Combined suicide gene therapy for human colon cancer cells ... - Nature

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

Combined suicide gene therapy for human colon cancer cells using adenovirus-mediated transfer of Escherichia coli cytosine deaminase gene and Escherichia coli uracil phosphoribosyltransferase gene with 5-fluorocytosine Fumikazu Koyama,1 Hidetomo Sawada,1 Tomoko Hirao,1 Hisao Fujii,1 Hirofumi Hamada,2 and Hiroshige Nakano1 First Department of Surgery, Nara Medical University, Nara, Japan;1 and Department of Molecular Biotherapy Research, Cancer Chemotherapy Center, Cancer Institute, Tokyo, Japan.2 The virus-directed enzyme/prodrug system using the Escherichia coli cytosine deaminase (CD) gene and 5-fluorocytosine (5-FC) suffers from a sensitivity limitation in many tumor cells. The E. coli uracil phosphoribosyltransferase (UPRT), which is a pyrimidine salvage enzyme, directly converts 5-fluorouracil (5-FU) to 5-fluorouridine monophosphate at the first step of its activating pathway. To improve the antitumoral effect of the CD/5-FC system, we investigated a combined suicide gene transduction therapy for human colon cancer cells using two separate adenovirus vectors expressing the E. coli CD and E. coli UPRT genes and systemic 5-FC administration (the CD, UPRT/5-FC system). The present study demonstrates that the CD, UPRT/5-FC system generates a co-operative effect of CD and UPRT, resulting in dramatic increases in both RNA- and DNA-directed active forms, including 5-fluorouridine triphosphate incorporated into RNA, 5-fluorodeoxyuridine monophosphate, and the thymidylate synthase inhibition rate, compared with the CD/5-FC system. Furthermore a significant increase in the 5-FC sensitivity of colon cancer cells was demonstrated in the CD, UPRT/5-FC system compared with the CD/5-FC system in vitro and in vivo. These results suggest that the CD, UPRT/5-FC system is a powerful approach in gene therapy for colorectal cancer. Cancer Gene Therapy (2000) 7, 1015–1022

Key words: Gene therapy; cytosine deaminase; uracil phosphoribosyltransferase; colon cancer.

C

olorectal cancer has recently become a leading cause of death in Japan as well as in Western countries. 5-fluorouracil (5-FU) has been a key drug in the treatment of colorectal cancer since the late 1950s. It is anabolized to the active metabolites 5-fluorouridine triphosphate (FUTP) and 5-fluorodeoxyuridine monophosphate (FdUMP) by several steps of an enzyme reaction series and is catabolized to dihydrofluorouracil (FUH2) by dihydropyrimidine dehydrogenase (DPD) (Fig 1).1,2 FUTP is incorporated into RNA (F-RNA) and impairs the multiple functions of RNA, whereas FdUMP blocks the catalytic activity of thymidylate synthase (TS) by forming a ternary covalent complex with its cosubstrate, 5,10-CH2-FH4, which inhibits DNA synthesis. Conversely, DPD is the first and rate-limiting enzyme of the chain of reactions that regulate 5-FU catabolism. The recent biochemical modulation therapies of 5-FU, which targeted TS or DPD, increased the efficacy rate from 20% for a single use of 5-FU to 30 –50%, but could not improve the prognosis of coloReceived April 9, 1999; accepted February 11, 2000. Address correspondence and reprint requests to Dr. Hidetomo Sawada, First Department of Surgery, Nara Medical University, 840 Shijo-cho, Kashihara-city, Nara 634-8522, Japan.

Cancer Gene Therapy, Vol. 7, No 7, 2000: pp 1015–1022

rectal cancer patients with unresectable lesions.3– 6 The development of a new strategy is expected for the treatment of colorectal cancer. The cytosine deaminase (CD)/5-fluorocytosine (5FC) system is a suicide gene therapy for malignant tumors using a virus-directed enzyme/prodrug system.7–13 CD converts the nontoxic prodrug 5-FC to the cytotoxic agent 5-FU. Transduction of the CD gene and administration of 5-FC could produce a high concentration of 5-FU in the cells expressing the CD gene and the local milieu. This strategy has the potential to overcome the low therapeutic index of 5-FU and reportedly is effective in vitro and in vivo. Despite CD expression, a number of tumor cells were 5-FC-resistant, which may be attributable to the lack of an active cytosine transport system in mammalian cells14 and to the degradation of the formed 5-FU by DPD.2 In the gene transfer strategy, to improve the effect of the CD/5-FC system, it might be possible to transduce the enzyme gene that converts 5-FU to its active forms. One of the candidates is Escherichia coli uracil phosphoribosyltransferase (UPRT). It is a pyrimidine salvage enzyme and is characteristic to bacteria. It directly converts 5-FU to 5-fluorouridine monophosphate (FUMP) at the first step of 5-FU 1015

1016

KOYAMA, SAWADA, HIRAO, ET AL: COMBINED SUICIDE GENE THERAPY FOR HUMAN COLON CANCER

Figure 1. Metabolism of 5-FU and modulation mechanism of CD and UPRT. The plain, dotted, and bold arrows indicate the activating pathway, the inactivating pathway, and enzymatic steps by CD and UPRT, respectively. Mammalian cells cannot convert 5-FC to 5-FU and cannot directly convert 5-FU to FdUMP because of a lack of CD and UPRT, respectively. FUrd, 5-fluorouridine; FUDP, 5-fluorouridine diphosphate; FdUDP, 5-fluorodeoxyuridine diphosphate; FdUrd, 5-fluorodeoxyuridine.

activation and has the potential to enhance the activating pathway against DPD.15 Actually, adenovirusmediated UPRT gene transfer into the tumor cells can increase human cancer cell sensitivity to the cytotoxic effect of 5-FU in vitro and in vivo.16 Although it was reported in bacterial cells that concomitant expression of E. coli CD and UPRT improved the cytotoxity of 5-FC,17 there are no reports about combined gene transfer of CD and UPRT into human cancer cells with 5-FC. In the present study, we investigated the effect of the CD, UPRT/5-FC system on the growth of two human colon cancer cell lines in vitro and in vivo and analyzed its effect on both the RNA- and DNA-directed actions of 5-FU by measuring the intracellular levels of F-RNA and FdUMP and the TS inhibition rate. MATERIALS AND METHODS

Cell culture Two human colon cancer cell lines (HT29 and KM12) and a human embryonal kidney cell line (293) were cultured in Dulbecco’s modified Eagle’s medium/nutrient mixture F-12 (Ham 1:1) containing 10% fetal bovine serum at 37°C in a humidified atmosphere of 5% CO2 and 95% air. HT29 and 293 were obtained from the American Type Culture Collection (Manassas, Va), and KM12 was kindly provided by Dr. R. Lotan (MD Anderson Cancer Center, Houston, Tex).

Preparation of adenovirus vectors A replication-defective adenovirus lacking the E1 and E3 regions was used to construct the following three vectors: AdUPRT (adenovirus containing a CAG promoter (composed of a cytomegalovirus immediate early enhancer and a modified chiken ␤-actin promoter) and the E. coli UPRT coding domain), AdCD (adenovirus containing the above CAG promoter and the E. coli CD coding domain), and AdlacZ (adenovirus containing the CAG promoter and the lacZ gene).

Construction was performed by homologous recombination between the expression cosmid and the parental virus genome as described previously.8,16,18,19,20 In brief, the UPRT (upp) gene (GenBank accession number X57104) was amplified from E. coli K12 DH␣ F⬘ genomic DNA by polymerase chain reaction using the following primers: sense primer, 5⬘-gcgaattccaccATGAAGATCGTGGAAGTCAAACAC-3⬘; antisense primer, 5⬘-ggcggatccTTATTTCGTACCAAAGATTTTGTCACCGG-3⬘. The amplified upp gene fragment was subcloned to the EcoRI and BamHI site of pBluescript II SK(⫹) (Stratagene, La Jolla, Calif.), resulting in pBluescript II SK(⫹)-upp. The EcoRI- and BamHI-digested upp DNA fragment from pBluescript II SK(⫹)-upp was subcloned into the EcoRI and BglII site of pCAGGS,20 resulting in pCA-UPRT. The CAGUPRT-poly(A) expression subunit was cloned into the SwaI site of the pAx1cw cosmid,18,19 resulting in pAx1CA-UPRT. The pAx1CA-CD cosmid, which contains the CD gene driven by the ␤-actin-based CAG promotor,20 was constructed by inserting the CAG-CD-poly(A) expression unit into the ClaI site of pAx1cw. A recombinant adenovirus with the lacZ gene driven by the CAG promoter, AxCAlacZ (AdlacZ), was constructed as described previously.18 To produce a recombinant adenovirus, the cassette cosmid bearing the expression unit was cotransfected into 293 cells together with the adenovirus DNA-terminal protein complex digested at several sites to allow homologous recombination, followed by replication and encapsidation of the recombinant adenovirus DNA into infectious virions.19 Incorporation of the expression cassette into the isolated recombinant virus was confirmed by digestion with restriction enzymes.19 The recombinant viruses were subsequently propagated into 293 cells and recovered after 3 days of sonication. The viral solutions were stored at ⫺80°C until use. Viral titers were determined by plaque assay using 293 cells.18,19 The viral stocks used in these experiments did not contain detectable levels of replicationcompetent virus, as determined by polymerase chain reaction analysis using two pairs of primers specific to adenoviral E1A DNA, with coamplification of E2B DNA used as an internal control.21 We used each adenovirus vector at a multiplicity of infection (MOI) of 10 in in vitro studies for the two human colon cancer

Cancer Gene Therapy, Vol. 7, No 7, 2000


cell lines, for which the virus dose with a satisfactory transduction without severe cell toxicity had been determined by AdlacZ transduction assay22 (data not shown).

UPRT activity of AdUPRT-infected colon cancer cell lines UPRT activity was evaluated by quantifying the amount of F-RNA and FdUMP and the TS inhibition rate. Approximately 107 colon cancer cells infected with either AdUPRT or AdlacZ at a MOI of 10 were incubated at 37°C with 1 ␮M 5-FU for 24 hours. Next, the cells were collected and freezestored at ⫺80°C. The amount of F-RNA was measured by the gas chromatography-mass fragmentography method.23 FdUMP and TS assays were performed using the modified method of Moran and Spears.24,25 The results were calculated as the percentage of inhibition of TS, using the following formula: (1 ⫺ [free TS/total TS]) ⫻ 100.

1017

were injected with the conjugated viruses of AdCD (5 ⫻ 107 pfu) plus AdUPRT (5 ⫻ 107 pfu) in a total volume of 0.1 mL of PBS. Each tumor volume (TV) was measured in a blind fashion using calipers up to day 35. Especially in the AdCD with 5-FC group and AdCD plus AdUPRT with 5-FC group, the observation period was extended for 70 days after tumor inoculation. TV was estimated from the following formula: TV ⫽ a2 ⫻ b)/2, where a is the width (mm) of the tumor and b is its length (mm); tumor growth was evaluated from the relative TV (RTV) as follows: (RTV ⫽ TVn/TV7); TVn and TV7 are TVs at day n and day 7, respectively.

Statistical analysis Statistical analysis was performed with the Student’s t test (two-tailed) throughout the present study. A P value of ⬍.05 was considered statistically significant.

Combination effect of AdCD plus AdUPRT on 5-FC metabolism The combination effect of CD plus UPRT was assessed by quantifying the amount of F-RNA and FdUMP and the TS inhibition rate under administration of 5-FC. Approximately 107 colon cancer cells infected with AdlacZ, AdCD, or AdCD plus AdUPRT (combination rate; 1:1) at a total MOI of 10 were incubated at 37°C with 20 ␮M 5-FC for 24 hours. F-RNA, FdUMP, and the TS inhibition rate in the collected cells were measured by the methods described above.

In vitro growth assay in CD/5-FC, UPRT/5-FU, and CD plus UPRT/5-FC systems Human colon cancer cells were seeded and cultured in a 96-well plate at a density of 2.5 ⫻ 103 cells/well for 24 hours. Next, the cells were infected with either AdUPRT or AdlacZ at a MOI of 10 in the UPRT/5-FU system and with AdlacZ, AdCD, or AdCD plus AdUPRT (combination rate; 1:1) at a total MOI of 10 in the CD plus UPRT/5-FC system. After incubation for another 24 hours, the medium was replaced with fresh medium containing various concentrations of 5-FU in the UPRT/5-FU system and with fresh medium containing various concentrations of 5-FC in the CD plus UPRT/system. The cells were cultured at 37°C for another 5 days. The number of viable cells was assessed by the crystal violet staining procedure.22 Results are expressed as a growth ratio of the number of cells in plates containing drugs as a percentage of that in corresponding drug-free controls.

In vivo tumor treatment HT29 cells were taken from continuous culture and resuspended in phosphate-buffered saline (PBS) without fetal bovine serum for inoculation into athymic mice. Five-week-old male athymic BALB/cAnNCrj-nu/nu mice (Charles River, Kanagawa, Japan) were inoculated subcutaneously (s.c.) with 1 ⫻ 106 HT29 cells. At 7 days after tumor inoculation, the tumor-bearing mice were divided randomly into five groups with six tumors in each group, and the mice in each group were injected intratumorally (i.t.) with any of the following preparations: PBS, AdlacZ, AdUPRT, AdCD, and AdCD plus AdUPRT, with 5-FC administration. At 7 days after tumor inoculation, injections of the viruses of 108 plaque-forming units (pfu) in 0.1 mL of PBS were performed for 3 successive days, and 500 mg/kg 5-FC was injected intraperitoneally (i.p.) daily for 14 days. The mice in the AdCD plus AdUPRT group


RESULTS

Effect of AdUPRT on the 5-FU sensitivities of human colon cancer cell lines in vitro UPRT activity was evaluated by quantifying the intracellular F-RNA, FdUMP, and TS inhibition rate 24 hours after virus infection with 5-FU at a dose of 1 ␮M. Intracellular F-RNA levels in AdUPRT-infected cells were significantly higher than those in uninfected cells (14.0 times higher in HT29 cells and 13.3 times higher in KM12 cells) (P ⬍ .0001 in each cell line; Fig 2A). Intracellular FdUMP levels in AdUPRT-infected cells were also significantly higher than those in uninfected cells (9.4 times higher in HT29 cells and 8.8 times higher in KM12 cells) (P ⬍ .005 in each cell line; Fig 2B). Furthermore, TS inhibition rates in uninfected, AdlacZinfected, and AdUPRT-infected cells were 47.7 ⫾ 5.4, 53.0 ⫾ 3.5, and 57.1 ⫾ 3.5 in HT29 cells and 37.2 ⫾ 1.7, 36.0 ⫾ 2.6, and 47.0 ⫾ 5.8 (%) in KM12 cells, respectively (Fig 2C). There were small but significant differences between uninfected and AdUPRT-infected cells with regard to TS inhibition rates (P ⬍ .05 in each cell line). There were no differences in the levels of F-RNA and FdUMP and in TS inhibition rates between uninfected and AdlacZ-infected cells in both cell lines. Growth assays were performed to determine whether UPRT gene transduction by adenovirus vector increases the 5-FU sensitivity of human colon cancer cell lines. HT29 (Fig 3A) and KM12 (Fig 3B) cells transduced by AdUPRT increased 5-FU sensitivity; however, AdlacZ infection did not. The 50% inhibitory concentration (IC50) values of 5-FU in HT29 and KM12 cells were 0.30 and 0.080 ␮M in AdUPRT-infected cells and lower than 4.49 and 3.92 ␮M in uninfected cells, respectively. Thus AdUPRT infection induced sensitivities that were 15.1 times higher in HT29 cells and 48.8 times higher in KM12 cells compared with uninfected cells. These results demonstrate that adenovirus-mediated UPRT gene transduction sensitizes human colon cancer cells to a low concentration of 5-FU in vitro.

1018


Figure 2. Effect of AdUPRT infection on the 5-FU metabolism in vitro. Marked increases in the levels of intracellular F-RNA (A) and FdUMP (B) and in TS inhibition rates (C) were observed in both HT29 and KM12 cells after AdUPRT infection with 1 ␮M 5-FU compared with those after AdlacZ infection or no virus infection. Each column is the mean of three determinations, with the bar indicating the SD. P values indicate the differences between no virus infection and AdUPRT infection.

Effect of coinfection of AdCD and AdUPRT on the 5FC sensitivities of human colon cancer cell lines in vitro To observe whether CD and UPRT genes functioned effectively after coinfection of AdCD and AdUPRT, intracellular F-RNA (Fig 4A) and FdUMP (Fig 4B) and the TS inhibition rate (Fig 4C) were measured 24 hours after virus infection with 20 ␮M 5-FC in the following three groups: AdlacZ alone, AdCD alone, and AdCD plus AdUPRT. In

the AdlacZ alone group, F-RNA and FdUMP were under detectable levels in both cell lines. The levels of F-RNA and FdUMP and the TS inhibition rates were significantly higher in the group coinfected with AdCD plus AdUPRT compared with the group that received AdCD alone (9.5, 8.5, and 1.2 times higher in HT29 cells and 3.5, 16.7, and 1.8 times higher in KM12 cells, respectively). To evaluate the ability of the CD plus UPRT/5-FC system to suppress human colon cancer cell growth in comparison with the CD/5-FC system alone, in vitro growth

Figure 3. In vitro sensitivity to 5-FU after infection with AdUPRT. Significant sensitization to 5-FU was observed after AdUPRT infection in HT29 (A) and KM12 (B) cells. Each point on the ordinate corresponds to the percentage of surviving cells compared with uninfected cells without 5-FU as 100%. Results are expressed as the means ⫾ SD of four individual determinations.



1019

Figure 4. Effect of AdCD plus AdUPRT infection on 5-FC metabolism in vitro. Significant increases in the levels of intracellular F-RNA (A) and FdUMP (B) and in TS inhibition rates (C) were observed in HT29 and KM12 cells after coinfection with AdCD and AdUPRT at 20 ␮M 5-FC, compared with those after single infection with AdCD or AdlacZ. Each column is the mean ⫾ SD of three individual determinations. P values indicate the difference between AdCD infection and AdCD plus AdUPRT infection.

assays were performed after adenoviral infection at a total MOI of 10 with various concentrations of 5-FC (Fig 5, A and B). Virus treatment was divided into three groups: those infected with AdlacZ alone at a MOI of 10, with AdCD alone at a MOI of 10, and with AdCD plus AdUPRT, each at a MOI of 5. Marked cytotoxicities were observed in the AdCD treatment group even in absence of 5-FC, with a 29.3% and 41.0% suppression of cell growth in HT29 and KM12 cells, respectively, which revealed a viral toxity of AdCD. The AdCD and AdCD plus AdUPRT groups showed a 5-FC dose-dependent suppression of cancer cell growth; however, no suppression of cell

growth was observed in the AdlacZ alone groups even at 1000 ␮M 5-FC in both cell lines. The IC50 values of 5-FC in the AdCD and AdCD plus AdUPRT groups were 45.3 and 4.2 ␮M in HT29 cells and 11.4 and 1.2 ␮M in KM12 cells, respectively. Thus, the combination of AdCD plus AdUPRT infection greatly increased 5-FC sensitivities compared with AdCD alone in both cell lines. In vivo treatment of s.c. transplanted human colon cancer cells in nude mice The effect of the combination of AdCD and AdUPRT with 5-FC was also studied in an in vivo experiment using

Figure 5. In vitro sensitivity to 5-FC following coinfection with AdCD and AdUPRT. Significant sensitization to 5-FC was observed after coinfection with AdCD and AdUPRT in HT29 (A) and KM12 (B) cells. Each point on the ordinate corresponds to the percentage of surviving cells compared with uninfected cells without 5-FC as 100%. Results are expressed as the mean ⫾ SD of four individual determinations.


1020


Figure 6. Growth suppression of HT29-derived s.c. tumors in vivo after the treatment described in Materials and Methods. To simplify the results, the data for PBS with 5-FC, lacZ with 5-FC, UPRT with 5-FC, CD with 5-FC, and CD plus UPRT with 5-FC were expressed as the mean ⫾ SD of RTVs (each, n ⫽ 6). Arrows indicate the vectors or PBS injection into tumors at days 7, 8, and 9. The horizontal arrow indicates 500 mg/kg 5-FC i.p. administration into 5-FC-treated mice daily for 14 days from day 7. The treatment with CD, UPRT/5-FC suppressed the tumor growth significantly compared with that seen for CD/5-FC from day 25 to day 70 (day 70, P ⬍ .0001) as well as that seen for the other control groups from day 14 to day 35 (day 35, P ⬍ .005).

a nude mouse s.c. system. For the treatment, PBS, AdlacZ, AdUPRT, AdCD, and AdCD plus AdUPRT were injected i.t. with 5-FC. The representative growth curve of the HT29 s.c. tumor after treatment is shown in Figure 6. Injection of AdCD alone or a combination of AdCD and AdUPRT with systemic 5-FC administration resulted in suppression of HT29 tumor growth; this suppression was not evident in PBS, AdlacZ, or AdUPRT-treated mice. Moreover, the RTV during treatment with AdCD or AdUPRT/5-FC was significantly different from that observed with AdCD/5-FC from day 25 until the end of the experiment (day 70; P ⬍ .0001). Three of six tumors disappeared after treatment with AdCD plus AdUPRT with 5-FC; however, no tumor disappeared in the AdCD alone group. These results indicate the efficacy of combined virus treatment in vivo. Throughout these experiments, there were no accidental deaths of the mice, and the differences in the body weight changes in each group were limited to ⬍20%; however, the mice fell into cahexia with tumor growth in the PBS- and AdlacZ-treated groups, respectively (data not shown). DISCUSSION Suicide gene therapy is now being widely investigated for the therapy of malignant tumors, to render target cells susceptible to a nontoxic prodrug that is converted into a toxic form by expression of the transferred gene. The CD/5-FC system has been developed as one such suicide gene therapy, and its effectiveness

in vitro and in vivo had been demonstrated in many cancer cell lines derived from the colon, stomach, liver, pancreas, breast, brain, and thyroid.7–13 This system has been under phase I study for metastatic colon carcinoma of the liver in the United States.26 Recent studies of this system have been concentrated on tumor-specific gene delivery using a tumor-specific promoter8,9 in combination with radiation27–30 or cytokine gene therapy31–33 to enhance the antitumoral effect. However, this system is restricted by a sensitivity limitation in many tumor cells, which is probably due to the lack of an active transport system in 5-FC14 and the immediate degradation of the formed 5-FU by DPD.2 From the point of view of 5-FU resistance, additive transfer of the gene for the enzyme that has the potential to activate 5-FU would improve the effect of this CD/5-FC system. In the present study, we demonstrated that the simultaneous transduction of E. coli CD and UPRT genes generated a cooperative effect, resulting in a dramatic increase in the 5-FC sensitivity of human colon cancer cells in vitro and in vivo compared with the transduction of the E. coli CD gene alone. Before analyzing the combination effect of the simultaneous transduction of the E. coli CD and UPRT genes, we analyzed the effect of AdUPRT infection on the metabolism of 5-FU in human colon cancer cells. 5-FU requires enzymatic conversion to a nucleotide, ribosylation, and phosphorylation to exert its cytotoxic activity.1 The enzymatic activity of UPRT transduced by the E. coli UPRT gene, which converts 5-FU to FUMP directly in the first step of the 5-FU activating pathway, has been



reported previously.16,17 However, there were no reports about the effects of transduction of the UPRT gene on intracellular active metabolites of 5-FU, including FRNA and FdUMP. AdUPRT infection increased the intracellular F-RNA and FdUMP levels significantly in transduced cells; the latter induced marked TS inhibition. Thus, the UPRT/5-FU system had the potential to enhance both RNA- and DNA-directed actions of 5-FU, and could actually sensitize their cells to a low concentration of 5-FU in vitro. Furthermore, we also analyzed the combination effect of AdCD and AdUPRT infection on the metabolism of 5-FC in human colon cancer cells. In simultaneous infections with AdCD and AdUPRT (1:1, a total MOI of 10), the levels of F-RNA and FdUMP and the TS inhibition rates in 5-FC-treated cells were significantly higher than those seen for infection with AdCD alone at a MOI of 10 in both cell lines. These results indicated that 5-FU, which was converted from 5-FC by CD, was effectively metabolized into RNA- and DNA-directed actions by UPRT. In vitro growth assays showed that the CD, UPRT/ 5-FC system induced a 10.7-fold and 9.5-fold sensitivity in the IC50 values of 5-FC in HT29 and KM12 cells, respectively, compared with the CD/5-FC system. Furthermore, in the in vivo study, significantly higher growth inhibition of the s.c. tumor was observed in the CD, UPRT/5-FC system compared with the CD/5-FC system. To achieve an equivalent effect by the CD/5-FC system alone, a 10-fold higher dose of the virus than the CD, UPRT/5-FC system was required (data not shown). The protocols, which had been reported in the in vivo experimental CD/5-FC gene therapy by i.t. injection of adenovirus vector into s.c. transplanted cancer cells with or without radiation or cytokine gene therapy, had total virus doses ranging from 4 ⫻ 108 to 3 ⫻ 109 pfu combined with 5-FC administration i.p. ranging from 300 to 800 mg/kg/day for 10 –15 days.7,9,28,30,33 A total virus dose of 3 ⫻ 108 pfu combined with 5-FC administration at 500 mg/kg/day for 14 days, which we used in this system, might be one of the lowest doses among these protocols. In the present study, we used two separate vectors of AdCD and AdUPRT with the combination rate of 1:1, which could produce a dramatic increase in the 5-FC sensitivity of human colon cancer cells compared with AdCD alone. Further studies are necessary to investigate the most effective combination rate of the two separate vectors and the possibility of the adenovirus vector carring the fusion gene of E. coli CD and UPRT genes. Our data demonstrate that the association between CD and UPRT, which were transduced by the E. coli CD and UPRT genes, respectively, greatly sensitized colon cancer cells to 5-FC by channeling 5-FC to 5-FU and 5-FU to both RNA- and DNA-directed actions. The present study indicated the strong modulation effect of in vitro and in vivo tumor treatment, applying double suicide gene transduction to the 5-FU metabolism. We suggest the CD, UPRT/5-FC system as a powerful approach to gene therapy for colorectal cancer.


1021

REFERENCES 1. Pinedo HM, Peters GFJ. Fluorouracil: biochemistry and pharmacology. J Clin Oncol. 1988;6:1653–1664. 2. Eienne MC, Cheradame S, Fischel JL, et al. Response to fluorouracil therapy in cancer patients: the role of tumoral dihydropyrimidine dehydrogenase activity. J Clin Oncol. 1995;13:1663–1670. 3. Moertel CG. Chemotherapy for colorectal cancer. N Engl J Med. 1994;330:1136 –1143. 4. Casillas S, Pelly RJ, Milson JW. Adjuvant therapy for colorectal cancer: present and future perspectives. Dis Colon Rectum. 1997;40:977–992. 5. Blanke CD, Kasimis B, Schein P, et al. Phase II study of trimetrexate, fluorouracil, and leucovorin for advanced colorectal cancer. J Clin Oncol. 1997;15:915–920. 6. Levi F, Zidani R, Misset JL. Randomised multicentre trial of chronotherapy with oxaliplatin, fluorouracil, and folinic acid in metastatic colorectal cancer: International Organization for Cancer Chronotherapy. Lancet. 1997;350:681– 686. 7. Hirschowitz EA, Ohwada A, Pascal WR, et al. In vivo adenovirus-mediated gene transfer of the Escherichia coli cytosine deaminase to human colon carcinoma-derived tumors induces chemosensitivity to 5-fluorocytosine. Hum Gene Ther. 1995;6:1055–1063. 8. Lan K-H, Kanai F, Shiratori Y, et al. Tumor-specific gene expression in carcinoembryonic antigen-producing gastric cancer cells using adenovirus vectors. Gastroenterology. 1996;111:1241–1250. 9. Kanai F, Lan K-H, Shiratori Y, et al. In vivo gene therapy for ␣-fetoprotein producing hepatocellular carcinoma by adenovirus-mediated transfer of cytosine deaminase gene. Cancer Res. 1997;57:461– 465. 10. Evoy D, Hirschowitz EA, Naama HA, et al. In vivo adenoviral-mediated gene transfer in the treatment of pancreatic cancer. J Surg Res. 1997;69:226 –231. 11. Li Z, Shanmugan 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. 12. Ge K, Xu L, Zheng Z, et al. Transduction of cytosine deaminase gene makes rat glioma cells highly sensitive to 5-fluorocytosine. Int J Cancer. 1997;71:675– 679. 13. Nishihara E, Nagayama Y, Narimatsu M, et al. Treatment of thyroid carcinoma cells with four different suicide gene/prodrug combinations in vitro. Anticancer Res. 1998; 18:1521–1525. 14. Plagemann PG, Wohlhueter RM, Woffendin C. Nucleoside and nucleobase transport in animal cells. Biochim Biophys Acta. 1988;947:405– 443. 15. Andersen PH, Smith JM, Mygind B. Characterization of the upp gene encoding uracil phosphoribosyltransferase of Escherichia coli K12. Eur J Biochem. 1992;204:51–56. 16. Kanai F, Kawakami T, Hamada H, et al. Adenovirusmediated transduction of Escherichia coli uracil phosphoribosyltransferase gene sensitizes cancer cells to low concentrations of 5-fluorouracil. Cancer Res. 1998;58:1946 – 1951. 17. Tiraby M, Cazaux C, Baron M, et al. Concomitant expression of E. coli cytosine deaminase and uracil phosphoribosyltransferase improves the cytotoxicity of 5-fluorocytosine. Microbiol Lett. 1998;167:41– 49. 18. Kanegae Y, Lee G, Sato Y, et al. Efficient gene activation in mammalian cells by using recombinant adenovirus

1022

19.

20. 21. 22.

23.

24.

25.

26.


expressing site-specific Cre recombinase. Nucleic Acids Res. 1995;23:3816 –3821. Miyake S, Makimura M, Kanegae Y, et al. Efficient generation of recombinant adenoviruses using adenovirus DNA-terminal protein complex and a cosmid bearing full-length virus genome. Proc Natl Acad Sci USA. 1996; 93:1320 –1324. Niwa H, Yamamura KI, Miyazaki JI. Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene. 1991;108:193–200. Zhang W-W, Koch PE, Roth JA. Detection of wild-type contamination in a recombinant adenoviral preparation by PCR. Biotechniques. 1995;18:444 – 447. Sawada H, Hirao T, Watanabe A, et al. Efficiency of gene transfer into human gastric carcinoma cells using adenovirus vector. Progress in Gastric Cancer Research, Proceedings of The 2nd International Gastric Congress. Bologna, Italy: Monduzzi Editere; 1997:677– 681. Peters GJ, Noorduis P, Komissarov A, et al. Quantification of 5-fluorouracil incorporation into RNA of human and murine tumors as measured with a sensitive gas chromatography-mass spectrometry assay. Anal Biochem. 1995; 231:157–163. Moran RG, Spears CP, Heidelberger C. Biochemical determinants of tumor sensitivity to 5-fluorouracil: ultrasensitive methods for determination of 5-fluoro-2⬘-deoxyuridylate, 2⬘-deoxyuridylate, and thymidylate synthetase. Proc Natl Acad Sci USA. 1979;76:1456 –1460. Spears CP, Shahinian AH, Moran RG, et al. In vivo kinetics of thymidylate synthetase inhibition in 5-fluorouracil-sensitive and -resistant murine colon adenocarcinomas. Cancer Res. 1982;42:450 – 456. Crystal RG, Hirschowitz EA, Lieberman M, et al. Phase I study of direct administration of a replication-deficient adenovirus vector containing the E. coli cytosine deami-

27.

28.

29.

30.

31.

32.

33.

nase gene to metastatic colon carcinoma of the liver in association with the oral administration of the pro-drug 5-fluorocytosine. Hum Gene Ther. 1997;8:985–1001. Gabel M, Kim JH, Kolozsvary A, et al. Selective in vivo radiosensitization by 5-fluorocytosine of human colorectal carcinoma cells transduced with the E. coli cytosine deaminase (CD) gene. Int J Radiat Oncol Biol Phys. 1998;4:883– 887. Hanna NN, Mauceri HJ, Wayne JD, et al. Virally directed cytosine deaminase/5-fluorocytosine gene therapy enhances radiation response in human cancer xenografts. Cancer Res. 1997;57:4205– 4209. Szary J, Missol E, Tarnawski R, et al. Selective augmentation of radiation effects by 5-fluorocytosine on murine B16(F10) melanoma cells transfected with cytosine deaminase gene. Cancer Gene Ther. 1997;4:269 –272. Pederson LC, Buchsbaum DJ, Vickers SM, et al. Molecular chemotherapy combined with radiation therapy enhances killing of cholangiocarcinoma cells in vitro and in vivo. Cancer Res. 1997;57:4325– 4332. Mullen CA, Petropoulos D, Lowe RM. Treatment of microscopic pulmonary metastases with recombinant autologous tumor vaccine expressing interleukin 6 and Escherichia coli cytosine deaminase suicide genes. Cancer Res. 1996;56:1361–1366. Nanni P, De Giovanni C, Nicoletti G, et al. The immune response elicited by mammary adenocarcinoma cells transduced with interferon-␥ and cytosine deaminase genes cures lung metastases by parental cells. Hum Gene Ther. 1998;9:217–224. Cao X, Ju DW, Tao Q, et al. Adenovirus-mediated GM-CSF gene and cytosine deaminase gene transfer followed by 5-fluorocytosine administration elicit more potent antitumor response in tumor-bearing mice. Gene Ther. 1998;5:1130 –1136.
