• No results found

Original Article A fiber-modified adenovirus co-expressing HSV-TK and Coli.NTR enhances antitumor activities in breast cancer cells


Academic year: 2020

Share "Original Article A fiber-modified adenovirus co-expressing HSV-TK and Coli.NTR enhances antitumor activities in breast cancer cells"


Loading.... (view fulltext now)

Full text


Original Article

A fiber-modified adenovirus co-expressing HSV-TK and

Coli.NTR enhances antitumor activities in breast

cancer cells

Yang Zhan1,2, Bin Yu1,2, Zhen Wang1, Yu Zhang1, Hai-Hong Zhang1,2, Hao Wu1, Xiao Feng1, Ran-Shen Geng1, Wei Kong1,2, Xiang-Hui Yu1,2

1National Engineering Laboratory for AIDS Vaccine, 2Key Laboratory for Molecular Enzymology and Engineering,

The Ministry of Education, School of Life Sciences, Jilin University, 2699 Qianjin Street, Changchun 130012, China

Received April 7, 2014; Accepted May 26, 2014; Epub May 15, 2014; Published June 1, 2014

Abstract: Breast cancers especially in late and metastatic stages remain refractory to treatment despite advances in surgical techniques and chemotherapy. Suicide gene therapy based on adenoviral technology will be promising strategies for such advanced diseases. We previously showed that co-expression of herpes simplex virus thymidine kinase (HSV-TK) and Escherichia coli nitroreductase (Coli.NTR) by an hTERT-driven adenovirus vector resulted in additive anti-tumor effects in breast cancer cells in vitro and in vivo. As many tumor tissue and cancer cells express

low level of coxsackie-adenovirus receptor (CAR), which is the functional receptor for the fiber protein of human adenovirus serotype 5 (Ad5), novel Ad5 vectors containing genetically modified fiber are attractive vehicles for

achieving targeted gene transfer and improving suicide gene expression in these cancer cells. In the present study,

we first built a simplified Ad5 vector platform for fiber modification and quick detection for gene transfer. Then a fiber-modified adenovirus vector containing an RGD motif in the HI loop of the fiber knob was constructed. After recombined with HSV-TK and Coli.NTR gene, this fiber-modified Ad5 vector (Ad-RGD-hT-TK/NTR) was compared with

that of our previously constructed Ad5 vector (Ad-hT-TK/NTR) for its therapeutic effects in human breast cancer cell

lines. The anti-tumor activity of Ad-RGD-hT-TK/NTR was significantly enhanced compared with Ad-hT-TK/NTR both in vitro and in vivo. This new vector platform provided a robust and simplified approach for capsid modification, and the fiber-modified Ad5 with double suicide genes under the control of hTERT promoter would be a useful gene therapy

strategy for breast cancer.

Keywords: Adenovirus, suicide gene, breast cancer


Suicide gene therapy, or gene-directed enzyme prodrug therapy, is becoming increasingly pop-ular for cancer treatments. This approach is based on the introduction of viral or bacterial genes into target cells to express an enzyme to catalyze a non-toxic prodrug into a highly toxic metabolite that would subsequently cause tumor cell death [1]. As there are proven

bene-fits of this type of therapy, the present study is

focused on developing an improved suicide gene therapy strategy for the treatment of human cancer [2, 3].

One of the most important factors affecting the

efficiency of suicide gene therapy is the choice

of a delivery vector to effectively transfer the therapeutic genes into the target cells [4]. Replication-defective adenovirus serotype 5 (Ad5) has become a promising transfer vehicle due to its well-characterized biology, large transgene capacity, and gene transfer to a wide variety of cell types, both dividing and non-dividing. Moreover, because of its safety and optimized production, adenoviral vector has become the most widely-used viral vector for gene therapy, especially for cancer gene thera-py [5].


gene delivery [6]. Deficiency of the

coxsackie-adenovirus receptor (CAR) is considered to be the biological basis of this phenomenon, beca- use CAR has been shown to be a docking site for Ad and act as a key receptor for the

enhance-ment of the virus-to-host affinity and the initia -tion of the virus internaliza-tion to host cells

[7-10]. Significant variability in CAR expression

between normal tissue and human cancer cells has been reported [11-13]. A major hurdle for Ad5-mediated cancer gene therapy is that a

significant fraction of cancer cells lack the CAR

receptor [14, 15]. The use of genetically

modi-fied Ad5 to enhance viral tropism and entry is

one approach that can overcome the problem

of CAR deficiency in the cancer cells while potentially enhancing the therapeutic efficacy

of suicide gene therapy.

The HI-loop is a region on the outer surface of

the knob of Ad5. It has been identified as a readily modifiable site to incorporate heterolo -gous amino acid sequences without affecting

the correct folding of the fiber polypeptide and the biological functions of the fiber, leaving the

peptides exposed on the outside of the virion [16, 17]. However, the complexity of inserting foreign gene corresponding to the motif of inter-est into the HI-loop hampers the development of targeted vectors. In the present study, we

developed a simplified system, using shuttle

plasmids and through in vitro ligation, to rapidly

and efficiently insert a gene of interest in the

HI-loop. At the same time, the inclusion of the

firefly luciferase reporter gene in the construct allows easy monitoring of gene transfer effi -ciency. Using this approach, we constructed a

fiber-mutant Ad5 vector containing an Arg-Gly-Asp (RGD) peptide in the HI-loop of the fiber knob domain. The incorporation of the RGD

peptide has been shown to increase gene transfer to various types of tumors lacking

suf-ficient CAR expression [18-21].

The most commonly-used therapeutic suicide gene strategy is based on the herpes simplex virus thymidine kinase gene (HSV-TK) followed by ganciclovir (GCV) administration [22]. How-

ever, specificity is of concern due to gap-junc -tion-mediated bystander effect and activation against cycling cells by HSV-TK [23]. The Esch- erichia coli nitroreductase (Coli.NTR)/5-(aza- ridin-1-yl)-2, 4-dinitrobenzamide (CB1954) sys-tem is another suicide gene therapy approach. However, it also has a strong bystander effect

as a result of the generation of a potent DNA

cross-linking agent that induces apoptosis in both dividing and non-dividing cells [24]. To overcome the shortcomings of using a single suicide gene, we used the combination of two suicide gene systems in one vector in this study to enhance targeted tumor cell killing effects

[25, 26]. We have also used the tumor-specific

human telomerase reverse transcriptase (hT-

ERT) promoter to improve the specificity of sui -cide gene expression in cancer cells. Ultimately, this strategy may reduce the amount of pro-drugs required for therapy, and improve

anti-tumor efficacy.

Materials and methods

Cell lines

The human normal lung fibroblast CCD-11Lu

and human embryonic kidney HEK293 cells were purchased from the American Type Cu- lture Collection (Manassas, USA). The ZR-75-30, MCF-7 human breast carcinoma cells, the KB and Tca-8113 human head and neck

can-cer cells, and the HeLa human uterine can-cervical

cancer cells were obtained from the Cell Bank of the Chinese Academy of Science (Shanghai, China). All cell lines were cultured in compatible media supplemented with 10% fetal bovine serum (FBS) (Invitrogen, Carlsbad, CA), penicil-lin (50 U/ml), and streptomycin (50 U/ml) in the presence of 5% CO2.

Flow cytometry analysis

The expression of CAR, αvβ3 and αvβ5 in the

cells were determined by flow cytometry analy

-sis with a MoFlo XDP cell sorter (Beckman

Coulter). Cultured cells were washed and har-vested with PBS, resuspended in PBS contain-ing 1% BSA for 30 min at room temperature, and then incubated with primary antibodies for 1 h on ice. Subsequently, the cells were washed and incubated with FITC-conjugated IgG for an additional 1 h. After washing with 1% BSA/PBS,

the cells were analyzed by flow cytometry. The

same secondary antibodies alone served as the negative control for each cell line. The

pri-mary antibodies specific for CAR (Rmcb, Mil-lipore, USA), the anti-αvβ3 mAb LM609, or the anti-αvβ5 mAb P1F6 were purchased from Mil- lipore. The secondary antibody is FITC-conjug-

ated rabbit anti-mouse IgG antibody (Dingguo,


Construction of target Ad5 platform and ad-enovirus plasmids

The Adeno-X expression system (BD, USA), wh-ich contains a viral DNA vector (pAdeno-X) and

a shuttle vector (pShuttle), was used to struct the recombinant Ad5 platform. The con-struction procedure is as follows. First, site-directed mutagenesis was performed to delete the SpeI and XbaI sites in the pShuttle vector. Primers, 5’-TGACATTGATTATTGAGTACTTATTAAT-AGTATACAAT-3’ (forward) and 5’-ATTGTATACT- ATTAATAAGTACTCAATAATCAATGTCA-3’ (reverse) were used to abrogate the SpeI site in pShut- tle. Subsequently, primers to the luciferase gene 5’-ACTAGTACCATGGAAGACGCC-3’ (forw- ard) and 5’-GGTACCTTACACGGCGATCTTT-3’ (re- verse), were used to delete the Xba I site in pShuttle. The forward primer to the luciferase gene contains SpeI site, which was then ligated to the XbaI multi-cloning site in pShuttle. After ligation, neither XbaI nor SpeI site remained. By this approach, we generated a new pShuttle

vector (pShuttle-ΔSX-Luc) with the deletion of

the SpeI and XbaI restriction sites and the addi-tion of a luciferase reporter gene. The

lucifer-ase expression cassette in pShuttle-ΔSX-Luc

was then cloned into the E1/E3 deletion region by an in vitro ligation method using I-CeuI and

PI-SceI sites to construct the pAd-Luc platform


The gene fragment flanked by the SpeI and XbaI sites was PCR amplified with specific prim -ers 5’-GCGCTGACTCTTAAGGACTAG-3’ (forward) and 5’-CGTCGCATGCTCCTCTAG-3’ (reverse) fr-

om the pAd-Luc vector, and cloned into the

pGEM-T Easy vector (Promega, Wis, USA) to generate the second shuttle plasmid, pT-SX

(shuttle 2). Subsequently, the fiber knob region

located between StuI and MfeI sites in pT-SX

was PCR amplified with specific primers,

5’-C-TACAACAAAGGCCTTTACTTG-3’ (forward) and 5’- TGCAATTGAAAAATAAACACG-3’ (reverse), and cl- oned into pGEM-T Easy vector to generate the pT-SM plasmid. Finally, for easy cloning of a

for-eign motif into the HI-loop of the fiber,

site-directed mutagenesis was conducted using

specific primers, 5’-ACAGGAGACACAAC

TGGTA-CCGGATCCCTCGAGCCAAGTGCATACTC-3’ (forw- ard) and 5’-GAGTATGCACTTGG CTCGAGGGATC- CGGTACCAGTTGTGTCTCCTGT-3’ (reverse), to in- troduce Kpn I, BamH I and XhoI sites into the HI-loop to generate the pT-SM-KBX plasmid (shuttle 1). Shuttle 1 was then digested with

KpnI and XhoI, and ligated with an

oligonucle-otide corresponding to the RGD peptide, result

-ing in the pT-SM-RGD plasmid. The knob region

in shuttle 2 was subsequently replaced with

the knob-RGD region in shuttle 1 by using the StuI and MfeI sites to obtain pT-SX-RGD. Finally, the knob region in pAd-Luc was replaced with the knob-RGD region in pT-SX-RGD with the use

of the SpeI and XbaI sites to generate pAd-


The recombinant adenovirus plasmid pAd-hT-TK/NTR previously constructed in our lab con-tains the HSV-TK and Coli.NTR genes under the control of the hTERT promoter in the E1/E3 deletion region. Plasmid pAd-hT-TK/NTR was digested with I-Ceu I and PI-Sce I, and then ligated with the I-CeuI and PI-SceI-digested

pAd-RGD-Luc to generate the pAd-RGD-hT-TK/

NTR plasmid.

Production of recombinant Ad5

The recombinant Ad5 (Ad-Luc, Ad-RGD-Luc, Ad-hT-TK/NTR and Ad-RGD-hT-TK/NTR) were amplified in HEK293 cells and purified by CsCl

gradient centrifugation. The titers were

deter-mined by the TCID50 method [25]. The negative control Ad-Blank contains the Ad5 genome with no insert in the E1/E3 deletion region (Newg- ene, Shanghai, China).

Ad5 infection assay and cytotoxicity assay in vitro

To assess the fiber-modified Ad5 infection abil -ity, experimental cells (1 × 105 cells) were seed-ed into 24-well plates and incubatseed-ed overnight

to allow adherence. The virus, Ad-Blank, Ad-Luc, or Ad-RGD-Luc, infected the cells at MOI = 10.

The cells were incubated at 37°C in 5% CO2 for 3 h, and the medium was then replenished with medium containing 2% FBS. Forty-eight hours later, the cells were rinsed with PBS and ass- ayed for luciferase activity according to the manufacturer’s protocol (Promega) using a Fluo-

roskan Ascent FL luminometer (Thermo, Japan).

To assess the cytotoxicity of the suicide genes in Ad5, experimental cells (5 × 103 cells) were seeded into a 96-well plate overnight before infection with Ad-Blank, Ad-hT-TK/NTR, or Ad-

RGD-hT-TK/NTR at MOI = 5 for 3 h, followed by culturing with GCV at various concentrations


medium was replenished with the same

con-centrations of GCV (0.1, 1, 10, or 100 μM). On

the fourth day, the cells were cultured with GCV

and also with CB1954 (0.1, 1, 10, or 100 μM)

(Sigma, USA). On the sixth day, cell viability was determined by the MTT assay. MTT (5 mg/ml) were added to each well, and the cells were incubated for another 4 h. The medium

con-taining MTT was then discarded, and 150 μL DMSO were added to each well and shaken to

dissolve the crystals. Absorbance (A) values

were measured at OD490 nm. Cell survival

rates were calculated according to the following

formula: survival = [(average A value of experi -mental group-average A value of control group)/ (average A value of positive group-average A value of control group)] × 100%, where the con-trol group includes blank wells, and the positive group is comprised of cells cultured with medi-um containing sermedi-um only.

Evaluation of anti-tumor effects and toxicity in vivo

ZR-75-30 cells (5 × 105 cells) were

subcutane-ously inoculated into the flank of BALB/c(-/-)

nude mice on day 1. On days 2 and 3, PBS,

Ad-Blank, Ad-hT-TK/NTR, or Ad-RGD-hT-TK/NTR

(1 × 108 PFU) was injected into the inoculated

area. From days 4 to 8, the mice received daily injections of GCV (75 mg/kg) intraperitoneally. On day 4 and day 11, the mice received CB1954 (40 mg/kg) intraperitoneally. At the same time, the mice in the PBS group were injected intra-peritoneally with physiological saline, and ser- ved as the negative control. Tumor size was measured every 4 days from day 14. Tumor vol-umes were calculated according to the

follow-ing equation: tumor volume = π/6 ab2, where a is the larger diameter, and b the shorter diam-eter (mm). To measure potential liver toxicity induced by the adenovirus, an identical experi-ment was set up. However, the mice in this

experiment were sacrificed on day 15, and the liver was removed, fixed in formalin, embedded in paraffin, sectioned, and stained with hema -toxylin and eosin (H&E). Mice were euthanized by CO2 when tumor size reaches 1500 mm3, or

sooner if ambulation was significantly impaired


by tumor size or cutaneous ulcerations devel-oped. Body weight was another symptom as an endpoint. The mice were euthanized when the body weights were 20% less than the normal control. To alleviate suffering, we followed the standard protocol, and monitored the mice every day. Once the unhealthy symptoms were determined, we stopped the study on the sick mice and euthanized them by CO2. All animal

Figure 1. The strategy for constructing the Ad5 vector platform and shuttle vectors. A: The luciferase expression


Figure 2. Flow cytometry analysis of levels of CAR and integrins αvβ3 and αvβ5 in experimental cell lines. Cells were labeled with mouse anti-CAR monoclonal antibody,

mouse anti-human integrin αvβ3 (LM609) antibody, or mouse anti-human integrin αvβ5 (P1F6) antibody. The cells were then incubated with FITC-conjugated rabbit


procedures were approved by the Jilin University Institutional Animal Care and Use Committee and the animal experiment approval number was 2010011203.

Statistical analysis

One-way ANOVA followed by Fisher’s LSD was

employed using SPSS 13.0 software. P<0.05

was considered to represent a significant differ -ence. All data points are presented as means ±

SD (standard error).


Construction of the Ad5 vector platform and generation of recombinant fiber-modified adenoviruses

Given that the commercial Ad5 system Adeno-X

(> 32 kb) is difficult to operate directly for genetic modification, it is necessary to design

an easy construction process to insert or incor-porate foreign gene of interest through in vitro ligation. As described in Materials and Methods and illustrated in Figure 1, we created a novel Ad5 vector platform based on a shuttle plasmid

(pT-SX) including whole fiber gene for targeted modification, and a new genome plasmid (pAd-Luc), which contains the luciferase reporter

gene under the control of the CMV promoter to

allow easy identification of the tropism of fiber-modified adenovirus. Using this novel system, we first generated a pAd-RGD-LUC plasmid, which contains a modified knob with the inclu

-sion of an RGD peptide to improve gene trans

-fer to CAR-deficient cancer cells. Then the Ad5 plasmids pAd-Luc and pAd-RGD-Luc were then

digested with Pac-I, and transfected into

ses using antibodies specific for CAR or the

indicated integrins to investigate the expres-sion of CAR and integrins in a number of human cell lines. As shown in Figure 2F, the HeLa cells

expressed the highest levels of CAR with a ratio of 96.98%. In contrast, the levels of CAR in other cell models examined were greatly reduced, and the ratio in ZR-75-30 and MCF7 breast cancer cells, in KB and Tca-8113 head

and neck cancer cells, and in normal lung fibro

-blast CCD-11lu cells was 12.62%, 34.22%,

20.19%, 35.53%, and 0.36%, respectively (Figure 2J, 2N, 2R, 2V, 2B). These cells also express variable levels of integrin. As shown in Figure 2K, 2T, ZR-75-30 cells expressed the

highest level of αvβ3 integrin (77.15% ratio), and

KB cells the highest level of αvβ5 integrin (17.65%).

Enhancement of transduction abilities of fiber-modified Ad5

To investigate the infectivity of the fiber-modi

-fied adenoviral vector, cells with different CAR

and integrin expression levels, as described

above, were infected with Ad-Blank, Ad-Luc, or Ad-RGD-Luc, and the log10-transformed values of the luciferase activity were presented in Figure 3. When the unmodified vector Ad-Luc

was used to infect cells at MOI 10, the

infectiv-ity followed the order of HeLa > Tca-8113 > MCF-7 > KB > ZR-75-30 > CCD-11Lu, consis -tent with their CAR levels. The data suggested CAR dependence of the transduction, and indi-cated a requirement to reroute the virus via


non-CAR pathways to achieve efficient gene delivery to CAR-deficient cells. Compared with the unmodified Ad-Luc, the infectivity of the fiber-modified Ad-RGD-Luc on all cell models

Figure 3. Transfer efficiency of fiber-modified Ad5. Recombinant

adeno-virus Ad-Blank, Ad-Luc, or Ad-RGD-Luc was used to infect the cells at

10 MOI. The luciferase activity was determined 48 h later. All data

rep-resent the mean ± SD of three experiments. (*) statistically significant (P<0.05) from the Ad-Luc group.

HEK293 cells (not homologous recombination) to obtain the recombinant Ad5 to analyze the

infectivity of the fiber-modified ade -noviral vector.

CAR and integrin expression in various cell models

Ad5 accomplishes initial binding to target cells via the primary recep-tor CAR. After binding, the particle achieves internalization via

inter-action with integrins of the αvβ3 and αvβ5 class [27-29]. We carried


-was significantly increased, and the most dra -matic increase was observed in ZR-75-30 cells,

which are deficient for CAR but express the high level of αvβ3 integrin.

Enhancement of HSV-TK and Coli.NTR cyto-toxic effects by fiber-modified Ad5 in vitro

In order to investigate whether the fiber-modi

-fied Ad5 vector could be used to more effec

-tively deliver suicide genes to CAR-deficient cancer cells and improve the efficacy of gene

therapy, we constructed the Ad-RGD-hT-TK/

NTR as described in Materials and Methods, and performed MTT assay on experimental ce-

lls infected with either Ad-hT-TK/NTR or

AdRGD-hT-TK/NTR followed by treatment with the pro-drugs GCV and CB1954. Ad-hT-TK/NTR and

Ad-RGD-hT-TK/NTR contain double suicide

ge-nes under the control of the hTERT promoter. As shown in Figure 4, although in all cancer cell lines, both Ad-hT-TK/NTR/GCV+CB1954 and

Ad-RGD-hT-TK/NTR/GCV+CB1954 displayed

gr-eater cytotoxicity than the Ad-Blank/GCV+C-

B1954 negative control, the cell killing efficacy was much improved with the use of the fiber-modified vector. The greater enhancement of cy-totoxicity by Ad-RGD-hT-TK/NTR/GCV+CB1954

versus Ad-hT-TK/NTR/GCV+CB1954 against ZR-75-30 and KB cells as compared to MCF7 and Tca-8113 cells was likely due to the fact that ZR-75-30 and KB cells express higher lev-els of integrin than the other two cell types. It is of note that neither Ad-hT-TK/NTR/GCV+C-

B1954 nor Ad-RGD-hT-TK/NTR/GCV+CB1954 showed significant cytotoxicity against the nor

-mal CCD-11Lu cells (Figure 4A), in which mini-mal hTERT promoter activity has been detected

[25]. These results indicated that the RGD pep -tide in the HI-loop and the use of the hTERT pro-moter to drive the expression of the suicide genes could improve the sensitivity and

speci-ficity of cancer cells to GCV and CB1954.

Inhibition of tumor growth by HSV-TK and Coli. NTR in vivo

To confirm our in vitro data, we compared the in vivo efficacy of the Ad-RGD-hT-TK/NTR/GCV+


CB1954 system versus the Ad-hT-TK/NTR/GC- V+CB1954 system against the growth of ZR-75-30 xenograft tumors. The experimental proce-dure is depicted in Figure 5B. The average tumor volume of the group that received con-ventional Ad5 with the double suicide genes (Ad-hT-TK/NTR/GCV+CB1954) was 21.8% of

Figure 4.In vitro anti-tumor effects of suicide gene therapy. The CCD-11 Lu (A), ZR-75-30 (B), MCF-7 (C), KB (D), and Tca-8113 (E) cells were infected with Ad-Blank, Ad-hT-TK/NTR, or Ad-RGD-hT-TK/NTR (MOI = 5) for 3 h, and then cul -tured with medium containing various concentrations of GCV/CB1954 (0.1, 1, 10, 100 µM). Cell survival rates were

measured as described in Materials and Methods. Each experiment was performed in quintuplicate. (*) statistically significant (P<0.05) from the Ad-Blank control group. (**) statistically significant (P<0.05) from the Ad-hT-TK/NTR


that of the control group (PBS/physiological

saline) on Day 30 (Figure 5A). The tumor

vol-ume of the fiber-modified Ad5 vector group (Ad-RGD-hT-TK/NTR/GCV+CB1954) was 10.3%

of the control group and 47.2% of the

conven-tional Ad5 group on Day 30 (P<0.05). Mea-nwhile, the survival times of the fiber-modified

Ad5 vector group was 197.2% of the control and 134.6% of the conventional Ad5 groups (Table 1). These data suggest that fiber modifi -cation of the adenovirus also led to an improved

anti-tumor efficacy in vivo, reducing the tumor

volume to nearly half the size of that in the

unmodified group. While the tumor volume of

the Ad-Blank/GCV+CB1954 group was slightly

the concentration of each prodrug was low [30]. Therefore, by adopting a double suicide gene strategy into just one vector, the combina-tion can compensate for the limitacombina-tions and shortcomings of using only a single suicide gene, while reducing the dose of the Ad5 virus and the prodrugs needed.

In a study conducted by Yu et al., combining HSV-TK with Coli.NTR under the control of the hTERT promoter in one Ad5 vector produced an additive anti-tumor effect in cancer cells both

in vitro and in vivo [25]. However, despite of the


improved efficacy, significant difficulties remain in the transfer of specific therapeutic genes to

Figure 5. Enhanced inhibition of the growth of ZR-75-30 tumor growth

by Ad-RGD-hT-TK/NTR. A: Growth curve of ZR-75-30 xenograft tumors. B: Treatment schedule. Each point represents the mean ± SD of nine mice. (*) P<0.05. (*) statistically significant (P<0.05) from the PBS control group. (**) statistically significant (P<0.05) from the Ad-hT-TK/

NTR group and the control group.

lower than that of the PBS control group, the difference was no

statis-tically significant (P > 0.05), indi -cating that the prodrugs (GCV and CB1954) had minimal effect on tumor growth in the absence of the suicide genes.

To examine potential hepatotoxici-ty associated with the dissemina-tion of Ad5 into hepatic tissue induced by the HSV-TK/GCV and Coli.NTR/CB1954 system, histo-pathological studies were per-formed. Histological staining of the sections of the hepatic cord

col-lected at Day 15 after gene trans -fer showed that the cells in all groups were arranged neatly in a radiating pattern around the vein with no apparent hepatotoxicity or hepatic necrosis (Figure 6).


In an attempt to reduce the sys-temic toxicities and poor quality of life caused by conventional con-ventional chemotherapy, much of the research has been devoted to more cancer-targeted therapeutic approaches, including suicide gene therapy. Previous studies suggest-ed that expression of one suicide gene, even when the concentration of the prodrug was maximal, could not completely prevent tumor growth [30]. On the other hand, better anti-tumor effects could be achieved in tumor cells transfected with two suicide genes, even when Table 1. Survival of mice with ZR-75-30 tumor models by the


Treatment Dose (PFU) Prodrug Survival (Days)

PBS 1 ×108 Sodium Chloride 89.22±10.02 Ad-Blank 1 ×108 GCV+CB1954 94.3±8.7 Ad-hT-TK/NTR 1 ×108 GCV+CB1954 130.67±34.5*


target cells by Ad5 vector. One of the difficulties

is that many cancer cells express few CAR receptors for Ad5 on their surface, limiting the

entry of the vector and transduction efficiency

of foreign genes [31-33]. Many groups have

reported that Ad5 vector containing an RGD peptide in the HI-loop of the fiber knob binds with high affinities to integrins on the cell sur -face, mediating not only CAR-dependent gene

delivery but also CAR-independent, RGD-integrin (αvβ3 and αvβ5)-dependent gene deliv-ery [19, 20]. Since integrins are generally expressed abundantly on the surface of cancer

cells, fiber-modified Ad5 containing RGD pep -tide in the HI-loop therefore represents a poten-tially effective tool for the treatment of tumors

lacking sufficient CAR expression [34, 35].

While HI-loop is an optimal site for inserting

gene of interest into the fiber to expand tropism

of Ad vector, the complexity of inserting foreign

DNA of interest into this region hampers the

development of retargeted vectors. Therefore,

we developed a simplified system that requires only two shuttle plasmids to construct a

fiber-mutant Ad vector containing the gene of inter-est in the HI-loop through in vitro ligation. With this platform, we constructed an Ad5 vector

containing an RGD motif in the HI-loop and a

luciferase report gene for monitoring of gene

transfer efficiency. We showed increased gene

transfer to various types of cancer cells lacking

sufficient CAR expression, and the most dra -matic increase was observed in cells that lack CAR expression but have high levels of integ-rins. Subsequently, the luciferase gene was

replaced by double suicide genes, HSV-TK and

Coli.NTR, under the control of the hTERT pro-moter to allow the expression of the suicide genes in cancer cells. We showed that suicide

gene therapy delivered by the fiber-modified Ad

vector led to a ~50% reduction of the size of ZR-75-30 xenograft tumors. It is of note that the concentration of the prodrug CB1954 used in our in vivo experiment was only half of the dose used in the study by Yu et al. [25],

indicat-ing that the use of the fiber-modified Ad5 may

reduce dose-limiting toxicity by directing more viruses into target cancer cells.

Due to both high levels of CAR and large endo -thelial pores to allow Ad vectors easy access to the liver, unwanted foreign gene transfer into this organ is a concern [10]. We therefore per-formed histopathological studies to screen for potential hepatotoxicity. The livers from neither the Ad-hT-TK/NTR/GCV+CB1954 group nor the

Ad-RGD-hT-TK/NTR/GCV+CB1954 group dis

-played obvious signs of hepatotoxicity, such as

inflammatory cell infiltration. This could be

attributed to two features of the vector system.

First, the use of the tumor-specific promoter

hTERT in the Ad5 vector in this suicide gene sys-tem makes the expression of HSV-TK and Coli.

NTR to be more confined to the tumor tissues

and prevented their expression in normal liver. Second, because the double suicide genes in this system were combined in a single vector, the dose of Ad5 was reduced to 1 × 108 PFU. Tjuvajev and colleagues reported that, at this dose, an Ad vector expressing HSV-TK showed little or no elevation of serum aspartate

trans-Figure 6. Histological analysis of the liver of ZR-75-30-tumor-bearing mice. Treatment schedule was the same as depicted in Figure 5B. On Day 15, the liver was removed, and histological analyses were performed using H&E


aminase and alanine transaminase, clinically relevant markers of liver damage, while these markers were markedly elevated with HSV-TK-expressing Ad at the dose of 5 × 108 PFU [36]. In conclusion, we developed a system for rapid

construction of fiber-modified Ad5 vector con -taining foreign peptides in the HI loop of the

fiber knob. This system allows generation of

recombinant Ad5 vector by in vitro ligation and can be easily monitored by a reporter gene in vitro and in vivo. This study also demonstrated

that a fiber-modified Ad5 containing an RGD

motif in the HI-loop combining double suicide genes HSV-TK and Coli.NTR under the control of the hTERT promoter could enhance anti-tumor effects in vitro and in vivo with no apparent hepatic side effects. Thus, this new vector sys-tem provides a promising gene therapy app-

roach for treating CAR-deficient cancers.


This study was supported by the National Nature Science Foundation of China (No. 30872396, 31300756) and the Key Projects in the National Science & Technology Pillar Pr- ogram in the Twelfth Five-year Plan Period (Grant 2012ZX10001-009).

Disclosure of conflict of interest


Address correspondence to: Dr. Xiang-Hui Yu or Wei Kong, School of Life Sciences, Jilin University, 2699

Qianjin Street, Changchun 130012, China. Fax: +86 431 85167751; E-mail: xianghui@jlu.edu.cn (XHY); weikong@jlu.edu.cn (WK)


[1] Freeman SM, Abboud CN, Whartenby KA,

Packman CH, Koeplin DS, Moolten FL and

Abraham GN. The “bystander effect”: tumor regression when a fraction of the tumor mass

is genetically modified. Cancer Res 1993; 53:


[2] Both GW. Recent progress in gene-directed en-zyme prodrug therapy: an emerging cancer treatment. Curr Opin Mol Ther 2009; 11: 421-432.

[3] Schellmann N, Deckert PM, Bachran D, Fuchs

H and Bachran C. Targeted enzyme prodrug therapies. Mini Rev Med Chem 2010; 10: 887-904.

[4] Vassaux G and Martin-Duque P. Use of suicide

genes for cancer gene therapy: study of the

dif-ferent approaches. Expert Opin Biol Ther 2004; 4: 519-530.

[5] Doloff JC and Waxman DJ. Adenoviral vectors

for prodrug activation-based gene therapy for cancer. Anticancer Agents Med Chem 2014; 14: 115-126.

[6] Sewell DA, Li D, Duan L, Schwartz MR and

O’Malley BW Jr. Optimizing suicide gene

thera-py for head and neck cancer. Laryngoscope

1997; 107: 1490-1495.

[7] Miller CR, Buchsbaum DJ, Reynolds PN, Doug

-las JT, Gillespie GY, Mayo MS, Raben D and Curiel DT. Differential susceptibility of primary

and established human glioma cells to adeno-virus infection: targeting via the epidermal

growth factor receptor achieves fiber

receptor-independent gene transfer. Cancer Res 1998; 58: 5738-5748.

[8] Bergelson JM, Cunningham JA, Droguett G,

Kurt-Jones EA, Krithivas A, Hong JS, Horwitz

MS, Crowell RL and Finberg RW. Isolation of a

common receptor for Coxsackie B viruses and adenoviruses 2 and 5. Science 1997; 275: 1320-1323.

[9] Bergelson JM, Krithivas A, Celi L, Droguett G, Horwitz MS, Wickham T, Crowell RL and Fin -berg RW. The murine CAR homolog is a recep-tor for coxsackie B viruses and adenoviruses. J.Virol 1998; 72: 415-419.

[10] Campos SK and Barry MA. Current advances and future challenges in Adenoviral vector biol-ogy and targeting. Curr Gene Ther 2007; 7: 189-204.

[11] Sullivan DE, Dash S, Du H, Hiramatsu N, Aydin

F, Kolls J, Blanchard J, Baskin G and Gerber

MA. Liver-directed gene transfer in non-human

primates. Hum Gene Ther 1997; 8: 1195-1206.

[12] Hemminki A, Kanerva A, Liu B, Wang M, Alva

-rez RD, Siegal GP and Curiel DT. Modulation of

coxsackie-adenovirus receptor expression for increased adenoviral transgene expression. Cancer Res 2003; 63: 847-853.

[13] Huard J, Lochmuller H, Acsadi G, Jani A, Mass -ie B and Karpati G. The route of administration is a major determinant of the transduction

ef-ficiency of rat tissues by adenoviral recombi -nants. Gene Ther 1995; 2: 107-115.

[14] Giaginis CT, Zarros AC, Papaefthymiou MA,

Pa-padopouli AE, Sfiniadakis IK and Theocharis

SE. Coxsackievirus and adenovirus receptor expression in human endometrial adenocarci-noma: possible clinical implications. World J Surg Oncol 2008; 6: 59.

[15] Sachs MD, Rauen KA, Ramamurthy M, Dodson JL, De Marzo AM, Putzi MJ, Schoenberg MP


[16] Krasnykh V, Dmitriev I, Mikheeva G, Miller CR, Belousova N and Curiel DT. Characterization of

an adenovirus vector containing a

heterolo-gous peptide epitope in the HI loop of the fiber

knob. J Virol 1998; 72: 1844-1852.

[17] Xia D, Henry LJ, Gerard RD and Deisenhofer J.

Crystal structure of the receptor-binding

do-main of adenovirus type 5 fiber protein at 1.7 A

resolution. Structure 1994; 2: 1259-1270. [18] Dmitriev I, Krasnykh V, Miller CR, Wang M,

Kashentseva E, Mikheeva G, Belousova N and

Curiel DT. An adenovirus vector with genetical

-ly modified fibers demonstrates expanded tro -pism via utilization of a coxsackievirus and adenovirus receptor-independent cell entry mechanism. J Virol 1998; 72: 9706-9713. [19] Kanerva A, Wang M, Bauerschmitz GJ, Lam JT,

Desmond RA, Bhoola SM, Barnes MN, Alvarez RD, Siegal GP, Curiel DT and Hemminki A.

Gene transfer to ovarian cancer versus normal

tissues with fiber-modified adenoviruses. Mol

Ther 2002; 5: 695-704.

[20] Kasono K, Blackwell JL, Douglas JT, Dmitriev I, Strong TV, Reynolds P, Kropf DA, Carroll WR, Peters GE, Bucy RP, Curiel DT and Krasnykh V.

Selective gene delivery to head and neck can-cer cells via an integrin targeted adenoviral vector. Clin Cancer Res 1999; 5: 2571-2579. [21] Haviv YS, Blackwell JL, Kanerva A, Nagi P,

Krasnykh V, Dmitriev I, Wang M, Naito S, Lei X, Hemminki A, Carey D and Curiel DT. Adenoviral

gene therapy for renal cancer requires retar-geting to alternative cellular receptors. Cancer Res 2002; 62: 4273-4281.

[22] Salama SA, Kamel M, Christman G, Wang HQ, Fouad HM and Al-Hendy A. Gene therapy of uterine leiomyoma: adenovirus-mediated her-pes simplex virus thymidine kinase/ganciclovir treatment inhibits growth of human and rat leiomyoma cells in vitro and in a nude mouse model. Gynecol Obstet Invest 2007; 63: 61-70.

[23] Patterson AV, Saunders MP and Greco O. Pro-drugs in genetic chemoradiotherapy. Curr

Pharm Des 2003; 9: 2131-2154.

[24] Djeha AH, Thomson TA, Leung H, Searle PF, Young LS, Kerr DJ, Harris PA, Mountain A and

Wrighton CJ. Combined adenovirus-mediated nitroreductase gene delivery and CB1954 treatment: a well-tolerated therapy for estab-lished solid tumors. Mol Ther 2001; 3: 233-240.

[25] Yu B, Zhang Y, Zhan Y, Zha X, Wu Y, Zhang X,

Dong Q, Kong W and Yu X. Co-expression of

herpes simplex virus thymidine kinase and Escherichia coli nitroreductase by an hTERT-driven adenovirus vector in breast cancer cells results in additive anti-tumor effects. Oncol Rep 2011; 26: 255-264.

[26] Sekar TV, Foygel K, Ilovich O and Paulmurugan R. Noninvasive Theranostic Imaging of

HSV1-sr39TK-NTR/GCV-CB1954 Dual-Prodrug Ther

-apy in Metastatic Lung Lesions of MDA-MB-231

Triple Negative Breast Cancer in Mice. Ther-anostics 2014; 4: 460-474.

[27] Howitt J, Bewley MC, Graziano V, Flanagan JM and Freimuth P. Structural basis for variation in

adenovirus affinity for the cellular coxsackievi -rus and adenovi-rus receptor. J Biol Chem 2003; 278: 26208-26215.

[28] Ruoslahti E. RGD and other recognition se

-quences for integrins. Annu Rev Cell Dev Biol

1996; 12: 697-715.

[29] Wickham TJ, Mathias P, Cheresh DA and Nem -erow GR. Integrins alpha v beta 3 and alpha v beta 5 promote adenovirus internalization but not virus attachment. Cell 1993; 73: 309-319. [30] Uckert W, Kammertons T, Haack K, Qin Z,

Ge-bert J, Schendel DJ and Blankenstein T. Double

suicide gene (cytosine deaminase and herpes simplex virus thymidine kinase) but not single gene transfer allows reliable elimination of tu-mor cells in vivo. Hum Gene Ther 1998; 9: 855-865.

[31] Anders M, Rösch T, Küster K, Becker I, Höfler

H, Stein HJ, Meining A, Wiedenmann B and Sarbia M. Expression and function of the cox-sackie and adenovirus receptor in Barrett’s esophagus and associated neoplasia. Cancer Gene Ther 2009; 16: 508-515.

[32] Stecker K, Koschel A, Wiedenmann B and

An-ders M. Loss of Coxsackie and adenovirus re -ceptor downregulates alpha-catenin expres-sion. Br J Cancer 2009; 101: 1574-1579. [33] Wu PC, Wang Q, Dong ZM, Chu E, Roberson

RS, Ivanova IC and Wu DY. Expression of cox -sackie and adenovirus receptor distinguishes transitional cancer states in therapy-induced

cellular senescence. Cell Death Dis 2010; 1:


[34] Ria R, Vacca A, Ribatti D, Di Raimondo F, Mer

-chionne F and Dammacco F. Alpha(v)beta(3)

integrin engagement enhances cell invasive-ness in human multiple myeloma. Haemato-logica 2002; 87: 836-845.

[35] Desgrosellier JS and Cheresh DA. Integrins in

cancer: biological implications and therapeutic opportunities. Nat Rev Cancer 2010; 10: 9-22. [36] Koehne G, Doubrovin M, Doubrovina E, Zanzo -nico P, Gallardo HF, Ivanova A, Balatoni J, Ter-uya-Feldstein J, Heller G, May C, Ponomarev V, Ruan S, Finn R, Blasberg RG, Bornmann W,

Riviere I, Sadelain M, O’Reilly RJ, Larson SM

and Tjuvajev JG. Serial in vivo imaging of the targeted migration of human


Figure 1. The strategy for constructing the Ad5 vector platform and shuttle vectors. A: The luciferase expression cassette with abrogated SpeI and XbaI sites (pShuttle-ΔSX-Luc) was inserted into the E1/E3 deletion region by an in vitro ligation method usin
Figure 2. Flow cytometry analysis of levels of CAR and integrins αvβ3 and αvβ5 in experimental cell lines
Figure 3. Transfer efficiency of fiber-modified Ad5. Recombinant adeno-virus Ad-Blank, Ad-Luc, or Ad-RGD-Luc was used to infect the cells at 10 MOI
Figure 4. In vitro anti-tumor effects of suicide gene therapy. The CCD-11 Lu (A), ZR-75-30 (B), MCF-7 (C), KB (D), and Tca-8113 (E) cells were infected with Ad-Blank, Ad-hT-TK/NTR, or Ad-RGD-hT-TK/NTR (MOI = 5) for 3 h, and then cul-tured with medium conta


Related documents

The Elementary code inside the DataVerse in presence of Agent and Motion Element moves to produce patterns.. The Agent chooses some of the patterns that make sense to

While ABC technologies are being widely adopted throughout the world, the use of ABC in moderate to high seismic areas has been somewhat limited because of the seismic forces that

From the above analysis, constructing a high-tech enterprise performance index system should not only consider the profitability of the enterprise but also to consider the growth

In conclusion, the Intrector ® was found to be safe and effective in our small case series, and it may offer a viable alternative to the tap and inject procedure

It can be concluded from the results of this study that these rangeland forbs are palatable and their CP content is high and promising to support different classes of livestock

For a spatially distributed circuit, KCL is applied to each locally referenced group one at a time, as the local reference terminals are electrically connected only through a

For the use of N -substituted phosphoramidic acids in the synthesis of pyrophosphate groups, see: Quin &amp; Jankowski (1994). For a corresponding dimer complex with similar P—O and

The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are