Use of a herpes thymidine kinase/neomycin ... - Nature

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

Use of a herpes thymidine kinase/neomycin phosphotransferase chimeric gene for metabolic suicide gene transfer Fabio Candotti,1 Riad Agbaria,2 Craig A. Mullen,1 Renaud Touraine,1 Jan Balzarini,3 David G. Johns,2 and R. Michael Blaese1 1

Clinical Gene Therapy Branch, National Human Genome Research Institute, and 2Laboratory of Medicinal Chemistry, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892; and 3Rega Institute for Medical Research, Katholieke Universiteit Leuven, Leuven, Belgium. Metabolic suicide gene transfer is widely applied for gene therapy of cancer, and retroviral vectors expressing the herpes simplex virus thymidine kinase (HSV-tk) gene are commonly used in clinical trials. Most of these vectors contain positive selectable markers that undoubtedly facilitate the determination of viral titer and the identification of high-titer producer clones. However, the presence of additional transcriptional units may result in reduced expression of the gene of interest. The use of fusion genes expressing bifunctional proteins may help to overcome this problem. We have constructed a retroviral vector carrying the TNFUS69 chimeric gene, which originates from the fusion of the HSV-tk and neomycin phosphotransferase II genes, and evaluated the functional expression of the encoded fusion protein. In vitro, expression of the fusion gene conferred to target cells both resistance to neomycin and selective sensitivity to the antiherpetic drugs ganciclovir and (E)-5-(2-bromovinyl)-2⬘-deoxyuridine. Cells transduced with the fusion gene, however, showed reduced ability to phosphorylate ganciclovir compared with cells expressing the native HSV-tk. Therefore, although the fusion gene may be used as a constituent of retroviral cassettes for positive and negative selection in vitro, its usefulness for suicide gene transfer applications in vivo may depend upon the possibility of using (E)-5-(2-bromovinyl)-2⬘deoxyuridine in a clinical context. Cancer Gene Therapy (2000) 7, 574 –580

Key words: Retroviral vector; gene therapy; suicide gene; ganciclovir; (E)-5-(2-bromovinyl)-2ⴕ-deoxyuridine.

I

n the last several years, “suicide” gene transfer systems have been extensively investigated as a therapeutic approach of cancer (reviewed in Refs. 1 and 2). These systems are based on the transfer of genes encoding for non-mammalian metabolic enzymes, which confer a novel and selective chemosensitivity to the target cells by providing intracellular conversion of a relatively nontoxic prodrug to a toxic metabolite. Several suicide gene/prodrug systems are currently under investigation.3– 8 Among these, the metabolic suicide strategy based on the use of the herpes simplex virus type 1 thymidine kinase (HSV-tk) in association with ganciclovir (GCV) is probably the most widely studied. In this system, the HSV-tk provides specific phosphorylation of GCV to GCV monophosphate; other cellular kinases complete the subsequent phosphorylation steps to diphosphate and triphosphate (GCV-TP) GCV derivaReceived May 10, 1999; accepted August 29, 1999. Address correspondence and reprint requests to Dr. Fabio Candotti, Clinical Gene Therapy Branch, National Human Genome Research Institute, Building 10, Room 10C103, National Institutes of Health, 10 Center Drive MSC 1851, Bethesda, MD 20892-1851. E-mail address: [email protected]

574

tives. GCV-TP, an analog of guanosine-TP, is the ultimate toxic metabolite of the prodrug that inhibits cellular DNA polymerase and DNA synthesis, leading to cell death.9,10 Preclinical models have demonstrated significant antitumoral activity against several tumors (reviewed in Ref. 11), leading to a series of clinical gene therapy trials that are currently testing the clinical efficacy of the HSV-tk/GCV system as a cancer treatment. Preliminary results have demonstrated that retroviral and adenoviral vectors can be used in vivo to confer GCV sensitivity to cancer cells,12,13 but have also indicated the need to improve the overall efficacy of this suicide system. In vitro studies of other antiherpetic compounds that are potentially more potent than GCV in arresting the growth of HSV-tk-transfected cells are underway, searching for more and more effective tools for the selective elimination of cancer cells. Among the possible alternative drugs, (E)-5-(2-bromovinyl)-2⬘-deoxyuridine (BVdU) was recently demonstrated to be a very efficient substrate for HSV-tk; in addition, it has a strong cytostatic activity against HSV-tk-transfected murine mammary carcinoma cells in vitro. The mechanism of action of BVdU differs from that described for GCV in that the

Cancer Gene Therapy, Vol 7, No 4, 2000: pp 574 –580

CANDOTTI, AGBARIA, MULLEN, ET AL: HSV-tk/neo FUSION GENE FOR RETROVIRAL VECTORS

principal toxic effect induced by BVdU is the inhibition of the cellular thymidylate synthase by the BVdU 5⬘monophosphate generated by the phosphorylation of BVdU by HSV-tk.10,14 In addition to their important potential therapeutic applications, suicide genes could be also considered as a safety feature for those gene transfer methods (retrovirus- or adeno-associated virus-based vectors) that involve integration of foreign sequences into the host genome. The presence of a negative selection system in these vectors would be highly beneficial in the unfortunate case of an insertional mutagenic event. However, the routine inclusion of a suicide gene as a regular constituent in the vector cassette would most likely affect the expression of other essential vector components such as a positive selectable trait (e.g., neomycin resistance gene (neo)) as well as the exogenous (therapeutic) gene of interest. One possible way to overcome this drawback is offered by the utilization of fusion genes that can provide expression of multifunctional proteins, avoiding the problems created by the coexistence of multiple transcriptional units. In the present work, we have explored the applicability of the TNFUS69 chimeric gene,15 which is created by the fusion of HSV-tk and neo genes, in a retroviral-based suicide gene transfer system by evaluating the extent of the chemosensitivity induced in target cells to GCV and BVdU. MATERIALS AND METHODS

Chemicals GCV was purchased from Syntex Laboratories (Palo Alto, Calif); BVdU was either obtained from the Rega Institute for Medical Research (Leuven, Belgium) or purchased from Sigma (St. Louis, Mo). [methyl-3H]thymidine ([3H]Thd) (specific activity 6.7 Ci/mmol) was obtained from DuPont-New England Nuclear Research Products (Boston, Mass); [8-3H]GCV ([3H]GCV) (specific activity 22 Ci/mmol) and [2⬘-3H]BVdU ([3H]BVdU) (3 Ci/mmol) were provided by Moravek Biochemicals (Brea, Calif).

Plasmids and cell lines The TNFUS69 plasmid containing the fused sequences of HSV-tk and neo genes has already been described in detail15 and was a gift of Dr. R. Kucherlapati (Department of Molecular Genetics, Albert Einstein College of Medicine, Bronx, NY); a native copy of HSV-tk was derived from the pSPTK1 plasmid, which was a gift of Genetic Therapy Inc. (Gaithersburg, Md). The pBabe-puro retroviral vector16 was kindly provided by H. Land (Imperial Cancer Research Fund, London, UK) and used to subclone the HSV-tk and TNFU69 genes to obtain the BTK and BTNfus constructs, respectively. The retroviral vector pLNL6 (encoding only for neo)17 and the PA31718 packaging cell line were a gift of A. D. Miller (Fred Hutchinson Cancer Research Center, Seattle, Wash); MC 38 and MCA 205 are murine methylcholanthrene-induced colon adenocarcinoma and fibrosarcoma cell lines, respectively, which originated from a C57BL/6 female mouse;19 both were kindly donated by S. A. Rosenberg (Surgery Branch, National Cancer Institute); L-M cellular TK-negative cells (L-M(tk⫺)) were obtained from the American Type Culture Collection (Manassas, Va). PA317 and L-M(tk⫺) cells were maintained in

Cancer Gene Therapy, Vol 7, No 4, 2000

575

Dulbecco’s modified Eagle’s medium (Biofluids, Rockville, Md) supplemented with 10% fetal calf sera and 2 mM Lglutamine (D10); MC 38 and MCA 205 cells were cultured in RPMI 1640 (Life Technologies, Rockville, Md) also supplemented with 10% fetal calf sera and 2 mM L-glutamine (R10).

Transfections and transductions Retroviral producer cell lines were obtained by conventional calcium phosphate coprecipitation of 20 ␮g of plasmid DNA on PA317 packaging cell lines. Cells producing the BTK and BTNfus retroviral vectors were obtained by selecting transfected cells in the neomycin analog G418 (Geneticin; Life Technologies; 1.0 mg/mL of active metabolite). Titers of BTK and BTNfus retroviral supernatants were assessed by transduction of NIH 3T3 cells followed by puromycin selection (2.5 ␮g/mL; Calbiochem, La Jolla, Calif) and were estimated to be 5 ⫻ 105 and 2 ⫻ 105 colony-forming units/mL, respectively. Viral particle-containing supernatants were used for transductions of target cells in the presence of 5 ␮g/mL protamine (Sigma). After transduction, cells were subjected to selection with 2.5 ␮g/mL puromycin. Cells transduced with the LNL6 vector were selected in 1 mg/mL G418.

Assessment of neo gene expression and biological activity All experiments were performed using MC 38 cells unmodified or subjected to transduction with either the BTK or BTNfus vectors. To evaluate the amount of neomycin phosphotransferase II (NPT II) produced by the neo and the HSV-tk/neo fusion genes, we used an enzyme-linked immunosorbent assay (ELISA) based on a rabbit polyclonal antibody specific to the NPT II protein encoded by the Escherichia coli Tn5 (5 Prime33 Prime, Boulder, Colo) according to the manufacturer’s specifications. In addition, NPT II enzymatic activity was determined by in situ phosphorylation of the antibiotic kanamycin after polyacrylamide gel electrophoresis of crude cell extracts as described previously.20 Detection and quantification of radiolabeled phosphorylated kanamycin were performed using a PhosphorImager SI (Molecular Dynamics, Sunnyvale, Calif). To evaluate the neo biological effect in vitro, unmodified or transduced MC 38 cells were plated (104 cells/well) in triplicates in 96-well plates, exposed to different concentrations of G418, and incubated at 37°C in 5% CO2 for 24 hours. Cell proliferation was estimated at the end of the treatment period as a function of radioactive thymidine incorporation into cellular DNA after a 6-hour pulse with [3H]Thd (0.5 ␮Ci/well). Results are expressed as the percentage of the cell proliferation observed in the control (untreated) population.

HSV-tk functional activity The kinase activity provided by the native HSV-tk and the TNFUS69 genes was investigated by high performance liquid chromatography (HPLC) analysis of [3H]Thd, [3H]GCV, and [3H]BVdU phosphorylated metabolites produced by retrovirally transduced cells. Experiments assessing the production of [3H]Thd phosphorylated metabolites were performed using L-M(tk⫺) cells, unmodified or transduced with the BTK or BTNfus vectors. The generation of [3H]GCV and [3H]BVdU phosphorylated metabolites was studied in MC 38 cells, also unmodified, or after transduction with BTK or BTNfus. Cells (106/mL) were cultured at 37°C in 5% CO2 in the presence of [3H]Thd, [3H]GCV, or [3H]BVdU (2 ␮Ci/mL) for 15 minutes,

576


24 hours, and 5 hours, respectively. Cells were then harvested, and pellets were extracted with 66% methanol and heated at 95°C for 2 minutes. After centrifugation, the clarified supernatants were evaporated under vacuum and redissolved in water. Reconstituted samples were subjected to HPLC using a Partisphere 5-SAX column (Whatman, Clifton, NJ) as described previously.21 Eluted, radiolabeled, phosphorylated thymidine, GCV, and BVdU metabolites were monitored with an in-line radioactivity flow detector and quantitated from the areas under the respective elution peaks. The precipitates of the methanol extractions (containing [3H]Thd, [3H]GCV, or [3H]BVdU incorporated into methanol insoluble cell material (e.g., nucleic acids)) were washed three times with 66% cold methanol and further assayed for radioactivity.10

HSV-tk biological activity HAT selection medium survival. L-M(tk⫺) cells were transduced with either the BTK or BTNfus vectors. A total of 500 cells were then plated in triplicate in 6-well plates, maintained in D10 or in Dulbecco’s modified Eagle’s medium supplemented with 100 ␮M hypoxanthine, 0.4 ␮M aminopterin, and 16 ␮M thymidine (HAT media supplement; Sigma), and incubated at 37°C in 5% CO2. After 7–10 days, colonies were fixed with methanol, stained with methylene blue, and counted. Inhibition of DNA synthesis and metabolic activity by GCV and BVdU. MC 38 cells transduced either with the BTK or BTNfus vectors were plated in 96-well plates (5 ⫻ 103 or 104 cells/well) in triplicates, treated with different concentrations of GCV or BVdU, and incubated at 37°C in 5% CO2 for 72 hours. Cell viability at the end of the treatment was assessed as function of their ability to reduce 3-(4,5-dimethylthiazol-2-yl)⫺2,5-diphenyltetrazolium bromide (MTT) according to the MTT colorimetric assay.22 Briefly, after 69 hours of incubation, 10 ␮L of a 10 mg/mL solution of MTT (Sigma) was added to the culture medium, followed by further incubation for 3 hours. After removing the supernatants, 100 ␮L of dimethylsulfoxide was added to each well to dissolve the water-insoluble formazan crystals. Optical density was determined with a microtiter plate reader using a 560- to 650-nm dual wavelength detector. All experiments were repeated at least twice. Inhibition of cell proliferation by GCV and BVdU. MC 38 and MCA 205 cells transduced with either the BTK or BTNfus vectors were plated to 6-well plates (2.5 ⫻ 105 cell/well) and allowed to grow overnight at 37°C in 5% CO2; cells were then exposed to GCV or BVdU at concentrations from 0.0005 ␮M to 200 ␮M. After 48 hours of incubation, cells were counted in a Coulter counter (Coulter Electronics, Luton, Beds, UK). The drug concentration required to inhibit cell proliferation by 50% was defined as the IC50.

Statistical analysis Analysis of significant differences among data groups was performed using the Student’s t test for unpaired observations.

RESULTS

Retroviral vector construction Using the polymerase chain reaction technique, the restriction enzyme sites BamHI (at the 5⬘) and EcoRI (at the 3⬘) were added to both TNFUS69 and HSV-tk genes to facilitate subcloning into the pBabe-puro cassette to obtain BTK and BTNfus retroviral vectors (Fig 1).

Figure 1. Schematic representation of the BTK and BTNfus retroviral vectors. The HSV-tk gene and the TNFUS69 fusion construct15 were inserted into pBabe-puro16 under the transcriptional control of the Moloney murine leukemia virus long tandem terminal repeat, using the BamHI-EcoRI sites. In both vectors, the expression of the drug resistance gene pac (puromycin N-acetyl transferase) is driven by the internal simian virus 40 early region promoter (SV40p). ␺, packaging signal.

Furthermore, the two ATG start codons were placed in the context of a “Kozak box” (GCCGCCGCCATGG), known as a favorable sequence for translation initiation in eukaryotic organisms.23 Desired changes of the nucleotide sequences were confirmed by sequencing of both BTK and BTNfus vectors.

The BTNfus vector confers neomycin resistance to transduced cells We first analyzed the efficacy of the fusion gene to provide expression of NPT II. Comparison between LNL6- and BTNfus-transduced MC 38 cells indicated that the fusion gene was generating a lower amount of NPT II protein, as determined by ELISA (⬃33% of the control LNL6-transduced cells, data not shown), and a decreased kanamycin phosphorylation activity (⬃21% of the control, Fig 2A). Surprisingly, no difference in the sensitivity to the neomycin analog G418 was demonstrated in a [3H]Thd DNA incorporation assay. Cells transduced with the fusion gene were just as resistant to inhibition by G418 as those carrying the original neo gene, even in the presence of relatively high drug concentrations (Fig 2B). Furthermore, we could easily recover G418-resistant colonies from cells retrovirally transduced with the BTNfus vector, thus allowing us to conclude that the fusion gene, when inserted into retroviral vectors, is able to provide efficient positive selection despite a somewhat lower NPT II expression.

Thymidine, GCV, and BVdU phosphorylation in BTNfus-transduced cells To assess whether the BTNfus retroviral vector was able to confer HSV-tk activity to target cells, we performed HPLC analysis of Thd, GCV, and BVdU phosphorylated metabolites generated in transduced cells. The results of this analysis are shown in Table 1. Compared with the fusion gene, cells transduced with the HSV-tk gene in its native form showed a markedly higher ability to phosphorylate Thd and GCV. Surprisingly however, cells transduced with the BTNfus vector were slightly more efficient than cells expressing native HSV-tk in the


577


Table 1. HPLC Analysis of Phosphorylated Thymidine, GCV, and BVdU Metabolites in Cells Transduced with BTK or BTNfus Vectors [3H]Thd (disintegrations per minute/106 cells) TP*

Bound†

0.0 4,666.5 387.0

97.5 19,075.5 3,118.0

Cell line ⫺

L-M(tk ) L-M(tk⫺) BTK L-M(tk⫺) BTNfus

[3H]GCV (pmol/ 106 cells)

[3H]BVdU (pmol/106 cells)

Cell line

TP

Bound

Monophosphate‡

Bound

MC 38 MC 38 BTK MC 38 BTNfus

0.0 527.3 13.8

9.6 41.3 20.1

0.8 58.6 77.5

2.3 16.2 11.6

*Cytoplasmic (methanol-soluble) TP metabolites. †Methanol-insoluble extracts (nucleic acid-bound radioactivity). ‡Cytoplasmic monophosphate metabolites.

vector, while in the case of BTNfus-transduced cells, ⬃55% of colonies were HAT-resistant (Fig 3). In addition, MC 38 cells transduced with the BTNfus vector showed metabolic inhibition when treated with GCV or BVdU (Fig 4), thus indicating the effective generation of phosphorylated GCV and BVdU metabolites. Interestingly, in GCV-treated cells, we observed significant differences between the effects of the native HSV-tk and the TNFUS69 chimeric gene. The reduced efficacy of the latter in providing metabolic suicide effects in the presence of GCV was less important at higher drug concentrations, but never negligible (Fig 4A). In contrast, BVdU treatment resulted in similar toxicity in both BTK- and BTNfus-transduced cells and Figure 2. A: In situ kanamycin phosphorylation activity of MC 38 cells unmodified or transduced with the LNL6 or BTNfus retroviral vectors. The total signal intensities of indicated areas are as follows: a1 ⫽ 172,999; a2 ⫽ 37,690; a3 ⫽ 23,202. Band “a3” probably represents the protein aggregate or enzymatically active degradation product of the TK-NEO fusion protein.20 B: In vitro neomycin resistance in MC 38 cells expressing the neo gene (MC 38 LNL6) or the TNFUS69 fusion gene (MC 38 BTNfus). Ability to proliferate in the presence of various concentrations of G418 is expressed as the percentage of incorporated radioactivity ([3H]Thd) in regular culture medium. Observations represent the mean values of triplicate cultures. Error bars indicate SDs (not visible if ⬍3%).

phosphorylation of BVdU. We subsequently evaluated the biological activity of the fusion gene in a series of functional assays indicative of Thd, GCV, and BVdU phosphorylation. Expression of the fusion gene resulted in efficient Thd phosphorylation, as demonstrated by the finding that L-M (tk⫺) cells transduced with the BTNfus retroviral vector were only slightly less efficient in forming colonies in HAT selection medium compared with BTK-transduced cells. On average, ⬃69% of colonies were rescued if cells were transduced with the BTK


Figure 3. Clonogenic potential of BTK- or BTNfus-transduced L-M(tk⫺) cells in HAT selection medium. Values are expressed as “% of unselected control” given by the following ratio: average number of colonies in HAT medium/average number of colonies in regular culture medium.

578


Figure 4. Metabolic inhibition induced by GCV (A) and BVdU (B) in MC 38 cells transduced with BTK or BTNfus retroviral vectors. Results (mean and SD of triplicates samples) are expressed as the percentage of MTT reduction observed in the absence of GCV or in the absence of BVdU. If SDs are ⬍3%, they are not visible. ⴱⴱⴱ, P ⬍ .0005; ⴱⴱ, P ⬍ .005; ⴱ P ⬍ .01; F, P ⬍ .05.

showed significant superiority of the fusion gene at lower BVdU concentrations (Fig 4B). Determination of the IC50 values of GCV and BVdU was performed in BTK and BTNfus MC 38 and MC 205 transduced cells. The results obtained are reported in Table 2 and are in line with the in vitro sensitivity assays. In particular, the reduced sensitivity of BTNfus-transduced cells to GCV (compared with BTK-transduced counterparts) was confirmed in both cell lines, thus demonstrating that this finding was independent of cell type. In addition, the IC50 values for BVdU were similar for BTNfus- and BTK-transduced cells, indicating that the suicide function of the fusion gene in combination with BVdU could be considered to be comparable with that of the native HSV-tk gene. DISCUSSION More than 30 cancer gene therapy clinical trials have been proposed that use retroviral vectors expressing the suicide gene HSV-tk.24 Most of these vectors also contain a selectable marker, because production and charTable 2. Inhibitory Effects of GCV and BVdU on BTK- or BTNfus-Transduced Cells IC50 (␮M)* Cell line

GCV


200 0.05 0.5–1


100–200 0.1–0.25 25

BVdU 5–10 0.01–0.05 0.001–0.005 50–100 0.01–0.05 0.05

*Drug concentration sufficient to achieve IC50 (see Materials and Methods).

acterization of retroviral producer cells in the absence of positive selection is cumbersome and extremely time consuming. However, the occurrence of transcriptional interference between multiple expression units within the same construct has been demonstrated in many systems25–27 and justifies the current common efforts to develop “simplified,” single-gene vectors. Clearly, the use of bifunctional fusion genes would represent an advantageous solution to these problems. We have taken advantage of the TNFUS69 plasmid,15 from which the HSV-tk/neo fusion gene was excised and subcloned in a retroviral cassette under the transcriptional control of the Moloney long tandem terminal repeat, in the context of a Kozak sequence. We show that the resulting retroviral vector, BTNfus, can efficiently confer G418-resistance to transduced cells and that, therefore, the fusion gene can be used as a regular positive selectable marker. Quantitation of the amount of NPT II protein and its ability to phosphorylate kanamycin indicated that there were lower levels of NPT II in cells transduced with the fusion gene compared with cells expressing native neo. However, the biological activity of the NPT II component of the fusion protein encoded by TNFUS69 was substantial, because BTNfus-transduced cells were as resistant to the inhibitory effects of G418 on cell proliferation as LNL6-transduced control cells. The reduced ability of the anti-NPT II antiserum used in the ELISA to recognize the NPT II “new” structure as part of a fusion protein, as well as a potentially decreased affinity of the TK-NEO protein for kanamycin, are possible explanations for these apparent discrepancies. Nevertheless, our data clearly demonstrate that the fusion gene can provide useful positive selection. The reduced ability of BTNfus-transduced cells to generate phosphorylated thymidine and GCV metabolites could similarly be ascribed to the modifications made to the HSV-tk carboxyl terminus to originate the TNFUS69 gene15 or to a lower affinity (or catalytic



efficacy) of the TK component of the fusion protein for the substrates thymidine and GCV. Verification of this supposition would require further biochemical studies involving enzyme purification, which are beyond the scope of this work. Although reduced when compared with HSV-tk-carrying cells, the phosphorylation of Thd and GCV provided by the fusion gene was biologically active in vitro, allowing mutant cellular TK-negative cells to survive HAT selection and conferring GCV sensitivity to BTNfus-transduced cells. Taken together, we conclude that the BTNfus retroviral vector can be used in vitro for both positive and negative selection of transduced cells. More commonly used approaches to achieve combined expression of positive and negative selectable markers by retroviral transduction are based on the inclusion in the retroviral cassette of two transcriptional units involving the need for two different promoter elements. Alternative strategies rely on the use of internal ribosomal entry site sequences, which allow the synthesis of two different protein products from the translation of a bicistronic mRNA transcribed under the control of a single promoter. In both of these approaches, the level of expression of the two marker genes can vary dramatically as a consequence of promoter interference or competition of gene expression. The use of a bifunctional protein may overcome these potential problems because of the presence of a single transcriptional unit driven by a single promoter and expressing a single gene product. Other fusion genes containing HSV-tk have been described. Among these, the tgLS(⫹)HyTK retroviral vector encodes a gene originated by the fusion of the hygromycin resistance gene (Hy) and HSV-tk,28 whereas the pETLGB expression cassette29 was generated by fusing the HSV-tk sequence to that of green fluorescent protein. Both the HyTK and the TK-green fluorescent protein fusion genes were demonstrated to originate bifunctional fusion proteins with potential applications that were similar to the TK-NEO protein encoded by the fusion gene inserted in our BTNfus retroviral vector. In addition, an attractive application for the same fusion gene used in our studies has been described recently by Bonini et al.30 The fusion gene was introduced in a retroviral vector used to transduced peripheral blood lymphocytes of bone marrow donors as part of a clinical protocol of allogeneic bone marrow transplantation for leukemia. These engineered lymphocytes were successfully used to treat leukemia relapse after bone marrow transplantation and were efficiently eliminated by GCV administration when they induced a severe graft-versushost reaction. Other very recent results of clinical trials using the HSV-tk/GCV suicide system for the treatment of cancer have underlined the need to improve the efficiency of the system.12,13 It is clear that both the delivery of the HSV-tk-carrying viral vector and the choice and/or the administration schedule of the active prodrug can be refined. Recent pharmacological studies have led to the identification of several compounds of demonstrated efficacy in the treatment of herpesvirus infection in


579

animal models and/or in human studies. Among these drugs, BVdU was found to be at least 100-fold more potent than GCV in inhibiting the proliferation of HSV-tk-transformed cells.10,31 In our studies, in contrast to GCV, BVdU was as efficiently phosphorylated by the fusion gene-transduced cells as it was by HSV-tk-transduced control cells. This observation, along with the remarkable efficiency of BVdU in inhibiting the in vitro proliferation of BTNfustransduced cells, suggests that the use of the fusion gene in association with BVdU may be useful in metabolic suicide gene transfer. This is particularly relevant in light of the impaired generation of phosphorylated GCV metabolites observed in BTNfus-transduced cells. The reduced sensitivity to GCV may result in inefficient metabolic suicide effects if this drug is used in combination with the fusion gene for in vivo applications. In this context, BVdU may represent a better choice. It is hoped that further biochemical studies, now made possible by the recent generation of [3H]-labeled BVdU, will help to better define the mechanism of action of BVdU and its derivatives, as well as to clarify the real possibility of its use for applications in vivo. ACKNOWLEDGMENTS We thank Drs. R. Kucherlapati, H. Land, A. D. Miller, and S. A. Rosenberg for kindly donating reagents and Dr. Hiroyuki Ishii for helpful discussions and advice.

REFERENCES 1. Blaese RM, Mullen CA, Ramsey WJ. Strategies for gene therapy. Pathol Biol. 1993;41:672– 676. 2. Blaese RM, Ishii-Morita H, Mullen CA, et al. In situ delivery of suicide genes for cancer treatment. Eur J Cancer. 1994;30A:1190 –1193. 3. Moolten FL, Wells JM. Curability of tumors bearing herpes thymidine kinase genes transferred by retroviral vectors. J Natl Cancer Inst. 1990;82:297–300. 4. Huber BE, Richards CA, Krenitsky TA. Retroviral-mediated gene therapy for the treatment of hepatocellular carcinoma: an innovative approach for cancer therapy. Proc Natl Acad Sci USA. 1991;88:8039 – 8043. 5. Mullen CA, Kilstrup M, Blaese RM. Transfer of the bacterial gene for cytosine deaminase to mammalian cells confers lethal sensitivity to 5-fluorocytosine: a negative selection system. Proc Natl Acad Sci USA. 1992;89:33–37. 6. Mroz PJ, Moolten FL. Retrovirally transduced Escherichia coli gpt genes combine selectability with chemosensitivity capable of mediating tumor eradication. Hum Gene Ther. 1993;4:589 –595. 7. Sorscher EJ, Peng S, Bebok Z, Allan PW, Bennett LL Jr, Parker WB. Tumor cell bystander killing in colonic carcinoma utilizing the Escherichia coli DeoD gene to generate toxic purines. Gene Ther. 1994;1:233–238. 8. Wei MX, Tamiya T, Chase M, et al. Experimental tumor therapy in mice using the cyclophosphamide-activating cytochrome P450 2B1 gene. Hum Gene Ther. 1994;5:969 – 978. 9. Mar EC, Chiou JF, Cheng YC, Huang ES. Inhibition of

580

10.

11. 12. 13.

14.

15. 16.

17.

18. 19.

20.


cellular DNA polymerase ␣ and human cytomegalovirusinduced DNA polymerase by the triphosphates of 9-(2hydroxyethoxymethyl)guanine and 9-(1,3-dihydroxy-2propoxymethyl)guanine. J Virol. 1985;53:776 –780. Balzarini J, Bohman C, De Clercq E. Differential mechanism of cytostatic effect of (E)-5-(2-bromovinyl)-2⬘-deoxyuridine, 9-(1,3-dihydroxy-2-propoxymethyl)guanine, and other antiherpetic drugs on tumor cells transfected by the thymidine kinase gene of herpes simplex virus type 1 or type 2. J Biol Chem. 1993;268:6332– 6337. Singhal S, Kaiser LR. Cancer chemotherapy using suicide genes. Surg Oncol Clin N Am. 1998;7:505–536. Ram Z, Culver KW, Oshiro EM, et al. Therapy of malignant brain tumors by intratumoral implantation of retroviral vector-producing cells. Nat Med. 1997;3:1354 –1361. Sterman DH, Treat J, Litzky LA, et al. Adenovirusmediated herpes simplex virus thymidine kinase/ganciclovir gene therapy in patients with localized malignancy: results of a phase I clinical trial in malignant mesothelioma. Hum Gene Ther. 1998;9:1083–1092. Balzarini J, Bohman C, Walker RT, De Clercq E. Comparative cytostatic activity of different antiherpetic drugs against herpes simplex virus thymidine kinase gene-transfected tumor cells. Mol Pharmacol. 1994;45:1253–1258. Schwartz F, Maeda N, Smithies O, et al. A dominant positive and negative selectable gene for use in mammalian cells. Proc Natl Acad Sci USA. 1991;88:10416 –10420. Morgenstern JP, Land H. Advanced mammalian gene transfer: high titre retroviral vectors with multiple drug selection markers and a complementary helper-free packaging cell line. Nucleic Acids Res. 1990;18:3587–3596. Bender MA, Palmer TD, Gelinas RE, Miller AD. Evidence that the packaging signal of Moloney murine leukemia virus extends into the gag region. J Virol. 1987;61: 1639 –1646. Miller AD, Buttimore C. Redesign of retrovirus packaging cell lines to avoid recombination leading to helper virus production. Mol Cell Biol. 1986;6:2895–2902. Restifo NP, Esquivel F, Asher AL, et al. Defective presentation of endogenous antigens by a murine sarcoma: implications for the failure of an anti-tumor immune response. J Immunol. 1991;147:1453–1459. Reiss B, Sprengel R, Will H, Schaller H. A new sensitive

21.

22. 23. 24. 25.

26.

27. 28.

29.

30. 31.

method for qualitative and quantitative assay of neomycin phosphotransferase in crude cell extracts. Gene. 1984;30: 211–217. Agbaria R, Mullen CA, Hartman NR, et al. Effects of IMP dehydrogenase inhibitors on the phosphorylation of ganciclovir in MOLT-4 cells before and after herpes simplex virus thymidine kinase gene transduction. Mol Pharmacol. 1994;45:777–782. Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J Immunol Methods. 1983;65:55– 63. Kozak M. The scanning model for translation: an update. J Cell Biol. 1989;108:229 –241. Human gene marker/therapy protocols. Hum Gene Ther. 1998;9:2805–2852. Wu X, Holschen J, Kennedy SC, Ponder KP. Retroviral vector sequences may interact with some internal promoters and influence expression. Hum Gene Ther. 1996;7:159 – 171. Yee JK, Moores JC, Jolly DJ, Wolff JA, Respess JG, Friedmann T. Gene expression from transcriptionally disabled retroviral vectors. Proc Natl Acad Sci USA. 1987;84: 5197–5201. Chen BF, Hsieh CL, Chen DS, Hwang LH. Improved gene expression by a U3-based retroviral vector. Biochem Biophys Res Commun. 1992;184:330 –337. Lupton SD, Brunton LL, Kalberg VA, Overell RW. Dominant positive and negative selection using a hygromycin phosphotransferase-thymidine kinase fusion gene. Mol Cell Biol. 1991;11:3374 –3378. Loimas S, Wahlfors J, Janne J. Herpes simplex virus thymidine kinase-green fluorescent protein fusion gene: new tool for gene transfer studies and gene therapy. Biotechniques. 1998;24:614 – 618. Bonini C, Ferrari G, Verzeletti S, et al. HSV-TK gene transfer into donor lymphocytes for control of allogeneic graft-versus-leukemia. Science. 1997;276:1719 –1724. Balzarini J, De Clercq E, Ayusawa D, Seno T. Murine mammary FM3A carcinoma cells transformed with the herpes simplex virus type 1 thymidine kinase gene are highly sensitive to the growth-inhibitory properties of (E)-5-(2-bromovinyl)-2⬘-deoxyuridine and related compounds. FEBS Lett. 1985;185:95–100.
