Targeted killing of carcinoembryonic antigen (CEA)- producing ...

© 2000 Nature America, Inc. 0929-1903/00/$15.00/⫹0 www.nature.com/cgt

Targeted killing of carcinoembryonic antigen (CEA)producing cholangiocarcinoma cells by polyamidoamine dendrimer-mediated transfer of an Epstein-Barr virus (EBV)-based plasmid vector carrying the CEA promoter Saiyu Tanaka,1 Masaki Iwai,1 Yoshinori Harada,1 Teruhisa Morikawa,1 Akira Muramatsu,1 Takahiro Mori,1 Takeshi Okanoue,1 Kei Kashima,1 Hiroko Maruyama-Tabata,2,3 Hideyo Hirai,3 Etsuko Satoh,3 Jiro Imanishi,3 and Osam Mazda3 1

Third Department of Internal Medicine, and Departments of 2Pediatrics and 3Microbiology, Kyoto Prefectural University of Medicine, Kamikyo, Kyoto, Japan.

The present study reports a novel nonviral method to efficiently and specifically target carcinoembryonic antigen (CEA)-producing cholangiocarcinoma (CC) cells in vitro. Epstein-Barr virus (EBV)-based and conventional plasmid vectors were constructed that possess the ␤-galactosidase (␤-gal) or herpes simplex virus-1 (HSV-1) thymidine kinase (Tk) genes as well as tandem repeats of the human genomic sequence ⫺82 to ⫺42 bp from the transcriptional start site of the CEA gene. The plasmids were transfected by means of polyamidoamine dendrimer into CEA-positive (HuCC-T1) or -negative cell lines. Transfection of the conventional plasmid vector with the CEA promoter and ␤-gal gene resulted in a very low or undetectable level of marker gene expression even in the CEA-positive cell line. Transferring the HSV-1 Tk gene by conventional plasmid did not affect the susceptibility of HuCC-T1 cells to ganciclovir. In marked contrast, strong ␤-gal expression was specifically obtained in HuCC-T1 cells by transfecting the EBV-based plasmid in which the CEA promoter and a ubiquitous promoter (SR␣) are employed to drive the EBV-encoded nuclear antigen 1 (EBNA1) and ␤-gal genes, respectively (pTES.␤). Furthermore, CEA-positive but not -negative tumor cells were rendered highly susceptible to ganciclovir when transfected with the EBV-based vector that carries the CEA promoter-EBNA1 and SR␣-HSV-1 Tk genes (pTES.Tk). These results strongly suggest that the EBV-based plasmid vector/cationic polymer system (EBV/polyplex) equipped with the CEA promoter provides an efficient nonviral method for the targeted gene therapy of CEA-producing malignancies. Cancer Gene Therapy (2000) 7, 1241–1249

Key words: Epstein-Barr virus-based vector; carcinoembryonic antigen promoter; polyamidoamine dendrimer; EBV-encoded nuclear antigen 1; cholangiocarcinoma; gene therapy.

T

he prognosis of cholangiocarcinoma (CC) is poor. Surgery is the only effective therapy, but in most cases, CC is diagnosed too late to be completely resected. Only 15–20% of patients have resectable tumors.1,2 Patients with unresectable tumor do not survive 5 years after surgery, and their median survival is 8 months.3 There is no effective chemotherapy or radiation therapy, and little progress has been made in the treatment of this cancer. A new strategy is needed to treat advanced CC, and gene therapy is an attractive choice. For effective and safe suicide gene therapy, it is necessary to induce strong, tumor-specific gene expression. To this end, the use of the tumor-specific promotReceived December 30, 1999; accepted April 15, 2000. Address correspondence and reprint requests to Dr. Osam Mazda, Department of Microbiology, Kyoto Prefectural University of Medicine, Kamikyo, Kyoto 602-8566, Japan. E-mail address: [email protected]

Cancer Gene Therapy, Vol 7, No 9, 2000: pp 1241–1249

ers, such as ␣-fetoprotein and carcinoembryonic antigen (CEA) promoters, shows promise. The ␣-fetoprotein promoter has been used successfully to target hepatocellular carcinoma (HCC) in combination with adenovirus or retrovirus vectors.4 –7 CEA is a widely used tumor marker, especially for cancers in the gastrointestinal tract. It is a 180-kDa glycoprotein with an oncofetal expression pattern.8 In some earlier studies, the CEA promoter was employed to target CEA-producing tumors.9 –13 Most of these studies employed a DNA fragment corresponding to the CEA gene upstream regulatory regions originally reported by Schrewe et al14 (⫺316 to ⫹107 from the transcriptional start site). Richards et al9 reported that the sequence between ⫺41 and ⫺18 was essential for transcription from the CEA promoter. Multimerization of the sequence ⫺89 to ⫺40 resulted in an increase in both the expression level and specificity for CEA-positive cells in a copy number-dependent manner. Tumor1241

1242

TANAKA, IWAI, HARADA, ET AL: NONVIRAL GENE TRANSFER TO CEA-POSITIVE TUMOR CELLS

specific gene expression was achieved with retrovirus or adenovirus vectors.9,15,16 In the case of nonviral vectors, however, transfection with plasmid vectors carrying the CEA promoter did not result in sufficient therapeutic outcome.17,18 Nonviral vector systems may provide safe gene delivery due to their low immunogenicity and cell toxicity. However, the nonviral systems devised to date have generally been insufficient in terms of intensity and longevity of gene expression. This may be the reason why nonviral systems engaging the CEA promoter were not successful in eliciting a therapeutic outcome specific for CEA-producing tumor cells. We have demonstrated previously that extremely efficient transfection can be achieved by using EpsteinBarr virus (EBV)-based plasmid vector.19 –23 The EBVbased plasmid vector carries the EBV-encoded nuclear antigen 1 (EBNA1) gene and origin of replication (oriP) region from the EBV genome.24,25 Through binding to oriP, EBNA1 facilitates nuclear localization, binding to the nuclear matrix, and replication of the plasmid DNA. Another function of EBNA1 is transcriptional up-regulation, and these functions make this vector highly efficient.26 –30 No infectious virus particle can be produced using this vector. Recently, we have reported that use of the EBV-based plasmid vector combined with polyamidoamine dendrimer (PAAD) (EBV/polyplex) resulted in a markedly high expression in human HCC and Ewing’s sarcoma cell lines, suggesting the potential usefulness of the EBV/polyplex system for gene therapy.31,32 However, no study has been reported in which a tumor-specific promoter was employed in the EBVbased plasmid vector to target tumor cells. In this study, we combined the multimerized CEA promoter with the EBV/polyplex system to establish an efficient nonviral strategy to target CEA-producing cancers. MATERIALS AND METHODS

Plasmid vectors The CEA promoter was constructed according to the published sequence.14 The upstream region ⫺89 to ⫹77 bp from the transcriptional start site of the CEA gene was amplified by polymerase chain reaction with a pair of primers (sense primer, 5⬘-TATGATATCAGGGGAGGGGACAGAGGACA; antisense primer, 5⬘-GGGATCGATCGAGAAGAGCTTGAGTTCCA). The resultant polymerase chain reaction product was inserted into the pBluescript II phargemid (Stratagene, La Jolla, Calif). Tetraplicated CEA promoter was constructed by a standard molecular biology technique (Fig 1). Four plasmid vectors, pSES.␤, pS.␤, pSES.Tk, and pS.Tk (Figs 2 and 3), have been described previously.21,31 The other plasmid vectors, pT.␤, pTES.␤, pSET.␤, pTET.␤, pT.Tk, and pTES.Tk (Figs 2 and 3), were constructed by substituting the tetraplicated CEA promoter for the SR␣ promoter in pS.␤, pSES.␤, pS.Tk, or pSES.Tk.

Cell lines and cell culture HuCC-T1 and SSP-25 cells were purchased from the Health Science Research Resources Bank (Osaka, Japan) and Riken

Figure 1. a: 5⬘ regulatory region of human CEA gene. b: Structure of the CEA promoter used in this study. Quadrupled sequences of the enhancer region (⫺82 to ⫺42 from the transcriptional start site) (filled box) and a single minimum promoter region (⫺42 to ⫹69) are included.

Cell Bank (Tsukuba, Japan), respectively. These cells were maintained in RPMI 1640 medium (Life Technologies, Grand Island, NY) supplemented with 10% fetal bovine sera, 2 mM L-glutamine, 100 U/mL penicillin, and 100 ␮g/mL streptomycin. A4573 cells were maintained in RPMI 1640 medium supplemented with 10% fetal bovine sera, 2 mM L-glutamine, 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, 100 U/mL penicillin, and 100 ␮g/mL streptomycin.

CEA production assay Cells were plated at a density of 2 ⫻ 105 cells/well in 24-well cell culture plates (Corning, Corning, NY); after 48 hours of cultivation at 37°C in 5% CO2/95% humidified air, the culture medium was renewed. Seven days later, the supernatant was collected and the CEA concentration was measured by solid phase radioimmunoassay using a CEA RIABEAD kit (Dainabot, Tokyo, Japan). CEA expression was standardized based on the protein content of cells measured by a protein assay kit.33

Transfection by PAAD Cells were seeded into wells of 6-well plates (Becton Dickinson Labware, Franklin Lakes, NJ) at a density of 1 ⫻ 105 (HuCCT1) or 3 ⫻ 105 (SSP-25) cells/well and incubated at 37°C in 5% CO2/95% humidified air. On the following day, 2 ␮g of DNA was mixed with 30 ␮g of PAAD (generation 5) (Qiagen, Hilden, Germany) in RPMI 1640 and incubated for 10 minutes at room temperature. The complex was added to 600 ␮L of complete medium and immediately transferred to cells that had been washed with phosphate-buffered saline (PBS). The cells were incubated at 37°C in 5% CO2/95% humidified air for 2 hours, and the medium was renewed.

5-bromo-4-chloro-3-indolyl ␤-D-galactoside (X-Gal) staining Cells were fixed with 1% glutaraldehyde/PBS for 10 minutes, washed three times with PBS, and subsequently incubated for 3 hours at 37°C in X-Gal staining solution (0.05% (vol/vol) X-Gal (Nacalai Tesque, Kyoto, Japan), 1 mM MgCl2, 150 mM NaCl, 3 mM K4 [Fe(CN)6], 3 mM K3[Fe(CN)6], 60 mM Na2HPO4, 40 mM NaH2PO4, and 0.1% Triton X-100). The reaction was terminated by replacing the solution with 1 mM ethylenediaminetetraacetic acid/PBS.

Cancer Gene Therapy, Vol 7, No 9, 2000


1243

Figure 2. Schematic diagram of ␤-gal expression vectors. Maps of pSES.␤, pS.␤, pT.␤, pTES.␤, pSET.␤, and pTET.␤ are shown. Amp, ampicillin-resistant gene; ␤-gal, Escherichia coli ␤-gal gene; EBNA1, EBNA1 gene; polyA, poly(A) additional signal; SR␣, SR␣ promoter; CEA, tetraplicated CEA promoter.

Figure 3. Schematic diagram of HSV-1 Tk expression vectors. Maps of pSES.Tk, pS.Tk, pT.Tk, and pTES.Tk are shown. Amp, ampicillin-resistant gene; EBNA1, EBNA1 gene; polyA, poly(A) additional signal; SR␣, SR␣ promoter; CEA, tetraplicated CEA promoter; Tk, HSV-1 Tk gene.


1244


Table 1. CEA production Cell lines

CEA secretion* (mean ⫾ SD)

HuCC-T1 SSP-25 A4573

7.0 ⫾ 0.7 BT† BT

*Nanograms of CEA/mg protein/7 days. †BT, below the threshold of 0.5 ng of CEA.

10,000 ␮M. After further incubation in 5% CO2/95% humidified air at 37°C for 120 hours, Alamar blue (Alamar Biosciences, Sacramento, Calif) was added according to the manufacturer’s protocol. The cells were further cultured for 4 hours, and the OD of each well was measured with a microplate reader at test and reference wavelengths of 570 and 600 nm, respectively.

RESULTS

␤-galactosidase (␤-gal) assay Cells were scraped off dishes, washed twice with PBS, and resuspended in 50 ␮L of tris(hydroxymethyl)aminomethaneHCl (pH 7.8). After two cycles of freezing and thawing, the lysate was centrifuged at 15,000 rpm for 5 minutes. The ␤-gal activity in the supernatant fraction was assayed with a ␤-gal assay kit (Invitrogen, San Diego, Calif) according to the manufacturer’s protocol. The optical density (OD) was measured at 420 nm. Activity was calculated using the following formula: ␤-gal units ⫽ ([OD420 ⫻ 380]/30)/mg protein, where 380 is a conversion factor and 30 is the time of incubation in minutes. The ␤-gal activity was standardized based on the amount of protein in the supernatant, as determined with a protein assay kit.33

Alamar blue assay The susceptibility of cells to ganciclovir (GCV) was examined by Alamar blue assay as described previously.34,35 Briefly, quadruplicate aliquots of cells were seeded in 96-well, flatbottom microtiter plates (Falcon, Lincoln Park, NJ) (5 ⫻ 103 cells in 200 ␮L of complete medium per well). After 24 hours, GCV was added at various concentrations ranging from 0 to

Production of CEA protein by tumor cell lines In the present study, we employed two CC cell lines, HuCC-T1 and SSP-25, as well as a Ewing’s sarcoma cell line, A4573. Earlier reports have shown that the transcription level of the CEA gene correlates with the CEA protein level36 –38 and that most of the CEA protein is secreted into medium.9 Therefore, we measured the CEA content in culture media. CEA was detected in the media of HuCC-T1 cells but not in the media of SSP-25 and A4573 cells (Table 1). The results are consistent with those published earlier.39

Comparison of transfection efficiencies with pSES.␤ and pS.␤ To test the effectiveness of the plasmid vectors carrying EBNA1 and oriP, we transfected the three cell lines with pSES.␤, a ␤-gal expression vector containing EBNA1 and oriP, or pS.␤, a conventional plasmid vector (Fig 2). A total of 2 ␮g of DNA was combined with 30 ␮g of PAAD to transfect the cells in each well of a 6-well plate.

Figure 4. ␤-gal expression in tumor cell lines transfected with pSES.␤ or pS.␤ /PAAD. HuCC-T1 (1 ⫻ 105 cells), SSP-25 (3 ⫻ 105 cells), and A4573 (3 ⫻ 105 cells) cells were plated on 6-well plates. After 24 hours, the cells were incubated for 2 hours with 2 ␮g of pSES.␤ (f) or pS.␤ (䡺) combined with 30 ␮g of PAAD. After 3 days, ␤-gal activities were measured. The data represent the mean ⫾ SE of triplicate determinations.



1245

Figure 5. X-Gal staining of transfected cells. HuCC-T1 (a,d), SSP-25 (b,e), or A4573 (c,f) cells were transfected with pSES.␤/PAAD (a– c) or pTES.␤/PAAD (d–f) and stained with X-Gal on day 3 posttransfection (original magnification is ⫻25).

On day 3 posttransfection, the ␤-gal activities were measured. As shown in Figure 4, the ␤-gal activities provided by pSES.␤ /PAAD were 3.4 to 13.4 times higher than those provided by pS.␤ /PAAD in all of the cell lines. X-Gal staining also revealed that HuCC-T1, SSP-25, and A4573 cells transfected with pSES.␤ /PAAD showed high-level ␤-gal expression. (Representative data are shown in Fig 5, a– c.) The data are compatible with our previous studies demonstrating the superiority of the EBV plasmid to conventional plasmid vector in various tumor as well as primary cells.23,31,32

Cell type-specific expression of the ␤-gal gene Because pSES.␤ has a strong universal promoter (SR␣), ␤-gal was strongly expressed in both CEA-positive and -negative cell lines transfected with pSES.␤. To target CEA-producing cells, we constructed three EBV-based plasmids, pTES.␤, pSET.␤, and pTET.␤, that contain the CEA promoter upstream of the EBNA1 gene, ␤-gal gene, or both (Fig 2). A conventional plasmid vector that


possesses the ␤-gal gene under the control of the CEA promoter (pT.␤) was also constructed. We tested the efficiency and specificity of these plasmids by transfecting them via PAAD into HuCC-T1, SSP-25, and A4573 cells. pT.␤ did not provide high ␤-gal activity in the cell lines tested (Fig 6d). This indicates that a conventional plasmid vector with CEA promoter is insufficient to induce high-level marker gene expression even in CEA-positive cells. When pSET.␤ was transfected, the ␤-gal activity in HuCC-T1 was about one-seventh of that given by pSES.␤ (Fig 6a). A considerable level of ␤-gal activity was also detected in A4573, indicating that pSET.␤ was lacking not only in activity but also in specificity. The ␤-gal activities provided by pTET.␤ were very weak in both CEA-positive and -negative cell lines (Fig 6c). In contrast, when pTES.␤ was transfected, both SSP-25 and A4573 cells exhibited very low levels of ␤-gal activity, whereas HuCC-T1 had strong enzyme activity (Fig 6b). The X-Gal staining also confirmed the strong expression of ␤-gal exclusively in the CEA-producing tumor cells (Fig 5, d-f). These

1246


Figure 7. Kinetics of ␤-gal expression given by pTES.␤. HuCC-T1 cells (1 ⫻ 105 cells) were transfected with pTES.␤ as described in the legend to Figure 4. ␤-gal activities were measured on the indicated days after the transfection. Means ⫾ SE of triplicate determinations are shown.

Figure 6. ␤-gal is specifically expressed in CEA-positive cells upon pTES.␤ transfection. HuCC-T1 (1 ⫻ 105 cells) (f), SSP-25 (3 ⫻ 105 cells) (u), or A4573 (3 ⫻ 105 cells) (䡺) cells were transfected with pSET.␤ (a), pTES.␤ (b), pTET.␤ (c), or pT.␤ (d) as described in the legend to Figure 4. After 3 days, ␤-gal activities were measured. Means ⫾ SE of triplicate determinations are shown.

results indicate the usefulness of pTES.␤ as a plasmid for tumor-specific transfection. We examined the kinetics of the ␤-gal activity in HuCC-T1 cells provided by pTES.␤ transfection. Strong ␤-gal activity was observed from 2 to 7 days after transfection (Fig 7). After day 7, the expression gradually diminished, but even on day 21 posttransfection, ␤-gal activity was detectable.

Targeted killing of CEA-producing cells by pTES.Tk To investigate whether CEA-positive tumor cells can be specifically killed by a nonviral vector system, four plasmid vectors carrying the herpes simplex virus-1 (HSV-1) thymidine kinase (Tk) gene were constructed (pSES.Tk, pS.Tk, pT.Tk, pTES.Tk) (Fig 3). They were transfected into HuCC-T1 or SSP-25 cells by means of

PAAD. At 72 hours posttransfection, cells were cultured in the presence of various concentrations of GCV, and 5 days later, their viabilities were measured by Alamar blue assay. Both cell lines transfected with pSES.Tk were rendered highly susceptible to GCV compared with mock-transfected cells (Fig 8, a and e). The transfection with pS.Tk moderately affected the susceptibilities of these cell lines to GCV (Fig 8b). These observations with pSES.Tk and pS.Tk are consistent with our previous results using HCC and Ewing’s sarcoma cell lines.31,32 The pT.Tk/PAAD did not significantly affect the susceptibility of either cell line to GCV; the susceptibility curves were similar to those of mock-transfected cells (Fig 8, c and e). This observation is compatible with the results for pT.␤ (Fig 6d). Compared with pS.Tk, the suicide effect provided by pT.Tk was weaker, indicating that the promoter activity of the CEA promoter is weaker than that of SR␣ (Figs 4 and 6). Therefore, a therapeutic effect cannot be expected with this plasmid vector. In marked contrast, the GCV sensitivity of HuCC-T1 was sufficiently elevated by pTES.Tk, whereas that of SSP-25 was not altered by the same plasmid (Fig 8d). The percent cell survival curve of pTES.Tk-transfected SSP-25 cells was similar to that of the mocktransfected cells (Fig 8, d and e), whereas HuCC-T1 cells transfected with pTES.Tk exhibited nearly the same percent cell survival curve as pSES.Tk-transfected HuCC-T1 cells (Fig 8, a and d). This finding indicates that pTES.Tk specifically affects CEA-positive cells, leaving the CEA-negative cells unaffected. In the aspects of clinical usage, the killing effects of these vectors were tested by culturing cells in the presence of 10 and 30 ␮M GCV after the transfection. It has been reported that patients systemically administered this prodrug exhibit a serum concentration of 2.7 to 22.3 ␮M.40 In the presence of 10 ␮M GCV, HuCC-T1 cells transfected with pTES.Tk showed a marked reduction in viability, whereas SSP-25 cells transfected with the same plasmid remained viable (data not shown). Even in the presence of



1247

Figure 8. Targeted killing of CEA-positive cells by pTES.Tk. HuCC-T1 (f) or SSP-25 (䡺) cells were transfected with pSES.Tk (a), pS.Tk (b), pT.Tk (c), or pTES.Tk (d) as described in the legend to Figure 4. Mock-transfected cells (e) were treated with PAAD alone. After 2 days, 5 ⫻ 103 cells were seeded into the wells of 96-well plates. On the following day, various concentrations of GCV were added to the cells. Five days later, the viabilities of the cells were assessed by Alamar blue assay. Each point represents the mean ⫾ SE of quadruplicate determinations.

30 ␮M GCV, which is beyond the peak serum level, the viability of the SSP-25 cells transfected with pTES.Tk was preserved (Fig 9), indicating CEA-negative cells were safe. Transfection with pT.Tk did not affect the viability of either

CEA-positive or -negative cells. The results strongly suggest that pTES.Tk but not pT.Tk can elicit significant therapeutic effects. DISCUSSION

Figure 9. Cell viability on day 5 of culture with a clinical dose of GCV. HuCC-T1 (left) or SSP-25 (right) cells were transfected with the indicated plasmids as described in the legend to Figure 8. Two days later, 5 ⫻ 103 cells were seeded into the wells of a 96-well plate. On the following day, GCV was added to a final concentration of 30 ␮M, and 5 days later, the viability of the cells was evaluated by Alamar blue assay. Each point represents the mean ⫾ SE of quadruplicate determinations.


In the present study, we have established an in vitro model system for targeted suicide gene therapy of CEA-producing cancer by using EBV/polyplex. Transfection with a non-EBV, conventional plasmid vector carrying the same promoter did not result in significant marker gene expression or effective killing of the CEApositive cells. Therefore, the present results demonstrate the usefulness of EBV-based but not conventional plasmid vectors in targeted suicide gene therapy for cancer. We and others have reported that plasmid vectors carrying oriP but not EBNA1 were effective in vitro in targeting EBV-associated lymphoma cells.20,41– 43 These particular tumor cells express endogenous EBNA1, which allows very strong expression when an oriPbearing plasmid is introduced. Because most normal cells do not express EBNA1, this strategy elicits specific killing of EBNA1-positive tumor cells. However, cancers not latently infected with EBV cannot be targeted by this method. To transfer the suicide gene into EBNA1negative cancer cells, we used an EBV-vector carrying

1248


both oriP and EBNA1 and succeeded in effectively killing the cancer cells both in vitro31 and in vivo.32 In previous studies, we employed a ubiquitous promoter and did not aim at targeting tumor cells. Therefore, the present study is the first in which an EBV-negative tumor was targeted by an EBV-based plasmid vector equipped with a tumor-specific promoter. To our knowledge, there have been two earlier studies in which targeted gene transfer into CEA-positive tumors was attempted by transfection with conventional plasmid vectors via nonviral methods. Fichera et al17 used liposome to deliver a plasmid vector in which the 490-bp sequence corresponding to the human CEA promoter was included to regulate the luciferase gene as a marker. Kurane et al18 used a plasmid vector containing a chimeric promoter consisting of the CEA promoter (a 424-bp fragment upstream of the translational start site originally reported by Schrewe et al14) and an enhancer from the immediate early gene of cytomegalovirus. Antibody against the Lewis Y antigen was also employed to improve the specificity of the vector.18 In both studies, however, only marker genes were transferred. Although specific expression was achieved in CEA-positive cells, the expression level was very low, with the duration of gene expression being short. No therapeutic effect was shown, probably due to the low level of expression obtained with the non-EBV-type plasmid vectors. PAAD is a positively charged synthetic polymer created recently.44 – 47 It is a macromolecule that is spherical in shape and composed of repeating polyamidoamino subunits. It has been shown that PAAD is an efficient vehicle to transduce genetic materials into mammalian cells. We have reported recently that by using PAAD, extremely efficient gene transfer can be achieved in tumor cells both in vitro31,32 and in vivo.32 Therefore, the EBV-plasmid vector/dendrimer complexes constructed in the present study may also be applied to experimental animal models, taking advantage of the in vivo effectiveness of PAAD. Transfection with pTET.␤ resulted in very poor marker gene expression in both CEA-positive and -negative cell lines. This may be ascribed to the relatively weak activity of the CEA promoter even in CEA-positive cells.13,48 pTES.␤ could be transduced effectively and exclusively into CEApositive cells, whereas pSET.␤ showed less selectivity. The results suggest that the expression level of the EBNA1 gene is critically important to the expression of the transgene. If a small amount of EBNA1 is first expressed in the transfected cells, the molecule may induce the expression of not only marker but also secondary EBNA1 molecules, which in turn induce further stronger expression of both genes. In this way, the EBNA1 gene driven by a weak promoter may elicit an intensive expression of the transgene, particularly if the marker gene is driven by a strong promoter. In conclusion, targeted gene transfer was achieved by using an EBV-based plasmid vector/PAAD equipped with a tumor-specific promoter. The system may be applicable not only to CC but to other CEA-producing malignancies, such as gastric and colon cancers. Further

studies are now underway to improve the vector system in terms of specificity and effectiveness and to evaluate the in vivo efficacy of the system. ACKNOWLEDGMENTS We thank Dr. M. Enjoji for cell lines. We also thank F. Hoffman-La Roche (Basel, Switzerland) for kindly providing GCV. This research was supported by a grant-in-aid for scientific research (No. 11670523) from the Ministry of Education, Japan.

REFERENCES 1. Chen MF, Jan YY, Wang CS, et al. Clinical experience in 20 hepatic resections for peripheral cholangiocarcinoma. Cancer. 1989;64:2226 –2232. 2. Schlinkert RT, Nagorney DM, Van Heerden JA, et al. Intrahepatic cholangiocarcinoma: clinical aspects, pathology, and treatment. HPB Surg. 1992;5:95–101. 3. Nakeeb A, Pitt HA, Sohn TA, et al. Cholangiocarcinoma: a spectrum of intrahepatic, perihilar, and distal tumors. Ann Surg. 1996;224:463– 475. 4. Kaneko S, Hallenbeck P, Kotani T, et al. Adenovirusmediated gene therapy of hepatocellular carcinoma using cancer-specific gene expression. Cancer Res. 1995;55:5283– 5287. 5. Ido A, Nakata K, Kato Y, et al. Gene therapy for hepatoma cells using a retrovirus vector carrying herpes simplex virus thymidine kinase gene under the control of human ␣-fetoprotein gene promoter. Cancer Res. 1995;55: 3105–3109. 6. Kanai F, Shiratori Y, Yoshida Y, et al. Gene therapy for ␣-fetoprotein-producing human hepatoma cells by adenovirus-mediated transfer of the herpes simplex virus thymidine kinase gene. Hepatology. 1996;23:1359 –1368. 7. Bui LA, Butterfield LH, Kim JY, et al. In vivo therapy of hepatocellular carcinoma with a tumor-specific adenoviral vector expressing interleukin-2. Hum Gene Ther. 1997;8: 2173–2182. 8. Gold P, Freedman SO. Specific carcinoembryonic antigens of the human digestive system. J Exp Med. 1965;122:467– 481. 9. Richards CA, Austin EA, Huber BE. Transcriptional regulatory sequences of carcinoembryonic antigen: identification and use with cytosine deaminase for tumor-specific gene therapy. Hum Gene Ther. 1995;6:881– 893. 10. Osaki T, Tanio Y, Tachibana I, et al. Gene therapy for carcinoembryonic antigen-producing human lung cancer cells by cell type-specific expression of herpes simplex virus thymidine kinase gene. Cancer Res. 1994;54:5258 –5261. 11. DiMaio JM, Clary BM, Via DF, et al. Directed enzyme pro-drug gene therapy for pancreatic cancer in vivo. Surgery. 1994;116:205–213. 12. Lan KH, Kanai F, Shiratori Y, et al. In vivo selective gene expression and therapy mediated by adenoviral vectors for human carcinoembryonic antigen-producing gastric carcinoma. Cancer Res. 1997;57:4279 – 4284. 13. Tanaka T, Kanai F, Lan KH, et al. Adenovirus-mediated gene therapy of gastric carcinoma using cancer-specific gene expression in vivo. Biochem Biophys Res Commun. 1997;231:775–779. 14. Schrewe H, Thompson J, Bona M, et al. Cloning of the complete gene for carcinoembryonic antigen: analysis of



15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25. 26.

27.

28.

29. 30.

31.

its promoter indicates a region conveying cell type-specific expression. Mol Cell Biol. 1990;10:2738 –2748. Cao G, Kuriyama S, Gao J, et al. In vivo gene transfer of a suicide gene under the transcriptional control of the carcinoembryonic antigen promoter results in bone marrow transduction but can avoid bone marrow suppression. Int J Oncol. 1999;15:107–112. Tanaka T, Kanai F, Okabe S, et al. Adenovirus-mediated prodrug gene therapy for carcinoembryonic antigen-producing human gastric carcinoma cells in vitro. Cancer Res. 1996;56:1341–1345. Fichera A, Michelassi F, Arenas RB. Selective expression of carcinoembryonic antigen promoter in cancer cell lines: targeting strategy for gene therapy in colorectal cancer. Dis Colon Rectum. 1998;41:747–774. Kurane S, Krauss JC, Watari E, et al. Targeted gene transfer for adenocarcinoma using a combination of tumor-specific antibody and tissue-specific promoter. Jpn J Cancer Res. 1998;89:1212–1219. Mazda O, Satoh E, Yasutomi K, et al. Extremely efficient gene transfection into lympho-hematopoietic cell lines by Epstein-Barr virus-based vectors. J Immunol Methods. 1997;204:143–151. Hirai H, Satoh E, Osawa M, et al. Use of EBV-based vector/HVJ-liposome complex vector for targeted gene therapy of EBV-associated neoplasms. Biochem Biophys Res Commun. 1997;241:112–118. Satoh E, Osawa M, Tomiyasu K, et al. Efficient gene transduction by Epstein-Barr-virus-based vectors coupled with cationic liposome and HVJ-liposome. Biochem Biophys Res Commun. 1997;238:795–799. Satoh E, Hirai H, Inaba T, et al. Successful transfer of ADA gene in vitro into human peripheral blood CD34⫹ cells by transfecting EBV-based episomal vectors. FEBS Lett. 1998;441:39 – 42. Tomiyasu K, Satoh E, Oda Y, et al. Gene transfer in vitro and in vivo with Epstein-Barr virus-based episomal vector results in markedly high transient expression in rodent cells. Biochem Biophys Res Commun. 1998;253:733–738. Sugden B, Marsh K, Yates J. A vector that replicates as a plasmid and can be efficiently selected in B-lymphoblasts transformed by Epstein-Barr virus. Mol Cell Biol. 1985;5: 410 – 413. Yates JL, Warren N, Sugden B. Stable replication of plasmids derived from Epstein-Barr virus in various mammalian cells. Nature. 1985;313:812– 815. Rawlins DR, Milman G, Hayward SD, et al. Sequencespecific DNA binding of the Epstein-Barr virus nuclear antigen (EBNA-1) to clustered sites in the plasmid maintenance region. Cell. 1985;42:859 – 868. Middleton T, Sugden B. Retention of plasmid DNA in mammalian cells is enhanced by binding of the EpsteinBarr virus replication protein EBNA1. J Virol. 1994;68: 4067– 4071. Gahn TA, Sugden B. An EBNA-1-dependent enhancer acts from a distance of 10 kilobase pairs to increase expression of the Epstein-Barr virus LMP gene. J Virol. 1995;69:2633–2636. Kirchmaier AL, Sugden B. Plasmid maintenance of derivatives of oriP of Epstein-Barr virus. J Virol. 1995;69:1280 – 1283. Jankelevich S, Kolman JL, Bodnar JW, et al. A nuclear matrix attachment region organizes the Epstein-Barr viral plasmid in Raji cells into a single DNA domain. EMBO J. 1992;11:1165–1176. Harada Y, Iwai M, Tanaka S, et al. Highly efficient suicide


32.

33.

34.

35.

36.

37. 38.

39.

40. 41. 42. 43.

44. 45.

46. 47.

48.

1249

gene expression in hepatocellular carcinoma cells by EpsteinBarr virus-based plasmid vectors combined with polyamidoamine dendrimer. Cancer Gene Ther. 2000;7:27–36. Maruyama-Tabata H, Harada Y, Matsumura T, et al. Effective suicide gene therapy in vivo by EBV-based plasmid vector coupled with polyamidoamine dendrimer. Gene Ther. 2000;7:53– 60. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem. 1976; 72:248 –254. Ahmed SA, Gogal RM Jr, Walsh JE. A new rapid and simple non-radioactive assay to monitor and determine the proliferation of lymphocytes: an alternative to [3H]thymidine incorporation assay. J Immunol Methods. 1994;170: 211–224. Hirai H, Shimazaki C, Yamagata N, et al. Effects of thrombopoietin (c-mpl ligand) on growth of blast cells from patients with transient abnormal myelopoiesis and acute myeloblastic leukemia. Eur J Haematol. 1997;59:38 – 46. Abbasi AM, Chester KA, MacPherson AJ, et al. Localization of CEA messenger RNA by in situ hybridization in normal colonic mucosa and colorectal adenocarcinomas. J Pathol. 1992;168:405– 411. Kim J, Kaye FJ, Henslee JG, et al. Expression of carcinoembryonic antigen and related genes in lung and gastrointestinal cancers. Int J Cancer. 1992;52:718 –725. Jothy S, Yuan SY, Shirota K. Transcription of carcinoembryonic antigen in normal colon and colon carcinoma; in situ hybridization study and implication for a new in vivo functional model. Am J Pathol. 1993;143:250 –257. Miyagiwa M, Ichida T, Tokiwa T, et al. A new human cholangiocellular carcinoma cell line (HuCC-T1) producing carbohydrate antigen 19/9 in serum-free medium. In Vitro Cell Dev Biol. 1989;25:503–510. Sommadossi J-P, Bevan R, Ling T, et al. Clinical pharmacokinetics of ganciclovir in patients with normal and impaired renal function. Rev Infect Dis. 1988;10:S507–S514. Judde JG, Spangler G, Magrath I, et al. Use of EpsteinBarr virus nuclear antigen-1 in targeted therapy of EBVassociated neoplasia. Hum Gene Ther. 1996;7:647– 653. Evans TJ, Brooks L, Farrell PJ. A strategy for specific targeting of therapeutic agents to tumor cells of virusassociated cancers. Gene Ther. 1997;4:264 –267. Kenney S, Ge JQ, Westphal EM, et al. Gene therapy strategies for treating Epstein-Barr virus-associated lymphomas: comparison of two different Epstein-Barr virusbased vectors. Hum Gene Ther. 1998;9:1131–1141. Haensler J, Szoka FC Jr. Polyamidoamine cascade polymers mediate efficient transfection of cells in culture. Bioconjug Chem. 1993;4:372–379. Kukowska-Latallo JF, Bielinska AU, Johnson J, et al. Efficient transfer of genetic material into mammalian cells using Starburst polyamidoamine dendrimers. Proc Natl Acad Sci USA. 1996;93:4897– 4902. Tang MX, Redemann CT, Szoka FC Jr. In vitro gene delivery by degraded polyamidoamine dendrimers. Bioconjug Chem. 1996;7:703–714. Tang MX, Szoka FC Jr. The influence of polymer structure on the interactions of cationic polymers with DNA and morphology of the resulting complexes. Gene Ther. 1997; 4:823– 832. Lan KH, Kanai F, Shiratori Y, et al. Tumor-specific gene expression in carcinoembryonic antigen-producing gastric cancer cells using adenovirus vectors. Gastroenterology. 1996;111:1241–1251.