Expression of Escherichia coli uracil phosphoribosyltransferase gene ...

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

Expression of Escherichia coli uracil phosphoribosyltransferase gene in murine colon carcinoma cells augments the antitumoral effect of 5-fluorouracil and induces protective immunity Kiyoko Kawamura,1 Kentaro Tasaki,1,2 Hirofumi Hamada,3,4 Keizo Takenaga,5 Shigeru Sakiyama,6 and Masastoshi Tagawa1 Divisions of 1Pathology and 5Chemotherapy, 6Chiba Cancer Center Research Institute, Chiba, Japan; 2 Department of Surgery (II), Chiba University School of Medicine, Chiba, Japan; 3Department of Molecular Biotherapy Research, Cancer Chemotherapy Center, Cancer Institute, Tokyo, Japan; and 4Department of Molecular Medicine, Sapporo Medical University, Sapporo, Japan. Uracil phosphoribosyltransferase (UPRT) of Escherichia coli origin can convert 5-fluorouracil (5-FU), a chemotherapeutic agent widely used for solid tumors, to an active intermediate, 5-fluorouridine-5⬘-monophosphate, as mammalian orotate phosphoribosyltransferase does. To examine whether the E. coli UPRT gene expressed in tumor cells can confer increased sensitivity to 5-FU, we retrovirally transduced Colon 26 cells, a murine colon carcinoma cell line, with the UPRT gene (Colon 26/UPRT cells) and tested the in vivo antitumoral effect of 5-FU in syngeneic immunocompetent mice. After 5-FU administration, tumors of Colon 26/UPRT cells regressed, whereas those of wild-type cells were unaffected. The mice that once eliminated Colon 26/UPRT tumors after 5-FU treatment rejected wild-type cells that were subsequently inoculated but not irrelevant syngeneic tumor cells. This suicide gene/prodrug system was less efficient in nude mice, suggesting that mature ␣␤ T cells play a role in the antitumoral effect. The cytotoxicity mediated by the bystander effect was marginal in this system, contrary to the herpes simplex virus-thymidine kinase gene/ganciclovir system. Therefore, expression of the UPRT gene in tumor cells followed by 5-FU administration is a possible strategy for cancer gene therapy, but potentiation of the bystander effect is required for its therapeutic application. Cancer Gene Therapy (2000) 7, 637– 643

Key words: Suicide gene; 5-fluorouracil; uracil phosphoribosyltransferase; protective immunity; gene therapy.

E

xpression of foreign gene(s) in tumor cells is a novel approach to cancer treatment. Enforcement of the immune surveillance system of the host using cytokine gene(s) is one such strategy, and can effectively kill cancer cells.1 Systemic immunity, when induced, inhibits the growth of metastatic or recurrent tumors. Another tempting procedure is the intratumoral expression of a suicide gene that is normally absent in mammalian cells. The suicide gene encodes an enzyme that can convert a nontoxic prodrug into a toxic drug and renders the transduced cells susceptible to corresponding prodrug(s).2 This system can achieve a high concentration of the toxic product in the local milieu while avoiding the

Received March 18, 1999; accepted September 9, 1999. Address correspondence and reprint requests to Dr. Masatoshi Tagawa, Division of Pathology, Chiba Cancer Center Research Institute, 666-2 Nitona, Chuo-ku, Chiba 260-8717 Japan. E-mail address: mtagawa@ chiba-cc.pref.chiba.jp © 2000 Stockton Press 0929-1903/00/$12.00/⫹0

Cancer Gene Therapy, Vol 7, No 4, 2000: pp 637– 643

systemic toxicity that limits the use of conventional chemotherapeutic agents. For the treatment of cancer, this suicide gene/prodrug system theoretically requires the expression of foreign gene(s) in all the tumor cells. However, the cytotoxic effect of prodrug was observed not only in the genetically modified cells but also in nontransduced cells. This paradoxical outcome, known as the bystander effect, contributes to the increased efficacy of suicide gene/ prodrug systems.3,4 Although the mechanism of the bystander effect has not been fully elucidated, the passage of a toxic substance(s) through intercellular gap junctions has been shown to play a crucial role.5 Uracil phosphoribosyltransferase (UPRT) (EC 2.4.2.9) of bacterial origin catalyzes the conversion to uridine-5⬘monophosphate from uracil6 and is functionally equivalent to orotate phosphoribosyltransferase (EC 2.4.2.10) or uridine-5⬘-monophosphate synthase of mammalian cells.7 These enzymes can also mediate the conversion of 5-fluorouracil (5-FU), a widely used chemotherapeutic agent for various solid tumors such as breast and colon 637

638

KAWAMURA, TASAKI, HAMADA, ET AL: SUICIDE GENE THERAPY USING UPRT AND 5-FU

tumors, to 5-fluorouridine-5⬘-monophosphate (FUMP).8 This conversion is one of the requisite pathways to achieve the cytotoxic effect of 5-FU, because FUMP is further metabolized into two major cytotoxic substances, 5-fluoro2⬘-deoxyuridine-5⬘-monophosphate, which inhibits thymidylate synthase, and 5-fluorouridine-5⬘-triphosphate, which is incorporated into RNA. The UPRT gene that is absent in mammalian cells, when expressed in tumor cells, can be expected to contribute to the enhanced cytotoxic effect of 5-FU in the transduced cells. In this study, we tested the efficacy of the UPRT gene/5-FU system using murine colon carcinoma cells (Colon 26). Retrovirally transduced cells with the UPRT gene (Colon 26/UPRT) showed high susceptibility to 5-FU; in addition, the immune system of the host was also involved in the antitumoral effect of this suicide gene/prodrug system. MATERIALS AND METHODS

Cells and animals Female BALB/c and BALB/c nu/nu mice (6 to 8 weeks of age) were purchased from Japan SLC (Hamamatsu, Japan). Colon 26, a carcinogen-induced undifferentiated adenocarcinoma cell line,9,10 and RL male-1, an irradiation-induced lymphoma line of BALB/c origin,11 were provided by Dr. J. Hamuro (Ajinomoto, Tokyo, Japan) and Dr. T. Ebina (Miyagi Cancer Center, Natori, Japan), respectively. Both ecotropic ␺2 and amphotropic PA317 packaging cells were obtained from the American Type Culture Collection (Manassas, Va).

Establishment of retrovirally transduced Colon 26 cells The polymerase chain reaction method was used to clone the Escherichia coli UPRT gene using K12 DH5␣-derived genomic DNA and two primers, 5⬘-GCGAATTCCACCATGAAGATCGTGGAAGTCAAACAC-3⬘ (as a 5⬘ primer) and 5⬘GGCGGATCCTTATTTCGTACCAAAGATTTTGTCACCGG-3⬘ (as a 3⬘ primer). Amplification was performed according to the manufacturer’s recommendations (Perkin-Elmer Cetus, Norwalk, Conn) and consisted of 30 cycles under the following conditions: 1 minute at 94°C for denaturation, 1 minute at 55°C for primer annealing, and 1 minute at 72°C for primer extension. The sequence of the polymerase chain reaction product was confirmed to be identical with the published sequence.12 The retrovirus vector LXSN (provided by Dr. A. D. Miller, Fred Hutchinson Cancer Research Center, Seattle, Wash)13 was used to harbor the cloned UPRT gene. The vector DNA with the UPRT gene was transfected into ecotropic ␺2 cells using Lipofectin reagent (Life Technologies, Gaithersburg, Md); after the drug selection with G418 (Life Technologies), cell-free supernatants of G418-resistant clones were used as a stock. The culture supernatants containing retrovirus were incubated with amphotropic PA317 cells in the presence of 8 ␮g/mL polybrene (Aldrich, Milwaukee, Wis) for infection. Among the G418-resistant PA317 cells, a clone that expressed the highest level of UPRT mRNA was selected, and the culture supernatants were used to infect Colon 26 cells. G418-resistant cloned Colon 26 cells were examined for their sensitivity to 5-FU. For the transduction with herpes simplex virus-thymidine kinase (HSV-TK) or ␤-galactosidase (␤-gal) gene, the retrovirus bearing HSV-TK or ␤-gal gene (Genetic Therapy Inc., Gaithersburg, Md) was used to infect Colon 26 cells as described above. A G418-resistant clone that showed

increased sensitivity to ganciclovir (GCV) and proliferated at the same rate as wild-type (WT) cells (Colon 26/HSV-TK) was used, as were G418-resistant cells transduced with ␤-gal gene (Colon 26/␤-gal).

Assay for FUMP synthesis The enzymatic activity of UPRT was measured based the method of Laskin et al.14 Cell extracts were incubated in a solution containing 10 mM tris(hydroxymethyl)aminomethane-HCl (pH 7.5), 10 mM MgCl2, 10 mM phosphoribosylpyrophosphate (Sigma, St. Louis, Mo), 2 mM 2-glycerophosphate, 75 ␮M ␣,␤-methylene-adenosine-diphosphate, 2 mM 5-FU, and 0.875 ␮M [6-3H]5-FU at 37°C. An aliquot of the reaction mixture was applied to a diethylaminoethyl-cellulose ion exchange filter, and the radioactivity of FUMP was measured by a liquid scintillation counting after extensive washing.

In vitro sensitivity to 5-FU WT or transduced cells were seeded in 96-well plates (5 ⫻ 102 cells/well) and incubated for 5 days with various concentrations of 5-FU. Cell viability was assayed with a tetrazolium-based colorimetric assay (Dojindo Chemicals, Kumamoto, Japan). The amounts of formazan produced were determined from the absorbance at 450 nm.

Animal studies WT or transduced Colon 26 cells (1 ⫻ 106) were injected subcutaneously (s.c.) into BALB/c or BALB/c nu/nu mice. From day 5 to day 19, 30 mg/kg 5-FU or phosphate-buffered saline (PBS) was administered intraperitoneally (i.p.) to mice every other day. Tumor volume was calculated according to the following formula: 1/2 ⫻ length ⫻ (width2). The statistical analysis was performed by one-way analysis of variance. The mice that had rejected the Colon 26/UPRT cells were challenged s.c. with WT cells (1 ⫻ 106) or syngeneic lymphoma RL male-1 cells (1 ⫻ 106). Mice were assessed for survival, and analysis was conducted by the Kaplan-Meier test. Statistical analysis was performed by the log rank test.

Bystander effect Mixed populations of WT and Colon 26/UPRT cells at various ratios (1 ⫻ 106 in total) were inoculated s.c. into BALB/c mice, and 5-FU (30 mg/kg) was administered i.p. as described above. For the in vitro bystander effect, variously mixed populations of WT and Colon 26/UPRT cells or WT and Colon 26/HSV-TK cells were seeded at 5 ⫻ 103 cells/well in 96-well plates and cultured for 4 days with 5-FU (0.1 ␮g/mL) or GCV (100 ␮g/mL), respectively. Viability of cells was determined with the tetrazolium-based colorimetric assay.

RESULTS In vitro analysis of the UPRT gene-transduced Colon 26 cells Colon 26 cells transduced with the UPRT gene were cloned, and the in vitro sensitivity to 5-FU was examined. We selected several clones that showed increased sensitivity to 5-FU (Fig 1). The sensitivity of the Colon 26/␤-gal cells remained the same as that of WT cells (Fig 1). All of the clones proliferated in vitro at the same rate as WT cells (data not shown), but the expression

Cancer Gene Therapy, Vol 7, No 4, 2000


Figure 1. In vitro sensitivity of WT cloned Colon 26/UPRT (Clones A–D) or Colon 26/␤-gal (␤-gal) to various concentrations of 5-FU.

levels of class I antigens (Ags) of major histocompatibility complex analyzed by flow cytometry were variable. Therefore, we selected a clone (Colon 26/UPRT, Clone A in Fig 1) whose expression level of class I Ags was comparable with that of WT cells (data not shown). The activities to convert 5-FU into FUMP of WT and Colon 26/UPRT cells were 2.0 ⫾ 0.18 and 4.6 ⫾ 0.55 nmol/ min/mg of protein, respectively. Expression of the UPRT gene in Colon 26/UPRT cells was also confirmed by Northern blot analysis (data not shown).

639

the tumor growth rate between the two groups was not different (Fig 2A). We also examined whether acquired immunity could be generated in the mice that had eliminated Colon 26/UPRT cells with 5-FU. After the tumors disappeared, the mice were challenged with a tumorigenic dose of WT Colon 26 cells or irrelevant syngeneic lymphoma cells (RL male-1). All of the mice rejected the WT Colon 26 cells but not the RL male-1 cells (Table 2). The survival of the mice that had rejected the Colon 26/UPRT cells and subsequently received RL male-1 cells was not different from that of naive mice that were inoculated with RL male-1 cells. The generation of cell-type specific, acquired immunity suggested that the amount of 5-FU used in this study did not impair T-cell-mediated immunity in immunocompetent mice. To examine a possible involvement of T cells in in vivo cytotoxicity by the UPRT/5-FU system, we used BALB/c nude that which lack mature ␣␤ T cells. Tumors of Colon 26/UPRT cells in nude mice grew slightly faster than those in immunocompetent mice, as did WT tumors (Fig 2, A and B), and the growth was significantly suppressed by the administration of 5-FU (Fig 2B). However, the tumors of Colon 26/UPRT cells in nude mice that were treated with 5-FU did not shrink, in contrast to those in 5-FU-injected BALB/c mice. The same results were obtained with separate experiment: injection of 5-FU was effective to suppress the growth of Colon 26/UPRT tumors in immunocompetent mice but was less effective in syngeneic nude mice (Fig 2C).

In vivo cytotoxic effect of 5-FU

Bystander effect by the UPRT/5-FU system

To examine the antitumoral effect of 5-FU on the UPRT gene-transduced cells, we inoculated WT or transduced Colon 26 cells into syngeneic BALB/c mice. After s.c. tumors became visible (day 5), 5-FU or PBS as a control was administered every other day from day 5 to day 19. The tumor growth of Colon 26/UPRT cells was slightly retarded compared with that of WT or Colon 26/␤-gal cells when the mice were injected with PBS (Fig 2A). However, the growth of Colon 26/UPRT tumors in the 5-FU-injected mice was significantly impeded compared with that in the PBS-injected group (P ⬍ .01) (Fig 2A). The tumors completely disappeared in five of eight mice without overt signs of toxicity and did not develop until the end of the observation period (day 100) (Table 1). In the rest of the mice, tumor growth remained suppressed until about day 35, but the tumors regrew again thereafter. In contrast, the tumor growth rate of WT or Colon 26/␤-gal cells was not statistically different between the 5-FU-injected and the PBS-injected groups. Consequently, the survival of the mice that received Colon 26/UPRT and were treated with 5-FU (Colon 26/UPRT plus 5-FU group) was significantly longer than that of the mice in the WT plus 5-FU group and that of the mice in the Colon 26/UPRT plus PBS group (Table 1). The survival of the mice in the WT plus 5-FU group was slightly longer than that of the mice in the WT plus PBS group (.01 ⬍ P ⬍ .05) due to the effect of 5-FU, although

We examined the bystander effect of the UPRT/5-FU system in vitro and in vivo. For the comparison of the efficacy of the bystander effect, another suicide gene system using HSV-TK gene and GCV was also tested in Colon 26 cells. Colon 26/UPRT or Colon 26/HSV-TK cells were mixed with WT cells at various ratios, and the viability of the mixed cells under confluent conditions was examined in the presence of 5-FU or GCV. The viability of a population containing a greater percentage of Colon 26/HSV-TK cells disproportionately decreased due to the bystander effect (Fig 3A). However, the bystander effect of the 5-FU/UPRT system was less noticeable than that of the HSV-TK/GCV system (Fig 3A). We also inoculated immunocompetent mice with the mixed population of Colon 26/UPRT and WT and treated them with 5-FU. Although the tumors from 100% of Colon 26/UPRT cells were eliminated, the tumor growth of the mixed populations was not suppressed proportionately to the percentage of Colon 26/UPRT cells (Fig 3B).


DISCUSSION In this study we examined, whether expression of the UPRT gene in tumor cells could confer increased sensitivity to 5-FU. Although we examined ⬎20 clones of

640


Figure 2. Tumor growth of WT, Colon 26/␤-gal (␤-gal), or Colon 26/UPRT (UPRT) cells in syngeneic BALB/c mice (n ⫽ 8) (A) or BALB/c nude mice (n ⫽ 8) (B). Mice received 5-FU (30 mg/kg/day) or PBS every other day from day 5 to day 19. Average tumor volumes with SE bars are shown. C: Change in the percentage of tumor volumes after the start of 5-FU administration in a separate experiment. The tumor volumes on day 5 are adjusted to 100%.

differential outcome between the in vitro and the in vivo study suggested that a mechanism(s) other than the pharmacological actions of 5-FU metabolites could be involved in the in vivo antitumoral effect. The immune system of the host is a possible mechanism, and in fact this suicide gene system is less effective in mature ␣␤ T-cell-defective nude mice (Fig 2, B and C). Generation

UPRT transcript-positive cells, all of them showed a ⬍10-fold increase in in vitro sensitivity to 5-FU compared with that of WT or Colon 26/␤-gal cells (Fig 1). The antitumoral effect of 5-FU, however, was evidently produced in vivo; 5-FU did not affect the tumor growth of WT or Colon 26/␤-gal cells but significantly suppressed the tumor growth of Colon 26/UPRT cells (Fig 2A). This

Table 1. Survival of Mice Inoculated with WT or Colon 26/UPRT Cells with or without 5-FU Treatment Cells inoculated

Treatment

n

WT Colon 26/UPRT WT Colon 26/UPRT

PBS PBS 5-FU 5-FU

7 7 8 8

Survival days 26, 21, 31, 53,

30, 28, 33, 72,

33, 30, 47, 88,

41, 48, 56, 63*‡ 31, 39, 54, 76*§ 57, 66, 68, 76, 77†‡ ⬎100, ⬎100, ⬎100, ⬎100, ⬎100†§

*␹2 ⫽ 0.032, df ⫽ 1, P ⫽ .858 (log rank test). †␹2 ⫽ 9.53, df ⫽ 1, P ⫽ .002 (log rank test). ‡␹2 ⫽ 4.12, df ⫽ 1, P ⫽ .0424 (log rank test). §␹2 ⫽ 11.4, df ⫽ 1, P ⫽ .0007 (log rank test).



Table 2. Ag-Specific Protective Immunity Generated in Mice that Rejected Colon 26/UPRT cells with 5-FU Challenged Cells inoculated

Survival days (after

Treatment cells (day 50)

the last inoculation)

Colon 26/UPRT

5-FU

Colon 26

⬎100, ⬎100, ⬎100, ⬎100, ⬎100

Colon 26/UPRT

5-FU

RL male-1

26, 28, 28, 30, 32*

—

—

Colon 26

27, 29, 30, 30, 49

—

—

RL male-1

24, 24, 25, 28, 31*

*␹2 ⫽ 1.29, df ⫽ 1, P ⫽ 0.256 (log rank test).

of tumor-specific protective immunity was also observed in the mice that had eliminated Colon 26/UPRT cells with 5-FU (Table 2), suggesting that Ag-specific cytotoxic T cells were induced.

Figure 3. The bystander effect in vitro (A) and in vivo (B). A: Variously mixed populations of Colon 26/HSV-TK and WT cells (5 ⫻ 103 cells in total) or Colon 26/UPRT and WT cells (5 ⫻ 103 cells in total) were tested for their sensitivity to GCV (100 ␮g/mL) or 5-FU (0.1 ␮g/mL). SE bars are also shown. B: Mixed populations of WT and Colon 26/UPRT cells (1 ⫻ 106 cells in total) were inoculated into BALB/c mice (n ⫽ 6), and 5-FU was injected i.p. every other day from day 5 to day 19.


641

The system described in this study is analogous to the combination of non-mammalian cytosine deaminase (CD) gene and the prodrug 5-fluorocytosine (5-FC).15–19 The CD gene of bacterial origin catabolizes 5-FC to 5-FU; consequently, the CD-positive mammalian cells can be sensitized to nontoxic 5-FC. These suicide gene/ prodrug systems share a similar mechanism of action; however, the properties of the enzyme-mediated products are different. The FUMP produced in the UPRT/ 5-FU system is membrane-impermeable, whereas the 5-FU synthesized by the CD gene is diffusible.19,20 The difference in permeability reflects the mode of the bystander effect. The 5-FU produced, which is released into the extracellular space, can damage neighboring cells without cell-to-cell contact, whereas the FUMP produced in the cytoplasm can kill the adjacent cells through cell-to-cell communication. The bystander effect of the HSV-TK/GCV system has been well documented.3 Although several mechanisms of the HSV-TK-mediated bystander effect have been postulated, the transfer of phosphorylated GCV through intercellular gap junctions seems to play a crucial role.21 For a comparison of the efficacy of the bystander effect between the HSV-TK/GCV and the UPRT/5-FU system, we also established Colon 26/HSV-TK cells and examined the bystander effect of each system (Fig 3A). A significant HSV-TK-mediated bystander effect using Colon 26 cells was also reported in another study.22 Compared with the bystander effect of the HSV-TK system, the UPRT-mediated effect was less significant in Colon 26 cells (Fig 3A). Moreover, an in vivo bystander effect in s.c. tumors was not observed (Fig 3B). The variance in diffusibility between phosphorylated metabolites of GCV and those of 5-FU might explain the difference in the bystander effect between the two systems. Because the expression of connexin 43, which constitutes a major part of gap junctions, is relatively weak in Colon 26 cells (data not shown), increased expression of connexin 43 gene may facilitate the transfer of FUMP and consequently enhance the bystander effect. The expression level of the transduced suicide gene in Colon 26/UPRT or Colon 26/HSV-TK cells is also possibly attributable to the differential bystander effect observed. Kanai et al23 recently reported the significant bystander effect achieved by adenovirus-mediated expression of the UPRT gene, suggesting that the elevated expression of the UPRT gene can increase the efficacy of the bystander effect. In contrast to our results, Rogers et al16 have shown that the bystander effect examined with human B-lymphoma lines was greater in the CD/5-FU system than in the HSV-TK/GCV system. This could be due to the different properties of the respective enzyme-mediated metabolites. The involvement of immunity in the CD/5-FC system has not been analyzed extensively, despite a number of reports regarding CD gene-mediated gene therapy. Although Consalvo et al15 suggested the requirement of host immune competence for effective antitumoral activity in the CD/5-FC system, effective treatment by the CD/5-FC system was also demonstrated in immunocom-

642


promised animals, including nude and severe combined immunodeficient mice.16 –18 This discrepancy is probably due to how efficiently the gene-expressing cells can metabolize 5-FC. If the conversion of 5-FC and 5-FU is efficient, the action by the immune system of the host is masked and the tumor cell killing appears to be mediated by the direct cytotoxic action of the metabolites. In that sense, the endogenous activity of orotate phosphoribosyltransferase in Colon 26 cells is relatively high compared with that of other cells;8,24 consequently, the UPRT/5-FU system is less efficient in Colon 26 cells. However, when the cytotoxic effect caused by active drug(s) can kill a part of tumors, the immune system of the host, stimulated by successful Ag presentation of putative tumor Ag(s), will effectively attack tumor cells. The immunogenicity of transduced tumor cells may be important to generate antitumoral immunity. The Colon 26 cells used in this study are relatively immunogenic, because mice that received irradiated WT Colon 26 cells could reject WT cells when subsequently challenged (data not shown). The generation of acquired immunity in host animals has also been reported when HSV-TKtransduced tumors were eliminated with GCV.25 Regardless of the method of tumor cell killing, protective immunity as a secondary immune response can be induced by the successful destruction of tumors, unless host immunity is not impaired by the process of tumor cell killing. The less significant bystander effect of the UPRT/ 5-FU system leads us to consider an additional strategy to attain complete elimination of tumor cells. The expression of both CD and UPRT genes in tumor cells, which facilitates the conversion of 5-FC to toxic metabolites of 5-FU, is a possible approach to enhance the cytotoxic effect. Stimulation of the host’s defense mechanism is another procedure (e.g., through the combinatory expression of suicide gene(s) and cytokine gene(s)) that can augment the activity of natural killer and/or cytotoxic killer cells.26 Efficient Ag processing is also a essential to obtaining an optimal activation of antitumoral immunity. Ramesh et al27 have shown that the tumor cell killing by the HSV-TK/GCV system upregulated the expression of B7 and intercellular adhesion molecule-1 molecules, which play a major role in T-cell activation, although tumor cell killing by suicide gene therapy may not always be favorable to Ag presentation. The activation of Ag-presenting cells, such as dendritic cells, through CD40 ligand may therefore contribute to a better therapeutic outcome in the application of suicide gene therapy for cancer.28 ACKNOWLEDGMENTS We thank Tamotsu Hashimoto (Pharmaceutical Research Institute, Kyowa Hakkou Kogyo, Shizouka, Japan) for technical assistance. This work was supported by a grant-in-aid for scientific research and a grand-in-aid for scientific research on priority areas from the Ministry of Education, Science, Sports, and Culture of Japan, as well as by a grant from the Hamaguchi Foundation for the Advancement of Biochemistry.

REFERENCES 1. Tepper RI, Mule´ JJ. Experimental and clinical studies of cytokine gene-modified tumor cells. Hum Gene Ther. 1994;5:153–164. 2. Freeman SM, Whartenby KA, Freeman JL, Abboud CN, Marrogi AJ. In situ use of suicide genes for cancer therapy. Semin Oncol. 1996;23:31– 45. 3. Freeman SM, Abboud CN, Whartenby KA, et al. The “bystander effect”: tumor regression when a fraction of the tumor mass is genetically modified. Cancer Res. 1993;53: 5274 –5283. 4. Pope IM, Poston GJ, Kinsella AR. The role of the bystander effect in suicide gene therapy. Eur J Cancer. 1997;33:1005–1016. 5. McMaster RA, Saylors RL, Jones KE, et al. Lack of bystander killing in herpes simplex virus thymidine kinasetransduced colon cell lines due to deficient connexin43 gap junction formation. Hum Gene Ther. 1998;9:2253–2261. 6. Jensen KF, Mygind B. Different oligomeric states are involved in the allosteric behavior of uracil phosphoribosyltransferase from Escherichia coli. Eur J Biochem. 1996; 240:637– 645. 7. Suttle DP, Bugg BY, Winkler JK, Kanalas JJ. Molecular cloning and nucleotide sequence for the complete coding region of human UMP synthase. Proc Natl Acad Sci USA. 1988;85:1754 –1758. 8. Peters GJ, Braakhuis BJM, de Bruijn EA, et al. Enhanced therapeutic efficacy of 5⬘deoxy-5-fluorouridine in 5-fluorouracil resistant head and neck tumours in relation to 5-fluorouracil metabolising enzymes. Br J Cancer. 1989;59: 327–334. 9. Corbett TH, Griswold DP Jr, Roberts BJ, Peckham JC, Schabel FM Jr. Tumor induction relationships in development of transplantable cancers of the colon in mice for chemotherapy assays, with a note on carcinogen structure. Cancer Res. 1975;35:2434 –2439. 10. Brattain MG, Strobel-Stevens J, Fine D, Webb M, Sarrif AM. Establishment of mouse colonic carcinoma cell lines with different metastatic properties. Cancer Res. 1980;40: 2142–2146. 11. Ebina T, Murata K. Antitumor effect of PSK at a distant site: tumor-specific immunity and combination with other chemotherapeutic agents. Jpn J Cancer Res. 1992;83:775– 782. 12. Andersen PS, Smith JM, Mygind B. Characterization of the upp gene encoding uracil phosphoribosyltransferase of Escherichia coli K12. Eur J Biochem. 1992;204:51–56. 13. Miller AD, Rosman GJ. Improved retroviral vectors for gene transfer and expression. Biotechniques. 1989;7:980 – 990. 14. Laskin JD, Evans RM, Slocum HK, et al. Basis for natural variation in sensitivity to 5-fluorouracil in mouse and human cells in culture. Cancer Res. 1979;39:383–390. 15. Consalvo M, Mullen CA, Modesti A, et al. 5-Fluorocytosine-induced eradication of murine adenocarcinomas engineered to express the cytosine deaminase suicide gene requires host immune competence and leaves an efficient memory. J Immunol. 1995;154:5302–5312. 16. Rogers RP, Ge J-Q, Holley-Guthrie E, et al. Killing Epstein-Barr virus-positive B lymphocytes by gene therapy: comparing the efficacy of cytosine deaminase and herpes simplex virus thymidine kinase. Hum Gene Ther. 1996;7:2235–2245. 17. Ge K, Xu L, Zheng Z, et al. Transduction of cytosine



18.

19.

20. 21.

22. 23.

deaminase gene makes rat glioma cells highly sensitive to 5-fluorocytosine. Int J Cancer. 1997;71:675– 679. Li Z, Shanmugam N, Katayose D, et al. Enzyme/prodrug gene therapy approach for breast cancer using a recombinant adenovirus expressing Escherichia coli cytosine deaminase. Cancer Gene Ther. 1997;4:113–117. Lawrence TS, Rehemtulla A, Ng EY, et al. Preferential cytotoxicity of cells transduced with cytosine deaminase compared to bystander cells after treatment with 5-fluorocytosine. Cancer Res. 1998;58:2588 –2593. Domin BA, Mahony WB, Zimmerman TP. Transport of 5-fluorouracil and uracil into human erythrocytes. Biochem Pharmacol. 1993;46:503–510. Elshami AA, Saavedra A, Zhang H, et al. Gap junctions play a role in the “bystander effect” of the herpes simplex virus thymidine kinase/ganciclovir system in vitro. Gene Ther. 1996;3:85–92. Link CJ Jr, Levy JP, McCann LZ, Moorman DW. Gene therapy for colon cancer with the herpes simplex thymidine kinase gene. J Surg Oncol. 1997;64:289 –294. Kanai F, Kawakami T, Hamada H, et al. Adenovirus-


24. 25.

26.

27.

28.

643

mediated transduction of Escherichia coli uracil phosphoribosyltransferase gene sensitizes cancer cells to low concentrations of 5-fluorouracil. Cancer Res. 1998;58:1946 – 1951. Inaba M, Mitsuhashi J, Sawada H, et al. Reduced activity of anabolizing enzymes in 5-fluorouracil-resistant human stomach cancer cells. Jpn J Cancer Res. 1996;87:212–220. Barba D, Hardin J, Sadelain M, Gage FH. Development of anti-tumor immunity following thymidine kinase-mediated killing of experimental brain tumors. Proc Natl Acad Sci USA. 1994;91:4348 – 4352. Bonnekoh B, Greenhalgh DA, Chen SH, et al. Ex vivo and in vivo adenovirus-mediated gene therapy strategies induce a systemic anti-tumor immune defense in the B16 melanoma model. J Invest Dermatol. 1998;110:867– 871. Ramesh R, Munshi A, Abboud CN, Marrogi AJ, Freeman SM. Expression of costimulatory molecules: B7 and ICAM up-regulation after treatment with a suicide gene. Cancer Gene Ther. 1996;3:373–384. Lanzavecchia A. Immunology: Licence to kill. Nature. 1998;393:413– 414.