Tel.: +81-86-235-7257, Fax: +81-86-221-8775
Genetically engineered oncolytic adenovirus induces autophagic cell death through an E2F1-microRNA-7-epidermal growth factor receptor axis
Article first published online: 30 APR 2012
Copyright © 2012 UICC
International Journal of Cancer
Volume 131, Issue 12, pages 2939–2950, 15 December 2012
How to Cite
Tazawa, H., Yano, S., Yoshida, R., Yamasaki, Y., Sasaki, T., Hashimoto, Y., Kuroda, S., Ouchi, M., Onishi, T., Uno, F., Kagawa, S., Urata, Y. and Fujiwara, T. (2012), Genetically engineered oncolytic adenovirus induces autophagic cell death through an E2F1-microRNA-7-epidermal growth factor receptor axis. Int. J. Cancer, 131: 2939–2950. doi: 10.1002/ijc.27589
- Issue published online: 23 OCT 2012
- Article first published online: 30 APR 2012
- Accepted manuscript online: 11 APR 2012 05:26AM EST
- Manuscript Accepted: 13 MAR 2012
- Manuscript Received: 5 AUG 2011
- Japan Science and Technology Agency
- Ministry of Health, Labour, and Welfare of Japan
- Ministry of Education, Culture, Sports, Science and Technology, Japan
- Top of page
- Material and Methods
- Supporting Information
Autophagy is known to have a cytoprotective role under various cellular stresses; however, it also results in robust cell death as an important safeguard mechanism that protects the organism against invading pathogens and unwanted cancer cells. Autophagy is regulated by cell signalling including microRNA (miRNA), a post-transcriptional regulator of gene expression. Here, we show that genetically engineered telomerase-specific oncolytic adenovirus induced miR-7 expression, which is significantly associated with its cytopathic activity in human cancer cells. Virus-mediated miR-7 upregulation depended on enhanced expression of the E2F1 protein. Ectopic expression of miR-7 suppressed cell viability and induced autophagy by inhibiting epidermal growth factor receptor (EGFR) expression. Our results suggest that oncolytic adenovirus induces autophagic cell death through an E2F1-miR-7-EGFR pathway in human cancer cells, providing a novel insight into the molecular mechanism of an anticancer virotherapy.
Autophagy is well known to have a cytoprotective role and to contribute to the maintenance of cell survival under various cellular stresses, such as deprivation of nutrients,1 hypoxia2 and interruption of growth signaling.3 The autophagic process has also been associated with the inhibition of tumor development. In fact, it has been reported that adenovirus infection induces autophagy-related cell death in infected cancer cells, leading to tumor suppression.4–6 Furthermore, oncolytic adenoviruses induce autophagic cell death in human malignant glioma cells7, 8 and in brain tumor stem cells.9 However, the molecular mechanism underlying virus infection-mediated autophagic cell death remains unclear.
MicroRNA (miRNA) is a small noncoding RNA consisting of 22 nucleotides, which post-transcriptionally suppresses the expression of many target genes by pairing with complementary nucleotide sequences in the 3′-untranslated regions of the target mRNA. A number of reports have indicated that miRNA can regulate diverse cell fates including cell proliferation,10 the epithelial-mesenchymal transition,11 apoptosis12 and senescence13 in cancer cells. Recently, Zhu et al. demonstrated inhibition of autophagy by miR-30a, which suppresses the autophagy-related beclin 1 gene in human cancer cells,14 suggesting the possible regulation of autophagy in cancer cells by miRNA. In addition, the Epstein-Barr virus15 and the human cytomegalovirus16 have been reported to modulate cellular miRNA expression in normal infected cells. The adenoviral E1A protein also downregulated miR-520h expression, resulting in an antitumor effect.17
These observations led us to examine whether genetically engineered telomerase-specific oncolytic adenovirus modulate cellular miRNA expression in human cancer cells. We previously developed an oncolytic adenovirus, OBP-301, which drives the expression of viral E1A and E1B genes linked with an internal ribosome entry site under the control of the human telomerase reverse transcriptase (hTERT) promoter for virus replication and, therefore, induces oncolytic cell death in human cancer cells with high telomerase activity, but not in human normal cells without telomerase activity.18 OBP-301 has an antitumor effect against a variety of human cancer cells in both in vitro and in vivo settings.18, 19 In this study, we investigated whether OBP-301- and wild-type adenovirus-mediated cytopathic activities are associated with autophagy induction in human cancer and normal cells. To address the molecular mechanism on the oncolytic adenovirus-induced autophagy, we assessed the global miRNA modulation in the infected cells and identified the miRNA-based autophagy induction system during adenovirus infection.
Material and Methods
- Top of page
- Material and Methods
- Supporting Information
The human nonsmall cell lung cancer cell lines H1299 and A549 were obtained from the American Type Culture Collection (Manassas, VA). The human esophageal cancer cell line T.Tn was purchased from Japanese Collection Research Bioresources (Osaka, Japan). The human normal lung fibroblast cell line NHLF was obtained from TaKaRa Biomedicals (Kyoto, Japan). The H1299 and T.Tn cells were maintained in RPMI 1640 medium, and A549 cells were maintained in Dulbecco's modified Eagle's medium containing a Nutrient Mixture (Ham's F-12). All media were supplemented with 10% fetal bovine serum, 100 U/ml penicillin and 100 mg/ml streptomycin. NHLF cells were cultured in the medium recommended by the manufacturer. The cells were routinely maintained at 37°C in a humidified atmosphere with 5% CO2.
Construction and characterization of the recombinant tumor-specific replication-selective adenovirus vector OBP-301 (Telomelysin) was previously reported.18, 19 Ad5 was the basal adenovirus for OBP-301 and was also used as another type of replication-competent adenovirus. Replication-deficient adenoviral vectors expressing E2F1 (Ad-E2F1) were used to induce E2F1 expression in infected cells, as previously reported.20 OBP-301, Ad5 and Ad-E2F1 were purified using CsCl step gradient ultracentrifugation followed by CsCl linear gradient ultracentrifugation.
Infection of cells with OBP-301 or Ad5 and cell viability assay
Cells were seeded on 96-well plates at a density of 1 × 103 cells/well 12 hr before infection and were infected with OBP-301 or Ad5 at MOIs of 0, 1, 5, 10, 50 and 100 plaque-forming units/cell. Cell viability was determined on day 3 after infection using the Cell Proliferation Kit II (Roche Molecular Biochemicals, Indianapolis, IN) according to the manufacturer's protocol.
Transfection of cells with siRNA or miRNA and cell viability assay
Cells seeded at a density of 5 × 102 cells/well in 96-well plates were transfected with either p62 siRNA (Applied Biosystems, Foster City, CA) or with control siRNA (Applied Biosystems) at a concentration of 0, 1, 5 or 10 nM using HiPerfect transfection reagents (Qiagen, Valencia, CA). miR-7 (Ambion, Austin, TX) or control miRNA (Ambion) was also transfected at the same concentrations. In contrast, EGFR siRNA (Applied Biosystems) or control siRNA (Applied Biosystems) was treated at a concentration of 0, 10 and 50 nM. Cells were pretreated with 3-methyladenine (3-MA) (200 nM) (Sigma–Aldrich, St. Louis, MO) for 2 hr before transfection to inhibit miR-7-mediated autophagy. Cell viability was determined on day 6 after transfection using the Cell Proliferation Kit II (Roche Molecular Biochemicals).
Western blot analysis
Cells were seeded in a 100-mm dish at a density of 1 × 105 cells/dish 12 hr before transfection and were transfected with either miR-7 (Ambion) or with control miRNA (Ambion) at a concentration of 10 nM, or were infected with OBP-301 at the indicated MOIs. On day 3 after treatment, whole cell lysates were prepared in a lysis buffer (50 mM Tris–HCl (pH 7.4), 150 mM NaCl and 1% Triton X-100) containing a protease inhibitor cocktail (Complete Mini; Roche). Proteins were electrophoresed on 6–15% SDS polyacrylamide gels and were transferred to polyvinylidene difluoride membranes (Hybond-P; GE Health Care, Buckinghamshire, UK). Blots were blocked with 5% nonfat dry milk in TBS-T (Tris-buffered saline and 0.1% Tween-20, pH 7.4) at room temperature for 30 min. The primary antibodies used were: rabbit antimicrotubule-associated protein 1 light chain 3 (LC3) polyclonal antibody (pAb) (Medical & Biological Laboratories (MBL), Nagoya, Japan), rabbit anti-Atg5 pAb (Cosmo Bio, Tokyo, Japan), mouse anti-p62 monoclonal antibody (mAb) (MBL), mouse anti-Ad5 E1A mAb (BD PharMingen, Franklin Lakes, NJ), rabbit anti-E2F1 pAb (Santa Cruz Biotechnology, Santa Cruz, CA), goat anti-wild-type EGFR pAb (R&D Systems Inc., Minneapolis, MN) and mouse anti-β-actin mAb (Sigma-Aldrich). The secondary antibodies used were: horseradish peroxidase-conjugated antibodies against rabbit IgG (GE Healthcare), mouse IgG (GE Healthcare) or goat IgG (Chemicon International Inc., Temecula, CA). Immunoreactive bands on the blots were visualized using enhanced chemiluminescence substrates (ECL Plus; GE Healthcare).
Quantitative real-time reverse transcription-PCR analysis
Cells were seeded on six-well plates at a density of 3 × 104 cells/well 12 hr before infection and were infected with either OBP-301 or with Ad5 at the indicated MOIs. Total RNA was extracted from cells using a miRNeasy Mini kit (Qiagen). Total RNA was extracted in dose-dependent experiments from cells infected at the indicated MOIs on day 3 after infection, and in time-course experiments from cells on days 0, 1, 2, 3 and 4 after infection. cDNA was synthesized from 10 ng of total RNA using the TaqMan MicroRNA Reverse Transcription kit (Applied Biosystems), and quantitative real-time RT-PCR was performed using the Applied Biosystems StepOnePlusTM real-time PCR system. The expression of miR-7 was defined from the threshold cycle (Ct), and relative expression levels were calculated using the 2−ΔΔCt method after normalization with reference to the expression of U6 snRNA.
The cells were seeded in 75T flasks at a density of 2.0 × 105 cells/flask 12 hr before infection and were infected with either OBP-301 or Ad5 using an MOI of 5. Total RNA, including miRNA, was extracted from the OBP-301-infected, Ad5-infected and mock-infected cells on day 3 after infection using a miRNeasy Mini kit (Qiagen) according to the manufacturer's protocol, and RNA concentrations were quantified using a NanoDrop spectrophotometer. The RNA samples were then used for microarray analysis, which was performed by Exiqon (Vedbaek, Denmark) (http://www.exiqon.com/). For this analysis, each RNA sample and a mixture of all samples were labeled with Hy3 or Hy5, respectively, and were hybridized with three dual-color miRNA microarray chips (miRCURY™ LNA Array version 10.0; Exiqon) in which 719 kinds of human miRNA probes were contained. Fifteen miRNAs showed more than a 50% difference in expression between the OBP-301- or Ad5-infected cells and the mock-infected cells (Supporting Information Fig. 4a). The expression levels of miR-33a, miR-183, miR-483-3p and miR-7 were evaluated using real-time RT-PCR, as described above.
Infection of cells with E2F1-expressing adenoviral vectors and treatment with E2F1 siRNA
H1299 and A549 cells, seeded at a density of 3 × 104 cells/well in six-well plates, were infected with Ad-E2F1 at an MOI of 100 for 2 days. The same cell lines, seeded at the same density in six-well plates, were transfected with E2F1 siRNA (Applied Biosystems) or control siRNA (Applied Biosystems) at a concentration of 10 nM and, 24 hr later, were infected with 5 or 50 MOI (H1299 and A549 cells, respectively) of OBP-301 for 3 days. Total RNA and whole cell lysates were prepared from the infected cells, and the expression levels of miR-7 and E2F1 were analyzed using real-time RT-PCR and western blotting, respectively.
Determination of autophagic cells using H1299-GFP-LC3 cells
H1299 cells stably transfected with GFP and LC3 fusion plasmid (GFP-LC3) were previously established.9 After transfection with 50 nM miR-7 (Ambion) or control miRNA (Ambion), GFP expression in the transfected cells was examined using a laser confocal microscope (Fluoview 300; Olympus, Tokyo, Japan). As a positive control, H1299-GFP-LC3 cells were serum-starved by culture in Hank's balanced salt solution for 4 hr before laser confocal microscopy (Olympus).
A549 cells, seeded at a density of 1 × 105 cells/dish in 100-mm dishes, were transfected with either 10 nM miR-7 (Ambion) or control miRNA (Ambion) for 3 days. Following staining with Acridine Orange solution (1.0 μg/ml; Sigma–Aldrich) for 15 min, the cells were trypsinized and were analyzed using a flow cytometer (FACSArray; Becton Dickinson, San Jose, CA).
Determination of significant differences was assessed using Student's t-test. Correlations between the expression levels of miR-7, the cytopathic activity of OBP-301 and the expression level of EGFR were analyzed using Pearson's correlation coefficient. p < 0.05 was considered significant.
- Top of page
- Material and Methods
- Supporting Information
The cytopathic effect of the oncolytic adenovirus OBP-301 is associated with induction of autophagy in human cancer cells
To investigate if the cytopathic effect of OBP-301 correlates with autophagy in human cancer cells, we used three human cancer cell lines (H1299, A549 and T.Tn), that showed different sensitivities to OBP-301.19, 21 The cytopathic effect of OBP-301 against each cell line was determined by assay of cell viability using the XTT assay (Fig. 1a). The H1299 and A549 cells showed high and moderate sensitivities, respectively, to OBP-301, but T.Tn cells were resistant. T.Tn cells showed lower expression level of coxsackie and adenovirus receptor (CAR) protein, but similar hTERT mRNA expression compared to H1299 and A549 cells (Supporting Information Figs. 1a and 2a). Consistent with CAR expression, T.Tn cells were less sensitive to adenovirus-mediated green fluorescent protein (GFP) induction compared to H1299 and A549 cells (Supporting Information Fig. 1b). In spite of high hTERT expression, the replication rate of OBP-301 was suppressed in T.Tn cells compared to H1299 and A549 cells (Supporting Information Fig. 2b). These results suggest that its resistance was due to impairment of virus infection and replication. Furthermore, as OBP-301 shows the tumor-specific cytopathic effect in a telomerase-dependent manner, the cell viability of human normal fibroblasts (NHLF), which show low CAR expression and no hTERT mRNA expression (Supporting Information Figs. 1a and 2a), was also determined after infection with OBP-301. As reported previously,18 NHLF cells showed the resistance to OBP-301-mediated cytopathic effect (Fig. 1a). The cytopathic activity and replication rate of wild-type adenovirus serotype 5 (Ad5) were also confirmed in all cell lines (Supporting Information Figs. 2 and 3). The data for the sensitivities to OBP-301 and Ad5 and expression levels of CAR and hTERT were summarized in Supporting Information Table 1.
Oncolytic adenovirus-mediated autophagy can be characterized by conversion of the microtubule-associated protein 1 light chain 3 (LC3)-I to the LC3-II form,7, 22, 23 by upregulation of the autophagy-related protein Atg59, 23 and by downregulation of the p62 protein.23 Therefore, to analyze OBP-301 induction of autophagy, we determined the expression levels of LC3-I/II, Atg5 and p62 proteins in OBP-301-infected cells by Western blot analysis (Fig. 1b). OBP-301-sensitive H1299 and A549 cells exhibited conversion of LC3-I to LC3-II, Atg5 upregulation and p62 downregulation after OBP-301 infection using more than five multiplicity of infections (MOIs). However, the OBP-301-resistant T.Tn cells and NHLF cells showed no induction of autophagy. Adenoviral E1A expression was detected in all four cell lines after infection with OBP-301. To further evaluate the relationship between OBP-301-induced autophagy and cytopathic effect, OBP-301-sensitive cells were infected with OBP-301 at an MOI of 50, and the morphological and autophagic changes were analyzed at 0, 24, 48 and 72 hr after infection using conventional microscopy, Western blot analysis and electron microscopy (Supporting Information Figs. 3 and 4). Although no morphological and autophagic changes were observed at 24 hr, the detachment of round shaped cells and autophagic markers were induced at 48 hr and more increased at 72 hr after infection. At 72 hr after infection, autophagic vesicles were also confirmed in the OBP-301-infected H1299 and A549 cells under electron microscopy. Furthermore, we found the significant correlations between cytopathic effect of OBP-301 and autophagy-related markers, such as LC3-II/LC3-I ratio and expressions of Atg5 and p62 (Fig. 1c). These results suggest that there is a relationship between the cytopathic activity of OBP-301 and induction of autophagy.
Autophagy is well known to show both cytoprotective and cytopathic effects in human cancer cells. Autophagy has recently been shown to suppress tumorigenesis through p62 downregulation.24 Furthermore, the accumulation of p62 proteins has been shown to be a critical factor for the survival of human cancer cells.25 We, therefore, determined if OBP-301-mediated autophagy, which leads to downregulation of p62, is associated with cell survival or cell death. For this purpose, we analyzed the effect of siRNA-mediated downregulation of p62, which mimics OBP-301-mediated p62 downregulation, on the viability of noninfected H1299, A549 and T.Tn cells (Supporting Information Fig. 5). Using Western blotting analysis, we first confirmed complete suppression of p62 protein expression by transfection of 10 nM p62 siRNA in all cell lines (Supporting Information Fig. 5a). Treatment with p62 siRNA significantly suppressed cell viability in all cell lines in a dose-dependent manner (Supporting Information Fig. 5b), suggesting that OBP-301-mediated downregulation of p62 induces cell death. In contrast to OBP-301-induced autophagic cell death, no apoptotic cell death, assessed by western blotting of caspase-3 cleavage, was observed in OBP-301-infected H1299 and A549 cells (Supporting Information Fig. 6). These results suggest that the cytopathic effect of OBP-301 is associated with autophagy-related cell death.
OBP-301 infection modulates miRNA expression in human cancer cells
To next investigate if OBP-301 induces autophagic cell death through modulation of miRNA expression in human cancer cells, OBP-301-sensitive H1299 cells were infected with OBP-301, and miRNA expression levels in the OBP-301-infected and mock-infected cells were analyzed using a miRNA microarray. Because wild-type Ad5 is the virus from which OBP-301 was generated, an Ad5-infected H1299 cell extract was also analyzed to clarify the candidate miRNAs modulated by infection with OBP-301 and/or Ad5. Fifteen miRNAs showed differences in expression that were higher than 50% in the OBP-301-treated and/or Ad5-treated cells compared to mock-treated cells (Supporting Information Fig. 7a). Of these 15 miRNAs, four miRNAs were downregulated and 11 miRNAs were upregulated. To further validate OBP-301-mediated modulation of miRNA expression, we further analyzed four miRNAs; two downregulated miRNAs (miR-33a and miR-183) and two upregulated miRNAs (miR-483-3p and miR-7), using same three RNA samples used for miRNA microarray by quantitative real-time RT-PCR (qRT-PCR) (Supporting Information Fig. 7b). Of these four miRNAs, the expression of miR-7 was upregulated 2.94-fold and 1.91-fold in the OBP-301-treated and Ad5-treated cells, respectively, compared to mock-treated cells. This result for miR-7 was consistent with the microarray data, whereas other three miRNAs showed different expression levels between microarray and qRT-PCR. Therefore, for further analysis, we focused on the role of miR-7 in OBP-301-mediated oncolytic cell death.
miR-7 upregulation is associated with the cytopathic activity of OBP-301
To further confirm OBP-301-mediated miR-7 upregulation, OBP-301-sensitive (H1299 and A549) and OBP-301-resistant (T.Tn and NHLF) cells were infected with OBP-301 at various MOIs, and the expression level of miR-7 was examined using qRT-PCR. miR-7 expression was dose-dependently upregulated in the OBP-301-infected H1299 and A549 cells, whereas T.Tn and NHLF cells showed no change in miR-7 expression after OBP-301 infection (Fig. 2a). Time-dependent upregulation of miR-7 expression was also observed in H1299 and A549 cells infected with OBP-301 at 5 and 50 MOIs, respectively (Supporting Information Fig. 8a). Furthermore, the level of miR-7 upregulation after OBP-301 infection significantly correlated with the cytopathic activity of OBP-301 (r = 0.954, p = 5.78E-13) (Supporting Information Fig. 8b). Similar to OBP-301, Ad5 infection also dose-dependently upregulated miR-7 expression, and this upregulation significantly correlated with the cytopathic activity of Ad5 (r = 0.933, p = 8.94E-6) (Supporting Information Fig. 9). These results suggest that miR-7 upregulation is implicated in oncolytic adenovirus-mediated cell death.
E2F1 activation is involved in OBP-301-mediated miR-7 upregulation
Adenovirus infection has been shown to modulate many kinds of protein-coding genes through activation of the transcription factor, E2F1, induced by adenoviral E1A26 and E4.27, 28 Furthermore, it has recently been shown that E2F1 regulates the expression of specific miRNAs in a transcription-dependent manner.29 Therefore, we sought to assess the role of E2F1 in OBP-301-mediated miR-7 upregulation. The Western blotting analysis revealed that OBP-301 infection at MOIs greater than five induced E2F1 protein expression in OBP-301-sensitive H1299 and A549 cells but not in OBP-301-resistant T.Tn cells (Fig. 2b). In contrast, NHLF cells showed slight increase in E2F1 expression after infection with high dose (more than 50 MOI) of OBP-301.The level of miR-7 upregulation in these cells significantly correlated with the level of E2F1 expression (r = 0.944, p = 4.48E-12) (Supporting Information Fig. 8c). Furthermore, to investigate the E2F1-mediated miR-7 upregulation, H1299 and A549 cells were infected with or without an E2F1-expressing replication-deficient adenoviral vector (Ad-E2F1) (100 MOI) for 48 hr. However, E1A-deleted control adenovirus was not used because other genes from E4 region might contribute to E2F1 activation.27, 28 Ectopic expression of E2F1 by infection with Ad-E2F1 significantly upregulated miR-7 expression 2.48- and 1.96-fold in H1299 and A549 cells, respectively, compared to mock infection (Fig. 2c). Overexpression of the E2F1 protein by Ad-E2F1 infection was confirmed by Western blot analysis. Conversely, specific downregulation of E2F1 by pretransfection of E2F1 siRNA (10 nM) significantly suppressed the level of miR-7 expression compared to mock or control siRNA treatment in both mock-infected and OBP-301-infected cells at 72 hr after OBP-301 infection following to siRNA treatment for 24 hr (Fig. 2d). Suppression of OBP-301-activation of E2F1 expression by pretreatment with E2F1 siRNA was also confirmed by Western blot analysis. These results suggest that OBP-301-mediated miR-7 upregulation depends mainly on activation of E2F1 expression.
miR-7 overexpression suppresses cell viability through induction of autophagy in human cancer cells
To determine if OBP-301-mediated miR-7 overexpression is associated with cell death in human cancer cells, we introduced exogenous miR-7, or control miRNA, into OBP-301-sensitive H1299, A549 and T.Tn cells and investigated the effect of miR-7 overexpression on cell viability (Fig. 3a and Supporting Information Fig. 10a). Ectopic expression of miR-7 significantly suppressed cell viability in a dose-dependent manner compared to control miRNA in H1299 and A549 cells. However, T.Tn cells showed less sensitivity to miR-7-mediated suppression of cell viability compared to H1299 and A549 cells. In contrast, miR-7-overexpression did not induce apoptotic cell death (caspase-3 cleavage) in any of these cells (Supporting Information Fig. 11). These results indicate that miR-7 overexpression suppresses cell viability through induction of nonapoptotic cell death in human cancer cells.
The observation of miR-7-mediated nonapoptotic cell death prompted us to investigate if miR-7 overexpression induces autophagic cell death, because we had previously shown that autophagy may be involved in OBP-301-mediated oncolysis of H1299 cells30 and observed the significant correlation between cytopathic activity of OBP-301 and autophagy induction (Fig. 1). To investigate miR-7-mediated induction of autophagy in cells, we used H1299 cells that were stably transfected with a GFP-LC3 fusion plasmid (GFP-LC3). miR-7 overexpression in H1299-GFP-LC3 cells induced the appearance of autophagic cells with a punctate pattern of GFP-LC3 expression in the cytoplasm, similar to that observed in serum-starved cells (Fig. 3b). To quantify autophagy induced by miR-7, A549 cells that were transfected with miR-7 or control miRNA were stained with acridine orange. The percentage of cells with stained acidic vesicular organelles (AVOs), which are indicative of autophagy, was then measured using flow cytometry. Transfection of A549 cells with miR-7 significantly increased the percentage of AVO-positive cells compared to control miRNA (Fig. 3c). These results indicate that miR-7 overexpression induces autophagy.
Because OBP-301 infection both upregulates miR-7 and downregulates p62 in association with autophagy, we determined if miR-7 overexpression might induce downregulation of p62 protein expression. miR-7 transfection suppressed p62 expression compared to control miRNA in H1299 and A549 cells as shown by Western blotting (Fig. 3d). These results indicate that miR-7 overexpression induces autophagy, resulting in p62 downregulation, in human cancer cells. To further examine if miR-7 overexpression suppresses cell viability through induction of autophagy, the effect of the autophagy inhibitor, 3-MA, on miR-7-mediated suppression of cell viability was determined. Treatment of H1299 cells with 3-MA significantly attenuated miR-7-mediated suppression of cell viability (Fig. 3e), suggesting that miR-7 does indeed mediate autophagic cell death.
EGFR downregulation by miR-7 overexpression is implicated in the OBP-301-mediated cytopathic effect
Recent evidence had shown that miR-7 functions as a tumor suppressor by suppressing the expression of the epidermal growth factor receptor (EGFR),31, 32 which is strongly associated with tumor progression and poor prognosis in human cancers.33 Furthermore, a recent report has shown that EGFR downregulation by siRNA induces autophagic cell death in human cancer cells.34 We, therefore, next sought to determine if OBP-301 suppresses EGFR expression through miR-7 upregulation. As shown in Figure 4a, OBP-301 infection suppressed EGFR expression in a dose-dependent manner in OBP-301-sensitive H1299 and A549 cells but not in OBP-301-resistant T.Tn and NHLF cells. The level of EGFR suppression was significantly associated with the level of miR-7 upregulation (r = −0.872, p = 2.64E-8) (Fig. 4b) and with the cytopathic activity of OBP-301 (r = -0.826, p = 6.73E-7) (Fig. 4c), suggesting the involvement of miR-7-mediated EGFR suppression in the cytopathic effect of OBP-301. The ectopic expression of miR-7 suppressed EGFR expression compared to control miRNA in H1299 and A549 cells (Fig. 4d). Furthermore, ectopic expression of E2F1 by infection with an Ad-E2F1 also downregulated EGFR expression compared to mock infection in H1299 and A549 cells (Fig. 4e). In contrast, treatment with EGFR siRNA significantly suppressed cell viability compared to control siRNA in H1299 and A549 cells (Fig. 4f). However, miR-7-resistant T.Tn cells showed about fivefold higher expression level of EGFR compared to H1299 and A549 cells (Supporting Information Fig. 10b). Even when T.Tn cells were transfected with miR-7 at 10 nM, high expression levels of EGFR and p62 were maintained. The combined results suggest that OBP-301 infection induces miR-7 expression through E2F1 activation and that E2F1-mediated miR-7 upregulation suppresses EGFR expression, resulting in the induction of autophagy-related cell death (Fig. 4g).
- Top of page
- Material and Methods
- Supporting Information
Tumor-specific replication-competent oncolytic virotherapy is emerging as a promising anticancer therapy for the induction of tumor-specific oncolytic cell death.4 Although the possible involvement of autophagy in oncolytic adenovirus-mediated cell death has recently been suggested,5–9 the molecular mechanism by which autophagic cell death is induced remains to be elucidated. In this study, we demonstrated that infection with the oncolytic adenovirus, OBP-301, upregulated miR-7 expression and that this upregulation was associated with its cytopathic activity in human cancer cells. Furthermore, OBP-301-mediated E2F1 activation was involved in miR-7 upregulation, which subsequently induced autophagy through suppression of EGFR expression in human cancer cells. Adenovirus infection is well known to induce the viral protein-mediated E2F1 activation and subsequent upregulation of many E2F1-target genes.35 Recently, E2F1 has been shown to induce autophagy through upregulation of autophagy-related genes in a transcription-dependent manner.36 In contrast, EGF is known to suppress autophagy through EGFR activation.37 Furthermore, it has been shown that EGFR downregulation by EGFR siRNA causes autophagic cell death in human cancer cells.34 Thus, oncolytic adenoviruses may activate E2F1 expression, resulting in the upregulation of autophagy-related genes and the downregulation of autophagy-suppressing genes via miRNA modulation. Subsequently, autophagy-related programmed cell death is induced.
OBP-301 induced higher levels of autophagy, replication rate and cytopathic activity than Ad5 in H1299 and A549 cells (Fig. 1 and Supporting Information Figs. 2 and 3). Recent report has shown that adenovirus-mediated autophagy induction is associated with viral replication and oncolysis.38 Autophagy inhibitor 3-MA has been suggested to inhibit the replication rate of Ad5 in A549 cells.38 However, our collaborators have shown that pretreatment with 3-MA or Atg5 siRNA did not affect the replication and oncolysis of fiber-modified OBP-301 in human brain tumor cells.22 Recently, we observed that OBP-301 infection upregulated hTERT mRNA expression and, subsequently, showed higher levels of replication rate and oncolysis than Ad5 in human sarcoma cells.39 Therefore, the replication of OBP-301 may be less sensitive to autophagy inhibitor compared to Ad5 because of enhanced viral replication by hTERT promoter activation.
On the molecular mechanism of adenovirus-induced oncolysis, recent report has suggested that adenovirus-mediated autophagy induces caspase-8 activation in association with oncolysis in human leukemia cells and normal fibroblasts.40 Recently, caspase-8 has been shown to be involved in not only apoptosis but also diverse cell fates including autophagy.41 In this study, we observed that oncolytic adenovirus induces autophagic cell death, not apoptotic cell death, in human cancer cells (Fig. 1 and Supporting Information Fig. 6). These results suggest the functional role of caspase-8 in adenovirus-mediated autophagic cell death. Atg5-mediated autophagic cell death has recently been shown to be induced through interaction with Atg5 and Fas-associated protein with death domain (FADD),42 which can also bind with caspase-8.41 Thus, oncolytic adenovirus may contributes to autophagic cell death through activation of Atg5-FADD-caspase-8 network. Furthermore, although transfection with p62 siRNA suppressed p62 expression more strongly than OBP-301, the inhibitory effect of p62 siRNA was lower than OBP-301 in the cell viability of H1299, A549 and T.Tn cells (Fig. 1 and Supporting Information Fig. 5). These results suggest that OBP-301-mediated p62 downregulation not only suppresses oncogenic p62 function but also contributes to autophagy-related cell death.
We demonstrated that OBP-301-mediated activation of E2F1 expression upregulated miR-7 expression in human cancer cells. E2F1 has recently been shown to regulate both oncogenic and tumor-suppressive miRNAs. The cluster of oncogenic miRNAs in the miR-19-72 polycistron has been shown to be upregulated by E2F1.43 In contrast, E2F1-inducible miR-449a/b has been shown to suppress cell proliferation and to induce apoptosis in human cancer cells.44 Furthermore, Brosh et al. have suggested that 15 p53-repressed miRNAs, including the miR-19-72 cluster and miR-7, are possibly regulated by E2F1,45 which is consistent with our results that show E2F1-mediated miR-7 upregulation. We previously reported that p53-inducible miR-34a suppresses E2F1 protein expression, resulting in downregulation of the E2F signaling pathway in human cancer cells.13 These reports suggest possible cross-talk between E2F1, p53 and miRNAs. As adenovirus infection is well known to induce E2F1 expression,35 but to suppress p53 expression,46 the E2F1-inducible miRNA network may mainly assist the induction of autophagic cell death by oncolytic adenoviruses.
OBP-301-resistant T.Tn cells showed no induction of the E2F1-miR-7-EGFR axis, resulting in a lack of OBP-301-mediated autophagic cell death. Adenovirus infection is known to modulate E2F1 expression via two main viral factors, E1A and E4. E1A interacts with the phosphorylated retinoblastoma protein, resulting in the release of free E2F1.26 In contrast, the adenoviral E4 19 kDa protein has been shown to enhance E2F1 protein levels through inhibition of proteasome-mediated E2F1 degradation.27, 28 Although the molecular basis for the lack of OBP-301-mediated E2F1 activation in T.Tn cells remains unclear, the cytopathic effect of an oncolytic adenovirus may mainly depend on E2F1 activation, leading to induction of autophagic cell death via modulation of E2F1-downstream target genes including miRNAs. On the role of another E2F family members during adenovirus infection, recent reports have suggested that adenovirus infection increases the E2F2 expression at the transcriptional level,47 whereas the E2F4 expression is decreased.48 Because it has been known that E2F2 is a transactivator as same as E2F1, but E2F4 functions as a transcriptional repressor, these E2F family members may function to induce the E2F-target gene network. Thus, further studies to address the role of E2F family members in OBP-301-mediated oncolytic cell death are warranted.
It has been recently shown that miR-7 functions as a tumor suppressive miRNA by suppressing the expression of various EGFR signaling-related genes including that of EGFR, insulin receptor substrate-2, Raf1 and p21-activated kinase 1 in human cancer cells.31, 32, 49 Consistent with these results, we observed that ectopic expression of miR-7 suppressed cell proliferation and subsequently induced autophagic cell death through suppression of EGFR expression in human cancer cells. Regarding miR-7-mediated cell death, Webster et al. have suggested that nonapoptotic cell death is induced by miR-7 transfection in human lung cancer A549 cells.31 In contrast, Kefas et al. have shown that miR-7 overexpression induces apoptotic cell death in human glioma cell lines.32 These contradictory results suggest that miR-7-mediation of autophagic cell death may depend on the type of cancer cell in which it is expressed.
Overexpression or amplification of several types of EGFR gene isoforms is frequently observed in human cancers.33 Recently, EGFR-targeting anticancer therapies, such as monoclonal antibodies and small molecule tyrosine kinase inhibitors, have been used to improve the clinical outcome of cancer patients. However, resistance to EGFR-targeting therapies is an issue that needs to be resolved. Furthermore, it has been recently reported that the EGFR regulates glucose transport that is required for the survival of cancer cells in an EGFR-kinase-independent manner.34 This result suggests that not only inhibition of EGFR-kinase activity but also downregulation of the EGFR itself will be required for complete eradication of cancer cells. Recent report has further suggested that combination therapy of EGFR kinase inhibitor erlotinib with autophagy inducer rapamycin synergistically decreased the cell viability through increased autophagy in H1299 and A549 cells.50 Our collaborators have also demonstrated that combination therapy of rapamycin with OBP-301 showed synergistic antitumor effect through activation of autophagy machinery in human brain tumor cells.22 Taking the oncolytic adenovirus-mediated EGFR suppression and autophagy via miR-7 induction into consideration, combination therapy of oncolytic adenoviruses with rapamycin may provide novel anticancer strategies that potentially have antitumor effects against cancer cells that are resistant to EGFR-targeting therapies.
In conclusion, we provide evidence, for the first time, that an oncolytic adenovirus induces autophagic cell death in human cancer cells through induction of miR-7 upregulation via enhancement of E2F1 expression and through suppression of oncogenic EGFR expression. An understanding of oncolytic adenovirus-mediated modulation of the cellular miRNA network would provide novel insights into the antitumor mechanism of oncolytic virotherapy.
- Top of page
- Material and Methods
- Supporting Information
The authors thank Drs. Hideki Matsui, Kazuhito Tomizawa and Yasutomo Nasu for helpful discussions, and Tomoko Sueishi and Mitsuko Yokota for their excellent technical support. This study was supported by grants from the Japan Science and Technology Agency (T.F. and H.T.); by grants from the Ministry of Health, Labour, and Welfare of Japan (T.F.) and by grants from the Ministry of Education, Culture, Sports, Science and Technology, Japan (H.T.). Y. Urata and M. Ouchi are employees of Oncolys BioPharma, Inc., the manufacturer of OBP-301 (Telomelysin).
- Top of page
- Material and Methods
- Supporting Information
- Top of page
- Material and Methods
- Supporting Information
Additional Supporting Information may be found in the online version of this article.
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.