Combination of vorinostat and adenovirus-TRAIL exhibits a synergistic antitumor effect by increasing transduction and transcription of TRAIL in lung cancer cells

Abstract

Soluble TRAIL and adenovirus (ad)-TRAIL exhibit a strong antitumor effect by inducing apoptosis. Vorinostat is the histone deacetylase (HDAC) inhibitor that induces cell death in cancer cell lines and regulates the expression of epigenetically silenced genes, such as Coxackie adenoviral receptor (CAR), the receptor for adenoviral entry. We propose a new strategy in which vorinostat will induce high expression of ad-TRAIL and a strong antitumor response, and investigated the mechanism involved. The effect of vorinostat on transcription and expression of TRAIL from ad-TRAIL-transduced lung cancer cells were confirmed by reverse transciption-PCR (RT-PCR), quantitative real time-PCR and western blot assay. Anti-tumor effects were measured after cotreatment of vorinostat and ad-TRAIL, and the drug interactions were analyzed. After combined treatment of vorinostat and ad-TRAIL, apoptosis and western blot assays for Akt, Bcl-2 and caspase were performed. Vorinostat increased the expression of CAR in lung cancer cell lines and increased the expression of luciferase (luc) from ad-luc-transduced cells and TRAIL from ad-TRAIL-transduced cells. RT-PCR and quantitative real time-PCR, after sequential vorinostat treatment, revealed that vorinostat may enhance TRAIL expression from ad-TRAIL by increasing transduction through enhanced CAR expression and increasing adenoviral transgene transcription. Combined vorinostat and ad-TRAIL treatment showed the synergistic anti-tumor effect in lung cancer cell lines. Combined vorinostat and ad-TRAIL induced stronger apoptosis induction, suppression of NF-κB activation and breakdown of the anti-apoptotic molecule Bcl-2. In conclusion, the vorinostat synergistically enhanced the anti-tumor effect of ad-TRAIL by (1) increasing adenoviral transduction through the increased expression of CAR and (2) increasing adenoviral transgene (TRAIL) transcription in lung cancer cell lines.

Introduction

Gene therapy was introduced in clinics more than a decade ago. However, in spite of initial enthusiastic expectations, gene therapy still remains an investigational therapeutic modality, especially in cancer treatment. There are many problems to be solved; however, a low gene transfer rate and low gene expression are the most serious problems, with respect to clinical application. Adenovirus is one of the most potent vectors that can infect replicating and non-replicating cells and induce the expression of a transgene.1 However, the gene transfer rate and rate of expression of adenovirus is still disappointing in obtaining a significant clinical response.

Coxackie adenoviral receptor (CAR) is a major cellular receptor for adenovirus and is known as a gate for adenovirus and infection by adenovirus of tumor cells by binding with CAR expressed on the cell membrane.2 Most lung cancer cell lines used in cancer research express variable amounts of CAR on cell membranes; however, the low expression of CAR in cancer tissues,3 including lung cancer patients,4 is a main obstacle to the clinical use of adenoviral gene therapy.

Histone deacetylase (HDAC) is known as a gate closer by gene silencing.5 Aberrant HDAC activity in cancer is known to induce epigenetic silencing of numerous genes, including CAR. Previously, our group reported that sodium butyrate, an HDAC inhibitor, induces CAR expression in CAR-deficient bladder cancer cell lines and increases an anti-tumor effect by adenovirus (ad)-p16.6 Numerous HDAC inhibitors have been shown to enhance CAR expression and enhance the anti-tumor effect of adenoviral gene therapy by increasing gene expression, and in turn increasing transduction efficiency.

In 1997, Dion et al.7 showed that sodium butyrate amplified adenoviral transgene expression at the transcriptional level. Furthermore, we found that the addition of an HDAC inhibitor after adenoviral transduction enhanced the expression, suggesting the role of HDAC inhibitors on transcription of the adenoviral transgene.

Vorinostat (suberoylanilide hydrozamic acid (SAHA); Zolinza; Merck Research Laboratories; Whitehouse Station, NJ) was the first HDAC inhibitor to be approved for clinical use in the treatment of cutaneous T-cell lymphoma.8 Several papers have focused on the combination of vorinostat and adenoviral gene therapy. VanOosten et al.9 reported that vorinostat enhances the ad-TRAIL anti-tumor effect by both increasing transduction through increased CAR expression and increased caspase-2 activity in TRAIL-resistant prostate cancer cells. Lillehammer et al.10 also reported that co-treatment of ad-TRAIL and vorinostat caused additive or synergistic growth inhibition by augmenting the apoptotic pathway through the death receptor pathway.

The combination of vorinostat and soluble TRAIL protein has also been investigated. Vorinostat sensitizes TRAIL-resistant breast cancer cells to TRAIL-induced apoptosis. Increased DR5 expression by vorinostat was suggested as the mechanism of sensitization.11 Recently, Carlisi et al.12 showed that vorinostat sensitizes hepatocellular carcinoma cells to TRAIL by increasing DR5 expression and decreasing c-FLIP. Also, sequential treatment of vorinostat followed by TRAIL is more effective in growth suppression of tumor formed by TRAIL-resistant breast cancer cells in BALB/c nude mice. Downregulation of nuclear factor (NF)-κB and its gene products (Bcl-2 and VEGF), and upregulation of DR4 and DR5 by vorinostat, were suggested as mechanisms.13

In the current study, we have focused on the role of vorinostat on adenoviral transduction and transcription of transfected genes in cancer cells. We hypothesized that the HDAC inhibitor vorinostat, may enhance the anti-tumor effect of ad-TRAIL by increasing transduction through the increased expression of CAR and by amplifying transcription of TRAIL by epigenetic regulation. The luciferase (luc) assay with ad-luc revealed that luciferase expression is increased by vorinostat, regardless of its timing. Initially, we expected that vorinostat before transduction would increase luciferase expression by increasing the transduction rate. However, addition of vorinostat after ad-luc transduction still increased the luciferase expression. Therefore, we reasoned that vorinostat would increase the transcription of the transduced TRAIL gene, as vorinostat increased the epigenetically silenced genes.

Materials and methods

Recombinant adenoviruses, vorinostat and lung cancer cell lines

Recombinant adenovirus expressing human full-length TRAIL was constructed in our laboratory. Briefly, a full length of human cDNA was cloned into the NotI restriction site on pAdTrack-CMV of the AdEasy system. Cloned pAdTrack-TRAIL and pAdEasy-1 were co-transfected into E. coli (BJ5183). After genetic recombination in E. coli, ad-TRAIL was generated and confirmed by DNA sequencing and production of TRAIL protein. Ad-TRAIL was cytomegalovirus-promoter driven, E1-deleted and replication-incompetent.14 Ad-null (E1-deleted adenovirus without any therapeutic gene) and ad-luc were also cytomegalovirus-promoter driven, E1-deleted and replication-defective adenoviruses, which were made in our laboratory.

Vorinostat, an HDAC inhibitor, was provided by Merck Research Laboratories. Vorinostat powder was dissolved with dimethyl sufoxide to 20 mM stock solution and kept in a −70 °C freezer. Vorinostat was diluted with media before the experiment.

Human lung cancer cell lines (A549 (human lung adenocarcinoma); NCI H460 (human lung large cell carcinoma)) were purchased from the American Tissue Culture Collection (Manassas, VA) and maintained in RPMI 1640+10% fetal bovine serum in albumin.

The pan-caspase inhibitor, z-VAD-fmk was purchased from R&D Systems (Minneapolis, MN).

Male BALB/c nude mice were purchased from Japan SLC (Hamamatsu, Japan). Animal experiments were approved by the Institutional Animal Care and Use Committee of Seoul National University Hospital in Seoul, Korea.

Antibodies against poly-ADP ribose polymerase, caspase-3 (3G2), cleaved caspase-3 (Asp175), Bcl-2 (50E3), phospho-Akt (Thr308) and DR5 were purchased from Cell Signaling Technology (Beverly, MA).

Antibodies against Bcl-2 (N-19), β-actin (I-19), TRAIL (H-257) and CAR (H-300) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).

Western blot assay and FACS for CAR

Lung cancer cell lines (A549, NCI H460) were treated with vorinostat (0.1–10 μM) for 48 h. A western blot assay for CAR was done with antibody to CAR. Western blot was developed with SuperSignal West Pico Chemiluminescent Substrate Kits (Thermo Scientific, Rockford, IL). Under the same conditions, lung cancer cells were stained with CAR antibody and secondary antibody (FITC-conjugated F(ab9)2 of antimouse IgG) for 30 min at 4 °C, and the cells were analyzed immediately on a flow cytometer (FACSCalibre; Becton Dickinson, San Jose, CA).

Luciferase assay

Lung cancer cells were transduced with ad-luc with an indicated multiplicity of infection (moi) and treated with vorinostat (0.1–10 μM) for 24 h before (pre-transduction) or for 48 h after (post-transduction), or 24 h before and 48 h after transduction (pre- and post-transduction). Luciferase assays were done 48 h after transduction, according to the manufacturer’s manual (Luciferase assay system; Promega, Madison, WI) and measured by relative light units using LMAX II384 (Molecular Devices, Sunnyvale, CA).

In vivo tumor imaging for luciferase expression was performed to determine the effect of vorinostat on luciferase expression in an animal tumor model. Briefly, lung cancer xenografts were established by injecting A549 cells (2 × 106 per mouse) into the subcutaneous tissue of nude mice (female BALB/c, 6 weeks old). When measurable tumors (7–8 mm in length) were formed after 2 weeks, vorinostat (30 mg kg−1) or phosphate-buffered saline in 100 μl was injected into the peritoneum (day 0). On day 1, ad-luc (2 × 108 pfu) was injected intratumorally and vorinostat (30 mg kg−1) or phosphate-buffered saline injections were repeated on day 2. On day 4, d-luciferin (5 mg per mouse) (Gold Bio Technology, St Louis, MO) was injected into the peritoneum. In vivo bioluminescence imaging was taken after 10 min with IVIS 100 (Caliper Life Sciences, Hopkinton, MA).15 Three different sets of experiments were done.

Western blot for TRAIL

Lung cancer cells were transduced with ad-TRAIL and treated with vorinostat before, after, and before and after transduction, as described previously. Western blot for TRAIL was done 48 h after transduction, using the same kit as described above.

RT (reverse transcription)-PCR for TRAIL mRNA

Lung cancer cells (A549 and NCI H460) were transduced with ad-TRAIL (10 moi) and treated with vorinostat (2 μM) for 24 h before (pre-transduction) or for 24 h after (post-transduction), or 24 h before and after transduction (pre- and post-transduction). Reverse transcription-PCR (RT-PCR) for TRAIL was done 24 h after transduction. Briefly, RNA was extracted by TRIzol reagent (Invitrogen, Carlsbad, CA). Reverse transcription was done for 1 μg of RNA with a PrimeScript 1st strand cDNA synthesis kit (Takara, Shiga, Japan). PCR for TRAIL was done with primers for TRAIL (hTRAIL forward: 5′-IndexTermCAGGATCATGGCTATGATGGAGGTC-3′, hTRAIL reverse 5′-IndexTermGCTGTTCATACTCTCTTCGTCATTG-3′). A total of 23 and 20 cycles of PCR were done for A549 and NCI H460 cells, respectively, to maximize the difference of the PCR product's amount. Densities of TRAIL band were measured and compared with cancer cells not treated with vorinostat.

Another RT-PCR was performed to investigate the role of vorinostat on the transcription of TRAIL gene after transduction. Lung cancer cells (A549 and NCI H460) were transduced with ad-TRAIL (10 moi) for 1 h and were washed with phosphate-buffered saline thoroughly. Then vorinostat (2 μM) was added to the medium (post-transduction). RT-PCR for TRAIL was measured at 1 and 24 h after the addition of vorinostat. The bands for TRAIL were compared with the band from cells not treated with vorinostat.

Quantitative real time RT-PCR

Real time-PCR for TRAIL was performed to quantify mRNA transcription from ad-TRAIL in cancer cells. Briefly, lung cancer cells (A549 and NCI H460) were transduced with ad-TRAIL (30 moi) for 1 h. Non-infected adenoviral particles were washed out thoroughly and cells were maintained with complete media for another hour and vorinostat (0.1, 1 and 2 μM) was added. Total RNA was isolated by TRIzol (Invitrogen), 24 h after addition of vorinostat and reverse transcription was done, as described previously. Quantitative real time-PCR for TRAIL was done with the primers for TRAIL (Hs00921974_m1 (TNFSF10); PE Applied Biosystems, Foster City, CA). Real time-PCR with 150 ng of cDNA was done by TaqMan Gene Expression Assays (PE Applied Biosystems) and the levels of PCR product were investigated by measuring relative quantitation.

Another set of real time-quantitative PCR for TRAIL was performed. Lung cancer cells were transduced with ad-TRAIL (30 moi) for 1 h and kept in complete media for an additional hour, and vorinostat (1 μM) was added to media. RNA was isolated 1, 4 and 24 h after vorinostat. After generation of cDNA by reverse transcription, quantitative real time-PCR for TRAIL was performed. The changes of TRAIL according to time and the presence of vorinostat were determined.

Growth inhibition assay

Lung cancer cells (A549 and NCI H460) were treated with vorinostat (0.1–5 μM) and ad-TRAIL (10-50 moi) at various combinations in 96-well plates (3 × 104 cells per well). MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) assay (CellTiter 96 Aqueous One Solution Cell Proliferation Assay; Promega Corporation) was performed at 72 h. Drug interactions were analyzed using Calcusyn software (Biosoft, Ferguson, MO) by measuring the combination index. Another set of experiments with ad-null and vorinostat was done to differentiate the effect of vorinostat on other genes of adenovirus, instead of the TRAIL gene.

NF-κB p65 subunit activation assay

We performed the NF-κB p65 subunit activation assay using the TransAM NF-κB p65 Activation Assay (Active Motif, Carlsbad, CA) to analyze the role of NF-κB activation for the combination of vorinostat and ad-TRAIL. Lung cancer cells (A549 and NCI H460 cells; 3 × 106 per plate) were treated with vorinostat (0.1–2 μM) for 24 h, then transduction with ad-TRAIL or ad-null (30 moi) and maintained with media containing vorinostat for another 48 h. At 48 h, cells were lysed and nuclear portions were extracted using a Nuclear Extract kit (Active Motif). An NF-κB p65 subunit activation assay was done according to the manufacturer's protocol.

Apoptosis and downstream pathway assays

We also determined the effect of ad-TRAIL and vorinostat on apoptosis. Lung cancer cells (A549; 2 × 105 cells per well in a 6-well plate) were treated with vorinostat (1 μM, pre- and post-transduction) and ad-TRAIL (single or combination). At 48 h after transduction, subG1 analysis was performed. Western blot assays for Akt, Bcl-2 and caspase-3 were performed in the same conditions, except the addition of pan-caspase inhibitor (40 μM of z-VAD-fmk after ad-TRAIL transduction) to investigate the role of the caspase pathway. Changes of DR5 by vorinostat were measured by western blot assay in lung cancer cells (A549 and NCI H460).

Results

Vorinostat increased the expression of CAR on lung cancer cell lines

Western blot revealed that CAR expression of lung cancer cell lines (A549 and NCI H460) were increased by the addition of vorinostat (Figure 1a). Fluorescence-activated cell sorting for CAR after vorinostat treatment confirmed again that vorinostat enhanced CAR on the cell surface in a dose-dependent manner (Figure 1b).

Figure 1
figure1

Change of Coxackie adenoviral receptor (CAR) expression on lung cancer cells by addition of vorinostat. Lung cancer cells (A549 and NCI H460) were treated with vorinostat at the indicated concentration for 48 h. Western blot assay for TRAIL and fluorescence-activated cell sorting (FACS) analysis for CAR on cell surfaces was performed. Both western blot (a) and FACS analysis for CAR (b) showed that vorinostat increased CAR expression in lung cancer cells in a dose-dependent manner.

Vorinostat increased luciferase expression from ad-luc-transduced lung cancer cells in vitro and in vivo

Addition of vorinostat (1–10 μM) to the media before and after adenoviral transduction increased luciferase expression from lung cancer cells transduced with ad-luc (Figure 2a). The enhancing effect of vorinostat is still present, whether or not vorinostat treatment was confined to the pre- or post-transduction period. The effect of vorinostat was the strongest when it was treated for the pre- and post-transduction period. Interestingly, the enhancing effect of post-transduction vorinostat was more potent than pre-transduction vorinostat (Figure 2b). The enhancing effect of pre-transduction vorinostat on luciferase expression should be linked to increased CAR expression, followed by increased adenoviral transduction, and that of post-transduction vorinostat should be linked to increased transcription of transduced luciferase gene.

Figure 2
figure2

Vorinostat increased the expression of luciferase (luc) from adenovirus (ad)-luc-transduced lung cancer cells. (a) Lung cancer cells (A549 and NCI H460) were treated with vorinostat (0.1–10 μM) for 12 h, then transduced with ad-luc (10–50 moi) for 1 h in RPMI. After a 1-h transduction, complete media with vorinostat was added. In vitro luciferase assay was performed 48 h after transduction. Addition of vorinostat dramatically increased luciferase expression from ad-luc-transduced lung cancer cells, according to dose. (b) To investigate whether vorinostat acted on transduction of ad-TRAIL or on transcription of TRAIL gene in cells, vorinostat was added in the pre- and post-transduction periods (group A), the post-transduction period (group B) and the pre-transduction period (group C) in the same setting as the previous experiment. An in vitro luciferase assay was performed 48 h after transduction. The effect of post-transduction vorinostat on luciferase expression was much stronger than pre-transduction vorinostat (P<0.01 by ANOVA). However, luciferase expression in cells treated with vorinostat pre- and post-transduction was the strongest (P<0.01). This finding suggests that the vorinostat effect on transcription of the TRAIL gene may be more potent than increasing the transduction rate of ad-luc by increased Coxackie adenoviral receptor (CAR) expression of vorinostat on cells (Y-axis: relative light unit). (c) In vivo bioluminescence-tumor imaging revealed stronger luciferase expression by vorinostat treatment. Lung cancer xenografts were established by subcutaneous injection of A549 (2 × 106 cells per animal) in subcutaneous tissues of nude mice (BALB/c). Vorinostat (30 mg kg−1 in 100 μl of phosphate-buffered saline) or 100 μl of phosphate-buffered saline injections on days 0 and 2. Intratumoral injection of ad-luc (2 × 108 pfu) was done on day 1. Tumor bioluminescence imaging was done on day 4. This experiment was repeated three times. Tumor treated with vorinostat showed much stronger luminescence than tumor treated with phosphate-buffered saline (details in the text).

These observations suggested that vorinostat can act on both adenoviral transduction and transcription; however, the effect on transcription was more potent.

The enhancing effect of vorinostat on ad-luc-transduced cells was confirmed in an in vivo tumor-imaging model. Intraperitoneal injections of vorinostat before and after intratumoral injection of ad-luc strongly increased the bioluminescence from tumors. However, tumors from mice, not treated with vorinostat, exhibited very weak bioluminescence (Figure 2c). This finding could be evidence that this combination strategy is also effective in an in vivo tumor model.

Vorinostat enhanced the expression of TRAIL from ad-TRAIL-transduced lung cancer cells

A western blot assay for TRAIL revealed that the addition of vorinostat, before and after transduction, increased TRAIL expression in a dose-dependent manner (Figure 3a). The addition of vorinostat before transduction or after transduction also increased the expression of TRAIL (Figure 3b). These findings suggested that vorinostat could act on adenoviral transduction and transcription of the adenoviral TRAIL gene.

Figure 3
figure3

Effect of vorinostat on the expression of adenovirus (ad)-TRAIL-transduced lung cancer cell lines. (a) Lung cancer cells (A549 and NCI H460) were transduced with ad-TRAIL (5 and 10 moi) for 1 h and vorinostat (0.1, 1, and 5 μM) was added to media both before and after transduction. Proteins were extracted after 48 h and western blot assay for TRAIL was done. Addition of vorinostat remarkably increased the expression of TRAIL in a dose-dependent manner. (b) Lung cancer cells (A549) were transduced with ad-TRAIL (10 moi) and the addition of vorinostat (2 μM) for 12 h before transduction, for 48 h after transduction, or both before and after transduction. Western blot assays were performed 48 h after transduction. Addition of vorinostat increased the expression of TRAIL, but the effect of vorinostat after transduction was much stronger than that of vorinostat before transduction.

RT-PCR and quantitative real time RT-PCR revealed that vorinostat enhanced the transcription of the TRAIL gene in ad-TRAIL-transduced lung cancer cells

The effect of vorinostat on transcription of the adenoviral TRAIL gene in lung cancer was investigated by RT-PCR and quantitative real time RT-PCR. RT-PCR for TRAIL showed that vorinostat increased the transcription of TRAIL, regardless of the duration and timing of addition. Vorinostat, after ad-TRAIL transduction, increased the transcription of TRAIL and the enhancement was stronger than that of vorinostat before transduction (Figure 4a). RT-PCR and quantitative real time RT-PCR for TRAIL also confirmed that addition of vorinostat 1 h after ad-TRAIL transduction increased the transcription of TRAIL, which directly confirmed that vorinostat increased the transcription of the TRAIL gene (adenoviral transgene) of lung cancer cells (Figures 4b and c).

Figure 4
figure4

(a) Reverse transcription-PCR (RT-PCR) for TRAIL. Lung cancer cell lines (A549 and NCI H460) were transduced with adenovirus (ad)-TRAIL (10 moi) for 1 h and incubated in complete media. Vorinostat (2 μM) was treated before, after, or before and after transduction. RT-PCR was performed 24 h after transduction. Both pre- and post-transduction vorinostat increased the mRNA expression level; however, addition of vorinostat after transduction was more effective than pre-transduction. (b) Sequential RT-PCR. Vorinostat (2 μM) was added 1 h after transduction and RT-PCR for TRAIL was performed 1 and 24 h thereafter. Increase of mRNA for TRAIL by vorinostat became evident at 24 h after vorinostat treatment. (c) Quantitative real-time PCR. Lung cancer cells were transduced with ad-TRAIL (30 moi) for 1 h. Vorinostat (0.1, 1, and 2 μM) was added after 1 h. Quantitative real-time PCR for TRAIL after RNA extraction was performed at 24 h. Another quantitative real-time PCR for TRAIL was done at 0, 1, 4 and 24 h after addition of vorinostat (1 μM). The TRAIL mRNA levels were significantly increased by vorinostat at 24 h compared with untreated controls (P<0.01 in A549, P<0.05 in NCI H460 by ANOVA).

Synergistic anti-tumor effect of vorinostat and ad-TRAIL in lung cancer cells

Co-treatment with vorinostat and ad-TRAIL exhibited a strong synergistic anti-tumor effect in two lung cancer cells (A549 and NCI H460). Combination indices determined by Calcusyn software (Biosoft) were <1.0 in most combinations (Figure 5a). However, combination of vorinostat and ad-null showed no interaction (Figure 5b). This finding suggests that the interaction of vorinostat and ad-TRAIL originated from the effect of vorinostat on the TRAIL gene instead of other adenoviral structural genes, as ad-null contained the same adenoviral genes with ad-TRAIL, except TRAIL gene.

Figure 5
figure5

(a) Synergistic interaction of vorinostat and adenovirus (ad)-TRAIL on lung cancer cytotoxicity. Lung cancer cells (A549 and NCI H460: 5 × 103 cells per well in a 96-well plate) were treated with vorinostat (0.1–5 μM) and ad-TRAIL (10–100 moi.). MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxy-methoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) assay was done after 96 h. Drug interactions were analyzed by Calcusyn software (Biosoft; right column). Vorinostat and ad-TRAIL showed strong synergistic cytotoxic interaction on both lung cancer cell lines (vorinostat: μM). (b) No interaction of vorinostat and ad-null on lung cancer cytotoxicity. To differentiate the effect of vorinostat on the TRAIL gene of adenovirus or any other adenoviral structural genes, we treated the cells with vorinostat and ad-null (E1-deleted adenovirus with CMVie promoter without any carrying gene) in the same experimental settings with (a). No interaction of vorinostat and ad-null was found on the cytotoxicity in lung cancer cell lines. This means that the synergistic interaction of vorinostat and ad-TRAIL arises from the effect of vorinostat on the TRAIL gene.

Increased induction of apoptosis and suppression of NF-κB activation by combination of vorinostat and ad-TRAIL

Sub-G1 analysis by fluorescence-activated cell sorting revealed that combined treatment of vorinostat (1 μM) and ad-TRAIL (10 moi) increased the sub-G1 fraction of A549 from 0.62% (vorinostat alone) and 6.52% (ad-TRAIL alone) to 12.16% (Figure 6a). NF-κB p65 subunit activity assay showed that ad-TRAIL induced marked activation of NF-κB p65, and vorinostat also induced weak activation of NF-κB p65. However, co-treatment of vorinostat and ad-TRAIL suppressed NF-κB p65 activation induced by ad-TRAIL (Figure 6b).

Figure 6
figure6

(a) Enhanced apoptosis by combination of vorinostat and adenovirus (ad)-TRAIL. A549 cells were treated with vorinostat and ad-TRAIL. Lung cancer cells (A549) were treated with vorinostat (1.0 μM) or ad-TRAIL (10 moi) or both. After 48 h, cells were detached and stained with propidium iodide (PI). The proportion of the sub-G1 population was analyzed by fluorescence-activated cell sorting (FACS) analysis. The sub-G1 proportions of vorinostat-treated or ad-TRAIL-treated A549 were 0.62 and 6.52%, respectively. However, the sub-G1 proportion was increased to 12.16% by the combination of vorinostat and ad-TRAIL. (b) Blockade of nuclear factor (NF)-κB activation by combination of vorinostat and ad-TRAIL. Lung cancer cells (NCI H460 and A549; 3 × 106 cells) were treated with vorinostat (0.1–2 μM) and ad-TRAIL (30 moi.) for 48 h. Ad-null (30 moi) or no virus was used as a control. Nuclear portion of cells were extracted by a Nuclear Extract kit (Active Motif, Carlsbad, CA). NF-κB activation was measured by NF-κB p65 ELISA assay kit (TransAM NF-κB p65 Activation Assay; Active Motif). Vorinostat induced moderate NF-κB p65 activation in untreated or ad-null treated lung cancer cells. Ad-TRAIL (10 moi) induced the marked activation of NF-κB p65 subunit. In contrast, co-treatment of vorinostat remarkably suppressed ad-TRAIL-induced NF-κB p65 activation. This finding strongly suggests that blocking of NF-κB activation, one of the major anti-apoptotic pathways, by vorinostat is one of the mechanisms that induce the synergistic effect of ad-TRAIL and vorinostat.

Analysis of apoptotic pathway after combination of vorinostat and ad-TRAIL

Western blot assay for apoptosis-related molecules revealed that co-treatment induced increased breakdown of poly-ADP ribose polymerase and caspase 3. Interestingly, the combination of vorinostat and ad-TRAIL induced the breakdown of Bcl-2 (anti-apoptotic protein), which was reversed by the pan-caspase inhibitor (Figure 7a). This may be another mechanism of the synergistic anti-tumor effect of this combination. As reported before,11 vorinostat increased the expression of DR5, the target for TRAIL in lung cancer cells (Figure 7b).

Figure 7
figure7

Analysis of apoptotic pathways. (a) Lung cancer cells (NCI H460: 2 × 105 cells per well) were treated with vorinostat (1 μM) and adenovirus (ad)-TRAIL (30 moi). After 48 h, proteins were extracted and proteins associated with apoptosis were analyzed by western blot assays. A pan-caspase inhibitor (z-VAD-fmk, 40 μM) 1 h after transduction was added to analyze the effect of caspase pathway activation. Both vorinostat and ad-TRAIL induced the cleavage of poly-ADP ribose polymerase and phospholylation of Akt, which was not blocked by z-VAD-fmk. Combined treatment of vorinostat and ad-TRAIL induced the breakdown of Bcl-2 and caspase 3, which was restored by z-VAD-fmk. (b) Vorinostat increased the expression of DR5 in two lung cancer cell lines.

Discussion

Poor gene transfer rate and poor gene expression are the major limitations for gene therapy for cancers. Adenoviral gene therapy has been used in cancer gene therapy extensively for its high gene transfer efficiency and high expression rate compared with other viral vectors;1, 16 however, many clinical trials with adenovirus have revealed that the infectivity and expression rate are still low.

Transduction of adenovirus into cells requires the binding of adenovirus and CAR on the cell surface, so the level of CAR on the cell surface is one of the major factors that determine the gene transfer rate.2 The expression of the adenoviral transgene is also determined by the transcription rate of the adenoviral transgene, which is successfully transfected into cells.17 Therefore, high transduction and transcription are very important factors for effective adenoviral gene therapy.

The combination of HDAC inhibitors to adenoviral gene therapy for cancer is an ideal combination option, because HDAC inhibitors could induce the expression of CAR in various cancer cell lines and facilitate the entry of adenovirus into cells in addition to the anti-tumor effect. We previously reported that sodium butyrate induces CAR expression in CAR-deficient bladder cancer cell lines and improve the anti-tumor effect of ad-p16.6

In addition to butyrate, several HDAC inhibitors were shown to be effective in increasing CAR expression. HDAC inhibitor (FK228: FR901228) increased the CAR RNA level and subsequently increased adenoviral transgene expression, preferentially in malignant cells.18 Hemminki et al.19 reported that trichostatin A and FR901228 enhanced adenoviral transgene expression mainly by increasing CAR expression.

Most reports on the effect of HDAC inhibitors on adenoviral gene therapy have focused on increased adenoviral transduction by increased CAR on cancer cells. However, we noticed that HDAC inhibitors could increase the transcription of the adenoviral transgene. In addition to the report of Dion,7 another HDAC inhibitor valproic acid, also enhanced gene expression from various viral vectors, including adenovirus, adeno-associated virus and herpes virus, by enhancing viral gene transcription.17

TRAIL, a member of tumor necrosis factor (TNF) superfamily is an attractive cytokine because of strong anti-tumor effect by inducing apoptosis and relative sparing of normal tissues.20, 21

Adenoviral delivery of TRAIL is an alternative method for TRAIL treatment, with local apoptosis-inducing effect and a systemic antitumor immune response.22, 23 We already reported that ad-TRAIL could overcome TRAIL resistance in several TRAIL-resistant cell lines.14 Recently, tumor-cell selective modification of soluble TRAIL by fusing epidermal growth factor receptor-directed antibody (scFv425) showed a more selective and potent antitumor effect,24 and a recombinant adenovirus that can express fusion protein (scFv425-TRAIL) demonstrated its clinical significance.25 However, resistance to TRAIL prevents its clinical use and requires the combination with other chemicals. Combination of TRAIL and HDAC inhibitors are an attractive trial, because some properties of HDAC inhibitors can ameliorate the mechanisms of TRAIL resistance.

Several previous studies have revealed that the combination of TRAIL and soluble vorinostat have a synergistic interaction by several mechanisms. Vorinostat induces the increased expression of death receptor 511 and the decreased expression of c-FLIP, and finally provoked the rapid production of TRAIL-DISC formation and activation of the caspase system.12 Real time-PCR for DR5 mRNA revealed that enhanced DR5 expression by vorinostat is due to increased DR5 mRNA transcription. Combined treatment of vorinostat and TRAIL enhanced the therapeutic effect on a TRAIL-resistant breast cancer model.13

In this study, we determined the target of the vorinostat effect on transgene expression by separating the timing of vorinostat. We treated cells with vorinostat before adenovirus infection to analyze the effect on transduction, or after adenoviral infection to analyze the effect on transcription.

At first, we performed the experiment with ad-luc, which contained luciferase gene as a reporter gene. Enhancing effect of post-transduction vorinostat on luciferase expression was more potent than that of pre-transduction vorinostat. Enhancing effect of post-transduction vorinostat should be linked to increased transcription, even though we didn’t perform RT-PCR for luciferase. However, we confirm that phenomenon by RT-PCR and quantitative RT-PCR of TRAIL gene.

Western blot assay, RT-PCR and quantitative real time RT-PCR confirmed that vorinostat act directly on the transcription of the TRAIL gene, which was already transferred to cells by adenoviral vectors, in addition to increasing transduction of ad-TRAIL by increased CAR expression.

Furthermore, we found several other mechanisms that might be responsible for the synergistic interactions. Ad-TRAIL induced the strong activation of NF-κB p65 subunit, one of the major anti-apoptotic pathways. Interestingly, co-treatment of vorinostat effectively suppressed NF-κB activation, even though vorinostat alone induced the weak activation of NF-κB p65 subunit. We did not reveal the mechanism of this finding. At least, two papers already reported same finding. Carlisi et al.12 reported that the addition of vorinostat to TRAIL induced remarkable decrease in both p65 and p50 NF-κB binding activity of hepatoma cells, which was enhanced by TRAIL alone. They suggested that activated caspase-3 cleaved both p65 and p50 subunit of NF-κB, because the pan-caspase inhibitor z-VAD completely abolish this phenomenon. Imre et al.26 reported that co-treatment of vorinostat suppressed the inducible NF-κB activation by TNF-α in same lung cancer cell lines we used (A549 and NCI H460). They suggested that it was related with downregulation of TNF-α receptor by HDAC inhibitor, followed by attenuated NF-κB nuclear translocation.

Bcl-2, known as an anti-apoptotic protein,27 was reduced by co-treatment of vorinostat and ad-TRAIL, which was reversed by a pan-caspase inhibitor. As the HDAC inhibitor-induced apoptosis was inhibited by Bcl-2 over-expression,28 this Bcl-2 reduction by vorinostat and ad-TRAIL may have considerable clinical significance. Effective inactivation of two anti-apoptotic molecules by vorinostat and ad-TRAIL combination might be responsible for the synergistic interaction to some extent.

In conclusion, an HDAC inhibitor (vorinostat) enhanced the anti-tumor effect of ad-TRAIL synergistically by (1) increasing adenoviral transduction through the increased expression of CAR and (2) increasing transcription of the adenoviral transgene in lung cancer cell lines.

References

  1. 1

    Hemminki A, Alvarez RD . Adenoviruses in oncology: a viable option? BioDrugs 2002; 16: 77–87.

  2. 2

    Bergelson JM, Cunningham JA, Droguett G, Kurt-Jones EA, Krithivas A, Hong JS et al. Isolation of a common receptor for Coxsackie B viruses and adenoviruses 2 and 5. Science 1997; 275: 1320–1323.

  3. 3

    Jee YS, Lee SG, Lee JC, Kim MJ, Lee JJ, Kim DY et al. Reduced expression of coxsackievirus and adenovirus receptor (CAR) in tumor tissue compared to normal epithelium in head and neck squamous cell carcinoma patients. Anticancer Res 2002; 22: 2629–2634.

  4. 4

    Wang Y, Wang S, Bao Y, Ni C, Guan N, Zhao J et al. Coxsackievirus and adenovirus receptor expression in non-malignant lung tissues and clinical lung cancers. J Mol Histol 2006; 37: 153–160.

  5. 5

    Gray SG, Ekstrom TJ . The human histone deacetylase family. Exp Cell Res 2001; 262: 75–83.

  6. 6

    Lee CT, Seol JY, Park KH, Yoo CG, Kim YW, Ahn C et al. Differential effects of adenovirus-p16 on bladder cancer cell lines can be overcome by the addition of butyrate. Clin Cancer Res 2001; 7: 210–214.

  7. 7

    Dion LD, Goldsmith KT, Tang DC, Engler JA, Yoshida M, Garver Jr RI . Amplification of recombinant adenoviral transgene products occurs by inhibition of histone deacetylase. Virology 1997; 231: 201–209.

  8. 8

    Mann BS, Johnson JR, Cohen MH, Justice R, Pazdur R . FDA approval summary: vorinostat for treatment of advanced primary cutaneous T-cell lymphoma. Oncologist 2007; 12: 1247–1252.

  9. 9

    VanOosten RL, Earel Jr JK, Griffith TS . Histone deacetylase inhibitors enhance Ad5-TRAIL killing of TRAIL-resistant prostate tumor cells through increased caspase-2 activity. Apoptosis 2007; 12: 561–571.

  10. 10

    Lillehammer T, Engesaeter BO, Prasmickaite L, Maelandsmo GM, Fodstad O, Engebraaten O . Combined treatment with Ad-hTRAIL and DTIC or SAHA is associated with increased mitochondrial-mediated apoptosis in human melanoma cell lines. J Gene Med 2007; 9: 440–451.

  11. 11

    Butler LM, Liapis V, Bouralexis S, Welldon K, Hay S, Thai le M et al. The histone deacetylase inhibitor, suberoylanilide hydroxamic acid, overcomes resistance of human breast cancer cells to Apo2 L/TRAIL. Int J Cancer 2006; 119: 944–954.

  12. 12

    Carlisi D, Lauricella M, D’Anneo A, Emanuele S, Angileri L, Di Fazio P et al. The histone deacetylase inhibitor suberoylanilide hydroxamic acid sensitises human hepatocellular carcinoma cells to TRAIL-induced apoptosis by TRAIL-DISC activation. Eur J Cancer 2009; 45: 2425–2438.

  13. 13

    Shankar S, Davis R, Singh KP, Kurzrock R, Ross DD, Srivastava RK . Suberoylanilide hydroxamic acid (Zolinza/vorinostat) sensitizes TRAIL-resistant breast cancer cells orthotopically implanted in BALB/c nude mice. Mol Cancer Ther 2009; 8: 1596–1605.

  14. 14

    Seol JY, Park KH, Hwang CI, Park WY, Yoo CG, Kim YW et al. Adenovirus-TRAIL can overcome TRAIL resistance and induce a bystander effect. Cancer Gene Ther 2003; 10: 540–548.

  15. 15

    Lee CT, Lee YJ, Kwon SY, Lee J, Kim KI, Park KH et al. In vivo imaging of adenovirus transduction and enhanced therapeutic efficacy of combination therapy with conditionally replicating adenovirus and adenovirus-p27. Cancer Res 2006; 66: 372–377.

  16. 16

    Bauerschmitz GJ, Barker SD, Hemminki A . Adenoviral gene therapy for cancer: from vectors to targeted and replication competent agents (review). Int J Oncol 2002; 21: 1161–1174.

  17. 17

    Fan S, Maguire CA, Ramirez SH, Bradel-Tretheway B, Sapinoro R, Sui Z et al. Valproic acid enhances gene expression from viral gene transfer vectors. J Virol Methods 2005; 125: 23–33.

  18. 18

    Goldsmith ME, Kitazono M, Fok P, Aikou T, Bates S, Fojo T . The histone deacetylase inhibitor FK228 preferentially enhances adenovirus transgene expression in malignant cells. Clin Cancer Res 2003; 9: 5394–5401.

  19. 19

    Hemminki A, Kanerva A, Liu B, Wang M, Alvarez RD, Siegal GP et al. Modulation of coxsackie-adenovirus receptor expression for increased adenoviral transgene expression. Cancer Res 2003; 63: 847–853.

  20. 20

    Wiley SR, Schooley K, Smolak PJ, Din WS, Huang CP, Nicholl JK et al. Identification and characterization of a new member of the TNF family that induces apoptosis. Immunity 1995; 3: 673–682.

  21. 21

    Griffith TS, Lynch DH . TRAIL: a molecule with multiple receptors and control mechanisms. Curr Opin Immunol 1998; 10: 559–563.

  22. 22

    Griffith TS, Anderson RD, Davidson BL, Williams RD, Ratliff TL . Adenoviral-mediated transfer of the TNF-related apoptosis-inducing ligand/Apo-2 ligand gene induces tumor cell apoptosis. J Immunol 2000; 165: 2886–2894.

  23. 23

    Griffith TS, Stokes B, Kucaba TA, Earel Jr JK, VanOosten RL, Brincks EL et al. TRAIL gene therapy: from preclinical development to clinical application. Curr Gene Ther 2009; 9: 9–19.

  24. 24

    Bremer E, Samplonius DF, van Genne L, Dijkstra MH, Kroesen BJ, de Leij LF et al. Simultaneous inhibition of epidermal growth factor receptor (EGFR) signaling and enhanced activation of tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) receptor-mediated apoptosis induction by an scFv:sTRAIL fusion protein with specificity for human EGFR. J Biol Chem 2005; 280: 10025–10033.

  25. 25

    Bremer E, van Dam GM, de Bruyn M, van Riezen M, Dijkstra M, Kamps G et al. Potent systemic anticancer activity of adenovirally expressed EGFR-selective TRAIL fusion protein. Mol Ther 2008; 16: 1919–1926.

  26. 26

    Imre G, Gekeler V, Leja A, Beckers T, Boehm M . Histone deacetylase inhibitors suppress the inducibility of nuclear factor-kappaB by tumor necrosis factor-alpha receptor-1 down-regulation. Cancer Res 2006; 66: 5409–5418.

  27. 27

    Reed JC . Bcl-2 family proteins. Oncogene 1998; 17: 3225–3236.

  28. 28

    Peart MJ, Tainton KM, Ruefli AA, Dear AE, Sedelies KA, O’Reilly LA et al. Novel mechanisms of apoptosis induced by histone deacetylase inhibitors. Cancer Res 2003; 63: 4460–4471.

Download references

Acknowledgements

This study is supported by grants from the Korea Science and Engineering Foundation (2008-00728) to C-T Lee.

Author information

Correspondence to C-T Lee.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kim, D., Park, M., Lee, C. et al. Combination of vorinostat and adenovirus-TRAIL exhibits a synergistic antitumor effect by increasing transduction and transcription of TRAIL in lung cancer cells. Cancer Gene Ther 18, 467–477 (2011). https://doi.org/10.1038/cgt.2011.11

Download citation

Keywords

  • vorinostat
  • adenovirus-TRAIL
  • CAR
  • transduction
  • transcription

Further reading