SEARCH

SEARCH BY CITATION

Keywords:

  • double-stranded decoy oligonucleotides;
  • gastric cancer;
  • ets-1 transcription factor;
  • gene therapy

Abstract

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The ets-1 transcription factor plays an important role in cell proliferation, differentiation, apoptosis and tissue remodeling. Aberrant ets-1 expression correlates with aggressive tumor behavior and poorer prognosis in patients with various malignancies. This study evaluated the efficacy of double-stranded decoy oligonucleotides targeting ets-1-binding cis elements for the suppression of ets-1 in treatment of a peritoneal dissemination model of gastric cancer. In vitro, MTT assay was performed to evaluate the effect of the ets-1 decoy on cell growth. Electrophoretic mobility shift assay (EMSA) was performed to determine ets-1 activity. In vivo, the effect of the ets-1 decoy was investigated in the peritoneal dissemination nude mice model. Disseminated nodules were analyzed immunohistochemically. Ets-1 decoy, but not scrambled decoy, significantly inhibited cell growth in 2 gastric cancer cell lines, which showed overexpression of ets-1 protein by inhibiting the binding activity of ets-1. In the peritoneal dissemination model, the ets-1 decoy significantly suppressed the disseminated nodules, and tended to prolong the survival rate. PCNA index, microvessel density and VEGF expression were also reduced in peritoneal tumors treated with ets-1 decoy. Intraperitoneal injection of ets-1 decoy inhibited peritoneal dissemination of gastric cancer in a nude mice model. The results indicate that the decoy strategy for ets-1 offers a promising therapy for patients with incurable peritoneal dissemination of gastric cancer, most of which show overexpression of ets-1 protein. © 2007 Wiley-Liss, Inc.

The prognosis of advanced gastric cancer, especially in serosa-invading tumors, remains poor even after curative resection, and in these cases peritoneal dissemination originating from free cancer cells seeded from the primary gastric cancer is the most common type of recurrence.1, 2 To date, various treatments have been used for peritoneal dissemination of gastric cancer, including aggressive surgery3, 4 and intraabdominal or systemic chemotherapy.5, 6, 7, 8, 9, 10 However, the contributions of these therapies to patient survival have been unsatisfactory.

The E26 transformation-specific (ets) family, known as a transcription factor, contains a conserved winged helix-turn-helix DNA binding domain, and regulates gene expression by binding to so-called ets-binding sequences found in promoter/enhancer regions of their target genes.11 This family is involved in a diverse array of biological functions, including cellular growth, migration and differentiation.12, 13, 14, 15 Ets-1 was the first member of the ets family to be identified.11 Recently, ets-1 has been shown to play a role in tumor progression in various kinds of malignant tumors.16, 17, 18, 19, 20, 21, 22, 23 Ets-1 is induced by and is required for the activation of several genes involved in angiogenesis and remodeling of extracellular matrix (ECM), such as vascular endothelial growth factor (VEGF), angiopoietins (Ang), fibroblast growth factor (FGF), epidermal growth factor (EGF), urokinase plasminogen activator (uPA), matrix metalloproteases (MMPs) and integrin β3.24, 25, 26, 27 Overexpression of ets-1 correlates with grade of malignancy and poorer prognosis in several types of tumors, including breast,16, 17 lung,18 ovarian,19, 20 colorectal21 and gastric cancers.22, 23 The correlation between ets-1 overexpression and aggressive tumor behavior suggests that ets-1 may be an attractive molecular target for cancer therapy.

The inhibition of transcription by using short sequence oligonucleotides such as antisense DNA and RNAi is a novel approach for molecular-targeted therapies.28, 29, 30 Recently, the decoy strategy has been developed and is considered as an effective tool for transcriptional suppression of downstream genes.31, 32 The double-stranded decoy oligonucleotides closely correspond to the response element within the promoter region of downstream genes. By achieving sufficient oligonucleotide concentrations in target cells, the authentic interaction between a transcription factor and its endogenous response element in genomic DNA is impaired with subsequent suppression of gene expression.33 The decoy strategy is also applicable for a loss-of-function approach at the pretranscriptional and transcriptional levels by directly targeting a transcription factor.31 Furthermore, decoy oligonucleotides provide novel methods for global control of the expression of genes that are regulated through an enhancer, unlike antisense oligonucleotides and RNAi, which only target ets-1 the mRNA of one specific gene.

The purpose of this study was to evaluate the efficacy of double- stranded decoy oligonucleotides in suppressing the transcriptional function of ets-1 and for the treatment of peritoneal dissemination in an animal model of gastric cancer.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Cells

For in vitro studies, we used gastric cancer cell lines (MKN28, MKN45 and MKN74) and human umbilical vein endothelial cells (HUVEC). Gastric cancer cell lines were cultured using RPMI 1640 (Life Technologies, Grand Island, NY) supplemented with 10% fetal bovine serum (GIBCO BRL, Grand Island, NY) plus 100 U/ml penicillin and 100 U/ml streptomycin (GIBCO BRL) at 37°C in a humidified atmosphere in the presence of 5% CO2. HUVECs were isolated and grown in MCDB-131 supplemented with 10 ng/ml EGF.

For in vivo studies, we used 2 gastric cancer cell lines, MKN45-EGFP and MKN45-P. We previously reported that the gastric cancer cell line, MKN-45-EGFP, which was transfected with an enhanced green fluorescent protein (EGFP)-expressing plasmid vector.34 The MKN45-EGFP cell line allowed us to visualize the microfoci of peritoneal tumors with higher detection sensitivity. The MKN45-P cells, which were extracted from ascetic fluid of patients with peritoneal carcinomatosa, were used to provoke ascites in the abdominal cavity of nude mice.

Immunohistochemical analysis for ets-1 expression in gastric cancer cell lines

Three gastric cancer cell lines cultured on slide glasses were paraformaldehyde-fixed and were subjected to immunohistochemical analysis. The endogenous peroxidase activity was blocked and nonspecific binding was blocked by incubating with blocking serum. Incubation with the primary antibody, rabbit polyclonal anti-human Ets-1 antibodies (1:4,000 dilution; Santa Cruz Biotechnology, Santa Cruz, CA), was performed overnight at room temperature. Subsequent immunodetection was done using the Elite Vector Stain ABC System (Vector Laboratories, Burlingham, CA). Color visualization was performed using 3,3′-diaminobenzidine tetrahydrochloride (DAB) as the chromogen substrate. The cells were counterstained with hematoxylin and were observed at 400× magnification.

Decoy oligonucleotide sequences

The sequences of double-stranded oligonucleotides were utilized as follows. Ets-1 decoy (consensus sequences are underlined):

5′-AATTCACCGGAAGTATTCGA-3′, 3′-TTAAGTGGCCTT CATAAGCT-5′ Scrambled decoy: 5′-ATACTACGAGCATATG CATG-3′, 3′-TATGATGCTCGTATACGTAC-5′. The scrambled decoy was designed as a 20-mer oligonucleotide without any consensus sequence for binding sites for transcription factors and used as a negative control. Each pair was annealed by heating to 80°C for 10 min followed by gradual cooling. The mixture was stored at 4°C after the reaction.

MTT assay

MTT [3-(4, 5-dimethylthiazol)-2, 5-diphenyltetrazolium bromide] (Sigma Chemical, St. Louis, MO) assay was performed to evaluate the effect of ets-1 decoy oligonucleotides on cell growth of 3 gastric cancer cell lines and HUVEC. Each cell line was incubated in 96-well microtiter plates (5 × 103 cells/well) with 100 μl medium for 24 hr. The medium was then changed for fresh medium containing ets-1 or scrambled decoy at the concentration of 0, 0.1, 0.2, 0.5, 1, 2, 5 or 10 μM. After incubation for 96 hr, MTT (10 μl/well, 5 mg/ml in phosphate-buffered saline; PBS) was added, and the plates were incubated for 4 hr. Subsequently, 100 μl of isopropanol/0.04 N hydrochloric acid solution was added to each well. The absorbance was measured at a test wavelength of 550 nm, and a reference was measured at the wavelength of 650 nm using a microplate reader for the calculation of cell growth.

Extraction of nuclear protein

MKN28, MKN45 and MKN74 cells were seeded onto 15-cm dishes and cultured until 80% confluent. The medium was then changed for fresh culture medium containing ets-1 or scrambled decoy at the concentration of 10 μM and incubation was continued for 24 hr.

Nuclear extraction was performed at 4°C according to the instructions provided by the manufacturer (CelLytic™ NuClear™ Extraction Kit; Sigma). Briefly, after rinsing with PBS, cells were collected by scraping and centrifuged for 5 min at 450g. Same aliquots of cell pellets were suspended with 1 ml of lysis buffer (10 mM N-2-hydroxyetylpiperazine-N′-ethanesulphonic acid [HEPES], pH 7.9, 1.5 mM MgCl2, 10 mM KCl), 10 μl of 0.1 M dithiothreitol (DTT) and 10 μl of protease inhibitor cocktail (containing 4-2-aminoethyl benzenesulfonyl fluoride, pepstatin A, bestain, leupeptin, aprotinin and trans-epoxysuccinyl-L-leucyl-amido-butane), incubated for 15 min and then centrifuged. The cell pellets were resuspended in 400 μl of lysis buffer and homogenized with a syringe with a narrow-gauge needle. The suspension was centrifuged for 20 min at 10,000g, and the crude nuclei pellets were mixed with 140 μl of extraction buffer (20 mM HEPES, pH 7.9, 1.5 mM MgCl2, 0.42 M NaCl, 0.2 mM EDTA, 25% glycerol) containing 0.1 M DTT and protease inhibitor. After gentle shaking for 30 min and centrifugation for 5 min at 20,000g, the supernatants were collected as nuclear extracts. The protein concentrations were determined by Bio-Rad Protein Assay (Bio-Rad Laboratories), and the extracts were stored at −70°C.

Electrophoretic mobility shift assay

To determine ets-1 activity, we performed electrophoretic mobility shift assay (EMSA) according to the protocol recommended by the manufacturer (Ets-1/PEA3 EMSA Gel-Shift Kits; Panomics, Redwood City, CA). For each binding reaction, 5 μg of nuclear extract was incubated with 1 μg poly-d (I–C), 2 μl of EMSA binding buffer (1.5% glycerol, 75 mM KCl, 0.375 mM DTT, 12.5 mM NaCl, 0.375 mM phenylmethylsulfonyl fluoride [PMSF]) and 1 ng of biotin-labeled transcription factor probe for 30 min at room temperature.

The reaction mixtures were loaded onto 6% TBE gels (Invitrogen, San Diego, CA) and electrophoresed at 120 V for 1 hr at 4°C in 0.5% Tris-Borate-EDTA (TBE). Probes were transferred to nylon membrane (Biodyne B; Pall) in an electroblotting device at 300 mA for 30 min. The membrane was baked for 30 min at 85°C in a dry oven. Biotinylated oligonucleotides were detected by probing with streptavidin-conjugated horseradish peroxidase and visualized by enhanced chemiluminescence. A biotin-labeled probe corresponding to the sequence of the ets-1 binding site was used with the sequence 5′-GATCTCGAGCAGGAAGTTCGA-3′. To confirm specific binding for ets-1/DNA complex, cold probe (biotin-unlabeled probe), which competes with the labeled probe in binding to ets-1 protein, was added and the corresponding band to the ets-1/DNA complex was confirmed.

The effect of ets-1 decoy on fluorescent intensity in MKN45-EGFP cells

The stability of fluorescent intensity in MKN45-EGFP cells was confirmed by the following procedures. MKN45-EGFP cells (1.5 × 106) were seeded onto 9-cm dishes and cultured for 24 hr. The medium was then changed for fresh medium containing ets-1 decoy at the concentration of 0, 0.5 or 5 μM. After incubation for 96 hr, the areas of total cells and fluorescent cells were calculated using stereomicroscopy and fluorostereomicroscopy (BZ 8000, KEYENCE, Osaka, Japan). The average ratio of fluorescent cell area/total cell area in 5 fields of each dish was compared in dishes with or without treatment of ets-1 decoy.

Animal model

Four-week-old female BALBc nu/nu mice were purchased from Japan Clea (Tokyo, Japan) and maintained in specific pathogen-free conditions. MKN45-EGFP cells (1 × 106) in 0.5-ml saline were injected into the peritoneal cavity of mice on Day 0. Ten mice were allocated into 3 groups: control, scrambled decoy and ets-1 decoy, respectively. The decoy oligonucleotide or PBS (control group) was injected into the peritoneal cavity at a concentration of 40 nmol/0.5 ml every other day from the next day after tumor cell injection (Day 1) until Day 13. The mice were sacrificed on Day 14. The antitumor effect of the decoy was assessed by the total fluorescent tumor weight under observation, using fluorostereomicroscopy. Disseminated nodules with fluorescence were fixed in formalin, embedded in paraffin and stored for immunohistochemical analyses. We also evaluated the effect of ets-1 decoy oligonucleotides on the survival of mice with peritoneal carcinomatosa. MKN45-P cells (5 × 106) in 0.5-ml saline were injected into the peritoneal cavity of mice on Day 0. Six mice were allocated into 3 groups: control, scrambled decoy and ets-1 decoy, respectively. The decoy oligonucleotide or PBS (control group) was injected into the peritoneal cavity at a concentration of 40 nmol/0.5 ml every other day from the next day after tumor cell injection (Day 1). Furthermore, the decoy oligonucleotide or PBS was injected 7 times into the peritoneal cavity each at a concentration of 40 nmol/0.5 ml every other day under the condition of no peritoneal carcinomatosa to evaluate the toxic effect of the decoy nucleotides (4 mice in each group). Body weight and oral intake were measured daily and major organs (heart, lung, liver, kidney, spleen) were assessed pathologically.

The experimental protocol was approved by the Ethics Review Committee for Animal Experimentation of Osaka University School of Medicine.

Immunohistochemical staining and evaluation

Paraffin-embedded sections (5 μm thick) were used for immunohistochemical staining with antibodies for proliferating-cell nuclear antigen (PCNA), platelet/endothelial cell adhesion molecule (PECAM)-1 (a marker of microvessels) and VEGF proteins using the streptavidin-biotin-complex method (Histofine SAB-PO kit, Nichirei, Tokyo). Briefly, the slides were deparaffinized and incubated with 10 mM sodium citrate buffer (pH 6.0) for 40 min at 95°C for antigen retrieval. After cooling to room temperature and washing with PBS, endogenous peroxidase activity was blocked by 1% hydrogen peroxide and 0.1% sodium azide in distilled water for 15 min. Nonspecific binding was blocked by incubating with blocking serum. After washing with PBS, the slides were incubated overnight at 4°C with each primary antibody: PCNA mouse monoclonal antibody (diluted at 1:400), PECAM-1 goat polyclonal antibody (1:750), or VEGF rabbit polyclonal antibody (1:100). After incubation with each biotinylated secondary antibody and streptavidin, proteins were visualized by incubating in 0.02% 3, 3-DAB in a 0.05 M Tris buffer with 0.01% hydrogen peroxidase. The sections were counterstained with hematoxylin.

PCNA expression in the tumor cells was evaluated as the percentage of positive tumor cells in a 400× field. The average of 5 fields was recorded as the PCNA index. VEGF expression was evaluated using a semiquantitative scoring system: 0 for absence of immunostaining, +1 for <20% positive tumor cells, +2 for 20–50% positive tumor cells and +3 for >50% positive tumor cells. The immunostaining score was evaluated as the average of five 200× fields in the slide section. Microvessel density was evaluated with PECAM-1 (CD31) immunostaining. Microvessels were counted in five 200× fields and the average count was recorded for scoring. Randomly chosen microscopic fields were evaluated by 2 independent investigators.

Statistical analysis

All values were expressed as mean ± SEM. Student's t test was adopted for statistical analyses of the tumor weight, and immunohistochemical staining scores in 3 groups: control, scrambled decoy and ets-1 decoy. Survival curves were drawn using the Kaplan–Meier method, and comparisons of survival distribution were made using the Log rank test. A p value less than 0.05 was considered significant.

Results

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Ets-1 expression in gastric cancer cell lines

Ets-1 protein expression was examined by the immunohistochemical analysis. Figure 1 showed ets-1 protein expression in 3 gastric cancer cell lines: MKN45, MKN28 and MKN74. Ets-1 expression was observed in all 3 cell lines. Highest expression was observed in MKN45 (Fig. 1a), and MKN28 showed weak expression in only cytoplasm of the cells (Fig. 1b).

thumbnail image

Figure 1. Ets-1 protein expression in 3 gastric cancer cell lines, MKN45 (a), MKN28 (b) and MKN74 (c) by immnohistochemical analysis (400× magnification). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Download figure to PowerPoint

Effect of ets-1 decoy on cell growth in vitro

MTT assay was performed to examine the direct effect of ets-1 decoy on cell growth using 3 gastric cancer cell lines, MKN28, MKN45, MKN74 and HUVEC. The growth rate of cells treated with scrambled decoy oligonucleotides was not affected compared to the control group in all 4 cell lines. In contrast, the growth rate of cells treated with ets-1 decoy was inhibited in a dose-dependent manner in MKN45 and MKN74 cells (Fig. 2). The efficacy of the ets-1 decoy in growth suppression differed between cell lines. In MKN45 cells, the ets-1 decoy was markedly effective and 5 nM of ets-1 decoy reduced the absorbance by about 80% of the untreated value (ED50 = 0.8257 nM). MKN74 cells also showed a dose-dependent reduction of absorbance (ED50 = 5.069 nM) after treatment with ets-1 decoy. However, the inhibition of growth was not significant in MKN28 cells and HUVEC.

thumbnail image

Figure 2. Effect of ets-1 decoy oligonucleotides on cell growth. MTT assays were performed on 3 gastric cancer cell lines (MKN28, MKN45 and MKN74) and HUVEC. Each cell line (5 × 103 cells/well) was incubated for 96 hr with ets-1 or scrambled decoy at the concentration of 0, 1, 2, 5 or 10 nM. Data represent the mean ± SEM of 3 different experiments performed in triplicate.

Download figure to PowerPoint

Inhibition of ets-1 activation by transfection of ets-1 decoy in vitro

We speculated that the treatment of cells with the ets-1 decoy would interfere with the binding of ets-1 to ets-1-specific DNA sequence in the promoter region of genes. To confirm this speculation, representative gastric cancer cell lines were treated with either ets-1 decoy or the scrambled decoy at 10 nM for 24 hr, and the binding efficacy was evaluated by EMSA using nuclear extracts from these cell lines. Ets-1 decoy-treated cells showed reduction of the binding activity of labeled ets-1 probe in the gel shift assay compared to the scrambled decoy-treated cells or control cells. The expression of ets-1 protein and the effect of ets-1 decoy on DNA binding varied among the 3 cell lines. In MKN45 and MKN74 cells, the binding activity was suppressed markedly by ets-1 decoy (Fig. 3). However, MKN28 cells showed weak expression of ets-1 compared to the other 2 cell lines, even without decoy treatment, and the difference in ets-1 DNA binding between with and without ets-1 decoy treatment was not significant.

thumbnail image

Figure 3. Inhibition of ets-1 binding activity on gastric cancer cell lines by ets-1 decoy oligonucleotides. Ets-1 EMSA of nuclear extracts from 3 gastric cancer cell lines, MKN28, MKN45 and MKN74, preincubated with 10 nM of ets-1 or scrambled decoy for 24 hr was performed as indicated. Arrows indicate migration of DNA-binding complexes. NS, nonspecific band; FP, free probe.

Download figure to PowerPoint

Effect of ets-1 decoy on peritoneal dissemination in the mouse model

In a preliminary experiment, mice were injected with fluorescein isothiocyanate (FITC)-labeled ets-1 decoy oligonucleotides (10 nmol/0.5 ml) into the peritoneal cavity under no peritoneal dissemination condition, and sacrificed at Days 1, 2 or 3 to examine the distribution and absorption of the decoy oligonucleotides in the peritoneal cavity. Intense FITC fluorescence was observed in the whole peritoneal cavity on Day 1. The following day, FITC fluorescence was observed in the whole peritoneal cavity despite the reduction compared to Day 1. However, at Day 3, the fluorescence level was markedly reduced and was limited to the peritoneum near the injection site (data not shown). On the basis of these results, we designed the decoy treatment schedule of intraperitoneal injection every other day from the day after tumor cell injection.

The effect of ets-1 decoy oligonucleotides was investigated in the peritoneal dissemination model. In this model, MKN45-EGFP cells (1 × 106) transduced with GFP-expressing plasmid vector, which enabled us to assess micrometastatic nodules with a high level of detection sensitivity, were injected into the peritoneal cavity of each mouse (n = 10).34 In this model, we confirmed the formation of peritoneal dissemination on the day after intraperitoneal injection of MKN45-EGFP cells (data not shown). At Day 14, after decoy injection, we evaluated fluorescent nodules, representing peritoneal disseminated tumors, under fluorescence stereomicroscopy. The total weight of fluorescent nodules was significantly lower in the ets-1 decoy group (24.2 ± 18.7 mg) than in the control group (60.1 ± 29.7 mg, p = 0.0046) and scrambled decoy group (43.5 ± 17.1 mg, p = 0.0274) (Fig. 4a). In the control and scrambled decoy groups, all mice were found to form large fluorescent nodules in the omentum. In contrast, in the ets-1 decoy group, the omental nodules were smaller than those of the other 2 groups and there were no or only few microfoci in the peritoneum other than omentum (Fig. 4b).

thumbnail image

Figure 4. Effect of ets-1 decoy oligonucleotides in the peritoneal dissemination model in nude mice. MKN45-EGFP cells (1 × 106) were injected into the peritoneal cavity of mice on Day 0. The decoy oligonucleotide (ets-1 or scrambled, 40 nmol) or PBS was injected into the peritoneal cavity every other day from Day 1 to Day 13. (a) Total weight of fluorescent nodules. Data represent the mean ± SEM. (b) Manifestations of peritoneal dissemination in mice. Green fluorescent nodules indicate peritoneal disseminated tumors.

Download figure to PowerPoint

With regard to body weight and food intake during the treatment, there were no significant differences among the 3 groups (data not shown).

Evaluation of inhibition of peritoneal dissemination by ets-1 decoy

We performed PCNA, CD31 (PECAM-1) and VEGF immunohistochemical staining with peritoneal microfoci obtained from mice treated with ets-1 decoy, scrambled decoy and controls to verify the mechanism of growth inhibition by ets-1 decoy. PCNA-positive cells, reflecting tumor growth, were significantly fewer in the ets-1 decoy group than in the other 2 groups. The PCNA index of the ets-1 decoy group (39.6% ± 9.4%) was significantly lower than those of the control group (61.9% ± 6.8%, p < 0.0001) and scrambled decoy group (57.4% ± 8.5%, p = 0.0003) (Fig. 5a). Microvessel density, represented by the number of PECAM-1-positive vessels, was also significantly lower in the ets-1 decoy group (13.4 ± 6.4 vessels/area) than in the control group (32.1 ± 11.8 vessels/area, p = 0.0003) and scrambled decoy group (23.3 ± 6.8 vessels/area, p = 0.0035) (Fig. 5b). Furthermore, VEGF expression was also significantly downregulated in the ets-1 group compared to the control and scrambled decoy groups (p = 0.0025, p = 0.0397, respectively, Fig. 5c).

thumbnail image

Figure 5. Immunohistochemical staining of peritoneal dissemination nodules of nude mice. (a) PCNA expression in tumor cells was evaluated by the percentage of positive tumor cells in a 400× field and the average of 5 fields was recorded as the PCNA index. Data represent the mean ± SEM. (b) Microvessels evaluated by PECAM-1 (CD31) were counted in 5 200× fields and the mean count was recorded for scoring. Data represent the mean ± SEM. (c) VEGF expression was evaluated using a semiquantitative scoring system as indicated and the mean value of 5 200× fields was recorded. Data represent the mean ± SEM.

Download figure to PowerPoint

No effect of ets-1 decoy on fluorescent intensity in MKN45-EGFP cells

Figure 6 showed the pictures of MNK45-EGFP cells cultured in medium containing 0 and 5 μM ets-1 decoy by stereomicroscopy (left), fluorostereomicroscopy (center) and both fusion (right). The average ratio of fluorescent cell area/total cell area in 5 fields of each dish containing ets-1 decoy at the concentration of 0, 0.5 and 5 μM was 94.6% ± 4.2%, 93.8% ± 2.9% and 94.0% ± 2.4%, respectively. These results confirmed that ets-1 decoy had no effect in the fluorescent intensity of MKN45-EGFP cells.

thumbnail image

Figure 6. The pictures of MNK45-EGFP cells cultured in medium containing 0 (a–c) and 5 μM ets-1 decoy (d–f) by stereomicroscopy (a, d), fluorostereomicroscopy (b,e) and both fusion (c,f). The average ratio of fluorescent cell area/total cell area in 5 fields of each dish containing ets-1 decoy at the concentration of 0 and 5 μM was 94.6 ± 4.2 and 94.0 ± 2.4. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Download figure to PowerPoint

Effect of ets-1 decoy on survival of mice with peritoneal dissemination

In this model, MKN45-P cells (5 × 106) were injected into the peritoneal cavity of each mouse (n = 6), and the decoy oligonucleotide or PBS was injected into the peritoneal cavity every other day from the day after tumor cell injection. The survival curves of the control and scrambled decoy groups were similar (p = 0.8671, Fig. 7). In comparison, the ets-1 decoy group showed a better prognosis than the control (p = 0.0391) and scrambled decoy groups (p = 0.058).

thumbnail image

Figure 7. Effect of ets-1 decoy oligonucleotides on survival rate in the peritoneal dissemination model. MKN45-P cells (5 × 106) were injected into the peritoneal cavity on Day 0. The decoy oligonucleotide (ets-1 or scrambled, 40 nmol) or PBS was injected into the peritoneal cavity every other day from Day 1.

Download figure to PowerPoint

Toxicity of decoy oligonucleotides in the mouse model

To evaluate the toxic effect of the decoy, each decoy oligonucleotide or PBS was injected 7 times into the peritoneal cavity (4 mice in each group). Such treatment did not produce significant changes in body weight and food intake in the 3 groups. Furthermore, no significant obstruction was detected pathologically in all major organs tested (data not shown).

Discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Recently, oligonucleotides corresponding to the consensus binding sequence of a specific transcription factor have been explored as tools for manipulating gene expression in living cells.32 The decoy strategy using double-stranded oligonucleotides for binding sequences has been developed as a new class of antigene strategy for clinical application.31 The occupation of the DNA binding site of the transcription factor by the decoy renders the protein incapable of subsequent binding to the promoter regions of target genes. Bielinska et al.35 were the first group to describe the utility of such decoys as a tool for investigating the function of transcription factors in cell lines. To date, decoy strategies have also been applied as therapeutic agents and the potential efficacy of decoy oligonucleotides that target other transcription factors has been described in cancer treatment,36, 37, 38, 39, 40 myocardial infarction41, 42 and rheumatoid arthritis.43

The ets family consists of ∼30 genes and the ets-1 gene was the first identified as the cellular homologue of the viral oncogene v-ets of the avian transforming retrovirus E26.11 The gene product of this family is a transcription factor, controlling various cellular functions in cooperation with other families of transcription factors and cofactors.11, 12, 13, 14 All members have an activation or a repression domain for transcription and an evolutionarily conserved ets domain, which can bind to the 5′-GGAA/T-3′ core motif.15 Each member have been found to bind specifically to each 10 bp of sequences containing the core motif,11 and we designed Ets-1 decoy including this specific sequence (ACC GGAAGTA).

The target genes for the ets transcription factor include oncogenes, tumor suppressor genes, apoptosis-related genes, differentiation-related genes, angiogenesis-related genes and invasion-related genes.12, 13, 14 With regard to ets-1, previous reports indicated that it regulates the expression of MMP, urokinase type-plasminogen activator,17, 18, 43 angiopoietin24, 25, 26 and VEGF.16, 20, 23

Overexpression of ets-1 has been reported to correlate with tumor malignancy and prognosis of patients with gastric,23 breast16 and ovarian cancers.20 In particular, ets-1 overexpression in gastric tumors correlated significantly with lymph node and distant metastasis and poorer prognosis of patients.44, 45 Furthermore, reduction of cell proliferation and MMP-9 expression led by ets-1 decoy have been reported in glioma cells.46 Therefore, ets-1 appears to a promising molecular target for gastric cancer therapy, since the targeting of this protein may suppress tumor proliferation and improve patient prognosis by inhibition of invasion and tumor angiogenesis.

In the present in vitro study, we performed MTT assays and EMSA using 3 gastric cancer cell lines (MKN28, MKN45 and MKN74), treated with ets-1 or scrambled decoy to assess the hypothesis that the treatment of cells with the ets-1 decoy would interfere with the binding of ets-1 to ets-1-specific DNA response elements and lead to inhibition of cell growth. Consistent with our hypothesis, treatment with ets-1 decoy led to marked inhibition of cell growth by interfering with the binding activity of ets-1 decoy in MKN45 and MKN74 cell lines. However, the effect was different between the cell lines and ets-1 decoy had no significant effect on cell growth and binding activity in MKN28 cells, which showed low expression of ets-1 protein compared to MKN45 and MKN74 cells. Furthermore, the ets-1 expression was limited in cytoplasm of MKN28 cells by immunohistochemical analysis. These results indicate that in MKN28 cells, the ets-1 signal transduction may not play an important role in cell growth.

We have established a peritoneal dissemination model of nude mice using MKN45-EGFP cells, produced by transducing MKN45 cells with an EGFP-expressing plasmid vector.34 We applied this model to assess the therapeutic effect of ets-1 decoy against peritoneal dissemination of gastric cancer cells. The stable GFP expression in cancer cells enabled us to evaluate micrometastasis in the abdominal cavity, which was otherwise not detectable by conventional microscopy. Furthermore, we confirm that ets-1 decoy has no effect on the reduction of the fluorescent intensity in MKN45-EGFP cells. In the ets-1 decoy group, GFP nodules, which indicated peritoneal dissemination, were significantly suppressed compared to the scrambled decoy and control groups. We also evaluated peritoneal dissemination nodules by immunohistochemical staining for PCNA, PECAM-1 and VEGF to clarify the effect of ets-1 decoy. The results of staining supported the hypothesis that ets-1 decoy inhibits tumor growth and progression by inhibiting tumor angiogenesis. With regard to its effect on prognosis, ets-1 decoy tended to improve survival, although the effect was not significant compared to the scrambled decoy group (p = 0.058). In this point, we need to change the dosage, treatment schedule and/or drug delivery system to apply the ets-1 decoy therapy for clinical use. We also demonstrated the lack of toxicity of oligonucleotides used in our ets-1 decoy as determined by the amount of food intake, body weight gain and pathological findings of major organs.

In our experiments, the scrambled decoy showed the tendency, though not significant, to suppress peritoneal dissemination in the assessment of total weight and microvessel density of fluorescent nodules in the MKN45-EGFP nude mouse model. This might be caused by the nonspecific side effects with high dose of oligonucleotides; however, there are few published studies referred to the toxicity of oligonucleotides in vivo.28, 29, 40, 41 We performed invitro assay with 0.1–10 μM of decoy, whereas the therapeutic dose employed in vivo was 80 μM. Higher doses of decoy oligonucleotides might raise other possible mechanisms for the inhibitory effects of the peritoneal dissemination.

In conclusion, we have demonstrated that intraperitoneal injection of ets-1 decoy inhibited the peritoneal dissemination of gastric cancer by suppressing tumor angiogenesis in a nude mouse model. This study provides the first evidence of application of decoy as a therapeutic strategy for peritoneal dissemination invivo. These results indicate that the decoy strategy for ets-1 offers a promising therapy for patients with incurable peritoneal dissemination of gastric cancer, in most of which, ets-1 plays an important role in cell growth.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We thank Dr Yutaka Yonemura (Shizuoka Cancer Center, Shizuoka, Japan) for providing MKN-45 P cell line.

References

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References