–307.
Shaib Y, El-Serag HB. The epidemiology of cholangiocarcinoma.
Semin Liver Dis 2004;24:115–25.
Patel T. Increasing incidence and mortality of primary intrahepatic chol-
angiocarcinoma in the United States. Hepatology 2001;33:1353–7.
Taylor-Robinson SD, Foster GR, Arora S, Hargreaves S, Thomas HC.
Increase in primary liver cancer in the UK, 1979–94. Lancet
Dixon E, Vollmer CM, Jr, Sahajpal A, Cattral M, Grant D, Doig C,
Hemming A, Taylor B, Langer B, Greig P, Gallinger S. An aggressive
surgical approach leads to improved survival in patients with gallblad-
der cancer: a 12-year study at a North American Center. Ann Surg
Jarnagin WR, Fong Y, DeMatteo RP, Gonen M, Burke EC, Bodnie-
wicz BJ, Youssef BM, Klimstra D, Blumgart LH. Staging, resectabil-
ity, and outcome in 225 patients with hilar cholangiocarcinoma. Ann
Surg 2001;234:507–17; Discussion 17–19.
10. Huang JL, Biehl TR, Lee FT, Zimmer PW, Ryan JA, Jr. Outcomes af-
ter resection of cholangiocellular carcinoma. Am J Surg 2004;187:
11. Nakagohri T, Asano T, Kinoshita H, Kenmochi T, Urashima T, Miura
F, Ochiai T. Aggressive surgical resection for hilar-invasive and
peripheral intrahepatic cholangiocarcinoma. World J Surg 2003;27:
12. Paul WE. Interleukin-4: a prototypic immunoregulatory lymphokine.
13. Prokopchuk O, Liu Y, Henne-Bruns D, Kornmann M. Interleukin-4
enhances proliferation of human pancreatic cancer cells: evidence for
autocrine and paracrine actions. Br J Cancer 2005;92:921–8.
14. Todaro M, Zerilli M, Ricci-Vitiani L, Bini M, Perez Alea M, Maria
Florena A, Miceli L, Condorelli G, Bonventre S, Di Gesu G, De Maria
R, Stassi G. Autocrine production of interleukin-4 and interleukin-10
is required for survival and growth of thyroid cancer cells. Cancer Res
15. Morisaki T, Yuzuki DH, Lin RT, Foshag LJ, Morton DL, Hoon DS.
Interleukin 4 receptor expression and growth inhibition of gastric car-
cinoma cells by interleukin 4. Cancer Res 1992;52:6059–65.
16. Gooch JL, Lee AV, Yee D. Interleukin 4 inhibits growth and induces
apoptosis in human breast cancer cells. Cancer Res 1998;58:4199–
17. Nelms K, Keegan AD, Zamorano J, Ryan JJ, Paul WE. The IL-4 re-
ceptor: signaling mechanisms and biologic functions. Annu Rev
18. Lundin J, Kimby E, Bergmann L, Karakas T, Mellstedt H, Osterborg
A. Interleukin 4 therapy for patients with chronic lymphocytic leukae-
mia: a phase I/II study. Br J Haematol 2001;112:155–60.
19. Whitehead RP, Lew D, Flanigan RC, Weiss GR, Roy V, Glode ML,
Dakhil SR, Crawford ED. Phase II trial of recombinant human inter-
leukin-4 in patients with advanced renal cell carcinoma: a southwest
oncology group study. J Immunother 2002;25:352–8.
20. Kawakami K, Kawakami M, Husain SR, Puri RK. Targeting interleu-
kin-4 receptors for effective pancreatic cancer therapy. Cancer Res
21. Shimamura T, Rotal RE, Kioi M, Nakajima A, Husain SR, Puri RK.
Interleukin-4 cytotoxin therapy synergizes with gemcitabine in a
mouse model of pancreatic ductal adenocarcinoma. Cancer Res
22. Perentesis JP, Miller SP, Bodley JW. Protein toxin inhibitors of pro-
tein synthesis. Biofactors 1992;3:173–84.
23. Garland L, Gitlitz B, Ebbinghaus S, Pan H, de Haan H, Puri RK, Von
Hoff D, Figlin R. Phase I trial of intravenous IL-4 Pseudomonas exo-
toxin protein (NBI-3001) in patients with advanced solid tumors that
express the IL-4 receptor. J Immunother 2005;28:376–81.
24. Rand RW, Kreitman RJ, Patronas N, Varricchio F, Pastan I, Puri RK.
Intratumoral administration of recombinant circularly permuted inter-
leukin-4-Pseudomonas exotoxin in patients with high-grade glioma.
Clin Cancer Res 2000;6:2157–65.
25. Fleming ID, Cooper JS, Henson DE, Hutter RVP, Kennedy BJ, Murphy
GP, O’sullivan B, Sobin LH, Yarbro JW, eds. AJCC cancer staging
manual, 5th ed. Philadelphia, Pennsylvania: Lippincott-Raven, 1997.
26. Knuth A, Gabbert H, Dippold W, Klein O, Sachsse W, Bitter-Suer-
mann D, Prellwitz W, Meyer zum Buschenfelde KH. Biliary adeno-
carcinoma. Characterisation of three new human tumor cell lines.
J Hepatol 1985;1:579–96.
27. Koike N, Todoroki T, Kawamoto T, Yoshida S, Kashiwagi H, Fukao
K, Ohno T, Watanabe T. The invasion potentials of human biliary
tract carcinoma cell lines: correlation between invasiveness and mor-
phologic characteristics. Int J Oncol 1998;13:1269–74.
28. Yano H, Maruiwa M, Iemura A, Mizoguchi A, Kojiro M. Establish-
ment and characterization of a new human extrahepatic bile duct
carcinoma cell line (KMBC). Cancer 1992;69:1664–73.
29. Iemura A, Maruiwa M, Yano H, Kojiro M. A new human cholangio-
cellular carcinoma cell line (KMC-1). J Hepatol 1992;15:288–98.
30. Kawamoto T, Shoda J, Irimura T, Miyahara N, Furukawa M, Ueda T,
Asano T, Kano M, Koike N, Fukao K, Tanaka N, Todoroki T. Expres-
sion of MUC1 mucins in the subserosal layer correlates with postsur-
gical prognosis of pathological tumor stage 2 carcinoma of the gall-
bladder. Clin Cancer Res 2001;7:1333–42.
IL-4-CONJUGATED PSEUDOMONAS EXOTOXIN
31. Kreitman RJ, Puri RK, Pastan I. Increased antitumor activity of a cir- Download full-text
cularly permuted interleukin 4-toxin in mice with interleukin 4 recep-
tor-bearing human carcinoma. Cancer Res 1995;55:3357–63.
32. Kawakami M, Kawakami K, Puri RK. Interleukin-4-Pseudomonas
exotoxin chimeric fusion protein for malignant glioma therapy. J Neu-
33. Husain SR, Obiri NI, Gill P, Zheng T, Pastan I, Debinski W, Puri RK.
Receptor for interleukin 13 on AIDS-associated Kaposi’s sarcoma
cells serves as a new target for a potent Pseudomonas exotoxin-based
chimeric toxin protein. Clin Cancer Res 1997;3:151–6.
34. Harder J, Riecken B, Kummer O, Lohrmann C, Otto F, Usadel H,
Geissler M, Opitz O, Henss H. Outpatient chemotherapy with gemci-
tabine and oxaliplatin in patients with biliary tract cancer. Br J Cancer
35. Kim ST, Park JO, Lee J, Lee KT, Lee JK, Choi SH, Heo JS, Park YS,
Kang WK, Park K. A Phase II study of gemcitabine and cisplatin in
advanced biliary tract cancer. Cancer 2006;106:1339–46.
36. Puri RK, Hoon DS, Leland P, Snoy P, Rand RW, Pastan I, Kreitman
RJ. Preclinical development of a recombinant toxin containing circu-
larly permuted interleukin 4 and truncated Pseudomonas exotoxin for
therapy of malignant astrocytoma. Cancer Res 1996;56:5631–7.
37. Puri RK, Mehrotra PT, Lelannd P, Kreitman RJ, Stiegel JP, Pastan I.
A chimeric protein comprised of IL-4 and Pseudomonas exotoxin is
cytotoxic for activated human lymphocytes. J Immunol 1994;152:
38. Kawakami K, Kawakami M, Puri RK. Overexpressed cell surface
interleukin-4 receptor molecules can be successfully targeted for anti-
tumor cytotoxin therapy. Crit Rev Immunol 2001;21:299–310.
ISHIGE ET AL.