School of Economics Study abroad opportunities 2015/16

Cancer Letters 302 (2011) 119–127
Contents lists available at ScienceDirect
Cancer Letters
journal homepage: www.elsevier.com/locate/canlet
Adeno-associated virus-mediated anti-DR5 chimeric antibody
expression suppresses human tumor growth in nude mice
Fujia Lv, Yuhe Qiu, Yaxi Zhang, Shilian Liu, Juan Shi ⇑⇑, Yanxin Liu ⇑⇑, Dexian Zheng ⇑
National Laboratory of Medical Molecular Biology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences & Peking Union Medical College,
5 Dong Dan San Tiao, Beijing 100005, China
a r t i c l e
i n f o
Article history:
Received 17 November 2010
Received in revised form 10 January 2011
Accepted 12 January 2011
Keywords:
Chimeric antibody
Death receptor 5
Gene transfer
Cancer therapy
a b s t r a c t
In the present study we demonstrate that adeno-associated virus (AAV)-mediated antiDR5 (death receptor 5) mouse–human chimeric antibody (shorten as Adximab) expression
significantly suppressed tumor cell growth by inducing apoptosis both in vitro and in vivo.
The viral-expressed and cell-secreted Adximab efficiently bound DR5 with an affinity of
0.7 nM and induced apoptosis of various tumor cells, but not normal cells. A single intramuscular injection of recombinant AAV particles resulted in a stable expression of Adximab in mouse serum for at least 70 days. AAV-mediated Adximab expression led to a
significant suppression of tumor growth in nude mice receiving xenografts of human liver
and colon cancer. These data suggest that chimeric antibody gene transfer may provide an
alternative strategy for the therapy of varieties of cancers.
Ó 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Tumor necrosis factor-related apoptosis inducing ligand
(TRAIL/Apo-2L), a member of the TNF family, can specifically induce cell death in various tumors but not in most
normal cells and tissues [1]. At least five TRAIL receptors
have been identified in humans: TRAIL receptor 1 (death
receptor 4, DR4), TRAIL receptor 2 (death receptor 5,
DR5), TRAIL receptor 3 (decoy receptor 1, DcR1), TRAIL
receptor 4 (decoy receptor 2, DcR2) and osteoprotegerin
(OPG). Both DR4 and DR5 contain an intracellular death
domain (DD), which triggers cell death signaling via activation of the caspase cascade and cleavage of downstream
caspase substrates [2–4]. The other three receptors –
DcR1, DcR2 and OPG – lack an intact death domain and
are unable to induce apoptosis, but they compete with
DR4 and DR5 for binding to TRAIL [5–7]. Upon TRAIL
stimulation, death receptors (DRs) rapidly recruit a
⇑ Corresponding author. Tel.: +86 10 65296409; fax: +86 10 65105102.
⇑⇑ Co-corresponding author. Tel.: +86 10 65296409; fax: +86 10
65105102.
E-mail address: [email protected] (D. Zheng).
0304-3835/$ - see front matter Ó 2011 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.canlet.2011.01.001
death-inducing signaling complex (DISC), which consists
of a Fas-associated death domain (FADD), procaspase-8
or -10, and/or receptor interaction protein (RIP) [8]. Therefore, procaspase-8 or -10 is cleaved into caspase-8 or -10,
which further activates the downstream procaspase-3, -6
and -7 [9]. This is known as the DR-mediated extrinsic signaling pathway. Bcl-2 interacting domain (Bid) could also
be cleaved by caspase-8 in certain types of cells, leading
to release of cytochrome c (Cyt C) from mitochondria and
formation of the apoptosome complex of procaspase-9,
Apaf1 and Cyt C, followed by successive activation of procaspase-9, -3, -6 and -7 [10]. This is known as the DR-mediated intrinsic signaling pathway.
DR5 expression is frequently detected on tumor cell
lines and various clinical cancer specimens, whereas there
is limited expression of DR5 in normal cells and tissues
[11]. In mice, DR5 deficiency enhances lymph node metastasis without affecting primary tumor development [12],
suggesting that DR5 plays an essential role in immune surveillance of malignancy under normal physiological conditions. Recently, it was reported that influenza and HIV-1
virus infection increased DR5 expression in NK cells, T
lymphocytes, and monocytes or monocyte-derived
120
F. Lv et al. / Cancer Letters 302 (2011) 119–127
dendritic cells [13,14]. Also DR5 could induce monocytemediated tumor cell apoptosis [15]. These data indicate
that DR5 plays a critical role in innate immunity. However,
the physiological role of DRs is not yet well understood.
Since DRs are expressed in a variety of normal cells and tissues, toxicity to normal cells is a concern when DRs are targeted. Indeed, some reports demonstrated toxicity to
normal hepatocytes, keratinocytes and neutrophils when
DR’s were triggered by recombinant TRAIL in vitro [16–
18]. However, Lawrence et al. demonstrated that the reported cytotoxic effect of recombinant TRAIL on normal
human liver cells was due to the tag linked with TRAIL, given that no cytotoxicity was seen with non-tagged TRAIL
[19].
A specific agonistic antibody against the death receptor
is considered to be a safer and more effective cancer therapy than DR ligands. Several specific antibodies against
DR4 or DR5 have been tested in clinical trials with promising outcomes [20–24]. At present, the immunogenicity of
mouse mAbs can be successfully reduced by generating
murine remodeling antibodies (mouse–human chimeric
antibody or humanized antibody), which have been clinically validated for use in a wide variety of applications in
immune therapy. The remodeling antibody has several
advantages, including an antigen binding affinity similar
to the intact parental murine antibody, the ability to induce antibody-dependent cell-mediated cytotoxicity
(ADCC) and complement dependent cytotoxicity (CDC) by
the human Fc domain, which may enhance the therapeutic
effect [25,26]. Despite the advantages of remodeling antibodies, antibody production, purification and half life
in vivo still create a bottleneck for both preclinical study
and clinical use. To overcome these problems, antibody
gene therapy has been demonstrated to be a viable therapeutic strategy for cancer and other chronic diseases. There
are three major factors to consider in gene therapy: the
transgene carrier, controllable transgene expression, and
safety. Considering the matter of the transgene carrier,
adeno-associated virus (AAV) vectors have been recognized as having distinct advantages, including long-term
transgene expression, low immunogenicity and more
safety in vivo [27].
Guo et al. previously reported that a novel mouse antihuman DR5 monoclonal antibody, AD5–10, induced apoptosis in various tumor cells with no toxicity to normal
hepatocytes or primary peripheral blood lymphocytes
in vitro. Systemic administration of AD5–10 in nude mice
with implanted human liver or lung cancer significantly
inhibited the tumor formation and growth without causing
toxicity to the liver, spleen or kidney [28]. These results
suggest that AD5–10 is a promising agonistic antibody
for cancer therapy. However, murine monoclonal antibodies can induce the human anti-mouse antibody (HAMA) response [29]. To circumvent this problem, Shi et al.
performed an antibody gene therapy study using AAVmediated expression of the single-chain Fv fragment (scFv)
of AD5–10 in transformed cell lines as well as human tumor-bearing mouse models [30]. The results showed that
viral expression of the scFv antibody induced significant
apoptosis in various tumor cells and prevented tumor
growth in the animal models. Since the life span of the
small molecule scFv in vivo is limited, in this study we further established an AAV-mediated human-mouse chimeric
antibody expression system by fusing the VH and VL cDNAs
of AD5–10 with the heavy and light chain of human Fc,
respectively. An anti-DR5 human–mouse chimeric antibody (Adximab) was successfully expressed both in cell
lines and animals by transformation with the expression
vectors or infection with the recombinant viral particles.
The expression, affinity and tumoricidal activity of the
chimeric antibody were examined, and the results demonstrated that AAV-mediated Adximab expression significantly suppressed tumor cell growth by inducing
apoptosis both in vitro and in vivo. This study provides further insight into the antibody gene therapy for clinical
application.
2. Materials and methods
2.1. Cell lines and culture
Human embryonic kidney cell line HEK293, breast cancer cell line MDA-MB-231, epithelial carcinoma cell line
Hela, colon cancer cell line HCT116, and lung cancer cell
line A549 were purchased from American Type Culture
Collection (Manassas, VA). Human glioma cancer cell line
U251 and liver cancer cell line SMMC7721 were purchased
from the Institute of Cell and Biochemistry, Chinese Academy of Sciences (Shanghai, China). Human colon cancer
cell line COLO205 was purchased from the Cell Culture
Center, Institute of Basic Medical Sciences (Beijing, China),
and human embryonic eye Tenon’s fibroblast cell line HFTF
from the Cell Bank, Chinese Academy of Sciences (Shanghai, China). These cells were cultured in RPMI 1640,
DMEM, or L-15 (Gibco, Grand Island, NY, USA) at 37 °C in
an atmosphere of 5% CO2 incubator. The entire medium
was supplemented with 10% heat-inactivated fetal calf serum (Hyclone, Logan, UT, USA), 100 units/ml penicillin, and
100 lg/ml streptomycin.
2.2. Constructs and generation of rAAV2 particles
VH and VL cDNAs of the chimeric antibody Adximab
were generated from the murine antibody AD5–10 [28].
The VH and VL cDNAs were linked with the Fc of human
IgG heavy and light chain to form the constructs Adximab
heavy chain (Adximab-HC) and Adximab light chain (Adximab-LC), respectively. The Adximab-HC and Adximab-LC
fragments were subcloned into an adeno-assoicated virus
serotype 2 (AAV2) expression vector pAM-CAG (cytomegalovirus enhancer plus chicken b-actin promoter) [31]. Enhanced green fluorescent protein (EGFP) subcloned into
the AAV2 vector was as the control. The generated expression vectors were depicted as pAM-CAG-Adximab-HC,
pAM-CAG-Adximab-LC, and pAM-CAG-EGFP, respectively.
Recombinant AAV viral particles (rAAV) were generated
in HEK293 cells using a three-plasmid packing system,
including one therapeutic plasmid and two helper
plasmids H22 and pFd6 [31]. The purification of rAAV
particles was performed by heparin affinity column
chromatography. The titers of rAAV particles were mea-
F. Lv et al. / Cancer Letters 302 (2011) 119–127
sured by real-time PCR and presented as virus genomes/ml
(vgs/ml).
2.3. Transfection and infection
HEK293 cells (5 105) were cultured in 6-well plates
and co-transfected with pAM-CAG-Adximab-HC and
pAM-CAG-Adximab-LC plasmids using lipofectamine
2000 (Invitrogen, CA, USA). The cells and cell medium were
collected separately at the indicated time for later on analysis. The cells transfected with pAM-CAG-EGFP plasmid
were used as control.
For rAAV infection, HEK293 cells (5 105) were cultured in 6-well plates in completed DMED medium for at
least 8 h, then rAAV-Adximab-HC (5 104 vgs per cell)
and rAAV-Adximab-LC (5 104 vgs per cell) particles were
added and cultured in DMED without fetal bovine serum
for an additional 6–8 h. The medium was switched to fresh
complete DMED medium and cultured for 96 h. The medium was then collected for the analysis of Adximab concentration by ELISA. The cells infected with rAAV-EGFP
(5 104 vgs per cell) were used as negative control.
2.4. Cell viability assay
Cells (1 104) were cultured in 96-well plates for at
least 8 h and treated with Adximab secreted from plasmids
co-transfected HEK293 cells medium for 24 h. The cell
viability was observed under a light microscope and
evaluated by 3-(4,5-dimethylthiazole-2-yl)-2,5-biphenyltetrazolium (MTT) assay according to the manufacturer’s
instructions (Sigma, St. Louis, MO, USA).
2.5. Western blot analysis
One hundred lg of total protein sample from cell or tissue lysate, and 10 ll of the cell culture medium were subjected to a 12% SDS–PAGE. The separated proteins in the
gel were transferred to a polyvinylidene difluoride (PVDF)
membrane (Amersham Biosciences, Little Chalfont, UK).
The membrane was then incubated with specific primary
121
antibodies and HRP-conjugated secondary antibodies, successively. Primary antibodies were directed against caspase-8 (Oncogene, La Jolla, CA, USA), caspase-3 and
caspase-9 (Cell Signaling, Beverly, MA, USA); poly ADP-ribose polymerase (PARP) (BD, San Jose, CA, USA); b-actin,
glyceraldehyde-3-phosphate dehydrogenase (GAPDH), human IgG Cj and horseradish peroxidase (HRP) linked antihuman IgG Fc (Sigma, St. Louis, MO, USA). HRP linked antimouse, anti-rabbit or anti-goat IgG secondary antibodies
were purchased from Zhongshan Biotech (Beijing, China).The proteins of interest were visualized by using the
ECL chemiluminescence system according to the manufacturer’s instructions (Amersham Biosciences).
2.6. ELISA
Ninety-six-well plates were coated with anti-human
IgG (H + L) chain antibody (KPL, Gaithersburg, Maryland)
(1:200 dilutions) and blocked with 5% nonfat dry milk.
Then medium or mouse serum was incubated in the wells
at indicated dilutions. The concentration of Adximab was
detected by following incubation with HRP-conjugated
anti-human IgG Fc antibody. The absorbance was measured at 492 nm on a microtiter reader (Thermo absystems, Finland). The concentration of Adximab in the cell
medium or mouse serum was calculated according to the
standard curve using normal human IgG.
To measure the affinity of Adximab binding to human
DR5, 96-well plates were coated with recombinant soluble
DR5 (3 lg/ml) [28] and blocked with 5% nonfat dry milk.
Then the cell medium was added at indicated dilutions
and incubated for 2 h, followed by incubation with HRPconjugated anti-human IgG Fc antibody. The absorbance
was measured at 492 nm on a microtiter reader (Thermo
absystems, Finland). The data were analyzed using GraphPad Prism Software.
2.7. Animal study
4- to 6-week-old male BALB/c nude mice were purchased from the Institute of Animal Sciences, Chinese
Fig. 1. Expression of Adximab in HEK293 cells. (A) Schematic diagram of the recombinant expression plasmids of pAM-CAG-Adximab-HC and pAM-CAGAdximab-LC. HC, heavy chain; LC, light chain. VH, variable region of heavy chain; VL, variable region of light chain; CH1, CH2, CH3, constant region 1, 2, and 3
of heavy chain; CK, constant region of light chain (j chain). (B) Expression of Adximab in HEK293 cells analyzed by Western blot. The cells were cotransfected with the recombinant plasmids of pAM-CAG-Adximab-HC and pAM-CAG-Adximab-LC. Ten microlitres of the cell culture media were collected
at indicated time points and subjected to a 12% SDS–PAGE followed by Western blot assay by using specific antibodies against human HC or LC, and
horseradish peroxidase (HRP)-conjugated secondary antibody, respectively. The cells transfected with pAM-CAG-EGFP were used as control. (C) Expression
of Adximab secreted in the cell culture medium. The cell culture media were collected at the indicated time points and subjected to a Sanwich ELISA to
determine the concentration of Adximab by using the anti-human IgG (H + L) antibody as capturer and the HRP-labeled anti-human IgG Fc antibody as the
second antibody. The data are expressed as the mean from at least three independent experiments.
122
F. Lv et al. / Cancer Letters 302 (2011) 119–127
Academy of Medical Science (Beijing, China) and housed
under specific pathogen-free conditions.
SMMC7721 tumor xenograft mouse model was established by subcutaneously injecting human liver cancer
cells SMMC7721 (2 106 cells per mouse) into the right
dorsal flank. When the tumor size reached about
50 mm3, the animals were divided into two groups
(n = 5) based on the tumor size. rAAV-Adximab (1 1011
particles) were injected into tumors in each mouse in the
experimental group, while an equal amount of rAAV-EGFP
particles were injected into tumors in each mouse in the
control group (n = 5). Tumor size was measured every
4 days over 32 days and calculated by the following formula: volume = (ab2)/2 (a: the longest axis, b: the shortest
axis).
HCT116 tumor xenograft mouse model was established
by s.c. inoculation of human colon cancer cells HCT116
(5 106 cells per mouse) into the right dorsal flank. The
mice were divided into four groups (n = 6–8) when the tumor size reached about 100 mm3. High dose of 1 1011
vgs and low dose of 2 1010 vgs of rAAV-Adximab and
rAAV-EGFP particles were injected into the tumors in the
experimental and control groups, respectively. Tumor size
was monitored every 4 days during a period of 32 days. At
the end of the experiment, the tumor was surgically excised and weighed.
For the in vivo expression study, 1 1011 vgs of rAAVAdximab particles were injected into the muscle of BALB/
c nude mice, and an equal amount of rAAV-EGFP particles
were injected as negative control. The blood samples were
collected once a week during a period of 70 days. The
expression of Adximab in mouse serum was analyzed by
ELISA. At the end of the experiment, the animals were sacrificed and tissues were surgically excised for Adximab
expression assay by Western blot or histochemical
analysis.
2.8. In situ cell apoptosis analysis
Paraffin-embedded sections of tumor tissues were prefixed in 4% paraformaldehyde and dewaxed. Cell apoptosis
was detected by using TUNEL (terminal deoxynucleotidyl
Fig. 2. Adximab efficiently binds DR5 and induces cell death by apoptosis in tumor cells. (A) The binding affinity of Adximab with DR5 determined by ELISA.
Wells of 96-well plate were coated with the recombinant soluble human DR5 protein, and incubated with culture medium of HEK293 cells transfected with
the recombinant plasmids or infected with rAAV-Adximab particles (5 104 vgs per cell) for 4 days. Horseradish peroxidase (HRP)-conjugated anti-human
IgG Fc antibody was used as the second antibody. The data are expressed as the mean from three independent experiments. (B) Cytotoxicity of Adximab in
various tumor cell lines. HCT116, SMMC7721, Hela, A549, MDA-MB-231, COLO205, U251, and HFTF cells were cultured in the 4th day’s medium (100 ll per
well) collected from HEK293 cells co-transfected with pAM-CAG-Adximab-HC and pAM-CAG-Adximab-LC. The cell viability was measured by MTT assay.
The data are expressed as the mean from three independent experiments. (C) Cytotoxicity of Adximab in HCT116 and SMMC7721 cells treated for 24 h with
the 4th day’s medium collected from the HEK293 cells co-transfected with pAM-CAG-Adximab-HC and pAM-CAG-Adximab-LC and observed under light
microscope (magnification, 200). (D) Caspase activation in HCT116 and SMMC7721 cells analyzed by Western blot. The cells were treated for 24 h with
the medium collected from HEK293 cells co-transfected with pAM-CAG-Adximab-HC and pAM-CAG-Adximab-LC and lysed with lysis buffer. The lysates
were subjected to SDS–PAGE followed by Western blot with the specific antibodies against caspase-8, caspase-9, caspase-3 and the caspase substrate PARP.
GAPDH was used as protein loading control.
F. Lv et al. / Cancer Letters 302 (2011) 119–127
123
transferase-mediated dUTP nick end labeling) assay
according to the manufacturer’s instructions (Roche,
Mannheim, Germany). Apoptotic cells in the defining regions of interest (ROI) were quantified by using automated
cell acquisition and the software for immunohistochemistry (Histoquest software, TissueGnostics GmbH, Vienna,
Austria).
2.9. Statistical analysis
All data were expressed as mean value ± standard deviation (SD), and the statistical differences between groups
were evaluated using Student’s t test. P < 0.05 was considered to be significant.
3. Results
3.1. Expression of Adximab in HEK 293 cells
Expression of the heavy chain (HC) and light chain (LC) of the chimeric antibody Adximab was first performed in HEK293 cells co-transfected
with the recombinant AAV vectors of pAM-CAG-Adximab-HC and pAMCAG-Adximab-LC (Fig. 1A). Western blots with antibodies against the
HC and LC of human IgG, respectively, showed that the soluble form of
Adximab was secreted into the medium of HEK293 cells in a time-dependent manner (Fig. 1B) and the concentration of Adximab, as determined
by ELISA, reached 11.03 ± 0.19 lg/ml on the fourth day post-transfection
(Fig. 1C), indicating that the chimeric antibody Adximab was expressed in
HEK293 successfully.
3.2. Adximab efficiently binds DR5 and induces apoptosis in various tumor cell
lines
To test the ability of Adximab to bind to DR5, an ELISA was performed
in the wells of a 96-well plate coated with recombinant soluble human
DR5 protein, and binding was detected with HRP-conjugated anti-human
IgG Fc antibody. As shown in Fig. 2A, the affinity of Adximab from the
medium of recombinant plasmid-transfected or rAAV particle-infected
HEK293 cells was 0.9 nM or 0.7 nM, respectively. To test the cytotoxicity
of Adximab to various tumor cell lines, the cells of HCT116 (colon cancer),
SMMC7721 (liver cancer), Hela (cervical carcinoma), A549 (lung cancer),
MDA-MB-231 (breast cancer), COLO205 (colon cancer), U251 (glioma)
and HFTF (human embryonic eye Tenon’s fibroblast cells) were cultured
for 24 h in medium (100 ll per well) collected from a 4-day culture of
HEK293 cells co-transfected with recombinant plasmids of pAM-CAGAdximab-HC and pAM-CAG-Adximab-LC, and the cell viability was measured by MTT assay. As shown in Fig. 2B, the viabilities of HCT116 and
SMMC7721 were dramatically reduced to 34.96 ± 4.76% and
33.56 ± 17.33%, respectively, and the viabilities of A549, MDA-MB-231,
COLO205, U251 and Hela cells were also reduced, to different levels.
Importantly, the viability of the normal cell HFTF was unaffected by Adximab, indicating that Adximab is specifically cytotoxic to tumor cells, but
not to normal cells. Morphological observation under a light microscope
showed many dead cells in tumor cell cultures treated with chimeric antibody (Fig. 2C). In addition, caspase activation analysis by Western blot
showed a decrease in full-length procaspase-3, -8, and -9, as well as an
increase in cleavage of downstream caspase substrate PARP (Fig. 2D), suggesting that the chimeric antibody induces tumor cell death by caspase
activation.
Collectively, these data indicate that AAV-mediated Adximab expression efficiently binds DR5 and induces caspase-dependent apoptosis specifically in various tumor cell lines.
3.3. AAV-mediates long-term gene expression of Adximab
To test the effects of stable Adximab expression in animals, 1 1011
vgs of recombinant AAV-Adximab viral particles were injected into the
muscle of BALB/c nude mice. An equal amount of recombinant AAV-EGFP
viral particles was used as control. Blood samples were collected once a
week over 70 days to analyze the expression of Adximab by ELISA. As
Fig. 3. Long-term expression of Adximab in BALB/c nude mice. (A)
Concentration of Adximab in the serum of mouse i.m. injected with rAAVAdximab particles. The serum was collected at indicated time and
analyzed by ELISA using anti-human IgG Fc antibodies. The serum of
mouse i.m. injected with rAAV-EGFP were used as control. (B) Expression
of Adximab in the mouse muscle detected by Western blot analysis. The
mouse muscle tissue was collected on the 70th day post-infection with
rAAV-Adximab or rAAV-EGFP particles and lysed with lysis buffer. The
lysate was subjected to SDS–PAGE followed by Western blot assay using
the specific antibodies against human HC and LC, respectively. Adximab
expression was representative of three specimens examined and two of
them were showed. The expression of GAPDH was used as protein loading
control.
shown in Fig. 3A, Adximab concentrations in the mouse serum reached
0.41 lg/ml on the 14th day post-infection, then increased to 1.5 lg/ml
on the 40th day and remained unchanged until the end of the experiment.
Western blot analysis at the end of the experiment confirmed significant
expression of Adximab in the injected muscle (Fig. 3B). Tissues including
heart, liver, spleen, lung, kidney and muscle were surgically excised from
the animals. The hematoxylin–eosin staining demonstrated that there
were no pathological changes in these tissues (photograph not shown).
Together, these data substantiate that recombinant AAV particles could
mediate long-term Adximab expression in vivo without systemic toxicity.
3.4. AAV-mediated Adximab expression suppresses tumor growth
To investigate the efficacy of chimeric antibody gene therapy for cancers, two human tumor xenograft models were established in BALB/c
nude mice by subcutaneously inoculating SMMC7721 liver cancer and
HCT116 colon cancer cells into the right dorsal flanks. When the tumor
size reached 50–100 mm3, the animals were divided into two groups
based on tumor size, and rAAV-Adximab and rAAV-EGFP viral particles
(1 1011 vgs) were injected into the tumors. The tumor size was measured every 4 days for a period of 32 days. The expression of Adximab
in the mice was examined by Western blot. As shown in Fig. 4A, Adximab
was highly expressed in the rAAV viral particle-injected SMMC7721 and
HCT116 xenografts. In the SMMC7721 xenograft model, the mean tumor
volume of the experimental group was 682 ± 204 mm3, whereas that of
the control group was 1821 ± 220 mm3 (Fig. 4B), suggesting that AAVmediated Adximab gene therapy in mouse markedly suppresses
SMMC7721 tumor growth (p = 0.00003, p < 0.01). In the HCT116
124
F. Lv et al. / Cancer Letters 302 (2011) 119–127
Fig. 4. Adximab expression suppressed human tumor growth in nude mice. (A) Western blot analysis of Adximab-HC and -LC expression in SMMC7721 and
HCT116 xenograft mice on the 32nd day post-infection with rAAV-Adximab. The animals were sacrificed and the tumor tissues were surgically excised. The
tumor tissues were smashed and lysed with lysis buffer. The lysate containing 100 lg of total protein was subjected to SDS–PAGE followed by Western blot
assay. Adximab expressions were representatives of 3–6 specimens examined and one of SMMC7721 and 3 of HCT116 tumor tissues were showed. The
rAAV-EGFP was used as negative control. The b-actin and GAPDH were used as protein loading controls. (B) rAAV-Adximab infection suppressed SMMC7721
tumor growth. The cells (2 106 cells per mouse) were inoculated into the right dorsal flanks of the mice. When the tumor size reached about 50 mm3, the
animals were divided into two groups (n = 5) based on the tumor size and rAAV-Adximab particles were injected into the tumor. rAAV-EGFP injection was
used as control. Tumor volumes were measured every 4 days during a period of 32 days. (C) rAAV-Adximab infection suppressed HCT116 tumor growth.
The cells (5 106 cells per mouse) were inoculated into the right dorsal flanks of the mice. When the tumor size reached about 50 mm3, the animals were
divided into four groups (n = 6–8) based on the tumor size. High dose (1 1011 vgs) and low dose (2 1010 vgs) of rAAV-Adximab particles were injected,
respectively. Equal dose of rAAV-EGFP was used as control. Tumor volumes were measured every 4 days during the period of 32 days. (D) HCT116 tumor
weight at the end of the experiment. ⁄⁄p < 0.01.
xenograft model, two experimental groups with injections of 1 1011 vgs
(high dose) and 2 1010 vgs (low dose) per mouse were performed. As
shown in Fig. 4C, the mean tumor volume was reduced from
5198.81 ± 1691.44 mm3 to 3059.20 ± 562.15 mm3 (p = 0.01478, p < 0.05)
in the high-dose group and from 4515.84 ± 820.54 mm3 to
2482.49 ± 393.06 mm3 (p = 0.00002, p < 0.01) in low-dose group. The tumor weight at the end of the experiment was reduced from
4.30 ± 0.97 g to 2.74 ± 0.49 g (p = 0.0056, p < 0.01) in the high-dose group
and from 3.77 ± 0.92 g to 2.07 ± 0.32 g (p = 0.0002, p < 0.01) in the lowdose group (Fig. 4D). Taken together, these data demonstrate that AAVmediated Adximab expression in mice suppressed tumor growth
significantly.
To determine whether the tumoricidal activity of AAV-mediated Adximab expression results in tumor cell death by apoptosis in vivo, TUNEL
analysis was carried out in tumor tissue sections. As shown in Fig. 5A, a
substantial number of apoptotic cells were detected in SMMC7721 liver
tumor injected with rAAV-Adximab particles, whereas few were observed
in the tumor injected with rAAV-EGFP control particles. A similar phenomenon was confirmed in HCT116 colon tumor injected with high dose
of rAAV-Adximab particles. After quantification by automated cell acquisition and Histoquest software, TUNEL mean intensities in SMMC7721 liver tumor injected with rAAV-Adximab particles was 81.20 ± 3.70%,
injected with rAAV-EGFP was 38.65 ± 3.04% (p = 0.0001, p < 0.01). In
HCT116 tumor tissue, the TUNEL mean intensities injected with rAAVAdximab particles was 55.19 ± 6.84%, whereas injected with rAAV-EGFP
was only 20.25 ± 5.50% (p = 0.0023, p < 0.01) (Fig. 5B, C).
Thus, these data suggest that AAV-mediated Adximab antibody gene
expression efficiently suppresses human tumor growth in nude mice by
inducing cell death by apoptosis.
4. Discussion
We have demonstrated in the present study that adenoassociated virus-mediated anti-DR5 mouse–human chimeric antibody (Adximab) gene expression significantly suppressed human tumor growth both in vitro and in vivo.
The affinity of the viral expressed Adximab to DR5 was
0.7–0.9 nM, similar to 0.3 nM of the parental murine antibody AD5–10 [28], but significantly higher than 4 nM of
the viral expressed scFv of AD5–10 [30], suggesting that
the AAV-expressed chimeric antibody adequately retained
the properties of the parental murine monoclonal
antibody.
The cell-secreted Adximab to DR5 induced cell death in
various tumor cells by apoptosis, but not in the normal
cells of HFTF, and triggered the activation of caspase -3,
-8, and -9 and the cleavage of PARP, the substrate of
F. Lv et al. / Cancer Letters 302 (2011) 119–127
125
Fig. 5. Apoptotic cell death in rAAV-Adximab infected tumor tissue. (A) Paraffin-embedded sections of SMMC7721 and HCT116 tumor tissues were
prepared from the mice at the end of the experiment. Apoptotic cells in the tumor tissues were visualized by TUNEL assay. Brown colored cells represent
apoptotic cells. rAAV-EGFP was used as control (magnification, 400). (B) Apoptotic cells in the defining regions of interest (ROI) were quantified by using
automated cell acquisition and the software for immunohistochemistry (Histoquest). (C) TUNEL mean intensities in tumor tissue from the representative
animal infected with rAAV-Adximab and rAAV-EGFP, respectively. ⁄⁄p < 0.01. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)
caspases. The tumoricidal activity of Adximab was confirmed in vivo. Single injections of recombinant AAV viral
particles in mouse muscle resulted in long-term (at least
70 days) and therapeutic expressions of the chimeric antibody in mouse serum and led to a remarkable suppression
of tumor xenograft growth. However, it was notably that
there was no significant differences in the ability of high
(1011 vgs) and low (2 1010 vgs) doses of viral Adximab
to inhibit HCT116 tumor growth. This result suggests that
that the viral dose might differ by more magnitudes, which
is remained to be evaluated.
The viral expressed chimeric antibody might be more
effective as a therapeutic than the scFv reported by Shi
et al. [30] because the chimeric antibody may have a longer
half life in vivo. The improved tumoricidal activity of Adximab versus the parental murine antibody indicates that
the chimeric antibody may elicit a more effective immunological response than the scFv fragment. Li et al. (unpublished data) have confirmed that the chimeric antibody
induced antibody-dependent cell-mediated cytotoxicity
(ADCC) and complement dependent cytotoxicity (CDC),
consistent with reports by Sánchez-Mejorada et al. [32]
and Nechansky et al. [33].
Importantly, AAV-mediated antibody gene expression
for the therapy of cancer and other chronic diseases could
resolve several serious bottlenecks, such as time, costs
and techniques involved in producing and manufacturing
reshaped monoclonal antibodies, because this strategy permits long-term antibody gene expression in vivo, leading to
significant inhibition of tumor growth in human liver cancer SMMC7721 and colon cancer HCT116 in xenograft
mouse models. Comparing with recombinant antibody,
126
F. Lv et al. / Cancer Letters 302 (2011) 119–127
which half life in vivo is usually from several days to no
more than 4 weeks [34], AAV-medaited antibody expression, which could be last at least for 70 days, may much
superior to administration of the purified recombinant
antibodies. However, several important problems need to
be resolved before antibody gene transfer becomes an
alternative supplement to the current therapies. First,
expression of the chimeric antibody following gene transfer
has to be controllable because certain normal cells may express the autologous DR5, which could be targeted by the
chimeric antibody and lead to autoimmune conditions. Second, the immunogenicity of chimeric antibodies during
long-term expression has to be clarified, because treatment
with infliximab, a chimeric monoclonal IgG1 antibody
against tumor necrosis factor, has been reported to result
in the formation of antibodies against infliximab [35]. And
finally, the therapeutic expression level of the chimeric
antibody in the serum needs to be further augmented.
In summary, we have reported that adeno-associated
virus-mediated an anti-DR5 chimeric antibody expression
suppresses human tumor growth in nude mice. This study
may enhance our understanding of antibody gene therapy
and its application in clinical trials, thus providing an alternative treatment strategy for varieties of cancers.
5. Conflicts of interest
The authors declare that there are no conflicts of
interest.
Acknowledgements
This work was partially supported by the Natural Science Foundation of China (Grant Nos. 30623009,
30772495, and 30972684) and the State Key Basic Research Program of China (Grant No. 2007CB507404).
References
[1] S.R. Wiley, K. Schooley, P.J. Smolak, W.S. Din, C.P. Huang, J.K. Nicholl,
et al., Identification and characterization of a new member of the
TNF family that induces apoptosis, Immunity 3 (1995) 673–682.
[2] G. Pan, K. O’Rourke, A.M. Chinnaiyan, R. Gentz, R. Ebner, J. Ni, et al., The
receptor for the cytotoxic ligand TRAIL, Science 276 (1997) 111–113.
[3] H. Walczak, M.A. Degli-Esposti, R.S. Johnson, P.J. Smolak, J.Y. Waugh,
N. Boiani, et al., TRAIL-R2: a novel apoptosis-mediating receptor for
TRAIL, EMBO J. 16 (1997) 5386–5397.
[4] P.M. Chaudhary, M. Eby, A. Jasmin, A. Bookwalter, J. Murray, L. Hood,
Death receptor 5, a new member of the TNFR family, and DR4 induce
FADD-dependent apoptosis and activate the NF-kappaB pathway,
Immunity 7 (1997) 821–830.
[5] S.A. Marsters, J.P. Sheridan, R.M. Pitti, A. Huang, M. Skubatch, D.
Baldwin, et al., A novel receptor for Apo2L/TRAIL contains a
truncated death domain, Curr. Biol. 7 (1997) 1003–1006.
[6] M.A. Degli-Esposti, P.J. Smolak, H. Walczak, J. Waugh, C.P. Huang, R.F.
DuBose, et al., Cloning and characterization of TRAIL-R3, a novel
member of the emerging TRAIL receptor family, J. Exp. Med. 186
(1997) 1165–1170.
[7] J.G. Emery, P. McDonnell, M.B. Burke, K.C. Deen, S. Lyn, C. Silverman,
et al., Osteoprotegerin is a receptor for the cytotoxic ligand TRAIL, J.
Biol. Chem. 273 (1998) 14363–14367.
[8] H. Walczak, T.L. Haas, Biochemical analysis of the native TRAIL
death-inducing signaling complex, Methods Mol. Biol. 414 (2008)
221–239.
[9] A. Thorburn, Death receptor-induced cell killing, Cell. Signal. 16
(2004) 139–144.
[10] S.J. Korsmeyer, M.C. Wei, M. Saito, S. Weiler, K.J. Oh, P.H. Schlesinger,
Pro-apoptotic cascade activates BID, which oligomerizes BAK or BAX
into pores that result in the release of cytochrome c, Cell Death
Differ. 7 (2000) 1166–1173.
[11] R.A. Daniels, H. Turley, F.C. Kimberley, X.S. Liu, J. Mongkolsapaya, P.
Ch’En, et al., Expression of TRAIL and TRAIL receptors in normal and
malignant tissues, Cell Res. 15 (2005) 430–438.
[12] A. Grosse-Wilde, O. Voloshanenko, S.L. Bailey, G.M. Longton, U.
Schaefer, A.I. Csernok, et al., TRAIL-R deficiency in mice enhances
lymph node metastasis without affecting primary tumor
development, J. Clin. Invest. 118 (2008) 100–110.
[13] J.P. Herbeuvala, G.M. Shearer, HIV-1 immunopathogenesis: how
good interferon turns bad, Clin. Immunol. 123 (2007) 121–128.
[14] E. Ishikawa, M. Nakazawa, M. Yoshinari, M. Minami, Role of tumor
necrosis factor-related apoptosis-inducing ligand in immune response
to influenza virus infection in mice, J. Virol. 79 (2005) 7658–7663.
[15] T.S. Griffith, S.R. Wiley, M.Z. Kubin, L.M. Sedger, C.R. Maliszewski,
N.A. Fanger, Monocyte-mediated tumoricidal activity via the tumor
necrosis factor-related cytokine, TRAIL, J. Exp. Med. 189 (1999)
1343–1354.
[16] M. Jo, T.H. Kim, D.W. Seol, J.E. Esplen, K. Dorko, T.R. Billiar, et al.,
Apoptosis induced in normal human hepatocytes by tumor necrosis
factor-related apoptosis-inducing ligand, Nat. Med. 6 (2000) 564–567.
[17] J.Z. Qin, P.E. Bacon, V. Chaturvedi, B. Bonish, B.J. Nickoloff, Pathways
involved in proliferating, senescent and immortalized keratinocyte
cell death mediated by two different TRAIL preparations, Exp.
Dermatol. 11 (2002) 573–583.
[18] S.A. Renshaw, J.S. Parmar, V. Singleton, S.J. Rowe, D.H. Dockrell, S.K.
Dower, et al., Acceleration of human neutrophil apoptosis by TRAIL,
J. Immunol. 170 (2003) 1027–1033.
[19] D. Lawrence, Z. Shahrokh, S. Marsters, K. Achilles, D. Shih, B.
Mounho, et al., Differential hepatocyte toxicity of recombinant
Apo2L/TRAIL versions, Nat. Med. 7 (2001) 383–385.
[20] L. Zhang, X. Zhang, G.W. Barrisford, A.F. Olumi, Lexatumumab
(TRAIL-receptor 2 mAb) induces expression of DR5 and promotes
apoptosis in primary and metastatic renal cell carcinoma in a mouse
orthotopic model, Cancer Lett. 251 (2007) 146–157.
[21] P.G. Oliver, A.F. LoBuglio, K.R. Zinn, H. Kim, L. Nan, T. Zhou, et al.,
Treatment of human colon cancer xenografts with TRA-8 anti-death
receptor 5 antibody alone or in combination with CPT-11, Clin.
Cancer Res. 14 (2008) 2180–2189.
[22] P.J. Frederick, J.E. Kendrick, J.M. Straughn Jr, D.L. Della Manna, P.G.
Oliver, H.Y. Lin, et al., Effect of TRA-8 anti-death receptor 5 antibody
in combination with chemotherapy in an ex vivo human ovarian
cancer model, Int. J. Gynecol. Cancer 19 (2009) 814–819.
[23] T.A. Luster, J.A. Carrell, K. McCormick, D. Sun, R. Humphreys,
Mapatumumab and lexatumumab induce apoptosis in TRAIL-R1
and TRAIL-R2 antibody-resistant NSCLC cell lines when treated in
combination with bortezomib, Mol. Cancer Ther. 8 (2009) 292–
302.
[24] M.R. Smith, F. Jin, I. Joshi, Bortezomib sensitizes non-Hodgkin’s
lymphoma cells to apoptosis induced by antibodies to tumor
necrosis factor related apoptosis-inducing ligand (TRAIL) receptors
TRAIL-R1 and TRAIL-R2, Clin. Cancer Res. 13 (2007) 5528s–5534s.
[25] A.M. Scott, Z. Liu, C. Murone, T.G. Johns, D. MacGregor, F.E. Smyth,
et al., Immunological effects of chimeric anti-GD3 monoclonal
antibody KM871 in patients with metastatic melanoma, Cancer
Immun. 5 (2005) 3–15.
[26] V. Ghetie, E.S. Ward, FcRn: the MHC class I-related receptor that is
more than an IgG transporter, Immunol. Today 18 (1997) 592–598.
[27] D. Grimm, J.S. Lee, L. Wang, T. Desai, B. Akache, T.A. Storm, et al., In
vitro and in vivo gene therapy vector evolution via multispecies
interbreeding and retargeting of adeno-associated viruses, J. Virol.
82 (2008) 5887–5911.
[28] Y. Guo, C. Chen, Y. Zheng, J. Zhang, X. Tao, S. Liu, et al., A novel antihuman DR5 monoclonal antibody with tumoricidal activity induces
caspase-dependent and caspase-independent cell death, J. Biol.
Chem. 280 (2005) 41940–41952.
[29] M. Yamashita, Y. Katakura, S. Shirahata, Recent advances in the
generation of human monoclonal antibody, Cytotechnology 55
(2007) 55–60.
[30] J. Shi, Y. Liu, Y. Zheng, Y. Guo, J. Zhang, P.T. Cheung, et al., Therapeutic
expression of an anti-death receptor 5 single-chain fixed-variable region
prevents tumor growth in mice, Cancer Res. 66 (2006) 11946–11953.
[31] Y. Zhang, H. Ma, J. Zhang, S. Liu, Y. Liu, D. Zheng, AAV-mediated TRAIL
gene expression driven by hTERT promoter suppressed human
hepatocellular carcinoma growth in mice, Life Sci. 82 (2008) 1154–
1161.
F. Lv et al. / Cancer Letters 302 (2011) 119–127
[32] G. Sánchez-Mejorada, C. Rosales, Signal transduction by
immunoglobulin Fc receptors, J. Leukoc. Biol. 63 (1998) 521–533.
[33] A. Nechansky, O.H. Szolar, P. Siegl, I. Zinoecker, N. Halanek, S.
Wiederkum, et al., Complement dependent cytotoxicity (CDC)
activity of a humanized anti Lewis-Y antibody: FACS-based assay
versus the ‘classical’ radioactive method – qualification, comparison
and application of the FACS-based approach, J. Pharm. Biomed. Anal.
49 (2009) 1014–1020.
127
[34] R.J. Keizer, A.D. Huitema, J.H. Schellens, J.H. Beijnen, Clinical
pharmacokinetics of therapeutic monoclonal antibodies, Clin.
Pharmacokinet. 49 (2010) 493–507.
[35] F. Baert, M. Noman, S. Vermeire, G. Van Assche, G. D’ Haens, A.
Carbonez, et al., Influence of immunogenicity on the long-term
efficacy of infliximab in Crohn’s disease, N. Engl. J. Med. 348 (2003)
601–608.
`