- Email: [email protected]
Enhancement of the Adenoviral Sensitivity of Human Ovarian Cancer Cells by Transient Expression of Coxsackievirus and Adenovirus Receptor (CAR)
Gynecologic Oncology 85, 260 –265 (2002) doi:10.1006/gyno.2002.6607, available online at http://www.idealibrary.com on
Enhancement of the Adenoviral Sensitivity of Human Ovarian Cancer Cells by Transient Expression of Coxsackievirus and Adenovirus Receptor (CAR) Jong-Sik Kim,* ,† Seung-Hoon Lee,‡ Yong-Suk Cho,‡ Jung-Joo Choi,‡ Young Ho Kim,† and Je-Ho Lee‡ ,1 *Clinical Research Center, Samsung Biomedical Research Institute, Seoul 135-710, Korea; ‡Molecular Therapy Research Center, Sungkyunkwan University, Samsung Medical Center Annex 8F, Ilwon-dong 50, Kangnam-ku, Seoul 135-710, Korea; and †Department of Microbiology, College of Natural Sciences, Kyungpook National University, Taegu, 702-701, Korea Received September 17, 2001
Objective. The coxsackievirus and adenovirus receptor (CAR) is known to be a primary receptor for attachment during adenovirus infection of target cells and thus is closely related to adenoviral infection efficiency. To extend this notion to human ovarian cancer, we investigated the difference in expression levels of CAR in human ovarian cancer cell lines and whether their adenoviral sensitivities are enhanced by transient transfection of the CAR gene. Methods. Adenoviral infection efficiency was examined by flow cytometry analysis, -galactosidase staining, and -galactosidase activity assay. Expression of the CAR-specific mRNA and protein was analyzed by reverse transcription polymerase chain reaction and flow cytometry, respectively. Results. Expression of CAR in human ovarian cancer cell lines (SKOV3, 2774, PA-1, and OVCAR3) appeared to be correlated with their susceptibilities to adenovirus-mediated gene delivery. The 2774 and PA-1 cells expressing an easily detectable level of CAR on the cell surface showed a higher susceptibility to infection with both AdCMVGFP and AdRSVgal than SKOV3 and OVCAR3 cells, both of which had hardly detectable levels of CAR. Ectopic expression of the CAR gene by transient transfection of these ovarian cancer cells resulted in a dramatic increase in their adenoviral sensitivities. Conclusion. These data show that the expression of CAR is closely related to susceptibility to adenovirus infection in human ovarian cancer cells. These results indicate that the CAR gene can be a useful tool in boosting the efficiency of adenoviral vector-mediated gene therapy against human ovarian cancer. © 2002 Elsevier Science (USA) Key Words: coxsackievirus and adenovirus receptor (CAR); adenovirus vector; infection efficiency; ovarian cancer cell lines.
INTRODUCTION Ovarian cancer is the most lethal gynecological cancer in Korea [1]. Despite intensive treatments including surgery, radiation, and chemotherapy, current therapy for ovarian cancer 1
To whom correspondence should be addressed. Fax: 82-2-3410-6829. E-mail: [email protected]. 0090-8258/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.
is often unsuccessful. Since cancer is a disease generated through dysregulated cell growth resulting from a process of genetic alterations, several recent studies have explored the potential application of gene therapy as an additional therapeutic method to improve the treatment of ovarian cancer [2– 4]. The basic concept of gene therapy is to replace a malfunctioning mutated gene with a normal wild-type gene or to express a therapeutic gene in target cells. A critical prerequisite for successful gene therapy is to determine a system for delivering the therapeutic gene into target cells with high efficiency and accuracy. In this regard, a recombinant adenovirus vector derived from subgroup C adenovirus is preferred to any other viral vector systems for cancer gene therapy, because it is highly infectious and is capable of delivering therapeutic genes to many different cell types. However, the susceptibility to adenoviral vector-mediated gene delivery is very low in some cell types and established cell lines [5–7]. Recently, several groups reported that the difference in the expression levels of the coxsackievirus and adenovirus receptor (CAR) on target cells was highly correlated with their sensitivities to adenoviral infection [8, 9]. CAR is a 46-kDa transmembrane protein that is composed of two extracellular immunoglobin-like domains [10]. Although the accurate cellular function of CAR has not been elucidated, CAR has been shown to function as a primary receptor for both coxsackie B virus and adenovirus. However, it is known that its cytoplasmic and transmembrane domains are not essential for coxsackievirus and adenovirus infection [11]. In this study, we demonstrate that the difference in susceptibility to adenovirus-mediated gene transfer among human ovarian cancer cell lines (SKOV3, 2774, PA-1, and OVCAR3) is correlated with the level of expression of CAR on the cell surface. We also demonstrate that ectopic expression of the CAR gene in these ovarian cancer cell lines by transient transfection dramatically increases their adenoviral sensitivities. These results indicate that the CAR molecule may be
260
TRANSFECTION OF CAR ENHANCES ADENOVIRAL SENSITIVITY
applicable as a useful tool to improve the efficiency of adenoviral vector-mediated gene therapy of human ovarian cancer. MATERIALS AND METHODS Cell Lines and Cultures Human ovarian cancer cell lines SKOV3, 2774, PA-1, and OVCAR3 were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum. Human embryonic kidney cell line 293 was maintained in Eagle’s minimal essential medium (EMEM) supplemented with 10% fetal bovine serum. All cell lines were obtained from American Type Culture Collection (ATCC, Rockville, MD). Adenoviral Vectors
261
formed using a -gal staining kit (Invitrogen, Carlsbad, CA) according to the manufacturer’s instruction. RT-PCR Analysis Total cellular RNA was extracted from each cell line using TRIzol (Gibco-BRL, Gaithersburg, MD). Two micrograms of total RNA was reverse-transcribed and total cDNA was subjected to a 50-l PCR using a CAR primer set (5⬘-ATCAGAAGGTGGATCAAGTG-3⬘ and 5⬘-TTACTGCCGATGTAGCTTCT-3⬘) or a GAPDH primer set (5⬘-TCGTGGAAGGACTCATGACC-3⬘ and 5⬘-TCCACCACCCTGTTGCTGTA-3⬘) by a RT-PCR kit (Promega, Madison, WI). The final PCR products were electrophoresed in 1.2% agarose gel and photographed under UV light. Transient Expression of CAR and Infection of Adenovirus
The adenovirus containing cDNA of green fluorescent protein (GFP) was constructed as described previously [12]. Briefly, the 0.8-kb cDNA of GFP was inserted at the HpaI site in the multiple cloning region of p␦ACMVpA to give rise to p␦ACMVGFP. The shuttle vector p␦ACMVGFP and an adenovirus vector pJM17 were used to cotransfect a packaging cell line, 293. One of the replication-deficient AdCMVGFP clones was isolated and propagated in 293 cells. Another recombinant adenovirus vector, AdRSVgal, a -galactosidase-expressing adenovirus, was provided by Wei Wei Zhang (GenStar Therapeutics Corp., San Diego, CA). Virus titers were determined by both plaque assay and optical observance at A 260 (1 A 260 unit ⫽ 10 12 particles/ml).
A full-length CAR cDNA was kindly provided by Dr. JhinGook Kim (Samsung Medical Center, Seoul, Korea). The CAR cDNA was cloned into pcDNA3.1 (Invitrogen) for mammalian expression with sense direction (pcDNA3.1-CAR). The nucleotide sequence of CAR cDNA was confirmed by automatic sequencing. For transient expression of the CAR gene, 8 ⫻ 10 5 cells of each cell line were seeded in 60-mm-diameter dishes 24 h before transfection. Cells were transfected with 5 g of empty vector or pcDNA3.1-CAR using FuGene6 reagent (Roche, Mennheim, Germany) according to the manufacturer’s protocol. After transfection for 24 h, cells were infected with Ad-gal or Ad-p53. And then, the infection efficiency was analyzed by Western blot analysis and -galactosidase activity assay.
FACS Analysis
Western Blot Analysis and -Galactosidase Activity Assay
To measure infection efficiency of AdCMVGFP in ovarian cancer cell lines, the cells were mock infected (PBS) or infected with 5 or 10 m.o.i. of AdCMVGFP. After infection for 24 hr, the cells were collected with trypsinization and washed twice with PBS. Expression of the GFP transgene product was assessed by flow cytometry. For analysis of the cell surface CAR molecule, the cells were washed twice with PBS and then incubated for 30 min at 4°C with RmcB monoclonal antibody (1:1000), obtained from Dr. Jeffrey M. Bergelson (Children’s Hospital of Philadelphia, Philadelphia, PA). Sequentially, the cells were washed with PBS and incubated for 30 min at 4°C with a secondary FITCconjugated goat anti-mouse antibody (1:50). For the negative control, only secondary FITC-conjugated goat anti-mouse antibody (Zymed, South San Francisco, CA) was used.
For Western blot analysis, cells were lysed in RIPA buffer [50 mM Tris (pH 8.0), 150 mM NaCl, 1% NP-40, 0.1% SDS, and 0.5% sodium deoxycholate]. Equivalent amounts of whole-cell extracts (10 g) were separated in 12% polyacrylamide gel and transferred to Hybond-ECL nitrocellulose filter paper (Amersham, Buckinghamshire, UK). Filters were blocked in 25 mM Tris (pH 8.0) containing 125 mM NaCl, 0.1% Tween 20, and 5% skim milk. Protein bands were probed with anti-p53 antibody (Novocastra, Newcastle on Tyne, UK) followed by labeling with horseradish peroxidase-conjugated anti-mouse antibody (Amersham). Bands were visualized using an ECL kit (Amersham) according to the manufacturer’s protocol. -Galactosidase activity assay was also performed using a -gal assay kit (Invitrogen) according to the manufacturer’s protocols.
-Galactosidase Staining -Galactosidase (-gal) staining was also used to measure the infection efficiency of Ad-RSVgal (Ad-gal) in four ovarian cancer cell lines. These cell lines were infected with 10 m.o.i. of Ad-gal. After infection for 24 h, cells were fixed with 0.5% glutaraldehyde and then -gal staining was per-
RESULTS Measurement of Sensitivity to Adenovirus Vectors in Ovarian Cancer Cell Lines In an attempt to examine the cellular mechanism responsible for the difference in the adenoviral sensitivity of human ovar-
262
KIM ET AL.
FIG. 1. Infection efficiency of AdCMVGFP in human ovarian cancer cell lines. Four ovarian cancer cell lines were treated with PBS (mock) or 5 or 10 m.o.i. of AdCMVGFP (Ad-GFP). After infection for 24 h, the cells were collected and the expression level of GFP transgene product in individual target cells was analyzed by flow cytometry.
ian cancer cell lines (SKOV3, 2774, PA-1, OVCAR3), we recently investigated the correlation between the infection efficiency of adenovirus and the expression level of the CAR gene using a recombinant adenovirus possessing the green fluorescent protein (GFP) gene, AdCMVGFP. Monolayers of individual ovarian cancer cells were mock infected (PBS) or infected with 5 or 10 m.o.i. of AdCMVGFP. The infection with AdCMVGFP appeared to have no significant effect on cell growth at the concentrations tested. To measure the infection efficiency of AdCMVGFP, the cells were collected by trypsinization 24 h after infection, and 10 5 cells were analyzed by flow cytometer. As shown in Fig. 1, both 2774 and PA-1 cells were highly susceptible to the infection of AdCMVGFP, whereas SKOV3 and OVCAR3 cells appeared to be relatively refractory to the adenoviral infection. These lower susceptibilities of SKOV-3 and OVCAR-3 cells to AdCMVGFP infection were consistently observed when these ovarian cancer cell lines were infected with 10 m.o.i. of another recombinant adenovirus, AdRSVgal, which expresses the -galactosidase gene (Fig. 2).
sion of the CAR gene. To test this hypothesis, we decided to examine the level of expression of CAR in four ovarian cancer cells. To quantify the expression level of CAR-specific mRNA in individual ovarian cancer cell lines (SKOV3, 2774, PA-1, and OVCAR3), RT-PCR was performed using total RNA
CAR Expression on Ovarian Cancer Cell Lines Since CAR is known to be the primary receptor molecule for the attachment of adenovirus to target cells and thus is believed to play a critical role in adenovirus entry into host cells [8, 9], it seems likely that the lower susceptibility of SKOV3 and OVCAR3 cells to the adenoviral infection compared with 2774 and PA-1 cells may reflect differences in the level of expres-
FIG. 2. Differences in infection efficiency of 〈dRSVgal in human ovarian cancer cell lines. Four ovarian cancer cell lines were infected with 10 m.o.i. of 〈dRSVgal. After infection for 24 h, the cells were fixed with 0.5% gluteraldehyde and the expression level of -galactosidase product was analyzed by -gal staining.
TRANSFECTION OF CAR ENHANCES ADENOVIRAL SENSITIVITY
263
cells, which showed an easily detectable level of the cell surface CAR molecule (Fig. 4). These results indicate that the low susceptibility of both SKOV3 and OVCAR3 cells to the recombinant adenoviral infection is possibly due to their low levels of expression of the CAR gene at the transcription level. Increased Infection Efficiency after Transient Expression of CAR
FIG. 3. Expression of CAR-specific mRNA in human ovarian cancer cell lines. Total RNA was extracted from four ovarian cancer cells. Expression of CAR in each cell line was analyzed by RT-PCR using either a CAR primer set or a GAPDH primer set. SM, 1-kb DNA ladder; (1) SKOV3; (2) 2774; (3) PA-1; (4) OVCAR3.
extracted from these cells. As shown in Fig. 3, the PCR product from CAR-specific mRNA was easily detectable in 2774 and PA-1 cells but was hardly detectable in SKOV3 and OVCAR3 cells. Under the same conditions, there was no difference in the level of the PCR product from GAPDH-specific mRNA. To confirm the results of RT-PCR analysis, the level of expression of the CAR molecule on the cell surface was also investigated by immunofluorescence assay using anti-CAR monoclonal antibody, RmcB. In accordance with the results of RT-PCR analysis which revealed the expression level of CAR mRNA, the CAR molecule was hardly detected on the surface of SKOV3 and OVCAR3 cells, as compared with 2774 and PA-1
To obtain direct evidence to support the belief that the low susceptibility of SKVO3 and OVCAR3 cells to the adenoviral infection is due to the low expression level of CAR, we examined whether the low susceptibility of these cells to the adenoviral infection can be enhanced by introducing the CAR gene into the cells. For this purpose, we intended to transiently introduce the CAR gene into four human ovarian cancer cell lines (SKOV3, 2774, PA-1, and OVCAR3) by transfection to improve CAR expression on the cell surface prior to AdRSVgal infection. After transfection with pcDNA3.1CAR for 24 h, the cells were infected with 50 m.o.i. of AdRSVgal. Subsequently, the cells were collected 24 h after AdRSVgal infection and then -galactosidase activity assay was performed. As shown in Fig. 5, the susceptibility of all ovarian cancer cells to AdRSVgal infection appeared to significantly increase after transfection with pcDNA3.1-CAR as compared with individual cells transfected with pcDNA3.1 vector only. This increase in adenoviral infection efficiency was more significant in SKOV3 and OVCAR3 cells than in 2774 and PA-1 cells. These results indicate that the level of
FIG. 4. Expression of surface CAR on human ovarian cancer cell lines. Individual cancer cells were analyzed by flow cytometry after the cells were treated with anti-RmcB antibody (1:1000 dilution) and then with FITC-labeled secondary antibody (1:50 dilution). For the negative control, only secondary antibody was used. (A) SKOV3, (B) 2774, (C) PA-1, (D) OVCAR-3.
264
KIM ET AL.
FIG. 5. Enhancement in the adenoviral sensitivity of human ovarian cancer cells after transient transfection of the CAR gene. Individual ovarian cancer cells were transfected with empty vector (䊐) or pcDNA3.1-CAR (■) before AdRSVgal infection (50 m.o.i.). After infection for 24 h, the cells were collected, and infection efficiency was determined by -galactosidase activity assay.
expression of CAR on target cells is directly associated with their susceptibility to adenoviral infection and this notion can be extended to human ovarian cancer cells. The results also demonstrate that transient introduction of the CAR gene into human ovarian cancer cells by transfection is sufficient to boost their susceptibility to adenoviral vector-based gene transfer. DISCUSSION A recombinant adenovirus derived from subgroup C is believed to be the most promising gene delivery system required for cancer gene therapy. However, the low sensitivity of the target cells to adenoviral vector-based gene transfer can be an obstacle to broad-range application of adenovirus-mediated gene delivery. Although a cellular mechanism of adenovirus infection has not been completely elucidated, it may be closely related to the coxsackievirus and adenovirus receptor (CAR) that binds adenovirus as the primary receptor [8, 9] and also to two integrins, ␣v3 and ␣v5, that facilitate the internalization of adenovirus as a secondary receptor [13–15]. These previous results suggest that the infection efficiency of adenovirus may be correlated mainly with the level of expression of CAR and the integrins on the target cells. In a previous study, we demonstrated that the adenovirusmediated delivery of wild-type p53 gene (AdCMVp53) into human ovarian cancer cell lines leads to growth arrest and apoptotic cell death [16]. For these studies, we employed four human ovarian cancer cell lines (SKOV3, 2774, PA-1, and OVCAR3) and found that the efficiencies of infection with AdCMVp53 of the individual cell lines were quite different. PA-1 and 2774 cells were more sensitive to AdCMVp53 than SKOV3 and OVCAR3 cells on treatment with the same dose of recombinant adenovirus. In this article, we analyzed CAR expression on these cell lines to reveal the cause of these differences in infection efficiency. Similar to their differences in susceptibility to AdCMVp53 [16], both 2774 and PA-1 cells
appeared to be more susceptible to infection with either AdCMVGFP or AdRSVgal than SKOV3 and OVCAR3 cells. These results suggest that the lower susceptibility of SKOV3 and OVCAR3 cells to the adenoviral infection compared with 2774 and PA-1 cells is not an artifact resulting from differences in the expression of the introduced genes, but is reliable. Two sequential experiments such as RT-PCR and immunofluorescence assay revealed that CAR-specific mRNA and cell surface expression of CAR were easily detectable in 2774 and PA-1 cells. However, under the same conditions these CAR expressions were undetectable or hardly detectable in SKOV3 and OVCAR3 cells, indicating that the failure to express the CAR gene at the transcription level in SKOV3 and OVCAR3 cells may lead to their low susceptibility to adenoviral infection. In addition, the low susceptibility of SKOV3 and OVCAR3 cells to AdRSVgal infection can be restored by transiently introducing the CAR gene through transfection prior to adenoviral infection. These results demonstrate that the loss of CAR expression in SKOV3 and OVCAR3 cells is the main cause of the lower susceptibility to adenoviral infection. Thus, transient expression of the CAR gene in SKOV3 and OVCAR3 cells is sufficient to boost their susceptibility to adenoviral infection. Recently, Kimura et al. showed that the infection with a recombinant adenovirus encoding CAR improves adenovirusmediated gene delivery to skeletal muscle [17]. Also, it has been reported that estradiol-induced enhancement of CAR expression in human ovarian cancer cell lines is paralleled by an increased infection efficiency of adenoviral vector [18]. More recently our study has demonstrated that the loss of CAR due to its downregulation at the transcription level is directly correlated with low efficiency of adenovirus infection in the cervical cancer cell line SiHa [12]. Transient expression of CAR in SiHa increases its sensitivity to adenovirus-mediated gene transfer. Although the physiological role of CAR in vivo still remains obscure, these previous results and our present results indicate that CAR expression on the cell surface is the most important prerequisite for efficient adenoviral vectorbased gene transfer into human ovarian cancer cells, and can thus be manipulated to enhance the susceptibility of cancer cells to adenoviral infection. These results also raise the possibility that as a novel strategy for successful adenovirusmediated gene therapy of human ovarian cancer, coexpression of the CAR gene and a therapeutic gene by an adenoviral vector may be employed together with a chemotherapeutic drug that can induce upregulation of CAR expression in target cells prior to adenoviral infection. However, we need to overcome many obstacles to apply this principle to human gene therapy. First, we should develop the proper methods to elevate CAR expression on ovarian cancer cells before adenovirusmediated gene delivery. There are three possible methods to increase CAR gene expression on target cells: viral-mediated CAR delivery, non-viral-mediated CAR delivery, and treatment with a chemotherapeutic drug that upregulates CAR expression.
TRANSFECTION OF CAR ENHANCES ADENOVIRAL SENSITIVITY
In a recent report, Okegawa et al. demonstrated that the expression of cell surface CAR molecule not only can facilitate the uptake of adenovirus but can also inhibit cell growth in bladder cancer, providing a clue to the physiological role of CAR in vivo [19]. Although more experiments are required to confirm the accurate function of CAR in vivo, these results indicate that CAR expression may be involved in virus attachment as well as cancer progression. In the future, we plan to investigate whether the susceptibility of primary ovarian cancer cells to the adenovirus-mediated delivery of wild-type p53 gene can be elevated by introducing CAR by employing animal models that have been established in our systems [20]. SUMMARY Our findings indicate that the infection efficiency of adenovirus differed in different ovarian cancer cell lines (SKOV3, 2774, PA-1, and OVCAR3) depending on their differences in expression of CAR. Moreover, ectopic expression of CAR significantly enhances sensitivity to adenoviral infection in ovarian cancer cell lines. These results indicate that CAR can be used to extend the efficiency of adenovirus-mediated gene delivery. ACKNOWLEDGMENTS
265
6. Nalbantoglu J, Pari G, Karpati G, Holland PC. Expression of the primary coxsackie and adenovirus receptor is downregulated during skeletal muscle maturation and limits the efficacy of adenovirus-mediated gene delivery to muscle cells. Hum Gene Ther 1999;10:1009 –19. 7. Watanabe T, Kelsey L, Ageitos A, Kuszynski C, Ino K, Heimann DG, Varney MT, Shepard HM, Vaillancourt MT, Maneval DC, Talmadge JE. Enhancement of adenovirus-mediated gene transfer to human bone marrow cells. Leuk Lymphoma 1998;29:439 –51. 8. Hemmi S, Geertsen R, Mezzacasa A, Peter I, Dummer R. The presence of human coxsackievirus and adenovirus receptor is associated with efficient adenovirus-mediated transgene expresssion in human melanoma cell cultures. Hum Gene Ther 1998;9:2363–73. 9. Li Y, Pong RC, Bergelson JM, Hall MC, Sagalowsky AI, Tseng CP, Wang Z, Hsieh JT. Loss adenoviral receptor expression in human bladder cancer cells: a potential impact on the efficacy of gene therapy. Cancer Res 1999;59:325–30. 10. Bowles KR, Gibson J, Wu J, Shaffer LG, Towbin JA, Bowles NE. Genomic organization and chromosomal localization of the human coxsackievirus B–adenovirus receptor gene. Hum Genet 1999;105:354 –9. 11. Wang X, Bergelson JM. Coxsackievirus and adenovirus receptor cytoplasmic and transmembrane domains are not essential for coxsakievirus and adenovirus infection. J Virol 1999;73:2559 – 62. 12. Kim JS, Lee SH, Cho YS, Kim YH, Lee JH. Ectopic expression of the coxsackievirus and adenovirus receptor increases susceptibility to adenoviral infection in the human cervical cancer cell line, SiHa. Biochem Biophys Res Commun 2001;288:240 – 4. 13. Goldman MJ, Wilson JM. Expression of ␣v3 integrin is necessary for efficient adenovirus-mediated gene transfer in the human airway. J Virol 1995;69:5951– 8.
We thank Dr. Sanghwa Yang (GenomicTree, Inc. Daejon, Korea) for reviewing this article and for his helpful discussion. This work was supported by G7 and SRC grants from the Korea Science and Engineering Foundation.
14. Takayama K, Ueno H, Pei XH, Nakanishi Y, Yatsunami J, Hara N. The levels of integrin ␣v5 may predict the susceptibility to adenovirusmediated gene transfer in human lung cancer cells. Gene Ther 1998;5: 361– 8.
REFERENCES
15. Wickham TJ, Mathias P, Cheresh DA, Nemerow GR. Integrin ␣v3 and ␣v5 promote adenovirus internalization but not virus attachement. Cell 1993;73:309 –19.
1. Lee SI. Annual Report of Cancer Registry Programme in the Republic of Korea, Jan. 1, 1994 –Dec. 31, 1994. Seoul: Ministry of Health and Welfare, 1996:3–33. 2. Casado E, Gomez-Navarro J, Yamamoto M, Adachi Y, Coolidge CJ, Arafat WO, Barker SD, Wang MH, Mahasreshti PJ, Hemminki A, Gonzalez-Baron M, Barnes MN, Pustilnik TB, Siegal GP, Alvarez RD, Curiel DT. Strategies to accomplish targeted expression of transgenes in ovarian cancer for molecular therapeutic applications. Clin Cancer Res 2001;7: 2496 –504. 3. Tsuruta Y, Mandai M, Konishi I, Kuroda H, Kusakari T, Yura Y, Hamid AA, Tamura I, Kariya M, Fujii S. Combination effect of adenovirusmediated pro-apoptotic bax gene transfer with cisplatin or paclitaxel treatment in ovarian cancer cell lines. Eur J Cancer 2001;37:531– 41. 4. Zang RY, Shi DR, Lu HJ, Cai SM, Lu DR, Zhang YJ, Qin HL. Adenovirus 5 E1a-mediated gene therapy for human ovarian cancer cells in vitro and in vivo. Int J Gynecol Cancer 2001;11:18 –23. 5. Goossens PH, Havenga MJ, Pieterman E, Lemckert AA, Breedveld FC, Bout A, Huizinga TW. Infection efficiency of type 5 adenoviral vectors in synovial tissue can be enhanced with a type 16 fiber. Arthritis Rheum 2001;44:570 –7.
16. Hwang ES, Kim J, Kim JS, Kao C, Ko SC, Chung L, Lee JH. The effects of the adenovirus-mediated wild-type p53 delivery in human epithelial ovarian cancer cell line in vitro and in vivo. Int J Gynecol Cancer 1998;8:27–36. 17. Kimura E, Maeda Y, Arima T, Nishida Y, Yamashita S, Hara A, Uyama E, Mita S, Uchino M. Efficient repetitive gene delivery to skeletal muscle using recombinant adenovirus vector containing the coxsackievirus and adenovirus receptor cDNA. Gene Ther 2001;8:20 –7. 18. You Z, Fischer DC, Tong X, Hasenburg A, Aguilar-Cordova E, Kieback DG. Coxsackievirus–adenovirus receptor expression in ovarian cancer cell lines is associated with increased adenovirus transduction efficiency and transgene expression. Cancer Gene Ther 2001;8:168 –75. 19. Okegawa T, Pong R-C, Li Y, Bergelson JM, Sagalowsky AI, Hsieh J-T. The mechanism of the growth-inhibitory effect of coxsackie and adenovirus receptor (CAR) on human bladder cancer: a functional analysis of CAR protein structure of CAR protein structure. Cancer Res 2001;61:6592– 600. 20. Kim J, Hwnag ES, Kim JS, You EH, Lee SH, Lee JH. Intraperitoneal gene therapy with adenoviral-mediated p53 tumor suppressor gene for ovarian cancer model in nude mouse. Cancer Gene Ther 1999;6:172– 8.
Enhancement of the Adenoviral Sensitivity of Human Ovarian Cancer Cells by Transient Expression of Coxsackievirus and Adenovirus Receptor (CAR) Jong-Sik Kim,* ,† Seung-Hoon Lee,‡ Yong-Suk Cho,‡ Jung-Joo Choi,‡ Young Ho Kim,† and Je-Ho Lee‡ ,1 *Clinical Research Center, Samsung Biomedical Research Institute, Seoul 135-710, Korea; ‡Molecular Therapy Research Center, Sungkyunkwan University, Samsung Medical Center Annex 8F, Ilwon-dong 50, Kangnam-ku, Seoul 135-710, Korea; and †Department of Microbiology, College of Natural Sciences, Kyungpook National University, Taegu, 702-701, Korea Received September 17, 2001
Objective. The coxsackievirus and adenovirus receptor (CAR) is known to be a primary receptor for attachment during adenovirus infection of target cells and thus is closely related to adenoviral infection efficiency. To extend this notion to human ovarian cancer, we investigated the difference in expression levels of CAR in human ovarian cancer cell lines and whether their adenoviral sensitivities are enhanced by transient transfection of the CAR gene. Methods. Adenoviral infection efficiency was examined by flow cytometry analysis, -galactosidase staining, and -galactosidase activity assay. Expression of the CAR-specific mRNA and protein was analyzed by reverse transcription polymerase chain reaction and flow cytometry, respectively. Results. Expression of CAR in human ovarian cancer cell lines (SKOV3, 2774, PA-1, and OVCAR3) appeared to be correlated with their susceptibilities to adenovirus-mediated gene delivery. The 2774 and PA-1 cells expressing an easily detectable level of CAR on the cell surface showed a higher susceptibility to infection with both AdCMVGFP and AdRSVgal than SKOV3 and OVCAR3 cells, both of which had hardly detectable levels of CAR. Ectopic expression of the CAR gene by transient transfection of these ovarian cancer cells resulted in a dramatic increase in their adenoviral sensitivities. Conclusion. These data show that the expression of CAR is closely related to susceptibility to adenovirus infection in human ovarian cancer cells. These results indicate that the CAR gene can be a useful tool in boosting the efficiency of adenoviral vector-mediated gene therapy against human ovarian cancer. © 2002 Elsevier Science (USA) Key Words: coxsackievirus and adenovirus receptor (CAR); adenovirus vector; infection efficiency; ovarian cancer cell lines.
INTRODUCTION Ovarian cancer is the most lethal gynecological cancer in Korea [1]. Despite intensive treatments including surgery, radiation, and chemotherapy, current therapy for ovarian cancer 1
To whom correspondence should be addressed. Fax: 82-2-3410-6829. E-mail: [email protected]. 0090-8258/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.
is often unsuccessful. Since cancer is a disease generated through dysregulated cell growth resulting from a process of genetic alterations, several recent studies have explored the potential application of gene therapy as an additional therapeutic method to improve the treatment of ovarian cancer [2– 4]. The basic concept of gene therapy is to replace a malfunctioning mutated gene with a normal wild-type gene or to express a therapeutic gene in target cells. A critical prerequisite for successful gene therapy is to determine a system for delivering the therapeutic gene into target cells with high efficiency and accuracy. In this regard, a recombinant adenovirus vector derived from subgroup C adenovirus is preferred to any other viral vector systems for cancer gene therapy, because it is highly infectious and is capable of delivering therapeutic genes to many different cell types. However, the susceptibility to adenoviral vector-mediated gene delivery is very low in some cell types and established cell lines [5–7]. Recently, several groups reported that the difference in the expression levels of the coxsackievirus and adenovirus receptor (CAR) on target cells was highly correlated with their sensitivities to adenoviral infection [8, 9]. CAR is a 46-kDa transmembrane protein that is composed of two extracellular immunoglobin-like domains [10]. Although the accurate cellular function of CAR has not been elucidated, CAR has been shown to function as a primary receptor for both coxsackie B virus and adenovirus. However, it is known that its cytoplasmic and transmembrane domains are not essential for coxsackievirus and adenovirus infection [11]. In this study, we demonstrate that the difference in susceptibility to adenovirus-mediated gene transfer among human ovarian cancer cell lines (SKOV3, 2774, PA-1, and OVCAR3) is correlated with the level of expression of CAR on the cell surface. We also demonstrate that ectopic expression of the CAR gene in these ovarian cancer cell lines by transient transfection dramatically increases their adenoviral sensitivities. These results indicate that the CAR molecule may be
260
TRANSFECTION OF CAR ENHANCES ADENOVIRAL SENSITIVITY
applicable as a useful tool to improve the efficiency of adenoviral vector-mediated gene therapy of human ovarian cancer. MATERIALS AND METHODS Cell Lines and Cultures Human ovarian cancer cell lines SKOV3, 2774, PA-1, and OVCAR3 were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum. Human embryonic kidney cell line 293 was maintained in Eagle’s minimal essential medium (EMEM) supplemented with 10% fetal bovine serum. All cell lines were obtained from American Type Culture Collection (ATCC, Rockville, MD). Adenoviral Vectors
261
formed using a -gal staining kit (Invitrogen, Carlsbad, CA) according to the manufacturer’s instruction. RT-PCR Analysis Total cellular RNA was extracted from each cell line using TRIzol (Gibco-BRL, Gaithersburg, MD). Two micrograms of total RNA was reverse-transcribed and total cDNA was subjected to a 50-l PCR using a CAR primer set (5⬘-ATCAGAAGGTGGATCAAGTG-3⬘ and 5⬘-TTACTGCCGATGTAGCTTCT-3⬘) or a GAPDH primer set (5⬘-TCGTGGAAGGACTCATGACC-3⬘ and 5⬘-TCCACCACCCTGTTGCTGTA-3⬘) by a RT-PCR kit (Promega, Madison, WI). The final PCR products were electrophoresed in 1.2% agarose gel and photographed under UV light. Transient Expression of CAR and Infection of Adenovirus
The adenovirus containing cDNA of green fluorescent protein (GFP) was constructed as described previously [12]. Briefly, the 0.8-kb cDNA of GFP was inserted at the HpaI site in the multiple cloning region of p␦ACMVpA to give rise to p␦ACMVGFP. The shuttle vector p␦ACMVGFP and an adenovirus vector pJM17 were used to cotransfect a packaging cell line, 293. One of the replication-deficient AdCMVGFP clones was isolated and propagated in 293 cells. Another recombinant adenovirus vector, AdRSVgal, a -galactosidase-expressing adenovirus, was provided by Wei Wei Zhang (GenStar Therapeutics Corp., San Diego, CA). Virus titers were determined by both plaque assay and optical observance at A 260 (1 A 260 unit ⫽ 10 12 particles/ml).
A full-length CAR cDNA was kindly provided by Dr. JhinGook Kim (Samsung Medical Center, Seoul, Korea). The CAR cDNA was cloned into pcDNA3.1 (Invitrogen) for mammalian expression with sense direction (pcDNA3.1-CAR). The nucleotide sequence of CAR cDNA was confirmed by automatic sequencing. For transient expression of the CAR gene, 8 ⫻ 10 5 cells of each cell line were seeded in 60-mm-diameter dishes 24 h before transfection. Cells were transfected with 5 g of empty vector or pcDNA3.1-CAR using FuGene6 reagent (Roche, Mennheim, Germany) according to the manufacturer’s protocol. After transfection for 24 h, cells were infected with Ad-gal or Ad-p53. And then, the infection efficiency was analyzed by Western blot analysis and -galactosidase activity assay.
FACS Analysis
Western Blot Analysis and -Galactosidase Activity Assay
To measure infection efficiency of AdCMVGFP in ovarian cancer cell lines, the cells were mock infected (PBS) or infected with 5 or 10 m.o.i. of AdCMVGFP. After infection for 24 hr, the cells were collected with trypsinization and washed twice with PBS. Expression of the GFP transgene product was assessed by flow cytometry. For analysis of the cell surface CAR molecule, the cells were washed twice with PBS and then incubated for 30 min at 4°C with RmcB monoclonal antibody (1:1000), obtained from Dr. Jeffrey M. Bergelson (Children’s Hospital of Philadelphia, Philadelphia, PA). Sequentially, the cells were washed with PBS and incubated for 30 min at 4°C with a secondary FITCconjugated goat anti-mouse antibody (1:50). For the negative control, only secondary FITC-conjugated goat anti-mouse antibody (Zymed, South San Francisco, CA) was used.
For Western blot analysis, cells were lysed in RIPA buffer [50 mM Tris (pH 8.0), 150 mM NaCl, 1% NP-40, 0.1% SDS, and 0.5% sodium deoxycholate]. Equivalent amounts of whole-cell extracts (10 g) were separated in 12% polyacrylamide gel and transferred to Hybond-ECL nitrocellulose filter paper (Amersham, Buckinghamshire, UK). Filters were blocked in 25 mM Tris (pH 8.0) containing 125 mM NaCl, 0.1% Tween 20, and 5% skim milk. Protein bands were probed with anti-p53 antibody (Novocastra, Newcastle on Tyne, UK) followed by labeling with horseradish peroxidase-conjugated anti-mouse antibody (Amersham). Bands were visualized using an ECL kit (Amersham) according to the manufacturer’s protocol. -Galactosidase activity assay was also performed using a -gal assay kit (Invitrogen) according to the manufacturer’s protocols.
-Galactosidase Staining -Galactosidase (-gal) staining was also used to measure the infection efficiency of Ad-RSVgal (Ad-gal) in four ovarian cancer cell lines. These cell lines were infected with 10 m.o.i. of Ad-gal. After infection for 24 h, cells were fixed with 0.5% glutaraldehyde and then -gal staining was per-
RESULTS Measurement of Sensitivity to Adenovirus Vectors in Ovarian Cancer Cell Lines In an attempt to examine the cellular mechanism responsible for the difference in the adenoviral sensitivity of human ovar-
262
KIM ET AL.
FIG. 1. Infection efficiency of AdCMVGFP in human ovarian cancer cell lines. Four ovarian cancer cell lines were treated with PBS (mock) or 5 or 10 m.o.i. of AdCMVGFP (Ad-GFP). After infection for 24 h, the cells were collected and the expression level of GFP transgene product in individual target cells was analyzed by flow cytometry.
ian cancer cell lines (SKOV3, 2774, PA-1, OVCAR3), we recently investigated the correlation between the infection efficiency of adenovirus and the expression level of the CAR gene using a recombinant adenovirus possessing the green fluorescent protein (GFP) gene, AdCMVGFP. Monolayers of individual ovarian cancer cells were mock infected (PBS) or infected with 5 or 10 m.o.i. of AdCMVGFP. The infection with AdCMVGFP appeared to have no significant effect on cell growth at the concentrations tested. To measure the infection efficiency of AdCMVGFP, the cells were collected by trypsinization 24 h after infection, and 10 5 cells were analyzed by flow cytometer. As shown in Fig. 1, both 2774 and PA-1 cells were highly susceptible to the infection of AdCMVGFP, whereas SKOV3 and OVCAR3 cells appeared to be relatively refractory to the adenoviral infection. These lower susceptibilities of SKOV-3 and OVCAR-3 cells to AdCMVGFP infection were consistently observed when these ovarian cancer cell lines were infected with 10 m.o.i. of another recombinant adenovirus, AdRSVgal, which expresses the -galactosidase gene (Fig. 2).
sion of the CAR gene. To test this hypothesis, we decided to examine the level of expression of CAR in four ovarian cancer cells. To quantify the expression level of CAR-specific mRNA in individual ovarian cancer cell lines (SKOV3, 2774, PA-1, and OVCAR3), RT-PCR was performed using total RNA
CAR Expression on Ovarian Cancer Cell Lines Since CAR is known to be the primary receptor molecule for the attachment of adenovirus to target cells and thus is believed to play a critical role in adenovirus entry into host cells [8, 9], it seems likely that the lower susceptibility of SKOV3 and OVCAR3 cells to the adenoviral infection compared with 2774 and PA-1 cells may reflect differences in the level of expres-
FIG. 2. Differences in infection efficiency of 〈dRSVgal in human ovarian cancer cell lines. Four ovarian cancer cell lines were infected with 10 m.o.i. of 〈dRSVgal. After infection for 24 h, the cells were fixed with 0.5% gluteraldehyde and the expression level of -galactosidase product was analyzed by -gal staining.
TRANSFECTION OF CAR ENHANCES ADENOVIRAL SENSITIVITY
263
cells, which showed an easily detectable level of the cell surface CAR molecule (Fig. 4). These results indicate that the low susceptibility of both SKOV3 and OVCAR3 cells to the recombinant adenoviral infection is possibly due to their low levels of expression of the CAR gene at the transcription level. Increased Infection Efficiency after Transient Expression of CAR
FIG. 3. Expression of CAR-specific mRNA in human ovarian cancer cell lines. Total RNA was extracted from four ovarian cancer cells. Expression of CAR in each cell line was analyzed by RT-PCR using either a CAR primer set or a GAPDH primer set. SM, 1-kb DNA ladder; (1) SKOV3; (2) 2774; (3) PA-1; (4) OVCAR3.
extracted from these cells. As shown in Fig. 3, the PCR product from CAR-specific mRNA was easily detectable in 2774 and PA-1 cells but was hardly detectable in SKOV3 and OVCAR3 cells. Under the same conditions, there was no difference in the level of the PCR product from GAPDH-specific mRNA. To confirm the results of RT-PCR analysis, the level of expression of the CAR molecule on the cell surface was also investigated by immunofluorescence assay using anti-CAR monoclonal antibody, RmcB. In accordance with the results of RT-PCR analysis which revealed the expression level of CAR mRNA, the CAR molecule was hardly detected on the surface of SKOV3 and OVCAR3 cells, as compared with 2774 and PA-1
To obtain direct evidence to support the belief that the low susceptibility of SKVO3 and OVCAR3 cells to the adenoviral infection is due to the low expression level of CAR, we examined whether the low susceptibility of these cells to the adenoviral infection can be enhanced by introducing the CAR gene into the cells. For this purpose, we intended to transiently introduce the CAR gene into four human ovarian cancer cell lines (SKOV3, 2774, PA-1, and OVCAR3) by transfection to improve CAR expression on the cell surface prior to AdRSVgal infection. After transfection with pcDNA3.1CAR for 24 h, the cells were infected with 50 m.o.i. of AdRSVgal. Subsequently, the cells were collected 24 h after AdRSVgal infection and then -galactosidase activity assay was performed. As shown in Fig. 5, the susceptibility of all ovarian cancer cells to AdRSVgal infection appeared to significantly increase after transfection with pcDNA3.1-CAR as compared with individual cells transfected with pcDNA3.1 vector only. This increase in adenoviral infection efficiency was more significant in SKOV3 and OVCAR3 cells than in 2774 and PA-1 cells. These results indicate that the level of
FIG. 4. Expression of surface CAR on human ovarian cancer cell lines. Individual cancer cells were analyzed by flow cytometry after the cells were treated with anti-RmcB antibody (1:1000 dilution) and then with FITC-labeled secondary antibody (1:50 dilution). For the negative control, only secondary antibody was used. (A) SKOV3, (B) 2774, (C) PA-1, (D) OVCAR-3.
264
KIM ET AL.
FIG. 5. Enhancement in the adenoviral sensitivity of human ovarian cancer cells after transient transfection of the CAR gene. Individual ovarian cancer cells were transfected with empty vector (䊐) or pcDNA3.1-CAR (■) before AdRSVgal infection (50 m.o.i.). After infection for 24 h, the cells were collected, and infection efficiency was determined by -galactosidase activity assay.
expression of CAR on target cells is directly associated with their susceptibility to adenoviral infection and this notion can be extended to human ovarian cancer cells. The results also demonstrate that transient introduction of the CAR gene into human ovarian cancer cells by transfection is sufficient to boost their susceptibility to adenoviral vector-based gene transfer. DISCUSSION A recombinant adenovirus derived from subgroup C is believed to be the most promising gene delivery system required for cancer gene therapy. However, the low sensitivity of the target cells to adenoviral vector-based gene transfer can be an obstacle to broad-range application of adenovirus-mediated gene delivery. Although a cellular mechanism of adenovirus infection has not been completely elucidated, it may be closely related to the coxsackievirus and adenovirus receptor (CAR) that binds adenovirus as the primary receptor [8, 9] and also to two integrins, ␣v3 and ␣v5, that facilitate the internalization of adenovirus as a secondary receptor [13–15]. These previous results suggest that the infection efficiency of adenovirus may be correlated mainly with the level of expression of CAR and the integrins on the target cells. In a previous study, we demonstrated that the adenovirusmediated delivery of wild-type p53 gene (AdCMVp53) into human ovarian cancer cell lines leads to growth arrest and apoptotic cell death [16]. For these studies, we employed four human ovarian cancer cell lines (SKOV3, 2774, PA-1, and OVCAR3) and found that the efficiencies of infection with AdCMVp53 of the individual cell lines were quite different. PA-1 and 2774 cells were more sensitive to AdCMVp53 than SKOV3 and OVCAR3 cells on treatment with the same dose of recombinant adenovirus. In this article, we analyzed CAR expression on these cell lines to reveal the cause of these differences in infection efficiency. Similar to their differences in susceptibility to AdCMVp53 [16], both 2774 and PA-1 cells
appeared to be more susceptible to infection with either AdCMVGFP or AdRSVgal than SKOV3 and OVCAR3 cells. These results suggest that the lower susceptibility of SKOV3 and OVCAR3 cells to the adenoviral infection compared with 2774 and PA-1 cells is not an artifact resulting from differences in the expression of the introduced genes, but is reliable. Two sequential experiments such as RT-PCR and immunofluorescence assay revealed that CAR-specific mRNA and cell surface expression of CAR were easily detectable in 2774 and PA-1 cells. However, under the same conditions these CAR expressions were undetectable or hardly detectable in SKOV3 and OVCAR3 cells, indicating that the failure to express the CAR gene at the transcription level in SKOV3 and OVCAR3 cells may lead to their low susceptibility to adenoviral infection. In addition, the low susceptibility of SKOV3 and OVCAR3 cells to AdRSVgal infection can be restored by transiently introducing the CAR gene through transfection prior to adenoviral infection. These results demonstrate that the loss of CAR expression in SKOV3 and OVCAR3 cells is the main cause of the lower susceptibility to adenoviral infection. Thus, transient expression of the CAR gene in SKOV3 and OVCAR3 cells is sufficient to boost their susceptibility to adenoviral infection. Recently, Kimura et al. showed that the infection with a recombinant adenovirus encoding CAR improves adenovirusmediated gene delivery to skeletal muscle [17]. Also, it has been reported that estradiol-induced enhancement of CAR expression in human ovarian cancer cell lines is paralleled by an increased infection efficiency of adenoviral vector [18]. More recently our study has demonstrated that the loss of CAR due to its downregulation at the transcription level is directly correlated with low efficiency of adenovirus infection in the cervical cancer cell line SiHa [12]. Transient expression of CAR in SiHa increases its sensitivity to adenovirus-mediated gene transfer. Although the physiological role of CAR in vivo still remains obscure, these previous results and our present results indicate that CAR expression on the cell surface is the most important prerequisite for efficient adenoviral vectorbased gene transfer into human ovarian cancer cells, and can thus be manipulated to enhance the susceptibility of cancer cells to adenoviral infection. These results also raise the possibility that as a novel strategy for successful adenovirusmediated gene therapy of human ovarian cancer, coexpression of the CAR gene and a therapeutic gene by an adenoviral vector may be employed together with a chemotherapeutic drug that can induce upregulation of CAR expression in target cells prior to adenoviral infection. However, we need to overcome many obstacles to apply this principle to human gene therapy. First, we should develop the proper methods to elevate CAR expression on ovarian cancer cells before adenovirusmediated gene delivery. There are three possible methods to increase CAR gene expression on target cells: viral-mediated CAR delivery, non-viral-mediated CAR delivery, and treatment with a chemotherapeutic drug that upregulates CAR expression.
TRANSFECTION OF CAR ENHANCES ADENOVIRAL SENSITIVITY
In a recent report, Okegawa et al. demonstrated that the expression of cell surface CAR molecule not only can facilitate the uptake of adenovirus but can also inhibit cell growth in bladder cancer, providing a clue to the physiological role of CAR in vivo [19]. Although more experiments are required to confirm the accurate function of CAR in vivo, these results indicate that CAR expression may be involved in virus attachment as well as cancer progression. In the future, we plan to investigate whether the susceptibility of primary ovarian cancer cells to the adenovirus-mediated delivery of wild-type p53 gene can be elevated by introducing CAR by employing animal models that have been established in our systems [20]. SUMMARY Our findings indicate that the infection efficiency of adenovirus differed in different ovarian cancer cell lines (SKOV3, 2774, PA-1, and OVCAR3) depending on their differences in expression of CAR. Moreover, ectopic expression of CAR significantly enhances sensitivity to adenoviral infection in ovarian cancer cell lines. These results indicate that CAR can be used to extend the efficiency of adenovirus-mediated gene delivery. ACKNOWLEDGMENTS
265
6. Nalbantoglu J, Pari G, Karpati G, Holland PC. Expression of the primary coxsackie and adenovirus receptor is downregulated during skeletal muscle maturation and limits the efficacy of adenovirus-mediated gene delivery to muscle cells. Hum Gene Ther 1999;10:1009 –19. 7. Watanabe T, Kelsey L, Ageitos A, Kuszynski C, Ino K, Heimann DG, Varney MT, Shepard HM, Vaillancourt MT, Maneval DC, Talmadge JE. Enhancement of adenovirus-mediated gene transfer to human bone marrow cells. Leuk Lymphoma 1998;29:439 –51. 8. Hemmi S, Geertsen R, Mezzacasa A, Peter I, Dummer R. The presence of human coxsackievirus and adenovirus receptor is associated with efficient adenovirus-mediated transgene expresssion in human melanoma cell cultures. Hum Gene Ther 1998;9:2363–73. 9. Li Y, Pong RC, Bergelson JM, Hall MC, Sagalowsky AI, Tseng CP, Wang Z, Hsieh JT. Loss adenoviral receptor expression in human bladder cancer cells: a potential impact on the efficacy of gene therapy. Cancer Res 1999;59:325–30. 10. Bowles KR, Gibson J, Wu J, Shaffer LG, Towbin JA, Bowles NE. Genomic organization and chromosomal localization of the human coxsackievirus B–adenovirus receptor gene. Hum Genet 1999;105:354 –9. 11. Wang X, Bergelson JM. Coxsackievirus and adenovirus receptor cytoplasmic and transmembrane domains are not essential for coxsakievirus and adenovirus infection. J Virol 1999;73:2559 – 62. 12. Kim JS, Lee SH, Cho YS, Kim YH, Lee JH. Ectopic expression of the coxsackievirus and adenovirus receptor increases susceptibility to adenoviral infection in the human cervical cancer cell line, SiHa. Biochem Biophys Res Commun 2001;288:240 – 4. 13. Goldman MJ, Wilson JM. Expression of ␣v3 integrin is necessary for efficient adenovirus-mediated gene transfer in the human airway. J Virol 1995;69:5951– 8.
We thank Dr. Sanghwa Yang (GenomicTree, Inc. Daejon, Korea) for reviewing this article and for his helpful discussion. This work was supported by G7 and SRC grants from the Korea Science and Engineering Foundation.
14. Takayama K, Ueno H, Pei XH, Nakanishi Y, Yatsunami J, Hara N. The levels of integrin ␣v5 may predict the susceptibility to adenovirusmediated gene transfer in human lung cancer cells. Gene Ther 1998;5: 361– 8.
REFERENCES
15. Wickham TJ, Mathias P, Cheresh DA, Nemerow GR. Integrin ␣v3 and ␣v5 promote adenovirus internalization but not virus attachement. Cell 1993;73:309 –19.
1. Lee SI. Annual Report of Cancer Registry Programme in the Republic of Korea, Jan. 1, 1994 –Dec. 31, 1994. Seoul: Ministry of Health and Welfare, 1996:3–33. 2. Casado E, Gomez-Navarro J, Yamamoto M, Adachi Y, Coolidge CJ, Arafat WO, Barker SD, Wang MH, Mahasreshti PJ, Hemminki A, Gonzalez-Baron M, Barnes MN, Pustilnik TB, Siegal GP, Alvarez RD, Curiel DT. Strategies to accomplish targeted expression of transgenes in ovarian cancer for molecular therapeutic applications. Clin Cancer Res 2001;7: 2496 –504. 3. Tsuruta Y, Mandai M, Konishi I, Kuroda H, Kusakari T, Yura Y, Hamid AA, Tamura I, Kariya M, Fujii S. Combination effect of adenovirusmediated pro-apoptotic bax gene transfer with cisplatin or paclitaxel treatment in ovarian cancer cell lines. Eur J Cancer 2001;37:531– 41. 4. Zang RY, Shi DR, Lu HJ, Cai SM, Lu DR, Zhang YJ, Qin HL. Adenovirus 5 E1a-mediated gene therapy for human ovarian cancer cells in vitro and in vivo. Int J Gynecol Cancer 2001;11:18 –23. 5. Goossens PH, Havenga MJ, Pieterman E, Lemckert AA, Breedveld FC, Bout A, Huizinga TW. Infection efficiency of type 5 adenoviral vectors in synovial tissue can be enhanced with a type 16 fiber. Arthritis Rheum 2001;44:570 –7.
16. Hwang ES, Kim J, Kim JS, Kao C, Ko SC, Chung L, Lee JH. The effects of the adenovirus-mediated wild-type p53 delivery in human epithelial ovarian cancer cell line in vitro and in vivo. Int J Gynecol Cancer 1998;8:27–36. 17. Kimura E, Maeda Y, Arima T, Nishida Y, Yamashita S, Hara A, Uyama E, Mita S, Uchino M. Efficient repetitive gene delivery to skeletal muscle using recombinant adenovirus vector containing the coxsackievirus and adenovirus receptor cDNA. Gene Ther 2001;8:20 –7. 18. You Z, Fischer DC, Tong X, Hasenburg A, Aguilar-Cordova E, Kieback DG. Coxsackievirus–adenovirus receptor expression in ovarian cancer cell lines is associated with increased adenovirus transduction efficiency and transgene expression. Cancer Gene Ther 2001;8:168 –75. 19. Okegawa T, Pong R-C, Li Y, Bergelson JM, Sagalowsky AI, Hsieh J-T. The mechanism of the growth-inhibitory effect of coxsackie and adenovirus receptor (CAR) on human bladder cancer: a functional analysis of CAR protein structure of CAR protein structure. Cancer Res 2001;61:6592– 600. 20. Kim J, Hwnag ES, Kim JS, You EH, Lee SH, Lee JH. Intraperitoneal gene therapy with adenoviral-mediated p53 tumor suppressor gene for ovarian cancer model in nude mouse. Cancer Gene Ther 1999;6:172– 8.