Highly Efficient Suicide Gene Expression in Hepatocellular Carcinoma Cells by Epstein-Barr Virus-Based Plasmid Vectors Combined

Home , Expression vector

Highly efﬁcient suicide gene expression in hepatocellular carcinoma cells by Epstein-Barr virus-based plasmid vectors combined with polyamidoamine dendrimer

Yoshinori Harada,1 Masaki Iwai,1 Saiyu Tanaka,1 Takeshi Okanoue,1 Kei Kashima,1 Hiroko Maruyama-Tabata,2,3 Hideyo Hirai,2 Etsuko Satoh,2 Jiro Imanishi,2 and Osam Mazda2

1Third Department of Internal Medicine, and Departments of 2Microbiology and 3Pediatrics, Kyoto Prefectural University of Medicine, Kamikyo, Kyoto, Japan.

The present study was aimed at devising an efficient nonviral strategy for suicide gene therapy of hepatocellular carcinoma (HCC). To improve the efficiency of DNA delivery and expression, we applied Epstein-Barr virus (EBV)-based plasmid vectors instead of conventional plasmid vectors and combined them with cationic liposome (EBV/lipoplex) or polyamidoamine dendrimer (PAAD) (EBV/polyplex). When the ␤-galactosidase gene was transferred to HuH7, PLC/PRF/5, or HLE cells, Յ50-fold higher ␤-galactosidase activities were demonstrated in the cells transfected with EBV vector compared with those transfected with conventional plasmid vectors. PAAD-mediated transfection of HCC with pSES.Tk (an EBV-based vector carrying the herpes simplex virus-1 thymidine kinase gene) resulted in a marked reduction in viable cell number by the addition of ganciclovir (GCV). The HCC cells transfected with pSES.Tk/PAAD showed 100- to 1000-fold higher susceptibilities to GCV than those transfected with pS.Tk (a conventional plasmid vector carrying herpes simplex virus-1 thymidine kinase gene)/PAAD. The pSES.Tk-transfected HCC cells were effectively killed by day 9 in culture with a clinically feasible concentration of GCV (25 ␮M), whereas the pS.Tk-transfected cells survived the culture. These results demonstrate highly efficient suicide gene transfer into various HCC cells by EBV-based plasmid vectors in vitro, suggesting the possible application of this nonviral vector system to gene therapy of HCC. Cancer Gene Therapy (2000) 7, 27–36

Key words: Epstein-Barr virus-based vector; polyamidoamine dendrimer; nonviral vector; hepatocellular carcinoma; suicide gene.

epatocellular carcinoma (HCC) is one of the most the transfection/expression efficiency of nonviral systems Hcommon and ominous malignancies in humans. remains relatively poor as long as conventional plasmid Despite recent advances in HCC therapy, including liver vectors are used. Less effort has been made to improve transplantation, the cure rate for HCC remains low.1 plasmid vectors, although some novel promoter/enhanc- Gene therapy modalities may provide an additional ers are being actively investigated.7–10 We substituted strategy for treatment. We focused on nonviral vector Epstein-Barr virus (EBV)-based plasmid vectors for the systems as a means of suicide gene therapy of HCC. conventional plasmid vectors and employed them as a Generally, a nonviral vector system consists of two component of practical nonviral vector systems in HCC. components: a plasmid vector (expression vector) and a EBV-based episomal vectors are plasmid vectors car- gene delivery vehicle (nonviral vector). Improvements of rying the EBV nuclear antigen 1 (EBNA1) gene and the both components are considered to be important to oriP region from EBV as trans and cis elements for DNA make a nonviral vector system sufficiently useful for replication, respectively.11,12 Through binding to oriP, human gene therapy. With regard to the gene delivery EBNA1 facilitates the retention, nuclear localization, vehicles, a number of methods have been devised, and replication of the plasmid DNA,13–17 giving the including those using liposomes, synthetic polymers, and EBV vector favorable features of an artificial chromo- physical means such as electroporation and a gene some. We have demonstrated that extremely efficient 2–6 gun. Despite the recent progress in these techniques, transfection can be achieved with EBV-based episomal vectors by means of electroporation in lymphoma cell lineage.18,19 In the present study, we transferred the herpes simplex Received October 20, 1998; accepted December 31, 1998. 20–22 Address correspondence and reprint requests to Dr. Osam Mazda, virus-1 (HSV-1) thymidine kinase (Tk) gene in vitro Department of Microbiology, Kyoto Prefectural University of Medicine, into three HCC cell lines via an EBV-based episomal Kamikyo, Kyoto 602-8566, Japan. E-mail address: [email protected] vector coupled with polyamidoamine dendrimer m.ac.jp (PAAD). The high sensitivity of the transfected cells to

Cancer Gene Therapy, Vol 7, No 1, 2000: pp 27–36 27 28 HARADA, IWAI, TANAKA, ET AL: SUICIDE GENE EXPRESSION IN HCC BY EBV-BASED VECTOR/PAAD

Figure 1. Plasmids used in this study. Maps of pSES.␤, pS.␤, pSES.Tk, and pS.Tk are shown. Prom, promoter; poly(A), SV40 poly(A) additional signal. ganciclovir (GCV) may suggest the possible application pS.Tk (Fig 1, lower right panel) were constructed from pSES.␤ of this vector system to the gene therapy of HCC. and pSES.Tk, respectively, by deleting SR␣-EBNA1 and oriP.

Cells MATERIALS AND METHODS The human HCC cell lines HLE, PLC/PRF/5, and HuH-7 Plasmid vectors (Health Science Research Resources Bank, Osaka, Japan), were passaged in Dulbecco’s modiﬁed Eagle’s medium (Life 21,22 Plasmid vectors were constructed as described previously. Technologies) supplemented with 10% fetal bovine sera (Life pSES.␤ (Fig 1, upper left panel) and pSES.Tk (Fig 1, upper right Technologies), 100 U/mL of ampicillin, and 100 ␮g/mL of panel) are composed of (a) the Escherichia coli ␤-galactosidase streptomycin (complete medium). (␤-gal) gene (derived from pSV␤ (Clontech, Palo Alto, Calif)) ␤ (pSES. )ortheHSV-1 Tk gene (derived from pHSV-106 (Life Transfection by cationic liposomes Technologies, Gaithersburg, Md)) (pSES.Tk), located between the SR␣ promoter (HindIII-XhoI 0.6-kilobase fragment from For cationic liposome-mediated transfection, cells were seeded pcDL-SR␣29623) and the simian virus 40 (SV40) poly(A) into the wells of a six-well plate (Becton Dickinson Labware, additional signal, (b) EBV oriP (derived from p220.212), (c) Franklin Lakes, NJ) and allowed to proliferate until they were the EBV EBNA1 gene (derived from p220.2) under the control 70–80% conﬂuent. Shortly before transfection, the culture of the SR␣ promoter (ClaI-ClaI 0.9-kilobase fragment from medium was replaced with 1 mL of fresh complete medium. pcDL-SR␣296), and (d) the ampicillin-resistant gene and the Plasmid DNA (1.5 ␮g) was mixed with various amounts of Col E1 replication origin. pS.␤ (Fig 1, lower left panel) and 1,3-di-oleoyloxy-2-(6-carboxy-spermyl)-propylamide (DOSPER)

Cancer Gene Therapy, Vol 7, No 1, 2000 HARADA, IWAI, TANAKA, ET AL: SUICIDE GENE EXPRESSION IN HCC BY EBV-BASED VECTOR/PAAD 29

Figure 2. ␤-gal expression in HCC cells transfected by cationic liposome. A: Transfection of HLE cells by cationic liposome. HLE cells (3 ϫ 105 cells) were plated in the wells of six-well plates; after 24 hours, the cells were incubated for 6 hours with 1.5 ␮g of pSES.␤ (closed bars) or pS.␤ (hatched bars) combined with the indicated amount of cationic liposome (DOSPER). At 60 hours posttransfection, ␤-gal activities were measured and normalized to 1 mg of protein. The results shown are the mean Ϯ SE of quadruplicate samples. B: Expression of ␤-gal in HLE, PLC/PRF/5, and HuH-7 cells transfected by cationic liposome. Cells were transfected with pSES.␤ (closed bars) or pS.␤ (hatched bars)/cationic liposome under optimal conditions (1.5 ␮g of DNA and 6 ␮g of cationic liposome per 3 ϫ 105 cells). ␤-gal activities were determined on the indicated days posttransfection. The results shown are the mean Ϯ SE of quadruplicate samples.

(Boehringer Mannheim, Mannheim, Germany) to a final vol- Transfection by PAAD ume of 100 ␮L. After incubation at room temperature for 15 minutes, 100 ␮L of the plasmid DNA-cationic liposome com- Cells were seeded into the wells of a six-well plate (Becton plex was added to the culture, which was then incubated at Dickinson Labware) at a density of 3 ϫ 105 cells/well and 37°C in 5% CO2/95% humidified air. After 6 hours, the culture incubated at 37°C in 5% CO2/95% humidified air. On the medium was replaced with fresh medium, and the cells were following day, 2 ␮g of DNA was mixed with various amounts of further cultured until they were subjected to the ␤-gal assay. PAAD (generation 5) (Superfect) (Qiagen, Hilden, Germany)

Cancer Gene Therapy, Vol 7, No 1, 2000 30 HARADA, IWAI, TANAKA, ET AL: SUICIDE GENE EXPRESSION IN HCC BY EBV-BASED VECTOR/PAAD

Figure 3. ␤-gal expression in HCC cells transfected by PAAD. A: Transfection of HLE cells by PAAD. HLE cells (3 ϫ 105 cells) were plated in the wells of six-well plates; after 24 hours, the cells were incubated for 2 hours with 1 or 2 ␮gof pSES.␤ (closed bars) or pS.␤ (hatched bars) combined with the indicated amount of PAAD. At 48 hours posttransfection, ␤-gal activities were measured and normalized to 1 mg of protein. The results shown are the mean Ϯ SE of quadruplicate samples. B: Expression of ␤-gal in HLE, PLC/PRF/5, and HuH-7 cells transfected by PAAD. Cells were transfected with pSES.␤ (closed bars) or pS.␤ (hatched bars)/PAAD under optimal conditions (2 ␮g of DNA and 15 ␮Lof PAAD per 3 ϫ 105 cells). At 4 or 8 days posttransfection, ␤-gal activities were measured. The results shown are the mean Ϯ SE of quadruplicate samples.

in Dulbecco’s modiﬁed Eagle’s medium and incubated for 10 according to the manufacturer’s protocol. The optical density minutes at room temperature. The complex was added with (OD) was measured at 420 nm. Activities were calculated ␮ ␤ ϭ ϫ 600 L of complete medium and immediately transferred to using the following formula: -gal units ([OD420 380]/30)/ the cells that had been washed with phosphate-buffered saline mg protein, where 380 is a conversion factor and 30 is the time (PBS). The cells were incubated at 37°C in 5% CO2/95% of incubation in minutes. The protein concentration of the humidiﬁed air for 2 hours, followed by replacement with fresh supernatant was assessed according to the method of Brad- complete medium. After further incubation for appropriate ford.24 periods, the cells were subjected to ␤-gal assay, 5-bromo-4- chloro-3-indolyl ␤-D-galactoside (X-Gal) staining, or Alamar blue assay. X-Gal staining

␤-gal assay Cells were ﬁxed with 1% glutaraldehyde/PBS for 10 minutes, washed three times with PBS, and subsequently incubated in Cells were scraped off dishes, washed twice with PBS, and X-Gal staining solution (0.05% (v/v) X-Gal (Nacalai Tesque, ␮ resuspended in 50 L of tris(hydroxymethyl)aminomethane- Kyoto, Japan), 1 mM MgCl2, 150 mM NaCl, 3 mM HCl (pH 7.8). After two cycles of freezing and thawing, the K4(Fe(CN)6),3mMK3(Fe(CN)6), 60 mM Na2HPO4,40mM lysate was centrifuged at 15,000 revolutions per minute for 5 NaH2PO4, and 0.1% Triton X-100) for 3 hours at 37°C. The minutes. The ␤-gal activity in the cell supernatant fraction was reaction was terminated by replacing the solution with 1 mM ␤ assayed with a -gal assay kit (Invitrogen, San Diego, Calif) Na2-ethylenediaminetetraacetic acid/PBS.

Cancer Gene Therapy, Vol 7, No 1, 2000 HARADA, IWAI, TANAKA, ET AL: SUICIDE GENE EXPRESSION IN HCC BY EBV-BASED VECTOR/PAAD 31

Figure 4. X-Gal staining of transfected cells. HLE cells were transfected with pSES.␤ (upper panels)orpS.␤ (lower panels)/PAAD, and ﬁxed and stained by X-Gal on day 4 (left panels) or day 10 (right panels) posttransfection (original magniﬁcation is ϫ32).

Alamar blue assay pS.␤ (see Fig 1) encapsulated with cationic liposome and ␤ The susceptibility of cells to GCV was examined by Alamar PAAD. The -gal activities were measured after appro- blue assay.25,26 Briefly, quadruplicate aliquots of cells were priate periods of culture. First, the optimal preparation seeded in 96-well, flat-bottom microtiter plates (Falcon, Lin- and transfection conditions were determined for cationic coln Park, NJ) (5 ϫ 103 cells in 200 ␮L of complete medium liposome-mediated transfection into HLE cells. Various per well). After 24 hours, GCV (Tanabe, Tokyo, Japan) was ratios of cationic liposome to DNA were tested. At every added at various concentrations ranging from 0 to 10,000 ␮M. ratio, transfection of the EBV-based episomal vector After further incubation in 5% CO2/95% humidified air at (pSES.␤) resulted in higher ␤-gal activity compared with 37°C for 96 hours, Alamar blue (Alamar Biosciences, Sacra- that of the conventional plasmid vector (pS.␤)at60 mento, Calif) was added according to the manufacturer’s hours posttransfection (Fig 2A). Maximal ␤-gal activity protocol. The cells were further cultured for 4 hours, and the ␮ ␮ OD of each well was measured with a microplate reader at test was obtained by mixing 1.5 g of DNA with 6 gof cationic liposome; at this optimal condition, ϳ12-fold and reference wavelengths of 570 and 600 nm, respectively. ␤ The percentages of viable cells were calculated according to higher enzyme activity was obtained with pSES. com- ϭ Ϫ ␤ the following formula: % viable cells (OD570 OD600)of pared with pS. . Ϫ GCV-treated cells/(OD570 OD600) of untreated cells. Next, we estimated the efficiency of transfection/ expression in three HCC cell lines using the EBV/ cationic liposome. Experimental conditions were opti- RESULTS mized for each cell line (data not shown). On day 4 posttransfection, the transfection efficiency by pSES.␤/ EBV-based episomal vector/cationic liposome complex cationic liposome was ϳ20-, ϳ2.8-, and ϳ11-fold higher (EBV/lipoplex) as an efficient nonviral system to than that provided by pS.␤/cationic liposome in HLE, transfect HCC cells in vitro PLC/PRF/5, and HuH-7 cells, respectively (Fig 2B). To examine whether EBV-based episomal vectors can Even on day 8 posttransfection, the levels of ␤-gal were be combined successfully with nonviral gene delivery still much higher in the cells transfected with pSES.␤ systems, we transfected HCC cell lines with pSES.␤ or than in the pS.␤-transfected cells. The results indicated

Cancer Gene Therapy, Vol 7, No 1, 2000 32 HARADA, IWAI, TANAKA, ET AL: SUICIDE GENE EXPRESSION IN HCC BY EBV-BASED VECTOR/PAAD

Figure 5. Susceptibilities to GCV of three HCC cell lines transfected with pSES.Tk or pS.Tk/PAAD. HLE, PLC/PRF/5, or HuH-7 cells were transfected with pSES.Tk (f), pS.Tk (F), or pS.␤ (‚)/PAAD as described in the legend to Figure 3. Mock-transfected cells (छ) were prepared as a control. After 3 days, 5 ϫ 103 cells were seeded into a 96-well plate and cultured in the presence of various concentrations of GCV for 4 days. The viabilities of the cells were measured by Alamar blue assay. The points on the ordinate corre- spond to the relative cell viability compared with the mock-transfected cells without GCV. Each point represents the mean Ϯ SE of quadruplicate determinations. that transfection by EBV-based episomal vectors com- ␤-gal expression was also demonstrated by the X-Gal bined with cationic liposome was much more efficient than staining of HLE cells transfected with pSES.␤ or pS.␤/ that by non-EBV conventional vector/cationic liposome. PAAD under optimal conditions. As shown in Figure 4, ϳ35% of the cells transfected with pSES.␤/PAAD ex- Highly efficient transfection into HCC cells by EBV- pressed ␤-gal on day 4 posttransfection, whereas only based episomal vector/PAAD complex (EBV/polyplex) 3% of the pS.␤/PAAD-transfected cells were positive for this marker gene product. Next, we tried to combine an EBV vector with PAAD, which has been reported to be more effective than Extremely high susceptibility to GCV of HCC cells cationic liposome in some experimental systems.27 In transfected with pSES.Tk/PAAD our first series of experiments, several different ratios of DNA to PAAD were tested for HLE cells (Fig 3A). The We hypothesized that the highly efficient EBV-based highest ␤-gal activity was observed when 15 ␮g of PAAD episomal vector system may be useful in transferring a and 2 ␮g of pSES.␤ were used. At this optimal prepa- suicide gene into HCC. To assess this possibility, we ration, the pSES.␤/PAAD yielded 30-fold higher ␤-gal constructed two plasmids (i.e., pSES.Tk, carrying the activity compared with the optimal preparation of pS.␤/ HSV-1 Tk gene expression unit, the EBNA1 gene, and PAAD. oriP, and pS.Tk, carrying HSV-1 Tk but not EBNA1 or We also estimated the efficiency of transfection/ex- oriP) (Fig 1). These plasmids were transfected into the pression in the other HCC cell lines by the EBV HLE, PLC/PRF/5, and HuH-7 cell lines by means of vector/PAAD. The experimental condition was also op- PAAD. At 72 hours posttransfection, cells were cultured timized for each cell line (data not shown). ␤-gal activity in the presence of various concentrations of GCV for 4 was ϳ50-fold (HLE), ϳ28-fold (PLC/PRF/5), and ϳ10- days; next, their viabilities were measured by Alamar fold (HuH-7) stronger in pSES.␤-transfected cells than blue assay. As shown in Figure 5, the transfer of pS.Tk pS.␤-transfected cells on day 4 posttransfection (Fig 3B). did not significantly affect the susceptibilities of the HLE

Cancer Gene Therapy, Vol 7, No 1, 2000 HARADA, IWAI, TANAKA, ET AL: SUICIDE GENE EXPRESSION IN HCC BY EBV-BASED VECTOR/PAAD 33

Figure 6. Sensitivity of HCC cells transfected with pSES.Tk/PAAD to a clinical dose of GCV. HLE, PLC/ PRF/5, or HuH-7 cells were transfected with pSES.Tk (f) or pS.Tk (F)/PAAD as described in the legend to Figure 3. Mock-transfected cells (छ) were prepared as controls. On day 2, cells were harvested and plated into a six-well plate at a concentration of 3 ϫ 105 cells/well. GCV was added at a ﬁnal concentration of 25 ␮M, and the number of viable cells was counted after 3, 6, and 9 days. Cumulative cell numbers were calculated and plotted along the vertical axis. Each point represents the mean Ϯ SE of quadruplicate determinations.

and PLC/PRF/5 cells to GCV, whereas the the suscep- Transfection of pSES.Tk did not influence proliferation tibility of HuH-7 to GCV was moderately elevated. In of HCC cells per se contrast, the transfection of pSES.Tk/PAAD rendered We examined whether cell proliferation was affected by all of the cell lines highly susceptible to GCV. The gene delivery with EBV-based vector. HLE, PLC/PRF/5, GCV-susceptibilities of HLE, PLC/PRF/5, and HuH-7 and HuH-7 cells were transfected with pSES.Tk or cells transfected with EBV-based vector/PAAD were pS.Tk/PAAD. After 1 day, cells were resuspended in the ϳ250-, ϳ100-, and ϳ1000-fold higher than that of the conditioned medium and plated at a density of 3 ϫ 105 cells transfected with non-EBV vector/PAAD, respec- cells per 35-mm well. The numbers of cells were tively. counted. As shown in Figure 7, all of the cell lines exhibited growth curves that were comparable with HCC cells transfected by pSES.Tk/PAAD were those of the mock-transfected cells, regardless of the effectively killed by a clinical dose of GCV plasmids transfected. These results indicated that the To examine whether the elevated sensitivity to GCV introduction of the HSV-Tk gene, the EBNA1 gene, given by pSES.Tk/PAAD had therapeutic significance, and oriP did not affect cell proliferation in the absence HCC cells that had been transfected as described above of GCV. were cultured in the presence of 25 ␮M of GCV; the numbers of cells were counted after 3, 6, and 9 days. We used 25 ␮M of GCV because this dose was considered to DISCUSSION be the standard sera GCV concentration in HCC patients administered ordinary clinical doses of GCV.28 As In the present study, we demonstrated that a suicide shown in Figure 6, HCC cells transfected with pSES.Tk/ gene could be quite effectively transferred into human PAAD were markedly reduced in cell number within 9 HCC cells in vitro by means of EBV-based episomal days of culture. In contrast, in the transfection by vectors coupled with synthetic macromolecules. Some pS.Tk/PAAD, the reduction in viable cell number was groups, including us, have reported that plasmid vectors less sufficient. carrying oriP but not EBNA1 were quite effective in

Cancer Gene Therapy, Vol 7, No 1, 2000 34 HARADA, IWAI, TANAKA, ET AL: SUICIDE GENE EXPRESSION IN HCC BY EBV-BASED VECTOR/PAAD

Figure 7. Influence of pSES.Tk transfection on the proliferation of HCC cells. HLE, PLC/PRF/5, and HuH-7 cells were transfected with pSES.Tk (f)or pS.Tk (F)/PAAD as described in the legend to Figure 3. After 2 days, cells were plated into the wells of a six-well plate at a concentration of 3 ϫ 105 cells/well and cultured in the absence of GCV. The cell number was counted on the appropriate days posttransfection. Cumulative cell numbers were calculated and are plotted along the vertical axis. Each point represents the mean Ϯ SE of quadruplicate determinations. transferring the HSV-1 Tk gene in vitro into EBV- cannot replicate,32 and (b) even in human cells, the associated lymphoma cells;29–31 these particular cancer oriP-bearing plasmids replicate only once in each S cells strongly express endogenous EBNA1, allowing very phase, so unlike the SV40-based plasmid vectors,33 the strong expression from oriP-bearing plasmids. However, copy number of the EBV plasmids does not increase to our knowledge, the present study is the first report after transfection. showing that EBV-episomal vectors carrying both oriP As far as we know, the present study is the first report and EBNA1 are quite effective in suicide gene transfer in which PAAD was used to deliver EBV plasmid DNA into HCC, which is an EBNA1-negative cancer. into cells. In the comparison of the cationic liposome In general, the advantage of EBV-based episomal (DOSPER) and PAAD, the marker gene was more vectors has been appreciated in long-term maintenance efficiently transferred by the latter than by the former. of the plasmid rather than in transient gene expression. These results are consistent with an earlier study by However, we demonstrated that even at a transient Kukowska-Latallo et al showing the efficacy of PAAD in phase, EBV-based episomal vectors combined with both gene delivery.27 However, many different cationic lipo- delivery vehicles resulted in a markedly high expression somes with different compositions have been devised, of marker gene in all of the cell lines tested. The high and some cationic liposomes other than DOSPER may transient expression by EBV-based episomal vectors work better with the EBV vectors. may result from the multiple functions of EBNA1, such The X-Gal staining experiments showed that ϳ35% of as nuclear transfer of the plasmid DNA, anchorage of the pSES.␤-transfected HLE cells expressed the marker the plasmid DNA to the nuclear matrix, and transcrip- gene product on day 4 posttransfection (Fig 4). Simi- tional enhancement.13–17 larly, the positivity in the pSES.␤-transfected HuH-7 The episomal replication may also contribute to the cells was ϳ45% (data not shown). In contrast, when high transient transfection/expression by the EBV plas- pSES.Tk-transfected HuH-7 cells were cultured in the mids. However, the contribution may be small because presence of 25 ␮M of GCV, virtually 100% of the cells (a) highly efficient transfection/expression can also be were killed during 9 days of culture (Fig 6). This seen in rodent cells, in which the EBV-based plasmids discrepancy may be explained by the so-called bystander

Cancer Gene Therapy, Vol 7, No 1, 2000 HARADA, IWAI, TANAKA, ET AL: SUICIDE GENE EXPRESSION IN HCC BY EBV-BASED VECTOR/PAAD 35 effect.34,35 Indeed, Kuriyama et al have shown that 5. Tang MX, Redemann CT, Szoka FC. In vitro gene delivery 10–20% of HSV-1 Tk-expressing cells was enough to kill by degraded polyamidoamine dendrimers. Bioconjug nearly 100% of HCC cells when HSV-1 Tk-positive and Chem. 1996;7:703–714. -negative cells were mixed at various ratios.33 Alterna- 6. Yang NY, Sun WH. Gene gun and other non-viral ap- tively, a small proportion of mock-transfected cells was proaches for cancer gene therapy. Nat Med. 1995;1:481– 483. killed by GCV treatment (Fig 5). A fraction of bystander 7. Weichselbaum RR, Hallahan DE, Beckett, MA, et al. cells might have been killed in a similar manner. Gene therapy targeted by radiation preferentially radio- The present X-Gal staining experiments also showed sensitizes tumor cells. Cancer Res. 1994;54:4266–4269. that in the case of HLE cells, the ␤-gal positivity was 8. Kaneko S, Hallenbeck P, Kotani T, et al. Adenovirus- ϳ12-fold higher in pSES.␤-transfected cells than in mediated gene therapy of hepatocellular carcinoma using pS.␤-transfected cells (35% vs. 3%) (Fig 4). In contrast, cancer-specific gene expression. Cancer Res. 1995;55:5283– the ␤-gal enzyme activity in the pSES.␤-transfected 5287. HLE cells was ϳ50-fold higher than that in the pS.␤- 9. Richards CA, Austin EA, Huber BE. Transcriptional transfected cells (Fig 3A). Thus, the high level of regulatory sequences of carcinoembryonic antigen: identi- expression by the EBV vectors may be attributed to both fication and use with cytosine deaminase for tumor-specific ␤ gene therapy. Hum Gene Ther. 1995;6:881–893. the high frequency of -gal-positive cells and the high 10. Paillard F. “Tet-on”: a gene switch for the exogenous intensity of marker gene expression in each positive cell. regulation of transgene expression. Hum Gene Ther. 1998; Specific targeting of cancer cells is important for the 9:983–985. clinical application of cancer gene therapy strategies. In 11. Sugden B, Marsh K, Yates JL. A vector that replicates as the case of HCC, the promoter/enhancer of the ␣-feto- a plasmid and can be efficiently selected in B-lymphoblasts protein gene has been used successfully in combination transformed by Epstein-Barr virus. Mol Cell Biol. 1985;5: with retroviral or adenoviral vectors.8,36–38 The ␣-feto- 410–413. protein promoter can be included in the EBV-based 12. Yates JL, Warren N, Sugden B. Stable replication of vectors to drive EBNA1 and/or HSV-1 Tk genes. plasmids derived from Epstein-Barr virus in various mam- It has been shown that the PAAD can be applicable to malian cells. Nature. 1985;313:812–815. 13. Rawlins DR, Milman G, Hayward SD, et al. Sequence- in vivo gene delivery without apparent immunogenic or 39 specific DNA binding of the Epstein-Barr virus nuclear toxic effects. Further studies in vivo are now underway antigen (EBNA-1) to clustered sites in the plasmid main- for the evaluation of the efficacy and safety of our tenance region. Cell. 1985;42:859–868. system. 14. Middleton T, Sugden B. Retention of plasmid DNA in mammalian cells is enhanced by binding of the Epstein- Barr virus replication protein EBNA1. J Virol. 1994;68: ACKNOWLEDGMENTS 4067–4071. 15. Gahn TA, Sugden B. An EBNA-1-dependent enhancer acts from a distance of 10 kilobase pairs to increase We thank Dr. Taiki Tamaoki (Department of Medical Bio- expression of the Epstein-Barr virus LMP gene. J Virol. chemistry, Faculty of Medicine, University of Calgary, Calgary, 1995;69:2633–2636. Alberta, Canada) for his helpful comments and critical review 16. Kirchmaier AL, Sugden B. Plasmid maintenance of deriv- of the manuscript and Drs. Akira Muramatsu and Takahiro atives of oriP of Epstein-Barr virus. J Virol. 1995;69:1280– Mori (Third Department of Internal Medicine, Kyoto Prefec- 1283. tural University of Medicine) for helpful discussions. We also 17. Jankelevich S, Kolman JL, Bodnar JW, et al. A nuclear thank Maki Hosoi and Satoko Watanabe (Third Department matrix attachment region organizes the Epstein-Barr viral of Internal Medicine and Department of Microbiology, Kyoto plasmid in Raji cells into a single DNA domain. EMBO J. Prefectural University of Medicine) for their excellent secre- 1992;11:1165–1176. tarial assistance. This research was supported by a grant-in-aid 18. Mazda O, Teshigawara K, Fujimoto S, et al. A reporter for scientific research from the Ministry of Education, Science, system using a flow cytometer to detect promoter/en- and Culture of Japan, by the Japan Heart Foundation, and by hancer activity in lymphoid cell lines. J Immunol Methods. an IBM Japan research grant. 1994;169:53–61. 19. Mazda O, Satoh E, Yasutomi K, et al. Extremely efficient gene transfection into lympho-hematopoietic cell lines by REFERENCES Epstein-Barr virus-based vectors. J Immunol Methods. 1997;204:143–152. 1. Colombo M. Hepatocellular carcinoma. J Hepatol. 1992; 20. Qian C, Bilbao R, Bruna O, et al. Induction of sensitivity 15:225–236. to ganciclovir in human hepatocellular carcinoma cells by 2. Felgner PL, Ringold GM. Cationic liposome-mediated adenovirus-mediated gene transfer of herpes simplex virus transfection. Nature. 1989;337:387–388. thymidine kinase. Hepatology. 1995;22:118–123. 3. Sorscher EJ, Logan JJ, Frizzell RA, et al. Gene therapy for 21. Moolten FL. Tumor chemosensitivity conferred by in- cystic fibrosis using cationic liposome-mediated gene serted herpes thymidine kinase genes: paradigm for a transfer: a phase I trial of safety and efficacy in the nasal prospective cancer control strategy. Cancer Res. 1986;46: airway. Hum Gene Ther. 1994;5:1259–1277. 5276–5281. 4. Boussif O, Lezoualch F, Zanta MA, et al. A versatile 22. Moolten FL. Drug sensitivity (“suicide”) genes for selec- vector for gene and oligonucleotide transfer into cells in tive cancer chemotherapy. Cancer Gene Ther. 1994;1:279– culture and in vivo: polyethylenimine. Proc Natl Acad Sci 287. USA. 1995;92:7297–7301. 23. Takebe Y, Seiki M, Fujisawa J, et al. SR␣ promoter: an

Cancer Gene Therapy, Vol 7, No 1, 2000 36 HARADA, IWAI, TANAKA, ET AL: SUICIDE GENE EXPRESSION IN HCC BY EBV-BASED VECTOR/PAAD

efficient and versatile mammalian cDNA expression sys- 31. Kenney S, Ge JQ, Westphal EM, et al. Gene therapy tem composed of the simian virus 40 early promoter and strategies for treating Epstein-Barr virus-associated lym- the R-U5 segment of human T-cell leukemia virus type 1 phomas: comparison of two different Epstein-Barr virus- long terminal repeat. Mol Cell Biol. 1988;8:466–472. based vectors. Hum Gene Ther. 1998;9:1131–1141. 24. Bradford MM. A rapid and sensitive method for the 32. Tomiyasu K, Satoh E, Oda Y, et al. Gene transfer in vitro quantitation of microgram quantities of protein utilizing and in vivo with Epstein-Barr virus-based episomal vector the principle of protein-dye binding. Anal Biochem. 1976; results in markedly high transient expression in rodent 72:248–254. cells. Biochem Biophys Res Commun. 1998;253:733–738. 25. Ahmed SA, Gogal RM, Jr, Walsh JE. A new rapid and 33. Yates JL, Guan N. Epstein-Barr virus-derived plasmids simple non-radioactive assay to monitor and determine the replicate only once per cell cycle and are not amplified proliferation of lymphocytes: an alternative to [3H]thymi- after entry into cells. J Virol. 1991;65:483–488. dine incorporation assay. J Immunol Methods. 1994;170: 34. Dilber MS, Smith CIE. Suicide genes and bystander 211–224. killing: local and distant effects. Gene Ther. 1997;4:273– 26. Hirai H, Shimazaki C, Yamagata N, et al. Effects of 274. thrombopoietin (c-mpl ligand) on growth of blast cells 35. Kuriyama S, Nakatani T, Masui K, et al. Bystander effect from patients with transient abnormal myelopoiesis and caused by suicide gene expression indicates the feasibility acute myeloblastic leukemia. Eur J Hematol. 1997;59:38– of gene therapy for hepatocellular carcinoma. Hepatology. 46. 1995;22:1838–1846. 27. Kukowska-Latallo JF, Bielinska AU, Johnson J, et al. 36. Ido A, Nakata K, Kato Y, et al. Gene therapy for Efficient transfer of genetic material into mammalian cells hepatoma cells using a retrovirus vector carrying herpes using Starburst polyamidoamine dendrimers. Proc Natl simplex virus thymidine kinase gene under the control of Acad Sci USA. 1996;93:4897–4902. human ␣-fetoprotein gene promoter. Cancer Res. 1995;55: 28. Sommadossi JP, Bevan R, Ling T, et al. Clinical pharma- 3105–3109. cokinetics of ganciclovir in patients with normal and 37. Kanai F, Shiratori Y, Yoshida Y, et al. Gene therapy for impaired renal function. Rev Infect Dis. 1988;10(suppl ␣-fetoprotein-producing human hepatoma cells by adeno- 3):S507–S514. virus-mediated transfer of the herpes simplex virus thymi- 29. Hirai H, Satoh E, Osawa M, et al. Use of EBV-based dine kinase gene. Hepatology. 1996;23:1359–68. vector/HVJ-liposome complex vector for targeted gene 38. Bui LA, Butterfield LH, Kim JY, et al. In vivo therapy of therapy of EBV-associated neoplasms. Biochem Biophys hepatocellular carcinoma with a tumor-specific adenoviral Res Commun. 1997;241:112–118. vector expressing interleukin-2. Hum Gene Ther. 1997;8: 30. Mucke S, Polack A, Pawlita M, et al. Suitability of Epstein- 2173–2182. Barr virus-based episomal vectors for expression of cyto- 39. Qin L, Pahud DR, Ding Y, et al. Efficient transfer of genes kine genes in human lymphoma cells. Gene Ther. 1997;4: into murine cardiac grafts by Starburst polyamidoamine 82–92. dendrimers. Hum Gene Ther. 1998;9:553–560.

Cancer Gene Therapy, Vol 7, No 1, 2000