Gene Therapy (2002) 9, 547–553
 2002 Nature Publishing Group All rights reserved 0969-7128/02 $25.00
www.nature.com/gt
RESEARCH ARTICLE
Human urinary bladder carcinomas express
adenovirus attachment and internalization receptors
A Loskog1, T Hedlund1, K Wester2, M de la Torre2, L Philipson3, P-U Malmström2
and TH Tötterman1
1
Clinical Immunology Division, Rudbeck Laboratory, University of Uppsala, Uppsala, Sweden; 2Experimental Urology Division,
Rudbeck Laboratory, University Hospital, Uppsala, Sweden; and 3Cell and Molecular Biology Division, Karolinska Institute,
Stockholm, Sweden
The use of adenoviral vectors as potent gene delivery systems requires expression of the Coxsackievirus/adenovirus
receptor (CVADR) on the target cell surface. This receptor
is important for virus attachment to the cell surface. For
effective internalization of the vector into the target cell the
integrins ␣v␤3 and/or ␣v␤5 are needed. Since there have
been reports of loss of CVADR in bladder cancer cell lines,
we wanted to investigate the expression of this receptor in
bladder carcinoma biopsies. Surgical biopsies, as well as
five human bladder cancer cell lines, were analyzed for
expression of CVADR, the integrins ␣v␤3 and ␣v␤5 and MHC
class I. Further, we studied the ability to transduce these cell
lines using adenoviral vectors. Immunohistochemistry
revealed that all biopsies (27/27) were positive for CVADR.
Some variation in expression was evident, and superficially
growing tumors stained more strongly than invasive ones.
Most human tumors expressed the integrin ␣v␤5 (14/24),
whereas integrin ␣v␤3 was less frequently seen (3/20). The
established cell lines were efficiently transduced with adenoviral vectors, and transduction could be reduced with antiCVADR antibodies. The abundance of appropriate viral
receptors on tumor biopsy cells is a further argument for
using adenoviral vectors in gene therapy of bladder cancer.
Gene Therapy (2002) 9, 547–553. DOI: 10.1038/sj/gt/3301689
Keywords: adenovirus; CVADR; CAR; integrins; bladder cancer
Introduction
Adenoviral vectors are commonly used as gene delivery
vehicles in different gene therapy approaches.1 Adenoviruses enter cells by receptor-mediated endocytosis
which is initiated by the interaction of the fiber protein
with a cell surface receptor called the Coxsackievirus
group B, adenovirus receptor (CVADR, formerly CAR).
Adenovirus serotypes of subgroups A, C, D, E and F all
use CVADR as a cellular fiber receptor for attachment to
the cell surface.2 Its cellular function, other than as a virus
receptor, remains unknown.
The CVADR mRNA expression is extremely variable
between different tissues such as liver, heart, lung and
prostate, whereas the ␣v integrins are more homogenously distributed among different organs.3,4 CVADR is
broadly expressed during embryonic development in the
central and peripheral nervous systems and in several
types of epithelial cells. In adults, the expression is more
restricted, but remains high mainly in epithelial cells.5
The adenovirus particle is not surrounded by a lipid
envelope and its penetration into the cytoplasm does not
depend on membrane fusion events. Instead, disruption
of the endosome is believed to permit entry. The penton
base protein contains the sequence RGD (arg-gly-asp)
Correspondence: A Loskog, Clinical Immunology Division, Rudbeck
Laboratory, University of Uppsala, S-751 85 Uppsala, Sweden
Received 16 August 2001; accepted 29 January 2002
that interacts with vitronectin receptors, integrins ␣v␤3
and ␣v␤5. Adenoviruses attach equally well to integrin
␣v positive or negative cells, but positive cells are more
susceptible to infection.6 Wickham et al7 showed as early
as 1993 that these integrins promote viral infection since
antibodies against these receptors or soluble pentone
base could participate in blocking of infection without
affecting attachment of the virus to the cell surface. In a
later study, they showed that both integrins ␣v␤3 and
␣v␤5 mediate adenovirus binding and internalization, but
integrin ␣v␤5 preferentially promotes virus-induced cell
membrane permeabilization and gene delivery as well.8
It has recently been shown that heparan-sulfate glycoaminoglycans (HS-GAGs) are involved in the binding
of adenovirus type 2 and 5.9 This binding is not essential
for adenovirus infection since HS-GAG-negative cells
could be transduced and reversibly blocking of HS-GAG
only partially inhibited transduction. In contrast,
addition of RmcB, a monoclonal antibody against
CVADR, completely blocked transduction.9,10 There are
several other structures that could influence adenoviral
transduction of the bladder. A thin layer called bladder
surface mucin or mucin-GAG coat covers the normal epithelium. This mucin is reported to represent an antibacterial defense mechanism and could therefore inhibit
adenoviral attachment.11 Ethanol pre-treatment of rat
bladders has been shown to enhance the transduction
rate possibly due to removal of the mucin layer.12
It has also been suggested that major histocompatibility complex (MHC) class I could act as a receptor for
Adenoviral receptors in bladder cancer
A Loskog et al
548
adenovirus infection.6 However, in a study by McDonalds et al,13 no correlation was seen between binding
of recombinant fiber knobs and MHC-I expression on
several tumor cell lines.
The expression of CVADR varies among human urinary bladder cancer cell lines.13,14 In this study we investigated not only cell lines, but actual tumor biopsies from
27 patients for the expression of both CVADR and the
integrins ␣v␤3 and ␣v␤5. We further studied the feasibility
to transduce human bladder cancer cells with an adenoviral vector containing the gene for the immunostimulatory protein CD154. This particular vector was successfully used in animal studies aiming to develop an
immunogene therapy for bladder cancer.15 The present
results indicate that adenoviruses are promising tools for
transferring genes into human bladder cancer cells.
Results
Human bladder cancer cell lines and biopsies are
broadly positive for CVADR
All 27 biopsies representing various tumor stages were
positive for CVADR as shown by immunohistochemistry
(Table 1 and Figure 1). Most superficial tumors (11/13)
were broadly CVADR positive (>80–100%), while onethird of the invasive tumors (5/14) only showed small
(⬍20%) weakly positive areas within the tumor. Tumor
cells close to blood vessels had a higher expression of
CVADR. The vessels themselves were, however, negative
Table 1
Biopsy
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
a
for CVADR expression. All five human bladder cancer
cell lines also stained positively for CVADR (Table 2). T24
was only weakly positive. The CVADR expression on J82
and T24 was not changed if the tumor cell lines were
cultured at high or low density (data not shown). As a
control for the staining method, 293 cells were positive
controls and CHO cells were used as negative controls
for CVADR expression. The cell lines were also analyzed
for CVADR mRNA expression by RT-PCR. This method
confirmed the immunohistochemistry data (Figure 2).
Integrin ␣v␤3 and ␣v␤5 expression on human bladder
cancer
The expression of integrins ␣v␤3 and ␣v␤5 was investigated by immunohistochemistry. Only 3/20 investigated
biopsies showed tumor cells positive for integrin ␣v␤3,
whereas more than half of the biopsies (14/24) expressed
integrin ␣v␤5 (Table 1 and Figure 1). Interestingly, the
vessels in the tumor areas were highly positive for both
integrins. Integrin ␣v␤3 has been associated with angiogenesis, which is a prerequisite for tumor growth.16 All
five investigated cell lines were positive for integrin ␣v␤5,
but only some of them had ␣v␤3 expression.
Adenovirus transduction of cell lines
Bladder cancer cell lines were readily transduced with
adenoviral vectors (Figure 3a and b), while the CAR
negative cell line CHO was resistant even with 15 times
p.f.u./cell of vector (Figure 3c). The T24 cells, which had
Expression of adenoviral attachment and internalization receptors in human bladder cancer biopsiesa
CVADR
␣v␤3
␣v␤5
Stage
Grade
+
+++
++
+++
+++
+++
++
+++
+
+++
+
+
++
+
++
+++
+++
+++
+
+++
++
++
++
+++
+
+++
+++
+
+
+
ND
ND
ND
ND
ND
ND
ND
+
+
+
++
+
++
+
+
++
++
+
ND
ND
ND
+
++
++
Ta
Ta
T1
Ta
Ta
Ta
Ta
Ta
T2
Ta
T1
T1
Ta
Ta
T1a
Ta
Ta
T1
T2
T2
Ib
T1
T1
T1
T2
Ta
T2
2B
2A
3-4
1
2B
2B
2B
2A
3-4
2A
3-4
3-4
2B
2B
2B
2A
2A
3-4
3-4
2B
b
3-4
2B
2B
3-4
2A
3-4
Twenty-seven human urinary bladder cancer biopsies of different stages (Ta, superficial; T1–T2, invasive) and grades were stained for
the adenovirus attachment receptor CVADR and the membrane permeabilization and internalization receptors integrin ␣v␤3 and ␣v␤5. All
tumors were positive for the CVADR (27/27), but the strength of expression varied. More than half of tumors stained positively for the
integrin ␣v␤5 (14/24), while only a few expressed integrin ␣v␤3 (3/20). +++, 100% of tumor cells strongly positive; ++, 80–100% of tumor
cells intermediate positive; +, small areas within the tumor positive; ND, not done.
b
Complete information on grade and stage missing, the tumor was considered as invasive.
Gene Therapy
Adenoviral receptors in bladder cancer
A Loskog et al
549
Figure 1 CVADR and integrin expression in bladder cancer. Human bladder biopsies broadly express CVADR (a–e), but the magnitude of expression
varies. Tumor cells close to blood vessels were often more strongly stained while the vessels themselves remained negative for CVADR. Panels a and
b exhibit an intermediate staining (++ in Table 1) at magnifications ×200 and ×400, respectively. Panel c, some biopsies had a more scattered intermediate
staining (++ in Table 1). 26% of biopsies were only weakly positive in some areas mainly surrounding vessels as shown in panel d (+ in Table 1). A
large proportion of biopsies (48%) were strongly positive for CVADR as shown in panel e (+++ in Table 1). The integrin ␣v␤3 was not found in most
tumors (f and g), while integrin ␣v␤5 stained positively in more than half of the patients (h and i). Vessels were strongly positive for both integrins.
Magnifications, a, ×200; b–I, ×400.
Table 2 Expression of adenoviral attachment and internalization
receptor proteins in bladder cancer cell linesa
Cell line
J82
T24
RT-112
VM-CUB1
KV-19-19
293
CHO
CVADR
␣v␤3
␣v␤5
+
+a
+
+
+
+
-
+
+
+b
+
-
+
+b
+b
+b
+
+
+
a
Immunocytochemistry revealed that all five human bladder cancer cell lines were positive for CVADR and integrin ␣v␤5. 293 cells
that are used for adenoviral vector production were also positive
for these proteins. CHO cells were negative for CVADR, but interestingly, positive for integrin ␣v␤5.
b
Weakly positive (⬍80–100%).
a low CVADR expression, needed twice the amount of
vector than the other cell lines. The transduction could
be reduced with anti-CVADR antibodies, indicating that
this receptor is crucial for an effective transduction
(Figure 4). Transduction of T24 cells was readily blocked
in a dose-dependent manner. Transduction of the other
four cell lines was only partially blocked with the same
amount (5 ␮g) of RmcB antibody. Increasing levels of
blocking antibodies, however, suppressed the transduction rate.
A peptide designed to block the interaction between
Figure 2 CVADR mRNA is present in human bladder cancer cell lines.
Bladder cancer cell lines were analyzed by RT-PCR using specific primers
for CVADR mRNA. The cell lines were strongly positive for CVADR,
except T24 that had a markedly lower expression. 293 cells commonly
employed for adenoviral production were used here as a positive control,
while CHO cells functioned as a negative control for CVADR expression.
The level of CVADR expression was compared with the expression of the
housekeeping gene ␤-actin (b).
the vector penton base and integrins (GRGDSP) was used
(Figure 4). We could partly inhibit the adenoviral transduction, but not to the same degree as using anti-CVADR
antibodies. Again, cells with lower integrin expression
were easier to block (T24, RT-112). Attempts to block both
CVADR and integrins simultaneously only had minimal
additional effects compared with blocking CVADR and
integrins separately. Blocking with irrelevant control peptide did not affect transduction. Since all cell lines are
Gene Therapy
Adenoviral receptors in bladder cancer
A Loskog et al
550
Figure 3 In vitro transduction of bladder cancer cell lines. Human bladder cancer cell lines are transducible with adenoviral vectors, as shown in (a)
(J82) and (b) (T24). CHO cells, which are CVADR-negative were not transducible (c). A vector containing CD154 (Ad-CD154) or an empty control
vector (Ad-dl30-7) was used. Three days after transduction, cells were stained with a PE conjugated anti-CD154 mAb and analyzed by flow cytometry.
Filled curves represent Ad-dl70-3 transduced cells negative for CD154 expression, and bold lines represent Ad-CD154 transduced cells.
highly MHC class I-positive, blocking experiments with
anti-MHC class I antibodies were performed. These
experiments revealed that the transduction of T24 cells
could be reduced by only 5–8%, whereas the other cell
lines remained fully transducible after using the same
amounts of antibodies (data not shown).
Discussion
Our present studies show that human bladder cancer
biopsies and cell lines express CVADR, a pivotal receptor
for adenovirus serotype 5 vector attachment to the cell
surface. In fact, all primary biopsies (n = 27) and cell lines
(n = 5) investigated in this study were positive for this
receptor, although some biopsies (7/27) only showed
small positive areas. The CVADR expression seems to
increase in human umbilical vein endothelial cells if the
cells are cultured at high density, while this did not seem
to affect CVADR expression on HeLa cells.17 The cell density in culture did not affect CVADR expression in our
experiments. The question remains what mechanisms
regulate CVADR expression, and whether the expression
could be elevated on target cells before adenovector
gene therapy.
Adenoviral transduction of tumor cell lines could be
reduced with antibodies against CVADR if low numbers
of vector particles were admixed with the tumor cells.
Transduction of the T24 cell line with a low-grade
expression of CVADR was easily blocked, while higher
amounts of blocking antibodies were required to block
other cell lines. CHO cells that lack CVADR expression
could not be transduced even with 15 times p.f.u./cell.
This indicates that CVADR is an essential receptor for
adenoviral transduction of bladder cancer cells. Since
integrins were earlier shown to be important for effective
internalization of the vector into the cells,7,8 we also
wanted to block this pathway for transduction. Antibodies that successfully block the biological activity of
integrins are lacking, and we therefore used a peptide
(GRGDSP) that blocks the interaction between the adenoviral penton base RGD sequence and integrins.8 These
studies showed minimal blocking effects, and only cell
lines with a low integrin expression (T24 and RT-112)
were affected using high amounts of peptide (2 mM).
Simultaneous blocking of integrins and CVADR was
without effect. CHO cells that are positive for integrin
Gene Therapy
␣v␤5 but negative for CVADR were not transducible,
which demonstrates that integrins alone are not sufficient
for transduction. In our studies, integrins did not seem
to have a major role under optimized transduction conditions in vitro, although in vivo infection may be more
effective in the presence of integrins.
We were unable to completely block adenoviral transduction in the majority of cell lines, and the presence of
other receptors for adenoviral uptake on bladder cancer
cell lines cannot be ruled out. However, the abundance
of CVADR expressed on these cells prohibits complete
blocking with antibodies. MHC class I molecules have
been suggested as alternative attachment receptors for
adenoviruses.6 All the investigated bladder cancer cell
lines in our study were highly positive for MHC-I, and
blocking studies using MHC-I antibodies had slight
effects on the transduction of the cell line T24, whereas
other cell lines remained unaffected.
There has been a vigorous debate about the use of serotype 5 adenoviral vectors in human gene therapy, since
these viruses are immunogenic and may cause an anaphylactic shock if administered in high titers directly into
the blood stream of preimmune patients.1 However,
Werthman et al18 injected adenoviral vectors both intraperitoneally (i.p.) and intravesically by instillation into the
bladder in mice. This study showed that i.p. injection produced viral infection of several organs besides the targeted tumor, while intravesical instillation into the bladder infected only a few epithelial cell layers of the
bladder. Thus, the bladder seems to be well isolated, and
may be the ideal organ for adenoviral vector-based
cancer gene therapy.
Taken together, urinary bladder cancer seems to be a
tumor well suited for adenoviral vector-based gene therapy due to the abundant expression of important receptors for attachment, internalization and membrane permeabilization. However, optimal transduction strategies
need to be developed that circumvent physical barriers
such as mucin.
Materials and methods
Culture of human urinary bladder cell lines
The human bladder cancer cell lines J82 (kind gifts of Dr
J Carlsson, Uppsala University, Sweden), T24 (gift of Dr
Adenoviral receptors in bladder cancer
A Loskog et al
551
Figure 4 Blocking of CVADR or integrins can inhibit the transduction. Transduction of tumor cells with Ad-CD154 vector can be partially blocked
by preincubation with RmcB antibody before exposure to with vector. Panel (a) describes the T24 cell line. 25% of cells blocked with irrelevant antibody
or control peptide were CD154-positive. Only 12% of cells blocked with RmcB antibody were CD154-positive (RmcB low) and by increasing the
concentration of RmcB the transduction rate could be further suppressed (RmcB high). Control vector transduced cells stained with the CD154 antibody
were negative for CD154. The use of integrin-specific peptides only had small effects on transduction. Panel (b) describes the RT-112 cell line. Note
that the CVADR and integrin blocking results represent different transduction experiments, which explains the difference in CD154 expression.
Gene Therapy
Adenoviral receptors in bladder cancer
A Loskog et al
552
A Richter-Dahlfors, Karolinska Institute, Sweden), RT112 (DSMZ, Braunschweig, Germany) and KV-19-19
(DSMZ) were cultured in RPMI supplemented with 10%
fetal bovine serum (FBS), 1% glutamine, 1% penicillinstreptomycin (PEST) (Life Technologies, Paisley, UK).
VM-CUB1 (DSMZ) was cultured using DMEM supplemented with 10% fetal bovine serum (FBS), 1% glutamine, 1% penicillin-streptomycin (PEST) and 0.1%
sodium pyruvate (Life Technologies). 293 human kidney
cells modified for production of adenoviral vectors
(ATCC, Bethesda, MD, USA) and Chinese hamster
ovarian (CHO) cells (ATCC) were cultured with DMEM
medium as described above.
Human biopsies
Biopsies were collected during transurethral resection
from patients with either superficially or invasively
growing bladder carcinoma. Samples were frozen and
stored at –70°C or short-time cultured (see below).
Adenoviral vectors
Replication-incompetent E1 + E3 deleted adenovirus (Ad)
serotype 5 vectors containing the murine CD154 (AdCD154) as a marker gene were used.15 As a negative control, the Ad-dl70-3 ‘empty’ control vector was used (a
kind gift from Dr Frank Graham, McMaster University,
Hamilton, Ontario).
Adenoviral transduction
The cells were transduced by pulsing with adenoviral
vectors (10–200 p.f.u./cell) for 1–2 h in serum-free cell
culture medium (200 (l). After this first incubation, 2 ml
ordinary medium was added and the vector pulse was
continued. Cells were analyzed for CD154 transgene
expression at different time-points after transduction by
flow cytometry or immunocytochemistry. For blocking
studies, different concentrations (1–10 ␮g) of monoclonal
antibodies against CVADR (RmcB)19or MHC class I
(DAKO, Carpinteria, CA, USA) were incubated with 200
000 cells on ice for 40 min. Adenoviral vectors were
added and the incubation was continued for 1 h at 37°C.
Cells were washed with PBS and further cultured with
ordinary medium. For blocking of vector–integrin interactions, 2 mM blocking peptide (GRGDSP) or control
peptide (GRGESP) (Bachem, Switzerland) was used in
the same manner as the antibody.
Flow cytometry
CD154 cell surface expression was detected by flow cytometry (FACSCalibur, Becton Dickinson, Sunnyvale, CA,
USA) using Armenian hamster anti-murine CD154 conjugated to phycoerythrin (PE) (PharMingen, San Diego,
CA, USA) Single cell suspensions of 1 × 105 cells were
incubated with antibodies for 10 min at room temperature, washed in 2 ml PBS, centrifuged at 200 g and the
supernatant was decanted. Cells were resuspended in 200
(l 1% paraformaldehyde in PBS before analysis.
Immunohistochemistry
Six-micrometer sections of tumor biopsies were placed on
ChemMate Capillary Gap Microscope Slides and stained
with TechMate 500 Plus (DAKO). Slides were fixated
with 4% paraformaldehyde and stained with the primary
antibodies rabbit anti-human/mouse-CVADR antibody,5
mouse anti-RmcB (CVADR),19 and mouse anti-human
Gene Therapy
integrin ␣v␤3 or ␣v␤5 (Chemicon International, Temecula,
CA, USA). Secondary antibodies were swine anti-rabbit
Ig/rabbit APAAP or rat anti-mouse Ig/rat APAAP
(DAKO). As a control of the staining irrelevant primary
antibodies was used, also 293 and CHO cells were used
as positive and negative controls, respectively. The tumor
material was further classified for grade and stage.
RT-PCR
RNA was isolated from cell lines using the TRIzol
Reagent (Life Technologies). The RNA pellet was suspended in diethyl pyrocarbonate-treated water (DEPCwater, RNAase free). The following CVADR primers
were used for PCR amplification: forward 5’-CCTGCGA
GATGTTACGTTGA and reverse 5’-CGCACCCAT
TCGACTTAGAT. For detection of ␤-actin the following
primers were used: forward 5’-GAGCAAGAGAGGCAT
CCTCA and reverse 5’-AGACAGCACTGTGTTGGCGT.
All primers were purchased from Cybergene, Stockholm,
Sweden. The PCR reaction was performed for 35 cycles
at 94°C for 45 s, 55°C for 45 s, 72°C for 1 min followed
by extension at 72°C for 10 min. PCR reactions were carried out in the presence of 5 mM each of dATP, dCTP,
dGTP and dTTP, 1.5 mM MgCl, 25 pg/mol of primer,
PCR buffer and 1 unit of Taq polymerase (Life
Technologies) in a total volume of 50 ␮l. PCR products
were analyzed on a 1% agarose-gel containing ethidium
bromide.
Ethical approval
The human ethical committee at Uppsala University
approved all experiments involving human material
(Dnr 99337).
Acknowledgements
This study was supported by The Swedish Cancer
Society, The Swedish Gene Therapy Program and the
Lions’ Cancer Fund at Uppsala University Hospital. The
authors thank Dr Frank Graham, McMaster University
for providing the Ad-dl70-3 vector, and Professor Jörgen
Carlsson, Uppsala University and PhD Agneta RichterDahlfors, Karolinska Institute, for providing the cell
lines, respectively.
References
1 Brenner M. Gene transfer by adenovectors. Blood 1999; 12:
3965–3967.
2 Roelvink PW et al. The Coxsackievirus-adenovirus receptor protein can function as a cellular attachment protein for adenovirus
serotypes from subgroups A, C, D, E and F. J Virol 1998; 10:
7909–7915.
3 Tomko RP, Xu R, Philipson L. HCAR and MCAR: the human
and mouse cellular receptors for subgroup C adenoviruses and
group B coxsackieviruses. Proc Natl Acad Sci USA 1997; 94:
3352–3356.
4 Fechner H et al. Expression of Coxsackie adenovirus receptor
and alphav-integrin does not correlate with adenovector targeting in vivo indicating anatomical vector barrier. Gene Therapy
1999; 6: 1520–1535.
5 Tomko RP et al. Expression of the adenovirus receptor and its
interaction with the fiber knob. Exp Cell Res 2000; 255: 47–55.
6 Bergelson JM. Receptors mediating adenovirus attachment and
internalisation. Biochem Pharm 1999; 57: 975–979.
Adenoviral receptors in bladder cancer
A Loskog et al
7 Wickham TJ, Mathias P, Cheresh DA, Nemerow GR. Integrins
␣v␤3 and ␣v␤5 promote adenovirus internalisation but not
virus attachment. Cell 1993; 1993: 73.
8 Wickham TJ, Filardo EJ, Cheresh DA, Nemerow GR. Integrin
␣v␤5 selectively promotes adenovirus mediated cell membrane
permeabilization. J Cell Biol 1994; 127: 257–264.
9 Dechecchi MC, Tamanini A, Bonizzato A, Cabrini G. Heparan
sulfate glycoaminoglycans are involved in adenovirus type 5
and 2-host cell interactions. Virol 2000; 268: 382–390.
10 Bergelson JM. Isolation of a common receptor for Coxsackie B
viruses and denoviruses 2 and 5. Science 1997; 275: 1320–1323.
11 Parsons CL, Shrom SH, Hanno PM, Mulholland SG. Bladder
surface mucin-examination of possible mechanisms for its antibacterial effect. Invest Urol 1978; 16: 196–200.
12 Engler H et al. Ethanol improves adenovirus-mediated gene
transfer and expression to the bladder epithelium of rodents.
Urology 1999; 53: 1049–1052.
13 McDonalds D et al. Coxsackie and adenovirus receptor (CAR)dependent and major histocompatibility complex (MHC) class
14
15
16
17
18
19
I-independent uptake of recombinant adenoviruses into human
tumour cells. Gene Therapy 1999; 6: 1512–1519.
Li Y et al. Loss of adenoviral receptor expression in human bladder cancer cells: a potential impact on the efficacy of gene therapy. Cancer Res 1999; 59: 325–330.
Loskog A et al. Potent antitumor effects of CD154 transduced
tumor cells in experimental bladder cancer. J Urol 2001; 166:
1093–1097.
Brooks PC, Clark RAF, Cheresh DA. Requirement of vascular
integrin ␣v␤3 for angiogenesis. Science 1994; 22: 569–571.
Carson SD, Hobbs JT, Tracy SM, Chapman NM. Expression of
the Coxsackievirus and adenovirus receptor in cultured human
umbilical vein endothelial cells: regulation in response to cell
density. J Virol 1999; 8: 7077–7079.
Werthman PE et al. Adenoviral–p53 gene transfer to orthotopic
and peritoneal murine bladder cancer. J Urol 1996; 155: 753–756.
Hsu K-HL, Lonberg-Holm K, Alstein B, Crowell RL. A monoclonal antibody specific for the cellular receptor for the group
B Coxsackieviruses. J Virol 1988; 5: 1647–1652.
553
Gene Therapy