Original Article

An artificially constructed radiation-responsive promoter is activated by doxorubicin

  • Cancer Gene Therapy volume 19, pages 345351 (2012)
  • doi:10.1038/cgt.2012.7
  • Download Citation


We previously developed an artificially constructed promoter that was activated in response to X-ray irradiation in LNCap, a prostate cancer cell line. Anticancer drugs were examined to see whether some of them could stimulate the activity of the promoter. It was found that doxorubicin (Dox) treatment to LNCap transfected with a gene cassette of the luciferase gene under control of the promoter-enhanced luciferase activity in a dose-dependent manner, indicating that the promoter could be controlled by Dox. When the luciferase gene was replaced with the fcy::fur gene whose product facilitates conversion of 5-fluorocytosine into 5-fluorouracil that is highly toxic, Dox stimulated the expression of the gene product, resulting in facilitation of cell killing effect in the presence of 5-fluorocytosine. These results suggest that therapeutic gene expression controlled with an anticancer drug may lead to a more effective cancer therapy with less hazardous side effects.


Prostate cancer is one of the most common types of cancer in men and a leading cause of cancer death.1 Against prostate cancer, endocrinotherapy is very effective.2 However, anticancer drugs are needed for recurrent prostate cancer treatment that is resistant to the therapy. In the case of recurrent prostate cancer treatment, docetaxel (Doc) is the drug of prime choice based on data gathered by two recently completed large-scale randomized studies.3, 4 However, anticancer drug treatment does not have much success with recurrent prostate cancer. A few months of life extension by the treatment would not be very satisfactory. Innovative combinations of drugs have been applied as another approach, resulting in an improved but still limited outcome. One reason for this is probably because of the hazardous side effects due to low specificity of these drugs. Therefore, developments of more specific remedies are needed.5

We previously developed artificially constructed promoters responsive to radiation stimulation in prostate cancer cells and reported our attempts to use them for artificial regulation of gene expression in vitro and in vivo.6 These promoters were constituted with randomly combined cis-elements of transcription factors that are activated in response to radiation in prostate cancer cells and are linked to a DNA fragment containing the TATA box signal. The most actively responsive promoter was designated clone 880-8 that was also activated in a living body by radiation. Oxidative stress was suggested to be involved in the activation as the activation was suppressed in the presence of anti-oxidants. It was pointed out that many of these anticancer drugs create conditions associated with oxidative stress. We thus considered that such drugs might be used to control gene expression mediated with clone 880-8 promoter, possibly leading to a novel cancer remedy for a combination of chemotherapy and gene therapy that can be more safe and effective.

The promoter of the egr-1 gene encoding for a zinc-finger transcription factor could be a useful tool for radiogenetic therapy,7 as it is responsive not only to growth factors and cytokines,8, 9 but also radiation stimulation.10 As this promoter is activated in response to hydrogen peroxide stimulation, oxidative stress is considered to be involved in the mechanism.11 In addition, activities of this promoter and its derivatives could be controlled by anticancer drug cisplatin (Cis) or doxorubicin (Dox), suggesting application of the promoter for a novel cancer therapy.12, 13 Quiñones and co-workers14 showed that mitomycin-C (Mmc), methylmethane sulfonate, 4-nitroquinoline oxide, MNNG (N-methyl-N′-nitro-N-nitrosoguanidin), topotecan, resveratrol and vincristine were also anticancer agents that stimulate the egr-1 promoter activity. They concluded that genotoxic stress and mitotic stress by the drugs might be triggering the promoter activation. These results suggest that there are signal-transduction pathways stimulated commonly by radiation and anticancer drugs, leading to activation of the promoter.

Prostate cancer is a particularly suitable malignancy to study as a target for gene therapy, because of some of its features. The prostate gland is easily accessible by transurethral, transperitoneal and transrectal approaches for the intratumoral administration of a therapeutic gene.15 In this study, we examine reactivity of clone 880-8 promoter to anticancer drugs just as the one that the egr-1 promoter was shown to be controlled by some of these drugs. Successfully using a reactive promoter to the drugs, a combination strategy of gene therapy and chemotherapy, is likely to result in an improved therapeutic outcome that will eventually contribute to patients’ quality of life. It is therefore the major aim of this study to examine possibility of clone 880-8 promoter that could be used for such a strategy.

Materials and methods

Cells and bacteria

A human prostate carcinoma cell line, LNCap, was purchased from the Health Science Research Source Bank (Tokyo, Japan). Cells were grown and maintained in RPMI 1640 medium supplemented with 10% fetal calf serum and appropriate antibiotics at 37 °C in a humidified 5% CO2 atmosphere.

The DH5α strain of Escherichia coli (Nippon Gene, Toyama, Japan) was used for the DNA manipulation experiments. The E. coli cells were grown in LB medium at 37 °C. All medium compositions were purchased from BD Diagnostics (Sparks, MD). DNA manipulation experiments with E. coli were performed according to the methods described by Sambrook and Russell.16

Constructions of clone 880-8 promoter

Clone 880-8 is an artificially constructed promoter reactive to radiation, increasing the expression of connected luciferase gene up to 10.4-fold 12 h after X-ray exposure at 10 Gy. Concrete processes to construct the promoter were described elsewhere.6 In brief, a promoter probe vector, pGL3-DU-TATA, was constructed by cloning a polymerase chain reaction (PCR)-amplified DNA fragment containing the TATA box signal out of the human heme oxygenase I gene promoter into upstream of the luciferase gene of pGL3-Control (Promega, Madison, WI). Equimolar amount of synthesized cis-elements of transcription factors that is responsive to radiation in prostate cancer cells, including NF-κB (5′-gggactttcc-3′), AP-1 (5′-tgactca-3′), Oct-1 (5′-atgcaaat-3′), p53 (5′-gaacatgtctaagcatgctg-3′) and Nrf-2 (5′-tgctgagtca-3′) in addition to 1/100 molar amounts of DNA fragments containing recognition sequences of HindIII and KpnI restriction enzymes, were randomly ligated. After the ligated fragments were digested with HindIII and KpnI, the resultant sequences were then cloned into the HindIII and KpnI sites of pGL3-DU-TATA that is immediate upstream of the TATA box signal, constructing 28 plasmids making up a promoter library. These plasmids were then transduced into LNCap cells and subjected to radiation stimulation. We chose the most reactive one showing 6.7-fold enhancement of luciferase activity 48 h after 10 Gy X-ray irradiation, designated as clone 880 promoter. Sequencing analysis showed this was 386 nucleotides in length and consisted of two NF-κB cis-elements, one Oct-1 cis-element, four p53 cis-elements and three Nrf-2 cis-elements, but no AP-1 cis-element. In the next step, random mutations were introduced into a DNA fragment containing clone 880 promoter by error-prone PCR, generating 30 mutated derivatives. After screening for the reactivity of the derivative promoters to radiation, clone 880-8 was chosen as the best promoter of all, showing 10.4-fold enhancement in response to 10 Gy X-ray irradiation. The chosen promoter was found to carry four point mutations. Sequence data of clone 880 and clone 880-8 were registered to GenBank under accession numbers of HQ418221 and HQ418222, respectively.

Recombinant retrovirus production and infection

A retrovirus generation vector, pRetroQ-AcGFP1-N-1, was purchased from Takara Bio (Otsu, Japan). The luciferase gene was obtained via PCR using pGL3-Control (Promega) as a template with a pair of primers of 5′-cgcgggcccaccatggaagacgccaaaaa-3′ (containing a SmaI site as a tag) and 5′-cctgaattctatcttatcatgtctgctcg-3′ (containing an EcoRI site as a tag). After digestion with EcoRI and SmaI, the amplified fragment (the luciferase gene) was purified and ligated into the EcoRI and SmaI sites of pRetroQ-AcGFP1-N-1, constructing a new plasmid designated pRetroQ-luc by replacing the GFP gene with the luciferase gene. Clone 880-8 promoter was amplified via PCR using pGL3-880-8-DU as a template with a pair of primers of 5′-ggagctcttacgcgtgctac-3′ (containing a BglII site as a tag) and 5′-tcttccagcggatagaatgg-3′ (containing an NcoI site as a tag). After digestion with BglII and NcoI, the PCR product was purified and inserted into the BglII and NcoI sites of pRetroQ-luc, creating a new plasmid, pRet-880-8-luc. A DNA fragment containing the SV40 promoter was also amplified via PCR using pGL3-Control as a template with a pair of primers of 5′-cgcagatctcatctcaattagtcagcaac-3′ (containing a BglII site as a tag) and 5′-gcgggatcctttgcaaaagcctaggcctc-3′ (containing an NcoI site as a tag). After digestion with BglII and NcoI, the PCR product was purified and inserted into the BglII and NcoI site of pRetroQ-luc, creating a new plasmid, pRet-SV40-luc, that was used as a control vector.

To construct recombinants expressing a suicide gene, the fcy::fur fusion gene encoding cytosine deaminase and uracil phosphoribosyl-transferase, a DNA fragment containing the fcy::fur fusion gene with the flag tag sequence (N-Asp-Tyr-Lys-Asp-Asp-Asp-Asp-Lys-C) was obtained via PCR using a plasmid, pORF5-Fcy::Fur (InvivoGen, San Diego, CA), as a PCR template with a pair of primers of 5′-ggctctagattatttagtagtatctgtccc-3′ (an XbaI recognition site and a flag tag coding region in addition to a stop codon was attached to the 5′ terminus) and 5′-gagacagaggagaccatggtcac-3′ (an NcoI recognition site was included). After digestion with XbaI and NcoI, the PCR products (a fcy::fur fusion gene fragment) was purified. The fragment was inserted into the XbaI and NcoI sites of pRet-880-8-luc to construct a new plasmid, pRet-880-8-fcy::fur, by replacing the fusion protein gene with the luciferase gene, and another plasmid expressing the fcy::fur fusion gene under control of the SV40 promoter for a control recombinant virus was similarly constructed by replacing the PCR-amplified fusion gene fragment with the luciferase gene of pRet-SV40-luc to create a new plasmid, pRet-SV40-fcy::fur.

In all, 100 000 AmphoPack293 cells (Takara Bio) were seeded onto a 60 mm collagen-coated cell culture dish and the following day they were transfected using CalPhosTM Mammalian Transfection Kit (Takara Bio) with the 10 μg of pRet-880-8-luc. The virus-containing conditioned medium was collected 48 h after transfection and passed through 0.45 μm filter to remove debris. Polybrene (Sigma-Aldrich, St Louis, MO) was added to the filtered medium at the final concentration of 7.0 μg ml−1. This prepared solution was used as a virus source to infect 1 × 106 LNCap cells. Infected cells were concentrated by puromycin treatment at 0.5 μg ml−1 to which infected cells were resistant, thus establishing a stably transfected cell line, LNCap-880-8-luc. A control cell line, LNCap-SV40-luc, was similarly constructed with pRet-SV40-luc and served as a control cell line.

Similar recombinant retroviruses were constructed and infected LNCap cells to establish stably transfected cell lines expressing the fcy::fur fusion gene under controls of clone 880-8 and SV40 promoters, LNCap-880-8-fcy::fur from pRet-880-8-fcy::fur and LNCap-SV40-fcy::fur cells from pRet-SV40-fcy::fur, respectively.

Drug treatment and single luciferase assay corrected with protein concentration of cell lysate

We examined six anticancer drugs that can be a choice for prostate cancer treatment, including Mmc, paclitaxel (Wako Pure Chemical Industries, Osaka, Japan), Dox, Cis, ifosfamide (Ifo), Doc (Sigma-Aldrich), 5-fluorocytosine (5-FC; Sigma-Aldrich) and 5-fluorouracil (5-FU; Sigma-Aldrich). These drugs were dissolved in dimethylsulfoxide at high concentrations (more than 1000 times to ones applied for the Figure 1 experiment) as stock solutions and stored at −20 °C. They were used for the treatment of cells after being diluted in cell growth medium to appropriate desired concentrations.

Figure 1
Figure 1

Responses of clone 880-8 to anticancer drugs. LNCap-880-8-luc cells integrated with a gene cassette of clone 880-8 and the luciferase gene were exposed to eight different kinds of drugs at various concentrations for 15 min. Luciferase activities were measured 12 h after the exposure by luciferase assay for assessing responses of clone 880-8 promoter to anticancer drugs. Enhancements of luciferase activities were expressed as fold activations that were ratios to that of the control. C: mock treatment (PBS); V: LNCap-880-8-luc exposed to 0.2% dimethylsulfoxide solution; DOX: doxorubicin; CIS: cisplatin; IFO: ifosfamide; DOC: docetaxel; MMC: mitomycin c; PAC: paclitaxel; 5-FC: 5-fluorocytosine; 5-FU: 5-fluorouracil. Error bars represent the s.d. (n=3 or 4). Statistically significant differences were indicated with asterisks (*P<0.05; **P<0.01).

Cells were treated with anticancer drugs by changing medium with one containing different kinds and various concentrations of drugs and incubated at 37 °C for 15 min. Cells were then washed three times with fresh medium and incubated again at 37 °C. At various times after drug treatment, the medium was removed and 400 μl of passive lysis buffer from the Dual Luciferase Assay Kit (Promega) was added to lyse the cells by shaking for 15 min at an ambient temperature on a platform shaker. A volume of 10 μl of cell lysate supernatant was mixed with 50 μl of luciferase assay reagent II of the kit to measure the luminescence generated by the firefly luciferase. Total protein concentration was assayed by Bradford method using Bio-Rad protein assay kit (Bio-Rad Laboratories, Hercules, CA). A luciferase expression value in a sample was corrected with protein concentration of the sample.

Quantitative real-time PCR

For evaluation of mRNA expression, quantitative real-time PCR was performed. Total RNA was collected from adherent tissue culture cells using RNeasy Mini Kit (Qiagen KK, Tokyo, Japan) with treatment with DNase I according to the manufacturer’s instructions. cDNAs were synthesized with the extracted RNA as templates, using PrimeScript RT reagent Kit (Takara Bio) according to the manufacturer’s instructions. Gene expression analysis was performed by Mx3000P QPCR System (Agillent Technologies, Santa Clara, CA) using the synthesized cDNA. Quantitative PCR measurement by real-time monitoring of SYBR Green integration into synthesized DNA was executed during a process of shuttle PCR: incubation at 95 °C for 10 s, and then 40 thermal cycles for reactions at 95 °C for 10 s and at 60 °C for 40 s, followed by reactions at 55 °C for 30 s and 95 °C for 30 s with SYBR Premix Ex Taq II (Takara Bio). After the shuttle PCR process, dissociation temperature of the synthesized DNA fragments were also observed by monitoring the release of SYBR Green from denatured DNA for confirmation of the integrity of the synthesized DNA fragment. The primer used for PCR reaction was selected by exploratory experiment from candidate primers designed by Primer 3 (v. 0.4.0) (http://frodo.wi.mit.edu/primer3/). We used two primer pairs, 5′-ttgatgagagaccccaggac-3′ and 5′-tccacaatctgcttctgcac-3′, to detect fcy::fur expression and, 5′-gagtcaacggatttggtcgt-3′ and 5′-ttgattttggagggatctcg-3′, glyceraldehyde-3-phosphate dehydrogenase expression. Relative standard curves representing several 10-fold dilutions of cDNA from a representative sample were used for linear regression analysis for other samples.

Western blot analysis

Various times after drug treatment, cells were harvested and lysed in RIPA buffer containing protease inhibitor cocktail (Nacalai Tesque, Kyoto, Japan). Cell lysate samples containing 30 μg of protein from LNCap-880-8-luc cells, LNCap-880-8-fcy::fur cells or LNCap-SV40-fcy::fur cells were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Fractionated proteins transferred on to nitrocellulose film were analyzed using ANTI-FLAG M2 monoclonal antibody as the primary antibody (Sigma-Aldrich) and goat anti-mouse IgG antibody conjugated with horseradish peroxidase as the secondary antibody (Millipore, Billerica, MA). Protein expression was visualized using ECL Western Blotting Detection Reagents (GE Healthcare UK, Buckinghamshire, UK) and the image was captured with LAS-4000 luminescence imaging analyzer (Fujifilm, Tokyo, Japan).

Cytotoxicity enhancement with enhanced suicide gene expression by Dox treatment

A stably transfected cell line treated with Dox at 5 μM for 15 min and after the cells were washed three times in fresh medium, the cells were collected. Each well of a 96-well plate was seeded with 2.5 × 104 of the cells in 50 μl medium. A measure of 50 μl of medium containing 5-FC at concentrations of 0, 2, 20 and 200 μM (so that the final concentrations of 5-FC are 0, 1, 10 and 100 μM, respectively) and the plates were incubated again at 37 °C for 24 h. After this period of incubation, the medium containing 5-FC was changed with fresh growth medium without 5-FC and incubated at 37 °C for 24 h. After addition of 10 μl WST-1 dye of Cell Counting Kit (Dojindo Laboratories, Kumamoto, Japan) to each well followed by incubation at 37 °C for 2 h, absorbance at 450 nm of the medium was measured as an indicator for viable or surviving cells with a microplate reader (Model 550; Bio-Rad Laboratories). Absorbance of the medium in a well without cells was also measured as a background that was subtracted from the measured absorbance of each well. The results were expressed as percentages of survival cell fraction (SC) calculated using the following formula: SC (%)=100 × AS/AC, where AS is absorbance of a sample at 450 nm and AC is absorbance of the control at 450 nm. The control was obtained with identically cultured cells without 5-FC treatment.

Statistical analysis

All values are expressed as means±standard deviations. Differences were assessed with the Student’s unpaired t-test. For comparison of more than two groups, one-way analysis of variance was used. Statistical significance was established at a value of P<0.05.


Responses of clone 880-8 to anticancer drugs

Clone 880-8 is a promoter responsive to radiation with the detailed mechanism; although considered unknown, oxidative stress was suggested to be involved.6 We thus investigated whether anticancer drugs that could induce oxidative stress in treated cells enhance clone 880-8 activity.

LNCap was transduced with the luciferase gene under control of clone 880-8 promoter by recombinant retrovirus vector, designated LNCap-880-8-luc. Six kinds of anti-cancer drugs including Dox, Cis, Ifo, Doc, Mmc and paclitaxel that could be choices for prostate cancer therapy in addition to two control drugs 5-FC and 5-FU were used to treat the cells. Concentrations of the drugs to treat cells were determined referring to their blood concentrations in patients when administered at their respective recommended clinical doses. Cells were treated with Dox at 1, 10 and 100 μM, Cis at 2, 20 and 200 μM, Ifo at 10, 100 and 1000 μM, Doc at 0.1, 1 and 10 μM, Mmc at 1, 10 and 100 μM, paclitaxel at 1, 10 and 100 μM, 5-FC at 1, 10 and 100 μM and 5-FU at 1, 10 and 100 μM for 15 min. Cells were incubated at 37 °C for 12 h and subjected to luciferase assay. Ifo, Doc, paclitaxel and 5-FC did not enhance the luciferase activity compared with that of LNCap-880-8-luc cells without drug treatment. In addition, although Mmc at 100 μM and 5-FU at 100 μM significantly increased luciferase activities, the values were minimal (1.34±0.05- and 1.27±0.11-fold, respectively). However, Dox increased luciferase activity up to 4.2- and 5.7-fold after treatments at 1 and 10 μM, respectively. When cells were treated with Dox at 100 μM, luciferase activity did not increase, but rather decreased. This may be because treatment with overdose of the drug affected cellular function for gene expression. Cis treatment also enhanced luciferase activity up to about 3.8-fold at 200 μM (Figure 1).

Activation of clone 880-8 by Dox

First, activation of clone 880-8 by various doses of Dox was investigated. When LNCap-880-8-luc cells were treated with 0, 1, 2, 3, 4, 5, 10 and 20 μM Dox for 15 min, luciferase activities increased as the concentration increased up to about 9.2±0.08-fold at 5 μM and then it gradually decreased (Figure 2a). As shown in Figure 2b, treatment time affected luciferase activity in LNCap-880-8-luc similar to those by concentration. When the cells were treated with Dox at 5 μM, luciferase activity increased in a treatment time-dependent manner up to 15 min, showing a peak of 8.9±0.22-fold compared with luciferase activity in LNCap-880-8-luc without Dox treatment. Luciferase activity gradually decreased as treatment time is more than 15 min and luciferase activity showed almost no enhancement when cells were treated with the drug for 60 min, suggesting that overstimulation of cells with Dox may disrupt the gene expression machinery. When LNCap-SV40-luc was treated with Dox at 5 μM for 15 min, luciferase activities were not significantly different over time (data not shown).

Figure 2
Figure 2

Reactivity of clone 880-8 to Dox treatment. (a) Luciferase activities enhanced by various concentrations of Dox in different exposure durations. LNCap-880-8-luc cells were exposed to 0, 1, 2, 5, 10 or 20 μM of doxorubicin for 15 or 30 min and luciferase activities were measured 12 h after the Dox exposure. Gray bars and black bars show results of Dox exposure for 15 and 30 min, respectively. Error bars represent the s.d. (n=3). (b) Luciferase expressions affected by exposure durations. LNCap-880-8-luc cells were exposed to Dox for 0, 4, 8, 15, 30 or 60 min. Error bars represent the s.d. (n=3). (c) Kinetic responses of luciferase activities in LNCap-880-8-luc exposed to 5 μM Dox for 15 min. Exposed cells were harvested to measure luciferase activity at 0, 4, 8, 12, 24 48 and 72 h after Dox treatment. Error bars represent the s.d. (n=3). (d) Enhancement of luciferase expression by 5 μM Dox was suppressed by an anti-oxidant, D-mannitol, in a dose-dependent manner. The bars represent the s.d. (n=3). Statistically significant differences were indicated with asterisks (*P<0.05; **P<0.01).

We then examined kinetics of luciferase activity after Dox treatment at 5 μM for 15 min. Enhancement was clearly detectable 8 h after treatment and it further increased up to 8.9±0.35-fold at the peak of 12 h after treatment. Although the activity then gradually decreased, it still showed about 1.9±0.06-fold of that observed without drug treatment even at 72 h after treatment. It was mainly regulated at the transcription level as we confirmed that the luciferase transcript increased up to 30-fold 12 h after Dox treatment at 5 μM (data not shown). Further, the luciferase enhancement by Dox treatment was suppressed dose-dependently with D-mannitol, an anti-oxidant, when it was added to cell culture immediately after the Dox treatment (Figure 2d), suggesting that oxidative stress is involved in the enhancement process.

Facilitation of cell killing effect by gene regulation with Dox treatment

Another recombinant retrovirus was constructed by replacing the luciferase gene with the fcy::fur gene that is a fusion gene whose product metabolizes 5-FC to 5-FU and facilitates the integration of 5-FU into the genome. It was transduced into LNCap cells to establish LNCap-880-8-fcy::fur. Levels of the fcy::fur transcripts detected with real-time PCR were shown to reach to its peak of about 11.5-fold enhancement 8 h after 5 μM Dox treatment for 15 min. It decreased thereafter, but even 36 h after drug treatment, it still showed about fourfold enhancement to that without drug treatment (Figure 3a). The Fcy::Fur protein synthesis was detected by immunoblotting. According to the data shown in Figure 3b, the protein was expressed to some extent without stimulation. However, increased amount was detectable from 6 h after Dox stimulation. It reached to its peak 12 to 24 h and even 48 h after Dox stimulation, the level of the protein still retained enhancement. On the other hand, the fcy::fur gene driven by the SV40 promoter showed constant amounts of enzyme proteins regardless of Dox treatment.

Figure 3
Figure 3

Increase of fcy::fur gene expression and augmentation of cell death by 5-FC after Dox treatment. (a) Kinetic responses of the fcy::fur transcript in LNCap-880-8-luc exposed to 5 μM Dox for 15 min. Exposed cells were harvested for real-time PCR to quantify fcy::fur transcript at 0, 6, 12, 24 and 36 h after Dox treatment. Error bars represent the s.d. (n=3). Statistically significant differences were indicated with asterisks (**P<0.01). Black line: fcy::fur transcript with Dox exposure; gray line: fcy::fur transcript with mock exposure. (b) Expression change of the Fcy::Fur proteins in LNCap-880-8-fcy::fur cells (880-8-Fcy::Fur) at various times after 5 μM Dox treatment for 15 min was detected by immunoblot analysis. Lysates of LNCap-880-8-luc cells (880-8-Luc) and LNCap-SV40-fcy::fur cells (SV40-Fcy::Fur) treated with or without 5 μM Dox were used as controls. β-Actin served as the internal control for protein loading. Three arrowheads and numbers in the right edge indicate molecular weights (MW). (c) and (d) Cytotoxicity enhancement of LNCap-880-8-luc cells (c) or LNCap-880-8-fcy::fur cells (d) in the presence of 5-FC after 5 μM Dox treatment for 15 min. Black line: cell survival ratio with 5 μM Dox treatment; dense gray line: cell survival ratio with 2 μM Dox treatment; thin gray line: cell survival ratio with mock treatment. Error bars represent the s.d. (n=4). Statistically significant differences were indicated with asterisks (**P<0.01). Panels (c) and (d) are plotted based on representative data of three independent experiments.

Lastly, we evaluated in vitro enhancement of cell killing effect by combination of Dox treatment and fcy::fur prodrug treatment. LNCap-880-8-luc cells cultured in the absence of 5-FC in the medium, 2 μM Dox decreased cell viability to 56.0±2.1% and 5 μM to 37.2±1.5% at 48 h after Dox treatments. On the other hand, in the case of LNCap-880-8-fcy::fur cells, 2 μM Dox treatment decreased cell viability to 55.1±4.5% and 5 μM treatment to 39.8±1.6% at 48 h after Dox treatment, indicating that either recombinant cell line was not different from each other in their sensitivity to Dox (data not shown).

When 5-FC was added to culture medium of LNCap-880-8-luc cells up to 100 μM, cell viability was not significantly changed regardless of Dox treatment (Figure 3c). In contrast, when 5-FC was added to the culture medium of LNCap-880-8fcy::fur cells, cell viability was significantly decreased in a dose-dependent manner (Figure 3d). When the cells were stimulated with 2 μM Dox, 5-FC addition of more than 1 μM caused a significant and dose-dependent viability decrease and 100 μM 5-FC diminished the viability to 41.5±5.3% of that of 2 μM Dox-treated LNCap-880-8-fcy::fur cells without 5-FC. When stimulated with 5 μM Dox, 5-FC addition of more than 1 μM decreased ratio of the viability to less than that observed with 2 μM Dox-treated LNCap-880-8-fcy::fur cells in the presence of the same concentration of 5-FC. The addition of 5-FC at 100 μM in the medium was shown to facilitate the decrease in the viability to 23.1±1.2%, clearly showing facilitation of cell killing effect by gene regulation with Dox treatment. However, without Dox stimulation, although 1 μM 5-FC did not significantly change the viability (P=0.476), 100 μM 5-FC treatment decreased the viability to 68.5±1.8% (P<0.01). This could be resulting from leaky expression of the Fcy::Fur protein even without Dox treatment as shown by immunoblot analysis in Figure 3b. Even so, these results show that the combination treatment was effective to some extent in this in vitro simulation, although further improvements to this system should be envisioned in the future.


In this report, we showed that a radiation-responsive artificial promoter that we developed was also activated by anticancer drug treatments. This activation enhanced the luciferase activity to about ninefold 12 h after Dox treatment at 5 μM for 15 min. However, higher concentrations or longer treatments interrupted the enhancement, presumably due to damage on the gene expression machinery by Dox. This promoter also enhanced expression of the fcy::fur gene, a potential therapeutic gene for cancer gene therapy, similarly to that of the luciferase gene in response to Dox stimulation. Moreover, it was shown that the addition of 5-FC that is converted to 5-FU by the fcy::fur gene product into LNCap-880-8-fcy::fur cell culture enhanced cell killing effect in vitro. Thus, such a promoter whose activity is controllable with anticancer drug treatment is a useful tool for a novel combination cancer therapy between chemotherapy and gene therapy.

Activation of clone 880-8 promoter by radiation was suppressed dose-dependently with dimethylsulfoxide and D-mannitol that are anti-oxidants.6 Similarly, enhancement of luciferase expression induced by Dox treatment was also suppressed with D-mannitol (Figure 1d). Therefore, oxidative stress is assumed to be involved in the activation of the promoter. Dox is an anthracycline antibiotic that induces DNA damage by inhibiting topoisomerase-2 activity and the drug inhibits DNA synthesis by intercalating in the double helix. During this process, a quinone ring of the Dox molecule functions as an electron acceptor to be converted to a semiquinone radical, which directly or after reaction with reactive oxygen species causes oxidative stress, bringing out its cytotoxicity.17 Cis that moderately activated clone 880-8 promoter is known to inhibit DNA synthesis by forming interstrand crosslinks of the double-stranded DNA. This is also known to generate reactive oxygen species by damaging electron transfer chains of the mitochondria.18 These oxidative stresses could possibly kindle common signal-transduction pathways to activate transcription factors that can bind to the clone 880-8 promoter. However, athough Ifo, Doc, Mmc and paclitaxel did not activate the promoter, these drugs are also reported to generate oxidative stresses.19, 20, 21, 22 Therefore, it could be assumed that kinds, locations and timings of reactive oxygen species generations differ among these drugs and any of these factors could significantly affect which pathway of signal transduction, transcription factors and then promoters may be specifically activated. In addition, we have also shown that promoters responsive to different anticancer drugs could be constructed by methods we used to construct clone 880-8, which is a radiation-responsive promoter.

Dox treatment for 15 min could induce LNCap cell killing up to 60.2–62.8% of cell populations at 5 μM and 44.0–44.9% at 2 μM. In addition, as shown in Figures 3c and d, only when LNCap-880-8-fcy::fur cells were stimulated with Dox that cell killing ratio increased even more in a manner dependent on 5-FC concentration within which cytotoxicity was not induced by 5-FC itself. The results suggest that this system could be useful and will help lead us to a specific mode of cancer treatment. However, there are still a few hurdles that we need to cross over. For instance, Figure 3b clearly shows that clone 880-8 surely expresses the fcy::fur gene product without Dox stimulation, presumably causing significant cell death at higher concentrations of 5-FC even without Dox treatment. We are currently tackling the problem.

Although Dox is not the primary choice in current clinical treatment for prostate cancer, it shows some degree of efficacy when used as a single drug or a component of multiple drug combinations.23 However, as this drug causes hazardous side effects, a safer mode of administration or a safer alternative application procedure has been desired. As we have shown in this study, clone 880-8 promoter was sensitively activated by Dox treatment. This promoter was originally developed for controlling gene expression with radiation. It is assumed that this promoter was controllable by Dox probably owing to similarity in the cellular responses to radiation and Dox. Thus, these results not only indicate that clone 880-8 could be a useful tool for a combination of gene therapy and chemotherapy, but also suggest that the methods we developed to obtain clone 880-8 could be applied for the development of a superior promoter responsive to an anticancer drug.

A combination between gene therapy and chemotherapy presumably leads to an effective cancer therapy with less hazardous side effects. This is likely because these two modalities could differently and mutually attack cancer tissue, and in addition, effective dose in either modality can be lower than when applied alone. Fervently, although more studies are needed, we hope that the findings in this study will lead us to a new modality for a more effective cancer treatment.




  1. 1.

    , , . International trends in prostate-cancer mortality in the ‘PSA era’. Int J Cancer 2001; 92: 893–898.

  2. 2.

    Immediate versus deferred treatment for advanced prostate cancer: initial result of the Medical Research Council Trial. Br J Urol 1997; 79: 235–246.

  3. 3.

    , , , , , et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 2004; 351: 1502–1512.

  4. 4.

    , , , , , et al. Docetaxel and estramustine compared with mitoxantrone and prednisone for advanced refractory prostate cancer. N Engl J Med 2004; 351: 1513–1520.

  5. 5.

    , , , , , et al. Phase-II study of docetaxel, estramustine phosphate, and carboplatin in patients with hormone-refractory prostate cancer. Eur Urol 2007; 51: 1252–1258.

  6. 6.

    , , , , , et al. Regulation of gene expression in prostate cancer cells with an artificially constructed promoter responsive to radiation. Gene Ther 2012; 19: 219–227.

  7. 7.

    , . Radiation and hypoxia inducible gene therapy systems. Cancer Metastasis Rev 2004; 23: 269–276.

  8. 8.

    , , , . Regulation of the Egr-1 gene by tumor necrosis factor and interferons in primary human fibroblasts. J Biol Chem 1992; 267: 1345–1349.

  9. 9.

    , , , , , . Growth hormone stimulates phosphorylation and activation of Elk-1 and expression of c-fos, egr-1, andjunB through activation of extracellular signal-regulated kinases 1 and 2. J Biol Chem 11998; 273: 31327–31336.

  10. 10.

    , , , , , et al. Ionizing radiation activates transcription of the EGR1 gene via CArG elements. Proc Natl Acad Sci USA 1992; 89: 10149–10153.

  11. 11.

    , , , , , . Reactive oxygen intermediates target CC(A/T)6GG sequences to mediate activation of the early growth response 1 transcription factor gene by ionizing radiation. Proc Natl Acad Sci USA 1993; 90: 2419–2422.

  12. 12.

    , , , , , et al. Transcriptional control of viral gene therapy by cisplatin. J Clin Invest 2002; 110: 403–410.

  13. 13.

    , , , , . Gene therapy vectors containing CArG elements from the Egr1 gene are activated by neutron irradiation, cisplatin and doxorubicin. Cancer Gene Ther 2005; 12: 655–662.

  14. 14.

    , , . The egr-1 gene is induced by DNA-damaging agents and non-genotoxic drugs in both normal and neoplastic human cells. Life Sci 2003; 72: 2975–2992.

  15. 15.

    , . Gene therapy for prostate cancer: where are we now? J Urol 2000; 164: 1121–1136.

  16. 16.

    , . Molecular Cloning: A Laboratory Manual, 3rd edn.Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY. 2001.

  17. 17.

    . A critical evaluation of the mechanisms of action proposed for the antitumor effects of the anthracycline antibiotics adriamycin and daunorubicin. Biochem Pharmacol 1999; 57: 727–741.

  18. 18.

    , , , , . Cisplatin-induced nephrotoxicity in porcine proximal tubular cells: mitochondrial dysfunction by inhibition of complexes I to IV of the respiratory chain. J Pharmacol Exp Ther 1997; 280: 638–649.

  19. 19.

    , , , . Ifosfamide nephrotoxicity in children: a mechanistic base for pharmacological prevention. Expert Opin Drug Saf 2009; 8: 155–168.

  20. 20.

    , . d-Limonene sensitizes docetaxel-induced cytotoxicity in human prostate cancer cells: generation of reactive oxygen species and induction of apoptosis. J Carcinogen 2009; 8: 9.

  21. 21.

    , , , . Role of the glutathione-glutathione peroxidase cycle in the cytotoxicity of the anticancer quinones. Pharmacol Ther 1990; 47: 359–370.

  22. 22.

    , , , , . Involvement of oxidative stress and caspase activation in paclitaxel-induced apoptosis of primary effusion lymphoma cells. Cancer Chemother Pharmacol 2004; 54: 322–330.

  23. 23.

    , , , , . The role of doxorubicin and epirubicin in the treatment of patients with metastatic hormone-refractory prostate cancer. Cancer Treat Rev 2008; 34: 710–718.

Download references


This work was supported in part by Grants-in-Aid for Scientific Research (C) (21500403) and for Young Scientist (B) (20377253 and 23791746) from Japan Society for the Promotion of Science.

Author information


  1. Department of Radiological Sciences, Graduate School of Medicine and Pharmaceutical Sciences for Research, University of Toyama, Toyama, Japan

    • R Ogawa
    • , N Doi
    •  & Q L Zhao
  2. Department of Urology, Graduate School of Medicine and Pharmaceutical Sciences for Research, University of Toyama, Toyama, Japan

    • A Morii
    •  & A Watanabe
  3. Department of Public Health, Graduate School of Medicine and Pharmaceutical Sciences for Research, University of Toyama, Toyama, Japan

    • Z-G Cui
  4. School of Allied Health Sciences, Kitasato University, Sagamihara, Japan

    • G Kagiya
  5. Department of Anatomy, School of Medicine, Fukuoka University, Fukuoka, Japan

    • L B Feril Jr


  1. Search for R Ogawa in:

  2. Search for A Morii in:

  3. Search for A Watanabe in:

  4. Search for Z-G Cui in:

  5. Search for G Kagiya in:

  6. Search for N Doi in:

  7. Search for Q L Zhao in:

  8. Search for L B Feril in:

Competing interests

The authors declare no conflict of interest.

Corresponding author

Correspondence to R Ogawa.