Improved retroviral suicide gene transfer in colon cancer cell lines after cell synchronization with methotrexate
- 3.2k Downloads
Cancer gene therapy by retroviral vectors is mainly limited by the level of transduction. Retroviral gene transfer requires target cell division. Cell synchronization, obtained by drugs inducing a reversible inhibition of DNA synthesis, could therefore be proposed to precondition target cells to retroviral gene transfer. We tested whether drug-mediated cell synchronization could enhance the transfer efficiency of a retroviral-mediated gene encoding herpes simplex virus thymidine kinase (HSV-tk) in two colon cancer cell lines, DHDK12 and HT29.
Synchronization was induced by methotrexate (MTX), aracytin (ara-C) or aphidicolin. Gene transfer efficiency was assessed by the level of HSV-TK expression. Transduced cells were driven by ganciclovir (GCV) towards apoptosis that was assessed using annexin V labeling by quantitative flow cytometry.
DHDK12 and HT29 cells were synchronized in S phase with MTX but not ara-C or aphidicolin. In synchronized DHDK12 and HT29 cells, the HSV-TK transduction rates were 2 and 1.5-fold higher than those obtained in control cells, respectively. Furthermore, the rate of apoptosis was increased two-fold in MTX-treated DHDK12 cells after treatment with GCV.
Our findings indicate that MTX-mediated synchronization of target cells allowed a significant improvement of retroviral HSV-tk gene transfer, resulting in an increased cell apoptosis in response to GCV. Pharmacological control of cell cycle may thus be a useful strategy to optimize the efficiency of retroviral-mediated cancer gene therapy.
KeywordsHT29 Cell Colon Cancer Cell Line Suicide Gene Cancer Gene Therapy Drug Removal
Cancer gene therapy by suicide gene transfer remains an alternative approach to increase selectivity in cancer treatment . The enzyme prodrug strategy, involving transfer of the suicide gene, i.e. HSV-tk, to tumor cells followed by ganciclovir (GCV) treatment, is the most widely used [2, 3, 4, 5]. HSV-TK phosphorylates GCV to its monophosphate form that is then converted by cellular kinases into GCV triphosphate, which causes DNA chain termination and cell death . In vivo, this strategy involves both a direct cytotoxic effect and a bystander effect . The bystander effect confers cytotoxicity to the neighboring nontransduced cells , and a distant anti-tumor immune response. These aforementioned ways for killing tumors are related to the quantitative efficiency of gene transfer [9, 10]. However, one of the major obstacles to successful cancer gene therapy is the inadequate transduction of the target cells . In vivo, several studies have shown that the number of cells transduced by retroviral vectors constitutes less than 10% of the target cell population [12, 13].
The transduction efficiency of defective murine-derived retroviral vectors requires target cells to be in division because integration of the great size viral DNA-protein complex needs the metaphasic breakdown of the nuclear membrane. Integration of the transgene thus depends on the phase of the cycle where the target cells are [14, 15, 16]. Consistently, the relationship between cell cycle and retroviral transduction has previously been shown [15, 17, 18]. The gene transfer efficiency was lower in cultured cells enriched in G0-G1 phase than that in similar cell populations enriched in S, G2 and M phases . The accumulation of cells blocked in a determined cell cycle phase which is the definition of synchronization, could thus improve the efficiency of gene transfer and finally the effectiveness of viral transduction. Consistently, cells need to be synchronized in S phase due to the intracellular half-life of murine retroviruses. Synchronization of cells in S phase can be obtained in vitro by serum starvation or by drugs inducing a reversible DNA synthesis inhibition. Methotrexate (MTX), aphidicolin or aracytin (ara-C) have been used to synchronize several cell lines in S phase. The effect of these drugs is reversible in respect with the micromolar concentrations used [19, 20, 21, 22]. Although synchronization has been used for improving the efficacy of chemotherapy [23, 24], the effect of synchronization on the efficiency of retroviral gene transfer has never been evaluated in colon cancer cells. The aim of this study was to evaluate whether transduction efficiency may be increased by the synchronization of target cells before retroviral gene transfer.
We used two colon cancer cell lines: the human HT29 and the murine DHDK12 pro-b (Pr. Martin, Dijon; France) cell lines. Cell lines were cultured in DMEM medium containing 10% calf serum/penicillin (50 units/ml)/streptomycin (50 μg/ml) at 37°C in 5% CO2. We used retroviral vectors carrying Escherichia-coli β-galactosidase (β-gal)  and herpes simplex thymidine kinase (HSV-tk) genes associated with pac and neoR gene respectively as positive selectable marker genes. Amphotropic packaging cells were generated from the human embryonic kidney cell line 293. The packaging cells stably express Friend Murine Leukemia Virus (F-MuLV) strain FB29 gag/pol genes and an amphotropic envelop gene derived from pPAM3 (A.D Miller Seattle, WA, USA). Packaging cells were transfected with plasmids pTG 5391 (FB29 LTR-lacZ-SV40-Puro-LTR, clone E17-12 -TG 5391) or pTG 9344 (FB29 LTR-PGK-TK-IRES-Neo -LTR clone E 17-21 pTG 9344) to isolate the retroviral producer clone E17-12 -TG 5391 and E 17-21 TG 9344 (Transgene S.A., Strasbourg, France). The retroviral producer clone were cultured in DMEM supplemented with 4.5 g/L of glucose, 1% non-essential amino acids, 40 μg/ml gentamycin (Sigma) and 10% calf serum. Culture supernatant was harvested, filtered through a 0.45 μm nitrocellulose filter (Sartorius, Goettingen, Germany) and used in the presence of polybrene (Sigma) at 8 μg/ml final concentration. NIH 3T3 fibroblasts were cultured in DMEM supplemented with 40 μg/ml gentamycin and 10% heat inactivated NBBS (GIBCO/BRL). Retroviral titration was determined by infecting NIH 3T3 fibroblasts with serial dilutions of the culture medium and staining respectively for β-galactosidase activity with X-gal protocol  or for HSV-TK expression using monoclonal antibody anti-HSV-TK as described below. All point titrations were performed four times. The titer of viral preparation was 4.9 (± 1.2) × 106 focus-forming units (FFU/ml) for TG 9344 and 1.7 (± 0.9) × 107 FFU/ml for TG 5391. The absence of competent replication helper retrovirus was checked by NIH 3T3 mobilization assay
Treatment of cells with MTX, ara-c or aphidicolin
IC50 of Methotrexate, Ara-C and Aphidicolin in DHDK12 and HT29 cell lines
Gene transfer into synchronized cells
DHDK12 and HT29 cells were transduced with the reporter gene β-gal. After removal of drug-containing medium, samples were taken every 8 hr during 72 hr. For each time, cells were infected with 1 ml of 0.45 μm filtered TG 5391 packaging cells supernatant in the presence of 8 μg/ml of polybrene.
Then, HSV-tk gene was used during optimal period determined with the reporter gene for each cell line. During this period, cells were infected with 1 ml of 0.45 μm filtered TG 9344 packaging cells supernatant in the presence of 8 μg/ml of polybrene at various time points after MTX removal.
For each time point, appropriate controls were performed. Transgene expression was determined 48 hr after transduction.
Transgene expression assay
For detection of β-galactosidase activity, cells transduced by TG 5391 were fixed for 15 min at 37°C with 0.5% of glutaraldehyde, then washed two times with PBS and stained with X-gal for cytochemical analysis, as previously described. The quantitative detection of β-gal expression was performed with the fluorescein-di-β-D-galactopyranoside (FDG) (Sigma) by flow cytometry . Cells were harvested (trypsin-EDTA), washed and resuspended at a concentration of 5.105/ml in 25 μl of PBS containing 2% fetal calf serum, at 37°C for 10 min. The β-galactosidase activity was obtained by cell incubation in 25 μl of 2 mM FDG solution for one min at 37°C, then for one hour at 0°C, in 1 ml of PBS. The fluorescence was analyzed by flow cytometry. Non-transduced cells formed the control group.
For HSV-TK expression analysis, cells transduced by TG 9344, cultured on slides (Labtek II-Nunc), were fixed for 15 min at 4°C with 4% paraformaldehyde and incubated with PBS containing 0.2% serum bovine albumin (SAB) and 0.1% saponin for 5 min. Cells were incubated with anti-HSV-TK mouse monoclonal antibody 4C8 (W. Summers, Yale University, USA) 1/50, for 30 min at room temperature. After washing in PBS, cells were incubated for 10 min in a secondary antibody solution of goat anti-mouse coupled to biotin (LSAB 2 System Peroxydase, Dako). Cells were washed in PBS and incubated 10 min with streptavidin-peroxydase. The revelation was achieved by incubation for 5 min with 3-3' diaminobenzidine (DAB) leading to cytoplasmic brown precipitates. Cells were counterstained with hematoxylin.
For flow cytometry analysis, cells were harvested, washed in PBS and fixed with 4% paraformaldehyde for 15 min at 4°C in PBS. Cells were washed in incubation buffer (0.2% SAB, 0.1% saponin in PBS containing 0.2% of sodium azide) then incubated in 200 μl of anti-HSV-TK monoclonal antibody 4C8, diluted to 1/50 in incubation buffer for 30 min at room temperature. Cells were washed three times with PBS. The pellet was resuspended 30 min at room temperature, in 200 μl of goat anti-mouse antibody coupled to FITC, diluted to 1/100 in incubation buffer. Cells were washed and resuspended in 1 ml of PBS for flow cytometry analysis.
Measurement of ganciclovir-induced cytotoxicity in synchronized cells
Flow cytometry was carried out on synchronized cell, transduced with TG 9344 at different periods, after 72 hr of 20 μM GCV treatment to quantitate cell apoptosis. Apoptosis was determinate by staining cells with annexin V-FITC and propidium-iodide (PI) labeling, because annexin V can identify the externalization of phosphatidylserine during the apoptotic progression and therefore detect early apoptotic cells . Cells were transduced with TG 9344 vector, on 12-well plates and treated after 24 hr by 20 μM GCV. Control cells were no transduced or untreated. After 72 hr of treatment, cells were harvested, and washed twice in PBS. The pellet was resuspended in 1 ml of 100 mM HEPES/NaOH, pH 7.5. Then 500 μl of the cell suspension were incubated in presence of 2 μg/ml annexin V-FITC, and 10 μl of PI (100 μg/ml) for 10 min. Samples were immediately analyzed by flow cytometry on a bi-parametric histogram giving the level of annexin V-FITC and PI fluorescence.
Apoptosis was assessed by DNA fragmentation assay. Samples of 5.105 pTG 9344 transduced cells with or without synchronization were treated 96 hr with 20 μM GCV. Cells then were centrifuged at 800 g for 5 min at 4°C. The pellet was resuspended in 20 μl of lysis buffer (EDTA 20 mM, Tris 100 mM, SDS 0,8%, pH 8). Then 10 μl of 500 UI/ml RNAse (Sigma) were added for 60 min at 37°C. The mix was incubated 90 min at 50°C with 10 μl of 20 mg/ml proteinase K. Migration was achieved on 1.8% agarose gel containing 0.5 μg/ml ethidium bromide at 35 V during 4 hr. MSP I digested PBR 322 was used as a size marker. Non-transduced cells treated with MTX or GCV constituted control groups.
Comparisons were made using the Student's t test. P < .05 was considered as significant.
Altered progression in the cell cycle by methotrexate, ara-C or aphidicolin
We first assessed the effect of drugs on DHDK12 and HT29 cell cycles to delineate the time for which a maximum of cells were in S phase after drug removal.
The effects of the three drugs, i.e. MTX, ara-C and aphidicolin, on the cell cycle were preliminary assessed in DHDK12 cells. After a 24 hr treatment with MTX, ara-C or aphidicolin, cells were analyzed between 0 and 72 hr after drug removal for DNA content by flow cytometry.
In HT29 cell line, the effect of MTX on cell cycle progression was slightly different. As illustrated in Figure 1B, cells began to accumulate in S phase almost immediately after MTX removal. While the rate of cells in S phase was 18% without treatment (Figure 1B), this rate reached 55% 6 hr after MTX removal and decreased thereafter to reach the ratio of untreated cells 24 hr after MTX removal.
Taken together, these observations indicate that the pattern of cell cycle synchronization after MTX removal is specific for each cell line. Because we hypothesize that gene transfer efficiency is improved by potent cell cycle synchronization, the time window for transduction experiments with the β-gal reporter gene should be different between the two cell lines.
Improvement of gene transfer efficiency in synchronized cell
Similar experiments were performed in HT29 cells. Accumulation of HT29 cells in S phase was observed almost immediately after drug washout. Accordingly, the highest transduction rate for β-gal gene was observed 6 hr after drug washout (Figure 2B). The efficiency of transduction was comparable to the control cells 12 hr after drug washout (Figure 2B).
For HT29 cells, transduction efficiency with HSV-TK was maximal at 6 hr after drug washout and reached 22% while it was 15% in untreated cells (Figure 4B). Therefore according to the host cell cycle, we found that pre-treatment with MTX resulted in improved gene transfer efficiency in these two cells lines.
Enhancement of apoptosis in synchronized cell
The percentage of MTX-treated DHDK12 cells undergoing apoptosis (Annexin V+, PI-) was two fold higher after MTX withdrawal (46% vs. 23% in the untreated cell population). The difference was maximal in cells transduced 20 hr after MTX withdrawal (Figure 6B).
In HT29 cells, the maximum percentage of MTX-treated cells undergoing apoptosis was 28% while it was 20% in untreated cells. The highest level of cell apoptosis was maximal 6 hr after MTX withdrawal (Figure 6C).
The objective of this work was to improve the efficiency of retroviral transfer of the suicide gene HSV-tk in colon cancer cells. This aim was achieved through the pharmacological control of the target cells cell cycle. Our results are consistent with previous reports showing that retroviral-mediated gene transfer depends on the cell cycle of target cells. The nuclear transfer of the pre-integrative viral complex is a strong limit to the efficiency of defective amphotrophic retroviral vectors derived from murine leukemia virus (MuLV). This step is possible only through the metaphasic breakdown of the nuclear membrane [14, 16, 30]. Therefore, the integration of retroviral DNA during cell division has only been evidenced when the doubling time of target cells was higher than the half-life of the virus . As the half-life of MuLV-derived vectors is between 5.5 and 7.5 hr  and as the DHDK12 and HT29 cell lines have a doubling time of 28 hr  and 24 hr , respectively, our model meet this criterion. Our experimental design thus was adapted to study the efficiency of retroviral gene transfer after pharmacological control of the cell cycle.
Cell synchronization has been used to increase the number of cells accessible to drug targeting DNA and to improve the action of several anti-proliferative chemotherapies [20, 23, 24]. In this regard, experimental works have studied the synchronization in S phase of cancer cell lines by MTX, aphidicolin or ara-C. Aphidicolin and ara-C are reversible inhibitors of DNA polymerases [18, 22]. MTX induces a reversible inhibition of dihydrofolate reductase, which is required for the de novo synthesis of nucleotides for DNA replication . Our study showed a limited efficiency of ara-C or aphidicolin in DHDK12 cells. Moreover, a significant toxicity of aphidicolin, not compatible with an in vivo application, has been observed on several cancer cell lines [19, 35]. We observed that non-toxic concentrations of MTX induced a reversible synchronization of DHDK12 and HT29 cells in early S phase (Figure 1). A 24 hr-treatment with MTX allowed increasing the rate of cells in S phase. The reversibility of MTX was confirmed as the cells returned to the normal cell cycle according to there doubling time. These results were in accordance to those obtained in others cell lines .
The reverse transcription of retroviral DNA can occur in several phases of the cell cycle . However, the cells should be stimulated to divide before infection for efficient gene transfer . According to the intracellular half-life of retroviral intermediates, the position of target cells relative to mitosis and the duration of S phase at the time of exposure both are critical to determine the efficiency of infection . This assumption was supported by the difference in retroviral gene transfer improvement between DHDK12 and HT29 cell lines after cell synchronization by MTX. These two colon cancer cell lines exhibit a different pattern of cell cycle distribution after synchronization (Figure 1). We have observed that in HT29 cells the level of transgene expression, which was lower than that observed in DHDK12 cells, was strictly related to the peak of cells in S phase (Figure 2B). In DHDK12 cell line, the peak of cells in S phase was located 10 hr after the recovery and the infection efficiency was improved by 2-fold 20 hr after MTX removal (Figure 2A). The time difference between the maximum level of DHDK12 cells in S phase and the maximum efficiency of transduction could be related to the reverse transcription and integration of the viral DNA. Thus, the period of internalization and reverse transcription, which lasts 4 to 8 hours , must correspond to the interval necessary for cells synchronized in S phase to reach the G2-M phase to obtain the optimal integration of viral DNA. Our results indicate that the pattern of synchronization in DHDK12 cells at 20 hr after MTX removal is adapted to these criteria. In contrast to DHDK12 cells, HT29 cells synchronized in S phase reach more rapidly the G2-M phase, which may prevent optimal internalization and reverse transcription of the viral DNA in HT29 cells. This hypothesis is consistent with a model analyzing the kinetic of short half-life retrovirus mediated gene transfer . Taken together, this allows delineating an optimal period for the retroviral gene transfer in synchronized target cells.
Quantitative detection of GCV-induced apoptosis was used to determine whether the increased efficiency of the HSV-tk retroviral gene transfer resulted in an increase in GCV-mediated cell death. The transduction rate of HSV-tk gene reached 30% in the DHDK12 cell line 20 hr after MTX removal, doubling the efficiency of retroviral gene transfer observed in untreated cells. Although the transduction rates of the β-gal reporter gene or the HSV-tk gene may appear rather low, they constitute a two-fold increase compared with the transduction rates previously described [12, 13]. Indeed, in the aforementioned studies, the fraction of infected cells was less than 10% whereas in our experimental design it reached 30% in the DHDK12 cell line 20 hr after MTX removal. Because Chen et al. have previously demonstrated that a higher level of HSV-TK expression correlates with greater bystander effect leading to increased cell killing, the increased transduction rate that we reached in our study could enhance GCV-mediated cell death. Consistently, our results show that the number of cells in apoptosis was higher than the number of cells expressing HSV-TK indicating greater bystander effect. Altogether, these observations indicate that improvement of transduction efficiency may represent a key step in retroviral suicide gene therapy by increasing both suicide gene expression and bystander effect.
We acknowledge nevertheless that this study has some limitations. Indeed, MTX was less efficient in HT29 cells than in DHDK12 cells in improving retroviral gene transfer and subsequently cell apoptosis after GCV treatment. This could be explained by an adverse effect of MTX metabolization leading to the inhibition of retroviral cycle. Indeed, the MTX metabolites have been shown to inhibit retroviral infection . However, the rate of HT29 transduced cells undergoing apoptosis after GCV treatment increased from 20% to 28% in cells pre-treated with MTX. We are currently investigating whether a rescue strategy could antagonize the inhibitory effect of MTX metabolites on retroviral infection.
We show here that cell synchronization may improve the efficacy of retroviral suicide gene transfer in a human and a murine colon cancer cell lines. Because the effect of cell synchronization on retroviral gene transfer differs between the two colon cancer cell lines used in this study, further investigations in more colon cancer cell lines are needed to draw definitive conclusion on the improvement of retroviral gene transfer after cell synchronization. Nevertheless, we demonstrate in the present study that this improvement increases the level of apoptosis induced with GCV treatment. This approach could be fruitful in colon cancer liver metastases because tumor cells are proliferating in a quiescent parenchyma. Therefore, we are currently assessing in a rat model of liver tumors whether this strategy could improve the antitumoral efficacy of cancer gene therapy using defective retroviral vectors.
This work was supported by Grants from the Fondation pour la Recherche Médicale, the Académie de Médecine, the Chancelleries de Paris and the Association de Recherche en OncoLogie Digestive (AROLD).
- 3.Sandmair AM, Loimas S, Puranen P, Immonen A, Kossila M, Puranen M, Hurskainen H, Tyynela K, Turunen M, Vanninen R, Lehtolainen P, Paljarvi L, Johansson R, Vapalahti M, Yla-Herttuala S: Thymidine kinase gene therapy for human malignant glioma, using replication-deficient retroviruses or adenoviruses. Hum Gene Ther. 2000, 11: 2197-2205. 10.1089/104303400750035726.CrossRefPubMedGoogle Scholar
- 4.Rainov NG: A phase III clinical evaluation of herpes simplex virus type 1 thymidine kinase and ganciclovir gene therapy as an adjuvant to surgical resection and radiation in adults with previously untreated glioblastoma multiforme. Hum Gene Ther. 2000, 11: 2389-2401. 10.1089/104303400750038499.CrossRefPubMedGoogle Scholar
- 6.Cheng YC, Huang ES, Lin JC, Mar EC, Pagano JS, Dutschman GE, Grill SP: Unique spectrum of activity of 9-[(1,3-dihydroxy-2-propoxy)methyl]-guanine against herpesviruses in vitro and its mode of action against herpes simplex virus type 1. Proc Natl Acad Sci USA. 1983, 80: 2767-2770. 10.1073/pnas.80.9.2767.PubMedCentralCrossRefPubMedGoogle Scholar
- 10.Smiley WR, Laubert B, Howard BD, Ibanez C, Fong TC, Summers WS, Burrows FJ: Establishment of parameters for optimal transduction efficiency and antitumor effects with purified high-titer HSV-TK retroviral vector in established solid tumors. Hum Gene Ther. 1997, 8: 965-977. 10.1089/hum.1997.8.8-965.CrossRefPubMedGoogle Scholar
- 29.Wei SJ, Chao Y, Hung YM, Lin WC, Yang DM, Shih YL, Ch'ang LY, Whang-Peng J, Yang WK: S- and G2-phase cell cycle arrests and apoptosis induced by ganciclovir in murine melanoma cells transduced with herpes simplex virus thymidine kinase. Exp Cell Res. 1998, 241: 66-75. 10.1006/excr.1998.4005.CrossRefPubMedGoogle Scholar
- 35.D'Anna JA, Crissman HA, Jackson PJ, Tobey R: Time-dependent changes in H1 content, H1 turnover, DNA elongation, and the survival of cells blocked in early S phase by hydroxyurea, aphidicolin, or 5-fluorodeoxyuridine. Biochemistry. 1985, 24: 5020-5026. 10.1021/bi00340a010.CrossRefPubMedGoogle Scholar
- 36.D'Incalci M, Erba E, Sen S, Rabbone ML, Perlangeli MV, Masera G, Conter V: Induction of partial synchronization of leukemia cells by continuous infusion of low-dose methotrexate followed by citrovorum factor. J Natl Cancer Inst. 1989, 81: 1509-1510. 10.1093/jnci/81.19.1509.CrossRefPubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.